CHAPTER 14

How fascinating! Insubordinate exclamations1

Marianne Mithun
University of California, Santa Barbara
ILCAA Joint Researcher

Abstract
It has been noticed that in language after language, exclamatives resemble some other sentence type, usually
interrogatives: How cute she is! How cute is she? Explanations for the similarities have generally been
couched in terms of shared abstract syntactic or semantic structure. Here another kind of explanation is
offered: intersecting histories of development through time. Previous work on the issue is surveyed, and the
emerging consensus summarized: exclamative constructions i) tend to be expressive rather than informative,
ii) convey a subjective judgment of the speaker, iii) describe a scalable property, and iv) assert an
unexpectedly high degree of that property. Not all exclamatives resemble questions in form, however. In
Mohawk, an Iroquoian language of Northeastern North America, exclamatives show these four
characteristics, but they resemble complements of declarative sentences. Exclamatives that resemble
interrogatives and those that resemble declarative complements can be seen to originate in different source
constructions, but they converge in the final steps of their development: a matrix clause expressing surprise
or an unexpectedly high degree of some property expressed in the complement disappears (insubordination),
but its meaning remains a part of the construction.

In a number of languages, exclamative constructions share some of their form with one or more
other constructions in the same language. Elliott (1974) cites counterparts from English, French,
Rumanian, German, Russian, Turkish, Mandarin, and literary Japanese. Alongside the German Wie
angenehm es ist! (‘How nice it is!’), for example, there is Wie machst du das nur? (‘How do you do
that?’). Russian has both Kak ona krasiva! (‘How beautiful she is!’) and Kak vy zhiv’ote? (‘How
are you?’) (1974:244-5). French contains such pairs as Comme vous avez de grandes oreilles!
(‘What big ears you have!’) and Fais-le comme tu veux (‘Do it however you want’), as well as
Qu’il est beau! (‘How handsome he is!’) and Je vois qu’il est beau (‘I see that he is handsome’). A
number of works have examined relationships between exclamative constructions, or exclamatives,
and other sentence types. These have generally focused on underlying synchronic syntactic and/or
semantic structures.
Explanations can also come from another approach: examining the processes by which the
constructions develop over time. Similarities across constructions are often due to intersecting
histories: both may have developed from the same source, one may have developed out of the other,
or they may have both been subject to some general development. In this chapter, modern work on
exclamatives, most from European languages, is first surveyed. Most of this work has focused on
similarities between exclamatives and interrogatives, explaining them in terms of abstract
underlying structure. Many exclamative constructions, however, show no similarities to questions,
but resemble declaratives. It will be shown that though the two types arise from different sources,
their developments through time converge in a final step. This step is the process termed by Evans
‘insubordination’ (2007), whereby an erstwhile matrix clause is omitted, and its subordinate
complement comes to be used as an independent sentence with conventionalized functions
attributable to the original matrix.

1
Thanks to Alex Bergs, Nick Evans, Andrej Kibrik, Heiko Narrog, Nigel Vincent, and Ellie Van Gelderen for useful
discussion of this topic.
2. Marianne Mithun

1 Previous work

Much work on exclamatives has taken as a point of departure the identification of sentence types.
Sadock and Zwicky define a sentence type as ‘a coincidence of grammatical structure and
conventional conversational use’ (1985: 155). They cite as major types declaratives, interrogatives,
and imperatives, and as a minor type exclamatives. (Exclamatives are sometimes classified as a
fourth major type, as in Quirk et al. 1985 and Biber et al. 1999.) Sadock and Zwicky note
similarities between exclamatives and interrogatives.

Since exclamations are, like interrogatives, non-assertive, exclamatory sentences often

resemble interrogative sentences in form, as in the English exclamatory type with how
and what a:

How tacky that is!

What a good syntactician she is! (Sadock and Zwicky 1985: 162)

They also note similarities between exclamatives and declaratives.

The function of exclamatory sentences is much like that of declarative sentences except
that exclamations are intended to be expressive whereas declaratives are intended to be
informative. Both represent a proposition as being true, but in an exclamation, the
speaker emphasizes his strong emotional reaction to what he takes to be a fact, whereas
in a declarative, the speaker emphasizes his intellectual appraisal that the proposition is
true. Because of this close relationship exclamatory sentences are often similar in form
to declarative sentences, as in the English exclamatory type with so and such a:

That’s so tacky!
She’s such a good syntactician! (Sadock and Zwicky 1985: 162)

Most work on exclamatives has focused on these parallelisms, especially that between exclamatives
and interrogatives.

1.1 Elliott

Early important work by Elliott (1971, 1974) identified syntactic and semantic similarities between
exclamatives and questions. Both can occur as sentences on their own: How attractive is she? How
attractive she is! (These are also called simple, root, matrix, direct, or absolute constructions). Both
can also occur as complements of higher verbs: I wonder [how attractive she is]; I know [how
attractive she is]. Both can contain a wh- word (how, what a). Elliott also noted differences. While
English questions can contain various wh- words (who, where, when, which), simple English
exclamatives can contain only how or what. Where simple interrogatives show subject inversion
(How attractive is she?), exclamatives do not (How attractive she is!). (Note, however, the more
recent How cool is that!) Elliott focused on exclamatives describing scalable properties (excluding
exclamations such as Hooray!). He noted that the word how pertains to either degree or manner in
interrogatives (How attractive is she? How did he do that?), but it pertains only to degree in
exclamatives (How attractive she is!).
Elliott also highlighted two special characteristics of exclamatives. Kiparsky and Kiparsky
(1970) had observed that certain English verbs, such as know and realize, presuppose the truth of
their complements; the speaker assumes the complement to be true. These verbs they termed
‘factive’. Elliott remarked that exclamatives can be embedded under factive verbs: (I know [how
 Chapter 14. Fascinating exclamations
3

