The Persistence of Language

The Persistence of Language
Culture and Language Use

Studies in Anthropological Linguistics
CLU-SAL publishes monographs and edited collections, culturally oriented

grammars and dictionaries in the cross- and interdisciplinary domain of
anthropological linguistics or linguistic anthropology. The series offers a forum for
anthropological research based on knowledge of the native languages of the people
being studied and that linguistic research and grammatical studies must be based
on a deep understanding of the function of speech forms in the speech community
under study.
For an overview of all books published in this series, please see
http://benjamins.com/catalog/clu
Editor
Gunter Senft
Max Planck Institute for Psycholinguistics,
Nijmegen
Volume 8
The Persistence of Language. Constructing and confronting the past and present
in the voices of Jane H. Hill
Edited by Shannon T. Bischoff, Deborah Cole, Amy V. Fountain
and Mizuki Miyashita
The Persistence of Language
Constructing and confronting the past and present
in the voices of Jane H. Hill
Edited by
Shannon T. Bischoff
Indiana University Perdue/University Fort Wayne
Deborah Cole
University of Texas-Pan American
Amy V. Fountain
University of Arizona
Mizuki Miyashita
University of Montana
John Benjamins Publishing Company

Amsterdamâ•›/â•›Philadelphia
TM
The paper used in this publication meets the minimum requirements of
8
theâ•¯American National Standard for Information Sciences – Permanence

of Paper for Printed Library Materials, ansi z39.48-1984.
Library of Congress Cataloging-in-Publication Data
The persistence of language : constructing and confronting the past and present in the
voices of Jane H. Hill / Edited by Shannon T. Bischoff, Deborah Cole, Amy V.
Fountain, Mizuki Miyashita.
p. cm. (Culture and Language Use, issn 1879-5838 ; v. 8)
Includes bibliographical references and index.
1. Language--History. 2. Hill, Jane H. honouree. I. Bischoff, Shannon T., editor of
compilation.
P26.H55P47 2013
497--dc23 2012050682
isbn 978 90 272 0291 8 (Hb ; alk. paper)
isbn 978 90 272 7224 9 (Eb)
© 2013 – John Benjamins B.V.

No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any
other means, without written permission from the publisher.
John Benjamins Publishing Co. · P.O. Box 36224 · 1020 me Amsterdam · The Netherlands
John Benjamins North America · P.O. Box 27519 · Philadelphia pa 19118-0519 · usa
Table of contents
Foreword vii
Kenneth C. Hill
Preface xi
Shannon T. Bischoff, Deborah Cole, Amy V. Fountain & Mizuki Miyashita
Introduction. The persistence of language: Constructing and confronting

the past and the present in the voices of Jane H. Hill xxi
Shannon T. Bischoff, Deborah Cole, Amy V. Fountain & Mizuki Miyashita
section 1. Approaches to the study of the indigenous languages

of the Americas
The diachrony of Ute case-marking 3

T. Givón
Language contact as an inhibitor of sound change: An Athabaskan example 29

Keren Rice
Stress in Yucatec Maya: Syncretism in loan word incorporation as evidence

for stress patterns 53
Emily Kidder
The phonetic correlates of Southern Ute stress 85

Stacey Oberly
Revisiting Tohono O’odham high vowels 107

Colleen M. Fitzgerald
Head-marking inflection and the architecture of grammatical theory:

Evidence from reduplication and compounding in Hiaki (Yaqui) 133
Jason D. Haugen & Heidi Harley
A case-study in grass roots development of web resources for language

workers: The Coeur d’Alene Archive and Online Language
Resources (CAOLR) 175
Shannon Bischoff & Amy Fountain
 The Persistence of Language
section 2. Approaches to the study of voices and ideologies
Language contact, shift, and endangerment – implications for policy
Spanish in contact with indigenous tongues: Changing the tide in favor

of the heritage languages 203
José Antonio Flores Farfán
How can a language with 7 million speakers be endangered? 229

Heidi Orcutt-Gachiri
A documentary ethnography of a Blackfoot language course: Patterns

of variationism and standard in the “organization of diversity” 257
Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
Syncretic speech, linguistic ideology, and intertextuality: Presenting

the Spanish translation of ‘Speaking Mexicano’ in Tlaxcala, Mexico 291
Jacqueline Messing & Ramos Rosales Flores
Racism in discourse – analyses of practice
Narrative discriminations in Central California’s indigenous

narrative traditions: Relativism or (covert) racism? 321
Paul V. Kroskrity
The voice of (White) reason: Enunciations of difference, authorship,

interpellation, and jokes 339
Barbra A. Meek
Double-voicing in the everyday language of Brazilian black activism 365

Jennifer Roth-Gordon & Antonio José B. da Silva
Uptake (un)limited: The mediatization of register shifting and the

maintenance of standard in U.S. public discourse 389
Deborah Cole & Régine Pellicer
The silken cord: An essay in honor of Jane Hill 415

Richard Delgado
Afterword: Jane Hill’s current work 425

Claire Bowern
(On behalf of the Dynamics of Hunter-Gatherer Language Change team:
Patience Epps, Russell Gray, Keith Hunley, Jack Ives & Patrick McConvell)
Language index 431
Subject index 433

Foreword
Kenneth C. Hill
Jane and I have always had trouble with the color word “pink.” For me, “pink” has
always been a pale reddish color, the color associated with clothing and toys for baby
girls. For Jane, this color word also applies to a purplish red, even in a saturated mani-
festation. For me, much of Jane’s “pink” looks very much like “red” and I often have
said something like “look at those pretty red flowers!” only to have Jane, in all sincerity,
say she doesn’t see any red flowers. This has happened often enough that she has taken
me as a sociolinguistic example of the sloppiness of the male color vocabulary. With
work, I have managed to keep things peaceful by consciously remembering to use the
word “pink” for reds that have a touch of purple or, better, using more technical terms
such as “magenta,” where we seem to have no problem.
A couple of years ago I discovered what may be at the root of the problem. It is
not a sociolinguistic matter but rather one of perception. We were walking in Catalina
State Park, near Tucson, looking at spring wild flowers. At one point Jane exclaimed
about the brilliance of a hillside covered with masses of purplish flowers, lupines I
believe. I hadn’t noticed. Not to be confrontational, I acknowledged that this was so,
but to tell the truth, for me the brilliance of the flowers was not remarkably different
from that of their green leaves, also all over the hillside. I have never found purplish
colors to have a brilliance anything like that of reds and yellows. This got me to think-
ing. For Jane, the visual stimulus provided by violet is evidently much more vivid than
it is for me. (I should mention that my color vision is clinically normal, so simple color
differentiation is not at issue.) In the red/purple “pink” color, the purple component
apparently is much more vivid for Jane than it is for me, and thus, for her, merging it
terminologically with “red” is strange.
I take the our differences with “pink” as a metaphor for much of Jane’s work. Jane
sees the same things that most people do, but many of them stand out better for her
than for many others.
Frances Jane Hassler was born in 1939 in Berkeley, California, the oldest of four
children. She was named Frances after her grandmother and was always called Jane –
except when she was naughty, then she was “Frances Jane.” Both her parents had
advanced degrees. Her father was Gerald L. Hassler, a physicist with a Cal Tech Ph.D.,
and her mother was Mildred E. Mathias, a botanist whose Ph.D. was from Washington
University, St. Louis. When Jane was born, her father was working for an oil company
 Kenneth C. Hill
but shortly thereafter, during the war, he was assigned to manage General Aniline
& Film, a company seized by the U.S. government as an enemy asset, and the family
moved to Binghamton, NY. After the war the family returned to California, this time
to the Los Angeles area and Jane’s parents joined the UCLA faculty. Her father was a
lecturer in engineering (Dr. Hassler was overqualified for the salary scale for a profes-
sorial appointment at the time) and her mother became the director of the botanical
garden. Throughout her long and distinguished academic career the botanical garden
always remained dear to her and it is now officially the Mildred E. Mathias Botanical
Garden.
The Hasslers were from southern Missouri. That was in the border area
between North and South and Jane claims ancestors on both sides in the Civil War.
Her grandfather was Robert Lee Hassler, named after the Civil War general for the
South. But in a generational turnaround, Robert Hassler’s first son, Jane’s uncle, was
named Karl Marx Hassler. Uncle Karl, in turn, became the right-wing businessman
of the family.
The Hasslers were a tight-knit family and while Jane was a child, they all –
independently it seems – had moved to California. When I first knew them, Uncle
Karl had just died and at the frequent family gatherings at Jane’s grandmother’s house
in Altadena, Uncle Karl’s absence was almost palpable.
Jane was a precocious child. She was an avid reader by the age of four and she
skipped a grade in elementary school. After her grandfather went blind from glau-
coma, Jane helped him keep up his study of Spanish by reading to him in Spanish, even
though she did not know the language.
After living in Altadena a few years, the family was able to move to West Los
Angeles, within walking distance of the UCLA campus, because they were able to pur-
chase a house from a woman who give them a special price because she believed that
that faculty should be able to live close to campus. Otherwise the neighborhood was
crowded with quite well-to-do people including a number of famous movie stars. Jane
used to babysit for Burt Lancaster.
Jane comes from a long line of secular-minded people. Her grandfather was an
ethical-culturist but when he was a young man he went to many churches because
in the late 19th century in southeast Missouri that was one of the best sources of
intellectual stimulation available. As a continuation of this, when Jane was a child
she attended Sundy school and learned much of the Biblical culture that permeates
American life. Though she has no belief in the supernatural, she regards herself as an
ethnic Christian and happily celebrates Christmas and, incidentally, makes the best
Christmas cookies I know of (also the best cheese cake). She has a low regard for
sanctimony and often remarks of politicians and other publically prominent people
that they didn’t learn their Sunday school lessons. I consider Jane a Sermon-on-the-
Mount Christian. As an anthropologist and simply a considerate good person, she
Foreword 
recognizes that the supernatural is an important part of the belief system of many
people and since those people deserve our respect, their beliefs should also be treated
respectfully.
After high school, Jane spent two years at Reed College in Portland, Oregon, and
then two years at the University of California, Berkeley. At Berkeley she did a massive
senior thesis on the narcotics complex in native California, an anthropological study
of jimson weed and tobacco. This study was a harbinger of things to come. Jane has
repeatedly done studies involving masses of information from disparate sources.
Jane went to graduate school near home, at UCLA. I met her there in 1960 in
Harry Hoijer’s class on historical linguistics. We got married in 1961 and went on our
first field trip to Mexico in January, 1962, to ascertain that Nahuatl was the language
spoken in Ostula, on the coast of Michoacán.
In the summer of 1962, Jane did her field work on Cupeño in Pala, CA, while I
taught introductory Yoruba to Peace Corps volunteers at UCLA.
Our first child was born in November, 1962. The following summer Jane worked
on her dissertation while I did field work on Serrano in Banning, CA.
Our graduate fellowships came to an end the following year and I was advised
that I must seek employment, that any delay in doing so would reflect badly on the
nascent Department of Linguistics which had sponsored my fellowhip. I was able to
get a one-year appointment in the linguistics department at Berkeley.
We were at Berkeley in 1964–65 when the Free Speech Movement erupted. That
was the first of the series of events that made the late 1960s the “Sixties.”
I got a position in the Department of Linguistics at the University of Michigan
and we moved to Ann Arbor.
Jane finished her dissertation on Cupeño in 1966 -I didn’t finish mine until the
following year- but employment at the University of Michigan was not a real option
for her. The old-fashioned minded faculty there dismissed her as simply a faculty wife.
Even years later after she had achieved quite a notable professional profile, I recall a
senior U of M professor politely asking her, at a social event, if she was still working.
She was still working. In 1968 she got a faculty position in the Department of
Anthropology at Wayne State University, where she thrived despite the long commute
to Detroit from Ann Arbor. She even became a labor negotiator. The faculty at Wayne
State were represented by the American Association of University Professors and Jane
was the chief negotiator. She had to leave that role when she was made department
head.
For her sabbatical year in 1974–75, Jane conceived of our project to work on
Mexicano, as the Nahuatl (“Aztec”) language is known by its speakers. I delayed my
own sabbatical a year so as to coordinate sabbaticals with Jane. The project was done
initially with support from the National Endowment for the Humanities. The research
site in central Mexico was chosen not only for the inherently interesting nature of the
 Kenneth C. Hill
project, but also because it was a good place to go with three small children. In fact,
taking the children along was good for the initial field work because they seemed to
demonstrate to the people we interviewed that we were fairly harmless people.
In the late 1970s Jane was head of the Department of Anthropology at Wayne
State University and I was head of the Department of Linguistics at the University of
Michigan. This was a dreadful period of inflation and extreme budgetary contraction.
When I would get home in the evening with horror stories about what the Dean’s office
at Michigan had just done to us, Jane’s horror stories from Wayne State would more
than match mine.
When Jane got the opportunity to move to the University of Arizona, we were
more than ready to move. Arizona was attractive for many reasons, not the least of
which was getting away from administrative duties. Jane moved to Arizona in 1983
and I remained two years more in Ann Arbor so our daughter could complete high
school there. I came to Arizona with no professional position though I soon found
myself immersed in the Hopi dictionary project.
At Arizona Jane became one of the stars of anthropology and linguistics. She
tried to avoid administration but administration found her anyway, and she served
as interim head of Linguistics and more than once acting head of Anthropology. She
was President of the American Anthropological Association, the recipient of several
awards both nationally and within the University of Arizona, and she was named a
Regents’ Professor.
Jane has also been a civic activist. She served on the citizens’ committee having
to do with the placement of telescopes on Mt. Graham. In this capacity she said she
was trying to preserve access to wild places by mobility-impaired old ladies and she
resisted efforts to make Mt. Graham an area closed off to the general public because
of the push to protect a sub-subspecies of red squirrel that had attracted the notice of
environmental activists. Currently she is working with Lend A Hand, a group orga-
nized among several Tucson neighborhoods to help keep seniors in their homes. Jane
provides transportation as to doctor’s appointments, grocery shopping and other
activities for seniors who can’t drive.
Now in retirement, Jane continues active work on research, much of which she
had to postpone because of her university duties. Last summer we celebrated our
fiftieth wedding anniversary.
Kenneth C. Hill
Tucson, January 2, 2012
Preface
Shannon T. Bischoff, Deborah Cole, Amy V. Fountain &

Mizuki Miyashita
As the author of eight books and over one hundred peer reviewed articles Jane
Hill has engaged in and led scholarly conversations across a wide range of social
scientific arenas, enabling interdisciplinary understanding and cross disciplin-
ary connections. She has contributed to the methods and theories in disciplines
as diverse as biological anthropology, historical linguistics, cultural anthropology,
formal phonology, formal syntax, education, applied linguistics, law and of course
linguistic anthropology. She has chaired more than thirty doctoral dissertation
committees and served on more than sixty more, and she has mentored scores of
bright young scholars in anthropology, linguistics and allied disciplines. Her former
students fill positions in numerous departments across the globe and the effects of
her work continue to be widely felt.
The number and prestige of her international and national awards as well as her
membership and leadership roles in various international and national o rganizations
are a further testament to the prominence of her voice in the social s ciences. She has
influenced the social sciences internationally as a fellow of the Royal Anthropological
Institute. In awarding her the Viking Fund Medal in A nthropology, the Wenner-Gren
Foundation wrote about her i nfluence in this way:
Professor Hill has made innovative contributions to several areas of research in
anthropology, including the historical linguistics of the Uto-Aztecan language
family, language contact and multilingualism in the U.S. Southwest and Mexico,
and the way in which popular ideas shape the uses of language in communities in
the Southwest, especially in the construction of white racism.
Her dedication to nurturing anthropology is reflected in Professor Hill’s
service as president of the American Anthropological Association, the Society
for Linguistic Anthropology and the Society for the Study of the Indigenous
Languages of the Americas. She has benefited many organizations in anthropology,
including the Wenner-Gren Foundation, the Linguistic Society of America and
the American Anthropological Association, through painstaking service on
important committees. Professor Hill’s honors include election to the American
Association for the Advancement of Science and the American Academy of Arts
and Sciences.
 Shannon T. Bischoff, Deborah Cole, Amy V. Fountain & Mizuki Miyashita
Even in retirement, Jane continues to garner awards. As we proof this volume, she has
just been elected to the Linguistic Society of America’s 2013 class of fellows. The work in
the pages of this volume pays tribute to Jane by demonstrating the ways her many voices
have shaped social scientific conversation about language for more than forty years.
The editors of this volume are all Jane’s students, and if we were competent Hillians
(and to paraphrase Jane (Hill 2008b: 318)), we would be looking for deep, compelling,
and disarming resemblances among the apparently divergent approaches represented
in this volume. Let us try. The authors of the included chapters have especially tried to
emulate three characteristics of Hill’s voices: (1) Her propensity to connect theory and
practice, linking people and ideas across social scientific inquiry (and thereby making
many friends), (2) Her ability to move seamlessly between a focus on the universal
(the human) and the particular (the cultural), and (3) Her continued commitment to
illuminating individual and collective responsibility in discourse. We take each of these
characteristics briefly in turn.
1. Interdisciplinary conversations: Jane’s many friends
Language permits you to have larger social groups. If you have to make friends by
grooming, it will take a long time to make lots of friends.
(Jane Hill on human evolution in an introductory linguistic
anthropology course, spring semester 1996)
The contributions to this volume reflect Jane’s interdisciplinary approach to linguistic

inquiry. During Jane’s career at the University of Arizona, this approach was in evidence
from the very beginning. In the early 1990’s, Jane was one of the major forces in the
development of the ground breaking joint Ph.D. in anthropology and linguistics at
the University of Arizona – a program that enabled students to pursue a dual degree in
the departments of Anthropology and Linguistics. The development of this program
was a particularly rich and multivocal choice, given Jane’s own history at the university.
Jane joined the anthropology faculty at the University of Arizona in 1983,
intent on continuing her research program on the languages native to the American
Southwest and Mexico (Zepeda & Hill 1998). At that time, the Linguistics department
at the university was still relatively new, having separated from Anthropology as part
of a general trend towards ‘biopsychological linguistics’1 – an approach “allied with
psychology and philosophy” rather than with the anthropological traditions of Boas,
Sapir and Hymes. Hill notes that it was not unheard of for biopsychological linguists
. Perhaps more generally known as ‘Chomskian Linguistics’ or ‘generative linguistics’.

Preface 
at Arizona – and elsewhere – to “flaunt their ignorance of anything cultural” (Zepeda &

Hill 1998: 138).
At the same time, the discipline of anthropology was beginning to turn away from
fieldwork with indigenous populations.
…over the years in the communities they [anthropologists] studied they had
also compiled a dismal record of offenses, ranging from countless instances of
inevitable intercultural clumsiness and individual poor judgment, through racist
and imperialist claims on what was not rightfully theirs, to blatant exploitation,
theft, and fraud motivated by the desire for career success and personal gain.
‘Anthropologists’ became the ‘white men’ indigenous people loved to hate…
(Zepeda & Hill 1998: 137)
In short, (biopsychological) linguists were not studying the language c ommunities

of interest for Jane, because their methodology focused on the ‘native speaker’s
intuition’ – and there were few native speakers from these language groups pursuing
linguistic training. But anthropologists were also withdrawing (or being ejected) from
these research programs, because of concerns over exploitation. Jane has written of
biopsychological and anthropological linguists as “having taken a vow not to read each
other” and as claiming that what the other guys do is as an “irrelevant or even hostile
pursuit” (Hill & Irvine 1992: 2). She notes that this tension was “especially acute at
Arizona” (Zepeda & Hill 1998: 138).
Lesser scholars than Jane may well have given up their research interests and
pursued an easier path, but to all of our great good fortune, Jane continued her
research, and added to it. Soon after Jane’s arrival at the University of Arizona, she
began sitting in on the elementary Tohono O’odham language classes then being
taught by a young graduate student in linguistics, Ofelia Zepeda. Jane and Ofelia’s
collaboration in the study of the Tohono O’odham language community began
almost immediately upon Ofelia’s dissertation defense (Zepeda & Hill 1998), and
the work undertaken by these two brilliant scholars over the intervening years may,
we think, be considered a national treasure.2 This collaboration crossed the divide
between anthropology and linguistics in exquisitely nuanced and interesting ways,
contributing crucial scholarship to both traditions.
The esteem with which the biopsychological linguists held, and continue to
hold, Jane was evidenced in the mid 1980s when she was asked to serve as head
of the Department of Linguistics. Departmental lore suggests that her headship
was a saving grace at a time when the department was still in its early phases of
. We use this particular metaphor in the spirit of Zepeda and Hill 1991, and also in ironic
acknowledgement of the pitfalls of commodification, as outlined in Hill 2002.
 Shannon T. Bischoff, Deborah Cole, Amy V. Fountain & Mizuki Miyashita
evelopment. Jane worked with colleagues in both departments to establish the Joint
d
Ph.D. program in Anthropology and Linguistics – we suspect that she did this both
because of, and in spite of, her own experiences in negotiating that divide. Students
in the Joint Ph.D. program (who include co-editors Cole and Fountain, as well as
contributors Kidder and Meek), have gone on to careers in the biopsychological and
anthropological traditions regardless of the name of the department in which they
ended up landing jobs – and exceptionlessly count Jane as their (our) guiding star.
It is sometimes the case in ‘interdisciplinary’ scholarship that interdisciplinarity
aligns with dilettantism. And dilettantism can breed contempt. This has never been
the case with Jane. Jane reads everybody. She reads all flavors of anthropologists and
all flavors of linguists, not to mention philosophers, educators, and biologists. And
she doesn’t just read them – she internalizes them. She can tell you what they said in
everyday language that summarizes it all in a neat little nutshell, or she can give you
the tripartite comparative analysis in perfect scholar speak. Jane makes lots of friends
by not only reading everybody, but by working to get everybody participating together
in conversation. In the opening sentences of “The grammar of consciousness and the
consciousness of grammar”, she writes:
Sociolinguistics should be a tool for the exploration of the role of human linguistic
capacities in the dynamic of the world system. However, while both the political
economic study of the world system and the structuralist study of language have
made important advances in recent years, there has seemed to be little possibility
of uniting them. (Hill 1985: 725)
This article proposed “one avenue toward such a union” (Hill 1985: 725).
Jane makes friends by listening to, reading and understanding others – across
disciplines, specializations, and time – and then by engaging them in conversa-
tion both scholarly and colloquial. By reading widely and making interdisciplinary
connections, Jane appears to be able to see into the future. You can read Jane’s early
work on apes and language and be perfectly in step with the research programs of
current language evolution scholars, who now have access to much more genetic
and experimental evidence (see Hill 1972, 1974, 1978, 1997). Or you can read Jane’s
work from the mid-80’s on consciousness and grammar (Hill 1985) and get a s uccinct
prophecy and précis of current work by leading philosophers’ on evolution and human
consciousness. Because of her vast interests and intellect, she is able to argue for and
against positions of others in a way that nevertheless recruits their solidarity rather
than incurs their wrath. Jane realized early on that Chomsky’s notion of universal
grammar did a lot to show that widespread ideas about language use and language
users were often fundamentally racist, since we all have the same biological endow-
ments. This realization helped her work to include diverse voices in the conversation
about human language.
Preface 
On one occasion, while waiting for one or another meeting to start, there was
some discussion involving faculty and graduate students from several departments
in the social and behavioral sciences, arguing amiably over their various academic
‘turfs’. After a pause, Jane said something like this: “I don’t know what you’re arguing
about. We’re anthropology – if it’s humans, it’s ours.”3
2. The universal and the particular: Jane’s human voices
Speech is by its very nature repeatable for any purpose, and thus is intrinsically
detachable from any autonomous, individual site of belief or commitment.
Bakhtin (1981) convincingly argues that no level of private commitment permits
speakers to fully purge their words of the traces of history and voices of others.
(Hill 2008a: 118)
Language is both a universal characteristic of the human species and a cultural

phenomenon. Part of Jane’s ability to make so many friends is her ability to stay
grounded in the absolute reality of both of these approaches to human language. Jane
cultivated an understanding of human biological evolution, which influenced her
focus on the human voice as a useful site for investigating both the formal aspects of
language and the ideological aspects of language usage. Jane was known to exhort her
students to “Read your articles!”. Those of us who did cannot begin to purge our words
of the traces of her voice, or of the history of contact and re-contextualization that her
voice represents.
During a lecture in 1996 in which Jane was talking about the Sapir and Whorf
Hypothesis and explaining linguistic relativity (the idea that the language you acquire
can actually alter your perceptual world), she noted that this hypothesis is highly
controversial for semantics and syntax but seems to be true of phonology. Those of
us lucky enough to have experienced language socialization into the culture of lin-
guistic anthropology under Jane’s instruction know the audible sound of Jane’s spoken
voice – the physical gestures that are specific to her – like the intonation contours of
her sentences, her unique discourse marker that sounds something like [aowah], the
way she holds her hand, thumb touching the tips of her index and middle fingers, palm
cupped upward when she is making an important point, and the sound of her laughter
after telling a good joke or giving a funny linguistic example. But it is not these unique
phonological and gestural elements that we have incorporated into our voices. No one
except Jane Hill really sounds like Jane Hill.
. Alas, her precise wording is forgotten. But the point was as represented here.
 Shannon T. Bischoff, Deborah Cole, Amy V. Fountain & Mizuki Miyashita
Rather, our own experience provides evidence that exposure to the particular
semantic and syntactic features of Jane’s voice has indeed affected our perceptual
worlds. For example, Jane’s work on mock Spanish directed our attention to exam-
ples of new words of the form “el noun-o” created by monolingual English speakers.
We notice them regularly now, uncannily often, in television media, for example,
even on programs that have an explicitly inclusive and liberal agenda, and we always
read the negative stereotypes of Spanish speakers that accompany these coinages.
Or sometimes upon encountering some new bit of linguistic input (perhaps an
anecdote from a friend or student, perhaps something read on the internet), we can
hear “Today there is no respect” (Hill 1998: 68). It’s always relevant and perfectly
à propos, p opping out of nowhere and surfacing into our conscious thought. That
sentence “Today there is no respect” which is both Jane’s voice and the voices of
the people she interviewed in Malinche is now a resource of ours, a voice we use to
frame our perceptual worlds.
What persists, what stays with us of Jane’s voice, is the realization that human
voices are immensely complex and important to a complete understanding of the
nature of our species. Jane’s voice reminds us over and over again that language pres-
ents us with a paradoxical truth about ourselves: we are many and we are one. We
may share a universal or proto-grammar but our individual voices and our perceptual
worlds were acquired in specific, multiple, and diverse social interactions. Looked at
one way, we all have the same vocal resources at our disposal – every speaker possesses
phonology, morphology, syntax, and semantics. Looked at another, everyone’s vocal
resources are individually specific, influenced by the sociocultural landscapes in which
we acquired and use our voices.
3. Responsibility in narrative: Jane’s discourses
“I contribute nothing but talk, but boy do I have symbolic capital!”

(Jane Hill, during a lecture on political economy, November 29, 1995)
In their co-authored Introduction to Responsibility and Evidence in Oral Discourse, Hill

and Irvine note that some of the works in that volume (including, of course, Jane’s)
“Call[ed] into question the authoritativeness – and even the possibility – of any strictly
individual voice and point of view” (1993: 2). To look at responsibility, Hill and Irvine
started by looking at speaker roles and “the interactional processes shaping the allo-
cation of responsibility for authorship of a message” (Hill & Irvine 1993: 4). Apply-
ing Bakhtin’s methods of literary analysis to conversational narratives, Hill developed
an approach for revealing how a speaker “claims a moral position among conflicting
ways of speaking, weighted with contradictory ideologies, by distributing these across
a complex of ‘voices’” (Hill 1995: 98).
Preface 
In looking at the way speakers used multiple voices to tell personal stories, Jane
advanced the hypothesis that “consciousness…may lie at the moral center of the
narrative” (Hill 1995: 137). She came to this hypothesis partly by looking at d
ysfluencies.
In “The Voices of Don Gabriel” (Hill 1995) she analyzed the narrator’s dysfluencies, like
hesitations and stammerings, which appeared in two contexts: When Don Gabriel was
doing some kind of accounting that required him to be accurate about a number or a
date and when he found himself needing to talk about business for profit, an ideology
that was in direct conflict with Mexicano culture. Jane wrote (Hill 1995: 137):
Dysfluencies are not easily assigned to narrative art, and it is too easy to adopt
a naïve Freudian approach to dysfluency that assumes that it reveals for us the
presence of an authentic subconscious locus of the affective, the self. But accounting
dysfluencies suggest a different interpretation. The self which produces these is
a responsible self, which attends to precise representation. Thus dysfluencies in
connection with terms for “business” may represent precisely responsibility, a
property not of the emotional unconscious but of the active, choosing consciousness
to which Bakhtin directs our attention…This is an important hypothesis, since
anthropological literature on the self has focused on this as the locus of some
continuity of emotional response and not a continuity of responsibility.
Jane’s hypothesis had important implications for formal linguistics too, which had
argued that hesitations, stammering, and other failures of fluency are uninteresting
because they lay outside of what a formal model of language would (and should) be
able to account for. Jane demonstrated these “uninteresting” bits of linguistic produc-
tion to be an important site for investigating what is perhaps the highest order pat-
terning of individual minds and thought: consciousness. Looking at narrative in this
way could, Jane argued, provide us with “a rigorously empirical investigation of the
‘practice’ of language” that linked “the systemic aspects of language” and “the study of
usage” (Hill 1985: 728).
For Jane, consciousness is the “symbolic practice of a structural position” (Hill
1985: 735). Or to put it slightly differently, the responsible self, the part of the brain that
chooses is consciousness. “Narrative” she wrote “give[s] us evidence of the integrity of
another self, the ‘responsible self ’ which we may call consciousness, and allows us a
privileged glimpse of the moment of ‘active choice’ when this consciousness orients itself
as a voice in a heteroglossic universe” (Hill 1985: 735). And it’s not just that Jane has
shown us where responsibility can be located in a person and how we can identify it in
discourse, she also models for us how to make those active choices in our own work. Jane
takes responsibility, even when she experiences resistance. She writes of her own work:
I have found on many occasions, in teaching and lecturing, that to question the
folk theory of racism elicits from my fellow White Americans a defense of it that
is acutely felt and even angry. To challenge this common sense is to become an
oddball or a divisive radical. (Hill 2008a: 5)
 Shannon T. Bischoff, Deborah Cole, Amy V. Fountain & Mizuki Miyashita
In his book, Freedom Evolves, the premise of which was already foreshadowed in
Jane’s work on Mexicano narratives, the philosopher Daniel Dennett (someone Jane
has also read) argues that consciousness evolves with our ability to make our selves
bigger, to incorporate more and more people, and perhaps eventually animals and
even plants, and the whole of our environment, into our conception of who and
what we are responsible for. Seen in this way, racism is a form of de-evolution, which
makes us smaller by allowing us to exclude other people from our sense of responsi-
bility and other voices from our repertoire of choices. Our ability to discriminate is
highly evolved and finely honed. We know, by the time the person on the other end of
the phone utters the word ‘hello’ who he is, or where she’s from (Purnell et al. 1999).
At least we think we know because we’re also very good at something else, which
is ignoring all the particular acoustic differences of individual voices and lumping
sets of features together. The truth is, the differences between individual voices are
infinite. But our brains don’t, and perhaps can’t, care. Sometimes, when we’re speak-
ing English, for example, we can perceive the voices to be all the same, one big set of
shared linguistic features: English – Until someone’s English doesn’t sound enough
like our own…
That Jane Hill, she’s a tough gal who talks straight talk, asks hard questions, and
speaks truth to power.4 Her self-imposed accountability to herself, to her students, to
academia, and to her culture “embiggens” us (to borrow a word from Lisa Simpson)
to take responsibility in our everyday talk and in our academic practice. Her work
challenges us to come to terms with our responsibility as English speaking elites
for addressing the double standard we apply to voice variation. We expect others to
use and acquire our forms of language (and we’ll continue to deny them access to
resources and enjoy the benefits if they don’t). But we certainly aren’t going to talk like
them (because by the logic of our received language ideologies, that could be offensive,
drawing attention to their lower class and status). In our everyday language usage, we
find ourselves constantly on what Jane has called “a translinguistic battle field upon
which [multiple] ways of speaking struggle for dominance” (Hill 1985: 731). Racism in
language is a way to keep some voices, and thereby some people, out of our heads and
therefore outside of ourselves.
Jane’s work invites us to confront the past and to actively work on constructing the
future. She invites us to be active, responsible choice makers in our selection of voices.
She encourages us to read and listen widely and to incorporate more and more voices
. This sentence riffs on a point Jane makes about “tough guys” in her piece about presidential
promising (Hill 2000).
Preface 
into our own. Doing so may enable us to predict and enact a more equitable future for
all humans.
In the foreword, Ken Hill notes that “Jane sees the same things that most people
do, but many of them stand out better for her than for many others.” And through her
voices, we are able to see things better. Thank you, Jane, for sharing your perceptual
world and your many voices with us.
References
Bakhtin, Mikhail. 1981. The Dialogic Imagination, Michael Holquist (ed.). Austin TX: The
University of Texas Press.
Hill, Jane H. 2008a. The Everyday Language of White Racism. Chichester: Wiley-Blackwell.
Hill, Jane H. 2008b. Otomanguean loan words in Proto-Uto-Aztecan maize vocabulary? In In
Hot Pursuit of Language in Prehistory: Essays in the Four Fields of Anthropology, John G.
Bengtson (ed.), 309–320. Amsterdam: John Benjamins.
Hill, Jane H. 2002. ‘Expert rhetorics’ in advocacy for endangered languages: Who is listening,
and what do they hear? Journal of Linguistic Anthropology 12(2): 119–133.
Hill, Jane H. 2000. Read my article: Language ideology and the overdetermination of promising
in American presidential politics. In Regimes of Language, Paul V. Kroskrity (ed.), 259–292.
Santa Fe NM: SAR Press.
Hill, Jane H. 1998. “Today there is no respect.” Nostalgia, ‘respect’, and oppositional discourse in
Mexicano (Nahuatl) language ideology. In Language Ideologies: Practice and Theory, Bambi
B. Schieffelin, Kathryn Ann Woolard, Paul V. Kroskrity (eds), 68–86. Oxford: OUP.
Hill, Jane H. 1997. Do apes have language? In Research Frontiers in Anthropology, Vol. 4:
Ethnology, Linguistic Anthropology, The Study of Social Problems, C.R. Ember & M. Ember
(eds), 114–132. Englewood Cliffs NJ: Prentice-Hall.
Hill, Jane H. 1995. The voices of Don Gabriel: Responsibility and self in a modern Mexicano
narrative. In The Dialogic Emergence of Culture, Dennis Tedlock (ed.), 97–147. Indiana IN:
University of Indiana Press.
Hill, Jane H. 1985. The grammar of consciousness and the consciousness of grammar. American
Ethnologist 12(4): 725–737.
Hill, Jane H. 1978. Apes and languages. Annual Review of Anthropology 7: 89–112.
Hill, Jane H. 1974. Possible continuity theories of language. Language 50(1): 134–150.
Hill, Jane H. 1972. On the evolutionary foundations of language. Language Evolution 74:
308–317.
Hill, Jane H. & Irvine, Judith. 1993. Introduction. In Responsibility and Evidence in Oral
Discourse, Jane Hill & Judith Irvine (eds), 1–23. Cambridge: CUP.
Purnell, Thomas, Idsardi, William & Baugh, John. 1999. Perceptual and phonetic experiments
on American English dialect identification. Journal of Language and Social Psychology
18(1): 10–30.
Zepeda, Ofelia & Hill, Jane. 1998. Collaborative sociolinguistic research among the Tohono
O’odham. Oral Tradition 13(1): 130–156.
Introduction
The persistence of language: Constructing and
confronting the past and the present in the voices of
Jane H. Hill
Shannon T. Bischoff, Deborah Cole, Amy V. Fountain &

Mizuki Miyashita
The diverse writers and topics of this book are linked together by the thread of Jane
Hill’s voice. This volume consists of two major sections. The first focuses on work
in the area of indigenous languages of the Americas. Jane Hill’s innovative work on
Uto-Aztecan has informed and inspired a generation of linguists working within that
language family and with indigenous languages of the Americas more broadly. These
chapters demonstrate Hill’s influence in the field of historical linguistics and on the
conduct of fieldwork and linguistic analysis. The second section presents work rooted
in or influenced by Hill’s groundbreaking and ongoing work on language ideology,
identity and linguistic racism. These chapters focus on the human voice in contempo-
rary social discourse in a range of geographic settings. Together, the chapters in this
volume demonstrate how Jane’s interdisciplinary and critical approach to the study of
language provides the social scientist with a useful array of tools for reconstructing
and confronting our linguistic past as well as our sociolinguistic present.
1. S
ection 1 – Approaches to the study of the indigenous languages
of the Americas
As noted in the preface, Jane began her scholarly career with research in indigenous
language communities in the Southwestern US and Mexico, and this has remained an
active strand of her research to this day. Her work on the Uto-Aztecan languages and
language family is so pervasive and influential that no responsible scholar of anything
Uto-Aztecan could be ignorant of it. Her work in this area encompasses documentary
linguistics, including a reference grammar of Cupeño (Hill 2005), historical linguistic
analysis reconstructing Proto-Uto-Aztecan and tracing the diffusion of maize through
the new world (Hill 2001, 2008b; Hill & Hill 1968), and generative phonology and
 Shannon T. Bischoff, Deborah Cole, Amy V. Fountain & Mizuki Miyashita
morphology (Hill & Zepeda 1992, 1998). The eight papers in this section build on Hill’s
pioneering work in these areas to reconstruct new knowledge about human history
and to confront some of the theoretical, methodological, and conceptual divisions that
continue to characterize contemporary language science.
Tom (Talmy) Givón is a long-time colleague of Jane Hill’s. Givón’s work is at the
heart of Jane’s scholarly interest in the history of the languages in the Uto-Aztecan
family and the methods and contributions of historical linguistics. He writes about
Ute, a Numic language of the Northern branch of the Uto-Aztecan language family.
Givón situates Ute pronouns within the functional domain of referential coherence.
He analyzes the absolutive grammatical relations of the optional clitic-anaphoric
pronouns, their discourse-functional distribution in text, and their diachronic trajec-
tory along the continuum of the rise of pronominal agreement systems on the verb
(so-called ‘argument indexing’). He argues for historical analysis based on internal
reconstruction, an approach taken to be more helpful than the comparative method for
resolving questions concerning the rise of pronominal systems and similar d iachronic
change.
Keren Rice, another of Hill’s colleagues, demonstrates the importance of
considering both grammatical features and socio-historical factors in r econstructing
and accounting for language change in the Fort Good Hope variety of Dene, also
known as Slavey. Dene is an Athabaskan language of the Mackenzie River valley of
northern Canada. Looking at phonological change and its relation to historical c ontact
between speakers of different varieties, Rice explores how in the past 150 years, the Fort
Good Hope variety of Dene has undergone several phonological shifts that appear to
have been at least partially conditioned by contact, as well as by grammatical pressures.
This analysis is in line with Jane’s recent work on the reconstruction of a Uto-Aztecan
homeland, for example (Hill 2001, 2008b), in which the forces of d iachronic change
and those of borrowing must be carefully teased apart.
Emily Kidder is a student of Jane Hill’s, and is currently completing her
dissertation on Yucatec Mayan under Jane’s mentorship. Kidder’s analysis of p rosody
in Yucatec Mayan also focuses on sound patterns, but in a synchronic frame. Yucatec
Maya, an indigenous language of Mexico’s Yucatán Peninsula, is studied in the
context of an ex-patriot community in San Francisco, California. Kidder’s work con-
vincingly argues for the existence of ‘syncretism’ (Hill 1999) in the manipulation of
pitch, tone, and duration by Yucatec Mayan speakers. In demonstrating how these
speakers navigate on a number of levels between and among identities associated
with Mayan and Spanish, Kidder's research has clear parallels to Jane's work on
Mexicano (Nahuatl).
Stacey Oberly, a member of the Southern Ute nation, came to the University of
Arizona to complete a Master’s degree in Native American Linguistics, with the goal of
assisting her community to promote and revitalize their language. Oberly completed
Introduction 
the MA, and stayed on to earn a Ph.D. as Jane’s student. On leave from her position as
an Assistant Professor in the Department of Linguistics at the University of Arizona,
Oberly spent a year as the Director of the Southern Ute Cultural Department. Oberly’s
chapter looks at several of the same acoustic features as Kidder’s, but in the context of
language documentation and revitalization in Southern Ute. Southern Ute, a variety
of the Numic branch of Northern Uto-Aztecan, is highly endangered. The commu-
nity, however, has an active and ongoing interest in pursuing language revitalization,
a program in which Oberly is also an important force. The analysis Oberly presents
was initiated by her observation that community elders were unhappy with the ways
in which younger speakers, language learners, were pronouncing their Ute words. The
paper is an example of linguistic research being undertaken at the behest of, and put
to use by, the community who needs it – that is, linguistic work in the pursuit of social
justice. Thus, it is a prime example of the type of work that Jane has so brilliantly
supported and promoted throughout her career.
Colleen Fitzgerald was first a student, and is now a colleague, of Jane Hill’s.
Fitzgerald’s chapter, which draws on twenty years of both documentation and the-
oretical work on the Tohono O’odham language, also highlights the importance
of documenting languages. Fitzgerald illustrates this by using a variety of types of
linguistic data from Tohono O’odham, a Uto-Aztecan language belonging to the
Tepiman branch of Southern Uto-Aztecan, spoken in and around Southern Arizona
and into Sonora, Mexico. Fitzgerald makes the case that the various types of language
documentation work undertaken to address community needs is also necessary for
the development of linguistic theory. The enormous usefulness of the vast range of
data collected during the ‘dialect survey’ collaboration between Jane Hill and O felia
Zepeda (Hill & Zepeda n.d.; Hill et al. 1994) is particularly highlighted. Fitzgerald
shows how data collected over multiple generations, and for multiple purposes,
shed new light on the fundamental structure of the phonological inventory of the
language – and in doing so, how this informs linguistic theory as a whole. In addi-
tion, given the context of language endangerment and the role and value of linguistic
documentation, it is timely to present the case for the importance of phonological
documentation both for linguists, and for the communities of speakers of endan-
gered and indigenous languages.
Jason Haugen worked with Jane Hill on his dissertation, and is now a
colleague of Jane’s. Heidi Harley has worked with Jane in a number of capaci-
ties as faculty at the University of Arizona – including the administration of the
Joint Ph.D. in A nthropology and Linguistics. Haugen and Harley’s contribution
illustrates the crucial role that endangered language data play in the development
of b iopsychological linguistic theory. Haugen and Harley’s analysis of Hiaki (also
known as ‘Yaqui’), a critically endangered Southern Uto-Aztecan language of the
Cahita branch spoken in Tucson, Arizona, outlines the important contribution of
 Shannon T. Bischoff, Deborah Cole, Amy V. Fountain & Mizuki Miyashita
the study of the morpho-syntax of that language to the theory of morpho-syntactic

and morpho-phonological interfaces in generative grammar. Within generative
grammar, noun incorporation and other compounding processes have traditionally
been the focus of morpho-syntacticians, while reduplication has been investigated
primarily by morpho-phonologists. Haugen and Harley demonstrate how the inter-
action of these two phenomena in Hiaki has significant implications that go beyond
the narrow concerns of these two sub-domains, bearing much more broadly on the
architecture of grammatical theory.
Jane Hill co-directed Shannon Bischoff 's dissertation, and Amy Fountain had
the honor of having Jane on her dissertation committee. Bischoff and Fountain’s
paper is methodological – f ocusing on the importance of finding ways in which the
members of endangered language communities can directly control the preservation
and presentation of their heritage resources. Bischoff and Fountain note that in the
past ten years much work and funding has been invested in projects that develop
web-based archives for language resources, specifically for endangered languages,
and that often the scope of these projects is identified based on community-external
criteria and funded via grant moneys awarded to ‘experts’ (linguists, anthropologists
and others) who bring their own goals and resources to the project. They find that
there is also much need for a grass roots approach to web-based language archiving,
and argue that such projects are significantly more feasible than is often thought.
The chapter presents the development of a set of web-based language resources for
the Coeur d A’lene language community as a case study to show how such projects
might be undertaken. Coeur d’Alene is a highly endangered Interior Salish language
of the Pacific Northwest that received a relatively large amount of linguistic and
anthropological attention under the Boasian linguist and anthropologist Gladys
Reichard and a group of speakers who worked with her. The Coeur d’Alene Online
Language Resources project was undertaken by two volunteers, with no previous
training or experience in web development, over a six week period. Bischoff and
Fountain align their work with Jane’s concerns for documentary linguistics, endan-
gered language revitalization, and social justice.
2. Section 2 – Approaches to the study of voices and ideologies
Jane’s oeuvre teaches us to confront the past’s power by questioning its right to apply to
the present. One of the key ways she has taught us to do this is by examining language
ideology, which functions in the present by maintaining a historical trajectory that
gives it solidity. The chapters in this section can be sub-divided into two general
foci. The first four look at language contact, shift, and endangerment and following
Jane’s propensity for connecting theory to practice, make specific recommendation
Introduction 
for policy. The final five focus on discourses of racism and inequality in a variety of
sociolinguistic contexts. Like Jane, the authors in this section demonstrate the impor-
tance for both linguists and l aypersons to acquire the analytical tools to recognize and
deconstruct the language ideologies we inherit so we may think, speak and write more
ethically about language and its users.
2.1 Language contact, shift, and endangerment – implications for policy

José Antonio Flores Farfán, a colleague and long time friend of Jane’s and a research
scientist at Centro de Investigaciones y Estudios Superiores en Antropología Social
(CIESAS) in Mexico City, opens this section by exploring the mutual effects that
indigenous languages and Spanish have had (and continue to have) on each other.
Building on Hill and Hill’s (1986) work on Mexicano, which argued for a syncretic
approach to language description and analysis, Flores Farfán presents data from
contact varieties of Nahuatl and Maya (Yucatec), together with (monolingual)
Mexican Spanish varieties. Contact varieties such as these are enmeshed in many
different linguistic ideologies, including an ideology of purism. In these cases, this
has favored language shift. In analyzing his data, Flores Farfán reflects on the vari-
able nature of languages against an ethnocentric idea of a single, abstract entity
called “language” (e.g. Spanish or Nahuatl). He demonstrates how any attempt to
describe and analyze language contact involves dealing with the undeniable pres-
ence of multilingualism. Such attempts also require the deconstruction of the
“monolingual voice” and its accompanying ideologies. Flores Farfán argues that
contact effects must be treated holistically, closing the gap between the historically
separate realms of linguistic analysis (historical, sociolinguistic, formal) to treat
“the ecology of d iscourses in its entirety”. Such an approach, he argues, includes
considering language phenomena from the speakers’ perspectives who play an
active role in situations of language contact by making performative choices to
achieve illocutionary success.
Heidi Orcutt-Gachiri, the last of Jane’s dissertation students to complete the degree
before her retirement, explores the changing landscape of Kenyan trilingualism. She
argues that the stable trilingualism formerly characteristic of Kenyans is not shared
by many of today’s Kenyan youth. Drawing on fieldwork in two secondary schools in
Kenya, Orcutt-Gachiri teases out the complex reasons for the current shift to bilin-
gualism, arguing that this shift is affecting not just smaller indigenous languages, but
much larger lingua francas. Orcutt-Gachiri presents the example of Gikuyu, which has
five million speakers, is widely spoken in important business spheres such as N airobi,
and counts presidents and prestigious scholars as native speakers and advocates.
Where other scholars have been dismissive that such a language could be at risk,
Hill has strongly supported this research noting that the Gikuyu case demonstrates
 Shannon T. Bischoff, Deborah Cole, Amy V. Fountain & Mizuki Miyashita
that the only safe languages are those that have institutional support. Looking at
Kenyan language ideologies in the discourses of nationalism, education, and develop-
ment, Orcutt-Gachiri shows how constellations of language repertoires that exclude
indigenous languages, in particular Gikuyu, have contributed to the striking shift to
bilingualism among young Kenyans. She argues that the complexity of language shift
implies the need for language planners to investigate and understand new reasons
(including new language ideological ones) that languages cease to be spoken if they
hope to make viable recommendations for language policy.
Annabelle Chatsis, Mizuki Miyashita, and Deborah Cole examine a language
ideological contact situation among speakers and heritage learners of Blackfoot, an
Algonquian language spoken in Alberta and Montana. Following Hill’s suggestions
for incorporating the study of language and culture into documentary linguistics
(Hill 2006), the authors document the development and implementation of a
university-level Blackfoot language course where students who are linguistic i nheritors
of Blackfoot encounter multiple varieties of the language, sometimes for the first time.
Their ethnography of classroom discourse reveals how the language classroom creates
an environment where the “ideology of Standard” (Hill 2008a) and variationism
(Kroskrity 2009) compete for dominance in the structuring of classroom practice. In
response to the widely noticed need for ideological clarification in language documen-
tation projects, the authors propose a model for representing ideological competition
and shift. Miyashita and Cole, former students of Jane, along with Chatsis (Miyashita’s
colleague and co-researcher) also draw on Zepeda and Hill (1998) in their explicit
reflexivity about the roles and responsibilities of academics and local language experts
in linguistic research and language pedagogy.
Jacqueline Messing’s chapter returns us to the concept of syncretism brought out
by Flores Farfán in this section’s opening chapter. As a doctoral student of Jane, Messing
returned to the Hills’ fieldsite in Central Mexico, and her chapter presents some of her
own research on Nahuatl in the same region. Messing documents the various language
ideologies surrounding the use of Mexicano and Spanish that were articulated by
participants at a book-launching event (at which Jane and Ken Hill were both present)
for the translation of Hill and Hill’s Speaking Mexicano into Spanish. Like the Hills,
Messing finds that syncretic Mexicano speech (“mixed speech”) still exists within a
local ideological landscape in which legítimo Mexicano – true Mexicano – is ideal-
ized as a form of the native language free from any trace of Spanish (although it is
no longer spoken as such). Her analysis of the various discourses articulated at the
book-launching event illuminates a disjuncture between the interpretation of syncre-
tism by locals, resident-scholars and outsider-scholars, adding an intertextual layer of
complexity to the academic interpretation of purism. Messing argues for a concept of
indigenous/postcolonial bilingualism, a broad language contact phenomenon that is a
direct result of colonialism and in need of further study.
Introduction 
2.2 Racism in (mediatized) discourse – analyses of practice

Paul Kroskrity, a long time colleague of Jane’s, explores mid-20th C. salvage era rep-
resentations of Yokuts and Western Mono narratives in an attempt to understand the
logic used by anthropologists and linguists who tended to characterize these narratives
in a disparaging manner. Scholars, including the anthropologist Anna G ayton and
the linguist Stanley Newman, represented these narratives as monotonous, deficient,
and as generally lacking in artful narrative qualities. Kroskrity explores two possible
explanations for what at first blush appears to be an unusually ethnocentric failure to
appreciate difference and an exercise in producing an aesthetic relativism that does not
explain or understand narrative difference but merely notes its existence. The first –
essentially a historical explanation – focuses on the lack of ethnographic methods, the
historical lack of a scholarly literature on ethnopoetics, and the comparative lack of a
critical literature on literacy available to these scholars. The second invokes Jane Hill’s
notion of covert racism to explain a pattern in which all representations of indigenous
narrative traditions appear to be negative and seem to rely on indirect indexical rela-
tions to stereotypes of primitivity and childishness. In order to assess the value of these
different and perhaps competing explanations, Kroskrity introduces the results of his
own ethnopoetic and ethnographic work on Western Mono.
Barbara Meek, also a former student of Jane’s, extends Hill’s work on White
racism in the U.S. to interrogate how unchanging images of Native Americans help to
maintain our taken for granted views of the American nation. Meek begins by arguing
that the language of White racism and the racialization of language have flourished
in the political economies of settler societies where the evaluation of difference has
had material consequences for the emerging nation-state and its citizenry. Despite
efforts to remedy such evaluations, the maintenance of racialized differences through
language endures in settler contexts because of the underlying economy of persons,
bodies, and attributes, which is driven by the need to recognize difference (the twin
projects of distinction and discrimination) in order to create and to maintain a par-
ticular regime of value, i.e. the status quo. Meek shows how the maintenance of such
regimes happens through subtle forms of authorship in jokes circulated electronically
(and often anonymously). She draws out the political, economic and moral valences of
the linguistic representations of Others in which a perduring discourse of difference
surrounds the representation of Native Americans within a wider representation of
citizenship.
Jennifer Roth-Gordon, Jane’s colleague at the University of Arizona, and J ennifer’s
student, Antonio José B. da Silva, write about the construction of race, racism and r acists
in Brazil. These authors bring together Hill’s work on multivocality, intertextuality, and
the reproduction of white racism to explore the discourse of various racial activism
groups in Brazil. In an era of redemocratization and heightened awareness of unequal
 Shannon T. Bischoff, Deborah Cole, Amy V. Fountain & Mizuki Miyashita
access to social, political, civil, and human rights, Afro-Brazilian activists have made
race a more explicit criterion for negotiating their access to positions of power. The
authors argue that in order to challenge a legacy of racial democracy and “cordial
racism,” these groups seek to raise individual awareness of dominant racial discourses
and provide daily strategies for questioning these often-implicit positions. In this
chapter, they ask: How can we understand the struggle to achieve “racial conscious-
ness” as a struggle that takes place among competing voices within daily discourse? To
answer this question, Roth-Gordon and da Silva bring together research on race-based
community organizations in Salvador in 2009–2010 and fieldwork conducted at the
height of politically conscious Brazilian hip hop in the late 1990s in Rio de Janeiro. The
chapter describes consciousness-raising racial activism as a fundamentally linguistic
struggle in which different voices are performed and (crucially) perceived as indexical
of conflicting ideological views.
Deborah Cole, a former student of Jane’s, and Régine Pellicer, a student of
Deborah’s, extend Hill’s work on language ideology and racism in the U.S. (Hill
2008a) Their chapter analyzes the language panic that followed Hillary Clinton’s
public performance of the gospel song “I Don’t Feel No Ways Tired” at a Black
church in Selma, Alabama during the U.S. presidential campaign in 2007. Build-
ing on previous scholarship that has demonstrated how style-shifting in American
English is evaluated negatively when speakers of varieties of Standard English con-
verge on Other than Standard varieties, the authors demonstrate how the ideologies
of personalism and referentialism (which Hill (2008a) argues support the ideology
of Standard) become denaturalized when Standard speakers perform out-group
adequation. Further, by destabilizing essentialist categorizing, cross-identity register
shifts threaten to denaturalize folk ideologies of language and race. They conclude
that the mainstream media’s defense of the ideology of personalism further obscures
the semiotic p rocesses that maintain unequal access to the public sphere and perpetu-
ate the systematic erasure of Other than Standard voices belonging to audiences who
respond to and evaluate public discourse.
The final chapter in this section, by critical race theorist and law professor
Richard Delgado, examines Official English laws in light of the region’s hidden
history of Latino lynching. Delgado demonstrates how Jane’s work helps us

reconstruct the h idden, unspoken and unheard nature of this history. For nearly a
century, Anglos in the Southwest lynched Latinos, mainly Mexicans, at a rate nearly
equal to that of African Americans. This is not generally known, in part because
accounts of these lynchings appeared in community newspapers in Spanish and
most mainstream historians were English readers. But the Mexican community
memorialized these lynchings in oral stories and corridos and passed them down to
their children. Delgado argues that English Only regimes and suppression of bilin-
gual education and ethnic studies, such as that taking place in Tucson, AZ, ensure
Introduction 
that the connection of Latino youth to these stories of their own past is cut. Unable
to communicate with their grandparents, they grow up with little knowledge of their
own histories, which include lynching, theft of lands, and other atrocities that now
account for their community’s low estate.
3. Closing
Jane Hill’s research career continues to inform work across a great variety of scholarly
disciplines. The afterword by one of Jane’s current collaborators, Claire Bowern
(representing the Dynamics of Hunter-Gatherer Language Change team), reminds
us of the ongoing and widespread applicability of Jane’s research and thinking for
scientific inquiry beyond the domains of linguistics and anthropology. The D ynamics
of Hunter Gatherer Change project, funded by the National Science Foundation, is
currently testing hypotheses about the ways in which language change proceeds in
hunter-gatherer communities worldwide. The team, comprised of linguists, anthro-
pologists, biologists, cognitive scientists, and geneticists, is producing scholarship
in the areas of historical linguistics, the development and distribution of numeral
systems, the study of ethnobiological nomenclature, change in material culture, and
genetic admixture and language/gene coevolution.
We would like to thank all the contributors to this book and all of those who sup-
ported us in producing this work. We would especially like to note that there were a
number of individuals who had wished to contribute to the volume, but due to time
constraints and other commitments, as well as our own oversight, were unable to do
so. We would also like to thank the publishers, who recognized that Jane’s influence
crosses so many boundaries and that it would be challenging to fit this collection into
any one series, subfield, or school of thought. We hope the readers of this book will
be as diverse as the voices who have written it.
References
Hill, Jane H. 2008a. The Everyday Language of White Racism. Chichester: Wiley-Blackwell.
Hill, Jane H. 2008b. Otomanguean loan words in Proto-Uto-Aztecan maize vocabulary? In In
Hot Pursuit of Language in Prehistory: Essays in the Four Fields of Anthropology, John G.
Bengtson (ed.), 309–320. Amsterdam: John Benjamins.
Hill, Jane H. 2005. A Grammar of Cupeño [University of California Publications in Linguistics
136]. Berkeley CA: University of California Press.
Hill, Jane. 2006. The ethnography of language and language documentation, In Essentials of
language documentation, Jost, Gippert, Nikolaus P. Himmelmann and Ulrike Mosel, eds.
DEU: Walter de Gruyter, p. 113–128.
 Shannon T. Bischoff, Deborah Cole, Amy V. Fountain & Mizuki Miyashita
Hill, Jane H. 2001. Proto-Uto-Aztecan: A community of cultivators in central Mexico? American

Anthropologist 103(4): 913–934.
Hill, Jane H. 1999. Syncretism. Journal of Linguistic Anthropology 9(1–2): 244–246.
Hill, Jane H. & Hill, Kenneth C. 1986. Speaking Mexicano: Dynamics of Syncretic Language in
Central Mexico. Tucson AZ: University of Arizona Press.
Hill, Jane H. & Hill, Kenneth C. 1968. Stress in the Cupan (Uto-Aztecan) languages. Interna-
tional Journal of American Linguistics 34(4): 233–241.
Hill, Jane H. & Zepeda, Ofelia. 1998. Tohono O’odham (Papago) plurals. Anthropological
Linguistics 40(1): 1–42.
Hill, Jane H. & Zepeda, Ofelia. 1992. Derived words in Tohono O’odham. International Journal
of American Linguistics 58(4): 355–404.
Hill, Jane H. & Zepeda, Ofelia. No date. Dialect survey data. Ms, University of Arizona.
Hill, Jane H., Zepeda, Ofelia, DuFort, Molly & Belin, Bernice. 1994. Tohono O’odham Vowels.
Ms, University of Arizona.
Kroskrity, Paul. 2009. Embodying the reversal of language shift: Agency, incorporation, and lan-
guage ideological change in the western mono community of central California. In Native
American Language Ideologies: Beliefs, Practices, and Struggles in Indian Country, Paul V.
Kroskrity & Margaret C. Field (eds). Tucson AZ: University of Arizona Press.
Zepeda, Ofelia & Hill, Jane. 1998. Collaborative sociolinguistic research among the Tohono
Zepeda, Ofelia & Hill, Jane. 1991. The condition of Native American languages in the United
States. Diogenes 39(45): 45–65.
section 1
Approaches to the study of the indigenous

languages of the Americas
The diachrony of Ute case-marking
T. Givón
White Cloud Ranch
Ignacio, Colorado
This paper deals with Ute case-marking and its reconstructed diachrony,
demonstrating once again that synchronic data from a single language can
serve, via Internal Reconstruction (IR) and a theoretically informed approach
to grammaticalization, to reconstruct older diachronic states. The synchronic
data of object and genitive marking in Ute yields sufficient clues to reconstruct
two or even three layers of older object and/or genitive markin. The persistent
connection between object and genitive points to several cycles of clausal (or VP)
nominalization, a well-documented phenomenon in Northern Uto-Aztecan
(Numic, Takic, Yaqui, Tarrahumara, Huichol). The post-positional markings
of indirect objects in Ute are also surveyed, showing that the vast majority of
them are verb-derived, through patterns that are consonant with serial-verb
constructions elsewhere.
Keywords: Ute; case-marking; diachrony
1. Orientation
This paper deals with the historical development of two major aspects of Ute case-
marking morphology: First, marking of the core grammatical roles of subject, direct
object and genitive; and second, the post-positions that mark indirect objects. Of the
two main methods most commonly used in historical reconstruction – the compara-
tive method (CM) and internal reconstruction (IR) – I will rely primarily on the latter.
The justification for this preference has been discussed elsewhere.1
2. Subject, object and genitive
2.1 Current state

Ute nouns can appear in two basic forms: (a) the nominative: subject or predicate
and (b) the oblique: object or genitive.
. See in Givón (2000).

 T. Givón
For most nouns, including both the majority that have noun-class suffixes and
the minority that are suffixless, the difference between the two forms appears to be a
trivial phonological adjustment: The de-voicing of the final vowel in the nominative
vs. the full voicing of the final vowel in the oblique. Thus compare:
(1) nominative oblique gloss

mama-chi mama-chi ‘woman’
tʉvʉ-pʉ tʉvʉ-pʉ ‘earth’
ma-sʉʉ-vʉ ma-sʉʉ-vʉ ‘finger’
tʉka-pi tʉka-pi ‘food’
toghoa-vi toghoa-vi ‘rattlesnake’
pɵ�ɵqwa-tʉ pɵ�ɵqwa-tʉ ‘book’
nʉa-rʉ nʉa-rʉ ‘wind’
kúchu kuchu ‘buffalo’
káni kani ‘house’
So far, one may conclude that the nominative vs. oblique contrast in Ute involves the
fortuitous exploitation, thus morphologization, of a phonetically-motivated variation,
the otherwise wide-spread process of devoicing unstressed word-final vowels. How-
ever, we are still left with the question: Why does such de-voicing occur in the subject/
predicate but not in the object/genitive positions?
One way of approaching this question is by noting that in all cases where an extra
suffix follows the noun, be it the plural or possessive pronoun for subjects, or the post-
position for indirect objects, the final vowel remains voiced, the way it is in the direct-
object and genitive cases. Thus compare:
(2) nominative with suffix

Plural: sari-chi ‘dog’ ==> sari-chi-u ‘dogs’
Possessor pronoun: tua-chi ‘child’ ==> tua-chi-n ‘my child’
Post-position: káni ‘house’ ==> kani-vaa-tʉ ‘at the house’
Taking a hint from such cases, one may now formulate a tentative diachronic hypothesis:
Ute had, at an earlier time, an object and/or genitive suffix, which shielded the
word-final vowel from de-voicing by making it non-final, but later disappeared.
But how does one go about substantiating such a hypothesis?
2.2 Traces of the oblique suffix -y/-i

2.2.1 Object or genitive nouns
A small group of suffixless nouns, all clearly native and all ending with the vowel /a/ in
the nominative form, take the suffix -y/i in the oblique form.
The diachrony of Ute case-marking 
(3) nominative oblique

tɵcay�a ‘prairie-dog’ tɵcay�a–y
páa ‘water’ páa-y
kuvúa ‘sharp edge’ kuvúa-y
táa ‘knee’ táa-y
piwa-n ‘my spouse’ piwa-yi-n
pía-�u ‘his/her mother’ pía-y-�u
múa-nʉmʉ ‘our father’ múa-y-nʉmʉ
Another small group of native nouns whose stem ends with /a/ show a substitution of
that final /a/ with /i/ when followed in compounds by either a verb or a noun, as in:
tavá-chi ‘sun’ tav-í-�iní-kya-tʉ ‘sun-made’, ‘sunshine’
sun-o-make-pass-nom
�awá-tʉ ‘long’ �aw-í-tavá-pʉ ‘for a long time’
long-o-day-nom
A number of Spanish-borrowed nouns whose stem ends with /a/ are suffixless in the
nominative but take the suffix -y/-i in the oblique, as in:
(5) nominative oblique form
kava ‘horse’ kava-yi
páana ‘bread’ páana-y
muruuna ‘melon’ murúuna-y
panaana ‘banana’ panaana-y
satiiya ‘watermelon’ satiiya-y
Several English-borrowed nouns with consonant-final stems in the nominative take
the suffix -i in the oblique form, as in:2,3
. All the example below are taken from an oral narrative told by Mollie B. Cloud.
. Abbreviations used here are:
adj = adjective; adv = adverb, adverbial; agt = agent (role); an = animate; ant =
anterior (aspect); asp = aspect; caus = causative; comp = complementizer, complement
clause; conc = closure (aspect); ass = associative (role); ben = benefactive (role); bkgr =
background (aspect); c = conjunction (suffix); caus = causative; comp = complementizer,
complement clause; conc = closure (aspect); dat = dative (role); def = definite; defun =
defunct; dem = demonstrative; dim = diminutive; dir = direction; du = dual (number);
emph = emphasis; excl = exclusive; fut = future (mode); g = genitive, possessor (case); hab
= habitual (aspect); hort = hortartive (mode); imm = immediate (aspect); imp = imperative
(speech-act); inan = inanimate; incep = inceptive (aspect); incl = inclusive; inst =
instrumental (role); int = intensive; invis = invisible; io = indirect object (case); irr =
irrealis (mode); loc = locative (role); mann = manner (role); mass = mass (number); mod =
modal (suffix); n = noun; neg = negative (mode); nom = nominal, nominalizer (suffix); np =
noun phrase; o = object (case); own = possessive-reflexive; par = participle (aspect); part =
 T. Givón
(6) a. …maay-pʉgay-�u doctor-i �uwa-y…

find-rem-him doctor-o that-o
‘…they found the doctor…’
b. …doctor-bag-i-av yáa�wa-vaci…
doctor-bag-o-own carry-asp
‘…he was carrying his doctor-bag…’
c. …�uru branch-i-vaa-tux…
that/o branch-o-loc-to
‘…(he moved) toward the branch…’
d. …�uru Mancos Creek-i-vaa…
that/o Mancos Creek-o-loc
‘…over there in Mancos Creek…’
It is well known that the morphological and phonological treatment of borrowed

nouns often reflects old native patterns.4
2.2.2 Demonstratives and pronouns

Traces of the same oblique suffix -y/i are also found in the demonstratives series,
whenever the stem ends with /a/. Thus consider:

Proximate:

�í-cha ‘this’ (inan.) �i-cha-y

�í-na ‘this’ (an. sg.) �i-na-y

�í-mʉ ‘these’ (an. pl.) �i-mʉ
Visible:
má-rʉ ‘that’ (inan.) ma-rʉ
máa ‘that’ (an. sg.) máa-y
má-mʉ ‘those’ (an. pl.) ma-mʉ
Invisible:

�ú-ru ‘that’ (inan.) �u-ru

�ú-wa ‘that’ (an. sg.) �u-wa-y

�ú-mʉ �u-mʉ
partitive; pass = passive (voice); pat = patient (role); pl = plural (number); poss = possession;
p = post-position; pp = post-positional phrase; pred = predicate (case); q = question (speech-
act); qu = quantity (question); rec = reciprocal (voice); red = reduplicated, repetitive
(aspect); refl = reflexive (voice); rel = relative marker; rem = remote (aspect); sg = singular
(number); s = subject (case); sub = subordinator (suffix); subj = subjunctive (mode, speech-
act); top = topic; v = verb; vis = visible; vp = verb phrase; wh = wh-question pronoun;
1du = first-person dual; 1p = first person plural; 1s = first person singular; 2p = second person
plural; 2s = second person singular; 3p = third person plural; 3s = third person singular.
. See Hyman (1970).
The diachrony of Ute case-marking 
More extensive traces of the same suffix are found in the personal pronoun series,
as in:
nʉ� ‘I’ nʉna-y ‘me’, ‘my’
támi ‘we’ (du, incl.) tami ‘us’, ‘our’
táwi ‘we’ (pl. incl.) tawi ‘us’, ‘our’
nʉmʉ ‘we’ (excl.) nʉmʉ-y ‘us’, ‘our’
�ʉmʉ ‘you’ (sg.) �ʉmʉ-y ‘you’, ‘your’
mʉni ‘you’ (pl.) mʉni ‘you’, ‘your’
máa-s ‘s/he’ (vis.) máa-y-a-s ‘him’, ‘his’, ‘her’
ma-mʉ-s ‘they’ (vis.) ma-mʉ-a-s ‘them’, ‘their’
�u-wa-s ‘s/he’ (invis.) �u-wa-y-a-s ‘him’, ‘his’, ‘her’
�u-mʉ-s ‘they’ (in vis.) �u-mʉ-a-s ‘them’, ‘their’
kʉma-s ‘other’ kʉma-y-a-s ‘other’, ‘other’s’
The only phonetic environment in (8) where the suffix -y is consistently absent is fol-
lowing the vowel /i/, a rather predictable deletion in Ute. The added oblique suffix
-a will be discussed further below. The pronominal suffix -s comes from a different
source and does not concern us here.5 Pronouns, including demonstrative pronouns,
are notorious diachronic graveyards where relics of older case-marking morphology
survive long after they have been lost in nouns (see e.g. English or German). Such
survival is due to the higher usage frequency of pronouns as compared to nouns.6
2.2.3 Nominalized clauses

Another context were the old object suffix -y is found is in several types of subordinate
clauses occupying the object position. Ute subordinate clauses are historically nomi-
nalized, a topic that has been discussed in considerable detail elsewhere.7 Thus, both
object REL-clauses and unreduced V-complements display two residual nominal/non-
finite characteristics: (a) the verb is marked by the nominalizing suffix -na and (b) the
subject appears in the oblique/genitive case
Thus consider:
(9) a. Main clause:
mamach sivaatuchi paqha-vaa-ni
woman/s goat/o kill-irr-fut
‘the woman will butcher the goat’
. The suffix -s is probably originally a clausal conjunction. Its current distribution is rather
complex; see Givón (2011, Chapter 18).
. See Zipf (1935).
. See Givón (2011, Chapters 8, 9, 10, 12, 17).
 T. Givón
b. Object REL-clause:
sivaatuchi mamachi paqha-vaa-na…
goat/s woman/gen kill-irr-nom
‘the goat that the woman will butcher’
(Hist.: ‘the goat of the woman�s future killing’)
c. V-complement:
puchuchugwa-y mamachi sivaatuchi paqha-vaa-na-y
know-imm woman/gen goat/o kill-irr-nom-o
‘(s/he) knows that the woman will butcher the goat’
(Hist.: ‘(s/he) knows the woman�s killing of the goat’)
When a noun-phrase with an object REL-clause is itself the object of a main verb, its
nominalized verb is marked with the object suffix -y, as in:8
(10) a. Indirect-object REL-clause:

…nʉ� �ʉmʉy-rugwa-paa-ni �uru pʉ-aa-m �uní-vaa-na-y…
I/s you/o-give-irr-fut that/o rel-o-with do-irr-nom-o
‘…I will give you what (you shall) to do it with…’
(Hist.: ‘I’ll give you that of your doing it with’)
b. Direct-object REL-clause:
… pʉnikya-pʉga tuachi-u-�u, �uway
…see-rem child-pl-3s he/gen
sinawavi kacha�na-qha-na-y…
Sinawav/gen chew-ant-nom-o
‘…(s/he) saw his children, (those) that Sinawav had chewed…’
(Hist.: ‘(s/he) saw the children of Sinawav’s chewing’)
c. Verb complement:
…púupa-aqh wachʉ-ka-na-y kach-ux puchuchugwa-wa…
manner-it put-pass-nom-o neg-it know-neg
‘…nobody knows the way it was put together…’
(Hist.: ‘nobody knows the way of (someone) putting it together’)
The nominalized clauses in (10) above were historically treated in Ute as objects
of the main verb. The survival of the old object suffix -y is again most consistent
following the vowel /a/, the way it was with nouns. One doesn’t find it following the
subject nominalizer -tʉ, even when the subject REL-clause is the object of the main
. Examples (10a, b, c) below are derived from recorded texts (Givón ed. 1985).
The diachrony of Ute case-marking 
verb. Thus, compare (10c) above with (11) below, in which the pronominal head of the
subject REL-clause is marked as object:
(11) �uway wacʉ-ka-tʉ-�u kachu-u� puchuchugwa-wa
3s/o put-pl-nom-3s neg-3s know-neg
‘nobody knows the one who put it together’
2.2.4 The ‘epenthetic’ suffix -y

Epenthetic consonants, inserted between vowels to break vowel clusters, seem at first
blush phonologically motivated. Quite often, however, a specific epenthetic consonant
traces back to a lost morpheme or part of a morpheme. The epenthetic /n/ in the
English indefinite article a(n) is a well-known case. As elsewhere, morpho-phonemic
variation is a notorious graveyard of old morphemes.9

Ute tense-aspect development is intimately associated with the nominalization
of verbal complements of erstwhile main verbs, verbs that grammaticalize into tense-
aspect-modality markers. In many complex T-A-M suffixes in Ute, one finds the epen-
thetic insertion of -y, particularly following the vowel /a/. As representative examples
taken from recorded texts, consider:
(12) a. …kukwi-kwa-pʉga-y-ku, �uwa-rugwa-pʉga-y-ku…

fill-go-rem-o-sub him/o-give-rem-o-it
‘… when it (the bowl) filled up, (he) gave it to him…’
b. …kh-�ura sinawavi-khu �uru magha-pʉa-y-agha-y-�u…
then-be Sinawav/o-it that/o feed-rem-o-asp-o-3s
‘…then he fed it to Sinawav…’
While the insertion of the epenthetic -y in such contexts is not fully predictable, its
verb-suffix position strongly suggests that it is historically derived from object mark-
ing of nominalized complement clauses. This topic will be discussed in more detail
below.
2.3 The oblique suffix -a

As noted in (8) above, the oblique form of many personal pronouns in Ute, in addition
to the suffix -y, an added suffix -a. In this section I will discuss evidence that points to
earlier use of the suffix -a to mark both the genitive and object grammatical roles. In
this respect, this suffix shows the same double distribution as -y.
. For a discussion of the grammaticalization cycle see Givón (1979), Dahl (2009).
 T. Givón
2.3.1 The suffix -a as a genitive marker

Consider first the partitive-genitive construction, as in:
(13) a. �umʉ-a-tʉ
hose-part-nom
‘one of them’
b. �umʉ-a-tʉ-mʉ
those-part-nom-pl
‘some of them’
c. máamachi-u-a-tʉ
women-pl-part-nom
‘one of the women’
d. nʉmʉy-a-tʉ-mʉ
us-part-nom-pl
‘some of us’
Next, many body-parts as well as other inalienably-possessed nouns, such as kin terms,
often display the suffixal vowel -a when followed by the possessor pronoun, as in:
(14) a. kwasí-vi ==> kwasí-a-�u
tail-nom tail-gen-3s
‘tail’ ‘his/her tail’
b. �ɵɵ-vi ==>
�ɵɵ-a-n
bone-nom bone-gen-my
‘bone’ ‘my bone(s)’
c. �uní-aa-�u
poss-gen-3s
‘his/her intimate possessions’
d. mú-a-n
father-gen-my
‘my father’
e. pí-a-n
mother-gen-my
‘my mother’
f. pʉ-a-�u
kin-gen-3s
‘his/her kin’
g. piw-a-mʉ
spouse-gen-2s
‘your spouse’
In the four kin terms in (14d,e,f,g), the suffix -a has already fused into the noun stem.
The following example from recorded text shows the suffix -a used as the genitive
marker before the object suffix -y following the suffixless noun túku ‘cougar’:
The diachrony of Ute case-marking 
(15) …kh-�ura �uwas �u tuku-a-y piwa-y

then-be he/s the/s cougar/o-gen-o spouse-o
chʉka-khwa-u-pʉga pina-khwa…
lead-go-inc-rem follow-go

‘…then finally he ran away with cougar’s wife…’
The suffix -a is also used in the formation of existential clauses, which are patterned
after alienable possession, with the location being the grammatical ‘possessor’.10 Thus
compare:
(16) a. Inalienable possession:
kani-gya-tʉ
house-have-nom/s
‘a home-owner’, ‘s/he has a house’
b. Existential-locative:
�i-vaa �ava�na-tʉ kani-aa-gha-tʉ
here-loc many-nom/o house-gen-have-nom
‘here there are many houses’, ‘this place has many houses’
c. Alienable possession:
�ava�na-tʉ kani �uni-aa-gha-tʉ
many-nom/o house/o do-gen-have-nom
‘(s/he) owns many houses’
2.3.2 The suffix -a as object marker

Consider the comparative pronouns meaning, roughly, ‘one like that’ or ‘some like
that’. When such an expression assumes the object role, it is marked by the suffix -a:
(17) subject object
a. �u-ra-tʉ-ni b. �u-ra-tʉ-aa-ni
that-be-nom-like that-be-nom-o-like
‘one like that’ ‘one like that’
c. �u-ra-tʉ-mʉ-ni d. �u-ra-tʉ-mʉ-aa-ni
that-be-nom-pl-like that-be-nom-pl-o-like
‘some like that’ ‘some like that’
e. �uwa-ra-tʉ-ni f.
�uwa-ra-tʉ-aa-ni
him-be-nom-like him-be-nom-o-like
‘one like him’ ‘one like him’
g. �uwa-ra-tʉ-mʉ-ni h.
�uwa-ra-tʉ-mʉ-aa-ni
him-be-nom-pl-like him-be-nom-pl-o-like
‘some like him’ ‘ones like him’
. See Givón (2011, Chapter 15).

 T. Givón
This usage is consonant with the use of -a in oblique – object or genitive – personal
pronouns.
The following examples are all taken from recorded texts. In all of them, the suffix
-a is used to mark the object noun:
(18) a. …nʉ-�ura pʉni-ti-paa-ni-mʉ �umʉ pʉa-u-aa-mʉ

I/s-be see-caus-irr-fut-2s them/o kin-pl-o-2s
‘…I will indeed show you your kin…’
b. …manu-khu-tʉ tʉvʉ-pʉ-aa �uní�ni-kya-tʉ-mʉ
all-o-nom/o earth-nom-o live-pl-nom-pl
�apagha-qa-na-y �apagha-pʉga…
speak-pl-nom-o speak-re
‘…he spoke the languages that all the peoples on earth spoke…’
c. …márʉ-�ura tʉvʉ-pʉ-aa-va pʉ-paa-y-�ura
that/su-be ground-nom-o-at rel-at-o-be
mama-khwa-nhka-miya, kwipa-ti-aa-va �uru…
woman-go-dance-hab hit-caus-o-loc that/o
‘…this is the place where they used to dance the Beardance, at the
baseball park…’
2.4 The object suffix -ku

A third, possibly older, suffix that has survived in a restricted set of contexts is the
object suffix -ku. Consider first its distribution in numerals and quantifiers, as in:
(19) subject object

sú-i-s ‘one’ sú-ku-s
súw-iini ‘other’ sú-ku(-nani)
wá-ini ‘two’ wá-y-ku(-nani)
pá-ini ‘three’ pá-y-ku(-nani)
manu-ni ‘all’ manú-khu-tʉ
Several other contexts where -ku is used may be related, historically, to its older role as
object marker. Consider first its use in DS verbal complements:
(20) a. SS-complements (modality verb):

�áapachi wʉʉka-vaa-chi �ásti�i-y
boy/s work-irr-comp want-imm
‘The boy wants to work’
b. DS-complements (manipulation verb):
mamachi �áapa-chi wʉʉka-vaa-ku máy-kya
woman/s boy-o work-irr-comp tell-ant
‘The woman told the boy to work’
The diachrony of Ute case-marking 
If the complement-marking -ku in (20b) is indeed to be traced back to its ear-

lier use as object marker, the connection is distant and indirect. Still, there are some
precedents for the subject vs. object case-marking contrast re-grammaticalizing as the
contrast between same reference (SS) vs. different referent (DS), respectively, in either
verbal complements or adverbial clauses.11
The suffix -ku is also used in Ute as the subordination marker of many adverbial
clauses, as in:
(21) …kukwi-kwa-pʉga-y-ku, �uwa-rugwa-pʉga-y-ku…

fill-go-rem-o-sub him/o-give-rem-o-it
‘… when it (the bowl) filled up, (he) gave it to him…’
A more marginal example is the sporadically-used inanimate suffix pronoun -ku/-ukh/

-khu ‘it’. Unlike its more frequent counterpart -aqh, -ku never appears as a subject or
genitive pronoun, only as an object pronoun. Thus compare:
(22) a. Subject: �uwa-y-aqh ‘it is raining’

fall-imm-it
b. Object: pʉnikya-pʉga-aqh ‘(s/he) saw it’
see-rem-it
c. Genitive: wiichi kuvua-aqh ‘the knife’s tip’
knife/gen tip-its
d. Subject: *�uway-ku
all-imm-it
e. Object: pʉnikya-pʉgay-ku ‘(s/he) saw it’
see-rem-it
f. Genitive: *wiichi kuvua-ku
knife/gen tip-its
While a diachronic path that could lead from object case-suffix to object pronoun is
not clear at the moment, the reverse direction is also possible. Still, the distributional
restriction on the use of the pronoun -ku is reminiscent of the restriction on the use of
the object suffix -ku – only object, never genitive.
. Such situations are found in the switch-reference grammar of some Australian l anguages
(Austin, 1980). A similar – if more restricted – case has been reported in Yuman languages
(Munro 1980, 1983). See also further discussion in Givón (2001, Vol. II, Chapter 18).
Thornes (2003) shows the affix ku in Northern Paiute as marking both objects and geni-
tives. This suggests a Numic provenance, with the limited distribution in Ute numerals and
quantifiers perhaps a relic feature.
 T. Givón
2.5 Interim summary

Of the three putative older oblique suffixes in Ute, the one with the widest distribution
is -y, appearing on nouns, pronouns and nominalized verbs, and as both object and
genitive marker. The distribution of the suffix -a is much more restricted, but it still
appears as both object and genitive marker. And when -a and -y appear together, the
order is invariably -a-y; that is, with -a closer to the stem. In contrast, the suffix -ku
appears in extremely restricted contexts, marking only the object, never the genitive;
and it does not combine with the other two. These facts suggest, on purely theoretical
grounds, two diachronic hypotheses:
(23) a. Of the three suffixes, -ku is the oldest, -a younger, -y the youngest.
b. The two younger suffixes, -a and -y, share a common diachronic
mechanism.
We will return to these hypotheses directly below.
2.6 Object and genitive: The hidden footprints of nominalization

The most plausible way in which either the subject or object case-marking can overlap
with the genitive is through nominalization, whether of a whole clause or of a verb
phrase. Nominalization as a process may be defined as the structural adjustment made
to a verbal clause, or to a verb phrase, when it is placed in a prototypically nominal
grammatical position, such as the subject or object of a clause. This adjustment is a shift
from the prototype finite verbal clause structure towards the prototype noun-phrase
structure. The main components of this structural adjustment may be summarized as
follows:12
(24) Adjustment from the prototype finite verbal clause to noun phrase:
a. The verb becomes a head noun.
b. The verb acquires nominal morphology.
c. The verb loses tense-aspect-modal marking.
d. The verb loses pronominal marking.
e. The subject and/or object assume genitive/possessor case-marking.
f. Determiners may be added.
g. Adverbs are turned into adjectives.
A simple example from English illustrates the general pattern emerging out of (24),
contrasting the finite clause in (25a) below with its nominalized versions in (25b)
or (25c):
. For details see Givón (2001, Chapter 11; 2009, Chapter 4). A more extensive discussion of
Ute nominalizations can be found in Givón (2011, Chapters 8, 9, 12, 17).
The diachrony of Ute case-marking 
(25) a. Finite verbal clause:

She knew mathematics extensively.
b. Non-finite nominalized clause:
[Her extensive knowledge of mathematics] surprised me.
c. Non-finite nominalized verb phrase:
[some knowledge of mathematics] is required.
Languages diverge immensely in how much of an adjustment is required when a verbal
clause is placed in a typical nominal position. Ute, much like the rest of No. Uto-Aztecan
languages, falls on the extreme nominalizing end of the continuum, where every sub-
ordinate clause is, at least historically, nominalized. Of the seven structural adjustments
listed in (24) above, two will concern us most here in reconstructing the history of Ute
case-marking – the appearance of nominal suffixes on the nominalized verb (24b), and
the genitive marking of subjects or objects in the nominalized clause (24e).
Of the many contexts where clauses or verb phrases are nominalized in Ute, we
will discuss here only three: Equi-subject verbal complements, full-clause verbal com-
plements, and sentential subjects. As illustrations consider:
(26) a. Main clause:
ta�wachi �u sivaatuchi paqha-qha
man/s the/s goat/o kill-ant
‘the man killed the goat’
b. Equi-subject V-complement:
ta�wachi �u sivaatuchi paqha-vaa-chi �ásti-kya
man/su the/s goat/o kill-irr-nom want-ant
‘The man wanted to kill the goat’
(Hist.: ‘the man wanted [the killing of the goat]’)
c. Full-clause V-complement:
puchuchugwa-y ta�wachi �uway sivaatuchi paqha-qha-na-y
know-imm man/gen the/gen goat/o kill-ant-nom-o
‘(s/he) knows that the man killed the goat’
(Hist.: ‘(s/he) knows [the man’s killing of the goat]’)
d. Clausal subject:
ta�wachi �uway sivaatuchi paqha-qha-na-y �ura-�ay
man/gen the/gen goat/o kill-ant-nom-o bee-imm
‘it is the case that the man killed the goat’
(‘Hist.: ‘[the man’s killing (of) the goat] is’)
These three syntactic contexts are the most common ones through which main verbs
become grammaticalized as tense-aspect-modality markers.13 In the process of such
. See Heine and Kuteva (2007), Gildea (1998), Givón (2001, Chapter 7; 2009, Chapter 4).
 T. Givón
grammaticalization, the nominalized complement (or subject) clause becomes

the main clause, and its old nominalized structural features now become the new
main-clause features. Of the six main tense-aspect suffixes in Ute, reasonable verbal
etymology can be assigned to five:
(27) Possible verbal etymology of main TAM suffixes:

tense/aspect form verbal source
habitual -mi(ya) -miya ‘walk/go’
anterior -ka -ga/-ka ‘have/be’
irrealis -va(a) -va/-pa ‘go’
future -ni -ni ‘do’
remote -pʉ-ga ga ‘have/be’ (-pʉ = nom)
immediate -y ???14
Given the extreme nominalizing nature of Ute – and No. Uto-Aztecan – syntax,
the marking of objects as genitives is a natural consequence of the rise of tense-
aspect-modal markers out of verbs in configurations such as (26b, c, d) above. It
is indeed just as natural as the marking of the subject as genitive in configurations
such as (26c, d) above. Since both the -a and -y suffixes mark both object and geni-
tive role, the most reasonable conclusion is that both must have started as g enitive
markers, and then spread on to object marking through the same diachronic
mechanism – VP nominalization.
2.7 Reconstruction
2.7.1 The suffixes -a and -y
Purely internal evidence in Ute suggests a multi-step diachronic scenario, whereby:
i. The suffix -a started as a genitive suffix.

ii. In the process of tense-aspect-modal grammaticalization, and due to the nomi-
nalized nature of Ute verbal complements, the suffix -a spread over to also mark
objects.
iii. The suffix -y was introduced later on to replace -a as the new genitive marker.
iv. A subsequent tense-aspect-modal grammaticalization cycle is responsible for
spreading the suffix -y into the object paradigm and replacing -a there too.
. A possible conjecture here is that the -y suffix of the immediate (‘present progressive’)
tense-aspect is another reflex of the old object/genitive suffix -y.
The diachrony of Ute case-marking 
It is of course nice that comparative evidence supports this relative chronology,

suggesting that the suffix -a is older and goes back to Proto Ute-Aztecan, while the
suffix -y is younger and is limited to Northern Uto-Aztecan.15
2.7.2 The demise of the suffix -y in Ute

The demise of the younger object/genitive suffix -y in Ute can be understood as the
result of purely phonological processes. First, the de-voicing of unstressed word-final
vowels is a natural phonetic process, applying in Ute only to noun stems that were not
followed by a suffix. This change created the current contrast in Ute between the silent-
vowel form of subject/predicate nouns and the rest of the nouns – genitive, object, and
indirect object – whose final vowels were shielded by suffixes.
Next, one may note that only two final vowels are found in Ute noun-class s uffixes:
/i/ in the suffixes -chi and -vi/-pi, and /ʉ/ in the suffixes -pʉ/-vʉ and -tʉ/-rʉ. Only
suffixless nouns, a small group, exhibit other final vowels – /a/, /u/ and /ɵ/.
One may note, next, that the loss of the suffix -y after the final vowel /i/ is a mun-
dane phonetic change. The text-frequency of this mundane phonetic change must
have been rather high, due to the high text frequency of animate nouns, which in Ute
take mostly the suffixes -chi and -pi/-vi.
The loss of -y following the other noun suffixes, those ending with /ʉ/, is observed
synchronically in rapid speech of younger Ute speakers. Thus, for example, the 2nd-
person-sg. oblique form ‘ʉmʉ-y in the common greeting expression ‘ʉmʉ-y-s-a? ‘and
how about you?’ is often pronounced ‘ʉmʉ-s-a? Likewise, tʉpʉy-chi ‘rock’ is often pro-
nounced tʉpʉ-chi. Again, the text frequency of animate nouns ending with the suffixes
-tʉ/-rʉ is high.
The noun-final vowels /i/ and /ʉ/ thus account, between them, for the overwhelm-
ing majority of genitive- or object-marked nouns. The most common phonetic context
in which relics of the suffix -y survive, following the vowel /a/, are found in suffixless
nouns, pronouns and the verbal suffixes of nominalized clauses.
2.7.3 The object suffix -ku

For the time being, the diachrony of the object suffix -ku in Ute cannot be easily recon-
structed. On the one hand, its size, larger than either -a or -y, suggests a more recent
history. On the other hand, if it ever was a generalized object suffix, its rather restricted
. Langacker (1977, pp. 82–83). Dakin (1985) has suggests that traces of the suffix -y can
also be found in Nahuatl, making it potentially as old as -a. And Hill (2011) shows -y to be the
regular marker of both the object and genitive case in Takic.
 T. Givón
current distribution – in numerals and quantifiers, switch-subject complements and

adverbial clauses – suggests, if anything, a more remote history. And unlike the two
other suffixes, a plausible diachronic scenario that would explain the current distri-
bution of -ku is not obvious. However, one finds -ku in Northern Paiute (Thornes
2003) as an oblique-case – object and genitive – marker, suggesting a common Numic
ancestry.
3. Post-positions and indirect objects
3.1 Introduction
The post-positions that mark indirect objects in Ute are heterogenous, often com-
plex, and clearly hint at repeated cycles of grammaticalization. At the one extreme,
one finds locative post-positions whose verbal, or in a few cases nominal, origin is
fairly transparent. They are large, their precursor verbs or nouns are still around, and
their meanings as post-positions are easily predicted from their meaning as verbs or
nouns. At the other extreme, one finds a few small-size post-positions – some locative,
one instrumental, one associative – whose etymology is rather opaque. In the middle
one finds several syllable-size post-positions whose meaning is either locative or easily
relatable to locative, and whose etymology may be still traceable.
3.2 Large-size locative post-positions

3.2.1 Noun-derived post-positions
The grammaticalization of body-part nouns or other positional nouns as locative
adpositions is well documented elsewhere.16 Such post-positions arise from a noun-
phrase configuration in which the locational head noun, already marked with an
adposition, becomes a complex locative adposition, while the erstwhile genitive
modifier becomes the new head noun. In Ute, there are at least four locative post-
positions that arose through this pattern, all derived from still-extant nouns. Thus
consider:
(28) a. ‘at the face of ’ ==> ‘in front of ’:
Kani qovaa-va wʉni-(y)
house/gen face-loc stand-imm
‘(s/he) is standing in front of the house’
. See Heine and Kuteva (2007).

The diachrony of Ute case-marking 
b. ‘to the sky of ’ ==> ‘on top of ’:

kani tugu-tukhwa tarugwa-y
casa/gen sky-go ascend-imm
‘(s/he) is climbing on (top of) the house’
c. ‘at the outside of ’ ==> ‘out of ’:
kani tʉʉra-va �uni�ni-(y)
house/gen outside-loc do/be-imm
‘(s/he) is outside the house’
d. ‘at the center of ’ ==> ‘in the middle of ’:
kani tʉragwaa-va �uni�ni-(y)
house/gen center-loc do/be-imm
‘(s/he) is in the middle of the house’
The rise of these post-positions is so recent that there is no phonological evidence that
they are cliticized to the preceding noun. That is, their form as post-position is neither
phonologically reduced nor de-stressed.
3.2.2 Verb-derived post-positions

The bulk of large-size locative post-positions in Ute are grammaticalized verbs, arising
in a verb-phrase configuration in which the main verb becomes the new grammatical-
ized suffix on its erstwhile locative object. This is, again, a well-known pattern.17 These
post-positions are listed below together with their verbal sources.
(29) Verb-derived post-positions:

post-position verbal source
-chukhwa ‘to’ (an. obj.) chugwa ‘go to’, ‘meet’ (an. obj.)
-tukhwa ‘to’ (inan. obj) tugwa ‘go to’ (inan. obj.)
-mana ‘from’ mana ‘leave’
-chawi ‘toward’ chawi ‘come to’
-naagha ‘in’ naagha ‘enter’
-tarukhwa ‘on (top)’ tarugwa ‘climb’
-pa�agha ‘on (top)’ pa�agha ‘ascend’
-tʉvwa ‘down’ tʉvwa ‘descend’
-rukwa ‘under’ rukwa ‘descend’
-yaakwi ‘down into’ yaakwi ‘descend into’
-pawi ‘down’ pawi ‘descend’
-pina ‘behind’, ‘after’ pina ‘follow’
-yukhwi ‘after’ yugwi ‘sit’ (pl.)
. See Heine and Kuteva (2007).

 T. Givón
Unlike the de-nominal post-positions in (28), which are invariably followed by

the older locative suffixes, the de-verbal post-positions tend to be the last e lement
in complex post-position sequences, attesting to their more recent rise. Thus
consider:
(30) a. kani-vaa-tukhwa qxáarʉ-pʉga

house/o-at-go/inan run-rem
‘(s/he) ran to the house’
b. mamachi-vaa-chukhwa qxáarʉ-pʉga
woman/ob-loc-go/an run-rem
‘(s/he) ran to the woman’
c. �uwa-rukwa-rukhwa kwica-kwa-pʉa-y-agha
3s-descend-go defecate-go-rem-o-asp
‘…(and he) defecated under him…’
d. kani-vina-kway tugwa-pʉga
house/o-follow-go go-rem
‘(s/he) went behind the house’
The relatively recent rise of these post-positions is also attested by the fact that if they
are followed by other suffixes, the full form of the source verb is preserved, as in the
following text-derived examples:
(31) a. �áa-vaya-vaa-chugwa-mʉ kani-vaghay-kwa-nʉ

new-side-at-go-2s house-walk-go-1s
‘…I’ll come and visit you in your new house…’
b. kani-vaa-chugwa-av �uwas �uni-�a-vaaci
house/ob-loc-go-refl 3s do-asp-asp
‘…he was going to his own house…’
These large de-verbal post-positions can still appear as main verbs, as in the following
text-derived examples:
(32) a. …kani �u-na-kwa-pa�agha tarugwa-y…

house/o there-on-go-ascend ascend-imm
‘…(s/he) is climbing up on the house…’
b. …�u-vwaa-tugwa-pʉga…
there-at-go/inan-rem
‘…he went that-a-way…’
c. …“�uwa-chugwa-qa-kʉ” máy-pʉga…
3s/o-go/an-pl-emph say-rem
‘…“let’s go to him” (he) said…’
The diachrony of Ute case-marking 
d. …�u-vwaa-chawi mʉrʉkáchi kani-vaa-cháwi…

there-at-come white.man/gen house/o-loc-come/imp
‘…(let’s) go over there to the white-man’s house…’
e. …mana-kwa-paa-chi-ukh!…
leave-go-irr-subjun-it
‘…get away from there!…’
Lastly, these locative post-positions can go on to assume more abstract functions. Thus,
for example, the verb ‘sit’ is in the process of becoming the temporal post-position
‘after’, in expressions such as:
(33) a. …wíi-tavachi-yukhwi…
old-sun/o-sit/pl
‘…after a long time…’
b. …�ʉvʉs, wíi-tava karʉ-ga, págha-kwa-pʉga…
end old-sun sit-par walk-go-rem
‘…finally, after a long time (of being there, he) left…’
(lit.: ‘…finally, sitting (there) for a long time, he left…)’
The participial construction in (33b) may be the serial-verb precursor of the more
reduced (33a).
3.2.3 Monosyllabic de-verbal post-positions

A small group of monosyllabic locative post-positions can still be traced to verbs.
Consider first two etymologically-related post-positions, the semantically bleached
-va ‘at’ and the semantically richer -pa ‘in the direction of ’. The suffix -pa may appear
by itself, as in:
(34) a. �úu-pa tʉga-kwa

there-dir arrive-go/imper
‘…come that-a-way!…’
b. máa-pa págha-kwa-pʉga
there-go walk-go-rem
‘…(s/he) went that-a-way (vis.; with gesture)…’
c. págha-kwa-pʉga, tʉvwa-tukhwa páa-pa
walk-go-rem descend-go water-dir
‘…(and he) took off, down to the water…’
d. �íi-pa-aa kuchu-mʉ pagha-y?
here-dir-q buffalo-pl walk-imm
‘…have the buffalo passed this way?…’
 T. Givón
Further, one can still find -pa used as the main verb, as in the nominalized plural forms
in (35a, b), or the finite form in (35c):
(35) a. …�i-vaa-chi-mʉ, tava-mawisi-paa-tʉ-mʉ

here-at-nom-pl sun-appear-dir-nom-pl
máa-pa-tʉ tʉvʉpʉ-aghaa-tʉ-mʉ…
there-dir-dir country-have-nom-pl
‘…the ones from here, from where the sun rises, (and) those that have
the country right there…’
b. …�ʉʉ-núuchi-u-�ura �íi-pa-tʉ-mʉ…
true-Ute-pl-be here-dir-nom-pl
‘…the true Utes from right around here…’
c. …�úu-pa-pʉga…
there-go-rem
‘…(he) took off that-a-way…’
The suffix -va by itself is used as ‘at’, as in:
(36) kani-vaa ‘at the house’

�u-vwa ‘there’ (invis.)
ma-vaa ‘there’ (vis.)
�i-vaa ‘here’
Likewise:
(37) …yúaa-va nagukwi-ta-pʉga…

open.county-va fight-pass-rem
‘…people were fighting (each other) in the open country…’
This suffix appears frequently as the semantically-empty first element in combinations

with other – semantically richer and diachronically younger – suffixes, as in:
(38) a. kani-vaa-tukhwa
house/o-at-go
‘(moving) to the house’
b. mamachi-vaa-chukhwa
woman/o-at-go
‘(moving) toward woman’
c. kani-vaa-tʉ-mana-kway
house/o-at-dir-leave-go/o
‘from the house’
d. �u-vwa-cawi mʉrʉkáchi kani-vaa-cawi
there-at-come white.man/gen house/ob-loc-come
‘…(let’s) over there to the white-man’s house…’
The diachrony of Ute case-marking 
The Ute consonant /v/ was probably an intra-vocalic variant of /p/.18 This and other
considerations point to an etymological connection between the two locative post-
positions -va and -pa. It is thus of interest to note that -pa, the semantically richer
post-position, can still be found as the main verb, albeit a morphologically defective
one, as in (35c) above.
The post-position -kwa is easily traced to an older motion verb ‘go’, which in Ute
has grammaticalized in multiple capacities. Most commonly, it supplements other
post-positions, endowing them with a sense of directional motion (allative), as in:
(39) a. táa-mana-khwa-�uru �ɵa-khwa-pʉga, tʉkuavi

knee-come-go-that pull.out-go-rem meat/o
‘…(he) pulled meat out of (from) his knee…’
b. kani-vaa-tʉ-mana-kway paghay�wa-y
house/o-dir-leave-go/o walk-imm
‘(s/he) is walking to the house’
The verbal origin of -kwa is fairly transparent, since one can still find it as the last ele-
ment in a main-verb complex, perhaps already grammaticalized as an aspectual or
directional marker, as in:
(40) a. �áa-vaya-vaa-chugwa-mʉ kani-vaghay-kwa-nʉ

new-side-at-go-2s house-walk-go-1s
‘…I’ll come and visit you in your new house…’
b. �u-vwaa-cawi-kwa-paa-ni
there-at-come-go-irr-fut
‘…(I’m) going to go there…’
c. …mana-kwa-paa-chi-ukh!…
leave-go-irr-nom-it
‘…get away from it!…’
d. táa-mana-khwa-�uru �ɵa-xwa-pʉga, tʉkuavi
knee-leave-go-it/o pull.out-go-rem meat/o
‘…he pulled meat out of his knee…’
e. páa-rukwa-n ta-yáakwa-kwa-vaa
water/o-descend-1s paw-push.down-go-irr
‘…you’re going to push me under-water with your paw…’
f. …págha-kwa-pʉga, tʉvwa-tukhwa páa-pa…
go-go-rem descend-go water-dir
‘…(and he) took off down to the water…’
. See Givón (2011, Chapter 2).

 T. Givón
The two directional post-positions -tʉ and -chʉ ‘in the direction of ’ have no clear
verbal source. There are some tantalizing hints, however, that they may be bleached
derivatives of the two motion verbs -tugwa and -chugwa, respectively. As noted ear-
lier above, both of the latter verbs have grammaticalized more recently as the allative
post-positions ‘to’. However, as such they still retain the sense of physical motion. The
directional suffixes -tʉ and -chʉ, on the other hand, impart a more abstract sense of
direction. Thus consider:
(41) a. �áapaci-vaa-chʉ nʉka�ni-kya

boy/o-at-dir/an listen-ant
‘(s/he) listened to the boy’
b. ta�wachi kani-vaa-tʉ máy-kya
man/o house/o-loc-dir tell-ant
‘(s/he) told the man about the house’
c. kani-vaa-tʉ pʉni�ni-kya
house/o-at-dir look-ant
‘(s/he) looked toward the house’
d. ta�wachi-vaa-chʉ �apagha-qa
man/o-loc-dir/an talk-ant
‘(s/he) talked to the man’
e. �u-vwaa-tʉ �uni�ni-(y)
there-at-dir be/move-imm
‘(s/he) is way over there’
f. …�u-vwaa-tʉ chichi�ni-pʉga…
there-at-dir peek-rem
‘…(and he) peeked that way…’
Lastly, a verbal etymology for these two post-positions is also suggested by the fact that
they show the same contrast as the two clearly de-verbal post-positions -tugwa and
-chugwa, of inanimate vs. animate location, respectively. Their comparatively bleached
state, both phonologically and semantically, suggests that they may be older deriva-
tives of the two allative motion verbs.
3.2.4 The older locative post-positions -na, -mi and -ma

Three mono-syllabic post-positions, -na, -mi, and -ma, have no discernible verbal
etymology. They can only be used in various combinations with other post-positions,
and such combinations are not altogether predictable. Among them, -na is commonly
associated with the stative locative sense of ‘on’ or ‘above’, as in:
(42) a. …�i-vaa-na-�uru tʉvʉpʉ-vwa-na-amʉ-�uru…
here-at-on-top earth/o-at-on-3p-it/o
‘…they (were) up here above the earth…’
The diachrony of Ute case-marking 
b. …�i-na-kwa-pa�agha tarukhwa-pʉga…
here-up-go-ascend ascend-rem
‘…(s/he) went up here…’
c. …�ichay tʉvʉpʉ-vwa-na pagha�ni-pʉga-vaci-mʉ…
this/o earth/o-at-over wander-rem-asp-pl
‘…they were wandering all over this earth…’
d. …�agha-pa-�ura ma-na-tarukhwa? máa-pa tʉna-khwa!…
q-dir-be there-on-ascend there-dir ascend-go/imper
‘…how can one climb up there? Climb that-a-way!…’
There is some evidence, found in high-frequency fixed expressions, that -na may have
been a verb meaning ‘be there’. Thus, in (43b) below it takes the normal verbal double
negation:
(43) a. �i-ya-na
here-???-be.at
‘(it/s/he) is here’
b. ka-�i-ya-na-wa-tʉ
neg-there-???-be.at-neg-nom
‘(it/s/he) is not here’
The post-position -ma often appears with the stative locative sense of ‘on’ or ‘above’,
as in:
(44) a. ‘…pʉ�i-av tu-túpʉna-pʉga, �u-ma-tugwa-aqh

eye-refl red-throw-rem there-on-go-it
�u-rukwa múkwi-pʉga…’
there-descend stick.head-rem
‘…(and he) kept throwing his eyes way up there and then stuck his
head under there…’
Lastly, the post-position -mi is rare and semantically opaque, and found primarily in
combination with other post-positions, as in:
(45) a. tʉka�napʉ-tʉvwa-mi-tukhwa
table/o-descend-loc-go
‘down off the table’
b. �i-mi-tukhwa
here-loc-go
‘(moving) this way’
One may as well mention, lastly, two non-locative post-positions, the instrumental -m
‘with’ and the associative -wa ‘with, ‘and’, that have no obvious etymology.
 T. Givón
4. Discussion
Both in the marking of the core case-roles object and genitive, and in the marking
of indirect objects by post-positions, the Ute synchronic data reveal the footprints of
repeated cycles of grammaticalization and re-grammaticalization. It is not surprising
that the diachrony of Ute case-marking morphology can be internally reconstructed
from irregularities and relic forms found in the synchronic data. The fact that com-
parative Uto-Aztecan data tend to confirm the scenarios generated by purely-internal
reconstruction is of course comforting.19
The grammaticalization of case-marking morphology in Ute is strongly
constrained by the language’s syntactic typology, in particular its historical SOV
word-order and its extreme propensity for nominalizing all subordinate clauses. Thus,
in the repeated waves of grammaticalization of main verbs into post-positions, the
OV order in verb phrases and the nominalized nature of verbal complements are key
predictive constraints on the way de-verbal suffixes in Ute pile one on top of the other
to yield complex post-positions.
One important reason why internal reconstruction is such a useful method
harkens back to its strong dependence on syntactic-typological universals, thus on a
theory of diachronic change and grammaticalization. In the main, the morphemes we
assume to be older are phonologically smaller and semantically more bleached. They
tend to cliticize closer to the lexical stem, and are distributionally less predictable.
Often, they may be found only in syntactic relic zones. For case-markers, such relic
zones – or diachronic graveyards – tend to be pronominal rather than nominal, and
nominalized (non-finite, subordinate) rather than finite main clauses. Though this
tendency is not absolute.20 The lesson to be drawn from this paper is that puzzling
synchronic facts are less puzzling when viewed from a diachronic perspective.21
References
Austin, Peter K. 1980. Switch reference in Australian languages. In Munro (ed.), 7–47.
Dahl, Östen. 2009. Two pathways of grammatical evolution. In Syntactic Complexity [Typologi-
cal Studies in Language 82], T. Givón & Masayoshi Shibatani (eds), 239–248. Amsterdam:
John Benjamins.
. See again discussion in Givón (2000).

. For some important exceptions to this, see Gildea (1998).
. See Givón (1971, 1979, 2009).
The diachrony of Ute case-marking 
Dakin, Karen. 1985. yi/ya, a Uto-Aztecan possessive suffix? Friends of Uto-Aztecan Conference.
Ms, University of Arizona, Tucson.
Gildea, Spike. 1998. On Reconstructing Grammar. Oxford: OUP.
Gildea, Spike (ed.). 2000. Reconstructing Grammar: Grammaticalization and the Comparative
Method [Typological Studies in Language 43]. Amsterdam: John Benjamins.
Givón, T. 1971. Historical syntax and synchronic morphology: An archaeologist’s field trip. In
CLS 7, 394–415. Chicago IL: Chicago Linguistics Society.
Givón, T. 1979. On Understanding Grammar. New York NY: Academic Press.
Givón, T. (ed.). 1985. Ute Traditional Narratives, Ignacio CO: Ute Press.
Givón, T. 2000. Internal reconstruction: As method, as theory. In Gildea (ed.), 107–159.
Givón, T. 2001. Syntax: An Introduction, 2 Vols, Amsterdam: John Benjamins.
Givón, T. 2009. The Genesis of Syntactic Complexity. Amsterdam: John Benjamins.
Givón, T. 2011. Ute Reference Grammar [Culture and Language Use 3]. Amsterdam: John
Benjamins.
Haiman, John & Munro, Pamela (eds). 1983. Switch Reference and Universal Grammar
[Typological Studies in Language 2]. Amsterdam: John Benjamins.
Heine, Bernd & Kuteva, Tania. 2007. The Genesis of Grammar. Oxford: OUP.
Hill, Jane H. 2011. Pronouns in the Cupan languages. Seminario de Compejidad Sintactica,
Universidad de Sonora, Hermosillo, Ms.
Hyman, Larry Michael. 1970. The role of borrowing in the justification of phonological
grammars. Studies in African Linguistics 1(1): 1–48.
Langacker, Ronald. 1977. An Overview of Uto-Aztecan Grammar, Studies in Uto-Aztecan
Grammar, Vol. 1. Dallas TX: UTA & SIL Publication 56.
Munro, Pamela (ed.). 1980. UCLA Working Papers in Syntax 8. Los Angeles CA: University of
California.
Munro, Pamela. 1980. On the syntactic status of switch-reference clauses: The special case of
Mojave comitatives. In Munro (ed.), 144–159.
Munro, Pamela. 1983. Where ‘same’ is not ‘not different’. In Haiman & Munro (eds), 223–244.
Thornes, Tim. 2003. A Northern Paiute Grammar with Texts. Ph.D. dissertation, University of
Oregon.
Zipf, G. 1935. The Psycho-Biology of Language: An Introduction to Dynamic Philology, paperback
edition, 1965. Cambridge MA: The MIT Press.
Language contact as an inhibitor
of sound change
An Athabaskan example*
Keren Rice
University of Toronto
In the past 150 years, the Fort Good Hope variety of Dene (also called Slavey),
an Athabaskan language of northern Canada’s Mackenzie River valley, has
undergone several phonological shifts. I focus on the change of nasals to r.
Not all nasals shift in the appropriate environment. At first, this failure to shift
appears attributable to functional factors like frequency and uniformity of
exponence. Another factor plays a major role: contact with a related language
where the n’s that shift to r in Fort Good Hope are distinct from those that do
not. Historical records indicate contact occurred around the time of the shift.
Both grammatical and social factors play an important role in blocking certain
n’s from shifting to r.
Keywords: Dene; phonological shift; phonology; historical linguistics
1. Introduction
A highly salient characteristic of the Dene variety (Athabaskan family) that is often
called Hare, spoken in Fort Good Hope and Colville Lake, Northwest Territories,
Canada is the presence of [r] where closely related varieties have [n]. This is illustrated
* I have always been a great admirer of Jane Hill, and the amazing balance that she achieves
in her work. It is in much part her inspiration that made me think beyond the grammatical
factors that might be involved in the language shift described in this chapter, and look to the
community of language use.
 Keren Rice
in (1), where Hare is compared with the Dene spoken in the community of Délįne.1
The sounds in question are bolded.2
(1) Hare Délįne
rake náke ‘two’
ráyuka náoka ‘Northern lights’
rɛshe nɛshe ‘potato’
rírɛhɬa nínεhtɬa ‘I arrived’
ruhshe nuhshe ‘I want to grow’ (optative)
Based on the forms in (1), one might think that everywhere a nasal occurs in Délįne,
[r] is found instead in Hare, with a general restructuring of /n/ to /r/. While for the
most part this is correct, this generalization fails in two ways. First, overall, [n] and
[r] are in complementary distribution, with [n] appearing in what I will call the nasal
environment and [r] elsewhere. This is an allophonic distribution involving a pair of
sounds that pattern together in many languages, and this distribution is not surpris-
ing. Second, and more interesting, there are nasals that occur in an oral environment,
exactly the environment in which the shift to [r] is expected.
In this chapter I examine the development of r from n in Hare. In the first s ection
I discuss the historical sources of both nasals and r. I then turn to the nasals that fail
to become r. I argue that these nasals fail to shift partially due to a constraint involving
uniformity of exponence, preferring that a morpheme have a single realization.
. Some notes on terminology are in order. First, I use the term ‘Hare’ as the name for the
primary language discussed in this chapter, although this term is not used as much today
as it once was. Second, I use the term Délįne, the current name of the community in which
the variety traditionally called Bearlake is spoken (see Rice 1989, for instance, for this usage
of terms). Third, the term Dene ‘people’ is used in two ways. First, it is the name for a group
of closely related languages in the Mackenzie area of Canada; this is the use in this chapter.
Second, it is also used more broadly, as an equivalent to Athabaskan. Third, language names
have shifted between the earliest dates referred to in this chapter and today. I try to use the
earlier names when talking about that time period.
. I generally use the practical orthography. In this, symbols such as b, d, g, dz are voiceless
unaspirated stops and affricates; p, t, k, ts are voiceless aspirated, and p’, t’, k’, ts’ are voiceless
ejective. Alveopalatal fricatives are written as sh (voiceless) and zh (voiced), and the voiced
velar fricative is written gh. The symbol ch represents a voiceless aspirated alveopalatal affri-
cate and the symbol ch’ is its ejective counterpart; wh represents a voiceless w. A glottal stop is
written with a raised comma. Nasalization of a vowel is written with a hook below the vowel.
The acute accent represents high tone. I use the symbol ‘e’ for a mid front closed vowel and ‘ε’
for an open one; in the orthography, these are written with ‘ә’ and ‘e’ respectively. Beyond this,
symbols correspond with a standard transcription system. Verb stem high tones in Hare do
not actually occur on the stem, but rather on the syllable that precedes it. I write these tones
on the stem for ease of comparison.
Language contact as an inhibitor of sound change 
owever, in and of itself, this does not provide a full account of why some n’s fail to
H
shift to r, as there are many other places where uniformity of exponence might be
expected, but is not found. I then discuss the language contact situation at the time
that the shift was ongoing, and suggest that one factor in maintaining /n/ was contact.
2. The historical source of Hare [n] and [r]
Leer (1996a) provides the following nasal inventory for Proto-Athabaskan.
(2) (*m) *n *ny/ŋ
I focus on the latter two nasals.