attractive she is]. Kiparsky and Kiparsky also identified a group of predicates they termed
‘emotive’, such as fascinating, amazing, fantastic, incredible, surprising, astonishing, significant,
important, sad, alarming, awful, terrible, tragic, etc., predicates ‘which express the subjective value
of a proposition rather than knowledge about it or its truth value’ (169). Elliott pointed out that
exclamatives also occur with emotive matrix predicates (It is amazing [how attractive she is]), but
not with negated emotive predicates (*It isn’t amazing [how beautiful this place is]), unless one is
contradicting an immediately preceding exclamation. He linked this pattern to the function of
exclamatives: ‘to talk about abnormal or unexpected situations’ (1974: 242).
Earlier Austin (1962) had identified a category of sentences he termed ‘performative’. Such
sentences overtly express illocutionary force: I promise you that I won’t squeal. They usually have
first person subjects, they often have second person objects, they are affirmative, and they are in the
present tense. The matrix verb is one of a class which also includes advise, answer, appoint, ask,
authorize, beg, bequeath, beseech, caution, cede, claim, command, condemn, counsel, dare,
declare, demand, empower, enquire, entreat, excommunicate, grant, implore, inform, instruct,
offer, order, pledge, pronounce, propose, request, require, say, sentence, vow, and warn. Around
the time Elliott was writing, it was proposed that the illocutionary forces of different sentence types
could be captured in a generative model of syntax by postulating that the sentences are embedded
under various higher performative verbs in deep structure (Ross 1970; Langacker ms; Sadock
1970). Thus the deep structure of the command Go!, for example, would be something like I order
you to go. The matrix clause (I order you) is eliminated by a ‘performative deletion’ transformation
in the course of the syntactic derivation. Elliott noted in passing that a performative treatment of
exclamatives might be possible.

It is reasonable to suppose that the notion of performatives, provided there is sufficient

independent evidence for it to motivate its inclusion in the theory of language, might be
extended to exclamations, and that their unique properties might be accounted for in the
grammar by an abstract performative verb, with no phonological realization and the
semantic content common to all exclamatory predicates, including amazing, terrible,
unbelievable, incredible, etc. (Elliott 1974: 231–232)

1.2 Grimshaw

In her work on predicate-complement combinations, Grimshaw (1977, 1979) compared sentences

containing embedded exclamatives (It’s amazing [how tall John is]), with those containing
embedded questions (Fred will ask [how tall John is]). She proposed that the constructions are
identical syntactically, but distinct semantically. She accounted for their syntactic similarity by the
fact that both occur with matrix predicates subcategorized for occurrence with a sentential
complement [ __ S]. Both are derived by a process of Wh Fronting within the complement. She
attributed their semantic difference to their occurrence with matrix predicates with different
semantic frames. Some predicates, such as be amazing and be surprised at, occur only with
embedded exclamatives [ __ E]; others, like ask and wonder, occur only with embedded questions
[ __ Q]; still others, such as know and find out, can occur with either [ __ E,Q]. Grimshaw echoed
Elliott in noting that sentences with embedded exclamatives, such as It’s amazing [how tall John
is], presuppose that John is extremely tall. Sentences with embedded interrogatives, like Fred will
ask [how tall John is], carry no such presupposition. Simple exclamatives differ from simple
questions in the same way: How tall John is! versus How tall is John? Exclamatives assert that his
tallness is unusual, while questions do not.
4. Marianne Mithun

1.3 Michaelis and Lambrecht

Michaelis and Lambrecht (1996a, 1996b) and Michaelis (2001) also address the issue of whether a
distinct exclamative sentence type can be identified for English. They cite characteristic properties
noted by previous authors.

Let us define exclamative utterances as those in which the speaker presupposes that
some entity is located at some point on a property scale, and asserts that the entity
manifests a remarkably high degree of this property. The affective stance involved is
generally expressed by a subordinating main predicator. These predicators evoke
expectation contravention; they contain AP complements like astonishing, incredible,
and unbelievable. (Michaelis and Lambrecht 1996b: 228).

They further point out that exclamatives are distinctive in their deictic anchoring.

Exclamatives involve both personal and temporal deixis. The notion of affective stance
entails the presence of someone making a judgment, and the speaker is the judge by
default. The speaker’s status as judge may be explicit (I can’t believe how much he’s
GROWN!) or implicit (It’s amazing how fast the WEATHER can change. (Michaelis and
Lambrecht 1996b: 239)

Thus a sentence like My mom can’t BELIEVE how much he’s GROWN! would not count for Michaelis
and Lambrecht as an exclamative, because it does not encode the speaker’s judgment.
Michaelis and Lambrecht model similarities among different sentence types, and among
constructions that function as exclamations, in a Construction Grammar framework. An inheritance
network specifies shared formal and semantic properties. The network involves two kinds of links:
subsumption links, which represent the formal and semantic overlap between two constructions, and
instance links, in which an abstract construction is linked to a more fully specified version.
Similarities between certain exclamatives and questions are explained in terms of a subsumption
link. The model is illustrated with the Indirect Exclamative construction.

(1) Indirect Exclamative construction (Michaelis and Lambrecht 1996b: 242):

i. I’m AMAZED at how much I SPENT.

The Indirect Exclamative construction is linked to the Bare Complement Question

construction by a subsumption link. . . . The Bare Complement Question
construction licenses indirect questions, as in ii.

ii. I wonder how much I SPENT.

Indirect questions, like direct questions, can be represented as open propositions: I

spent X amount, they hired X. The open proposition, minus the unbound variable,
represents the presupposed material. In ii, for example, the speaker presupposes that
she spent some amount.

The Indirect Exclamative construction is also linked to all other exclamative types through an
instance link: ‘Each of the exclamative subtypes inherits its semantic and pragmatic properties from
an abstract superconstruction: the ABSTRACT EXCLAMATIVE CONSTRUCTION (AEC)’ (1996b: 237).

(2) Semantico-pragmatic properties of the ABSTRACT EXCLAMATIVE CONSTRUCTION:

 Chapter 14. Fascinating exclamations
5

(Michaelis and Lambrecht 1996a: 378, 1996b: 239)

a. Presupposed open proposition

b. Scalar extent
c. Assertion of affective stance: Expectation contravention
d. Identifiability of described referent
e. Deixis

1.4 Continuing work

The late 20th and early 21st centuries have seen increased attention to exclamative constructions,
particularly in Germanic and Romance languages. The same characteristics continue to be
observed: they express surprise at the high degree of some scalar property, emotional content, and
factivity. The focus of most of this work as well has been on the relationships of exclamatives to
other sentence types, particularly content questions, but also to other constructions, including
isolated subordinate clauses. German, for example, contains both exclamatives with interrogative
words like Wie müde bin ich heute! ‘How tired I am today’ and those resembling dependent
clauses, like, Daß ich heute so müde bin! literally ‘That I am so tired today!’ (Näf 1996: 137).
Scandinavian languages contain exclamatives based on degree words like Nynorsk kor ‘how’, on
deictic degree forms like så ‘so’ and on complementizers like som ‘that’ (Abels and Vangsnes (eds.)
2010). Catalan contains exclamatives like Que car que és! ‘How expensive it is!’ alongside
exclamatives like Que n’es, de car!, introduced by the complementizer que without a matrix
(Villalba 2003: 714).
Some authors (Lahiri 1991; Rosengren 1992, 1997; Huddleston 1993; d’Avis 2001, 2002;
König and Siemund 2007; and others) have concluded that exclamatives do not constitute a distinct
sentence type, but simply function as exclamations under certain lexical, syntactic, and/or prosodic
circumstances. Others have proposed explanations of their differences in terms of features in formal
models of syntactic or semantic structure. Interest in the relationship of exclamatives to other
constructions has been stimulated by the cartographic approach to the Left Periphery of Rizzi
(1997) in which clause typing is built into syntactic structure. Summaries of the issues are in
Villalba (2008) and Abels and Vangsnes (2010).
Much recent work posits a special syntactic or semantic operator in the Left Periphery.
Gutiérrez-Rexach (1996, 2001, 2008) proposes that exclamatives have the same basic underlying
structure as interrogatives and the same denotations, but are distinguished by the presence of an
illocutionary exclamative force operator EXC, an operator on propositions. This operator, adjoined
to CP, triggers different syntactic operations in the course of the syntactic derivation. Zanuttini and
Portner (2003) counter that a special force element is unnecessary in the syntactic representation of
exclamatives. They argue that exclamatives and interrogatives are similar in that both denote a set
of alternative propositions. Questions like How cute is he? involve a choice among alternatives in
their possible answers: extremely, sort of, not very, not at all, etc. Exclamatives denote a degree that
is greater than the alternatives under consideration. They attribute this shared property of a choice
among alternatives to a common syntactic WH operator in Spec, CP in their underlying syntactic
structure. They ascribe the factivity of exclamatives to an abstract FACT morpheme, located in the
CP domain of exclamatives, in an extra layer of CP structure. The question How cute is he? is not
factive: there is no presupposition that he is cute. Accordingly, there is no FACT morpheme in the
syntactic structure of interrogatives, nor is there the extra layer of CP structure (2003:59).
Rett distinguishes exclamatives with an illocutionary force operator: ‘E-FORCE, like other
illocutionary force operators, is a function from propositions to expressive speech acts’ (2011: 430).
6. Marianne Mithun

Unlike most previous authors, Rett questions the factivity of exclamatives. She maintains that the
content of exclamatives may be new to the hearer.

Imagine that John goes to Crete and writes Mary to let her know how his visit is going.
It seems perfectly fine for him to begin his letter, What a magnificent place Crete is!,
thus informing Mary that Crete is magnificent to an unexpected degree, whether or not
Mary would have guessed as much before hearing from John. (Rett 2011: 435)

Quantitative work by Potts and Schwarz (2008) shows robust links between exclamatives and
intensity of affect. Working from a corpus of about 100,000 online product reviews, they find
strong correlations between exclamatives and both very positive (4-5 star ratings) and very negative
(1-star ratings).
Additional discussion of exclamatives can be found in the papers in Abels and Vangsnes
(2010) on Scandinavian languages; Postma (1995), Bennis (1998), Bennis et al. (1998), and den
Dikken (2006) on Dutch; Gérard-Naef (1980), Marandin (2008), and Aslanov (2009) on French;
Bosque (1984) and Villalba and Bartra-Kaufman (2010) on Spanish; Villalba( 2001, 2002, 2004),
Castroviejo Miró (2006, 2008), González-Rodriguez (2008), and Mayol (2008) on Catalan; and
Radford (1982), Benincà (1995, 1996), and Zanuttini and Portner (2000a, 2000b) on Italian.

1.5 Summary

Work on exclamatives, most of it directed at the issue of whether they constitute a distinct sentence
type, has revealed certain core characteristics of exclamative constructions.
Functionally, exclamatives are characterized as expressive rather than informative, conveying a
subjective judgment on the part of the speaker. They describe a scalable property. They are
subjective: they represent the judgment of the speaker at the time of the utterance. They assert that
the property holds to a remarkably high degree, beyond expectations; they thus contain an element
of surprise and can be embedded under emotive predicates.