In studying Athabaskan languages, it is important to distinguish between stem-
initial and prefix-initial positions, and between morpheme-initial and morpheme-
final positions as well (see, for instance, the discussion in Rice and Hargus 2005 for a
summary). The elements of interest in this chapter are prefixes. The development of
nasals in stems does not bear on the issues addressed here, so I set aside stems in the
following discussion.
In prefix-initial position, the Proto-Athabaskan nasals generally develop as n in
Délįne and as r in Hare in what I will call the non-nasal environment and as n in what I
will call the nasal environment. Hare and Délįne forms are compared in (3). The nasal
environment basically involves a nasalized vowel or a syllable-final nasal following the
consonant in question (3b); in the oral environment, the following vowel is oral (3a).
Proto-forms here and throughout are from Leer (1996a); Hare and Délįne Dene are
from my fieldwork.
(3) Development of Proto-Athabaskan nasals

Proto-Athabaskan Hare Délįne
Non-nasal environment: *nasal usually develops as n in Délįne
a.
and r in Hare
*nə- gender, assume position rɛ- nɛ-
*na’ˑ- ‘down, to ground; continuative’ rá- ná-
*na- ‘back, again, iterative’ ra- na-
b. Nasal environment: *nasal always develops as n
*nə+n gender, assume position nį- nį-
*na’ˑ- + na- continuative+iterative nǫ- (~rára) nǫ-
In (3), the same morphemes, first in the oral environment (3a) and then in the nasal
environment (3b), are illustrated. While these morphemes are invariantly [n]-initial
in Délįne, the consonant varies between [n] and [r] in Hare depending on what
follows.
 Keren Rice
In some cases, invariant [r] occurs in Hare: if no environment exists in which a

nasalized vowel follows, the sound is always realized is [r], as in (4).
(4) Invariant [r] arises when no nasal environment is possible.
*ncx(ə)- ‘1, 2 pl. object’ raxɛ- naxɛ-
The facts discussed so far are reasonably straightforward to account for: in Hare a nasal
developed from a Proto-Athabaskan nasal in the nasal environment while r developed
in the oral environment.
However, while in general this is the pattern of development for nasals, there are
also some morphemes in Hare that begin with an invariant n, with a nasal in the oral
environment. An exhaustive list, to my knowledge, of these morphemes is given in (5).
(5) Invariant prefix-initial nasals3

*ŋə- stative perfective nɛ- nɛ-
*ŋə- 2sg subject, object nɛ- nɛ-
? ne- ‘across’ ne- ‘across’
To summarize, the development of nasals in Hare and Délįne is compared in (6).
(6) Summary: Development of prefix-initial nasals

Hare Délįne
Nasal Non-nasal Nasal Non-nasal
environment environment environment environment
invariant n n n n n
variant n n r n n
invariant r – r n
In the next section, I briefly present a synchronic analysis for Hare.
3. Synchronic patterning of historical nasals
The synchronic analysis of Délįne Dene is straightforward: there is a nasal, and it

always surfaces as such in prefixes. Where Proto-Athabaskan is reconstructed with
. The data in (5) might suggest that the reconstructed velar nasal develops as invariant
n, with the reconstructed coronal nasal developing as the variable n~r. See Section 4.4 for
arguments against this position.
Language contact as an inhibitor of sound change 
both *n and *ŋ/ny in prefixes, these merge in Délįne Dene and develop as a nasal
consonant or nasalization, depending on the context.
The synchronic analysis in Hare is straightforward as well: in this dialect there are
two phonemes, /n/ and /r/. /n/ is always realized as [n], while /r/ has allophones [n]
in the nasal environment (before a nasalized vowel, syllable-finally) and /r/ elsewhere.
First consider /r/ in Hare. Many morphemes show variation between [n] and [r].
(7) illustrates a morpheme that I will call n- situation aspect, also called n- conjugation
in the Athabaskan literature. Again Hare and Délįne Dene are compared, with the
segment in question bolded. In the first person singular form, the situation aspect
marker is followed by an oral vowel; the consonant is realized as [r] in Hare. (Note
that there are further morphological complexities that lead to the other differences in
the paradigms illustrated here. They are outside the scope of what I want to deal with
in this chapter; see Rice 1989 for discussion.) In the second person singular form of
‘go into water’ in (7a), the first of the two n’s in Délįne, realized as r in the oral envi-
ronment in Hare, is the situation aspect marker and the second is the person marker;
in ‘arrive laughing’ in (7b), the situation aspect marker has the form n in the second
person singular form as it is in the nasal environment (created by the nasalization, the
realization here of the second person singular subject). Finally, in the third person, the
nasalization on the vowel in (7a) represents third person in the perfective; it creates
the nasal environment, and thus the situation aspect marker occurs in the form [n]. In
(7b), the situation aspect marker is realized as nasalization in the third person form.
(7) Variant prefixes: n~r alternations (/r/)

Hare Délįne
a. ‘go into water’, perfective n-situation aspect
Oral environment: r
tɛ-ri-ya tɛ-ni-ya ‘I went into water’
Oral environment: r
tɛ-rɛ-nɛ-ya tɛ-nɛ-nɛ-ya ‘you sg. went into water’
Nasal environment: n
tɛ-nį-ya tɛ-nį-ya ‘s/he went into water’
b. ‘arrive laughing’, perfective n-situation aspect
Oral environment: r
rídlodɛ́-rɛ-h-ɬa nídlodɛ́-nɛ-h-tɬa ‘I arrived laughing’
rídlodɛ́-nɛ̨-ɬa nídlodɛ́-nɛ̨-tɬa ‘you sg. arrived laughing’
Syllable-final environment: nasalization.
rídlo-d-ɛ̨-́ ɬa nídlo-d-ɛ̨-́ tɬa ‘s/he arrived laughing’
The n- ‘qualifier’ is illustrated in (8a). In this case the second person singular is realized
as nasalization on the vowel and the qualfier has the form [n]; elsewhere it is [r].
 Keren Rice
Finally the terminative is shown (8b). It differs from the situation aspect and quali-
fier morphemes in that it is less closely linked phonologically with the rest of the
verb word. It has two possible forms, with variation, rí- (more common) and nį́-;
thus the oral allophony holds, with [r] in the oral environment and [n] in the nasal
environment.
(8) Variant prefixes: n~r alternations (/r/)

Hare Délįne
a. n-‘qualifier’ (‘conjunct’ prefix)
Oral environment: r
rɛ-h-she nɛ-h-she ‘I grow’
nį-ye nį-ye ‘you sg. grow’
Oral environment: r
rɛ-ye nɛ-ye ‘s/he grows’
b. ní-‘terminative, to the ground’ (‘disjunct’ prefix)
Oral environment: r
rí-rɛ-h-ɬa ní-nɛ-h-tɬa ‘I arrive’
Oral environment: r
rí-nį-ɬa ní-nį-tɬa ‘you. sg. arrive’
ní ̨-ɬa ní ̨-tɬa
varies w. oral environment: r
rí-hɛ̨-ɬa ‘s/he arrives’
Two notes are in order. First, in the forms with second person singular subjects in (8),
second person singular is marked by nasalization on the vowel; the preceding n is
the morpheme in question here, a qualifier in (8a) and a morpheme often translated
as ‘terminative, to the ground’ in (8b). Second, in the Hare nasal environment forms
shown in (8b), there is variation between nį́ɬa and ríhɛ̨ɬa, with the syllable hɛ added
to host the nasalization. For speakers who use this form, this prefix is always rí-; the
n-initial form has been lost.
While there is alternation between [r] and [n] in Hare in the forms in (7) and (8),
in cases where the morphology never creates the nasalized environment, only [r]
occurs, as in the examples in (4), and, for the speakers who use rí- in the third person,
in (8b).
Finally, the second person has two allomorphs, nɛ- and nasalization. This
morpheme is never realized as [r], even in the oral environment. Examples are given
in (9), showing the second person singular as a subject, possessor, oblique object, and
direct object. While these particular examples are from Hare, the facts surrounding
second person allomorphs are identical in Délįne.
Language contact as an inhibitor of sound change 
(9) Second person singular allomorphs

nɛ- nasalization
nɛ-jɛ̨ ‘you sg. sing’ dį-jɛ̨ ‘you sg. start to sing’
nɛ-t’ é ‘you sg. cook’ dį-t’é ‘you sg. cook for yourself ’
wɛ-nɛ-da ‘you sg. sit’ wǫ-dá ‘you sg. optative sit’
k’ína-nɛ-ɬe ‘you sg. go around’ keyíyį-ɬa ‘you sg. are going up’
nɛ-tá ‘your sg. father’
nɛ-hɛ́ ‘with you sg.’
rá-nɛ-reyiht’u ‘I hit you sg.’
Similarly, the morpheme identified as perfective/stative in the literature is realized as

either nɛ- or nasalization; it never has an oral form.
(10) Perfective/stative allomorphs

nɛ- nasalization
nɛ-zǫ ‘it is good’ hį-bárɛ ‘it is rounded, scalloped’
gǫ-sɛ́lɛ ‘it (land, river) is small’
The nɛ- allomorph of the perfective/stative is rare in Hare, where this morpheme is,
with the exception of the word ‘good’, shown in (10), realized as nasalization; it is
common in Délįne; see below for additional discussion.
In order to capture the synchronic facts of Hare, it is important to distinguish
between the environment where an oral vowel follows and the environment where a
nasalized vowel or a consonant follows. I will not formalize this in this chapter; the
goal of this section was simply to show that a synchronic account, with both /n/ and
/r/, is possible.
4. An historical note: Timing of the shift to /r/
4.1 Introduction
There are excellent resources on Hare that allow us to understand the overall timing
of the shift from n to r in this language. First is a dictionary compiled by Petitot,
Dictionnaire de la langue dènè-dindjié. Dialects montagnais ou chippewayan, peaux de
lièvre et loucheux, renfermant en outré un grand nombre de termes propres à sept autres
dialects de la meme language, published in 1876. This volume includes a grammar and
verb paradigms as well as dictionaries of three languages, Dëne Sųɬiné (also called
Chipewyan; Petitot’s Montagnais), Hare (Petitot’s Peaux-de-Lièvre), and Gwich’in
(formerly called Kutchin and Loucheux; Petitot’s Loucheux); the dictionary contains
information on related varieties in the larger region as well, but focuses on these three.
 Keren Rice
Petitot was a priest in the Mackenzie area between 1862 and 1874, and he did this
work at that time. Second is a word list gathered by Fang-Kuei Li in the 1929. Third is
current material, compiled over the past thirty years.
A comparison between the Peaux de Lièvre of Petitot’s time and the Hare recorded
by Li shows that where there are r’s in Hare today and at the time of Li’s fieldwork, Peti-
tot usually wrote n. Petitot’s French translations are included. The relevant segments
are bolded.
(11) 1876: Petitot (dictionary); 1920’s: Fang-Kuei Li (word list)

Petitot 1876 1920’s/current
’klo yanêchié ‘cultiver’ tl’o rɛhshe ‘I grow grass’
na-déinlin ‘chute’ rádɛyįlį ‘waterfall’
natsézé ‘chasse’ ráts’ɛzé ‘hunt’
naêta ‘cheminer’ ráyɛhda ‘I walk along’
nonta ‘cheminer, 3e pers’ nǫda ~rárada ‘s/he walks along’
nâ-nél’u ‘coudre’ rá’ɛrɛhlu ‘I sew’
nâtl’a ‘courir’ ráhɬa ‘I go (fast)’
ne ‘tu’ nɛ- ‘you sg.’
4.2 Language at the time of Petitot

The Peaux de Lièvre forms in Petitot in many ways are more similar to current Délįne
forms than to current Hare forms. It is worth noting that the Hare and the Délįne
peoples are generally considered more distinct now than they were at the time that
Petitot wrote. Helm (2000: 15–16) remarks that in 1865 Petitot identified two regional
bands of Peaux de Lievre who used the northern shores of Great Bear Lake, but that by
the early twentieth century, these two groups were no longer identifiably Hare, but had
become part of a people that identified themselves as Sahtugot’ine (Bear Lake people),
today called Sahtúot’ine or Délįneot’ine. Howren (1970) notes that the change from
/n/ to [r] took place between the time that Petitot recorded the language and Li did
his work.
Petitot, in addition to the general Peaux de Lièvre, has forms in the Peaux de
Lièvre column of the dictionary from what he calls Bâtard Loucheux and others that
he calls Peaux de Lièvre du Grand Lac des Ours (Hareskin of Great Bear Lake), clearly
varieties of Peaux de Lièvre. In many cases only a general Peaux de Lièvre form is
given. Petitot often lists more than one form, sometimes labeled for variety and some-
times not, and he notes that the first form is the most frequent, or the correct term
(1876: xvii). Terms that follow are equally appropriate, but used less frequently.
The Bâtard Loucheux variety contains many characteristics that clearly identify
the language that is called Hare today, spoken in Fort Good Hope and Colville Lake
(the differences between these are not relevant to the current chapter). For example,
Language contact as an inhibitor of sound change 
Hare as recorded in the 20th and 21st centuries has [l] where Déline has [ɬ], and gen-
erally has [w] where Délįne has the voiceless counterpart. This is reflected in much
of the vocabulary in Petitot, with the general Peaux de Lièvre and the specific Bâtard
Loucheux often distinguished in this way. In (12), I give Petitot’s transcription. Petitot
uses l’ for a voiceless lateral fricative and notes that ll has the sound of English ‘well.’
n’ indicates nasalization of the preceding vowel (written with a hook under the vowel
today; Petitot puts a dot under this n – I do not include the dot).
(12) Peaux de Lièvre Bâtard Loucheux Délįne Hare

l’u llu ɬu lu ‘ice’ (glace)
fwén wén whɛ̨ ́ wɛ̨́ ‘star’ (étoile)
In addition, voiceless aspirated affricates in Délįne generally appear as fricatives in

Hare, and this is reflected in many words in Bâtard Loucheux.

tsén sen tsɛ̨ sɛ̨ ‘dirt’
tsa sa tsá sá ‘beaver’ (castor)
tchô chô cho sho ‘big’ (grand)
(Petitot’s ‘tch’ is a voiceless aspirated affricate, now written ‘ch’; his ‘ch’ is a voiceless
fricative, now written ‘sh’.)
The word ‘dog’ differs in Hare and Délįne today in the same way as in Petitot’s
time. Petitot’s notes suggest that the raised comma here is basically velarization.
t’lin llin tɬį lį ‘dog’ (chien)
Where today Délįne Dene has kw, Hare has f, and this is reflected in Petitot.
kfwè fwè kwe fe ‘rock’ (pierre)
kfwè-kfwin fwè-fwin gohkwį gohfį ‘axe’ (hache de pierre)
Assuming that Petitot’s writing is systematic, there appears to have been considerable
variation at the time (variation that continues in many ways today). For instance, the
form ‘big’ in (13) begins with an alveopalatal affricate, now written ch, in Délįne, and
with an alveopalatal fricative, now written sh, in Hare. The Petitot dictionary includes
another form of ‘big’, nétchay, with an affricate – this is labeled Bâtard Loucheux.
Petitot also includes words that he labels specifically as Peaux de Lièvre du Grand
Lac des Ours as well; this is now known as Délįne. For instance, the word ttsè-k’u
‘woman (femme)’4 is labeled for this dialect, and this is the word that is used in Délįne
. Petitot’s symbol tts would be written as ts’ now – it is an ejective consonant.
 Keren Rice
Dene today; it is not used in Hare. Petitot gives as the general Peaux de Lièvre word
ttsè-liṇe, a word that is not used today to my knowledge. He gives the word used in
Hare today (and not used in Délįne) as a word meaning ‘femme mariée’, yé-nnéné
(today yenene), without a specific label. The word for ‘five’ (cinq) is identified as Délįne
(su-sinla; current sǫlái), and the general word that is given, lla-kké is not identified for
variety; this is the form used in Hare now (lák’ɛ).
Words that are found in Hare today but not in Délįne also appear unlabelled.
For example, for ‘meat’ (viande), Petitot gives iṇé (contemporary/įyɛ̨), a form
used in Hare but not in Délįne. In addition, the words for ‘red’, ‘trout’, and ‘caribou’
differ between Hare and Délįne, and Petitot gives either only the Hare form (not
labeled for variety), or both the Hare and Délįne forms, with the current Hare form
first. For instance, for ‘trout’ (truite), he gives pièrè, sapa, biéré (B), with the first
two without a specific variety and the last specified as Bâtard Loucheux. The first
and third of these are probably the same; this is the form that is used in Hare now,
while the middle form is used in Délįne. A similar example is the word ‘tomorrow’
(demain). The current word ‘tomorrow (demain)’ in Hare is ’ɛk’ɛ̨; Petitot gives
êkkèén, ekkiné (B); these are variants of the same word. The current Délįne word,
sachǫ, he gives as sa-tchon, saying it is from Tlicho Yatiì (Dogrib). Many other
similar examples can be cited, and I give just one more. For ‘muskrat’ (rat-musqué)
Petitot gives dzén = zén (B) = t’ è-kkpáe. The first form is used in Hare and the last
one in Délįne today.
Petitot further remarks for the language generally that “n est susceptible de se
changer en t=d=r” – n is susceptible to vary with t, d, and r (d is pronounced as in
dame and dernier; t is pronounced as in temps), and that r is always ‘doux’ and is some-
times replaced by d, n, or l.
It thus appears that the variation that we see between Hare and Délįne Dene today
was present at the time that Petitot recorded vocabulary. Petitot gives some examples
that are labeled Bâtard Loucheux that have r (16a), but many others have n where r
occurs (16b) today.
(16) Petitot current

a. ta-gorétté ‘combien’ tágoret’e ‘how many’
ara-k’e ‘esquimaux’ ’aráke ‘Inuit’
b. ttsé na k’age ‘extérieurement’ ?
énat’u ‘larme’ (tears) ratú ‘tears [eye+water]’
In other cases, r is found, but the word is not labeled ‘Bâtard Loucheux’ – kfwi
êrég’é ‘nod ones head’ (hocher la tête); békkè ra-êtsi ‘I clean’ (nettoyer). I do
not know the first word, but for both words, they would have r in Hare and n in
Délįne.
Language contact as an inhibitor of sound change 
4.3 The problem

With some sense of the variation present in the language at the time that Petitot wrote,
we are now in the position to understand the problem posed by the Hare sound shift.
Why, did the second person singular and a few other morphemes fail to shift their
initial n to r in the oral environment? (I do not address the very interesting question of
what might have triggered the shift from a nasal to [r].) Both the current /n/ and the
current /r/ are represented identically, generally by n, in Petitot, yet in only some mor-
phemes did the shift to /r/ occur between the time that Petitot compiled the dictionary
and Li did his work.
4.4 Possible accounts

One possibility is that the different developments of the nasal can be accounted for
historically. In reconstructing Proto-Athabaskan, Krauss and Leer (1981) and Leer
(1996a) propose that there are two nasals, reconstructed as *n and *ny/ŋ, as in (2). In
prefixes, *ny/ŋ is found in the second person singular and the perfective/stative. Thus
it could be that *n developed into /r/ while *ny/ŋ remained /n/.
A few facts suggest that this is not likely the case, at least not directly. First, *n and
*ny/ŋ had merged in both Hare and Délįne Dene in stem-initial position by the time
of Petitot, so, if this analysis were correct, a merger would have occurred in stems but
not in prefixes. In general, mergers are more common in prefixes than in stems in this
language family. For instance, in the Dene language complex under discussion, *d and
*t merge in certain types of prefixes, while they remain distinct in stems. For example,
Leer 1996a reconstructs the inceptive prefix as *tə-, while in the Dene complex, this pre-
fix is realized as dɛ-, merged, for instance, with *də- self-benefactive, also realized as dɛ-.
Second, there is no apparent way of distinguishing *n and *ny/ŋ at Petitot’s time.
These nasals had almost certainly merged by the time that Petitot recorded the lan-
guage. Petitot notes the presence of a palatal nasal in Gwich’in (his Loucheux), a
related language (see the discussion below), comparing it directly with Spanish ñ. Peti-
tot remarks that this nasal occurs in Fort Good Hope, but only very rarely. He writes
this nasal as n�, and I have not found words with it in Hare. As we will see below, Petitot
systematically writes the second person singular with a palatal nasal in Gwich’in. Peti-
tot thus was clearly aware of the difference between these, and, if a difference existed
between the two nasals in Hare, it is very likely that he would have recorded it.
An analysis differentiating two nasals at the time of Petitot might still be pos-
sible if there were differences between them in patterning. For instance, in the very
well-known analysis of Yowlumne (also called Yawelmani), a Yokutsan language of
California, vowels are proposed that never surface; however their patterning is such
that they are argued to be present underlying, with a non-local effect on n eighbouring
 Keren Rice
vowels; see, for instance, the overview in the Kenstowicz and Kisseberth 1979
phonology textbook.
In Hare, there is no systematic difference in the phonological patterning of the
nasals that vary with r and those that do not, and there does not appear to have been
any systematic difference between them at Petitot’s time that might have led speakers
to differentiate them. For instance, both the second person singular and the situation
aspect can appear as a syllable or as nasalization on a vowel, as illustrated in (7–8)
above. While the particular morphological conditions are different for the different
morphemes, the surface effects are the same.
In general then, it appears that an abstract analysis that appeals to the historical
origins of the sounds is not appropriate for these data, and the question remains as to
why some nasals failed to shift to /r/.
5. Other diachronic changes: Restructuring and invariant nasals
In this section I examine the invariant nasals, arguing that the failure to shift involves
a single morpheme, the second person singular. However, the second person singular
is not the only morpheme that is reconstructed with *ny/ŋ, or that has an invariant
nasal in Hare.
As noted in Section 3, there is a morpheme that is called perfective/stative in the
Athabaskan literature. This morpheme has the form nɛ- in one word in Hare.
(17) Délįne Hare Petitot
nɛ-zǫ nɛ-zǫ ‘it is good’ nézin (bon)
Otherwise, the perfective/stative has the form nasalization in Hare; it is nɛ- in Délįne.
Further, as also shown in (18), it had the form nasalization at the time that Petitot
recorded the language. These forms are not labeled for any particular variety in Petitot.
(I write tones on the verb stem [final syllable] in Hare in order to make the comparison
with Délįne clear. However, in Hare these tones actually surface on the syllable before
the verb stem.)
(18) Délįne Hare Petitot
nɛ-chá / hį-shá / ‘it is big’ / intcha (grand)
nɛh-chá hɛh-shá ‘I am big’
nɛ-tsɛ́lɛ / hį-sɛ́lɛ / ‘it is small’ / intsélé (petit)
nɛh-tsɛ́lɛ hɛh-sɛ́lɛ ‘I am small’
nɛ-ká hį-ká ‘it is wide’
nɛ-ghalɛ hį-ghalɛ ‘it is narrow’
hįt’alɛ ‘it is flat’ inttàlé (plat)
hįhxǫ́ ‘be no good, moldy’ inχun (moisi)
hįlɛgɛ ‘be slippery’ inlléyé (glissan)
hįkį ‘be heavy’ ink’in (lourd)
Language contact as an inhibitor of sound change 
With the exception of ‘good’, a word that is very common, the initial consonant of the
perfective/stative does not occur syllable-initially, and thus is not in the environment
for reinterpretation as [r].
There are two other items with invariant n. One is the initial n- of na- k’ína-,
a preverb meaning roughly ‘around about’. This preverb has two syllables and in all
likelihood is historically k’ɛ́- ‘around’ followed by na- iterative. However, in both Hare
and Délįne Dene, the first of these, on this meaning, does not occur independently of
na-. It thus appears likely that k’ína- is regarded as a single morpheme, with the n being
morpheme internal rather than morpheme initial. It is only morpheme-initial nasals
that were restructured to /r/ in Hare; morpheme-final and morpheme-internal n’s did
not shift, as the examples in (19) illustrate.
(19) Délįne Hare

tɛnɛ tɛnɛ *tɛrɛ ‘trail’
kǫ́, -kón-ɛ́ kǫ́, -kón-ɛ́ *-kórɛ́ ‘fire, non-possessed, possessed’
Morphological restructuring that shifts morpheme boundaries has occurred in the

Athabaskan family. For instance, Leer (1996a) provides the Proto-Athabaskan recon-
structions in (20), while these are analyzed either as single morphemes or CVC stems
followed by a vocalic suffix in the language now.
(20) Proto-Athabaskan Dene (generally)

*tə-ŋəɬ ‘bucket’ (water-pour) tɛni, tɛnɛ ‘bucket, pail’
*də-neˑ ‘person’ dɛn-ɛ ‘person’
Finally, the morpheme ne- ‘across’ remains as such. I have no account for this; the
morpheme is not very common in my corpus and is, perhaps, a more recent borrowing
into Hare from another variety.
I have suggested that the perfective/stative failed to shift to /r/ in Hare because,
with a single exception, ‘good’, it did not occur in the appropriate environment for
the nasal to be reinterpreted as oral. For the two other invariant n’s, in one case
I suggested that a morphological restructuring occurred, with a single morpheme
resulting from an historical sequence. I do not have an account for the final case,
ne- ‘across.’
I now turn to the second person singular, again asking why that morpheme escaped
the sound shift. As subject, this morpheme patterns very much like the n- situation
aspect, a morpheme that shifted to /r/. It remains a mystery given that the initial
consonants of theses morphemes do not appear to be phonetically or phonologically
distinct at the time of Petitot.
As discussed already, the second person singular has two forms: it is either a
nasal-initial syllable or it nasalization on a vowel. The different forms are distributed
as described in (21), and exemplified in (9).
 Keren Rice
(21) a. Second person singular subject

Always nasal: syllable-initial [n], syllable-final nasalization
b. Second person singular direct object, possessor
Always n-, in an onset
Assuming that there were not two nasals at Petitot’s time, as discussed above, another
possible account of the failure of the initial consonant of the second person singular
must be sought. I pursue here an account based on avoidance of homophony,
uniformity of exponence, and frequency. In short, once the nasals merged, the amount
of homophony in prefixes was greatly increased, often with only subtle differences
between persons. This creation of homophony could, perhaps, have created pressure to
change. What might be more vulnerable to restructuring? The second person singular
most commonly occurs in the nasal environment, while the other morphemes in ques-
tion most commonly appear in the oral environment. Assuming a general constraint
on uniformity of exponence, requiring that, all other things being equal, a morpheme
be realized in the same form, the nasal would prevail in the second person singular,
blocking it from shifting. See Albright (2011) and Urbanczyk (2011), for instance, for
recent discussion of homophony avoidance and of uniformity and its potential role in
phonology.
In order to test this idea, I did frequency counts of the second person singular and
of the variant nasals based on dictionary entries and on texts.
With the second person singular, it appears that the nasal environment is more
common than the oral environment. Beginning with grammatical role, the second
person singular is realized as either a full syllable or nasalization when it is the sub-
ject; in other grammatical roles, it is a full syllable. While I do not have texts that are
appropriate to do counts of the distribution of the second person singular in various
grammatical roles, intuitively it seems that subjects, where the form varies between nɛ
and nasalization, are more common than objects and possessors, where the second
person singular is always nɛ-.
What then is the more common form with subjects? If frequency is an important
factor in inhibiting sound change, following the line of reasoning above, it should be
the case that the nasalized form of the second person singular subject is more common
than the full syllable. I looked at paradigms and asked which form is more common
paradigmatically. There are two major factors to consider. One is aspect – whether the
verb is imperfective, perfective, or optative. Further, in the perfective, it is necessary to
distinguish two major classes, labeled perfective 1 and perfective 2 in (22). The second
factor is the type of morpheme, if any, that immediately precedes the subject marker.
Prefixes in Athabaskan languages are frequently divided into two types, called ‘dis-
junct’ and ‘conjunct,’ with the former basically being derivational and the latter overall
inflectional. Putting these factors together, we find the following distribution.
Language contact as an inhibitor of sound change 
(22) Distribution of 2 singular subject

Imperfective Optative Perfective 1 Perfective 2
nɛ - Word-initial – – √
after disjunct
Nasalization After conjunct √ √ –
I used the Hare dictionary (Rice 1978) and did a count of the number of verbs in which
each of the different forms occurs. I focused on one initial letter, ts’. This is because
dictionary entries are organized by what is called the ts’ form, the form representing an
unspecified subject (similar to French on). All instances of the second person singular
in word-initial position fall under this letter. I counted 50 words where the subject
would be word initial in the imperfective aspect and 17 where it would directly follow
a disjunct prefix out of a total of approximately 190 verbs. This means that in around
120 of the verbs, the second person subject appears as nasalization in the imperfective,
or around 2/3 of the time. The number of verbs in which it appears as nasalization will
be increased with other letters just because there is not the word-initial possibility.
By my count, then, an overwhelming number of verbs use nasalization in the
imperfective. This subject is always marked by nasalization in the optative. The per-
fective is more complex; in this aspect it is likely that the nɛ- form is more common
paradigmatically.
These rough counts based on paradigms suggest that the nasalized form of the
second person singular subject is more frequent than the full syllable form. It is
important to note that I do not have information about which form is more frequently
used; I simply did lexical counts.
For the variable nasals, I did two types of counts. First I looked at paradigms
with n- situation aspect (7) and n- qualifier (8a). In both imperfective and perfective
paradigms, the oral environment occurs in the first person singular, first person plural,
and second person plural subject forms, and, depending on a variety of morphological
factors, in the second person singular and third person as well. Only the oral environ-
ment occurs in the optative. I also counted oral and nasal environments in a story, and
the oral environment is more common: in the story I looked at, the oral environment
was about three times as frequent as the nasal environment.
Assuming that these rough counts based on paradigms and a single text are
representative of the language as a whole, a possible account of the failure of the second
person singular to shift to /r/ is that frequency is important in determining whether
a nasal underwent this shift or not: when the morpheme occurs more commonly in
the nasal environment, it retained its nasal form; when it is more common in the oral
environment, it underwent the shift.
It has been amply demonstrated that factors of frequency and uniformity of
exponence play important roles in language change (e.g. Albright 2011; Blevins 2004;
 Keren Rice
Bybee 2001; Urbanczyk 2011), but this account of the shift, like an abstract nasal
account which would attribute the invariant nasal to the historical ny/ŋ and the vari-
ant nasal to the historic *n, faces some challenges. First note that there is not full
uniformity – the second person appears as either a full syllable beginning with a nasal
or as nasalization, depending on morphological context, and the /r/-initial morphemes
vary allophonically between [n] and [r], showing less uniformity than they did before
the shift. Second, I have suggested that pressure for uniformity prevents restructur-
ing of the second person singular, even when it is not in the nasal environment. One
might think that if uniformity were important, it might also drive restructuring. This
does not appear to be the case, however. For instance, the first person singular subject
has the form h- in the imperfective and optative, and in some perfectives; in other
perfectives it has the form i-. The first person singular direct object, indirect object,
and possessor is sɛ-. While in Délįne there is some restructuring of the subject to be
h- no matter what the aspect is, this is not the case in Hare, and there is no move
towards regularizing the forms so that the same consonant occurs in all grammatical
roles. Similarly, the second person plural has three forms. The subject is generally ah-,
but it is a- in some perfectives; the direct object, oblique object, and possessor is raxɛ-.
Again, there does not appear to be a move towards uniformity of exponence. The third
person plural has the form kɛ- as a subject; the third person plural object and possessor
is ku-, go-, or ki-, depending on the person and number of the subject. Again, there is
no move to leveling to yield a uniform realization.
Uniformity of exponence might provide a partial account for why the second
person singular failed to shift to r. In the next sections I argue that while this might
be a factor, another factor may well also have been at work, namely contact with other
languages.
6. Nasals in Gwich’in
In this section I look at nasals in Gwich’in, a neighbouring Athabaskan language to

Hare. While these languages are spoken in areas that are geographically proximate,
the languages are not closely related to each other within Athabaskan in genetic terms.
Gwich’in is usually grouped together with languages of eastern Alaska and the Yukon,
while Hare is grouped with Northwest Mackenzie languages, based on phonological,
morphological, and lexical developments (e.g. Goddard 1996).
Recall that both *n and *ny/ŋ are reconstructed in Proto-Athabaskan. As discussed
in Section 2, these sounds appear to have merged in Hare and closely related v arieties
by the time that Petitot wrote. This was not the case in the Loucheux that Petitot
recorded. Leer (1996b: 199) reports that Proto-Athabaskan *n in stem-initial position,
before a non-nasal rhyme, became nd or d in Gwich’in, while *ny/ŋ developed as ndʒ
in the nasal environment. Leer further remarks that “The nasals n and ny have since
Language contact as an inhibitor of sound change 
merged – apparently quite recently, since ny has persisted into historical times in at
least one morpheme: the second person pronominal prefix. Archdeacon McDonald
transcribes this sound as ny, and according to John Ritter, it is still found in the speech
of some Canadian Gwich’in, usually in free variation with n” (1996b: 200).
Gwich’in (Loucheux) is one of the languages included in the Petitot dictionary.
As discussed above, Petitot notes a sound that he writes as n�, saying that the sound is
palatal, pronounced as in Spanish señor, and he distinguishes it from n, pronounced
as in French nation, nez. Of the former, Petitot says that it is used regularly only in
Loucheux.
The entries for the second person singular in Petitot are shown in (23), using
Petitot’s transcription and language names, focusing on the nasal in syllable-initial
position.
(23) Second person singular

Loucheux n̄e, n̄i
Peaux de Lièvre né
In Loucheux (Gwich’in), the palatal nasal occurs in this morpheme, while in Peaux
de Lièvre (Hare), n is found. There is some variation in how Petitot writes this nasal
in Gwich’in, sometimes using n, but he often uses the symbol for the palatal nasal. It
appears from his examples that the palatality of the second person may appear on a
nasal that is not in fact the second person singular (e.g. n̄an-djet ‘you sg. fear’ (tu as
peur), where the second person singular is the nasal following the vowel, realized as
nasalization on the vowel; even in this case the second person has a palatal nasal while
other persons have a plain nasal, n-, as in nell-djet ‘I fear’, na-djet ‘s/he fears’, nâ-djet
‘we fear’, nô-djet ‘you pl. fear’, k’na-djet ‘they fear’), representing an n- initial prefix (the
cognate prefix in Hare occurs with [r]). The second person singular pronoun contrasts
with the first/second person plural non-subject, shown in (24), again using Petitot’s
transcription and names.
(24) First/second person plural non-subject

Loucheux nurwé
Peaux de Lièvre naχe, naχo
In this case in both Loucheux (Gwich’in) and Peaux de Lièvre (Hare) n occurs;
current Hare realizes this morpheme as raxɛ.
In Loucheux forms in Petitot with the perfective/stative as a full syllable, the
palatal nasal is also found in some forms (25a), although not in others (25b).
(25) a. n̄étsell ‘small’ (petit)

n̄itié ‘heavy’ (lourd)
b. nizjin ‘good’ (bon)
nittsik ‘narrow’ (étroit)
 Keren Rice
Could the existence of a contrast between a dental and a palatal nasal in Gwich’in
have provided speakers of Hare with evidence for the different nasals? Even if these
sounds had merged in Hare by the time of Petitot, could people have been exposed to
Gwich’in, and heard the nasal in the second person singular as different from the nasal
in other morphemes?
7. Contact between the Hare and the Gwich’in
Based on the ethnographic literature, it is evident that Hare and Gwich’in speakers
were in contact, and the retention of /n/ in the second person singular might be at
least partially attributable to this contact. The summary below is based on Hara (1980)
and Savishinsky and Hara (1981), anthropologists who worked in the Hare area in
the 1960’s and 1970’s, as well as on work by Krech, an anthropologist who carried
out research in the Gwich’in area in the lower Mackenzie in the 1970’s. I also draw on
research by the anthropologist June Helm (2000) on the Dene people more broadly.
The first written records by early explorers document that there was contact
between the Hare and the Gwich’in. Alexander Mackenzie travelled through the
area occupied by the Hare and the Gwich’in in 1789. Mackenzie writes of encounters
with a group that he calls the Hare Indians and a group that he calls the Deguthee
Dinees, a branch of the Gwich’in, with dates of these encounters a day apart, July 8,
1789 and July 9, 1789 (Mackenzie 1927). As Hara (1980: 31) points out based on
Mackenzie’s account, “we may assume that the Hare shared a considerable range
of cultural traits with the adjacent Slave and Dogrib Indians, as well as with the
Loucheux.”
During the fur trade era, a fort was established in the Fort Good Hope region in
1804, and was, as Krech (1979: 100) notes “in the territory of the contiguous Hare”
(writing from the perspective of the Gwich’in). The fort was moved further north in
1823 to better facilitate trade with the eastern Gwich’in and Inuit (also called Eskimo);
at the same time, Krech notes this post “was located near the Kutchin-Hare border”,
and many Gwich’in participated in trade. The post was moved again in 1827 for bet-
ter provisioning and access to the Hare, and in 1840–1850, posts were established in
Gwich’in territory (Krech 1979: 100). Krech (1979) reports heavy use of the Fort Good
Hope post by the Gwich’in.
This is confirmed by Savishinsky and Hara (1981), who write that the trading
post at Fort Good Hope attracted Mountain Indians, Gwich’in, and other peoples as
well as the Hare, with the result that, during the first half of the nineteenth century,
members of several ethnic groups were drawn into the fort’s population. Explorer John
Franklin also writes of encountering Gwich’in near Fort Good Hope in 1825–1826
(Krech 1979: 112).
Language contact as an inhibitor of sound change 
Savishinsky and Hara (1981: 323) give the following population figures for the
Fort Good Hope area in 1829–1830.
Leaders, Men Boys Girls Children Women Total Identity

chiefs hunters
1 51 32 28 39 151 Mackenzie River Indians

or Rapid Indians
2 57 51 33 44 187 Mackenzie River Indians
or Outer Hare Indians
1 43 43 14 30 131 Mackenzie River
Indians, Lower or
Bastard Loucheux
(Kutchin)
1 33 10 18 16 78 Mackenzie River
Indians, Upper
Loucheux (Kutchin)
1 30 37 25 93 Mackenzie River
Indians, or Lower Rat
Hungers, Loucheux
(Kutchin)
Mackenzie River Indians or Rapid Indians are Hare. As noted earlier, Gwich’in,
Loucheux, and Kutchin are variant names. These figures show that there was clearly
contact between Hare and the Gwich’in in this time period through the use of a
common trading post.
Contact continued after a trading post opened in the Gwich’in area. Ross, in a
manuscript held in the Bureau of Indian Affairs cited in the 1906 Handbook of American
Indians (Hodge 1907), wrote that in 1859 the Hare lived in the country around Fort
Good Hope, extending beyond the Arctic circle, where they came into contact with the
Gwich’in, with whom they intermarried. Helm (2000: 16), in discussion of joint land
occupancy and intermarriage, talks of the 19th century Bâtard Loucheux, descended
from marriage between Gwich’in and Hare, with their subsequent absorption into the
Hare trading post at Fort Good Hope.
Petitot lived in the region between 1862 and 1874, and did extensive travelling.
Petitot (1876) identifies the Hare as living on the lower Mackenzie from Fort N orman
(south of Fort Good Hope) to the Arctic ocean, an area that overlaps with the
Gwich’in. Krech (1979: 112) writes of Petitot that he encountered Gwich’in in various
areas, including one night distant from Fort Good Hope. Savoie (1971: 193), quotes
Petitot as saying “All the Dindjié [Loucheux] … living in the steppes along the coast
of the Arctic Ocean speak and understand the Hareskin dialect.” With respect to the
relationships between the Gwich’in and the Hare, Savoie (1971: 192–192) further
quotes Petitot: “The Loucheux of Dindjié apply to the Hareskins the abusive epithet
 Keren Rice
Hatchen, meaning enemies. The Hareskins, who have no craving for deadly warfare,
keep pretending that they are being called Itchun, i.e. Rosebuds. Thanks to this subtle
pretense, Loucheux and Hareskins are the best of friends in the world.”
I have found less in the academic literature on the later part of the 19th century
and the early part of the 20th century on relationships between these groups.
However, an excellent resource is available on the Gwich’in, compiled by Heine,
Andre, Kritsch, and Cardinal. This 2007 book compiles stories of the Gwichya
Gwich’in, Gwich’in who today live in the community of Tsiigehtshik, down river
from Fort Good Hope. There is much talk of the Hare (often called ‘Slavey’ by the
elders) as well as of the Gwich’in. Father Séguin, who was in Fort Good Hope in
the 1860’s, “received the first Gwichya Gwich’in families in the m ission at Fort
Good Hope in April 1864” (Heine et al. 2007: 206). Father Séguin travelled between
Fort Good Hope and Gwich’in territory, overseeing the building of a mission at
Tsiigehtchic, in Gwich’in territory, and continued his life split between these
communities until 1890; this suggests that there was likely continued contact
between the peoples.
Stories from elders in the book speak to relationships with the Hare. Julienne
Andre, born in 1887, tells of life on the land around 1900. She talks of travelling
towards Fort Good Hope, and of there being a good trail, as well as travelling to other
areas. She also talks about gambling with the Slavey, by which she means the Hare. In
her story there is frequent reference to meeting with the Slaveys, and living with the
Slaveys. Other elders talk of meeting with the Hare and relations with the Hare, includ-
ing marriage.
Elders also talk of their languages. Julienne Andre spoke both Gwich’in and
Slavey (Heine et al. 2007: 246). Hyacinthe Andre, born in 1910, spoke Gwich’in, Slavey,
French, and English (247). Others also spoke both Gwich’in and Slavey (as well as
English), and many spoke Gwich’in and English. Those born somewhat later (around
1930 and later) tend to speak Gwich’in and English, although some speak Slavey as
well. While the number of elders is not large, many of them spoke Hare as well as
Gwich’in, suggesting contact.
Just as the Gwich’in talk of relations with the Hare, the Hare speak of relations with
the Gwich’in. When I lived in Fort Good Hope, people often talked of their relations
with the Gwich’in.
All of this suggests that speakers of Hare were exposed to Gwich’in on a con-
tinuing basis, with bilingualism also being likely. Helm (2000: 19) quotes Krauss
and Golla 1981 on the structure of linguistic relationships across the northern Dene
(meaning northern Athabaskan peoples in general) domain: “… intergroup com-
munication has ordinarily been so constant, and no dialect or language was ever
completely isolated from the other for long. … Whatever the immediate, short-term
language boundaries, the network of communication in the northern Dene linguistic
Language contact as an inhibitor of sound change 
complex is open-ended” (Krauss & Golla 1981: 68–69). Krauss (2000: li), speaking of

Alaska, reinforces this, remarking on multilingualism among Athabaskans: “In the
Athabaskan family, …, people traditionally did not just learn their own language and
learn to recognize and deal with all its dialects, but they also learned the languages
and dialects of their neighbors as well, at least passively, depending on travel pat-
terns and need. It might have been difficult or impossible, therefore, to find an adult
who had not often heard neighboring varieties of Athabaskan, within a radius of a
hundred miles or so, who had not become at least passively multilingual.”
Given what is understood of the contact situation, then, it seems quite reason-
able that the nasal in the second person might have failed to shift in part based on the
contact of Hare speakers with speakers of Gwich’in, from whom they heard a distinct
nasal in the second person singular.
8. Summary
In linguistic terms, the sound patterns discussed in this chapter are summarized in (26).
(26) Délįne Hare

mid 1800’s early 1900’s late 1900’s
/n/ /n/, /r/ /n/, /r/
2 sg. nɛ-~V̨ nɛ- ~ V̨ nɛ- ~ V̨ nɛ- ~ V̨
stative nɛ- hį- hį- hį-
variable n~r n- n- n- ~ r- n- ~ r-
invariant r n- n- r- r-
It is reasonable to assume that Proto-Athabaskn *n and *ny/ŋ had merged in Hare

before Petitot’s time, given that he does not distinguish them for Hare or other closely
related languages, but does for Gwich’in. Given this, an explanation is needed for why
some /n/’s at Petitot’s time failed to shift to /r/. I have suggested that this basically
involves accounting for why the /n/ of the second person singular failed to shift, and I
suggested that two factors might be involved.
First, pressures for uniformity of exponence coupled with the higher frequency of
the nasal environment for the second person singular facilitated its maintaining this
form, while others shifted in the oral environment.
Second, the Hare were in regular and ongoing contact with the Gwich’in, with
intermarriage. In Gwich’in, the initial nasal of the second person singular was distinct
from other prefix-initial nasals at the time that Petitot recorded the language. It appears
that exposure to Gwich’in, where the second person singular was distinct from the
other n-initial morphemes, could well have been a critical factor in blocking the nasal
of this morpheme from shifting its form.
 Keren Rice
References
Albright, Adam. 2011. Paradigms. In The Blackwell Companion to Phonology, Marc van Oostendorp,
Colin Ewen, Elizabeth Hume & Keren Rice (eds). pg. 1972–2001. Oxford: Blackwell.
Blevins, Juliette. 2004. Evolutionary Phonology. Cambridge: CUP.
Bybee, Joan. 2001. Phonology and Language Use. Cambridge: CUP.
Goddard, Ives. 1996. Introduction. In Languages. Handbook of North American Indians, Vol. 17,
Ives Goddard (eds), 1–16. Washington DC: Smithsonian Institution.
Hara, Hiroko Sue. 1980. The Hare Indians and their World [National Museum of Man Mercury
Series]. Ottawa: National Museums of Man. (1964 dissertation).
Heine, Michael, Andre, Alestine, Kritsch, Ingrid and Cardinal, Alma. 2007. Gwichya Gwich’in
Googwandak/The History and Stories of the Gwichya Gwich’in. As Told by the Elders of
Tsiigehtshik, revised edn. Tsiigehtshik & Fort McPherson NT: Gwich’in Social and Cultural
Institute.
Helm, June. 2000. The People of Denendeh. Iowa City IA: University of Iowa Press.
Hodge, Frederick Webb. 1907. Handbook of American Indians north of Mexico [Smithsonian
Institution, Bureau of American Ethnology Bulletin 50]. Washington DC: GPO.
Howren, Robert. 1970. A century of phonological change in a northern Athapaskan dialect. Ms.
Kenstowicz, Michael & Kisseberth, Charles. 1979. Generative Phonology. New York NY:
Academic Press.
Krauss, Michael E. 2000. Koyukon dialectology and its relationship to other Athabaskan
languages. In Koyukon Athabaskan Dictionary, Jules Jetté & Eliza Jones, l-lxv. Fairbanks
AK: Alaska Native Language Center, University of Alaska Fairbanks.
Krauss, Michael E. & Leer, Jeff. 1981. Athapaskan, Eyak, and Tlingit sonorants. Fairbanks AK:
Alaska Native Language Center, University of Alaska.
Krauss, Michael E. & Golla, Victor K. 1981. Northern Athapaskan languages. In Handbook of
North American Indians, Vol. 6: Subarctic, June Helm (ed.). pg. 67–85. Washington DC:
Smithsonian Institution Press.
Krech III, Shepard. 1979. Interethnic relations in the lower Mackenzie River region. Arctic
Anthropology 16: 102–122.
Leer, Jeff. 1996a. Comparative Athabaskan Lexicon [Item CA965L1996]. Ms, Alaska Native
Language Archive. 〈http://www.uaf.edu/anla/collections/ca/cal/〉
Leer, Jeff. 1996b. The evolution of the stem syllable in Gwich’in. In Athabaskan Language Studies:
Essays in Honor of Robert W. Young, Eloise Jelinek, Sally Midgette, Keren Rice & Leslie
Saxon (eds), 193–234. Albuquerque NM: University of New Mexico Press.
Li, Fang-Kui. 1929. Hare file slips.
MacKenzie, Alexander. 1927. Voyages from Montreal on the River St. Laurence through the
Continent of North America to the Frozen and Pacific Oceans in the Years 1789 and 1793
with a Preliminary Account of the Rise, Progress, and Present State of the Fur Trade of the
Country, John W. Garvin (ed.). Toronto: The Radisson Society of Canada.
Petitot, Emile-Fortune-Stanislas-Joseph. 1876. Dictionnaire de la langue dènè-dindjié dialects
montagnais ou chippewayan, peaux de lièvre et loucheux, renfermant en outre un grand
nombre de termes propres à sept autres dialectes de la même langue; précédé d’une monogra-
phie des Dènè- Dindjié, d’une grammaire et de tableaux synoptiques des conjugaisons. Paris:
Ernest Leroux; San Francisco: A.-L. Bancroft.
Language contact as an inhibitor of sound change 
Rice, Keren & Hargus, Sharon. 2005. Introduction. In Athabaskan Prosody [Current Issues
in Linguistic Theory 269], Sharon Hargus & Keren Rice (eds), 1–45. Amsterdam: John
Benjamins.
Rice, Keren. 1978. Hare Dictionary. Ottawa: Northern Social Research Division, Indian and
Northern Affairs.
Rice, Keren. 1989. A Grammar of Slave. Berlin: Mouton de Gruyter.
Savishinsky, Joel & Hara, Hiroko Sue. 1981. Hare. In Handbook of North American Indians,
Vol. 6. Subarctic, June Helm (ed.). pg. 314–325. Washington DC: Smithsonian Institution
Press.
Savoie, Donat. 1971. The Amerindians of the Canadian North-west in the 19th Century, as seen
by Emile Petitot, Vol. II: The Loucheux Indians. Preceded by General Observations on the
Dèné-dindjié Indians. Ottawa: Northern Science Research Group, Department of Indian
Affairs and Northern Development.
Urbanczyk, Suzanne. 2011. Root-affix asymmetries. In The Blackwell Companion to Phonology,
Marc van Oostendorp, Colin Ewen, Elizabeth Hume & Keren Rice (eds). pg. 2490–2515.
Oxford: Blackwell.
Stress in Yucatec Maya
Syncretism in loan word incorporation as evidence
for stress patterns
Emily Kidder
University of Arizone
Yucatec Maya (YM) is an indigenous language of Mexico which displays

phonemic tonal distinctions and phonemic vowel length. These features are
primarily associated with the phonetic cues of pitch and duration, which are also
considered the primary correlates of stress in language. Though scholars have
noted the existence of stress or accent since it was first documented centuries ago,
no detailed account of stress as either a separate or related entity to tone or length
has been made. In this paper, I analyze the phonetic changes evident in a series of
loan words into YM from Spanish, a language without tone or length distinctions,
found in elicited and conversational data as an initial step in diagnosing stress
patterns in YM.
Keywords: Maya; pitch; stress; accent
1. Introduction
Stress, or accent, is a cross-linguistically variable linguistic entity that is thought to

rely on the acoustic cues of pitch, duration, and to some extent intensity, as its main
building blocks. Languages utilize these and other cues in different ways to create a
multi-tiered system of emphasis that has a variety of semantic functions. From word
level semantic distinctions to intonational contours, emphasis can convey a wide
range of meanings from syntactic scope to emotional state. The phonetic and phono-
logical coding of stress, or accent (depending on the definition being used), plays a
large role in the structuring of language. Attempting to ‘diagnose’ the metrical p attern
of a language is not a simple task, and is the subject of vast amounts of scholarly effort.
Some languages have proven especially problematic for the task of untangling the
prosodic and metrical systems, particularly those in which the main acoustic cues of
stress (pitch and duration), are already in use for separate phonological means.
 Emily Kidder
Yucatec Maya, an indigenous language of Mexico’s Yucatán Peninsula, is just such

a case, and though many aspects of the language have been thoroughly documented,
it is currently lacking in a comprehensive account of stress. Yucatec Maya, (hence-
forth YM), has been documented as having a tonal system, as well as a differentia-
tion between long and short vowels. Although the tonal system has been robustly
analyzed in the literature, many studies mention the likely presence of stress, but do
not systematically analyze it as a separate entity from tone (see Bricker et al. 1998;
Gussenhoven & Teeuw 2008; Frazier 2009). The presence of tonal features in YM, as
well as the presence of a meaningful distinction between long and short vowels, com-
plicates the prosodic landscape in a way that makes a difficult task of determining what
stress patterns might be present in the language, if indeed it has any at all.
Native YM words almost always have features that obscure the presence of stress,
making it a difficult to determine if a rhythmic structure utilizing either pitch or dura-
tion even exists. The presence of a pitch peak or a lengthening of syllables may not be
due to stress alone, but instead be due to the presence of a separate phonological pro-
cess. In order to get a clearer window into how stress works in YM, data was collected
from 8 native speakers, who (like the majority of speakers of YM), are also bilingual in
Spanish. During this data collection, instances of switching between YM and Spanish
were common, as were loan words incorporated into YM from Spanish.
When Spanish words were incorporated into Yucatec morphological and syntac-
tic processes, they were often, but not always, modified to fit into the native Yucatec
prosodic system. This phenomenon of linguistic ‘syncretism’, in which bilingual com-
munities of speakers can choose to make their utterances follow the rules of either lan-
guage as an act of claiming membership in one community or another, was described
for the Nahuatl (or Mexicano) speaking communities by Hill and Hill (1986) and Hill
(1999). These works describe syncretic practices as falling along a continuum from
‘more Mexicano’ to ‘more Spanish’, and where an utterance lies can give insight into how
much indigenous identity is being claimed by the speaker. The same process appears
to be occurring in YM, where Spanish words are incorporated to varying degrees into
YM phonological and morphological systems. Instances of phonological syncretism
(when Spanish loan words are prosodically modified) are especially interesting to the
study of stress, specifically because the Spanish input words do not have underlying
tone, nor do they have phonemic length. When they are uttered within the YM ‘accent’,
the prosodic changes they undergo can provide crucial information on the default
stress patterning of YM, information which may be just the clearer window needed
to understand the intricacies of YM stress. When the rules of a particular system are
unclear, looking at syncretic practices when utterances shift from one set of rules to
another can be an invaluable tool in the discernment of the underlying form. This
paper presents data from natural syncretic practices of Spanish loan word adaptation
Stress in Yucatec Maya 
into Yucatec Maya in order to both provide insight into the complicated and under-
documented phenomenon of the phonetic realization of stress, and to demonstrate the
value of looking at syncretic processes for such purposes.
2. Syncretism in language
According to Hill (1999: 1), “syncretism” is a term that was imported into linguistics
from the study of the history of religion, and is defined as “suppression of a relevant
opposition under certain determined conditions”. This act of suppressing one side
of an opposition or the other plays out linguistically when communities are highly
bilingual, and therefore have a choice between two linguistic repositories when speak-
ing. In Hill and Hill’s (1986) work on syncretic practices among Mexicano (Nahuatl)
speakers, they describe a linguistic situation in which a dichotomy between “legitimo
Mexicano” and Castellano (Spanish) is highly relevant to members of this bilingual
speech community. They were found to use syncretic practices in all aspects of speech,
including prosodically. Hill (1999) describes an instance of phonological syncretism
in the following excerpt.
For instance, the Spanish word cajon ‘coffer’, ‘chest’ could be pronounced in
Mexicano as [kaxón], or it could be shifted toward the Mexicano end of the
syncretic continuum by shifting the stress to [káxon], reflecting the invariant
penultimate stress of indigenous Mexicano words, or, even further, by
pronouncing the [x] as [•],1 reflecting the phonological pattern of the earliest
borrowings from Spanish. Thus [ká•ón] might be a self-conscious performance
of indigenous identity, especially on the part of a young or middle-aged male
speaker, while [kaxón] might be an equally self-conscious gesture of political
potency and forward-looking urbanity (Hill 1999: 245).
Although they are culturally and linguistically distinct from Mexicano speakers, an
analogous situation can be argued to exist among Yucatec Mayan speakers. YM speak-
ers interviewed for this study overtly made the distinction between utterances that
were ‘puro Maya’, from those that were not, and recognized and commented on the
influence of Spanish on their language. In incorporating loan words into their speech,
they can choose whether to shift them into a more Mayan style prosodic system, or
maintain the Spanish accent, in order to identify with one or the other dichotomies of
the syncretic situation.
. The symbol [•] is taken directly from (Hill 1999, p. 245), and represents a voiceless alveolar
fricative.
 Emily Kidder
In a similar vein to the example given above, when words from Spanish are
incorporated into YM utterances, they can undergo prosodic shifting from a ‘Spanish
pronunciation’ to one that obeys YM rules of prosody. However, unlike the situation in
Mexicano described by Hill (1999), the rules of YM prosody, particularly the rules of
stress and accent, are not clear-cut and well understood. However, by looking at words
we know to have undergone a prosodic shift from Spanish to YM accent structure, we
can work backwards to better understand what rules might have been applied to create
the changes we see.
Spanish has its own default system of lexical stress and phrasal accent that is unaf-
fected by either a tonal system (which is not present in Spanish), or a phonological
length distinctions (Spanish does not distinguish between long and short vowels) as
in YM. Studying the output of syncretic processes that act on Spanish loans into YM,
i.e. looking into exactly what happens to Spanish words when they are made to sound
more Mayan, can give us a different and possibly greater understanding of the system
of stress in YM than we might otherwise get from looking at native words alone.
3. Stress and accent in language
Before looking deeply into the stress system in YM as evident through its rules of
loan word incorporation, we may want to ask the question, what is or are the underly-
ing properties that scholars have used to define stress? This question has no simple
and straightforward answer, the loosest definition that might work to capture all the
various ways in which stress manifests in different languages, is to say it is some kind
of emphasis, a differentiation in prominence between elements that is structurally
or linguistically meaningful, that is realized in a variety of ways phonetically across
languages. As discussed above, this prominence might be cued by pitch, duration,
intensity, as well as a contrast in precision between it and the surrounding syllables
(particularly the following), or by a specific combination of these features that is often
unique to a particular language or dialect.
4. Lexical stress vs. metrical stress
Some types of emphasis are used on a lexical level, and it is the only characteristic that
differentiates one word from another. This type of stress, typically called lexical or
distinctive stress, is similar to a phonemic distinction, in that minimal pairs of words
are created that would otherwise be functionally the same. For example, the follow-
ing are often considered minimal stress pairs in English: (a) perMIT versus PERmit
and (b) conVICT versus CONvict. No scholars of YM have claimed that it possesses
Stress in Yucatec Maya 
a lexically distinctive stress system of this sort, i.e. there are no documented cases in
which two semantically distinct words differ only in stress (though there is extensive
documentation of lexically distinctive tone).
Metrical stress, on the other hand, does not provide meaningful distinctions
between words, but instead provides the units of a language’s rhythmic structure.
Examples of metrical stress in English can be seen in the following: (a) HOCKey,
(b) MISSiSSIpi and (c) FANCiEST.
Minimal pairs cannot be found for stress of this sort, and instead of being speci-
fied on the lexical level, it can often be reliably predicted by context. Hayes (1995)
describes this distinction as ‘free’ versus ‘fixed’ stress. “Fixed stress is predictable in its
location, and may be derived by rule, while free stress is unpredictable, and must be
lexically listed” (Hayes 1995: 31). As English shows, languages can have a combination
of lexical and metrical stress, or they can have only one or the other. Fixed stress is
often termed the ‘default’ stress system, or the pattern that can often be generalized
to the whole language (commonly with exceptions). Although languages have been
claimed to lack a metrical system, the presence of some sort of metrical stress seems to
be the norm cross-linguistically.
The literature on YM that does look at stress is not consistent as to exactly what
the ‘default’ stress system looks like. Bricker et al. (1998) claim that a default system
does exist in YM, and they describe stress as falling on the final syllable if a word
has no ‘long vowels’,2 if one long vowel is present the stress falls on that syllable,
and if two long vowels are present it falls on the first long syllable. Gussenhoven
and Teeuw (2008), who analyzed YM prosody in order to discern the nature of the
H-tone, have a different assessment of the default stress pattern. They argue that YM
has “three long syllables, Long High, Long Low, and Glottalized, which are stressed,
and a word-initial short syllable, which is stressed” (Gussenhoven & Teeuw 2008: 17).
This specifically differs from Bricker et al. (1998) in two ways: (1) that the tone or
glottalization features are what attract stress, rather than length, and (2) that words
with only short syllables have initial, rather than final stress. Gussenhoven and Teeuw
do not state what happens when there is more than one syllable in a word with tone
or glottalization, and this is admittedly a rare situation, because the majority of YM
stems are monosyllabic.3
. Long vowels described in YM consist of either high-toned long, low-toned long or
glottalized long. These are distinguished from short vowels, which normally carry no tone or
glottalization. This will be discussed further in the sections below.
. The favored syllable type in YM seems to be CVC, where the vowel can be any of the four
‘vowel types’, Long Low, Long High, Long Glottalized or Short.
 Emily Kidder
5. Syntagmatic nature of stress
This monosyllabic nature of YM lexical items also makes identifying a default lexical
stress pattern difficult. Crucially lexical stress differs from other phonological features
that can create a minimal pair and hence prove the existence of a phoneme (i.e. back-
ness, rounding, continuant) in that it has to exist in relation to its surrounding. Looking
at stress on a purely lexical level, a stressed syllable cannot be found on words consist-
ing of only one unit of measurement (whether that is the syllable or mora), there must
be another syllable to compare to the stressed one for prominence to exist, because in
essence, prominence is contrast. And to have contrast, more than one ‘unit’ has to be
present. Without the unstressed syllable, the stressed syllable no longer stands out, and
prominence loses its power to be perceived.
Beckman (1986) and Lindstrom and Remijsen (2005: 2) describe this charac-
teristic of stress as a syntagmatic feature, versus a paradigmatic feature.4
Syntagmatic Paradigmatic
Figure 1. Syntagmatic features versus paradigmatic feature
As is shown in the schematic of Figure 1, “a syntagmatic feature distinguishes a

syllable from those preceding or following it, while a paradigmatic feature contrasts
a syllable from other syllables that may appear in the same position” (Lindstrom &
Remijsen 2005: 2). In other words, syntagmatic features look horizontally at what
comes before and after a segment in question. Paradigmatic features differ in that they
look at one segment, and compare it with the other possible segments that might fill
that position. Stress, as discussed above, is syntagmatic because it requires the pres-
ence of an unstressed element in order to be recognized. Tone, however, is usually
thought to be paradigmatic, because tones are contrasted with a set of other possible
tones, which are present in the phonological inventory, but do not have to be present
in the word in question.
Tone can, and often does exist in one syllable words, because a tonal distinction is
perceived paradigmatically, i.e. in relation to other possible tones that could occur in
its place. It does not necessarily need to be in contrast horizontally, or syntagmatically.
. These concepts originated in de Saussure’s Course in General Linguistics (1974), and
Hjelmslev (1938).
Stress in Yucatec Maya 
One consequence of stress being syntagmatic is that it can and often does effect the
surrounding syllables as well, in that the addition of a stress to a syllable may not raise
its pitch and duration, but shorten those of the surrounding syllables, hence enhancing
the contrast.
6. Grouping properties of stress
Since stress must have surrounding context to be realized, it is natural that scholars
have analyzed stress in terms of its grouping properties. The notion of metrical ‘feet’,
defined by Hayes (1995) as the ‘minimal bracketed units of metrical theory’, is crucial
to the analysis of the rhythmic structure of a language. Feet are generally groups of 2
syllables or moras,5 with stress falling on either the initial (trochee) or the final (iamb).
The language specific patterns and constraints on how syllables or moras are grouped
into feet, and how feet are grouped into larger constituents, determines the fundamen-
tal rhythmic properties of a language. Some languages are sensitive to syllable weight,
meaning a foot can be constructed of a single syllable if the syllable is heavy, while
others are not and feet must consist of 2 separate syllables despite the presence of a
coda or a long vowel.
Krämer (2001) looked at the phenomenon of blocking of particular vowel har-
monic alternations in YM, and claims that it is indeed quantity sensitive, a claim which
he states that Mayan scholars would find uncontroversial. Certain suffixes undergo
vowel harmonic alternations, but harmony is blocked when more than one consonant
intercedes between the suffix and the root vowel. Examples of this phenomenon can
be seen in (1).
(1) Subjunctive suffixes with vowel harmony
a. ah-ak wake.up-subj
b. ok-ok enter-subj
c. lub’-uk fall-subj
d. wen-ek sleep-subj
e. kíim-ik die-subj
Subjunctive suffixes where vowel harmony is blocked
f. tùukul-n-ak think-n-subj
g. hèek’-n-ak break-n-subj
h. ts’íib’-n-ak write-n-subj(Krämer 2001: 11)
. In moraic languages, light and heavy syllables are treated differently, therefore the
quantitative unit of ‘mora’ is used. Light syllables contain one mora, while heavy syllables
contain two.
 Emily Kidder
In (1a) through (1e), the suffixal vowel corresponds to the root vowel, but in (1f)
through (1h), an intervening consonant appears to block the harmony from occurring,
in which case the default vowel /a/ is used.
In other words, coda consonants are argued to act as a blocking mechanism for
harmony, and therefore must be counted as having quantity. In the examples above
which do undergo harmony, the final consonant of the root is syllabified as the onset
of the final syllable, (i.e. a.hak, o.kok, etc.), and onsets do not usually count as bearing
quantity. Therefore, according to Krämer’s analysis, the syllabification of (1f) through
(1h) would create a heavy syllable preceding the vowel harmonic suffix, (i.e. túu.kul.
nak, héek.nak, etc.), and the presence of this extra mora is what blocks the harmonic
process.
A consequence of this analysis is that final syllables in Yucatec would always
be counted as heavy, because of a phonotactic constraint that exists in YM; that all
words must end in a consonant.6 Additionally, Krämer (2001) claims a stress system
exists in YM that is quantity sensitive, so it follows that because of the constraint that
causes heavy final syllables, that final syllables must bear stress in YM. In addition to
this observation, Krämer also proposes that initial syllables attract stress. In a way,
Krämer’s argument combines that of Gussenhoven and Teeuw (2005) and Bricker et
al. (1998), in that it claims stress is attracted to both the initial and final syllable in an
utterance given the right environments.
Using the data seen in (2) below, which are intonational contours7 taken from
Blair and Vermont-Salas (1967), Krämer claims that stress is assigned to the initial and
final syllables in a phrase, as well as to any heavy syllables intervening between them.
(2) Intonational contours (Krämer 2001: 12)8

a. 2ku. me. 2yah2→
pr.3sg work
‘he works’
. Gussenhoven and Teeuw (2008), Blair and Vermont-Salas (1967), and Bricker et al. (1998)
all describe YM as having a system in which word initial onsets are obligatory, as are word
final codas. If a vowel is found in word final position due to phonological processes, a glottal
stop or glottal fricative is epenthesized, creating an obligatorily heavy final syllable.
. Intonational emphases are different from lexical stress, however the syllables on which
intonational accents appear are usually those that bear lexical stress, i.e. the intonational
accent cannot be placed on a syllable that is lexically unstressed.
. The superscript numbers in this figure represent the following: 3 indicates high pitch and
intensity, 2 indicates medially high, 1 lower than medial, and ∅ neutral intonation. The arrow
at the end of each phrase in the examples below indicates whether the terminal intonation
contour of a phrase is stable, falls or rises.
Stress in Yucatec Maya 
b. 2h me. 2yah.-n-a. 2k-en1↓

past work-n-subj-1sg
‘I have worked’
c. 2way. ka. me.ya. 2h-eʔ2↑
q pr-2 work-term
‘Are you working here?’
Superscript numbers, representing levels of intonational accent assignment on the

phrase, are always assigned to the initial and final syllables in both single words (as in
Example (2a)), or more complex phrases, (in Examples (2b) and (2c)). In b, the syllable
‘yah’, which is heavy, is assigned an intonational accent, while in Example (2c), many
intervening light syllables are not assigned an accent.
7. Foot structure in YM
Krämer (2001) claims that foot structure in YM consists of trochaic, bimoraic,

symmetric feet which can is represented visually in (3).
(3) Yucatec Trochaic Feet (Krämer 2001: 13)
(s = strong, w = weak, μ = mora)
a. F b. F
s w s w
µ µ µ µ
(C V V/C) (C V C V)
According to Krämer’s analysis, bimoraic feet are built in YM using either two light
syllables (as in (3a)) or one heavy syllable (3b), at both the left and right boundaries
of the word and of the whole phrase. Heavy syllables that lie in between these bound-
ary feet are also footed and attract stress. When light syllables appear in between the
initial and final feet they are left un-footed, and therefore do not receive any stress. The
foot structure of the examples in (3) above, using Krämer’s analysis, can be seen in (4).
(4) Foot structure (Krämer 2001: 12)
a. 2ku. me. 2yah2→ (ku.me)F (yah)F
pr.3sg work
‘he works’
b. 2h me. 2yah.-n-a. 2k-en1↓ (h.me)F (yah)F na (ken)F
past work-n-subj-1sg
‘I have worked’
 Emily Kidder
c. 2way. ka. me.ya. 2h-eʔ2↑ (way)F ka.me.ya (heʔ)F

q pr-2 work-term
‘Are you working here?’
Interestingly, Krämer (2001) purposefully omits data in which tones are present,
because the analysis makes no claims on whether or not tone would also attract stress,
as it does in many languages. This topic, along with whether or not Krämer’s proposals
are supported in the YM data collected for this study will be discussed in the sections
to follow.
8. Overview of Yucatec phonology
Before moving onto the analysis of stress in YM, it will prove useful to give a gen-
eral overview of YM phonology. Figures 2 and 3 show the basic phonemic inventory
of YM.
Apico- Centro- Dorso-

Bilabial Alveo- Lamino Palato- Palato- Laryngeal
Dental Alveolar Velar Velar
voice - + - + - + - + - + - +
Stop p b t (d) k () ʔ
Fricative (f) s š h
Affricate ts č
Ejective p’ t’ d’ ts’ č’ k’ ’
Nasal m n
Lateral l
Flap r
Glide w y
Figure 2. Consonantal phoneme inventory (Straight 1976: 22)9
. Phonemes in parentheses indicate those which are only present in loan words.
Stress in Yucatec Maya 
front central back
high i
u
Rounded
low e a o
Figure 3. Vowel inventory (Bricker et al. 1998: xiii)
The vowel nuclei of syllables in Yucatec are analyzed by Mayan scholars as falling
into the patterns seen in (5).
(5) Vowel Articulations (Bricker et al. 1998: xiii)

Neutral Low High Glottalized10
a aa áa áʔa
i ii íi íʔi
e ee ée éʔe
o oo óo óʔo
u uu úu úʔu
The pitch distinctions between Neutral, High, Low and Glottalized vowels are pho-
nemic, i.e. they create minimal pairs which differ semantically. Examples of minimal
pairs can be seen in (6).
(6) Vocalic Minimal Pairs in YM11

High Low Glottalized Neutral
a. áak’ – ‘turtle’ aak’ – ‘reed’
b. cháak – ‘rain’ chaak – ‘boil’ cha’ak – ‘starch’ chak – ‘red’
c. kaan – ‘snake’ ka’an – ‘sky’ kan – ‘four’
The acoustic difference between a High and Low tone, áak’ – ‘turtle’ and aak’ – ‘reed’,
can be seen in the spectrograms in Figure 4.12
. Glottalized vowels are also called Rearticulated in YM literature, and are often realized
with creaky voice rather than by a complete glottal closure.
. Finding complete sets of YM words that differ only in vowel type are rare, more often a
set will exist between High and Low, Low and Glottalized, etc. The example in 5b is one of the
few with each type represented.
. The ejective [k’] is clearly seen at the end of each utterance by the period of closure
followed by a short burst.
 Emily Kidder
áak’ aak’

Figure 4. F0 of High versus Low tone13
High tones can be realized in YM as steadily rising, as in Figure 4, or as contours

which rise and then fall (or vice versa), depending on the intonation contours pres-
ent in speech. Low vowels are more uniformly level in F0, though they can also fall
or rise depending on location in the intonational phrase. Neutral tones are realized
with shorter durations than the other three articulation types, and lack any phonemic
tonal distinctions, though they can appear in surface forms as the result of reduction
of High or Glottalized vowels. Figures 5 and 6 show the acoustic features of Neutral
and Glottalized vowels.
bak’

Figure 5. Neutral vowel – bak’ – ‘meat’14
. This spectrogram was created using Praat software. The lower line represents f0, and the
upper line represents amplitude.
. The length distinction between the vowels in Figure 4 through Figure 6 are not read-
able from these spectrograms. The length of the vowels are as follows: aak’ – 0.255243,
áak’ – 0.230341, bak’ – 0.092760, ba’al – 0.209306 (all measurements are in seconds). The
neutral vowel is less than half as long as the other three vowel types, which echoes findings
in Frazier (2009).
Stress in Yucatec Maya 
ba’al

Figure 6. Glottalized vowel – ba’al – ‘thing’15
Glottalized vowels have long durations, but the initial rising vowel is interrupted
medially by a glottal stop or more often by creaky voice, as it is in Figure 6, which
is documented in Frazier (2009) as well. In rapid speech, these can be shortened to
High Long vowels, or shortened further to High short vowels (Blair & Vermont-Salas
1967: 15), or are realized with creaky voice (Frazier 2011).
Vowels that are marked for either Low or High tone are phonetically longer, and
orthographically are written as geminates. In this paper, Low tones will be designated
as geminate vowels with no diacritic, while High tones will be designated by a gem-
inate vowel with a rising diacritic (this is consistent with the orthography adopted
by the Mayan community in the Yucatán for its language materials), as in Blair and
Vermont-Salas (1967), Limón-Rojas (1997), Navarrete (2009). As can be seen in the
list of vocalic minimal pairs in figure (6), Low, High and Glottalized vowels always
appear orthographically as long vowels, and this is representative of the way these
vowels have been documented. There are no long vowels that are not considered by
YM scholars to carry one of these three phonological features.
9. Methodology
Data for this study was collected from 8 native speakers, all bilingual in Spanish, who
resided in Northern California. In recent years, there has been a significant growth in
immigration from YM speaking communities of the Yucatán to the United States, and
the main locus of this new community of Yucatecos is in the San Francisco Bay Area.
As of 2010, there were an estimated 10,000 Yucatec Mayans resettled there, the major-
ity of them having come during the last 6–7 years (Delugan 2010). Two main commu-
nities have developed, one in the Mission District of San Francisco, and the other in the
. In this instance the glottalized vowel appears with creaky voice or glottalization, and
therefore the f0 stops being readable after a short time.
 Emily Kidder
Canal District of San Rafael, north of San Francisco, though both have ties with each
other and participate in community events and gatherings. The neighborhood cen-
tered around the intersection of 16th and Mission in the Mission District has become
known as Little Yucatán, and Yucateco restaurants and cultural events are common
in this area, although it remains an incredibly diverse population with immigrants
from cultures all over the world. In both San Francisco and San Rafael, grassroots
organizations have developed which advocate for the community, organize cultural
events both for the enjoyment and maintenance of Yucateco culture and language, and
to educate those outside of the community. Asociación MAYAB (Maya Yucateca de la
Area de la Bahia) in San Francisco, and Chan Kahal in San Rafael provide a variety of
important services to members of this community, and help to maintain cultural ties
to the Yucatán.
The San Francisco community of Yucatecos primarily come from the area sur-
rounding Oxcutzcab, Mexico, while those in San Rafael mainly come from nearby
Peto. There is also a separate community of immigrants from Southern Mexico and
Guatemala, who speak the related but quite distinct Mayan languages of Quiche,
Tzeltal, Chol, Tzotzil and Mam, that have settled in the Mission District and across
the San Francisco Bay in Oakland (Delugan 2010). The participants in this study were
originally from the city of Oxcutzcab, and the surrounding towns, including Chumayel,
Akil and Peto. The following map shows where these cities are located in the Yucatán.
Yucatán, Mexico
Mérida
Valladolid
Chumayel
Oxcutzcab
Akil Peto

Figure 7. Map of the Yucatán Peninsula, México
Frazier (2009) describes this geographical area as the home of the ‘western dialect’
of YM, which can be distinguished from the eastern dialect by the presence of tonal
distinctions.
Stress in Yucatec Maya 
10. Subjects
All 8 speakers were fully bilingual in Spanish, ranged in age from 30 to 60, and
reported that they speak Spanish outside the home and to their children, grand-
children, nieces and nephews, while they speak YM to their spouses, parents, aunts
and uncles, and older friends. YM itself is not considered endangered, but Spanish
is quickly overtaking YM as the language spoken by the younger generations, as
it is the language of education. The majority of adult YM speakers are fully bilin-
gual in Spanish, and elderly speakers make up the vast majority of monolingual YM
speakers.
All participants for this study were contacted through the Asociación MAYAB
community center in San Francisco and the Chan Kahal Mayan Association in the
Canal District of San Rafael, and recording took place at the headquarters of these
organizations, which was deemed to be the most comfortable and convenient location
for participants.
11. Elicitation materials
During recording sessions, the types of data seen in (7) were elicited from speakers.
(7) Data
a. 98 elicited words followed by the word used in a sentence and on
optional frame sentence that was identical for all words (7 speakers,
4 using a frame sentence, and 3 without the frame sentence)
b. Conversations elicited through given topics (2 speakers)
c. Stress placement intuitions for 55 words (4 speakers)
The data was recorded using a Mini DV video recorder, enhanced with a Sennheiser
shotgun microphone. Participants were asked whether they were comfortable being
videotaped as well as audiotaped, and 5 opted to not be recorded visually. All inter-
views were conducted in Spanish.
For the data collected of type (7a), YM words and phrases were displayed in both
YM orthography and in Spanish translations, as in the following example.
(8) Sample of elicitation materials

-chúuk carbon
-Juane’ ku konik chúuk Juan vende carbon
The words and sentences were taken from a Yucatec Maya-Spanish Dictionary,
Navarrete (2009), and were chosen to be representative of different syllable types and
number of syllables, different parts of speech, and different tones and vowel nuclei.
 Emily Kidder
All speakers were literate in Spanish, and the majority of the speakers were more
comfortable reading the Spanish words and sentences, and translating them into YM.
The YM orthography is not in common use among the communities, and though
they could make it out with effort, they all either preferred to read and translate the
Spanish or have the words and sentences read aloud to them in Spanish for them to
translate into YM.
12. Tasks
After reading the words alone, and in a sentence, a subset of the participants were asked
to read an additional frame sentence in order to determine whether or not phrasal
intonation might have an effect on the word readings. The frame sentence used was
él dice la palabra ___ en todo el tiempo (in English: ‘he says the word ___ all the time’).
To put this sentence in some context for participants, they were told to imagine a child
was just learning to speak and that he was repeating the same word over and over. Data
from these different environments was collected in order to document lexical items
on their own, in a semantically relevant phrasal context, and in a phrasal context that
remains consistent throughout words.
Each speaker was given a practice example and instructions, and were then
asked to say aloud each word, sentence, and frame sentence in YM, or to translate
the Spanish into YM if that was preferred. In order to elicit the most natural speech
patterns, they were also given the instruction that they should translate the words and
sentences in a way that felt natural (i.e. if the words chosen were not commonly used
or archaic, which turned out to be the case for a subset of the chosen words, they were
encouraged to use a different word that was more natural). They were also instructed
that it was acceptable to substitute a Spanish word if it was what they would normally
use in a particular context. During recording sessions, speakers often gave their own
intuitions on which words were the most natural, which were uncommon, and which
were not used at all, and in which cases the Spanish loan word was the most natu-
ral. Instances in which a speaker opted to use a Spanish word will be analyzed in the
following sections.
Conversational data was also collected from 2 speakers, who were asked to con-
verse on topics related to language use and education in YM. Specifically they were
asked to discuss which situations they speak Spanish versus Maya, what kind of edu-
cational aspirations they have for their children, and whether they plan or desire to
return to the Yucatán in the future. This also provided many instances of loan word
usage, and there was robust use of code-switching and syncretic processes occurring
during this conversation.
Stress in Yucatec Maya 
13. Loan word usage
The use of Spanish loan words is not a rare occurrence in most Mayan communi-
ties, in fact, according to Bricker et al. (1998: xiv) “Spanish loans are so common in
Maya that an unlimited number of them occur in everyday speech.” Syncretic prac-
tices are to be expected in a language with such a degree of mixing and bilingualism.
Loan words were found to be introduced sporadically throughout the speech data
collected for this study, often the speakers explicitly called out the words as being of
Spanish origin, though they were not always acknowledged as such, and some may
have undergone incorporation to such a degree that the speakers were unaware they
originated in Spanish. As in Hill’s (1999) description of the variable realizations of
the Spanish word cajón by Mexicano speakers described above, there appear to be
a range of different levels of incorporation of Spanish words into Yucatec. In some
cases, the words retain Spanish morphology, while in others they occur with Mayan
morphology.
The following table shows some examples of these cases.
(9) Loans With Spanish and Mayan Morphology

Word YM context Gloss
a. pecados taak in kopesatik pecados ‘I must confess my sins’
b. odiarme jach odiarme ‘(he/she) really hates me’
c. nietos in nietoso’obe’ ‘my grandchildren’
d. pretender baax ka pretendertik ‘what are you intending’
As can be seen in Examples (9a) and (9b) above, some of the Spanish loans retain their
morphology, i.e. pecado is pluralized by adding -s instead of the YM -o’ob, and the
Spanish reflexive morpheme me remains in place. While in Examples (9c) and (9d),
the YM plural (/-o’ob/) and YM causative morphology (/-tik/) are incorporated onto
the loan word.
The most interesting instances of loan word incorporation in this data include
examples in which individual speakers naturally uttered the same word on both sides
of the syncretic continuum between Spanish and Yucatec. This happened in a subset of
cases, which will be discussed in detail below.
14. The syncretic continuum in YM
Three separate instances of one speaker uttering a single word twice, once on both
sides of the syncretic continuum, were present in this data. There were obvious
prosodic differences between the two pronunciations. Two of these instances came
 Emily Kidder
from one female speaker during the spontaneous conversation portion of the data
collection, and the third came from a different male speaker during the single word
elicitation portion. The examples in (10) show the three words and their standard
Spanish pronunciations, while (11) shows the contexts in which they surfaced in
the data.
(10) Loan Words16

Word Meaning Spanish Pronunciation
a. trabajo ‘job’, or ‘I work’ /tra.bá.ho/
b. abuelo ‘grandfather’ /a.bwé.lo/
c. palabra ‘word’ /pa.lá.βra/
(11) YM versus Spanish Context

Context English Gloss
a. trabajo k’áax ta meyaj waye’ ‘I work (in the) jungle, you work
here’
yo puedo porque tiene trabajo ‘I can because he has a job’
b. in abuelo desde uuche maya ‘my grandfather (spoke) maya since
long ago’
mi abuelo desde antiguamente ‘my grandfather since long ago’
c. leti’e’ ku yaak’ le palabra chúuk ‘he says the word charcoal all the
time’
el dice la palabra carbón en todo ‘he says the word ‘charcoal’ all the
el tiempo time’
These three words all happen to be trisyllabic and have stress on the second syllable
in the origin language. If the assertions of Krämer (2001) are accurate, and YM has
the trochaic quantity sensitive system described above, words of this type would be
expected to undergo prosodic shifts when incorporated into the foot structure of YM.
The predictions assumed by Krämer’s model can be seen in (12).
(12) Predictions for YM stress

a. The initial syllable will be stressed
b. A glottal stop or fricative will be epenthesized word finally creating a
final heavy syllable
c. The final syllable will be stressed
. Accent marks denote the stressed syllables, and periods denote syllable boundaries.
Stress in Yucatec Maya 
Figure (13) shows how the words trabajo, abuelo and palabra should fit into Krämer’s
model for YM foot structure.
(13) Predicted Foot Structure for incorporated Spanish Loans

F1 F2
s w s w
µ µ µ µ
1. ( tra ba ) ( ho ʔ/h )
2. ( a bwe ) ( lo ʔ/h )
3. ( pa
la ) ( bra ʔ/h )
If these predictions were attested, it follows that the first syllable in the word should
undergo a phonetic change to signal the shift of foot structure rules from the Spanish
to YM sides of the syncretic continuum. Also, the last syllable should undergo the
epenthesis of either a glottal stop or a glottal fricative, in order to satisfy the final
consonant constraint posited in the literature. The question now arises whether or
not this change will be signaled by a rise in pitch, as is common in many languages
(including in the origin language of Spanish according to Ortega-Llebaria 2006), by
lengthening of the syllable, by intensity, or by a combination of all three. Because
these words originate in Spanish, they should not have an underlying tonal specifica-
tion, which can act as an obstacle when looking at native YM words. Measuring both
the relative pitch levels and the syllable length of each word, as well as the intensity
when uttered on either side of the continuum sheds light on how foot structure is
signaled in YM.
15. Pitch in YM syncretic shifts
First we will investigate what role pitch might play in YM stress placement. Figure 8
shows the pitch measurements for each word pronounced on both sides of the
continuum.17
. Pitch measurements were taken using Praat software from the vowel of the nucleus of
each syllable. The first data point was taken from the initial pitch of the vowel, and the second
was taken from the highest or lowest F0 point reached.
 Emily Kidder
Word Language Pitch Range in HZ

S1 S2 S3
trabajo Spanish 2416–1808 1861–2495 1570–829.8
YM 1596–1491 1491–1279 1279–1094
abuelo Spanish 988–935 1121–1332 1306–1173
YM 1226–1464 1358–1226 1306–1253
palabra Spanish 456–282 352–559 594–594
YM 535–470 557–492 427–514
Figure 8. Pitch measurements for Spanish loans
You can see this data represented visually in Figures 9 through 11.
3000
2500
2000
Spanish
1500
Yucatec
1000
500
0
S1 Start S1 End S2 Start S2 End S3 Start S3 End
Figure 9. Syllable pitch in trabajo
1600
1400
1200
1000
Spanish
800
Yucatec
600
400
200
0
Figure 10. Syllable pitch in abuelo

Stress in Yucatec Maya 
700
600
500
400
Spanish
Yucatec
300
200
100
0
Figure 11. Syllable pitch in palabra
As is evident in these graphs, there appears to be a pattern that when words are
pronounced with Spanish prosody, the pitch rises on the stressed second syllable, the
syllable preceding this falls slightly, and the third syllable falls or stays level. This find-
ing is in agreement with work on Spanish prosody that argues that stress is primarily
cued by a rise in pitch on the stressed syllable. In YM, there appears to be no regular
pitch pattern discernible from these three examples. In Figure 9, the three syllables
appear to be all roughly declining at similar rates, in Figure 10 the pitch rises on the
first syllable and falls on the second, and finally in Figure 11 the pitch in YM declines
slightly on the first two syllables and rises slightly on the final, while in Spanish the first
syllable sharply falls, the second sharply rises, and the third remains level. From this
initial investigation, pitch appears to be an unclear representative of stress in Spanish
loan words into YM.
16. Intensity in YM syncretic shifts
Even though intensity has been demonstrated by many scholars (Fry 1955; Fry 1958;
Morton & Jassem 1965; and Nakatami & Aston 1978) to play a subsidiary role to pitch
and duration in the perception of stress cross-linguistically, it should not be com-
pletely discounted as a possible factor in the cueing of stress in YM, in case it might
be a predictable indicator of the stressed position. Figures 12 through 14 below shows
the difference in intensity levels between the Spanish and YM pronunciation of abuelo,
trabajo and palabra.
 Emily Kidder
982.200308 982.930938
100
Intensity (dB)
50
982.2 982.9
Time (s)
Figure 12a. Syllable intensity – abuelo Spanish
978.141648 978.623256
100
Intensity (dB)
50
978.1 978.6
Time (s)
Figure 12b. Syllable intensity – abuelo YM

Stress in Yucatec Maya 
541.503089 542.122504
100
Intensity (dB)
50
541.5 542.1
Time (s)
Figure 13a. Syllable intensity – trabajo Spanish
510.635656 511.182464
100
Intensity (dB)
50
510.6 511.2
Time (s)
Figure 13b. Syllable intensity – trabajo YM

 Emily Kidder
100
Intensity (dB)
50
126.2 126.7
Time (s)
Figure 14a. Syllable intensity – palabra Spanish
100
Intensity (dB)
50
370.9 371.4
Time (s)
Figure 14b. Syllable intensity – palabra YM
The data shown in Figures 12 through 14 don’t appear to show any particularly
strong instances of syllable intensity peaks differing greatly between the first and
second syllable of each word. They also don’t appear to show any noticeable distinc-
tion in peak level between the Spanish and YM pronunciations. Though a study with
more data points would be necessary to determine with certainty, it appears as if
intensity is not an obvious indicator of potential stress placement, which would be as
expected given that intensity has been shown to play a minimal role in signaling stress
cross-linguistically.
Stress in Yucatec Maya 
17. Length in YM syncretic shifts
Since in this preliminary investigation, neither pitch nor intensity appear to be clear
markers of stress shift into the YM syncretic continuum, the next logical assumption
is that duration may be a more robust cue. In fact, it might be a more logical predic-
tor than pitch, since YM has been demonstrated to be phonemically tonal, while long
vowels always co-occur with the additional features of High tone, Low tone or Glot-
talization. Figure 15 shows the measurements of length for trabajo, abuelo and palabra
in both Spanish and YM.
Word Language Length in seconds

S1 S2 S3
trabajo Spanish 0.139039 0.180492 0.194309
YM 0.199454 0.123954 0.139730
abuelo Spanish 0.171481 0.294948 0.241789
YM 0.233391 0.137071 0.161768
palabra Spanish 0.056 0.144439 0.070080
YM 0.100279 0.060474 0.091859
Figure 15. Loan word syllable length and pitch range measurements
This data conforms to a more regular pattern than the pitch data described above.
In the Spanish cases, each conforms to the general rule that S2 is longer than S1, while
in YM, the opposite is the case. Figure 16 represents the length data visually.
0.3
0.25
0.2
Syllable 1
0.15 Syllable 2
Syllable 3
0.1
0.05
0
Spanish - Maya - Spanish - Maya - Spanish - Maya -
tra.ba.jo tra.ba.jo a.bue.lo a.bue.lo pa.la.bra pa.la.bra
Figure 16. Visual representation of length measurements

 Emily Kidder
The graphs in Figures 17 through 19 show the same data for each individual word.
It is evident from the data on syllable length that when moving from Spanish
pronunciation to YM pronunciation, the length ratio between the first and second
syllables changes distinctly. If length was a cue for stress, and if stress is placed in YM
according to the structure in (13) above, we would expect the first and last syllable in
these words to be longer than the second, and this prediction is borne out in these
examples.
0.25
0.194309 0.199454
0.2
0.180492
0.139039 0.13973
0.15 Syllable 1
0.123954
Syllable 2
Syllable 3
0.1
0.05
0
Spanish Yucatec
Figure 17. Syllable length – trabajo
0.35
0.294948
0.3
0.241789
0.25 0.233391
0.2 Syllable 1
0.171481
0.161768 Syllable 2
0.15 0.137071 Syllable 3
0.1
0.05
0
Spanish Yucatec
Figure 18. Syllable length – abuelo

Stress in Yucatec Maya 
0.16
0.144439
0.14
0.12
0.100279
0.1 0.091859 Syllable 1
0.08 Syllable 2
0.07008
0.060474 Syllable 3
0.06 0.056
0.04
0.02
0
Spanish Yucatec
Figure 19. Syllable length – palabra18
18. Testing the predictions
Of the three predictions in (12) for how this data would surface after shifting to the
YM side of the syncretic continuum, when incorporated into the YM prosodic system,
both (12a) and (12c) appear to be validated in this data set, i.e. the initial and final
syllables appear to be more prominent, if length is the cue for prominence. However,
prediction (12b), that the final syllable would undergo final glottal epenthesis, has not
yet been looked at. According to the both the attested prosodic rules of YM and the
predicted foot structure seen in (12) above, all of these YM words should be realized
with final glottalization in order for the final syllable to be counted as heavy. The data,
however, does not appear to show any signs of final glottalization in any of the three
instances described thus far.
Final glottalization can be seen in the spectrogram as an abrupt stop, or more
commonly in YM as a series of glottal pulses known as creaky voice. An example of the
YM word ‘wayeʔ ‘here’, which has a final glottal stop, can be seen in Figure 20 below. 19
. The speaker who uttered the word ‘palabra’ on both sides of the syncretic continuum did
this in repeated instances in the data, in most cases with the YM morpheme ‘-oʔ’ attached to
the end. The instance of ‘palabra’ used for these measurements was the only one in which this
morpheme was not attached, and was chosen so that the analyses would be unobscured by the
addition of another YM morpheme.
. The upper line in the spectrogram denotes intensity and the lower line denotes
fundamental frequency.
 Emily Kidder
If the Spanish loan words in question undergo final glottal epenthesis, we would expect
to see similar evidence of creaky voice or closure at the end of the words, but as you can
see in Figures 21 through 23, this is not evident.
waye’
Figure 20. Spectrogram of YM wayeʔ
pa la bra
Figure 21. Spectrogram of YM palabra
‘a bwe lo
Figure 22. Spectrogram of YM abuelo

Stress in Yucatec Maya 
tra ba jo
Figure 23. Spectrogram of YM trabajo
These spectrograms show no signs at all of glottalization in final position akin to

that found in the native YM word wayeʔ from Figure 20, although abuelo in Figure 22
does show evidence of an initial glottal stop before the first syllable. The glottal pulses
in initial position in abuelo show that for this speaker the YM rule for glottal epenthe-
sis appears to be valid in initial position, but not word finally.
However, as was seen in the measurements from Figures 17 through 19 as well
as in the spectrograms in Figures 21 through 23, the final syllable is lengthened in all
three instances. This requires a slight revision to the foot structure posited in (13),
which is shown in (14).
(14) Attested Foot Structure for incorporated Spanish Loans20

F1 F2
s w s w
µ µ µ µ
1. ( tra: ba ) ( ho :)
2. ( ʔa: bwe ) ( lo :)
3. ( pa:
la ) ( bra :)
This final lengthening differs from the expectation that a final consonant would be
added. Further study of this process with a larger data set would be needed in order
. Another possible model of the attested foot structure could be ‘(tra:) ba (jo:)’, ‘(a:) bwe
(lo:)’, and ‘(pa:) la (bra:)’ in which the initial and final lengthening cause the middle syllable to
be unfooted. This would also be in line with Kramer’s analysis, which states that stress occurs
on the initial and final position, and that any non-heavy syllables that appear between them
remain unfooted.
 Emily Kidder
to determine whether or not the omission of the final glottal in exchange for the
lengthening of the final vowel is limited to Spanish loans, or whether it is a dialec-
tal difference only present in the community of YM immigrants living in the San
Francisco Bay Area. However this might give further evidence that duration is the
salient cue for stress in YM, since the language seems to have an obvious system in
place that functions to adapt Spanish words into the YM prosodic system, and this
system adds length to the segment that was predicted to be stressed by Krämer’s
(2001) model.
19. Conclusion and implications for further study
The preliminary investigation into the prosodic changes between sides of the syncretic
continuum between Spanish and Yucatec Maya supports Krämer (2001) in the view
that YM stress consists of a bimoraic and trochaic foot structure. As can be seen in
the examples above, when Spanish loan words were incorporated into YM prosodic
rules, the duration of the first and last syllable was markedly longer than the second,
a shift which is predicted by utilizing Krämer’s posited foot structure. Although this
data comes from a small set of incorporated loan words, this suggests that duration
may be the most salient cue for stress in YM, rather than pitch or intensity. More in
depth work on this subject is needed in order to determine whether or not the pattern
found in this small data set is held up throughout the YM lexicon, however utilizing
the notion of the syncretic continuum has provided an invaluable resource for the
investigation of stress in this language. Looking at the prosodic changes between one
side of the syncretic continuum and the other in this bilingual community can help to
elucidate the factors at play that would otherwise be very difficult to untangle. Not only
do syncretic practices give insight into which part of their culture a speaker chooses
to identify with in particular utterances, but they can also be a looking glass into the
underlying processes at work that may only be clearly seen when moving across the
syncretic continuum.
References
Beckman, Mary. 1986. Stress and Non-Stress Accent. Dordrecht: Foris.

Blair, Robert W. & Vermont-Salas, Refugio. 1967. Spoken (Yucatec) Maya. Chicago IL: Depart-
ment of Anthropology, University of Chicago.
Bricker, Victoria, Yah, Eleuterio Po’ot & Dzul de Po’ot, Ofelia. 1998. A Dictionary of the Maya
Language as Spoken in Hocabá, Yucatán. Salt Lake City UT: The University of Utah Press.
Delugan, Robin Maria. 2010. Indigeneity across borders: Hemispheric migrations and cosmo-
politan encounters. American Ethnologist 37(1): 83–97.
Stress in Yucatec Maya 
Frazier, Melissa. 2009. The Production and Perception of Pitch and Glottalization in Yucatec
Maya. Ph.D. dissertation, University of North Carolina at Chapel Hill.
Frazier, Melissa. 2011. Tonal dialects and consonant-pitch interaction in Yucatec Maya. In New
Perspectives on Mayan Linguistics, Heriberto Avelino (ed.), 21–55. Newcastle upon Tyne:
Cambridge Scholars.
Fry, Dennis B. 1955. Duration and intensity as physical correlates of linguistic stress. Journal of
the Acoustical Society of America 27(4): 765–69.
Fry, Dennis B. 1958. Experiments in the perception of stress. Language and Speech 1(2): 120–152.
Gussenhoven, Carlos & Teeuw, Renske. 2008. A moraic and a syllabic H-tone in Yucatec Maya.
In Fonología Instrumental: Patrones Fónicos y Variación, Esther Herrera Zendejas & Martin
Pedro Butragueño (eds), 49–72. México DG: El Colegio de México.
Hayes, Bruce. 1995. Metrical Stress Theory: Principles and Case Studies. Chicago IL: The Univer-
sity of Chicago Press.
Hill, Jane H. 1999. Syncretism. Journal of Linguistic Anthropology 9(1–2): 244–246.
Hill, Jane H. & Hill, Kenneth C. 1986. Speaking Mexicano: Dynamics of Syncretic Language in
Central Mexico. Tucson AZ: The University of Arizona Press.
Hjelmslev, Louis. 1938. Essai d’une theorie des morphemes. In Actes du Quatrieme Congres
International de Linguisties: Tenu a Copenhague due 27 au l er septembre 1936, 140–151.
Copenhague: E Munksgaard.
Krämer, Martin. 2001. Yucatec Maya vowel alternations. Harmony as syntagmatic identity.
Zeitschrift für Sprachwissenschaft 20(2): 175–217.
Limón Rojas, Miguel, Secretaría de Educación Pública. 1997. Diccionario de la lengua Maya.
Instituto Nacional para la Educación de los Adultos.
Lindstrom, Eva & Remijsen, Bert. 2005. Aspects of the prosody of Kuot, a language where into-
nation ignores stress. Linguistics 43(4): 839–870.
Morton, John & Jassem, Wiktor. 1965. Acoustical correlates of stress. Language and Speech 8(3):
159–181.
Nakatami, Lloyd & Aston, Carletta H. 1978. Acoustic and linguistic factors in stress perception.
Ms, ATT Bell Laboratories.
Navarrete, Javier Abelardo Gómez. 2009. Diccionario introductorio Español – Maya Maya –
Español. Quintana Roo MX: Universidad de Quintana Roo.
Ortega-Llebaria, Marta. 2006. The phonetic cues to stress and accent in Spanish. In Selected
Proceedings of the 2nd Conference on Laboratory Approaches to Spanish Phonetics and
Phonology, Manuel Díaz-Campos (ed.), 104–118. Somerville MA: Cascadilla Proceedings
Project.
de Saussure, Ferdinand. 1974. A Course in General Linguistics. Glasgow: Fontana/Collins.
(Transl. by Wade Baskin from the 3rd edn of Cours de Lingquistica Generale. Paris: Payot,
1931, first edn. in 1916).
Straight, H. Stephen. 1976. The Acquisition of Maya Phonology: Variation in Yucatec Child
Language. New York NY: Garland.
The phonetic correlates of Southern
Ute stress
Stacey Oberly*
Endangered languages are in need of urgent documentation. Following the

phonetic tradition of Ladefoged and Maddieson, this chapter reports results
of the first phonetic instrumental analysis of stress in Southern Ute, a severely
endangered Uto-Aztecan language. The phonetic correlates of stress could
include changes in duration, pitch and/or intensity (Ladefoged 2006). This
analysis is based on a list of 100 words designed to determine the phonetic
correlates of stress. Five speakers (three female and two male) were recorded
using a frame sentence to control intonation. Stress placement data was elicited
from one male and one female speaker. This prosodic information forms the
basis of revitalization and documentation efforts as well as further phonological
investigation.
Keywords: Uto-Aztecan; stress; phonetic correlates; pitch; Ute
1. Introduction
As a scientific field, phonetics systematically analyzes human speech sounds using

segmental distinctions and state of the art technology. Ideally, these analyses are
based on cross-linguistic data from a wide variety of language families. This c hapter
provides a phonetic analysis of Southern Ute, a severely endangered Uto-Aztecan
language on the Sothern Ute reservation, located in Southwestern Colorado. Specifi-
cally it describes the phonetic correlates of Southern Ute stress. Southern Ute is from
the southern Numic branch of Northern Uto-Aztecan, which was one of the language
families that Dr. Jane Hill studied. In the theoretical domain, Numic languages provide
rich data for phonetic and phonological analysis.
* I would like to thank Dr. Natasha Warner, Dr. Amy LaCross and Ethan Dickenson for
running the statistical analysis. I am grateful to the Ute speakers that have shared their
knowledge with me. All errors are my own.
 Stacey Oberly
From a phonetic/phonological perspective, stress in Southern Ute is interesting

because it appears to interact with vowel devoicing and seems to be affected by its agglu-
tinative morphology. Describing Southern Ute stress is imperative as it will p rovide the
Ute community members who are learning Ute as a second language with important
prosodic information. This research topic arose when a fluent Ute speaker heard some
Ute second language learning children speak Ute on the local tribal radio station. The
speaker stated that “the children have all the right words in the right order but did not
“sound Ute.” This comment raised the question about what prosodic information do
the children need to learn to “Sound Ute.” My hypothesis is prosodic information such
as duration, pitch and stress would help. This prosodic information will form the basis
of revitalization efforts as well as further p honological investigation. This research is
important because although there are approximately 1,400 enrolled members of the
Southern Ute tribe, according to a 2002 informal language survey, there are only forty
remaining speakers, who are all over the age of sixty.
This research benefits the Southern Ute community, the linguistic community and
other indigenous communities by providing a model for phonetic analysis of an endan-
gered language utilizing fluent speaker intuition about stress. It serves as an example
of an indigenous community determining the direction of research that services
community needs as Dr. Jane Hill advocates in her work in linguistic anthropology
(1985, 1989, 1995, 1998; Hill & Hill 1980), and language rights (2002). It is imperative
that Southern Ute phonetic properties are analyzed, documented and archived before
the small number of fluent speakers die, leaving no digital audio recordings behind for
future generations.
There are relatively few instrumental acoustic analyses of Native A merican
languages, making this research important. The previous acoustic studies of Indig-
enous languages include: McDonough’s work on Navajo (2003); Gordon’s analysis of
word-level stress in Chickasaw (2004); Gordon et al.’s work on the phonetic structures
of Western Apache (2001); the work on Cahuilla by Seiler (1957); Michelson’s
dissertation on Mohawk and Oneida (1983); Everett’s analysis of the acoustic cor-
relates of Pirahã stress (1998); an analysis of Cayuga stress by Doherty (1993); Tuttle’s
dissertation on Tanana (1998); a stress analysis of Witsuwit’en by Hargus (2001); and
Martin & Johnson’s work on tonal accent in Creek (2002). The work of Ladefoged
and Maddieson (1996) also adds to the instrumental acoustic analysis of the world’s
languages.
1.1 The Southern Ute tribe

Before European contact, the Ute people hunted and gathered on foot in small family
groups in Colorado, Utah, Arizona, New Mexico, Wyoming and Oklahoma (Jefferson
et al. 1972). Occasionally they would plant corn, beans, and squash in mountain
The phonetic correlates of Southern Ute stress 
meadows and harvest them in the autumn (Southern Ute Tribal Website 2011). In
1896, the Ute tribe was divided onto three different Ute reservations: Southern Ute and
Ute Mountain Ute, both in Southwestern Colorado; and Northern Ute in central Utah
(Jefferson et al. 1972: 95). Currently, the Southern Ute reservation is a checkerboard
piece of land consisting of 1,058,785 square miles in three counties, La Plata, A
rchuleta
and Montezuma (Southern Ute Tribal Website 2011). A checkerboard reservation
occurred in 1899 when un-allotted reservation lands were opened up to settlers which
resulted in reservation lands being scattered among plots of non-reservation land.
A map of the Southern Ute reservation is shown in Figure 1.
The headquarters for the Southern Ute tribe is located in Ignacio, Colorado.
HUNTING
GROUNDS
Colorado
WHITE
RIVERS
W
EE
MI UNCOMPAHGRE HUNTING Colorado
NU (TABEGUACHES) UINTAH
CH GROUNDS
E OURAY
CAPOTE
MUACHE
HUNTING
GROUNDS UTE SOUTHERN UTE
MT
Figure 1. Southern Ute traditional territory (left) and current reservation (right)
(Jefferson et al. 1972: xi)
1.2 Language information

Southern Ute is a member of the large Uto-Aztecan language family (Lewis 2009).
It belongs to the Southern Numic branch, which also includes Southern Paiute,
Chemehuevi and Kawaiisu as shown in Figure 2.
Northern Uto-Aztecan
Hopi Numic Takic Tübatulabal
Western Central Southern
Mono Comanche Ute
No. Paiute Panamint So. Paiute
Shoshone Chemehuevi
Figure 2. Northern Ute-Aztecan branch (Lewis 2009)

 Stacey Oberly
The previous work on Southern Ute includes three dictionaries (Goss 1961; Givón
1979; Charney 1996), two grammars (Givón 1980 & 2011), two dissertations (Goss
1972 & Oberly 2008) and a collection of traditional narratives (Givón 1985).
1.2.1 Southern Ute language status

According to Fishman’s (1991) Graded Intergenerational Disruption Scale, Southern
Ute is at stage 8, which is the most advanced stage of language attrition. At stage 8,
the degree of attrition is so advanced that the few elder speakers are socially isolated
that the language “needs to be re-assembled from their mouths and memories and
taught to demographically unconcentrated adults” (88). As a result English is the first
language of the current generation of parents which means the children do not learn
Southern Ute as a first language. These facts make linguistic analysis, documentation
and revitalization imperative. According to Grenoble and Whaley (2006),
“Simply put, a seriously endangered language should be documented as quickly

and as thoroughly as possible. The more extensive the documentation, the easier
revitalization (or even reclamation) will be in the future should a community
desire it…revitalization efforts rely on dictionaries and descriptive grammars,
recorded speech and so on… (5)”
This research effectively advances the linguistic knowledge and understanding of the
Southern Ute language.
1.2.2 Southern Ute Consonants

Out of respect for tribal sovereignty, the data is this chapter is written in the official
Southern Ute orthography developed by Charney in 1996. Data from Givón (1979,
1980, 2001a, 2001b) has been rewritten into the 1996 orthography. Chart 1 is a
consonant chart with the International Phonetic Alphabet symbols in brackets.
Notice that obstruents make up the largest class of Southern Ute Consonants.
Chart 1. Southern Ute Consonants (Givón 1980: 3–5)

Place of Articulation
MANNER VOICE Bilabial Dental Palatal Velar Uvular Glottal
Obstruent Stops Voiced g’[g]

Voiceless p [p] t [t] ch [tʃ] k [k] k’ [q] ’ [ʔ]
kw[kw]
Fricative Voiced v [v] g [ɣ] g’ [ʁ]
Voiceless s [s] g [x] k’ [χ] H [h]
Flap Voiced r [ɾ]
Sonorant Glide w [w] y [j]
Nasals m [m] n [n]
The phonetic correlates of Southern Ute stress 
1.2.3 Southern Ute Vowels

The Southern Ute vowels are shown in Chart 2.
Chart 2. Southern Ute Vowels (Givón 1980: 4–5)

Front Central Back
High Rounded u [u]

i [i]
Unrounded ü [ɯ]
High-Mid Rounded ö [ø] o [o]
Unrounded e [e]
Mid-Low Rounded
a [æ]
Unrounded
Low Rounded
a [ɑ]
Unrounded
There is an equal distribution between front and back vowels, four in each. Notice the
absence of central vowels. Four of the vowels are unrounded i, ü, e, a and three rounded
u, ö, o. Ute vowels can either be long or short. If a vowel is longer in duration, it will be
doubled in the Southern Ute orthography. If a vowel is devoiced, it will be underlined.
Southern Ute is phonetically, phonologically, morphologically and syntactically
complex (GivÓn 1979, 1989, 1985, 2001a, 2001b; Charney 1996 and Oberly 2008). For
example, the voicing and devoicing of vowels is an important phonological property
of Southern Ute (GivÓn 1979, 1989, 1985, 2001a, 2001b; Charney 1996 and Oberly
2008). This phonological alternation is difficult for learners to master as the devoiced
vowels are not audible. The learner must watch the speaker’s mouth to determine the
placement of the voiceless vowel. Additionally it is difficult to determine which vowels
will be devoiced in which environments.
k w a a t ü
°
Figure 3. The high back unrounded devoiced “ü”: kwaatü, ‘car’
 Stacey Oberly
With phonetic instrumentation, devoiced vowels appear as a gap or aspiration in

spectrograms between the two dark lines as shown in Figure 3. Compare it with the
voiced version in Figure 4.
k w a a t ü n a g’
Figure 4. The high back unrounded voiced “ü”: ‘kwaatünag’, ‘in the car’
(example from R ivera & McKinley 1995)
In the first spectrogram of the devoiced “ü’ of kwaatü ‘car,’ (marked by the vertical
lines) there are no formants (which would appear as dark bands in the spectrogram),
or voicing (which appears as a dark band at the bottom of the spectrogram). There
seems to some very light aspiration after the [t]. In the second spectrogram the ü is
voiced as the formants and voicing are visible.
2. Phonetic correlates of stress
A stressed segment or syllable is pronounced with extra muscular energy than adjacent
unstressed segments or syllables which may result in greater length, increased inten-
sity and/or increased pitch (Ladefoged 2006). For example, the phonetic correlates for
English stress are, in order of importance, higher pitch, increased duration, and, least
important, increased intensity (Adams 1979; Lehiste 1970). This investigation seeks to
determine which of these three phonetic correlates (length, intensity or pitch) are used
to express stress in Southern Ute.
2.1 Descriptive research on Ute stress

According to GivÓn (1980), “the main stress in Ute words can come only on the first or
second vowel of the word, but never further back (5).” During compounding, “…the
The phonetic correlates of Southern Ute stress 
main stress of the first stem/word in the compound becomes the main stress for the
entire compound word…[while] the second or third elements… have their main stress
reduced (GivÓn 1980: 7).” GivÓn does not provide further discussion on the phonetic
correlates of stress. Similarly, Charney (1996) states,
Ute words normally have their main stress on the second syllable. For that reason,
stress is only marked when it falls on an unexpected syllable, which is generally
the first vowel of the word. Stress is also marked when it falls on one part of a long
vowel (x–xi).
Stress can mark the difference between minimal pairs as (1) illustrated below.
(1) Minimal Pairs Marked by Stress (GivÓn 1980: 6)

Stress on 1st Syllable Stress on 2nd Syllable
a. súwa- ‘straight out’ suwá- ‘almost’
b. pü ḱ a- ‘persistently’ püká- ‘vigorously’
c. págü ‘trout’ pagü ́ ‘fish’
It is interesting to note that these minimal pairs have closely related meanings.
In this research, only one set of minimal pairs (c), págü versus pagü, was
recorded as it was the only minimal pair consisting of two independent words,
instead of two prefixes as the other two minimal pairs. Only female speaker one
had the distinction between págü, ‘trout’ and pagü ,́ ‘fish’. Female speaker two and
male speaker two said that pagü ́ only means ‘fish.’ Male speaker one said there was
not a general word for ‘fish,’ instead one has to use a specific name for the particular
fish. This speaks to the realities of linguistic fieldwork in an endangered language
community where the few speakers are isolated and rarely speak the language. It
also speaks to variation among speakers and voice which may reflect the differing
dialects used in the seven family bands. The spectrograms of female speaker one’s
págü, ‘trout’ and pagü ,́ ‘fish’ are shown in Figure 4. The pitch is marked with the
dotted line and intensity with the white line. Waveforms are included to show the
difference in vowel duration.
Figure 5 shows that there is a difference in peak pitch and duration between
the stressed and unstressed vowels. While intensity is shown in the spectrograms in
Figure 5, it will not be discussed further. In a pilot study of the phonetic correlates
of Southern Ute stress (Oberly 2007), it was found that intensity is not correlated
with stress. The study showed that the difference in intensity between stressed and
unstressed vowels varies from 0.35 to 4 dB. This difference is not significant, and
suggests that intensity is not a phonetic correlate of Ute stress. It will not be dis-
cussed further. This chapter focuses on duration and peak pitch using speaker intu-
ition. The duration and peak pitch information for this minimal pair is summarized
in Chart 3 below.
 Stacey Oberly
p a g üʹ
p á g ü
Figure 5. Spectrograms of minimal pair pagü,́ ‘fish’ and págü, ‘trout’
Chart 3. Stress minimal pair pagü,́ ‘fish’ Versus págü, ‘trout’
a ü Difference
Peak Pitch Duration Peak Pitch Duration Peak Pitch Duration

(Hz) (ms) (Hz) (ms)
pagü ,́ ‘fish’ 172 95 241 260 ü ́ 69 Hz higher ü ́ 70 ms longer

than a than a
págü, ‘trout’ 221 377 172 285 á 49 Hz higher á 92 ms longer
than ü than ü
In prose, for pagü ,́ ‘fish’ the stressed vowel ü  has a peak pitch which is 69 Hz
higher and a duration that is 70 ms longer than the unstressed a. For págü, ‘trout’
the stressed á has a peak pitch which is 49 Hz higher and a 92 ms gain in dura-
tion compared to the unstressed ü. For this minimal pair, the difference in length
between the stressed and unstressed vowel is an average of 131 ms which is even
The phonetic correlates of Southern Ute stress 
greater than the difference between short and long vowels. These results show that
peak pitch and duration are the phonetic correlates of stress for these short vowels.
In the GivÓn (1979) and Charney (1996) dictionaries, there is disagreement
about stress placement making this phonetic documentation vital. Stress placement in
Southern Ute resources provides essential prosodic information. Example (2) illustrates
differing stress placement by Givon and Charney for the same words.
(2) Differing Stress Placement
GivÓn1 (1979)
Charney (1996)
a. ʔáa-ci aachi ‘bow’
b. cíi-ci chiichi ‘flea, bed bug’
c. ʔáapa-ci aapachi ‘boy’
d. kác-may-kH kachmaikH ‘to refuse, reject’
The first two examples in (2) consist of only one syllable. GivÓn places stress on
the first vowel of the words but Charney does not mark stress. In other words, the
learners would not know which vowels is stressed if using the Charney dictionary.
It is important to know which of the two vowels is stressed. This may be the result of
an orthographic oversight by Charney but it is important to know which of the two
vowels is stressed. In the last two examples in (2), GivÓn marks stress on the first vowel
and Charney places stress on the second syllable of the word. The differing placement
of stress causes difficulties for the Southern Ute community. There may be several
causes for differing stress placement. It may be due to varying dialects of the speakers,
individual speaker variation or transcription error of the linguist. From a phonetic and
language learner point of view, accurate stress placement is necessary.
During a pilot study (Oberly 2007) of the phonetic correlates of Southern Ute
stress, three important discovers were made. First, as noted earlier, intensity is not
correlated with stress. Second, the placement of stress may have been transcribed
incorrectly in both previous Southern Ute dictionaries. Third, the placement of stress
may vary from speaker to speaker. The last two discoveries require that speaker
intuition of stress placement is necessary.
3. Methods
This study used two methods to gain speaker intuition of stress placement. The first
method was tapping, in which the speaker tapped on the stressed syllable. The second
was repetition, in which I repeated the word with the stress on a different syllable
. Givon used 〈c〉 for IPA [tʃ] while Charney used “ch.” GivÓn marks morpheme boundaries
with a dash -.
 Stacey Oberly
each time. Repetition was the primary method used. Once stress placement was iden-
tified two measurements were made. First, the duration of the vowels were measured
by the onset and offset of the first formant (F1). Second, the peak pitch of each vowel
was measured.
The present analysis is based on a list of 100 words designed to determine the
phonetic correlates of stress in Southern Ute. A total of 347 vowels in 299 syllables
were measured. The 347 vowels occurred in many different environments including:
word-initial, non-final, before a word-final syllable with devoiced vowel and word
final. The word list was recorded by five speakers: three female and two male on the
Southern Ute reservation in Ignacio, Colorado. The speakers were recorded indoors.
Stress placement data was elicited from one male and one female speaker.2 During
elicitation, I read the English definition and gave the speakers the context and they
translated them into Southern Ute.
The speakers were asked to translate words in two different environments. First
the words were spoken in isolation. Due to the agglutinative nature of the language,
it was difficult to elicit some words in isolation without additional morphemes. The
speakers would add morphemes onto the words both in the isolation and the carrier
sentence environment. For example, female speaker two was asked for iipa, ‘this way’
she produced iipaküü, ‘carry it this way.’ Male speaker one was asked for chag’ai, ‘to
sew’ and he produced chag’aivaachi, ‘to want to sew.’ Again this speaks to the reality of
conducting linguistics research in an endangered language community.
For the second environment, the words were recorded in the carrier sentence,
Maas taani ___ maachu. ‘He always says _____.’ which was used to control for
intonation contours. This carrier sentence varied for each speaker and even within the
same speaker depending on the grammatical category or the word. Due the relatively
free word-order in Ute, the target word may have appeared phrase-finally which may
have affected the vowel length. Each speaker recorded each word and sentence once.
The recordings were made using a Marantz PMD671 portable solid state recorder
and a Countryman Isomax E6 omni-directional head mounted microphone with a
frequency response of 30 Hz to 20 kHz and sensitivity of 6.0 mV. The digital audio
was exported as uncompressed.wav files into the PRAAT software which was used to
conduct the phonetic analysis.
Two methods were used to obtain speaker intuition of stress placement; t apping
and repetition. Stress placement intuition data was elicited from two speakers;
female speaker one and male speaker two. The tapping method has been used for
field research and experimental phonology for languages such as Banawá by Everett,
. Speaker intuition of stress placement was elicited from only two speakers due to the
health concerns of the three other speakers.
The phonetic correlates of Southern Ute stress 
Ladefoged and Everett (1996) and Tohono O’odham by Fitzgerald (1997) and
Miyashita (2002). During the tapping method, the speakers learned to tap using
English words then transferred the skills to Southern Ute. There are three steps
involved in the tapping method. They are shown in (3).
(3) Steps in Tapping Method

a. Tap on every syllable while saying the word.
b. Tap only on one syllable while saying the word.
c. Repeat with tap on different syllables for comparison.
These steps are show visually for pog’oi, ‘to be bumpy/lumpy’ in Chart 4.
Chart 4. Tapping for pog’oi, ‘to be bumpy/lumpy’

a. Tap on all syllables while saying ‘to be bumpy/lumpy’. po g’oi
TAP TAP
b. If you can only tap once, which syllable would you tap on? po g’oi
TAP
c. Which sounds better po g’oi
TAP
or po g’oi
TAP
In all but two words out of 100 both speakers tapped in the same place. The tap-
ping method was not as natural as the repetition method.
The repetition method was the second method used for stress placement. The
repetition method has proven to be extremely useful for gaining speaker intuition of
stress placement. It has been used by Lindblom and Rapp (1973), Everett and Everett
(1984), Fitzgerald (1992) and Miyashita (2002). In this method, the word was repeated
back to the speaker with stress placed on differing syllables each time. This method was
more natural for the two speakers who often joined in the repetition and used hand
motions to accentuate the pronunciation. The steps used in the repetition method are
shown in Chart 5.
Chart 5. Repetition method steps
Elicitor Speaker
a. How does ö́apog’ochi sound? ö́apog’ochi That’s bad.

b. Let’s try öapóg’ochi. öapóg’ochi No.
c. How about öapog’óchi? öapog’óchi. No.
d. Is it öápog’ochi? Yes, it’s öápog’ochi.
 Stacey Oberly
These two methods were used in determining the stress placement in Southern
Ute. Of the two methods, the repetition method was the most natural and interactive.
These are proven field methods for assessing speaker intuition of stress placement.
3.1 Phonetic analysis

Kent and Read (2002) define prosody “as the suprasegemental features of speech that
are conveyed by the parameters of [1] fundamental frequency (perceived p rimarily
as vocal pitch), [2] intensity (perceived primarily as loudness), and [3] duration
(perceived primarily as length) (229).” This investigation attempts to determine which
of the three phonetic correlates (pitch, intensity and duration) are utilized to express
stress in Southern Ute. Recall from the two spectrograms for the minimal pair, págü,
‘trout’ and pagü ́, ‘fish’, it appears that pitch and duration may be the phonetic correlates
of Southern Ute stress.
3.1.1 Duration
As discussed above, the duration of the vowels was measured by the onset and offset
of the first formant. The data consists of vowels in many different environments
including: word-initial, non-final, before a word-final syllable with devoiced vowel
and word final.3 A sampling of the elicitation list is given in (4) below.
(4) Sampling of Elicitation Items

Word-Initial Word-Medial Word-Final
[i] ichei ‘this, inan’ asti’i ‘to like, want’ peiki ‘Come here!’
[u] u’nikH ‘to do’ pukun ‘my horse’ doesn’t occur _#
[ɯ] üaka ‘to plant past’ pükwi ‘to blow’ pagü ‘fish’
[o] doesn’t occur #_ tog’oevi ‘rattlesnake pog’o ‘bumpy/lumpy’
[ø] öapog’ochi ‘orange’ töna ‘to hit’ pö’ö ‘to write’
[ɛ] doesn’t occur #_ peiki ‘Come here!’ ive ‘here’
[ɑ] aapachi ‘boy’ nag’ami ‘to be sick’ paa ‘water’
The vowels are divided into short and long vowels. The duration of short vowels ranged
from 43–214 ms while long vowels range from 66–420 ms. The average duration of ten
stressed and unstressed tokens of each vowel for the three female speakers is shown
in Figure 6.
. The data was not controlled for word-final lengthening. In word-final lengthening, “…a
word or syllable that precedes the end of a major syntactic unit is lengthened (230).” This
lengthening occurs as a boundary cue.
The phonetic correlates of Southern Ute stress 
250
200
150
Stressed
ms
Unstressed
100
50
0
Female 1 Female 2 Female 3
Figure 6. Short vowel duration: Female speakers
For female speakers two and three, the short stressed vowels are longer than
short unstressed vowels. For female speaker one the short stressed vowels are shorter
than unstressed. The data from the two of the three female speakers is evidence that
duration may be a phonetic correlate of stress.
Next consider the duration of the short vowels for the two male speakers shown
in Figure 7.
120
100
80
Stressed
ms
60
Unstressed
40
20
0
Male 1 Male 2
Figure 7. Short vowel duration: Male speakers
The short stressed vowels for male speaker two patterns the same as the female
speakers’ short vowels. His short stressed vowels are an average of 11 ms longer than
 Stacey Oberly
the unstressed. But male speaker one does not pattern like the three other speakers.
His stressed short vowels are an average of 5 ms shorter than the unstressed vowels.
Male speaker one is the oldest speaker of the four. Since three out of four speakers’
short stressed vowels are longer than the unstressed vowels, duration may be the major
phonetic correlate for short vowels.
Next consider the duration of the long vowels for the three female speakers shown
in Figure 8.
300
250
200
Stressed
150
ms
Unstressed
100
50
0
Figure 8. Long vowel duration: Female speakers
For female speakers one and two, the long unstressed vowels are longer than the
long stressed vowels. Specifically, the long unstressed vowels for female speaker one
are an average of 53 ms longer. For female speaker two, the long unstressed vowels
are an average of 48 ms longer. The stressed long vowels are shorter than unstressed
long vowels which suggest that long vowels are shorter when stressed. This is opposite
to the stressed short vowels which were longer than unstressed short vowels. Female
speaker three’s long stressed vowels are longer than the unstressed vowels.
Next consider the duration of the two male speakers shown in Figure 9.
For male speaker two, the duration of long vowels pattern similar to the
female speakers with the long stressed vowels are an average of 9 shorter than long
unstressed vowels. But for male speaker one, the long stressed vowels are an average
of 2 ms longer that the long unstressed vowels. For long vowels, the stressed vowel
was slightly longer than the unstressed vowel for only one male speaker. For three
speakers this data is evidence that duration of long vowels may be in an inverse
The phonetic correlates of Southern Ute stress 
correlation with stress. In other words, to mark stress on long vowel the vowel is
shorter than if it was unstressed.
180
160
140
120
100 Stressed
ms
80 Unstressed
60
40
20
0
Male 1 Male 2
Figure 9. Long vowel duration: Male speakers
To summarize the data presented so far, the short stressed vowels are longer than
the short unstressed vowels for three out of four speakers. But the long stressed vowels
are shorter than the unstressed long vowels for three out of four speakers. This suggests
that duration may be an important phonetic cue for Ute stress for short vowels but not
for long vowels. A repeated measures analysis of variance (ANOVA) was conducted in
order to determine whether the effects of stress and length had a significant effect on
the duration of individual subject’s vowel productions. There was not a significant effect
of stress on the duration of subjects’ vowel productions (F1(1,3) = 2.81, p = 0.192). This
means that whether a vowel was stressed or not failed to affect the duration of subjects’
vowel productions. Additionally, there was not a significant interaction of length and
stress on duration (1,3) = 3.311, p = 0.166). These are preliminary results and further
research with more subjects is necessary.
3.1.2 Pitch
As discussed above, pitch is often a phonetic correlate of stress (Kent & Read 2002).
This section presents the peak pitch averages for each of the five speakers. The vowels
are divided into short and long vowels. The pitch data for the three female speakers is
presented in Figure 10.
For short vowels, the peak pitch is higher for stressed vowels for the female
speakers. Specifically, for female speaker one, stressed vowels are an average of 22 Hz
higher than unstressed short vowels. For female speaker two, her stressed short vowels
are an average of 23 Hz higher than the unstressed short vowels. This suggests that for
female speakers, pitch is an important phonetic cue of stress.
 Stacey Oberly
300
250
200
Stressed
150
Hz
Unstressed
100
50
0
Stressed Female 1 Female 2
Figure 10. Short vowel peak pitch: Female speakers
Next consider the peak pitch of short vowels of the male speakers shown in
Figure 11.
200
180
160
140
120
Stressed
100
Hz
Unstressed
80
60
40
20
0
Male 1 Male 2
Figure 11. Short vowel peak pitch: Male speakers
For both male speakers, the peak pitch is higher for short stressed vowels than
short unstressed vowels. The peak pitch of male speaker one’s short stressed vowels are
on average 27 Hz higher than unstressed. This pattern continues with male speaker
two whose peak pitch of short stressed vowels are an average of 29 Hz higher than
short unstressed vowels. This data is evidence that peak pitch is an important phonetic
correlate for short vowels for all speakers.
The phonetic correlates of Southern Ute stress 
Next consider the peak pitch of long stressed vowels of the two female speakers is
shown in Figure 12.
250
200
150
Stressed
Hz
Unstressed
100
50
0
Figure 12. Long vowel peak pitch: Female speakers
For female speaker one, her long unstressed vowels had a peak pitch of an average
of 21 Hz higher than stressed vowels. This female speaker’s peak pitch data patterns
with the duration data of long vowels discussed above. For female speaker two, the
stressed long vowels are an average of 16 Hz higher than unstressed long vowels.
Figure 13, displays the peak pitch data for the two male speakers.
200
180
160
140
120
Stressed
100
Hz
Unstressed
80
60
40
20
0
Male 1 Male 2
Figure 13. Long vowel peak pitch: Male speakers

 Stacey Oberly
For male speaker one, the long stressed vowels have a peak pitch an average of
18 Hz higher than long unstressed vowels. The long stressed vowels have a peak pitch
an average of 24 Hz higher than long unstressed vowels for male speaker two. For
four out of five speakers, the stressed vowel has a higher peak pitch than the long
unstressed vowel.
To summarize the peak pitch data, for all but one speaker the peak pitch of long
stressed vowels are higher than long unstressed vowels. For female speaker one, long
unstressed vowels have a higher peak pitch than long stressed vowels. This peak pitch
data suggests that higher peak pitch signals stress for both short and long vowels for
“most” Southern Ute speakers. A two factor within-subjects ANOVA was run on the
data, comparing pitch with vowel stress (stressed or unstressed) and vowel length (short
or long). There is a significant correlation between pitch and stress (F(1,4) = 12.05,
p < 0.03), no correlation between pitch and length (F < 1), and no interaction between
the effects of stress and length (F(1,4) = 3.09, p > 0.15). This means that the pitch of
subjects’ productions of vowels was affected by whether a vowel is stressed or not as
shown in Figure 14.
200.00
150.00
Mean pitch
Length
Short
100.00 Long
50.00
.00
Stressed Unstressed
Stress
Figure 14. Mean pitch for stressed long and short vowels
The phonetic correlates of Southern Ute stress 
4. Conclusion
This is the first systematic instrumental and experimental phonetic analysis of the
phonetic correlates of Southern Ute stress. The duration data presented here suggests
that duration is not a phonetic correlate of Southern Ute stress. The duration data for
all speakers as shown in Figure 15.
300
250
200
Short stressed
Short unstressed
150
ms
Long stressed
Long unstressed
100
50
0
Male 1 Male 2 Female 1 Female 2 Female 3
Figure 15. Duration data summary
For Southern Ute, duration is a phonetic correlate for stress for short vowel. Since
long stressed vowels are shorter than long unstressed vowels, shorter duration may
signal stress in long vowels.
The pitch data suggests that peak pitch is an important phonetic cue for most
speakers for both short and long vowels. The peak pitch data is shown in Figure 16.
For short vowels, peak pitch of stress vowels is higher than short unstressed vow-
els. With three of the four speakers’ peak pitch higher for stressed long vowels which
supports the claim that peak pitch is a phonetic cue of stress for the Southern Ute
speakers.
To conclude, peak pitch is the phonetic correlate for stress in for these five
Southern Ute speakers. The intensity data suggests that intensity is not a phonetic cor-
relate for Ute stress. The prosodic information is vital for the Southern Ute community
as it struggles to revive and document the language while the few speakers are health
enough to work with the language.
 Stacey Oberly
300
250
200
Short stressed
Short unstressed
150
Hz
Long stressed
Long unstressed
100
50
0
Male 1 Male 2 Female 1 Female 2 Female 3
Figure 16. Peak pitch data summary
This research is significant because it is an example of linguistic research designed

and conducted by an endangered language community member to ensure that the
children “Sound Ute.” It answers the call put forth by Dr. Jane Hill to empower
Indigenous community members to design and conduct linguistic research for the
good of the community versus an outside Academic determining the research agenda
for the community.
Further research is necessary regarding the interaction between stress and
voiceless vowels. Also the interaction between stress and intonation is in need of
further phonetic investigation.
References
Adams, C. 1979. English Speech Rhythm and the Foreign Learner. The Hague: Mouton.
Charney, Jean. 1996. Ute Dictionary. Ignacio CO: Southern Ute Tribe.
Doherty, Brian. 1993. The Acoustic-phonetic Correlates of Cayuga Word-stress. Ph.D. disserta-
tion, Harvard University.
Everett, Keren M. 1998. The acoustic correlates of stress in Pirahã. Journal of A mazonian
Linguistics 1–2: 104–162.
Everett, Daniel & Everett, Karen. 1984. Syllable onsets and stress placement in Pirahã. Proceed-
ings for the West Coast Conference in Formal Linguistics 3: 101–116.
Everett, Daniel L., Ladefoged, Peter & Everett, Karen. 1996. Native speaker intuitions and the
phonetics of stress placement. Paper presented at 1996 Annual Meeting of the LSA.
Fishman, Joshua. 1991. Reversing Language Shift. Theoretical and Empirical Foundations of
Assistance to Threatened Languages. Clevedon: Multilingual Matters.
The phonetic correlates of Southern Ute stress 
Fitzgerald 1992/ Fitzgerald, Colleen M. 1997. O’odham Rhythms. Ph.D. dissertation, University
of Arizona.
Givón, T. 1979. Ute Dictionary. Ignacio CO: Southern Ute Tribe.
Givón, T. 1980. Ute Reference Grammar. Ignacio CO: Southern Ute Tribe.
Givón, T. 1985. Ute Traditional Narratives. Ignacio CO: Southern Ute Tribe.
Givón 1989/Givón, Thomas. 1980. Ute Reference Grammar. Ignacio, CO: Ute Press, Southern
Ute Tribe.
Givón, Thomas. 2001a. Syntax, volume I. Philadelphia, PA: John Benjamins P ublishing
Company.
Givón, Thomas. 2001b. Syntax, volume II. Philadelphia, PA: John Benjamins P ublishing
Company.
Givón, T. 2011. Ute Reference Grammar [Culture and Language Use 3]. Amsterdam: John
Benjamins.
Gordon, Mathew. 2004. A phonological and phonetic study of word-level stress in Chickasaw.
IJAL 70: 1–32.
Gordon, Mathew, Potter, Brian, Dawson, John, De Reuse, William & Ladefoged, Peter. 2001.
Phonetic structures of western apache. International Journal of American Linguistics 67(4):
415–448.
Goss, James A. 1961. A Short Dictionary of Southern Ute. Ignacio CO: Southern Ute Tribe.
Goss, James A. 1972. Ute Lexical and Phonological Patterns. Ph.D. dissertation, University of
Chicago.
Grenoble, L. A. & Whaley, L. J. 2006. Saving Languages: An Introduction to Language
Revitalization. New York: Cambridge University Press.
Hargus, Sharon. 2001. Quality sensitive stress reconsidered. University of Washington Working
Papers in Linguistics 20: 25–56.
Hill, Jane H. & Hill, Kenneth C. 1980. Mixed grammar, purist grammar and language attitudes
in Modern Nahuatl. Language in Society 9: 321–348.
Hill, Jane H. 1985. The grammar of consciousness and the consciousness of g rammar. American
Ethnologist 12: 725–737.
Hill, Jane H. 1989. The role of theory in language description (Conference Report). Current
Anthropology 30: 119–123.
Hill, Jane H. 1995. The voices of Don Gabriel. In The Dialogic Emergence of C ulture, Bruce
Mannheim & Dennis Tedlock (eds), 96–147. Urbana IL: University of Illinois Press.
Hill, Jane H. 1998. Today there is no respect: Nostalgia, respect, and oppositional discourse in
Mexicano (Nahuatl) language ideology. In Language Ideologies, Practice and Theory, Bambi
B. Schieffelin, Kathryn A. Woolard & Paul V. Kroskrity (eds), 68–86. Oxford: OUP.
Jefferson, James, Delaney, Robert W. & Thompson, Gregory C. 1972. The Southern Utes: A Tribal
History, ed. by Floyd A. O’Neal. Ignacio CO: Southern Ute Tribe.
Kent, Ray D. & Read, Charles. 2002. The Acoustic Analysis of Speech. Albany NY: Thomson
Learning.
Ladefoged, Peter. 2006. A Course in Phonetics. Boston MA: Thomson Higher Education.
Ladefoged, Peter & Maddieson, Ian. 1996. The Sounds of the World’s Languages. Oxford:
Blackwell.
Lehiste, Ilse. 1970. Suprasegmentals. Cambridge MA: The MIT Press.
Lewis, M. Paul (ed.). 2009. Ethnologue: Languages of the World, 16th edn. Dallas TX: SIL
International. 〈http://www.ethnologue.com/〉
 Stacey Oberly
Lindblom, Björn E.F. & Rapp, Karin. 1973. Some Temporal Regularities of Spoken Swedish
[Publication No. 21]. Stockholm: Institute of Linguistics, University of Stockholm.
Martin, Jack & Johnson, Keith. 2002. An acoustic study of ‘tonal accent’ in Creek. IJAL 68: 28–50.
McDonough, Joyce M. 2003. The Navajo Sound System: Studies in Natural Language and
Linguistic Theory. Norwell MA: Kluwer.
Michelson, Karin. 1983. A Comparative Study of Accent in the Five Nations Iroquoian
Languages. Ph.D. dissertation, Harvard University.
Miyashita, Mizuki. 2002. Tohono O’odham Syllable Weight: Descriptive, Theoretical and
Applied Aspects. Ph.D. dissertation, University of Arizona.
Oberly, Stacey. 2007. A preliminary phonetic analysis of Southern Ute stress. Ms, University of
Arizona, Tucson.
Oberly, Stacey. 2008. A Phonetic Analysis of Southern Ute with a Discussion of Southern Ute
Language Policies and Revitalization. Ph.D. dissertation, University of Arizona.
Rivera (Oberly), Stacey & McKinley, Georgia. 1995. Conversational and Everyday Ute. Ignacio
CO: Southern Ute Tribe.
Seiler, Hansjakob. 1957. Die phonetischen Grundlagen der Vokalphoneme des Cahuilla.
Zeitschrift für Phonetik und allgemeine Sprachwissenschaft 10: 204–223.
Southern Ute Tribal Website. 2011. 6 September 2011. 〈http://southern-ute.nsn.us〉
Tuttle, Siri. 1998. Metrical and Tonal Structures in Tanana Athabaskan. Ph.D. dissertation,
University of Washington.
Revisiting Tohono O’odham high vowels*
Colleen M. Fitzgerald
The University of Texas at Arlington
Tohono O’odham is a language of much phonological interest from a typological

perspective. For example, the language has pre-aspiration, multiple patterns of
reduplication, morphological truncation, and displays prosodic inconsistency
with a quantity-insensitive stress system that nonetheless employs multiple
strategies in the prosodic morphology to enhance that prominent syllable. This
paper utilizes a variety of sources of evidence, spanning multiple generations
of speakers, including Hill and Zepeda’s massive work documenting Tohono
O’odham dialects in the 1980s, which remains unpublished for the most part.
I focus on the phonological features and distribution of vowels in Tohono
O’odham – and argue that the features [high] and [front] interact in this language
in previously undocumented ways.
Keywords: Tohono O’odham; phonology; vowel inventory; CV interactions
1. Introduction
Tohono O’odham, a Uto-Aztecan language spoken in Arizona, is a language of much

phonological interest from a typological perspective. For example, the language has pre-
aspiration (Fitzgerald 1996), multiple patterns of reduplication (Hill & Zepeda 1992,
1998; Fitzgerald 2000, 2002, 2012; Miyashita 2004), and morphological truncation
(Fitzgerald 1997, 2012; Fitzgerald & Fountain 1995), among many other interesting
features and patterns. No bibliography of the phonological work that has been done on
the Tohono O’odham language is complete without several of the collaborative papers
* A few notes on phonetic symbols will be helpful: /č J�/ represent affricates; /ñ/ the palatal
nasal; /d�, s�/ for the two retroflexes; / / represents a palatal lateral flap; and /ɨ/ a high central
ɼ
unrounded vowel. Voiceless or “extra-short” and long vowels are typically marked (the former
with the / ̥/ diacritic underneath and the latter with a colon). It should be noted there is
considerable variation in the voiceless/extra-short vowels among different dialects. I have left
stress unmarked; for more detail on this aspect of O’odham phonology, Fitzgerald 1997, 1999,
2002 on primary and secondary stress patterns. Any errors are my responsibility.
 Colleen M. Fitzgerald
published by Jane Hill and Ofelia Zepeda, which analyze data they collected in their
dialect survey in the 1980s. A wonderful discussion of this project comes in Zepeda
and Hill (1998), where they outline their collaboration and how they collected data
from most of the villages on the Tohono O’odham reservation.
Here Jane Hill outlines why it was important to study Tohono O’odham dialects
in their collaboration:
There were several important reasons to do this. First, Tohono O’odham

people are interested in dialects, and the existing literature on the regional
variation in the languages was both contradictory and perfunctory. Second,
dialect differences in the language complicated bilingual education programs
in reservation schools because parents did not like their children to be taught
by instructors who spoke a dialect different from their own. Since bilingual
education is one of the main avenues for language maintenance available to
O’odham people (before the 1991 Native American Languages Act) it was the
only way to get funding to teach the language), it seemed important to develop
a sound understanding of dialect differences that could provide a basis for
training parents and teachers. Finally, a dialect survey allowed us to obtain a
sample of the usage of elderly people, speakers who could provide not only
the basis for a mapping of the most conservative regional variation, but also
samples of what was generally accepted as “good” usage. Such samples would be
useful for language curricula. There were also “sneaky” reasons. For example,
my Mexican fieldwork had given me very good experience with the necessary
methodology: I knew how to study the language variation that regional dialects
embody. Also, research on dialects would not step all over Ofelia’s own research
(which was then on morphology, especially derivation). In fact, it could even
help it by providing a large sample of usage. Further, since Ofelia used O’odham
language in her poetry, I thought she might like the idea of being able to go out
and listen to turns of phrases from the best speakers. Finally, I knew I didn’t
have a prayer of ever getting onto the reservation as a researcher unless I had
her help! I needed protective coloration, both as a collaborator with a member
of the O’odham community, and, not least, as a “linguist” instead of as an
“anthropologist.” (Zepeda & Hill 1998: 139)
As a graduate student working on Tohono O’odham at the University of Arizona

with both Ofelia Zepeda and Jane Hill serving on my dissertation committee, I was
fortunate to have both of them serve as examples and mentors for how to work in an
ethical way with Native American language communities. There is a way to conduct
linguistic research that is of value to communities and to linguists, and the dialect sur-
vey conducted by Hill and Zepeda at its heart represents a research project designed
with larger social justice goals in mind, with its potential to impact e ducational access
and language maintenance resources for an indigenous language community. These
Revisiting Tohono O’odham high vowels 
principles have inspired my own work with communities, particularly the Tohono
O’odham. Finding ways to learn more about languages to better support Native
teachers with materials and teacher training, for example, has been the subtext of
much of my recent work, especially in the Oklahoma Native language context.
In this paper, I focus on the phonological features and distribution of vowels in
Tohono O’odham. The focus lies on the behavior and distribution of the vowels. The
vowel inventory is particularly interesting, with three high vowels, including a high
centralized unrounded vowel.
Inspired by the depth of Jane Hill’s work in terms of its contribution to historical
and comparative work in Uto-Aztecan, looking at the past and the present of many
languages in this family, especially Tohono O’odham, this paper uses that perspective
to interrogate the O’odham vowel inventory. I draw on several data sources from
different points in time, including some discourse data that shows patterns previously
unanalyzed in the literature, and patterns that emerge from dialect variation. This
study relies on several sources to do so: Mathiot (1973), Hill and Zepeda (unpub-
lished dialect survey data) and Fitzgerald et al. (2012). The first of these is the Mathiot
(1973) dictionary. Madeleine Mathiot worked with speakers aged 50 and older from
1958–1961, and thus the dictionary represents the language of speakers born in the
late 19th century. The dictionary includes extensive examples of sentences with con-
nected speech pronunciations indicated.
Secondly, I have drawn information from dialect patterns found by Hill and
Zepeda in their analysis of the Tohono O’odham dialect survey, which was conducted
in the 1980s. In that study speakers were at least fifty-five years old, and so these
data represent language one or even two generations removed from that used in the
Mathiot dictionary.
Finally, some of the data collected more recently, from a narrative collected in
2002 (published in Fitzgerald et al. 2012), comes from a speaker born two generations
after those in the O’odham dialect survey and so is even more diachronically removed
from the speakers whose forms were recorded in the Mathiot dictionary.
What I explore in this paper is a range of patterns unified by how they often oppose
/i/ to the rest of the vowel inventory, as well as in that they fail to display a unified
three-member high vowel group. The front high vowel rarely surfaces in a group with
the other two high vowels, /ɨ u/, as target or triggering elements for phonological
processes.
I present three main sets of patterns in making this claim: synchronic data
on the fronting of the retroflexed consonants, the ban on high vowels in vowel-
vowel sequences in the Totogwañ dialect, and patterns of total assimilation by
all vowels except /i/ by function words in connected speech. These patterns
argue against grouping /i ɨ u/ in the same natural class. However, I present some
 Colleen M. Fitzgerald
c ontrasting patterns that present a contrary view favoring treating the high vowels
as a group.
The paper is organized in the following way. Section two presents the vowel
inventory of Tohono O’odham. Section three outlines distributional patterns that show
retroflex consonants are restricted from preceding the high front vowel. Section four
recaps dialectal evidence from Hill et al. (1994) and other sources showing how only
a subset of the high vowels group together. Section five presents the connected speech
patterns, most of which are drawn from Mathiot (1973), showing how the front high
vowel /i/ acts in opposition to all the remaining vowels, which are non-front. Section
six turns to a pattern more commonly discussed in the literature on Tohono O’odham
phonology, the interaction of high vowels and coronal consonants. In this pattern,
unlike the other two, the three high vowels finally group together as a clear natural
class. Overall, the range of data and phonological patterns examined shows variability
in the groupings that can occur among the high vowels, suggesting that these vowels
are not so unified as a natural class as the last set of data – the one most explored in
the existing phonological literature on Tohono O’odham – suggests. The final section
concludes the paper.
2. The Tohono O’odham vowel and consonant inventory
As a preliminary to discussions of vowel patterning and consonant-vowel interac-

tion, this section presents the vowel and consonant inventory of Tohono O’odham.
The Tohono O’odham vowel inventory consists of five vowels, including three high
vowels. The complete inventory appears in (1), note that there is only one front
vowel in the set:
(1) Tohono O’odham vowel inventory

front central back
high i ɨ u
mid o
low a
Five vowel inventories occur in other Uto-Aztecan languages, such as C omanche and

Southern Paiute. The vowel inventory above is also reconstructed by Langacker (1970) as
the Proto-Uto-Aztecan vowel inventory. Wrestling with the O’odham vowel inventory,
both in terms of the distribution and co-occurrence patterns and in terms of how these
patterns influence phonological behavior, has preoccupied a number of researchers,
including Hill et al. (1994), Miyashita (2002, Forthcoming), Fitzgerald (2000), and
Jackson (2003).
Revisiting Tohono O’odham high vowels 
The consonant inventory of Tohono O’odham appears in (2).