(3) Observed characteristics of exclamatives

i. Expressive rather than informative
ii. Convey subjective judgment of the speaker
iii. Describe a scalable property
iv. Assert an unexpectedly high degree of that property

Many authors specify that they presuppose a proposition (factivity) on the part of the speaker and
can therefore be embedded as complements of factive matrix clauses, but Rett provides examples of
exclamatives that assert the basic proposition.
Most proposed analyses have accounted for similarities and differences between exclamatives
and other sentence types in terms of underlying syntactic or semantic structure. Many of these
approaches are quite similar, though the analyses are couched in different formal frameworks. The
most frequently-cited feature shared by exclamatives and interrogatives is an element with a
meaning like ‘how’ or ‘what’. In many cases the two sentence types are assumed to have the same
basic syntactic structure, with just one difference, usually an additional, higher layer of structure.
For Elliott this might be a higher performative verb. For Gutiérrez-Rexach it is an illocutionary
force operator EXC on propositions, adjoined to CP, which triggers different syntactic operations in
the course of the syntactic derivation. For Zanuttini and Portner it is an additional higher layer of
CP structure, though the feature distinguishing exclamatives is an abstract FACT morpheme,
responsible for the factive property of exclamatives, rather than a specific force feature. For Rett it
is a special illocutionary force operator E-Force on exclamatives.
 Chapter 14. Fascinating exclamations
7

2 Exclamatives and declaratives

Not all exclamatives resemble questions. Villalba (2008) takes the existence of exclamatives
without a word meaning ‘how’ or ‘what’ as evidence against the accounts proposed in the last
section. Consider, for example, the French Qu’il est beau! (‘That he is handsome!’), German Daß
ich heute so müde bin! (‘That I am so tired today!’), Norwegian Som de kranglar! (‘That you
quarrel!’), and Catalan Que n’es, de car! (‘That it is expensive!’). Similar exclamative
constructions can be seen in Mohawk, an Iroquoian language of northeastern North America. (All
examples here are from unscripted speech.)

(4) Basic Mohawk exclamative: [Billy Two Rivers, speaker]

Á:ke tsi nikanó:ron.
á:ke tsi ni-ka-nor-on
gee how PRT-N-dear-ST
‘Gee how expensive it is!’

This construction consists of the particle tsi ‘how, so’ and a following verb, often but not always
prefixed with the partitive ni-/na’-/n-. The Mohawk exclamatives display the characteristics noted
for European exclamatives by previous authors, as set out in (3) above.
In Mohawk, property concepts are expressed by stative verbs, like - nor- ‘be expensive, dear,
precious’ above. The same construction occurs with verbs denoting events to indicate an unusually
high degree of intensity.

(5) Exclamative with action verb: [Vina Loft, speaker]

Á:ke kia’ tsi nihaki’tanonhwarià:kon.
á:ke kia’ tsi ni-hak-i’t-anonhwar-ia’k-on
gee wow how PRT-M.SG>1SG-excrement-brain-hit-ST
‘Gee how he whipped me!’
But this Mohawk particle tsi, glossed in (5) as ‘how’, does not occur in questions. Degree questions
are built on the particle tó:.

(6) Degree/quantity question: [Leatrice Beauvais, speaker]

Tó: nikanó:ron?
to: ni-ka-nor-on
how PRT-N-be.expensive-ST
‘How expensive is it?’ = ‘How much does it cost?’

‘How’ questions about manner are built on the particle óh.

(7) Manner question: [Skawén:nati Montour, speaker]

Óh niiawèn:’en tsi rawenhé:ion?
oh ni-iaw-en’-en tsi raw-enhei-on
how PRT-N.PAT-happen-ST how M.SG.PAT-die-ST
‘How did he die?’ (‘How did it happen that he died?’)

Neither tó: nor óh appears in exclamatives.

An account of Mohawk exclamatives in terms of structures shared with interrogatives thus
holds little appeal. But like the French, German, Norwegian, and Catalan exclamatives seen above,
Mohawk exclamatives resemble another construction: the complements of declaratives.
8. Marianne Mithun

(8) Complement: [Karihwénhawe Dorothy Lazore, speaker]

Ion’wé:sen tsi shé:kon tetsiá:ron kiahrónkha’.
i-on’wesen tsi shekon tetsiaron ti-ahronk-ha’
N-be.pleasant how still we both 1INCL.DU.AGT-speak.a.language-HAB
‘It’s nice [that both of us still speak [Mohawk]]’.

To explain this recurring parallelism, we could adopt the strategy seen in the previous section,
searching for an abstract underlying structure shared by constructions built on tsi, then adding some
feature unique to exclamatives. The fact that similar parallelisms recur across multiple languages,
even languages that are not genetically or areally related, suggests that there may be some general
principles in play. But these principles may not all be synchronic. Shared features among
constructions may be the result of intersecting histories. Recurring similarities may point to general
processes of grammatical development. Since there is no ancient record of Mohawk connected
speech, it is not possible to trace the histories of constructions philologically. But there is another
tool: internal reconstruction. As we learn more about the recurring pathways by which constructions
develop cross-linguistically, we can compare related constructions within a single language and
make hypotheses about the processes by which they evolved. The following sections trace likely
steps in the evolution of the Mohawk exclamative.

2.1 Basic Mohawk morphological structure

Mohawk contains just three lexical categories identifiable in terms of their internal morphological
structure: verbs, nouns, and particles. Verbs consist minimally of a pronominal prefix identifying
the core arguments of the clause, and a verb stem. They can be complex morphologically, with up
to five prepronominal prefixes, a complex stem potentially containing a reflexive or middle prefix,
an incorporated noun, and one or more derivational suffixes, followed by one or two inflectional
aspect suffixes. The internal structure of the verb in (5) is repeated below.

(9) Verb: [Vina Loft, speaker]

Nihaki’tanonhwarià:kon.
ni-hak-i’t-anonhwar-ia’k-on
PRT-M.SG/1SG-excrement-brain-hit-ST
‘He really beat my brains out.’

Morphological nouns show simpler internal structure, with a prefix indicating the gender of the
referent or the gender and number of a possessor, followed by a noun stem, and a noun suffix.

(10) Nouns
ohwísta’ raohwísta’
o-hwist-a’ rao-hwist-a’
N-money-NS M.SG.AL.POSS-money-NS
‘money’ his money’

Morphological particles by definition have no internal structure.

(11) Particles
iáh NEGATIVE ken INTERROGATIVE
á:ke ‘gee’ tó:kens ‘certainly’
 Chapter 14. Fascinating exclamations
9

Lexical categories defined by morphological structure do not always coincide with those
defined by syntactic function. Verbs can function as predicates, as would be expected. But they can
also function as clauses or sentences on their own, and as referring expressions with no explicit
marking. They are often lexicalized as such.

(12) Morphological verb, lexicalized nominal

kawistóhtha’
ka-wisto-ht-ha’
N.AGT-be.cold-CAUS-HAB
‘it chills’ = ‘refrigerator’

2.2 Place

Neither the comparative method nor internal reconstruction provides evidence of a lexical source
for the particle tsi. Perhaps its most concrete use is as a locative nominalizer, which serves to form
terms for places from clauses. The term for ‘school’ does not itself specify a grammatical role
within the clause.

(13) Place nominalization: [Josie Day, speaker]

Wats kí:ken wakkaratóntie’ ne
wats kiken wak-kar-at-ontie’ ne
wait this 1SG.PAT-story-stand-PROG aforementioned

tsi ionterihwaienstáhkhwa’.
tsi ie-ate-rihw-a-ien-st-ahkw-ha’
LOC.NMLZ INDEF.AGT-MID-word-LK-lay.down-CAUS-INST.APPL-HAB
at one causes to lay down one’s words with it = one teaches to read with it

‘Wait a minute, I’m discussing the school.’

The fact that an expression refers to a place can be made more explicit with the addition of the
noun nón:we ‘place’, often shortened to nón:, as in E’thó tsi nón: nikonnitskwahronnion’. ‘That is
the place where they would perch’. References to places can also be made more precise with such
phrases as tsi niió:re’ ‘up to, until’, containing the verb niió:re’ ‘it is so far’: … kwáh nek tsi
niió:re’ iotenénhstate’ ‘[We removed the husk] just up to the base of the cob’.

2.3 Time

A well-known process of grammatical development is the extension of place expressions to time.

The distinction between place and time is not crisp in many utterances, providing bridging contexts
for the space > time development. The reference to the wedding below could be to the place of the
wedding, the time of the wedding, or even the event. (Material providing context is given with just
free translations.)

(14) Place/time nominalization: [Josephine Horne, speaker]

Khe:re’ kati’ ken nè:’e aa,
‘I suppose then, ah,

iontatena’konnì:re’ kwi’ ken thí:ken ne:

she is actually going to show the
10. Marianne Mithun

nahò:ten’ rotiráhston tsi iakoniákon.

naho’ten roti-rahst-on tsi iako-niak-on
what M.PL.PAT-draw-ST LOC/T.NMLZ INDEF.PAT-marry-ST

pictures they took at her wedding.’

But many tsi nominalizations now unambiguously refer to time.

(15) Simultaneity: Kaia’titáhkhe’ Jacobs, speaker

Í: kwi' tehsekkà:nere' tsi né: sá:ton
i’i ki’ wahe’ te-hsek-kahner-e’ tsi ne s-aton
1 in.fact.you.know DV-2SG/1SG-look.at-ST T.NMLZ that 2SG.AGT-say

‘Well you’re looking at me while you’re saying that.’

Some tsi time expressions are now lexicalized, such as tsi wa’ó:rhen’ne’ ‘when it dawned’ = ‘the
next day’.
Reference to time can be made more specific with the particle ó:nen ‘the time, now, then’, as
in Saionawì:rehte’ tsi (ó:nen) wa’thotihèn:rehte’ ‘It slithered back into its hole at (the time when)
they yelled.’ The phrase tsi niió:re’ ‘as far as, up to, until’ seen above with reference to place
occurs even more frequently with reference to time. Some uses could be interpreted either spatially
or temporally: Ià:ia’k na’kahwistà:’eke’ tiio’kehà:keionkwahonwì:sere’ tsi nii:ió:re’ saiákwawe’
Tiohtià:ke ‘We rode on the train for six hours until we arrived back in Montreal.’ Others,
however, are unambiguously temporal: Karì:wes tsi niió:re’ ientewanénhstake’ ‘It will be a long
time before we eat corn.’

2.4 Manner

The particle tsi now also appears in manner constructions. This could have come about via an
extension of its coincidence meaning: two situations coinciding in space and time. Possible bridging
contexts might have involved structures like that in (16). The men performed two actions at the
same place and time: ‘they formed a circle while/how/as they were sitting’.

(16) Simultaneity > Manner: [Sonny Edwards, speaker]

Wahatikahkwèn:taron tsi wahónkien.
wa-hati-kwahkwen’ta-r-on tsi wa-hon-at-ien-’
FACT-M.PL.AGT-wheel-be.in-DIST as FACT-M.PL.AGT-MID-set-PFV
‘They sat around in a circle.’

But many tsi constructions now unambiguously indicate manner, as in (17).

(17) Manner subordinate clause: [Wilfred Jaimison, speaker]

Tokat ioiánere’ tsi tesatháhahk tsi
tokat io-ianer-e’ tsi te-s-at-hah-ahk tsi
if N.PAT-be.good-ST how DV-2SG.AGT-MID-path-pick.up LOC.NMLZ

ionhwentsiá:te’
i-onhwentsi-at-e’
 Chapter 14. Fascinating exclamations
11

N.PAT-earth-extend-ST

skén:nen’k tsi sanonhtónnion.

skennen’-ok tsi s-anonhton-nion
peaceful-only how 2SG.AGT-think-DIST

‘If the way you walk on earth is good, the way you think will be peaceful.’

The reference to manner can be made more specific. A verb based on the root -o’ten ‘be a
way, be a kind of’ can be added: tsi niiakwawennò:ten’ ‘how so.our.language.is.a.kind.of’ = ‘the
way we speak’= ‘how we speak’. A verb based on the root -ht ‘be so, be such’, along with the
particle tsi can have a similar effect. The phrase tsi ní:ioht, often shortened to tsi ní:, is literally ‘as
so it is’: tsi ní: tsi rótston ‘as so.it.is as he.is.dressed’ = ‘the way he was dressed’ = ‘how he was
dressed’.