(2) Tohono O’odham consonant inventory
bilabial dental alveolar palatal retroflex velar glottal
stop p b t d d� k g �
affricate č
ȷ�
fricative s s� h
nasal m n ñ
lateral ɼ
glide w y
While the precise nature of the contrast in the stop system has been the subject of
considerable discussion, the pairings given above for the stops (i.e. /p/ and /b/) are
uncontroversial. An additional pairing relevant here comes in the retroflexes and their
front counterparts. The retroflex stop /d�/ is partnered with its palatal counterpart / /. ɼ
The retroflex fricative /s�/ has as its front counterpart the alveolar fricative /s�/. Note also
that the palato-alveolar (or post-alveolar) affricates are grouped in the palatal column
on the chart.
Of particular interest in this paper is how the distribution of the front high vowel
forces certain analyses on the system. In an unpublished collaborative manuscript by
Jane Hill Ofelia Zepeda, Molly Dufort and Bernice Belin, (Hill et al. 1994) the authors
use the distribution of palatalized consonants and a pattern of dialect variation in diph-
thongs to lay out the claim that /i/ is differentiated from its central high counterpart /ɨ/
(and the rest of the inventory) by the feature [+ATR]. Miyashita (forthcoming) also notes
additional unique phonetic and phonological properties associated with /i/, including its
frequency as the most-attested voiceless vowel. Her proposal is subdivide the diphthongs
in a way that isolates a set of ‘heavy diphthongs’, in which “neither vowel is /i/.”
In the next several sections, I lay out the argument that the vowel /i/ displays
asymmetric behavior in a number of phonological contexts. Only one distributional
pattern, described in section six, illustrates /i/ grouping with the other high vowels as
a single natural class. Overall, what we will see is a body of evidence where /i/ fails to
group consistently with the other two high vowels in a single natural class.
3. Retroflex consonants and the front vowel
In this section, I will show that there is a constraint against retroflexed consonants pre-
ceding the high front vowel, /i/, but not the other high vowels. The goal of this section
is to sketch out some preliminary splits in the vowel system in terms of which vowels
behave as a natural class. The two retroflex consonants, /d� s�/, in Tohono O’odham have
 Colleen M. Fitzgerald
several phonological and morphological restrictions on their phonological distribution;

we will look at their distribution with vowels, as well as the distribution of their ‘fronted’
counterparts, /s /.1 I will establish the opposition between /i/ and the remaining vowels
ɼ
in terms of the phonological restrictions on the distribution of these retroflexes.
In (3), the leftmost column has a set of verbs ending in retroflex consonants in
their base forms. The remaining columns illustrate how each verb appears (or fails
to appear) in a number of paradigms with gaps. These columns show that the suffixal
morphology results in an absence of retroflexes in the plural imperative and causative
paradigms. Even more intriguingly, the hortative and perfective causative forms result
in an opaque contrast, where the word-final non-retroflex consonants provide surface
contrast with possible retroflexes in the same position.
(3) Relevant paradigms for the retroflex alternation2

Imp. (Base) Perf. Pl. Hort. Caus. Perf. of Gloss of “base”
Imper. Caus.
a. mɨḍ mɨ: mɨ iñ
ɼ mɨ ɼ ‘to run, drive, flow,
blow, crawl’
b. hidoḍ hido hidoḍo hidoḍ hido id
ɼ hido ɼ ‘to cook obj in a
single cooking pot’
c. tonoḍ ton tono id
ɼ tono ɼ ‘to shine’
d. ko:ṣ koi ko:siñ ko:s ko:sid ko:s ‘to fall asleep, to sleep’
e. but: kɨ iw
ɼ kɨ ɼ kɨ iw
ɼ kɨ iwid
ɼ ‘to shuck obj (corn), to
pick small berries off
the bush’
In (3a–d), retroflexes /s� d�/ alternate with non-retroflexes (or ‘fronted’) counterparts
/s / before /i/. Note that the hortative and perfective of causative forms in (b, d)
ɼ
appear with fronted final consonants, fronting being triggered by the presence of a
following /i/ in the plural imperative forms. The plural imperative form of (b) does
not place /i/ after the retroflex, and the hortative retains its word-final retroflex.
The causative forms of (b), by contrast, surface with the fronted / / before /i/. For
ɼ
comparison, (3e) presents a case where the fronted consonant remains constant
throughout the paradigm.
It is challenging to find true imperfective bases that are both retroflex-final and
have these particular paradigmatic tokens (perfective, plural imperative, hortative,
. For more on the phonetic detail about the retroflexes, I refer the interested reader to
Dart (1993).
. Full forms of column headings are as follows: Imperfective (base), Perfective, Plural
Imperative, Hortative, Causative, and Perfective of Causative.
Revisiting Tohono O’odham high vowels 
causative, and perfective of the causative). Mathiot (1973) has the most detailed lexical
entries for the paradigms, but as is clear from (3), not all verbs have fully fleshed out
paradigms. Our focus in presenting the data in (3) is to highlight the alternation of /d�/
and / / and also, that /s�/ and /s/ alternate, and that in both cases, the retroflexes do not
ɼ
precede the high front vowel.
First, it is important to note that the there is a deceptiveness about the surface con-
trast between /s/ and /s�/, as well as / / and /d�/. This is in part due to the complexities of
ɼ
the morphophonological system of Tohono O’odham, which marks perfective aspect
by truncation, uses several processes of reduplication to mark different categories of
aspect and number, and that also includes suffixation. Morphological truncation is
used to mark perfective aspect in a large set of verbs (Zepeda 1988; Saxton 1982), with
that truncation typically deleting the final consonant, or the final consonant and vowel
if final consonant deletion would result in a coronal-high configuration at the right
edge of the perfective (Fitzgerald & Fountain 1995; Fitzgerald 1997). This prohibition
against a final coronal/high sequence removes any high vowel, not just /i/ – and this
phenomenon is discussed in section six below.
To flesh out the details for this pattern a bit more, I present the distributional slots
and conditions where each element of the two pairs of consonants occurs. Table 4
shows the word-initial distribution of the two retroflexes, /s� d�/, and their correspond-
ing non-retroflexes, /s /. Several facts emerge. The retroflex /d�/ never appears word-
ɼ
initially. Its fronted counterpart, / /, occurs initially before any vowel – but it is only
ɼ
found in Spanish loanwords, or a small set of emotive words which begin / ɨ-/.3 The ɼ
/s�/, by contrast, appears word-initially before any vowel except /i/, while its fronted
counterpart, /s/, occurs word-initially only before /i/, or in Spanish loanwords.
(4) Word-initial distribution of /s /, /s� d�/

ɼ
/s/ and /s�/
/si/ siw ‘bitter’ /s�i/ –
/sɨ/ – /s�ɨ/ s�ɨ ig
ɼ ‘prairie dog,
ground
squirrel’
/su/ – /s�u/ s�u:dagi� ‘water’
/so/ – /s�o/ s�opo k ɼ ‘(being)
short’
/sa/ sa:nto ‘a Catholic /s�a/ s�a�i ‘grass, hay’
(>Sp santo)’
. The main exception to this are words that begin / ɨ…/; Footnote 22 in Hill et al. (1994)
ɼ
suggests these words have an “affective” quality by using / / in initial position, rather than /d/.
ɼ
 Colleen M. Fitzgerald
/ / and /d�/
ɼ
/ i/
ɼ ɼi:wa ‘jacket, coat (>Sp levita)’ /d�i/ –
/ ɨ/
ɼ ɼɨ�ɨJ�ɨ ‘used to refer to a child that /d�ɨ/ –
does not behave’
/ u/
ɼ u: si
ɼ ɼ ‘candy (>Sp dulce)’ /d�u/ –
/ o/
ɼ ɼo:go ‘(being) crazy (>Sp loco)’ /d�o/ –
/ a/
ɼ ɼa:nJ�u ‘ranch (>Sp rancho)’ /d�a/ –
In (5) the two pairs of consonants are compared in terms of their word-medial
distribution intervocalic position, if such examples exist. This set, and those in
Example (6) below, by necessity, include truncated and reduplicated words to
instantiate some patterns where a less morphologically-influenced token does not
appear to exist. Spanish loanwords have also been included, as needed, to flesh out
the sets.
(5) Word-medial, intervocalic distribution

/Vsi/ wosibaḍ ‘deceased paternal /Vs�i/ –
grandfather, deceased
paternal great uncle’
/Vsɨ/ – /Cs�ɨ/ naks�ɨ ɼ ‘scorpion’
/Vsu/ – /Cs�u/ s�oñs�ud� ‘to crack obj
(such as a hard
nut or a stone)
with a single
blow’
/Vso/ gi:sobi� ‘a bird, verdin, oriole’ /Vs�o/ s�os�obbid� ‘doll, pl’
/Vsa/ �usaga ‘double ball stick; /Vs�a/ ha:s�añ ‘saguaro (a type
staff ’ of cactus)’
/V i/
ɼ s�ɨ in
ɼ ‘to straighten’ /Vd�i/ –
/V ɨ/
ɼ – /Vd�ɨ/ –
/V u/
ɼ u uya
ɼ ɼ ‘gray horse (<Sp /Vd�u/ kud�ut ‘to disturb, an-
grulla ‘ash-colored – noy bother obj’
of a horse’)’
/V o/
ɼ pi: os
ɼ ‘pear, pear tree /Vd�o/ kod�og ‘to make a
(<Sp pera)’ gargling noise,
to gurgle’
/V a/
ɼ čiñwo a ɼ ‘one with a beard’ /Vd�a/ �ɨd�a ‘inside, in’
Revisiting Tohono O’odham high vowels 
The examples in (5) show that intervocalic retroflexes are not found before /i/. The
retroflex /ṣ/ precedes any other vowel. Non-retroflex /s/ precedes /i/, but is not
found before /ɨ, u/. Retroflex /d�/ is not found before /ɨ/, but neither is its fronted
counterpart, / /. ɼ
In terms of the word-final distribution, shown in (6), both the retroflex and
fronted consonants appear. This is because fronting is triggered by a following /i/.
(6) Word-final distribution
/is/ �u:d�wis ‘grape vine, grapes’ /is�/ mu is� ɼ ‘to break off object by
bending it reiteratedly
(cf. mu iñ ‘to break obj)’
ɼ
/ɨs/ �ɨ�ɨs ‘a plant, a crop /ɨs�/ �ɨs� ‘chin’
(of plants)’
/us/ �u:s ‘yard’ /us�/ wu:s� ‘to come out, perf ’
/os/ ko:s ‘to sleep, hort’ /os�/ ko:s� ‘to fall asleep, to sleep’
/as/ ha�as ‘to be through, /as�/ tas� ‘sun’
stop, doing
something’
/i /ɼ �añi: ɼ ‘bluing (>Sp ‘añil)’ /id�/ wɨhɨȷ�id� ‘for, in place of, with obj’
/ɨ /ɼ hɨwɨ ɼ ‘wind’ /ɨd�/ mɨd� ‘to run, drive, flow,
blow, crawl’
/u /ɼ �u ɼ ‘to extend perf ’ /ud�/ ñu:kud ‘to take care of oneself ’
/o /ɼ ko�oko ɼ ‘chile’ /od�/ hidod� ‘to cook obj in a single
cooking pot’
/a /ɼ ha: ɼ ‘squash, pumpkin’ /ad�/ tad� ‘foot’
Overall, the pattern is that the retroflexes can precede any vowel except /i/, which is
where their non-retroflex counterparts instead occur. This trend comes into sharper
focus if the layers of the lexicon are cleaned up, such that loanwords, m orphological
truncations, reduplicated words, and suffixed words (especially those suffixes that
begin with /i/) are eliminated from consideration. One way to schematize the division
in the vowel inventory between the two conditions is in (7), with the high front vowel
/i/ pitted against the others, so that the non-front vowels act as a natural class:
(7) Evidence for the vowel natural classes from the patterning of retroflexes
Front Non-front
i ɨ u
o
a
 Colleen M. Fitzgerald
In the next section, I explore another phenomenon that suggests a somewhat different
grouping in the vowel inventory, based on data from dialect variation.
4. Dialect variation in vowel clusters
Another set of data that raises some doubt about on the status of high vowels as a
natural class in Tohono O’odham comes from a comparison of permissible vowel
clusters in two different dialects. Hill et al. (1994: 17) note the following:
Mathiot (1973), Saxton (1982), and Saxton, Saxton and Enos (1983[1989]) all
record a set of dialect variants in TO /Tohono O’odham/, whereby the central
dialect tolerates a narrower range of vowel clusters than do peripheral dialects.
The central dialect permits only clusters of /i(L)a/, /i(L)o/, /u(L)a/, and rules out
clusters */e(L)a/, */e(L)o/, and */o(L)a/ (where L is a laryngeal, /�/ or /h/). /Author
note: In the official writing system, /e/ is the orthographic version of the IPA
symbol for the /ɨ/ vowel and /l/ is the orthographic version of the IPA symbol for
the / / consonant./
ɼ
While the behavior of the laryngeal in Tohono O’odham is remarkably complex

and perplexing (and beyond the scope of this paper),4 the description of the vowel
groupings holds considerable interest. Notably, the target of this process fails to group
vowel /ɨ/ with the high front vowel /i/, but instead groups it with the back mid vowel
/o/ in terms of the central (Totogwañ) dialect described above. Thus, in the Totogwañ
vowel sequences noted above, /ɨ, o/ are prohibited, and /i, u/ appear in their r espective
places. In (8), the comparison between the Totogwañ and other dialect patterns is
given.
(8) Totogwañ dialect vowel sequences versus the “standard” or peripheral

patterns
Totogwañ/Central Dialect Standard/Peripheral Dialects
a. čiad�agi� ‘Gila Monster’ čɨad�agi� ‘Gila Monster’
čihan ‘commanding’ čɨhan ‘commanding’
hiosig ‘flower’ hɨosig ‘flower’
ñiad� ‘to sponge from obj’ ñɨad� ‘to sponge from obj’
ñiok ‘speaking’ ñɨok ‘speaking’
but: miabid ‘coming near’ miabid ‘coming near’
. For more discussion on the topic of laryngeals in O’odham, see Hale (1970), Hill and
Zepeda (1992), and Hill et al. (1994), among others.
Revisiting Tohono O’odham high vowels 
b. tuaya ‘towel (>Sp toalla)’ toaya ‘towel (>Sp toalla)’

�uam ‘yellow’ �oam ‘yellow’
tuha ‘white’ toha ‘white’
du�ag ‘mountain’ do�ag ‘mountain’
s�uak ‘to cry; to rattle; to s�oak ‘to cry; to rattle; to
moo; to bleat’ moo; to bleat’
but kuadčid ‘to peek in or out kuadčid ‘to peek in or out
looking for obj’ looking for obj’
Characterizing this pattern in terms of natural classes or groupings is somewhat

challenging. Perhaps the easiest way is in terms of outputs: the preferred outputs have
vowel sequences that constitute the peripheral high vowels of Tohono O’odham, /i u/.
The vowels dispreferred by Totogwañ in this configuration are /ɨ o/, a high central
unrounded vowel and a mid back round vowel respectively. One approach might be to
treat the central vowel in Tohono O’odham as [+back], making the contrast between
/ɨ/ and /u/ dependent upon their different values for /round/. The shift in Totogwañ
might then plausibly be characterized as fronting /ɨ/ and raising /o/. However, this
leads to a position where the characterization of this output lacks phonological unity
in terms of features; one vowel, /ɨ/, changes from [+back] to [–back], while the other
vowel dispreferred in these pairs, /o/, changes from [–high] to [+high].
Visualizing the vowel inventory in terms of the groupings of banned vowels and
permissible outputs offers an interesting counterpoint, where /ɨ o/ are the impermis-
sible first vowels of the sequence, and /i u/ are permissible outputs as the first vowel
of the sequence. In (9), I present a visual representation of these two groupings (plus
the vowel /a/), drawing on the representation of the division of the vowel system in
terms of retroflex-influenced patterning shown in (7). To put the peripheral vowels
in a grouping by themselves, the high central vowel needs some distance from /i u/,
almost as if the vowel represented by /ɨ/ is not phonologically a high vowel.
(9) Evidence for the vowel natural classes from the patterning of Totogwañ
vowels
i u
ɨ
o
a
It is not clear that this is the right representation of the overall vowel system, but it
might be helpful to come up with some phonologically-driven grouping that divides
up the vowel space in a way that groups those banned vowels in the sequences of
 Colleen M. Fitzgerald
/ɨo /, /ɨa/, and /oa/ and then in a different group, categorizes the permissible vowels in
the sequences of /io/, /ia/, and /ua/. It is important to note that Totogwañ dialect does
allow sequences that represent close proximity in terms of movement from the first to
the second vowel of the sequence. Therefore, both central and standard d ialects permit
/ɨi/, with a somewhat rarer distribution of words with /iɨ/ and /ɨu/. The sequence /ao/ is
tolerated in the standard/peripheral dialects, but not in the Totogwañ dialect.
The solution taken by Hill et al. (1994) is to argue that /i u/ stand as [+ATR] vowels,
as opposed to the other three vowels that are [–ATR]; vowel sequences without [+ATR]
are not permitted in Totogwañ. But in addition to presenting the four disrupted vowel
sequences in the Totogwañ dialect (/ɨo/, /ɨa/, /oa/, /ao/), they also suggest that there
are two vowel sequences that are rare, /ɨu/ and /iɨ/. One shortcoming of the [+ATR]
solution is there is no easy way to address that. In fact, only one vowel sequence with
the high central vowel is well-attested in all dialects of Tohono O’odham: /ɨi/.
Consider how clustered /ɨ u o a/ are away from the front vowel space. An approach
that draws on unifying the less well-attested vowel sequences and those that are subject
to different realizations in Totogwañ might make more sense. Maximizing the distance
from the two vowel sequences makes each part of the diphthong more perceptible and
produces a sequence with more movement from the initial target to the final target in
the vowel sequence. Relevant to this is a study on vowel dispersion in Jackson (2003),
using one speaker of Pima, a related dialect. Although the small sample size is prob-
lematic for generalizing, it is interesting to see how the study’s measurement of the first
and second formants of the five vowel system in Pima shows the front vowel distanced
away from the central and back vowels, with considerable crowding in the back vowel
space. The preliminary phonetic data also suggest, then, that /i/ is in opposition to the
other vowels.
(10) “Measurements of the first and second formant frequencies of the canonical
vowels of Pima” Figure 1 in Jackson (2003: 35)
200
300
i u
400
e
F1 (Hz)
500
o
600
700
a
800
2400 2000 1600 1200 800 400
F2 (Hz)
Revisiting Tohono O’odham high vowels 
Reconceptualizing the vowel system somewhat by questioning the phonological

behavior and characterization of /ɨ/ could lead to a better solution of the dialectal data.
A phonologically aggressive solution, like treating /ɨ/ as a phonologically mid vowel
would mean the banned, dialectally-targeted and rare vowel sequences are all those
which occur in close proximity to each other, while the attested vowel sequences all
have more distance and less proximity to each other, and thus involve more m ovement
from one vowel position to the final vowel position. The occurrence of /ɨi/ makes
more sense in this system, although it might occur as a sequence more than is actually
attested. Reexamining the high central vowel and its phonological features may yield
a more satisfactory answer to the entire set of vowel sequences than appealing to
a [±ATR] active distinction in the vowel inventory. At the least, there is additional
evidence that /i/ and /ɨ/ do not act like members belonging to the same natural class.
5. Total vowel assimilation in connected speech
The previous two sections worked to establish the disparate behaviors of /i/ and
/ɨ/, and this section works to bolster that, drawing on data coming from connected
speech. What this section will do is establish facts regarding total assimilation in con-
nected speech, drawing nearly exclusively from tokens in the example sentences from
the Mathiot (1973) dictionary. Most tokens involve the back vowels as undergoers,
although the presence of /i/ can trigger the assimilation in a following element. The
data will show that /i/, along with /u/, fails to undergo this process. Total assimilation
in connected speech provides another example where the high vowels fail to act as a
unified natural class, with /i u/ resisting assimilation, but /ɨ/, along with other vowels,
undergoing total assimilation.
While the patterns in the previous two sections have received additional attention
in the O’odham literature, the process presented in this section has received little, if
any, discussion from researchers, although there is robust attestation of the phenom-
enon in the Mathiot dictionary. Furthermore, drawing from discourse data like this
is consistent with the approach taken by Jane Hill in her research, where more than
just the typical ‘word list’ is used. In the collaborative work on Tohono O’odham with
Ofelia Zepeda, the two used word lists and interviews. In a twist to the typical word
list task, they elicited their data via pictures as stimuli, so as to better collect dialectal
variants in singulars and plurals. Finally, examining discourse data with an eye to what
it tell us about the phonology of O’odham brings in data that represents “aspects of lan-
guage produced by the processes of human social life,” a take on Dell Hymes’ notions
of historical linguistics, as channeled by Jane (Zepeda & Hill 1998: 133).
The largest single set of published documentation produced on the Tohono
O’odham language is the dictionary by Madeleine Mathiot (1973), which consists of
 Colleen M. Fitzgerald
nearly 1,000 pages of entries. The entries are detailed, both in terms of grammatical
information and in terms of the wide range of sentences that Mathiot provides to
illustrate different lexical uses. A number of the example sentences come from texts
collected by Mathiot, such as coyote stories told by Jose Pancho. This means the
dictionary is a treasure trove of naturalistic data coming from context, although the
sentences themselves appear out of context from their original narratives. The nar-
ratives remain unpublished as of this writing. In (11), I give a sample lexical entry
together with its example sentences, using one of the entries for the word mu�i:5
(11) mu�i
Mod Subst – much, many, a lot, a lot of
ex: Wo ma: g mu�i ia / Mu�i wo ma: g ia . Give him a lot of money. • Mu�i
ɼ ɼ ɼ ɼ
�o �om(<�am) ču:č g �u�us. Many trees are (standing) there. • �Am �atp wɨ:č g
mu�i ia . There lies a lot of money. • … hɨghɨkaȷ� mo mu�i �i waidasč abs�aba
ɼ ɼ
pi mu�i wo �ɨ-�ui. … because many are called but few are chosen (taken). •
Mu�i ñ(<�añ) ia ga. I have a lot of money.
ɼ ɼ
This entry represents two examples of assimilation of speech in connected discourse,

with each separated from within the sentence as a parenthetical. The first is the
occurrence of �om, from the locative �am, where the vowel of the function word displays
total assimilation to the vowel that precedes it, the �o of the third person imperfective
auxiliary. The second example is ñ, the clipped version of the first person imperfective
auxiliary, �añ. In both cases, Mathiot gives the dictionary user the attested form (what
speakers produced), but then also provides the standard lexical representation to
connect with the connected speech output. She uses forms like (<�añ) throughout the
dictionary, typically linked with either a shortened function word, or an assimilated
function word. As a result, Mathiot (1973) provides a significant corpus of transcribed
connected speech, allowing for a number of possible reanalyses and investigations.
Because the examples are given as if they occurred as isolated sentences, the contex-
tual details of triggering environments can be analyzed within, but not beyond the
sentence.
Following the line of investigation in the previous two sections, I will now work
through which vowels undergo this total assimilation. The interesting aspect of this
. Examples used in this paper were taken from an electronic version of Mathiot (1973) in
which the entries are in process of being converted into the official orthography of the Tohono
O’odham Nation, developed by Albert Alvarez and Kenneth Hale. Electronic copies of the
Mathiot dictionary have been deposited with the Tohono O’odham Community College and
the Tohono O’odham Nation Cultural Center & Museum Himdag Ki: Hekĭhu, Hemu, Im B
I-Ha’ap for dissemination. To be consistent with this paper, they have been presented using
the APA, as described in the first footnote of the paper. For more on the dictionary project,
see Fitzgerald (2009).
Revisiting Tohono O’odham high vowels 
pattern of total assimilation lies in the set of vowels that do and do not undergo assimi-
lation. Nearly every example of a vowel target of total assimilation is a back vowel.
From nearly 1,000 pages, only ten examples in the Mathiot dictionary involve the front
vowel /i/, and those examples will be argued further below to reflect a process other
than total assimilation. Thus what the next few pages show is reinforcement of /i/ as
apart from the rest of the vowel inventory (and the importance of the front-back dis-
tinction in the inventory).
The assimilation pattern exemplified in the first case in (11), which is repeated
in a slightly different format in (12), suggests a preliminary description of this
phenomenon. The topmost line, Connected speech, is the actual occurring speech
form given as in the dictionary, with the assimilatory examples of connected speech
underlined. The next line, Standard speech, presents the standard (or unshortened
and unassimilated) language version of the example sentence, which puts it closer to
the representation that would be found in other dictionaries and grammars of the
language (Saxton et al. 1989; Zepeda 1988), as well as how the O’odham language
often appears in more literary contexts, such as in the poetry of Ofelia Zepeda
(i.e. Zepeda 1995, 1997). In these contexts, fewer of the shortened forms and assimi-
lations are represented, and the official orthography is used as the writing system.6
Finally, Translation shows Mathiot’s characterization of the O’odham sentence in
English. Underlining in both the connected and standard speech allows an easy com-
parison of the assimilated and unassimilated versions of the same word. Finally, in the
translation line, the lexical entry information is given for where the example occurs in
Mathiot (1973).
(12) �om < �am

Connected speech: Mu�i �o �om ču:č g �u�us.
Standard speech: Mu�i ‘o �am ču:č g �u�us.
Translation: Many trees are (standing) there. (entry: mu�i ).
In this example, assimilation is progressive and total with the targeted word, �am, a
function word that contains the low central unrounded vowel, being realized as �om.
The trigger in this case is the third person auxiliary, �o, suggesting that this harmony is
function word to function word, from left to right.
As (13) below shows, the high front unrounded vowel can also trigger the same
assimilatory pattern. In this case, the first person imperfective auxiliary �añ surfaces as
�iñ when it is preceded by the negative marker, pi.
. It should be noted that there is also a considerable written tradition using the o
rthography
developed and used in the Saxton et al. (1989) dictionary. This written tradition includes
traditional O’odham legends and portions of the Bible.
 Colleen M. Fitzgerald
(13) �iñ < �añ

Connected speech: Pi �iñ �ɨ:bid mant o mu:.
Standard speech: Pi �añ �ɨ:bid mant o mu:.
Translation: I am not afraid of dying. (entry: mu:k-)
In (14), the triggering vowel is the high back rounded vowel, /u/, while (15–16) show
the /o/ and the low central unrounded vowel, /a/, respectively, serving as the t riggers
for assimilation. The example in (15) shows that function elements that begin with
laryngeal /h/, primarily the third person plural marker, are also targeted for this
assimilation. Finally, in (16), a different vowel gets targeted. The full range of triggers
seem to be all of the vowels except /ɨ/ (represented by e in the official orthography, as
noted earlier).
(14) �ut < �at

Connected speech: Hɨmu �ut o mu:.
Standard speech: Hɨmu �at o mu:.
Translation: He is going to die soon. (entry: mu:k-)
(15) ho < ha
Connected speech: Nt o ho-�ɨ�ɨs�a g ha�iču kai.
Standard speech: Nt o ha-�ɨ�ɨs�a g ha�iču kai.
Translation: I’ll plant several kinds of seeds (entry: �ɨ�ɨs�a)
(16) �a < �o
Connected speech: Ṣa �a wa s-namkig.
Standard speech: Ṣa �o wa s-namkig.
Translation: It is quite expensive. (entry: namkig)
While /i/ is included in the set of triggers, it is rarely attested in the set of undergoers.
Examples (12–15) showed that the vowel /a/ can undergo total assimilation;
Example (16), along with the set of examples in (17), shows that other vowels can also
be t argeted for total assimilation. In (17), the high central unrounded vowel /ɨ/ surfaces
as an assimilated connected speech form before each of the four triggering vowels,
/i o u a/. In these cases, the function word that is targeted is the reflexive marker �ɨ-, and
the triggering elements include the directional mood marker (17a), and indefinite pro-
noun in (17b). The final example in this set, (17d), is intriguing because the targeted
vowel is identical to the first vowel in the word to which it is affixed, but it instead
assimilates to the vowel on its left.
(17) a. Connected speech: �I �i- ñɨi.

Standard speech: �I �ɨ-ñɨi.
Translation: He looked at himself.
(entry: ñɨid)
Revisiting Tohono O’odham high vowels 
b. Connected speech: Mt o ha�iču �u-wɨpogidad.

Standard speech: Mt o ha�iču �ɨ-wɨpogidad.
Translation: Each one of you will be imitating something.
(entry: wɨpogid)
c. Connected speech: Nap �ɨ-na:tokč mapt o �o-čɨčɨgaim
si�a im? ɼ
Standard speech: Nap �ɨ-na:tokč mapt o �ɨ-čɨčɨgaim
si�a im? ɼ
Translation: Are you ready for your examination
tomorrow?
(entry: čɨčɨgaimɨd�)
d. Connected speech: S-kɨg �ats� hi wa �a-�ɨñgad�ad.
Standard speech: S-kɨg �ats� hi wa �ɨ-�ɨñgad�ad.
Translation: He was indeed dressed up (hearsay).
(entry: �ɨñgad�ad)
In (18) examples of assimilation occur where the targeted vowel is the mid round
vowel of Tohono O’odham. While forms like (18a) are heavily attested, and forms
showing total assimilation to /u/ are reasonably robust, there do not seem to be any
cases where an /o/ assimilates to /i/, despite the proliferation of environments with a
preceding function word with /i/.
(18) a. Connected speech: Ṣa �a wa �i s-ȷ�uhu�uȷ�u him. ɼ
Standard: Ṣa �o wa �i s-ȷ�uhu�uȷ�u him ɼ
Translation: He is zigzagging (he walks in zigzags).
(entry: ȷ�uhu�uȷ�u ) ɼ
b. Connected speech: Ṣa:ču �u hab kaiȷ�?
Standard: Ṣa:ču �o hab kaiȷ�?
Translation: What’s that noise?
(entry: kaiȷ�)
We find a few examples of this process in the narrative in Fitzgerald et al. (2012) that
show that this process persists in contemporary speech for at least some speakers.
These appear in (19)
(19) Connected speech: �Ant �i �as� o ho-to�akčid.7

Standard: �Ant �i �as� o ha-to�akčid.
Translation: ‘I will just keep them here.’
. For ease of presentation, this example is given without additional fast speech effects.
 Colleen M. Fitzgerald
Finally, in (20), one of the ten total examples of /i/ behaving in what seems to be an
assimilating fashion appear. In all cases, the apparent triggering vowel is an /o/. As
I show further below, there are several aspects of this that suggest that these examples
may not truly be connected speech as identified by Mathiot.
(20) a. Connected speech: Nt o �op ba�ioka�i.

Standard: Nt o �ip ba�ioka�i.
Translation: First, I’ll swallow it. (entry: ba�a)
b. Connected speech: Nt o �opki� gd� hu čɨkioka�i!
Standard: Nt o �ipki� gd� hu čɨkioka�i!
translation: First, I’ll put it away. (entry: čɨka-/čɨki-)
There are several function words that use /i/ and that have the same phonologi-
cal structure of the true assimilating function words, primarily a laryngeal onset
followed by the target vowel. However, there are only these ten tokens; the directional
mood marker �i never shows assimilation to the preceding vowel, for example. Also of
interest is a comparison with the Saxton et al. (1989: 26) dictionary, where the lexical
entry for conjunction �ip lists it with �op (and �op refers the reader back to �ip). This is
not the case for the other cases of assimilating function words, where the Saxton dic-
tionary does not list attestations similar to what has been shown here. Furthermore,
the high front vowel /i/, if it is truly assimilating in these ten cases, is not displaying
assimilation to any other vowel in the set, unlike all the other assimilators. This range
of differences from the others suggest that while (12)–(19) represent total assimila-
tion characteristic of connected speech, the cases in (20) represents allomorphy at a
lexical level.
If the arguments against regarding /i/ as an actual undergoer of the assimila-
tory process hold up, then it appears /ɨ o a/ all function as targets for assimilation.
The triggering vowels are /i o u a/, although /o/ fails to assimilate to /i/. Overall,
the vowels /i u/ resist assimilation, with the remaining vowels all acting together
and as undergoers of assimiliation to varying degrees. Additionally, /ɨ/ never trig-
gers assimilation, although this may in part be because its distribution as a function
word is t ypically before content words (as when it is a reflexive marker), rather than
before other function words, where it would appear in the triggering environment
as a p
ossible catalyst. It is difficult to tease apart the lack of triggering environment
from other aspects of /ɨ/ that might be at play. While /i u/ can trigger total assimila-
tion, they essentially never undergo it. The apparent absence of /u/ as an assimilator
follows from its nonoccurrence in the person prefixes and the fact that its distribution
elsewhere in function vocabulary is usually as either a second element or in conjoined
auxiliaries like kut, which reduce to forms like t rather than displaying vowel assimi-
lation. The absence of /i/ as an undergoer cannot be explained by these factors, and so
must be a result of the phonology.
Revisiting Tohono O’odham high vowels 
The pattern of distribution from the connected speech assimilatory patterns

suggests once again that the full set of high vowels fail to form a natural class. Instead,
/i u/ fail to undergo total assimilation, while /ɨ/ and the remaining vowels do p
articipate.
In (21), the chart models the natural classes suggested by total assimilation patterns in
connected speech:
(21) Vowel patterning based on assimilation in connected speech

i u
ɨ
o
a
6. High vowels and coronal consonants
In this section, I turn to a discussion of a more well-known consonant-vowel pattern

in Tohono O’odham, one that is displayed by the high vowels and the coronal conso-
nants. In particular, the palatal series of consonants have a restricted distribution before
/o a/. Like the retroflexed pattern discussed in section three above, morphophonologi-
cal considerations complicate the pattern considerably in noninitial contexts. However,
unlike the patterns in the previous sections of the paper, this phenomenon is the single
one presented in this paper in which /i ɨ u/ pattern as a class. This section thus serves as
a counterpoint to the arguments in prior sections for treating /i/ as asymmetric.
Tohono O’odham high vowels and coronal consonants exhibit several interesting
quirks, as noticed by various researchers (For example, Mason 1950; Saxton 1963;
Hale 1965; Hill et al. 1994; Fitzgerald & Fountain 1995; Miyashita Forthcoming). Hill
et al. (1994) show that only a subset of the coronal consonants, the palatal series,
appear before the three high vowels. Word-initially, the palatal consonants never
appear before the nonhigh vowels /a o/, save for a small set of borrowings like ča:ŋgo
‘monkey’. Noninitially, however, the generalizations are not as clear-cut, as is seen
further below. I will work through an initial set of data that shows the opposing pat-
terns of palatal consonants before high and nonhigh vowels, but there are also several
sets of data that make it challenging to treat this as a clear case of complementary
distribution, and I review those further below. The data in (22) provide the first set of
surface patterns suggesting that all three high vowels act together as a natural class.
Note that I am using ‘alveolar’ to group the dentals and alveolars, and ‘palatal’ to
group the post-alveolar (or palato-alveolars) and palatals, since these two groupings
act in opposition to each other.
 Colleen M. Fitzgerald
(22) Word-initial distribution of alveolars and palatal, before each vowel

Alveolars Palatals
Gloss Gloss
/ti/ tianna ‘store (>Sp. tienda)’ /či/ čikpan ‘is/are working’
/tɨ/ – /čɨ/ čɨhia ‘daughter’
/tu/ tu:ŋgo ‘dress (>Sp. túnica)’ /ču/ ču:wi� ‘jackrabbit’
/to/ toka ‘doubleball’ /čo/ –
/ta/ ta:pañ ‘to have a split, /ča/ –
crack or crevice’
/di/ – /ȷ�i/ ȷ�iwa ‘to arrive’

/dɨ/ – /ȷ�ɨ/ ȷ�ɨwud� ‘earth, dirt’
/du/ – /ȷ�u/ ȷ�ud�um ‘black bear’
/do/ do�iču ‘unripe or unbaked /ȷ�o/ –
one’
/da/ daikud� ‘chair’ /ȷ�a/ –
/ni/ – /ñi/ ñi�oki� ‘words, language’
(dial.)
/nɨ/ – /ñɨ/ ñɨm ‘liver (of someone)’
/nu/ – /ñu/ ñu:kud ‘to take care of oneself ’
/no/ nowi� ‘hand, arm’ /ño/ –
/na/ nakog ‘to stand, endure obj’ /ña/ –
The pattern in (22) appears to be a classic pattern of complementary distribution,

whereby alveolars appear only before the non-high vowels, and palatals precede high
vowels. However, the data in (23) shows there is a full set of contrasts for both places
of articulation in word-final position. In order to provide a full range of distributional
slots, each consonant is given following each of the five vowels to show that the preced-
ing vowel does not affect this pattern at all.
(23) Word-final distribution of alveolars and palatal, following each vowel

Alveolars Palatals
/it/ ma:kait ‘to teach obj the secrets of /č/ wɨ:č ‘to be lying
curing’ somewhere’
/ɨt/ �apɨt ‘to get to be ready’ /ɨč/ �apɨč ‘to prepare obj, to
get obj ready, perf ’
Revisiting Tohono O’odham high vowels 
/ut/ ču:t ‘to reduce dry obj to /uč/ ču:č ‘to blow out, put
powder, to grind, obj’ out, extinguish
(fires), to turn off
(lights) perf ’
/ot/ wo�ot ‘to make a charco’ /oč/ mo�oč ‘to have one’s head
on obj (such as a
pillow)’
/at/ ga:t ‘bow, gun, rifle’ /ač/ ma:č ‘to learn
something’
/id/ ñɨid ‘to see, look at, watch obj’ /iȷ�/ ha�asiȷ� ‘to be of about a
certain (large) size’
/ɨd/ gagɨd ‘to roast obj (such as /ɨȷ�/ gɨ�ɨȷ� ‘to be big, to be too
several pieces of chicken, much’
several chickens), durative’
/ud/ ñɨ�ičud ‘to sing and dance for obj’ /uȷ�/ čukuȷ� ‘to blacken obj,
perf ’
/od/ wopod ‘to have down, fur, hair on /oȷ�/ nawoȷ� ‘friend, brother,
one’s body, durative’ male cousin
(of a man)’
/at/ mas�ad ‘moon’ /aȷ�/ kɨgaȷ� ‘to be pretty, nice,
good’
/in/ sihowin ‘to rummage into, poke /iñ/ wawiñ ‘to be sated, to
at obj’ have quenched
one’s thirst’
/ɨn/ ñɨn ‘to wake up (same as ñia)’ /ɨñ/ �ɨ�ɨbɨñ ‘to be afraid of obj
repeatedly, unitive’
/un/ �o hun
ɼ ‘unbranded cow or horse’ /uñ/ mu:ñ ‘beans’
/on/ to:n ‘knee’ /oñ/ toñ ‘heat’
/an/ �a�an ‘a pair of wings’ /añ/ ha:s�añ ‘saguaro cactus’
In order to maintain the case that this is complementary distribution, a set of abstract
underlying short high vowels typically have been posited to account for the surface
palatals, with those vowels deleted by rule following palatalization. In many cases,
these vowels never surface in other contexts, resulting in a highly abstract set of under-
lying representations. Additionally, it necessitates a set of additional assumptions on
vowel length that result in positing a set of vowel distribution patterns that is larger
than those that occur in surface forms. The result is a set of highly abstract underlying
 Colleen M. Fitzgerald
forms essentially to predict that final palatals are derived and to predict the comple-
mentary distribution found in clear cases throughout the word preceding high vowels.
The data in (23) also present a set of challenges because not all vowel-consonant
combinations are equally well-attested in word-final position. Commonly occurring
patterns above do not carry additional morphological information in their gloss
above. The least common or most challenging combinations to find involve a final
/t/ or a sequence where the final consonant is preceded by /ɨ/. Where possible, I have
avoided using forms above that are reduplicated (i.e. plural or repetitive, among other
glosses), truncated (perfective), dialectal variants, and suffixed forms (such as the -t
causative) by choosing more simple forms wherever possible, but it is impossible to fill
out the full range of patterns and avoid all such forms.
An additional set of facts involving the perfective auxiliaries challenges the under-
lying high vowel proposal. The auxiliaries shown in (24) are most easily analyzed by
positing alveolar assimilation to the perfective -t, and with no (abstract) intervening
high vowel.
(24) Short auxiliaries (imperfective/perfective)

1s ñ/nt 1p č/tt
2s p/pt 2p m/mt
3s �o 3p �o
An analysis assuming underlying high front vowels is also complicated by the

patterns in (25), when again the ‘palatalized’ series appears to depalatalize preced-
ing a n
onpalatal coronal like /t/. Related morphologically complex forms can precede
noncoronal consonants and still surface as a palatal, as in nawoȷ��ma ‘to be friendly to
object.’ The hypothesized presence of an abstract underlying /i/ that never surfaces
moves the analysis to more complicated levels with these types of patterns.
(25) Assimilation of Palatals to Alveolars
Alveolars Palatals
/d/ nawod-t ‘to make a friend’ /ȷ�/ nawoȷ� ‘friend’
gɨ�ɨ-d-ta ‘to get big’ gɨ�ɨ-ȷ� ‘to be big’
/n/ kotont ‘to make a shirt out of /ñ/ kotoñ ‘shirt’

obj’
ȷ�ɨnto ‘to finish smoking obj ȷ�ɨ:ñ ‘to smoke obj’
once, unitive’
ȷ�unt ‘to make (a cake of) ȷ�uñ ‘dry saguaro fruit
dry saguaro fruit formed into a cake’
candy out of obj’
n-na:k ‘my ear’ ñ-pi kañ
ɼ ‘my ear’
The pattern in (25) is found in other paradigms that include high-vowel-initial

suffixes, (such as -id or -in), which then trigger the retroflexed consonants to move to
Revisiting Tohono O’odham high vowels 
their plain coronal counterparts. Saxton (1982) characterizes this as part of a process
that includes a slightly larger set of coronals, /d d� s� n/, which either palatalize or lose
retroflexion preceding the high front vowel.
Finally, there appears to also be an asymmetric distribution between the high
front vowel /i/ and the high central vowel /ɨ/ when it comes to surface forms of prefixes
and suffixes. Aside from reduplication, which involves an identity relationship of seg-
ments in the base and the reduplicant, the high central vowel only appears to surface
in a single affix, a prefix, the reflexive, �ɨ-. In contrast, the high front vowel /i/ appears
in numerous suffixes, thus serving as a relatively frequent trigger for its associated
phonological processes, such as palatalization.
Overall, the initial set of data in this section, the opposition between alveolars and
palatal consonants in word-initial position, suggests that the high vowels act together
as a natural class. However, this pattern is mitigated by the highly complicated rela-
tionship and distribution of two phonetic subcategories of coronal place elsewhere in
the word, as well as by problematic aspects of hypothesizing an abstract underlying
high vowel in other places in the word. In the subsequent sets of data, /i/ acts as a more
dominant trigger in terms of forcing phonological processes. While those sets of data
suggest that /i/ is not a natural class with /ɨ u/, there was evidence from the first set of
patterns in this section suggesting that the high vowels do form a group, as illustrated
in (26):
(26) Vowel patterning based on alveolar and palatal distributions (cf. data in (22))
i ɨ u
o
a
7. Conclusion
Tohono O’odham’s unusual and asymmetric vowel inventory represents one of many
ways in which Uto-Aztecan phonology serves as a challenging domain in which to
work. The inventory itself is quite light on front vowels, but robust in terms of its
apparent set of three high vowels, overall more than half of its entire vowel inven-
tory. The intrigue continues on into the distributional patterns of consonants and the
phonological processes throughout the language. As shown throughout this paper,
the set of high vowels rarely acts as a unified natural class. Complicating the details
throughout are the rich and varied processes of prosodic morphology, like truncation
and reduplication, which together with suffixation mean that any given snapshot of a
word must consider its morphological structure and derivational history in order to
 Colleen M. Fitzgerald
get a sense of its distribution in its “true” local environment. The surface is not what
it seems and in order to get a sense of the “present” state of a word, it is key to know
its past.
Jane Hill’s work as a scholar in Uto-Aztecan languages puts these languages in
their familial context, their historical context, and considers the neighboring lan-
guages even as she looks at the individual language. The long and respectful collabo-
ration between Jane Hill and Ofelia Zepeda allows a richness of context and culture
to permeate their joint work. Her work has a deep sense of history and past imbuing
the present, somewhat like these patterned constellations of different groups of high
vowels influencing the consonants of Tohono O’odham.
References
Dart, Sarah N. 1993. Phonetic properties of O’odham stop and fricative contrasts. International
Journal of American Linguistics 59(1): 16–37.
Fitzgerald, Colleen M. 1996. Aspiration in Tohono O’odham. Paper presented at Winter Meeting
of the Society for the Study of the Indigenous Languages of the Americas.
Fitzgerald, Colleen M. 1997. O’odham Rhythms. Ph.D. dissertation, University of Arizona.
Fitzgerald, Colleen M. 1999. Loanwords and stress in Tohono O’odham. Anthropological
Fitzgerald, Colleen M. 2000. Vowel hiatus and faithfulness in Tohono O’odham reduplication.
Linguistic Inquiry 31(4): 713–722.
Fitzgerald, Colleen M. 2002. Tohono O’odham stress in a single ranking. Phonology 19(2):
253–271.
Fitzgerald, Colleen M. 2009. Pathways for accessing legacy materials in Tohono O’odham. Paper
presented at the 1st International Conference on Language Documentation and Conserva-
tion, University of Hawai’i, Honolulu HI.
Fitzgerald, Colleen M. 2012. Prosodic inconsistency in Tohono O’odham. International Journal
of American Linguistics 78(4).
Fitzgerald, Colleen M. & Fountain, Amy V. 1995. Subtractive morphology, Tohono O’odham
and optimality theory. Paper presented at the 1995 Meeting of the Linguistic Society of
America, New Orleans.
Fitzgerald, Colleen M., Miguel, Phillip & Tucker, Stella. 2012. Contemporary storytelling in
Tohono O’odham. In Inside Dazzling Mountains: Contemporary Translations of Southwest
Native Verbal Arts. David Kozak (ed.), 594–617. Lincoln NB: University of Nebraska Press.
Hale, Kenneth. 1965. Some preliminary observations on Papago morphophonemics. Interna-
tional Journal of American Linguistics 31: 295–395.
Hale, Kenneth. 1970. On Papago laryngeals. In Languages and Cultures of Western North
America: Essays in Honor of Sven S. Liljeblad, Earl H. Swanson, Jr. (ed.), 54–60. Pocatello
ID: Idaho State University Press.
Hill, Jane H. & Zepeda, Ofelia. No date. Dialect survey data. Ms, University of Arizona.
of American Linguistics 58: 355–404.
Revisiting Tohono O’odham high vowels 
Linguistics 40: 1–42.
Hill, Jane H., Zepeda, Ofelia, DuFort, Molly & Belin, Bernice. 1994. Tohono O’odham vowels.
Ms, University of Arizona.
Jackson, Eric. 2003. Dispersion in the vowel system of Pima. Paper presented at the 2003 Meeting
of the Linguistic Association of the Southwest.
Langacker, Ronald. 1970. The vowels of Proto Uto-Aztecan. International Journal of American
Mason, J. Alden. 1950. The Language of the Papago of Arizona. Philadelphia PA: University of
Pennsylvania Museum.
Mathiot, Madeleine. 1973. A Dictionary of Papago Usage. Bloomington IN: Indiana University.
Miyashita, Mizuki. 2002. Tohono O’odham Syllable Weight: Descriptive, Theoretical and
Applied Aspects. Ph.D. dissertation, University of Arizona.
Miyashita, Mizuki. 2004. O’odham collateral reduplication. Paper presented at the Workshop on
American Indian Languages, Santa Barbara CA.
Miyashita, Mizuki. 2011. Diphthongs in Tohono O’odham. Anthropological Linguistics 53(4):
323–342.
Saxton, Dean. 1963. Papago phonemes. International Journal of American Linguistics 29(1):
29–35.
Saxton, Dean. 1982. Papago. In Studies in Uto-Aztecan Grammar 3, Ronald Langacker (ed.),
92–266. Arlington TX: SIL Publications in Linguistics.
Saxton, Dean, Saxton, Lucille & Enos, Susie. 1989. Dictionary, Papago/Pima-English,
O’otham-Milgahn. Tucson AZ: University of Arizona Press.
Zepeda, Ofelia. 1988. A Papago Grammar. Tucson AZ: University of Arizona Press.
Zepeda, Ofelia. 1995. Ocean Power: Poems from the Desert. Tucson AZ: The University of
Arizona Press.
Zepeda, Ofelia. 1997. Jeweḍ ‘I-Hoi: Earth Movements. Tucson AZ: Kore Press.
Zepeda, Ofelia & Jane Hill 1998. Collaborative sociolinguistic research among the Tohono
Head-marking inflection and the architecture
of grammatical theory
Evidence from reduplication and compounding
in Hiaki (Yaqui)
Jason D. Haugen & Heidi Harley

Oberlin College / University of Arizona
Within generative grammar, noun incorporation and other compounding

processes have traditionally been the focus of morpho-syntacticians, while
reduplication has been investigated primarily by morpho-phonologists.
The interaction of these two phenomena in a single language has significant
implications that go beyond the narrow concerns of these two sub-domains,
bearing much more broadly on the architecture of grammatical theory. This
paper investigates the interactions of reduplication and compounding within one
language, Hiaki (Yaqui). Reduplication for aspectual inflection in Hiaki occurs
inside of compounds and other derived words, marking the head of the word.
We demonstrate the major architectural issues resting on the analysis of these
phenomena by examining how different theoretical perspectives can (or cannot)
accommodate the Hiaki data.
Keywords: inflection; head-marking; reduplication; noun incorporation;

Hiaki (Yaqui)
1. Introduction
Two areas that have received an extraordinary amount of attention in contemporary

theory-building have been reduplication, at the interface of morphology with pho-
nology, and noun incorporation (NI), at the interface of morphology with syntax.
Given the wide cross-linguistic distribution of these two phenomena in addition to
their great theoretical importance, it is relatively surprising that the interaction of the
two has received extremely little (and in fact, almost no) attention in the literature.
The purpose of this paper is to provide extensive empirical documentation of how
reduplication interacts with noun incorporation and other types of verb compound-
ing in one language, Hiaki (Yaqui). We will show that V(erb)-V(erb) compounds
 Jason D. Haugen & Heidi Harley
allow reduplication to apply to either (or both) of the verbal elements, whereas noun
incorporation constructions, which are in essence N–V compounds, only allow for
reduplication as a prefix to the verbal element. Inflectional reduplication thus appears
inside the compound in noun incorporation contexts.
We then go on to address the theoretical ramifications that these data have for a
variety of different frameworks in theoretical morphology. The central issue raised by
the interaction of reduplication with compounding in Hiaki is word-internal head-
marking inflection. As we will show, the Hiaki reduplication + compounding data, and
word-internal head-marking more generally, raise interesting problems for competing
theories. We will argue that those theories whose architectures contain inflectional
and derivational processes in a single grammatical component (e.g. Strong Lexicalism
or syntacticocentric theories like Distributed Morphology) fare far better in account-
ing for these data than those which would separate the two into separate components
(e.g. Weak Lexicalist theories). We will also argue that a recent theory designed explic-
itly to account for head-marking inflection, Paradigm Function Morphology (PFM)
(Stump 2001), can straightforwardly account for the basic Hiaki reduplication facts
but runs into empirical problems arising from the interaction of reduplication and
compounding with yet other aspects of Hiaki grammar.
Before proceeding to our main discussion, however, we would first like to say
how thrilled we are to include our paper in this volume in honor of the one and
only Jane Hill. She has been an exemplary scholar and wonderful personal friend to
both of the present authors (not to mention her service as co-advisor on Haugen’s
2004 University of Arizona dissertation). No single paper could hope to encompass
all of the many voices of Jane Hill – indeed, it is our belief that no single person
could match the range of topics that she has covered with the depth and insight
that she has contributed to so many branches of linguistics, anthropology, and fields
beyond. The particular “voice” of Jane Hill with which we hope to connect our pres-
ent paper is the one that applies linguistic data collected from the documentation of
threatened and endangered languages to empirical considerations crucial to linguis-
tic theorizing – i.e. what Ken Hale (2000) has called the “confirmatory function” of
linguistic diversity. Along these lines Jane’s own work on prosodic morphology and
word derivation in Tohono O’odham (Hill & Zepeda 1992) and the morphology and
semantics of reduplication in Tohono O’odham (Hill & Zepeda 1998), as well as her
work on the theoretical ramifications of the complex verb morphosyntax of Cupeño
(Hill 2003), spring immediately to our minds. We hope that our remarks on the
theoretical ramifications of reduplication and compounding in Hiaki will be taken
in a similar spirit.
We’d also like to thank Jane for all of the work that she has done to foster at the
University of Arizona a unique and exciting research community of scholars working
Head-marking inflection and the architecture of grammatical theory 
on all aspects of Uto-Aztecan linguistics, through which we have both profited

greatly – intellectually, personally, and otherwise. This one’s for you, Jane!
This paper is organized as follows. Section 2 outlines how the two different mor-
phological processes of reduplication and noun incorporation (N–V compounding)
each work in Hiaki. Section 3 then discusses the interaction of Hiaki reduplication and
compounding processes in several different contexts: NI constructions (§3); in hybrid
verb constructions (§3.1); with verbal suffixes (§3.2); and in “pseudo”-compounds
(§3.3). Section 4 then focuses on the implications of the word-internal reduplica-
tion process that occurs specifically in NI constructions (and pseudo-compounds),
in standard theoretical architectures: Weak and Strong Lexicalist approaches and the
syntacticocentric approach to incorporation proposed by Baker (1988). We conclude
that the order of the inflectional reduplicative morpheme in Hiaki, which appears
word-internally marking the verbal head of the compound, cannot be accounted for
in the Weak Lexicalist theory or in Baker’s theory without additional architectural
accommodations.
Section 5 then discusses some architectural accommodations that have been
proposed which can account for these data, in both Lexicalist and syntacticocen-
tric frameworks. We argue that while the Hiaki reduplication data specifically are
consistent with the predictions of the Paradigm Function Morphology proposed by
Stump (2001), its interaction with other cases of inflection violate a crucial principle
of PFM – namely, inflectional object clitics exhibit edge-marking behavior in the
same forms in which reduplication exhibits head-marking, in violation of PFM’s Par-
adigm Uniformity Generalization. We close by discussing how the Hiaki data can be
accounted for in a syntacticocentric framework by amending Baker’s theory with the
notion of Local Dislocation, within the theory of Distributed Morphology. Section 6
concludes.
1.1 Background: Reduplication and compounding in the literature

Reduplication interacting with the general phenomenon of compounding has
received some recent attention, and we identify two general types of previous research
in this area. The first, of particular importance in the present context, is reduplication
plus compounding. Recent work in this area includes discussion of reduplication in
nominal compounds in Pima (Munro & Riggle 2004), a Uto-Aztecan language of
the Tepiman sub-group (see Footnote 12 below), as well as Haugen (2010), which
provides a more general discussion of reduplication in compounding contexts
cross-linguistically.
A second strand of research has examined the process of reduplication itself
as a kind of compounding (e.g. Zoll 2002; Downing 2003; Inkelas & Zoll 2005).
 Jason D. Haugen & Heidi Harley
For example, Downing (2003) proposes that reduplicated verb stems in Bantu are
composed of a compound stem formed by a reduplicated stem combining with the
base stem, as in (1):
Compound Structure for Reduplicated Bantu Verb Stems

(1)
(Downing 2003, p. 7 [7b])
Verb Word
infl MacroStem
(OM) [Compound Stem]Stem1
[red Stem]Stem2 [Base Stem]Stem3

Inkelas and Zoll (2005) provide a similar representation in Morphological Doubling

Theory (MDT), but they propose instead that a reduplicant and its base are simply
potentially heterogeneous daughters of a morphological mother node. The representa-
tion that they give for reduplication structures is shown in (2):
Schematic for Reduplication in Morphological Doubling Theory

(2)
(Inkelas & Zoll 2005, p. 19 [27])
[zzz]
Co-phonology Z
[xxx] [yyy]
Co-phonology X Co-phonology Y
/Stemi/ /Stemi/

Our focus in this paper will involve only the first kind of investigation, as we will
be examining the interactive processes of inflectional reduplication occurring in the
contexts of noun incorporation and other compound verbs in Hiaki.
2. Reduplication and noun incorporation in Hiaki
Hiaki (Yaqui) is a Southern Uto-Aztecan language of the Taracahitic sub-group, and

is indigenous to northwestern Mexico (Sonora) but also spoken in southern A rizona,
USA. Hiaki allows for the productive compounding of verbal roots to indicate various
semantic notions. These compounds do not generally allow for any other constituent
or affix to intervene between their components, with one exception: reduplication.
Head-marking inflection and the architecture of grammatical theory 
2.1 Reduplication in Hiaki

Reduplication in Hiaki is a productive inflectional process typically marking habitual
or progressive aspect (also, in some cases, emphasis) on verbs, and it is generally pre-
fixal. There are several forms of reduplication which can be used for any of the above
semantic functions: a light syllable reduplicant; a heavy syllable reduplicant that trig-
gers gemination of the onset of the base into the coda of the reduplicant; a disyllabic
reduplicant; and a word-internal pattern of morphological gemination. In general,
neither the form nor the meaning of a reduplication type is fully predictable based
on the phonological makeup of the stem to which it attached. We will not concern
ourselves here with this rampant reduplicative allomorphy; for further discussion see
Harley and Amarillas (2003), Haugen (2003), and Harley and Florez Leyva (2009).
Examples illustrating the three most common meanings associated with one of the
reduplication types, a light syllable, are presented in (3) (the reduplicant will appear in
bold in all examples henceforth):1
(3) a. Habitual (hab)

Itepo hunum ke-ke’ewe
1.pl there red-gather.firewood
‘We gather firewood there.’
b. Progressive/continuative (‘in progress’), (prog)
Uu hamut totoi kava-m bwa-bwata
The woman chicken egg-pl red-stir
‘The woman is mixing the eggs.’
c. Emphatic (emph) (often in Imperative (imp) examples)
Kat=ee uka soto’i-ta hunum ma-mana
neg.imp=2.sg the.acc pot-acc there red-put
‘Don’t put that pot there.’
(Harley & Leyva 2009, p. 253 [13])
. The abbreviations for our glosses are as follows: 1 = 1st person; 2 = 2nd person; 3 = 3rd
person; acc = accusative; af = unsepecified affix; ag.nom = agentive nominalizer; appl =
applicative; caus = causative; cont = continuative; desid = desiderative; det = determiner;
dir = directive; emph = emphatic; fut = future; hab = habitual; incep = inceptive; inst =
instrumental; intr = intransitive; neg.imp = negative imperative; obj = object; obl = oblique;
perf = perfective; pass = passive; past = past tense; pl = plural; ppl = past participle; pret.
aug = preterite augmentative; prosp = prospective; red = reduplication; rev = reveren-
tial; sg = s ingular; subj = subject; to = directional postposition; tran = transitive; WH =
question word.
 Jason D. Haugen & Heidi Harley
2.2 Noun incorporation in Hiaki

Noun incorporation involves the compounding of a noun with a verb, where in the
usual case the noun satisfies the verb’s internal argument selectional properties. A typi-
cal Hiaki example is given in (4) below:
(4) Peo maso-peu-te-k

Peo deer-butcher-intr-perf
‘Peo deer-butchered’ (Jelinek 1998, p. 213 [48])
Hiaki is an SOV language, so objects normally precede their verbs in any case. There
are, however, clear diagnostics that distinguish incorporated objects from non-
incorporated O–V juxtapositions. First, an incorporated nominal in Hiaki does not
indicate number or case, which contrasts with object nouns in verb-external object
noun phrases, which must do so – compare the inflection on the nominal stem maaso
‘deer’ in (5) with the absence of inflection on its incorporated counterpart in (4).2
(5) a. Peo maso-ta peu-ta-k
Peo deer-acc.sg butcher-tran-perf
‘Peo butchered a deer’
b. Peo maso-m peu-ta-k
Peo deer-pl butcher-tran-perf
‘Peo butchered some deer’
Second, no constituent may intervene between a noun and the verb in an NI con-
struction, as seen by comparing (6a′) with its overtly transitive counterpart in (6b′);
the adverbial aman in the latter can intervene between the verb and its inflected
object, but not between the verb and the incorporated, uninflected nominal in the
former.
(6) a. kuta-siu-te a′. *kuta-aman-siu-te
stick-tear-intr stick-there-tear-intr
‘wood-split’ ‘wood-split over there’
b. kuta-m siu-ta b′. kuta-m aman siu-ta
stick-pl tear-tran stick-pl there tear-tran
‘split wood.’ ‘split wood over there’
The above examples show that the integrity of the noun-verb complex in Hiaki NI
cannot be disrupted by another syntactic constituent, such as a locational adverb.
. The underlying vowel in maaso ‘deer’ is long and would surface as such in, e.g. the
nominative case, which is unmarked; long vowels in Hiaki stems regularly shorten under
suffixation.
Head-marking inflection and the architecture of grammatical theory 
Third, in some cases, these Hiaki NI constructions involve nominal stems rather
than fully free nominals. In Hiaki, some stems have special forms which are used
when the stem is subject to derivational affixation; these are distinct from the free
stems which are typically used in inflectional affixation (see Tubino Blanco & Harley
2010 for a full discussion). The NI examples in (7) show that these special deriva-
tional stems are used in NI constructions, and are hence easily distinguished from
the corresponding verb phrases with independent NP objects, which would use the
free stem form for the object. The incorporated forms in (7) are derived using the
bound stems of the nominals whose free stems are chichi ‘saliva’ and hipetam ‘bed’,
respectively:
(7) a. chit-wat-te ‘spitting’

saliva-throw-intr derived from chichi ‘saliva’ n. + watta ‘throw (tr.)’
b. hipe-teka ‘make bed’
bed-lay.across.tran derived from hipetam ‘bed’ + teéka ‘lay sthg. across’
Fourth, word-internal phonological processes apply inside these NI constructions

in Hiaki, showing that they form a single phonological word. For example, they
undergo the word-internal phonological rule of /s/ → [h] in word-medial syllable final
position. (8) illustrates this process in a normal affixation context, using the verb stem
bwasa/bwase ‘cook (tr./intr.)’; when suffixed with the future suffix -ne in (8b)(8), the
stem-final /s/ becomes [h]:
(8) a. Haisa intok bwa-bwasa’a-wa?

how and red-cook-pass
‘And how are they cooked?’ (Conversation 8 [15])
b. abwe, oowa-m ae-t mo-monto-wa hunak veha bwah-ne
well coal-pl it-on red-pile-pass then then cook-fut
‘Well, coals are placed on it and then it will cook’ (Conversation 8 [16])
This rule also applies in NI constructions, illustrated in (9). The noun stem lioh ‘God’
in (9a) is derived from Hiaki Lios, which in turn was borrowed from the Spanish dios
‘God’. It appears as Lios in non-compounded contexts (9b):
(9) a. lioh-bwania
God-promise
‘giving thanks’
b. Lios enchi ania.
God you.acc help
‘(May) our creator help you’ (a traditional greeting)
Fifth, these N–V compounds are conducive to idiomatic interpretation, and are often
used for common, culturally-relevant activities, as is characteristic of compounding NI
 Jason D. Haugen & Heidi Harley
constructions more generally (Escalante 1990, p. 105; Dedrick & Casad 1999, p. 161)
(cf. Mithun 1984):
(10) a. tekil-maka ‘commissioning, making responsible’ (< lit. ‘work-giving’)
b. kuchu-sua ‘fishing’ (< lit. ‘fish-kill.pl.obj.’)
Hiaki noun incorporation, then, appears to produce normal N–V compounds, exhib-
iting all the appropriate word-like properties expected from the result of a deriva-
tional process. Mithun (1984) differentiates two sub-types of what she calls Type 1 NI:
true compounding NI, which creates a single verbal word, and composition by jux-
taposition, which exhibits a much looser morpho-phonological connection between
the two elements of the compound. Hiaki NI constructions must be regarded as of
the former type, since NI compounds meet all the identificational criteria laid out by
Mithun.
2.3 Hiaki NI as detransitivizing?

Rosen (1989) proposes a major distinction among NI types along the lines of transitiv-
ity – i.e. intransitive versus transitive NI constructions (compound vs. classifier NI, in
her terminology). Hiaki NI has generally been taken to be intransitivizing (Escalante
1990; Jelinek 1998), as illustrated in the following contrasting examples presented by
Jelinek (1998 p. 213 [48]), repeated from (4) and (5) above:
(11) a. aapo maso-ta peu-ta-k

3sg deer-acc butcher-tran-perf
‘He butchered a deer.’
b. aapo maso-peu-te-n
3sg deer-butcher-intr-past
‘He was deer butchering.’
c. *aapo bwe’uu-k maso-peu-te-n
3sg big-acc deer-butcher-intr-past
[*‘He was [big deer]-butchering.’] or [*‘He was deer-butchering a
big one.’]
(11a) presents a transitive sentence with an accusative case-marked direct object nomi-
nal (masota ‘deer-acc’) and a verb marked with the transitive suffix -ta. (11b) provides
the corresponding NI example, where the nominal is compounded with the verb (and
is no longer marked with the accusative marker), and the verb takes the intransitive
suffix -te. That such NI verbs are truly intransitive is further demonstrated in (11c),
which illustrates the ungrammaticality of external modifiers (adjectives, numerals,
determiners, etc.) with incorporated nominals.
However, in contrast to Jelinek’s data, in some cases it appears to be possible to
have “stranded” or (“null-head”) modifiers with incorporated nominals in Hiaki.
Head-marking inflection and the architecture of grammatical theory 
Molina et al. (1999) list pan hoa as an intransitive verb meaning ‘to make bread’ (< pan
‘bread’ + hooa ‘make’). In (12) we present new empirical evidence that this is actually
a transitive NI construction:
(12) a. Irene panim am-hoo-ria

Irene bread-pl 3.pl-make-appl
‘Irene is making bread for them.’
b. Irene am=pan-hoo-ria
Irene 3.pl=bread-make-appl
‘Irene is making bread for them.’
c. *Irene pan am=hoo-ria
Irene bread 3.pl=make-appl
d. Irene oficiom sii kiam pan-hoo-ria
Irene oficio very delicious-pl bread-make-appl
‘Irene is making very delicious bread for the ceremonial officials’
(12b) and (12c) show that the nominal root pan ‘bread’ must be incorporated onto the
verb, given both the inability of the third person plural agreement clitic am= (which
specifies the benefactee argument introduced by the applicative suffix -ria) to inter-
vene between it and the verb hooa, and the absence of inflection on pan. However, in
(12d), the intensifier and adjective modifying pan show that the incorporated nominal
can still be externally modified in some instances, which is a sign of the transitivity of
this construction in Rosen’s (1989) typology.
In theories which posit that an incorporated nominal forms a constituent with its
external modifiers, either through movement (e.g. Baker 1988) or co-analysis (Sadock
1991), such modifiers are considered to be “stranded”. Indeed, such “stranding” of
modifiers is presented as a motivation for these theories. Theories like that of Rosen
1989, on the other hand, regard incorporation as a (non-syntactic) morphological
compounding process, and regard the modifiers in the external NP to be “null-head”
phrases. We’ll return to the differing predictions of these two contrasting approaches
below. In any case, the evidence suggests that Hiaki NI is at least sometimes transitive
NI, in Rosen’s terms, though often intransitive NI in other cases.
We now turn to illustrate the interaction of NI and reduplication in the language:
What happens when a compound V is inflected for habitual aspect by means of
reduplication?
3. Reduplication with noun incorporation in Hiaki
Above we saw that nothing could intervene between the nominal and the verb stem
in Hiaki NI. The only exception to this generalization is the case of reduplication,
 Jason D. Haugen & Heidi Harley
when reduplication is used as the means to express a verbal inflection, as illustrated

in (13):
(13) Peo maso-peu-peute
Peo deer-red-butcher
‘Peo is always butchering deer’
Here, inflection appears to disrupt the lexical integrity of the N–V compound. In NI in
Hiaki, reduplication cannot target the left edge of the compound:
(14) *Peo *ma-maso.peute / *Peo *maso-maso.peute

Peo red-deer.butcher
‘Peo is always butchering deer’
A number of cases of reduplication in Hiaki noun incorporation (NI) constructions

are listed in (15). These examples all involve a nominal stem being compounded with
an identifiable verbal stem, i.e. a verbal element that can also appear as a free verb
without an incorporated nominal.
(15) Verb Meaning Reduplicated Form

a. chit-wat-te ‘spitting’ chit-wat-watte
saliva-throw-intr
b. hiavih-muke ‘gasping, short of breath, hiavih-mu-muke
breath-die (sg. subj.) suffocating’
c. hipe-teka ‘make bed’ hippe-te-teka
bed-lay.across (t.v.)
d. kuchu-sua ‘fishing’ kuchu-su-sua
fish-kill (pl.obj.)
e. kova-hamti ‘deep in thought, kova-ham-hamti
head-broken concentrating, thinking’
f. kupi-tomte ‘lose sight temporarily’ kupi-tom-tomte
eye-blossom
g. lioh-bwania ‘giving thanks’ lioh-bwa-bwania
God-promise
h. lio-noka ‘praying’ lio-no-noka
God-talk
i. Mao-noka ‘speaking in the Mayo Mao-no-noka
Mayo-speak language’
j. masa-vaite ‘flapping wings’ masa-vai-vaite
wing flap
In particular, note that this process applies to cases in which the N–V compound
involves a bound nominal stem, as in (15a) and (15c); that no inflection appears on the
Head-marking inflection and the architecture of grammatical theory 
nominal half in the reduplicated form; and that word-internal phonological processes
still apply to the entire N–V compound with intervening reduplication, as in (15g).
The resulting complex reduplicated NI form continues to behave as a V with
respect to other inflectional processes of the language. To illustrate, consider the fol-
lowing example, which shows that the preverbal object clitics attach to the left edge
of the compounded N–V form at the same time that reduplication is marked internal
to the compound (16a); reduplicating at the left edge of the compound N–V form
is ungrammatical (16b). Similarly, attaching the object clitic to the reduplicated V,
between the incorporated N and its sister V, is ungrammatical (16c):
(16) a. Irene am=pan-ho-hoo-ria
Irene 3.pl-bread-red-make-appl
‘Irene is always making bread for them.’
b. *Irene am=pa(n)-pan-hoo-ria
Irene 3.pl- red-bread-make-appl
c. *Irene pan-am=ho-hoo-ria.
Irene bread-3.pl=red-make-appl
Reduplication in combination with NI in Hiaki, then, applies to the head V within the
otherwise apparently completely atomic N–V compound.
We next turn to illustrate extensively the interaction of reduplication with other
verbal compounding and derivation processes in Hiaki, first in derivations with
“verb-affix hybrids” (§3.1); then in other cases of verb-derivation, such as occurs with
the affixation of derivational suffixes on verbs (§3.2); and finally in words with uniden-
tified word-internal structure (‘pseudo-compounds’; §3.3).
3.1 Reduplication and verb-affix hybrids

In addition to the reduplication of N–V compounds that occurs in NI constructions,
Hiaki has a small, closed class of free verb roots that can also be used as verbal suffixes.
Under one possible analysis, such constructions could be considered to be verb-verb
compounds. We refer to this closed class of verbs as “verb-affix hybrids”.3 Examples
showing both free and compounded uses of maachi ‘appear’ are presented in (17), and
a pair illustrating free and bound uses of siime ‘go’ are presented in (18):
(17) a. Hai=sa maachi huu’u ’em sa’awa
how=wh appear that your sore
‘How does your sore seem?’ Or ‘how is your sore?’
(Dedrick & Casad 1999, p. 67 [39])
. These behave identically to verbal affixes (suffixes) with respect to binding and the
assignment of case in subordinate clauses (see Tubino et al. 2009 for discussion).
 Jason D. Haugen & Heidi Harley
b. Kaita-e mo’iti-machi
nothing-inst plow-appear
‘There is nothing with which to plow’
(Dedrick & Casad 1999, p. 67 [40])
(18) a. Yoko=ne potam-meu sim-ne
tomorrow=I Potam-to go-fut
‘I’m going to Potam tomorrow.’ (Dedrick & Casad 1999, p. 293 [1])
b. Inepo ili hu’unee-sime
I little know-go
‘I’m beginning to understand a little bit.’
(Dedrick & Casad 1999, p. 294 [7])
These V–V compounds exhibit properties similar to those illustrated above for N–V
compounds with respect to the use of bound stem forms for the left-hand member,
word-internal phonological alterations, and absence of internal inflectional material;
they show no evidence of being derived by a distinct word-formation process from the
NI cases.
Reduplication with such verb-affix hybrids functions in the same way as with the
NI compounds: it appears word-internally, reduplicating the rightmost member of the
V–V compound. We illustrate the word-internal, head-reduplication pattern with our
example hybrid verbs maachi ‘seem, appear’ and siime ‘go’ in (19) below; Escalante
(1990, p. 78) also emphasizes similar cases:
(19) a. Vempo si kuhti-ma-machi

3pl emph angry-red-seem
‘They really seem like hateful people’.
b. Hita=sa empo hoo-si-sime
what=wh you do-red-go
‘What are you going around doing?’
Unlike the case for reduplication with NI, however, reduplication in these V–V com-
pounds can target either member of the compound, so long as the semantics of redu-
plication can be applied to either member. An example of reduplication applying to
the first verbal element of a verb-affix compound is given in (20), and an example with
both members reduplicating is given in (21).
(20) Vempo si kuh-kuhti-machi

3pl emph red-angry-seem
‘They seem like really hateful people.’
(21) Vempo si kuh-kuhti-ma-machi

3pl emph red-angry-red-seem
‘They really seem like really hateful people.’
Head-marking inflection and the architecture of grammatical theory 
3.2 Reduplication and verbal affixes

We see the same word-internal, head-reduplication pattern on certain affixes, where
reduplication occurs between a verb stem and a derivational suffix. In these cases, it
is the suffix which receives the reduplication, despite the suffix being a bound ele-
ment which otherwise does not occur as an independent verb in the language. Some
examples are presented in (22):
(22) a. Directive (-sae)(Escalante 1990, p. 78 [41])

inepo a=nok-sae → inepo a=nok-sas-sae
1sg 3s=talk-dir 1sg talk-red-dir
‘I am telling him to speak up’ ‘I tell him to speak up’
b. Desiderative (-’ii’aa)(Escalante 1990, p. 78 [42])
inepo a=nok-’ii’aa → inepo a=nok-’ii-’ii’aa
1sg 3s=talk-desid 1sg 3sg=talk-red-desid
‘I want him to talk’ ‘I would like him to talk (more)’
c. Inceptive (-taite)(Escalante 1990, p. 79 [43])
aapo nok-taite → aapo nok-tai-taite
3sg talk-incep 3sg talk-red-incep
‘She is starting to talk’ ‘S/he repeatedly starts to talk (hesitates)’
d. Prospective (-vae)(Escalante 1990 p. 79 [44])
aapo nok-vae → aapo nok-vav-vae
3sg talk-prosp 3sg talk-red-prosp
‘He feels like talking’ ‘From time to time he gets the urge to talk’
‘S/he gets the desire to talk a lot’
These verbal affixes are always bound, and cannot appear as independent verbs at all,
at least in the synchronic grammar. That such affixes may not occur independently as
the main verb in a sentence is illustrated for a subset of them in (23) below:
(23) a. *Inepo apo’ik ii’aa

1.sg 3.sg.acc want.
Intended reading: ‘I want it.’
b. *Aapo taite-k.
3.sg begin-perf
Intended reading: ‘S/he began.’
c. *Aapo a=vae.
3.sg 3.sg =prosp
‘Intended reading: ‘He’ll do it/He feels like doing it.’
When reduplication applies to such suffixes, as above, it takes scope over the whole
complex verb form. In (24a), for example, the meaning is one of habitual wanting-
him-to-talk, not one about habitual talking. If the speaker wishes to indicate habitual
 Jason D. Haugen & Heidi Harley
semantics for just the leftmost member of the compound, reduplication applies to that
leftmost member and takes scope only over the lower verb; in (24b), the speaker wants
him to habitually talk, but doesn’t habitually want anything. If the situation warrants
it, habitual semantics and reduplication can apply to both the stem and the suffix of
the complex verb form (24c):
(24) a. Inepo aa=nok-ii-’ii’aa ne vetchi’ivo
I him=speak-red-want me for
‘I always want him to speak for me.’
b. Inepo aa=no-nok-ii’aa
I him=red-speak-want
‘I want him to be the speaker/the one who habitually speaks.’
[e.g. at a council meeting]
c. Inepo aa=no-nok-ii-’ii’aa
I him=red-speak-red-want
‘I always want him to be the speaker.’
It is worth noting, however, that not all derivational verbal affixes can reduplicate.
There are some affixes that only allow reduplication to apply to the verbal stem. These
include the applicative -ria (25a), the causative -tua (25b), and a morpheme (-te) that
derives verbs of creation from nominal stems (25c):
(25) a. Applicative (-ria)
lu-luuta-ria *luuta-ri-ria
red-finish-appl
‘(habitually) use up on someone’
b. Causative (-tua)
mahhai-tua *mahai-tu-tua
“red”(μ-infix)-afraid-caus
‘really scare someone’
c. make (-te)
hi-hipe-te *hipe-te-te
red-mat-make
‘(habitually) make mats’
We return to these in Section 5.2 below in which we discuss available theoretical

treatments.
In sum, many complex verbs in Hiaki can exhibit word-internal reduplication, as
with the N–V compounds illustrated above. However, unlike the complex verbs cre-
ated by noun incorporation, reduplication of verb stems with certain verbal suffixes
allow three possibilities for reduplication, with the reduplication taking scope over the
morphological target: i.e. the verb stem, the suffix, or both.
Head-marking inflection and the architecture of grammatical theory 
3.3 Word-internal reduplication in pseudo-compounds

There are also a variety of forms which display the pattern of internal reduplication,
but which do not involve otherwise free nominal or verbal stems. In fact, the syn-
chronic internal make-up of these forms is not known to us, since we only find the
morphological elements in these “pseudo-compound” contexts. In other words, these
constructions are composed of cran-morphs. A list of such pseudo-compounds is
given in (26), along with the reduplicated form, and in some cases, a possible etymo-
logical connection:4
(26) Verb Meaning Reduplicated Form Possible etymology
a. bwah-suma ‘braid’ bwah-su-suma <suma ‘tie’?
b. bwal-wotte ‘feel weak’ bwal-wot-wotte <bwala ‘sheep’?
c. bwal-wotta ‘make to feel bwal-wot-wotta <bwala ‘sheep’?
weak’
d. chiki-pona ‘tickling chiki-po-pona <poon a ‘strike,
someone’ knock’
e. le-siki-le ‘itching, ele-si-sikile <siki ‘red’?
tickles’
f. ha’a-chih-te ‘sneezing’ ha’a-chih-chihte <chitei ‘mash’?
g. haawahsa’a-te ‘steaming’ haa-wa-wahsa’ate <haawa ‘steam’
h. hun-hiawa ‘make fun hun-hi-hiawa
of, tease’
i. hu’u-nakte ‘created hu’u-na-nakte
deliberately’
j. iva’a-chaka ‘embrace, iva’a-cha-cha’e <iva’a ‘hug’?
hugging’
k. iva’a-nama ‘cradling, iva’a-na-nama <iva’a ‘hug’?
embracing’
l. kuhtiachi ‘Hateful, mean, kuh-ti-tiachi
awful person
or animal’
. While the majority of these forms may only reduplicate on the rightmost member, with
left-edge reduplication ruled flatly ungrammatical, our consultant felt that a couple of these
forms could also accept reduplication at the edge when prompted, e.g. (26h). However, there
was no semantic differentiation between the two reduplicated forms (as expected given their
semantic opacity) and the most natural, spontaneously produced form was always the internal
reduplication of the rightmost member. One possibility is that those which accept leftmost
edge reduplication have a V–V compound as their historical source (rather than an N–V com-
pound), although etymological information is not available which would allow us to confirm
this hypothesis.
 Jason D. Haugen & Heidi Harley
m. kutsaite ‘dusk, early kut-sai-saite <kut- ‘dark’?

evening’
n. machu’unama ‘hold in machu’u-na-nama
hands, grasp’
o. maukaroa ‘rise in the mau-ka-karoa
early dawn or
before dawn’
p. naamuke ‘drunk, dizzy’ naa-mu-muke <naamu ‘cloud’?
q. tekipanoa ‘work’ teki-pa-panoa <tekil- ‘job’?
Reduplication in these pseudo-compounds appears to target an internal head, but our

native speaker consultants, queried about the possible sources of these expressions,
indicated that they were fixed forms with no internal analysis. When further asked
about the potential partial etymologies noted above (which we proposed based on
phonological relatedness and semantic plausibility) they explicitly denied the related-
ness of the forms.
It is important to note that while most of the words in (26) are composed of three
or four syllables, this kind of internal reduplication is not required for all words above
two syllables; i.e. this internal reduplication is not strictly phonological. Verbs ending
in the derivational suffixes illustrated in (25) above frequently reach lengths of 4–5
syllables but nonetheless reduplicate on the left edge; some other trisyllabic forms with
initial reduplication are illustrated in (27):
(27) a. Itepo aman si hu-hu’uwaasu

3.pl there very red-freeze
‘Over there we really freeze!’
b. A’apo siime taewata hi-hiwihte
3.sg always day red-saw
‘S/he saws (wood) all day long.’
c. Uu uusi muni-m chive-chivehta
The child bean-pl red-spread.out
‘The child is spreading out the beans.’
As shown in (27), the initial reduplicant can vary between one syllable and two
syllables, which for the most part is a stem-specific lexical choice.
With the above considerations in mind, we conclude that the internal reduplica-
tion in the examples in (26) indicates some kind of complex morphological structure,
where the reduplication process apparently targets the head of an opaque compound.
As in the NI cases from the beginning of Section 3, the edge of the phonological
word/derived stem is not itself the domain for the attachment of the reduplicative
morpheme.
Head-marking inflection and the architecture of grammatical theory 
3.4 Interim conclusion: Head-marking

Above we have seen the interaction between reduplication and a variety of compound-
ing and derivational processes creating complex verb forms in Hiaki. In each case, we
have seen that reduplication targets the rightmost member of the complex verb. We
propose that this internal inflection is a type of head-marking, in which the redu-
plicative affix applies to the head of a complex morphological form, even within a
single phonological word. We now turn to consider the theoretical implications of this
phenomenon, as well as discussions of similar data in other languages.
4. Implications for standard theoretical architectures
The empirical evidence surveyed above represents a case in which lexical integrity
appears to be violated. The Hiaki compound and derived forms we have surveyed are
indubitably single phonological words, exhibiting many characteristics which both
language-internally and cross-linguistically are canonical hallmarks of wordhood.5
Nonetheless, a regular inflectional process appears to target a subconstituent within
these derived forms – a process which would normally target the left edge of a verb,
but which in this case targets the left edge of the rightmost element in a derived verb.
This results in an inflectional affix which intervenes between the components of the
derived form.
This situation is not unique to Hiaki, of course. Similar facts are observed for
inflectional reduplication in four languages that we know of. In two other Uto-Aztecan
languages, Hopi and Classical Nahuatl, examples analogous to the Hiaki cases are
attested, as illustrated in (28). Example (28a) shows reduplication inside a nominal-
ized N–V compound from Classical Nahuatl; (28b) shows reduplication on the head
of a V–V compound from Hopi:
(28) a. ixmjmjqujnj
Ø-ix-mi-miqui-ni
3.sg.subj-eye-red-die-ag.nom
‘It is one which is blinded (by strong light)’ (lit. ‘It is one whose eyes
die’), (re: the gopher/toçan)
(Classical Nahuatl, Florentine Codex, Book 11, p. 16)
. There is a large literature on the issue of “wordhood”. With respect to the notion ‘phono-
logical word’, see the review in Hall 1999. For relevant discussion about the relationship of the
phonological word to the lexeme and/or syntactic terminal ‘word’, see among many others such
works as Haspelmath 2011; Aronoff 1976; Lieber 1980; Williams 1981; Selkirk 1981; Farmer
1980; Lapointe 1980, 1981; Newmeyer 1986; DiSciullo & Williams 1987; Carstairs-McCarthy
1992; and Marantz 1997.
 Jason D. Haugen & Heidi Harley
b. lavay-ho-honaq-lawu
talk-red-be.erratic-cont
‘He was just jabbering’ (Hopi, Hopi Dictionary 1998, p. 202)
Both of these language also allow internal reduplication of certain suffixal heads of
complex forms:
(29) a. Auh in jtlaqual mjchtepitzitzin mjchcocone
auh in i-tlaqual mich-tepi-tzi-tzin mich-co-cone
and det its-food fish-offspring-red-rev fish-red- child
‘(and) its food is small fish, baby fish’ (re: the tlacamichin, a type of fish)
(Classical Nahuatl, Florentine Codex, Book, 11 p. 58)
b. ööqa-to-to-yna
bone-red-caus-caus
‘She fastens a new stem [in a basket]’
(Hopi, Hopi Dictionary 1998, p. 362)
Outside Uto-Aztecan, similar instances of reduplication appearing inside derived

forms are analyzed for Sanskrit by Stump (2001); similar cases are discussed from
Bahasa Indonesia by Sato (2010). A Sanskrit example is presented in (30):
(30) pary-a-da-dhat < pari-dha- ‘put around’
around-pret.aug-red-put
‘S/he was putting (it) around (something)’
(Stump 2001, p. 110; our translation)
These internal reduplication cases are a sub-case of the larger phenomenon of internal
inflectional head-marking of derivationally complex forms, which, although uncom-
mon, are far from unattested cross-linguistically; see the extensive documentation in
Stump (2001, pp. 96–137), as well as related discussion by Harris (2000, 2002), Sato
(2010), and others. Here, we will specifically consider the theoretical implications of
the Hiaki data described above, but many of the issues raised will of course be relevant
to the analysis of head-marking generally; we do not, however, propose to consider the
whole range of head-marking facts here.
First we explore the possibilities within established standard frameworks, both
Lexicalist and syntacticocentric, and conclude that neither family of approaches can
implement head-marking without supplementation by additional mechanisms.6 We
. We will mainly discuss theories for which we can find specific proposals dealing with the
relationship of inflection, derivation and compounding in the architecture. Lieber and Scalise
(2006) propose a revised Lexicalist architecture to allow the morphology limited access to the
output of phrasal syntax, but we are unclear on whether their architecture implements any
strict restrictions on the interaction of derivational and inflectional processes, so we do not
Head-marking inflection and the architecture of grammatical theory 
then go on to consider what such supplementation might consist of for each approach,
reviewing the proposal of Stump (2001) within a word-based approach, and exploiting
the mechanisms available in Distributed Morphology to propose an account within a
syntacticocentric framework.
We will assume below, without argument at this point, that reduplication involves
the affixation of a morpheme (“RED”) that triggers some kind of copying process on
a stem. In this we follow many others working within a variety of different theoretical
frameworks: e.g. Moravcsik (1978); Marantz (1982); McCarthy and Prince (1986, 1993,
1995); etc. Other views are possible, e.g. reduplication occurring as a phonological rule
(e.g. Aronoff 1976; Raimy 2000; Frampton 2009). For a defense of the piece-based
view of reduplication over process-based views, see Haugen (2008), among others. We
think that our ultimate conclusions do not ride on this particular assumption; see our
further discussion in Section 6 below for what is ultimately at stake, and how theories
such as those proposed by Raimy (2000) and Frampton (2009) could be compatible
with the analyses that we present in the sections to follow.
4.1 Weak Lexicalist architectures

In Lexicalist theories (e.g. Sapir 1911; Mithun 1984, 1986; DiSciullo & Williams 1987;
Rosen 1989; Mithun & Corbett 1999), NI is the morphological (“lexical” or pre-
syntactic) process of compounding a nominal root onto a verbal stem. Under such
a view, this morphological process creates a noun-verb compound that names some
culturally-significant concept and which is stored as its own listeme in the lexicon (see
e.g. Mithun & Corbett 1999, p. 68).
Given a division of labor between morphology and syntax like that of Aronoff
(1976) or Anderson (1982), where derivation is lexical and inflection is syntactic (the
“Weak Lexicalist Hypothesis”), the conception of NI as a pre-syntactic lexical process
makes a strong prediction about how NI should interact with an inflectional process of
reduplication: this inflectional process, like others, should not be able to operate inside
this compound. The Weak Lexicalist Hypothesis derives peripherality of inflectional
affixes as a natural consequence, and thus predicts that inflectional reduplication will
apply to the edge of a compound. Under the Weak Lexicalist Hypothesis, reduplica-
tion of NI structures should only operate on the edges of a compound stem (i.e. on
the left-edge for prefixes, or on the right-edge for suffixes), rather than targeting some
sub-constituent of the N–V compound (e.g. the noun or the verb only). This theory
thus incorrectly predicts that Hiaki NI will yield *ma-maso-peute ‘*red-deer-butcher’
rather than the actual attested form, maso-peu-peute ‘deer-red-butcher’.
discuss it here. Insofar as Lieber and Scalise’s approach is strongly Lexicalist, as is DiSciullo
and Williams (1987), the remarks about the latter may be relevant to the former as well.
 Jason D. Haugen & Heidi Harley
One language that potentially illustrates this expected interaction of reduplica-

tion with NI is Paiwan, an Austronesian language spoken in Taiwan. Chang and Wu
(2006) argue that Paiwan has two types of NI, lexical and syntactic, which are identi-
fied by two independent diagnostics. Incorporated nominals undergoing lexical NI do
not have case-markers, and such noun-verb compounds can undergo reduplication
(31). In syntactic NI, on the other hand, case-markers are incorporated with the incor-
porated nominal, and the noun-verb compound cannot undergo reduplication (32).7
Both of these diagnostics make the first type look like a lexical process: case-inflection
cannot occur inside the N–V compound, cf. (31b), and the process of reduplication
appears to target the edge of the compound, cf. (31b) and (33):
(31) a. s〈em〉a-’uma=aken
go.to-〈af〉-home=1.sg.nom
‘I went home’
b. s〈em〉a-’uma-’uma=aken
go.to〈af〉-home-red=1.sg.nom
‘I am going home’
(32) a. s〈em〉a-tjaj-palang=aken
go.to〈af〉-obl-Palang=1.sg.nom
‘I went to Palang’s place’
b. *s〈em〉a-tjaj-palang-*palang =aken
go.to〈af〉-obl-Palang-*red=1.sg.nom
‘I am going to Palang’s place’
(33) s〈em〉a-pana-pana=aken
go.to〈af〉-river-red=1.sg.nom
‘I am going to the riverbank’
An important observation about the data provided by Chung and Wu (2006) is that
the reduplicant seems to always be co-extensive with the nominal root, which in turn
seems to always be disyllabic. Chung and Wu do not present cases of monosyllabic
nominal roots, with which we might expect to see the reduplication of the monosyl-
labic nominal and the last syllable of the verb (assuming that the progressive redu-
plicant in Paiwan is consistently a disyllabic foot). For example, with a hypothetical
nominal stem ba in the present progressive motion constructions above, we would
expect a reduplicated form something along the lines of (34), where underlining indi-
cates the hypothetical nominal stem, and where brackets indicate the portion of the
compound verbal stem that is copied by the reduplicant (in bold):
. An unusual trait of NI in Paiwan is that it only occurs with spatial verbs (e.g. verbs of
motion or location). Reduplication that occurs with such verbs indicates progressive aspect,
and appears as a suffix.
Head-marking inflection and the architecture of grammatical theory 
(34) Hypothetical Paiwan Monosyllabic Nominal Reduplication

s〈e[m〉a-ba]-maba=aken
‘I am going to 〈whatever is denoted by √ba〉)
Forms like the hypothetical example in (34) would provide strong evidence in favor
of the view that the N–V compound behaves as a unified stem that gets inflected after
syntax, therefore supporting the idea that it was created prior to syntax, in the lexicon.
In the absence of such data, however, the possibility that reduplication in Paiwan is
only targeting the nominal root cannot yet be ruled out (though its progressive mean-
ing implies that it must at least semantically apply to an eventive verbal concept, not a
stative nominal one).8 In any case, the interaction of reduplication with NI in Paiwan
looks like a reasonable candidate for a Weak Lexicalist analysis.
On the other hand, the interaction of reduplication and NI in Hiaki, as noted
above, is a problem for a theory which predicts edge-marking as a consequence of
its architecture. In order for this critique to have teeth, it is particularly important to
establish that the Hiaki derived forms which exhibit head-marking do count as ‘lexical’
in the relevant sense. We therefore pause here to comment on two established crite-
ria for distinguishing lexical processes from syntactic ones: (non-)productivity (4.1.1)
and non-compositionality (4.1.2).
4.1.1 Productivity
Smirniotopoulos and Joseph (1998) “take the absence of (a high degree of) produc-
tivity as a clear indicator of a lexical rule” (p. 452, emphasis in original). On these
grounds, Hiaki NI should clearly be considered to be a lexical process, since N–V
compounding in that language is relatively restricted (as noted by, e.g. Jelinek, p.c.),
being not nearly as productive as in some other Uto-Aztecan languages, e.g. Nahuatl
(Merlan 1976) or Hopi (K. Hill 2003).
In contrast, the reduplication patterns that we see in Hiaki compound verbs
conform to the otherwise regular and productive morpho-phonological patterns of
prefixal reduplication in the language, in terms of both form and function. In terms
of function, reduplication in Hiaki (in compound verbs and otherwise) typically has a
habitual aspectual meaning, which is usually characterized as verbal inflection, cross-
linguistically as well as in Hiaki itself. Like other inflectional processes, reduplication
. Of course, just because the stem to which the RED affix attaches morphosyntactically
is verbal, it need not be the case that the entire stem with its internal morphological com-
plexity is therefore defined as the Base for the morphophonological process of reduplication
which the RED affix triggers. We can distinguish between the “Target”, i.e. the stem to which
the RED affix attaches, and the “Base”, i.e. that potential sub-set portion of the stem which
is delimited as the morphophonological sub-domain marked as available for copying (for
further d iscussion, see Haugen 2008).
 Jason D. Haugen & Heidi Harley
is fully productive, and it can be applied to any verb that is compatible with the result-
ing meaning. In terms of form, reduplication in Hiaki usually involves syllabic redu-
plication (Haugen 2003, 2008; Harley & Leyva 2009), e.g. kupi-tom-tom.te (15f) vs.
lioh-bwa-bwa.ni.a (15g).9 In addition, we also see distinct patterns of reduplicative
allomorphy, e.g. morphological gemination (cf. 17). This kind of allomorphy is also
attested with other verb forms, and it appears to be the case that which reduplicative
morpheme (or dupleme in the terminology of Spaelti 1997) goes with which verb stem
is unpredictable and often must be lexically listed (Haugen 2003, 2008; Harley & Leyva
2009); this is similar to other irregular inflectional processes cross-linguistically.
By the productivity criterion, then, Hiaki NI, which is not productive, is clearly
lexical, and reduplication, which is productive, is clearly inflectional.
4.1.2 Compositionality
Similarly, Smirniotopoulos and Joseph (1998) write, “The output of a syntactic rule
should show compositional semantics, so that the meaning of the whole is composed
from the meaning of its parts. By contrast, the output of a lexical rule can be non-
compositional in its semantics and thus can show meanings that differ in ways that
are unpredictable in relation to the meanings of the individual parts composing it”
(p. 452). On this criterion, too, these Hiaki compound forms which exhibit head-
marking, are clearly noncompositional; as we have shown above, many of them have
idiomatic meanings – see, for example, (15d), (15e), and (15f) above – and some are
even composed entirely of cran-morphs, whose sub-parts do not contribute any detect-
able independent meaning to the meaning of the whole, as illustrated in (26) above.10
This point is also made by Dedrick and Casad (1999, p. 161) in their d iscussion of
lexicalization and non-predicability of meaning in N–V compounds.
In contrast, the semantic contribution of reduplication to these compound forms
is entirely consistent with the contribution of reduplication to verb forms elsewhere in
the language. As emphasized in Section 2.1, the pluractional semantics of reduplica-
tion on verbs is identical in compound and non-compound forms, with three primary
interpretations, the most common being that of habitual aspect. There is no significant
idiomaticity involved in interpreting productively reduplicated verbs in the language.
Hiaki noun incorporation, then, is clearly lexical by the compositionality criterion
as well, while reduplication is just as clearly inflectional.
. Just as is also the case with other instances of verbal reduplication in Hiaki, we never see
copy into the second syllable to create a coda for syllabic reduplication; e.g. *li.oh.-bwan-bwa.nia.
. Anticipating some discussion below, we would like to point out here that much recent
work within Distributed Morphology (DM) does not take the notion of (non)composition-
ality (nor that of (non)productivity) to be indicative of syntactic vs. lexical processes (see e.g.
Marantz 1997).
Head-marking inflection and the architecture of grammatical theory 
4.1.3 Interim conclusion for weakly lexicalist frameworks

In sum, theories involving a strict division between lexical derivational processes and
syntactic inflectional processes would strongly predict edge-marking behavior, in con-
trast to the facts described above. We conclude that without supplemental mechanisms
such frameworks are not equipped to account for the interaction of reduplication and
noun incorporation in Hiaki.
4.2 Syntacticocentric architectures: Baker (1988) on noun incorporation

We now turn to consider the predictions concerning the interaction of incorporation
and reduplication made in a basic syntactic analysis of incorporation, like that initially
proposed by Baker (1988). Baker argued that noun incorporation was simply syntactic
head-movement, by which the internal object nominal head-adjoins to the verb that
is its sister, and is then carried with the verb through the syntactic tree in any fur-
ther head-movements in which the verb may participate. In such an approach, head-
movement is assumed to create a syntactically complex head which corresponds to a
single phonological word at Spell-Out. What would such a theory predict concerning
the interaction of reduplication and noun incorporation?
If the incorporated element, when nominal, originates in the object position (i.e.
sister to the verb, as proposed by Baker), it will be closer to the verb, structurally, than
any higher functional morpheme, such as we assume RED to be, since it is an aspec-
tual morpheme denoting a plurality of eventualities. Thus, RED likely heads an Aspect
projection in the inflectional complex. If the usual incorporation via head-to-head
movement is the only mechanism assumed (al a Baker 1988), it will be impossible to
situate RED linearly within the N–V compound which adjoins to it. The problem for
a entirely syntactic analysis using only head-to-head movement is illustrated in (35)
below, using our original example of reduplication with NI in (13) above:
(35) a. TP
DP T′
AspP T0
VP Asp0
NP V V0 Asp0+hab
N0 V0
Aapo t t maso- peute -RED Ø

He deer- butcher-RED Ø
 Jason D. Haugen & Heidi Harley
Here, the habitual Asp0 head, which is spelled out by RED, forms part of the
complex verbal head resulting from syntactic head-to-head movement. The sister
to Asp0 is the complex N–V constituent formed by incorporation of the N0 into the
V0. If RED is morphophonologically specified to be a prefix, and hence is ‘flipped’
to appear to the left of its sister at linearization (or alternatively simply triggers
right-adjunction of the N0–V0 complex to Asp0), it should be prefixed to that com-
plete N–V constituent. On that analysis, barring further assumptions, RED would
not be able to intervene between the verb and its incorporated object under the
head-movement analysis. That is, just as is the case in the weakly lexicalist account,
the basic syntacticocentric approach predicts edge-marking inflection for Hiaki
inflectional reduplication. This is clearly inconsistent with the facts shown above,
and we conclude again that such a framework, absent supplemental mechanisms, is
also unable to account for the interaction of noun incorporation and reduplication
in Hiaki.
4.3 Strong lexicalist architectures: DiSciullo and Williams (1987)

Interestingly, the strongly Lexicalist framework articulated by DiSciullo and
Williams (1987) provides the tools to accommodate head-marking phenomena
within compounding more easily than the frameworks described above. In contrast
to the Weak Lexicalist architecture described in Section 4.1, in the strong lexical-
ist architecture all word-formation operations, both inflectional and derivational,
are contained within a single module of the grammar. This allows the framework
to more easily accommodate apparent interleaving of these processes; indeed, the
apparent inflection/derivation distinction is argued to be epiphenomenal within
the framework.
DiSciullo and Williams (1987, p. 25) write, “The real generalization about
inflectional affixes is that they must appear in head position, not that they must appear
‘outside’ all other word formation – the latter is partly a consequence of the former,
although there are cases in which the former holds but the latter does not.” The Hiaki
data would appear to be one such case. Let us consider how it might be treated in this
framework.
In the Strong Lexicalist architecture proposed by DiSciullo and Williams
(1987), compounding involves the creation of a morphological unit that derives
its features from the percolation of the features of the head. For example, consider
the underlying structures for the meaning contrast inherent to the English
compounds part-supplier vs. parts-supplier (36a–b), on the one hand, and parts-
supplier vs. parts-suppliers (36c–d), on the other (following Di Sciullo & W illiams
1987: 24–5):
Head-marking inflection and the architecture of grammatical theory 
(36) a. N[sg] b. N[pl]
N[sg] N[sg] N[sg] N[pl]

part supplier part suppliers
[partsg-suppliersg]sg vs. [partsg-supplierspl]pl
c. N[sg] d. N[pl]
N[pl] N[sg] N[pl] N[pl]

parts supplier part suppliers
[partspl-suppliersg]sg vs. [partpl-supplierspl]pl
Fully inflected nominals, already bearing number features, are the input to the com-
pounding process; this is possible because of the non-segregated nature of the lexical
module in this architecture. Verbal inflection will agree with the number of the entire
compound, based on the number inherited from the head of the compound. In no case
should a verb be able to “see into” the compound to note that there is a second nominal
with a potentially conflicting number specification; however, such inflection-within-
derivation is perfectly legitimate in the framework. For the purposes of the syntax, the
feature specifications for the non-head of the compound are completely irrelevant.
We can see how such an analysis could approach the case of reduplication in
N–V compounding in Hiaki. The rightmost component of the compound, the verb,
would enter the compound already inflected for habitual aspect, i.e. reduplicated; the
non-head would then compound with the inflected verbal head to produce the head-
marked compound V0. The aspectual features of the head would then determine the
aspectual properties of the whole by percolation, appropriately.
(37) a. Step 1: Inflection for habitual aspect: b. Step 2: N–V compounding:
V[+Hab] V[+Hab]
Asp[+Hab] V0 N V[+Hab]
RED- sua kuchu susua
(su-) kill.PL.OBJ fish RED-kill.PL.OBJ
Despite the ability of the framework to generate the relevant form, however, we see at
least two significant problems with the approach. First, the framework does nothing to
prevent the non-head element in a compound from exhibiting inflectional affixation,
 Jason D. Haugen & Heidi Harley
thereby permitting the generation of examples like parts-supplier or parks commis-

sioner.11 Such affixation cannot affect the featural properties of the whole compounded
word, of course, since it is on the non-head element, but no mechanism for outlaw-
ing such affixation is provided or, presumably, desired. This leaves us without an
account for the strictly uninflected character of the left-hand member of the Hiaki
N–V compounds: incorporated forms like *kuchu-m-sua ‘fish-pl.obj-kill’ are always
ungrammatical.
The second question that the Hiaki data raise for the DiSciullo and W illiams
approach has to do with the fact that reduplication is prefixal in Hiaki, and yet
it behaves as a head, contributing its features to the complex form in which it is
contained. This contravenes the Right-Hand Head Rule of both Williams (1981) and
DiSciullo and Williams (1987, p. 26 & 81). We take the Right-Hand Head hypothesis
to be c ounter-exemplified by the Hiaki data as well as numerous other cases from lan-
guages around the world, but we will not consider the implications of this further here.
4.3.1 Interim conclusion for standard architectures

In this section we have argued that two standard theoretical architectures, the Weak
Lexicalist approach and the syntactic approach of Baker (1988), are unable to appro-
priately place Hiaki aspectual reduplication inside of the word formed by NI and other
compounding processes. These architectures would need some kind of supplemental
machinery to account for these phenomena. In the next section we will discuss some
architectural accommodations that have been proposed which can better model the
Hiaki data.
The Strong Lexicalist architecture designed by DiSciullo and Williams (1987),
on the other hand, is able to account for these data since it does not posit a strict
separation of derivational and inflectional processes. However, the Hiaki data raise
other problematic issues for the Strong Lexicalist Hypothesis: namely, that inflec-
tion is strictly forbidden on the non-head member of compounds in Hiaki, and that
reduplication can form a Left-Hand head. Neither of these facts seem to accord with
the predictions made by the Strong Lexicalist theory proposed by DiSciullo and
Williams.
. DiSciullo and Williams (1987) do not, apparently, regard examples such as *choirs-boy or
*rats-eater to be ungrammatical, as their framework allows the free generation of such forms.
However, a significant literature on when and why inflectional marking is (im)possible within
English compounds, beginning with the level-ordering work of Kiparsky (1982), has been
concerned with exactly how to rule out such cases, which seem flatly ungrammatical to most
English speakers. It is perhaps worth noting that the Kiparsky level-ordering framework faces
the same architectural issues as the Weak Lexicalist frameworks do with respect to the Hiaki
data, since it posits a strict ordering between earlier level processes (such as compounding)
and later inflectional processes (such as Hiaki aspectual reduplication).
Head-marking inflection and the architecture of grammatical theory 
5. Architectural accommodations
Both weakly-lexicalist word-based models and syntax-only models, then, face prob-
lems in coping with head-marking reduplication inside N–V and V–V compounds
in Hiaki. Below, we review a proposal within the word-based Paradigm-Function
Morphology (PFM) framework by Stump (2001) to accommodate similar cases in
Sanskrit (§5.1); he introduces a distinction between Root-Root compounding and
Word-Word compounding, which enables him to account for both edge-marking and
head-marking inflectional behaviors. We then turn to consider what type of supple-
mentation is needed in a syntax-based model to account for the Hiaki data above
(§5.2), proposing that the operation of Local Dislocation (Embick & Noyer’s (2007)
updated implementation of Marantz’s 1984 Morphological Merger) can provide the
necessary tools to capture the patterns observed.
5.1 Word-based approaches: Stump (2001) and PFM

The most developed word-based approach to inflectional morphology is the g eneral
family of Word and Paradigm (WP) models. Several alternative versions of such
models have been proposed, including prominent proposals by Mathews (1972) and
Anderson (1992). We will consider here the more recent implementation developed by
Stump (2001), Paradigm Function Morphology (PFM), which has an advantage over
previous WP theories in that it provides a more satisfactory account for the morpho-
logical inflection of heads than earlier theories.
In PFM, word-forms are derived through the application of rules in a p aradigm
function to a lexeme, which generates inflected forms for the lexeme correspond-
ing to each cell in a language’s paradigm space. Stump introduces a three-way
distinction between the types of rules which produce derived lexemes. The head-
marking behavior (or lack thereof) in the inflected forms of a paradigm function
applied to a given derived lexeme depends on the particular type of rule which
produced the lexeme in the first place. “Word-to-word” rules derive lexemes
which exhibit head-marking; “Root-to-Root” rules derive lexemes which exhibit
edge-marking. A third type, “Word-to-Stem” rules, produce lexemes which exhibit
double marking (i.e. the inflectional rules apply to both the head and the edge of
the derived lexeme).12
. These two patterns suggest a potentially larger typology of inflectional patterns: if edges
and heads can each individually be marked, and they can also be marked simultaneously, one
might in addition expect to see cases where either head, edge, or both could be marked for
inflection. Such a case may be illustrated by the Pima reduplication data presented by Munro
and Riggle (2004). In Pima nominal compounds marked for plural, reduplication can appar-
ently target any member of the compound or any subset of members of the compound, up to
 Jason D. Haugen & Heidi Harley
An example of a Root-to-Root derivative given in Stewart and Stump (2007,

p. 407) are the Breton forms in -ad ’-ful’; when a form like ti-ad, ‘house-ful’, is inflected
for plural, the plural rule applies to the edge, giving ti-ad-où, ‘house-ful-pl’, rather
than to the head, which would produce *tiez-ad, ‘house.pl-ful’. In contrast, Sanskrit
preverb-verb compounding is a Word-to-Word rule. Hence after a preverb such as vi-
‘away’ is compounded with a verb like gam, ‘go’, to produce vi-gam, the derivative will
exhibit head-marking behavior, so, e.g. the prefixal augment a- as well as tense/aspect
inflection is attached to the head verb -gam, giving vi-a-gacchat ‘s/he goes away’.
The intervention of the inflectional augment prefix inside the complex derivative is
predicted by the fact that it was produced by a Word-to-Word rule.
With respect to inflectional reduplication, PFM thus predicts the pattern seen in
Example (30) above, repeated below for convenience:
(38) pary-a-da-dhat < pari-dha- ‘put around’
around-pret.aug-red-put
(Stump 2001, p. 110; our translation)
Here, within a preverb-verb compound, prefixal reduplication applies to the verbal
head, intervening between the preverb and the verb along with the prefixal augment
a-. The Hiaki data are equally amenable to this approach. This analysis would entail
that the N–V compounds, the V–V compounds, and the V-suffix derived forms which
exhibit head-internal reduplication are all produced by Word-to-Word rules; their
head-marking behavior with inflectional reduplication would then be expected.
While the PFM approach can indeed account for the basic patterns described
above, certain of Stump’s claims about the properties of head-marking within his
framework may be called into question when additional data from Hiaki are con-
sidered. A central plank in the theoretical platform Stump espouses is the Paradigm
Uniformity Generalization (PUG), according to which “head marking is an all-or-
none phenomenon: If a root ever exhibits head marking in its inflectional paradigm, it
always does” (Stump 2001, p. 109).
and including each member of the compound. A 5-stem nominal compound like that illus-
trated in (i) below has, in theory, 31 possible plural forms. The form illustrated below shows
the case in which all stems receive the inflectional marking:
(i) [lil-mìmida] -hoahas-hàha’a] -[dádagkuanakud:]

[glass] -[baskety-jar] -[wiper]
‘glass dish cloth’ (Pima, Munro & Riggle 2004, [18])
Interestingly, there are no scope effects for plural marking in Pima. Rather, there is free
variation: reduplication of any or all of the stems in the compound makes the entire compound
plural. We will not discuss this pattern further here except to note that it is not clear to us how
the PFM account would extend to these facts.
Head-marking inflection and the architecture of grammatical theory 
For example, the inflectional paradigm function for the Sanskrit verb above speci-
fies both inflectional reduplication and inflectional prefixation. Due to the PUG, both
of those processes must apply to the verb which is the head of the complex lexeme.
The PUG predicts that a form behaves uniformly with respect to all inflectional rules:
one rule cannot head-mark while another edge-marks. Stump (2001, p. 133) is help-
fully explicit about the type of evidence which would genuinely disconfirm the PUG.
Such potential counterevidence is offered under the guise of Pseudo-Sanskrit, in which
reduplication targets the verb while the augment a- targets the left edge of the complex
form. If the PUG is a valid universal generalization about word-formation, and spe-
cifically the structure of inflectional paradigms, then this kind of disparate inflectional
marking should be impossible. Stump illustrates this hypothetical illicit pattern with
the following form:
(39) “*Pseudo-Sanskrit”
*a-pary-da-dhat < pari-dha- ‘put around’
pret.aug-around-red-put
In fact, however, Hiaki compound verbs containing object clitics provide a case
essentially identical to the Pseudo-Sanskrit counterexample above. In Hiaki, object
clitics are inflectional elements which must appear prefixed to the main verb of the
clause in which they occur, as illustrated in (40a). They may not be separated from
the main verb by any nonverbal material (40b), even particles which are important to
the entire predicate’s meaning and which otherwise must appear adjacent to the main
verb themselves (40c) (see discussion in Dedrick & Casad 1999, p. 269).
(40) a. Vempo aman aa=vicha-k

3.pl.nom there 3.sg.obj =see-perf
‘They saw him/her over there.’
b. *Vempo aa=aman vicha-k
3.pl.nom 3.sg.obj =there see-perf
c. nat am=totta-ka
on.top 3.pl.obj=pile-ppl
‘piling them on top of one another’ (Dedrick & Casad 1999, p. 271)
The object clitics, then, are inflectional elements which prefix to their verb, just like
reduplication does; when attached to a non-compound verb form, both appear as
prefixes to the verb stem, as expected:
(41) Kat=ee unna kusisi aa=‘e-‘eta

neg.imp=2.sg.nom too loudly 3.sg.obj =red-close
‘Don’t close it too loudly!’
 Jason D. Haugen & Heidi Harley
However, when reduplication and an object clitic are attached simultaneously

to a compound verb, they behave exactly like Pseudo-Sanskrit: i.e. the object
clitic attaches to the left edge of the complex form while reduplication targets the
head. We saw this pattern illustrated above in Example (16), which we repeat for
convenience below as (42). The grammatical form in (42a) is equivalent to the
Pseudo-Sanskrit example above. (42b) would be expected in a uniformly edge-
marking Pseudo-Hiaki which conformed to the PUG; (42c) would be expected in
a uniformly head-marking Pseudo-Hiaki, also conforming to the PUG; both are
flatly ungrammatical.
(42) a. Irene am=pan-ho-hoo-ria

Irene 3.pl-bread-red-make-appl
‘Irene is always making bread for them.’
b. *Irene am=pa(n)-pan-hoo-ria
Irene 3.pl- red-bread-make-appl
c. *Irene pan-am=ho-hoo-ria.
Irene bread-3.pl=red-make-appl
So, while the head-marking mechanism proposed by Stump can accommodate the
basic Hiaki cases, assuming that the verb compounds are produced by a Word-
to-Word rule,13 the way the system interacts with paradigm functions would have
to be relaxed in order to allow for some inflectional processes to be specified as
edge-marking with even Word-to-Word derivatives, while others are specified as
head-marking.
We conclude that the PUG cannot be maintained in its current form in the face of
examples like (42). This represents a serious challenge to the PFM architecture, which
conspires to derive the PUG as a sub-case of the Head Application Principle (HAP), a
supposed universal of morphological structure. As Stump himself indicates, one can
. Stump claims that both the root and the resulting derivative of a Word-to-Word Rule
must be either both roots or both nonradical words (Stump 2001, p. 117). Nonradical words
are forms extracted from the fully-inflected paradigm of a lexeme. In the case of most Hiaki
examples, we assume that within PFM, the root and the derivative would themselves both
have to be roots. However, in Example (24c) above, we exhibit a case where the root, as well as
the derivative, exhibits reduplication as an inflectional marker of habitual aspect. In Stump’s
terminology, one can conclude that this case would have to involve two nonradical words;
the inflected root form would be drawn from the output of the paradigm function applied
to the basic root, which produces nonradical words. However, the leftmost member of this
complex form is not a nonradical word itself; rather, it is still a bound form: no-nok-; the cor-
responding habitually inflected free form is no-noka, ‘RED-speak’. This may prove puzzling for
the definition of ‘Word-to-Word Rule’ in PFM.
Head-marking inflection and the architecture of grammatical theory 
conceive of evidence which genuinely disconfirms the PUG, hence also the HAP. We
submit that the Hiaki pattern of object cliticization and reduplication constitute such
evidence. We therefore suggest that the PUG cannot be a universally valid generaliza-
tion about the structure of inflectional paradigms.
We now turn to consider a potential account of this data in a syntacticocentric
framework supplemented with certain purely morphological operations, Distributed
Morphology (Halle & Marantz 1993 et seq.).
5.2 Syntacticocentric approaches: Distributed Morphology

Above, we sketched the syntactic account of noun incorporation proposed by Baker
(1988), and showed that, like the Weak Lexicalist account, it too produces the incor-
rect prediction that reduplication in Hiaki should be edge-marking, rather than head-
marking, since the incorporated N–V form is adjoined as a whole to the Asp0 head
realized by the RED prefix. The RED prefix should then attach to the entire N–V
subconstituent, not just to the V. Similar remarks apply to the V–V compound and
V-suffix cases in their interaction with reduplication. A Baker-style analysis of a V–V
compound like that in (24a) above is illustrated below; the embedded verb raises to
adjoin to the matrix verb, and again both verbs move to Asp0, predicting edge-mark-
ing behavior:
(43) AspP
VP1 Asp0
DP V1′ RED
HAB
Nee VP2 V10
1sg.NOM
DP V′ V20 V10
aa tV2 nok- ii′aa

3sg.ACC speak want
Nee aa=nok-i’-ii’aa
1sg.nom 3sg.acc=speak-red-want
‘I (habitually) want him to speak.’
We will propose that the interaction of reduplication and verb-compounding in Hiaki

is best analyzed, within the Distributed Morphology framework (Halle & Marantz
1993, 1994; Harley & Noyer 1999; Embick & Noyer 2007, a.o.), as an example of the
 Jason D. Haugen & Heidi Harley
post-syntactic operation Local Dislocation, of the type discussed in Embick and Noyer
(2007).
Following head-to-head movement of the complex verb to Asp0, the Asp0 head
will have the internal structure illustrated below:
(44) Asp0
V10 Asp0
V20 V10
nok- -ii′aa -RED

speak want HAB
At Morphology, this structure undergoes Vocabulary Insertion and Linearization. The

insertion of the RED morpheme will trigger a morpheme-specific operation of Local
Dislocation, in the sense of Embick and Noyer (2007). This operation prefixes the
RED morpheme to the first lexical verb to its left. The structural operations within the
complex Asp0 head are represented in bracketed notation in (45) below:
(45) a. [[V0 V0]V0 Asp0]Asp0 (Complex Asp0 head – Output of syntax)
b. [[nok [ii’aa-RED]]Asp0 (Insertion of Vocabulary Items,
Linearization)
c. [[nok [RED-ii’aa]Asp0]Asp0 (Local dislocation of red and ii’aa)14
d. [[nok [i’-ii’aa]] (Phonological content of red computed
by copying from the base.)
That is, in order to accommodate the Hiaki facts in a syntacticocentric approach,

the syntax must be supplemented by some operations specific to the morphology
component; syntax alone does not suffice. Under the version of the Y-model of gram-
mar assumed by DM, (45a) is the output of syntax proper, while (45b) and (45c) occur
. Within DM, this example illustrates an interesting feature of the relationship between
the Headedness Parameter (see, e.g. Baker 2001), which linearizes syntactic terminal nodes,
and the prefixal or suffixal status of particular Vocabulary Items (VIs), which we take to
be specified by Alignment constraints operating on those specific VIs. Linearization must
have applied before Local Dislocation so that the heads undergoing Local Dislocation are
linearly adjacent. We conclude that the prefixal or suffixal status of an affix has its effect
within a structured, morphological string that is already linearized, but the notions “prefix”
or “suffix” do not drive linearization itself. That is, Alignment seems to be separate from
Headedness, as we would expect: Hiaki is head-final (a general Linearization constraint),
but RED is prefixal.
Head-marking inflection and the architecture of grammatical theory 
in the post-syntactic module of Morphological Structure. The operation in (45c),

which, in conjunction with (45b), resolves the reduplicant linearization conundrum,
is taken to be a PF operation.15,16
The same sequence of derivation can apply to derive the examples in (24b) and
(24c) above, where reduplication applies to the embedded verb, on the assumption
that in those cases, the complement to -ii’aa is not a VP, but instead includes an entire
AspP with a habitual RED morpheme in its head. That embedded RED morpheme will
scope only over the embedded verb, and will undergo local dislocation to the right of
the embedded verb in such a case, deriving the patterns with low-scope reduplication
on the leftmost element of the complex predicate, or (if there is habitual aspect in both
the matrix and the embedded clauses) reduplication on both, interpreted, as indicated
above, in both places.
Below we show the same series of operations as it would apply to our initial exam-
ple of reduplication interacting with Noun Incorporation, in (13) above, based on the
Baker-style tree in (35), maso-peu-peute, ‘deer-RED-butchering’:
(46) a. [[N0 V0]V0 Asp0]Asp0 (Complex Asp0 head – Output of

syntax)
b. [[maso [peute-RED]]Asp0 (Insertion of Vocabulary Items,
Linearization)
c. [[maso [RED-peute]Asp0]Asp0 (Local dislocation of RED and peute)
d. [[maso [peu-peute]] (Phonological content of RED
computed by copying from the base.)
The tree illustrating the syntactic structure of the above example in (35) represents a
much earlier view of phrase structure than is commonly assumed in syntacticocentric
. In fact, different DM proposals distinguish several different operations similar to Local
Dislocation, including “Merger Under Adjacency” (Bobaljik 1994), which itself is similar to
Mithun’s (1984) Compounding by Juxtaposition. The proposal in the text represents just one
of several possible DM analyses of these constructions, each with its own set of consequences
for the analysis of hyponymous objects, object clitics, etc. We will not consider alternatives
here, but the proposal in the text represents the optimal account given the larger empirical
picture, according to our best current understanding. See Harley (2010) for a review of some
of the range of analytical possibilities made available by the interaction of these operations
with syntactic head-movement.
. In the end, then, our account does not ascribe ‘head-marking’ reduplication to Hiaki;
rather, the reduplicant in Hiaki morphologically selects or aligns itself with the closest avail-
able root to its left. As shown by the mochik (in (50) below) case, the root need not in fact be
verbal in character. The head of the constituent to which the aspectual meaning is attached is
actually a v0, projecting vP, in all of the above, while the √ that ends up bearing the inflection
is itself further embedded.
 Jason D. Haugen & Heidi Harley
analyses today. Most importantly from our current point of view, the representation
of peute as heading a simplex V node is inaccurate, given the considerable develop-
ments in the theory of argument structure syntax in the decades since Baker’s original
proposal. Agentive verbs are commonly now understood to consist of a minimum of
two projections: a lexical root which selects the internal argument (corresponding to
the original ‘V’), and an external-argument introducing ‘light verb’ functional projec-
tion, which we will notate here as v0. Indeed, the verb pair peu-te/peu-ta ‘butcher-
INTR/butcher-TR’ contains an overt morpheme which itself is plausibly analyzed as
an instantiation of that v0 node (though see Jelinek 1998 and Tubino Blanco 2010 for
more detailed discussion). That is, the actual base-generated structure, without any
syntactic head-movement, in this approach, does not look like (35) but rather should
minimally include the functional projections below:
(47) TP
T′
AspP T0
Asp′
vP Asp0
DP v′
√P v0
NP √
N0
Aapo maso peu te -RED Ø

He deer- butcher- INTR RED Ø
Following movement of the various heads (and the subject for nominative case), the
output of the syntax at Spell-Out, prior to Morphology, will have (minimally) the
following form:
Head-marking inflection and the architecture of grammatical theory 
(48) TP
DPi T′
AspP T0
Asp′ Asp0 T0
vP tAsp v0 Asp0
ti v′ v0
√
√P tv
N √
NP t√
tN
Aapo maso- peu- te- RED- Ø

He deer- butcher- INTR -RED- Ø
Of interest, of course, is the structure under the complex T0 head, containing the result
of head-movement through the verbal extended projection. The reduplicative mor-
pheme, RED, which is inserted to realize the Asp0 node, is subject to Local Dislocation.
However, this dislocation does not apply to RED and its immediate neighbor, which is
the light verbal morpheme -te. Rather, RED is dislocated leftward until it is prefixed to
the first available lexical root morpheme, √peu-. In Footnote 14 above, we suggested
that Local Dislocation could be conceived of as the application of a Vocabulary-Item-
specific Alignment constraint; this would be a perspicuous implementation of what
appears to be a straightforward subcategorization requirement of the RED morpheme:
specifically, it requires a lexical root as its host.
This implementation also predicts that the RED affix will not necessarily be sensi-
tive to the lexical category of the item which hosts it; under standard DM assumptions,
it is looking for an acategorial lexical √ to attach to, not a particular syntactic c ategory.17
. Because of the syntactic and semantic requirements of the Asp0 head which RED is the
realization of, this RED morpheme will necessarily appear in predicative contexts in Hiaki;
however, under this treatment, the RED vocabulary item itself does not select for verbs, but
rather subcategorizes for/aligns with √ morphemes.
 Jason D. Haugen & Heidi Harley
This prediction appears to be borne out in another interesting corner of Hiaki gram-
mar: the possession/use construction. In this construction, a nominal root is inflected
with verbal morphology, receiving an interpretation of ‘possessed N’. An example is
given in (49) below:
(49) Huan mochik-e-k
Juan turtle-PHAVE-perf
‘Huan has a turtle/turtles.’ (Lit: ‘Huan is turtled’.)
Reduplication in this construction targets the √ of the nominal mochik ‘turtle’:

(50) Huan mo-mochik-e
Huan red-turtle-PHAVE
‘Huan usually has turtles.’ (Haugen 2004, p. 264)
This supports the notion that the RED affix subcategorizes for √ morphemes: when
the closest available √ in the complex head happens to be a nominalized root, rather
than a verbalized one, that is the √ with which RED undergoes Local Dislocation/
Alignment.18
Under the present account, we can conclude that the difference between the
verbal affixes which support reduplication such as -ii’aa ‘want’, described above in
Section 3.2, and those such as -tua ‘CAUS’ or -ria, ‘APPL’, which do not, is that the
former retain their lexical roots, despite being on a grammaticalization path which
has restricted them to bound positions and which potentially could ultimately result
in their reanalysis as heads of functional categories such as v0. The latter affixes, which
do not support independent reduplication, we presume to head functional projections
in the synchronic grammar.19
The above analysis represents our current best understanding of the optimal
approach to this complex array of facts within a syntacticocentric analysis. Other
avenues of analysis obviously remain, but we hope to at least have shown that the
interaction of reduplication and head-marking is amenable to treatment given reason-
ably non-contentious assumptions within such frameworks. The analytical key which
the syntacticocentric approach makes available is that the internal structure of the
. Note that within the DM framework’s assumptions, all √s must occur within the scope
of a categorizing morpheme. In many cases this morpheme is null, as is the case with the
nominalizing head that presumably intervenes between the √ and the ‘PHAVE’ marker here. In
other cases, it is overt, as with the intransitive verbalizer -te in Hiaki siute, ‘tear.intr’ or English
-ify as in clarify, stupefy.
. These affixes most likely originally arose from independent lexical verbs which u
nderwent
this same grammaticalization process.
Head-marking inflection and the architecture of grammatical theory 
c omplex word-form remains accessible throughout the derivation, given the single-
engine architecture of the framework.
6. Conclusion
It should be clear that important architectural issues ride on the correct theoretical
approach to the range of facts described above, and on similar little-studied facts from
languages around the world. Many questions and issues, however, remain open. Here
we comment on some general and specific implications of and questions raised by the
discussion above, and indicate briefly some future directions that the current line of
analysis opens up.
We have considered how several distinct grammatical architectures could in prin-
ciple approach the Hiaki facts. A Strong Lexicalist approach like that of DiSciullo and
Williams (1987) seems to offer a set of tools which could allow an account of the inter-
action of the derivational and inflectional processes we are considering. The relevant
feature of the Strong Lexicalist architecture is the lack of differentiation between deri-
vational and inflectional processes, which allow us in principle to interleave the two.
Interestingly, a unified approach to inflection and derivation is a feature which Strong
Lexicalism shares with single-engine syntacticocentric analyses: i.e. all word-building
operations, inflectional and derivational, occur in the same generative subcomponent
of the grammar. Thus, architectures which group all types of word-formation opera-
tions into a single component do not encounter the problems that we have identified
for Weak Lexicalist and word-based architectures.
In the latter frameworks, on the other hand, if derivation is treated as creating
word-forms which are input to an inflectional word-formation process, such as a Para-
digm Function, head-marking phenomena become difficult to account for. Given cer-
tain additional assumptions, such as those concerning the formation of derivatives
that Stump (2001) puts forward, such frameworks can accommodate head-marking
phenomena. The specific proposal of Stump, however, runs aground on the mixed
character of inflection in Hiaki. Stump’s PUG requires that a derived form will consis-
tently head-mark or edge-mark (or doubly-mark) with respect to all inflectional pro-
cesses of the language. Inflectional systems in which some inflection is head-marking
and some edge-marking within the same form are outlawed. However, exactly such a
case is presented by the Hiaki NI + reduplication cases in combination with the object
cliticization properties of the language. Object cliticization is edge-marking in the very
same forms in which reduplication is head-marking. Consequently, we conclude that a
PFM analysis of these facts is untenable as things currently stand.
The syntacticocentric approach that we advocate also requires some supplemen-
tal assumptions in order to correctly position the inflectional reduplicant inside the
 Jason D. Haugen & Heidi Harley
compound verb form. However, these assumptions that we utilize are not novel ones;
rather, they are well-established mechanisms within the Distributed Morphology
framework. The strength of a DM approach is that, like the Strong Lexicalist approach,
derivational and inflectional affixation is accomplished in the same component – in
the case of DM, this is the syntax. Hence, the internal structure of complex derived
forms remains visible even following further inflectional affixation. In contrast to
PFM, no requirement that complex forms are restricted to either head-marking or
edge-marking behavior is either expressed or implied. The prefixation of the Hiaki
object clitic to the finite tensed verb form, for example, can coexist simultaneously
with the head-marking behavior of the reduplicative prefix. We assume, in fact, that
Hiaki object clitics are positioned by a syntactic mechanism similar to that which posi-
tions Romance object clitics next to the tensed verb: i.e. syntactic clitic movement to
TP and subsequent merger of the clitic with the T0 root node (cf. Matushansky 2006)
apparently captures the central facts concerning Hiaki object clitic distribution.
Unsurprisingly, many questions remain. One issue which we leave for future
investigation is the status of reduplication as a process or an affix. We have assumed
an affix-based approach here, following argumentation in Haugen (2008, 2010, 2011).
However, an approach treating reduplication as a readjustment operation triggered
by null affixation like that proposed in Raimy (2000) or Frampton (2009) could also
be entertained. Under this latter view, the aspectual head involved in triggering redu-
plication would be realized by a null affix which would trigger stem readjustment.20
Just as in the account proposed here, the stem readjustment approach would also have
access to the internal structure of the derived verb form, and in fact it would require
this: the null-affixation and stem-adjustment approach would still have to make refer-
ence to the ‘left edge of the closest lexical root’ as the domain for the relevant read-
justment. We leave open here the ultimate question of which of these approaches to
reduplication should be preferred within a syntacticocentric framework.
To conclude, the phenomena considered above graphically illustrate the impor-
tance of a ‘big-picture’ view of grammatical systems. Noun incorporation has tradi-
tionally (within generative grammar) been the province of morphosyntacticians, while
reduplication has been investigated primarily from the perspective of morphophonol-
ogy. The interaction of these two phenomena in a single language has significant impli-
cations that go beyond the narrow concerns of these two sub-domains, and bear much
more broadly on the architecture of grammatical theory.
. Under Raimy’s (2000) approach, this readjustment operation would involve re-lineariza-
tion, i.e. altering the precedence relationships between segments of the stem. For Frampton
(2009), the operation involves inserting ‘duplication junctures’ into the timing tier, ultimately
resulting in multiple links to the segmental material of the stem.
Head-marking inflection and the architecture of grammatical theory 
Acknowledgments
Our first thanks go, as always, to our wonderful Hiaki consultants, Maria Florez
Leyva and S antos Leyva, who patiently provided and explained much of the key data
presented here. Rosario Amarillas also generously contributed much data and com-
mentary. We would also like to thank the National Science Foundation, which funded
the research reported here, under grant BCS-0446333. Examples from Classical
Nahuatl were collected from the Florentine Codex (English translation by Anderson &
Dibble 1963) during the course of the Classical Nahuatl Project at Oberlin College,
via funding from the OC Office of Sponsored Projects, with the help of OC Research
Assistants Miriam Rothenberg and Anne Thompson. Thanks also to Shannon Bischoff
and Mercedes Tubino Blanco for valuable feedback.
References
Anderson, Arthur J.O. & Dibble, Charles E. (eds). 1963. Florentine Codex: General History of the
Things of New Spain by Fray Bernardino de Sahugún, Book 11: Earthly Things. Santa Fe NM:
The School of American Research and The University of Utah.
Anderson, Stephen. 1982. Where’s morphology? Linguistic Inquiry 13: 571–612.
Anderson, Stephen. 1992. A-morphous Morphology [Cambridge Studies in Linguistics 62].
Cambridge: CUP.
Aronoff, Mark. 1976. Word Formation in Generative Grammar. Cambridge MA: The MIT Press.
Baker, Mark C. 1988. Incorporation: A Theory of Grammatical Function Changing. Chicago IL:
University of Chicago Press.
Baker, Mark C. 2001. The Atoms of Language: The Mind’s Hidden Rules of Grammar. New York
NY: Basic Books.
Barragan, Luis M. & Haugen, Jason D. (eds). 2003. Studies in Uto-Aztecan. Cambridge MA: MIT
Working Papers on Endangered and Less Familiar Languages #5.
Bobaljik, Jonathan David. 1994. What does adjacency do? In The Morphology-Syntax Connec-
tion [MIT Working Papers in Linguistics 22], Heidi Harley & Colin Philips (eds), 1–32.
Cambridge MA: MITWPL.
Carstairs-McCarthy, Andrew. 1992. Current Morphology. London: Routledge.
Chang, Henry Y. & Wu, Chun-Ming. 2006. Nominal incorporation in Paiwan and its implica-
tions. Paper presented at the Conference on Nominal Incorporation and Its Kind (NIK),
University of Ottawa, February 21.
Dedrick, John M. & Casad, Eugene H. 1999. Sonora Yaqui Language Structures. Tucson AZ:
University of Arizona Press.
DiSciullo, Anna-Maria & Williams, Edwin. 1987. On the Definition of Word. Cambridge MA:
The MIT Press.
Downing, Laura J. 2003. Compounding and tonal nontransfer in Bantu languages. Phonology
20: 1–42.
Embick, David & Noyer, Rolf. 2007. Distributed morphology and the syntax-morphology inter-
face. In The Oxford Handbook of Linguistic Interfaces, Gillian C. Ramchand & Charles Reiss
(eds), 289–324. Oxford: OUP.
 Jason D. Haugen & Heidi Harley
Escalante, Fernando. 1990. Voice and Argument Structure in Yaqui. Ph.D. dissertation,
University of Arizona.
Farmer, Ann. 1980. On the Interaction of Morphology and Syntax. Ph.D. dissertation, MIT.
Frampton, John. 2009. Distributed Reduplication [Linguistic Inquiry Monographs 52].
Cambridge MA: The MIT Press.
Hale, Ken. 2000. A Uto-Aztecan (O’odham) reflection of a general limit on predicate argument
structure. In Uto-Aztecan: Structural, Temporal, and Geographical Perspectives; Papers in
Memory of Wick R. Miller by the Friends of Uto-Aztecan, Eugene H. Casad & Thomas L.
Willett (eds), 155–169. Hermosillo, Sonora: Universidad de Sonora, Editorial UniSon.
Hall, T. Alan. 1999. The phonological word: A review. In Studies on the Phonological Word
[Current Issues in Linguistic Theory 174], T. Alan Hall & Ursula Kleinhenz (eds), 1–22.
Amsterdam: John Benjamins.
Halle, Morris & Marantz, Alec. 1993. Distributed morphology and the pieces of inflection. In
The View from Building 20: Essays in Linguistics in Honor of Sylvain Bromberger, Kenneth
Hale & Samuel Jay Keyser (eds), 111–176. Cambridge MA: The MIT Press.
Halle, Morris & Marantz, Alec. 1994. Some key features of Distributed Morphology. In Papers
on Phonology and Morphology [MIT Working Papers in Linguistics 21], Andrew Carnie &
Heidi Harley (eds), 275–288. Cambridge MA: MITWPL.
Harley, Heidi. 2010. Affixation and the Mirror Principle. In Interfaces in Linguistics [Oxford
Studies in Theoretical Linguistics 31], Rafaella Folli & Christiane Ullbricht (eds), 166–186.
Oxford: OUP.
Harley, Heidi & Amarillas, Maria. 2003. Reduplication multiplication in Yaqui: Meaning X
form. In Barragan & Haugen (eds), 105–139.
Harley, Heidi & Florez Leyva, Maria. 2009. Form and meaning in Hiaki (Yaqui) verbal redupli-
cation. International Journal of American Linguistics 75: 233–272.
Harley, Heidi & Noyer, Rolf. 1999. State-of-the-article: Distributed Morphology. Glot Interna-
tional 4: 3–9.
Harris, Alice C. 2000. Where in the word is the Udi clitic? Language 76: 593–616.
Harris, Alice C. 2002. Endoclitics and the Origins of Udi Morphosyntax. Oxford: OUP.
Haspelmath, Martin. 2011. The indeterminacy of word segmentation and the nature of
morphology and syntax. Folia Linguistica 45(1): 31–80.
Haugen, Jason D. 2003. Allomorphy in Yaqui reduplication. In Barragan & Haugen (eds), 75–103.
Haugen, Jason D. 2004. Denominal verbs in Yaqui. In Estudios en lenguas amerindias: Homenaje
a Ken L. Hale, Zarina Estrada Fernández, Ana Fernández Garay & Albert Álvarez González
(eds), 229–267. Hermosillo, Son: Editorial Unison.
Haugen, Jason D. 2008. Morphology at the Interfaces: Reduplication and Noun Incorporation in
Uto-Aztecan [Linguistik Aktuell/Linguistics Today 117]. Amsterdam: John Benjamins.
Haugen, Jason D. 2010. Reduplication in compounding contexts: Morphological doubling vs.
correspondence. In the Proceedings of NELS 39. 〈http://ling.auf.net/lingBuzz/001368〉
Haugen, Jason D. 2011. Reduplication in Distributed Morphology. In Proceedings of the 4th
Annual Arizona Linguistics Circle Conference (ALC 4). Coyote Papers, Vol. 18. Tucson AZ:
University of Arizona Department of Linguistics. 〈http://arizona.openrepository.com/
arizona/handle/10150/138590〉
Hill, Jane H. 2003. Subject number agreement, grammaticalization, and transitivity in the
Cupeño verb construction. In Formal Approaches to Function in Grammar: In Honor of
Eloise Jelinek [Linguistik Aktuell/Linguistics Today 62], Andrew Carnie, Heidi Harley &
MaryAnn Willie (eds), 207–226. Amsterdam: John Benjamins.
Head-marking inflection and the architecture of grammatical theory 
of American Linguistics 58(4): 355–404.
Hill, Kenneth C. 2003. Hopi denominal and noun-incorporating verbs. In Barragan & Haugen
(eds), 215–244.
Hopi Dictionary Project. 1998. Hopi Dictionary/Hopìikwa Lavàytutuveni: A Hopi-English
Dictionary of the Third Mesa Dialect. Tucson AZ: University of Arizona Press.
Inkelas, Sharon & Zoll, Cheryl. 2005. Reduplication: Doubling in Morphology [Cambridge
Studies in Linguistics 106]. Cambridge: CUP.
Jelinek, Eloise. 1998. Voice and transitivity as functional projections in Yaqui. In The Projection
of Arguments: Lexical and Compositional Factors, Miriam Butt & Wilhelm Geuder (eds),
195–224. Stanford CA: CSLI.
Kiparsky, Paul. 1982. Word formation and the lexicon. In Proceedings of the 1982 Mid-America
Linguistics Conference, 3–29. Lawrence KS: University of Kansas.
Lapointe, Steven. 1980. A theory of Grammatical Agreement. Ph.D. dissertation, MIT.
Lapointe, Steven. 1981. General and restricted agreement phenomena. In The Scope of Lexical
Rules, Michael Moortgat, Harry van der Hulst & Teun Hoekstra (eds), 125–159. Dordrecht:
Foris.
Lieber, Rochelle. 1980. On the Organization of the Lexicon. Ph.D. dissertation, MIT.
Lieber, Rochelle & Scalise, Sergio. 2006. The Lexical Integrity Hypothesis in a new theoretical
universe. Lingue e Linguaggio 5(1): 7–32.
Marantz, Alec. 1982. Re reduplication. Linguistic Inquiry 13: 435–482.
Marantz, Alec. 1984. On the Nature of Grammatical Relations. Cambridge MA: The MIT Press.
Marantz, Alec. 1997. CAT as a phrasal idiom. Ms, MIT.
Matthews, Peter H. 1972. Inflectional Morphology: A Theoretical Study Based on Aspects of Latin
Verb Conjugation. Cambridge: CUP.
Matushansky, Ora. 2006. Head movement in linguistic theory. Linguistic Inquiry 37: 69–109.
McCarthy, John J. & Prince, Alan S. 1986. Prosodic morphology. Ms, University of Massachu-
setts at Amherst & Rutgers University.
McCarthy, John J. & Prince, Alan S. 1993. Prosodic Morphology I: Constraint interaction and
satisfaction. Ms, University of Massachusetts at Amherst & Rutgers University.
McCarthy, John J. & Prince, Alan S. 1995. Faithfulness and reduplicative identity. In U niversity
of Massachusetts Occasional Papers in Linguistics 18: Papers in Optimality Theory, Jill
Beckman, Laura Walsh Dickey & Suzanne Urbanczyk (eds), 249–384. Amherst MA: GLSA.
Merlan, Francesca. 1976. Noun incorporation and discourse reference in Modern Nahuatl.
International Journal of American Linguistics 42: 177–191.
Mithun, Marianne. 1984. The evolution of noun incorporation. Language 60: 847–895.
Mithun, Marianne. 1986. On the nature of noun incorporation. Language 62: 32–38.
Mithun, Marianne & Corbett, Greville G. 1999. The effect of noun incorporation on argument
structure. In Boundaries of Morphology and Syntax [Current Issues in Linguistic Theory
180], Lunella Mereu (ed.), 49–71. Amsterdam: John Benjamins.
Molina, Felipe S., Valenzuela, Herminia & Leedom Shaul, David. 1999. Yoeme/English-English/
Yoeme Standard Dictionary. New York NY: Hippocrene Books.
Moravcsik, Edith. 1978. Reduplicative constructions. In Universals of Human Language, Vol.
3: Word Structure, Joseph H. Greenberg (ed.), 297–334. Stanford CA: Stanford University
Press.
 Jason D. Haugen & Heidi Harley
Munro, Pamela & Riggle, Jason. 2004. Productivity and lexicalization in Pima compounds. In
Proceedings of BLS 30. Berkeley CA: Berkeley Linguistics Society.
Newmeyer, Frederick. 1986. Linguistic theory in America, 2nd edn. New York NY:
Academic Press.
Raimy, Eric. 2000. The Phonology and Morphology of Reduplication. Berlin: Mouton de Gruyter.
Rosen, Sara Thomas. 1989. Two types of noun incorporation: A lexical analysis. Language 65:
294–317.
Sadock, Jerrold M. 1991. Autolexical Syntax: A Theory of Parallel Grammatical Representations.
Chicago IL: University of Chicago Press.
Sapir, Edward. 1911. The problem of noun incorporation in American languages. American
Anthropologist 13: 250–282.
Sato, Yosuke. 2010. Complex phrase structures within morphological words: Evidence from
English and Indonesian. Lingua 120: 379–407.
Selkirk, Lisa. 1981. English compounding and the theory of word structure. In The Scope of
Lexical Rules, Michael Moortgat, Harry van der Hulst & Teun Hoekstra (eds), 229–277.
Dordrecht: Foris.
Smirniotopoulos, Jane C. & Joseph, Brian D. 1998. Syntax vs. the lexicon: Incorporation and
compounding in Modern Greek. Journal of Linguistics 34: 447–488.
Spaelti, Philip. 1997. Dimensions of Variation in Multi-Pattern Reduplication. Ph.D. disserta-
tion, University of California at Santa Cruz.
Stump, Gregory T. 2001. Inflectional Morphology: A Theory of Paradigm Structure [Cambridge
Studies in Linguistics 93]. Cambridge: CUP.
Stewart, Thomas & Stump, Gregory T. 2007. Paradigm function morphology and the
morphology-syntax interface. In The Oxford Handbook of Linguistic Interfaces, Gillian
Ramchand & Charles Reiss (eds), 383–421. Oxford: OUP.
Tubino Blanco, Mercedes. 2009. Contrasting Causatives: A Minimalist Approach. Ph.D. disser-
tation, University of Arizona.
Tubino Blanco, Mercedes & Harley, Heidi. 2010. Dos tipos de base verbal en Hiaki (yaqui): Hacia
una clasificación morfológica. In Memoria del IX Encuentro Internacional de Lingüística en
el Noroeste. Hermosillo: Editoriales UniSon.
Tubino Blanco, Mercedes, Harley, Heidi & Haugen, Jason D. 2010. The syntax of hybrid verb/
affix lexemes and clause fusion in Hiaki (Yaqui). Rice Working Papers in Linguistics 1: 79–91.
Williams, Edwin. 1981. On the notions ‘lexically related’ and ‘head of a word’. Linguistic Inquiry
12: 245–274.
Zoll, Cheryl. 2002. Vowel reduction and reduplication in Klamath. Linguistic Inquiry 33:
520–527.
A case-study in grass roots development
of web resources for language workers
The Coeur d’Alene Archive and Online
Language Resources (CAOLR)
Shannon Bischoff & Amy Fountain*

Indiana University-Purdue University Fort Wayne / University of Arizona
“Community language workers, speakers, and other members of local groups

are both participants and overhearers in a global conversation about language
endangerment in which the voices of academics and policymakers are especially
prominent. How might this global conversation resonate for members of
communities that are custodians of endangered languages – communities that
are themselves a diverse audience? Do they find it empowering and encouraging,
unintelligible and alienating, or something in between?” (Hill 2002: 119). We
discuss ‘grass roots digital archiving’ as a method of addressing Hill’s questions.
Specifically, we discuss the Coeur d’Alene Archive and Online Language
Resources as a case study of digital archiving undertaken in the spirit of grass
roots archiving and digital resource creation.
Keywords: Coeur d’Alene; archive; digital resources; online dictionary
1. Introduction
During the past ten years much work and funding has been invested in projects that
develop or help in the development of web-based archives for language resources.
Important projects of this sort include the Open Language Archives Community
(OLAC),1 the Pacific and Regional Archive for Digital Sources in Endangered Culture
* The authors wish to thank Philipp Brandeburg, Raymond Brinkman, Debbie Cole, Colleen
Fitzgerald, Musa Yasin Fort, Terry Langendoen, Susan Penfield, and Mizuki Miyashita for
useful discussion and development of the ideas presented in this paper. All errors are our own.
1. 〈http://www.language-archives.org/〉
 Shannon Bischoff & Amy Fountain
(PARADISEC),2 the Digital Endangered Languages and Musics Archives Network

(DELAMAN),3 the Alaska Native Language Center (ANLC),4 the Archives of the
Indigenous Languages of Latin America (AILLA),5 the Sam Noble Museum of Natural
History’s Language Collection (SNOMNH),6 and the Electronic Metastructure for
Endangered Languages Data (EMELD),7 among others. These projects have focused in
great part on the development of complex data sets and/or online archiving resources
devoted to making linguistic and cultural material easily searchable, locatable, and
responsibly stored. In the same vein, more traditional archives, such as those housed
at the National Anthropological Archives at the Smithsonian, have focused efforts on
creating microfilm copies of works and placing the catalogs online (see Macri et al.
2009, for example, for discussion of this focus with regard to the Harrington collection
materials). The investment in financial and human resources for such endeavors illus-
trates the widely accepted value among academics and policy makers of digital and
web resources for endangered and minority languages. In many ways, we might think
of these efforts as the realization of Krauss’ (1992: 8) call for “a network of repositories
and centers for safeguarding and using…documentation”. Competition for resources
for these very important and useful projects is typically based on a set of expert rheto-
rics that can sometimes render their development “unintelligible” and “alienating” for
communities who are the custodians of the language resources themselves.
This paper presents another means, less costly and quite complementary to
the types of projects mentioned above, for small-scale, community-based p rojects
to improve the accessibility and preservation of endangered language materials
using d igital and web resources, in ways that respect and express their unique local
understandings of the world.8 This follows Hill’s (2002) exhortation that acknowl-
edgement of the “genius of ordinary people” may provide a useful and important
venue for advancing the cause of endangered language documentation and revitaliza-
tion. We suggest that grass roots archive projects can be remarkably successful, even
if they begin with very little in terms of technical knowledge, and no outside funding.
. 〈http://www.paradisec.org.au/〉
. 〈http://www.delaman.org/〉
. 〈http://www.uaf.edu/anlc/〉
. 〈http://www.ailla.utexas.org/site/welcome.html〉
. 〈http://www.snomnh.ou.edu/collections-research/nal.htm〉
. 〈http://emeld.org/〉
. In this paper we do not outline in detail the steps to building the COALR (e.g. specific
HTML, PHP code). However, the about page of the COALR does provides limited discussion
of such issues. Presently, the authors are working on a paper to serve this purpose.
A case-study in grass roots development of web resources for language workers 
Crucially, they require motivated and dedicated language workers willing to work
with publically available technical training resources. We describe the Coeur d’Alene
Archive and Online Language Resources (CAOLR) as an example and starting point
for what can be done to make legacy materials and other documentation readily
available for maintenance or revitalization efforts, and, whenever appropriate, the
larger linguistic community, without external funding and without technical experts.
Additionally, we believe that such efforts, when undertaken by communities using
their own legacy materials, can serve as a bridge between documentation and revi-
talization efforts in much the same way as Flores Farfán and Ramallo (2010) suggest
sociolinguistic approaches may do. Further, they can lead to more collaborative
projects between linguists in academic settings and community members as outlined
in Yamada (2011).
The process of digitizing, storing and publishing materials online is often felt to
be too technical or difficult for non-specialists to undertake. The primary goal of this
paper is to show that much of this is possible, even with relatively minimal overhead
and little or no previous technical knowledge, if a small number of highly motivated
language workers are willing and able to invest their time in such an endeavor. While
these projects should not (and could not) replace larger-scale, externally funded
archiving initiatives – and may not be perfectly suited to the goals of the scholarly
community – they can be feasible and useful to communities who want access to legacy
materials about their languages. Most importantly, small-scale projects undertaken by
non-specialists can allow room for the many voices of today’s language workers, their
communities, and of the language workers of the past, to be heard.
This case study suggests some important questions that should be addressed in
the initial stages of such projects in order to ensure the long-term maintenance and
preservation of digital resources. Addressing these questions can insure that grass
roots archives are able to meet the high standards of digital archiving exemplified in
the Open Language Archive Communities’ TAPS checklist (Chang 2010), outlined in
Bird and Simons (2003), and the Protocols For Native American Archive Materials
(First Archivists Circle 2007). Importantly, they can and should do so in a way that
includes room for the voices of community language workers and those of the aca-
demic experts and policy makers, to be instantiated in the “product”, which can then
be seen as a truly shared resource.
We begin by elaborating the terms ‘grass roots’ and ‘archiving’ as we intend them
to be used here. We then describe the CAOLR as a case study that can inform efforts
in grass roots web-based language archiving projects. While the CAOLR is not a ‘grass
roots’ project, it does share important properties with true grass roots efforts. We con-
clude by highlighting important lessons learned in the development of the CAOLR,
and summarizing ways in which the project ought to serve as an encouraging example
to language workers interested in pursuing projects like this one.
 Shannon Bischoff & Amy Fountain
2. What makes a project a ‘grass roots’ development effort?
We use the term ‘grass roots’ to identify at least the following two properties of any
language documentation project. First, a ‘grass roots’ project is one that is initiated
and controlled by community members – local language workers – rather than some
external agency or ‘expert’. While outsiders (including linguists, anthropologists and
perhaps others) may be invited to participate, the goals and trajectory of a grass roots
project are decided upon by the language workers themselves. If language workers
choose to enlist the assistance of experts in these projects, it is with the strict under-
standing that the appropriate role of those experts is to provide support and assistance
to the workers, not to determine the goals and trajectories for the project.
Second, a ‘grass roots’ project typically begins its life as an under- or un-funded
initiative. These two properties – the lack of an externalized project management or
oversight structure and the paucity of financial support at least in the initial stages of
a project – are sometimes viewed as insurmountable obstacles to the project’s success,
and become barriers in themselves. Language workers interested in developing these
projects may represent multiple generations, and often come from communities
which are economically and politically disadvantaged. But these workers often have
tremendous energy and passion for the work that they are doing. Language workers
in today’s minority or endangered linguistic communities are often highly motivated
and extraordinarily resourceful, and scholarship (i.e. Johnstone 2011) on digitization
of materials documenting minority voices indicates that the very act of digital pub-
lication may help these workers assert their own expertise and authority about their
heritage language resources. In addition, such community language workers would
be well positioned to incorporate outcomes into revitalization efforts and local school
curriculums in the spirit of ‘Culturally Based Education’ and ‘Culturally Responsive
Schooling’ as outlined in McCarty and Wiley Snell (2011).
Communities may have documentary materials collected in the past by linguists
and anthropologists, but their conflicted history of working with outside academ-
ics sometimes results in an imperative that access to these materials, and any new
materials to be created based on them, should be controlled community-internally,
primarily to serve the needs of the community itself (cf. Yamada 2011 for relevant
discussion).
A grass roots development effort, then, is typically characterized by an initial lack
of resources and expertise as to the how of the project but not as to the what of the
project. A grass roots web development effort turns out to be an extremely fortuitous
kind of project – because the how of web-development, though shrouded in mystery
for most people who’ve not participated in such endeavors before, turns out to be
remarkably accessible for precisely those highly committed community members who
are likely to form the core of community-generated language documentation projects.
A case-study in grass roots development of web resources for language workers 
This accessibility is not immediately apparent upon review of the standards for p rojects
like EMELD, OLAC, or PARADISEC, which have been elaborated on a model that
is productive and effective in the environment of academics and policy makers. The
application of the literature on best practices in online archiving (i.e. Bird & Simons
2003 and Chang 2010) to grass roots projects is similarly unclear about how such
projects can be undertaken. Our goal is to show that grass roots efforts can meet many
of these standards, in spite of their lack of ‘expert’ leadership and agency funding.
In recent years, there has been a proliferation of endangered language websites –
both community developed and externally driven – and we anticipate that this will (and
should) continue in the foreseeable future. Buszard-Welcher (2001) notes that many
of these sites contain few or no examples of the endangered language itself, however.
This dearth of language content may be the result of perceived (or actual) barriers in
terms of web accessibility, orthographic difficulties, and other technical challenges (see
Galla 2009, for example, for discussion of some of these challenges). Development of
the CAOLR required that developers grapple with many of these problems, and find
manageable solutions to them.
The goal of the CAOLR project was to find out whether, and how, the creation of
an online language resource archive could be accomplished in a reasonably short time
frame using only publicly available, low-cost or cost-free resources and training. This
project was undertaken to discover whether language workers without specialized
training in linguistics or web development could successfully launch a useful and
viable online language resource while adhering to the strict guidelines required of
a digital archive.
3. What makes a project an ‘archive’?
An ‘archive’ is a collection of materials that is intended to safely store those materials

over a relatively extended timespan, and support appropriate access to them for at
least the foreseeable future (Austin et al. 2005). A ‘language archive’ is a collection of
language materials – these may be written texts, audio recordings, video files, images,
and/or other artifacts related to the structures and functions of the language of a
particular community of speakers. Language materials that are good candidates for
an online archive project are those that can be stored and accessed over the World
Wide Web – that is, materials that are already in, or can be readily converted to, some
standard digital format.
‘Safe storage’, in terms of digital archives, requires a combination of data security
measures, data backup plans, and digital format decisions (see, for example, Simons
2003). Externally funded projects often include support for server space and server
administration, which may be contracted out to an academic institution, library,
 Shannon Bischoff & Amy Fountain
or other external agency. Grass roots projects may or may not have access to these
resources (which can be quite expensive, and require long term funding commitments).
For grass roots projects, approaches to the issues of ‘safe storage’ may need to be scaled
down to very simple, locally controllable protocols for providing multiple, regular
backups of files. However, many communities in the U.S. and Canada have the relevant
resources and it may be a matter of contacting, for example, a tribal council or direc-
tor of tribal web resources. It may also be in the interest of some project developers to
establish partnerships with Tribal Colleges in order to gain access to such resources as
server space.
‘Appropriate access’ is an extraordinarily complex and nuanced issue for
endangered language communities. Language materials vary tremendously in their
size, scope and sensitivity. Language workers must develop an appropriate approach to
language materials in terms of intellectual property, cultural sensitivity, and personal
sensitivity, and must be judicious in the ways in which they determine what appropriate
access is, and how it might be controlled. These factors vary widely among various
language communities, within communities, and across time periods – and there is
a growing and robust literature on the ethical, legal and practical issues surround-
ing appropriate access (see, for example Anderson & Koch 2003; Boynton et al. 2006;
Fitzgerald 2005 & 2009; Lewis et al. 2006; Liberman 2000; O’Meara & Good 2010;
Penfield et al. 2008; Tatsch 2004 and Warner et al. 2009, to name just a few).
Grass roots projects are well-positioned to develop and maintain appropriate
access plans because the communities who manage the project are also those who
best understand and most appropriately determine the meaning of ‘appropriate
access’ to their own materials.9 However, grass roots projects can learn much from the
experiences of those outside the community who have done much work in such areas.
Establishing ‘appropriate access’ for materials available on the web is a thorny
issue. It is relatively simple to ‘secure’ access to web-delivered materials against being
discovered by casual web users, for example, by calling for simple password-protection
on various pages (for the lowest level of security), or by publishing materials on ‘local’
web servers that are not connected to the entirety of the World Wide Web (for much
higher security). Such ‘protection’, however, can easily be construed as ‘censorship’ –
particularly when it renders materials more difficult for authorized users to access.
It is not so simple, however, to ‘secure’ access to web-delivered materials such
that the community would be protected against very determined hackers, or others
with nefarious uses in mind. Language workers should understand that it is to some
. Many of the larger, externally-funded projects mentioned in the introduction to this paper
have done an extraordinary job of ensuring appropriate access, of course. It is our interest to
demonstrate that even small-scale, grass-roots efforts can do so.
A case-study in grass roots development of web resources for language workers 
extent inherent in their decision to develop a web-accessible language archive that

they may not be fully in control of who views materials they have posted, or of the
secondary uses to which such materials might be put. They must pay careful attention
to appropriate rules of usage, copyright (see, especially, Newman 2007 for details), and
cultural or individual sensitivities as they make decisions about what material is, and
what is not, appropriate for a web-based language archive.
Issues of ‘appropriate access’, therefore, will be appropriately subject to much
discussion, debate, and careful consideration by community members and language
workers in preparation for any web-based language archive project. Fortunately, many
communities have existing web resources for tribal affairs and agencies. Local experts
working in these areas can be an excellent resource in the early planning stages and
throughout any project developing online language resources.
4. A
case study in grass roots archiving: The Coeur d’Alene Archive and
Online Language Resources (CAOLR)
With these issues in mind, we turn to what we believe can serve as an example or
starting point for those wishing to pursue grass roots, web-based, language archiving
projects – the development of the CAOLR. We begin by discussing the background
of this project, and then we outline its relationship to ‘grass roots language archiving’
as we defined it above. We describe the goals identified in developing the CAOLR,
and how the project addressed several kinds of issues often at the heart of such
development efforts for indigenous language communities. We outline the steps
undertaken in building the CAOLR – including the creation of navigation struc-
tures that were developed in order to make the resources intelligible and useful to
community members.
4.1 B
ackground: The Coeur d‘Alene Language and legacy materials
and the content of the CAOLR
At the time of this writing, The Coeur d’Alene Archive and Online Language Resources
(CAOLR) exists as a proof-of-concept of the kind of grass-roots archiving project we
believe to be possible.10 The CAOLR was not produced by community members, but
it was produced in a manner we believe could be readily undertaken by community
members who are not trained in linguistic theory or web d evelopment techniques.
The CAOLR was created in consultation with the then Director and the Project
. 〈http://academic.uprm.edu/~sbischoff/crd_archive/start1.html〉
 Shannon Bischoff & Amy Fountain
oordinator of the Coeur d’Alene Tribe Language Program. The archive c ontains a
C
significant amount of documentary material – integrating previously unpublished
legacy materials with published scholarly resources – in a way that is s uggestive of how
much can be done with relatively few resources and just a few committed volunteers.
It is certainly not complete. It is available online to the Coeur d’Alene community, and
is being utilized by community members. We provide some very brief background on
the Coeur d’Alene language community, and describe the resources that were used to
begin development of the CAOLR here.
Coeur d’Alene (CRD) is a Southern Interior Salishan (USA/Idaho) language
no longer learned by children in the home. The community has an active language
program and the language is taught in a few schools as a separate subject. A number
of unpublished documentary materials on Coeur d’Alene were provided to one of the
authors (Bischoff) when he was a graduate student in linguistics who was working
on a masters’ thesis on that language. In the course of events, it was discovered that
these materials were not accessible to the Coeur d’Alene community itself. The project
described here was a result of efforts to ensure that the resources available to gradu-
ate students in linguistics would also be available to the community. These resources
were rich and varied, springing from linguistic and anthropological research carried
out in the first half of the twentieth century by Franz Boas’ student Gladys Reichard
and a team of community scholars including Dorothy Nicodemus, Tom Miyal, Julia
Antelope, and Lawrence Nicodemus. Dorothy Nicodemus and Tom Miyal provided
the majority of narratives recorded while Julia Antelope was the primary editor of the
recordings. Lawrence Nicodemus typed up many of the narratives and provided vari-
ous editorial support and analysis.
The first extensive documentation of the language was undertaken by Reichard
and the team of community scholars beginning in the late 1920’s and continuing
through the 1950’s (see Reichard 1938, 1939, 1947, 1958–1960).11 Reichard worked
closely with Lawrence Nicodemus,12 who went on to consult with most scholars
working on the language up until his death in 2004. Nicodemus created the tribal
orthography, a Coeur d’Alene to English dictionary, and an English to Coeur d’Alene
dictionary (Nicodemus 1975). Nicodemus also worked in the development of peda-
gogical materials (see for example Nicodemus et al. 2000).
Over 1,200 pages of unpublished field notes and typed manuscripts of CRD were
left by Reichard and the team of community scholars in hard copy, and r endering
these materials accessible to the community was the impetus behind this project.
. For an excellent account of Reichard’s career and work on Salish, see Faulk 1997, 1999.
. For an account of Nicodemus’ life and role as language scholar, see Brinkman 2003.
A case-study in grass roots development of web resources for language workers 
The hard copy manuscripts were subject to decay and neglect, and – because they
were only in hard copy – it was difficult to make them fully accessible to a geo-
graphically distributed readership. Digitizing these materials, both to make them
accessible and to ensure their preservation, seemed like an appropriate first step.
It was determined that published scholarly materials should be integrated with
these legacy resources in order to improve their usability for community members.
Reichard’s published work, including a list of the verb stems found in this highly
polysynthetic language (Reichard 1939) and a reference grammar (Reichard 1938),
were at the start of development of this project already available in digital formats,
but were discoverable only through advanced searches of scholarly resources. The
published works were therefore marked for inclusion in the archive in a way that
would allow visitors to interact with the unpublished field notes as they relate to the
published works.
The previously published works also contained information that was organized
in a way that served scholarly interests, but that was not readily accessible to a non-
academic audience. For example, the affixes used to create fully inflected verbs were
all documented somewhere in the various published resources – but there was no
centralized tool containing the affixes and their uses. Therefore, a searchable affix list
containing roughly 200 affixes from the 1938 grammar was compiled for inclusion in
the archive.
Published resources by scholars working from Reichard and Nicodemus’ original
notes, and independently of Reichard, were also found in the scholarly literature, and
marked for inclusion in the archive. In particular, Lyon and Greene-Wood’s (2007)
edition of Nicodemus’ Coeur d’Alene dictionary in root form was also identified early
on for inclusion in the CAOLR. The resources outlined above form the basis for the
CAOLR today. Teit’s 1917 publication of Coeur d’Alene narratives, recorded before
Reichard’s work had begun, included some narratives that corresponded to stories
collected separately by Reichard and the Coeur d’Alene team. The benefit of their
inclusion in the resource was obvious.
Thus as of the time of this writing, the COALR includes:
1. Searchable Root Dictionary (Lyon & Green-Wood 2007 modified)

2. Searchable Stem-List (Reichard 1939)
3. Searchable Affix List
4. Some 48 narratives (over 1,200 pages of field notes, typed manuscripts and E
nglish
translations in various digital formats)
5. Links to Reichard’s original Grammar
6. Links to earlier scholarly work
7. Various other resources
 Shannon Bischoff & Amy Fountain
There remains a large body of digitizable materials that could – and we hope will –
be included in the archive in future.13 The most prominent of these is a set of more
than 1,000 of Nicodemus’ file slips, a variety of published and unpublished peda-
gogical materials developed by Nicodemus and others over the years, and early audio
recordings. The inclusion of these additional legacy materials in the CAOLR will be an
important step in ensuring that the community has appropriate access to all available
documentation about their language. The project of ensuring that the CAOLR meets
or exceeds the scholarly community’s expectations of best practices in online archiving
is also an important goal for the future of the CAOLR.
4.2 Th
e Coeur d’Alene Archive and Online Language Resources project
and its relation to grass-roots archiving
The primary goal of the CAOLR was to make the known Coeur d’Alene language
resources (published and not) available to the Coeur d’Alene community in an online
format. The project began as a ‘secondary’ outcome of externally-driven scholarly
research. Since the community was itself for the most part unaware of the existence
and extent of the legacy materials available to linguists, the community did not initially
drive the development of the CAOLR. In this regard, this project is different than true
grass-roots language documentation efforts. But it shares striking similarities with
such projects in other ways.
The project began as a response to an obvious need – for community members
to be able to access the legacy materials collected from them over the years. The
project was not designed primarily to facilitate scholarly research, and in this way it
is also similar to many grass roots efforts. The project was undertaken on a v olunteer
basis – without external support, and in addition to the normal daily duties of the
developers. Neither of the project’s developers (Shannon Bischoff and Musa Yasin
Fort, an interested and motivated undergraduate student at the University of Puerto
Rico Myagüez) had any previous training or experience in web development or
digital archiving. Although one of the developers (Bischoff) had significant training
in linguistics, the CAOLR’s development was not guided by concerns of linguistic
analysis. For these reasons, we believe that the effort might be a good test-case for
grass roots projects of this sort, possibly serving as a model to build on for commu-
nity-based efforts, and could help us to better understand whether or not the ‘best
practices’ advocated for by the scholarly community could be attained under these
conditions.
. At the time of writing an NSF grant has been awarded to expand the CAOLR. This grant
is a collaborative effort with members of the Coeur d’Alene community and Tribal Language
Program.
A case-study in grass roots development of web resources for language workers 
4.3 Notes on the development of the CAOLR

The developers had a number of goals for the CAOLR. First, all field notes and typed
manuscripts that had never been published needed to be digitized, and safely stored.
Safe storage was not enough, of course – the next concern was providing appropriate
access to these materials. Appropriate access to the materials involved not only mak-
ing it possible for community members to view the materials, but also for them to be
able to put those materials to use, by providing a context in which the materials could
be understood. Putting materials into a meaningful context demanded finding ways
to link together related materials within the original set of documents (for example,
allowing users to view the hand-written field notes alongside the typed versions of
those same notes, in order to be able to understand how the notes were revised in the
process). It also meant linking these materials to other published work already publicly
available on the web. It required that users be able to search the materials so that they
could find the kinds of information they were interested in.
The materials available use at least three different orthographic systems, and because
orthographic choices are reflective of users various linguistic ideologies (see, for exam-
ple, Kroskrity 2009) it was necessary to determine how these d ifferent o
rthographies
should be represented. One orthography, used by the tribal community today, was
developed by Nicodemus, and so is referred to as the ‘Nicodemus orthography’. Lyon
and Greene-Wood, in the dictionary they created based on Nicodemus’ (2007) work,
used what they refer to as the ‘Salish orthography’, and this differs from Nicodemus’
system. Reichard used the orthographic conventions adopted by Franz Boas for the
transcription of many American Indian languages. We shall refer to this as the ‘Reichard
orthography’. The Reichard orthography and the Salish o rthography where developed
primarily for linguists, and they are not generally accessible to community members.
Following the suggestion of the tribal Language Program Project Coordinator,
where it was possible we made resources available in the Nicodemus orthography (e.g.
stem-list, root dictionary, and affix list). This decision followed directly from the fact
that the goal of the project was to make materials accessible to the community. It was
not possible, however, to produce all of the materials in the Nicodemus orthography,
because many of them were hard copy originals of field notes – both handwritten and
typed, which would be digitized directly. Therefore, it was important that translation
tables between the orthographies be available to users from any area of the CAOLR,
so that they could go back and forth between the orthographies if needed. It was
also important that users be able to execute successful searches using any of these
orthographies, so the developers determined to create search functions that could
easily incorporate any of the orthographies.
For this collection, understanding how to provide appropriate access of the sort
described above was complex – but the question of appropriate access to the collection
as a whole (that is, whether or not the collection should be made available via the
 Shannon Bischoff & Amy Fountain
web in an unrestricted manner) was simple. In consultation with the then Director
of Language Programs, it was determined that the materials were not sensitive, and
broad access was appropriate. Furthermore, much of the previously published and
unpublished material was available in various other formats in the public domain
(though it was challenging to locate).
The volunteer developers utilized only publicly available, online web develop-
ment training, primarily using a resource called ‘w3schools.com’.14 This is one of many
readily available online resources devoted to learning web programming. The coding
tools needed were ‘hypertext markup language’ or HTML, which is the core resource
for creating static web pages, and two scripting languages: ‘JavaScript’ and ‘Hypertext
Preprocessor’, or PHP. Scripting is needed to provide dynamic content such as search
functions. In order to create a web resource that would provide access to a very large
number of digitized documents, as well as complex linkages among them, in a way that
would be seamless in appearance and also relatively efficient to maintain, the develop-
ers utilized ‘Cascading Stylesheets’ (CSS) as part of the HTML design for the site.
Fundamental understanding of these tools sufficient to develop the CAOLR were
all learned via w3schools.com, and through trial and error, during a single six week
period. Because the creators had worked with the materials contained in the resource
over several years, in various formats, and for various purposes, designing the content
and functionality of the site was relatively straightforward. Language workers inter-
ested in conducting similar projects using their own materials would benefit similarly
from their own familiarity with the specific resources at their disposal. In addition, the
then Director of Language Programs at the tribe had also considered how the material
might be organized in an online format and provided insightful suggestions for the
development of the CAOLR. We believe that a process like this one – in which workers
first identify the resources to be included, then establish a vision of how those resources
should be interlaced in order to maximize their usability for their communities, and
then go about acquiring a relatively limited set of technical skills needed to accomplish
that, can be repeated by most linguists and community members (although if develop-
ment teams are beginning with less familiarity with the resources that they hope to
archive, the design portion of the process would take significantly longer).
4.4 Building the CAOLR

The first step in creating the CAOLR was to digitize hard copy materials. Over 1,200
pages of unpublished field notes and typed manuscripts required digitization. These
were scanned to ‘portable document format’ (PDF), and stored in multiple l ocations, to
. 〈http://w3schools.com〉
A case-study in grass roots development of web resources for language workers 
ensure their safe storage. This format was selected because of its s tability over time, its
universal accessibility, and its omnipresence in the digital world (see Simons 2003, and
others for similar recommendations). Scanning hard copy documents to PDF requires
only readily accessible equipment (scanners can be p urchased for minimal cost, and
many schools, libraries and other public institutions can provide access) and time.
In addition, included in the site are several versions of Reichard’s 1939 Stem List.
Like the field notes, the stem list was first scanned to PDF. The PDF version of the
list is most useful to users who wish to download the resource, but it presents some
interoperability problems depending on users’ internet browser configuration and
download speeds. To accommodate these difficulties, PNG image versions of the
resources were created (using a freely available PDF to PNG converter). PNG image
files are accessible to nearly all web browsers (even some that do not gracefully sup-
port PDF), and so ensured that users wishing to view the list in a browser could do so.
Optical scanning of documents to PDF typically creates an ‘image’ file – like a
photograph of each page – rather than a ‘text’ file. In order for the contents of a file to
be searchable, ‘text’ format is needed, and for this project, creation of the text versions
of materials was undertaken by hand. Each of Reichard’s stems (the scanned stem list
was written in the Reichard orthography) were manually entered into HTML tables
in the Reichard, Nicodemus and Salish orthographies. This required only access to a
text editor (such as MS Notepad or Aquamacs Emacs, which are freely available) and
a familiarity with the conventions for writing the set of special characters required for
Coeur d’Alene using standard Unicode character encoding. The developers located
and used free tutorials for creating HTML tables at w3schools.com, and found and
utilized information about Unicode conventions provided online by the Unicode
Consortium.15
Coeur d’Alene is a polysynthetic language, and this makes a searchable affix list
quite desirable. No affix list had previously been compiled, but the morphological
analysis provided in Reichard’s (1939) stem list and her grammar (1938) provided all
of the information necessary to build one. As with the searchable stem list, an HTML
table containing roughly 200 affixes taken from these resources was manually created
and populated. Hand-entered HTML links were created to connect each of the affixes
to their respective entries in the original publication of Reichard’s 1938 grammar,
which was available online at the Internet Archive.16 Similarly, we developed search-
able version of Lyon and Greene-Wood’s (2007, based on Nicodemus’ 1975) dictionary
in root form.
. 〈http://unicode.org〉
. 〈http://www.archive.org〉
 Shannon Bischoff & Amy Fountain
Initially, it was expected that a good deal of material beyond the 1,200 pages of
field notes and manuscripts would need to be digitized for inclusion in the CAOLR.
However, an extensive search on the internet turned up the fact that much work had
already been digitized and made available online at various digital archives. For e xample
Reichard’s 1938 grammar in its original context in The Handbook of American Indian
Languages (Boas 1938) was discovered online at the Internet Archive. This allowed the
developers to simply link to the existing work rather than scan it ourselves and post
it online. We suspect that many communities may find early documentation of their
language, if any occurred, at online archives such as the Internet Archive. Linking
resources already online on a single website is one aspect that makes the development
of such projects easier than assumed.
These materials were all in the public domain, and so copyright issues were less
relevant than they might have been otherwise. In an abundance of caution, however,
the developers contacted the publishers of all incorporated work to ensure that this
usage was acceptable, and that it did not contravene any copyright or fair use issues.
The Lyon and Greene-Wood root dictionary (2007) was used by permission.
Included in the archive is also a call for suggestions, comments and other
thoughts regarding the archive. The archive is open in that there is no login or pass-
word protection. At this time we find no need for such measures as all of the material
is available in the public domain or traditional archives with no restrictions. However,
such m echanisms for protection could easily be created in a variety of ways – password
protection using HTML or JavaScript is a simple measure that can be used to protect
sensitive cultural material.
Finally, a list of orthographic conventions and comparisons was created. Arriving
at orthographic conventions was perhaps the most challenging step in the develop-
ment of content for the site. The difficulty was compounded by the fact that there
are no standardized spelling conventions for Coeur d’Alene, and thus the correct or
appropriate written representation of the language is still very much disputed (see
Kroskrity 2009 for similar examples). A page was created which shows the correspon-
dences among the systems (this is shown in Figure 1), and a system for searching and
replacing characters was generated to allow other resources to be transliterated from
one orthography to another.
This is one of the few development steps that seemed to benefit from the
developers’ expertise in linguistics. The Coeur d’Alene Language Programs played
a crucial role in determining the strategy that should be taken, in that they use the
Nicodemus orthography exclusively – and have found initial materials in the Salish
orthography to be “useless”. With this clear community preference for the Nicodemus
orthography, it was selected as the primary system for which to present examples in
the HTML pages created for this project. Transliteration to the Nicodemus orthog-
raphy from the Salish system used by Lyon and Greene-Wood was facilitated by the
latter’s description of their process (which went in the other direction). This allowed
A case-study in grass roots development of web resources for language workers 
us a method of going from the Salish to the Nicodemus orthography using a simple
search and replace algorithm.

Figure 1. Orthographic conversions
Transliteration from Reichard’s work, which used the Reichard orthography,

was not so simple. Reichard’s work was transcribed, as was much early fieldwork on
American Indian languages, with a great deal of phonetic detail not readily agreed
upon by contemporary users of the Salish orthography. The Nicodemus o rthography,
as it was meant to serve as a practical orthography for the community, captures less
phonetic variation, but perhaps better approximates a phonemic transcription of
words – which can approximate a standard spelling. Later scholars such Sloat (1966,
1968) and Doak (personal communication) have noted that some of the vowel
alternations that Reichard transcribed are not likely to be represented in the practical
system of Nicodemus or the Salish orthography. Thus, in simply transliterating from
Reichard’s texts to the Nicodemus and Salish orthographies, a given word may gener-
ate several different spellings.
A system for transliterating from the Salish system to the Reichard orthography
had already been done by Barthmeir (1990). This process was applied, in reverse, to
go from the Reichard orthography to the Salish one. Once that was accomplished,
the result could be transliterated into Nicodemus’ orthography. As noted above, the
result of this process was imperfect. Since users could also access the scanned original
files (in PNG or PDF) from Reichard’s field notes, as well as the translation table, they
could at least access all the steps taken in order to transliterate examples from any of
the three systems to any other.
For some communities, this sort of issue may well be a significant roadblock
in development of web-based archives. In this case, the benefit of having previously
unavailable resources made accessible outweighed the problems introduced because
of the complexity of orthographic conversions.
 Shannon Bischoff & Amy Fountain
Each ‘start’ page in the CAOLR is organized into frames. Frames are one way in
which a web page can be divided into a small number of areas, each of which can be
scrolled separately from the others. While frames are not the only, or necessarily the
most up-to-date, way of organizing web pages, they are easy to code and do not require
special PHP scripting. This is why they were selected for the first pass at the CAOLR.17
The start page of the CAOLR (Figure 2) is divided into two frames. The name
and copyright information for the site itself appear in the upper left corner. Below are
links to the various pages within the CAOLR. The second frame provides information
about the archive, links to various resources in the archive, brief information regarding
how the archive was constructed, and information regarding other resources available
online about Coeur d’Alene. The start page also includes a link to a website created
by linguist Ivy Doak18 which provides morpho-phonological and morpho-syntactic
analysis of a number of Reichard’s texts and other information about the language such
as an excellent grammatical sketch. A bibliography link is included so that users can
readily find complete information about where the content of the site originates, and
other resources available on Coeur d’Alene.
Figure 2. CAOLR start page19
. ‘Frames’ have become a ‘deprecated’ or ‘forbidden’ element in current versions of HTML,
meaning that it is not a recommended strategy for developers to use in creating resources
under current and future standards for HTML development. w3schools.com offers tutorials
for developers to update code to meet the specifications of more current versions of HTML.
The standard at the time of this writing, HTML4.1, calls for different strategies to accomplish
the same goals for which the CAOLR utilizes frames.
. 〈http://ivydoak.com/Coeurd’Alene/grammar/crgrammar.htm〉
. 〈http://academic.uprm.edu/~sbischoff/crd_archive/start1.html〉
A case-study in grass roots development of web resources for language workers 
The root dictionary start page (Figure 3) is divided into three frames. The top
frame includes copyright information, and on embedded pages it also includes an
alphabetic list of orthographic characters linked to sections of the dictionary for
browsing, and a search box that allows for searching the dictionary in the Salish or
Nicodemus orthography or English gloss.
Figure 3. Root dictionary start page20
The CAOLR required a search mechanism with functions not typically found in
freely available or built-in search features. For example, visitors needed to be able to
enter special characters (such as š, ʷ, č, ʔ, x̣, ɫ, ɛ, ə, á, é, í, ó, and ú) into the search box.
The developers used Javascript to present ‘clickable’ characters below the search box.
When clicked by the user, the characters are added to the search string in the box. The
search engine was then created using a combination of Javascript and PHP coding.
The lower frame is divided into two subframes. To the left is a frame contain-
ing a number of links: search, browse, and archive. The search link explains how the
search box at the top can be best used. The browse link brings up the dictionary itself
(Figure 4) and the archive link returns the user to the main page of the CAOLR.
The right frame provides information about the dictionary and a link to the
publisher’s website where the original paper copy can be ordered in book format. It
also includes a brief discussion of how the entries are organized and a further link
to greater detail regarding the organization of entries. Finally, a link to Ivy Doak’s
grammatical sketch (based on Doak 1997) is provided.
. 〈http://academic.uprm.edu/~sbischoff/crd_archive/start2.html〉
 Shannon Bischoff & Amy Fountain

Figure 4. Root dictionary ‘Browse’
When a user searches or browses the dictionary, the lower right frame opens
to display the dictionary entries (see Figure 4). The dictionary is organized by roots
indicated by the root symbol “√ ” followed by the root. Under each root are the r elevant
entries that contain the root. Each entry is numbered for ease of reference. The entries
begin with an entry in the Salish orthography in a column on the left, then in a center
column is the form in Nicodemus’ orthography, and finally the English gloss appears.
This was done using HTML tables. This is not how the forms are organized in Lyon
and Greene-Wood’s (2007) original work as they have been modified for the web.
Figure 5. Dictionary search
One useful feature of the Root Dictionary in web format is the ease of
searchability. As noted, we created our own search mechanism with PHP. The site’s
A case-study in grass roots development of web resources for language workers 
search mechanism locates all forms and returns them highlighted in their entry
format, with root, as illustrated in Figure 5, where a search for the lexical affix =ílt
‘offspring, child’ has been conducted. The search mechanism, in essence, turns the
dictionary into a searchable corpus that can be used for data analysis or pedagogical
purposes.
The affix list start page is formatted just like the root index start page (Figure 3).
Continuity in the design of each section of the resources (e.g. affix list, root dictionary,
stem list) is provided by utilizing CSS, and contributes both to the usability and ease
of maintenance for the site.
Selecting the browse link in the lower left frame of the affix list page brings
up the page found in Figure 6. Again, for consistency throughout the archive, the
entries to the far left are in the Salish orthography, the Nicodemus orthography is
used in the center column and the English gloss appears in the far right column.
In addition to the English gloss a page number and section number (in parenthe-
ses) are included and serve as a link to the entry of the form in Reichard’s original
grammar. For example the first entry is glossed ‘first, before’ and includes a link to
page 599 Subsection 420 of Reichard’s original work posted at the Internet Archive.
The link opens in a new tab. Figure 7 provides an image of the relevant entry brought
up after clicking the link.

Figure 6. Affix list ‘Browse’
Reichard’s 1939 Stem List was organized in the same fashion as the root d
ictionary
and the affix list. The layout and search mechanism followed exactly those of the
dictionary and affix list with the original article available as a downloadable PDF for
reference or a PNG file that opened within the website.
 Shannon Bischoff & Amy Fountain
Figure 7. Page 559 Subsection 420 of Reichard’s (1938) Coeur d’Alene Grammar
The Reichard texts were organized in a different manner. Selecting the Text link
on the main ‘about’ page brings up a similar image as that of the ‘about’ page. The
leftmost frame includes the same code as the same frame ‘about’ page. The right
frame contains information about the texts, links to Doak’s website with analysis of
a small number of the texts, and a link to Tiet’s 1917 publication of a limited number
of narratives c ollected by earlier informants that are similar to those c ollected by
Reichard. Links to Teit’s texts were incorporated to facilitate comparison between
them and Reichard’s.
Choosing the text list returns the page shown in Figure 8. In Figure 8 it can be
seen that there are two frames. The left frame contains the list of texts, forty-eight
in all. The texts are numbered in accordance to Reichard’s 1947 English publication
of the works. Below each title are links, where possible, to Reichard’s original field
notes, typed manuscripts, both, and the English translation published by Reichard
in 1947.
Figure 8. The text list

A case-study in grass roots development of web resources for language workers 

Figure 9. Both field notes and typed manuscripts
Each link opens in a new tab with a PNG file of the original. PDF versions of each
file are also available to be downloaded. Examples of both the field notes and typed
manuscript for the text Cricket rides Coyote are shown in Figure 9. The user can open
each file separately, but allowing the user to view both files simultaneously allows for
easier comparison of the texts, for philological purposes for example. Scroll bars for
each text, field notes and manuscripts, allow the viewer to move through the texts
comparing glosses, transcription, and other elements.
This effect was created by using an upper and lower frame. By selecting the English
link a scanned copy of the pages containing Reichard’s English publication of the text
appears in a new tab in PNG format as shown in Figure 10.
Figure 10. Reichard’s English versions

 Shannon Bischoff & Amy Fountain
Recently, Reichard’s original 1947 publication of the English translations and

analysis of the texts was discovered to be available online at the Internet Archive.
A link to the site will be added to the archive in the near future including links for each
individual narrative.
Developing the CAOLR resources described here required concentrated time and
attention by two highly motivated volunteers over the span of six weeks. The volun-
teers utilized a variety of freely available online technical documentation and training,
much of which might appear daunting to language workers. It is important to note,
however, that neither volunteer began the process with any previous knowledge of, or
training in, web development. Language workers who are willing and able to devote
the time to a project like this, and who are willing to overcome initial worries about
the feasibility of such a project, can accomplish a tremendous amount in a relatively
short time frame.
5. Conclusion
The above discussion by no means covers all the material in the archive, which will
continue to grow as more resources are discovered and developed. However, it does
give a general idea about the presentation and organization of the archive and the types
of materials available. More importantly, it serves as a case study in the d evelopment of
web-based archives for language communities that could be replicated (and improved)
in grass roots language projects.
The process of archive creation is often shrouded in mystery, and can be
extraordinarily time and resource intensive. There is certainly great value in large-
scale, externally funded and supported archiving projects and resources like OLAC,
PARADISEC and EMELD. But it is also the case that smaller-scale, grass roots
projects can make great strides towards online language archiving with fairly minimal
overhead, and in relatively short periods of time. These efforts can greatly improve
the safe storage and appropriate access to legacy materials – especially those materials
that are readily digitizable, and that are not particularly sensitive. Perhaps even more
importantly, they can allow communities to feature their own voices, intermingled
with those of the past, to preserve and protect resources in a way that best fulfills
their needs.
There are at least two key resources required for such efforts, however. First,
and most crucially, enthusiastic and energetic language workers can be successful in
these endeavors – provided they are able to devote an appropriate measures of time
and effort to it. As noted, the entire process (e.g. learning how to and b uilding the
CAOLR), took six weeks. They do not need to be technical specialists, and they do
A case-study in grass roots development of web resources for language workers 
not need p revious experience in web development. Second, workers need to have
access to a stable and relatively modern computer environment, with a reliable,
preferably high-speed, web connection. This is important because the freely avail-
able training and technical support resources they need are themselves online. This
makes community schools, which may have such resources, an ideal place to begin
such projects.
The CAOLR was developed and housed on server space provided by a
University, and we have not addressed the issue of finding and securing a web server
environment in this discussion in any detail. Accessibility of a secure and reliable
web server can be a roadblock for grass roots projects – but nearly all of the resources
we’ve d iscussed here (all those other than PHP scripting) could be developed and
displayed on a simple desktop computer. If the files were saved to portable media
such as CDs, thumb-drives, or external hard drives, they could be shared among
multiple computers within a community. For projects in which general publication
on the world wide web is not seen by the community as appropriate, this type of pub-
lication may even be preferable over actual web deployment. For projects in which
free publication is desired, language workers may pursue server space by approach-
ing local government (language and culture departments, for example), schools or
libraries.
In sum, the goal of this discussion has been to illustrate that grass roots web
archiving of endangered language material is possible, and that materials produced
in this way can go far in meeting basic requirements for safe storage and appropriate
access recommended by such large-scale projects as OLEC, PARADISEC and
EMELD. Community members with rapidly decaying paper, or other old media,
records need not feel that they have to wait for an external expert to take on their
cause, and they do not need to cede control over these resources to outside efforts
in order to protect, preserve, and appropriately share them. The energy, creativity,
and capability to do this work are not restricted to “experts” or outside agencies;
nor should it be. Technologies such as web development have the potential to sup-
port the expression of local knowledge in ways that serve the needs of community
members, academics, and policy makers without excluding local voices from the
conversation.
We count ourselves among those who have found the expert rhetorics
surrounding language resource preservation and access to be sometimes alienating,
and sometimes unintelligible. We intend this work to be a demystification of at least
some of these processes, and in doing so we hope the project will serve to empower
and encourage language activists, and academics and policy makers, to collaborate
in ways that respect and advance all of these heterogeneous communities’ distinct
varieties of local genius.
 Shannon Bischoff & Amy Fountain
References
Anderson, Jane & Koch, Grace. 2003. The politics of context: Issues for law, researchers and the
creation of databases. In Researchers, Communities, Institutions, Sound Recordings, Linda
Barwick, Allan Marett, Jane Simpson & Amanda Harris (eds). Sydney: Sydney University
Press.
Austin, Peter K., Munro, Robert & Nathan, David. 2005. Archives, linguists and language
speakers. Presented at the annual meeting of the Linguistic Society of America as part of
the Open Language Archives Community tutorial sessions, January 6, Oakland CA.
Barthmeir, 1990. A Dictionary of Coeur d’Alene from Gladys Reichard’s File Slips. MA thesis,
University of Montana.
Boas, Franz (ed.). 1938. Handbook of American Indian Languages, Vol. 3, New York NY:
J.J. August.
Boynton, Jessica, Moran, Steven, Aristar, Anthony & Aristar-Dry, Helen. 2006. E-MELD and
the school of best practices: An ongoing community effort. In Sustainable Data from
Digital Sources: From Creation to Archive and Back, Linda Barwick (ed.). Sydney: Sydney
University Press.
Brinkman, Raymond. 2003. Etsmeystkhw khwe snwiyepmshtsn: ‘You Know How to Talk Like a
Whiteman’. Ph.D. dissertation, University of Chicago.
Bird, Steven & Simons, Gary. 2003. Seven dimensions of portability for language documentation
and description. Language 79(3): 557–582.
Buszard-Welcher, Laura. 2001. Can the Web help save my language? In The Green Book of
Language Revitalization in Practice, 3rd edn, Leanne Hinton & Ken Hale (eds), 331–345.
Bingley: Emerald.
Chang, Debbie. 2010. TAPS: Checklist for Responsible Archiving of Digital Language Resources.
MA thesis, Graduate Institute of Applied Linguistics.
Doak, Ivy. 1997. Coeur d’Alene Grammatical Relations. Ph.D. dissertation, University of Texas
at Austin.
Falk, Julia S. 1997. Territoriality, relationships, and reputation: The case of Gladys Reichard.
Southwest Journal of Linguistics 16: 19–36.
Falk, Julia S. 1999. Women, Language and Linguistics: Three American Stories from the First Half
of the Twentieth Century. New York NY: Routledge.
First Archivists Circle. 2007. Protocols for Native American Archive Materials. 〈http://www2.
nau.edu/libnap-p/protocols.html〉 (16 May 2012).
Fitzgerald, Colleen. 2005. Documenting the documentation: The case of O’odham. Paper
presented at the Conference on Endangered Languages and Cultures of Native America,
University of Utah, April 8.
Fitzgerald, Colleen. 2009. Pathways for accessing legacy materials in Tohono O’odham.
Paper presented at the 1st International Conference on Language Documentation and
Conservation, University of Hawai’i at Manoa, March 14.
Flores Farfán, Antonio José & Ramallo, Fernando. 2010. Exploring links between documenta-
tion, sociolinguistics and language revitalization. In New Perspectives on Endangered Lan-
guages: Bridging Gaps between Sociolinguistics, Documentation and Language Revitaliza-
tion [Culture and Language in Use 1], José Flores Farfán & Fernando Ramallo (eds), 1–13.
A case-study in grass roots development of web resources for language workers 
Galla, Candace. 2009. Indigenous language revitalization and technology: From traditional to
contemporary domains. In Indigenous Language Revitalization: Encouragement, Guidance
and Lessons Learned, John Reyhner & Louise Lockard (eds), 167–182. Flagstaff AZ:
Northern Arizona University.
Hill, Jane H. 2002. ‘Expert Rhetorics’ in advocacy for endangered languages: Who is listening,
and what do they hear? Journal of Linguistic Anthropology 12: 119–133.
Johnstone, Barbara. 2011. Making Pittsburghese: Communication technology, expertise,
and the discursive construction of a regional dialect. Language & Communication
31: 3–15.
Kraus, Michael. 1992. The world’s languages in crisis. Language 68(1): 4–10.
Kroskrity, Paul V. 2009. Language renewal as sites of language ideological struggle: The need
for ‘ideological clarification’. In Indigenous Language Revitalization: Encouragement,
Guidance & Lessons Learned, Jon Reyhner & Louise Lockard (eds), 71–83. Flagstaff AZ:
Lewis, William D., Farrar, Scott & Langendoen, D. Terrence. 2006. Linguistics in the internet
age: Tools and fair use. In Proceedings of E-MELD 2006 Workshop on Digital Language
Documentation: Tools and Standards: The State of the Art.
Liberman, Mark. 2000. Legal, ethical and policy issues concerning the recording and p ublication
of primary language materials. In Proceedings of the Workshop on Web-based Language
Documentation and Description.
Lyon, John & Greene-Wood, Rebecca (eds). 2007. Lawrence Nicodemus’s Coeur d’Alene
Dictionary in Root Format. Missoula MT: UMOPL.
Macri, Martha, Golla Victor & Woodward, Lisa, J.P. 2009. Harrington project: Academic
and community participation. In Proceedings of the 11th annual Stabilizing Indigenous
Languages Conference [Report 14: Survey of California and Other Indian Languages],
Wesley Y. Leonard & Ethel B. Gardner (eds). Berkeley CA.
McCarty, Teresa & Wiley Snell, Alice. 2011. State of the field. Policy brief. 〈http://center-for-
indian-education.asu.edu/sites/center-for-indian-education.asu.edu/files/McCarty,%20
Role%20of%20Native%20Lgs%20&%20Cults%20in%20AI-AN-NH%20Student%20
Achievement%20%5B2%5D%20%28071511%29.pdf〉
Newman, Paul. 2007. Copyright essentials for linguists. Language, Documentation and
Preservation 1: 28–43.
Nicodemus, Lawrence G. 1975. Snchitsu’umshtsn: The Coeur d’Alene Language, Vol. I & II.
Spokane WA: University Press.
Nicodemus, Lawrence, Matt, Wanda, Hess, Reva, Sobbing, Gary, Wagner, Jill Maria & Allen,
Dianne. 2000. Snchitsu’umshtsn: Coeur d’Alene reference book. Coeur d’Alene: Omni
Graphics.
O’Meara, Carolyn & Good, Jeff. 2010. Ethical issues in legacy language resources. Language and
Communication 30: 162–170.
Penfield, Susan D., Serratos, Angela, Tucker, Benjamin V., Flores, Amelia, Harper, Gilford, Hill,
Johnny Jr. & Nora Vasquez. 2008. Community Collaborations: Best practices for North
American indigenous language documentation. International Journal of the Sociology of
Language 191: 187–202.
Reichard, Gladys A. 1938. Coeur d’Alene. In Handbook of American Indian Languages, Vol. 3,
Franz Boas (ed.), 515–707. New York NY: J.J. August.
 Shannon Bischoff & Amy Fountain
Reichard, Gladys A. 1939. Stem-list of the Coeur d’Alene language. International Journal of
American Linguistics 10: 92–108.
Reichard, Gladys A. 1947. An Analysis of Coeur d’Alene Indian Myths [The American Folklore
Society Memoir 41]. Philadelphia PA: American Folklore Society.
Reichard, Gladys A. 1958–1960. A comparison of five Salish languages. International Journal
of American Linguistics 24: 293–300; 25: 8–15; 25: 90–96; 25: 154–167; 25: 239–253; 26:
50–61.
Simons, Gary. 2003. The electronic encoding of text resources: A roadmap to best practice.
Paper presented at the EMELD Workshop on Digitizing and Annotating Texts and Field
Recordings, July 11–13, Lansing MI.
Sloat, Clarence. 1966. Phonological Redundancy Rules in Coeur d’Alene. Ph.D. dissertation,
University of Washington.
Sloat, Clarence. 1968. A skeleton key to Reichard’s Coeur d’Alene transcriptions. Anthropological
Tatsch, Sherry. 2004. Language revitalization in Native North America – Issues of intellectual
property rights and intellectual sovereignty. Collegium Antropologicum 28 (Supplement 1):
257–262.
Teit, James. 1917. Coeur d’Alene tales. In Folk-tales of Salishan and Sahaptin Tribes,
Franz Boas (ed.), 119–128. New York NY: American Folk-Lore Society. Kraus Reprint,
New York 1969.
Warner, Natasha, Luna, Quirina, Butler, Lindsay and van Volkinburg, Heather. 2009.
Revitalization in a scattered language community: Problems and methods from the

perspective of Mutsun language revitalization. International Journal of the Sociology of
Language 198: 135–148.
Yamada, Racquel-Mariá. 2011. Integrating documentation and formal teaching of Kari’nja:
Documentary materials as pedagogical materials. Language Documentation & Conservation
5: 1–30.
section 2
Approaches to the study of voices

and ideologies
Language contact, shift, and endangerment –
implications for policy
Spanish in contact with indigenous tongues
Changing the tide in favor of the heritage languages
José Antonio Flores Farfán

Centro de Investigaciones y Estudios Superiores en Antropología Social
In this chapter effects of indigenous languages on Spanish and vice versa are
discussed, raising a number of issues. These include a reflection on the variable
nature of languages against an ethnocentric idea of a single abstract entity called
(e.g. the Spanish, Nahuatl or Maya) “language”, which stems from monolingual
approaches to linguistic phenomena. Such diverse configurations of Spanish
and indigenous languages allows a characterization of different contact varieties
in their social, ideological and political realms. Therefore contact effects will be
treated holistically, closing the gap between different realms of the sociolinguistic
analysis, including a critique of previous reductionist approaches and its
implications from an actors’ perspective and their educational possibilities for
(e.g. Mexican) society as a whole.
Keywords: indigenous languages; Spanish; language contact; educational

implications
1. Introduction
Although presenting high indexes of bi and multilingualism, in Mexico contact stud-

ies are in a marginal, underdeveloped situation. This is due to a number of reasons.
For instance, historically, there was never a real need of a pidgin or the development
of a Creole1 in the history of Mexican multilingualism. In fact, there were a couple
of lingua francae in Prehispanic times, outstandingly Nahuatl, still in use as a lin-
gua franca up to the 18th century. In modern times, at least since the inception of
the Mexican nation, a monolingual ideology has been threatening the existence of
Mexican linguistic diversity. Ever since, the inroads of Spanish have been changing
. There is only one Creole language in Mexico, Seminol, which was not developed in Mexico
but rather is the result of Afro-American immigration to the country escaping from slavery
in the US. It is telling that this language has not received any attention from Mexican scholars.
 José Antonio Flores Farfán
the multilingual ethos that prevailed before Spanish invasion (and even in colonial
times). Even when contact studies are few, they include very important works,
mostly with reference to Mexicano (i.e. Nahuatl: e.g. Hill & Hill 1986; Karttunen &
Lockhart 1976; Lockhart 1992) and to a lesser extent Yucatec Maya (Karttunen 1985),
which have opened key queries to understand the future of endangered languages,
such as language syncretism, maintenance, resistance and shift (cf. Hill & Hill 1986;
Hill 1993).
Spanish in contact with indigenous languages includes a number of different vari-
eties, ranging from bilinguals’ to monolinguals’ speech. Efforts to capture this com-
plexity exist in the form of a theory of different contact phases which present specific
characteristic features, yet refer exclusively to indigenous languages. At least three
phases have been postulated to understand language contact history (cf. Lockhart
1992). In a nutshell, when compared to 16th century and colonial Nahuatl, contem-
porary Nahuatl is characterized by a stage demonstrating a wealth of (socio) linguistic
changes, including the blurring of a series of distinctions such as (1) The early leveling
of the plural form to all nouns; i.e. not exclusively limiting plural to animated entities
as in Prehispanic and early colonial Nahuatl e.g. te-meh ‘stones’, (2) The recent blurring
of possessive and absolutive paradigms: no-kone-w vs. no-kone-tl, ‘my son’, (3) The shift
from polysynthetic to analytic forms of the language: n-axka-w vs. in de newa, ‘this
is mine, my property’, (4) The incorporation of a series of Spanish phonemes to the
Nahuatl phonology (e.g. /n/) substituting Nahuatl ones (e.g. ʔ); (cf. nemi-n vs. nemi-ʔ
‘they live’). This is not to speak, even, of massive borrowing and extensive code switch-
ing (for more details see Example (1) and Flores Farfán 1999, 2008).2 Moreover, there
is a rich continuum between bilinguals and monolingual Spanish varieties.3 Analyz-
ing these varieties entails a number of implications in terms of research and so-called
applied issues. Thus contact history between indigenous people and Spanish invaders
is expressed in a series of continua between bilingual and monolingual varieties of
e.g. Spanish, Nahuatl or Maya, with different degrees of integration of for instance
morpho-phonological and lexical items. Different ideologies revolving around such
contact varieties abound. Outstandingly common perspectives as purism, which
condemn contact variability, frequently arise. In the case of the heritage languages,
in general such ideologies have favored language shift (cf. Hill & Hill 1986: passim;
Flores Farfán 2009).
. For one of the few contact linguistics works in Mexico, including the analysis of
indigenous Spanish cf. Hekking and Bakker (1999), who study Hñahñu (Otomi).
. Since there are very few published works available in English or even Spanish regarding
such varieties, especially indigenous Spanish varieties, I provide an appendix with the most
important features of Nahuas’ Spanish.
Spanish in contact with indigenous tongues 
Against such perspectives, in this chapter the actual educational possibilities

carried out by such rich and diverse variability, stemming from the influence of indig-
enous languages on e.g. the hegemonic languages, specifically Spanish, will be high-
lighted. For this purpose, after reviewing some of the main contact effects of indigenous
languages on Spanish, I will discuss some educational materials we have developed to
revaluate the influence of Nahuatl on Mexican Spanish, specifically a DVD and a book
entitled Las Machincuepas del Tlacuache ‘The Opossum’s Somersaults’, (see http://www.
youtube.com/watch?v=Drzu0eT8wUk).
2. Spanish in contact with indigenous languages
What effects have Spanish produced on indigenous languages? It is indeed possible to

point to a typology of languages in terms of their contact effects, ranging from cases
in which indigenous languages have very few (early) borrowings from Spanish yet
nativized: horse, sheep, etc.) such as Sierra Totonac, situations in which Spanish has
almost not influenced or its influence is only starting to be felt in the language (e.g. Seri
or even monolingual varieties of several indigenous languages such as Nahuatl, Maya
or Totonac) to cases in which heavy or massive impact of Spanish on (almost extinct)
specific varieties is attested (for an illustration see Example 1), passing through several
bilingual varieties. In a context of permanent endangerment, these very different situ-
ations can be organized as a continuum of survival and shift which pose a series of
dilemmas, as those that have been appropriately described as a syncretic project; i.e.
a situation in which the dilemma between language maintenance and shift is at stake
(Hill & Hill 1986).
In general contact phenomena have been interpreted from a one-sided approach,
reducing their complexity to one stream of the maintenance – shift continuum, most
of all emphasizing obsolescence and shift against the survival possibilities also posed
by the syncretic project. Speakers do find ways to give continuity to their endangered
languages, developing strategies in which Spanish has not always had a negative
impact on their languages; as an illustration, consider the bilingual pairs that have
emerged in Balsas Nahuatl with borrowings such as kristiano ‘one of us, person’
vs. kiixtiano, ‘stranger, foreigner, exploiter’ (stemming from cristiano “Christian”);
michiin ‘edible fish’ vs. peskaadoh ‘carved craft in the form of a painted fish for the
tourist market’ (derived from pescado ‘fish’), xaaxayaakatl ‘ritual mask’ vs. maskarah
‘carved painted mask for the tourist market’ (which comes from máscara ‘mask’),
chieempoh ‘village time’ tiempoh ‘city time’ (derived from tiempo ‘time’), etc. Yet, as
we will see the impact of Spanish on Nahuatl or Maya (or in general on endangered
languages) is much higher quantitatively and qualitatively speaking than the other
way around.
 José Antonio Flores Farfán
2.1 Linguistic characteristics of Spanish in contact with Nahuatl

Nahuatl is the most documented language of the American continent. Together with
its current use, its historical and literary legacy is comparable with any of the planet’s
classical heritages. Therefore, it has been possible to reconstruct the history of contact
between Nahuatl and Spanish with sufficient systematicity, especially in terms of the
influence that Spanish has had on Nahuatl in the different stages of their long mutual
contact (cf. among others Karttunen & Lockhart 1976; Lockhart 1992; Flores Farfán
1999, 2008).
The most outstanding characteristics of this contact have also left a clear cut mark
on Mexican Spanish, having an outstanding impact on the lexical ambit and also play-
ing a role in other levels of the linguistic analysis in monolingual varieties of Mexican
Spanish, outstandingly in central Mexico, not to speak of indigenous speakers’ Spanish.
From a synchronic point of view, the influence of Spanish on Nahuatl is stronger in
the most Hispanized varieties of Nahuatl, producing highly syncretic varieties of the
language. This can reach up to 50% in terms of different types of loans, resembling a
mixed language, as illustrated by the following example in which massive borrowing
from Spanish into Nahuatl is attested (Flores Farfán 1999: 141)
(1) paisanos de aqui tinochimeh. a ve tiwaalaweh de Copalillo sitio de
cabecera municipal neke neechilis oome información e tahameh pan dia
veintitrés veinticuatro de ne otii. ootiinenkeh o puerto de Zihuatanejo
que see c ompromiso titlaalikeh pa reuniones anteriores de que… lema
nodifundiroska que iipan lugares kampa tipian gente como partido waan
como organización campesina ugose kwakon otiakeh oke waan dirigentes de
más de trescientos comerciantes… nochi see acuerdo de que timoreuniroskeh
pan iipuestos kitlaaliskia mantas solo de que neextlatlaaliniaya see mapa
no…martes mas bien waallase compañeros Chilpancingo iiwan noche
tinmamakaskeh propaganda no sé si naan naan kipian Consejo para welis
mas teechapoyaroske, ya no kitokeh de que de acuerdo kidifundiroskeh
kiteeemachtiskeh iika gringos iika mexicanos con fines de semana… tan
de vacaciones pero gringos después de vacaciones normales de nikaan
México waalloweh hasta en… puerto como Zihutane. por esa parte
meechinformaroa de que no difundiroskia yehun dee. o problema nikaan
nikaan de un dee. ribera de río Balsas. seke cosas tlen keno nimeechilis
paisanos. tahameh nepan tomunicipio ee kipiakeh dificultades ee normales
de de elecciones iipan comunidades…kimateh xuun katka para…comisarios.
sin embargo ee como por ese lado tahameh ti de kse partidos de oposición el
perrete iiwan eh por otro lado de me seke comisarios del pri kinekia ¿verda?
kinekia cas peewaskia iigente iipan iipan puestos de comisarías.4
. Spanish materials are highlighted in black. The situation in which this text is produced
is in the context of the struggle against a hydroelectric dam that the Nahuas were able to
Spanish in contact with indigenous tongues 
Another much less studied subject is that of bilingual and even monolingual Nahuas’
Spanish and of indigenous people in general, in this case at all levels of linguistic analy-
sis (see nevertheless Flores Farfán 1998, 1999). As we will see, it is important to distin-
guish between different contact varieties as well as the intensity of contact phenomena
among different groups within Mexican society.
2.1.1 Phonology
The phonological phenomena of Nahuatl that are most problematic to a Spanish-
speaker are /λ/ (orthographically tl), /¢/ (orthographically ts in modern Nahuatl, tz in
classical and colonial Nahuatl), and the glottal stop /ʔ/ (orthographically written as ’ in
modern writing, rarely written as h in ancient documents, baptized by missionaries as
a saltillo ‘little jump’), /š/, and vowel length. In colonial times, these phenomena have
only been systematically noted by Carochi (1983).
We will briefly analyze the evolution of these phones in the history of the con-
figuration of so-called Mexican Spanish. Before going into details, we reiterate that it
is important to clearly distinguish different varieties of “Mexican” Spanish, such as the
inter-languages of speakers for whom Spanish is a second language, mostly indigenous
people, in contrast to speakers of monolingual Spanish varieties in different regions of
the country. Due to the extensiveness of this subject, I will only provide a few selected
examples from this enormous and almost unexplored diversity. In the first place, let us
consider the influence of Nahuatl on Mexican Spanish.
s uccessfully arrest. The discourse revolves around the political solidarity that another regional
indigenous organization lends to the Balsas Nahuas’ opposition to the hydroelectric dam
(cf. Flores Farfán 1999). Spanish materials are highlighted in black. Folks of this part of the
region, we come from Copalillo, head of the Municipio. We want to provide some piece of
information. The 24th of … we met at the Zihuatanejo port where we ratified the commitment
we acquired to disseminate the lemma [information identifying the political movement] in
[different] places where we have comrades affiliated as a party or as peasants’ organization
UGOSE, which has more than 300 hundred affiliated vendors… All agreed on meeting on
their stands and to put announcements [with political consigns] and a map too. It will be on
Tuesday that fellowmen from Chilpancingo will arrive and provide us with the whole propa-
ganda. I don’t know if you have a council that could give us a hand here. In this way we are
agreeing on disseminating information to inform gringos and Mexicans every weekend …
when they are on vacation although the gringos spend their holiday after our normal vacation
down here in Mexico. We came all the way to the Zihuatanejo port. There they are informing
me to be passing the word about the problem [you are going through] here in the Balsas river
[the potential construction of a dam]. There are other things to inform fellowmen. In our
Municipio we are experiencing normal difficulties in the communities as represented by the
elections… We don’t know who is going to be the Comisarios [head of the local communal
authority]. Nevertheless on the one hand … we [belong to] the opposition party PRD while
on the other hand as you know [candidates for] Comisarios from the PRI are also contending,
maybe they want to introduce their own people in the Comisarias.
 José Antonio Flores Farfán
One of the earliest records of Spanish-speaking in contact with Nahuatl, which

extends to contemporary Mexican Spanish speakers, referring to the use of /λ/, is pro-
vided in Luces del Otomí, published in the 19th Century:
El [nombre] Otomí, que dan los españoles, parece ser el mismo que dan los
mexicanos, aunque diminuto o mutilado. Es la razón de que los españoles no
pronuncian con perfección todos los términos mexicanos, principalmente los
que tienen la partícula tl… Y así se advierte… en la palabra flor… xochitl…xuchil.
Y esto puede haber acaecido [con la palabra otomí] que diciendo el mexicano
Otomitl, el castellano haya dicho Otomí (Anónimo 1893: 6).5
This is a marginal but significant mention of the use of /λ/ in terms of a phonological
substitution: /λ/ for /l/.6 The other option is its complete omission, as suggested with
the word otomí. Both forms (especially the former) are still used by Mexican Spanish
speakers to a greater or lesser extent. Another possibility is where the phoneme /λ/
unbinds into two segments, technically a dephonologization, resulting from the pho-
notactics of groups that are allowed in Spanish. In this sense, historically, most of the
Mexicanisms that are integrated into Mexican Spanish reinterpret the final segment
/λ/ in terms of an open syllable: -te: tomate, aguacate, chocolate, petate, metate, amate,
etc. Moreover, even other Nahuatl final segments are also at times reinterpreted as the
open syllable -te, over-generalizing this strategy, as in Nahuatl chan-tli which becomes
chan-te ‘home’, another well integrated loan in Mexican Spanish. On the other hand,
λ is at times reinterpreted as the sequence /kl/, of course a common consonant cluster
in Spanish, as in clavo ‘nail’, etc. For example, the food combination of corn cake stuffed
with, among other edible things, black beans or broad beans called tlacoyo is some-
times articulated as clacoyo. Notice that it is especially true in initial position where /λ/
is produced closer to its articulation in Nahuatl, as in the last example (tlacoyo), or in
locatives such as Tlalpan. As can be observed, even though the reinterpretation of /λ/
as /kl/ can be produced in initial position, it is typically words in Nahuatl that end in
-tli that are more subject to this. For example, the most common form of articulating
the Nahuatl word /i¢kwi:nλi/, “Mexican dog” (for more details see below), in Mexican
. The [name] Otomí, that the Spanish provide, seems to be the same as that provided by
the Mexicans, although diminutive or mutilated. This is the reason why the Spanish do not
pronounce all the Mexican terms perfectly, mainly those that have the particle tl… And thus it
is noted… in the word flower… xochitl… xuchil. And this could have occurred [with the word
Otomí] that while the Mexican would say Otomitl, the Castilian would say Otomí (Anonymous
1893: 6). (my translation).
. By the way, it is interesting that in certain varieties in which Nahuatl is reaching the brink
of extinction this phonological substitution is being deployed by Nahuatl speakers, getting
Nahuatl closer to Spanish – a clear index of language shift (cf. Flores Farfán 2008).
Spanish in contact with indigenous tongues 
Spanish is escuincle, meaning ‘child’. Chile chipoctli becomes chipocle or ‘sandal’ cactli
becomes cacle, even though these two examples could also be interpreted as a substitu-
tion, which in any case produces the same result in terms of a group of consonants that
is allowed in Spanish.
It is possible that /λ/ already forms part of the repertoire of Mexican Spanish
varieties from Mexico City, or at least is more integrated to the articulatory man-
ner of the Mexican way of speaking, popularly known as chilango (Mexico City
dweller), compared to the peninsular (i.e. Spaniards) ways of speaking, for example,
from Madrid, of whose inhabitants find it very hard to pronounce words such as
Tlatelolco – a neighbourhood in Mexico City.
Another phenomenon presented in the ways of speaking in Mexico City
that involves /λ/, and which is produced especially among the middle – mainly
intellectual – class, is its voicing in final position, which is nonexistent in Nahuatl.
It is common to hear this overcorrection with regard to this segment on the radio,
when the announcer says that a concert will be held in the sala Netzahualcoyotl. As
suggested by this example, the overcorrection is extended to /s/ that becomes /¢/,
“ts”, and so instead of the word in Nahuatl, pronounced as /nesa:walko:yoλ/ (with the
final voiceless λ) ‘fasting coyote’ (name of the famous poet from Texcoco from the
pre-Hispanic era), we have /ne¢awalkoyoλ/ (with /¢/ and the final voiced /λ/) and of
course no vowel length whatsoever.
With regard to /¢/, apart from the overcorrection mentioned earlier, the most recur-
ring phenomenon is its phonological substitution; /¢/ /č/: /ma:¢inkwepa/ /mačinkwepa/
“somersault, pirouette”, /malin¢i:/ /malinče/ “la Malinche (the interpreter of Cortés)”.
It has also been suggested that the retention and reinforcement of the final sibilant
-s is influenced by Nahuatl, once again in Mexican Spanish of central Mexico (Lipski
1994). Mexican Spanish is seseante (the Castilian phonemes /s/ and /θ/ are both pro-
nounced as [s]). This could have been partially reinforced by the nonexistence of such
phonological opposition in Nahuatl; i.e. the contrast between /θ/ and /s/.
It is also considered that /š/ has become integrated into Mexican Spanish,
though only in the form of loans, as in “xoloizcuintle o xoloescuintle o xoloes-
cuincle o soloescuincle o soloescuintle (from the Nahuatl word xoloitzcuintli,
literally= “monster-dog”, from Xolotl “monster” + itzcuintli “dog”) PRONOUNC.
The x is pronounced as /š/. Mute pre-colonial dog, having no hair”.7 Note that in
this quotation, the first forms that are enlisted are those with /š/, which is rather the
least frequent form in other cases of initial /š/ in Nahuatl, as in Xochimilco, “field of
flowers”, the floating gardens in Mexico City, generally pronounced as /sočimilko/.
. Gómez de Silva (2001) 〈http:www.academia.org.mx/dicmex.php〉. In the sake of space,

quotes are presented in English – all translations are mine.
 José Antonio Flores Farfán
An example in which the palatal form /š/ contrasts with /s/ is /šola/ ‘place name’:
/sola/ ‘alone’, even though this is a marginal example. All in all, unlike this situation,
from a historical viewpoint and compared to peninsular or Latin American Spanish,
it can be said that both /λ/ and /š/ have been gradually integrated into the repertoire
of Mexican Spanish, or rather, varieties of Mexican Spanish, since this is not the same
as, for example, the Spanish spoken in Yucatan, of which we will speak briefly later
on (cf. Lipski 2005).
2.1.2 Morphosyntax
When compared to the influence of Spanish on Nahuatl, not much can be said about
the influence of Nahuatl on Spanish in this field, especially when compared to the
influence of Spanish on Nahuatl, as Example (1) suggests. This reflects significantly
the asymmetry of the effects in one or other direction, a mirror of the dilemma posed
by the Mexicano syncretic project in terms of language retention or shift (Hill & Hill
1986). In other words, Spanish has had an impact at all levels of the linguistic analysis
of Nahuatl, notably yet not exclusively in highly Hispanized varieties; from phonology,
with the emergence of new phonological distinctions such as voice: kimaka ‘gives’ vs.
kimaga ‘hits’; and the loss of other phonological distinctions, such as vowel length or
specific phonemes, such as the glottal stop as witnessed also in the morphology; for
example, in terms of the convergence of the Nahuatl pluralizer with the Spanish one:
nemi-n instead of nemi-h ‘they live’ substituting the glottal stop, the adoption of the
Castilian agentive suffix substituting the Nahuatl one: tlawaan-k-ero for tlawaan-ketl
‘drunkard’, and even affecting the formerly obligatory clear cut distinction between
the absolutive and the possessive nominal markers: no-konee-tl for no-konee-w ‘my
son’. In the case of morphosyntax, a change from a polysynthetic structure to a more
analytical one is confirmed (e.g. kal-itik shifts to itik kahli ‘inside the house’). Among
other phenomena, at the pragmatic level we witness the emergence of salutations cop-
ied from Spanish, the borrowing of all types of functional particles, and the extensive
presence of code-switching and mixing. With regard to lexicon, it is not only massively
borrowed (recall Example 1), but complete subsystems have been displaced, such as
the twentieth based numeral system, of which only glimmers remain, giving way to the
Castilian decimal numeral system as in most Mesoamerican languages (for a detailed
analysis of the contemporary Nahuatl Spanish contact situation cf. Flores Farfán 1999,
2000a, 2007a, 2008).
In contrast, the influence of Nahuatl on Spanish is practically limited to the lexi-
con, and through it, to some morpho-phonological features, speaking of course of
the Mexican Spanish standard monolingual varieties. The morphological influence of
Nahuatl on Spanish is related to the reinforcement or predilection of certain lexical
forms or trends that are more often used in Mexican Spanish, such as the diminutive
form, a use probably reinforced by the Nahuatl diminutive -tsiin. For example, without
Spanish in contact with indigenous tongues 
denying the Latin origin of forms such as chico (small), in Mexico, the use of the word
chiquito might have a close relationship to the Nahuatl word tsitsiikitoon, ‘small, little,
child’.
Since these phenomena are relatively well-known in the literature concerning
Spanish in general and Mexican Spanish in particular, I will not dwell upon them
here. Instead, I will present the less approached elements, particularly those pertain-
ing to Spanish in contact with indigenous languages.8 In what follows, I introduce an
approach to the lexicon, in which I explore conflicting interpretations of the etymol-
ogy of the word gachupín in Mexican Spanish, which from a monolingual perspective
is attributed to Spanish (or vice versa only to Nahuatl). As we will see, the origin of
gachupin has at least a concurrent Nahuatl contribution and it is better understood
when appealing to a bilingual or syncretic approach, which is after all the point of view
of the actor, at least historically speaking.
2.1.3 Lexicon
Stating that the contribution of Nahuatl – and many other indigenous languages, nota-
bly Maya in the case of Spanish spoken in Yucatan – to the shaping of American –
Mexican – Spanish is limited mostly to the lexicon must not be interpreted as if its
contribution is minor or unimportant. On the contrary, the presence of indigenous
lexicon not only provides a sui generis physiognomy to American dialectal varieties,
but through it, other characteristic features have been added, like those succinctly
reviewed in the case of the phonology of Mexican Spanish, not to mention what it
represents in terms of sociolinguistic identity and conscience of ethnic differentiation
and social class, among other non minor facts.
In what follows, I will present an example that should be explained as part of a
Nahuatl-Spanish converging origin, an effect of bilingualism. Stemming from a mono-
lingual perspective, the word gachupín is a lexical item not recognized as a loan from
Nahuatl by the main gatekeepers of Spanish, the Real Academia Española (RAE).9
According to the Academia Mexicana de la Lengua: “gachupín, gachupina. (From the
Americanism cachupín a Spaniard who settles in America or from the Spanish word
cachopo “hollow or dried trunk”, from the word cacho, “broken pot: piece”.) masculine
. In this sense, given that the pronominal system of Mexican Spanish, as far as I can see,
has no relationship whatsoever with contact, I will not deal with it. Mexican Spanish uses the
familiar form of addressing tú, contrary to the more formal pronoun usted. In some regions
of Chiapas, and apparently also in Tabasco, people use the form of addressing vos. While stu-
dents are still being taught the form vosotros, in practice, ustedes is used for all cases of second
person plural.
. For other examples of converging interpretations of other loans from Nahuatl into
Spanish and vice versa, cf. Flores Farfán (2006).
 José Antonio Flores Farfán
and feminine pejorative. Spanish”.10 The etymology provided by the RAE website does
not coincide with the previous version: “cachupín, na. (From the diminutive Portuguese
cachopo, child). Masculine and feminine pejorative. Colloquial Americanism Spaniard
who settles in America”.11
The first case speaks of two possibilities, either it comes from the Americanism
(Mexicanism?) cachupin or from the Spanish cachopo, from cacho, “broken pot, piece”,
whereas the RAE establishes that it simply comes from Portuguese (!) cachopo.
Notice that neither entry explains the voicing of the voiceless velar plosive /k/, a
voicing that could have been exerted by a Spanish speaker or even a Nahuatl-Spanish
bilingual speaker as an overcorrection. On the other hand, the indistinctness of the
Spanish phonemes /o/ and /u/ is probably derived from the treatment they receive in
Nahuatl, i.e. as allophones, a phenomenon very well documented in the literature on
Nahuatl-Spanish contact (cf. e.g. Karttunen & Lockhart 1976 and Infra). If you search
for the word gachupín at the RAE website, it will redirect you to cachupín, with the
voiceless plosive, and that is where the word’s presumed origin is posted: “gachupín,
na. Masculine and feminine. Derogative. Cuba. Honduras and Mexico Cachupín”.12
In Mexico, it is gachupín and never cachopín, a form on which only the official
dictionary insists, probably so as to make its etymology more credible.13 Curiously, the
devoicing that occurs in cachupín would be typical of a Nahuatl-speaker – recall that in
Nahuatl, the voiced/voiceless opposition is a recent phonological innovation induced
by Spanish, as illustrated in passing in the morphosyntax section (also cf. Flores Farfán
1999, 2000a, 2007a). This is an indication that at least partially supports the idea of
gachupín being a double loan, as suggested by Karttunen and Lockhart (1976) and
Karttunen (1983). Moreover, Karttunen and Lockhart (1976) find gachupín as a loan
in a text from the mid 18th Century. In a note (1976: 138) in which they mention Lucas
Alamán, a historian from the 19th Century, they suggest a plausible Nahuatl origin,
based on which they postulate the idea of a double loan. As a matter of fact, this expla-
nation seems to be much more likely than that provided by the RAE: the voicing could
have been produced by Spanish-speakers, in addition to imposing the already pre-
sented substitution of /¢/ by /č/, resulting in gachupín, originally derived from /kak.’¢o.
pi:.nia:/ literally “to prod, prick with cacles ‘shoes, boots’”, from cactli, “sandals” by
extension, shoes, footwear and /¢o.’pi:.nia:/ [tzopinia], “to prick, pique”; that is, to prod
. 〈http://www.academia.org.mx/dicmex.php〉
. 〈http:///buscon.rae.es/draeI/SrvltConsulta?TIPO_BUS=3&LEMA=cachupín〉
. Although the form gachupín also appears in the website, cf. 〈http://buscon.rae.es/draeI/
SrvltConsulta?TIPO_BUS=3&LEMA=gachipín〉
. 〈〈cachopín. 1. m. Disus. A Spaniard who settles in America〉〉. 〈http://buscon.rae.es/drael/
SrvltConsulta?TIPO_BUS=3&LEMA=gachipín〉
Spanish in contact with indigenous tongues 
or pique with shoes or boots (which can be combined with spurs)”, something that the
Spanish probably did to the indigenous people, for example, based on the encomienda,
a semi-slavery institution on which the Spaniards based the exploitation of indigenous
populations. In Molina (1970 [1571]: 153), an early colonial friar who produced the
most outstanding lexical Nahuatl 16th century dictionary we find: “ Tzopinia.nite.pun-
çar a otro…” “to prick, pique someone”. The Castilian section of this source states:
“Punçador…tetzopiniani…” “Prickler”. In Karttunen’s (1983: 19) Nahuatl dictionary,
the word is also attributed a plausible Nahuatl origin, for historical reasons as well as
the viability of its phonological adaptation from Nahuatl to Spanish, as we have also
maintained with the elements expressed here.
The RAE ignores all these sources, together with the dictionary of Mexicanisms of
Francisco Santamaría (2000: 541), in which he also profusely documents the Nahuatl
origin of gachupín, which he identifies in much older documents than those sug-
gested by Karttunen and Lockhart (1976), going back to the 16th Century and passing
through the 17th Century (Santamaría 2000: 542; cf. Alatorre 1992).
Let me stress the possibility that forms that are analogical from a point of view of
phonetic homophony and semantics, concur to provide validity and strength to a spe-
cific use, which I assume is the case of the word in question. For this purpose I return to
the word gacho, which seems to me a clear example of the case in point. From an actors’
centered perspective, it can be related to gachupín, as with agachar (to cower), which
is where supposedly the Mexicanism gacho comes from, according to both the RAE
and the Academia Mexicana de la Lengua, respectively: “gacho,cha. (Related to cower-
ing). 5. Adjective. Mexico. Bad, nasty, unpleasant”;14 “gacho, gacha. (From the Spanish
gacho “bent, crooked”, from the verb agachar “to bend, contract”.) adjective. Bad, ugly,
unpleasant. bien gacho. … Very unpleasant.|| ¡qué gacho! …. How nasty!”15
Why is the word gacho only used in Mexico to refer to something bad or unpleas-
ant? One concurring possibility is that it is, in fact, related to gachupín, apocopate,
given that Spaniards were not precisely seen in good light in their history with the
Mexican population. But at the same time, the word could have had the same meaning
or an analogous one by independent routes, as suggested by the official sources. We
can say that the two possibilities, as in other cases that I have presented for Nahuatl
elsewhere, are concurrent, consonant with a bilingual approach. I claim that this type
of explanation is much richer, closer to the complexity of these matters, and therefore
agrees more with the history of Nahuatl-Spanish bilingualism (cf. Flores Farfán 1998,
1999, 2000a, 2006, 2007a) and their contemporary syncretic realities (cf. Hill & Hill
1986; Hill 1991).
. 〈http:///buscon.rae.es/draeI/SrvltConsulta?TIPO_BUS=3&LEMA=cachupín〉
. 〈http://www.academia.org.mx/dicmex.php〉
 José Antonio Flores Farfán
2.2 Some linguistic characteristics of Spanish in contact with Maya

Consider that a couple of generations ago – at the time of my grandparents – it was
pretty common to have Maya nannies (one of my grandparents had Maya as his first
and the other as his 2nd language). This trend has being swiftly changing, a clear
index of the advance of Spanish together with Maya urbanization and migration. I will
not go into the influence of Spanish on Maya here (for a discussion of such topic see
Karttunen 1985), which as a whole seems less than that of the hegemonic language on
Nahuatl. Yet Nahuatl has had more influence on the overall picture, i.e. on Mexican
Spanish in the face of other national varieties. Typologically we could argue that since
Maya, in contrast to Nahuatl, has most of all an analytic structure, it converges and
does not conflict morpho-syntactically with Spanish. The impact is rather to be found
in losses of lexicon and classifiers, most typically among young people.
All in all, Yucatec Maya is the language that has had more influence on regional
monolinguals’ varieties of all the country’s indigenous languages, competing with vari-
eties that also have a strong impact on National varieties of Spanish, notably the case of
Paraguayan and Andean Spanish (cf. Palacios Alcaine 2005a, 2005b; Cerrón-Palomino
2003). As suggested, we should make a clear distinction between different varieties of
the language, both on a national level as within these varieties. Therefore, it is appropri-
ate to speak of Mexican Spanish and distinguish it from Yucatec Spanish, and within it,
to speak of Yucatec Spanish in a substratum vs. an adstratum situation; that is, distin-
guish the monolingual varieties from the bilingual ones, even when one nurtures each
other, a practice that is not well established in the respective studies (cf. Lope Blanch
1987). For instance, Mexican Spanish is identified with the “prestige” variety of the
country, defined as the language of the media, high written literature, as the “standard”
central varieties of the Mexican Received Pronunciation (which are also in contact with
Yucatec Spanish via television and the radio among other ambits).
Some examples of the need for the distinction between monolingual and bilin-
gual speech are provided by Lipski (2005: 123–24), although with no explanation of
their origin: “Bilingual Yucatecs can use the redundant possessive in constructions
such as me dieron un golpe en mi cabeza (I received a strike to my head), te cortaste
tu dedo (you cut your finger)… su papá de Pedro (his father of Pedro)”. Yet, according
to my observations and introspection as a Spanish Yucatec speaker, the redundant
possessive is probably one of the features that has gone beyond bilinguals’ speech and
reached at least some Spanish monolingual varieties, especially the most colloquial
ones. As in many other Mesoamerican indigenous languages, the redundant posses-
sive is repeated, like in the expression from modern Nahuatl i-chaan de no-nan ‘my
mom’s house’ (lit. her house of my mother). These structures have become stronger
due to their use by bilinguals, at times having an effect on monolingual varieties, either
of Mexican or Yucatec Spanish, as in the case in point.
Spanish in contact with indigenous tongues 
In the case of Yucatan, Lope Blanch (1987) acknowledges the adstratum situation
that prevails in the peninsula, along with what he calls a pronounced polymorphism
and originality in the phonetic realization that acknowledges the influence of Yucatec
Maya. Yet, he declares that preference must be given to “systematic” (sic, read systemic)
explanations, fairly telling not only of a specific Hispanic centered approach which
considers the lexical vitality of Maya in Yucatec Spanish as “exotic” (1987: 8), but also
of a close knit language perspective, which implies a linguistics without speakers.
The Maya phonemes that are more problematic for a Spanish-speaker are the
series of glottalized segments (orthographically represented as ′), and the glottal stop
itself, as well as vowel length and its tone system. There is still a lack of systematic stud-
ies that account for the effects of these characteristics on monolingual as well as Maya
bilingual Spanish, especially the latter varieties and its features. The studies conducted
by Lope Blanch (1987) constitute revealing advances on the subject at issue. I will sum
up his conclusions here, so as to introduce some novel aspects for the development of
future research. In other words, introducing a few albeit key clarifications will allow us
to continue advancing with a better and greater understanding of the aforementioned
varieties of Yucatec Spanish, which I cannot exhaust here.
It is precisely through the lexicon that sounds characteristic of Maya have entered
regional Spanish, and this justifies granting them a joint treatment. Yucatec Spanish
allows segments considered as “exotic” in all positions; for instance, in final positions,
which in varieties of standard Spanish are prohibited, notably -m: Voy del Colóm a la
colonia Alemám por el pam ‘I’m on my way from the Colon [ice cream shop] to the
Aleman [a Merida’s neighbourhood] to fetch bread’. This is the only Maya phonologi-
cal segment (in final position) that is presented in Spanish words. Otherwise, similar
“exotic” executions entail loans: hach ‘the last, the one from the house (referring to
alcoholic beverages at a canteen)’, xix “crumbs, leftovers”, xtup ‘the smallest of the fam-
ily, youngest child’ (xocoyote in Mexican Spanish; i.e. from central Mexico). The vital-
ity of Maya lexicon in Yucatec Spanish is thus a fact. It is expressed not only by the
preference that its speakers have for Maya words, in contrast with the “normal” (“stan-
dard”) forms of Spanish, but also by their non-literal, pragmatic, versatile usages. For
example, the word xix, although literally refers to the “leftovers, residues” of food or
beverage, can also be applied to other situations: se pasó con el ultimo xix del semáforo,
‘he took advantage of the xix of the green traffic light’; i.e. he drove through the last
glint of amber traffic light, almost red.
Even though the analysis of the influence of glottalization and of the glottal
stop itself entails a significant advance in explaining aspirated and full-fledged
stops’ articulations, respectively, characteristic of the series of plosives in Yucatec
Spanish, once again it is necessary to distinguish between features of monolingual
from those of bilingual Yucatec Spanish. Lope Blanch’s (1987) research includes both
varieties without making a clear distinction, producing a homogenizing effect of the
 José Antonio Flores Farfán
ifferences, which exist in practice and cry out for a separate treatment. In this sense,
d
cases that are considered marginal can be attributed to the Spanish of bilinguals.
For example, devoicing the voiced plosives constitutes contact phenomena due to
the impact of indigenous language on Spanish as a second language; i.e. constitutes
part of bilinguals’ speech, specifically a phenomenon of under-differentiation of
the voiced: voiceless opposition, not productively exploited in Maya languages. For
example, in Tzotzil, a Maya language spoken in the state of Chiapas, such forms are
also established, such as supcomandante Marcos (one of the main leaders and port
parole of the Zapatista movement). Other phenomena that are likewise produced in
Yucatec and Tzotzil, are the dephonologization of the palatal nasal, with the crucial
difference that in Tzeltal it is restricted to bilinguals, whereas in Yucatec it has passed
the threshold of monolingual Spanish speakers, which makes it so unique, going
from niño to ninio ‘boy’.
Regarding the two forms in which polymorphism can occur according to Lope
Blanch (1987: 14), it is necessary to rank, if not discard, the second one, and analyze
in depth the hierarchal relationship and mutual determination into which he sub-
divides the first: conditioned polymorphism vs. free polymorphism. The first dis-
tinction, in turn, is subdivided into historical and linguistic. A deeper analysis will
probably reveal that the lexical variation between forms such as xtup, xtupito, tup,
tupito ‘youngest child’, includes not only historical and linguistic aspects, but is first
and foremost sensitive to sociolinguistic features such as type of interlocutor, inter-
actional context, degree of bilingualism and of course sociolinguistic competence.
In this specific case, as probably in several others, the differences might be pro-
ductively established as a speaker continuum, ranging from a more Maya (quasi)
monolingual (xtup) to a S panish monolingual one (tup, tupito). This suggests that
although such differences can be plotted on an individual level, they do not depend
on the speaker’s will, thus as with free variation, no such thing as free polymor-
phism exists. In this sense, syncretic forms such as tulish-e ‘dragonfly’, mulish-e
‘curly’, which have a final epenthesis to reproduce the Spanish pattern of the open
syllable, thus avoiding prohibited final consonants in standard Spanish, actually
constitute the least common form. It remains to be determined what type of speak-
ers use these forms (if at all!) and in which contexts. As a Yucatec native speaker
and a linguist, I suspect these forms were obtained via elicitation, corresponding to
induced lexicon with middle upper class Spanish s peakers with which Lope Blanch
probably had more contact.
The idea that systemic explanations must prevail constitutes a dogma in practice.
In my opinion, this stems from a Hispanic-centrism that is hard to sustain in situa-
tions of intense contact, linked to sociolinguistic over-determinations that award a
high value to the adstratum language, as in the case we are dealing with. For instance,
stating that the influence of Maya on Yucatec Spanish is “limited, weak”; i.e. marginal,
Spanish in contact with indigenous tongues 
secondary, to understand its evolution, contradicts the empirical evidence, not only
considering the volume and preference for lexicon of Maya origin and its impact on
the phonetics that prevail in Yucatec Spanish, but also its influence on other analytic
levels, such as semantics, in which the system has been partially reorganized accord-
ing to the semantic nature of specific Maya verbs. Consider for instance no lo busqué
‘I couldn’t find it; (lit. ‘I didn’t look for it’, in Maya there is only one verb for ‘to look
for’ and ‘to find’) – an expression only found in Yucatec Spanish. Even though these
types of elements do not allow speaking of the emergence of a semi-Creole, in some
ways they resemble one, as they imply a partial reorganization of the linguistic system
according to the adstratum language, no matter how minimal it may be. Although it
is important to search for complementary, concurrent explanations, in the sense that
there may be internal forces in Spanish that could allow to shed some light on the
presence of a phenomenon such as final -m, the “indirect, weak, limited” influence is
rather the one that stems from the internal Spanish drift, contrary to what Lope Blanch
sustains in several passages of his work (1987: 32–47, 59 & passim), probably due to
his own Hispanic (monolingual) voice and consequent ideology and biases. In this
sense, it becomes clear that academic work is also submitted to power and ideologi-
cal relationships, which are better captured considering a translinguistic approach to
language (cf. Hill & Hill 1986).
All this suggests that there is still a need to develop not only a methodology that
considers the specific weight of forces involved in the emergence of a specific phe-
nomenon, but also the need for deconstructing different antagonistic voices at work
in the production of academic (or other) voices, as part of the ecology of discourses in
its entirety. For instance, against the reductionist Lope Blanch’s and followers idea of a
prevailing systemic explanation, the phonotactic reinterpretation of the alveolar nasal
[n] as a bilabial [m], which could have originated in Maya, and later on entered to
Spanish through bilinguals’ speech and finally adopted by monolinguals, is most of all
due to matters of sociolinguistic order, as acts of identity in terms of speakers’ differen-
tiation from other Spanish varieties (i.e. Spaniard and Central Mexico varieties). This
would have led to the consequent weakening of the standard forms and its contiguous
“romance” (Romance)” (i.e. Yucatec Spanish). Limiting the explanation to “internal”
or even “external” linguistic forces constitutes an explanation which denies speak-
ers’ agency in the configuration of their own specific varieties and looks to impose
one single solipsist voice for phenomena which require complex and rich approaches
closer to much more ‘realistic’ focuses on languages. Deconstructing such monolin-
gual voices has a series of implications, including debating descriptive, analytical and
applied issues, as advanced in this chapter, not to speak of its political implications. Let
me provide another example, regarding the use of the clitics in Spanish, as conceived
and treated from different points of view, which often times entail antagonistic voices,
contradicting each other.
 José Antonio Flores Farfán
3. Investigating clitics in Spanish
The use of clitics lo and le in Mexican Spanish neatly differentiate Mexican from other
national varieties of Spanish (e.g. Madrid Spanish). For instance, consider the col-
loquial forms typical of Mexican Spanish óra-le ‘Ok, yeah, wow!’, quiúbo-le “What’s
up?”, ánda-le ‘come on!, alright!, OK, let’s go!’, hijo-le ‘Jesus!, Oh-oh!’, etc. Against pur-
ist approaches which consider these uses “barbarisms” or bad usages, López Austin
(1989) has convincingly established the Nahuatl origin of these forms. The author
actually suggests an explanation along the lines favored by the convergence of Spanish
dative form -le with Nahuatl exhortative ones, composed with the interjection c ue-le!,
such as in tlayecue-le! ‘Hurry up!’. As López Austin mentions, it is likely that the
Spaniards used -le as an emphatic form, as in trabája-le!, dá-le, ‘Work!’, péga-le! ‘Hit
it!’, to give orders to indigenous people, in which -le is attached to verbs marking an
indirect object and producing an emphatic effect. Under the influence and identifica-
tion with Nahuatl interjections, it was later on extended to the above mentioned forms
with different interjectional, exhortative or desiderative functions, contributing to the
wealth of expressive resources of the Mexican Spanish variety and the reinforcement
of its identity.
Even when the Nahuatlism cuele! ‘Get lost!’ or the syncretic form axcale (equiva-
lent to órale, but with the Nahuatl form axcan ‘now, today’) have probably become
obsolete, the use of the above mentioned forms is still robust in Mexico’s (mostly
central) Spanish varieties, in which as suggested (and insistently denied or at least
undermined by unilateral Hispanic viewpoints) Nahuatl has modulated its diversity.
If historically the influence of Nahuatl or other indigenous languages on Spanish has
been the condemnatory target of the Hispanic scholarly tradition, the situation of
indigenous Spanish – not to mention indigenous languages themselves – is even more
highly stigmatized and their varieties correspond to what is considered the ‘lowest’,
most ‘vulgar’ forms of speech in Spanish. Take for example the use of the accusative
clitic lo in indigenous Spanish (Nahuatl but also in several other indigenous groups
such as Totonac, Hñahñu and Maya), summarized in Table 1.
Table 1. Different functions of lo in indigenous (Nahuatl) Spanish
Accusative Dative Pronoun Article

(ac) (dat) (pro) (art)
LO LE TE/SE EL
LO LO LO LO
(The first row corresponds to clitics use in Mexican monolingual Spanish, while
the second refers to its deployment by the Nahuas).
Spanish in contact with indigenous tongues 
Respective examples are:

(2) ¿Lo quieres café? ‘Do you want coffee?’ (ac) instead of the standard form
¿Quieres café?
(3) Lo da Juan ‘He gives it to Juan’ (dat) instead of the standard form Le da
a Juan
(4) Lo vinieron traer Juan ‘They brought it to Juan’ (pro) instead of the standard
form Vinieron a traérselo a Juan
(5) Pero por lo momento ‘For the time being’ (art) instead of the standard form
Pero por el momento
Contrary to Spanish and English which are pro-drop languages, notice that Nahuatl
(and several other Mesoamerican languages) obligatorily marks object in the verb,
a fact that explains the replica to which transitive Spanish verbs are submitted to,
anchoring it to lo as the replica object (for one of the few discussions of this phe-
nomena see Hill 1987 and Flores Farfán 1999: 187–193). There are a number of
multi-vocal focuses regarding this issue. We have already alluded to the prescrip-
tive condemnatory gate keepers’ view which simply considers such uses a plain
error, calling it loismo, defined as the use of the AC instead of the DAT, as in (3).
From a totally different perspective, that of the actor and not any more of a (biased)
observer, the focus is on the performative attitude, which basically pursues the
achievement of illocutionary success. Thus while such forms as Usted los haces
pescados? ‘Do you (Honorific) carve (IT) fish?’ is considered perfectly valid from
speakers’ usage, who point to a specific speech act (a directive), the Hispanic aca-
demic gate keeper – who by the way can become quite influential regarding the lay
man – would consider such forms ungrammatical or rather – as we have done – at
best explain it as an unconscious effect of Nahuatl on Spanish; i.e. a consequence
of language contact.
It is almost exclusively the Nahuas (or as suggested several other indigenous
(e.g. Maya) speakers) who use the redundant clitic. Yet some Spanish indigenous
monolinguals ironically present several such Nahuatl traces in their Spanish eth-
nolects. This is an almost uninvestigated topic, raising several queries dealing not
only with Nahuas’ (or in general indigenous Spanish), but also the question of the
role and contribution of Nahuatl to Mexican monolingual Spanish varieties at this
level, to which we have only alluded to here. In any case, it remains to be determined
to which extent ethnolects have had an influence on monolingual Spanish-speaking
varieties.
A final example will enable us to vindicate a multi-vocal approach to language,
which encompasses linguistic research (e.g. description and documentation), lan-
guage use and linguistic theory, as has been developed by Hill and Hill (1986) and
in Hill (1993), among other works (cf. Flores Farfán 2009). This example involves
the Mexicano lexical form maaskeh ‘notwithstanding, no matter that, although’, or its
 José Antonio Flores Farfán
variants maaskih, maasih, maaseh, or even simply mas in Malinche Mexicano (Hill &
Hill 1986: 181). As I have depicted elsewhere (Flores Farfán 1998), maaskeh is a clear
example of convergence between adversative forms in Nahuatl and Spanish, stemming
from Nahuatl sixteenth century māciuhqui which in turn probably converged with the
old Spanish expression manque expressed in contemporary forms such as (por) mas
que ‘no matter how much…’. Let us stress that this interpretation neatly corresponds
to the perspective and usage of the Mexicano speaker; i.e. a bilingual one, against
unilateral viewpoints which consider it deriving from one single origin. Due to Hill
and Hill’s following quote (1986: 180), I have been able to reach the above mentioned
conclusion regarding the origin of maaskeh.
The adversative conjunction mas ‘but, although, even’ is very common in

Malinche Mexicano, but we have not heard it in Spanish. Mas often appears with
que or qui … The source of this element, māzqui/māzque may be an example of
syncretism between this form and a Spanish word. Suárez (1977) takes this form
to be from Spanish. However, Classical Mexicano had a closely similar form,
māciuhqui ‘nonetheless, anyhow’, and the coincidence between this form and
Spanish mas que, in nearly the same meaning, may have led speakers to merge
the two forms. (1986: 180)
Notice that as they recognize, the Hills did not register maaskeh in use in local Spanish
in la Malinche. Yet in fact Mexicano speakers in the Balsas region (where I have done
most of my research) do, especially in female speech from conservative Mexicano
communities, such as San Agustin Oapan:
(6) Esa mujer es chilapeña maaskeh esté viviendo aquí
‘That lady is from Chilapa [a City not in the Balsas region], no matter
she’s living here [in San Agustin Oapan]’
The use in (6) as well as the profuse use of maaskeh for instance in Nahuatl riddles or
as a leaving taking courtesy rule suggests that the form is more Nahuatl than Spanish;
yet, one cannot deny its analogous function in Spanish, which bilinguals ‘naturally’
concur with their native form. Its vitality is so crystal clear that it is calqued when
Balsas Nahuas greet each other in Spanish while leaving an interactional scene. When
leaving in Balsas Nahuatl one says (7.1) and replies with (7.2):
(7.1) ye niaw ‘I’m leaving’
(7.2) maaskeh teh ‘Goodbye’
Notice that the literal translation of maskeh teh into Spanish is aunque pues, a loan
translation which Balsas Nahuas directly replicate when speaking Spanish. Apart
from such robust usages, consider the ideological maneuvers which these forms
entail when judged from the position of a (racist) Mestizo, who could point to
the use of indigenous aunque pues as an impolite form which defines indigenous
Spanish in contact with indigenous tongues 
eoples’ rudeness, lack of “education” or “culture”, stressing his-her indifference

p
for the use of polite courtesy rules. From such ideological standpoints covert rac-
ism is also more commonly indirectly expressed, while mocking indigenous ways
of speaking, similar to those judgments associated to “junk” or “mock Spanish” as
depicted in Hill (1995). These conclusions have not only research implications, but
also applied ones. Let us now turn to consider some practical implications of these
and similar issues.
4. Building strategies for the defense of endangered languages
Consider some implications that such research as presented above, like the use of
no le aunque ‘it doesn’t matter’ in Mexican especially rural and indigenous Spanish
(which is probably related to the above discussion of both -le and maaskeh), entails
not only of and for the field of (contact) sociolinguistics. Think for example of the
need to develop an intercultural intervention which would contribute to solving the
potential interethnic misunderstandings nurturing conflicts that as suggested such
usages convey, together with the educational resources that knowing the character-
istics of indigenous or Mexican Spanish entail. For instance, at least part of a course
for teaching indigenous languages could be designed for Spanish-speaking students
of e.g. Nahuatl as a second language, or for indigenous bilinguals, emphasizing fea-
tures such as the obligation to mark object in the Nahuatl (transitive) verb (or the
restriction to possess kinship terminology or distinguish different paradigms for the
absolutive and the possessive, among many other facts). Based on this knowledge, it
is predictable that Spanish or English speaking students would drop objects following
their own native linguistic pro-drop pattern, producing forms such as -kwa tlaxkahli
instead of ki-kwa tlaxkahli, ‘(s)he eats tortilla’, a fact to which the instructor would
call his-her attention in order to start acquiring a native (e.g. Nahuatl or Maya, etc.)
competence.
Together with becoming an effective pedagogical tool, this type of reflection would
also allow starting to dignify such marginal, still highly stigmatized varieties, which by
the way are only beginning to be vindicated in the literature in the form of literary
works, a fact that could also be pinpointed to potential learners of these languages –
even favoring more analyses and development of these types of productions.16
. One of the few works that productively exploit indigenous Spanish in literary works is
Arguedas (2001) for Peruvian Spanish and Subcomandante Marcos and Taibo (2005) for the
case of Mayas’ Spanish.
 José Antonio Flores Farfán
As we have seen, received monolingual approaches have obscured the origin of

a series of words in Mexican Spanish which are not well established or are a matter
of debate, including pulque, ‘Drinking (maguey) sap’, taco ‘taco’, gachupín ‘Spaniard’,
machincuepa ‘somersault’ and many others. Another most obvious educational impli-
cation sustained on this diversity includes reflecting on such integrated indigenous
lexicons (outstandingly Nahuatl) in Mexican Spanish varieties, including famous
words also present in English (tomato, avocado, tamale, taco). We have already worked
on and are facing the practical facet of this knowledge in a variety of ways including a
book and a DVD, entitled the Machincuepas del Tlacuache ‘The opossum somersaults’.
This product combines live image and animation in the form of opossum’s cartoons
in which selected Nahuatl names of the Mexico City metro are depicted and played
around with, instigating viewers’ curiosity and interest, looking to attract a wide audi-
ences’ attention. The basic idea is that in an enjoyable way viewers build upon preexist-
ing words already integrated in their Mexican Spanish repertoire, such as numerous
names in maps of cities in Mexico, especially in Central Mexico. Visit 〈http://www.
youtube.com/watch?v=Drzu0eT8wUk〉
The Tlacuache ‘Big eater, the opossum’, is one of the most important Mesoameri-
can tricksters, the host of the Machincuepas and other products we are working
with. It is a Mesoamerican marsupial with several varieties, which represents the
Mesoamerican Prometheus, since (s)he was responsible for donating fire to human-
ity. Its name derives from tla-kwaa “(s)he eats something” and -tsiin, the honor-
ific, affective diminutive suffix, turned into -che (recall that Nahuatl /¢/ becomes
Spanish //č/). Thus elaborating upon the actors’ own linguistic competence, Mexican
users, especially children, are invited to actively reflect on the lexicon which they
use on an everyday basis, opening a viewpoint in which their own repertoire is con-
ceived as an opportunity to approximate and reflect on a knowledge which is at the
same time so close but yet so far. In this sense, with these materials the user is made
aware for instance that Nahuatl /¢/ becomes Spanish /č/ and that (s)he can recon-
struct the original form of the word applying a very simple, at hand methodology.
Thus tepa-che ‘indigenous sap made of pineapple’ becomes tepatsiin, nan-che, ‘type
of edible fruit’, becomes nantsiin, etc.
The same is true for /λ/, a segment that as we have seen becomes an open syl-
lable when entering Spanish; the inverse route, reaching the indigenous language,
is what the user is invited to stroll through: toma-te: toma-tl ‘tomato’, aguaca-te:
awaka-tl ‘avocado’, etc. alongside explaining with somersaults on the syllables that
Nahuatl stress is on the penultimate syllable, with such words as Chapultépec,
Cuitláhuac, Xochimílco, all metro stations which meaning our trickster deciphers.
Together with these joyful brief invitations, with which we hope to inspire curiosity
on the part of the user, short but very powerful Nahuatl texts are introduced in the
Spanish in contact with indigenous tongues 
Tlacuache’s tour through the metro guts, such as tongue twister-riddles and a short
song in Nahuatl:
Otlica tecuatica ca titotecunia
‘When along the way you go, it will bite and make you fall!’
Ya tiawe compañero tipaxiaalo te Maria, oome yehuaalootsiin waan tonaali,
Santa Maria Guadalupe
‘We are leaving companion, we are strolling te17 Maria, two venerable
rounds and one day, [Virgin] Holly Mary Guadalupe!’
A project oriented to revaluate this legacy through such initiatives should be devel-
oped of course with mainstream Spanish speakers as well as with the Nahuas (or other
indigenous populations) themselves, a project we have been working with for over a
decade (see Flores Farfán 2011). Recall that it is indigenous people that have to learn
Spanish (or English!), not the other way around, a unidirectional inter-culturality that
the production of such materials looks to overcome while at the same time vindicating
endangered languages at the indigenous community level.
5. Concluding remarks
It is important to emphasize that unilateral positions that bet on a single explanation,

much less translated into sterile polemics in the sense of opposing a Hispanic
tradition vs. an Indigenous one for the aforementioned phenomena, have inhibited
the development of a methodology for understanding the hierarchies of concur-
ring forces, as I have suggested here and elsewhere. Thus, for example, in the case
of performing the series of plosives in Yucatec Spanish without fricativization, as
full plosives, the concurrence of the effect of the adstratum – the presence of the
glottal stop that brakes the fricative articulations of the plosives, re-syllabification
and links – can be postulated along with an internal development detonated by
that same adstratum effect, together with the archaistic conservatism of Mexican
Spanish, and particularly of the Yucatec Peninsula (cf. Lope Blanch 1987: 91) where
the overall sociolinguistic determinations not considered in such works play an
outstanding role.
Opposing the multiple forces which in practice play a role in the configuration of
specific phenomena is partially responsible for the poor advancement of the state of
the art in contact studies, at least in Mexico, which still require more and better inves-
. Te is translated as “pues” in Spanish, something like “well” in English.