2.5 Degree

The feature of coincidence also appears in tsi nominalizations used in comparisons, where two
manners or qualities coincide in degree.

(18) Coinciding degree: Sha’tekenhátie’ [Marian Phillips, speaker]

Thó na’teiotskenné:nonte’
tho na’-te-io-atskennenont-e’
there PRT-DV-ZOIC.PAT-be.pale-ST

tsi ní:ioht ne:.. ki: kahiatónhsera’.

tsi ni-io-ht ne kiken ka-hiaton-hser-a’
as PRT-N.PAT-be.so the this N-write-NMLZ-NS

‘She’s as pale as this paper.’

2.6 Event nominalizations

The particle tsi that marks clauses as referring to places, times, manners, and degrees also
nominalizes clauses referring to situations: events and states.

(19) Complement of immediate perception: [Margaret Lazore, speaker]

Waháttoke’ ki’ tsi rotihnekatárion.
wa-ha-attok-e’ ki’ tsi roti-hnek-atar-ion
FACT-M.SG.AGT -notice-PFV in.fact how M.PL.PAT-liquid-be.in-DIST

‘He noticed [how they had been drinking].’

It is not uncommon cross-linguistically for a manner word meaning ‘how’ to serve as a

complementizer with matrix verbs of immediate perception. Popova (2011a, 2011b), working with
the Russian National Corpus and citing earlier work in Arutjunova (1988), Kobozeva (1988),
Padučeva (1986, 2004), and Kobozeva and Popova (2013), shows that the Russian counterpart kak
‘how’, used in questions about manner and in manner nominalizations, also serves as a
complementizer of matrix verbs of direct perception, such as videt’ ‘see’, smotret’ ‘watch/look at’,
12. Marianne Mithun

and slušat’ ‘listen to’. Russian verbs of dreaming, such as snit’sja also occur with kak complements.
Mohawk dreaming verbs also occur with tsi complements.

(20) Complement of ‘dream’: [Raymond Miller, speaker]

Ne:’ wakatetshen:
ne:’ wak-atetshen
it.is 1SG.PAT-dream

tsi kató:rats ka’ok nón: karhakónhshon.

tsi k-atorat-s ka’ok nonwe ka-rha-kon-hshon
how 1SG.AGT-hunt-HAB some place N-forest-interior-DIST

‘I dreamt I was hunting someplace through the bush.’

Popova argues that the equivalent Russian verbs involve a kind of inner vision. Russian kak also
occurs with pomnit’ ‘remember’ and vspominat’ ‘recollect’, which could be understood similarly as
a kind of inner vision. Mohawk tsi serves as a complementizer with such verbs.

(21) Complement of ‘remember: [Jim Maracle, speaker]

Nok kè:iahre’
ne ok k-ehiahr-e’
the and 1SG.AGT-remember-ST

tsi rake’nihkén,
tsi rake-’nih=kenha’
how M.SG/1SG-be.father.to=DEC

wahatkáhtho’ ken’ nikontarò:ten …

wa-ha-atkahtho-’ ken’ ni-ka-ront-o’ten
FACT-M.SG.AGT -see-PFV just PRT-N-log-be.a.kind.of

‘And I remember how my late father saw such a log …’

The many Mohawk tsi complements that could be translated into English with either ‘how’ or ‘that’
would provide bridging contexts, by which the manner nominalizer tsi could be extended to other
kinds of complements: ‘I remember how/that my father saw such a log’.
Popova notes that Russian kak also serves as a complementizer with matrix predicates of
knowledge and acquisition of knowledge, a development shared by Mohawk tsi.

(22) Complement of ‘find out’: [Josie Day, speaker]

Thó ki’ non: onkhtó:kenhse’
tho ki’ nonwa wa’-wak-token-hs-e’
there in.fact place FACT-1SG.PAT-be.certain-BEN.APPL-PFV

tsi iáh ki’ tehanákere’.

tsi iah ki’ te-ha-naker-e’
how not in.fact NEG-M.SG.AGT -reside-ST

‘That’s actually when I found out how/that he [Santa Claus] wasn’t real.’
 Chapter 14. Fascinating exclamations
13

(23) Complement of ‘know’: [Billy Two Rivers, speaker]

Wakaterièn:tare’ ni: tsi ó:nen wakekstèn:ha.
wak-ate-rien’tar-e’ ne i: tsi onen wake-ksten=ha
1SG.PAT-MID-know-ST the 1 how now 1SG.PAT-be.old=DIM

‘I know that I’m old.’

Mohawk tsi has now been extended to complements of some more, though not all, of the
matrix types described by Noonan (2007). These include utterance predicates such as -ihron ‘say’, -
hrori ‘tell’, -hiaton ‘write’, - henreht ‘brag, shout’, -rihwenht ‘condemn’; predicates of propositional
attitude such as -anonhtonnion ‘think’; of emotional reaction such as -atshennonni ‘be happy’,
- nehrako ‘be surprised’, -atera’swiio ‘be lucky’; of judgment such as -ianer- ‘be good’, -iio ‘be
nice’, -ahskats ‘be pleasant’; and of fearing such as -tshahni ‘be afraid’.

(24) Complement of ‘be nice’: [Karihwénhawe Dorothy Lazore, speaker]

Ion’wé:sen tsi shé:kon tetsiá:ron kiahrónkha’.
i-on’wesen tsi shekon tetsiaron ti-ahronk-ha’
N-be.pleasant how still both 1INCL.DU.AGT-speak.a.language-HAB

‘It’s nice [how both us still speak [Mohawk]]’.

(25) Complement of ‘be lucky’: [Watshenní:ne Sawyer, speaker]

Shé: ionkwatera’swí:io
shekon ionkw-ate-ra’sw-iio
still 1PL.PAT-MID-luck-be.good

tsi shé:kon ionkwá:ien’ nonkwehonwehnéha’ aetewatá:ti’.

tsi shekon ionkwa-ien-’ ne=onkwe=honwe=hneha’ aa-etew-atati-’
how still 1PL.PAT-have-ST the=person=real=style OPT-1INCL.PL-talk-PFV

‘Still, we’re lucky [that we have the Indian language to speak].’

Popova observes that Russian kak complements are assumed by the speaker to be true. This is
largely true of the Mohawk complements set off by tsi seen above. The various speakers cited
above, for example, were presupposing that the men were drunk, that the father saw such a log, that
Santa Claus is not real, that the speaker was old, and that the two friends speak Mohawk.

(26) Factive complements

a. Waháttoke’ ki’ tsi rotihnekatárion.
‘He noticed [how they had been drinking].’

b. Nok kè:iahre’ tsi rake’nihkén, wahatkáhtho’ ken’ nikontarò:ten.…

‘And I remember [how my late father saw such a log] …’

c. Thó ki’ non: onkhtó:kenhse’ tsi iáh ki’ tehanákere’.

‘That’s when I found out [how he [Santa Claus] wasn’t real].’

d. Wakaterièn:tare’ ni: tsi ó:nen wakekstèn:ha.

14. Marianne Mithun

‘I know [that I’m old].’

e. Ion’wé:sen tsi shé:kon tetsiá:ron kiahrónkha’.

‘It’s nice [how both us still speak [Mohawk]]’.

f. Shé: ionkwatera’swí:io tsi shé:kon ionkwá:ien’ nonkwehonwehnéha’ aetewatá:ti’.

‘Still, we’re lucky [that we have the Indian language to speak].’

Generally, when the complement is not assumed to be true, no tsi appears.

(27) Complement of thinking: [Joe Deer, speaker]

Ì:rehre’ kati’ken’ óksa’k ken
i-hr-ehr-e’ kati’ken oksa’=ok ken
PROTHETIC-M.SG.AGT-think-ST supposedly immediately=just Q

enthonwaiaróthahse’ ne ohwísta’
en-t-honwa-iar-ot-hahs-e’ ne o-hwist-a’
FUT-CISLOC-M.PL/M.SG -bag-stand-BEN.APPL-PFV the N-money-NS

‘Did he think [that right away they would give him a bag of money]?’

Matrix verbs indicating modality occur without a complementizer. They do not presuppose
their complements.

(28) Complement of ability: [Billy Two Rivers, speaker]

Hátskwi enhskwé:ni á:re' thé:nen' wahsì:ron ken?
hatskwi en-hs-kweni-’ are’ thenen’ wa-hs-ihron-’ ken
come.on FUT-2SG.AGT-be.able-PFV again something FACT-2SG.AGT-say-PFV TAG

‘Come on now, you can manage [to say something different], can’t you?’

(29) Complement of necessity: [Kaia’titáhkhe’ Jacobs, speaker]

Kwah entà:’on ensehsenóhare’.
kwah enta’on en-se-hse-nohar-e’
even necessary FUT-REP-2SG.AGT-wash-PFV

‘You even have [to wash it again].’

Tsi complements containing scalar verbs can also indicate degree.

(30) High degree: [Sonny Edwards, speaker]

Wahshakohró:ri’ tsi nahotera’swí:iohste’,…
wa-hshako-hrori-’ tsi n-a-ho-ate-ra’sw-iio-hst-e’
FACT=M.SG/M.PL -tell-PFV how PRT-FACT-M.SG.PAT-MID-luck-be.good-CAUS-PFV

‘He told them how lucky he was.’

Such sentences are ambiguous between degree and event interpretations: ‘He told them how lucky
he was’ presupposing that he was lucky, or simply ‘He told them that he was lucky.’ (The English
counterparts show similar ambiguity.) The degree interpretation can be emphasized in Mohawk
with the phrase tsi niió:re’ ‘so it is far’ seen earlier with locative and temporal nominalizations.
 Chapter 14. Fascinating exclamations
15

(31) Degree with tsi niió:re’: [Cecelia Peters, speaker]

Né: ki’k wa’kehià:ra’ne’ kí:ken
ne: ki’=k wa’-k-ehiahr-a’n-e’ kiken
that in.fact=just FACT-1SG.AGT-remember-INCH-PFV this

‘In fact I just remembered

tsi niió:re’ tsi kiótkon ionkwahskéhnhen …

tsi ni-io-r-e’ tsi tiotkon ionkw-ahskenh-en
how PRT-N.PAT-be.far-ST how always 1PL.PAT-fight.for-ST

how very hard we are always fighting [to speak our language].’

2.7 The evolution of the Mohawk exclamatives

Viewed in the light of their development over time, the similarity between the exclamative and
complement constructions makes sense. If this hypothesized series of events is correct, the
similarities among the various Mohawk tsi expressions can be explained in terms of diachronic
layers, rather than a single synchronic abstraction. The tsi constructions that currently coexist within
Mohawk are related by a series of changes over time, each of which has been observed to occur in
many other, unrelated languages. The particle tsi may have been extended from use as a place
nominalizer to time, manner, and degree, and ultimately to events that function as complements.
The final step appears to have been insubordination: ‘the conventionalized main clause use of what,
on prima facie grounds, appear to be formally subordinate clauses’ (Evans 2007: 367). The
meanings of the exclamatives often include semantic elements that might be conveyed in a matrix
clause, such as surprise or a subjective assessment of intensity: ‘be amazing’, ‘surprising’, ‘intense’,
etc., in combination with a complement that could denote a gradable property such as ‘pleasant’,
‘angry’, ‘funny’, or a situation of gradable intensity, such as ‘be tired’ or even ‘eat’. At a certain
point, the original matrix verbs apparently disappeared from these constructions, but their semantic
flavor remained as part of the construction.