 José Antonio Flores Farfán
tigations, not to speak of its applied implications, expressed in a huge vacuum in this
respect to which I have briefly referred to, exploring links between so called “basic”
and “applied” research. In other words, opposing ideologies revolving around contact
Spanish have inhibited the advancement of contact studies, not even scratching the
surface of its applied issues (a similar situation is reported for countries such as Peru,
cf. Cerrón Palomino 2003). This has also implied the lack of a clear methodology to
unravel the specific weight of internal-external forces in the configuration of such
phenomena. A concurrent rather than a one-sided approach is much more realistic
for language contact studies.
As suggested, critically reviewing such one-sided approaches, allows developing
a perspective which goes beyond monolingual conceptions of language, often linked
to purism (Hill & Hill 1986: passim), especially in the case of endangered languages.
In turn this allows debating the concept of language itself. Developing a more real-
istic approach to specific “languages” in terms of varieties is sensitive to all types of
contextual determinations and in fact closer to an actor’s perspective approach. This
phrasing seems much more appropriate to unravel the complexity of the syncretic
project and its dilemmas – as the Hills and I following them have try to suggest in
this contribution.
References
Anónimo. 1893. Luces del idioma otomí o gramática del idioma que hablan los indios otomíes en
la República Mexicana. México: Imprenta del Gobierno.
Alatorre, Antonio. 1992. Historia de la palabra gachupín. In Scripta philologica in honorem
Juan M. Lope Blanch a los 40 años de docencia en la UNAM y a los 65 años de vida, II:
Lingüística española e Iberoamericana, Elizabeth Guadalupe Luna Traill (ed.), 275–302.
México: UNAM.
Arguedas, José María. 2001. El Zorro de Arriba y el Zorro de Abajo. Lima: Horizonte.
Carochi, Horacio. 1983. Arte de la Lengua Mexicana: con la declaración de los adverbios della.
México: UNAM-Facultad de Ciencias Políticas y Sociales.
Cerrón, Palomino. 2003. Castellano andino. Aspectos sociolingüísticos, pedagógicos y gramatica-
les. Lima: Fondo Editorial PUCP.
Flores Farfán, José Antonio. 2011. Keeping the fire alive: A decade of language revitalization in
Mexico. Small Languages and Small Languages Communities 69. International Journal of
the Sociology of Language 212: 189–209.
Flores Farfán, José Antonio. 2009. Variación, ideologías y purismo lingüístico. El caso del mexicano
o náhuatl. México: CIESAS.
Flores Farfán, José Antonio. 2008. The Hispanisation of Modern Nahuatl varieties. In Hispan-
isation. The Impact of Spanish on the Lexicon and Grammar of the Indigenous Languages
of Austronesia and the Americas, Thomas Stolz, Dik Bakker & Rosa Salas Palomo (eds),
27–48. Berlin: Mouton de Gruyter.
Flores Farfán, José Antonio. 2007a. Buenas prácticas en la revitalización de lenguas. In La
Romania en interacción: Entre historia, contacto y política. Ensayos en homenaje a Klaus
Spanish in contact with indigenous tongues 
Zimmermann, Martina Schrader-Kniffki & Laura Morgenthaler García (eds), 675–689.

Frankfurt & Madrid: Vervuert & Iberoamericana.
Flores Farfán, José Antonio. 2006. La Malinche: Portavoz de dos mundos. Estudios de Cultura
Náhuatl 37: 119–139.
Flores Farfán, José Antonio. 2000a.Transferencias náhuatl-español en el Balsas. Reflexiones sobre
el desplazamiento y la resistencia lingüística en el náhuatl moderno. Amerindia 25: 87–106.
Flores Farfán, José Antonio. 1999. Cuatreros Somos y Toindioma Hablamos. Contactos y Conflictos
entre el Náhuatl y el Español en el Sur de México. México: CIESAS.
Flores Farfán, José Antonio. 1998. On the Spanish of the Nahuas. Hispanic Linguistics 10(1): 1–41.
Gómez de Silva, Guido. 2001. Diccionario breve de mexicanismos. URL: http:www.academia.
org.mx/dicmex.php
Hekking, Ewald & Bakker, Dik. 1999. El Otomí y el español de Santiago Maxquititlán: Dos
lenguas en contacto. Foro Hispánico 13: 45–74.
Hill, Jane H. 1993. Spanish in the indigenous languages of Mesoamerica and the Southwest:
Beyond stage theory to the dynamic of incorporation and resistance. Southwest Journal of
Linguistics 12: 87–108.
Hill, Jane H. 1995. Junk Spanish, covert racism, and the (leaky) boundary between public and
private spheres. Pragmatics 5: 197–212.
Hill, Jane H. 1991. In Neca gobierno de Puebla: Mexicano penetrations of the Mexican state. In
Indians, Nation-States and Culture, Greg Urban & Joel Sherzer (eds), 72–94. Austin TX:
University of Texas Press.
Hill, Jane H. 1987. Spanish as a pronominal argument language: The Spanish interlanguage of
Mexicano speakers. Coyote Papers 6: 68–87.
Hill, Jane H. & Hill, Kenneth. 1986. Speaking Mexicano. Dynamics of a Syncretic Language in
Central Mexico. Tucson AZ: The University of Arizona Press.
Karttunen, Frances. 1985. Nahuatl and Maya in contact with Spanish. Texas Linguistic Forum
26: 1–135.
Karttunen, Frances. 1983. An Analytical Dictionary of Nahuatl. Austin TX: University of Texas
Press.
Karttunen, Frances & Lockhart, James. 1976. Nahuatl in the Middle Years: Language Contact
Phenomena in Texts of the Colonial Period. Berkeley CA: University of California Press.
Lipski, John. 2005. El español de América: los contactos bilingües. Historia de la lengua española,
1117–1138. Barcelona: Ariel.
Lipski, John. 1994. Latin American Spanish. New York NY: Longman.
Lockhart, James. 1992. The Nahuas after the conquest. A Social and Cultural History of the
Indians of Central Mexico, Sixteenth Through Eighteenth Centuries. Stanford CA: Stanford
University Press.
López Austin, Alfredo. 1989. Sobre el origen del falso dativo -le del español de México. Anales
de Antropología 26: 407–416.
Lope Blanch, Juan M. 1987. Estudios sobre el español de Yucatán. México: UNAM.
Molina, Fray Alonso de. 1970 [1571]. Vocabulario en lengua castellana y mexicana y mexicana y
castellana. Edición facsimilar. México DF: Editorial Porrúa.
Palacios Alcaine, Azucena. 2005a. Lenguas en contacto en Paraguay: Español y guaraní. In Var-
iedades lingüísticas y lenguas en contacto en el mundo de habla hispana, 1st ed., Carmen
Ferrero Pino & Nilsa Lasso-von Lan (eds), 35–43. Bloomington IN: Books Library.
Palacios Alcaine, Azucena. 2005b. La influencia del quichua en el español andino ecuatoriano.
In Variedades lingüísticas y lenguas en contacto en el mundo de habla hispana, Carmen
Ferrero Pino & Nilsa Lasso-von Lan (eds), 44–52. Bloomington IN: Books Library.
 José Antonio Flores Farfán
Santamaría, Francisco. 2000. Diccionario de mexicanismos. México: Porrúa.

Suárez, Jorge A. 1977. La influencia del español en la estructura gramatical del náhuatl. Anuario
de Letras 15: 115–164.
Subcomandante Marcos & Ignacio Taibo II, Paco. 2005. Muertos Incómodos. México: Joaquín
Mortiz.
Websites
Flores Farfán, José Antonio. CIESAS

〈http://www.ciesas.edu.mx/jaff/multimedia.html〉
Gómez Silvia. 2001. Academia Méxicana de la Lengua
〈http://www.academia.org.mx/dicmex.php〉
Real Academia de la Lengua Española
〈http:///buscon.rae.es/draeI〉/
Machincuepas del Tlacuache. 2009
〈http://www.youtube.com/watch?vDrzu0eT8wUk>
Appendix
Main characteristics of Nahuas’ Spanish

Phonological features
– [o ~ u]: cumu tipubres for como tipobres ‘’Since we are poor’ (these vowels are allophones
in Nahuatl)
– [γ ~ k]: amiko for amigo ‘Friend’ (under-differentiation of the voice: voiceless Spanish
distinction since Nahuatl originally does not make this distinction)
Míyel for Miguel (substitution) ‘Michael’
–
Límoon for limón (nativization) ‘lemon’, méloon for melón ‘melon’
–
Anke for aunque ‘although, no matter’ (there are no diphthongs allowed in the Nahuatl
–
syllable) regunion for reunión ‘meeting’ quilavo for clavo ‘nail’ polatano (polan) for plátano
‘banana’ (epenthesis)
Morpho-syntactic-semantic replicas from Nahuatl into Spanish

cien peso for cien peso-s ‘100 pesos’ (numerals already pluralize in Nahuatl)
–
no soy de acuerdo for no estoy de acuerdo, ‘I don’t agree’ (there’s only one verb for ser and
estar, like in English)
su-s problema for su problema ‘their problem’ (Nahuatl has in- as the third plural
–
possessive and does not require concordance)
su-s casa for su casa ‘their house’, (Nahuatl in-kal); su-s mole, su-s atole for su mole, su atole
–
‘their mole, their atole’
Spanish in contact with indigenous tongues 
¿usted los vendes pescados? for ¿usted vende pescados?

–
‘Do you sell fish?’ (Nahuatl obligatorily marks object in the verb as in the equivalent
Nahuatl form:
ti-k-tlanamaka peskaados?
–
you-OBJECT-sell fish
–
– el costumbre for la costumbre (no gender distinction)
– ‘The tradition’
el violinist-o for el violinista ‘the violinist’ (there are no violinist females either)
–
un mula for una mula ‘a mule’ derived from the double loan see, originally the numeral
–
one, which has become an indefinite article in Nahuatl, as in see chichi ‘a dog’)
– está queriendo mujer for quiere casarse
– ‘He wants to get married’
Derived from ki-nek-tok siwaatl
– Object-want-PROGRESSIVE female
díselo Juan for díselo a Juan ‘Tell Juan’ (no prepositions: deletion)
–
– ¿Qué haces a Holanda? for ¿Qué haces en Holanda?
– ‘What are you doing in Holland’ (alternation of prepositions)
LOISMO (redundant use of the clitic lo, considered the most vulgar form by the RAE)
(see Table 1)
– yo lo conozco su hija for yo conozco a su hija
– ‘I know her daugther’
– lo platicó nada más a Cesar for le platicó nada más a César
– ‘He only talked to Cesar’
– los vas a ir poniendo las botas for te vas a ir poniendote las botas
– ‘You are going to put your boots on’
– por lo momento… for por el momento
– ‘For the time being’
– aunque pues for adiós
– ‘Good bye’ (derived form maaske teh, a leaving taking courtesy rule)
– escuchar for entender, comprender
– ‘Listen, understand’ (derived from -kaki one single verb for such verbs in English and
Spanish)
– como no escuchan nada de español for no entienden el español
– ‘She doesn’t understand any Spanish’.
Nahuas’ Spanish is also rich in ARCHAISMS, for instance:

barreta ‘sowing instrument’
–
culantro for cilantro ‘coriander’
–
de presto for rápido ‘quick, fast’
–
moza, mozo for novia, novio ‘girl friend, boy friend’
–
silleta for silla ‘chair’
–
How can a language with 7 million
speakers be endangered?
Heidi Orcutt-Gachiri
The stable trilingualism characteristic of older Kenyans is not shared by many

young Kenyans. The shift away from Kenyan indigenous languages is affecting
even large languages such as Gĩkũyũ. Where other scholars have been dismissive,
claiming there is no threat to Gĩkũyũ, Jane Hill has strongly supported this
research and has noted that the case of Gĩkũyũ shows the prescience of
Krauss’s (1992) point that the only safe languages are those that have institutional
support. Using Lieberson and McCabe’s (1982) domains of use in an analysis
of Kenyan language ideologies in the discourses of nationalism, education, and
development, I show how constellations of language repertoires that exclude
indigenous languages have contributed to the striking shift to bilingualism among
young Kenyans.
Keywords: indigenous languages; language endangerment; languages of Kenya
1. Introduction
Batibo (2005) identifies a triglossic structure for sub-Saharan Africa – 3 ± 1, with a

language of wider communication, a regional lingua franca (plus possibly a second
regional lingua franca), and an indigenous language. Sub-Saharan Africans typically
speak a predictable triglossic combination. In Kenya, English is the official language,
and Kiswahili is the national language. Until several decades ago, an educated Kenyan
was likely to speak English, Kiswahili, and at least one indigenous language.1 Stable
triglossia (Abdulaziz 1982: 100), which held from Independence in 1963 through the
1980s, is no more. Currently, a diglossic structure of English and Kiswahili is all that
is needed for most Kenyan young people if they are educated; those who are not need
only one language if they speak Kiswahili.
. Indigenous languages are called by the general term “mother tongue” in Kenya, which
does not signify that they are first languages. I use the terms interchangeably here.
 Heidi Orcutt-Gachiri
I conducted ethnographic fieldwork in two secondary schools2 in Kenya in 2004

to understand the reasons for the shift to bilingualism. My data collection process
consisted of conducting participant observation in classrooms, the teachers’ lounge,
and special school events in which I was able to participate; taking fieldnotes as part
of this process; audiotaping multiple classes; videotaping a limited amount of classes
in the rural school and videotaping only one event in the urban school; audiotaping
interviews with most of the students with whom I worked, some of their teachers, and
two of their parents; and collecting newspaper articles on a wide variety of topics. The
students I worked with ranged in age from early teens to early twenties. The students
in the rural school all spoke an indigenous language (not always Gĩkũyũ because some
students came from other regions or had parents who spoke other languages), whereas
many of the students in the urban school did not speak an indigenous language.
Between the teachers and students, approximately 25 indigenous languages were
represented at the urban school.
An important part of my analysis consisted of examining language ideologies3
of English, Kiswahili, and Kenya’s 53 indigenous languages4 in the context of Kenyan
discourses of nationalism, education, development, and tribalism.5 English indexes6
education, development, and Westernization, Kiswahili indexes nationalism, and
mother tongue indexes tribalism and parochialism.7 Here, I will show how language
ideologies are contributing to the endangerment of Kenya’s indigenous languages, with
a specific focus on Gĩkũyũ.8
How can a large language like Gĩkũyũ, with 7,180,000 speakers (SIL 2012), be
at risk? It is important to note that the size of the speaker population is not the only
condition for a language to be considered endangered (Nettle & Romaine 2000: 9).
Some of the most striking evidence of language shift away from Gĩkũyũ comes from
. Utheri, a rural public school in Kirinyaga District, and Elewa, an urban private school in
Nairobi.
. Language ideologies are ideas about a language that are at heart ideas about the people
who speak a language (Woolard 1998: 3).
. Summer Institute of Languages (SIL) (2009).
. The Kenyan discourse of tribalism is a political discourse that has at its heart the cry that
one group is hoarding the national cake over another group; the intention is to incite ethnic
groups against one another, particularly during elections, so that they will not form alliances
that could be politically disadvantageous to other ethnic groups (divide and conquer).
. See Ochs (1990) for the concept of indexicality.
. For a detailed discussion of these discourses, see Orcutt-Gachiri (2009).
. Gĩkũyũ [ɣɛkɔyɔ] is sometimes called Kikuyu [kɪkuyu]. The terms are used interchangeably
in this paper.
How can a language with 7 million speakers be endangered? 
language use in the family setting. Years ago, even in urban settings, children of my
husband’s generation (he was born in the 1960s) were raised speaking mother tongue,
Kiswahili, and, if the parents were educated, English. The children also spoke Sheng (a
hybrid language used by young people that has Kiswahili as the matrix).9 The language
of socialization was mother tongue in both rural and urban areas to such an extent
that I do not know any Kenyans over 40 who do not speak at least one mother tongue.
Today, in the urban areas in particular but also to some degree in rural areas, in
those same households, the parents and grandparents speak mother tongue to each
other but then speak exclusively in Kiswahili to the grandchildren. This happens even
in families where the grandparents made a big effort to speak mother tongue to their
children, and it happens even when the grandchildren are frequent visitors to the
rural areas.
When I was in a rural area, I heard the children speaking in Kiswahili. I w
ondered
how much Kiswahili these children in the rural area knew because they were quite
young and the area was quite remote. I asked one of the parents of an urban, visiting
child, “What languages does your child use to play with the children here?” The parent
thought for a moment and said, “Kiswahili.” I asked whether the child ever used
Gĩkũyũ, and the parent said, “I hadn’t thought about it, but no, just Kiswahili.” Rather
than the urban child accommodating the rural children’s use of mother tongue, the
rural children accommodated by speaking Kiswahili. This fact had gone unnoticed by
the adults because it was taken for granted as the natural thing, which is usually the
case with language ideologies.10
The language ideology that children are supposed to speak Kiswahili, not mother
tongue, is making inroads into rural areas because there is so much back and forth
between urban and rural. Some rural parents are beginning to wonder why their
children need to know mother tongue, since it is not serving any purpose in areas with
jobs (their cousins in town do not speak Gĩkũyũ, for example), and because it does
not help their children in school, where English is the language of instruction. When
I asked Mrs. J, a parent at Utheri, why so many parents are not using mother tongue
with their children, she replied as follows.
. For more on Sheng, see, among others, Abdulaziz and Osinde (1997), Bosire (2008),
Githinji (2006), King’ei (2001), Mazrui (1995), and Samper (2002).
. This is in contrast to situations in which, such as in much of North America, parents who
had been punished for speaking an indigenous language (such as in boarding schools) made a
conscious decision to not speak an indigenous language to children because they did not want
their children to also be punished. There are punishments in Kenyan schools for speaking
mother tongue, and I will discuss these later on in the chapter, but I do not believe the punish-
ments are a major force in the language shift in the Kenyan situation, although surely they do
not help to bolster the position of indigenous languages.
 Heidi Orcutt-Gachiri
Mrs. J: Because even the little ones, we we want them to know Kiswahili first.
If mother tongue ceases to be the language of socialization in both urban and rural areas
and it has no other role in Kenyan society because language ideologies of nationalism,
education, and development, and tribalism have chased it from the Kenyan language
scene, this will create huge problems for the future of mother tongue.
Here, I will evaluate the status of Gĩkũyũ, and I will also explore ways in which
the language ideologies rather than the languages could shift. I begin with a h istorical
sketch and then move into a more detailed discussion of Gĩkũyũ, explaining the factors
contributing to language endangerment and illustrating the significant language
ideologies in play.
2. Gĩkũyũ in colonial and post-Independence language policies
The colonial administration, which established itself in 1904 in Kenya, left all education
for Africans in the hands of the missionaries, and when school expansion did not occur
rapidly enough, the Gĩkũyũ formed their own school committees and built and staffed
schools (Afigbo 1985: 491; Furley & Watson 1978: 72, 160). Mission education was
conducted in mother tongue; only European and Asian children, who were a minority
in Kenya, were allowed to learn in English. The Gĩkũyũ wanted the same educational
privileges European and Asian children had, including being allowed to attend second-
ary school, being allowed to take the secondary school exams, and being allowed to
attend university (Furley & Watson 1978: 178; Mafela & Mgadla 2000: 4). Slowly, they
and other fairly Westernized groups, like the Luo, fought for and obtained these rights
more than a decade before Independence (Furley & Watson 1978: 253). Being denied
the right to education in the language the European and Asian children were learning,
which also was the language of university, did not sit well with the Gĩkũyũ. As soon
as the Gĩkũyũ were able, they introduced education in English in their independent
schools. Access to English was very important to being able to compete on the same
level as Europeans and Asians. The inequality in educational access and opportunities
between Europeans, Asians, and Sub-Saharan Africans was symbolized by access to
English, and this is something that must be kept in mind when undertaking language
planning in Kenya. Policies encouraging use of mother tongue must not be seen to hark
back to colonial days, when use of mother tongue in e ducation was a symbol of unequal
access to education and employment possibilities.
At the time of this study, the educational system in Kenya consisted of an 8-4-4
system, with 8 years of primary school (Standards 1 through 8), 4 years of secondary
school (Forms 1 through 4), and 4 years of university. Education is exclusionary in
Kenya, and advancement to the next phase hinges in large part on two national exit
How can a language with 7 million speakers be endangered? 
exams, one at the end of primary school, and one at the end of secondary school.
The Kenya Certificate of Primary Education (KCPE) determines whether students have
a chance of advancing to secondary school, and the Kenya Certificate of S econdary
Education (KCSE) does the same for tertiary education. There are not enough places
for students at any level, and even those students who pass the national exam have no
guarantee of advancing. For example, in January 2004, 317,796 students who passed
the KCPE were not able to find places in a secondary school (Daily Nation 2004). Their
education ended abruptly at Standard 8.
Current language policies have Gĩkũyũ being used in Standards 1 through 3 in
linguistically homogeneous areas of Gĩkũyũland.11 Gĩkũyũ is not used after Standard 3,
and it is not examined on the KCPE or KCSE. English is the language of instruction
in Kenya from Standard 3 on, with the exception of one subject, Kiswahili. In urban
areas, English and Kiswahili are the languages of instruction from Standard 1 on. Both
the KCPE and the KCSE are in English, and all students are examined as if they were
native speakers of English (yet only the wealthy exclusively use English as the language
of socialization). There is a subsection of the exam in which students are examined
on Kiswahili as an L2, and that is the only part of the exam in which students are
allowed to use Kiswahili. In Kenya today, because of the exam-oriented system, com-
petence in English is the priority. Because indigenous languages are not examined
on the KCPE or KCSE, they are not a serious subject of study in the way that English
and Kiswahili are. The modicum of institutional support in Standards 1 through 3 for
mother tongues has no real weight because parents, teachers, and students care only
about what is on the two exams.
During the colonial and post-Independence periods, there were many steps that
could have been taken to protect and promote mother tongues but were not (Sow &
Abdulaziz 1993: 522). During colonialism, European languages were the vehicle
for acculturation. Indigenous languages were not promoted in institutions or in
government. Although indigenous languages were used in mission schools, Kenyans
saw the absence of English in their schools as a distinct disadvantage and advocated
for instruction in English so they could obtain employment after graduation (Sow &
Abdulaziz 1993: 527–528).
During Independence, a second opportunity was missed for mother tongues
(Sow & Abdulaziz 1993: 528–530). Independence for African countries could have
been a time for assessing and redesigning the colonial language, educational, and
cultural policies. But rather than transforming colonial policies, the new governments,
including Kenya’s, adopted them wholesale (Sow & Abdulaziz 1993: 530).
. In linguistically homogeneous areas where another mother tongue predominates, that
mother tongue is used in education through Standard 3.
 Heidi Orcutt-Gachiri
Additionally, in the post-Independence era, language ideologies promoted the

idea that mother tongues divide rather than unite nations (Sow & Abdulaziz 1993: 530).
The language ideologies of mother tongue languages as divisive and the people who
speak them as tribalist are extremely harmful for these languages and their speakers.
Since Independence, the status of mother tongues has not improved. The
emphasis is on English and Kiswahili instead, even though mother tongues could be an
important force in socioeconomic development with local solutions (Sow & A bdulaziz
1993: 550–551). Because of language ideologies of mother tongue as divisive, language
shift is happening quite quickly in urban contexts because of the multiethnic setting.
Kiswahili is being adopted in place of mother tongue, in contrast to earlier urban
generations, who spoke mother tongue and Kiswahili.
Rural areas tend to be seen in too homogeneous a light, as safeguarding the language,
when in fact these areas are undergoing transformations similar to what is occurring
in the urban area, since these language ideologies and discourses operate nationally.
Rural and urban areas are not separate islands in Kenya – there is an intense amount of
back and forth travel and habitation in Kenya, as elsewhere in sub-Saharan Africa. For
example, Mr. X, a teacher at Utheri, lives in Kirinyaga, but his son lives in an urban area.
He is not using Gĩkũyũ as a language of socialization with his son even though Mr. X is in
a Gĩkũyũ-speaking area. Rather, he is accommodating his son’s use of Kiswahili because
his son is in an urban area. Mr. X would like for his son to learn Gĩkũyũ but is planning
to wait until the child is fluent in Kiswahili before introducing another language. He then
plans to move him to school in the rural area so he can learn mother tongue.
Mr. X: So what I am planning to do because I would hate for him not actually to talk in
the vernacular, mmm, after he has mastered one language, I’m waiting for him to
master one language, after he’s mastered one language, then I’ll bring him here to
school in the the lower forms I mean lower standards so that he pick the vernacu-
lar now from there
This practice of bringing the child to the rural area later in life risks having the child
miss the critical period for language acquisition (Newport 2006), so that the child
would never acquire native fluency in Gĩkũyũ. In addition, the rural children may
well accommodate the urban child’s use of Kiswahili. In that case, the child would not
learn mother tongue. Looking at residential and travel patterns is an important part of
understanding the future of mother tongue in Kenya and the language i deologies and
discourses at work in rural and urban areas.
3. Gĩkũyũ
When I tell researchers that I work on Gĩkũyũ as an endangered language, I am

generally painted as delusional, as exemplified by the following comment: “Gikuyu is
How can a language with 7 million speakers be endangered? 
a strong African language and will be among the last African languages to disappear.”
In fact, Gĩkũyũ is undergoing observable language shift and is endangered. If this is
happening to Gĩkũyũ, which 29 years ago had a rock-solid position in Kenya, this
suggests a problem for indigenous Kenyan languages as a whole. It is a source of hope,
however, that the shift is in the early stages and could be reversed if there is a will to
do so and if the factors contributing to endangerment are addressed, beginning with
language ideologies.
How can Gĩkũyũ be following the path of the many African languages that are
endangered? Does prestige factor into this shift? Batibo (1992: 86) showed that in
Tanzania, the triglossic structure has a hierarchy of prestige of the languages such
that in some cases, the national language overwhelmed local languages and they
died out. I argue that prestige does not explain the language shift away from Gĩkũyũ
because Gĩkũyũ has a great deal of prestige, for the following reasons: (1) one of, if
not the, most widely read Kenyan scholars, Ngũgĩ wa Thiong’o, writes novels exclu-
sively in Gĩkũyũ and advocates passionately for the use of indigenous languages in
Kenya and elsewhere;12 (2) two of Kenya’s three presidents have been Gĩkũyũ speakers;
(3) Gĩkũyũ speakers have been very well placed politically and economically, which
has led members of other ethnic groups to learn Gĩkũyũ for business or employment
purposes (Abdulaziz 1982: 115); (4) Gĩkũyũ is a language in which literacy has been
achieved, and the Bible is widely read in Gĩkũyũ, along with novels and print media;
and (5) there are Gĩkũyũ radio stations and newspapers. I do not believe prestige lies
at the heart of the general shift away from indigenous languages in Kenya, nor does it
explain the shift from Gĩkũyũ specifically.13
Jane Hill (personal communication, 2009) has noted the case of Gĩkũyũ shows the
prescience of Krauss’s (1992: 7) point that the only safe languages are those that have
institutional support. I note above that Gĩkũyũ’s institutional support is too short lived
to benefit it because it ends at Standard 3 and is seen as a hindrance in s econdary school
(students are not allowed to speak mother tongue in secondary school). So, I argue that
Gĩkũyũ does not in fact have sustained institutional support of the kind Krauss states
makes languages safe. What has happened in just 29 years in Kenya p rovides evidence
that languages are extremely sensitive to political economic contexts when they do not
. wa Thiong’o is regularly referred to in Kenyan newspapers, and this lends prestige to
Gĩkũyũ even if not all are reading his work in Gĩkũyũ. A Standard (2007) article titled “Why
mother tongues are dying” mentions wa Thiong’o’s work in Gĩkũyũ and states, “Ngugi had
demonstrated by word and deed that he was willing to go to great lengths to keep African
languages alive. For him, using foreign languages in literature was a mark of neo-colonisation.”
Nettle and Romaine (2000: 6) write: “Ngũgĩ Wa Thiong’o’s decision to write in his native
language, Gikuyu, resulted in his imprisonment and eventual exile.”
. Dr. Salikoko Mufwene first suggested to me that we need to find a new framework other
than prestige for analyzing this shift. See also Mufwene (2003).
 Heidi Orcutt-Gachiri
have the support Krauss describes. The lack of official, and specifically institutional,
support for mother tongue, along with the language ideologies that discourage use
of mother tongue with children, is causing the shift in Kenya from trilingualism (or
multilingualism) to bilingualism.
The study of language shift for large Kenyan languages spans several decades.
Lieberson and McCabe (1982) studied mother tongue language shift in Nairobi
29 years ago, when the numbers were quite reassuring but when they felt there was
still cause for concern because of the nature of the shift. Lieberson and McCabe (1982)
look at language use in terms of the spaces in which languages are used, the ‘domains
of language use.’ Of relevance to this argument is that Lieberson and McCabe (1982: 83,
93) note that language shift in certain domains, such as the domestic space, c ontributes
more toward overall language shift than does language shift in other domains. The
language ideologies of English, Kiswahili, and mother tongue in discourses of nation-
alism, education, and development have so negatively affected Gĩkũyũ, it has caused
language shift in the most intimate and vital domain for a language, the domestic
space, which is a red flag for Gĩkũyũ.
The composition of Kenyan households varies, depending partly on socio-
economic status. It can include several generations within a household or just a parent-
child grouping. Kenyans who are able to hire a domestic worker (called a house help)
to care for children and cook and clean do so. The house help may come daily for
families who cannot afford or house live-in staff or may live in for those who have
the means and space. Abdulaziz (1982: 112) studied wealthy residents of Nairobi and
their domestic workers. He noted that house helps often come from the parents’ home
areas, which are often rural and poor, are known to the parents (possibly relatives), and
speak mother tongue as a first language, and that the parents hope the children will
learn the language from the house help. This was not something I observed, although
it is not unusual to hire a person from a rural area one knows well, which would lead
to hiring workers who know the same mother tongue. Abdulaziz (1982) says that the
effort to try to have house helps teach the children mother tongue is backfiring for
language ideological reasons. Abdulaziz asks what the effects would be for children to
see their wealthy parents using only English and the house helps using only mother
tongue. He suggests that children would link mother tongue with lower class. This is
in line with the language ideologies of mother tongue and English in the discourse of
development.
Obondo (1996) explores language shift away from Luo. I did not know of O bondo’s
research when I formulated my research question – Why are trilingual parents raising
bilingual children? – but she and I were of one mind, as her dissertation is titled From
Trilinguals to Bilinguals? A Study of the Social and Linguistic Consequences of Language
Shift on a Group of Urban Luo Children in Kenya. The shift is so dramatic that two
How can a language with 7 million speakers be endangered? 
of the largest and most politically and economically connected languages in Kenya,
Gĩkũyũ and Luo, have become the subjects of parallel studies about 10 years apart.
Obondo (1996: 20, 86), using narratives from a group of urban Luo and a group of
rural Luo children, studies language shift in the urban group. She argues that language
use in the domestic sphere is contributing to language shift away from Luo (Dholuo).
This is in line with Lieberson and McCabe’s (1982) point that shift in this space has
the most severe consequences for mother tongue. Obondo (1996: 86) writes that while
adults use English, Kiswahili, and Dholuo among themselves in her data, “whenever
a toddler was a participant in a conversation, there was often a tendency to switch to
Kiswahili or a repetition in Kiswahili if the utterance was first spoken in Dholuo.”14
Obondo (1996: 91, 206) believes that the shift to Kiswahili as the language of socializa-
tion is contributing to language shift, and she warns that although the numbers are
still very good for Dholuo, there is need for caution because the urban areas may shift
completely away from Dholuo. Given the dynamic relationship between urban and
rural areas, it is not clear that rural areas would remain a stronghold. Dorian (1989: 9)
illustrates the phenomenon of “tip,” “abrupt transmission failure,” when language shift
happens almost overnight. Obondo’s study of Luo and my study of Gĩkũyũ show that
tip away from indigenous languages may be imminent in Kenya. It is significant that
scholars are exploring the very real possibility that some of sub-Saharan Africa’s largest
languages could quite easily and rapidly disappear.15 The language ideologies held by
the teachers, parents, and students I worked with echo disturbing trends across Africa.
I heard three arguments from Kenyans for why language shift is occurring,
none of which I feel gets to the heart of the issue. The first is that urbanization is
causing language shift. However, the children who grew up in Nairobi between the
1950s and the 1980s largely speak mother tongue. The shift is a lot more recent than
urbanization. Second, I was told it is because people hire domestic workers who do
not speak the same mother tongue as the parents, so the children do not learn it. A
mother of two, Nancy Mackenzie, is quoted in the Standard (2007) article: “‘Most of
a child’s life is spent with a nanny who does not come from the same background as
the child’s parents,’ she says. ‘The nanny will speak to the child in either Kiswahili or
English and in the end, the child ends up speaking the same language as the nanny.’”
Mackenzie believes this gives rise to language shift. The article continues, “According
to her, the buck stops with the parents. ‘If they hire a nanny who is not from their
background, they will have to accept the fact that their child will grow up not knowing
. Kulick (1992) found a similar pattern for Taiap.

. For example, Frederikse (1992: 70) looked at Shona in educational contexts in Zimbabwe
and concluded, “Shona might end up a dead language.”
 Heidi Orcutt-Gachiri
his mother tongue especially if they spend little time with their child,’ says Mackenzie”
(Standard 2007). Given Abdulaziz’s (1982) argument that children would link mother
tongue with lower class if they only hear it spoken by house help, this is not a con-
vincing reason for why language shift is occurring. Third, I was told it is because of
marriages between people of different ethnic groups because their children do not
learn either of the parents’ mother tongues. An article titled “Mixed marriages that
speak in tongues” articulates this view: “Mixed marriages, where both parents do not
speak the same mother tongue, is among the greatest causes of this language erosion”
(Daily Nation 2008). In contrast, Zubedi, a student at Elewa, told me that she expected
she might marry a man from a different ethnic group and that her future children
would know both parents’ mother tongues. Some of the students whose parents spoke
different mother tongues did know both languages, although they might be more com-
fortable speaking one or the other. I believe that because Kiswahili is the language of
socialization in the urban area and even in some of the rural areas, it does not mat-
ter whether the nanny speaks mother tongue or whether the parents speak different
mother tongues. No one is going to use mother tongue to the children no matter the
circumstances. It is an ideologically driven shift.16
Lieberson and McCabe (1982: 87) provide data with which to reflect on
changes in language transmission patterns. In 1982, 88.4 percent of Gĩkũyũ parents
in Nairobi spoke to their children exclusively in Gĩkũyũ, and another 2.1 percent
spoke Gĩkũyũ and another language to their children. This finding fits perfectly
with the pattern I found in which the oldest adult children are fluent Gĩkũyũ
speakers. Language shift in Gĩkũyũ is so recent, members of the same household
have stratifiable levels of c ompetence in Gĩkũyũ, from fluent (grandparents and
older adult children) to semi-speaker (younger adult children) to non-speaker
(grandchildren under 25). Within a household, at the same moment in time, there
is tremendous difference in use of mother tongue. There is an unspoken agreement
not to use Gĩkũyũ with this youngest generation, and it is a pattern that is repeat-
ing itself over and over in urban areas and some rural areas. These are families who
brought up their children to speak Gĩkũyũ (and still speak to their children in it
almost exclusively) but not the grandchildren. Where before it was not correct to
speak Kiswahili to elders, it is now assumed in many cases that this is what the
grandchild will use to address the elder.
. There is a pragmatic component as well, as a reviewer pointed out, because there is no
point in learning a language that no one else speaks. However, I would argue that at heart,
any and all “pragmatic” reasons for this shift away from indigenous languages are ultimately
ideological.
How can a language with 7 million speakers be endangered? 
4. The need for language planning
I do not pretend to have solutions to the complex issues raised here. Instead, I hope
to aid language planners by showing the form and effect of the language ideologies
and discourses that are significant at the national level and provide analytical tools for
them and for Kenyans who care deeply about the future of these languages to use in
developing policies for Kenya’s government and educational institutions. Linguistic
anthropology offers a unique set of tools with which to understand in great detail
and at multiple levels the meaning of language use – the language ideologies and the
discourses with which they intersect – in both everyday and institutional life. I believe
that language planning needs these tools to succeed because people will only partici-
pate in policies that take into account the specific impact of languages on their lives.
Without the participation of Kenyans, the future of mother tongues cannot be altered.
With their participation, it is more likely that language policies can be created that
address the beliefs, wishes, and desires of Kenyans and work with them.17
One of the big challenges that language planners will need to overcome is the
ideology that emerged through my interviews with students that mother tongues do
not need to be taught – they will simply be picked up, it is thought. This is in line with
indigenous education more generally, which often involved learning through partici-
pation (Otiende 1990: 145). There were of course also formal contexts, but thinking
of indigenous education as something that will simply be picked up because of the
opportunities to learn may have bled into current thinking about how children learn
indigenous languages. However, since there is language shift in the domestic sphere,
mother tongue is no longer being learned in many households. Also, as Abdulaziz
(1982) notes, it matters who is doing the teaching of mother tongues and where it is
occurring. Is it in a context that reinforces existing language ideologies that disad-
vantage mother tongues or in transformed contexts with a sea change in language
ideologies? In addition, given the language ideologies of English, Kiswahili, and mother
tongues in discourses of nationalism, development, education, and tribalism, parents,
teachers, and children de-emphasize mother tongues and focus on the languages that
are thought to ensure educational and socioeconomic success for children – English
and Kiswahili. Those languages are thought to be languages that need to be formally
taught, and this again puts indigenous languages at a disadvantage. If Kenya’s 53 indig-
enous languages do not become the explicit focus of language planning at the national,
regional, and family level, they will be at more risk than ever.18
. See Hinton and Hale (2001) for foundational work on language planning.
. Admittedly, at the institutional level, this would be difficult in an already stretched
economy because a great deal of resources would be needed to do this for all indigenous
languages.
 Heidi Orcutt-Gachiri
Another problem for mother tongues that language planning needs to address
is the fact that students are often punished, sometimes physically, for using mother
tongue in schools. Mutahi (1996: 6) writes that his teacher made him and “those who
found intimacy with that language [English] carry a block of wood inscribed, ‘I am a
fool’ round the neck for speaking one of ‘those native languages.’” While this incident
happened many years ago, I did see students being punished at Utheri for speaking
mother tongue in 2004. I saw students digging trenches – quite hard labor – and was
told they were being punished for speaking mother tongue.19
Scholars of language planning in sub-Saharan Africa protest the major roles of
ex-colonial languages in multiple domains of use. Not using indigenous languages in
educational contexts hinders their chances for being passed on to the next generation
and contributes to social reproduction of socioeconomic differences based on access
to and command of the ex-colonial language (Prah 2005: 33). Westaway (1995: 6)
writes, “I would like to question the degree to which Kenyan citizens are content to
accept a state of affairs in which English holds such a strong position and in which the
demise of the vernaculars, if not Kiswahili, is inevitable.” Concern about the future
of mother tongue languages is expressed fairly regularly in the Kenyan media. Also,
my consultants of all ages were happy to hear that I was working on this issue, as they
agreed that something is happening with mother tongues and that they are not being
learned as regularly by young people.
Whatever position Kenyan language planners take, it is important to plan with an
understanding of the language ideologies and discourses of nationalism, development,
education, and tribalism (Orcutt-Gachiri 2009). The policies must be practical and
consistent with the current language ideologies and discourses in Kenyans’ lives. The
policies can actively try to shift these ideologies, but they ignore them at their peril.
A position that advocated for total rejection of English would not find traction in the
current language ideological climate in Kenya.20
It is also possible to incorporate mother tongue use in the educational sys-
tem beyond Standard 3 and to give these languages, at least the larger ones, more
weight by making them examinable. The same way students in some high schools
in Kenya now learn French, they could learn an indigenous language, one that need
not be of their ethnic group(s). Or, as Bunyi (1996) argues, mother tongues could
be used as media of instruction to complement English and Kiswahili throughout
. Although generations of Kenyans have been punished in some form for speaking mother
tongue, I do not believe this has been a major contributor to the language shift. Rather, it
reinforces prevailing language ideologies of mother tongues as without utility and as not
belonging in the educational sphere.
. See Senanu and Williams (1995: vi).
How can a language with 7 million speakers be endangered? 
grade school and high school in linguistically homogeneous areas. It is also possible
that K iswahili could be used as the language of instruction and English taught as a
foreign language. Bunyi (1996) argues that the quality of the English learned and
the level of comprehension would be higher. It is possible, as Fuller (1991) advo-
cates, to make education a more local phenomenon that takes local experiences and
knowledge into account.
Another way to harness the energy of young people and get them and their elders
speaking indigenous languages is to creatively use the media. Jaffe (1999: 30) views
creative media practice as the most successful of all the activities intended to promote
the Corsican language because it does not focus on the language itself but makes it
part of the activity. Jaffe (1999: 30) feels that any explicit attempt to discuss language, or
any contexts that have language as their focus, will be shunned by Corsicans because
of the ideological and political loading of any act of language. The only contexts in
which Corsican will be regularly used without self-consciousness and without political
motivation are contexts in which it is a medium of communication but not a focus of
ideological attention. Because of the politicization of mother tongue in Kenya, Jaffe’s
research may prove very useful for Kenyan language planners.
In order to reverse the trend of eroding mother tongues, sea changes21 in language
ideologies are necessary in the same way that ideologies about local cultural practices
have undergone a sea change (see discussion below and also Orcutt-Gachiri 2009 and
Samper 2002). Local practices have become embraced by young people, where before
they were shunned (Mutahi 1996).
5. Young people
Since the 1990s, young people have developed an increasingly local take on global
issues. Although young people respect and celebrate local perspectives in music,
mother tongues are not the medium for this because of the pull of nationalism, which
demands Kiswahili or Sheng as the form of expression for these Kenyan identities in
music. English is used much more rarely in music by Kenyan artists who cater to young
people. This trend is positive in that Kenyan young people are proud of their hybrid
cultures (Samper 2002), which reinterpret global elements with local p erspectives,
especially their music, clothing, and dance styles. However, this does not contribute
positively toward mother tongue.
While the hybrid identities (Samper 2002) being forged by Kenyan young people
may be highly beneficial in lessening the rancor that exists between some members
. Sea changes are broad, sweeping changes that occur rapidly and dramatically.
 Heidi Orcutt-Gachiri
of different ethnic groups because of past events in Kenya, unfortunately, this desire
for pan-Kenyanism has a high cost in terms of the 53 Kenyan indigenous languages
because it intersects with language ideologies about speaking mother tongue in ways
that are quite problematic for the continued survival of these languages if the trends
continue.
Many young people I spoke with have very positive attitudes toward mother
tongue. For example, Josiah, a student at Elewa, enjoys the diversity of languages in
Kenya.
Josiah: yeah [Heidi: okay] I enjoy the diversity of language [Heidi: okay] because if I hear
a Kikuyu talking like you know [Heidi: yeah] it sounds good yeah [Heidi: okay] so
I think the diversity [Heidi: yeah] is good
Albert also expresses positive feelings toward mother tongues.
Albert: if only our traditional languages were worldwide
The problem, then, is not language attitudes, it is language ideologies. The positive
feelings these young people articulated about mother tongues does not translate into
language use. In contrast, language ideologies of young people present a fairly unified
picture that gets at the heart of some of the issues facing Kenyan young people today;
poverty, joblessness despite advanced academic degrees, effects of globalization, and
socialization with their peers.
One problem is that in the local interpretations of global contexts discussed above,
the focus is on urban interpretations, rather than rural. Because of language ideolo-
gies that mother tongues and their speakers are unsophisticated and backward, these
languages are seen as “left behind” in the move to “development.” A second problem is
that young people often see speaking mother tongue in a situation of ethnic heteroge-
neity as an expression of ethnic identity, even akin to an expression of tribalism.
Jackson: yeah [Heidi: ok] so the fact that we we speak in in in Kikuyu I have brought up
@ it’s it’s the fact that we speak in speak in Kikuyu in in uh both of us like uh so
we speak with mostly Kikuyu [Heidi: yeah] it’s it’s what it it’s it is so illustrative of
[Heidi: yeah] of of tribalism [Heidi: yeah]
This contrasts with the hybrid identity many urban young people strive for, a form of
hyper-anti-tribalism.
Jackson: yeah the the unison between [Heidi: yeah] and but all l– l– like it like in a in
a in a school we have we’re from tr– tr– different tribes we speak we speak one
main language [Heidi: yeah] eh, maybe Sheng or Kiswahili [yeah] so it has
removed that tribalism [Heidi: ok] that unison and and turned it to more eh
better unisons [Heidi: ok] than that tribe [Heidi: ok] and that unison also yeah
〈clicks tongue〉
How can a language with 7 million speakers be endangered? 
Ralph thought it is going a bit far to say it is tribalist, but he did agree that it is bad to
speak mother tongue in a heterogeneous group.
Ralph: That one is bad but I think it’s {xx} cause when you talk like him when we’re
talking in Kikuyu yeah it’ll be like {xxx} yeah [Heidi: mm] so, it’s kind of—ok
but not really tribalism
Because of the language ideology that speaking mother tongue equals being tribalist,
there is a tendency for these young people to avoid speaking mother tongue, if they
indeed know it.
Students were not always clear about whether they wanted their children to learn
mother tongue, so I prompted when necessary. Speaking with Angela, Paula, and
David, I asked what languages they wanted their children to know.
Angela: Kiswahil-
Angela: [-i]
Paula: [E-]
Paula: -nglish
David: know English and Kiswahili
Heidi: what about Kikuyu?
David and Paula explain that Gĩkũyũ will be the first language the children learn, but
Angela says no to its being important.
David: Kikuyu they
Angela: No
David: they it’ll be the first
Paula: @@
David: language of them
Heidi: ok so that
Heidi: [one they’ll]
Angela: [I]
Heidi: know
However, as the students continued speaking, it did not sound as if Gĩkũyũ were
definitely going to be the first language of the children. Angela and Paula explain that
they do not intend to put their time toward teaching mother tongue. David, who plans
to stay in a Gĩkũyũ area, might make more of an effort for them to learn, although he
does not commit to putting energy toward it. Angela continues from her “I” in the
excerpt above, once I gave her the floor.
Angela: when they are b– when they start talking [Heidi: mm] I’ll tell I’ll teach them
Kiswahili and English 〈claps hands together〉 [Heidi: ok] so that when they grow
up they’ll be used to Kiswahili and English
 Heidi Orcutt-Gachiri
Heidi: ok would you also teach them Kikuyu as they grow?

Angela: Kikuyu? No. I’d like to—I would teach them, but it is not all that important but
I can teach them
Heidi: You will just prefer that they know English and Kiswahili
Angela: yes
Heidi: and Paula?
Paula: I prefer they to know English because when they’ll be at school they’ll be be
communicating mostly in English [Heidi: ok, and what about Kikuyu?] it is not
that much but I’ll be niní22 telling them the words [uhmhm] to how to {xxx}
Heidi: ok, but Angela, if your kids end up speaking Kikuyu, that’s ok with you?
Angela: y[es]
Paula: [yes]
Heidi: you don’t not want them to know but it’s just you won’t put your time and energy
into it
Paula: yeah
Angela: I can like them to know [Heidi: uhmhm] but it is not that important
Heidi: you won’t focus on it
Paula: yeah
Heidi: and David
Angela: yeah it’s just for enjoyment
Heidi: I’m sorry, Gichuhi23
David: I can like them to know it
Heidi: I’m sorry?
David: I can like them to learn
When I probed a bit more, David responded as follows.
David: yeah they they will know it because some of the—where they can stay they—it is
Kikuyus communities
Heidi: ok so you’re planning on living in a Kikuyu area and you want them to know it
David: yes
Whereas this group of young people had told me that English and Kiswahili needed to
be taught and practiced, the most commitment Gĩkũyũ received was that the children
. Paula uses a common verbal tic of young people, ‘nini.’ In this sentence, ‘nini’ has stress
on the second syllable, which makes it the English equivalent of ‘um.’ If it had normal stress
(on the first syllable), it would be Kiswahili for ‘what.’
. Gichuhi is David’s middle name, and he had just finished telling me he prefers it to
David, so I corrected myself.
How can a language with 7 million speakers be endangered? 
can know it. These are quite different ways of thinking about language socialization,
and given the trend toward use of Kiswahili as the language of socialization in urban
but also in rural areas, this would make it even more difficult for children to learn and
use Gĩkũyũ.
Brenda, a student at Elewa, explains that she feels children are being discouraged
from speaking mother tongue by their friends.
Brenda: but you see, at the moment, guys have been discouraged mother tongue [Heidi:
hm], I don’t know why [Heidi: ok] ‘cause fine you know my grandmother can’t
understand English Swahili so I’ll have to speak to her [Heidi: yeah] in this
language [Heidi: mm] but as time goes by you know your friends keep telling
you, ‘Why are you talking that language,’ you know, ‘it’s not worth it,’ so you tend
to like forget it [Heidi: ok] and
Heidi: do you think there’s peer pressure not to speak it? I mean
Brenda: yeah
Heidi: people get discouraged
Brenda: actually I think it’s peer pressure mostly yeah
Brenda, who does not speak mother tongue herself, explains that children have been
discouraged from speaking mother tongue because their friends see it as a useless
language and deride them from speaking it. She noted that the original motivation for
speaking it was to be able to communicate with elders in mother tongue, but that as
friends become more important in a young person’s life, their peer influence grows,
and peers do not understand the need for mother tongue. Brenda’s statement highlights
a couple of language ideologies of mother tongue that I mentioned earlier. The first is
that mother tongue is thought of as static – its use, in Brenda’s characterization of peer
groups, is with the elderly – it has no place among the young. Second, she blames peer
pressure for why children abandon these languages. This speaks to the language ideolo-
gies that motivate children to see the languages as useless more than it does to actual
peer influence, which I think is not the major factor in language shift in Kenya. Rather,
the children see the writing on the wall and understand that English and Kiswahili
are the languages with linguistic capital24 in Kenya, whereas mother tongue has none.
Parents and children together embrace Kiswahili as the language of socialization.
Annette and Celeste, at Utheri, gave me another piece of the language ideological
puzzle when I asked them what language or languages they wanted their children to
know.
. See Bourdieu (1977) for the concept of linguistic capital.

 Heidi Orcutt-Gachiri
Annette: Kiswahili
Heidi: Anything else?
Annette: No
Celeste: English and Kiswahili
These were very typical answers from students at both schools. I then asked whether
they wanted their children to know Gĩkũyũ.
Celeste: Yeah
Annette: No
Heidi: that’s not important to you
Celeste: @@
Annette: Yeah
Heidi: Do you want them not to know it or it’s ok if they know it? Or you prefer they
don’t even know it?
Annette: Yeah @
Heidi: Annette ok, and Celeste, how about you?
Celeste: they know it [Heidi: you want them to] but [Heidi: {x}] is not [Heidi: {x} got] it
Heidi: now can you guys tell me a bit about why
Annette: because if they speak in vernacular they’ll translate and talk bad English [Heidi:
ok] their English [Heidi: is not]
Heidi: ok, so it can affect their ability to speak the other languages
Annette’s was an unusually strong rejection of the language. Celeste gave a more typical
answer, that they will or can know it, and suggested that it was not important to her,
although the details of that are not clear. Annette said Gĩkũyũ hurts their English when
they translate from Gĩkũyũ, so she would rather the children grow up speaking English
straight off, rather than learning Gĩkũyũ and then English.
What is interesting is that there is an all-or-nothing feeling in these answers.
Trilingualism is not an ideal for these young people. They do not see Gĩkũyũ as adding
a dimension to their lives. Rather, it is subtractive, and this is a big change from their
parents’ generation. Even where Gĩkũyũ and other mother tongues are seen positively,
as preservers of culture or anti-globalization tools, they are relegated to the sidelines.
These young people do not see mother tongue as a rich, dynamic force in their lives
that makes a strong contribution to their quality of life and their identities.
6. Young people and the events of the 2007 presidential election
To briefly summarize the discourse of tribalism, when one ethnic group is seen as
favoring members of its own ethnic group through nepotism or by developing
areas of the country in which members of that group live, there is a public outcry of
How can a language with 7 million speakers be endangered? 
“Tribalism!” Tribalism is seen as bestowing privilege, wealth, or favoritism that could

put one group at an advantage politically and economically (Orcutt-Gachiri 2009). In
2004, Kibaki’s second year in office, accusations of tribalism were at a fever pitch over
filling government positions.25
The first election since Kibaki took office in 2003 took place on December 27,
2007. In the run-up to the election, National Public Radio reporter Gwen Thomp-
kins went to Kenya and began reporting on the pre-election political climate. She filed
three reports, and pointed out routinely that Kenyans appeared to care more about
the ethnicity of the candidate than any other factor. She spoke with Kenyans of differ-
ent ethnic backgrounds, but what they had in common was a strong ethnic alliance
to a particular ethnic group and not much positive to say about other ethnic groups.
Thompkins paints Kenyans as naïve and out of touch with issues, completely focused
on ethnicity (Thompkins 2007).
Thompkins: But there is little belief here that corruption or any other issue matters to
voters
Abongo: People are ethnically aligned
Thompkins: That’s pollster Alan Abongo. Word on the street is that ethnicity will decide
this vote
This is not typical of my conversations with Kenyans. The discourse of tribalism is part
of the ideological nexus in Kenya, but for many Kenyans, young and old, it is not part
of their beliefs.
Thompkins’ analysis was historically shallow and decontextualized, and I found
myself disappointed. While this may be particularly true of young people, there are also
older people who reject the discourse of tribalism. You Tube had some video messages
from a person planning to vote outside his ethnic group (Gĩkũyũ) and encouraging
others to do the same (Ngatia 2007). I began to note the difference between what the
people Thompkins was interviewing were saying about ethnicity and what the young
people I worked with in the two secondary schools had said. The young people I
worked with were often attempting to forge shared identities of Kenyan nationalism
that attempted to meld different ethnic experiences, creating hybrid identities (Samper
2002) rather than continuing to see different ethnic groups as diametrically opposed,
polarized entities.
The December 27, 2007, election has led to disastrous consequences for Kenya
because the election, which was between Kibaki, a Gĩkũyũ, and Odinga, a Luo whose
father was one of the main proponents of decentralized government back in the 1960s,
had results that were not up to international election standards. A further issue that
. See Rothchild (1973: 11) on “ethnic arithmetic.”

 Heidi Orcutt-Gachiri
created controversy and heated emotions was that Kibaki was only supposed to hold
office for 5 years and was then supposed to step aside and not run again (Encyclopedia
of World Biography 2005). When the time came, he ran for re-election.
Because this new layer of ethnic tension built quite clearly on historical tensions
that have simmered quite close to the surface for more than 40 years before coming
to the fore during the 1992 and 1997 elections, Kenya exploded into violence that was
understood both locally and internationally as organized along ethnic lines. Accord-
ing to some reports, the violence would have happened whatever the election results.
In 2008, under pressure from Kenyans and international leaders to stop the violence,
Kibaki and Odinga agreed on a power-sharing arrangement in which a Prime Min-
ister position would be created that Odinga would hold. This was in fact very similar
to the new constitution that had been waiting to be voted on since 2004. The violence
ceased in March 2008 and left more than 1,000 Kenyans dead and 500,000 internally
displaced (Human Rights Watch 2008: 2).
I believe that things are going to get worse for Gĩkũyũ and other mother tongues
because of the events of the 2007 presidential election in Kenya. I believe that this
ethnically targeted violence will have an effect on mother tongues because since young
people use Sheng as a way of showing solidarity between ethnic groups (anti-tribalist)
and avoid mother tongue to not appear divisive (tribalist), then the traumatic events
of the 2007 election, during which in some areas neighbors were pitted against neigh-
bors, will only make it even more likely that young people will avoid speaking mother
tongue.
The following illustration of the politicization of mother tongue in Kenya helps
to explain what young Kenyans are reacting against by using Sheng and Kiswahili
to show strategies of neutrality (Orcutt-Gachiri 2009).26 A news article titled
“Moi seeks to restrict language broadcasts” discusses President Moi’s banning of
indigenous language radio broadcasts. The article states, “President Moi said private
radio stations must use English or Swahili to promote national unity – but the
proposed new law has been interpreted as a means of containing criticism of the
government” (BBC News 2000). Mother tongues are seen as secretive and as under-
mining national unity in the language ideologies of these indigenous languages.27
The article continues,
. See Myers-Scotton (1976) for the concept of strategies of neutrality.

. A reviewer asked, “Could it be that tribalism is a threat to the government?” I argue that
tribalism is portrayed in the discourse of tribalism as a threat to national unity but that it is
in fact a political tool to keep groups from forming strong alliances and gaining a sustained
majority in the government.
How can a language with 7 million speakers be endangered? 
Speaking in Mombasa, the president charged that vernacular radio promoted

tribal chauvinism and undermined national cohesion. He said that speakers of
the main vernacular languages were adequately served by the Kenya Broadcasting
Corporation, ‘which ensures that national unity is not undermined.’ Tribalism, he
said, was the root cause of instability in many African countries and should be
fought at all costs. (BBC News 2000)
I could not ask for a clearer illustration of the language ideology in which the mere
act of speaking mother tongue is seen as a declaration of tribalism. As long as this
language ideology is hegemonic in Kenya, mother tongue will be at risk because of
the very personal, very difficult effects of political violence. Unless and until mother
tongue is depoliticized and seen just as an affirmation of identity and not also as a
declaration against other ethnic groups, mother tongue will be at risk. This would
be a tremendous loss. Hill (2001: 176) writes, “Small languages seem to provide for
their speakers deeply embodied and very local ways of being-in-the-world, highly
economical alignments of knowledge and rationality with emotional and aesthetic
life that are part of a sense of belonging to a place and a community and living in
these with skill and relative ease.”
7. Conclusion
The young people I worked with in my fieldwork are the first generation of Kenyan
young people who are not learning mother tongue, and my research (Orcutt-Gachiri
2009) examines the causes of this phenomenon through the historically contextualized
study of language ideologies in discourses of nationalism, education, development,
and tribalism. English is privileged in the discourses of education and development,
and Kiswahili in the discourses of nationalism and education. Mother tongues are left
on the margins in every case and are at serious risk. In addition, there is a language
ideology operating that children will simply pick up mother tongue, when in fact the
language shift that is taking place in the domestic sphere – language shift that is occur-
ring even in rural areas – ensures that they will not. Worse still, rather than being
perceived as neutral languages, mother tongues are seen as separatist and divisive in
multi-ethnic contexts, which are common across all areas of Kenya, rural and urban,
and are not embraced in the language ideologies of Kenyan young people. In response
to ethnic political and economic tensions and the dangerous conflicts in Kenya’s past,
young people have created a multilingual fusion of local and global that unites as it
excludes.
The political climate has led to hyper-correction against tribalism, expressed
through use of Sheng and avoidance of mother tongue, for many Kenyan young
people, who want above all to share a sense of nationalism and unity that contrasts
 Heidi Orcutt-Gachiri
with the 1992, 1997, and 2007–2008 ethnically targeted violence. The recent post-
election violence these young people have witnessed in which neighbors attacked
neighbors may give them even more motivation to shun what they view to be symbols
of tribalism, including speaking mother tongue. For younger Kenyans, use of mother
tongue indexes tribalism. The events of 2007–2008 have the potential of intensifying
the language ideologies sketched here and placing mother tongues at even greater risk
than they were before those events.
Mother tongues are thus thrice damned in language ideologies of English,
Kiswahili, and mother tongues in the discourses of nationalism, development,

education, and tribalism. This is contributing to the language shift underway in
Kenya, and, combined with the lack of substantive official or institutional support
for Gĩkũyũ, has led to its not being spoken by many young Kenyans under 25. These
Kenyans will not be able to pass the language on to their children. That is how a
language with 7 million speakers can be endangered.
However, while it is the goal of this work to lay out specifically what the c urrent
trend is in Kenya with respect to Gĩkũyũ, along with other mother tongues, it is
important to note that the story of language ideologies, discourses, and institutional
and official support is not the whole story of Kenyan mother tongues. Indigenous
languages have their champions in Kenya as elsewhere in the world, and it is my hope
that those voices will help shift language ideologies in favor of indigenous language
use. For example, Mr. Q, who as a teacher could choose to emphasize Kiswahili and
English with his children as other parents are doing, instead chooses to make sure his
children have a strong foundation in mother tongue and says they can learn English
and Kiswahili later on.
Mr. Q: In fact uh with my children, I speak in Kikuyu all the time
This is positive, and it shows that language ideologies alone do not tell the whole story.
News reports on the language shift occurring within Kenya appear every so often,
and these add voices of concern to the discussion about mother tongue. For example,
an article titled “Why mother tongues are dying” explores reasons for and feelings
about the shift (Standard 2007). A young woman of 18 was interviewed, saying she is
now learning Kikamba at age 18. The article interviewed a father who made sure his
children learned Kisii. The father says, “‘It is my duty to make sure that the children
stay in touch with their culture and mother tongue plays a big part in their upbringing’”
(Standard 2007). The article states, “Language experts are concerned that children are
not mastering their mother tongue as before” (Standard 2007). Another article, titled
“Mother tongue and cultural identity,” notes that “the fact that mother tongues are
never taught after primary three renders them vulnerable to extinction” (Kabaji 2007).
Kabaji (2007) continues, “The bottom line is that if the government does not intervene,
some Kenyan mother tongues are bound to die at great cost.” I believe this is true.
How can a language with 7 million speakers be endangered? 
In another article, titled “Mixed marriages that speak in tongues,” the head of a local
culture center is interviewed (Daily Nation 2008). The article states, “Multilingualism
opens doors to a deeper understanding of a people across the board, Ms Kimani says”
(Daily Nation 2008). Culture centers are being opened in urban areas to encourage
urban children to learn mother tongues in the center. This would provide parents with
more options for exposing their children to languages and supporting their a cquisition
if the domestic sphere is not a place in which mother tongue language learning is
happening. However, careful assessment and follow-up must be undertaken to evalu-
ate the utility of these programs.
Another hope comes from young people who do not accept the language
ideologies that put mother tongue at a disadvantage, even while they are not rejecting
the discourses. One of them, Rachel, a student at Elewa, said that all languages are
important in development. She continued her thought.
Rachel: You don’t need English to build a road
An article after the 2007–2008 violence countered claims that mother tongues were
used to inflame violence in radio and TV broadcasts; the article was titled “Use of
vernacular not to blame.” In contrast to a friend of his who “observed that ethnic chau-
vinism and vernacular language are the root causes of Africa’s problems,” the author
writes of those who hold that theory, “I agree with them that ethnic chauvinism is
a big problem, but a blanket condemnation of languages misses the point” (Ayuma
2008). He continues, “Language is a vehicle through which we convey our thoughts,
fears, aspirations, and even prejudices. If our thoughts are well-intended, so will be
the language we use, be it Dholuo, Giriama, Gikuyu, or Kiswahili” (Ayuma 2008). The
author concludes,
Language per se is not to blame. It is merely a resource which can be used or

abused to fit the whims of a user. The recent spate of hate-mail can attest to this.
Over 90 per cent of the hate-mail was in languages that we all understand –
English and Kiswahili. Does that make English and Kiswahili dangerous?
(Ayuma 2008)
People who champion these languages argue for policy changes and against the
language ideologies that put indigenous Kenyan languages at risk, and these are
positive steps. In addition, language ideologies change over time and can be trans-
formed in favor of mother tongues. It is possible to create a transformed view of
mother tongues that does not subscribe to the view that speaking these languages is
akin to an act of tribalism. For example, several students mentioned the language ide-
ology that mother tongues bolster local traditions and are a tool of resistance against
the encroachment of W esternization through globalization. Brenda spoke to this
notion of resistance.
 Heidi Orcutt-Gachiri
Brenda: ok, they’re fine, I really don’t mind them [Heidi: ok] a way of communicating,
keeping up your culture [Heidi: uhm-hm] so, I think of them as a way to get as
in a way to keep your traditions instead of eroding them [Heidi: ok] especially at
this era where uhm Western Western ways are eroding most of African t raditions
[Heidi: uhm-hm] so if they could actually still keep a hold on their mother
tongue then the traditions would most likely stay with them also
It could be that the language ideologies of mother tongue as preserving culture and
local identities could grow in importance in a sea change of priorities in Kenya. That
could most certainly happen, although to assume it will happen on its own without
campaigning for that sea change would be naïve, given the importance placed on
English and Kiswahili in Kenya today. In order for the language ideologies rather than
the languages to shift, the context and causes of the language shift must first be under-
stood at a deep level, in the tradition of Jane Hill’s ethnographic work on language
endangerment. In this chapter, I have attempted to lay out as clearly as possible the
social and historical context and causes for the language shift so that the efforts of
language planners and community activists to reverse the language shift can be as
effective as possible.
Appendix A: Transcription conventions
Truncated word –
Truncated sentence —
Speech overlap []
Word continuing into the overlap -
Laugh syllable @
Author comment 〈〉
Indecipherable syllable {x}
Pause ,
References
Abdulaziz, Mohamed H. 1982. Patterns of language acquisition and use in Kenya: Rural-urban
difference. International Journal of the Sociology of Language 34: 95–120.
Abdulaziz, Mohamed H. & Osinde, Ken. 1997. Sheng and Engsh: Development of mixed codes
among the urban youth in Kenya. International Journal of the Sociology of Language 125:
43–63.
Afigbo, Adiele E. 1985. The social repercussions of colonial rule: The new social structures.
In Africa under Colonial Domination 1880–1935 [UNESCO General History of Africa,
Vol. 7], A. Adu Boahen (ed.), 487–507. Berkeley CA: University of California Press.
How can a language with 7 million speakers be endangered? 
Ayuma, Isaac. 2008. Use of vernacular not to blame. Daily Nation. April 23, 2008.
Batibo, Herman. 1992. The fate of ethnic languages in Tanzania. In Language Death: Factual and
Theoretical Explanations with Special Reference to East Africa, Matthias Brenzinger (ed.),
85–98. Berlin: Mouton de Gruyter.
Batibo, Herman. 2005. Language Decline and Death in Africa: Causes, Consequences and
Challenges.Clevedon: Multilingual Matters.
BBC News. 2000. Moi seeks to restrict language broadcasts. 〈http://news.bbc.co.uk/2/hi/
Africa/906166.stm〉 (June 23, 2007).
Bosire, B. Mokaya. 2008. Sheng: The Phonology, Morphology and Social Profile of an Urban
Vernacular. Ph.D. dissertation, State University of New York at Albany.
Bourdieu, Pierre. 1977. The economics of linguistic exchanges. Social Sciences Information 16:
645–668.
Bunyi, Grace Wangarĩ. 1996. Language, Education and Social Selection in Kenya: An
Ethnographic Study of Two Schools. Ph.D. dissertation, University of Toronto.
Daily Nation. 2004. Over 317,000 to miss Form One admission. January 6, 2004. Nairobi.
Daily Nation. 2008. Mixed marriages that speak in tongues. September 2, 2008. Nairobi.
Dorian, Nancy C. 1989. Introduction. In Investigating Obsolescence, Nancy C. Dorian (ed.),
1–10. Cambridge: CUP.
Encyclopedia of World Biography. 2005. Mwai Kibaki. 〈http://www.encyclopedia.com/topic/
Mwai_Kibaki.aspx〉 (7 August 2009).
Frederikse, Julie. 1992. All Schools for All Children: Lessons for South Africa from Zimbabwe’s
Open Schools. New York: Oxford University Press.
Fuller, Bruce. 1991. Growing Up Modern: The Western State Builds Third-World Schools. New
York NY: Routledge.
Furley, Oliver W. & Watson, Thomas. 1978. A History of Education in East Africa. New York:
NOK Publishers.
Githinji, Peter. 2006. Bazes and their shibboleths: Lexical variation and Sheng speakers’ identity
in Nairobi. Nordic Journal of African Studies 15(4): 443–472.
Hill, Jane H. 2001. Dimensions of attrition in language death. In On Biocultural Diversity: Link-
ing Language, Knowledge, and the Environment, Luisa Maffi (ed.), 175–189. Washington
DC: Smithsonian Institution Press.
Hinton, Leanne & Hale, Ken (eds). 2001. The Green Book of Language Revitalization in Practice.
New York NY: Academic Press.
Human Rights Watch. 2008. Ballots to bullets: organized political violence and Kenya’s crisis of
governance. 〈http://www.hrw.org/sites/default/files/reports/ kenya0308web.pdf〉 (7 August
2009).
Jaffe, Alexandra. 1999. Ideologies in Action: Language Politics on Corsica. Berlin: Mouton de
Gruyter.
Kabaji, Egara. 2007. Mother tongue and cultural identity. Standard. February 24, 2007. Nairobi.
King’ei, G. Kitula. 2001. Language and identity: The challenge of Sheng in Kenya. Paper pre-
sented at the Association of Third World Studies Conference, Egerton University, Kenya,
September 17–19.
Krauss, Michael. 1992. Endangered languages. Language 68(1): 1–42.
Kulick, Don. 1992. Language Shift and Cultural Reproduction: Socialization, Self, and Syncretism
in a Papua New Guinean Village. Cambridge: CUP.
Lieberson, Stanley J. & McCabe, Edward J. 1982. Domains of language usage and mother-tongue
shift in Nairobi. International Journal of the Sociology of Language 34: 83–94.
 Heidi Orcutt-Gachiri
Mafela, Lily & Mgadla, Part T. 2000. The historical context of education in British colonial
Africa. In Social Studies in African Education, Michael Bamidele Adeyemi (ed.), 1–11.
Gaborone, Botswana: Pyramid Publishing.
Mazrui, Alamin M. 1995. Slang and code-switching: The case of Sheng in Kenya. Afrikanistische
Arbeitspapiere 42: 168–179.
Mufwene, Salikoko S. 2003. Language endangerment: What have pride and prestige got to do
with it? In When Languages Collide, Brian D. Joseph (ed.), 324–346. Columbus OH: Ohio
State University Press.
Mutahi, Wahome. 1996. How to Be a Kenyan. Nairobi: Kenway Publications.
Myers-Scotton, Carol. 1976. Strategies of neutrality: Language choice in uncertain situations.
Language 52(4): 919–941.
Nettle, Daniel & Romaine, Suzanne. 2000. Vanishing Voices: The Extinction of the World’s
Languages. Oxford: OUP.
Newport, Elissa L. 2006. Critical periods in language development. 〈http://www.mrw.
interscience.wiley.com.ezproxy1.library.arizona.edu/emrw/9780470018866/ecs/article/
s00506/current/abstract〉 (1 August 2009).
Ngatia, Martin. 2007. Raila Odinga is fit to lead Kenya. 〈http://www.youtube.com/watch?
v=ZJL13ivOISQ&feature=PlayList&p=F222560BE0ED44F4&playnext=1&playnext_
from=PL&index=11〉 (12 April 2009).
Obondo, Margaret Akinyi. 1996. From Trilinguals to Bilinguals? A Study of the Social and
Linguistic Consequences of Language Shift on a Group of Urban Luo Children in Kenya.
Stockholm University: Centre for Research on Bilingualism.
Ochs, Elinor. 1990. Indexicality and socialization. In Cultural Psychology: Essays on Comparative
Human Development, James W. Stigler, Richard, Richard A. Shweder & Gilbert H. Herdt
(eds), 287–308. Cambridge: CUP.
Orcutt-Gachiri, Heidi Ann. 2009. Kenyan Language Ideologies, Language Endangerment, and
Gikuyu (Kikuyu): How Discourses of Nationalism, Education, and Development Have
Placed a Large, Indigenous Language at Risk. Ph.D. dissertation, University of Arizona,
Tucson.
Otiende, James E. 1990. Education since the early times. In Themes in Kenyan History, William
R. Ochieng’ (ed.), 145–155. Nairobi: Heinemann.
Prah, Kwesi Kwaa. 2005. Language of instruction for education, development and African
emancipation. In Languages of Instruction for African Emancipation: Focus on Postcolonial
Contexts and Considerations, Birgit Brock-Utne & Rodney Kofi Hopson (eds), 23–49. Dar
es Salaam: Mkuki na Nyota Publishers.
Rothchild, Donald. 1973. Racial Bargaining in Independent Kenya: A Study of Minorities and
Decolonization. Oxford: OUP.
Samper, David Arthur. 2002. Talking Sheng: The Role of a Hybrid Language in the Construction
of Identity and Youth Culture in Nairobi, Kenya. Ph.D. dissertation, University of
Pennsylvania.
Senanu, Kwadzo & Williams, Drid. 1995. Introduction. In Creative Use of Language in Kenya,
Kwadzo Senanu & Drid Williams (eds), v–ix. Nairobi: The Jomo Kenyatta Foundation.
Sow, Alfa I. & Abdulaziz, Mohamed H. 1993. Language and social change. In Africa since 1935
[UNESCO General History of Africa, Vol. 8], Ali A. Mazrui (ed.), 522–552. Berkeley CA:
University of California Press.
How can a language with 7 million speakers be endangered? 
Standard. 2007. Why mother tongues are dying. April 24, 2007. 〈http://www.mail-archive.com/
africanlanguages@yahoogroups.com/ msg00507.html〉 (25 April 2009).
Summer Institute of Languages. 2012. Ethnologue: Gikuyu. 〈http://www.ethnologue.com/
show_language.asp?code=kik〉 (8 January 2012).
Summer Institute of Languages. 2009. Ethnologue: Languages of Kenya. 〈http://www.ethnologue.
com/show_country.asp?name=KE〉 (25 April 2009).
Thompkins, Gwen. 2007. Ethnicity likely to play part in Kenya elections. All Things Considered
on National Public Radio. Filed from Kenya. 18 December 2007.
Westaway, Gill. 1995. The English language in Kenya: Development or decline, acceptance or
rejection. In Creative Use of Language in Kenya, Kwadzo Senanu & Drid Williams (eds),
1–8. Nairobi: The Jomo Kenyatta Foundation.
Woolard, Kathryn. 1998. Introduction: language ideology as a field of inquiry. In Language Ide-
ologies: Practice and Theory, Bambi B. Schieffelin, Kathryn A. Woolard & Paul V. Kroskrity
(eds), 3–47. Oxford: OUP.
A documentary ethnography of a Blackfoot
language course
Patterns of variationism and standard
in the “organization of diversity”
Annabelle Chatsis1, Mizuki Miyashita1 & Deborah Cole2

1The University of Montana / 2The University of Texas – Pan American
This chapter documents the development of a university-level Blackfoot

language course in which many of the students are linguistic inheritors
(Rampton 1990) of Blackfoot. In attempting to integrate “the study of the
culture of language into documentary linguistics” (Hill 2006: 113), we observe
how varied language ideological patterns among speakers and learners of
different linguistic repertoires came to be organized for the purposes of formal
language instruction. Analysis of classroom discourse reveals conflicting
language ideologies between variationism (Kroskrity 2009b) and standard
(Hill 2008). We propose a model of “Language Ideological Variation and
Emergence” (LIVE) to clarify how participant affiliations to competing language
ideologies can emerge and shift as different language ideologies come into contact
during discourse.
Keywords: Blackfoot; variationism; standard; language ideology;

language teaching
1. Introduction
A second foundational presumption of the ethnography of language is, of

course, that speech communities are not linguistically homogeneous, but are
“organizations of diversity.” The idea of the speech community as an “organization
of diversity” is a very useful one for students of minority languages who encounter
communities that are at the very least bilingual. Especially important, of course,
is the distribution of the linguistic resources of the minority language versus the
other language or languages across the repertoire of possible speech events and
acts, across genres, across the kinds of speakers and addressees, across channels,
across affective keys, and the like. This organization of diversity has very practical
consequences for our work. (Hill 2006: 117–118)
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
In this chapter, we document the development of a course for teaching Blackfoot, an

Algonquian language spoken in Alberta and Montana. The Blackfoot course was first
offered at the University of Montana in 2009. The students in the course included
linguistic inheritors (Rampton 1990) of different varieties of Blackfoot. Because of
the language’s particular sociolinguistic history (summarized below), the creation of
this language curriculum provided a currently rare opportunity for the Blackfoot lin-
guistic inheritors to be exposed to varieties other than their own.1 Unlike languages
typically taught in institutionalized settings, Blackfoot does not have standard forms
pre-selected. The need for explicit selection highlighted the existence of language ideo-
logical variation across the different user communities who had a stake in the language.2
Creating and teaching a Blackfoot language curriculum thus required the organization
of multiple diversities, ranging from formal diversity (phonetic, phonological, mor-
phological and lexico-semantic variation) to language ideological diversity, i.e. varying
ideas about which and whose forms should be used when and where. Both organiza-
tions of diversity were directly relevant for selecting what would be taught and learned
in the Blackfoot classroom. Our analysis demonstrates how emerging organizations of
ideological variation are directly linked to organizations of linguistic diversity.
In presenting and analyzing the ways that linguistic and ideological diversity came
to be organized for language teaching, we also try to “think through” the “three require-
ments” for “integrat[ing]…the study of the culture of language into documentary
linguistics” (Hill 2006: 113). We have organized the chapter according to these require-
ments: We begin by drawing on ethnographic practice using Hymesian “ethnography
of speaking” (Section 2). We then pay attention to locally emergent interactional norms
(Sections 3 and 4) and attend to the semiotics of language ideologies (Section 5). Hill
(2002) also noted that the ways in which naturally occurring linguistic variation gets
categorized into discrete languages have tended to benefit the categorizers and/or the
speakers of the variants that get selected as category prototypes. Thus, language catego-
rization processes not only sort variation into “typical” and “a-typical” variants, they
also sort people into “representative” and “marginal” speakers. In our data, we notice
how classroom talk influenced how variant selections were made for classroom instruc-
tion and conclude that the integrated approach Hill proposes can raise our awareness
of the effects linguists can and do have on how linguistic diversity becomes organized.
. We adopt Rampton’s proposal of a three-way distinction between language expertise, lan-
guage inheritance, and language affiliation as a more accurate and useful way of organizing
differences between language learners than the concepts of mother tongue or native speaker.
As re-presented in Leung, Harris and Rampton (1997), a person may inherit a language by
being born into a community which speaks it, whether or not she has the ability to use it
herself (expertise) or the desire to be identified with it (affiliation).
. This “lack” of standard has been similarly observed in language projects aiming to
establish standardized orthographies (e.g. Hinton 2010).
A documentary ethnography of a Blackfoot language course 
Within this integrated approach, we respond to the need within documentary

linguistics for language ideological clarification, specifically in collaborative projects
between linguists and community language experts (Hill 2006; Kroskrity & Field
2009). Encounters between the various participants in this curriculum development
project involved language ideological contact between two broad ideologies: Varia-
tionism, in which “dialectal variation is the expected outcome of family and individual
differences” (Kroskrity 2009b: 193), and the ideology of standard (Hill 2008), in which
one language variety is reified at the expense of all others. In proposing a model which
represents possible organizations of language ideological diversity, we hope to provide
linguists and community language experts with a concrete tool for clarifying language
ideological diversity. Given that linguists are almost exclusively linguistic inheritors
of standardized languages and standard language ideologies (Dorian 2010a), we hope
this tool can be used to recognize patterns of language ideological difference in order
to initiate and maintain more equitable interactional norms between people with
diverse language inheritances, interests, and training. Our work is also explicitly
linked to Zepeda and Hill (1998), who model collaborative best practice in linguistic
documentation and ethnographic writing.
2. E
thnography: The geographic, historical and ideological contexts
for linguistic diversity in Blackfoot
Speech communities will differ not only in manifesting different kinds of

language structures, but in manifesting different patterns of use. An ethnography
of the distribution of registers, speech-act types, and the like across the contextual
landscape is critical to linguistic documentation. (Hill 2006: 114)
There are four Blackfoot speaking tribes: Siksiká ‘Blackfoot’, Kainai ‘Many Chief ’,
Apátohsipikani ‘North Piegan’, and Aamsskáápipikani ‘South Piegan’ (Frantz 1991).
The English version of their tribe names, though not necessarily their translations,
are: Blackfoot, Blood, North Piegan, and Blackfeet, respectively. These English terms
refer to reservations (or reserves in Canada) as well as to tribal members. The first
three reserves are located in Alberta, Canada. The Blackfeet reservation is in the U.S.
“Blackfoot” is a cover term used to refer to these four tribal groups as well as to the
mutually intelligible dialects spoken by tribal members.3 For political reasons, these
. Two other tribes are recognized under Treaty 7 along with these four Blackfoot tribes
in Canada, but they are not in the Confederacy and do not speak Blackfoot. These are Tsuu
T’ina (Sarcee), who speak Dene language, and Stoney (Morley), who speak Nakota language.
Treaty 7 was an agreement made between the Crown and tribes in Southern Alberta in 1877 to
declare reserve land for the tribes within this region. This included the Blackfoot C
onfederacy,
Tsuu T’ina, and Stoney. Each tribe would have their own land base (or “reserve”) recognized by
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
four groups have been grouped together as the Confederacy of Blackfoot, in large part
on the basis that they share a “language”. Figure 1 shows the geographic locations of
the tribal lands of each of the four groups with respect to the political border between
Canada and the US.
ALBERTA
SASKATCHEMWAN
Calgary
B.C.
A. Detail
MONTANA
IDAHO WYOMING
B. Lethbridge
C.
Waterton
NP Cardston CANADA
USA
Glacier D. Cut Bank
NP
Browning
Heart
Flathead Butte
Lake
Great Falls
© Kevin McManigal
Figure 1. A. Blackfoot Reserve, Alberta; B. North Piegan Reserve, Alberta; C. Blood Reserve,
Alberta; D. Blackfeet Reservation, Montana. Cartography by Kevin McManigal, The University
of Montana.
2.1 Historical separation of land, language, and people

As far back as I (Annabelle) can remember, my grandparents told me stories of how
Niitsítapiiksi (the Blackfoot people, lit. the real people) visited one another for weeks
on end. They would talk about how large the camps were and how it would take days
for them to travel throughout the territory to get to where they were going. However,
that all changed when the U.S. and Britain entered into a war (declared in 1812 over
the Canadian government. There are eleven signed treaties throughout Canada with various
tribes. See (Dempsey 1978) for more. Hugh A. Dempsey, Indian Tribes of Alberta. Calgary:
Glenbow/Alberta Institute. 1978.
A documentary ethnography of a Blackfoot language course 
trading and water rights) that would divide the land base. This division would greatly
impact the Niitsítapiiksi and their way of life. The change caused a lot of confusion
for Niitsítapiiksi because they couldn’t understand how the government could divide
their hunting and camping grounds. This division also meant that the Niitsítapiiksi
would lose their freedom to travel throughout an area known as Blackfoot territory.
The Niitsítapiiksi continued to go back and forth for a while, not paying attention to
the “invisible line” dividing their territory.
My great grandfather was most affected by this change because he was known
among the Niitsítapiiksi as a great medicine man. He would travel throughout Black-
foot territory and beyond in search of various medicines to help the Niitsítapiiksi
in their time of need. My great grandfather was originally from Aamsskáápipikani
(Blackfeet tribe in Montana), and on one of his many visits to his relatives in the north
camps, the areas known as Kainai (C in the map) and Aapátohsipikani (B in the map)
in Canada, he became involved in a situation he did not anticipate. The Canadian gov-
ernment was taking a census of the tribes along the border when he was asked of his
name at a gathering. He told the census taker his name, and the census taker placed his
name on a list, which contained names of Kainai members. He was from that moment
considered a Kainai member and a Treaty Indian of Canada. At this time the Canadian
government assigned Indian Agents to each tribe, and this agent would be responsible
for each member’s coming and going and well-being. My great grandfather, who had
no intention of staying in Canada, continued to go back and forth throughout the ter-
ritory as he had always done. Due to the rapid growth of settlers occupying land areas
not assigned to various tribes, the Indian Agents decided to implement permits for
travel so as to better monitor the mobility of the tribal members. It was difficult for
Canadian tribal members to obtain this permit because the Indian Agent (assigned by
the Department of Indian Affairs) would determine the viability of the travel.4
During this time, my great grandfather believed he was still able to travel back and
forth without question, and on a visit to relatives, again in the north camps, he was
told by family there that he would have to obtain a permit to go back to his home land.
Since my great grandfather’s name was on the list of Kainai tribal members, the agent
didn’t believe that my great grandfather was from the U.S. side. The Agent continued
to insist he apply for a permit to travel. When my great grandfather gave his reason to
travel as gathering medicines and wishing to see his family, the agent didn’t see this as
a viable reason. So he was denied the permit. Only those men who were hunting or
looking for work could obtain a permit to leave the reserve. My great grandfather then
agreed to take work as a scout and monitor the invisible line for the Canadian govern-
ment. He continued to travel back and forth until it became too difficult for him to
travel. He stayed in Canada and raised his family.
. Obtaining such permits has become even more difficult since September 11, 2001.
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
2.2 Ongoing effects of geographic separation on language

It was the fall of my fifth year when my mother told me I needed to get ready as
I would be going to school the next day. I was excited and couldn’t wait. This meant I
would see my cousins who had already left for school. Going to school meant I was a
big girl now and that I would be entering a new adventure. Little did I know just what
kind of an adventure it would be. I really didn’t understand that I would be leaving
my mother completely for months and that I wouldn’t be seeing my father or grand-
parents for a long time. It wasn’t until I arrived at the residential school that I was told
that I would be there for some time and would only see my family on holidays and
summer months.5 The suitcase my mother packed for me with my best clothes was
taken away, and I was given different clothes to wear while I attended school. My hair
was cut short, and I had to shower with a very strong smelling soap that burned when
it was used. I was then given a number and told to remember that number because
that number would be on everything I had. The matron in charge told me to speak
only English and never to speak Blackfoot again. If I didn’t listen I would be punished.
My whole idea of the excitement of going to school and being a big girl soon faded
away. In the summer of my tenth year, it was time for me to leave the school and
return home because my father felt he wanted me home. There were a lot of changes
then. Some of my family moved to different areas of the reserve and others left alto-
gether. My grandmother and mother still spoke to me in Blackfoot, but encouraged
me to speak English. They said it would help me get a job, and I would be better off if
I didn’t speak Blackfoot as much.
My family was always busy preparing for the seasons and would always have
family members from other reserves like Siksiká and Aamsskáápipikani visit for long
periods. Throughout my life I was always aware that my relatives who lived in Siksiká
and Aapátohtsipikani and Aamsskáápipikani spoke a little differently. But it was still
Blackfoot. It wasn’t uncommon for me to be able to recognize where people were from
just by their dialect. It also wasn’t uncommon to tease one another on how we spoke
Blackfoot, because as young people it didn’t matter. It was always understood with
family members that being able to speak fluent English would mean you would sur-
vive and speaking Blackfoot was a way of communicating to one another in private. I
was speaking both English and Blackfoot, and I noticed that at home family members
would speak Blackfoot pretty freely, but when we were out of the home English was
spoken. At the time I never questioned the reason why. I thought that was the way it
was. It was only when I started teaching Blackfoot that I became aware that not all
Niitsítapiiksí knew there were different dialects among our people.
. These were called “boarding schools” in the U.S.

A documentary ethnography of a Blackfoot language course 
I soon left home and continued my education in a city just north west of my com-
munity. Speaking Blackfoot became less and less frequent for me, and when I married
a Cree man who spoke his language fluently I became familiar with the Cree language
and was able to say some words in Cree. When I called home or spoke with family
members, they would speak Blackfoot to me and I would respond in English, mostly
because I didn’t want to be rude to my husband but also I now felt uncomfortable
speaking Blackfoot. When my husband passed away, I made the decision to move back
home with my children and reconnect with family. My mother was much older now,
and it was time for us to be together again. I had been away a long time and was com-
fortable speaking English, so when I spoke Blackfoot to my mother, who spoke more
Blackfoot than English, I wasn’t confident to speak Blackfoot.
We often ran into conflict when I spoke Blackfoot because my mother would tell
me I wasn’t saying the words right or I would shorten the sentences. My mother would
tell me “just stick with English.” One conversation I remember was when we were
leaving to go shopping and I told her to get in the car. I used the word iitáísapópao’p
and she said “that is not how you say it”. She then corrected me by saying áínaka’si. At
first I was confused and a little hurt by her comments, but now I understand her rea-
sons for correcting me. I was saying words that were different from my mother’s way
of saying them, but I felt they still had the same meaning. I had little confidence now
when it came to speaking Blackfoot to my mother and older members of my family.
I felt more comfortable speaking Blackfoot to family members who were more my age,
because nearly all of us were getting reacquainted with Blackfoot and were using words
that made more sense to us like iitáísapópao’p or aiksisstomatoo for “car” rather than
áínaka’si which the older people would say.6
2.3 Linguistic diversity in Blackfoot

When I (Mizuki) first started researching in Blackfoot, I wondered why I kept hearing
and seeing both the name “Blackfoot” and “Blackfeet”. Soon I learned that “Blackfoot”
was a direct English translation of the word Siksiká. The morpheme sik means “black”
and -ika is a medial form meaning “leg or foot.” The medial form differs from the seman-
tically related independent form, mohkasts “leg or foot”. The [s] in the middle sik-s-ika
surfaces as a result of phonology as described in Taylor (1969) and Frantz (1991). The
English translation of siksiká appeared as a cover term for all tribes that speak one of
several mutually intelligible language varieties. By visiting Darrell Kipp, the director of
the Piegan Institute (the non-profit research organization on the Blackfeet reservation),
. While we were confirming the spelling of the word iitáísapópao’p for this chapter,
Annabelle produced several forms for the same word. The representation we give here is the
form found in Frantz and Russell 1995, which is one of the forms Annabelle gave me (Mizuki).
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
I learned that attendees at an intertribal Blackfoot teacher conference had agreed that
the word “Blackfoot” should be the term used to refer to the language. But this is not
widely known. I learn from people who I interview on the Blackfeet reservation in the
U.S. that the correct form is “Blackfeet”. My students who are from that reservation tell
me that they are “Blackfeet” and that their ancestral language is “Blackfeet”. For them,
“Blackfoot” refers specifically to the reserves in Canada. When I say I study Blackfoot,
people will often ask me, “Which is the correct form, Blackfoot or Blackfeet?” I say,
“both and neither.”
The varieties I study belong to the Algonquian family, specifically to the Plains
Algonquian branch (Mithun 2001). Based on a 2001 Canadian Census, Enthnologue
cites 4,500 speakers in Canada. The youngest fluent speaker generation is in its mid
50s. According to Goddard (2001), the speaker population in the U.S. is about 100.
Today, according to the research conducted by the Piegan Institute, the number of
speakers is probably around 60 and speaker age range is 80 and higher (p.c. Darrell
Kipp). Linguistic variation across Blackfoot dialects can be identified at the pho-
netic, phonological, morphological and lexico-semantic levels. Linguistic variants
exist along geographic and generational scales (Blommaert 2010). Geographic varia-
tion occurs at two levels: International variations, between the border-separated
Canadian Blackfoot and American Blackfeet and inter-tribal variations, regardless
of the national boarder. Examples in (1) show international geographic variation. An
example of phonetic variation appears in (a): The word for ‘potato’ starts with [m] in
Canada and [p] in the U.S. Example (b) also shows a sound difference in the root nins/
ninih for ‘sing.’ Examples (c) and (d) show vocabulary differences. The word for ‘cof-
fee’ is niita’psiksikimi (truly-black-water) in Canada, and aisiksikimi (black-water) in
the U.S.; the word for ‘tea’ is siksikimi (black-water) in Canada and áísoyoopoksiikimi
(leaf-black-water) in the US.
(1) Canada vs. U.S.

Canada U.S.
a. maatááki paataaki ‘potato’
b. ninskssini ninihkssini ‘song’
c. niita’psiksikimi aisiksikimi ‘coffee’
d. siksikimi áísoyoopoksiikimi
‘tea’
Inter-tribal variations within the same national boundaries are also found. Examples
are given in (2) and (3) below (Frantz & Russell 1995). As shown in (2), in Siksiká, near
Calgary in Canada, the word samákinn means ‘lance’ or ‘spear’ while it means ‘large
knife’ in Kainai, west of Lethbridge in Canada.7 Example (3) shows a grammatical
. Semantic information for samákinn in other tribes has not been described.
A documentary ethnography of a Blackfoot language course 
difference. In Siksiká, ‘it flew’ is naipottaawa, but it is íípottaawa in other Blackfoot

tribes. Thus, the form of the past tense prefix varies: na- vs. íí-.8
(2) Inter-tribal: semantics
samákinn ‘lance, spear’ Siksiká
‘large knife, machete’ Amsska’piipikani
(3) Inter-tribal: morphosyntax
naipottaawa ‘it flew’ Siksiká
íípottaawa ‘it flew’ Apatohtsipikani,
Kainai,
Amsska’piipikani
Linguistic variations are also observed across generations. Some cross-tribe members
refer to the form used by elderly people as High Blackfoot, and newer versions Low
Blackfoot. There are other terms for these variations: Old Blackfoot and New B lackfoot
respectively. The latter terms are preferred here as they are considered to be value
neutral. Old Blackfoot has been described by Uhlenbeck (1938), Taylor (1969), and
Frantz (1991). New Blackfoot has not been extensively described, but it is now spo-
ken by some members of the younger generations. Forms vary between Old and New
Blackfoot, especially in morphology and syntax. For example, New Blackfoot speakers
tend to drop suffixes at the end of verbs. As a result, New Blackfoot speech may not
indicate proximity such as fourth person and fifth person, while Old Blackfoot exhibits
these distinctions.
There are also differences in choice of lexical items, as appeared in Annabelle’s
anecdote in the previous section. Annabelle used the word, iitáísapópao’p, referring
to a car. Then her mother “corrected” her by suggesting she use áínaka’si instead. After
she wrote that passage above, I asked her about this exchange. Annabelle recalled
that the vehicle she was referring to was a wagon model car, and Annabelle chose to
use the word iitáísapópao’p which can variously mean “a car, container, or receptacle
one sits in.” However, her mother insisted on using the word that refers specifically
to a wagon, áínaka’si. That word had a different meaning for Annabelle. The literal
meaning of áínaka’si is “it rolls,” and for Annabelle, this refers to a smaller wagon
with a person dragging it. Here we observe semantic diversity between Annabelle’s
and her mother’s Blackfoot. Table 1 presents the multiple words that can be used to
refer to a type of vehicle in Blackfoot as included in the dictionary by Frantz and
Russell (1995).
. There is a variety of strategies to express past expression in Blackfoot. The examples
shown here use a prefix to indicate a past expression. Having no affix which indicates tense/
aspect tends to be interpreted as past expression. See Frantz (1991) for more information.
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
Table 1. Words for vehicle from Frantz and Russell (1995)
iitáísapópao’p car/container or receptacle one sits in
áínaka’si wagon, lit: it rolls
aapátataksáakssin truck box or a type of wagon, lit: box in the back
ikkstáínaka’si buggy/wagon used for leisure purposes
omahkainaka’si wagon used for utilitarian purposes
iitawáí’pihtakio’p truck/hauling vehicle
áíksisstoomatokska’si automobile
aiksisstoomatapisttsipatakkayayi automobile
aiksisstoomatomaahkaa automobile
aiksisstoomatoo automobile
Presumably, which of these words are selected depends on an individual’s

language experience. When users’ experiences do not overlap, it is understandable
that users of the same language will use different variants. Further, when users have
less opportunity to use the language together, it is understandable that they might
miss opportunities to find naturally occurring consensus in grammar and lexicon.
2.4 V
arying awareness of linguistic diversity in Blackfoot and varying
language ideologies
As can be seen from the examples above, the existence of linguistic variation is an
essential, defining characteristic of Blackfoot that is inescapable in any attempts
to accurately describe the language. However, tribal members (speakers and non-
speakers) are not uniformly aware of this fact. Whereas the geographic and gen-
erational diversity of forms (such as that between Annabelle and her mother or
between Annabelle and her family members from other reserves) were relatively
salient to Annabelle growing up (and would be quite noticeable to scholars), such
differences are not always transparent to tribal members. Such varying awareness of
Blackfoot’s diversity is of course grounded in individuals’ particular sociolinguistic
experiences, but it also due in large part to the particular history and politics of
the language (described above) that have created patterns of usage that obscure
Blackfoot’s diversity for many users and learners. Specifically, evolved patterns of
usage has meant that speakers no longer have contact with others who would use
different forms.
One pattern is that speakers tend to only use the language with members of their
own generation. In Canada, people fifty years and older still speak the language, but
A documentary ethnography of a Blackfoot language course 
their spoken variety is distinct from that of the elderly speakers (ages eighty and
higher). Elderly speakers in the U.S. are usually the only member of their family to
speak Blackfoot. The language is not used in the home, and there are a very few oppor-
tunities for them to use the language with others in their age group.
Another pattern is that Blackfoot speakers will most often engage in conversations
in Blackfoot only when they are certain that their interlocutor also speaks Blackfoot.
With everyone else, they tend to exchange forms of English. Because of this, Blackfoot
speakers could be speaking English without knowing their conversational partner also
speaks Blackfoot.
Current young tribal members who are English monolinguals thus have very
little opportunity to hear Blackfoot in their daily lives. It follows, then, that younger
generations typically would not have opportunities to hear different varieties of Black-
foot and therefore would not be familiar with the co-existence of inter- or intra-tribal
linguistic variation. This is especially true of younger generations in the U.S. which is
where most of the students in this study are from. They also seemed unaware of the
fact that there are not very many fluent speakers at all left in their communities, and
those that are left belong to the oldest generation. Further, students who had a grand-
parent who spoke Blackfoot also tended to believe that everyone in the grandparent’s
generation also speak Blackfoot.
Older Blackfoot community members, on the other hand, may be well aware that
linguistic variation exists within and across their communities. For example, during
an interview for another project a fluent elderly speaker from the Blackfeet reservation
expressed his perception of Blackfoot language variation. He stated that there is one
Blackfoot language and there is “no difference” in Blackfoot among the four tribes.
However, in response to the follow-up question, “Are there differences in how people
pronounce words or can people use different words to refer to the same thing?”, he
acknowledged the existence of variation. We take his “no difference” to mean “mutual
intelligibility”. For him, variation was normal, and all those variations were to be
equally respected by speakers.
Not all awareness of variation in Blackfoot is uniformly neutral or positive,
however. While some elderly speakers simply see linguistic differences as facts about
the language, other elderly speakers will comment that younger speakers’ varieties
are “not correct” or “inappropriate”. For some, saying “one language” does not mean
“one form”, for others, it does, however. Some elderly speakers pick on New Blackfoot
speakers for using “incorrect expressions”, and their comments discourage younger
speakers from using the language. We discovered that the language ideology of varia-
tionism appears to be “normal” only for the generation of people who are highly pro-
ficient in the language (and who are often well respected for it), but not for others
including current learners.
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
3. Locally emergent interactional norms 1: Roles and resources
So the notion of the community of practice teaches us that the ethnography of

language in documentary linguistics must take as its site for study not only the
organization of diversity in the speech community, but also organization and
patterning that is emergent, including emergent in the context of elicitation and
language learning itself. (Hill 2006: 124)
3.1 Initiating collaboration and inhabiting reciprocal roles

I (Mizuki) vividly remember my first meeting with Annabelle. It was in the fall
of 2007. She came to my office and asked a few questions. I don’t remember what
exactly she was asking me but I think it was about my research. She was a student
majoring in Native American Studies then, finishing in a year. Someone told her
about me – that I was someone who researches her native tongue. I explained to
her that I had studied Tohono O’odham from Ofelia Zepeda as a graduate student
and started to study Blackfoot when I came to Missoula. I told her I didn’t speak
the language but would like to. Some people assumed I spoke Blackfoot because I
could say oki tsa nita’pii (hello, how are you?). I also think I said I am interested in
doing research that would both contribute to linguistics and have practical applica-
tions, like for teaching. In the middle of our conversation, Annabelle had a phone
call from someone and she answered in a language that was not English. I assumed
it was Blackfoot, and she told me right away that it was and that she spoke it with
her family and friends.
Later, I met her again. This time, it was during a Blackfoot class offered by the
Blackfeet Community College to students at the University of Montana via teleconfer-
ence (Annabelle was serving as the onsite teaching assistant for the U of M students).
There I asked her whether she would be interested in helping me with my research.
She was busy with her class and teaching assistantship, so we did not discuss exact
days or times. I hoped to see her again to ask her a big favor. Then I saw her on campus
in the summer. Classes weren’t in session, so we both had more time than during the
semester. She had just finished her degree and told me that she was working again as
an onsite teaching assistant. I asked her if she would teach me some Blackfoot phrases,
and she said yes. We would meet at the University Center, and talk over a cup of coffee.
That was the time I learned that istopiit ‘sit down!’ and apiit ‘sit down!’ had separate
functions: provide a seat for a visitor vs. direct a person in the room, respectively.
She tutored me a few times – I was a happy student.
Towards the end of the summer, I asked her if she would help me with my
research as a language consultant. She said “yes” again. Annabelle and I met almost
every week to listen to Blackfoot conversations which I had recorded, and we tackled
the task of transcribing the speech together. During our transcribing sessions, I had
A documentary ethnography of a Blackfoot language course 
to ask her many questions about the same word multiple times. These included ques-
tions about phonological segments, word boundaries, general meaning, and literal
meaning. I felt that it was uncomfortable for her when she had to give me transla-
tions and literal meanings of the phrase we were listening to. It seemed to her that
the English translation was so different from the literal meaning of the corresponding
Blackfoot, and yet the literal meaning did not make sense at all in English. She was
almost feeling bad about giving unintelligible literal translations. For example, the
phrase used as “equivalent” to good-bye is kiatamattsin which is a short form for
kitaakitamattsin or more formally, kitaakotamattsinoo. When we analyzed the literal
meaning of the morphological pieces in order, the morphemes are “you-will-later-
again-see-I.” I must have sounded like a weird person wanting to know what part
of kitaakotamattsin ‘see you later again’ means ‘again’ because no Blackfoot speaker
takes these pieces apart. Annabelle did not know then that I was immensely enjoying
those literal bits and pieces in order.
Then in the beginning of the following year during the spring semester, Annabelle
visited my office and told me that the Blackfoot course she had been TA-ing was not
going to be in a teleconference format anymore. Instead, she had an opportunity to
teach the course as a face-to-face class. She asked me if I would be willing to help her in
developing a course. I said yes! We immediately started meeting every day, as we had
about a week to prepare for the class before it started. After we started developing the
course, I also had opportunities to meet Annabelle’s friends and family. At those meet-
ings, she introduced me as her teacher, which was rather flattering because I never
felt I was her teacher. I was sharing with her what I learned about Blackfoot structure
and orthography, but to me she was my teacher because she knew how to say things in
Blackfoot. Thus began our reciprocal relationship in which we took turns taking on the
roles of “teacher” and “student”.
3.2 Learning about teaching languages

A year after I (Annabelle) had been graduated, I was approached to teach the “Intro-
duction to Blackfoot I” course by the Department of Native American Studies. The
Chair of the department at the time was aware that I could speak Blackfoot fluently.
I was hesitant because it had not been my plan to teach. I was more interested in
research in history, and I never even considered teaching the language. Teaching the
language as opposed to speaking it was a whole different story, and I had so little time
to prepare. I realized I did not have any idea where to start or what the lesson plans
should be. I decided to talk with Mizuki, my colleague in the Linguistics Program, with
whom I was already working as a language consultant for her B lackfoot transcription
project. When I asked if she would be willing to work with me to develop teaching
materials appropriate for a college level language course, she agreed. So I accepted
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
the offer to teach the course with the understanding that she and I would collaborate
to develop the curriculum. Since I was assigned to teach the course the first time
without any teaching experience, the types of support Mizuki needed to provide were
beyond simple course development. It also involved general preparations for teach-
ing a course. She shared with me course syllabi for her linguistics courses, and we
talked about course objectives, learning outcomes, testing, and grading rubrics she
had used.
I (Mizuki) was aware, through personal communication, that the language teach-
ing method known as “total physical response” or TPR (Asher & Adamski 2003) was
viewed highly by Darrell Kipp, the director of Piegan Institute. The method involves
the teacher speaking only in the target language to give students commands to act, and
from which they deduce the meanings of specific lexical items. I talked with Annabelle
about the possibility of using TPR in her Blackfoot class. One day, I demonstrated (to
be specific, I pretended) a TPR lesson in Japanese for her in my office. I used things
that were in my office such as books and mugs to “teach” the words for colors and
numbers, showing how these words can be combined with nouns and verbs in Japa-
nese. Moreover, I began to think that in order for Annabelle to get a better idea of how
to teach a second language in a college level course it would be a good idea to observe
foreign language courses. I contacted instructors of Russian and Japanese at our insti-
tution and asked permission for us to audit a few class sessions. I chose these languages
because Russian and Japanese are considered to be “uncommonly taught languages”
in the U.S. (and because I already knew them). Visiting these classes seemed to help
Annabelle think about using different teaching methodologies. It was important to me
that she would be able to imagine her own instruction practices, not just the general
structure of Blackfoot, so she would be able to develop course materials that best fit
her classroom environment.
We (Mizuki and Annabelle) attended a second year Russian and a first year
Japanese course. In both classes, we sat in the back corner of the classroom. It was
a sensational experience for both of us, who were in sore need of language class-
room experience. The instructor in the second year Russian class spoke exclusively
in Russian, except on a few occasions during grammatical explanations. The students
answered the instructor in Russian and carried on short conversations in the target
language. In the first year Japanese class, the instructor also spoke mostly in Japanese
and students responded in Japanese. They appeared to be enjoying studying these
languages. As neither of us had any language teaching background, we furiously took
lengthy notes. From these visits we noted the use of the following methods: (i) the
use of the target language as much as possible during class, (ii) the use of props (such
as photographs and drawings) to supplement the necessary grammatical explana-
tions, (iii) the use of role play among the students, (iv) a focus on reading and writing
practices, and (v) the mixing of these various teaching methods within a single class
A documentary ethnography of a Blackfoot language course 
meeting. We took these observations back to our curriculum development meetings,

and we both thought these methodologies could be applied to teaching Blackfoot
language.
3.3 Finding unequal distributions of resources

Typically, second language instruction and testing makes use of an accepted s tandard
version of a language. From our extensive prior conversations regarding variations in
Blackfoot, we knew that dialectal differences existed and could sometimes be used
to identify which part of the tribe a speaker is from. We did not realize at this point
how these variations would bring ideological conflict to our teaching tasks, though
our observations in the Russian and Japanese courses made us realize that having no
standardized Blackfoot dialect would make the presentation of course material com-
plicated.9 This realization highlighted the differences between teaching two types of
languages: languages with a long history of institutionalized instruction and languages
without such a history. Blackfoot is the latter type. Although B lackfoot has been taught
as a subject, it has not been used as the language of instruction within the institutiona
lized setting of school except for at the Cuts Wood School at the Piegan Institute.10
From observing the courses in languages of the first type, I (Annabelle) also noticed
that languages with a long history of formal language teaching have ample materials to
support instruction, even if they have a status of being “uncommonly taught” in U.S.
universities. Such ample instructional resources are available when languages have
been taught formally to children of the language community and to second language
learners. For both Russian and Japanese, for example, there are quite a number of
ready reference materials that instructors can utilize and students can access.
When it came to Blackfoot teaching materials, there were not many options.
Initially, we knew of resources such as descriptive grammars by Uhlenbeck (1938),
Taylor (1969), and Frantz (1991), and a dictionary (Frantz & Russell 1995) because
these were the main sources used for linguistics research. In further research, we
were able to gather materials created by the tribes: a book by the Kainai Board
of Education, a Siksiká language book for Siksiká elementary students, language
books for members of the Kainai tribe (Lena Russell), CDs and notes for the course
offered by Blackfeet Community College (Weatherwax 2007), and several other pri-
vately created materials. The most accessible resources were the B lackfoot Dictionary
. In the Russian class we visited, we remember hearing the instructor mention some
dialectal variations in class, but the discussion was limited to the information, and no more
extensive questions were raised by the students.
. Piegan Institute, which was founded in 1987, established the immersion program in 1995.
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
(Frantz & Russell 1995) and The Blackfoot Grammar (Frantz 1991) because they are
still in print; they are also well-received by the tribes. From among these materials,
we did not pick one or exclude any for the purposes of course development, but
we decided to create our own materials using bits and pieces from this collection.
Although we did not have a textbook, we decided to use Don Frantz’ grammar book
and dictionary as references in class, particularly because we are in contact with
the author and could consult his expertise in Blackfoot grammar for clarification
if needed.
In developing instructional materials, we were reminded of the lexical differences
between tribes (e.g. nita’psiksikimi vs. aisiksikimi ‘coffee’ as described in Section 2).
This fact was a challenge for the development of course material: Were we to present
multiple versions of the same word? Once we moved into the classroom we noted that
this challenge also affected the students, who were already familiar with the idea of
“standard language” from their prior experiences with institutionalized English lan-
guage instruction. Should we introduce them to multiple forms without making them
acquire a “single correct form” in Blackfoot language instruction? This would prove to
be the most challenging aspect of taking on the roles of instructor and student in our
Blackfoot language classroom.
4. Locally emergent interactional norms 2: Awareness and evaluation
Work on communities of practice, specifically learning communities, provide

very useful theoretical foundations for understanding what is likely to go on in
these most dynamic of local systems, where goals and routines are negotiated at
the level of distinct individuals. (Hill 2006: 113–114)
4.1 Encountering other awarenesses of Blackfoot diversity in the classroom

My first class experience was a bit nerve wrenching because I (Annabelle) had never
taught in a classroom before let alone the Blackfoot language. But once I got over the
initial shock and started to teach, it became easier as time went on. As noted above, I
had grown up in contact with relatives in Siksiká, Aamsskáápipikani, and Pikani, so my
sociolinguistic background had familiarized me with the dialectal diversity present in
Blackfoot. I started providing instruction based on my own dialect, Kainai. But some
of the students (particularly those who were from the Aamsskáápipikani or “Blackfeet”
reservation) soon made it quite clear to me that they were confused by my usage of
some words. Several students who were linguistic inheritors of Blackfoot let me know,
“Well, that isn’t what I hear in my community, why is it that you’re teaching it this
way?” I had entered the classroom taking it for granted that anybody from the reserves
A documentary ethnography of a Blackfoot language course 
or reservations would be aware of the difference in dialects. It hadn’t occurred to me

that there were some Niitsítapiiksi, especially among the younger generations, that
didn’t know dialectal variation existed.
When the forms I gave were different from forms students had heard previously,
they asked questions. For example, one student who had grown up on the Blackfeet
reservation in the U.S. confronted me with, “My grandmother says this, why do you
say it that way?”. My initial reactions were to feel unsure that I was saying the word
correctly and to think that maybe the student was misunderstanding. But as we went
along, as we talked about it, I realized that the student was not aware of the difference
in dialect. She had grown up with a grandmother who was the only fluent Blackfoot
speaker in their home at the time. Often her grandmother would ask for tea, and she
would say áísoyoopoksiikimi. I had used the word siksikimi for tea. I went on to explain
the difference in dialects and the different meanings for the two words: Siksikimi
literally means “black water”, and it is most often used in Kainai. Áísoyoopoksiikimi
literally means “leaves turn water black”, and it is often used by older Aamsskáápipikani
(Blackfeet) speakers.
When the student continued to show interest and asked, “Why are they so differ-
ent?”, I explained that at one time the various tribes were separated by territory only,
and they could visit one another freely. There probably hadn’t been much difference
in the dialects before. But now with the Canada-U.S. border separating the Northern
Blackfoot from the Southern Blackfoot and with the introduction of English, there
are significant differences in how members of each tribe speak Blackfoot. The student
said she thought the Blackfoot she grew up hearing was how all Blackfeet/Blackfoot
spoke. There were at least seven students from the Blackfeet reservation in my class
that first semester I instructed the course who were surprised to learn about dialectal
diversity. They so seldom heard Blackfoot spoken in their community that even those
who had living relatives who still spoke the language in one of the three northern
tribes didn’t realize there were differences. Once students realized there were multiple
versions, some asked why a Kainai speaker (from Canada) was teaching the class and
not someone from Aamsskáápipikani (the Blackfeet reservation in the U.S.).
Students also asked questions when the forms and pronunciations I provided
orally in class differed from the written forms students had in their course reference
materials, which were based on multiple dialects. Many students noticed such differ-
ences, and both Blackfoot and non-Blackfoot alike would ask questions like “Which
is the correct form?” or “Which one should we remember?” Once, an L2 English
speaker from Europe with a language studies background asked “The dictionary says
this (iiksisstoyiiwa, ‘it’s cold’), why do you say it differently?” (I say iiksisstoyii, ‘it’s
cold’). That time I explained that the dictionary lists words used by elderly speakers
and the difference was generational.
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
I wanted the students to explore every aspect of the language, so when they asked
these questions I would say, “There is no correct form or one way of saying a word.”
For me, the main thing was for them to try to learn the language, and I felt frustrated
because they expected more concrete answers to their questions about the language’s
diversity. Their questions made me wonder: “Why is there dialectal variation? We’re
all speaking the same language, so what makes the differences?”. I thought maybe it
had something to do with the location, the area where speakers live, or maybe it had
something to do with the way the language is starting to be lost and how we’re losing
parts of the language. Their questions made me realize that there have been changes
in the way certain words are said in my generation from the way my mother would
say them.
All this caused me to want to do more research on how my generation uses
the language. For class, I had to think about what I was doing with the dialects
and decide how I was going to talk about them with students. I wondered whether
I was teaching the class appropriately. I found myself using the dictionary more
to check the spelling and pronunciation, not so much because I worried whether
I was speaking the language “correctly”, but because I became aware of my teach-
ing methods and ways of presenting the material, i.e. the language. As the course
went along, it became obvious that I would have to teach students explicitly about
variation in Blackfoot, but there were no models for teaching multiple dialects of
a language.
4.2 Teaching and evaluating Blackfoot diversity

Two facts about the student population became particularly important to classroom
practice. First, the university students entering the Blackfoot course did not have expe-
rience with variation in Blackfoot, and second, students brought to the classroom the
idea that if multiple variants did exist, only one would be correct. I had to work hard
to help them understand the mindset that variation was the norm when I was growing
up. The first time I taught, I wasn’t very clear about this. I took it for granted that they
would accept my way of teaching the language because I am a Blackfoot speaker. It
wasn’t until several students let me know that they noticed differences between what
I said and what they’d heard before that it became almost a challenge for me to revisit
my language and readdress the way I had planned on instructing it.
So the next time I taught the course I made it clear. I introduced who I was,
what my dialect was, and that I was very respectful of the other dialects and that we
would be learning them as well. I included an introduction about dialectal varia-
tions. I talked about the differences in speech among native speakers on the first day
of class. I thought it was necessary to do so up front. I needed the students to realize
A documentary ethnography of a Blackfoot language course 
that the dialects are very important, because the dialects for me as I grew up were a
way of identifying where you were from. It wasn’t a matter of whether or not someone
was speaking the language properly or correctly, or whether it was an Old Blackfoot
or a New Blackfoot form. As an instructor it was important for me to ensure that
every student recognized the different dialects in the various regions and could be
respectful of that fact.
I pointed out that the same could be said about other languages: You have dialects
within any European language which identify where the people speaking them are
from. I kept introducing the students to the differences, and in every class there were
discussions or remarks about different ways of saying the same word. I did my best
to find the variations and the examples I would use for certain words so they could
see the difference in the dialects. It didn’t take them long to understand it once they
saw it. Some variants would have the suffix -wa at the end, some wouldn’t. Even the
word for “good bye”, kitaakotamattsino (see you later again) could be used as a way to
recognize dialectal differences, because some people use kitaakotamattsin and some
drop off the -o suffix (probably reflecting a generational difference). The way I modi-
fied my teaching style was received well by the students, and since the variations were
introduced in class, I could be flexible with how the students were graded. When I told
them that the variants being taught in class that day would be the ones used on the
exams, they were satisfied with my answer.
As I had informed my students, the exams made use of forms that were used in
class. However, when grading the exams I also accepted not only the forms introduced
in class, but also included forms that students produced which were acceptable by my
own intuition. Examples from the final exams from the spring semester 2011 illustrate
the variety of accepted answers. The exam included translation items from English
to Blackfoot. The anonymous students (A, B and C) are all members of Niitsítapiiksi:
A and B are from Aamsskáápipikani and C is from Siksiká. Item 5 asks students to
translate the phrase “Stand up!”. The form that I had given in class is “iipoopoyit!”.
The dictionary gives the form “niipóípoyit!” (Frantz & Russell 1995: 76). Student A
has niipoipoyit! as it appears in the dictionary, only without accent markers. B has
nii’poypoyit. The differences in this case include the use of y instead of i and the addi-
tion of a glottal stop. Student C wrote iipaipooyit. This is not the form given by instruc-
tion or the book, but the fact that I can understand it lead to my decision to give the
student a point.
Additionally, item 9 asks students to translate “it’s a nice day”. Here, students
A and B gave the form of i’taamiksistsikowa, and C gave iiki’taamiksistsikowa. The
difference between the two answers is the choice to have iik at the beginning or not.
Morphologically iik used to have the meaning of “very” and functioned as an inten-
sifier, but today this morpheme does not necessarily have this function. Of course,
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
either form is accepted. These examples show the diversity represented by the stu-
dents’ answers and how this diversity was evaluated within the formal instructional
context.
Student A: (Aamsskáápipikani)
Student B: (Aamsskáápipikani)
A documentary ethnography of a Blackfoot language course 
Student C: (Siksiká)
5. Clarifying the ideological patterns of variation and standard
History speeds up at the margins…[and] oppression and marginalization…

produces a special intensification of language ideological projects…Recent
advances in our understanding of the semiotics of language ideologies provide
very useful tools for documentary linguists, who must be able not only to identify
and work among clashing ideological discourses, but assist communities with
what Nora and Richard Dauenhauer (1998) have called “ideological clarification”
to bring these discourses into line with what a community truly desires for
endangered-language resources. (Hill 2006: 114)
In the preceding sections, we took on the first two of Hill’s requirements for integrating
studies of “language and culture” into documentary linguistic practice (i) by providing
an ethnographic account of the sociolinguistic contexts and events in which this Black-
foot course unfolded and (ii) by describing some interactional norms that emerged
within the research and teaching communities made up of the individuals participating
in the project. Drawing on my (Debbie) interest in language ideologies, we turn now to
Hill’s third suggestion and take up the “arcana” of “the semiotics of ideology formation”
in an attempt to clarify the language ideologies at stake in this project (and in the hopes
of demonstrating how such an understanding can be useful).
The Blackfoot language course we documented above is a case study for under-
standing how patterns of competition between the language ideologies of standard
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
and variationism can emerge, shift, and spread over time and across speech events as
language users come into contact with others whose ideological patterns differ from
their own. Such contact has important consequences for the linguistic repertoires
(Blommaert 2010) to which they are attached, often determining which forms will
continue to be learned and spread and which ones will not (Hill 1995; Reynolds 2009;
Dorian 2010b). Thus the ability of linguists and community language experts to recog-
nize language ideological patterns is potentially a very important factor in the success
of documentary linguistic projects as well as for the survival of particular linguistic
varieties (Kroskrity & Field 2009; Flores Farfán & Ramallo 2010).
Drawing loosely on Optimality Theory (or OT), which models how the macro-
level patterns we call “languages” emerge from various rankings of a set of universal
constraints (Hammond 1995, 1997), we model how the macro-level patterns we call
“language ideologies” emerge in discourse. In this section, we propose a model of ide-
ological patterning and use it to (i) organize the ethnographic observations presented
above and (ii) analyze the effects that contact between different language ideological
patterns had on the discourse, selection, and evaluation practices of the participants
in the Blackfoot course. Following analytic convention, we abstract away from the
personal specificity of the ethnographic style used previously and refer to the partici-
pants solely in terms of the roles they inhabited in the particular contexts and events
we documented.
5.1 M
odeling patterns of language ideological variation
and emergence (LIVE)
Language ideologies are universally present in human language (Irvine & Gal 2000).
They are produced in discursive practices common to language use in general and are
acquired through socialization into communities of language users.11 In our model,
standard refers to the robustly attested language ideology that there is a single lan-
guage variety that is “best” and which is associated with an “appropriate” or sometimes
“superior” sociolinguistic identity (Silverstein 1998, 2003; Hill 2008; Lippi-Green
2012). variation refers to another widely-attested but competing language ideol-
ogy, variationism, that sees linguistic variation as normal and expected (Kroskrity
2009b) and attaches little to no social/hierarchical significance to formal difference
(Dorian 2010a). We represent these widely observed, shared ideas about language that
are applied by language users to a range of linguistic resources (including d ialects,
registers, voices, and genres) with the formalizations variation and standard

(written in small caps like the grammatical constraints of Optimality Theory).
. By general discursive practices we mean patterned semiotic behaviors that link language
to identities, such as in the tactics of intersubjectivity (Bucholtz & Hall 2004) by which l anguage
users identify themselves and others as members of different groups.
A documentary ethnography of a Blackfoot language course 
A more substantive borrowing from the OT model is the formal process of rank-
ing. In OT, a constraint which appears in a column to the left of another is said to
“outrank” or “dominate” any constraints which appear to its right. By ranking and
re-ranking these constraints, OT models of grammars can represent different lan-
guage patterns, or “languages”, all of which can be derived from the same universal
set of constraints. We use this simple formalization to rank variation and standard
with respect to each other. This formalization can be used to model two possible pat-
terns: One in which variation is higher-ranked, i.e. variation >> standard (read,
“variation outranks standard”) and one in which standard is higher-ranked, i.e.
standard >> variation. Compare, for example, patterns A and B in the model in
Table 2. These two possible rankings alone, however, are insufficient for representing
the language ideological complexity we observed in the Blackfoot language course.
An important variable in our observations was the degree to which variation and
standard were implicit or explicit.12 By introducing a formal feature to differen-
tiate between language ideologies which were present implicitly (im) versus those
which were articulated explicitly (ex), we can now represent eight possible patterns
(as shown by A–H in our model).
Note that the LIVE model represents the clashing language ideologies of
variation and standard as being continuously available to users: Both play an ongo-
ing role in structuring the specific ideological pattern that characterizes discourse and
practice. The ranking formalism enables the modeling of how one of these ideologies
exerts more influence than the other in particular contexts, at particular moments in
time, and in particular discursive exchanges.
Table 2. LIVE model: Patterns of language ideology and variation emergence
Pattern Higher-ranked Lower-ranked Participant affiliations
ideology ideology
A variation (im) standard (im) instructor before the language course

B standard (im) variation (im) instructor enters the classroom
C variation (ex) standard (im) instructor transitions in the classroom
D standard (ex) variation (im) students in and out of the classroom
E variation (im) standard (ex) not observed
F standard (im) variation (ex) not observed
G variation (ex) standard (ex) instructor in the classroom now
H standard (ex) variation (ex) not observed
. Michael Silverstein (1981) famously noted the importance of metalinguistic awareness to
theories of language ideologies in general.
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
5.2 Language ideological patterns and participant affiliations

We are now ready to use our model to organize our documented observations of the
various ideological positions taken up by the teacher and the students in the Black-
foot language course (summarized in the “participant affiliations” column of the
model). In pattern A, variation and standard are both implicit, but variation is
dominant: variation (im) >> standard (im). This pattern characterizes the instruc-
tor’s ideological position with respect to Blackfoot before she began preparing for
the course. Her language background included contact with older speakers of mul-
tiple varieties of Blackfoot, and like them she viewed dialectal variation as normal
while seeing all varieties as belonging to the same language. This particular ideologi-
cal pattern was not something that was explicitly articulated, however. It would only
emerge explicitly later when the instructor reflected on classroom interactions with
her students. As the instructor began to prepare to teach her language in a classroom
setting, the relative ranking of standard and variation was reversed (pattern B).
In looking for written instructional resources to provide a specific model for student
acquisition, standard (im) outranked variation (im). Both ideologies remained
implicit during this process, as the instructor had no overt plan to teach one par-
ticular dialect at the expense of others and no such a thing as a standard Blackfoot
dialect exists.
Patterns C and D can be observed in classroom dialog between instructor and
students. Students brought to the classroom the expectation of finding pattern D: An
explicitly articulated Standard, standard (ex), dominating an implicit variationism,
variation (im). They had no doubt previously encountered and acquired this pattern
in institutionally-provided English language courses.13 This pattern also characterizes
the way that students who had had prior contact with Blackfoot viewed Blackfoot out-
side the classroom. These students typically had only encountered one variety, and
sometimes only a single speaker. Without contact with Blackfoot diversity, students
assumed a single form: the one their grandparent spoke. Unlike her students, how-
ever, the instructor was well-aware of the diversity present in Blackfoot communi-
ties outside the classroom. The contact and conflict that arose in the classroom as she
addressed students’ desire to learn the “right” form forced reflection and conscious
decision-making to be explicit about her own higher ranking of variationism with
respect to Blackfoot. As she selected and presented students with multiple possible
forms and explained the socio-historical reasons for linguistic diversity in Blackfoot,
the instructor maintained the relative ranking of variation over standard from her
. See Cole and Meadows 2013 and Dorian 2010b for discussion of this widespread
phenomenon in language education contexts.
A documentary ethnography of a Blackfoot language course 
experiences with Blackfoot outside the classroom (as in pattern A), but she was now
explicit about variationism (pattern C, variation (ex) >> standard (im)).
Over time in multiple iterations of the course and across various discursive
exchanges in Blackfoot classroom communities of practice, pattern G emerged for the
instructor. As she became aware that a higher-ranked standard (ex) characterizes
the “typical” language classroom and that her students will not have had contact with
Blackfoot diversity, she knows her students will expect pattern D for Blackfoot as
well. However, she maintains her own relative ranking of variation over standard
in her instruction of Blackfoot by being up front about the formal differences that
characterize different geographic locations and varying generational usages, i.e.

variation (ex). She is now also explicit about how course materials are selected and
how evaluation procedures will be structured to reflect and accept the multiple forms
available to speakers and learners of Blackfoot, i.e. standard (ex). For the instruc-
tor in the classroom, an explicitly articulated variationism now outranks an explicitly
articulated Standard: variation (ex) >> standard (ex).
Our model includes three logically possible patterns that we did not observe:
Patterns E, F and H. We speculate briefly about (i) why we did not observe these
patterns in the community of practice we describe and (ii) where we might expect to
observe them. We begin with pattern H, standard (ex) >> variation (ex), which is
the exact converse of the pattern which now characterizes the instructor’s ideological
affiliation in the classroom context (i.e. pattern G). This pattern would characterize
contexts in which sociolinguistic variation is explicitly noted but participants decide
to select single, representative forms for their shared repertoire. It would also charac-
terize joint enterprises for which the selection of unique forms from among diverse
options is the stated focus, as in the creation of a dictionary or other language materi-
als. It is possible that the students who completed the Blackfoot course we describe
could exhibit this pattern after contact with the instructor, i.e. they might add varia-
tion (ex) to their ideological repertoires. However, their more frequent and lengthier
contact with and uptake of pattern D suggests they will continue to prefer patterns
with a higher-ranked standard. Several factors in the instructor’s own sociolinguistic
trajectory mitigate against her own adoption of pattern H. These include (i) her lan-
guage socialization with an older generation of speakers who exhibit pattern A, (ii) her
raised meta-linguistic awareness of the facts of Blackfoot’s diversity from observation
and study, (iii) the continued lack of language-teaching resources for Blackfoot, and
(iv) the absence of a standard version of Blackfoot.
We also did not observe patterns E or F, which are characterized by having an
implicit ideology outrank an explicit one. We hypothesize that such patterns are inher-
ently unstable. It is likely that a lower-ranked, explicit ideology would quickly move to
being higher ranked by virtue of its explicitness and consequent availability to metalin-
guistic awareness (resulting in patterns C and D). Similarly, an implicit ideology that
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
is higher ranked would need to quickly become explicit to maintain its hierarchical
advantage (resulting in patterns G and H). We predict, however, that patterns E and
F might be observed briefly during a period of rapid change in which a lower-ranked
implicit language ideology became explicit in response to some sociolinguistic pres-
sure. Such may be generally the case for languages like Blackfoot that have historically
had a higher ranked variation (im), but whose speakers are now in contact with lan-
guages with explicit ideas about standardization, like English. Patterns E and F could
also characterize ironic performances by metalinguistically-aware speakers intending
to comment on the more widely available ideological pattern currently enjoying the
implicitly dominant status quo. Our analysis predicts, then, that implicit ideologies
will generally not dominate explicit ones. When (im) and (ex) ideologies come into
contact, we predict (ex) will soon dominate (im).14
5.3 Conclusions and implications

Today we are finding…that…ideological systems can evolve and spread in
communities with astonishing rapidity. (Hill 2006: 125)
It has been repeatedly demonstrated how the linguistic preferences of more power-
ful speakers come to dominate and even erase the linguistic preferences of the less
powerful (Lippi-Green 2012; Dorian 2010b; Hill 2008). Language ideological prefer-
ences are not exempt from this process, and in fact are central to hierarchical catego-
rizations of language forms and language users (Irvine & Gal 2000). Language users
who come into contact with others who prefer different language ideological patterns
are faced with making a choice: Either to maintain their own preferred pattern or
to move towards an other’s. A metalinguistic awareness of the diverse ideological
patterns that are available in micro-interactions thus provides the opportunity for
(but does not ensure the success of) language ideological re-organizations (Kroskrity
2009: 208). Which pattern ultimately emerges has everything to do with the relative
. In some versions of OT, constraints can also be unranked with respect to each other.
We can imagine situations where two explicit ideologies (or two implicit ideologies) were
balanced. For example, patterns G and H could vary freely, with regular shifts between
variation (ex) and standard (ex). Such may be the case for the semiotic registers at work in
contemporary usage of Bahasa Indonesia, the national language of Indonesia (Goebel 2011;
Cole 2010; Smith-Hefner 2009). Such language ideological varying could only be termed
“free”, however, if speaker-hearers were “free to choose” which ideology to invoke in a given
interaction. The moment contextual elements conspired to “prefer” or “force” the choice of one
ideology over another, they would become ranked with respect to each other (at least as long
as some minimal number of typical contextual elements were in play or until an interlocutor
purposefully chose to invoke the ideology that was a-typical for that context).
A documentary ethnography of a Blackfoot language course 
status of the participants, what resources they have access to, and what choices they
make for how to use those resources to identify themselves and others as members
of language using groups. In the case of Blackfoot, it was precisely the absence of an
agreed upon standard variety in the form of classroom resources, like textbooks and
teaching aids, that provided the possibility of raising the question “Which variety
would be taught and learned?”
The Blackfoot language course we document here highlights several important
factors at work in language (ideological) shift. The first is the degree to which a partic-
ular ideological pattern is consistent with shared social norms in a particular context.
Although Annabelle and her mother sometimes disagreed about which forms to use in
particular instances, at home Annabelle shared a pattern of higher-ranked variation
with the wide range of family members from different tribes who spoke the language
and with whom she had contact. It was only in the university language classroom
where a higher-ranked standard was more readily available (given the institutional
context) that the possibility for ideological change with respect to Blackfoot arose
for Annabelle. A second factor is the degree to which linguistic forms vary from
prior convention and the availability of this variation to metalinguistic awareness.
Students in the Blackfoot class asked questions because they heard (and saw) differ-
ences between the instructor’s forms and ones they had encountered previously. If
they had not noticed such differences, it is probable that there would have been no
discussions or negotiations about which forms should be learned. Finally (and echoing
a repeated observation recently central to contemporary second language acquisition
scholarship), a key factor in determining the uptake of new language patterns is the
language user’s particular sociolinguistic history (Leung et al. 1997; Hall 2011). Here
we have shown that individuals’ trajectories of socialization (Wortham 2005) affect
their willingness to re-organize their language ideological patterns when they come
into contact with the patterns of others.15
As language ideological patterns in which standard (ex) dominates variation
(patterns D and H in the model above) are naturalized in discourses that justify the
erasure of diversity for reasons of appropriateness (Lippi-Green 2012) and practi-
cality (Benesch 1993), we need to train ourselves to become even more aware of
the variety of ideological patterns that are potentially available in collaborative
encounters. We should also keep in mind that standardization is linked to language
loss (Dorian 2010b), and taken to the extreme, standard can become a kind of
purism, which “will quickly kill a language, unless the linguistic community has
the resources to back their purist position” (Hill 2006). Participants in linguistic
. See also Kroskrity 2009b for a similar conclusion grounded in an understanding of
speaker agency.
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
documentation and language teaching projects may want to discuss these facts and
tendencies with relation to the design, selection and distribution phases of collab-
oratively produced language resources.16 We hope that others will find the model we
have proposed here to be a useful tool for keeping such facts in mind in attempts to
measure a language’s stability or health and in finding some clarity in dealing with
language ideological clash.
Our use of the LIVE model above to track the diverse language ideological pat-
terns at work in the Blackfoot language course support an argument that a higher
ranked variation produces a language ideological environment in which linguistic
diversity can flourish in ways that standard does not. If prior studies of endan-
gered language contexts where language users with ideological systems of the type
standard >> variation come into contact with users of ideological systems exhibit-
ing variation >> standard are indicative, we predict standard >> variation to
emerge as the preferred (even only) pattern within the Blackfoot community in the
near future (see Dorian 2010b: 42). What would be truly astonishing is if users of
ideological systems in which variation outranks standard could obtain and make
use of the resources they would need to ensure the continued availability of such
systems for contemporary and future users. Alternatively, inheritors of ideological
systems in which standard outranks variation could willingly adopt patterns with
a higher-ranked variation. This too would be astonishing.
6. Ethnographic coda: Practical heteroglossia
Consciousness finds itself inevitably facing the necessity of having to choose a

language. With each literary-verbal performance, consciousness must actively
orient itself amidst heteroglossia. (Bakhtin 1981: 295, cited in Hill 1995)
In documenting the development of a Blackfoot language course, we have also docu-

mented how choosing between competing language ideologies is linked to choosing
between available voices, genres, dialects, and/or languages. Creating a university-
level Blackfoot course in a context where no standard exists highlighted the avail-
ability of vocal and ideological diversity, which made the possibility of choosing
salient. Ironically, the “challenge” presented by the lack of standardized materials also
provided the opportunity to orient to heteroglossia itself. To choose consciously in
the ways we document here is to take responsibility for deciding which voices are
included and whose voices get heard (Hill 1995). It has been argued that this kind of
. See also Flores Farfán and Ramallo (2010) for recent discussions in this vein.
A documentary ethnography of a Blackfoot language course 
conscious responsibility is especially important in collaborative endeavors between

people with different interests and different access to resources, particularly in docu-
mentary linguistic projects (Hill 2006; Flores Farfán & Ramallo 2010) and classroom
contexts (Philips 2001). This is so because a lack of such consciousness easily results
in the reproduction of the power relations leading to language loss (Dorian 2010b)
and the reproduction of identity-based social inequality (Bucholtz & Hall 2008; Hill
2008; Gaudio 2009).
As co-authors, we brought different interests, training, resources, and repertoires
to this project. In the process of writing of this chapter, our own orientations towards
heteroglossia meant that we took turns inhabiting different roles and making use
of different voices, including “teacher”, “student”, “interviewer”, “interviewee”, “col-
league”, “friend”, “tribe member”, “language expert”, “reader of academic texts”, and
“writer of academic texts”. In our effort to stay conscious of the diversity of resources
and repertoires represented by each author in this process, we also found ourselves
conscious of “the construction and reproduction of power relations, social exclusion
and inequality that also operate in the academic world” (Ramallo & Flores Farfán
2010: 147). We offer this coda to clarify some of the choices we made in organizing
our diverse voices in the hopes of modeling a kind of “practical recommendation” we
see as aligning with any “theoretical penetration” (Hill 2006: 127) that may be present
in our analysis.
Since the time when I (Annabelle) moved back to the reserve to live with my
mother, I have learned a great deal about my language that I may not have known
had I not gone on to study my language from an academic perspective. I’ve learned
that the Blackfoot language has evolved since my grandmother’s time, as all lan-
guages do. I have also learned how to write the language, which is something I had
never done before. I have learned that the Blackfoot language is being lost at a rapid
rate, and that teaching it is important for keeping it alive. In teaching Blackfoot at
the University of Montana, I’ve learned that even though my dialect differs from
other Blackfoot speakers, it is a conscious decision to recognize the differences, to
help students become aware of these differences, and to encourage them to never be
afraid of them.
What matters most to me through all this is that even though the language is
changing with each generation, we can never lose sight of who the Niitsítapiiksi
are and how they lived. And it is the language that will enable us to understand
how valuable that is. The Blackfoot language has so much to offer. It is more than
just learning a unique language. It is a way of life. It gives us an appreciation of a
people and their world view. The language establishes a connection to commu-
nity and family as well as identity. It is the foundation for our ceremonies which
constitute a renewal of life. It is something to be proud of. As Dr. Bastien (2004)
states so eloquently, “for the Siksikatsitapi, knowledge is experiential, participatory,
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
and ultimately sacred, rather than objective and inert”. The Blackfoot language is
essential for our being and our way of life. And as my mother would often say,
“Iksokapii, kitasskinimatsin manistapis, ikakimat, iskatsim kisto” (It’s good you are
teaching Blackfoot, try hard, take care of yourself).
Helping Annabelle develop her course materials to address variation in Blackfoot
has been a great challenge for me (Mizuki). To write about it from the perspective
of linguistic anthropology was also challenging. These challenges were good though.
Working with both Annabelle and Debbie kept reminding me what Jane has taught
me: any variation must have its reason. The things I learned as a student of Jane are
still strong fuel for getting me going. Jane was a member of my dissertation committee
along with Ofelia Zepeda and Mike Hammond (who was my chair). While learning
about O’odham through a series of conversations with Jane, as well as by taking classes
from her, I naturally came to learn about the wonderful relationship between Jane and
Ofelia. As they write about their collaboration in Zepeda and Hill (1998), a teacher
and a student can take on reciprocal roles. I got to have a similar experience work-
ing with Annabelle: I was informally teaching Annabelle how Blackfoot orthography
works and how to do morphological analysis while she was teaching me new phrases
and pronunciation in Blackfoot. Writing this paper has heightened my sensitivity as a
linguist to linguistic variation and its contexts.
What matters to me most as a linguist trained in theoretical linguistics who works
with Native American languages is that I cannot forget that languages are spoken by
people. It is very easy for me to get excited about unique language structures and get
into a mode of puzzle-solving in theoretical frameworks. But I have had opportuni-
ties to hear the voices of speakers of Native American languages who talked about the
value their language has to them. Their voices resonate in my mind as I consider how
speakers and heritage learners may differently value the “puzzle pieces” that excite
me. When that happens, I feel the urge to return the results of my research to the lan-
guage community. Creating something that is returnable and usable in the community
often requires a completely different process and yields different products, however.
A linguist’s research process may produce databases, descriptions, journal articles, and
books. These are not always what members of the language community want. Some
speakers shared with me that they feel that their language is just being taken apart into
pieces, and it becomes useless. As a linguist, I believe that linguistic research can be
useful to speakers and learners, but I see that our studies must be reinterpreted and
refashioned in order for them to be useful in the community. How to do this, how-
ever, is not usually included in the curricula of formal linguistics training, though I
am aware of many linguists who are striving to enact this “returning process” in their
own work. Participating in curriculum development for Blackfoot language teaching
provided me with a very modest way of pursuing my urge to return my research to the
language community.
A documentary ethnography of a Blackfoot language course 
In the short time that I (Debbie) have been involved in this project, I have
barely scratched the thinnest surface of what can be learned about Blackfoot
and its speakers. The closest I have been to a Blackfoot speaker is through video
chats with Annabelle on my computer. Despite this distance and despite the fact
that it has only been about a year since I was invited to collaborate on this paper,
Annabelle has invited me to visit her in Montana during the yearly spring festival.
She assures me that we will eat well together when I arrive. Mizuki and I studied
together at the University of Arizona, and it was after reading a draft of a paper
on standardization, diversity, and national identity in classroom discourse which
I was co-authoring with yet another one of Jane’s students that Mizuki had the
idea for this chapter.17 Jane taught me about the methods and insights of linguistic
anthropology, historical linguistics, and sociolinguistics, and she supported and
guided my research on language ideological diversity in Indonesia. It was for this
alternative training that Mizuki and Annabelle thought I could be useful to them
for writing this chapter.
What has mattered most for me throughout this collaborative process is realizing
how attempting to maintain an orientation to heteroglossia that preserves diversity
takes work. Because our voices, training, vocabularies and repertoires are relatively
diverse, Annabelle, Mizuki, and I spent many hours together online talking through
possible wordings to ensure mutual understanding. I also found I needed to make use
of skills I’m accustomed to using in other contexts (like in data collection or fieldwork)
but not used to using when “sitting down to write”. For example, because there was so
much background knowledge I needed, Annabelle responded to questions I had by
recording herself talking about the details of the course or the history of Blackfoot,
sometimes in an interview format with Mizuki. I then transcribed the sections of these
recordings that were most relevant for the argument we wanted to make and wove
them together with the other text we individually wrote.
Our collaborations sometimes highlighted how exclusion and inequality
are inherent to institutionalized academic endeavors. In the final editing, “native
speaker” judgments about word choice and sentence structure were left to me, for
example. My “authority” was mitigated with constant help from Mizuki, who sug-
gested jargon-light translations for my lexical choices and less-heavily embedded
alternatives to my lengthy sentences. We hope these changes will make the inter-
pretive task of reading this chapter easier, especially for second language readers of
English and/or non-linguists. What was harder to mitigate was the pressure from
prescriptive rules of standardized English to change structures Mizuki and Anna-
belle produced for convention’s sake despite the fact that leaving their choices intact
. Cole and Meadows 2013.

 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
would result in no ambiguity or loss of meaning for any reader. A writing practice
which was better aligned with a higher ranked variation (ex) would have more
willingly ignored this pressure against diversity exerted by academic English’s higher
ranked standard (ex) and/or produced versions of this chapter in Blackfoot and
Japanese.
References
Asher, James John & Adamski, Carol. 2003. Learning Another Language through Actions: The
Complete Teacher’s Guidebook. Los Gatos CA: Sky Oaks Productions.
Bakhtin, Mikhail Mikhailovich. 1981. The Dialogic Imagination, Michael Holquist (ed.). Austin
TX: University of Texas Press.
Bastien, Betty. 2004. Blackfoot ways of knowing: The Worldview of the Siksikaitsitapi. University
of Calgary Press.
Benesch, Sarah. 1993. ESL, Ideology, and the politics of pragmatism. TESOL Quarterly 27(4):
705–717.
Blommaert, Jan. 2010. The Sociolinguistics of Globalization. Cambridge: CUP.
Bucholtz, Mary & Kira Hall. 2004. Language and identity. In A companion to linguistic anthro-
pology, Alessandro Duranti (ed.), 369–394. Oxford: Blackwell.
Bucholtz, Mary & Hall, Kira. 2008. Finding identity: Theory and data. Multilingua 27: 151–163.
Cole, Debbie. 2010. Enregistering diversity: Adequation in Indonesian poetry performance.
Journal of Linguistic Anthropology 20(1): 1–21.
Cole, Debbie & Meadows, Brian. 2013. Avoiding the essentialist trap in intercultural education:
Using critical discourse analysis to read nationalist ideologies in the language classroom. In
Linguistics for Intercultural Education [Language learning and Language Teaching 33], Fred
Dervin & Anthony Liddicoat (eds), 29–47. Amsterdam: John Benjamins.
Dorian, Nancy. 2010a. Investigating Variation: The Effects of Social Organization and Social
Setting. Oxford: OUP.
Dorian, Nancy C. 2010b. The private and the public in documentation and revitalization. In
Flores Farfán & Ramallo (eds), 29–47.
Flores Farfán, José Antonio & Ramallo, Fernando F. (eds). 2010. New Perspectives on Endangered
Languages: Bridging Gaps between Sociolinguistics, Documentation and Language Revital-
ization [Culture and Language Use 1]. Amsterdam: John Benjamins.
Frantz, Donald G. 1991. Blackfoot Grammar. Toronto: University of Toronto Press.
Frantz, Donald G. & Russell, Norma J. 1995. Blackfoot Dictionary of Stems, Roots, and Affixes.
2nd edition. Toronto: University of Toronto Press.
Goebel, Zane. 2011. Language, Migration, and Identity: Neighborhood talk in Indonesia.
Cambridge: CUP.
Goddard, Ives. 2001. The Algonquian languages of the Plains. In The Handbook of North
American Indians, Vol. 13: Plains, William Sturtevant (ed.). Washington DC: Smithsonian
Institution.
Hall, Graham. 2011. English Language Teaching. New York NY: Routledge.
Gaudio, Rudolf Pell. 2009. Allah Made Us: Sexual Outlaws in an Islamic African City. Chichester:
Wiley-Blackwell.
A documentary ethnography of a Blackfoot language course 
Hammond, Michael. 1995. Metrical phonology. Annual Review of Anthropology 24: 313–342.
Hammond, Michael. 1997. Optimality theory and prosody. In Optimality Theory, Diana
Archangeli & D. Terence Langendoen (eds), 33–58. Oxford: OUP.
Hill, Jane H. 1995. The voices of Don Gabriel. In The Dialogic Emergence of Culture, Dennis
Tedlock & Bruce Mannheim (eds), 96–147. Urbana IL: University of Illinois Press.
Hill, Jane H. 2006. The ethnography of language and language documentation. In Essentials of
Language Documentation, Jost Gippert, Nikolaus P. Himmelmann & Ulrike Mosel (eds),
113–128. Berlin: Walter de Gruyter.
Hill, Jane H. 2008. The Everyday Language of White Racism. Chichester: Wiley-Blackwell.
Hinton, Leanne. 2001. New writing systems. In The green book of language revitalization in
practice, Leanne Hinton & Ken Hale (eds), 239–250. San Diego, CA: Academic Press.
Dempsey, Hugh A. 1978. Indian Tribes of Alberta. Calgary: Glenbow/Alberta Institute.
Dauenhauer, Nora & Richard Dauenhauer. 1998. Technical, emotional, and ideological issues
in reversing language shift: Examples from Southeast Alaska. In Endangered languages:
Current issues and future prospects, Lenore Grenoble & Lindsay Whaley (eds), 57–99.
Cambridge, UK: Cambridge University Press.
Irvine, Judith T. & Gal, Susan. 2000. Language ideology and linguistic differentiation. In Regimes
of Language: Ideologies, Polities, and Identities, Paul V. Kroskrity (ed.), 35–84. Santa Fe NM:
School of American Research.
Kroskrity, Paul V. 2009a. Language renewal as sites of language ideological struggle: The need
for “ideological clarification”. In Indigenous Language Revitalization: Encouragement,
Guidance & Lessons Learned, Jon Reyhner & Louise Lockard (eds), 71–83. Flagstaff AZ:
Kroskrity, Paul V. 2009b. Embodying the reversal of language shift: Agency, incorpora-
tion, and language ideological change in the Western Mono Community of Central
California. In Native American Language Ideologies: Beliefs, Practices, and Struggles in
Indian Country, Paul V. Kroskrity & Margaret C. Field (eds). Tucson AZ: University of
Arizona Press.
Kroskrity, Paul V. & Margaret C. Field (eds). 2009. Native American L anguage Ideologies: Beliefs,
Practices, and Struggles in Indian Country. Tucson AZ: U niversity of Arizona Press.
Leung, Constant, Harris, Roxy & Rampton, Ben. 1997. The idealized native speaker, reified eth-
nicities, and classroom realities. TESOL Quarterly 31(3): 543–560. (Special issue: Language
and Identity).
Lippi-Green, Rosina. 2012. English with an Accent. Language, Ideology, and Discrimination in the
United States. New York NY: Routledge.
Mithun, Marianne. 2001. The Languages of Native North America. Cambridge: CUP.
Philips, Susan U. 2001. Participant structures and communicative competence: Warm springs
children in community and classroom. In Linguistic Anthropology: A Reader, Alessandro
Duranti (ed.). Oxford: Blackwell.
Ramallo, Fernando F. & Flores Farfán, José Antonio. 2010. Linking three agendas: Opening a
debate and directions for the future. In New Perspectives on Endangered languages: Bridging
Gaps between Sociolinguistics, Documentation and Language Revitalization, José Antonio
Flores Farfán & Fernando F. Ramallo (eds), 147–149. Amsterdam: John Benjamins.
Rampton, Ben. 1990. Displacing the “native speaker”: Expertise, affiliation and inheritance.
ELTJournal 44: 97–101.
 Annabelle Chatsis, Mizuki Miyashita & Deborah Cole
Reynolds, Jennifer. 2009. Shaming the shift generation: Intersecting ideologies of family and
linguistic revitalization in Guatemala. In Native American Language Ideologies: Beliefs,
Practices, and Struggles in Indian Country, Paul V. Kroskrity & Margaret C. Field (eds),
214–237. Tucson AZ: University of Arizona Press.
Silverstein, Michael. 1998. Monoglot “standard” in America: Standardization and metaphors of
linguistic hegemony. In The Matrix of Language: Contemporary Linguistic Anthropology,
Donald Brenneis & Ronald K.S. Macaulay (eds), 284–306. Boulder CO: Westview Press.
Silverstein, Michael. 2003. Indexical order and the dialectics of sociolinguistic life. Language
and Communication 23: 193–229.
Silverstein, Michael. 1981. The limits of awareness [Sociolinguistic Working Paper 84]. Austin
TX: Southwest Educational Development Laboratory.
Smith-Hefner, Nancy. 2009. Language shift, gender, and ideologies of modernity in Central Java,
Indonesia. Journal of Linguistic Anthropology 19(1): 57–77.
Taylor, Allan R. 1969. A Grammar of Blackfoot. Ph.D. dissertation. University of California,
Berkeley.
Uhlenbeck, Christian C. 1938. A Concise Blackfoot Grammar [Verhandelingen der K oninklijke
Akademie van Wetenschappen. Afdeeling Letterkunde, Nieuwe Reeks, Deel 41].
Amsterdam: KNAW.
Weatherwax, Marvin. 2007. Blackfeet Language Lessons. Browning MT: Blackfeet Community
College.
Wortham, Stanton E.F. 2005. Socialization beyond the speech event. Journal of Linguistic
Anthropology 15(1): 95–112.
Zepeda, Ofelia & Hill, Jane H. 1998. Collaborative sociolinguistic research among the Tohono
Syncretic speech, linguistic ideology,
and intertextuality
(Re)Presenting the Spanish translation
of ‘Speaking Mexicano’ in Tlaxcala, Mexico
Jacqueline Messing & Ramos Rosales Flores

Georgetown University and University of Maryland-College Park & Escuela
Xochitekali and Municipio de San Bernardino Contla de Juan Cuamatzi, Tlaxcala
“Syncretism” describes the structural incorporation of indigenous languages

like Mexicano (Nahuatl) from Central Mexico with majority languages like
Spanish. Building on research of Hill and Hill (1986), and collaborations with
local scholars including teacher Ramos Rosales Flores, We analyze a 1999 public
linguistic event celebrating the Spanish publication of “Speaking Mexicano”
in Tlaxcala. Syncretic Mexicano, so-called “mixed speech,” exists within a
local ideological landscape in which legítimo Mexicano – true Mexicano – is
an idealized, largely not-spoken form of the native language, free of Spanish.
We analyze multiple ideologies and metadiscursive practices at this event.
I (Messing) further explore interpretation of syncretism by locals, resident-
scholars and outsider-scholars, adding intertextual complexity to the academic
and local interpretations of purism.
Keywords: Mexicano; syncretism; purism; language ideology
1. Introduction: Writing about syncretic speech in Central Mexico
The concept of “syncretism” in linguistic anthropology was developed to describe a

phenomenon observed in indigenous Latin American communities, that is, the mix-
ing of an indigenous language such as Mexicano (Nahuatl) from Central Mexico with
loan words and grammatical constructions from colonial languages like Spanish that
became incorporated into the structure of the language. Hill and Hill (1986, 1999)
made the case that speakers in this Mexican region have survived years of cultural and
political infiltrations by integrating elements of Spanish into their Mexicano syncretic
speech; they described syncretic speech as an alternative to views of “mixed languages,”
 Jacqueline Messing & Ramos Rosales Flores
thus honoring the linguistic purism they found in Mexicano indigenous communi-
ties.1 Linguistic syncretism has proven to be a productive concept cross-culturally
(Makihari 2004; Fader 2007).
This chapter analyzes a public linguistic event that celebrated the Spanish
publication of “Speaking Mexicano” in Tlaxcala, in 1999, and builds on Jane Hill’s
work, my research, and collaborations with local scholars, including teacher Ramos
Rosales Flores who contributed to the chapter.
I (J. Messing) organized the presentation of the Spanish translation of Hill and
Hill’s (1999) book, Hablando Mexicano: La dinámica de una lengua sincrética en el
centro de México. In June 1999, a panel consisting of Tlaxcalan intellectuals, teachers,
and scholars, and the authors and translator came together in the county seat of San
Bernardino Contla, Tlaxcala in the Casa de Piedra/Casa de Cultura. The presentation
drew a sizeable and diverse audience, who engaged with the speakers and the linguistic
content of the event both during and after the day itself. The interest expressed by
many who were present offered a rare glimpse into an event in which the value of
the Mexicano language was publicly legitimated, and the importance of its use was
proclaimed by local officials and residents.
My role in the book presentation was that of facilitator. I did not comment on
the book at the time, when I was a graduate student of Jane Hill’s doing dissertation
research, but rather made the organization of the event an element of my ethnographic
fieldwork, undertaking an ethnographic inquiry into this type of speech event.
Collaborators Ramos Rosales Flores, a bilingual teacher and teacher trainer, and
Refugio Nava Nava, a professor at the Autonomous University of Tlaxcala were key
in the planning. Both presented commentaries on the book Hablando Mexicano, and
Rosales Flores hosted a post-event dinner at his home in San Felipe Cuahutenco.
The book presentation itself was held in the Casa de Piedra – ‘Stone house,’ part of
a network of a Mexican “Casas de Cultura,” or houses of culture, on the central town
square of San Bernardino Contla, Tlaxcala. Contla is the head town of the county in
which I have conducted much of my Mexico fieldwork, along with the San Isidro/
Canoa towns on another flank of the Malinche mountain.
In the book presentation event multiple ideologies of syncretism and purism
were expressed by locals, resident-scholars and outsider-scholars, on purism
in indigenous language ideology. Indeed language ideologies in this region are
. Preliminary analysis of the data in this paper was presented at the 2009 Conference in
honor of Jane Hill’s retirement, University of Arizona, and in a 2011 session on language mixing
at the American Anthropological Association. Messing and Rosales Flores are c ollaborating
with Ethel Xochitiotsin Perez on a study of oral traditions in the Malinche region.
Syncretic speech, linguistic ideology, and intertextuality 
ultiple, as I’ve observed them Messing 2007). These ideologies, both tacit and
m
explicit, surfaced within the discourses I analyze below. The metadiscursive
practices in this event, convened to discuss language mixing, purism, and the
publication of a book, clearly show the “capacity of discourse to both represent
and regulate other discourses” (Bauman & Briggs 2000: 142). My representation of
the event is in this chapter, informed by conversations with collaborator, teacher
Ramos Rosales Flores.
I’ve been interested in how to describe varieties of “mixed languages” and
linguistic phenomena. My view on Tlaxcalan bilingualism is that it doesn’t conform to
European notions of diglossia and bilingualism, as we observe them in Latin A merican
indigenous, sociohistorical contexts of speaker inequality and racism, given the legacy
of colonialism; today menosprecio, racist discourses abound (Messing 2007), in which
Native languages such as Mexicano are iconized, coming to stand for indigenous
identities that are too often denigrated. For work on language contexts such as the
Mexicano-speaking region of Central Mexico, the concepts of semi-speaker (Dorian
1977) and quasi-speaker (Flores Farfán 1999) have helped advance our descriptions
of the language skills of speakers in a helpful direction. Hill’s work on language and
racism in broader contexts (cf. Hill 2004, 2010) has invited further analysis of notions
of identity stigmas and “authenticity.”
The Mexicano communities that skirt the Malinche volcano of Central Mexico
speak a native language that has incorporated Spanish loan words and grammatical
constructions into its structure. Most often this syntactic convergence involves
prepositions and conjunctions, Spanish language numbers and various other lexical
items were borrowed into Mexicano and have been adapted to Mexicano grammar.
For example, de, the Spanish ‘from’ becomes den in local syncretic Mexicano speech)
(Hill & Hill 1986). An example of syncretic speech from a Mexicano narrative
I recorded in Tlaxcala is:
(1) AMO KA mas este posibilidad para para TI-vivir-OZ-KE

‘there is not much more um possibility for us to make a living’
Above, Mexicano is marked by CAPS, and Spanish is marked by italics. You’ll note
that in “ti-vivir-oz-ke” the Spanish verb ‘to be’ (vivir) is embedded in a Nahuatl agglu-
tinative syntactic construction. The utterance refers to economic challenges that are
often described in conjunction with a meta-discourse of salir adelante (forging ahead,
bettering one’s socioeconomic situation) that I’ve described elsewhere (Messing 2003,
2007a, to appear a). There are instances of syncretic Mexicano dating back to Tlaxcala
in the sixteenth century Messing, to appear (b).
Despite the commonality of speakers’ linguistic mixing and maintenance of
Mexicano across many centuries, the most salient language ideology in 20th and
 Jacqueline Messing & Ramos Rosales Flores
21st century Tlaxcala is the ideology of “legítimo Mexicano,” legitimate or true

Mexicano. This language ideology surfaces in a discourse of nostalgia about the past
(Hill & Hill 1986) that:
suggests that the Mexicano language, especially in some “pure” form, is a p
eculiarly
appropriate vehicle for the social forms of long ago, in achto, and especially for
“respect.” On the other hand, Spanish, and the mixing of Spanish and Mexicano,
are peculiarly associated with the social forms of today, a:xa:n, and with the loss
of respect. (Hill 1998: 69)
Mexicano is then a language through which locals can express great respect for people,
which is linguistically marked through use of the levels of honorific speech, and by the
absence of speaking Spanish, which in high Mexicano-maintenance communities is
the language of business for profit, drunkenness, and profanity (Hill 1995). In regions
with lesser degrees of language maintenance today, speakers do express legítimo
mexicano ideology, whereby the Native language is itself considered to be “pure” and
untainted with the colonial language.
2. The chapter
This chapter analyzes the metapragmatic commentary that surfaced in the book
presentation, an event in which authors Jane Hill and Ken Hill were themselves
participants. I highlight the voices of those who expressed their viewpoints on Speak-
ing Mexicano at our 1999 event, by including substantial narrative quotes from several
participants, and a section on audience reception. Recent, reflective commentaries by
collaborator Ramos Rosales Flores are incorporated into the chapter, within a context
of intertextuality and metadiscursivity.
3. Presentation de libro
Complex ideologies of purism surfaced in talk of “legítimo Mexicano,” true Mexicano

and syncretic Mexicano. In 1999, while I was doing dissertation fieldwork in Tlaxcala,
I was asked to organize a local academic celebration of the publication of the translation,
from English to Spanish, of Hill and Hill’s 1986 book “Speaking Mexicano.” The book is
considered the most extensive ethnolinguistic description of any Mexican indigenous
speech community to date, and to have the book appear in Spanish, published by a
Mexican press was a matter of great interest to many scholars. The book describes the
sociolinguistic situation in the Malinche region, and the most salient language ideol-
ogy – the ideology of “legítimo Mexicano” (legitimate Mexicano), in which speakers’
purist attitudes encourage speech which is completely Mexicano, without any trace
Syncretic speech, linguistic ideology, and intertextuality 
of its syncretic elements whose source is the Spanish language, with the associated
discourse of nostalgia about the past.
This “presentación de libro” – book presentation – was a very formal academic
celebration of the type held in Mexico to comment on, and toast, a new book and its
authors, and one of the translators. Eight commentators and an audience of seventy,
comprised of local people, including teachers, their students with families, local schol-
ars, oral historians and Mexicano language promoters, and the extended family I lived
with, academics from the cities of Tlaxcala, Puebla and Mexico City, along with the
authors and one of the translators, José Antonio Flores Farfán, offered commentary on
the new publication.
The event created a site for metapragmatic commentary – an expression of
language ideologies – that was both explicit and implicit. Speakers raised many issues
that are not usually discussed in a markedly “public” forum in this region (see Messing
2007b). Mexicano is usually reserved for contexts of great mutual trust, such as within
family and similar circles of trust, or between peers. Some town meetings are held in
Mexicano in parts of the Contla region that have maintained this custom, and this is the
notable exception to this linguistic custom. Ramos Rosales Flores, in preparing his con-
tribution to this chapter, indicated having experienced instances of negative reactions
(menosprecio- denigration), or outright racist comments, after public speeches he has
given in Mexicano, outside the aforementioned circles of trust. Language and racism is
indeed a key topic for students of ideology in this region (Rosales Flores, personal com-
munication; Messing 2003, 2007). In the San Isidro Buensuceso, Tlaxcala/San Miguel
Canoa, Puebla region where the Hills concentrated many of their observations, the
language use habits along the public/private continuum are different; language mainte-
nance is stronger and shift has a different cultural character than in the Contla county
region, just a two hour bus ride (or one-hour car ride) away.
The goal of this speech event was a formal commentary on the new translation,
resulting in a conversation about syncretism and purism, creating a site for the explicit
expression of linguistic ideologies. Ten speeches were given. As per local convention,
formal invitations were printed and delivered in person to many interested people,
accompanied by a verbal, personal invitation to those within the local area. Professors
Nava Nava and Rosales Flores accompanied me to some of these visits to distribute
the invitations. Invitations were also extended to scholars in the region by email. We
secured small grants to support the event, from the Instituto Tlaxcalteca de la Cultura
and the municipal county of San Bernardino Contla, by order of the Municipal Presi-
dent – the Mayor who attended the event as a special guest. The event was recorded by
local videographers, and me (Messing).
The event, an unusual one of its kind outside of a university, served to open a
discursive space in which metapragmatic commentary, language shift, and, in the
commentary period, the politics of translation of foreigners’ scholarly work into
 Jacqueline Messing & Ramos Rosales Flores
S panish were discussed among specialists and nonspecialists. Spanish was the p
rimary
language spoken, with both substantive and symbolic uses of Mexicano. The seventy-
person audience included Mexicano-Spanish bilinguals, semi-speakers, Spanish
monolinguals, and three of us norteamericanos- the Hills and me (Messing).
4. Purist ideologies and syncretic speech
My work has focused on the interplay of multiple ideologies of language, identity and
economics that surfaced in discourses I observed and taped in this region. P urism is
considered a dominant ideology – if not the dominant ideology in academic discourses
on indigenous languages in Mexico. The work in Mexican sociolinguistics has been
much influenced by Speaking Mexicano and its description of purism, alongside uses of
syncretic speech. The ‘legítimo Mexicano’ ideology that Hill and Hill (1986) described
for the period of the 1970’s and 80’s has continued in ensuing years.
One of the first presentation commentators from Contla, a teacher, gave his
view that “Mexicano isn’t totally authentic anymore…. all the words aren’t used to
100 percent, but rather that they are mixed …. it’s a little harder to be able to say that
it’s an authentic Nahuatl anymore.”
Questions of authenticity abound and language is often at the center of identity
debates in Tlaxcala and other parts of indigenous Mexico. In the data that follows,
several perspectives on the sociolinguistic situation in Tlaxcala are presented, ending
with the speech given by co-author Ramos Rosales Flores, and Jane Hill’s speech.
5. Intertextuality and the participants
The presentation participants were:

Municipal President Pablo Flores Galicia
Prof. Diego Xochitemol
Dr. Refugio Nava Nava
Prof. Leonor Cuamatzi
Dr. Elsie Rockwell Richmond
Prof. Ramos Rosales Flores
Dr. José Antonio Flores Farfán
Prof. Alberto Zepeda Serrano
Drs. Jane Hill and Kenneth Hill
Audience commentary: Dr. Luis Reyes Garcia
Emcee: Jacqueline Messing
Syncretic speech, linguistic ideology, and intertextuality 
The book presentation was unique because it consisted of an academic event in

a small town setting, with participants from several schools of thought, across social
classes, and institutions in rural, semi-urban, and urban spaces, indigenous and
nonindigenous, with Mexicans and foreigners, in sum, across several boundaries of
difference and power. I use the terms “resident scholars” and “outsider scholars” to
describe a distinction that, alongside nationality and ethnicity/race, is salient to the
people I’ve worked with in both rural and urban Mexico.
Some metapragmatic comments were direct and some more implicit. Language
ideologies, as Susan Philips describes, “can be both multisited and site-specific,
and therefore partial rather than whole in their diverse manifestations” (Philips
1998: 255–56).
The Municipal President, at the head of the large county of Contla, used M
exicano
to present his speech as part of his official opening of the event. Because this was
not a typical, or even likely speech event in Contla, his use of Mexicano served to
indexically praise and support both the event and the book publication, at the same
time that it subverted expected linguistic norms and symbolically expressed support
for the public use of the language. Indeed, this same politician was known to hold
official audiences in his Mayoral chambers in the native language, something which
had not been heard to happen in many decades. His presence lent an official stamp
of approval to the event, and his language of choice brought the indigenous language
into the geographic sphere of the “center” of town and the center of the county. There
is a history to this symbolic action, as the “core” and “peripheries” of Contla have long
had distinct social, economic and ethnic identities mapped onto them (see Messing
2003). Amid these symbolic subtexts, there were many interests, experiences, and
agendas among the speakers, and the audience, leading to a discussion with multiple
intertextual levels. The multiplicity of ideologies of language, identity and economics
surfaced.
Some discussants had read the book for the first time and commented on the idea
of purism and syncretism which so many of Jane Hill’s writings have described for this
region.
(2) Diego Xochitemol, High school teacher, local bureaucrat, collector

of Mexicano narratives in his spare time.
The speaker whose quote introduces my presentation of data is the teacher I quoted
above on the inauthenticity of today’s Mexicano, a comment that illustrated the central
ideology in Speaking Mexicano. Prof. Xochitemol explained that the importance of the
book, in Spanish, was to open a space for speakers to reflect about their own language,
and its survival.
 Jacqueline Messing & Ramos Rosales Flores
Quiero decirles que este libro, creo que está inspirado en una intención de querer
que los hablantes de el náhuatl, el mexicano, tengan un espacio de reflexión respecto
a su propia lengua, es decir, lo que normalmente nosotros consideramos como un
aspecto que paulatinamente se va perdiendo, el libro nos hace una reflexión acerca
de que que la vida económica, la vida social, la vida política del pueblo, no solo
Tlaxcalteca, sino del pueblo de San Miguel Canoa, que es del estado de Puebla
y otros pueblos circunvecinos que fueron objeto de studio. Pues obviamente se
presentan a través de una situación, la situación ritual, característica del pueblo
náhuatl, del pueblo indígena, se ve mezclada ahora con la cuestión también de la
influencia mútua entre dos lenguas que son el español y el mexicano.
Esta situación nos permite abordar una cuestión que para mí es importante.
Hablamos primeramente de una degradación de una lengua, es decir de que el
náhuatl, el mexicano ya no es totalmente auténtico. Me refiero en el sentido de que
no se utilizan todas las palabras al 100 por ciento, sino que se van mezclando en
la medida que la gente es mas joven o en la medida en que la gente habita no tan
arriba de la montaña, es decir, entre mas cerca de la ciudad está la gente, poco a
poco se va viendo una situación un poquito más difícil de poder decir que ya es un
auténtico náhuatl. Entonces, tampoco podemos decir que la cuestión pura está en
Cuahutenco [pueblo], está en San Isidro Buen Suceso, o está en San Miguel Canoa,
de hecho los hablantes al 100 por ciento ya no existen. Lo que me llama la atención
del estudio, desde el punto de vista práctico es que es un estudio ecológico.
I want to tell you all that this book, I think that it’s inspired by an intention of
wanting for the speakers of Nahuatl, Mexicano, to have a space to reflect about their
own language, that is, what we normally consider to be an aspect that gradually is
becoming lost, the book reflects about economic life, social life, political life of the
people, not only the Tlaxcalans but of the people of San Miguel Canoa, which is of
the state of Puebla and other neighboring towns that were the object of study. Well
obviously they present about a situation, the rural situation, characteristic of the
Nahuatl people, of the indigenous people, this is mixed now with the question of the
mutual influence between two languages that are Spanish and Mexicano.
This situation permits us to approach a question that for me is important.
We speak first of a degradation of a language, that Nahuatl, Mexicano isn’t totally
authentic anymore. I refer to [the fact] that all the words aren’t used to 100 percent,
but rather that they are mixed to the extent that the people are younger, and to the
extent that people live not that far up the mountain, that is to say, the closer people
are to the city, little by little a situation is seen in which it’s a little harder to be able to
say that it’s an authentic Nahuatl anymore. So, neither can we say that the “cuestión
pura” [pure question] is in Cuahutenco [town], is in San Isidro Buen Suceso, or is
in San Miguel Canoa, in fact speakers who are 100 percent don’t exist anymore.
What calls attention to me in this study, from the practical perspective, is that it’s
an ecological study.2
. A note on transcription and translation in this chapter: Messing taped the event, subse-
quent interviews, and did the fine-grained transcriptions of the Spanish data, after a research
Syncretic speech, linguistic ideology, and intertextuality 
This metapragmatic commentary is an example of the ideological multiplicity that

I encountered in this region. The speakers draw Barthian boundaries between the states
of Tlaxcala and Puebla in this narrative excerpt.3 I want to point out that the p urist
ideology that belies local comments on lack of authenticity of native language doesn’t
exclude their use of syncretic Mexicano, or a positive review of the book that described
the “syncretic project.” As Hill and Hill (1986) so importantly point out, there is much
ambivalence in the region, and speakers are themselves quite aware of this.
In this region, syncretic Mexicano, as “mixed speech,” exists within a local
ideological landscape in which legítimo Mexicano – true Mexicano- is idealized as a
form of the native language free from any trace of Spanish, although it is no longer
spoken as such. My view is that the role of respect in social relations in the Malinche
region is so strong that speakers of Mexicano show great respect to their ancestors via an
iconicized Mexicano, by not wanting this language to be “tainted” with the i nfiltrating
(colonial) language. Paradoxically, the infiltrating language is one that locals speak,
for the most part, on a daily basis to the high degree of language shift in much of this
region. Ideological struggle is inherent in both local structure and practice. Mexicano
speakers in the Malintzi region may call for respect of legítimo Mexicano, the ‘true’
variety, but then choose whether or not to speak their modern-day syncretic version of
Mexicano to their children. Competing ideas about authenticity of code and identity
are at the heart of these language socialization and maintenance issues.
Among the multiple ideologies, in the book presentation event, there was a dis-
juncture between the interpretation of syncretism by locals, resident-scholars and
outsider-scholars, adding an intertextual layer of complexity to the academic inter-
pretation of purism in indigenous language ideology. Consider the following three
examples. In the next example, the late Alberto Zepeda Serrano, teacher and school
principal, field assistant and key informant to Jane and Ken in their Malinche research
begins his commentary in Mexicano:
(3) tlaso’kamati’ nikan o no nechon tlayekolti’ke itech i nin iluitl

Thank you for having invited me to this anniversary party
assistant had done a basic transcription. The Mexicano transcription and translation was a
combined effort between Rosales Flores and Messing. The translations to English are Messing’s.
Writing systems are ideologically-laden, and the Mexicano transcribed in this chapter follows
one of the local conventions in which Mexicano writing is partially calqued on Spanish, rather
than Nahuatl agglutinating syntax.
. The cultural area known as the Malinche encompasses communities in the states of
Tlaxcala and Puebla. Mexican regionalism and state identity that is so salient elsewhere is
less developed in the San Isidro Buensuceso and San Miguel Canoa region where a barranca
(ravine) separates the two towns and states, otherwise the quite similar in traditions,
economics, language and strong language maintenance.
 Jacqueline Messing & Ramos Rosales Flores
kampa ki tsakua’ome xiuitlnin kaltetl-kaltetl kampa tlamachtiloyan

where it is being celebrated, two years of the house of culture (stone house/
casa de piedra) – school where it is taught
o tia’ke ken mi i’totia ken tltaoma.’ Nochi kual miek tlaso’kamati’
dance, embroidery. All of this I am very thankful for
amo’o nik matia’ kox i nin ma amoxtli’ yo mo chi’chi’i’kin español
I didn’t know that this book had already been done in Spanish
paki no yolo’kuak i’kin nik mati’ tlen o mo chi den sempoali uan ma’tlaktl.
I was very happy when I find out about this type of things, that were done
30 years ago.
(Speaker continues in Spanish)

Me da mucho gusto que me hayan invitado a participar en esta fiesta donde cumple
dos años la Casa de Piedra, casa donde se enseña a bailar, a cocer, y a todo tipo
de conocimientos. Jamás me imaginé que lo que se hizo hace veinte-cinco años,
aproximadamente, junto con mis amigos, hermanos y padrinos, los Hill, sea tan
conocido, sea tan aceptado. Más que trabajo duro donde anduvimos en el calor, en
el agua, también fue algo muy bonito por que la gente con la que platicamos nos
abrió las puertas de sus casas. Con ellos comimos, bebimos, cantamos, jugamos,
por que en realidad para mi fue como un juego. El hacer todo esto fue algo
divertido por que efectivamente lo que decía el Señor eh José Antonio, hablamos
en mexicano con ellos, desde que llegábamos mi misión era hablar el mexicano
con ellos, hacerlos sentir en confianza y posteriormente todo salía tan libremente,
tan fácil, que inclusive la gente nos pedía que regresáramos para volver a platicar
con ellos. Entonces todo este trabajo fue algo divertido, gracias a los Hill, porque
hemos convivido con todos ellos. Gracias a todos ustedes por la aceptación y
efectivamente y ojalá pues sigamos todo esto que nos dicen ellos y acertadamente
han dicho, no hay mexicano puro, todos tratamos de incluir, hasta yo, aunque a
veces, pues lo hago, mentalmente trato de no meter ninguna palabra, pero eso
es imposible. Sencillamente a veces se piensa en las dos lenguas y sale la palabra
que está más cercana, la palabra que más se facilita. Entonces muchísimas gracias
y felicidades a los traductores y a toda la gente que ha hecho posible todo esto y
especialmente a mis padrinos los Hill. Muchísimas gracias.
I’m very glad that you invited me to participate in this party in which the Casa
de Piedra (Culture house name), a house in which dancing, sewing and all sorts of
knowledge is taught. I never imagined that what was done twenty-five years ago,
approximately, along with my friends, brother and sister and godparents, the Hills,
would be so well known, so well accepted. More than hard work, where we walked
around in the heat, in the rain, it was also something very lovely because the people
with whom we spoke opened for us the doors to their houses. With them we ate,
drank, sang, played, because in reality it was fun. Doing all of this was something
fun because, as what Mr. José Antonio says, we spoke in Mexicano with them, from
when we would arrive my mission ws to speak Mexicano with them, to make them
feel in “confianza” [mutual trust], and afterwards everything would come out so
Syncretic speech, linguistic ideology, and intertextuality 
freely, so easily, that people would also ask us to return to continue chatting with
them. So all of this work was something fun, thanks to the Hills, because we have
lived and shared with all of them. Thank you to you all for the acceptance and truly
and hopefully we will continue all of what they say and certainly have said, there
is no pure Mexicano, we all try to include, even me, although sometimes, well I do
it, mentally I try not to put any word [of Spanish], but that is impossible. Simply
sometimes one thinks in the two languages and the word comes out that is the closest,
the word that most facilitates. So thank you very much and congratulations to the
translators and to all the people that have made all of this possible and especially my
godparents the Hills. Thank you very much.
The late Alberto Zepeda, high school principal from Puebla, and research assistant to the Hills
in the 1970’s. Also in photo: Pablo Flores Galicia, Refugio Nava Nava, and Jacqueline Messing
The late Alberto Zepeda Serrano was a respected school principal in the state of
Puebla. His work as the Hills’ research assistant influenced his professional t rajectory,
as he explained to me after the event. In addition to the gratitude and respect addressed
to the Hills, Alberto points out his own attempt to keep Spanish out of his Mexicano
discourse. His Mexicano discourse is noticeably devoid of any Spanish language
infiltrations, with the exception of the word for Spanish – español.
The next speaker excerpted here is Refugio Nava Nava, an anthropological
linguist and Nahuatlato from the Contla county in Tlaxcala who has written on
Mexicano language socialization and levels of respect in the Mexicano honorific
system. His comment contains a tacit critique of the notion of “pure” Mexicano,
which he surrounds by the proverbial, nonverbal anthropological quotation marks
(entre comillas, ?no?, in question marks, no?) He states that speakers and teachers
are closing themselves off to a reality, and we can surmise that the reality is one of a
disconnect between actual usage and that which is found in the available language
textbooks he mentions.
 Jacqueline Messing & Ramos Rosales Flores
(4) Dr. Refugio Nava Nava, Professor of Nahuatl, Anthropology,

and Linguistics
Me parece que [el libro] nos ofrece una herramienta increíble para quienes nos
dedicamos a la enseñanza de lenguas…, los que diseñamos material para enseñar
lenguas, los que enseñamos libros, ejercicios, lecturas, para enseñar este idioma
en esta región, debemos tomar en cuenta este factor. Tenemos que sentarnos
a reflexionar en torno a qué tanto en nuestros textos, como diseñadores de
materiales, estamos aceptando esta realidad. O por el contrario, en este afán de
pretender un náhuatl “puro” -entre comillas, ¿no? Nos estamos cerrando a una
realidad.
It seems to me that [the book] offers us an incredible tool for those of us
who are dedicated to the teaching of languages…, those who design curriculum to
teach languages, those who teach books, examples, radings, to teach this language in
this region, we need to take into account this factor. We have to sit and rflect on how
much our texts, as developers of (curricular) materials, we are accepting this reality.
Or on the contrary, this eagerness to pretend a “pure” Nahuatl – in quotation marks,
no? We are closing ourselves off to a reality.
When the speaker says “pretender un náhuatl “puro” – entre comillas, ¿no?/ to pretend
a “pure” Nahuatl – in quotation marks, no?” he broke from his narrative commentary
to insert a Bakhtinian sidelong glance that, to the academic members of the audience,
referenced prior discussions of linguistic purism among scholars. These prior conver-
sations -part of a metadiscourse on purism and indigenous languages – took place in
writing via publications, and between people at conferences and in classrooms. This
moment is an instance of what Tannen terms ‘intertextuality in interaction’ (Tannen
2006: 598), when public and private discussions collide through the repetition of
“words and topics” that are “recycled, reframed, and rekeyed” in everyday speech.
This discursively highlighted moment served both to respect the local purism of the
scholar’s community, and simultaneously to critique the notion of “purism” from an
academic linguistic perspective. This descriptive-linguistic perspective is highlighted
in the next excerpt.
(5) Dr. José Antonio Flores Farfán, University scholar, translator
In the following commentary by ethnolinguist and Speaking Mexicano t ranslator José

Antonio Flores Farfán he picks up on the book’s theme of how speakers themselves
deal with the dilemma of language shift and maintenance. The highlight of his
commentary is the statement that: “Pure languages do not exist and Mexicano is not
the exception”; this is the position of descriptive linguistics par excellence. A larger
excerpt follows:
El objeto de “Hablando Mexicano” es precisamente entender como una lengua
se encuentra ante ante el dilema, o los hablantes de una lengua, en este caso los
Syncretic speech, linguistic ideology, and intertextuality 
hablantes de la Malinche se encuentran ante ante el dilema de la pérdida o del

mantenimiento de su lengua de su lengua materna. Esta obra, que por cierto es
la obra mas completa de sociolingüística con la cual contamos hasta la fecha,
celebra esas voces y esa posibilidad de continuidad del mexicano en un contacto
muy añejo con el español. Esto los lleva como han dicho ya los compañeros, a
hacer una reflexión que me parece muy importante. No existen lenguas puras
y el mexicano no es la excepción. No debemos, lo malo es que ésta es es una
concepción muy generalizada de que las lenguas se degradan, de que las lenguas
decaen y se empobrecen incluso por la mezcla con otras lenguas.
The object of “Speaking Mexicano” is precisely to understand how a language
or the speakers of a language, in this case the speakers of the Malinche, find
themselves facing, facing the dilemma of the loss or maintenance of their mother
tongue. This work that is certainly the most complete work of sociolinguistics that
we have to date, celebrates these voices and that possibility of continuity of Mexicano
in a very mature way with Spanish. This brings them as the colleagues have said,
to make a reflection that seems to me very important. Pure languages do not exist
and Mexicano is not the exception. We shouldn’t, the bad thing is that this is a very
generalized concept that languages degrade, that languages experience a downturn
and become impoverished including by the mixture with other languages.
An ideology of descriptive language versus prescriptive language, the mainstay of con-

temporary linguistics, belied these comments. As linguistic anthropologists working
on Native languages in Native communities, we are often faced with the desire to fly
the flag of descriptivism, and point out the dangers of prescriptivism. These ideologies
of language brought by us, the outsider-scholars, are part of the ideological mix in
indigenous communities. Some local residents of Contla, after the presentation, made
comments that revealed purist language ideologies about the very syncretic speech
that Hill and Hill (1986, 1999) had described.
In the section that follows, my collaborator Ramos Rosales Flores’ insightful
commentary about syncretism and globalizing languages is reproduced, followed by
his commentaries in 2012. The intertextual connections between the ideas in these
two commentaries and the larger discourse on language, identity and authenticity are
explored, followed by Jane Hill’s 1999 comments, and audience reception.
6. Intertextual conversations
The book presentation, the planning of it, the event itself, and post-event d
iscussions
comprised a set of conversations between collaborators and co-authors. Ramos Rosales
Flores offered a striking commentary at the book presentation, based on his critical
reading of Speaking Mexicano, He found the book to be an eye opener and it served
as a catalyst his thinking and work on language revitalization from within the state
indigenous education system. In his commentary, which we reproduce in its entirety
 Jacqueline Messing & Ramos Rosales Flores
below, proclaims the vitality of Mexicano, as a language that exists still in many hiding
places that are not always observable to others.
(6) Ramos Rosales Flores, Bilingual-intercultural school teacher and teacher

trainer
Hace unos días yo pensaba que para rescatar nuestra lengua sería necesario iniciar
por una investigación, donde la búsqueda sería en aquellos escondites donde se
guarda la lengua de manera pura – los archivos, los museos. Pero al leer este libro
ha cambiado mi manera de pensar. Me mueve a pensar en otro sentido, y creo
que el pretender hacer una investigación de este tipo si tendría sentido pero creo
yo a través de la lectura de este libro, creo yo que es mas importante hacer una
investigación dentro de mi misma lengua, en la que yo estoy hablando, la que
practico, ahora creo que la forma en que hablo, mezclando algunas palabras del
español dentro de mi diálogo, mexicanizando dichos términos también tienen
un valor y permiten de alguna manera que el idioma siga existiendo de manera
dinámica. Porque si yo me adentro en una investigación documental de archivos,
de documentos, por ahí bien escondidos, resguardados en museos también,
como que esta investigación no tendría sentido, porque es una investigación que
me llevaría tal vez a saber la lengua clásica, pero no es práctica para mí. Bueno,
todo esto ha permitido que ahora surjan en mi mente algunas inquietudes que
tal vez ponga en práctica con mis compañeros nuevamente de trabajo, con mis
amigos, alumnos del salón de clases y creo que nuevamente valdrá la pena hacer
este trabajo.
El náhuatl no es una lengua en proceso de desaparecer como por ahí
algunos de mis compañeros lo manejaban, ya que al igual que todas las lenguas
del mundo esta también se encuentra en constantemente movimiento, y que al
igual que el español recibe la influencia de idiomas homogenizantes, el náhuatl
recibe primeramente la influencia del español, mexicanizando los términos con
una reacción defensiva con una estrategia tal vez de sobrevivencia.
Como dice en el último párrafo del prólogo, a mí me llamó mucho la
atención, dice: “Esperamos que ‘Hablando Mexicano’ resulte estimulante para
una nueva generación de lectores;” a mí me ha resultado bastante estimulante
porque estoy pensando iniciar una investigación y a la vez una aventura
dentro de mi propia manera de hablar, el mexicano sincrético, creo que es más
importante eso.
Si nosotros vemos al mexicano, está en su diaria lucha de sobrevivir,
el mexicano sincrético sigue en su lucha por sobrevivir, sigue en la lucha por
permanecer en los diálogos que se generan alrededor de una buena medida de
pulque, o como dicen los autores, alrededor del compadrito nectzintli.
Creo que el mexicano es parte de nuestra indumentaria, yo lo entiendo
así, y que aún no hemos permitido que nos desnuden, y mucho menos que nos
impongan de manera total otro idioma, lo que sí hemos permitido es remendar
estas ropas lingüísticas con aportes del español, y esto no significa que nuestro
idioma cambie o nuestro mexicano cambie, solo se renueva y se actualiza de
Syncretic speech, linguistic ideology, and intertextuality 
manera constante, ya que es mejor traer cada día un vestuario adecuado a estos
tiempos que caminar cada día con un mexicano roído por las ausencias y los
olvidos de los inusuales términos que forman estancadas expresiones, éstas se
han quedado atrapadas en documentos de museos, o en una concepción purista
de un lingüista conservador. Con la ayuda de este texto, creo ahora que es mejor
imprimir palabras vivas a nuestra expresión de nuestro cotidiano hablar, que
incluir en nuestro discurso palabras sin una viva significación.
Mirando un poco mas allá del español sincretizador del mexicano, también
es importante observar el bombardeo a través de los actuales medios de
comunicación de otros idiomas que enriquecen el aspecto sincrético del mexicano.
Uno de estos idiomas es el inglés, y que primeramente este idioma sincretiza al
español, o sea que lo oprime, y el español a su vez sincretiza al náhuatl, esto resulta
que el náhuatl es doblemente oprimido.
Finalmente creo que es importante tomar otra foto, como ellos lo
manifestaban por ahí en alguna línea del libro, que ellos solamente habían
tomado una foto instantánea de un momento histórico de esta región, pero yo
creo que es necesario volver a tomar otra foto instantánea de esta historia, de
estos pueblos ahora, y hablar no solamente desde la perspectiva del español como
idioma sincretizador, sino también de la perspectiva cultural universalmente
sincretizante y que al mexicano y a cualquier otro idioma del mundo no se le
puede inmunizar contra estas influencias.
Creo que este fenómeno de la sincretización de las lenguas no se puede
separar del continuo movimiento de todos los idiomas del mundo, en alta o
baja medida todos los idiomas han sido sincretizados por varios idiomas. Y si
hablamos por ejemplo en los idiomas que se manejaron antes de la invasión
española también había sincretización entre estos idiomas, entonces, es un
fenómeno que no podemos separar del movimiento continuo social y por lo
tanto lingüista. Es todo mi comentario. Gracias.
A few days back I was thinking that in order to rescue our language it would
be necessary to begin a research Project, where the search would be in those far
away, hidden places where the pure language is kept – the archives, the museums.
So on reading this book it moves me to think in another way, and I think that to
undertake research of this type would make sense because I, through the reading
this book, I think that it’s more important to do research within my own language,
in what I am speaking, what I practice. Now I think that the way in which I speak,
mixing some words from Spanish into my dialogue, Mexicano-izing said terms
also has value and permits in some way that the language continue existing in
a dynamic manner. Because if I get into documentary research in archives, of
documents that are well hidden, guarded in museums perhaps, so that research
would not make sense, because well, it’s research that would perhaps bring me
to know the classical language, but that isn’t [Mexicano in] practice for me.
Well, all this has permitted that now some concerns have arisen in my mind
that perhaps I’ll put into practice with my work colleagues once again, with my
friends, students in the classroom and I think that again it will be worthwhile to
do this work.
 Jacqueline Messing & Ramos Rosales Flores
Nahuatl is not a language in process of disappearing as some of my colleagues

affirm, since the same as all of the world’s languages, Nahuatl finds itself in constant
change and motion. The same as Spanish receives the influence of homogenizing
languages, Nahuatl receives primarily the influence of Spanish, Mexicano-izing
terms as a defensive reaction and perhaps as a strategy for survival. Yes.
As the last paragraph of the [Speaking Mexicano’s] prologue says, “We
hope that Speaking Mexicano will be just as stimulating for a new generation of
readers.” It has been very stimulating for me. Because perhaps I am thinking of
initiating a research project and at the same time an adventure within my own
way of speaking syncretic Mexicano as it was said a moment ago, I think that this
is more important.
If we see Mexicano, it’s in its daily struggle for survival, syncretic Mexicano
continues in it’s struggle to survive, it continues in the fight to remain in the
dialgoues that are generated by a good measure of pulque (fermented agave drink),
or as the authors say, around the “compadrito nectzintli” dear compadre pulque [in
Mexicano with honorific].
I think that Mexicano is part of our costume, I understand it this way, and
we still haven’t allowed that they unclothe us, and much less that they impose in
a complete manner another language, what we have indeed allowed is to patch
these linguistic [articles of] clothing with contributions from Spanish, and this does
not mean that our language changes or our Mexicano changes, just that it renews
itself and updates itself in a constant manner, since it is better each day to bring
a wardrobe appropriate to these times than to walk each day with a Mexicano
[language] that is corroded by the absences and the oblivion of unused terms that
form stilted expressions, these have remained trapped in museum documents, or in
a purist concept of a conservative linguist. With the help of this text, I think that it is
better to print live words of our daily expression, of speaking, which includes in our
discourse words without a real meaning.
Looking a bit beyond the Spanish that syncretizes Mexicano, it’s also important
to observe the bombardment via today’s means of communications of other languages
that enrich the syncretic aspect of Mexicano. One of these languages is English, and it
is primarily this language syncretizing Spanish, that is that it oppresses it, and Spanish
in turn syncretizes Nahuatl, this results in Nahuatl being doubly oppressed.
Finally I think that it is important to take another photo, the way that they
express it there in one of the lines of the book, that they only had taken an instant
photo of a historical moment in this region, but I think that it is necessary to go
and take another instant photo of this history, of these towns now, and to speak
not only from the perspective of Spanish as a syncretizing language, but rather also
the cultural perspective that is universally syncretizing and that Mexicano and any
other language of the world cannot be immunized against these influences.
I think that this phenomenon of syncretization of languages cannot be separated
from the continual movement of all the languages of the world, to a high or low
extent all languages have been syncretized by several languages. And if we speak for
example of the languages that were used before the Spanish invasion there was also
Syncretic speech, linguistic ideology, and intertextuality 
syncretism between these languages, so then, it is a phenomenon that we cannot

separate from the continuous movement that is social and therefore linguistic. This
my entire commentary. Thank you.
R. Rosales Flores found the Speaking Mexicano book invited him to shift his
perspective on “mixed languages.” His statement illustrates how a published mono-
graph like Speaking Mexicano, can have the power to cause speakers to challenge their
own views of their native language, in this case, regarding the questioning of a purist
ideology of ‘mixed speech.’ This reading of Hablando Mexicano inspired a discussion
of language change and genesis similar to Mufwene’s comments (2004) that languages
are constantly dynamic. As R. Rosales Flores pointed out in 1999, “Nahuatl finds itself
in constant change and motion. The same as Spanish receives the influence of homoge-
nizing languages, Nahuatl receives primarily the influence of Spanish, Mexicano-izing
terms as a defensive reaction and perhaps as a strategy for survival.” The idea that
a native language is constantly in a state of change, and can accommodate external
linguistic influences is one that carries with it some hope for language revitalization
work.
Follow up discussions continued during collaborative research with J. Messing.
The book presentation also became the starting point for a working group called
Matitlatohcan Mexicano: Comite Promotor de la Lengua Mexicana – Let’s Speak
Mexicano: Committee for the Promotion of the Mexicano Language. The idea for the
group came from José Antonio Flores Farfán, and included many of the participants
in the book presenation, and additional teachers and townspeople, some of them
university students, and outside researchers including Elsie Rockwell.
In 2012, while working on this chapter, I had some conversations with R. Rosales
Flores regarding what he now thought of his earlier comments on the book, from 1999.
Some excerpts from his responses follow, in Mexicano, followed by Spanish and my
English translation:
(7) Ramos Rosales Flores, 2012

In Mexicano:
Nik pia’miek tla’tolme uan ika non nik neki ni mechon tlapuis, axan ni mo ueimati’
miek, tlika ne ni kualti ni mechon tlapuis ika no mexikatla’tol, nomeuantsitsi
xon mo yolmiktikan uan xon mo tsa’tsilikan kuak to techon investigaroa, kuak
nomeuantsitsi non ko neki to techon entenderoske, ne ni mo yolmiktis uan ni mo
tsatsilis i nauak no ikniuan kuak yokmo ki neki nahuatlatoske kuak ki neki yesque
kaxtiltlatlaka uan mo tlapuiske solo ikan kaxtil tla’tol uan ki neki ki po’poloske i
min xayak, i min tlakayo, i min neluayo’.
In Spanish:
Tengo muchas palabras y con eso quiero dirigirme a ustedes, ahora me siento muy
orgulloso de poder platicarles con mi lengua, ustedes preocúpense y discutan
 Jacqueline Messing & Ramos Rosales Flores
cuando nos investigan, cuando ustedes tratan de entendernos, yo me preocuparé

y discutiré con mis hermanos cuando ellos ya no quieren usar el mexikano,
cuando quieren ser solo hablantes del español y quieren ser como el español o
mestizo, y quieren perder la lengua, quieren cambiar su rostro, quieren ser otras
personas, quieren perder sus raíces.
[El libro] Me influyó de distintas maneras, primeramente en otorgarle
un valor a mi náhuatl de ese momento, antes de esto sentía y pensaba que el
conocimiento acerca de mi lengua era muy pobre y que de alguna manera no
tenía valor. Con la lectura del libro di un salto gigantesco para tener el valor
de mostrarme con mis pensamientos y mis ideas en relación a la lengua, la
conversación, y la lectura me iniciaron en mi proceso de autorecuperación.
El trabajo de José Antonio fue otro grano de arena en este proceso. Ahora
estoy seguro de que lo poco o mucho que se tiene un valor y es algo que vale la
pena mostrarse y que coadyuva o contribuye a la reactivación.
[…]
Ahora lo que más he hecho es que, desde mi familia, he tratado de que
primeramente le den valor profundo a nuestra cultura. Posteriormente he hecho
que [mi hijo] poco a poco se involucre en el uso de la lengua, he logrado que
participe en fechas importantes de nuestra cultura con un discurso o una poesía
en forma bilingüe, resaltando la importancia de la lengua. De igual manera lo he
hecho yo, en este momento tengo la figura de cronista municipal y por lo tanto
he aprovechado los espacios oportunos para manifestarme como un hablante
bilingüe ante diferentes públicos. De esta manera te ejemplifico el cambio de mi
pensamiento.
En cuanto al respeto del mexicano puro, mi manera de ver este aspecto de
la lengua es la misma, ya que para conservar o mantener un nivel de pureza es
importante el uso; no tiene ningún sentido mantener, conservar o discutir en
relación a la conservación pura de la lengua si ésta no se usa. Tener esta posición
ante la lengua es tratar de convertir a la lengua en un objeto inerte o en una pieza
de exhibición de museo; ésta no es mi preocupación en este momento y tampoco
será en otro momento. Justifico el sincretismo, y este fenómeno se reducirá solo
con el uso y la recreación de la lengua.
[…]
Después de aproximadamente trece años de haber tenido el honor de ser
uno de los invitados para hacer un comentario en torno a la presentación del
libro “Hablando Mexicano”, mi pensamiento es totalmente diferente, en cuanto
al rescate de las lenguas, ahora estoy seguro de que para hacer que mi lengua
permanezca el mayor tiempo y en el mayor espacio posible de las interacciones
de mi entorno social, permanente, ocasional, lejano y cercano es, que yo sea un
elemento activo tanto en el uso y en la promoción de esta esplendorosa forma de
hablar.
Estoy cierto en este momento de que además de hacer investigación, es vital
convertirme en un elemento reactivador de mi propia lengua.
Syncretic speech, linguistic ideology, and intertextuality 
Por otro lado en cuanto a la influencia o la opresión de las lenguas, además

de ser un fenómeno inevitable, es también un fenómeno que sucede de manera
automática por la debilidad que presentan las lenguas en algún momento, esta
debilidad de identidad puede ser ocasionada por diversos factores, sobre todo
internos, es solo una forma de manifestarse la aspiración de “salir adelante” como
una actitud de negación a lo que uno realmente es, ésta es solo una forma de
debilitar la fortaleza cultural y es cuando se origina o se provoca que suceda la
opresión en mayor o menor medida de las lenguas.
Por lo tanto los únicos responsables de que una lengua se ponga en peligro o
en riesgo de extinción somos los que pertenecemos a una u otra cultura.
Translation:
I have many words and so I want to address you all, now I feel very proud to be
able to converse with you all with my language, you all concern yourselves with and
discuss when you research us, when you all try to understand us, I will work on and
discuss with my brothers/sisters when they don’t want to use Mexicano anymore,
when they only want to be speakers of Spanish and want to be like the Spanish or
Mestizo, and want to lose the language, they want to change their face, they want to
be other people, they want to lose their roots.
[The book] influenced me in various ways, first in granting a value to my
Nahuatl of that moment, before that I felt and thought that the knowledge about my
language was quite poor and that in some way it did not have value. On reading the
book I took a huge leap to have the courage to show through my thoughts and ideas
about language, conversation, and reading was initiated a process of self-recovery.
The work of José Antonio was another grain of sand in this process. Now I am
sure that the little or much that has value is something that is worth showing and
that assists or contributes to revitalization.
[…].
Now what I have done most is from my family, I have tried to [see to it that],
first off, they give profound value to our culture. Subsequently I have seen to it that
my son has little by little become involved in the use of the language, I’ve achieved his
participation in important dates of our culture with a speech or a poem in bilingual
form, highlighting the importance of the language. In the same way I have done it,
at this time I have the post of Municipal Oral Historian and therefore I have taken
advantage of the opportune spaces to show myself as a bilingual speaker in front of
different audiences. In this way I illustrate my change in thinking.
Regarding respect towards pure Mexicano, my way of seeing this aspect of
the language is the same, since to conserve or maintain a level of purity the use is
important; there is no point to maintain, conserve o discuss the pure conservation of
the language if it isn’t used. To have this position regarding the language is to try and
convert language into an inert object or an exhibition ítem; this isn’t my concern
in this moment and it won’t be in another moment. I justify syncretism, and this
phenomenon will only diminish with the use and recreation of the language.
[…]
 Jacqueline Messing & Ramos Rosales Flores
After approximately thirteen years since having had the honor of being one of
the invited guests to give commentary on the presentation of the book “Hablando
Mexicano,” my thinking is totally different, with regard to the rescue of languages.
Now I am sure that in order for my language to remain the longest time possible and
in the largest possible space in the interactions of my social environment, permanent,
occasional, far reaching and close-by, it’s if I am an active element as much in the
use and the promotion of this splendid form of speaking.
I am certain that in this moment that in addition to doing research, it is vital to
convert myself to a revitalizing element [agent] for my own language.
On the other hand, regarding the influence or the oppression of languages, in
addition to being an inevitable phenomenon, it’s also a phenomenon that happens
automatically due to the weakness that languages present sometimes, this identity
weakness that can be due to diverse factors, especially internal ones, it’s only a
manner of showing the aspiration to “salir adelante” (forge ahead) as an attitude
of negation of that what one truly is, this is just a way of debilitating strength of a
culture and it’s what leads to or provokes the oppression of languages in a major or
minor measure.
Subsequently the only responsible parties for a language to become endangered
or at risk of extinction are we, those who pertain to one culture or another.
J. Messing’s commentary: Opening in Mexicano and transitioning to Spanish, my

collaborator addresses an international audience with his words. The identities of
researchers and locals are highlighted early on “when you resarch us, when you all
try to understand us” and indeed this has been a topic of much conversation between
R. Rosales Flores and myself. In this chapter I label these identities as “insider scholars’
and “outsider scholars.”
I am particularly struck by the description of Rosales Flores’ “brothers/sisters” as
sometimes wanting “to be like the Spanish or Mestizo, and want to lose the language.”
He asserts that they “want to lose the language, they want to change their face, they
want to be other people, they want to lose their roots.” This is an interpretation of
local ideology as quite active, an active choice to lose the language, rather than an
ambivalence. Later in the commentary, my collaborator highlights the importance
of being an “active element” to promote the use of Mexicano, “this splendid form of
speaking.” He promotes Mexicano language use with his young son, and symbolically
through the use of Mexicano in public speeches where Spanish would otherwise be
used, and places blame of linguistic oppression on “identity weakness,” a negation
of “what one truly is.” On the idea of local intellectuals and outside researchers can
work towards language revitalization, in addition to ethnographic descriptions of
linguistic ideologies, I am reminded of some insightful comments made by my col-
leagues in a forthcoming volume about the importance of research in native language
communities as informing praxis (McCarty, Wyman & Nichols forthcoming). In the
next section I present Jane Hill’s speech to the book presentation.
Syncretic speech, linguistic ideology, and intertextuality 
Photo: Authors Jane Hill, Kenneth Hill, and one of the translators José Antonio Flores Farfán.
(8) Jane Hill

Mexicano:
Pero deveras non amoxtli ka amo tekiu
But truly this book is for your work
de nochtin amehuan uan
axan ye non de amehuan oksepa’
of all of you, and now it is yours once again
para que (n?)ankile
for you to respond
Huan ojala que han nechpaktis nonamochtli. Occepa miek tlazokamati para
otechinvitaroa.
And hopefully you will like this book. Once again many thanks for the invitation.
Lo que traté de decir, muchas gracias a todos por invitarnos y ahora tenemos
que dejar esta obra “Hablando Mexicano” a ustedes y ojalá que les guste. En
escribir “Hablando Mexicano” tratamos de reconocer la riqueza que es el habla
bilingüe de la gente de la Malinche y reconocer la inteligencia, el trabajo duro
y la capacidad no solo de sobrevivir, pero ir adelante al próximo siglo con toda
su herencia que incluye “Hablando Mexicano”. Y tenemos que decir que no
existe ninguna lengua que pare de desarrollarse, hasta mi idioma inglés sigue
desarrollándose por pedir préstamos de todos los idiomas del mundo, bueno
hasta del mexicano y hemos pedido prestado del mexicano una de mis palabras
favoritas que es chocolate, que viene de chocolatl, y este chocolatl, su bebida de
cacao y agua sus antecesores echaban chile y nosotros echamos azúcar y canela.
Y azúcar, vino a nosotros desde India por las manos de los Árabes, hasta Europa
y hasta las Américas y la canela vino desde Indonesia, y ahora en México y en los
Estados Unidos los mezclamos, bueno la canela y el azúcar con chocolate y si es
mejor, pues quien sabe, pero es nuestra herencia. Muchas Gracias.
 Jacqueline Messing & Ramos Rosales Flores
What I tried to say, thank you very much to everyone for inviting us and now
we have to leave this work “Speaking Mexicano” to you and hopefully you like it. In
writing “Speaking Mexicano” we tried to recognize the richness that is the bilingual
speech of the people of the Malinche and to recognize the intelligence, the hard work
and capacity not only to survive, but to go forward to the next century with all of
your heritage that includes “Speaking Mexicano.” And we have to say that there
does not exist a single language that stops developing itself, as my language English
continues developing by taking loans from all of the languages of the world, well also
from Mexicano and we have taken as loans from Mexicano one of my favorite words
that is chocolate, that comes from ‘chocolatl,’ and that ‘chocolatl,’ your drink of cacao
and water whose antecedents were ‘chilli’ and we add sugar and cinnamon. And
sugar, came to us from India from the hands of the Arabs, from Europe and to the
Americas, and cinnamon came from Indonesia and now in Mexico and the United
States we mix them, well cinnamon and sugar with chocolate and it is better, well
who knows, but that is our heritage. Thank you very much.
So really thank you for everything and now today you again invite us and
hopefully you will enjoy our book. Again many thanks for inviting us.
Local Newspaper photo, June 1999. Pictured are all the speakers and the mayor. Translation of
caption: “The book “Hablando Mexicano”: La dinámica de una lengua sincrética en el centro de
México was presented. In one of the photos, Pablo Flores Galicia, municipal president of Contla,
with the writers Jane Hill and Kenneth Hill
7. Audience reception
After Jane Hill completed her commentary, the discussion was opened to the audience.
There was a high attendance for an event of this type, and loud applause indicated
audience appreciation. There was a ribbon cutting ceremony to open the event, and
Syncretic speech, linguistic ideology, and intertextuality 
newspaper reports covered the story in a local paper (see photo). The presence of the
mayor was notable. I heard lots of subsequent comments, most notably among teach-
ers and students in the school I was observing, plus people I knew in the community.
Just after the event finished, I experienced an interesting instance of someone
showing me the book Hablando Mexicano, and pointing to a passage on a page,
questioned its content. In contrast to the previous examples in which both community-
insider and community-outsider speaker-scholars described Mexicano as in a state of
flux, in this last example, a person I knew well took the book Hablando Mexicano and
pointed to one of the Hills’ linguistic examples, where the Spanish was embedded into
Mexicano speech, and told me: “Esto está mal,” “This is wrong.” He went on to say that
the authors should not have put Spanish into the text, that the very act of publish-
ing mixed speech rather than the legitimo mexicano was disrespectful to the Nahuatl
language. This comment, questioning the validity of publishing a book that contains
“tainted” Mexicano, embodied the very purism that was the ideological backdrop to
the discussion on syncretic speech, and at the same time the importance of outsider-
scholars understanding the very respect that is afforded to the ancestral language.
8. On meta-discursive practices, translation and publication
The late Professor Luis Reyes Garcia, a resident of Tlaxcalan and highly respected
scholar in international, national and local circles, attended the book presentation and
during the discussion period made a strong, compelling speech about the importance
of translating work produced outside of Mexico and publishing it to make it available
to students and scholars within Mexico.

The late Luis Reyes García, well known Nahuatl language, history, and art scholar, during the
commentary period
 Jacqueline Messing & Ramos Rosales Flores
(9) Luis Reyes Garcia, Nahuatl scholar, Art Historian, Anthropologist

…yo creo que tenemos un compromiso, como casa de cultura, como
universidades, como centros de investigación y de docencia, para presionar, para
hacer un plan de edición de las obras que nos interesa leer porque son de aquí.
Porque queremos entenderlas, queremos verlas, queremos aprender si es que
tenemos que aprender o tenemos que rebatir algo. Muchas gracias.
…I think that we have a commitment, as House of Culture, as universities, as
centers of research and teaching, to pressure, to make a plan of editing the works
that interest us to read because they are from here. Because we want to understand
them, we want to see them, we want to learn if it is that we have to learn or we have
to resist something. Thank you very much.
I have heard these sentiments voiced on several occasions among researchers and stu-
dents in Mexico, about making research about Mexico available within Mexico, and
in the national language rather than English (or French, German, etc.). The fact that
one of the most important books about Mexican linguistics and anthropology had
been translated and published was lauded. I know from personal experience that the
paucity of non-English academic writing in the U.S. is driven in large part by the North
American tenure system, one that prioritizes peer-reviewed work published by recog-
nized (and ranked) journals within the U.S. Funding is altogether nonexistent for such
translations. Despite these challenges, Professor Reyes Garcia eloquently raised a criti-
cal point that has a powerful ideology of language and power at its core, and we North
American academics need to find ways to change or circumvent our system to make
publications in our field languages available, as Jane and Ken Hill were able to do.4
As Bauman and Briggs (2000: 142) point out
Metadiscursive practices shape, both positively and negatively, processes of

producing and receiving texts, affecting who is authorized to speak or write, or to
be listened to or read, and in what sorts of social institutional spaces.
The publishing of Hablando Mexicano has had far-reaching effects, (re-)opening a dia-
logue in rural Tlaxcala and Latin American Anthropology and Linguistics classrooms,
challenging people to consider syncretism from all its angles, and to offer an in depth
. I agree strongly with Dr. Reyes Garcia, and at the same time have yet to publish an article
in a language other than English during my pre-tenure years. After completing my disserta-
tion I wrote an article-length précis of my dissertation and distributed it to several dozen sup-
porters of my research in Tlaxcala. Some people told me that they were surprised that I came
back and gave them these copies. This is definitely an issue that we North American academics
need to keep in mind. Today, online publications may be the best way to increase accessibility
and rapid transmission of research data to readers in our field languages.
Syncretic speech, linguistic ideology, and intertextuality 
ethnographic view of one of Mexico’s majority indigenous languages. The fact that the
book’s contents are discussed and debated by students of language, culture and Native
Studies across regional and national borders is what forms the metadiscursive practice.
The metadiscourse on the day of the book presentation was indexical of the linguistic
awareness shared by speakers, semi-speakers and analyzers of contemporary forms of
speaking Mexicano.
The event I’ve described in this chapter was simultaneously a local event, held in a
local cultural and educational space, and an academic event to discuss the contents
of a scholarly book. The relationship between Mexicano and Spanish, and the uses of
both languages as well as the endorsement of such discussion by well-known North
American scholars were the reason that seventy people from diverse backgrounds
came together in June 1999. After the event, I became interested in the role that out-
sider-scholars can play in conjunction with the work of insider-scholars, and how
these identities can overlap.
One of my strongest findings in the six months after this book presentation is
that community-outsiders can convene meetings among community-insiders that
bring together diverse stake-holders who might not otherwise have gotten together
to work on language-related issues. I think that this is due to a combination of
factors, including ambivalence, customary communication barriers (“I don’t talk
to that person because I don’t know them or their family”) or histories of disagree-
ments between locals, or their ancestors (i.e. “My ancestors in my town used to
be discriminated by your ancestors in your town”). Ethnographers and applied
researchers must of course be sensitive to existing local ideologies of language,
identity, and socioeconomics as well as local histories to achieve fruitful collabora-
tive research projects.
The study of syncretic speech highlights what happens when a colonial language
collides with a native language, after the speakers go through power conflicts over
time. In comparing Tlaxcalan linguistic history to the one in my mother’s homeland
of Switzerland, where there exists bilingualism that looks a lot like classic Fergusonian
diglossia, with languages on equal footing. For Latin America, we need a new linguis-
tic description of bilingualism, one that is more fitting for indigenous contexts that
produced broad language contact phenomenon, such as syncretic speech. Syncretic
speech was formed as part of a process of speaker survival of racism and structural
inequality that were direct results of colonialism.
Syncretic speech seems to have become enregistered (Agha 2006), socially recog-
nized as indexical of the mixing of languages and traditions after conquest. Perhaps
 Jacqueline Messing & Ramos Rosales Flores
syncretic speech emerged as an enregistered voice, a type of 16th century social voice
that emerged during the period of conquest. This enregistered voice is open to change
over time.
Today in Mexico, English has been added to the mix, with Tlaxcalan immigrants
to the U.S. or Canada returning with language skills, and musical tastes favoring
English or Spanish/English hybrids such as reggaeton. Increasingly, since the early
2000’s, in Tlaxcala you see T-shirts emblazoned with English words or place names.
As R. Rosales Flores pointed out in his 1999 commentary, no language is exempt from
the inclusion of English in today’s speech practices in Mexico and elsewhere. His com-
ment is reminiscent of Mufwene’s assertion that languages are always undergoing a
process of linguistic creation and recreation.
R. Rosales Flores has asked what the larger point is to academic research, even
as he was quite influenced in his teaching and cultural work by the reading of the
Hills’ book. The genesis of my thinking about outsider-scholars and insider-scholars
comes from the commentary given by Rosales Flores at this event. Perhaps what
is so powerful about an event like this one is that it opened a dialogue between
people interested in language who didn’t otherwise communicate, and between
people who communicate with each other but not about linguistic issues. The late
Professor Reyes Garcia’s argument for increasing publication in Spanish by North
American academics suggests another method for increasing communication

between students of sociolinguistic issues. Through this article and its attempt to
translate and publish the ideas of a Tlaxcalan insider-scholar in an English-language
publication, my goal is increase this type of communication. Ideologies will often
differ between outsider-academics, insider-academics and non-academic speakers,
and it is through collaboration and discussion that the multiplicities become appar-
ent and collaborations can take place.
Acknowledgements
The authors extend a miek tlazokamati to Jane Hilltzi for her scholarship and research
in the Malinche/Malintzi region, to Kennethtzi and all the participants of the event
discussed in this chapter. Special thanks to Professor Refugio Nava Nava for assist-
ing in the planning of the event, and to the Instituto Tlaxcalteca de la Cultura, and
Presidencia Municipal de Contla, Casa de Piedra-Contla for supporting the event,
and to José Antonio Flores Farfan for conceiving the idea in the first place. JM
thanks the Fulbright-Garcia Robles program, CIESAS, and José Moreno and Edgar
Amador at the University of South Florida for transcribing recordings of this event.
Finally, thank you to the editors for helpful commentary, and for putting together
this volume.
Syncretic speech, linguistic ideology, and intertextuality 
References
Agha, Asif. 2006. Language and Social Relations. Cambridge: CUP.

Bauman, Richard, & Charles Briggs. 2000. Language philosophy as language ideology: John
Locke and Johann Gottfried Herder. In Regimes of Language: Ideologies, Polities, and Identi-
ties, Paul V. Kroskrity (ed.) 139–204. Santa Fe: School of American Research Press.
Bauman, Richard & Briggs, Charles. 2003. Voices of Modernity: Language Ideologies and the
Politics of Inequality. Cambridge: CUP.
Dorian, Nancy. 1977. The problem of the semi-speaker in language death. I nternational Journal
of the Sociology of Language 12: 23–32.
Fader, Ayala. 2007. Redeeming sacred sparks: Syncretism and gendered language shift among
Hasidic Jews in New York. Journal of Linguistic Anthropology 17(1): 1–23.
Flores Farfán, José Antonio. 1999. Cuatreros Somos y Toindioma Hablamos. México: CIESAS
Press.
Flores Farfán, José Antonio. 2003. Nahuatl purism: Between language innovation, maintenance
and shift. In Purism in the Age of Globalization, Joseph Brincat, Winfried Boeder & Thomas
Stolz (eds), 281–313. Bochum: Universitätsverlag Dr. N. Brockmeyer.
Gal, Susan. 1998. Multiplicity and contestation among linguistic ideologies. In Language
Ideologies: Practice and Theory, Bambi B. Schieffelin, Kathryn A. Woolard & Paul V.
Kroskrity (eds), 317–331. Oxford: OUP.
García Canclini, Néstor. 1994. De lo local a lo global: Perspectivas desde la antropología. Mexico
City: Universidad Autónoma Metropolitana.
Garrett, Paul. 2004. Language contact and contact languages. In A Companion to Linguistic
Anthropology, Alessandro Duranti (ed.), 46–72. Oxford: Blackwell.
Hill, Jane, & Hill, Kenneth. 1986. Speaking Mexicano: Dynamics of Syncretic Language in Central
Mexico. Tucson AZ: University of Arizona Press.
Hill, Jane, & Kenneth Hill 1999. Hablando Mexicano: La dinámica de una lengua sincrética en
el centro de México (J. A. Flores Farfán and G. López-Cruz, Translators.). Mexico: CIESAS
Press.
Hill, Jane H. 1991. In Neca Gobierno de Puebla: Mexicano Penetrations of The Mexican State. In
Indians, Nation-States and Culture, Greg Urban & Joel Sherzer (eds). Austin TX: University
of Texas Press.
Hill, Jane H. 1995. “The voices of Don Gabriel: Responsibility and self in a modern Mexicano
Hill, Jane H. 1998. “Today there Is no respect:” Nostalgia, ‘respect,’ and oppositional discourse in
Mexicano (Nahuatl) language ideology. In Language Ideologies: Practice and Theory, Bambi
B. Schieffelin, Kathryn A. Woolard & Paul V. Kroskrity (eds). Oxford: OUP.
Hill, Jane H. 2004. Discussant commentary, invited session on language, racism and discourse,
organized by Jacqueline Messing, American Anthropological Association Annual Meeting.
Irvine, Judith & Gal, Susan. 2000. Language ideology and linguistic differentiation. In Regimes
of Language: Ideologies, Polities, and Identities, Paul V. Kroskrity (ed.). Santa Fe NM: School
of American Research Press.
McCarty, Teresa L., Wyman, Leisy & Nichols, Sheilah E. Forthcoming 2013. Introduction.
In Indigenous Youth and Bi/Multilingualism: Language Identity, Ideology, and Practice in
Dynamic Cultural Worlds, Leisy Wyman, Teresa L. Mccarty & Sheilah E. Nicholas (eds).
London: Routledge.
 Jacqueline Messing & Ramos Rosales Flores
Messing, Jacqueline. 2003. Ideological Multiplicity in Discourse: Language Shift and Bilingual
Schooling in Tlaxcala, Mexico. Ph.D. dissertation, University of Arizona.
Messing, Jacqueline. 2007a. Multiple ideologies and competing discourses: Language shift in
Tlaxcala, Mexico. Language in Society 36(4): 555–577.
Messing, Jacqueline. 2007b. Ideologies of public and private usages of language in Tlaxcala,
Mexico. International Journal of the Sociology of Language, 187–188: 211–227.
Messing, Jacqueline. Forthcoming 2013. “I didn’t know you knew Mexicano!:” Shifting ideolo-
gies, identities and ambivalence among youth in Tlaxcala, Mexico. In Indigenous Youth and
Bi/Multilingualism: Language Identity, Ideology, and Practice in Dynamic Cultural Worlds,
Leisy Wyman, Teresa L. Mccarty & Sheilah E. Nicholas (eds). London: Routledge.
Messing, Jacqueline. To appear. Speak to me in Nahuatl, I’ll answer in Spanish: Ideologies of
Language and Modernity in Tlaxcala, Mexico. Ms.
Messing, Jacqueline & Rockwell, Elsie. 2006. Local language promoters and new discursive
spaces: Mexicano in and out of schools in Tlaxcala. In Mexican Indigenous Languages at
the Dawn of the Twenty-First Century, Margarita Hidalgo (ed.). Berlin: Mouton de Gruyter.
Makihari, Miki. 2004. Linguistic syncretism and language ideologies: Transforming sociolin-
guistic hierarchy on Rapa Nui (Easter Island). American Anthropologist 106(3): 529–540.
Mufwene, Saliko. 2004. Language birth and death. Annual Review of Anthropology 33: 201–222.
Philips, Susan U. 1998. Language ideologies in institutions of power: A commentary. Language
Ideologies: Practice and Theory, Bambi B. Schieffelin, Kathryn A. Woolard & Paul V.
Kroskrity (eds). Oxford: OUP.
Philips, Susan U. 2000. Constructing a Tongan nation-state through language ideology in the
courtroom. In Regimes of Language: Ideologies, Polities, and Identities. Santa Fe NM: School
of American Research Press.
Rosales Flores, Ramos & Messing, Jacqueline. 2009. NIN TLA’TOL-These words: In honor of
Jane Hill’s forthcoming retirement. Anthropology Newsletter, January.
Tannen, Deborah. 2006. Intertextuality in interaction: Reframing family arguments in public
and private. Text and Talk 26(4–5): 597–617.
section 2
Approaches to the study of voices

and ideologies
Racism in discourse – analyses of practice
Narrative discriminations in Central
California’s indigenous narrative traditions
Relativism or (covert) racism?
Paul V. Kroskrity
UCLA Department of Anthropology
This chapter explores salvage era representations of Yokuts and Western Mono
Narratives in an attempt to understand the logic in use of anthropologists
and linguists who tended to characterize these narratives in a disparaging
manner. I explore two possible explanations for what at first blush appears to
be an unusually ethnocentric failure to appreciate difference and an exercise in
producing an aesthetic relativism that does not explain or understand narrative
difference but merely notes its existence. The first is essentially a historical
explanation and the second invokes Jane Hill’s notion of covert racism. In order
to assess the value of these different and perhaps competing explanations,
I introduce the results of my own ethnopoetic and ethnographic work on
Western Mono.
Keywords: Western Mono; linguistic racism; ethnopoetics; ethnographic;

narrative discrimination
In this chapter I explore mid-20th C. salvage era representations of Yokuts and Western
Mono narratives in an attempt to understand the professional language ideologies
(Kroskrity 2000) and academic practice of anthropologists and linguists who tended
to characterize these traditional narratives in a disparaging manner. Scholars, includ-
ing the anthropologist Anna Gayton and the linguist Stanley Newman, represented
these narratives as monotonous, deficient, and as generally lacking in artful narra-
tive qualities. I explore two possible explanations for what, upon first examination,
appears to be an unusually ethnocentric failure to appreciate difference and an exer-
cise in producing an aesthetic relativism that does not explain or understand narra-
tive and discursive difference but merely notes their “otherness”. The first – essentially
a historical explanation – focuses on the lack of ethnographic methods, the historical
paucity of a scholarly literature on ethnopoetics, and the comparative dearth of a
 Paul V. Kroskrity
critical literature on academic literacies available to these scholars. This account

would understand and ultimately apologize for an apparent scholarly disinterest,
among “salvage” researchers, for seeking to understand and approximate a “native
perspective” on narrative aesthetics. That same account would also offer an inter-
pretation for the failure of these scholars to have acknowledged the role of their own
culture of literacy in misrecognizing the cultural practice of indigenous Californian
discourse traditions.
The second, and alternative explanation, invokes Jane Hill’s (2001, 2008) notion
of covert racism to explain a pattern in which all representations of indigenous narra-
tive traditions appear to be negative and seem to rely on indirect indexical relations to
stereotypes of primitivity and “inevitable disappearance” (Deloria 2004: 10). In order
to assess the value of these different and perhaps non-mutually exclusive explanations,
I assess their salvage linguistic representations in light of my more recent ethnopo-
etic and ethnographic research on Western Mono (Kroskrity, Bethel, Reynolds 2002;
Kroskrity 2009).
My goals then are, first, to deconstruct the narrative discriminations of these
scholars by disclosing the ethnocentric basis of their apparent aesthetic taste and by
exposing their then current professional practice of decontextualized appreciation
that would only further contribute to the discursive marginalization of these Central
California narrative traditions. By narrative discrimination, I want to argue for the
theoretical utility of a concept similar to Hymes’s (1996) “narrative inequality” – the
important observation that stories, like the languages they are told in and the speakers
who tell them, are decidedly unequal, say, in the support they receive from state insti-
tutions. My preference for discursive discrimination, as I have elaborated elsewhere
(Kroskrity 2012: 152), stems from the lexical ambiguity which captures the multiple
and simultaneous indexing of aesthetic appreciation and political economic stratifica-
tion. As common sense actors we find it possible to talk positively about people having
“discriminating” taste. Such people are aesthetically informed and possess knowledge
sufficient to render sophisticated artistic evaluations. But such aesthetic discrimi-
nation is also connected to the indexing and reproduction of social hierarchies as
Bourdieu (1984) has demonstrated. By “narrative discriminations” here, I mean to
focus attention on the simultaneity of aesthetic appreciation and the (re-)production
of socioeconomic inequality.
My second goal, which is not fully realizable within the limited scope of this
chapter, is to contribute to a restoration of Western Mono voices by using ethnographic
and multimedia resources that help provide an alternative, community-based under-
standing (Kroskrity 2009; Kroskrity, Bethel & Reynolds 2002; Kroskrity & Reynolds
2001). Though Gayton and Newman base their analyses on both Yokuts and Western
Mono narratives, I will represent their analysis of each. Though I examine a Yokuts
language text, I am able to provide additional ethnographic and language ideological
Narrative discriminations in Central California 
restudies only for the Western Mono and language ideological restudies only for the
Western Mono communities I was able to study in original research from 1980–1994.
1. Western Mono: Hegemonic erasure and counter hegemonic responses
Today the Western Mono, by their own reckoning number 1800 in North Fork,
Auberry, and other Central California Rancherias. This total includes perhaps
25 highly fluent speakers – all in the oldest generation, making it a severely endan-
gered language according to Krauss’s (2007) classification of endangered lan-
guages. Precluding the need to provide extensive background here, I have treated
elsewhere the history of language contact, shift, and ideological change (Kroskrity
2009). Much of this can be efficiently summarized by saying that the commu-
nity went from a classic residual zone, in Nichols’s (1999) sense, as an adaptation
involving multilingualism, seasonal movement, and intermarriage to one that fea-
tured the aggressive spread of English, forceful suppression of the Mono language,
and a hegemonic pressure following a massive language shift that was facilitated
in part by indigenous language ideologies that prioritized language as an adaptive
tool (Kroskrity 2009).
Though early popular representations of these and comparable indigenous
verbal art traditions, such as Judson’s 1912 [1994] Myths and Legends of California
and the Old Southwest and Gifford and Block’s 1930[1990] California Indian Nights
contributed to their discursive marginalization by fetishizing their referential con-
tent as once local “lore” now in the control of the dominant society, recontextual-
izing them solely in English, and by erasing locally valued rhetorical features such
as “repetition” (Gifford & Block 1990[1930]: 43) and “long conversations” (Judson
1994 [1912]: 15) – both essential features of indigenous narrative to be sure – my
focus in this chapter is on “expert” academic representations. Rather than examine
the representational practices of de- and re-contextualization of indigenous storytell-
ing traditions designed with the primary goal of popularizing narrative “content” for
a reading public, I want to devote focal attention to specific works on the Mono (and
the neighboring Yokuts), that were written by professional academics for their peers
and meant to address not just content but aesthetic and stylistic issues as well. I think
these attempts at literary criticism of an oral tradition are an especially appropriate
site for understanding the professional and other language ideologies that further
contributed to the discursive marginalization of these narrative traditions.
After more than 70 years, much has changed since Anna H. Gayton and Stanley
Newman (1940) characterized what they called the “narrative style” of Yokuts and
Western Mono myths, and proceeded to supply a “deficit” image of these Central
Californian indigenous traditions. Though this collaborative research was performed
 Paul V. Kroskrity
and published as part of the salvage research orchestrated by Alfred Kroeber and his
University of California, Berkeley colleagues, the work continues to merit scholarly
attention for at least two reasons. One, the work became at least semi-canonical to
the evolving subfield of linguistic anthropology. Its inclusion in Hymes (1964) famous
red book Language in Culture and Society – arguably the first anthology of our sub-
field – both expresses Hymes’s evolving interest in what he would later call anthro-
pological philology and prefigures his scholarly quest to develop a notion of “style”
at a time when so many scholars were obsessed with a rage for grammatical order
(whether Bloomfieldian “structuralism” or Chomskyan “transformational-generative
grammar” (Bloomfield 1933; Chomsky 1957, 1965). It is important to remember the
historical chronology of linguistics and linguistic anthropology and its impact on
research on. Native American narrative traditions.
A tradition of textual collection had emerged since Boas’s (e.g. 1894, 1918,
1928) foundational work but scholars working around mid-century did not have
the benefits of several significant scholarly movements relevant for the study
of narratives. These include: (1) verbal art as performance (e.g. Bauman 1975),
(2) ethnopoetics (of Hymes 1981; Tedlock 1983), and (3) poetics and politics
(Bauman & Briggs 1990; Briggs & Bauman 1992), and (4) the ethnography of
communication (Gumperz & Hymes 1972; Bauman & Sherzer 1974). Lacking
these important resources, salvage linguists and anthropologists also labored
under pervasive paradigms that emphasized acculturation and assimilation, and
consequently they regarded their “salvage” efforts as directed almost exclusively
to academic elites and not toward the communities whose linguistic and cultural
heritage they collected and analyzed. But though these scholars never imagined the
heritage language persistence or the cultural interest shown by these c ommunities
in some form of language revitalization, the texts they created for consumption
by academic elites now circulate freely in digital form to internet-savvy heritage
language community members. This fact alone makes this chapter more than a
critique of the shortcomings of past scholarship; it is rather an attempt to correct
the record and improve the commentary that accompanies heritage language nar-
ratives as they circulate back to their home communities.
2. B
rief sketches of two salvage era researchers of indigenous California:
Anna Hadwick Gayton (1899–1977) and Stanley Newman (1905–1984)
A “salvage” anthropologist, and student of both Kroeber and Lowie at UC-B, special-
izing in Central California groups like the Mono, Ann H. Gayton was more familar
with trait-list ethnography, or with folkloristic “motifs” than with the rigorous study
of language (Gayton 1948). Her Mono stories were “taken in English” translation and
Narrative discriminations in Central California 
submitted to the same kinds of areal analysis of cultural elements as were other parts
of Mono culture (Gayton 1935). Though her emphasis was on content and not based
on a careful analysis of the Mono originals, Gayton nevertheless contributed a list of
comparative observations that suggested the use of benchmarks derived from the liter-
ary qualities of expository or essayist discursive prose so highly valorized in academic
settings (Collins 1996, 2009: 334).
She went on to a major career in Folklore and became the President of the
American Folklore Society in 1950 (Boyer 1978). Certainly Stanley Newman was
similarly accomplished. A student of Sapir, he followed his mentor from Chicago to
Yale. Newman abandoned his early career in English composition and literature and
committed to a professional focus on anthropological linguistics, earning his Ph.D.
from Yale on the basis of an outstanding grammar of Yokuts (Newman 1944; Silverstein
1989). His collaboration with Gayton consisted mainly of his providing an annotated
Yokuts text and a section on the “linguistic style” of Yokuts narratives. These sections
were added to Gayton’s synthetic discussion of “narrative style” for these two groups,
plot summaries of the collected narratives, and a discussion of areal themes and motifs
suggesting the influence of culture contact. For purposes here, I combine the remarks
of Gayton and Newman but it should be observed that all of Newman’s remarks were
focused upon Yokuts and though Gayton commented about stories in both languages,
her only knowledge of them was gleaned from English language translations.
3. Relativism or covert racism: The case for relativism
Collectively, both Gayton and Newman offered a putatively descriptive character-

ization that assumed the form of a list of contrastive expectations. They (Gayton &
Newman 1940) present seven partially overlapping negative features of Central
California indigenous narratives. These include: lexical deficiency, lack of figurative
language, simplicity, redundancy, lack of explication, lack of variation, lack of formal
structure. Though it is easy to prejudge Newman and Gayton harshly for what looks
like their ethnocentric English literacy based judgments, a closer examination reveals
that they were, comparatively speaking, attempting a relativism in their appreciation
of distinct “narrative styles”. Newman, for example, wrote (1940):
But by the same token, the stylistic features of English cannot appeal to the
intuitions of a Yokuts native. To him English must appear erratic, lacking in
those qualities of restraint and consistency that he finds in his own language.
He will see no uniformity in the pattern of English sentences…Behind this
unevenness of expression there seems to be a strident and feverish energy
obsessed with the need of expressing nuances that could best be left to
contextual inference.
 Paul V. Kroskrity
Relativism, I suppose, represents an improvement over treating discursive difference

in an even more discriminating manner but though Newman achieves a measure of
relativism by attempting to view English speakers’ literacy based expectations from
an other’s perspective, neither he nor Gayton ever succeeds in imagining a system-
atic alternative in which the cluster of normally dispreferred traits would make sense.
Does an appeal to relativism rationalize a reluctance to search for an alternative pat-
tern in structural form, performance practice, ethnographic detail, or speakers’ own
accounts? While I concede that advances in linguistic anthropology have considerably
broadened the scope of our investigations and that it is unfair, in many respects, to
judge these scholars by today’s methodological and theoretical standards, I still find it
completely appropriate to interrogate their unwillingness to attempt an understand-
ing of this array of “dispreferred” narrative norms. Classic statements on linguistic
relativity by Whorf that are contemporaneous with the study by Gayton and Newman,
clearly do not limit relativity to an isolation of difference and offer an alternative model
for displaying relativism. In works such as “The Relation of Habitual Thought and
Behavior to Language” (1939) and “Science and Linguistics” (1940), Whorf did not
merely detail the lack of tense-aspect markers in Hopi or note the lack of a general-
ized verb “to clean” in Shawnee. Whorf would also detail the richness of Hopi linguis-
tic resources for representing “preparation” and relate this linguistic complex to Hopi
ceremonial practice (Whorf 1939) or detail the precise alternative coding of specific
cleaning actions in Shawnee (Whorf 1940). But the relativism of Gayton and Newman
is mute with regard to locating an alternative pattern. Is there something else at work
that makes sense of both their willingness to note “difference” without understand-
ing it and their contentment with displaying hegemonic narrative forms without also
attempting to denaturalize them? I will have suggestions in my conclusion.
To exhaustively deconstruct the Gayton-Newman analysis and the apparently
pejorative characterizations it rests on, we would need to examine all seven negative
claims. Reserving a more comprehensive treatment for another publication, I will limit
the present treatment to the detailed analysis of only two of Gayton and Newman’s
partially overlapping claims: (1) repetition (or the lack of variation) and (2) the lack
of explanation.
4. (Artless) repetition
Newman [1940: 373] unpacks the “bareness” of Yokuts narratives when he explicates,

“in these stories there is no tendency to indulge in elaboration of concrete details. The
notions expressed remain at a highly generalized level.” Another key manifestation
of “simplicity and bareness” is the preference for “repetition” rather than “variation”.
Here the authors fail to appreciate the orality of Mono and Yokuts narratives and
Narrative discriminations in Central California 
thus misrecognize textual artifacts that display the artistic work of parallelism or
dramatic repetition in orally performed narrative with mere, “artless” repetition in
entexted form. Contrasting cultural perspectives, Newman (Newman & Gayton
1964[1940]: 374) compares Eurocentric to Native Californian narrative aesthetics,
when he writes: “although we may regard variety as an absolute virtue of style and rep-
etition as a universal sin, it is obvious that Yokuts cannot be driven in this direction. …
When a notion is to be repeated there is no need to avoid verbal repetition.”
But such words fail to appreciate the role of repetition in the parallelism of oral
discourse and to fully appreciate its patterned use.
We can view an example of repetition from Chawchilla Yokuts by Johnny Jones in
1931, collected by Newman himself. I am relying on its interlinear representation in
Geoffrey Gamble’s (1994: 17) collection of Yokuts texts.
(1) ‘ama’ ‘amin nophop ‘amil’ay ne:ye: ’ay ‘amam, Sa:liki’ki put’uh.
And his father came he:shook him wake up son
And his father came and shook him, “Wake up, my Son!”
(2) ‘ama’ ‘ohom’ ‘okot’oy.
And not he:got:up.
And he didn’t get up.
(3) ‘ama’ thawtham’ nim pu c’on wil’ay
and is dead my son he-says
And he said, “My son died.”
(4) ‘ama’ thah paxat’xo ‘am ’an
And then mourned they
And then they mourned.
(5) ‘ama’ yuk’ulhal’ thah ’ama’ pu c’on ‘amin tha:with’ay
And he was buried that one son his died
And they buried that one – his dead son.
(6) ‘ama’ he:te’
And that’s all.
And that is all.
Viewed strictly from a perspective that seems to regard English prose conventions as
the appropriate benchmark, it is true that the concluding passage (lines 1–6 above) is
“repetitious” and lacking in the syntactic diversity that is conventionally prescribed.
But Newman and Gayton never explore whether this pattern conforms to an a lternative
aesthetic or whether the oral performance needs to be regarded as a shaping factor.
It is also noteworthy that these scholars were so preoccupied with the transcribed
texts that they failed to relate the features of those texts to the oral performances that
were recontextualized in them. Newman and Gayton’s type of salvage linguistics did
 Paul V. Kroskrity
not attempt to collect narratives in culturally appropriate contexts even when this was
possible. To exacerbate the problem of more naturalistic collection, unobtrusive and
accurate recording devices were not yet a conventional part of the anthropological lin-
guist’s technological toolkit. Narrative texts thus could not be fluently performed; they
needed to be dictated to the novice linguist. And, perhaps unfortunately, none of the
Yokuts and Mono consultants exhibited the refusal to cooperate with the linguist or eth-
nologist that was displayed to K. David Harrison (2007) by a Tuvan (Siberian) storyteller
who sneered, “Do you expect me to tell stories to that thing”, meaning the linguist’s tape
recorder, requiring Harrison to assemble an improvised human audience by scouring
the neighboring area in order to provide the audience prerequisite to his telling stories.
Thus it is no wonder that Newman and Gayton would misrecognize their text artifacts
as “the stories” rather than viewing them, as we might today, as e ntextualizations of
embodied performances of verbal art (Bauman 1975; Kroskrity 2009).
This is remarkable since in both Yokuts and Western Mono narratives there is
a very detectable pattern of textual cohesion created by repeated use of initial lexi-
cal items meaning “and” and “and then.” While Euro-American scholars in the
pre-ethnopoetic period found little value in this type of repetition, it is clear that tra-
ditional storytellers, operating within a different discursive regime, regarded it as an
authenticating feature of proper performance.
Western Mono storytellers employ a similar organization for creating a basic
pattern of sentences linked by parallel use of the initial onnoho-yaisi “and then.” The
examples below, (7–11), are taken from the story “Coyote and the Moles” performed
in 1993 by Rosalie Bethel and audio and video-recorded by me at that time (Kroskrity,
Bethel & Reynolds 2002).
(7) Onnoho yaisi onnoho miya-t, niimi-boyo-naapaa miya-t.
Then and then go-tns Indian-trail-along go-tns
And then he went; he went along the Indian trail.
(8) Onnoho yaisi qwena’a-diya miya-t.
Then and far-also go-tns
And then he also went far.
(9) Onnoho yaisi na’mihoowi-t.
Then and tire-tns.
And then he got tired.
(10) Onnoho yaisi onnoho paya-ibo’ huu’i-di
Then and then water-emph flow-tns
And then a little river was flowing there.
(11) Onnoho yaisi mannoho paya-na hibi-kus sunawi-t.
Then and there water-obl drink-while think-tns
And then he thought that he would drink water there.
Narrative discriminations in Central California 
This passage occurs early on in the story as Coyote is introduced and represented as
waking up in the spring after his winter hibernation. Though Rosalie Bethel uses this
as her preferred linkage between story clauses she also displays two instances of varia-
tion in the same story, each building off of this basic pattern to create a meaningful and
dramatic variation. In one of these, there is a stanza of four clauses with the first three
all displaying the usual onnoho-yaisi. These sentences describe Coyote lying down and
falling asleep. But in the fourth sentence the initial mowaho “now” is used to surprise
the hearer as Coyote is abruptly and mysteriously woken up.
Later in the narrative this alternation is exploited not to create the “single-effect”
surprise but as a temporary replacement of “and then” with “now” to create an immedi-
acy and intensity that is highly appropriate for that point in the narrative. This intensity
is used in describing a race between Coyote and Mole that is just reaching its climax.
In this passage, (Examples 12–16), Coyote is tripped up by some tree roots as he nears
the finish line, finally manages to rise up and run toward the finish line only to witness
Mole popping up across the finish line. The patterned alternation in sentence-initial
time adverbs is thus similar to the role of tense variation in English language narratives
in which the switch from simple past to conversational historical present would be used
to delineate an episode and/or emphasize its immediacy (Schiffrin 1981).
(12) Mowahu iweehu hani’isU sumaiqaabina-t

Now here why confuse-tns
Now he became confused here.
(13) Tipudi’i-wai-n tsinipipoosa’ipa’i-t.
escape-fut-sub flip-tns
In trying to get free, he flipped over and over.
(14) Pidisi-yaisi mowa winikiya-t.
finally-and now stand-tns
And finally he stood up
(15) Mowa onnohO miya-t.
Now then go-tns
And now he went.
(16) Mowahu yaisi a-tiwadiqa-qwee mowahu pitihu-gaa-wai-s miyu-tsi’
Now and it-finish-at now arrive-go-fut-as mole-dim
“AAAA, nii ti-ponaa-t!,” inee-t.
“AAAA, I it-win,” say-tns
And now just as Coyote was approaching the finish line, Mole said,
“AAAA, I won!”
To summarize the repetition critique, I think it is clear that Newman and Gayton
appear to be assuming a narrative ideal that approximates Western academic literacy
 Paul V. Kroskrity
conventions of varying sentence organization. This aesthetic is ethnocentric both in

seeming to prescribe a specific literacy convention as an ideal for narratives from the
oral tradition and in its failure to consider alternative aesthetic principles. I think it
is reasonable to contend in the Yokuts example that repetition – as dramatically per-
formed – would approach the parallelism we find in rhetorically powerful speeches or
in poetry. For both Western Mono and Yokuts storytellers, what Gayton and N ewman
called “repetition” was viewed as an essential trope of traditionalizing the narratives.
However offensive to the sensibilities of the dominant society, this structural rep-
etition was approvingly regarded as an authenticating trope by Natives. The Mono
example also demonstrates that this use of repetition was not invariant but rather sub-
ject to narrative strategy and the need to create discursive immediacy or distinguish
episodes within the narrative. Far from artless, repetition in these Central Califor-
nian indigenous narrative traditions, was often deployed in accord with culturally and
linguistically recognizable patterns. The Gayton-Newman critique seems to think it
unnecessary to consider indigenous notions of genre and performance and, lacking
these, imposes a model of narrative evaluation from the dominant society.
5. The lack of explication
My second critique of Gayton and Newman concerns their claim of a supposed lack
of explication in Western Mono. Both Gayton and Newman find Yokuts and Western
Mono narratives lacking in regard to the inclusion of details regarding characters and
cultural background. Newman and Gayton (1964[1940]: 378–379) finds that there is:
no tendency to be explicit in regard to character or cultural features. If an episode

has to do with several characters, the speaker or actor in each instance is not
always named. The listener is expected to know who is talking or acting, for it is
apparently assumed that he is already familiar with the details of the story.
This resembles an appeal to a notion akin to a more recent educational concept of

“restricted code” (Bernstein 1977) in which there is a strong cultural preference for
relying heavily on taken for granted, shared background knowledge. On this point,
the Gayton-Newman observation of difference is indisputable but I do reject their
uncritical acceptance of the hegemonic normativity of what Jim Collins (1996, 2009)
calls “schooled” literacy conventions and their failure to understand this practice
within the context of Mono storytelling discourse. Certainly one of the reasons why
storytellers sometimes do not specify who is speaking is that performance features
like voice quality and facial expression often disambiguate such information. Beyond
such p erformance features, Keith Basso (1990: 152–153) has eloquently described an
Apache narrative aesthetic in which it is very inappropriate to tell people what they
Narrative discriminations in Central California 
already know or can readily figure out. Based on interviews with Mono elders, I must
also conclude that a similar value appears to have prevailed among the Western Mono
since narrators did not typically provide morals or evaluative coda-type conclusions or
even explanations of apparent incongruous endings.
One such story typically told without explanation is the story “Coyote Races
Mole” – a story of a race in which the diminutive Mole somehow beats his much
larger, and presumably faster, rival. I have personally heard five performances of this
story over the course of two decades of field work in Mono communities and only
one of those performances contained an explanatory ending. Rosalie Bethel (North
Fork Mono) and other elders explained to me that endings containing moral conclu-
sions were not customarily provided for two reasons. One, they were not provided
to show respect for those audience members who knew the details (and would be
offended by being told something they already knew) and, two, to compel children
(and other novices) who were hearing it for the first time to ask their parents or
older siblings. In earlier times, Western Mono narrators embedded their storytell-
ing practices into a larger metanarrative discourse of “talk about stories.” Children
who did not understand how Coyote could be beaten in a race by Mole could be
“home-schooled” by their parents, older siblings, or other socializing agents by ask-
ing questions and getting answers in sidebar interactions. In comparing contem-
porary practices with those of older times and ultimately in rationalizing her own
innovation of an explanatory coda, North Fork Mono elder Rosalie Bethel said,
“Nowadays we can’t rely on [families to explain stories] so that maybe our stories
will have to tell it more completely since the children cannot always find someone
to ask.” This is what motivated her to include the explanatory coda below in her
1993 performance – a performance that amounted to a recontextualization of a story
that was performed as a model indigenous-language pedagogical discourse before a
collection of assembled North Fork Mono elders. At this event, Rosalie Bethel per-
formed this story as a demonstration to other elders of how Mono language stories
might be adapted to the needs of younger audience members who lack either prior
knowledge of the story or a personal network that connects them to an informed
elder capable of explaining that story. A more complete description and analysis
of that recontextualized telling of the story is beyond the scope of this article but is
available to interested readers in (Kroskrity 2009).
In the following passage (17–26), Rosalie Bethel takes a page from Bakthin’s
(1986: 62) playbook in her creation of an instance of what he would call a “secondary
genre” of conventional storytelling via her intertextual linking of story with metanar-
rative commentary. In her performance, a brief pause occurs between the conclud-
ing action of the narrative and the explanatory code below. As noted below, she leans
toward the audience available through the camera (and the video it will ultimately
produce) as if to provide confidential information.
 Paul V. Kroskrity
(17) Uhu miyu-tsi’ nihi tisumiya-daa-pI.

that mole-dim very think-hab-compl
[Leaving narrative posture, leans forward] Contrary to expectations, that
little mole had been very thoughtful.
(18) Uhu miyu-tsi’ ti-poso’o-hotU yaduha-s ihi sunawi-t.
that mole-dim own-friend-with talk-sub he thought
That little mole had talked with his friends and planned it.
(19) “Ii-bo’ ‘a-wiya!,’ a-inee-s, ii-bo’ iga-gaa-wai”
he-emph imp-go, him-say-sub you-emph enter-go-fut
“When he says, ‘GO!,’ you enter (the burrow).”
(20) “Taaqwa-bo’ yaisi simi’-a a-na-wadiqa-qwee-dugu a-digi’i-wai.”
We:incl-emph and one-obl it-pas-finish-loc-through him-place-fut.
“And we will place another one across the finish line”.
(21) “Mowahu yaisi isa’ pitihuu-gaa-wai-s ii-bo’, ‘Nii
Now and coyote arrive-go-fut-sub you-emph ‘I
ti-wusu’a-t’ inee-wai.”
it-win’ say-fut.
“And now just as Coyote approaches (the finish), you will say, ‘I won.’”
(22) Taaqwa-bo’ nasimi-tU miyu simi’i-nisU sunawi-di.
We:incl-emph all-subj mole one-like appear-prog.
“We moles all look alike.”
(23) Qadu’u yaisi uhu isa’ sutabihi-duwa-t.
neg and that coyote know-can-tns
“And that Coyote cannot figure it out.”
(24) Uwamaqahuu ihiwi ti-ponaa-t.
In:this:way they it-win-tns
That’s how they won.
(25) [Audience (off-screen)] Ano’o[tU.
(It is) enough
That’s all.
(26) Onno a-na-wadiqa-nU.
then its-pas-finish-nom
Then, it’s finished.
In this passage, Rosalie Bethel explicates how the Mole, or actually a team of moles, won
the race. By using more than one mole they are able to position a second mole, across
the finish line. That mole merely has to wait for Coyote and anticipate his c rossing the
finish line. Though she could have conveyed this entirely in the narrator’s voice that
she begins with in lines (19–23), she artfully selects to perform part of the explanation
Narrative discriminations in Central California 
in previously unheard dialog with appropriate iconic gestures about the placement of
the mole and even a replay of the victory pose struck by the “winning” Mole.
What Rosalie Bethel’s member-analysis, as well as her innovation of explanatory
story coda suggest is that Gayton and Newman’s perception of a “lack of e xplanation”
in what Bakhtin would call the “simple” traditional narrative is actually the failure
to recognize how Western Mono ideologies of intertextuality (Bauman 2004) shaped
these narratives – by not recognizing either their recipient-designing for c hildren,
their oral performance orientation, or their embedding in Mono metanarrative
discourses. By erasing these and other connections to Mono and Yokuts social life
through decontextualizing the resulting text artifacts and failing to adequately
appreciate indigenous understandings of aesthetics and intertextuality, Gayton and
Newman found in their own unexamined, literacy-based and Western literary biases
toward verbal art a readily available basis for invidious comparison and ethnocentric
judgment.
6. Relativism or covert racism: The case for covert racism
For Gayton and Newman, while the role of their own schooled literacy practices is
particularly apparent in pre-structuring their “expectations” about narrative form
and content, it provides only a partial account for their apparent failure to appreci-
ate the aesthetics of the narratives of indigenous Central California. Similarly, then
current professional ideologies of language, largely tied to Boasian models of culture
(Bauman & Briggs 2003: 257–282) and derivative “acculturation” theory (Linton 1940;
Kroskrity 2000) deterred these scholars from either attempting to study narratives in
more naturalistic contexts or from producing materials that might be more recipient-
designed for community consumption. Those theories represented indigenous cul-
tures as delicate, uniform and unified wholes and pathologized culture contact and
change as disintegrative and destructive. Anthropological theory could then ratio-
nalize further marginalization and erasure of indigenous peoples by declaring them
“acculturated” and projecting their complete assimilation into the dominant society.
Anthropological linguistic methods did not need to involve ethnographic investiga-
tion because there was really no “authentic” indigenous culture to observe. But though
such debilitating assumptions might apologize for or excuse specific instances of mis-
interpretation, the consistent and totalizing pattern of negative evaluation seems to
require either an alternative or a supplementary explanation.
I would like to suggest that the profusion and consistency of negative characteriza-
tions, the unwillingness to look for alternative cultural pattern for anomalous traits,
and the satisfaction with an empty relativism are underlain by something analogous to
what Jane Hill (2008) has termed “covert linguistic racism”. As in the case of Hill’s mock
Spanish examples – there is nothing explicitly racist in either the single use of Spanish
 Paul V. Kroskrity
words like “macho” or “mañana” or in any single negative evaluation but rather in the
enveloping pattern in which one finds a consistent, pejorative design. As Hill (2008: 150)
indicates, such usages among members of the dominant society presuppose stereotypes
of Spanish speakers as lazy, irrational, immoral, and/or otherwise culturally “other”. I do
not want to suggest that the scholarly evaluations of Gayton and Newman are formally
equivalent to the production of mock Spanish lexical forms but I do want to want to
suggest that both patterns of consistently negative evaluation similarly rely on indexical
connections to stereotypes. In the case of Spanish, the consistent p attern of pejorative
usage by non-hispanics contributes to a covert racist project of creating hierarchies and
of constructing racial “others” who lack the cultural virtues (and political-economic
standing) dominant groups associate with “Whiteness” (Hill 2008).
But racism and analogous efforts to exclude do not simply follow a singular
model. Etienne Balibar (1991), in his discussion of the interaction of nationalism and
racism, has suggested the value of thinking about a “spectrum of racisms.” The species
of racism directed at California Natives is likely to have a different character in part
because of the political economy that underlies intergroup relations. Though there are
aspects of linguistic racism such as the development of the epithet “digger” ( Hinton
1994: 165–179) – to refer to California Indians who practiced traditional hunting and
gathering economies – that suggest similarity to the racisms directed against African-
Americans and others, the racism directed at California Indians has some expected
differences. Not surprisingly, people who have experienced the brutality of settler
colonialist (Wolfe 1999) adaptations are represented by different types of stereotypes.
As Philip Deloria has suggested, such stereotypes often include primitivity and “inevi-
table disappearance” (Deloria 2004: 10). Settler colonialist nations have a special need
to rationalize their displacement and dislocation of indigenous communities. The
construction of such tropes both racializes California Indians and rationalizes their
dispossession, marginalization, and erasure. These tropes work in a similar way to
what Barbra Meek (2006, and this volume) has called “Hollywood Injun English” – a
stereotypical form of linguistic representation designed by the culture of the dominant
society to subvert the “indigenousness” of Native Americans by making them appear
foreign. Such stereotypical images both subordinate cultural others and invalidate
their claims to indigenous identities.
I want to conclude by highlighting two issues – linguistic racism and narrative dis-

crimination. Were Gayton and Newman merely attempting a contrastive presentation
of narrative norms in an attempt to explore “narrative relativity” or were they con-
tributors to a project of linguistic racism directed at California Indians? As I have
Narrative discriminations in Central California 
detailed above, there is ample evidence to say yes to both questions. Clearly Gayton
and Newman appear to have seriously attempted a relativistic appreciation of cultural
differences in narrative style and certainly at the time they wrote they lacked the ben-
efit of working with breakthrough theories of language use, performance, intertex-
tuality, and language ideologies. But the consistency and comprehensiveness of their
negative evaluation of California Indian narratives strongly suggests the operation of
a totalizing pejoration consistent with the kind of covert linguistic racism that would
be directed at Native Americans, the kind that indexes “inevitable disappearance” and
further promotes their marginalization and erasure.
As for narrative discrimination, I think it is an especially appropriate tool for
understanding how linguistic and cultural experts – ones who are overtly advo-
cates of the languages/cultures they describe – can be recruited to participate
in racializing projects that are much larger than their individual contributions.
While Hymes’s narrative inequality does a significant job in calling attention to the
way narrative difference is managed, (re)produced, evaluated and institutionally
inscribed, discursive discrimination may play a further role in emphasizing the
inevitability of imposing standards and the ease with which attempts to appreci-
ate the narrative conventions of others’ are saturated with discursive expectations
and evaluations, typically located at the level of practical consciousness, that often
prevent either an informed understanding or a constructive representation. In the
case of Newman and Gayton, the role of their own schooled literacy practices is
particularly apparent in pre-structuring their “expectations” and aesthetic feelings
about narrative form and content. Academic scholars may be experts capable of
focusing a bright light on the limited regions of their expertise – whether it be
verbal morphology or comparative myth motifs – but they are elsewhere common-
sense social (i.e. national) actors who are likely to (re-)produce familiar cultural
patterns stored in their practical consciousness. Narrative discrimination may
help us as scholars to further illuminate the ease with which ethnocentrism and
social hierarchy enter into overtly “simple” acts of aesthetically appreciating stories
from other c ultures. Narrative discrimination can help us understand how people
who are decidedly non-racist can participate in racist projects as an unintended
consequence of aesthetic appreciation.
References
Bakhtin, Mikhail Mikhailovitch. 1986. Speech Genres and Other Late Essays, Caryl Emerson &
Michael Holquist (eds), Vern McGee (transl.). Austin TX: University of Texas Press.
Balibar, Etienne. 1991. Racism and nationalism. In Race, Nation, Class: Ambiguous Identities,
Etienne Balibar & Immanuel Wallerstein (eds), 37–67. London: Verso.
 Paul V. Kroskrity
Basso, Keith H. 1990. Western Apache Language and Culture. Tucson AZ: University of Arizona
Press.
Bauman, Richard. 1975. Verbal art as performance. American Anthropologist 77: 290–311.
Bauman, Richard. 2004. A World of Other’s Words: Cross-cultural Perspectives on Intertextuality.
Malden MA: Blackwell.
Bauman, Richard & Briggs, Charles L. 1990. Poetics and performance as critical perspectives on
language and social life. Annual Review of Anthropology 19: 59–88.
Bauman, Richard & Briggs, Charles L. 2003. Voices of Modernity: Language Ideologies and the
Politics of Inequality. Cambridge: CUP.
Bauman, Richard & Sherzer, Joel. 1974. Explorations in the Ethnography of Speaking. Cambridge:
CUP.
Bernstein, Basil. 1977. Class, Codes, and Control, Vol. 3: Towards a Theory of Educational
Transmissions. London: Routledge & Kegan Paul.
Bloomfield, Leonard. 1933. Language. New York NY: Holt.
Boas, Franz. 1894. Chinook Texts. (Bureau of American Ethnology, Smithsonian Institution).
Washington, DC: USGPO.
Boas, Franz. 1918. Kutenai Tales. (Bureau of American Ethnology, Smithsonian Institution)
Washington, DC: USGPO.
Boas, Franz. 1928. Keresan Texts. New York: American Ethnological Society.
Bourdieu, Pierre. 1984. Distinction: A Social Critique of the Judgement of Taste. (Translated
Richard Nice). Cambridge, MA: Harvard University Press.
Boyer, Ruth M. 1978. Anna Hadwick Gayton (1899–1977). Journal of American Folklore 91:
834–41.
Briggs, Charles L. & Bauman, Richard. 1992. Genre, intertextuality, and social power. Journal of
Linguistic Anthropology 2: 131–172.
Chomsky, Noam. 1957. Syntactic Structures. The Hague: Mouton.
Chomsky, Noam. 1965. Aspects of a Theory of Syntax. Cambridge MA: The MIT Press.
Collins, James. 1996. Socialization to text. In The Natural History of Discourse, Michael
Silverstein & Greg Urban (eds), 203–228. Chicago IL: University of Chicago Press.
Collins, James. 2009. The place of narrative in human affairs: The implications of Hymes’s
Amerindian work for understanding text and talk. Text and Talk 29: 325–345.
Deloria, Philip J. 2004. Indians in Unexpected Places. Lawrence KS: University of Kansas
Press.
Gamble, Geoffrey. 1994. Yokuts Texts [Native American Texts Series]. Berlin: Mouton de
Gruyter.
Gayton, Anna H. & Newman, Stanley S. 1940. Yokuts and Western Mono Myths [Anthropologi-
cal Records 5: 1]. Berkeley CA: University of California Press.
Gayton, Anna H. 1935. Areal affiliations of California folktales. American Anthropologist 37:
582–599.
Gayton, Anna H.1948. Yokuts and Western Mono Ethnography. University of California
Anthropological Records, 10: 1–302. Berkeley: University of California Press.
Gifford, Edward W. & Block, Gwendoline Harris. 1990 [1930]. California Indian Nights. Lincoln
NB: University of Nebraska Press.
Gumperz, John J. & Hymes, Dell H. 1972. Directions in Sociolinguistics: The Ethnography of
Communication. New York NY: Holt, Rinehart and Winston.
Harrison, K. David. 2007. When Languages Die: the Extinction of the World’s Languages and the
Erosion of Human Knowledge. Oxford: OUP.
Narrative discriminations in Central California 
Hill, Jane H. 2001. Mock Spanish, covert racism, and the (leaky) boundary between public and
private spheres. In Languages and Publics: The Making of Authority, Susan Gal & Kathryn
A. Woolard (eds), 83–102. Manchester MA: St. Jerome.
Hill, Jane H. 2008. The Everyday Language of White Racism. Malden MA: Wiley-Blackwell.
Hinton, Leanne. 1994. Flutes of Fire: Essays on California Indian Languages. Berkeley, CA:
Heyday Press.
Hymes, Dell H. 1964. Language in Culture and Society. New York NY: Harper.
Hymes, Dell H. 1981. “In Vain I Tried to Tell you”: Essays in Native American Ethnopoetics.
Philadelphia PA: University of Pennsylvania Press.
Hymes, Dell H. 1996. Ethnography, Linguistics, Narrative Inequality: Toward an Understanding
of Voice. London: Taylor and Francis.
Judson, Katherine B. 1994[1912]. Myths and Legends of California and the Old Southwest.
Lincoln NB: University of Nebraska Press.
Krauss, Michael. 2007. Classification and terminology for degrees of language endangerment.
In Language Diversity Endangered, Matthias Brenzinger (ed.), 1–8. Berlin: Mouton de
Gruyter.
Kroskrity, Paul V. 2000. Language ideologies in the expression and representation of Arizona
Tewa ethnic identity. In Regimes of Language: Ideologies, Polities, and Identities. Santa Fe
NM: School of American Research.
Kroskrity, Paul V. 2009. Embodying the reversal of language shift: Agency, incorporation and
language ideological change in the Western Mono Community of Central California. In
Native American Language Ideologies: Beliefs, Practices, and Struggles in Indian Country,
Paul V. Kroskrity & Margaret Field (eds), 190–210. Tucson AZ: University of Arizona
Press.
Kroskrity, Paul V. 2012. Telling Stories in the Face of Danger: Language Renewal in Native
American Communities. Norman OK: University of Oklahoma Press.
Kroskrity, Paul V. & Reynolds, Jennifer F. 2001. Using Multimedia in Language Renewal:
Observations from Making the CD-ROM TAITADUHAAN: WESTERN MONO WAYS
OF SPEAKING. In Green Book of Language Revitalization, Kenneth Hale & Leanne H inton,
(eds), 312–325. New York: Academic Press.
Kroskrity, Paul V., Bethel, Rosalie & Reynolds, Jennifer F. 2002. Taitaduhaan: Western Mono
Ways of Speaking (CD-ROM). Norman OK: University of Oklahoma.
Linton, Ralph. 1940. Acculturation in Seven American Indian Tribes. New York NY: D. Appleton-
Century Company.
Meek, Barbra. 2006. And the injun goes how!: Representations of American Indian English in
(white) public space. Language in Society 35(1): 93–128.
Newman, Stanley S. 1944. Yokuts Language of California [Viking Find Publications in Anthro-
pology 2]. New York NY: Viking Fund.
Newman, Stanley S. & Gayton, Anna H. 1964[1940]. Yokuts Narrative Style. In Language in
Culture and Society, Dell H. Hymes (ed.), 372–82. New York NY: Harper.
Nichols, Johanna. 1999. Linguistic Diversity in Space and Time. Chicago: University of Chicago
Press.
Schiffrin, Deborah. 1981. Tense variation in narrative. Language 57: 45–62.
Silverstein, Michael. 1989. Singularly like our ideal of a scientist. In General and Amerindian
Ethnolinguistics: In Remembrance of Stanley Newman, Mary Ritchie Key & Henry M.
Hoenigswald (eds), 45–63. Berlin: Mouton de Gruyter.
 Paul V. Kroskrity
Tedlock, Dennis. 1983. The Spoken Word and the Work of Interpretation. Philadelphia PA:
University of Pennsylvania Press.
Whorf, Benjamin Lee. 1939. The relation of habitual thought and behavior to language. In
Language, Culture, and Personality, Leslie Spier (ed.), 75–93. Menasha WI: Sapir Memorial
Fund.
Whorf, Benjamin Lee. 1940 Science and linguistics. Technology Review (MIT) 42: 229–231:
247–248.
Wolfe, Patrick. 1999. Settler Colonialism and the Transformation of Anthropology. London:
Cassell.
The voice of (White) reason
Enunciations of difference, authorship,
interpellation, and jokes
Barbra A. Meek
University of Michigan
This chapter argues that the power of “covert racist discourses” lies in the
obscurity of authorship and the interpellation of readership along with the
tacit preconditions of their enunciation. Drawing on Jane H. Hill’s concern
with the practices of enunciation (2008), this chapter explores the ways in
which conceptualizations of difference and unity are enunciated beyond clearly
defined institutional domains. It analyzes the semiotic elements deployed in
electronically-circulating jokes with American Indian characters and shows how
such jokes re-inscribe tropes of conquest. Furthermore, the discourse emanating
from such characterizations maintain a particular type of citizen as quintessential
and perpetuate the already difficult struggle people of color, especially American
Indians, face with respect to recognition, legitimation, and citizenship in “White”
domains.
Keywords: racist discourse; discourse analysis; language ideology
“…. Covert racist discourses have authors, influential Whites who appropriate
new linguistics resources from ways of speaking associated with people of color,
and reshape these to serve their own purposes.” – Jane H. Hill (2008)
The language of “White racism” and the racialization of language have flourished
in political-economic contexts where the evaluation of difference has had material
consequences for the emerging nation-state and its citizenry. An exemplary case in
point is the history of rhetoric depicting American Indians in the United States and
Canada and the corresponding “Indian” problem (Berkhofer 1978; Deloria 1998,
2004; Garroutte 2003). Despite efforts to remedy such rhetoric, the maintenance of
racialized differences through language endures in such “settler” contexts because
of the underlying economy of persons, bodies, and attributes (cf. Hall 1986). Such
economies are driven by the need to recognize difference and establish unity (the
twin projects of distinction and discrimination) in order to create and to maintain
 Barbra A. Meek
a particular regime of value, or status quo, the maintenance of which happens

through subtle enunciations of difference/unity and covert forms of authorship (cf.
de Certeau 1984).1
As Jane Hill succinctly expresses above, “covert racist discourses” are authored by
privileged individuals with resources, networks, and institutions for maintaining their
entitlements (authorship) in part through the on-going disenfranchisement of people
of color through appropriations and modifications of associated speech styles and
semiotic registers (2008: 177).2 Regimes of value gain currency and endure through
the institutionally-managed, state-sanctioned practices of authorship/authoring.3
The audience invested in these practices presupposes a schema of difference covertly
articulated through such rhetoric and offers interpretive rubrics when other audience
members fail to recognize them (the schema). At the same time, such rhetorics of dif-
ference/discrimination entailed in these practices of authorship circulate strategically
within and beyond the authored context, seeping into other everyday, parochial texts.
This seepage begins to blur the origins and attribution of authorship. Social
theorists and anthropological scholars have recognized this seepage and b lurredness
by extending analytic attention from the discrete, individualized act of textual and dis-
cursive creation to the sociality of such projects and their institutional hubs.4 Jane Hill
in particular has theorized this seepage in terms of “voice” and “mock-ness” (Hill 1995,
2001, 2008), emphasizing individual agency in the construction, maintenance, and
transformation of such regimes. However, authorship – and origins – cannot always
be neatly discovered or unpacked, especially in relation to genres of folklore that have
circulated and continue to circulate beyond the moment of inception. New mediums
. A la Hill (2001: 80), I’m using “covert” to mean “actively suppressed, hidden and
inadmissible, ‘closeted’” rather than subconsciously “subtle” or “tacit” in a taken-for-granted,
everyday cultural practice kind of way.
. For a politically current example of such appropriation, see Newt Gingrich’s shifting
stance on Spanish that has been circulating over the internet via “The Daily Show” (Winter
2012; 〈http://www.thedailyshow.com/watch/tue-january-31-2012/indecision-2012 –
pander-express〉).
. Asif Agha’s recent work on “mediatization” acknowledges the circulation of semiotic
forms and their perduring significations, but he undertheorizes (as does Silverstein 1996) the
role of the author/producer (individuals) and the force or impact of such mediatized s emiosis
in the social and historical contexts of its circulation (2011a, 2011b; for an exception, see
Kockelman 2007 on agency).
. As “indexicality,” see Ochs 1990; Silverstein 1979, 2003; as “voice,” see Bakhtin 1981,
Bauman & Briggs 2003; Hill 1995; see also special issue of Journal of Linguistic Anthropology,
December 2011. For some institutional ethnographic examples, see Jaffe 1999; Meek 2010;
Morgan 2009; Richland 2008.
The voice of (White) reason 
of circulation (email, the internet) further confound and render opaque authorship
and individual agency in the maintenance of these regimes.5 However, drawing on
Hill’s insights regarding appropriation and reshaping, this chapter investigates the leg-
acy of racialized difference through linguistic resources in a particular folkloric genre
of English language media – jokes. To narrow the analytic abundance of this genre, I
focus on jokes that circulate electronically and racialize characters as A merican Indian.
The analysis draws out the political, economic and moral valences of the semiotic ele-
ments that monopolize on the tacit expectations of difference tied to these characters,
implied by and relied on by these jokes.
Similar analyses of difference have largely focused on institutionalized d iscourse
and imagery in contexts of education, politics, advertising, and bureaucracy (for
example, Agha 2011a, b; Foster 2008; Gal & Woolard 2001; Irvine & Gal 2000;
Mazzarella 2003). However, unlike such state-centered, market-driven domains of
discourse where “images of linguistic phenomena gain credibility when they create
ties with other arguments about aspects of aesthetic or moral life” (Gal & Woolard
2001: 3), the “credibility” of a joke relies on its ability to play with incongruity and
interrupt such aesthetic and moral expectations (Oring 2003), resulting in laughter or
at the very least a smile. While this ability often relies on some presupposition(s) about
“aesthetic or moral life” associated with the semiotic personification of some charac-
ter (in this case, American Indians), this process of semiotic extension is not merely
fractal recursion (Irvine & Gal 2000), the mapping of difference from one domain or
context onto another in a reductive-like fashion. The semiotic processes at work in
joking depend upon the sociological and ideological proclivities of the audience or
individual reader. For jokes to succeed, certain mappings, or a multitude of mappings,
need already to have happened so that jokes may then furiously unravel them or vig-
orously reinforce them – depending on the recognition of incongruity and thus the
reader.
As with any text, some reader/audience mediates its existence (interpretation,
circulation, and to some extent authority). Jokes seem especially vulnerable to this
mediation in part because their effectiveness (interpretation and circulation) depends
upon their efficiency of expression (along with an element of surprise or unexpected-
ness). Efficiency in this case means maximal signification with minimal lexification,
and thus a substantial reliance on the indexical connections presupposed and entailed
by the content words of a joke’s text. Failure to realize all relevant indexical c onnections
. More insidious is the remediatization of powerful discursive events. Such remediati-
zations reframe, disrupt, and neutralize their discursive framing, and message, in order to
preserve dominant “White” rhetorics (for an analytically precise example, see Cole & Pellicer
2012).
 Barbra A. Meek
may result in a failed joke. Even in indexical fruition, a joke may fail because the
social backdrop of its signification is not sufficiently entailed or the reader is insuffi-
ciently impressed (or is sufficiently offended). Gal and Woolard (2001: 4) additionally
point out that “representations of language phenomena gain social authority … from
the institutional locations from which their proponents speak.” Jokes circulating
electronically remain partially unmoored in this case; their institutional “location”
shifting, their (original) authorship obscured, and their “authority” distributed and
dispersed. A joke’s social authority, or provenance, emerges in relation to the con-
texts of its circulation, or lack thereof, and only in relation to its interpellative efficacy
(its successful hailing of like-minded, reasonable and humor-full individuals) and its
capacity to motivate a reader/audience to carry it forward, perhaps even virally.
The interpellation of the joke’s reader, subject to the ideological framing of the
text and current socio-political context, encourages or discourages circulation.
Are we affronted? Ashamed? Unexpectedly delighted? The ideological mapping of
persons/traits/patterns conditions our response. As Althusser discussed, “ideology
‘acts’ or ‘functions’ in such a way that it ‘recruits’ subjects among the individuals… or
‘transforms’ the individuals into subjects by that very precise operation […] interpel-
lation or hailing” (1971: 174). To “get” a joke, a reader must be open to its ideological
framing and interpellative effect. If an individual is insufficiently “hailed” by a joke, she
might be less inclined to participate in its circulation.
Thus unlike other forms of media that gain legitimacy and authority through the
roots of their authorship and institutionality, jokes gain value through the context(s)
and networks of their circulation. Their circulation, or recontextualizability (cf.
Briggs & B auman 1992, 1999), endures through their adaptability (to changing
social circumstances) and strategically (or ambiguously) valenced interpellation(s) of
readership. Unexpected junctures and disjunctures – “acceptable incongruities” – keep
jokes alive.
Additionally, a joke’s textual malleability and contextual adaptability will
contribute to its circulation, or its demise as spam/trash. When jokes resonate with
readers/hearers, and thus circulate, they create a network of participants who share
in some mutually intelligible social frame. “Humor allows us to create, exchange, and
sustain various interpretive commentaries on social life by mobilizing shared frames”
(Paolucci & Richardson 2006: 335; see also Goffman 1974). In the mobilization of
frames through the often unremarkable strategies of obscured authors and inter-
pellated readers-circulators, the creation, exchange, and sustaining of certain well-
entrenched and hyper-salient elements and figures (sign-vehicles) remain intact and
effective as markers of difference and distinction. The social work accomplished by
such simplicity and efficiency is the focus of the rest of this paper. In particular, I
analyze the various ways in which jokes can “enunciate” difference and discursively
interpellate readers as both “native” and “non-Native,” obscuring potential offense
The voice of (White) reason 
through the intersubjective composition of an idealized, “normal” American citizen.6

This is especially true of the joke that anchors this paper. It begins like this,7
Three strangers strike up a conversation in the airport passenger lounge in

Bozeman, Montana, while awaiting their respective flights.
One is an American Indian passing through from Lame Deer, Another is a
Cowboy on his way to Billings for a livestock show, And the third passenger is
a fundamentalist Arab student, newly arrived at Montana State University from
the Middle East.
Their discussion drifts to their diverse cultures. Soon, the two Westerners
learn that the Arab is a devout, radical Muslim and the conversation falls into an
uneasy lull.
The cowboy leans back in his chair, crosses his boots on a magazine table and
tips his big sweat-stained hat forward over his face. The wind outside is blowing
tumbleweeds around, and the old windsock is flapping; but still no plane comes.
This joke first appeared in my email inbox in 2005, circulated through many inboxes
prior to mine and probably many hence. It arrived by way of a non-Native acquaintance.
The joke relies on historical dichotomies of difference (cowboy/Indian; “Western/
Eastern), underscored by the “strange(r)ness” of our three interlocutors above that
slides gracefully across a veneer of multiculturalism (a discussion of their cultural
diversity). Such enunciations reinforce old tropes of difference, resurrecting them in
new contexts. But why these characters, and why an American Indian in particular?
What “old” tropes frame this joke?
1. From “Red” to “Dead”: Indians in “White” public space
Media depicting Indians, or Native Americans, while pervasive, is oftentimes subtle:

a reference by Virginia Woolf to comparing children dancing on a beach to Native
Americans dancing elsewhere or the final line of a Dave Egger’s novel (again on a
beach, although somewhere in California) referring to the ancestral pedestrian t raffic
and the fading of their imprints, their history, and their existence from this same sandy
terrain. While such literary forms contain passing references to Indians, other textual
domains highlight and intentionally represent Indians in especially demeaning ways,
. The difference in capitalization of the “n” in “native” indicates the difference between
the unmarked form of “native” and the marked form, “Native” which indicates a political
difference between “native-born” Americans and “Native Americans” who are also often
“native-born.”
. This is only the first third of this joke.
 Barbra A. Meek
from racist declarations supporting the Redskins or Chief Illiniwek to humorous and
not-so-humorous greeting cards picturing Indians and Pilgrims in peaceful collusion
out in their natural environment, amidst a holiday parade (Figure 1). These images
portray Indians as ancient, fantastic, sometimes evil, often elderly, and frequently silent
(and silenced). In these portrayals, however, a limited, generic set of elements coordi-
nate our/the reader’s interpretation. These elements reflect and resurrect past tropes
for humor, irony, and ethos, especially in service to the construction and p romotion
of a national morality.
Figure 1. Greeting card of Thanksgiving Day parade
As this card portraying a Thanksgiving Day parade reflects, American Indians can
serve as the figure and the ground that defines “Americanness.”
The voice of (White) reason 
Several scholars have examined the history of American Indian representations in

media such as comic books and other literature (Berkhofer 1978, 1988; Green 1988a;
see also Strong 2004, 2005), advertisements (Green 1988b; M erskin 2001), and film
(for examples, see Aleiss 2008; Kilpatrick 1999; Marsden & Nachbar 1988; Rollins &
O’Connor 1998; Strong 1996). This scholarship reveals a dual or tripartite configura-
tion of personality in representations of American Indian identity, all in juxtaposition
to some “White” standard, as (1) “noble,” respectable yet vanquished, as (2) redeem-
able and partially assimilatable/adaptable, and as (3) inherently bloodthirsty, uncivi-
lized and irredeemable. As Bonnie Urciuoli has noted more generally (1996: 16), “At
the base of U.S. assumptions about ethnicity and race is the idea of the normative or
generic American, white, middle class, English-speaking…It stands in opposition to
all categories of …difference.” While particular iconic features attributed to Indians are
certainly juxtaposed with some normalized White populace and have remained in cir-
culation for several centuries, analysis of American Indian characterizations found in
internet jokes primarily, though not exclusively, reproduce images associated with the
“noble savage” and “ecological Indian” (cf. Oring 2003), underscored throughout by
an uncivilized, if not primitive, demeanor. As part of this characterization, the Indian
characters are also typically generic (no tribal affiliation specified), male, elderly, often
solitary, and linguistically different, to the point of “sounding” inarticulate and dysflu-
ent. This convergence of features indexes not only the character’s difference, the age
and singular presence might also underscore the prevalent belief in the disappear-
ance of Indians from an “American” landscape; it potentially signals the extinction of
Native Americans, though not necessarily through genocide (anymore) but merely an
individual lifespan.
The textual “artifacts” that are the focus of this paper are jokes circulating
electronically, by email and Internet.8 I focus on the opening joke in particular because
of the conceptual work that it accomplishes in only a few lines. It continues as follows,
Finally, the American Indian clears his throat and softly speaks, “At one time here,
my people were many, but sadly, now we are few.”
The Muslim student raises an eyebrow and leans forward, “Once my people
were few but now we are many. Why do you suppose that is?”
While the text of this joke is remarkable in several ways, most relevant to this analysis is
the clarity and directness with which this narrative laminates conventionalized features
of the American Indian character, the old enemy of the state, onto the Arab Muslim
. Scholarship on jokes covers a range of approaches, examining the structural elements
of jokes, the psychology of jokes, and the social and political functionality of jokes (Dundes
1987; Oring 2003; Rappoport 2005).
 Barbra A. Meek
student, the new enemy of the state, through the use of poetic parallelism across their
utterances.9 This lamination renders the “new enemy” interpretable, including the pre-
supposed conceptual baggage trailing the imagery of American Indians.
Figured in relation to a presupposed audience there are two general categories of
jokes involving American Indian characterizations, those directed toward a unmarked
reader and those penned for a Native reader. My focus is on the ambiguously authored
jokes for an unmarked readership. However, I will briefly remark on the latter in
relation to the former. A dichotomy between these two kinds of jokes becomes appar-
ent in several ways, most saliently through the framing of the Internet joke page and
its location. Natively-oriented websites often have an (auto)biographic page detailing
the website builder’s heritage, that is, Native heritage and the jokes themselves u
sually,
if not always, contain references to particular tribes and provide narratives about
everyday scenarios that a Native American audience is likely to have experienced and
find humorous (think Keith Basso’s Portraits of the “Whiteman” and the “Whiteman’s”
bumbling ways (1979). These sites are often embedded within other websites, such as
news sites, sites focused on cultural revitalization, and personal websites with links
to other Native American sites, including tribal websites. They are not found in web
archives of jokes nor do they appear to have been the object of analysis in terms of
“cycles” or themes by folklorists and literary scholars.
The jokes under investigation here could be categorized as “ethnic jokes,” but
Oring (2003) offers a more precise category: colonizing or frontier humor. Within this
category, he points out that few jokes exist involving American Indians and indigenous
people more generally (Oring 2003: 108). Of those that do exist, Oring notes that these
populations are portrayed as
“not hostile but naïve and unsophisticated [and] they also possess natural wisdom
and discernment. Despite or rather because of, the relative simplicity of their way
of life, they often see issues clearly. They are not deceived by the manners and
forms of colonial society and culture… They are portrayed as recognizing the
political and economic forces they are confronting”. (2003: 108)
The “frontier” jokes I analyze are found in web archives of jokes and/or joke-only
commercial sites, framed by pop-up ads for car insurance, travel, weight loss, credit
scores and the like. Some of these jokes denote authorship, but many are re-circulated,
recycled jokes that have longer histories than the sites or authors acknowledge and
are often re-circulated (and recontextualized) by email.10 That is, they form a “joke
. Silliman (2008) notes this pattern in relation to U.S. military metaphors.
. Episode of “Family Guy”, season finale (May 6, 2010; nineteenth episode of the eight
season), entitled “The Splendid Source” and based on a short story by Richard Matheson of
the same name (published in 1956). In both, the protagonists are searching for the origin of
the world’s dirtiest jokes and the (near) impossibility of establishing their origins.
The voice of (White) reason 
cycle.” I discuss this frontier category a bit more below, but for now suffice it to say that
these jokes fall into two basic sub-categories: jokes emblematizing Indians directly and
jokes employing Indian characters to emblematize some other (ethnically-indexed)
group, all of which plays against some “white” public. I analyze both kinds of jokes to
investigate the subtle strategies of “voice,” of privileged authorship, and the mainte-
nance of “white/American” status quo such jokes entail while they mask or erase the
traumatic, genocidal history of relations between the colonized and the colonizer in
North America.
2. And the Indian says to the White man: A note on method
To examine the representational repertoire deployed in jokes with American Indian

characters, I began by analyzing jokes I received by email and then by surveying jokes
on the Internet, searching “Indian” and “joke” (see Hill 2005 for a similar method-
ology). As expected, this search generated several pages of website links – 79 with
10 links per page (except for the final page), resulting in 784 links in all. Obviously,
many of these links led to jokes with South Asian or Southeast Asian Indian characters.
Approximately four links per page dealt with American Indians. With respect to the
sites themselves, most of them were and are public archival-like sites where authorship
is unknown, but a few sites from this search were smaller, more personal and less
commercial. While most of the jokes have unknown authorship, many of the jokes in
some way specify or denote the identity of the “submitter” of the joke to that site. Also,
in the more commercial sites, advertisements framed the content of the jokes and also
“popped up,” obscuring the reader’s view of the joke-text at times.11 Once a particular
title or typifying phrase was identified, I then searched particular phrases, resulting in
fewer pages of links wherein most, if not all, of the jokes used a similar joke template
with minor alterations to elements such as character referents, beverage consumed,
location, and degree of specificity of these referents.12
. I also uncovered a distinctive difference between these extremely commercialized joke
sites and geographically-locatable, personal business sites managed by individuals identi-
fying as Native American. Two major differences were the presence/absence of site-external
advertising and the presence/absence of tribe-specific tropes.
. This approach is analogous to “purposeful snowball sampling” methods. It is relevant to
this case because we are dealing with mediatization where the uptake and spread of jokes is
crucial to their role in mediating the everyday discourse of “White” racism.
 Barbra A. Meek
The jokes found on these commercial sites, as well as those sent to me by email
from colleagues, friends, and relatives,13 identify Native American characters with the
following generic terms: American Indian, Native American, Red Indian, and Indian.
Other phrases indexing American Indian descent were elders, tribe, and Chief. On
singular occasions specific tribal groups/nations were named (Blackfeet, Navajo,
Cherokee) – an example of which is given below. In general, most of the terms
remained generic and thus socially, culturally, historically, and linguistically unmarked
and ambiguous. Within the jokes themselves, several additional phrases contributed
to the interpretability of these terms and the characterological figure referenced by
them: geographical framing or setting of the joke (deserts and caves, North America,
reservation, Times Square, saloon), other characters in the joke (especially one or two
cowboys, a bartender), and style of speech (elements of “Hollywood Injun English”
(Meek 2006)).
These elements in conjunction with the narratives of the jokes evoke and
perpetuate some of the dominant tropes associated with Indians found across a
range of media genres noted above: the “noble” respectable yet vanquished Indian,
the partially adaptable and ecologically-minded yet primitive Indian, and the
blood-thirsty foreign Indian – all of whom are nearing extinction. Another signifi-
cant trope demanding of attention is the significant ambiguity that underlies these
portrayals, the ambiguity of citizenship. While American Indians were granted U.S.
citizenship in 1924 and have been reckoning their own tribal membership since tribal
governments were established under the IRA (Indian Reorganization Act (Wheeler-
Howard Act) of 1934), the duality – or presupposed duality – of their allegiance
presents an opportunity to portray Native Americans as non-citizens, as foreigners,
and thus enemies of the state. Additionally, the changing relationship between tribal
nations and the U.S. government complicates this duality further, juxtaposing ward-
ship and dependence (financial, institutional) – where American Indians have been
institutionally managed as wards of the state – with self-determination and (partial)
independence. The jokes to follow allow for an interpretation of tribal “dependence”
and individual incompetence over that of tribal “sovereignty” and individual self-
sufficiency while still potentially acknowledging federally-recognized tribes as
dependent sovereign nations and A merican Indians as contemporaries, all the while
hailing the victors (as it were).
. I received jokes from colleagues and students who were aware of my research interests.
The ones I received from friends and family, however, were forwarded for other reasons.
The voice of (White) reason 
3. “It’s so easy a [blank] could do it”: Modern-primitive iterability
Several jokes utilized many of these generic characteristics in order to allude to

or reference these dominant themes. One of the most pervasive is the modern-
primitive distinction. The “Me Want Coffee/Beer” joke effectively illustrates the use
of this theme. I received a version of this joke from a colleague of mine in F ebruary
2005. After searching this phrase on the internet in conjunction with two modifying
lexemes, Indian and joke, I uncovered 64 sites with the “coffee” wording and 14 sites
with the “beer” wording, for a grand total of 78 sites – the most salient differ-
ence across sites and jokes being degree of vulgarity. The jokes begin with a man
walking into a place. Depending on the beverage desired, the location might be a
café or a saloon, but in all cases the Indian character carries with him a shotgun and
something to target.
Version A: “Me want coffee.”14

An Indian walks into a cafe with a shotgun in one hand pulling a male buffalo with
the other.
He says to the waiter, “Me want coffee.”
The waiter says, “Sure chief, coming right up.”
He gets the Indian a tall mug of coffee. The Indian drinks the coffee down in one gulp, turns
and blasts the buffalo with the shotgun, causing parts of animal to splatter everywhere,
then just walks out.
The next morning the Indian returns. He has his shotgun in one hand pulling another
male buffalo with the other.
He walks up to the counter and says to the waiter, “Me want coffee.”
The waiter says, “Whoa, Tonto! We’re still cleaning up your mess from yesterday.
What was all that about, anyway?”
The Indian smiles and proudly says, “Me training for upper management position:
Come in, drink coffee, shoot the bull, leave mess for others to clean up,
disappear for rest of day.”
Version B: “Me want beer.”15

An American-Indian walks into a saloon with a shotgun in one hand and a 10-litre bucket
of manure in the other.
The Indian says to the bartender, “Me want Lager!”
The bartender says, “Sure, Chief, coming right up.”
. http://www.dwmbeancounter.com/acctjokes.html
. http://www.searchablejokes.com/jgov.htm;
http://www.jokebuddha.com/Shotgun/recent/10
 Barbra A. Meek
He then serves the Indian a tall glass of Tennents Lager. The Indian drinks it down in one
gulp, picks up the bucket, throws the manure into the air and blasts it with the shotgun. He
then walks out.
Five days later, the Indian returns. He has his shotgun in one hand and another bucket
of manure in the other.
He struts up to the bar and tells the bartender, “Me want beer!”
The bartender says, “Whoa there Chief, we’re still cleaning up
from the last time
you were here.”
“What was that all about, anyway?” he asked.
The Indian explained, “Me training for job as government employee. Drink beer,
shoot the shit, disappear for a few days, then come back and see if somebody else
has cleaned up the mess me left behind.”
While both of these jokes are conceivably poking fun at contemporary business or
management practices, they do so through the imagined body and voice of an Indian
character – a juxtaposition of the modern and the primitive for humorous effect. If
Indians were already conceptualized as modern or as MBAs, then the humor of the joke
would be diminished. This difference is marked linguistically as well. Here the Indian
character’s style of speech differs significantly from the speech of the bartender/waiter
character; the Indian speaks “broken” English – a style of English typical of H
ollywood
Indian characters (me as subject pronoun, deletion of articles, and omission of verb
copulas). This grammatical primitiveness subsequently indexes a social primitiveness,
now linked to American Indians. Underlying this analytic trajectory, an additional
indexical link that appears in dominant discourses about tribes and American I ndians
relates these primitive habits to a failure and inability to conform rather than a
choice not to adhere to dominant cultural norms and practices (Meek 2011; see also
Csordas 1999; Darian-Smith 2004; Mihesuah 1996; Strong 2005). The intellegibility
of the joke hinges on the subtly indexed, conventionalized primitiveness of the Native
American character. Consider importing a different ethnically-inflected character
(a Jewish or Asian-American character); does the joke fit?16 The efficacy of the joke
then is g enerated by the contrast and acceptable incongruity between a supposedly
uneducated, unrefined character in an educated, entitled position. The contradiction
between social persona and social position thusly elicits a chuckle by evoking the
image of an u nsophisticated, strange-talking savage as an upper-management execu-
tive and at the same time portraying an upper-management executive as a prehistoric
. As Debbie Cole has pointed out (p.c.), this joke will work with a range of marginalized
personae with some variations. Examples can be found on the Internet.
The voice of (White) reason 
imbecile.17 Through the figure of the Indian character, this incongruity allows us the
audience to safely undermine authority.
So why not swap out the Indian character in this generic joke with another symbol
of primordiality such as a caveman?18 Commercials certainly have drawn on the
caveman figure to avoid the protest and acerbic criticism that any other figure might
invite while reiterating yet blurring the distinction between modern and primitive.
However, unlike the Indian character in the above joke, the Geico commercials for car
insurance portray cavemen as modern contemporaries in all ways but one. The follow-
ing commercial opens with the announcer stating that applying for Geico insurance
is so easy “a caveman could do it (ha, ha, ha).”19 The scene then shifts to an upscale
restaurant where entrees such as roasted duck with mango salsa are served.
Announcer: Seriously, we apologize. We had no idea you guys were still around
(chuckles).
Caveman 1: Yeah, next time maybe do a little research.
Caveman 2: [snorts in disgusted agreement]
Waiter: Gentlemen, are we ready to order?
Caveman 2: I’ll have the roast duck with the mango salsa.
Caveman 1: I don’t have much of an appetite, thank you.
In this example, as well as throughout these commercials, the caveman c haracters speak
a rather standard variety of English, wear pants and collared shirts, use contemporary
contraptions such as moving walkways and airplanes without remark, play sports such
as tennis, baseball and bowling, sunbathe with girlfriends, and eat at upscale restau-
rants. In fact, the only elements signifying their primitive heritage are phenotypic –
extended brow ridge and hirsute appearance. Along with the incongruity between
appearance and performance, the apology dialogue above serves as a criticism of
“political correctness,” in other words, attempts to disrupt dominant representations
of people of color, if not American Indians in particular (note the line, “we didn’t know
you guys were still around”).
The incongruities in both the joke and the commercial work in relation to the
conventionalized personae that each joke requires as framing. The incongruity in
the “upper management” joke is the primitive Indian and the modern manager; the
incongruity in the caveman skit is the primitive appearance and the modern lifestyle.
Unlikely though this may seem, the Geico commercial might even be more profound
. Several colleagues have also pointed out the use of “American Indian” terms in corporate
culture, referring to meetings as “powwows” and directors as “chiefs.”
. The “cavemen” in Geico commercials performed a similar task.
. http://www.youtube.com/watch?v=iVvBXBZEhkw; March 11, 2012.
 Barbra A. Meek
than poking fun at those who make assumptions based on an individual’s appear-
ance (or political stance), suggesting through this series of advertisements that none
of us has ever been modern. Even so, the Indian persona – more so than the caveman
character – remains dysfluent, un-modern, and simple.
4. “The Indians are collecting firewood like crazy!”:

Primitive-incompetent motif
This theme of “backwardness” filters through many of these Indian jokes. However,
when comparing several versions of the next joke, not all iterations overtly emphasize
the Indian character’s incompetence; some jokes simply take this characteristic for
granted. This next joke – titled in some contexts the “cold winter” joke – was one that
I received over email in the Fall of 2001. It is a joke about an Indian character calling
the National Weather Service, and it can be found across (at least) 53 different joke
sites. Across these multiple iterations, the main character is the “Chief ” of a tribe, but
the name and location of the group are not always specified. This joke also has various
introductions that differentially juxtapose the primitive and the modern.20
Version A:
The Blackfeet (Native American Tribe) asked their Chief in autumn, if the winter
was going to be cold or not. Not really knowing the answer, the chief replies that
the winter was going to be cold and that the members of the village were to collect
wood to be prepared.
Version B:
It was October and the Indians on a remote reservation asked their new Chief if
the coming winter was going to be cold or mild. Since he was a Chief in a modern
society he had never been taught the old secrets. When he looked at the sky he
couldn’t tell what the winter was going to be like. Nevertheless, to be on the safe
side he told his tribe that the winter was indeed going to be cold and that the
members of the village should collect firewood to be prepared.
Version C:
The Indians asked their Chief, in the autumn, if the winter was going to be cold
or not. Not really knowing an answer, the chief replies that the winter was going
to be cold and that the members of the village were to collect firewood, to be
prepared.
. 〈http://www.weatherimages.org/wxhumor.html〉
The voice of (White) reason 
While the structure of these jokes remains consistent across versions, Version A
specifies a particular cultural or tribal entity, which suggests a particular geographic
location (perhaps somewhere in or near Montana). Version B and C, on the other
hand, simply identify the characters in the joke as Indians.
The rest of the joke continues as follows:
“Being a good leader, he then went to the phone booth and called the National
Weather Service and asked, “Is this winter going to be cold?” The man on the
phone responded, “This winter was going to be quite cold indeed.” So the Chief
went back to speed up his people to collect even more wood, to be prepared. A
week later he called the National Weather Service again, “Is it going to be a very
cold winter?” “Yes,” the man replied, “It’s going to be a very cold winter.” So the
Chief goes back to his people and orders them to go and find every scrap of wood
they can. Two weeks later he calls the National Weather Service again, “Are you
absolutely sure, that the winter is going to be very cold?” “Absolutely” the man
replies, “The Indians are collecting fire wood like crazy!”
These versions also explicitly construct the Chief as being not “traditional,” motivating
his need to call the National Weather Service from a phone booth, but also portraying
him as not quite modern because he used a pay phone. This disconnection from both
“traditional” and “modern” practices interdiscursively plays upon the generic image of
the “halfbreed” who is neither quite “Indian” nor “white” and therefore cannot fully
participate or know either culture (cf. Garroutte 2003; Mihesuah 1998; Trechter 2001).
He inevitably fails to be competent in any realm.
Version B illustrates most directly the necessary presupposing of this frame,
contrasting “new” and “old,” “remote” and “modern.” Because the new Chief had not
learned the “old” ways, he wasn’t able to predict the immediate meteorological future.
Instead, he resorted to contacting the National Weather Service for such information.21
The instrument for this task was a pay phone. Not only has the Chief character not
acquired the traditional knowledge of his ancestors, but he also hasn’t fully acquired
the modern trappings of Euro-American society – his own private phone.
As with the above “management” joke, the humor of this joke requires a famil-
iarity with the conceptualization of Native Americans as incompetent, primitive and
close to nature. Beginning with the fact that the National Weather Service “believes”
“Indians” to have greater predictive ability than their own meteorological equipment
and reinforced by the gathering of wood rather than the paying of gas bills to keep
warm for the winter, which would of course require a furnace. (How many teepees
. And, naturally, the NWS is equally incompetent, having focused on the habits of
“Indians” rather than their own meteorological instruments and training.
 Barbra A. Meek
do you know that can fit both a furnace and a campfire?) Drawing on this trope, an
advertisement for a Linux furnace states that “it’s like a tipi, no gates, no windows,
apache inside.”22
Figure 2. Furnace advertisement
These jokes rely on this primordial framing. These texts also underscore popular
ideas about contemporary Native Americans – as being incompetent in “both worlds,”
Native and non-Native alike (see Meek 2011 for discussion).
All of the jokes presented reify conceptions of Indians as the antithesis of an
idealized “American” citizen. As Toni Morrison (1992) and others have pointed out,
such racialized images are logical (and unremarkable) because their opposition to
some taken-for-granted idealized “whiteness” is exactly the counterpoint that r enders
them understandable. Yet the image of Native America defines America and the
. http://www.fmlftw.com/2010/11/26/linux-its-like-a-tipi-images/
The voice of (White) reason 
American citizen in a way that no other category of difference does. The racialized
image of an American Indian isn’t merely a counterpoint to “whiteness,” but a measure
of progress and success.
Of course the relationship between American Indians and U.S. progress has a long
history, beginning with imagery reflecting the emergence of the U.S. nation-state and
perpetuated through its industrial development and the defining of citizenship. With
respect to citizenship, progress and oppression go hand in hand. In some “White”
public spaces, the ideal citizen would be English-speaking and claim descent from
both Captain Smith and Pocahontas. He would be fully assimilated, yet appreciate
American Indian values. He would be the man Pratt had hoped to create at Carlisle
(Pratt 2004; see also Child 2000). He would be a symbol of modernity tempered with a
touch of environmental and spiritual awareness. In this way, Native American ancestry
has become uniquely valued and valuable as a defining feature of American citizen-
ship, personhood, and progress.
5. The “Arab” qua “Indian” problem: Laminations of difference
The jokes presented so far display a version of Indianness that continually

reconstitutes Indian subjects as remedial, partial, poorly adapted and un-adept. The
ability of American Indians to participate as responsible citizens is made available for
questioning through these frames and, unlike the cavemen of the Geico commercials,
they have remained primitive in contemporary times. Returning now to the joke in the
introduction, we see linguistic, generic, and visual cues being used to index p articular
essentialized ethnic attributes. The Indian’s speech is marked with three commas –
orthographically indexical of pauses and a slowed cadence. The Arab Muslim student’s
speech has zero commas, and the cowboy’s speech has one. The Indian character
doesn’t use contractions, reminiscent of a more oratorical style, and the statement itself
points out the extreme population decline suffered by all Native American groups. The
Arab student’s representation has no contractions as well, but is also void of colloquial-
isms and English slang. It is linguistically differentiable in its lack of distinctiveness. By
contrast, the cowboy character is most stylistically distinct not only in behavior and
remark, but he is distinguished through image as well. The photographic finale of a
cowboy is a crucial image in that it facilitates, or rather is, the punch line of the joke.
Returning now to the joke and its punch-line, recall that these three characters are
stranded at the Bozeman airport, waiting for their airplane, and so they engage in con-
versation, discussing “their diverse cultures” until the student character discloses that
he is a “devout radical Muslim.” At this point, the conversation lulls until the Indian
character notes “softly,” that “at one time here, my people were many, but sadly, now
we are few.” The student character responds, “Once my people were few, […] and now
 Barbra A. Meek
we are many. Why do you suppose that is?” Here we get a “quiet,” noble, elderly Indian
juxtaposed with a young Arab character. Historically, the younger character might
have been “played” by a young Indian warrior intent on causing bloodshed rather than
succumbing to the “White” man’s ways. This old “noble-savage” trope is being recon-
stituted through a shift in mapping, from a young Indian warrior confronting a male
elder to a young Arab Muslim student.
So, what’s the punch line? Well, after hearing the Arab student’s response,
The Montana cowboy shifts his toothpick to one side of his mouth and from
the darkness beneath his Stetson says in a smooth drawl…
“That’s ‘cause we ain’t played Cowboys and Muslims yet, but I do believe it’s
a-comin’.”
Following this final statement, an image of former President George W. Bush appears
on the screen as part of the text of the joke, wearing a cowboy hat and smiling.
The interpretability of this joke relies just as equally on a person’s ability to access
a stereotypical image of an American cowboy (and all the potential social valences
accompanying that image) as it does on the hearer-reader’s ability to access personae
for American Indians and Arab Muslims.
Not only does this joke situate Muslims of Arab ancestry as foreign enemies, but
it reifies the historical image of American Indians as hostile savages and trivializes the
genocidal impact the U.S. nation has had on American Indian populations, a fate, the
joke suggests, soon to befall “Muslims.” Or, to put this in historical bureaucratic terms,
this joke transposes the past nineteenth and twentieth century “Indian Problem” into
the new twenty-first century “Arab Problem.”
Our attention to jokes reveals the malignant ways in which humor serves as “an
index to the development of a national character” (Mintz 1977: 17). All of the jokes
presented here, many of which have been circulating since at least the beginning of
this century, illustrate the ways in which semiotic elements associated with dominant
discourses of Whiteness are covertly maintained and transformed to accommodate
changing socio-political contexts and categories/typifications of difference. These
jokes in particular perpetuate dominant stereotypes of Native North Americans – both
in terms of citizenship (equating them with foreigners, and hence, not U.S. citizens)
and in terms of temporality (imagining them as solitary, elderly figures in an airport
or as ageless figures on horseback riding off into a historical sunset). This final joke has
illustrated the semiotic process of recursion most cogently through the transference of
an old trope onto a new figure through the equation of a child’s game – from “cowboys
and Indians” to “cowboys and Arabs” – and the merging of these two Others as the
playful counterpart to the cowboy’s figure. By mediating the joke through reference
to a child’s game, the more dramatic framings for the joke are diminished (forced
The voice of (White) reason 
a ssimilation, genocide), covertly masked by child’s play. On the one hand, such j ocular
texts are the etched reminders of an American Indian presence and their erasing,
revealing an underlying Whiteness after all. On the other, and perhaps more precisely,
this is a joke about genocide.
6. Conclusion: Deniability, accountability, and covert racist discourse
In de Certeau’s explication of institutional dominance, he proposes a semiotically-

driven theory of consumption that hinges on “the problematics of enunciation”
(1984: 33) – a set of preconditions for a successful (interpretable) communicative
exchange, an exchange through which concepts, or meanings generally, become
conventionalized, (re-)interpretable, and changed. His concern is with the “strate-
gies” of dominant institutions to maintain the status quo and the “tactics” of ordinary
people (consumers) to subvert such dominance. To subvert such dominance requires
partial recognition of the institutional practices of “enunciaton,” their preconditions,
and the circulation and consumption of these institutionally-driven meanings. This
is the first step of our intervention – an articulation of some of the preconditions and
semiotically-mediated generic practices of “enunciation”23 within particular (insti-
tutional/media-producing) locations. Through generic enunciations of “American
Indian-ness,” this paper unpacks and problematizes the ways in which conceptual-
izations of American Indians are entextualized and entitled (rendered interpretable)
discursively, showing that the semiotic elements deployed across media can and do
re-inscribe old-fashioned stereotypes.24
In doing so this analysis attends to both the text of the jokes themselves as well as
the entextualized statements of the characters in the jokes. The linguistic d imension
of these characterizations can be strikingly subtle, and fail to attract the attention of
. Similar to what Asif Agha has called “enregisterment” (1998, 2005; see also Irvine 1990)
or Kenneth Burke’s “entitlement” (1962) or more recently, Paul Garrett’s use of “enfigurement”
(2008).
. In thinking about the social life of texts and their interpretability (through the very
nature of their interdiscursivity), Michael Silverstein encourages his readers to ask “about
those text processes presupposed in a reading by interrogating their traces in the artifactual
form of interest [here, jokes]. We can engage in reading a text, as it were, to shed ethnographic
light on an earlier, otherwise secret discursive life of the text(s) therein” (1996: 81). Similarly,
Richard Bauman and Charles Briggs’ concepts of “de-/re-contextualization” (1990, 1999) and
“intertextuality” (and Silverstein’s “interdiscursivity”) provide a methodology for investigating
the circulation of certain tropes, themes, or meanings in the constitution of a genre and the
emergence of generic form.
 Barbra A. Meek
the reader, perhaps because it resonates with the expectations of a “White” English-
speaking audience. Linguistic characterizations in general – whether textual or per-
formed – frequently seem unremarkable and un-remarked upon by audiences, or
publics, unless the linguistic image interrupts or disengages with the sociolinguistic
assumptions and semiotic expectations of the consumers. Linguistic images seldom
seem to be used to disrupt popular portrayals because of the particular images they are
meant to inspire in the minds of readers and the socializing discourses (norms, values,
etc.) they ultimately reinforce, though even in disruption such indexical disconnects
may be reinforcing. As Hill suggests (2008: 19), such sociolinguistic stereotyping per-
sists as common sense through “White racism as culture, as discourse, as world view,
or as a generative frame for thought.” She continues (2008: 19, 31),
“Each time this common sense plays out in talk and behavior, these fundamental
ideas [of White racism] become available anew, and people use them to understand
what has happened and to negotiate interaction. This constant feedback is
dynamic… and [t]hese stereotypes … circulate among [White Americans] in
discourse, in everyday language, made public in talk and text.”
The majority of this analysis, then, has addressed the linguistic and textual elements
and presumed shared frames deployed in jokes circulating on the internet that rein-
force the derogatory, subordinate, and endangered dimensions of American Indian
enfigurement in public spaces. My reason for focusing on ethnic and frontier jokes
in particular has been that these jokes make public in talk and text stereotypes. They
link to images and discourses that reverberate with the social and cultural values (or
ideologies) of “White” consumers. They likewise provide commentary on national
sentiments and moral trends (Oring 2003). Oring notes that many analysts have
focused on the nationally distinctive elements of humor;
“[T]he content and style of a people’s humor is usually assumed to be peculiar

to the people to whom it belongs. This sense of peculiarity extends beyond the
recognition that the humor of a nation will be expressed in its own language, and
depend upon the idiosyncrasies of the nation’s history, belief, and custom. For it
is further believed that humor is an index of a people’s opinions and character.
It is held to express its temper and embody its spirit. Consequently, the humor
of different nations will of necessity be different because their people and their
temperaments differ.” (2003: 97)25
. Oring goes on to argue that in fact there are similarities across nations and illustrates this
by comparing what he calls “frontier” humor, i.e. humor in three “settler” societies (Australia,
the United States, Israel).
The voice of (White) reason 
Thus, jokes in reflecting the humor of a nation equally reflect or express its historical
and ideological idiosyncrasies, in our case, the fundamental ideas of White racism
in the United States. As shown above, some of the most common jokes are those
that rely on and buttress ultra-established, deeply entrenched features that index par-
ticular stereotypic or regimented images in order to be interpretable, humorous, and
iterable.
Furthermore, the ideologies of citizenship and nationalism emerging from
these media contradictorily promote distinction but mock diversity, maintaining
a particular type of English-speaking citizen as quintessentially “American.” Along
with reinforcing conventional alignments and composites of ethnolinguistic differ-
ence, these semiotic laminations perpetuate the already difficult struggle American
Indians face for individual and tribal recognition. They maintain particular kinds of
inequalities derived from a conceptualization of U.S. citizenship and nationalism as
a “melting pot,” thusly erasing and fossilizing contemporary Indians and creating/
reinforcing a conceptual paradox. What we have seen here are the ways in which par-
ticular features indexical and iconic of (American) Indianness can be and have been
incorporated, embedded and layered into jokes for humorous effect. By continually
representing Indians as historical and to varying extents foreign, particularly through
linguistic images, these jokes constantly obfuscate and erase from public imagina-
tion the reality of American Indians’ citizenships. By representing them as existing
only in the past, the on-going fossilization of American Indians, evidenced most
saliently by Thanksgiving educational traditions, further reduces Indians to a mere
fissure on the historical landscape of America. These antiquated conceptualizations of
Indianness reinforce the erasure of contemporary Indians from a modern U.S. land-
scape. Furthermore, media images that do recognize contemporary Indian citizens
– situating Indian characters in everyday public spaces (like airports) – portray them
as alone, weak, linguistically challenged, and male, in juxtaposition to an enlightened,
civilized and enduring (“white,” non-Indian) nation. Finally, the ideologies of citizen-
ship and nationalism emerging from these humorous genres unsurprisingly under-
score the conception, and the perception, of American Indian citizens as culturally
and linguistically assimilated – that is, as neither “red” nor “white,” but incompetent
and incomplete (contemporary) persons.
Such enunciations reproduce the preconditions that make these images, these
jokes, comprehensible in White public space. They also obscure the location of
Native-ness, of indigeneity; any person born in the U.S. could be recognized as
“native,” but not all persons born in the U.S. can be recognized as “Native.” In fact,
very few can be and are. The ambiguity of the term allows blurredness in inter-
pellation such that were a person identifying as “Native” to take offense at such
ethnically-valenced humor, a differently claimed “native” interlocutor could readily
call his/her sensibilities into question and demand that s/he “get a sense of humor”
 Barbra A. Meek
(cf. van Dijk 1993). The i nterpellative effect of such enunciations results in recogni-

tion, ideally a recognition of “our” American-ness.
American Indianness is crucial to, even emblematic of, our relationship to each
other as U.S. citizens. Through such intersubjective instantiations, this artifact of
Indianness provides and defines a national and a moral sense of belonging and unity
that becomes uniquely American.
That Native Americans continue to be the rube in American humor, and not an
ethnic target in and of themselves, suggests that they have become a significant e lement
of the discourse of “America,” of elite “White” rhetorics. Such generic strategies invite –
if not overtly coerce – a reader’s complicity, which then allows the interlocutor some
deniability and un-accountability. “Through such strategies… [individuals/authors/
interlocutors] preserve deniability about their racist views while at the same time
reproducing them” (Hill 2001b: 79). What Native Americans symbolize in these jokes
is the ideal generic character, the perfect figure and ground to American discourses of
citizenship, dominance, and morality. They are the orienting framework that invokes
a set of expectations whereby the joke becomes readable and humorous. Any other
ethnic or racialized character in American culture would bring along additional
interpretive baggage that would affect the “economy” of the joke, requiring additional
elaboration or a failed response. That the “Indian” goes “how” needs no articulation;
that covert racist Briggs & discourses continue to exist does.
Thus, while Bauman reminds us that “perspectives [such as Hanks’ on genre]
provide a framework for the historical investigation of generic innovation and trans-
formation as speakers manipulate generic conventions in response to and in service of
social change” (1992: 58), the figure of the American Indian has remained remarkably
unchanged in these public forums. The Indian figure has become so generic that it
has become more than just a symbol of American-ness; it is the symbol – the every-
day, taken-for-granted image of mutually constitutive elements that bind the U.S.
nation together: wealthy yet impoverished, democratic yet oppressive, dominant yet
vulnerable, and of course noble yet savage.
Acknowledgements
Heartfelt thanks to Mindy, Trevon, Anna, Rocio, Heidi and Debbie for helping to
breathe life back into a tired, old, fading paper. Thanks to Kathe Managan and John
Thiels for organizing the AAA panel where a much earlier iteration of this paper
first appeared and to Paul Garrett for his wonderful remarks as discussant. Thanks
to Sherina Feliciano-Santos and Sonia Das for commenting on earlier versions, and
Grace Cichy for comments on the most recent one. Mercy buckets to Jane Hill for
inspiration and a few jokes along the way. And the utmost appreciation and gratitude
The voice of (White) reason 
to the editors of this volume for their patience, skill, and encouragement. All errors,
mistaken incongruities, or lack thereof, are my own.
References
Agha, Asif. 1998. Stereotypes and registers of honorific language. Language in Society 27(2):
151–193.
Agha, Asif. 2005. Voice, footing and enregisterment. Journal of Linguistic Anthropology 15(1):
38–59.
Agha, Asif. 2011a. Commodity registers. Journal of Linguistic Anthropology 21(1): 22–53.
Agha, Asif. 2011b. Meet mediatization. Language and Communication 31(3): 163–170.
Aleiss, Angela. 2005. Making the White Man’s Indian: Native Americans and Hollywood movies.
Westport: Praeger.
Althusser, Louis. 1971. Lenin and Philosophy. New York NY: Monthly Press.
Bakhtin, M. M. 1981. The Dialogic Imagination: Four essays. Austin: University of Texas Press.
Basso, Keith. 1979. Portraits of the “Whiteman.” Cambridge: CUP.
Bauman, Richard & Charles L. Briggs. 1990. Poetics and performance as critical perspectives
on language and social life. Annual Review of Anthropology 19: 59–88.
Bauman, Richard & Charles L. Briggs. 2003. Voices of Modernity: Language ideologies and the
politics of inequality. Cambridge: CUP.
Berkhofer, Robert F. 1978. The White Man’s Indian: Images of the American Indian from
Columbus to the Present. New York: Knopf.
Berkhofer, Robert F. 1988. White conceptions of Indians. In History of Indian‐White rela-
tions, Handbook of North American Indians, vol. 4, Wilcomb E. Washburn (ed.), 522–47.
Washington D.C.: Smithsonian Institution.
Briggs, Charles and Richard Bauman. 1992. Genre, Intertextuality, and Social Power. Journal
of Linguistic Anthropology 2(2): 131–172.
Briggs, Charles and Richard Bauman. 1999. The Foundation of All Future Researches; Franz
Boas, Native American Texts, and the Construction of Modernity.” American Quarterly
51(3): 479–528.
Burke, Kenneth. 1962. What are the signs of what? (A Theory of Entitlement). Anthropological
Linguistics 4(6): 18. (See also Burke, Kenneth. 1966. Language as Symbolic Action:
Essays on Life, Literature, and Method, 359–379. Berkeley CA: University of California
Press).
Child, Brenda. 1998. Boarding School Seasons: American Indian families, 1900–1940. Lincoln:
University of Nebraska Press.
Cole, Debbie and Régine Pellicer. 2012. Uptake (un)limited: The mediatization of register shift-
ing in US public discourse. Language in Society 41: 449–470.
Csordas, Thomas J. 1999. Ritual Healing and the Politics of Identity in Contemporary Navajo
Society. American Ethnologist 26(1): 3–23.
De Certeau, Michel. 1984. The Practice of Everyday Life. Berkeley CA: University of California
Press.
Darian‐Smith, Eve. 2003. New Capitalists: Law, Politics, and Identity surrounding Casino Gaming
on Native American Land. Wadsworth.
Deloria, Philip. 1998. Playing Indian. New Haven: Yale University Press.
 Barbra A. Meek
Deloria, Philip. 2004. Indians in Unexpected Places. Lawrence: University Press of Kansas.
Dundes, Alan. 1987. Cracking Jokes: Studies of Sick Humor Cycles and Stereotypes. Berkeley CA:
Ten Speed Press.
Foster, Robert. 2008. Coca‐globalization: Following soft drinks from New York to New Guinea.
New York: Palgrave Macmillan.
Gal, Susan & Kathryn Ann Woolard. 2001. Languages and publics: The making of authority.
Manchester: St. Jerome Publishing.
Garroutte, Eva Marie. 2003. Real Indians: Identity and the survival of Native America. Berkeley:
University of California Press.
Green, Rayna. 1988a. The tribe called wannabe: Playing Indian in America and Europe. Folklore
99(1): 30–55.
Green, Rayna. 1988b. The Indian in Popular American Culture.” In Handbook of North American
Indians, History of Indian‐White Relations (volume 4), Wilcomb Washburn (ed.), 587–606.
Washington D.C.: Smithsonian Institution.
Goffman, Erving. 1974. Frame Analysis: An Essay on the Organization of Experience. Boston
MA: Northeastern University Press.
Hall, Stuart. 1986. Gramsci’s relevance for the study of race and ethnicity. Journal of Communi-
cation Inquiry 10: 5–27.
narrative. In The Dialogic Emergence of Culture, Dennis Tedlock & Bruce Mannheim (eds),
97–147. Urbana IL: University of Illinois Press.
Hill, Jane H. 2001a. Mock Spanish, covert racism, and the (leaky) boundary between public and
private spheres. In Languages and Publics: The Making of Authority, Susan Gal & Kathryn
A. Woolard (eds), 83–102. Manchester: St. Jerome.
Hill, Jane H. 2001b. Discussion (special issue). Journal of Linguistic Anthropology, 78–82.
Hill, Jane H. 2005. Intertextuality as source and evidence for indirect indexical meanings.
Hill, Jane H. 2006. The ethnography of language and language documentation. In Essentials
of Language Documentation, Jost Gippert, Nikolaus Himmelmann & Ulrike Mosel (eds),
113–128. Berlin: Walter de Gruyter.
Irvine, Judith T. 1990. Registering affect: Heteroglossia in the linguistic expression of emotion.
In Language and the politics of emotion, Catherine Lutz & Lila Abu‐Lughod (eds), 126–161.
Cambridge: CUP.
of Language: Ideologies, Politics, and Identities, Paul V. Kroskrity (ed.), 35–83. Santa Fe NM:
School of American Research Press.
Jaffe, Alexandra. 1999. Ideologies in action: Language politics on Corsica. Berlin and New York:
Mouton de Gruyter.
Kilpatrick, Jacqueline. 1999. The celluloid Indian: Native Americans and film. Lincoln: University
of Nebraska Press.
Kockelman, Paul. 2007. Agency: The relation between meaning, power, and knowledge. Current
Anthropology 48(3): 375–401.
Marsden, Michael T. & Jack G. Nachbar.1988. The Indian in the movies. In History of Indian‐
white relations, Handbook of North American Indians, vol. 4, Wilcomb E. Washburn (ed.),
607–616. Washington D.C.: Smithsonian Institution.
Mazzarella, William. 2003. Shoveling Smoke: Advertising and globalization in contemporary
India. Durham: Duke University Press.
The voice of (White) reason 
Meek, Barbra A. 2006. And the Injun goes how!: Representations of American Indian English in
White Public Space. Language in Society 35: 93–128.
Meek, Barbra A. 2010. “We Are Our Language”: An ethnography of language revitalization in a
Northern Athabaskan community. Tucson: University of Arizona Press.
Meek, Barbra A. 2011. Failing American Indian languages. American Indian Culture and
Research Journal 35(2): 43–60.
Merskin, Debra. 2001. Winnebagos, Cherokees, Apaches, and Dakotas: The persistence of
stereotyping of American Indians in American advertising brands. The Howard Journal of
Communications 12: 159–169.
Mihesuah, Devon. 1996. American Indians: Stereotypes and realities. Atlanta: Clarity.
Mintz, Lawrence. 1977. American humor and the spirit of the times. In It’s a Funny Thing,
Humor, Anthony J. Chapman & Hugh C. Foot (eds), 17–21. London: Pergamon.
Morgan, Mindy. 2009. The bearer of this letter: Language ideologies, literacy practices and the Fort
Belknap Indian community. Lincoln: University of Nebraska Press.
Morrison, Toni. 1992. Playing in the Dark: Whiteness and the literary imagination. Cambridge:
Harvard University Press.
Ochs, Elinor. 1990. Indexicality and socialization. In Cultural psychology, James Stigler et al.
(eds), Cambridge: CUP.
Oring, Elliott. 2003. Engaging Humor. Urbana IL: University of Illinois Press.
Paolucci, Paul & Richardson, Margaret. 2006. Sociology of humor and a critical dramaturgy.
Symbolic Interaction 29(3): 331–348.
Pratt, Richard. 2004[1964]. Battlefield and Classroom: Four decades with the American Indian,
1867–1904. Norman: University of Oklahoma Press.
Rappoport, Leon. 2005. Punchlines: The Case for Racial, Ethnic and Gender Humor. Westport
CT: Praeger.
Richland, Justin 2008. Arguing with tradition: The language of law in Hopi tribal court. Chicago:
University of Chicago Press.
Rollins, Peter C. & John E. O’Connor. 1998. Hollywood’s Indian: The portrayal of the Native
American in film. Lexington: University Press of Kentucky.
Silliman, Stephen W. 2008. The “Old West” in the Middle East: U.S. Military metaphors in real
and imagined Indian Country. American Anthropologist 110(2): 237–247.
Silverstein, Michael. 1979. Language structure and linguistic ideology. In The elements: A para-
session on linguistic units and levels. P. Clyne, W. F. Hanks, & C.L. Hofbauer (eds), 193–247.
Chicago: Chicago Linguistic Society.
Silverstein, Michael. 1996a. The secret life of texts. In Natural Histories of Discourse, Michael
Silverstein & Greg Urban (eds), 81–105. Chicago IL: University of Chicago Press.
Silverstein, Michael. 2003. Indexical order and the dialectics of sociolinguistic life. Language and
Communication 23: 193–229.
Strong, Pauline Turner. 1996. Animated Indians: Critique and contradiction in commodified
children’s culture. Cultural Anthropology 11(3): 405–424.
Strong, Pauline. 2004. Representational practices. In A Companion to the Anthropology of Amer-
ican Indians, Thomas Biolsi (ed.) 341–359. Malden, MA and Oxford UK: Blackwell.
Strong, Pauline Turner. 2005. Recent ethnographic research on North American Indigenous
peoples. Annual Review of Anthropology 34: 253–268.
Trechter, Sarah. 2001. White between the lines: Ethnic positioning in Lakhota discourse. Journal
of Linguistic Anthropology 11(1): 22–35.
Urciuoli, Bonnie. 1996. Exposing Prejudice. Boulder CO: Westview Press.
van Dijk, Teun. 1993. Elite Discourse and Racism. Newbury Park CA: Sage.
Double-voicing in the everyday language
of Brazilian black activism
Jennifer Roth-Gordon & Antonio José B. da Silva

In this study of the daily linguistic practices of Brazilian black activists, we

draw on Jane Hill’s well-known research on voice to interrogate how speakers
metalinguistically invoke “competing” points of view. Bringing together research
conducted at the height of politically conscious hip hop’s success in the late
1990s in Rio de Janeiro and fieldwork conducted with race-based community
organizations in Salvador, Bahia in 2009–2010, we argue that speakers actively
counterpose “racist” and “anti-racist” voices – often within a single translinguistic
word – in their quest to display racial consciousness. Embracing similar linguistic
processes, political opponents of race-based policies draw different battle lines
within the same words, interpreting the struggle as one between North American
and Brazilian understandings of race and racism.
Keywords: anti-racism; heteroglossia; language ideology; discourse analysis; Brazil
1. Introduction
In the mid-1980s, Jane Hill began to write extensively on the work of Mikhail
Bakhtin (made available to the English-speaking public starting in the late 1960s,
see 1968, 1981, 1984), introducing his view of language as characterized by disor-
der, struggle, and the cacophony of disparate voices or points of view to the field
of linguistic anthropology (Hill 1985, 1986). Applying Bakhtin’s methods of lit-
erary analysis to conversational narratives, Hill revealed how a speaker “claims a
moral position among conflicting ways of speaking, weighted with contradictory
ideologies, by distributing these across a complex of ‘voices’ ” (Hill 1995: 98). At
a 2008 AAA panel honoring Jane Hill’s nearly 50-year career in academia, Paul
Kroskrity drew on her classic study of Don Gabriel to remark that Hill had inspired
generations of linguistic anthropologists to hear voices – and lots of them (see
also Kroskrity 2011). In what follows, we turn to interrogate how voices are not
just “heard” – by informants or linguistic anthropologists – but also constructed,
 Jennifer Roth-Gordon & Antonio José B. da Silva
counterposed, and made recognizable. In our study of the “everyday language” of

Brazilian black activists,1 we are inspired by Hill’s research on voice systems to
interrogate how speakers metalinguistically use voice to invoke “competing” points
of view and position themselves in relation to them.
We bring together research conducted at the height of politically conscious hip
hop’s success in the late 1990s in Rio de Janeiro, which corresponded with an upsurge
in black political mobilization, and fieldwork conducted with race-based community
organizations of the Brazilian black movement in an era of new affirmative action
policies in Salvador, Bahia in 2009–2010. Examining daily conversations, we argue
that speakers actively set “racist” and “anti-racist” voices against one another, while
identifying themselves with anti-racist ones, in their quest to display racial conscious-
ness. We further illustrate how critics of race-based policies in Brazil embrace simi-
lar linguistic processes, albeit drawing different battle lines within the same words, to
interpret the struggle as one between North American and Brazilian understandings
of race and racism.
In an era of redemocratization and heightened awareness of unequal access to
social, political, civil, and human rights, some Afro-Brazilian activists have made race
a more explicit criteria for negotiating their access to positions of power. In order
to challenge a legacy of racial democracy and “cordial racism,” these groups seek to
raise individual awareness of dominant racial discourses and propose new interpre-
tations of racial inequality. However, these reinterpretations are often challenged
in the dominant media as inappropriate imports from U.S. racial politics that are
uncritically adopted by intellectuals and Brazilian activists. Our emphasis is on the
metalinguistic activity of black Brazilian activists who often reflexively engage with
the words used in their daily activities and who focus, in particular, on words that
are used to describe them in racializing terms. Within a context of antiracist activ-
ism, participants are often quite invested in metalinguistic maneuvering – including
the resignification of terms – to advance their political goals. And yet, as we have
found, the linguistic processes they draw on can be picked up by their critics, who in
turn “hear” different voices in the race-conscious arguments of anti-racist activists.
Attributing these voices to U.S. racial politics and counterposing them with more
authentically Brazilian ones, they reframe the conflict through the construction of
an alternate set of opposing perspectives. We thus argue that the battle over racial
“consciousness” in Brazil is, in no small measure, a linguistic battle that is fought at
the level of the translinguistic word.
. Here our use of the term “everyday language” is intended to create an intertextual link to
Hill’s substantive contributions to the study of race and language (see, in particular, 2008).
Double-voicing in the everyday language of Brazilian black activism 
2. Debating racial politics in Brazil
In order to understand today’s fierce debate over racial politics in Brazil, it is n ecessary
to understand the persistent comparison made between Brazilian and U.S. race
relations – two countries that share much in their racial histories. This includes a Native
population quickly decimated by European colonization; the importation of millions
of African slaves over several centuries; and the subsequent assimilation of millions of
“whiter” Europeans. And yet, despite these similarities, Brazil and the United States
came to eventually be pitted against one another as stark racial o pposites. Brazil, in
particular, received international attention in the 1950s, when post-WWII studies
funded by UNESCO sought to understand its unique situation of “racial harmony.”
North American researchers, among others, arrived in Brazil to investigate how the
two countries could have started out so similarly and wound up in such different
places: The U.S. was known for a history of violent racial conflict; legal struggles over
definitions of race (who is white?), relationships between whites and non-whites, and
the rights of people of color; and a strict system of racial categorization (and segre-
gation) based largely on ancestry. Brazil, by comparison, described itself as kinder,
gentler, and more flexible – in everything from its master-slave relationships, to its
peaceful (albeit delayed) abolition of slavery and its embrace of miscegenation, to its
status as a model of interracial harmony for the rest of the world.
The results of the UNESCO studies revealed, however, the existence of prejudice
and discrimination coexisting with miscegenation (Costa 1985), such that darker-
skinned individuals clearly fared worse in Brazilian society. By the 1980s, the role
of race in Brazilian social stratification was well documented both qualitatively and
quantitatively (Hasenbalg 1979; Hasenbalg & Silva 1988). At this point, some schol-
ars turned the focus of their research to study Afro-Brazilian “consciousness,” or the
awareness of “the conditions of one’s existence, imagining alternatives and striving to
actualize them” (Hanchard 1991: 99). Still grounding their research in a comparison
between Brazilian and U.S. racial politics, they sought to explain the alleged “weak-
ness” of antiracist mobilization among Brazilians of African descent. These studies,
conducted in the 1990s by mostly North American academics, suggested that Brazil’s
dominant racial ideology of a shared national background that included Europeans,
Africans, and Indigenous peoples and its reputation for “cordial” racism had led blacks
in Brazil to self-identify as racially mixed, rather than black. This, they further argued,
had prevented Brazilian blacks from discussing race or white supremacy, from orga-
nizing politically around being “black,” and ultimately from redressing racially s pecific
patterns of inequality (Hanchard 1994, 1998; Sheriff 2000; Twine 1998; W arren &
Twine 2002). Hanchard’s work, in particular, attracted criticism for its Brazil-U.S.
comparisons and the suggestion that racism could only be addressed through the overt
racial struggles found in the U.S. (Bairros 1996; Fry 1995/96; Silva 1998).
 Jennifer Roth-Gordon & Antonio José B. da Silva
Fifty years after the UNESCO studies, at the turn of the twenty-first century,
Brazil took a sharp turn in state policy to recognize the existence of Brazilian racism
and to embrace government-sponsored positive actions against racial discrimina-
tion. Due in part to this institutional support, North American inspired ideas of
racial consciousness and identity politics have now become a primary idiom through
which contemporary black organizations articulate their struggle for the economic,
social, and civil rights of Brazilians of African descent. While some academics laud
these new political shifts and have documented how race-based NGOs have capital-
ized on the increased attention to racial inequality (French 2009; Vargas 2006), crit-
ics of race-based activism accuse the black movement of Americanizing Brazilian
race relations and ignoring Brazil’s unique racial situation (Fry 2000; Fry et al. 2007;
Sansone 2003).
Kamel’s (2006) book, Não Somos Racistas: Uma Reação aos que Querem nos Trans-
formar numa Nação Bicolor (We Are Not Racist: A Reaction to Those Who Want to
Convert us into a Bi-color Nation), serves as an excellent introduction to Brazil’s most
recent racial debates. In this widely distributed book, Ali Kamel, executive director
of journalism for Globo, Brazil’s largest media network, harshly criticizes the intro-
duction of U.S.-based affirmative action policies into Brazil’s higher education system
and the state sectors, targeting especially the initiation of racial quotas in public uni-
versities throughout the country. He argues that Brazil has always been proud of its
mixed-race population and that racism is non-existent as a general characteristic of
the Brazilian nation. Labeling people based on race, Kamel warns his readers, will only
produce racism in Brazil. He accuses the black movement and policy makers of misin-
terpreting census statistics to prove that racism is the cause of the disenfranchisement
of the Brazilian black population. Instead, Kamel contends that racism in Brazil is not
institutional but consists of isolated, individualized incidents. Voicing a longstanding
common opinion in Brazil, Kamel argues that the country’s problems are based in
social inequality and Brazilian poverty and that they reflect the misguided priorities of
the country’s political and economic elites.2 Arguing that racism should be combated
through anti-racialism, i.e. with the rejection of the concept of race altogether, Kamel
blames the current government and black activists for formalizing ideologies of race
as a criterion for defining public policy. Thus even as racism is now widely recognized
within Brazil, front page news stories and everyday conversations include heated dis-
cussions over how to talk about race in Brazil and what to make of the steady influx of
North American ideas.
. See Miranda-Ribeiro (2006) for a scholarly review of Kamel’s argument.

Double-voicing in the everyday language of Brazilian black activism 
3. Negro vs. preto: Double-voicing blackness
One of Brazil’s most salient racial struggles includes the various linguistic terms that
have been employed to describe Brazilians of African descent. Racial classification
figures strongly in the current debate over affirmative action (which we will discuss in
a later section), and yet the struggle over racial categories can be traced back at least 80
years, as black activists attempted to semantically shift the term negro (black) from a
negative term used by slave masters to describe their less docile and more “rebellious”
slaves to a positive term that signified black pride. In a recent history of the black
movement, Garcia explains:
The expression negro used to humiliate, discriminate and attack the African
descendents in Brazil. … The Black movement rewrote this expression. Negro
became the word of order, of reconstructing dignity, of self-esteem development.
We have transformed the disqualification into the greatest qualification of our
identity.(Garcia 2006: 24)
Thus, even though the term was not commonly used by the Brazilian government, in
daily speech, or as a term of self-reference, black activists continued to argue for the
use of negro instead of more common terms including: preto (black, person with black
skin color), mulato (mixed, offspring of a black person and a white person), moreno
(brown, person with brown skin color), or pardo (brown or mixed, person whose dark
or brown skin color indicates racial mixture).3 In 1931, as one of the earliest exam-
ples, the Frente Negra Brasileira (FNB, The Black Front) was created in São Paulo as a
black press that later became an official political party. In 1945, Abdias do Nascimento
founded the Teatro Experimental do Negro (TEN, Experimental Black Theater) in Rio
de Janeiro, and then in the 1970s, the Movimento Negro Unificado (MNU, The Unified
Black Movement) marked the reemergence of black activism during the Brazilian dic-
tatorship and into the start of redemocratization – this time originating in Salvador,
Bahia. In part due to its association with black activism, negro has come to be seen
as a term associated with raça (race) in contrast to a description of cor (color), which
is readily associated with the term preto (Sansone 2003; Sheriff 2001). A distinction
between race and color has also been informed by decades of scholarship on Brazilian
race relations – which often sought to contrast the system of racism (associated more
strongly with what was found in the United States) with Brazilian experiences of color
prejudice (Frazier 1942; see also the history of this comparison in Guimarães 2004).
. Definitions are taken from Stephens’ (1989) Dictionary of Latin American Racial and
Ethnic Terminology.
 Jennifer Roth-Gordon & Antonio José B. da Silva
Here we draw on research conducted by Jennifer Roth-Gordon to show how

Brazilian rap fans discussed these terms at the height of politically conscious hip
hop, in the late 1990s. While these youth would not technically identify themselves
as black activists, they were deeply inspired by rappers who may be quite politically
active: Rio rapper M.V. Bill has been involved with various NGOs and helped create a
new Brazilian political party. Following legendary U.S. rappers such as Public Enemy,
politically conscious rap in Brazil (what Pardue (2008) calls “marginal rap”) differs
from other rap genres in its overt discussion of race as a primary factor of Brazilian
social inequality. Predating the advent of race-based governmental reforms and the
majority of affirmative action programs, politically conscious rappers drew on North
American hip hop to encourage fans to recognize institutional racism and “assume
their blackness” (assumir a sua negritude). Rappers often spoke in and publicized a
language of “consciousness,” as when Rio rappers M.V. Bill and D.J.T.R. named their
community radio rap program “S.O.S. Consciência: O programa para alguém que
tem algo a dizer, algo a fazer.” (S.O.S. Consciousness: The program for someone who
has something to say, something to do). Rappers and some rap fans would actively
identify themselves as racially conscious, contrasting this with the figure of the negro
comportado/acomodado (unassuming, assimilated, and well-behaved black person).
Roth-Gordon’s research explores the racialization of the Brazilian body through
linguistic practice based on fieldwork conducted with a male peer group who lived
in a favela (shantytown) in the wealthy South Zone of Rio de Janeiro. Male youth in
this community often used their participation in global hip hop to help them navi-
gate their position between First and Third-World living, as they resided in run-down
government buildings (akin to U.S. projects) that were a mere stone’s throw from some
of Rio’s most exclusive neighborhoods (see Roth-Gordon 2012, 2013). In Excerpt 1, we
present an interview in which two poor male youth – both rap fans – explain how they
racially identify, drawing on the terms also given by well-known Brazilian rappers (see
Appendix for transcription conventions):
Excerpt 1
1 CW: tu é preto, are you preto ((black)),
2 ou é branco? or are you white?
3 heim? huh?
4 a pergunta que the question she’s asking you,
ela tá te fazendo,
5 KLJ: se eu sou 〈F o quê F〉? if I’m 〈F what F〉?
6 CW: preto ou é branco? preto or are you white?
7 KLJ: eu sou 〈F negro F〉. I’m 〈F negro F〉 ((black)).
8 preto é o 〈F asfalto F〉. 〈F asphalt F〉 is preto.
9 CW: 〈@ lápis de cor @〉. 〈@ a colored pencil @〉
Double-voicing in the everyday language of Brazilian black activism 
10 Jen: preto é o quê? what is preto?

11 KLJ: preto é é é cor, preto is is is a color,
12 CW: (0) lápis de cor. (0) a colored pencil.
13 KLJ: (0) é asfalto, (0) it’s asphalt,
14 é pneu, it’s a tire,
15 a minha cor é negra. my color is negra.
16 cor cor ..negra. color color ..negra.
17 preto é chão, preto is the floor,
18 chinelo, a sandal,
19 pneu, a tire,
20 CW: 〈isso aqui〉 〈like this〉
21 tim sempre falava isso, tim always said that,
22 “preto? “preto?
23 eu sou é negro.” I’m negro.”
24 ele arrumou uma he created a scene in the
confusão no hospital. hospital.
25 porque era “〈preto〉, because it said “〈preto〉,
26 preto pardo e branco.” ((se black brown and white.”
referindo ao formulário que tim ((referring to a form tim had
teve que preencher no hospital)) to fill out in the hospital))
27 KLJ: é negro, rapaz. it’s negro, man.
28 CW: “preto? “preto?
29 eu sou é negro. I am a negro.
30 pode ir tirando essa you can get rid of this
porra 〈F daí F〉. crap 〈F here F〉.
31 eu sou é negro. I am negro.
32 〈tira essa〉 〈get rid of〉
33 preto? preto?
34 não. no.
35 bota 〈L negro L〉. write 〈L negro L〉.
36 bota aí.” write it down.”
37 […] […]
38 Jen: você vai preencher o quê? what would you mark?
39 KLJ: negro. negro.
40 Jen: preto ((se referindo preto ((referring to
ao formulário)) the form))
41 〈que você vai〉 〈what would you〉
42 KLJ: eu risco preto, I would cross out preto,
43 e boto negro. and put negro.
44 CW: 〈F é F〉 pô 〈F right F〉 damn it
45 〈porque é mais〉 〈because it’s more〉
46 lá em são paulo, there in são paulo,
47 〈já é〉 〈it’s already〉
 Jennifer Roth-Gordon & Antonio José B. da Silva
48 não adianta – you don’t have a choice –

49 é preto tipo A. it’s preto tipo A. ((class A black))
50 〈negro já〉 〈negro already〉
51 Gue: 〈F preto tipo A F〉? 〈F preto tipo A F〉?
52 CW: 〈@ tipo A @〉 〈@ class A @〉
53 KLJ: 〈@ é @〉 〈@ right @〉
54 Jen: que que quer dizer isso? what does that mean?
55 KLJ: preto tipo A é 〈aquele〉 preto tipo A is 〈someone〉
56 aquela música que that song I was
eu tava escutando. listening to.
57 é uns que são melhor it refers to those who think
do que o outro. they’re better than other people.
58 CW: (0) 〈F não F〉 (0) 〈F no F〉
59 KLJ: não? no?
60 CW: não, no,
61 KLJ: não? no?
62 CW: nada a ver. not at all.
63 KLJ: você you
64 CW: (0) é aquele que assume que é (0) it refers to someone
65 preto mermo e.. who really embraces being black..
66 Gue: (0) então não é preto. (0) so then it shouldn’t be preto.
67 é negro. it’s negro.
68 não tem negócio de preto. there’s no such thing as preto.
69 preto é.. 〈F chão F〉. preto is.. 〈F the floor F〉.
In this excerpt, Rio youth demonstrate their awareness of ongoing debates over racial
terminology, including the common Brazilian distinction between race and color.
This comparison has historically entrenched notions of Brazilian exceptionalism –
suggesting that Brazilians notice color, but do not make racial distinctions or par-
ticipate in the extreme examples of racism found in the United States. However, in
this interview with a North American researcher, these youth actively construct them-
selves as “racially conscious” or as blacks who take pride in their race. To this end, they
dismiss preto as a color that applies to objects (such as colored pencils, sandals, asphalt,
the floor, and a tire, in lines 8–19 and 69) – rather than to people – and they agree that
the more appropriate term for people of African descent is negro. Drawing on Hill’s
Bakhtinian analysis of dialogism and voicing, we turn to interrogate how these kinds
of metalinguistic discussions among Brazilians who identify as “racially conscious”
disperse conflicting points of view across a system of voices.
While all utterances are dialogic according to Bakhtin, some are “more d ialogic”
than others, and Bakhtin showed particular interest in situations of active double-
voicing (Bakhtin 1981, 1984; Morson & Emerson 1990). In the case of Brazilian racial
categorization, the unmarked census category and more common everyday term for
blackness is preto. In Bakhtin’s terms, the use of preto can be single-voiced; that is, as
Double-voicing in the everyday language of Brazilian black activism 
the unmarked term, its use for many Brazilians does not call to mind a linguistic strug-
gle between options or different perspectives. As Morson and Emerson note, single-
voiced words “speak as if there were no ‘spectral dispersion’ of the word; they simply
name their referent” (1990: 148). Indeed, CW and KLJ frequently use the term preto
in their daily speech when they wish to describe someone as dark-skinned. However,
in their metalinguistic discussion above, they reveal the potential for an utterance,
or translinguistic word, to turn from single to double-voiced. As they reject the term
preto for its “racist” point of view (from an imagined author who considers them to be
objects, rather than empowered racial subjects), they seek to remind others that, in the
words of Bakhtin, “there are no ‘neutral’ words and forms – words and forms that can
belong to ‘no one’ ” (1981: 293). While the origins of unmarked or single-voiced words
may be less visible or audible, these words do not, in fact, speak “from nowhere”
Irvine & Gal (2000: 36).
By contrast, marked terms such as negro are actively double-voiced and are more
easily traced back to a source such as the Brazilian black movement. These terms also
readily call to mind unmarked terms (such as preto) that have been avoided or rejected.
Providing an excellent (if perhaps hypothetical) visual illustration of this point, KLJ
explains in lines 42–43 that to use the term negro is to “cross out” the option of preto.
This type of linguistic contrast has been previously theorized by Kathryn Woolard,
who argues in her study of Spanish/Catalan bilinguals that speakers do not always
choose between languages or between linguistic options but may instead productively
juxtapose linguistic elements to allow them to “thrive in tense intersection” (1999: 5).
Drawing on Bakhtin’s interest in simultaneity, she argues that “live, unresolved copres-
ences” (1999: 6) create new meaning: “Contrast and opposition do not have to do all of
their semantic work in absentia, through mutual exclusion” (Woolard 1999: 5). Thus to
choose to identify as negro is not just to cite the Brazilian black movement (or North
American hip hop): It is to know that preto is a term to be avoided as the voice of those
less “ligado” (a slang term meaning connected or “attuned”). When KLJ responds in
lines 42–43, “I would cross out preto and put negro” (Eu risco preto e boto negro), he
capitalizes on oppositional copresence to make simultaneous statements of rejecting
the color term given to him by dominant Brazilian society and actively choosing to
identify with a more controversial or “empowered” term.
Bakhtin’s work on dialogism allows us to see not only how voices may “collide”
within the single word, but how these voices reveal the perspectives of embodied
authors who speak from “a definite position” (1984: 184). Through active double-
voicing, speakers thus produce “objectified and finalized images of people” (1984: 182).
Drawing on related theories of citationality, Inoue’s work suggests that voices are not
merely heard or recognized but “rendered audible” and created through a produc-
tive act of juxtaposition (2006: 39). We suggest that, through this process, racially
conscious Brazilians actively and intentionally construct the voices they hear in these
translinguistic terms as binary racial positions. In this conversation, KLJ suggests
 Jennifer Roth-Gordon & Antonio José B. da Silva
that we can hear in these racial terms two recognizable and contrasting voices: Preto
exposes the voice of dominant racist ideology with its more docile and assimilated
black person who accepts this racial logic, while the use of negro reveals a voice of
black awareness and pride. Through a process of erasure (Irvine & Gal 2000), the
various voices that collide are reduced to two: the racist and the anti-racist.
And yet the confusion at the end of Excerpt 1 belies a convenient reality of one-to-
one correspondences. The term preto tipo A (a class A black person) was popularized
by Brazil’s most successful politically conscious rap group, Racionais MC’s (The Ratio-
nals) in their song Capítulo 4, Versículo 3 (Chapter 4, 3rd Verse – referring to their 4th
album and the 3rd track). In this song, they criticize a preto tipo A who used to look
and act like them but then “sells out” and aspires after white culture. In this coinage,
Racionais MC’s engage in active double-voicing as they intend to juxtapose the estab-
lished but hidden meaning of preto as degraded and low-ranking with the affirmative
addition of tipo A (grade A, first-class – used in Brazil, as in the U.S. for food items
like milk or eggs to denote higher quality). The term is thus designed to encapsulate
the linguistic struggle between those who would use preto to demean and discriminate
and those who use preto tipo A to protest this stigmatization and valorize blackness.
Thus this seemingly clear-cut distinction between “racist” (preto) and “anti-racist”
(negro) points of view does not hold up: Brazilian rappers, many of whom espouse
messages of black pride and anti-racism, are often divided over racial terms pertain-
ing to blackness. And everyday terms that are recognized as obviously racist likewise
can take either form: for example, negro safado (an insult approaching the n-word
in English) or coisa de preto (only a black person would do a thing like that). Due to
his own ideas about these terms, KLJ proceeds to “correct” Racionais MCs’ use of the
term preto, ignoring the active double-voicing the rap group is engaging. KLJ’s con-
fusion over this term’s “point of view” was not unusual: In the article “O Rap Sai do
Gueto” (Rap Comes Out of the Ghetto) published in the Brazilian mainstream maga-
zine, Época (Rodrigues 1998), the term preto tipo A is listed in a side bar entitled “A
Voz do Mano” (The Voice of the Brothers). Translating for their mostly white middle-
class audience, reporters explain that preto tipo A means “aquele que virou mauricinho”
(a guy who imitates a white preppy kid) – a far cry from the guy who takes pride in
being black. Dialogism allows us to see that Racionais MC’s do not control the
representations of their own voice. Just as KLJ and his friends can “out” single-voiced
terms such as preto when they appear on hospital forms, here the intentionally transling-
uistic expression attributed to Racionais MC’s is reinterpreted by the dominant B razilian
media to give the opposite meaning of what they intend. In all of these situations, the
linguistic struggle over translinguistic words spills beyond the intentions of its speakers.
In the sections that follow, we describe how metalinguistic negotiations that take up the
question of the representation of Brazilians of African descent allow for reinterpreta-
tions by black activists that are themselves further reinterpreted by their critics.
Double-voicing in the everyday language of Brazilian black activism 
4. Escravo vs. escravizado: What’s in a word?
First, the existence of what is called an ‘escravo’ (slave) is not a reason to accept
slavery. Under any circumstance, slavery is a dehumanizing institution and it must
be condemned. Men are born free until someone enslaves them. Therefore, the
concept itself is wrong. The correct term is ‘escravizado’ (enslaved), not ‘escravo’
(slave). There is no natural category of a slave.4 – Kabengele Munanga 2010
In this interview, Kabengele Munanga, a black activist and anthropology professor at

Universidade de São Paulo, illustrates how the black movement has recently sought to
reinterpret and rewrite the history, culture, and position of blacks in Brazilian society
(see also Munanga & Gomes 2004). By the time of Antonio José B. da Silva’s fieldwork
in 2009–2010, many self-identified black activists in the city of Salvador articulated
similar didactic and linguistic strategies. Silva observed a range of classes in commu-
nity-based organizations associated with the black movement. Some of these classes
focused on consciousness-raising, such as one entitled “Cidadania e Consciência
Negra” (Citizenship and Black Consciousness), while other classes taught anti-racism
through content-based instruction, such as the one entitled “História da África, Cul-
tura Negra, e o Negro no Brasil” (African History, Black Culture, and Blacks in Brazil).
These classes were attended by youth and adults in the evening and on weekends and
had the objectives of providing supplemental education for high school students,
preparation for college admissions exams, and training for public school teachers5 or
members of the general public.
Activists often taught these anti-racist perspectives through what we have pre-
viously called “affirmative language practices” or linguistic strategies promoting
black consciousness (Silva 2011). For example, as Munanga suggests above, teachers
routinely encouraged participants to embrace the term escravizado (enslaved) and
avoid the use of escravo (slave) in order to highlight a process through which Africans
were forcibly enslaved in colonial Brazil. This shift from escravo to escravizado
. “Em primeiro lugar, a existência do chamado ‘escravo’ não é razão para aceitar a es-
cravidão. Em qualquer circunstância, a escravidão é uma instituição desumanizante e deve
ser condenada. O homem nasce livre até que alguém o escravize. Portanto, o próprio conceito
está errado. O correto é ‘escravizado,’ não ‘escravo.’ Não há uma categoria de escravo natural”
(Munanga 2009).
. In 2003, Brazil passed a law that mandates that schools incorporate Afro-Brazilian history
and culture into their curriculum. Teachers who attended these classes were generally inter-
ested in networking with other educators who were also teaching these new courses on Afro-
Brazilian history and culture. They wanted to learn about available teaching and learning
resources and to improve their abilities to discuss racial issues with their s tudents.
 Jennifer Roth-Gordon & Antonio José B. da Silva
embraces double-voicing to point out the hidden and competing perspectives on slav-
ery, contrasting the racist voice of the white slave master who recognized Africans as
natural slaves with the anti-racist voice of one opposed to this unnatural and unjust
enslavement. In Excerpt 2, a teacher named Lúcio initiates a discussion about present-
day black struggles for land rights based on a school fieldtrip in which adult students
(ranging in age from 18–70) visited a nearby quilombo (a rural community of descen-
dants of fugitive slaves). Despite the physical proximity of these communities and
students’ awareness of well-publicized controversies surrounding these lands, many
Brazilians living in Salvador have never actually visited a quilombo or spoken with
current day residents who are fighting for the legal recognition and title of their lands.
As Lúcio explains that quilombos have been the target of land disputes between “legal”
owners and quilombo descendants (since only some quilombos benefit from federal
recognition), his use of the term escravo is corrected by one of his students:
Excerpt 2
1 Luc: então o nosso intuito hoje aqui, so our goal here today,
2 […] […]
3 fazendo uma relação, establishing connections,
4 entre o que fala os documentários, between what the documentaries
os filmes, os livros, os textos que say, the films, the books, the texts
a gente viu aqui that we studied in class,
5 e..a realidade do pessoal de and..the reality of the people
lá do são francisco do paraguaçu. there in são francisco do
paraguaçu [quilombo].
6 ou seja, that is,
7 ah a gente sabe que a luta we know that it is an..
é..árdua arduous struggle,
8 né? right?
9 em relação ((cough)) a I’m talking about ((cough))
politicas publicas, public policies,
10 e a direitos principalmente and the rights of our people,
do nosso povo,
11 dos nossos ancestrais, of our ancestors,
12 e aquele pessoal que tá lá, and those people there,
13 vocês viram, that you saw,
14 os senhores contar que the men say that they are the
eles são filhos dos filhos children of the children
de escravos, of slaves,
15 〈filha〉 〈the daughter〉
16 a mãe foi filha de escravo, the mother was the daughter
of slaves,
17 o pai foi filho de escravo, the father was the son of slaves,
18 foi neto de escravo, he was the grandson of slaves,
19 Nat: 〈F escravizado F〉 X 〈F enslaved F〉 X
Double-voicing in the everyday language of Brazilian black activism 
20 Luc: for- wer-

21 foram filhos de escravo were children of slaves
ou escravizados, or enslaved ones,
22 a gente aprendeu o we have always learned to say
tempo todo a falar escravo, slave,
23 né? right?
24 claro que escravizados. of course enslaved ones.
25 então é – so uh –
26 e eles permanecem lá, and they are still there,
27 né? right?
28 pouco tempo atrás é que vieram, a little while ago they came,
29 a saber que eram quilombolas. to know that they were quilombolas.
((residents of a quilombo))
30 que até então a história negou until then history completely
isso pra eles o tempo todo. denied this to them.
31 não é? right?
32 e de certo modo ainda continua in some ways the history books
tentando negar. still try to deny it.
33 né? right?
34 a ancestralidade deles, their ancestry,
35 e tudo mais. and all that.
In lines 14–18, Lúcio uses the term escravo four times as he reports, using indirect
speech, what a family of descendants of slaves told them during their visit to the
quilombo. In line 19, Natanael, a student in the class, interrupts Lúcio and, from the
back of the room, boldly launches a repair of his use of escravo, offering the word
escravizado as a replacement. Here, as above, we are interested in how this metalin-
guistic conversation foregrounds the unmarked term as double-voiced. While the past
participle escravizado from the verb escravizar (to enslave) is common grammatical
knowledge, its use in place of the nominal form escravo politically appropriates the
term to mark participation in or affiliation with the anti-racist movement. In offering
escravizado as a correction for escravo, Natanael is double-voicing his teachers and
other black activists who have made him aware of this contrast and indicated that
escravizado is the preferred term. In other words, in Natanael’s citation of escravizado,
the anti-racist voice he has been exposed to in previous contexts predominates. This
example illustrates how translinguistic words are a “powerful effect of a system of
citations” (Inoue 2006: 281). His teacher Lúcio engages Natanael’s correction and ini-
tiates a self-repair with an eventual uptake in line 21, which can be understood as a
citation of Natanael’s citation. Lúcio then continues the metalinguistic discussion to
reinforce that the term escravo is internally dialogized. He seeks to excuse his use of
escravo by noting that such usage has been ingrained in them, and thus it is difficult
to change (line 36). At the same time, agreeing with Nataneal, Lúcio reiterates the
importance of replacing escravo with escravizado (line 38).
 Jennifer Roth-Gordon & Antonio José B. da Silva
As part of their anti-racist teachings, black activists constantly engage in discur-

sive and ideological work to represent conflicts like this as two-sided battles between a
voice that ignores Brazilian racial inequality and another that is knowledgeable about
race and racism or likewise between a voice that uncritically repeats the dominant
representation of Brazilian society and history (what “we have always learned” in
line 22) and another that uses what is being taught in consciousness-raising classes
to offer contrasting representations (of quilombo residents as legitimate land own-
ers, for example).6 In examining metalinguistic commentaries such as these, in which
contrasting voices are made audible and recognizable as encapsulating ideological
opposites (e.g.: racists vs. anti-racists), we pay close attention to how contemporary
black activists stake out the boundaries of these voices (see also Inoue 2006). Indeed,
activists often reflected on these affirmative language practices, recognizing that
they commonly found themselves embroiled in “conflicting ways of speaking” (Hill
1995: 98). As Marta, an activist participating in the classes remarked: “Sometimes we
stick to the idea that we were slaves- we were enslaved, as those who are knowledge-
able say, but we were kings and queens in Africa.”7 We suggest that by recognizing the
translinguistic nature of these words, activists do more than just point to a drama of
conflicting viewpoints; they shape these perspectives into opposing personae, such
that one should be able to hear and identify the “racist” voice that is to be condemned
and, at the same time, be able to speak in the anti-racist voice that is to be embraced.
As Mannheim has noted of Hill’s work, a voice system does not “merely identify social
personages, but … identify social personages in a relational assemblage” (Mannheim
2008). The anti-racist depends on the figure of the racist for its very existence. Trans-
linguistic words thus allow black activists to embody the spirit of anti-racism.
Staking out the boundaries of the “racist” voice entails a complicated process
of pointing out the paradoxes of Brazilian national narratives – the ways in which
Brazilians of African descent are acknowledged in national history, often for their
cultural contributions (Davis 1999; Fry 1982; Santos 2005), and yet simultaneously
dismissed or delegitimized as bearers of full political rights and belonging (Holston
2008). At the core of Brazilian lore is the notion that brasilidade (Brazilianness) is
comprised of a mixture of African, European, and Indigenous influences that have
combined to create a country proud of its racial harmony. The majority of Brazilians
have historically maintained that there is a unique Brazilian national identity that is
shared by all. Due to this harmony, there is no need for separate racial identities or
. Along similar lines, rapper M.V. Bill liked to remark: “A gente não é marginal; a gente é
marginalizado” (We are not marginals/criminals; we are marginalized/criminalized).
. “Às vezes a gente ainda fica com aquela ideia de que a gente foi escravo. a gente foi
escravizado, como dizem aí os entendidos, mas a gente também era rei e rainha na na África.”
Double-voicing in the everyday language of Brazilian black activism 
political groups based on race – which were understood as a challenge to Brazilian

identity and national unity (Guimarães 1995). In their current struggles for politi-
cal and cultural representation, then, black activists seek to encourage participants to
identify themselves as part of the Afro-Brazilian population, to revalorize the place of
blacks in Brazilian society, and to interrogate the processes through which blacks have
been relegated to the bottom of the social hierarchy.
Ana, a consciousness-raising workshop leader for a different anti-racist organi-
zation that Silva observed, noted that participants are frequently unable to recognize
or describe many of the significant quotidian aspects of racism, such as systemic dis-
crimination in hiring and in other employment practices, racially motivated police
harassment, and racially constructed patterns of authority and deference. The over-
all goal of her course was to help participants critique the organization of Brazilian
society and to raise their political consciousness against racial stigmas, stereotypes,
discrimination, and inequality. In the lesson presented in Excerpt 3, Ana discusses
with students an academic article that critically analyzes the relationship between rac-
ism and the Brazilian press. She challenges the idea that black Brazilians played only
a secondary (and more passive) role in the construction of the Brazilian nation. Most
importantly, she encourages the class to problematize the discursive construction of
racial cooperation in national narratives.
Excerpt 3
1 Ana: 〈que〉 nós não contribuímos, 〈that〉 we didn’t contribute,
2 nós construímos esse país, we built this country,
3 né? didn’t we?
4 〈nós influenciamos〉 〈we influenced〉
5 〈como temos〉 〈as we have〉
6 as populações negras – the black populations
influenciaram influenced –
7 não influenciamos, we didn’t influence,
8 construímos. we built it.
9 construímos. we built it.
10 e construímos a partir de and we built it based on
conhecimento secularmente knowledge that we gained over
trabalhado, the centuries,
11 né? right?
12 não foi? isn’t it true?
13 〈não viemos né?〉 〈we didn’t come here right?〉
14 não viemos do processo we didn’t come here
de escravida-, during slaver-,
15 escravização, enslavement,
16 nem de diáspora a passeio. nor during diaspora on recreation.
17 né? did we?
 Jennifer Roth-Gordon & Antonio José B. da Silva
In the segment above, Ana gives examples of lexical items that illustrate the
ideologically weighted positions from which the history of Brazil is told and
portrayals of blacks are constructed. Through double-voicing (with deliber-
ate exaggeration for ironic effect) of the words contribuímos (we contributed)
in line 1, and influenciamos (we influenced) in line 4, Ana creates and simulta-
neously critiques the dominant voice that constructs “Brazilianness.” This racist
voice ignores and undermines the indispensable role that blacks have played in
building the Brazilian nation – a position acknowledged by the anti-racist voice
behind her correction construímos (we built it), in lines 2 and 8–10. Through
double-voicing, Ana parodies voices of the elite to highlight a discursive process
through which they minimize the critical role of blacks in Brazil. Ana alludes
to the fact that this dominant voice has imperceptibly impregnated the voice of
people who uncritically repeat these words. Through the correction construímos
(we built it), Ana gives voice to an alternative perspective, through which blacks
can challenge the racism of the Brazilian elite – those who have produced the
narrative of B razilian history that is found in textbooks and in common lore.
Towards the end of this excerpt, Ana also self-corrects her own use of the word
escravidão (slavery), changing it to escravização (enslavement) in lines 14–15.
Through the interplay of clashing voices, teachers emphasize how even a single
word (the translinguistic word, to use Bakhtin’s term) can be a powerful tool in
reproducing or fighting racism.
Black activists thus sought to teach Brazilians to distinguish between racist and
anti-racist positions in the everyday use of language. Discursive practices such as these
have been widely used in contemporary black activism as part of community projects
that are increasingly engaging race as a deliberate political strategy in their struggles
for social justice. These organizations (NGOs, networks, and forums) continue to
advocate strongly in support of racial quotas in higher education, to promote p ublic
health initiatives, particularly for black women, to publicly denounce police violence
against black male youth, and to work to increase the political participation and
representation of blacks in city and state councils and elected offices (Bairros 2008).
Present day black activism thus incorporates not only new ways of thinking and speak-
ing about race and racism, but also a vast array of race-based political strategies for
countering inequality and effecting change.
5. North American voices in Brazil?
Brazil’s new racial politics, in which Afro-descendants voice collective demands for
rights, remain a highly contested issue among the Brazilian public and B razilian
academics alike. While studies have shown tangible successes for these political
Double-voicing in the everyday language of Brazilian black activism 
strategies (Hooker 2008), strong criticism has led to legal challenges of these
policies, including multiple supreme court cases debating the legality of racial quotas
in higher education. Thus the future of these policies remains unclear (Sousa &
Nascimento 2008). Academics and intellectuals – who are deeply divided on these
issues – have been highly influential in these debates. On one side, “ multiculturalists” –
who tend to align with activists in the affirmation of distinct racial identities – view
race as a critical element in the formation of anti-discriminatory policies (Guimarães
2006; Oliveira 2004; Vargas 2006; Warren 2001). As Vargas emphasizes,.
historically-concerned, locally-based, internationally-connected, inclusive, trans-

formative Afrodiasporic social movements constitute, not only the strongest
antidote to the various forms of fascist [racist] hierarchies, but the best references
as we dream and work toward constructing … another country.
(Vargas 2006: 497)
Speaking from the other side of the debate, critics of race-based affirmative action
policies acknowledge that Brazilian racism exists, but they argue that it should be
combated through the rejection of the concept of race altogether – allowing Brazil
to turn their myth of racial democracy into a reality.8 Seeking to move the nation
towards a state of post-racialism, these “non-racialists” blame the multicultural-
ists for embedding ideologies of race in Brazilian public policy (Fry 2000; Fry et al.
2007; Sansone 2003). They believe that race-based policies head in the wrong direc-
tion to address racism, and they object to drawing on North American models that
have altered Brazilian laws in ways that are unprecedented and, according to them,
“un-Brazilian.” In a widely discussed polemic, Bourdieu and Wacquant offer the
. We wish to clarify here that the Brazilian non-racialist position is not the same as U.S.
colorblindness. The legality of U.S. affirmative action, for example, is being reconsidered by
the U.S. Supreme Court at the same time as it undergoes legal challenge in Brazil. In the
U.S., the dominant political positions on this issue (which do not necessarily fall along
the lines of the major political parties) include those who think that racism is a thing of
the past, rendering affirmative action unnecessary, and those who recognize the continued
existence of racial inequality (and may agree or disagree with affirmative action as a legal
governmental policy). In Brazil, non-racialists and multiculturalists both fall into the latter
group: That is, they recognize and seek to diminish racial inequality in Brazil, but they dis-
agree about whether race-based programs are the solution. Of course, many other Brazilians
see inequality as strictly a class issue, and therefore they identify neither with the multi-
culturalist nor with the non-racialist positions that we describe here. Note that this belief
in Brazil as a classist society is still different from colorblindness, which does not officially
acknowledge any form of discrimination and explains all inequality through a belief in
individual merit.
 Jennifer Roth-Gordon & Antonio José B. da Silva
“Americanization” of Brazilian race relations as a quintessential example of “cultural

imperialism:”
What are we to think, indeed, of those American researchers who travel to Brazil
to encourage the leaders of the Movimento Negro to adopt the tactics of the Afro-
American Civil Rights Movement and to denounce the category of pardo (an
intermediary term between branco, white, and preto, black, which designates
people of mixed physical appearance) in order to mobilize all Brazilians of African
descent on the basis of a dichotomous opposition between ‘Afro-Brazilians’ and
‘whites’ … (Bourdieu & Wacquant 1999: 47–48)
Like Brazilian black activists, critics of race-based programs and policies recognize
that language has become a critical site for anti-racist or multiculturalist efforts. They
also engage in overt and clearly metalinguistic discussions over how Brazilians should
talk about race. Bourdieu & Wacquant are not alone in their strong critique of the
introduction of a lexicon inspired by North American racial terms – one that is viewed
by critics as “foreign” and “imposed.” Kamel bristles at the proposal of a black-white
opposition that is framed in “a terminology that is not ours” (uma terminologia que
não era nossa) (2006: 20), and Fry suggests that such a sharp racial dichotomy not
only simplifies the vocabulary of Brazilians from multiple terms down to two, but also
imposes “the difficulty of making people let go of a way of life to which they are accus-
tomed” (a dificuldade de fazer com que as pessoas abram mão de um modo de vida ao
qual estão habituados) (1995/1996: 132). Non-racialists therefore dive into these meta-
linguistic debates – proposing their own linguistic strategies and rejecting others –
suggesting that one’s choice of words is more than mere semantics. Word choice is
also, to borrow Fry’s words, “a way of life.”
We are most interested, however, in how non-racialists also work to identify
particular voices and points of view that they pit in active opposition to each other.
Thus while the black activists we describe above – and the multiculturalists who
support them – render audible racist and “anti-racist” voices, non-racialists (who
decry race-based policies and reforms) embrace a divide between “North American”
and “Brazilian” voices. This discursive strategy has the effect of portraying particular
voices as less “authentic” than others. Just as black activists and multiculturalists
seek to undermine the voice of dominant Brazilian society, non-racialists attempt
to devalorize voices that embody North American perspectives. We suggest that
non-racialists thus engage Hill’s system of voices to negatively evaluate the contribu-
tions of North American researchers (Brazilianists) and activists associated with the
Brazilian black movement, who they claim speak in the voice of the foreigner. Fry
engages this strategy below, as he critiques Michael Hanchard, an African American
political scientist:
Double-voicing in the everyday language of Brazilian black activism 
Even in this text by Hanchard, who is one of the most sophisticated authors, we
find … racial group, race and racial difference without italics, without quotation
marks. Hanchard’s text, like so many others, is harmed even more by the fact
that many of the terms utilized to describe and analyze the Brazilian situation
in an article written in English and published in the United States are also native
categories of U.S. ‘identity politics.’ (Fry 1995/1996: 125)9
Fry objects to the introduction of “identity politics” into Brazil, but he voices his criti-
cism of this political strategy in linguistic terms. Fry stresses that Hanchard writes
“in English” and publishes “in the United States” (for a North American audience)
in order to make Hanchard’s voice recognizable as the outside voice of the “North
American.” Words such as “racial group, race and racial difference” – which Fry leaves
in English though he writes in Portuguese – are not only marked through his pro-
ductive use of linguistic contrast, but offered by Fry as clear indicators of Hanchard’s
North American voice. Here Fry “outs” the single-voiced manner in which Hanchard
uses these terms; his suggestion to include italics and/or quotation marks reminds
Fry’s readers that these are, in fact, double-voiced words that speak from a particular
point of view (to which one can, as he does, object). Non-racialists thus protest that
black activists and multiculturalists create race, racism, and racists10 in Brazil – in
part through their linguistic strategies of racial “empowerment.” Embracing Bakhtin-
ian analysis and Hill’s system of voices, we suggest that both sides engage in strate-
gies of double-voicing that create “objectified and finalized images of people” (Bakhtin
1984: 182) to support their respective political positions.
6. Conclusion
Through her intricate analyses of polyphony in daily discourse, Hill has illustrated how
language can become “a translinguistic battlefield, upon which two ways of speaking
struggle for dominance” (1985: 731). In her canonical study of an oral narrative by a
speaker of modern Mexicano, Hill (1995) meticulously examines Don Gabriel’s various
. “Mesmo neste texto de Hanchard, que é um dos autores mais sofisticados, encontra-se …
racial group, race e racial difference sem itálico, sem aspas. O texto de Hanchard, como
tantos outros, é prejudicado ainda mais pelo fato de que muitos dos termos utilizados para
descrever e analisar a situação brasileira num artigo escrito em inglês e publicado nos Estados
Unidos também são categorias nativas da ‘política de identidade’ dos Estados Unidos” (Fry
1995/1996: 125).
. Note Kamel’s (2006) response to the message of multiculturalists in his provocatively
titled book, “Não Somos Racistas” (We are not Racists).
 Jennifer Roth-Gordon & Antonio José B. da Silva
discourse strategies to show how a peasant consciousness is at least partially consti-

tuted through ongoing ideological struggle between peasant and capitalist ideologies.
This ideological struggle is constructed through the speaker’s codeswitching between
Spanish, which represents for him the language of profit, and Mexicano, which indexes
peasant values. Through his narrative of the death of his son, Don Gabriel positions
himself in relation to these two conflicting ideological views. In our investigation of
the metalinguistic debates over how to talk about race in Brazil, we have been struck
by how “two ways of speaking” are actively constructed by speakers in order to impose
order and uphold particular perspectives. Thus even though speakers invoke many
voices through the act of speaking, metalinguistic analysis (conducted by a linguistic
anthropologist, in the case of Hill) and/or metalinguistic discussion (by participants
themselves) reveals how speakers often organize this plurality of voices into two
ideological camps.
In this chapter, we have shown how youth inspired by politically conscious hip
hop and Brazilian activists associated with the black movement engage translinguistic
words to make speakers and listeners aware of a racist perspective associated with
dominant Brazilian society – a voice that they locate in history books, on census
forms, and in the many daily repetitions of nationalistic lore. This voice speaks of
racial harmony and national unity. In response, racially conscious Brazilians counter
with a voice of anti-racism – one that identifies this historical narrative as a perspec-
tive by and for the Brazilian elite. Activists can thus demonstrate their race conscious-
ness by speaking in a voice of racial empowerment and black pride as they transform
translinguistic words into the site of struggle between “racists” and “anti-racists.”
Speaking from a different vantage point, non-racialists loudly protest that this cre-
ates “racists” and racism in Brazil. And yet, through their metalinguistic challenges
of the black activist and multiculturalist political position, they too engage in active
double-voicing to “hear” North American perspectives that they deem less authentic
than contrasting “Brazilian” ones. The translinguistic battlefield now has two new
voices struggling for dominance. Each “side” has turned language to its purpose, not
only through the words it utters, but also through the distribution of voices that it has
called into being.
Our analysis draws on Hill’s inspiration to further the study of how speakers
embrace language to claim moral positions within situations of significant political
instability and change. Running through Hill’s scholarship is a commitment to study
the people and the daily linguistic practices that do not merely respond to larger politi-
cal economic contexts but actively produce these dynamics in the first place. We thus
engage in our own double-voicing in the title of our chapter as we have sought to
contribute not only to Hill’s study of voice systems, but also to her body of work on
the “everyday language” of racism (Hill 2008). In this, we take up Hill’s call to examine
how language plays an integral role in the construction of racial meaning and racism,
Double-voicing in the everyday language of Brazilian black activism 
often in “covert” ways that are revealed only through detailed linguistic analysis. We
would agree that Brazil offers a fascinating contrast to race relations in the United
States, and yet we have worked to critically interrogate how transnational racial “con-
trasts” and similarities are created, maintained, and made meaningful, in part through
linguistic forms that are made to represent coexisting, and often competing, perspec-
tives. If voices are the embodiments of various points of view, as Bakhtin and Hill have
so carefully demonstrated, then racial ideologies become audible and open to response
through the voices we hear.
Appendix
Transcription Conventions (adapted from Du Bois et al. 1993):

Each intonation unit appears on a separate numbered line
(–) a break of the intonational unit
(-) truncation of a word
X inaudible speech
[] overlap between utterances of two speakers
(.) final intonational contour
(,) continuing intonational contour
(?) an appeal to the listener for a response
(!) a higher than expected pitch on an accented word (a “booster”)
(…) long pause
(..) short pause
(0) no pause or (latching) between two speakers’ turns
@ pulse of laughter
(COUGH) non-verbal noise produced by speakers
〈words〉 “false starts”
〈X Y〉 marked quality or prosody of some sort
〈F F〉 loud speech
〈P P〉 soft speech
〈L L〉 slow speech
(( )) comment from the transcriber or researcher
[…] lines of the transcript have been omitted
References
Bairros, Luiza. 1996. Orfeu e poder. Afro-Ásia 17: 173–186.

Bairros, Luiza. 2008. A community of destiny: New configurations of racial politics in Brazil.
Souls 10(1): 50–53.
Bakhtin, Mikhail Mikhailovitch. 1968. Rabelais and His World. Cambridge MA: The MIT Press.
Bakhtin, Mikhail Mikhailovitch. 1981. The Dialogic Imagination, Caryl Emerson & Michael
Holquist (transl.). Austin TX: University of Texas Press.
 Jennifer Roth-Gordon & Antonio José B. da Silva
Bakhtin, Mikhail Mikhailovitch. 1984. Problems of Dostoevsky’s Poetics, Caryl Emerson (ed. and
transl.). Minneapolis MN: University of Minnesota Press.
Bourdieu, Pierre & Wacquant, Loïc. 1999. On the cunning of imperialist reason. Theory,
Culture & Society 16(1): 41–58.
Costa, Emília Viotti da. 1985. The Brazilian Empire: Myths and Histories. Chicago IL: University
of Chicago Press.
Davis, Darién J. 1999. Avoiding the Dark: Race and the Forging of National Culture in Modern
Brazil. Aldershot, England: Ashgate.
Du Bois, John W., Schuetze-Coburn, Stephan, Cumming, Susanna & Paolino, Danae. 1993.
Outline of discourse transcription. In Talking Data: Transcription and Coding in Discourse
Research, Jane A. Edwards & Martin D. Lampert (eds), 45–89. Hillsdale NJ: Lawrence
Erlbaum Associates.
Frazier, E. Franklin. 1942. Some aspects of race relations in Brazil. Phylon (1940–1956) 3(3):
249+287–295.
French, Jan H. 2009. Legalizing Identities: Becoming Black or Indian in Brazil’s Northeast. Chapel
Hill NC: The University of North Carolina Press.
Fry, Peter. 1982. Para Inglês Ver: Identidade e Política na Cultura Brasileira. Rio de Janeiro: Zahar
Editores.
Fry, Peter. 1995/96. O que a cinderela negra tem a dizer sobre a ‘política racial’ no Brasil. Revista
da USP 28: 122–135.
Fry, Peter. 2000. Politics, nationality, and the meanings of “race” in Brazil. Daedulus 129(2):
83–118.
Fry, Peter, Maggie, Yvonne, Chor Maio, Marcos, Monteiro, Simone & Ventura Santos, Ricardo.
2007. Divisões Perigosas: Políticas Raciais no Brasil Contemporâneo. Rio de Janeiro:
Civilizção Brasileira.
Garcia, Januário. 2006. 25 Anos 1980–2005: Movimento Negro no Brasil. Brasilia DF: Fundação
Cultural Palmares.
Guimarães, Antonio Sérgio Alfredo. 1995. Racism and anti-racism in Brazil: A post modern
perspective. In Racism and Anti-Racism in World Perspective, Benjamin P. Bowser (ed.),
208–226. Newbury Park CA: Sage.
Guimarães, Antonio Sérgio Alfredo. 2004. Preconceito de cor e racismo no Brasil. Revista de
Antropologia, USP 47(1): 9–43.
Guimarães, Antonio Sérgio Alfredo. 2006. Depois da democracia racial. Tempo Social: Revista
de Sociologia da USP 18(2): 269–287.
Hanchard, Michael. 1991. Racial consciousness and Afro-diasporic experiences: Antonio
Gramsci reconsidered. Socialism and Democracy 7(14): 83–106.
Hanchard, Michael George. 1994. Black cinderella? Race and the public sphere in Brazil. Public
Culture 7(1): 165–185.
Hanchard, Michael George. 1998. Orpheus and Power: The Movimento Negro of Rio de Janeiro
and São Paulo, Brazil, 1945–1988. Princeton NJ: Princeton University Press.
Hasenbalg, Carlos Alfredo. 1979. Discriminação e Desigualdades Raciais no Brasil. Rio de
Janeiro: Graal.
Hasenbalg, Carlos A. & Silva, Nelson do Valle. 1988. Estrutura Social, Mobilidade e Raça. Rio de
Janeiro: Instituto Universitario de Pesquisas do Rio de Janeiro, Vertice.
Hill, Jane H. 1985. The grammar of consciousness and the consciousness of grammar. American
Ethnologist 12: 725–737.
Hill, Jane H. 1986. The refiguration of the anthropology of language. Cultural Anthropology 1(1):
89–102.
Double-voicing in the everyday language of Brazilian black activism 
narrative. In The Dialogic Emergence of Culture, Dennis Tedlock & Bruce Mannheim (eds),
97–147. Urbana IL: University of Illinois Press.
Holston, James. 2008. Insurgent Citizenship: Disjunctions of Democracy and Modernity in Brazil.
Princeton NJ: Princeton University Press.
Hooker, Juliet. 2008. Afro-descendant struggles for collective rights in Latin America: Between
race and culture. Souls 10(3): 279–291.
Inoue, Miyako. 2006. Vicarious Language: Gender and Linguistic Modernity in Japan. Berkeley
CA: University of California Press.
of Language: Ideologies, Polities, and Identities, Paul V. Kroskrity (ed.), 35–83. Santa Fe NM:
Kamel, Ali. 2006. Não Somos Racistas: Uma Reação aos que Querem nos Transformar numa
Nação Bicolor. Rio de Janeiro: Editora Nova Fronteira.
Kroskrity, Paul V. 2011. Facing the rhetoric of language endangerment: Voicing the conse-
quences of linguistic racism. Journal of Linguistic Anthropology 21(2): 179–192.
Mannheim, Bruce. 2008. The voices of Jane Hill. Paper presented at the Annual Meetings of the
American Anthropological Association, San Francisco CA.
Miranda-Ribeiro, Paula. 2006. Somos racistas. Revista Brasileira de Estatística Populacional
23(2): 375–377.
Morson, Gary Saul & Emerson, Caryl. 1990. Mikhail Bakhtin: Creation of a Prosaics. Stanford
CA: Stanford University Press.
Munanga, Kabengele & Gomes, Nilma L. 2004. Para Entender o Negro no Brasil de Hoje: História,
Realidades, Problemas e Caminhos. São Paulo: Global.
Munanga, Kabenguele. 2010. Nova legislação e política de cotas desencadeariam ascensão
econômica e inclusão dos negros, diz professor. Pambazuka News. <http://pambazuka.org/
pt/category/features/62676> (1 March 2010)
Oliveira, Luís R.C. 2004. Racismo, direitos e cidadania. Estudos Avançados 18(50): 81–93.
Pardue, Derek. 2008. Ideologies of Marginality in Brazilian Hip Hop. New York NY: Palgrave
Macmillan.
Rodrigues, Cinthia. 1998. O rap sai do gueto. Época Edição 12, 10/08/98.
Roth-Gordon, Jennifer. 2012. Linguistic techniques of the self: The intertextual language of
racial empowerment in politically conscious Brazilian hip hop. Language & Communica-
tion 32(1): 36–47.
Roth-Gordon, Jennifer. 2013. Racial malleability and the sensory regime of politically conscious
Brazilian hip hop. The Journal of Latin American and Caribbean Anthropology.
Sansone, Livio. 2003. Blackness without Ethnicity: Constructing Race in Brazil. New York NY:
Palgrave Macmillan.
Santos, Jocélio Teles dos. 2005. O Poder da Cultura e a Cultura no Poder: Disputa Simbólica da
Herança Cultural Negra no Brasil. Salvador: EDUFBA.
Sheriff, Robin E. 2000. Exposing silence as cultural censorship: A Brazilian case. American
Anthropologist 102(1): 114–132.
Sheriff, Robin E. 2001. Dreaming Equality: Color, Race, and Racism in Urban Brazil. New
Brunswick NJ: Rutgers University Press.
Silva, Antonio José B. da. 2011. Dialogism in the Brazilian black movement’s affirmative lan-
guage practices. Paper presented at the Annual Meetings of the American Anthropological
Association. Montreal, QC, Canada.
 Jennifer Roth-Gordon & Antonio José B. da Silva
Silva, Denise Ferreira da. 1998. Facts of blackness: Brazil is not (quite) the United States … and
racial politics in Brazil? Social Identities 4(2): 201–234.
Sousa, Leone C. de. & Nascimento, Paulo. 2008. Brazilian national identity at the crossroads:
The myth of racial democracy and the development of black identity. International Journal
of Politics, Culture and Society 19: 129–143.
Stephens, Thomas M. 1989. Dictionary of Latin American Racial and Ethnic Terminology.
Gainesville FL: University of Florida Press.
Twine, France Winddance. 1998. Racism in a Racial Democracy: The Maintenance of White
Supremacy in Brazil. New Brunswick NJ: Rutgers University Press.
Vargas, João H.C. 2006. Black radical becoming: The politics of identification in permanent
transformation. Critical Sociology 32(2–3): 475–499.
Warren, Jonathan W. 2001. Racial Revolutions: Antiracism and Indian Resurgence in Brazil.
Durham NC: Duke University Press.
Warren, Jonathan W. & Twine, France W. 2002. Critical race studies in Latin A merica: Recent
advances, recurrent weaknesses. In A Companion to Racial and Ethnic Studies, David Theo
Goldberg & John Solomos (eds), 538–560. Malden MA: Blackwell.
Woolard, Kathryn A. 1999. Simultaneity and bivalency as strategies in bilingualism. Journal of
Linguistic Anthropology 8(1): 3–29.
Uptake (un)limited
The mediatization of register shifting and the
maintenance of standard in U.S. public discourse*
Deborah Cole & Régine Pellicer

The University of Texas–Pan American
This chapter analyzes the “language panic” (Hill 2008) following Hillary Clinton’s
register-shifting performance of the gospel song “I Don’t Feel No Ways Tired”
during the U.S. presidential campaign in 2007. We observe that mediatization
(Agha 2011b) creates and maintains the conditions by which some messages
and uptake patterns remain unavailable to wider audiences while others are
continuously recycled and increasingly accessible. We argue that the maintenance
of unequal divisions of semiotic labor can be facilitated by mediatization as
currently practiced. We observe that value projects attached to mediatized
fragments work to maintain the hierarchy of perduring semiotic registers (Goebel
2010) in U.S. public discourse in which Standard (Hill 2008) continues to
dominate all others.
Keywords: Black preaching style; register shifting; Standard
* An earlier version of this chapter appears in Language in Society 41(4), copyright 2012.
Reprinted with permission. Some of the thinking for this chapter began in preparation for
the “I ♥ Jane Hill” conference in Tucson to celebrate Jane’s retirement from the University
of Arizona in 2009. We appreciate the engaged feedback from the audiences who responded
to versions of this paper presented at the annual meetings of the American Anthropological
Association in November 2010 and the applied linguistics colloquium at the University of
Texas-Pan American in the fall of 2010. We extend our thanks to Shannon Bischoff, Amy
Fountain Minhee Eom, Barbara Johnstone, Bryan Meadows, Xiaojing Sheng, and anony-
mous reviewers from the Journal of Language in Society for their helpful comments and
positive evaluations of our manuscript. This piece would have never existed if it hadn’t
been for Jamee Cole, who kept up with the U.S. news while living in Indonesia in 2007 and
noticed the incongruity between the media’s representations and the audience’s response
to Clinton’s speech in Selma.
 Deborah Cole & Régine Pellicer
1. Introduction
To ignore the performativity of value projects, their felicity conditions and

fragility under conditions of uptake, is to allow anxieties about hegemony to
infect moments of decontextualized reflection, even as our actual conduct
remakes what we fear. (Agha 2011a: 28)
In the spring of 2007, Barack Obama and Hillary Clinton were contending for the
nomination of the Democratic Party during the U.S. presidential campaign that ended
with the election of Barack Obama in November 2008. In competing for s upport from
African American voters, Clinton and Obama both gave speeches on Sunday, March
4, 2007 at churches in Selma, Alabama. Congregations in Selma would later that day
participate in a march that would commemorate the forty-second anniversary of the
Bloody Sunday march led by Rev. Dr. Martin Luther King, Jr. in 1965. Although playing
at reproducing others’ voices in the United States can be a risky business, readily inter-
preted as a negative form of stereotyping in public discourse, both candidates made
stylistic choices in the delivery of their speeches to adopt elements of black preaching
style (Britt 2011). Although the local congregations favorably evaluated their speeches
and use of this style, the evaluations of Clinton’s speech that became widely available
through processes of mediatization (Agha 2011b) generally delegitimated (Bucholtz
& Hall 2004) her performance, calling it “fake”, “pandering”, “mocking”, “laughable,”
“fraudulent,” “shameless,” and even “racist”.
The following discussion aims to clarify a contradiction that arises in the enregis-
terment of dialectal difference in American English. Speakers who have linguistically
inherited (Leung, Harris & Rampton 1997) a language variety other than Standard
(Silverstein 1998), and in using it habitually perform racially or ethnically marked
identities, are encouraged and expected to acquire Standard as well as perform its
indexically linked unmarked identity in public space (Urciuoli 1996; Lippi-Green
2012; Urciuoli 2003). The converse is not true, however. Citizens who typically per-
form a ground (Agha 2003) identity, signaled by the habitual speaking of standard
versions of American English, and thereby perform racially unmarked identities in
the public sphere (Urciuoli 2003; Mendoza-Denton 2008; Meadows 2010; Lippi-Green
2012) are discouraged from acquiring and performing varieties other than Standard.
Repertoire expansion of the first type (from others to include Standard) is evaluated
as a “willingness to assimilate” or as a necessary if unfortunate process that politi-
cally and economically disadvantaged people undergo to gain access to national and
global resources (Urciuoli 1996; Ullman 2004; Lippi-Green 2012). Meanwhile, the
performance of a repertoire expansion of the second type (from Standard to include
others) can be variously evaluated as “inappropriate”, “inauthentic”, “condescending”,
“mocking”, “racist” or as a failure to recognize group boundaries (Cutler 2003;
Uptake (un)limited 
Hill 2008; Schwartz 2008).1 We demonstrate how mediatization can facilitate the
maintenance of this contradiction.
Mediatization, or the institutional practices “that reflexively link processes of
communication to processes of commoditization” (Agha 2011a: 163) relies on the abil-
ity of users to identify and detach fragments of communicative behavior from their
semiotic contexts and recycle them into new ones. We apply this theoretical insight to
the mediatization of Clinton’s speech to demonstrate how returning our gaze to the prior
contexts from which mediatized fragments are detached reveals how m ediatization
limits the opportunities for uptake of some value projects more than it does for others.
In what follows, we pay attention to the performativity of the value projects of Stan-
dard and Other than Standard American Englishes in a case where register shifting is
involved. We note that felicity conditions on uptake can be different in a performed
fragment’s pre-mediatized environment than in its post-mediatized one. Defined most-
simply by Agha as “an act from which other acts can follow” (Agha 2011a: 167), uptake
refers to a kind of perception or awareness of an isolatable or identifiable piece (frag-
ment) of semiotic behavior that can lead to the recycling or reinterpretation of the frag-
ment. Blommaert (2003: 616) talks about uptake as something that must “be granted
by others, on the basis of the dominant indexical frames and hierarchies” [emphasis in
original]. We also notice that some value projects are more fragile than others when
mediatization is one of the conditions of uptake, as mediatization delimits the frag-
ments and voices that can become more widely available for future use.
In examining the particular fragments, voices, and value projects at stake in our
data, we take on several expositional goals. First, we demonstrate how a theory of
mediatization makes possible a critique of the practices and behaviors that enable
and maintain unequal representational economies (Keane 2002) on increasingly global
scales. Second, by applying Jane Hill’s work on the ideologies underpinning the pro-
liferation of Standard (Hill 2008) to a case where personalism failed, we demonstrate
how Standard’s dominant position could be maintained when referentialism came
to its defense. Finally, we hope to demonstrate some practical implications of recent
sociolinguistic theory for shifts in the disciplinary practices of sociolinguistic inquiry,
shifts that could point to a way out of our propensities to participate in the remaking
of familiar hegemonies.
. Further, how one might go about acquiring an Other than Standard variety is less
explicitly articulated in public or scholarly discourse and has little if any institutional support.
For example, there are no widely circulated grammars of or courses on how to do this,
though Johnstone’s work on locally produced resources for local ways of speaking English in
Pittsburgh (e.g. Johnstone 2011) may point to changes in the availability of such educational
materials.
 Deborah Cole & Régine Pellicer
2. B
ackground: (Semiotic) register-shifting, role alignment,
and construable contexts
As sociolinguistic scholarship in the United States (like that cited above) has dem-
onstrated again and again, in numerous contexts, and with multiple language vari-
eties, the current structure of the representational economy enacted in U.S. public
discourse relies upon the maintenance of an evaluative opposition between two
semiotic registers (Agha 2007). In particular, the semiotic register (SR) evoked by
standard varieties of American English is authorized (Bucholtz & Hall 2004) while
the one evoked by other-than-Standard varieties of English (as well as by the use
of other languages) is delegitimated (Bucholtz & Hall 2004). Here, we will refer to
the SR evoked by the truncated repertoires (Blommaert 2010) that become enreg-
istered as Standard American English(es) as SAME. In much U.S. institutionally-
situated public discourse, SAME is opposed to the SR evoked by all other linguistic
repertoires (Lippi-Green 2012; S ilverstein 1998; Hill 2008), which includes all the
other language skills and practices that become enregistered as the many dialects
and languages other than Standard. We call these other-than-standard repertoires
LOTS.2 Our categorization of this wide variety of linguistic repertoires and semiotic
behaviors into two opposing SRs is not meant to obscure the descriptively apparent
grammatical and pragmatic diversity present in each. Rather this categorization is
meant to refer to the well-documented semiotic opposition that emerges both prac-
tically and metadiscursively in U.S. public discourse as a particular, well-known and
often recycled value project (Agha 2011a). A summary of the signs and functions that
constitute SAME and LOTS appear in Table 1.3
Table 1. Semiotic registers (SAME and LOTS) and their constellations of signs
SAME LOTS
Standard English, American citizenship, Regional dialects and accents, “foreign”
nation, formality, objectivity, rationality, languages (Spanish, Chinese, etc.), intimacy,
education, upward mobility, phonological family, ethnicity, talk about personal life
patterns of elites, sociolinguistically “neutral” worlds, distinctive prosodic and gestural
patterns, sociolinguistically “distinctive”
. Here we explicitly draw on Blommaert’s (2010) suggestion that we abandon the term
“language” in favor of “truncated repertoires” to more accurately refer to the abilities of users
and the nature of the phenomena we study. Also, as disambiguation of /l/ and /ɹ/ does not
occur syllable initially in several varieties of English spoken internationally, we leave the
choice of pronunciation of the name for this SR as /lats/ or /ɹats/ to the reader.
. We follow Goebel’s presentations of SR models in his work on Indonesian and Languages
Other Than Indonesian (Goebel 2008, 2010).
Uptake (un)limited 
Critical investigations of ideology and practice surrounding shifts between these

registers, as studies of codeswitching or style shifting for example, have been well
established (e.g. Valdés-Fallis 1978; Zentella 1997; Wolfram & Schilling-Estes 2006).
This field of scholarship has provided robust evidence of the performativity of iden-
tity categories and helped to enable the identification of a set of common tactics of
intersubjectivity (Bucholtz & Hall 2004), or universally available, evaluative, semiotic
processes by which membership in sociolinguistically-salient groups is performed.
Where these studies have been concerned with speaker-hearer evaluations, they have
tended to focus on evaluations of switches out of LOTS and into SAME as performed
by linguistic inheritors of LOTS. Such switches tend to be evaluated positively by lin-
guistic inheritors of SAME (Baugh 2000; Chun 2004), especially when the performer
is able to maintain strict linguistic partitioning between the distinguishing features of
different SRs (Urciuoli 2003). These same types of switches can be evaluated negatively,
however, by other heritage speakers of LOTS, who may perceive such a codeswitching
facility as a rejection of local identity-based values (Mendoza-Denton 2008). Work
that has examined switches into LOTS by heritage speakers of SAME as studies of
crossing (the performance of linguistic forms not typically indexical of the perform-
er’s ground identity), on the other hand, are less numerous, and have tended to find
that this type of switching tends to be negatively evaluated by hearers (Cutler 2003).
Related studies of the use of some elements of LOTS linguistic repertoires by heritage
speakers of SAME in studies of mocking (like crossing but involving a definite negative
evaluation of the performed forms and linked identities) find that heritage speakers
of LOTS negatively evaluate these types of shifts too (Hill 2008; Schwartz 2008; Alim,
Lee & Carris 2010).
That performance shifts of the first type (out of an other-than-Standard repertoire
and into a Standard one) should be authorized and those of the second type delegiti-
mated within any particular structured representational economy where Standard is
present is not necessary or natural, however. This was noted four decades ago when
Labov wrote about evaluations of the uptake of standard forms by “lames” within
communities in the U.S. where Black English vernacular was also spoken (Labov
1973), and it has been re-demonstrated very clearly in recent work on switches out
of S tandard Indonesian and into Languages (and dialects) Other Than (Standard)
Indonesian (Goebel 2010; Smith-Heffner 2007, 2009; Cole 2010). Neither is the par-
ticular evaluative configuration of semiotic registers heretofore robustly exemplified
in the U.S. case privileged theoretically. Given these facts, perhaps one reason why
examples of the first type have served as the primary source of evidence for under-
standing linguistic differentiation in the U.S. is alluded to in Agha’s (2005) piece
“Voice, Enregisterment, Footing”:
Effects of register token use are not always consistent with the stereotypic values
associated with the register’s form types. This flies in the face of a common folk
 Deborah Cole & Régine Pellicer
t heory about registers, a kind of folk assumption of contextual invariance, typically

subscribed to by language users and often adopted uncritically by linguists as
well. Taken very strictly, this view implies that the construable context, or co-text,
of any particular token use is always irrelevant to the overall construal of that use.
(Agha 2005: 47)
The fact that context is always relevant for construing language in use is well
documented and firmly established (Duranti & Goodwin 1992; Bucholtz & Hall 2008).
Yet the fact that context remains relevant to evaluations of register token use is some-
how easy to forget. We are concerned with why this is so and with how this kind of
forgetting (Mohamad 2002) about contextual variance is linked to erasures (Irvine &
Gal 2000) of adequative evaluations (Bucholtz & Hall 2004) of register-shifts out of
SAME and into LOTS.
The Clinton speech we examine below is interesting analytically because the same
register shifting performance was evaluated differently depending on who was doing the
construing in which context. The pre-mediatization audience, the congregation at the
Selma church, evaluated Clinton’s register token use as symmetric role alignment (Agha
2005), i.e. they had a positive evaluation of the characterological figure she p resented in
her message against the ground of her self-identity. But the producers and consumers of
mainstream electronic media tended to evaluate Clinton’s performance as asymmetric
role alignment (Agha 2005). In mediatized contexts, language users produced negative
evaluations of the same register token use, consistent with stereotypic values. That the
same token use can be used in different value projects is possible, Agha tells us, because
semiotic processes involving voicing phenomena occur simultaneously at two levels. At
the tier of entextualized individuation, we simply notice the metrical or acoustic differ-
ences between the voices or personae being performed (that person says /a:/ where I
would say /aɪ/, for example). At the tier of descriptive identification and characterization,
the tier to which role-alignment applies, we go on to describe and classify these differ-
ences, (“she must be from Texas,” or “that accent sounds sweet”, etc.). Thus, the same
physical phenomenon indexed by a decontextualized fragment can be classified and
described differently in circulation as part of varying language ideological structures.
For us, the most striking aspect of the language panic (Hill 2008) that followed
Clinton’s speech in 2007 is that negative, asymmetric, evaluations of her register-shifting
performance were widely circulated through mediatization despite the obviously posi-
tive, symmetric, evaluations of her performance by her intended, f ace-to-face audience
at the church in Selma. To better understand how this happened, we take an ethno-
graphically informed approach to illuminating the construable context from which the
register-shifting fragment that interests us was lifted for use in mediatized value proj-
ects. In doing so, we provide further evidence of the observation that “an account of
mediatization is…an account of the social processes that the media c onstruct obscures”
(Agha 2011b: 164). In addressing ourselves to the question “What happened to the
Uptake (un)limited 
Selma congregation’s symmetric evaluations?”, we demonstrate how the systematic era-

sure of other-than-Standard voices in the U.S. public sphere includes the obscuring of
how these voices recognize and evaluate public discourse, a process that enables the
advancement of the familiar value project in which Standard (and its indexable figures
of personhood) occupy the highest position in a structured SR economy.
3. Clinton’s black preaching style and reconstructing

congregational uptake
The conventional idea that meaning resides in the speaker, and that interpretation
is thus a process of correctly recovering a speaker’s intentions, is rooted in a
characteristically Western way of attributing internal mental states to others. It
has been repeatedly observed that audiences are always in one way or another
“coauthors”…, sometimes contributing to the construction of form…, sometimes
to the determination of meaning. (Johnstone 2000: 406)
In this section, we describe the formal properties of the mediatized fragment that
became widely available for uptake during the U.S. presidential campaign in 2007:
the C-SPAN recording that documented Hillary Clinton’s speech in Selma (C-SPAN
2007a).4 In particular, we pay attention to the voicing structures and participant
frameworks that Clinton took up in her speech. We will see that although the church
congregation ratified the range of uptake behaviors Clinton performed, fragments of
her almost nineteen-minute long speech that were later made most widely a vailable
through mediatized recycling limited the range of Clinton’s repertoire expansion
available to media audiences and “erased” the congregation’s ratifications. We begin
with a brief review of the descriptive scholarship on black preaching style (Britt 2011),
which will serve as a model for evaluating the degree to which Clinton can be read to
have entered into the participation frameworks appropriate for this speech event. We
then turn to the congregation’s uptake of her register shifting, reconstructing their
ratification from the same mediatized fragment from which we can observe Clinton’s
performance.
. C-SPAN stands for Cable-Satellite Public Affairs Network. It is a network dedicated to inform
the general public of political talks and hearings taking place in the U.S. Congress. C-SPAN
also covered the 2008 presidential campaign and recorded public speeches of the main con-
tenders for the election. The video used for Method 1 is available at http://www.c-spanvideo.org/
program/196941-1. Fragments from this video were recycled on other networks, for example by
John Stewart on The Daily Show and Wolf Blitzer on CNN. We used a longer video which also
included Obama’s speech 〈http://www.c-spanvideo.org/program/196943-1〉 to count audience
responses to Obama’s and Reverend Lowery’s speeches under Method 2.
 Deborah Cole & Régine Pellicer
3.1 Black preaching style

The sociolinguistic characteristics of the semiotic register Clinton shifted into at the
Selma First Baptist Church have been previously well described. Recently, Britt (2011)
has called the coherent set of linguistic practices commonly performed in churches
belonging to African American communities in the U.S. black preaching style (BPS).
A review of previous studies of the patterns and practices that typify this linguistic
repertoire reveals that many of the defining characteristics of this register can be gen-
erally categorized as requirements on dialogism and explicit intertextuality. One of the
most salient requirements observable by researchers of this style is that its successful
performance relies on the ability of the participants (the preacher and the members
of the congregation) to co-construct the discourse event of the church service. This
is true even of the sermon, the central text of church service events, which though
technically “delivered” by the preacher depends crucially for its success and effect
on the audience’s sustained audible response. The requirement that the preacher and
the audience co-construct the event is formally realized within the sermon’s “call and
response” structure (Wharry 2003; Britt 2011). Successful participation by the audi-
ence or congregation requires participants to not only actively attend to the content
of the sermon, but to be acutely attuned to the patterns of the preacher’s delivery that
signal when and for how long they will have turns to respond. Silence on the part of
the congregation is interpreted by participants as a lack of communication between
the preacher and the audience, audience boredom, or audience disagreement with the
preacher (Wharry 2003), so much so that participants characterize church services
that “lack lively interaction” as “dead” (Zeigler 2001). The dialogic requirement in BPS
is also evident in joke telling and in “voice merging”, whereby the preacher incor-
porates the voices of previous orators or texts as they perform their pulpit identities
(Mitchell 1970; Vail 2006; Britt 2011).
Other types of intertextuality also define the genre of the sermon. The sermon’s
content must interleave sacred themes with secular ones, which are directly relevant
to contemporary people and their needs (Wharry 2003). Biblical references are used
to frame the sermon as divinely inspired (Wharry 2003; Britt 2011), and Gospel music
is used to motivate social action (Barnes 2005). Throughout the performance, the lin-
guistic styles and discourse practices common to the congregation’s daily experience
(outside of the context of the church service) must be apparent (Mitchell 1970; Wharry
2003). The use of lexical and syntactic parallelism to link later utterances to prior ones
is prevalent (Wharry 2003; Britt 2011), and each particular sermon is linked to oth-
ers in its genre through formal conventions that require sermons to cohere around a
central theme (with textual boundary markers to signal conceptual changes within the
text) and then elevate to a textual climax following a gradual rise in intonation and
volume at the end of the sermon (Wharry 2003).
Uptake (un)limited 
3.2 M
ethod 1: Listening to Clinton’s and the congregation’s
co-performance ethnographically
The publicly available C-SPAN recording of Clinton’s speech documents that she
was well aware of these elements of style and actively adopted them. To demon-
strate the degree to which this is true, we watched the entire speech and docu-
mented when and where Clinton performed elements of BPS. As the recording
also documents the congregation’s responses (in so far as their voices are audible
on the recording), we also listened for their participation in the speech event and
cataloged where and for how long their voices are audible. The time markers used
in the rest of this section refer to those on the C-SPAN recording (C-SPAN 2007a),
which begins as Clinton takes the pulpit and ends 18.46 minutes later. The dialogic
co-construction of her speech begins the moment the sound file fades in to join
the video feed of Hillary behind the podium flanked by her hosts and faced by a
standing and clapping audience (0.03). Their c lapping lasts for 15 seconds, during
which the amplitude of the recording reaches a maximum level with the audience
voices cheering and calling. Her first words, uttered over the roar of the audience
are “thank you”, which she repeats three times and to which someone calls out, /la:d
blɛs hɪlare/ (Lord bless Hillary!). By (0.20), the backs of people’s heads, visible up to
this point, disappear as the audience sits down, and Clinton opens with the lyrics
from a well-known hymn: “This is the day the Lord has made”. By the time she
utters the word “made”, audience voices have risen loudly, and they continue to call
as she utters the song’s second line: “Let us rejoice and be glad in it”. She continues
with her “thank yous”, including the thanking of “the First Baptist Church family
for opening [their] hearts and [their] home to [her]” (0.47), and then she moves
into her first joke in which she acknowledges the religious and racial differences
between herself and her audience. In the following transcript, audible audience
participation is in ((parentheses)).
Transcript 1 – Opening joke

(0.55) And I have to confess that I did seek dispensation from Reverend Armstrong
to come. Because you know, I’m a Methodist ((audience laughter)), and I’m in one of
those mixed marriages ((audience laughter)). And my husband who sends greetings to
all of you today ((Alright!)), felt it necessary to call the Reverend to make sure that that
was alright ((laughter and clapping)). And thank you Reverend for being so broad-
minded and understanding ((laughter and clapping)) (1.27).
Within the first minute and half, Clinton has successfully performed several of
the key elements of BPS, including quoting gospel music, telling a joke, and eliciting
active audience participation. Following these introductory remarks, Clinton goes
on to introduce the march theme [at (4.31) and (5.40)] and to take up contemporary
 Deborah Cole & Régine Pellicer
secular issues [like Bloody Sunday marches past (5.10) and present (10.39), Hurricane
Katrina (9.30), healthcare reform (10.08), and voter discrimination against African
Americans (15.31)]. Her use of syntactic parallelism in framing these issues is evident
throughout as is her ability to use intonation to highlight the parallelism and elicit
audience participation. Transcript 2 begins immediately after Clinton invokes Rev.
Martin Luther King Jr.’s voice (at 9.53) and to which someone calls out “Speak! Speak!”.
This transcript illustrates Clinton’s use of parallelism and documents her choice of a
central theme, i.e. “the march”. The word “march” occurs nine times in her speech, and
Clinton uses the phrase “we have a march to finish” to mark four s ectional transitions
(at 7.25; 16.32; 16.56; 18.04). In the transcript, HC refers to Hillary Clinton and A to the
audience. Other transcription conventions are provided in the appendix.
Transcript 2 – Parallelism, contemporary issues, march theme

1 HC: (10.02) Well I’m here to tell you
2 HC: poverty and growing inequality matters.
3 A: ((slight applause))
4 HC: healthcare matters.
5 HC: the people of the Gulf Coast and New Orleans /nuw.ór.lɪnz/ matter.
6 A: ((applause and cheering begin))
7 HC: OUR SOLDIERS MATTER.
8 HC: OUR STANDING IN THE WORLD MATTERS.
9 HC: OUR FUTURE MATTERS
10 HC: AND IT IS UP TO US TO TAKE IT BACK PUT IT INTO OUR HANDS
11 HC: START MARCHING TOWARD A BETTER TOMORROW (10.27)
12 A: (applause and cheering last until 10.42)
We turn now to the climax of Clinton’s speech, which occurred during the final two
minutes and did indeed follow a gradual rise in intonation and volume. In Transcript 3,
several of the register-specific elements of BPS converge. Clinton signals a textual tran-
sition by re-invoking the march theme (line 2) and returns to the gospel lyrics with
which she opened her speech (line 3). She directs the audience to participate with her
in voice merging (line 4) and tells them which voice to perform (line 5). In line 6, she
begins reciting the lyrics from another gospel song and this time also performs some
grammatical elements of southern dialects of U.S. English (monophthongization in
“I” and “tired” in line 4 and final consonant deletion on “don’t” and “I’ve” in lines 6
and 8), following the stylistic requirement that patterns of daily speech be present
in the sermon. These are displayed in phonetic transcription. In lines 6 through 10,
Clinton is quoting the lyrics of a hymn written and performed by the gospel singer and
songwriter James Cleveland.
Uptake (un)limited 
Transcript 3 – Climax, gospel music, voice merging, and style-shifting (16.56)

1 HC: we have to stay awake
2 HC: we have a march to finish.
3 HC: on this Lord’s day
4 HC: let us say with one voice.
5 HC: the words of James Cleveland’s great freedom hymn.
6 HC: I don’t /a: don/ feel no ways tired /ta:jrd/?
7 A: ((cheering and clapping begin))
8 HC: I’VE COME /a: kʌm/ TOO FAR /fa:r/ FROM WHERE I STARTED FROM.
9 HC: NOBODY TOLD ME THAT THE ROAD WOULD BE EASY?
10 HC: I DON’T BELIEVE HE BROUGHT ME THIS FAR TO LEAVE ME.
11 A: (17.30) ((cheering and clapping))
12 HC: (17.34) AND WE KNOW
13 HC: WE KNOW
14 HC: WE KNOW IF WE FINISH THIS MARCH WHAT AWAITS US.
(both speech and clapping end at 17.58)
During and following this section of her speech, Clinton receives one of her six stand-
ing ovations. The one audience member’s face visible on the podium behind her can
be seen nodding and smiling as she recites. At 17.30, Clinton had finished reciting
the lyrics and had tried to move on, but the crowd was still clapping and calling eight
seconds later when they sit down. Clinton presses on to end her speech, citing a pas-
sage from the Bible while the audience continues to call. She enjoins them to “not lose
heart” as they “finish the march” together, and as she utters her penultimate sentence,
“We have a march to finish”, driving home the theme of her speech, audience members
can be heard joining her, timing their own utterances of the words “march to finish”
to match hers.
3.3 Method 2: Listening to congregational uptake quantitatively

Clinton’s ability to elicit audience participation in the construction of her speech is
striking throughout as evidenced in the transcripts we provide. The clapping, cheer-
ing, and standing ovation that erupts from the audience as she concludes and that
erupted from them during her voice merging with James Cleveland provide ample
evidence of how the audience in Selma, Alabama evaluated Clinton’s speech. It is
clear from their responses captured on the CSPAN recording of this event that they
read her performance as symmetric role-alignment. To bolster this analysis, however,
we can listen to the audience in another way, by quantitatively measuring audience
responses. In what follows, we adapt a method familiar in experimental studies
within the field of communication. Following Mottet, Beebe, Raffeld and Medlock
 Deborah Cole & Régine Pellicer
(2004), who looked at effects produced by student audiences’ verbal and non-verbal
responses on teacher self-efficacy and job satisfaction, we listened to the C-SPAN
recording (C-SPAN 2007a) and counted audience responses during Clinton’s speech.
As the camera did not capture visuals of the audience (except for the one face other
than Clinton’s visible throughout the speech and the occasional appearance of the
backs of people’s heads during standing ovations), we do not have evidence of audi-
ence responses that require the visual perception of the responders (head nodding,
smiling, etc.). Only auditorily perceivable responses were counted. For this analysis, a
response is a vocalization or clapping that was audible on the recording and was pro-
duced by someone other than Clinton during the time frame of her speech. Since at
many points during the speech individual audience members’ responses overlapped
and since it is impossible in most of these cases to determine how many people were
responding, many voices responding together were counted as a single response.
A response, then, is a time period on the recording during which a person other than
Clinton is audible. Some responses are relatively short and are produced by a single
audience member, while other responses are relatively long and are p roduced by
multiple people.
We began counting from (0.18) when Clinton begins her speech with the
words “This is the day the Lord has made” and we stopped counting at (18.17)
when she says her final “thank you”. During that time frame, we counted 106 dis-
crete a udience responses. We also counted the approximate length of each of these
audience responses by counting the number of second markers that elapsed during
each response. A udience responses were as short as 1 second or less (responses
that took less than 1 second were counted as a second) and as long as 26 seconds.5
We then used these measurements to calculate the average number of audience
responses per minute as well as to calculate the response density, or the percentage
of the time during the speech when the audience is audibly participating. These
measurements and calculations for Clinton’s speech appear in the “Clinton” column
in Table 2.
. We did not include the audience welcoming standing ovation, as the speech had not
yet begun. We counted the final standing ovation as a single response in our count of total
responses, but counted it as only 1 second (though it lasted from 18.20 till 18.41) in order to do
a balanced comparison between Clinton’s and Obama’s audiences (to follow). The r ecording of
Obama’s final standing ovation is cut off on the CSPAN recording so its actual length cannot
be accurately measured.
Uptake (un)limited 
Table 2. Audience responses and response densities

Clinton Obama Lowery
Length of speech in minutes 17.98 min 34.68 min 4.52 min

Total # of audience responses 106 124 47
Average # of audience responses/minute 5.89 3.57 10.40
Length of speech in seconds 1079 sec 2081 sec 271 sec

Total combined length of audience responses 402 sec 356 sec 117 sec
Response density 37.26% 17.11% 43.17%
To get an idea for how well Clinton’s audience responded to her speech given the
expectations and standards of the speech event, we compared how Clinton’s audience
responded to how Obama’s audience responded to his speech at the African Methodist
Episcopal church in Selma that same morning. These data appear in the “Obama” col-
umn, and a comparison reveals that Clinton’s audience was more actively engaged in
co-authoring her speech than Obama’s was in co-authoring his. Clinton’s audience has
a higher average number of responses per minute (5.89 compared to 3.57), and they
were active 37.26% of the time compared to Obama’s, who were active only 17.11%. To
get an idea for how well Clinton succeeded at performing BPS, we also compared these
numbers to Obama’s audience’s response to the Rev. James Lowery’s short speech.
Lowery is a widely recognized and highly regarded African American minister and
orator, who represents a model speaker for this type of speech style. He spoke just
prior to Obama on the theme of how people working across religious boundaries for
social good were “good crazy” people. His short speech was so well-received by the
AME congregation that when Obama later took the podium and extended his “thank
yous”, he jokingly refused to thank Rev. Lowery because Lowery had already “stolen
the show”.
The table reveals that Clinton’s speech elicited more audience responses than
Obama’s, but fewer audience responses than Lowery. Further, the response density
for Clinton’s audience is closer to that for Rev. Lowery (37.26% for Clinton and
43.17% for Lowery) than it is to Obama’s (17.11%). These facts are significant for
understanding the use of BPS in general as well as its use by Clinton in particular,
for as Britt (2011) notes [quoting earlier work by Holt (1972:192)] “the intensity and
volume of audience response signals the preacher that he is getting across, that he’s
telling the truth, that the audience is enjoying what he says and appreciates how he
says it.” (Britt 2011: 217).
 Deborah Cole & Régine Pellicer
3.4 Summary
In reconstructing how the Selma First Baptist Church congregation responded to
Clinton’s performance of black preaching style, we can see that Clinton and her audi-
ence successfully entered into a speech event in which the felicity conditions and rights
and obligations for this semiotic register were met. In this sense, Clinton’s perfor-
mance meets with previously documented conventionalized expectations for the use
of this SR. On the face of it, the methodological pains we have taken to demonstrate
that the Selma First Baptist Church congregation liked Clinton’s speech seem hardly
necessary, as it should be obvious even to the casual viewer. It was necessary, however,
to “reconstruct” the congregation’s co-authorship for this speech event and their sym-
metric evaluations, because, as we will see, congregational voices tended to disappear
from subsequent mediatized projects, making possible evaluations of her performance
as asymmetric role-alignment. In the following section, we document this disappear-
ance and in doing so demonstrate how Clinton’s performance of BPS varied from two
other conventionalized interpretative expectations that prevail in U.S. public discourse
contexts, i.e. that context and co-text are irrelevant to interpreting register token use,
the folk theory of contextual invariance (Agha 2005), and that co-authors are irrel-
evant to the determining of speaker intentions, the ideology of personalism (Hill 2008).
4. Th
e congregation’s indurable response under conditions
of mediatization
In some cases, what is important to a social process is precisely how uptake

transforms in durability that to which it responds. (Agha 2011b: 166)
As members of the Selma First Baptist Church congregation are generally not the same
people who produce mass media for national media-consuming audiences, it is per-
haps not surprising that their explicit, meta-linguistic evaluations of Clinton’s speech
were not widely represented in post-performance mediatized value projects. But that
a directly opposing evaluation of Clinton’s register-shifting performance emerged and
replaced the congregation’s evaluation in increasingly available fragments of p ublic
discourse begs an explanation. This replacement, or erasure (Irvine & Gal 2000), is all
the more striking given that the congregation’s immediate reactions are amply docu-
mented and continuously available to anyone wanting to report on or opine about this
speech event in the days and weeks that followed Clinton’s speech (as the preceding
analysis revealed). An entire article could be devoted to analyzing what was said by
whom in the mediatized projects that recycled fragments of Clinton’s speech in Selma.
But as the voices who produced them have already had the o pportunity for wider
dissemination and uptake, controlling as they do the modes of media production,
Uptake (un)limited 
we will not linger too long over the many details of how they re-evaluated Clinton’s
style-shifting as “inauthentic” here. Instead, we provide a general summary of these
texts to document the diversity of voices that participated in replacing the symmetric
evaluations produced by Clinton’s face-to-face audience in the context of the church
service with their own asymmetric evaluations produced for media consuming
audiences in mediatized contexts.
4.1 Method
A Google search for “Clinton Selma Alabama” and for the lyrics of James Cleveland’s
song “I don’t feel no ways tired” returns a mixed bag of links to relevant mediatized
fragments including video clips from television shows that discussed Clinton’s speech
on U.S. television networks, text-based news articles and op-ed pieces distributed by
news corporations, and discussions posted by online viewers and readers. These links
include texts meant for local consumption and discussion, like the Selma Times–
Journal (Alabama) and The Bayou Buzz (Louisiana) as well as texts intended for
national or global audiences, appearing on websites for the New York Times, CNN,
Fox News, The John Stewart Show, and YouTube. Before turning to examples where
the erasure of symmetric role-alignment evaluations were evident in a mediatized
language panic, we present the only two mediatized projects we were able to find in
which the congregation’s evaluation was preserved and the metalinguistic evaluation
of language users in the speech event’s local context were explicitly consulted.
4.2 Uptake preserved

On March 7, 2007, three days after Clinton and Obama visited Selma, a local Selma
newspaper ran an article that provided an evaluation of Clinton’s speech matching
the evaluation we can hear from her audience on the C-SPAN recording. The article
cites “an assistant school principal in Selma” who was quoted as saying this about
Clinton’s speech: “She’s a great speaker, and she’s willing to make a change. She con-
nected with her audience, and connected with me.” (Selma Times-Journal 2007). The
report c ontained no negative evaluations of Clinton’s speech and no discussion of the
(in)authenticity of her language use or identity performance, themes that dominated
nationally distributed mediatized projects.
We found one example on the national scale. On March 19, 2007, two weeks
after Clinton’s and Obama’s Bloody Sunday speeches, Newsweek magazine published
a half-page table under the title “Preaching to the Choir” that provided explicit evalu-
ations by three Selma pastors. Although the report is limited to the reactions of three
people, these individuals represent precisely the social type indexed by the register
Clinton and Obama performed: They are ministers of Black Churches in the South,
and two of them were the host pastors of the churches where the candidates spoke.
 Deborah Cole & Régine Pellicer
The pastors were asked to give each of the speakers a grade on an “A” to “F” scale
(with plus and minus modifications), where “A+” was excellent and “F” was failing,
for the categories “biblical acumen”, “delivery”, and “crowd response”. Both candidates
received high marks for all three categories. The lowest grades were Bs given by the
same pastor to both Clinton and Obama for “delivery”.
For each of the grades given, the Newsweek table also included short comments
from each of the pastors explaining their ratings. With respect to “crowd response”,
Clinton earned two “A’s” and one “B+”, and all three ministers explicitly stated that
Clinton’s audience responded positively. They commented, “She definitely kept the
crowd’s attention. She got a great response”, “The crowd stood and applauded three
or four times while she spoke”, and “The crowd responded to her. She’s a naturally
comfortable speaker” (Alston 2007: 10). For the category “delivery”, Clinton received
a “B”, a “B+” and an “A”. One pastor commented, “Her delivery was good, but not as
good as a real southern preacher” (Alston 2007: 10). Another Selma minister directly
addressed the negative evaluations of Clinton’s “delivery” that had already been cir-
culating in other mediatized projects by asserting, “I didn’t pay any mind to whether
she was trying to mimic our accent” (Alston 2007: 10). The third minister is quoted as
saying, “She did what she knew best. She’s just a down-home Arkansas girl.” (Alston
2007: 10).6
These two mediatized texts provide corroborating evidence for the analysis that
members of Clinton’s intended audience read Clinton’s register shifting as symmetric
role-alignment. As the only mediatized texts that recycled this local value project, how-
ever, they surely had a much smaller likelihood of becoming part of further mediatized
(and otherwise mediated) value projects than the multiple and varied other texts we
found in which Clinton’s register shifting was evaluated as asymmetric-role alignment.
We turn now to examples of these texts.
4.3 Uptake transformed

Although print articles that appeared on websites for CNN and the New York Times
on March 5, 2007 (the day immediately after the Selma speeches) made no mention
of Clinton’s “fake accent”, reports which followed these generally framed her perfor-
mance as inauthentic. A clip from the CNN television show “The Screening Room”
. Although Clinton’s “biblical acumen” is not as directly relevant to the discussion here, one
comment under this category is worth noting: “She did a really good job relating to a religious
audience” (Alston 2007: 10). Whereas Clinton made stylistic choices characteristic of sermons
delivered in BPS, like using biblical references to frame a sermon as divinely inspired (Wharry
2003), these did not garner mainstream media reactions.
Uptake (un)limited 
hosted by Wolf Blitzer posted on March 5, for example, problematized both Obama’s
and Clinton’s performances, pointing out that politicians keep changing their accents
depending on their audiences. The reporter for this segment can already refer to
Clinton’s performance of James Cleveland’s song as “the example that everyone’s been
citing”. By March 6, “local” news outlets as far away from Selma as Denver, Colorado
and Chicago, Illinois had begun to run stories that criticized Clinton’s register shifting
as “bizarre” (Colorado Media Matters 2007) and “disagreeably phony and affected”
(Zorn 2007). Also on March 6, Fox News personality E.D. Hill accused Clinton of
strangely “affecting a ‘Southern drawl’” (Mediamatters March 6, 2007). The wider
uptake of this evaluation of decontextualized register shifting as inauthentic is also evi-
dent in the titles of some of the links our search surfaced, for example “Hillary Fakes
Southern Accent At Black Church” (Barber 2007) and “Hillary Clinton Fakes Southern
Accent/ Mocks South at Church Rally” (Simpson 2007). Some writers even speculated
about the socio-political implications of performing others’ voices and identities by
considering what might have happened in other contexts if Clinton had “pretended”
to be Jewish (Parker 2007), Asian, or Native American (Willams 2007) or gay (Media
Matters 2007a). These imagined, potential register shifts were asserted to be offensive
and readers were enjoined to use these hypothetical scenarios to inform their own
evaluations of Clinton’s actual register shifting in Selma.
A frequently recycled lexical item in our data set that contributed to this
renormalizing of media consumer evaluations of Clinton’s shift into BPS is the repeated
use of the word “pander”, which showed up 51 times in the 18 electronic text sources
our search found. Its use is interesting because of how it not only directs consum-
ers to evaluate Clinton’s performance but also because of how it directs them to (re)
evaluate the original evaluations of the Selma congregation, should they happen to run
across them. Forms of the word “pander” appeared in professionally written articles
and in online reader discussion forums. One author specifically invited discussion on
the topic, titling his posting, “Stop me before I pander again (audio clip of Hillary
Clinton’s Selma speech)” (Lifson 2007). A closer look at how the lexical item “pander”
showed up in two nationally distributed op-ed pieces provides further evidence of
how the Selma congregation’s symmetric evaluations were explicitly transformed and
disappeared.
In an article entitled “The Rev. Hillary’s Tin Ear” by Kathleen Parker that appeared
on the website for the Washington Post Writers Group three days after Clinton’s speech,
for example, the author ridicules Clinton’s voice itself, saying it made listeners “cringe”
and “recoil”. Parker, aware that that this analysis is not supported by the evidence on
the recordings of Clinton’s speech, nonetheless bolsters her argument by pointing out
that “Her audience, nevertheless, was polite and affirming (Southerners are like that),
even as she turned on the worst fake accent since Kevin Costner played Robin Hood”
 Deborah Cole & Régine Pellicer
(Parker 2007). In the final sentences of her piece, when Parker explicitly uses the word
“pander”, she evaluates Clinton’s style shifting as revealing of her inauthenticity:
It is her fault that she panders – badly – to her audiences. Her performance
last weekend in Selma revealed more than atonality. Like a warped bell, Hillary
Clinton rings untrue. [Parker 2007]
Six months later after the initial language panic had dissipated, Walter Williams, a
professor of economics at George Mason University, published another piece that fea-
tured “pandering” entitled “Insulting Blacks” as part of his regular column, estimated
to be “regularly published in at least 79 different newspapers” (Media Matters 2007b).
In it, Williams opens with the by-then-infamous James Cleveland lyrics, goes on to
analyze Clinton’s register shifting performance as producing negative, racist effects,
and then ends with this question and answer: “What does it say about blacks who can
be taken in by pandering, alarmist nonsense from both whites and blacks as a means
to get their votes? As a black man, I don’t find the most obvious answer very flattering”
(Williams 2007).
4.4 Summary
In demonstrating how mediatization processes that recycled fragments of Clinton’s
speech produced multiple communicative events, we observe that mediatization
produced not only (re)readings of the original communicative event itself but pre-
scriptions for how the localized, pre-mediatized event should be (re)evaluated in
future uptake. Such mediatized forms of uptake tended to ignore local conventions
for evaluating register-shifting phenomena and served to maintain the wider, folk-
theoretic conventions of interpreting non-stereotypic uses of registers as inauthentic
regardless of the original context of their usage. They also followed the convention of
ascribing intentions to text producers while ignoring the collaborative co-authorship
of their audiences. Thus asymmetric evaluations of Clinton’s performance became
available for wider uptake (despite the robust evidence of local symmetric evaluations)
as fragments of the original speech event were decontextualized from their co-text and
recycled without the contributions of Clinton’s co-authors, whose voices disappeared
from more durable mediatized value projects.
5. W
hen the ideology of personalism fails, authentic referentialism
comes to its defense
In her recent book on the persistence of racism through language in the U.S., Jane
Hill demonstrates how the implicit ideologies of referentialism and personalism inter-
act with the explicit ideology of Standard to mutually enforce folk theories of race.
Personalism is the name for a language ideology, widespread in the United States,
Uptake (un)limited 
“which holds that the most important part of linguistic meaning comes from the
beliefs and intentions of the speaker” (Hill 2008: 38). Hill’s work demonstrates how
members of the white middle class evade charges of racism by appealing to personalist
ideology to reframe racist language with respect to the intentionality of speakers. But
she suggests an alternative to her project: She invites us to “think of incidents of rac-
ism as providing an opportunity to defend personalism, rather than thinking of per-
sonalism as an ideological framework within which to evade charges of racism” (Hill
2008: 117). Clinton’s style-shifting performance in Selma, Alabama enables us to take
steps towards this kind of thinking because attempts to apply the logic of personalism
failed to produce the usual results in this case. Considering the role of personalism in
the Clinton case helps us to identify what is at stake for mainstream elites when the
identity categories of race and language are denaturalized.
Much of what goes on in language panics around racist language is the exoneration
of speakers who are argued to not have intended racism when uttering potentially racist
language even though audiences can and do evaluate their language as racist. In cases of
potential racism in U.S. public discourse, the logic of personalism, as Hill (2008) has laid
it out for us, works like this. Speaker X uttered lexical item Y, and Y is a member of a set
of lexical items that have been categorized as “racist language”. Therefore, speaker X can
potentially be categorized as racist. Luckily for speaker X, personalist ideology (Keane
2002) holds that the most important factor in determining linguistic meaning is the
speaker’s intentions and beliefs. If X can be shown to have non-racist beliefs, X cannot
be a member of a set of individuals that can be categorized as racist. Appeals to person-
alism enable the evasion of racist charges, as Hill’s book convincingly demonstrates, for
examples of racist language that can be publicly labeled as gaffes or slurs.
Folk theorists cannot easily apply personalist ideology to the Clinton example
presented here, however, primarily because no potential gaffe or slur occurred. None
of the lyrics in Rev. James Cleveland’s song belong to the set of lexical items that
have been categorized as racist. Instead, Clinton performed phonetic and syntactic
features of a register indexically associated with a particular speech style typically
performed by African Americans in the context of Black Churches in the South.
Yet, despite the absence of the usual types of linguistic evidence for racism about
which folk theorists are typically metalinguistically aware and vocal, the folk theorist
senses racism nonetheless, and following the conventions of Western, personalist
logic must ask herself, “What were Clinton’s intentions?”. She must consider at least
two answers. Either Clinton did not intend racism by style-shifting to converge on a
persona with which she imagined her audience to be familiar, or Clinton did intend
racism by style-shifting to converge on a persona with which she imagined her audi-
ence to be familiar. The s econd option is clearly illogical, given the context of the
presidential race, for if C linton had taken this option knowingly, she surely would
have been committing an act of political suicide. The first option must be right,
but leaves the folk theorist, embarrassingly, without any supporting evidence. Thus,
 Deborah Cole & Régine Pellicer
the conventional logical of personalist ideology provides no purchase for evaluat-

ing the potentially racist nature of Clinton’s performance. If Clinton style-shifted
intentionally in the hopes of encouraging her audience to support her bid for presi-
dential nominee, she could not have logically intended racism. From the point of
view of personalism, with no racist language and no racist intent, there is technically
no racist.
Personalism’s inability to make common sense racism out of Clinton’s cross-
identity style-shifting throws it into crisis, denaturalizing its ability to apply to the
world of discourse. Hill points out that personalism is constantly in a crisis, which
“requires that those who live most fully within the representational economy which it
anchors – which includes such dimensions of contemporary thought as the possibility
of rational choice, the possibility of authentic belief, and the possibility of valid asser-
tion about the value of money (Keane 2002, 2007) – must find occasions to defend and
reinscribe it” (Hill 2008: 118).
In this case, folk theorists jumped to personalism’s defense, and sidestepped the
issue of the missing racist language and therefore the missing racist by instead c alling
into question Clinton’s ability to successfully and authentically imitate a language
register publicly associated with Southern African Americans. The side-stepping argu-
ment must go something like this: Clinton is clearly not a member of the Southern
African American identity category, so her attempts to perform the linguistic ele-
ments that commonly index that identity category were inauthentic. Thus, instead
of questioning Clinton’s intentions and by focusing instead on the “inauthenticity”
of her performance, the media-generated panic diverted public attention away from
personalist ideology’s failure to apply to the Clinton-Selma congregation speech event
and towards the way that Clinton violated referentialist ideology by presenting herself
in an inauthentic or untrue way.
Thus, Clinton’s convergence on the language patterns she imagined members of
her Southern, African American audience to use and recognize produced multiple
threats to the ideological underpinnings that naturalize race and language categories.
Her adequative performance not only threatened to denaturalize personalism and
its “useful” role in addressing potential accusations of racism by mainstream, elite
Americans, it also threatened the naturalness of the ideology of Standard and called
into question essentializing folk categories of race. These threats necessitated a pre-
emptive defense, characterized by a flurry of ideologically authorized, veiled (and not
so veiled) accusations of racism that worked to divert public attention away from the
crises of personalism in an attempt to convince media audiences, yet again, of the
naturalness and necessity of the dominant representational economy (Keane 2007) in
which Standard occupies the highest position in the linguistic hierarchy.
By performing the Southern speech found in African American sermons and
hymns, Clinton enabled the interpretation of all linguistic identities as performances.
This possibility implies that the hierarchy of voices in which Standard occupies the
Uptake (un)limited 
highest slot is socially constructed by all the voices, Standard and Other than Standard,
which agentively collaborate to give Standard its representational value. In selecting
the persona of James Cleveland, and perhaps of Reverend King, and in performing
a register that would convey a social range (Agha 2005: 39) that would include them,
Clinton’s adequative act of style-shifting itself put out of focus her other personal
alignments, that may well be oppositional to her “imagined” alignment with the audi-
ence. This is exactly the point of adequative performance. By focusing instead on what
the speaker and the audience currently have in common, in this case an appreciation
for the sound of James Cleveland’s voice and the rhythms and cadences of sermons
in Southern Black Churches, Clinton and her appreciative audience denaturalize the
categorization processes that keep people of different identities apart.
It becomes clear in the moment of performance and perception of a metrically
different voice that a shift towards the audience has occurred. And because it can hap-
pen at this particular moment of production and perception, the implication arises
that at each and every moment of production and perception of voices, persons make
choices to speak and perceive in familiar or unfamiliar ways, vis-à-vis their audience.
Thus, in denaturalizing performances, the force of authentication becomes momen-
tarily de-emphasized. Further, by writing for and to a media consuming audience
whose responses took precedence over the responses of the congregation who had
collaborated in the construction of Clinton’s text, the media generated diversion away
from the (il)logic of personalism also worked to re-naturalize mainstream hierarchies
of race and language.
The data presented here demonstrate again the potential that a rethinking of
both language and identity with special attention to audience and context has for
denaturalizing ideologies of Standard and Race. In responding to Hill’s musing about
an alternative project of inquiry mentioned above, it does in fact appear that “incidents
of racism” are useful sites for periodically stirring up a panic in which to do defend
Standard and its supporting ideologies of personalism and referentialism. In her anal-
ysis of the roles played by these ideologies in upholding essentialized categories of
Standard and Race, Hill points out that “both the idea that speaker intention is pri-
mary and the idea that words have inherent meanings leave out a third possibility: that
if language is found to be racist by its targets, then it is racist language” (Hill 2008: 96).
These data suggest a fourth possibility, the converse of Hill’s third: that if language is
found to not be racist by its targets, then it ain’t.
6. Implications and conclusions
By focusing our analysis on the ways that different value projects are more or less
limited in their potential for wider uptake under conditions of mediatization, the
foregoing analysis has also contributed to “clarify[ing] how persons and groups
 Deborah Cole & Régine Pellicer
ifferentiated by this process become recognizable to each other as social types”

d
(Agha 2011a: 51). In particular, these data train our attention on how the indexical
values of register-shifting phenomena do not “inhere in speech diacritics (as norma-
tive discourses often insist)” (Agha 2011a: 51) but rather emerge in competing perfor-
mances of value p rojects whose wider dissemination and uptake are constrained by the
degree to which the felicity conditions for particular register usage and interpretation
are met in different contexts.
As there is nothing about the forms used in the performance of value projects
that dictates their evaluations, the uptake of value project performances are fragile
because the opportunities for re-valuation that arise with each recycling are potentially
unlimited, as speakers and hearers make choices about how to value particular per-
formances in each new encounter, i.e. role-alignments are “inherently agentive” (Agha
2005). At the same time, however, the potential for uptake of any given value proj-
ect can become quite limited as institutional discourses channel them “in ways that
routinely restrict them to certain participation frameworks…creating very interesting
methods for ‘zoning’ ideas and ideologies” (Agha 2011b: 176). We have seen above
that when forms of voicing and personhood are detached from general mediation
phenomena and recycled through mediatization, the value projects that were being
performed using these forms are not necessarily recycled along with them. In fact,
the value projects being constructed in pre-mediatized contexts can be diametrically
opposed to the value projects to which the very same formal fragments are pressed
into service post-mediatization.
A theory of mediatization thus enables us to identify and isolate those special
cases of mediation that most clearly foreground the linguistic forms, functions, and
ideologies that contribute to and maintain dominant hierarchies of semiotic registers.
At the same time, by categorizing mediatized semiotic phenomena as a special case of
mediation in general (Agha 2011), a theory of mediatization makes visible the mul-
tiple and varied forms, functions, and ideologies that continue to exist alongside their
dominant counterparts. Such recessive memes become visible as viable alternatives
to the hegemonies we critique even as they point to ways out of remaking what we
fear. It might be worth thinking about, for example, the effects we could produce if as
sociolinguists we were to actively take up the forms and functions of semiotic register
shifting valued in black preaching style “to manag[e] the reception of the political and
social messages” (Britt 2011: 212) we present and “to dis-identify with the normative
balance” (Myers-Scotton 1985: 109, cited in Britt 2011: 215) of register hierarchies. As
a discipline that has had significant difficulty communicating its value project of uni-
versal language competence and descriptivism to a wider evaluating public (Hill 2008;
Battistella 2010), we “might could” benefit from having more registers at our disposal.
Who could be more qualified to take up multiple and varied value projects than those
with contact with and awareness of the wide variety of uptake formulations competing
for survival in the public sphere?
Uptake (un)limited 
As we “look beyond dominant institutional discourses about language to consider

other discourses that may be in play” and “think about the technological constraints
on whose voices are heard when” (Johnstone 2011: 13), we can note that opportu-
nities for any given sociolinguistic value project to be continuously recycled can be
enhanced through upscaling, or by being made available through mediatization to
wider audiences. The fact that mediatized value projects go on to serve as the input
for uncountably many post-mediatized mediation projects (Agha 2011b) certainly
magnifies this effect. However, although institutions find ways to successfully control
the mediatziational means of production, they cannot always control “what…people
effectively do with language” (Blommaert 2010: 187). As mediation is ongoing, con-
stant, and unlimited, a better understanding of its relationship to mediatization can
indeed provide a “tonic for anxieties about the ‘mass media’” (Agha 2011c: 173) and its
(in)ability to maintain dominant hegemonies by obscuring or invalidating the perfor-
mance of alternative value projects.
Appendix
The transcription system used in this analysis follows Britt (2011: 230).

(( )) Characterizations of stretches of talk, vocalizations, and speaker/audience
behavior or applause are described between double parentheses
? “question” intonation (i.e. rising pitch)
. “period” intonation (i.e. falling pitch)
Underline Stretches of talk delivered with stress or emphasis by the speaker
CAP Stretches of speech delivered loudly and with emphasis
References
Agha, Asif. 2003. The social life of cultural value. Language & Communication 23: 231–273.
Agha, Asif. 2005. Voice, footing, enregisterment. Journal of Linguistic Anthropology 15(1): 38–59.
Agha, Asif. 2007. Language and social relations. Cambridge: CUP.
Agha, Asif. 2011a. Commodity registers. Journal of Linguistic Anthropology 21(1): 22–53.
Agha, Asif. 2011b. Meet mediatization. Language & Communication 31: 163–170.
Agha, Asif. 2011c. Large and small scale forms of personhood. Language & Communication 31:
171–180.
Alim, Samy H., Lee, Jooyoung & Carris, Lauren Mason. 2010. “Short fried-rice-eating Chinese
MCs” and “Good-Hair-Havin Uncle Tom Nigas”: Performing race and ethnicity in freestyle
rap battles. Journal of Linguistic Anthropology 20(1): 116–133.
Alston, Joshua. 2007. Preaching to the choir. Newsweek, March 19, 149(12): 10.
Barber, La Shawn. 2007. Hillary fakes southern accent at black church. 〈http://lashawnbarber.
com/archives/2007/03/05/hillary-fakes-southern-accent-at-black-church/〉 (29 December
2009).
 Deborah Cole & Régine Pellicer
Barnes, Sandra L. 2005. Black church culture and community action. Social Forces 84(2):
967–994.
Battistella, Edwin. 2010. Ideologies of language, art, and science. In Linguistics at School:
Language Awareness in Primary and Secondary Education, Kristin Denham & Anne Lobeck
(eds), 13–23. Cambridge: CUP.
Baugh, John. 2000. Beyond Ebonics: Linguistic Pride and Racial Prejudice. Oxford: OUP.
Blommaert, Jan. 2003. Commentary: A sociolinguistics of globalization. Journal of Sociolinguistics
7(4): 607–623.
Blommaert, Jan. 2010. The Sociolinguistics of Globalization. Cambridge: CUP.
Bucholtz, Mary & Hall, Kira. 2004. Language and identity. In A Companion to Linguistic
Anthropology, Alessandro Duranti (ed.), 369–394. Oxford: Blackwell.
Bucholtz, Mary & Hall, Kira. 2008. Finding identity: Theory and data. Multilingua 27: 151–163.
Britt, Erica. 2011. “Can the church say amen”: Strategic uses of black preaching style at the State
of the Black Union. Language in Society 40: 211–233.
C-SPAN. 2007a, March 4. Civil rights issues. 〈http://www.c-spanvideo.org/program/196941–1〉
C-SPAN. 2007b, March 4. Road to the White House. 〈http://www.c-spanvideo.org/
program/196943–1〉
Chun, Elaine. 2004. Ideologies of legitimate mockery: Margaret Cho’s revoicings of Mock Asian.
Pragmatics 14(2–3): 263–289.
Cutler, Cecilia. 2003. Keepin’ it real: White hip-hoppers’ discourse of language, race and
authenticity. Journal of Linguistic Anthropology 13(2): 211–233.
Cole, Debbie. 2010. Enregistering diversity: Adequation in Indonesian poetry performance.
Colorado Media Matters. 2007. Boyles accused Hillary Clinton of speaking in “a blackface voice” to
an African-American audience. 〈http://colorado.mediamatters.org/items/200703060008〉
(14 June 2010).
Duranti, Alessandro & Goodwin, Charles (eds). 1992. Rethinking context: Language as an
Interactive Phenomenon. Cambridge: CUP.
Goebel, Zane. 2008. Enregistering, authorizing, and denaturalizing identity in Indonesia.
Goebel, Zane. 2010. Language, migration, and identity: Neighborhood talk in Indonesia.
Cambridge: CUP.
Hill, Jane H. 2008. The Everyday Language of White Racism. Chichester: Wiley-Blackwell.
Holt, Grace Sims. 1972. Stylin’ outta the black pulpit. In Rappin’ and Stylin’ outta the Black Pulpit:
Communication in Urban Black America, Thomas Kochman (ed.), 189–204. Urbana IL:
University of Illinois Press.
Irvine, Judith & Gal, Susan. 2000. Language ideology and language differentiation. In Regimes
of Language, Ideologies, Politics, and Identities, Paul V. Kroskrity (ed.), 35–84. Santa Fe NM:
Johnstone, Barbara. 2000. The individual voice in language. Annual Review of Anthropology 29:
405–424.
Johnstone, Barbara. 2011. Making Pittsburghese: Communication technology, expertise, and
the discursive construction of a regional dialect. Language & Communication 31: 3–15.
Keane, Web. 2002. Sincerity, “modernity,” and the Protestants. Cultural Anthropology 17: 65–92.
Labov, Wiliam. 1973. The linguistic consequences of being a lame. Language in Society 2(1):
81–115.
Uptake (un)limited 
Leung, Constant, Harris, Roxy & Rampton, Ben. 1997. The idealized native speaker, reified
ethnicities, and classroom realities. TESOL Quarterly 31(3): 543–560. (Special issue:

Language and Identity).
Lifson, Thomas. 2007. Stop me before I pander again (audio clip of Hillary Clinton’s Selma
speech). 〈http://www.freerepublic.com/focus/f-news/1795893/posts〉 (29 December 2009).
Lippi-Green, Rosina. 2012. English with an Accent. Language, Ideology, and discrimination in the
United States. New York NY: Routledge.
Mitchell, Henry. 1970. Black Preaching. Philadelphia PA: Lippincott.
Meadows, Bryan. 2010. “Like my tutor and stuff, people I would talk to”: Laying claim to
imagined national communities of practice in language learner discourse. Critical Inquiry
in Language Studies 7(2–3): 88–111.
Media Matters. 2007a. Fox’s Hill on Clinton’s “Southern drawl”: “[I]f she was attending,
say, a GLAAD convention, would she speak with a lisp?” 〈http://mediamatters.org/
research/200703060009〉 (14 June 2010).
Media Matters. 2007b. Accusing Clinton of “mimicking black dialect”. Walter Williams
didn’t note she was reciting hymn in Selma speech. 〈http://mediamatters.org/research/
200709130001〉 (29 December 2010).
Mendoza-Denton, Norma. 2008. Homegirls: Language and Cultural Practice among Latina Youth
Gangs. Malden MA: Blackwell.
Mohamad, Goenawan. 2002. Forgetting: Poetry and the nation, a motif in Indonesian literary
modernism after 1945. In Clearing a Space: Postcolonial Readings of Modern Indonesian
Literature, Keith Foulcher & Tony Day (eds), 183–211. Leiden: KITLV Press.
Mottet, Timothy P., Beebe, Steven A., Raffeld, Paul C. & Medlock Amanda L. 2004. The effects
of student verbal and nonverbal responsiveness on teacher self-efficacy and job satisfac-
tion. Communication Education 53(2): 152–165.
Myers-Scotton, Carol. 1985. What the heck sir: Style shifting and lexical colouring as features of
powerful language. In Sequence and Pattern in Communicative Behavior, Richard Street &
Joseph Cappella (eds), 103–119. Baltimore MA: Edward Arnold.
Parker, Kathleen. 2007. The Rev Hillary’s tin ear. 〈http://www.postwritersgroup.com/archives/
park0307.htm〉 (15 November 2009).
Selma Times-Journal. 2007. Clinton urges citizens to “stay awake”. March 7.
Schwartz, Adam. 2008. Their language, our Spanish: Introducing public discourses of ‘Gringoism’
as racializing linguistic and cultural reappropriation. Spanish in Context 5(2): 224–245.
Silverstein, Michael. 1998. Monoglot “Standard” in America: Standardization and metaphors
of linguistic hegemony. In The Matrix of Language: Contemporary Linguistic Anthropol-
ogy, Donald Brenneis & Ronald K.S. Macaulay (eds), 284–306. Boulder CO: Westview
Press.
Simpson, Jackson. 2007. Hillary Clinton fakes southern accent: Mocks South at church rally.
〈http://www.nationalledger.com/cgi-bin/artman/exec/view.cgi?archive=10&num=11927〉
(14 June 2010).
Smith-Hefner, Nancy. 2007. Youth language, gaul sociability, and the new Indonesian middle
class. Journal of Linguistic Anthropology 17(2): 184–203.
Smith-Hefner, Nancy. 2009. Language shift, gender, and ideologies of modernity in Central Java,
Indonesia. Journal of Linguistic Anthropology 19(1): 57–77.
Urciuoli, Bonnie. 1996. Exposing Prejudice: Puerto Rican Experiences of Language, Race, and
Class. Boulder CO: Westview Press.
 Deborah Cole & Régine Pellicer
Urciuoli, Bonnie. 2003. Boundaries, language, and the self: Issues faced by Puerto- Ricans and
other Latino/a college students. The Journal of Latin American Anthropology 8(2): 152–173.
Ullman, Char. 2004. English Matters? Undocumented Mexican Transmigration and the
Negotiation of Language and Identities in a Global Economy. Ph.D. dissertation, U niversity
of Arizona, Tucson.
Valdés-Fallis, Guadalupe. 1978. Codeswitching and the classroom teacher. Language and
Education: Theory and Practice 4: 1–31.
Vail, Mark. 2006. The integrative rhetoric of Martin Luther King Jr’s “I have a dream” speech.
Rhetoric & Public Affairs 9(1): 51–78.
Wharry, Cheryl. 2003. Amen and hallelujah preaching: Discourse functions in African American
sermons. Language in Society 32(2): 203–225.
Williams, Walter E. 2007. Insulting blacks. 〈http://www.creators.com/opinion/walter-williams/
insulting-blacks.html〉 (29 December 2009).
Wolfram, Walt & Schilling-Estes, Natalie. 2006. American English: Dialects and Variation,
2nd edn. Oxford: Blackwell.
Zentella, Ana Celia. 1997. Growing up Bilingual. Malden MA: Blackwell.
Zeigler, Mary B. 2001. Something to shout about: African American Vernacular English as a
linguistic and cultural treasure. In Sociocultural and Historical Contexts of African A
merican
English [Varieties of English around the World G27], Sonja L. Lanehart (ed.), 169–185.
Zorn, Eric. 2007, March 6. Hillary don’t know nothin’ ‘bout birthin’ no babies. 〈http://blogs.chi-
cagotribune.com/news_columnists_ezorn/2007/03/hillary_dont_kn.html〉 (14 June 2010).
The silken cord
An essay in honor of Jane Hill
Richard Delgado
Seattle University School of Law
In honor of Professor Hill’s courageous stand against mock Spanish in a state,

Arizona, that has been taking anti-Latino nativism to new extremes, this selection
examines Official English laws in light of the Southwest’s hidden history of
Latino lynching. It posits that suppression of Spanish severs the connection of
citizens, especially ones of Mexican descent, with the past. Not only do these laws
contribute to cultural ignorance, they leave young Latinos and Latinas without
defenses against hateful stereotypes – in effect, a second form of lynching.
Keywords: stereotype; racism; discourse analysis
1. Introduction: Symbols of threat and their meaning
1.1 What do tangible things mean, and for whom do they hold meaning?
A conservative Supreme Court in Virginia v. Black1 had little trouble in appreciating
the intimidating nature of a burning cross for an African American. Other courts
have been equally quick to discern the overtones of threat in a noose left in a locker
room, workplace area, or hanging from a tree on a high school grounds (Nossiter
2007). In Jena, Louisiana, for example, one such symbol figured prominently in the
controversy over the right to sit in a certain spot on a public high school’s campus
and the ensuing prosecution. When someone hung a length of rope, doubled back in
a familiar shape, on a branch of a tree beneath which white students had traditionally
congregated, no one doubted its meaning. Blacks stay away. Gather here and expect
trouble (Krugman 2007).
The noose acquired its powerful symbolic force during the Jim Crow era (Holden-
Smith 1996: 31, 35–37, 39–40, 77–78). But blacks today are still not entirely free of
the threat of racial violence, so the reappearance of a noose or burning cross carries
. 538 U.S. 343 (2003) (upholding major portions of a Virginia law that criminalized
cross-burning with the intent to intimidate).
 Richard Delgado
for them much the same meaning that it did in the age of Southern violence and
repression.
To whom else would a noose carry the same meaning? Gays and lesbians? Jews?
Gypsies? Odious as the symbol may be, it lacks the same meaning for them that
it does for blacks. Other symbols – a swastika, perhaps, or an epithet like “fag” or
“kike” – certainly might. (Delgado & Stefancic 2004: 47–93). But the noose is linked,
in the public mind at least, with this one group alone.
Yet that seeming singularity has recently come under question. Research by repu-
table historians shows that Latinos, particularly Mexican Americans in the Southwest,
were lynched in substantial numbers during roughly the same period when lynching
of blacks ran rampant (Carrigan & Webb 2003: 413). Every schoolchild knows that
blacks suffered that fate. Why do so few know about the lynching of Latinos?
1.2 Latino lynching

A recent casebook (Delgado et al. 2007: 207–209) summarizes much of what is known
about Latino lynching. Most of the reports, all relatively recent, are also remarkably
consistent, listing about 597 lynchings, most dating to the same heyday as black
lynching, Reconstruction and the years immediately following it. Moreover, the rea-
sons that prompted the lynchings were similar for the two groups – acting “uppity,”
taking away jobs that whites coveted, making advances toward a white woman, cheat-
ing at cards, practicing “witchcraft,” and occupying land that Anglos desired – with
one exception: Mexicans were lynched for acting “too Mexican,” speaking Spanish
too loudly or reminding Anglos too defiantly of their Mexicanness (Carrigan & Webb
2003: 418–422). Mexican women, particularly those from the poorer classes, came
in for lynching for offenses such as resisting an Anglo’s advances too forcefully or,
conversely, for engaging in prostitution (Carrigan & Webb 2003: 419–421).
The reader versed in civil rights history may recall that a similar but less deadly
form of violence took place during World War II when U.S. servicemen in Los Ange-
les attacked young Mexican-American men who congregated on street corners
wearing distinctive “Zoot suits,” gold watch chains, and slicked-back hair (Acuña
2000: 254–258). Although the violence amounted to beatings, public depantsing, and
other forms of nonlethal humiliation, the attacks went on for days without official
intervention. As with lynching, the World War II-era attacks targeted youths who
displayed their Mexican identity too proudly and openly.
The number of African Americans lynched during this period (1865 to 1920) was
of course higher, around 3400 to 5000 (Holden-Smith 1996: 155–156). But the popula-
tion of Latinos was much smaller then so that the rate of lynching for the two groups
was similar (Carrigan & Webb 2003: 414, 430–432, app.). As with blacks, Latino
lynching went on with the knowledge and sometimes active participation of Anglo
The silken cord 
law enforcement authorities, especially the Texas Rangers, many of whom displayed a
special animus toward the group (Carrigan & Webb 2003: 416, 420–422).
As with blacks, the lynching of Mexicans was often marked by hilarity and an
atmosphere of righteous celebration. One authority even described Anglo vigilantism
toward the group as a means by which whites increased solidarity and civic virtue
(Carrigan & Web 2003: 415–416); indeed, the few English-language newspapers that
reported the events often praised them for those reasons. As with black lynching, the
participants would often mutilate the bodies of the victims and leave them on display,
or cut off body parts for bystanders to take home as souvenirs (Carrigan & Webb
2003: 416–419, Gonzales-Day 2006: 175–178).
Most lynchings of Latinos took place in the states or territories of Texas,
California, Arizona, and New Mexico, all of which had substantial Mexican or
Mexican-American populations. Fewer took place in Colorado and Nevada, with
cases as far afield as Nebraska, Oklahoma, Oregon, Kentucky, Louisiana, Montana,
and Wyoming (Carrigan & Webb 2003: 415).
Many lynchings began with a mob snatching a Mexican from the hands of the
authorities, removing him from a prison cell or courthouse, and stringing him up.
In June 1874, a Latino man named Jesús Romo was arrested for robbery and other
crimes in La Puente, California (Carrigan & Webb 2003: 416). Shortly thereafter, a
group of masked men seized him from the arresting officers, took him outside, tied a
rope around his neck, and hanged him to death. Local opinion celebrated the event,
describing Romo as a “hardened and blood-stained desperado” who richly deserved
his fate (Carrigan & Webb 2003: 416, quoting a local newspaper).
Other lynchings occurred in isolated mining camps or sparsely settled ranch
areas, often with the assistance of groups such as the Texas Rangers. Historians William
D. Carrigan and Clive Webb write that the number of Mexicans murdered by members
of that organization may have run into the thousands and that very few, if any, Anglos
were made to stand trial for taking part in the lynching of a Mexican (Carrigan &
Webb 2003: 417).
Lynching was so common that the Mexican government, civil rights organiza-
tions, and even the U.S. consul in Matamoros, Mexico lodged official complaints
(Carrigan & Webb 2003: 417, 427–438; Carrigan & Webb 2005: 266, 282–285). Others,
including Juan Cortina, Gregorio Cortez, and mythic outlaw Joaquin Murietta, took
matters into their own hands, avenging the lynchings of compatriots by murdering
the Anglos responsible (Rosenbaum 1981: 53–67). Still other Mexicans organized
secret, conspiratorial societies, such as the Plan de San Diego, which called for the
overthrow of Anglo society (Johnson 2005). Anglos met all of these acts of resistance
with r uthless, organized force (Navarro 2005: 100, 317).
The few Anglo-American historians who have written about lynching of Latinos
ascribe it to battles over turf and Yankee nationalism left over from the Mexican War
 Richard Delgado
(Carrigan & Webb 2003: 416–417). Rodolfo Acuña, on the other hand, compares it to
anti-black racism and the form of special hatred that accompanies unjust wars (Acuña
2005). The abovementioned casebook, after reviewing the evidence, concludes that
Latino lynchings are a relatively unknown chapter in United States history whose
invisibility is part of a common pattern of shaping discourse by the dominant group
(Delgado et al. 2007: 208–209).
2. W
hy these events are so little known – language orthodoxy
and official English
Why are these events not better known? One reason is that the primary accounts
of the linchamientos appeared in community newspapers, which were in Spanish.
Since relatively few mainstream historians consulted these sources, if indeed they
read the language at all, Latino lynching remained beyond the ken of most main-
stream readers. Mexicans and Mexican Americans, of course, knew about it, either
from newspapers like La Opinion or corridos, actos, and cantares – forms of oral
culture – that told of the deaths of brave Mexicans who defied Anglo authority and
paid the price (Carrigan & Webb 2003: 425–426). Sung at parties, funerals, and
other ritual occasions, these laments kept alive the memory of the events genera-
tion after generation, celebrating figures like Juan Cortina who stood up for their
rights or avenged the murder of a friend (Carrigan & Webb 2003: 426; Navarro
2005: 116–117).
But if many in the Mexican and Mexican American community knew of the lynch-
ings, most mainstream Anglo historians did not. A few early scholars, such as Carey
McWilliams and Arnoldo De Leon, mentioned them (Carrigan & Webb 2003: 412).
But in contrast to black lynching, that of Mexicans is largely absent from America’s
collective record and memory.
The oversight, as mentioned, may be a product of many historians’ simply not
speaking Spanish. Others may have come across evidence of Latino lynching but
chosen to focus on the black variety which alone fell within the dominant paradigm of
American racial thought (Perea 1997: 1219–1252).
But America’s unique history with the group may also have played a role in
the silence surrounding Latino lynching. An emerging school of postcolonial juris-
prudence describes how colonial societies almost always circulate accounts of their
invasions that depict them as bearers of justice, science, and humanism (Delgado
2007: 1703–1718). The native or colonial subjects, by contrast, emerge as primitive
or hapless and in need of the civilizing force of the invading power with its superior
administration, better use of the land and its resources, and, of course, a higher form
of justice (Delgado 2007: 1703–1712).
The silken cord 
Edward Said, for example, writes that the West paints the Orient as exotic,
unknowable, and foreign in order to legitimize its hegemony over that region and to
persuade itself that its history of imperialism is entirely justifiable (Said 1993). Scholars
from other previously colonized regions, including India, Pakistan, Africa, and Latin
America, sound many of the same themes. The colonial subject emerges as unruly,
dark, untrustworthy, and in need of tutelage, discipline, and punishment (Ashcroft
et al. 1995). In such accounts, if the conqueror oversteps, beating a native to death, for
example, or punishing an innocent victim, the action is often excused as merely the
right idea gone awry (Carrigan & Webb 2003).
Written from the perspective of the conqueror, the official history thus always
shows him in the best possible light. Chicano historian Rodolfo Acuña writes that
Latinos living in the United States are, for all practical purposes, an internal colony
of the United States (Acuña 1972; Acuña 1981: vii–ix). If so, postcolonial theory may
explain elements of the relationship between the conquering Anglos and the subju-
gated Latinos, particularly in the Southwest, including the prevalence of lynching
and vigilante justice and their near-absence from the history books (Acuña 1972;
Blauner 1972).
3. English-only movements and their connection to Latino lynching
Movements to declare English the official language of the United States sprang up with
the early advocacy of U.S. senator and former university president S.I. Hayakawa and
political operative John Tanton (Perea 1992: 335–349). Currently, more than half of
the states declare English their official language, while Congress periodically consid-
ers national legislation to the same effect (Delgado et al. 2007: 237–240). Meanwhile,
many cities and states – including Jane Hill’s Arizona – gripped by anti-immigrant
fervor have enacted ordinances forbidding behavior associated with Latino immi-
grants, including the speaking of Spanish.2 A number of workplaces have begun
requiring that their employees speak English when interacting with the public or each
other,3 and even some taverns and other places of public accommodation have begun
requiring patrons to do the same (Delgado et al. 2007: 652–670).
The movement’s supporters argue that declaring English the official language will
promote Americanism and civic values, while encouraging immigrants to assimilate
. See Ruiz v. Hill 191 Ariz. 441, 957 P.2d 984 (1998), striking down an unusually harsh
Official English law.
. See, e.g. Garcia v. Spun Steak Co., 998 F.2d 1480 (9th Cir. 1993); Rivera v. College of
DuPage, 445 F.Supp.2d 924 (N.D. Ill. 2006).
 Richard Delgado
(Hayakawa 1990). At the same time, a related effort urges the abolition of bilingual
education in public schools (Delgado et al. 2007: 250). Its backers warn that allow-
ing immigrant schoolchildren to learn in both languages will slow their acquisition of
English and send the message that adoption of American ways is a choice rather than
a civic necessity.4 Both movements sprang up around 1990 when Latino immigration
began gathering force.
Given that the policy underpinnings of these measures seem alarmist, if not flatly
wrong, one wonders why they persist. In addition to sending signals about who belongs
to America, these laws regulate history and knowledge of the past. For example, they
convey the impression that the United States is an inherently English-speaking coun-
try, when, in fact, it is a product of many different streams of immigration, ethnicities,
and tongues (Perea 1992: 272). But they also shield from view historical events that the
dominant group might like to keep hidden.
Consider, for example, how a Latina child brought up in a Spanish-speaking
household is apt to acquire knowledge about the group’s treatment in the United
States through discussion with parents and grandparents. That treatment is full of
matters such as: a war of aggression; seizure of lands in the Southwest; broken treaty
obligations (Perea et al. 2007: 288, 296–302, 308–320); Jim Crow laws directed against
Mexicans (Perea 2004: 1426, 1439–1446); brutal Texas Rangers (Rosenbaum 1981);
crooked lawyers and land surveyors who conspired to deprive Latinos of their ances-
tral lands (Perea et al. 2007: 308–323); one hundred ten years of colonial status for
Puerto Rico (Perea et al. 2007: 376–379) – and, finally, lynching, the most lethal form
of mistreatment of all.
But imagine such a child who does not speak Spanish or only speaks it haltingly
because society punishes her for employing it and the school authorities do not pro-
vide the education that would enable her to preserve fluency in the two languages as
she matures. Such a child would be unable to converse with her grandparents about
life in the Southwest or Puerto Rico. Unable to absorb the cultural record through
newspapers like La Opinion, such a child could easily grow up believing that lynching
was largely a problem for blacks, that civil rights and the struggle for equal dignity
were largely black affairs, and that current racism, mock Spanish (“no problemo”) and
stereotypes maligning Spanish-speaking people as dirty, tricky, or stupid were simply
examples of tasteless humor unconnected with a history of colonialism and oppression.
. See Lau v. Nichols 414 U.S. 563 (1974) (requiring bilingual language instruction in a public
school system). See also Kevin Johnson & George A. Martinez, Discrimination by Proxy: The
Case of Proposition 227 and the Ban on Bilingual Education, 33 U.C. Davis L. Rev. 1227 (2003)
(discussing a California referendum limiting bilingual education in public schools).
The silken cord 
Postcolonial scholars from several continents have identified language as a prime

field on which the subaltern struggles for recognition and equal treatment. Writers
such as Trinh Minh-ha (Minh-ha 1989), Chinua Achebe (Achebe 1952), Haunani-Kay
Trask (Trask 1993), Ngugi wa Thiong’o (Ngugi wa Thiong’o 1993), and Frantz Fanon
(Fanon 1967) point out how the colonial subject forced to speak the colonizer’s tongue
loses contact with her own people. Indeed, until recently, American and A ustralian
administrators required young Indian or aboriginal children to attend boarding
schools where they learned to reject their own culture, acquire English, and forget
their native language (Dussias 2001: 819).
A similar phenomenon can set in, these scholars point out, when a writer like
Ngugi wa Thiong’o or Fanon writes for an audience of Western readers. The author,
who may be a staunch anti-colonialist and spokesperson for the native resistance,
unintentionally chooses terms, topics, and metaphors that will resonate with Western
readers living in both the occupied land and the home country of the occupying
force. She softens her punches and tries to find common ground with them. Ngugi wa
Thiong’o considers this a serious mistake, for the colonial language subtly incorporates
the worldview of the conquering society through terms like “folk medicine,” “tribe,”
“responsible leadership,” “civilization,” and “advance” (Ngugi wa Thiongo 1986: 4, 23,
27–29; Ngugi wa Thiong’o 1993: 285, 287). The writer finds it difficult to make an argu-
ment for liberation in a language whose very terms de-legitimize the native culture
and make any such effort sound quaint, sentimental, or even ridiculous (Ngugi wa
Thiong’o 1995: 438).
Braj Kachru, for example, writes that “the English language is a tool of power,
domination, and elitist identity” (Kachru 1995: 291); Simon During, that “for the post-
colonial to speak or write in the imperial tongues is to call forth a problem of iden-
tity, to be thrown into mimicry and ambivalence” (During 1995: 125). These scholars
find inspiration in Frantz Fanon, who insists that the native must forcefully assert
her identity on penalty of succumbing to despair and spiritual annihilation (Fanon
1967: 141–210). Recent studies of the second generation of Latino immigrants in the
United States show that their children who are born here exhibit much higher rates of
depression, drug-taking, and crime than their parents, who immigrated to the United
States as adults (Portes & Rumbaut 2001: 276–281). Might the severed connection with
their culture and history, compounded by failure to learn Spanish, be contributing to
this increase in pathology and social distress?
If so, English-Only laws and practices emerge as much more than misguided
efforts to achieve national uniformity or a pleasing linguistic sameness. Reminiscent
of former harsh practices, they inhibit adults in the ordinary business of work and con-
versation and convey the message that outsiders are not welcome unless they behave
according to standards set by others. They sever the cultural cord from one generation
to the next and hide histories of aggression, unprovoked war, lynching, segregated
 Richard Delgado
schools, and stereotypical treatment going back at least 150 years. They inhibit righ-
teous indignation and efforts to achieve redress, while leaving the young defenseless
against mistreatment they are ill-equipped to understand or counter.
English-Only orthodoxy is, thus, a form of lynching in at least two senses.
Although not physically lethal, it can inflict psychic and cultural damage. It also con-
ceals from view events – including actual lynchings – that call out for exposure and
reparative justice. English-Only laws and workplace rules are, then, aspects of the law
of the noose. As with a silken cord that tightens the more one struggles, they will not
choke if one stands still and does not resist. But the price is to go through life with
a silken rope around one’s neck. As with the teenagers at Jena, the cord operates as
a highly coercive sorting mechanism. We belong here, you there.
We should disavow any such conduct. Scholars should unearth other laws and
practices, such as the mock Spanish that Jane Hill criticized, which operate on dis-
tinct minorities the way language regulation operates on Latinos (Hill 1995). They
should oppose book banning in Tucson, Arizona, and crackdowns on Mexican
American Studies programs in public schools (Biggers 2012). Otherwise, marginal-
ized groups will find themselves in a condition similar to that which the postcolonial
scholars describe – alienated from themselves, co-opted, and unable to mount serious,
concerted resistance to illegitimate authority, if not dead.
References
Achebe, Chinua. 1952. Things Fall Apart. New York NY: Anchor.
Acuña, Rodolfo. 1972. Occupied America: A History of Chicanos. Saddle River NJ: Pearson.
Acuña, Rodolfo. 1981. Occupied America: A History of Chicanos, 2nd edn. New York NY: Harper.
Acuña, Rodolfo. 2000. Occupied America: A History of Chicanos, 3rd edn. New York NY: Harper.
Acuña, Rodolfo. 2005. Crocodile Tears: Lynching of Mexicans. For ChicanaChicanoStudies, July
18, 2005. 〈http://forchicanachicanostudies.wikispaces.com/Acuna,+%E2%80%9CCrocodi
le+Tears〉 (21 July 2011).
Ashcroft, Bill, Griffiths, Gareth & Tiffin, Helen (eds). 1995. The Postcolonial Studies Reader.
London: Routledge.
Biggers, Jeff. 2012. Who’s afraid of the tempest? Salon, 13 January 2012 〈http://www.salon.
com/2012/01/13/whos_afraid_of_the_tempest/〉 (13 January 2012).
Blauner, Robert. 1972. Racial Oppression in America. New York NY: Harper Collins.
Carrigan, William D. & Webb, Clive. 2003. The lynching of persons of Mexican origin or descent
in the United States, 1848 to 1928. Journal of Social History 37: 411–438.
Carrigan, William D. & Webb, Clive. 2005. “A dangerous experiment”: The lynching of Rafael
Benavidez. New Mexico Historical Review 80: 265–292.
Delgado, Richard. 2007. Rodrigo’s corrido: Race, postcolonial theory, and U.S. civil rights.
Vanderbilt Law Review 60: 1691–1745.
Delgado, Richard & Stefancic, Jean. 2004. Understanding Words that Wound. Boulder CO:
Westview.
The silken cord 
Delgado, Richard, Stefancic, Jean & Perea, Juan. 2007. Latinos and the Law: Cases and Materials.
St. Paul MN: Thomson/West.
During, Simon. 1995. Postmodernism or postcolonialism today. In The Post-Colonial Studies
Reader, Bill Ashcroft, Garth Griffiths & Helen Tiffin (eds), 125. London: Routledge.
Dussias, Allison M. 2001. Let no Native American child be left behind: Re-envisioning Native
American education for the twenty-first century. Arizona Law Review 43: 819–903.
Fanon, Frantz. 1967. Black Skin, White Masks. New York NY: Grove Press.
Gonzales-Day, Ken. 2006. Lynching in the West: 1850–1935. Durham NC: Duke University Press.
Hayakawa, Sessue I. 1990, March 20. English Is key to opportunities of American life. Reading
(PA) Eagle.
Hill, Jane H. 1995. Mock Spanish: A site for the indexical reproduction of racism in American
English. Language & Culture 10: 58. Language & Culture Symposium 2, 〈http://language-
culture.binghamton.edu/symposia/2/part1/index.html〉 (21 July 2011).
Holden-Smith, Barbara. 1996. Lynching, federalism, and the intersection of race and gender in
the progressive era. Yale Journal of Law and Feminism 8: 31–78.
Johnson, Benjamin Heber. 2005. How a Forgotten Rebellion and Its Bloody Suppression Turned
Mexicans Into Americans. New Haven CT: Yale University Press.
Kachru, Braj B. 1995. The alchemy of English. In The Post-Colonial Studies Reader, Bill Ashcroft,
Garth Griffiths & Helen Tiffin (eds), 291. London: Routledge.
Krugman, Paul. 2007, September 24. Politics in black and white. New York Times, A23.
Minh-ha, Trinh. 1989. Woman, Native, Other. Bloomington IN: Indiana University Press.
Navarro, Armando. 2005. Mexicano Political Experience in Occupied Aztlan. Lanham MD:
Alta Mira.
Ngugi wa Thiong’o. 1986. Decolonizing the Mind: The Politics of Language in African Literature.
Portsmouth NH: Heinemann.
Ngugi wa Thiong’o. 1993. Moving the center. In The World of Ngugi wa Thiong’o, Charles
Cantalupo (ed.), 219–220. Trenton NJ: Africa World Press.
Ngugi wa Thiong’o. 1995. On the abolition of the English department. In The Post-Colonial
Studies Reader, Bill Ashcroft, Garth Griffiths & Helen Tiffin (eds), 438. London: Routledge.
Nossiter, Adam. 2007, September 22. Black youth, conviction in beating voided, will stay jailed.
New York Times, A12.
Perea, Juan F. 1992. Demography and distrust: An essay on American languages, cultural
pluralism, and official English. Minnesota Law Review 77: 269–373.
Perea, Juan F. 1997. The black/white binary paradigm of race: The “normal science” of American
racial thought. California Law Review 85: 1213–1258.
Perea, Juan F. 2004. Buscando America: Why integration and equal protection fail to protect
Latinos. Harvard Law Review 117: 1420–1469.
Perea, Juan F., Delgado, Richard, Harris, Angela P., Stefancic, Jean & Wildman, Stephanie M.
2007. Race and Races: Cases and Resources for a Diverse America, 2nd edn. St. Paul MN:
Thomson/West.
Portes, Alejandro & Rumbaut, Ruben G. 2001. Legacies: The Story of the Immigrant Second
Generation. Berkeley CA: University of California Press.
Rosenbaum, Robert J. 1981. Mexicano Resistance in the Southwest. Dallas TX: SMU Press.
Said, Edward. 1993. Culture and Imperialism. New York NY: Vintage Books.
Trask, Haunani-Kay. 1993. From a Native Daughter. Honolulu HI: University of Hawaii Press.
Afterword
Jane Hill’s current work
Claire Bowern
(On behalf of the Dynamics of Hunter-Gatherer Language
Change team: Patience Epps, Russell Gray, Keith Hunley,
Jack Ives & Patrick McConvell)
Yale University
It gives us great pleasure to add an afterword to this volume concerning Jane’s current
(and future!) research. Some people use their retirement to start a new hobby – but
Jane got involved with a new NSF grant in addition to continuing her own research
program.
Jane is currently involved in a three-year (2009–2011) NSF funded project on the
dynamics of hunter-gatherer language change.1 The aim of the grant is to test hypoth-
eses about how language change occurs in hunter-gatherer communities, both within
regions and across the world, in order to draw generalizations about the ways in which
social structures might influence language change. We are testing generalizations in
specific case studies in three parts of the world: Northern Australia, Amazonia, and
California and the Great Basin (the latter being Jane’s particular domain of exper-
tise). Since hunter-gatherer communities vary extensively in their social complexity,
size/density, relations with their neighbors, and internal organization (Arnold 1996),
we need to make sure that any generalizations hold across case studies as well as
within them.
A further part of the project involves testing existing generalizations which have
been made about hunter-gatherers and the properties of their languages (Nettle 1999;
Trudgill 2010). For example, hunter-gatherer flora and fauna nomenclature systems
have been characterized as non-hierarchical, with few hierarchical levels (Brown
[forthcoming]; Berlin 1992). Preliminary work is strongly indicating that lack of hier-
archical structuring in hunter-gatherer nomenclature systems is areal, rather than a
property of hunter-gatherer systems more generally. The same is true for many of the
. BCS-902114 “Dynamics of Hunter-Gatherer Language Change”.

 Claire Bowern
other properties we have so far examined (see further Bowern et al. 2011). That is,
generalizations that have been taken to apply to hunter-gatherers as a class can now
be seen to be the result of specific areal and historical processes, rather than reflecting
universal processes and correlations.
This project owes much of its intellectual background to Jane’s earlier work (e.g.
Hill 1996, 1978). Her 1996 paper “Languages on the Land”, for example, raised impor-
tant questions about how we view language spread in history, how different types of
language spread might be associated with different cultural features, and how we might
model these differences in linguistics. She also highlights nuances in the interaction of
demographic features such as population density and mobility. In the introduction to
this volume, the editors describe the importance of Jane’s work in bridging the divide
between linguistics, linguistic anthropology, and social and cultural anthropologi-
cal studies; the breadth of contributions to this volume from students and colleagues
speaks to the success of this program, and has made it much easier for us to investigate
these topics.
The breadth of Jane’s contribution has included those which reach a wider public
such as penetrating analyses of how racism works in contemporary America – but
always with an eye to the linguistic detail that builds discourses and ideologies. Her
love for digging out the detail of languages is also given full scope in her work on the
Hunter-Gatherer project, where she has returned to fieldnotes and published works of
early pioneers who recorded the languages and cultures of western America. Her work
with C Hart Merriam’s materials, for example, show how valuable such records are
today, especially when integrated into a project which allows systematic comparison
across cultures. Jane reaches back to this often neglected work and forwards to a new
generation of researchers, situating herself as a prime link in continuing the linguistic
anthropological tradition.
This approach shines through not just in contemporary and synchronic linguis-
tic ethnography, but in prehistory as well. We might characterize this approach as
pragmatic uniformitarianism, in which assumptions about language change in the
past must be grounded in plausible theories of social interaction. In this view, proto-
languages are languages which were spoken by real people. While the past is another
country, it is still a country where children need to be carried and fed, where a day’s
walk is still a day’s walk, and where humans do not suddenly radically alter their
behavior without good reason. Using and borrowing words from others or refraining
to do so is part of a cultural stance by individuals and groups, which in turn is under-
pinned by needs of groups to share and interact or conversely to be independent
and symbolize that stance. Theories about linguistic prehistory often make specific
predictions about (and are bounded by) human behavior, and good theories make
plausible predictions.
Afterword 
As mentioned above, the Dynamics of Hunter-gatherer Language Change project

draws on detailed research into the three case study areas in order to shed light on
the way in which local and regional patterns might be relevant to our understanding
of hunter-gatherer languages. In one strand of work, we have been investigating loan
patterns within and between case study areas (Bowern et al. 2011). Hunter-gatherer
groups are often characterized either as isolated from their neighbors, or as highly
multilingual and therefore high borrowers, even into basic vocabulary. We took a list
of 204 items of basic vocabulary (that is, common, culturally neutral items) and coded
them for etymological status in 122 languages, including 47 languages of California
and the Great Basin, from 7 major lineages. This coding has proven very revealing
about contact patterns in the languages under study. Since we have information not
only about loan levels, but also about heritability of vocabulary items, it is possible to
tease apart the differences between stability vs loanability in this area of vocabulary.
This part of the project has also enabled us to look at word borrowing in language
shift. Our data indicate that language shift can go hand in hand with high rates of
borrowing, contra typologies such as Thomason and Kaufman (1988).
A second area of lexical research involves flora and fauna nomenclature systems.
We have compiled a list of functionally equivalent items across the three case study
areas and have researched their etymologies. Items have also been coded for their
function (that is, whether the materials are traded, eaten, used for weapons, or are
ritually important), whether they are dangerous and/or psychotropic, and whether
they are domesticated. Loan rates in flora/fauna are broadly correlated with loan
rates in basic vocabulary; they also differ by area, and by whether the culture is
primarily hunter-gatherer or agricultural. However, neither population size, density,
nor mobility appear (at this point) to drive loans in this domain. This implies that
loans in this area are driven by different factors from those which drive loans in basic
vocabulary.
Work on flora/fauna vocabulary has also given us a chance to investigate claims
about the structure of this lexical domain across languages. Work by Berlin (1992),
Brown (1974), Brown et al. (1985) and Urban (2010) has led to a picture of hunter-
gatherer nomenclature systems which are simpler than agriculturalist systems, which
contain few taxonomical terms at the generic rank, which all treat biological terms in
the same way, and which are less stable over time. This last point builds on work by
Hill (2011), following Balée’s (2000) observation that in Bolivian languages, names
for wild plants tend to be less linguistically stable than words for domesticates and
semi-domesticated plants. Preliminary work is showing that “flora and fauna” is not a
single domain when it comes to loans; that is, there are quite clear differences in dif-
ferent parts of this domain. Some of our results confirm those from previous research.
Terms for dangerous items are more likely to have undergone semantic shift; this is
 Claire Bowern
unsurprising, given the frequent taboo replacement of such names (see for example
Emeneau 1948). Other results are more surprising. For example, items of food are
more likely to be referred to by words which are inherited than items which are not
eaten; flora terms are more likely than fauna terms to have unetymologizable names.
Work like this requires detailed knowledge of the area, not only of the linguistics, but
also the trade routes, the flora and fauna, and ethnographic context in which such
items are used. We are very lucky to have Jane on the project and that the North
American case study is in such good hands.
The team who are working with Jane feel exceptionally privileged to be colleagues
of such an original and enquiring leader in linguistic anthropology and prehistory,
who is still doing great work and providing sound advice for those who are continuing
in this tradition.
References
Arnold, Jeanne. 1996. The archaeology of complex hunter-gatherers. Journal of Archaeological

Method and Theory 3(1): 77–126. doi:10.1007/BF02228931 (23 June 2009).
Balée, William. 2000. Antiquity of traditional ethnobiological knowledge in Amazonia: The
Tupí-Guaraní family and time. Ethnohistory 47(2): 399–422. doi:10.1215/00141801-47-2-
399 (17 January 2012).
Berlin, B. 1992. Principles of Ethnobiological Classification. Princeton NJ: P rinceton University
Press.
Bowern, Claire, Epps, Patience, Gray, Russell, Hill, Jane H., Hunley, Keith, McConvell, Patrick &
Zentz, Jason. 2011. Does lateral transmission obscure inheritance in hunter-gatherer
languages? PLoS ONE 6(9). e25195. doi:10.1371/journal.pone.0025195 (22 January 2012).
Brown C.H. 1974. Unique beginners and covert categories in folk biological taxonomies.
American Anthropologist, 325–327.
Brown C.H., E.N. Anderson Jr, R. Bulmer, P. Drechsel, R.F. Ellen, T.E. Hays, T.N. Headland et al.
1985. Mode of Subsistence and Folk Biological Taxonomy [and Comments and Reply].
Current Anthropology 26(1): 43–64.
Brown C.H. forthcoming. Ethnobiology and the hunter-gatherer/food-producer divide: How
hunter-gatherers differ from farmers in folk biological classification. In The Languages of
Hunter-gatherers: global and historical perspectives, Patrick M
cConvell, Richard Rhodes &
T. Güldemann (eds). Cambridge: Cambridge University Press.
Emeneau, Murray B. 1948. Taboos on animal names. Language 24(1): 56–63. doi:10.2307/410287
(17 January 2012).
Hill, Jane H. 1978. Language contact systems and human adaptations. Journal of Anthropological
Research 34: 1–26.
Hill, Jane H. 1996. Languages on the Land: Toward an Anthropological Dialectology. Bloomington
IN: Indiana University.
Hill, Jane H. 2011. Borrowed names in Northern Uto-Aztecan. In Information and its role in
Hunter-Gatherer Bands, Robert Whallon, William Lovis & Robert Hitchcock (eds). Los
Angeles CA: Cotsen Institute of Archaeology, UCLA.
Afterword 
Nettle, Daniel. 1999. Linguistic Diversity. Oxford: OUP.

Thomason, Sarah Grey & Kaufman, Terrence. 1988. Language Contact, Creolization, and Genetic
Linguistics. Berkeley CA: University of California Press.
Trudgill, Peter. 2010. Social structure, language contact and language change. The SAGE
Handbook of Sociolinguistics, Ruth Wodak, Barbara Johnstone & Paul E. Kerswill.

London: Sage.
Urban, Matthias. 2010. Terms for the unique beginner: Cross-linguistic and cross-cultural
perspectives. Journal of Ethnobiology 30(2): 203–230.
Language index
A G P
Algonquian 264 Gĩkũyũ 229–238, 243–251 Paiute see Northern Paiute
Athabaskan 29–33, 39–42, 44, Paiwan 152, 153
48–49 H Petitot’s Montagnais 35
Austronesian 152 Hare 29–41, 43–49 Proto-Athabaskan 31, 32, 39,
Hiaki 133–146, 149–151, 41, 44
B 153–165, 167–170 See also Athabaskan
Bahasa Indonesia 91 Hopi 149–150, 153, 326
See also Indonesian Huichol 3 Q
Bantu 136s Quiche 66
Bearlake 30 I
Blackfoot 258–275, 280–287 Indonesian 393 S
Breton 160 Salish 182, 185, 187–189, 191,
K 192, 193
C Kawaiisu 87 Sanskrit 150, 159–162
Cahuilla 86 Kiswahili 229–234, 236–246, Serrano 296, 299, 301
Castellano 55, 208 248, 250–252 Slavey 29, 48
Cayuga 86 Kutchin 35, 46 Southern Interior Salishan 182
Chemehuevi 87 Southern Numic 85, 87
Chickasaw 86 M Southern Paiute 87, 110
Chol 66 Mam 66 Southern Ute 85, 86, 88, 90,
Classical Nahuatl 149–150, 171 Mexican Spanish see Spanish 93, 102
Coeur d’Alene 175–197 Mexicano ix, xvii, xviii, Spanish 5, 39, 45, 53–56, 65–70,
Comanche 87, 110 54–56, 69, 204, 206, 208, 72, 74, 77, 78, 81, 113, 114,
Corsican 136 210, 219–220, 291–315, 383 203–224, 291–316, 333, 340,
Creek 86 Mohawk 86 373, 384, 392, 415–422
Cupeño 134 Montagnais see Petitot’s Mexican Spanish 205–211,
Montagnais 214–215, 218, 221–223
D
Dene 29–41, 46, 48 N T
Dogrib 38, 46 Nahuatl 54, 55, 153, 203–214, Takic 3, 17, 87
218–223, 291, 293, 296, 298, Tanana 86
E 302, 304–309, 313, 314 Tarrahumara 3
English 5, 7, 9, 14, 37, 48, 56–57, See also Classical Nahuatl Tepiman 135
61, 63, 68, 70, 88, 90, 94, Nahuatlato 301 Tlaxcala 291–316
156, 158, 168, 204, 219, Navajo 86, 348 Tzeltal 66, 216
221–223, 226, 229–234, Northern Paiute 18 Tzotzil 66, 216
236–237, 239, 245–246, Northern Uto-Aztecan 3, 17, 85
248–251, 262–263, 267–269, Numic 3, 18, 85 U
273, 280, 282, 287–288, 306, See also Southern Numic Ute 3–26, 85–104
312, 314, 316, 323–326, 329, Uto-Aztecan 3, 15–17, 26, 85,
345, 350, 355, 383, 390–392, O 87, 107, 109, 110, 128, 130,
398, 415, 417, 418–422 Oneida 86 135, 136, 149, 150, 153
 Language index
W Y Yokutsan 39
Western Apache 86 Yaqui 3, 133–171 Yowlumne 39
Western Mono 321–323, See also Hiaki Yucatec Maya 53–82, 204,
328–333 Yawelmani 39 214, 215
Yokuts 321–335 Yucatecos 65, 66
Subject index
A assimilation case-marking 3, 7, 13–15, 26

Aamsskáápipikani 259, 261, cultural 324, 333, 357, 367 causative 5, 69, 112–113, 128,
262, 272–273, 275–276 phonological 109, 119, 137, 146
accent 54, 55, 392, 394, 121–125, 128 Chan Kahal 66–67
404–405 audience response 400–401 Chinua Achebe 421
accent stress 53, 56, 60–61 authentic viii, 298, 333, 382, Chomsky xiv, 324
accent tonal 86 384, 406, 408 Chumayel 66
acoustic 53, 63–64, 86, 394 authenticity 293, 296, 299, clitic 135, 141, 143, 161–162, 165,
adjective 5, 14, 140–141, 213 303, 403 170, 217–219, 227
adstratum 214–217 authenticating 328, 330 cliticize 19, 26
adverb 5, 14, 138, 329 auxiliaries 124, 128 coda 284–285, 331, 333
adverbial clauses 13, 18, 25, 183 syllable coda 59–60, 137, 154
affix 13, 129, 136, 143–146, B code-switching 68, 210
168, 170 back vowel 89–90, 118–119, 122 collaborative 107, 111, 119, 177,
affix list 183, 187, 193 Bakhtin 284, 302, 333, 365, 184, 259, 283, 285, 287, 307,
Akil 66 372–373, 380, 383, 385 323, 406
Alaska 44, 49, 176 benefactee 141 colonial 204, 207, 209, 213,
alienable possession 11 Berkeley vii, ix, 324 232–233, 240, 291, 294,
allative 23–24 best practices 179, 184, 259 299, 315, 346, 375, 418–422
allomorphy 124, 137, 154 bilingual 54–55, 65, 67, 82, colonialism 293, 315
allophones 33, 212, 226 204–205, 207, 212–217, colonialist 334
allophonic distribution 30 220–221, 257, 304, 309, Colorado 85–87, 94, 405, 417
allophony 34 312, 373 Colville Lake 29, 36
alternation 33–34, 59, 89, bilingualism 48, 69, 211, 230, complementary
112–113, 189, 227, 329 236, 293, 315 distribution 30, 125–126
alveolar 62, 111, 125–126, bilingual education 108 compositionality 153–154
128–129, 217 black activism 368–369, 380 compound 5, 91, 133–171
analytic 204, 214 Blackfeet 259–261, 263–264, conditioned polymorphism 216
anthropologist 46, 108, 178, 267, 272–273, 348, 352 confirmatory function 134
303, 314, 321, 324, 365, 384 Blood 259, 260 conjunct 34, 42–43
anti-Latino nativism 415 Bolivian 427 conjunction 5, 7, 124, 220, 293
Apátohsipikani 259, 261 borrowed nouns 5–6 connected speech 109–110,
appropriate access 179, 180, borrowing xxii, 41, 55, 125, 119–125
184–185, 196–197 204–206, 210, 426–427 consciousness xiv, xvii–xviii,
archive 175–177, 179, 181–184, Braj Kachru 421 284–285, 370, 384
188–193, 196, 305, 346 Brazil 365–385 black consciousness 375
Archuleta 87 consciousness raising 375,
argument 138, 141, 166 C 378–379
Arizona 86, 107, 136, 417, California 39, 65, 321–325, practical consciousness 355
419, 422 333–355, 417, 420, racial consciousness
Arnoldo De Leon 418 425, 427 366–368
aspect 5–6, 9, 14–15, 23, Canada 29–30, 180, 259–261, consonant 5, 9, 23, 31, 33, 35,
33–34, 40–44, 113, 137, 264, 266, 273, 316, 339 41–42, 44, 59–60, 62, 71,
141, 152–155, 157–158, 162, case study 175, 177, 181, 277, 81, 88, 109–116, 125–126,
165, 326 427–428 128–130, 208–209, 216, 398
 Subject index
consonant inventory 62 demonstratives 5–7 diversity 207, 218, 222,

constituent 59, 136, 138, 141, dental 46, 62, 88, 111, 125 242, 257–259, 265, 266,
156, 163, 165 depoliticized 249 268, 272–276, 280–281,
sub-constiutuent 149, 151 derivational 42, 129, 134, 139, 283–285, 287–288
constraint 26, 30, 42, 59, 60, 143, 145–146, 148–149, documentary 178, 182, 305
71, 111, 164, 167, 278–279, 150, 155–156, 158, documentary linguistics
282, 411 169–170 257–259, 268, 277
contact 29, 31, 44, 46–49, 86, determiner 14, 137, 140 documentation 57, 85, 88, 93,
203–208, 210–212, 214, 216, de-verbal suffixes, see suffix 119, 133–134, 150, 176–178,
219, 221, 223–224, 259, 266, devoiced vowel 4, 17, 89–90, 182, 184, 188, 196, 219,
272, 278, 280–284, 315, 410, 94, 96 259, 284
421, 427 diachronic graveyards 7, 26 Don Gabriel xvii, 365,
culture contact 325, 333 diachrony 3, 17, 26 383–384
contextualized 249 dialect(s) 33, 35, 37, 47–49, Dorothy Nicodemus 182
decontextualized 247, 322, 56, 66, 91, 93, 107–111, duration 53–54, 56, 59, 64–65,
390, 394, 405–406 116, 118–119, 128, 211, 77, 82, 85, 86, 89–98, 99,
recontextualized 327, 259, 262, 264, 271–274, 101, 103
331, 346 278, 280, 284–285, 390,
contrast 392–393, 398 E
of sociolinguistic dichotomy 55, 346, 382 edge-marking 135, 153, 155–156,
identity 242, 251, 350, dictionary 35–37, 39, 42–43, 159, 162–163, 169–170
353, 355, 369–370, 373, 374, 45, 67, 93, 109, 119–121, education(al) xi, xxvi,
376–378, 383, 384 124, 182–183, 185, 187–188, xxviii, 67, 68, 108, 178,
phonological 46, 56, 58–59, 191–193, 212–213, 265, 203, 205, 221–222,
110–113, 117, 126, 209–210 271–275, 281, 369 229–230, 232–234,
morpho-syntactic 4, 13–14, digital archiving 175–179, 184 236–237, 239–241,
17, 24, 45, 138, 140, 146 digital resource creation 175 249–250, 263, 271, 280,
contrastive expectations 325 digitizing 177–183 303, 315, 330, 341, 359, 368,
contrasting cultural diphthong 111, 118, 226 375, 380–381, 391–392, 420
perspectives 327 directional 23–24, 137 Edward Said 409
conversational directional mood 122, 124 ejective 30, 37, 62, 63
data 53, 68 discourse(s) 207, 217, 230, elicitation 67, 70, 94, 96,
historical present 329 234, 236, 239–240, 216, 218
narratives 365 246–251, 257, 277–279, elicited 53, 67, 85, 94, 119, 401
partner 267 283, 287, 293–296, emotive words 113
cordial racism, see racism 301–303, 306, 322, 327, emphasis 137, 411
coronal consonant 110, 330–333, 339, 341, 347, and stress 53, 56
112–113, 125, 128–129 350, 356, 358, 360, 366, emphatic 137, 218
corpus 41, 120, 193 383–384, 389–392, 395, empirical 133, 134, 141, 149, 217
covert racism, see racism 396, 402, 407–408, endangered (language) 67,
creaky voice 63, 65, 79–80 410–411, 418, 426 85, 86, 88, 91, 94, 104, 134,
CSS (Cascading connected 120 175–176, 178–180, 197,
Stylesheets) 186, 193 covert racist 339–340, 357 204–205, 221, 223–224,
culture(s) 82, 130, 175, 221, analysis 339, 367, 415 229–230, 234–235, 250,
241, 246, 250–252, 258, data 109, 119 277, 284, 310, 323, 358
292, 300, 309–310, 314, discursive discrimination epenthetic 9
322, 324–325, 333–335, 343, 322, 335 erasure 283, 323, 333–335, 359,
346, 351, 353, 355, 358, 360, disjunct 34, 43 374, 394–395, 402–403
374–375, 418, 421, 426–427 Distributed Morphology Eskimo 46
CV interactions 107 134–135, 151, 154, 163, 170 ethnographic 46, 230, 258–259,
distribution 9, 12–14, 18, 30, 277–278, 284, 292, 310, 315,
D 42–43, 89, 107, 109–115, 321–322, 326, 333, 340, 357,
Deguthee Dinees 46 118, 124–130, 133, 170, 257, 394, 397, 428
delegitimated 390, 392 259, 271, 284, 384 ethnopoetics 321, 324
Dell Hymes 119 distributional pattern 110–111 evolution xii, xiv
Subject index 
F Head Application Principle ideology 203, 217, 231, 239, 243,

field notes 182–183, 185–189, (HAP) 162 249, 251, 257, 267, 277–279,
194–195 head noun 14, 18 281–282, 291–297, 299,
fieldwork 31, 36, 91, 108, head-marking 133–135, 303, 307, 310, 314, 339,
189, 230, 249, 287, 292, 149–150, 153, 156, 159–160, 342, 365, 367, 393, 402,
365–366, 370, 375 162, 165, 168–170 406–408
finite 14, 15, 22, 26, 170 head-movement 155–156, ideology of standard 259
flora/fauna vocabulary 425, 165–167 Ignacio, California 87, 94
427–428 heavy diphthong 111 image file 187
fluency 9, 133, 226 heavy syllable 59, 60–61, 70, imperative 112, 137
fort 46 81, 137 imperfective 40, 43, 44, 112,
Fort Good Hope 30, 36, 39, hegemonic 205, 214, 249, 323, 120–121, 128
46–48 326, 330 inanimate 24
Fort Norman 47 heritage language 178, inanimate suffix 13
frame sentence 67–68, 85 203–204, 324 indexicality 340
Frantz Fanon 421 Heteroglossia 284–285, 287, 365 indigenous 378
Franz Boas 182, 185 high vowel 107, 109–111, 113, aesthetics 333
free polymorphism 216 116–117, 119, 125–130 communities 86, 104, 108,
frequency 7, 17, 25, 29, 42–43, Hillary Clinton 289–390, 395, 181, 223, 291–293, 297,
49, 111 398, 405, 406 303, 334
fricativization 223 Hispanic-centrism 216 culture(s) 333
fronting 109, 112, 115, 117 historical linguistics xxi, 29, discourse (narrative
function words 109, 120–124 119, 287 traditions) 322–323,
fundamental frequency 79, 96 homogeneous 233–234, 325, 330
fur trade 46 241, 257 education 239, 303
Honorific 219, 222, 294, language(s) 53–55, 86, 136,
G 301, 306 203–207, 211, 214, 216,
geminate 65 hortative 112 218–222, 229–231, 233, 235,
gemination 137, 154 HTML 186–188, 190, 192 238–242, 248, 250–251,
generative grammar 133, hunter-gatherer 425–427 291, 296, 302, 315, 331
170, 324 language change 425, 427 language ideology 292,
genitive 4, 5, 7, 9–10, 12–18, 26 nomenclature systems 425 299, 323
Gerald L. Hassler vii Hyacinthe Andre 48 people 213, 218, 223, 298,
Gladys Reichard 182 hybrid verb constructions 346, 367
globalization 242, 246, 251 135, 144 populations xiii, 223
glottalized 57, 63–65, 215 hyper-correction 249 indigenousness 334
Graded Intergenerational inflection 133, 138, 141–142,
Disruption Scale 88 I 149–150, 151, 153, 156–158,
grammar 88, 132, 145, 156, 164, iamb 59 160, 169
168–170, 183, 187–188, 193, Idaho 182, 260 inflectional 41, 134, 144, 149,
266, 272, 279, 293, 324–325 ideological 203, 217, 230–231, 153–154, 156–162, 169
grammatical factors 29 236, 238, 240–241, 247, affixation 139, 149, 156,
grammatical theory 133, 170 257–259, 271, 277–284, 287, 161, 170
grammaticalization 3, 16, 18, 291, 299, 303, 313, 322–323, case 152
26, 168 341–342, 359, 378, 380, 384, reduplication 135–137, 149,
grass-roots 175, 181, 184 394, 407–408 160–161, 169
ideologies 185, 204, 224, intensity, phonetic
H 229–237, 239–245, loudness 71, 73–76,
habitual 5, 16, 137, 141, 145–146, 248–252, 257–259, 266, 90–91, 93, 96, 103
153–154, 156–157, 162, 165, 277–280, 282, 284, narrative 319
326, 390 291–292, 294–297, 299, of audience response 401
harmony 59, 60, 121, 367, 303, 310, 315–316, 321, 333, of contact 207
378, 384 335, 358–359, 366, 368, interdisciplinary xi, xii, xiv
Harry Hoijer ix 381, 384–385, 391, 406, Internal Reconstruction 3, 26
Haunani-Kay Trask 421 409–410, 426 interpellation 342, 359
 Subject index
intonation iv, 60, 68, 369, language learning 86, 251, 268 loan word(s) 53–56, 68–70, 73,
398, 411 language maintenance 108, 77, 80, 82, 113–115, 291, 293
intonation(al) contour 54, 205, 294, 295, 299 Local Dislocation 135, 159,
60, 61, 64, 94 language mixing 293 164–165, 167–168
intonation(al) phrase 64 language rights 86 locative 11, 18–21, 23–25,
intonation unit 385 language shift 29, 204, 208, 120, 208
intransitive 140–141, 160 230, 234, 235, 236–239, lynching 415–422
intransitivizing 140 245, 249–250, 252, 295,
Inuit 46 299, 302, 323, 427 M
invariant [r] 32 language socialization xv, 245, Mackenzie River 30, 47
invisible 6 281, 299, 301 main verb 8–9, 15, 19, 20,
language workers 175, 177, 22–23, 26, 145, 161
J 179–181, 186, 196–197 maintenance–shift
JavaScript 186, 188, 191 laryngeal 116, 122, 124 continuum 205
Jim Crow 415, 420 law 248, 375, 381, 415, 417, Malinche xvi, 209, 220,
Joaquin Murietta 417 420–422 292–294, 299, 303,
Jose Pancho 120 Lawrence Nicodemus 182 311–312, 316
Juan Cortina 417 legacy materials 177, 181–182, mediatization 340–341, 347,
Julia Antelope 182 184, 196 389–391, 394, 402, 406,
Julienne Andre 48 lexeme 149, 159, 161–162, 349 409–411
lexical stress 56, 58, 60 metalinguistic 281–283,
K Lexicalist 134–135, 150–151, 365–366, 372–374, 377–378,
Kainai 259, 261, 264–265, 153, 155, 156, 158, 159, 163, 382, 384, 403, 407
271–273 169, 170 metrical 53, 397
Karl Marx Hassler viii Strong Lexicalism 134, 169 feet 59
Kenya(n) 229, 230, 232–237, Weak Lexicalist stress 56–57, 59
239–242, 245, 247–250, 252 Hypothesis 151 Mexico ix, xi–xii, 53–54,
kin terms 10 light syllable 61, 137 66, 135, 203, 206–207,
linearization 156, 164–165 209, 211–213, 215–218,
L lingua franca 204, 229 222–223, 291–293,
La Opinion 418, 420 linguistic anthropology xi, 295–297, 311–316, 417
La Plata (county) 87 xv, 86, 239, 286–287, 291, mid vowel, see vowel
language archive 179, 181 324, 326, 366, 426, 428 migration 214, 420
language attitudes 242 linguistic diversity 134, 203, minimal pairs 56, 57, 63, 65, 91
language attrition 88 258–259, 263, 266, 280, 284 minority language 104, 144
language change 43, 306–307, linguistic fieldwork 91 mock Spanish 221, 333,
425–427, 429 linguistic inheritors 257, 420, 422
language community 91, 94, 258–259, 272, 393 mock Spanish lexical
104, 108, 182, 271, 286, 324 linguistic prehistory 426 forms 334
language contact 29, 31 linguistic repertoires 257, 278, monosyllabic 21, 57, 58, 152
language endangerment 175, 392–393, 396 Montezuma 87
232, 252, 283–287 linguistically inherited 390 mora 58, 59
language ideological 258, 259, linguistics ix–xvii, 55, 94, 134, extra mora 60
277–280, 322, 323, 394 179, 184, 188, 215, 258–259, bimoraic 61, 82
language ideologies xvii 269–271, 286, 302–303, morpheme 9, 26, 30, 32–35, 39,
229–237, 239–242, 245, 314, 324–325, 327, 426, 428 41–46, 49, 69, 94, 146, 148,
248–252, 257–258, 266, descriptive linguistics 302 151, 155, 167, 263, 269, 275
277–278, 297, 299, 303, documentary aspectual morpheme 155
322–323, 335 linguistics 257, 268 morpheme boundaries
language ideology 248–249, historical linguistics 30, 41, 62
267, 282, 291–292, 294, 119, 287 morpheme-final 31, 41
297, 339, 365, 394 Uto-Aztecan linguistics 135 morpheme-initial 31, 41
language learner(s) 93, 258, 271 LIVE 257, 278–279, 284 morpheme-internal 41
Subject index 
morpheme-specific conversational Ofelia Zepeda 108, 111, 119,

operation 164 narratives 366 121, 130
overt morpheme 166 metanarrative 331, 333, 334 official orthography 121, 122
reduplicative national narratives 378–380, Oklahoma 86, 109, 417
morpheme 135, 154, 167 383, 384 online 175–177, 179, 181, 182,
RED morpheme 164, narrative 184, 186–188, 190, 196, 197,
165, 167 discrimination 183, 335 287, 403, 405
√ morphemes 168 narrative norm 326 onset (syllable) 42, 60, 124, 137
Morphological Doubling narrative relativity 334 onset (phonetics) 94, 96
Theory (MDT) 136 narrative style 325, 335 Optimality Theory 278
morphologization 4 nasalization 30, 33–35, 40–45 oral environment 30–34, 39,
morphology 12, 17, 19, 23, 29, nasal 29–32, 39, 40–46, 49, 217 42, 43
33, 34, 50, 59, 69, 70, 72, nasal-initial syllable 41 orthography 65, 67, 68,
81–83, 91, 95, 98, 100–103, palatal anasal 39, 45, 46, 216 88, 89, 121, 122, 182,
108, 121, 148, 183 nasal environment 30–34, 185, 187–189, 191–193,
agglutinative 42–44, 49 269, 286
morphology 86 non-nasal environment 31 Other than Standard
case-marking nationalism 229, 230, 232, 236, 390–392, 409
morphology 3, 7, 26 239, 240, 241, 247, 249, other-than-standard 392,
Mayan morphology 69 250, 334, 359, 417 393, 395
Spanish morphology 69 natural class 109–111, 115–117, overcorrection 209, 212
suffixal morphology 112 119, 125, 129 Oxcutzcab 66
morphosyntax 134, 210, 212 negation 25
motion verb 23, 24 double negation 25 P
movement, negation (attitude) 310 palatal 45, 111, 125–129, 210
seasonal mobement 323 New Mexico 86, 417 palatal nasal 39, 45, 46, 216
scholarly movement 324 NI constructions, see noun palatalization 127, 129
movement syntactic 141, 166 incorporation palatalized 111, 128
clitic movement 170 nominalization 3, 9, 14, 16 paradigm 16, 33, 35, 42, 43, 112,
head-movement 155, 156, nominative 3–5, 166 113, 128, 160, 161, 163, 204,
166, 167 noose 415, 416 221, 234, 418
head-to-head North Piegan 259 Paradigm Function
movement 155, 156, 164 Northwest Territories 29 Morphology 134, 135, 159
movement noun incorporation 133, 135, PFM 134, 135, 159, 160, 162,
phonological 118, 119 136, 138, 140, 146, 154–156, 169, 170
movement political 216, 163, 165, 170 Paradigm Uniformity
306, 419, 420 NI constructions 135, 139, Generalization 135, 160
anti-racist movement 377 140, 142, 143 PUG 160, 161, 162, 163, 169
black movement 366, 368, noun stem 10, 139 paradigmatic feature 58
369, 373, 375, 382, 384 noun-class suffixes, see suffix paradigmatic token 112
Civil Rights Movement 382 PDF 186, 187, 189, 193, 195
movement social 381 O peak pitch 91–94, 99, 100–103
Multilingualism 49, 203, 236, Oakland 66 perfective 33, 35, 40–45,
251, 323 object 3, 4, 7–9, 11–19, 26, 32, 113, 128
multilingual fusion 249 34, 42, 44, 128, 138–140, perfective of causative 112
Musa Yasin Fort 184 155, 156, 218, 219, 221 Personalism 391, 406–409
object cliticization 163, 169 ideology of
N object clitic 135, 143, 161, personalism 402
narrative 88, 109, 120, 123, 162, 170 Peto 66
182, 183, 194, 196, 237, oblique 3–7, 9, 12, 14, 17, phonemic 53, 54, 56, 62,
293, 294, 297, 299, 302, 18, 34 63, 189
321–331, 333, 335, 344, oblique object 34, 44 morpho-phonemic 9
346, 348 obstruents 88 phonemic tone 64, 77
 Subject index
phonetic 7, 17, 41, 53, 55, 56, 65, prosodic morphology 107, right-adjunction 156
71, 85, 86, 88–91, 93, 94, 129, 134 Right-Hand Head Rule 158
96–100, 103, 104, 111, 118, prosodic shift 56, 70
129, 189, 213, 215, 217, 258, prosodic system 54, 55, 79, 82 S
264, 398, 407 prosody 56, 57, 73, 96 safe storage 179, 180, 185, 187,
phonetic correlates of Protocols for Native American 196, 197
stress 91, 93, 94 Archive Materials 177 salvage era representations 321
phonetic instrumental proximate 44 San Francisco 65–67, 82
analysis 85 purism 204, 224, 283, 291–297, San Rafael 66, 67
phonology 40, 42, 62, 94, 107, 299, 302, 313 second person singular 33, 34,
110, 119, 124, 129, 133, 204, puro Maya 55 39–46, 49
210, 211, 263 semantics 134, 144, 146, 154,
phonological 4, 6, 17, 19, Q 213, 217, 382
33, 40, 41, 44, 54–56, 58, qualifier 33, 34, 43 lexico-semantic 258, 264
65, 85, 86, 89, 107, 109, quantity sensitive 59, 60, 70 semantic 53, 136, 137, 148,
110–113, 117, 119, 124, 125, 154, 217, 265, 373
129, 137, 139, 140, 143, 144, R semiotic register 340, 390, 392,
148, 151, 153, 156, 164, 165, racial violence 415 393, 396, 402, 410
190, 208–210, 212, 213, 215, racism 293, 295, 315, 319, sentential subjects, see subject
258, 264, 269 322–335, 366–372, 378–383, Siksiká 259, 262, 263, 264, 271,
phonological features 58, 406–409, 415, 418, 420, 426 275, 277
65, 107, 119, 210 anti-racism 374, 375, 378, 384 situation aspect marker 33
phonological shift 29 cordial racism 366, 367 Slave 46
phonological word 139, 148, covert racism 321–322, slave(s) 367, 369, 375–379
149, 155 333–335, 339, 385 slavery 367, 375–376, 380
PHP 186, 190–192, 197 White racism 339, 347, semi-slavery 213
phrasal accent 56 358, 359 social factors 29
physiognomy 211 reciprocal relationship 269 social justice 108, 380
Piegan Institute 263, 264, RED 151, 155, 156, 163–165, socioeconomic 234, 236,
270, 271 167, 168 239–240, 293, 315, 322
Pima 118, 135 reduplication 107, 113, 129, sociolinguistic
pitch 53, 56, 59, 63, 71, 73, 77, 133–137, 141–163, 165, competence 216
82, 86, 90–94, 96, 99–103 168–170 sociolinguistics 221, 287,
plural 4, 22, 43–45, 69, 119, 122, reduplicated 114, 115, 128, 296, 303
128, 141, 160, 204 136, 143, 147, 152, 154, 157 sound change 30, 42
plural imperative 112 reduplicating 143, 144 sound shift 39, 41
PNG 187, 189, 193, 195 referentialist ideology 408 sovereignty 88, 348
polysynthetic 183, 187, 204, 210 relativism 321, 325, 326, 333 spectrograms 63, 81, 90–92, 96
postcolonial 418, 419, 421, 422 REL-clauses 7 Spell-Out 155, 166
post-positions 3, 18–21, 23–26 relic zones 26 standard(s) 116, 120, 121, 135,
pre-aspiration 107 repetition 93–96, 237, 302, 323, 149, 150, 151, 167, 177, 179,
predicate 3, 4, 17, 161, 165 326, 327, 328, 329, 330, 384 187, 190, 210, 214–219,
prefix(es) 31–34, 39, 42, 43, 45, representational 232–240, 247, 257–259,
91, 124, 129, 134, 151, 152, economy(ies) 391–393, 408 270, 277, 278–288, 326,
160, 161, 163, 164, 170, 231 reservation 85, 87, 94, 108, 335, 345, 389–409, 421
present progressive 152 259, 263, 264, 267, 272, standard dialect 116, 118
productivity 153, 154 273, 348 standard language 259, 272
progressive 121, 137, 152, 153 Response density 400, 401 standard pronunciation 70
prominence 56, 58, 79 restructuring 30, 41, 42 standard speech 121–124
pronoun(s) 4, 7, 9–14, 17, 45, retroflex 109–113, 115, 117, standard spelling 189
122, 350 125, 128 standard transcription 30
prosodic 53, 54, 69, 70, 79, 82, revitalization 86, 88, 176–178, stative 24, 25, 32, 35, 39, 40, 41,
85, 86, 93, 103 303, 307, 309, 310, 324, 346 45, 49, 153
Subject index 
stem(s) 5, 6, 14, 31, 39, 57, symmetric role-alignment 399, tribalism 230, 232, 239, 240,
91, 137, 144, 146, 148, 151, 402, 403 242, 243, 246, 247, 248,
160, 170 synchronic data 3, 26, 109 249, 250
compound stem 136, 151 syncretic continuum 55, 69, 71, trigger(ed/ing) 39, 109, 112,
CVC stems 41 77, 79, 80 115, 119–124, 128, 129, 137,
lexical stem 26 syncretic speech 191–316 151, 153, 156, 164, 170
noun stem 10 syncretism 53–55, 204, 220, triglossia 229
nominal stem 138, 139, 291–292, 295, 297, 303, triglossic 229, 235
146, 152 307, 309, 314 Trinh Minh-ha 421
verb stem 30, 40, 136, 141, syntacticocentric 134–135, trochee 59
142, 145, 146, 154, 183 150–151, 155–156, 163–170 truncated repertoires 392
verbal stem 142, 146, 151, syntagmatic 58–59 truncation 107, 113, 115, 129, 385
152 syntax 16, 133, 150, 159, typological 26, 107
stem list 183, 193 164–166, 170, 265, 299 Typologically 214
stem-initial 31, 44 typology 26, 141, 159, 205
stress 53–62, 67, 70, 73, 82, 85, T
90–104, 107, 222, 244, 411 Taiwan 152 U
Strong Lexicalism, see tapping 93, 94 underlying form 54
Lexicalist TAPS checklist 177 unicode 187
style(s) 55, 241, 275, 278, 324, Taracahitic 136 uniformity of exponence 29,
327, 355, 358, 397 Target(s) 109, 116, 118, 121, 30, 31, 42, 43, 44, 49
linguistic style 325 124, 142, 144, 146, 148, unstressed 4, 17, 58, 60, 90, 91,
narrative style 323, 325, 355 152, 159, 162, 168, 218, 92, 96–104
preaching style 390, 395, 360, 376, 409 upscaling 411
402, 410 target word 94 uptake 281, 283, 347, 377,
speech style 340, 348, 350, target language 270 389–391, 393, 395, 399,
401, 407 tense-aspect 9, 14–16, 326 402–406, 409, 410
style-shifting 393, 399, 403, tense-aspect-modal urbanization 214, 237
406–409 marking 14 Utah 86, 87
subject 3–18, 32–34, 41–45, 137, terminative 34
139, 166–167, 350 Texas Rangers 417, 420 V
sentential subjects 15 texts 8, 42, 120, 189, 190, value project 389–395, 402,
subjunctive 6, 59 194–196, 222, 302, 406, 409–411
subordinate clause 7, 15, 314, 324, 327, 328, variation(s) 4, 265, 267, 271,
26, 143 340, 354, 357, 376, 274, 277–284, 286, 288
subordination marker 13 396, 403, 404 phonetic/phonological
suffix(es) 4–27, 41, 59, 60, 115, recorded texts 9, 12 4, 9, 33, 34, 37–39, 45,
128–129, 140–143, 145–147, written texts 179, 340 91, 107–109, 111, 116, 189,
210, 222, 265, 274–275 third person 6, 33–34, 43–44, 216, 264
de-verbal suffixes 20–21, 120–123, 141 dialect variation 259,
24, 26 tonal accent 86 273, 274
noun-class suffixes 17 tone 53–54, 57–58, 62–67, free variation 160, 216
suffixal 112, 150, 164 77, 215 geographic variation 264
suffixation 113, 138 tones 30, 36 ideological variation 257, 258
suffixed 128, 139 TPR (total physical lexical variation 216
verbal suffixes 17, 135, response) 270 linguistic variation 258, 264,
143, 146 transitive 137, 138, 140, 141, 265, 266
swastika 416 219, 221 V-complements 7
syllable 18, 30–34, 40–45, translinguistic 217, 365, 366, velarization 37
54–63, 67, 70–82, 90–96, 373, 374, 377, 378, 380, verb(s) 4–26, 30, 35, 42, 43, 112,
107, 137, 139, 148, 152, 208, 383, 384 113, 133, 134, 135, 137–170,
216, 222 translinguistic approach 217 213, 217–221, 265, 270, 293,
syllable weight 59 transnational 385 326, 350, 377
 Subject index
verb compounding 133 vowel 4, 7–10, 30–35, 37, 39–41, web archive 346
verb stem 40, 136, 183 45, 54, 56, 57, 59, 60, web-based 175, 177, 189,
verb word 34, 136 63–65, 67, 71, 77, 82, 86, 190, 196
verbal suffixes, see suffix 90–104, 107–130, 138, 226 web-delivered 180, 181
de-verbal suffixes, see suffix mid vowel 116, 119 web development 175–197
vigilante justice 419 vowel cluster 9, 116 web resources 175–197
Virginia v. Black 415 vowel devoicing 86 website(s) 179, 188, 190–191,
visible 6, 90, 170, 373, 397, 399, vowel harmony 59, 60 193, 194, 212, 346, 347,
400, 410 vowel inventory 63, 89, 110 403, 404, 405
vocabulary 37, 38, 124, 264, vowel length 53, 94 Westernization 230, 251
382, 427 word final vowel 4, 17 White racism 339, 347,
Vocabulary Insertion 164 358, 359
voice(s) 4, 62, 63, 65, 79, 81, W word list 36, 94, 119
88, 134, 175, 177, 178, 196, Wayne State University ix, x word-final vowels, see vowel
197, 201, 210, 217, 250, 278, Weak Lexicalist Hypothesis, Wyoming 86, 260, 417
284–287, 294, 303, 316, 318, see Lexicalist
322, 330, 332, 339, 340, 347, Web (world wide) 179, 180, Y
350, 366, 372–385, 390–411 185, 186, 187, 190, 197 Yukon 44

The Persistence of Language - Constructing and Confronting The Past and Present in The Voices of Jane H. Hill 2013

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Persistence of Language - Constructing and Confronting The Past and Present in The Voices of Jane H. Hill 2013

Uploaded by

Copyright:

Available Formats

Culture and Language Use

CLU-SAL publishes monographs and edited collections, culturally oriented

John Benjamins Publishing Company

theâ•¯American National Standard for Information Sciences – Permanence

Library of Congress Cataloging-in-Publication Data

© 2013 – John Benjamins B.V.

Introduction. The persistence of language: Constructing and confronting

section 1. Approaches to the study of the indigenous languages

The diachrony of Ute case-marking 3

Language contact as an inhibitor of sound change: An Athabaskan example 29

Stress in Yucatec Maya: Syncretism in loan word incorporation as evidence

The phonetic correlates of Southern Ute stress 85

Revisiting Tohono O’odham high vowels 107

Head-marking inflection and the architecture of grammatical theory:

A case-study in grass roots development of web resources for language

section 2. Approaches to the study of voices and ideologies

Language contact, shift, and endangerment – implications for policy

Spanish in contact with indigenous tongues: Changing the tide in favor

How can a language with 7 million speakers be endangered? 229

A documentary ethnography of a Blackfoot language course: Patterns

Syncretic speech, linguistic ideology, and intertextuality: Presenting

Racism in discourse – analyses of practice

Narrative discriminations in Central California’s indigenous

The voice of (White) reason: Enunciations of difference, authorship,

Double-voicing in the everyday language of Brazilian black activism 365

Uptake (un)limited: The mediatization of register shifting and the

The silken cord: An essay in honor of Jane Hill 415

Afterword: Jane Hill’s current work 425

Language index 431

Subject index 433

Shannon T. Bischoff, Deborah Cole, Amy V. Fountain &

1. Interdisciplinary conversations: Jane’s many friends

The contributions to this volume reflect Jane’s interdisciplinary approach to linguistic

. Perhaps more generally known as ‘Chomskian Linguistics’ or ‘generative linguistics’.

at Arizona – and elsewhere – to “flaunt their ignorance of anything cultural” (Zepeda &

In short, (biopsychological) linguists were not studying the language c­ ommunities

2. The universal and the particular: Jane’s human voices

Language is both a universal characteristic of the human species and a cultural

3. Responsibility in narrative: Jane’s discourses

“I contribute nothing but talk, but boy do I have symbolic capital!”

In their co-authored Introduction to Responsibility and Evidence in Oral Discourse, Hill

Shannon T. Bischoff, Deborah Cole, Amy V. Fountain &

the study of the morpho-syntax of that language to the theory of morpho-­syntactic

2. Section 2 – Approaches to the study of voices and ideologies

2.1 Language contact, shift, and endangerment – implications for policy

2.2 Racism in (mediatized) discourse – analyses of practice

Hill, Jane H. 2001. Proto-Uto-Aztecan: A community of cultivators in central Mexico? American

Approaches to the study of the indigenous

Keywords: Ute; case-marking; diachrony

2. Subject, object and genitive

2.1 Current state

. See in Givón (2000).

(1) nominative oblique gloss

(2) nominative with suffix

But how does one go about substantiating such a hypothesis?

2.2 Traces of the oblique suffix -y/-i

(3) nominative oblique

(6) a. …maay-pʉgay-�u doctor-i �uwa-y…

It is well known that the morphological and phonological treatment of borrowed

2.2.2 Demonstratives and pronouns

(7) nominative oblique

The diachrony of Ute case-marking 3

Language contact as an inhibitor of sound change: An Athabaskan example 29

The phonetic correlates of Southern Ute stress 85

Revisiting Tohono O’odham high vowels 107

How can a language with 7 million speakers be endangered? 229

Double-voicing in the everyday language of Brazilian black activism 365

The silken cord: An essay in honor of Jane Hill 415

Afterword: Jane Hill’s current work 425

Language index 431

Subject index 433

In short, (biopsychological) linguists were not studying the language c ommunities

the study of the morpho-syntax of that language to the theory of morpho-syntactic

2. Subject, object and genitive

2.1 Current state

2.2 Traces of the oblique suffix -y/-i

2.2.2 Demonstratives and pronouns

2.2.3 Nominalized clauses

2.2.4 The ‘epenthetic’ suffix -y

2.3 The oblique suffix -a

2.3.1 The suffix -a as a genitive marker

2.3.2 The suffix -a as object marker

2.4 The object suffix -ku

2.5 Interim summary

2.6 Object and genitive: The hidden footprints of nominalization

2.7.2 The demise of the suffix -y in Ute

2.7.3 The object suffix -ku

3. Post-positions and indirect objects

3.2 Large-size locative post-positions

3.2.2 Verb-derived post-positions

3.2.3 Monosyllabic de-verbal post-positions

3.2.4 The older locative post-positions -na, -mi and -ma

(2) (m) n *ny/ŋ