(32) Exclamative: [Vina Loft, speaker]

Á: tsi nahonà:khwen’.
a: tsi n-a-ho-na’khwen-’
ah how PRT-FACT-M.SG.PAT-angry-PFV

‘Oh how angry he was.’

(33) Exclamative: [Watshenní:ne Sawyer, speaker]

Á: tsi niiohsteristòn:ne.
a: tsi ni-io-hsterist-on=hne
ah how PRT-N.PAT-funny-ST=PAST

‘Oh how funny it was.’

(34) Exclamative: [Watshenní:ne Sawyer, speaker]

16. Marianne Mithun

Ó: sens tsi na’teionkwahwishenhé:ion kí:ken

o: sens tsi n-a’-te-ionkwa-hwish-enhei-on kiken
oh formerly how PRT-FACT-DV-1PL.PAT-strength-die-ST this

iakwaksa’okòn:’a …
iakwa-ksa’=okon’a
1PL.EXCL-be.child=DIST

‘Oh how tired we were, us kids.’

(35) Exclamative: [Kaia’titáhkhe’ Jacobs, speaker]

Ótsta’ tsi nihonahiá:kon
otsta’ tsi ni-hon-ahi-ak-on
scab how PRT-M.PL.PAT-fruit-eat-ST

ki: kátshe’ káhi.

kiken ka-tshe’ ka-ahi
this N-jug N-fruit

‘Gee how they were gobbling up those pears.’

Mohawk shows evidence of a process by which such matrix clauses might fade. A common
way of expressing intensity in modern Mohawk is with the particle kwahí:ken ‘very’. For some
speakers this is now the regular, unmarked intensity construction.

(36) Intensifier: [Beatrice Francis, speaker]

Kwahí:ken tsi enkhní:rate’.
kwahiken tsi en-k-hnir-at-e’
very how FUT-1SG.AGT-strong-CAUS-PFV
‘I’ll make it very tight.’

(37) Intensifier: [Beatrice Francis, speaker]

Kwahí:ken tsi tehahnenhsaká:ron.
kwahiken tsi te-ha-hnenhs-a-kar-on
very how DV-M.SG.AGT-shoulder-LK-wide-ST
‘He had very wide shoulders.’

These kwahí:ken ‘very’ constructions all include the particle tsi. Closer examination of kwahí:ken
reveals the reason. It originated as a clause which would normally introduce a complement: ‘It is
intensely [that I will tighten it]’(36).

(38) kwáh í:ken

kwah i-ka-i
quite PROTHETIC-N-be
quite it is

Over time, the clause kwah í:ken has become reduced to what most speakers perceive simply as an
unanalyzable adverbial particle, often pronounced very briefly. But the nominalizer/complementizer
tsi remains in constructions with it. Other particles appear in modern exclamatives which
 Chapter 14. Fascinating exclamations
17

apparently also originated as full predicates serving as matrix verbs. One is é:so’ ‘much/many’,
which still occurs with tsi.

(39) Exclamative with é:so’ ‘much’: [Watshenní:ne Sawyer, speaker]

É:so’ tsi teió’tson.
eso’ tsi te-io-i’ts-on
much how DV-N.PAT-be.dirty-ST
‘How dirty it was!’

(40) Exclamative with é:so’ ‘much’: [Watshenní:ne Sawyer, speaker]

É:so’ tsi teionkhinónhware’.
eso’ tsi te-ionkhi-nonhwar-e’
much how DV-INDEF>1PL-whip-ST
‘How they whipped us!’

In some Mohawk dialects and related languages this word shows full verbal morphology, with the
Neuter pronominal prefix w- found on verbs and final glottal stop, a form of the Stative aspect
suffix: w-é:so-’ NEUTER-be.much-STATIVE = ‘It is much’.
Neither of these constructions, with kwahí:ken ‘very’ and é:so’ ‘much’, is technically
insubordination in Evans’ (2007) sense: traces of the erstwhile matrix remain. They do provide
examples, however, of steps in a kind of process which could ultimately result in insubordination
proper.
As seen earlier, a frequently-cited feature of exclamatives is factivity. Most Mohawk tsi
complement constructions, from which the exclamatives apparently developed, are factive.2 But the
modern Mohawk exclamatives are not necessarily factive. In fact most exclamatives in spontaneous
speech, like Á: tsi nahonà:khwen. ‘How angry he was!’ and Ó: sens tsi na’teionkwahwishenhé:ion
kí:ken iakwaksa’okòn:’a ‘Oh how tired we were, us kids’ ((32) and (34)) involved no
presupposition when they were uttered; the speakers were communicating the information that the
man was angry and the children were tired for the first time. This is exactly the kind of usage
discussed by Rett (2011) with What a magnificent place Crete is! It is well known that intensifying
constructions undergo change especially rapidly, as speakers seek new ways for adding emphasis
(Hopper and Traugott 2003: 122). This is apparently what has happened to both Mohawk and
European exclamatives. They are typically used simply to assert a property forcefully: Tsi
niion’wé:sen’ ‘How pleasant it is!’.
It should of course be noted that reconstruction of the source of a specific exclamative
construction in a language does not imply that there was no exclamative construction earlier.
Multiple exclamative constructions often coexist within languages, and as one gains ground,
another may recede.

3 Exclamatives, interrogatives, and declaratives

The fact that recurring cross-linguistic similarities between exclamatives and declarative
complements in many languages can be understood in terms of recurring processes of grammatical
change raises the question of whether the similarities between exclamatives and interrogatives seen
earlier might be understood in a comparable way. In fact they might. Assembling work based on
philological investigations of various European languages over centuries, Heine and Kuteva (2006)

2
A possible exception is ‘dream’, seen in (20) ‘I dreamt I was hunting in the bush’. This might be attributable to a
generalization of tsi to use with verbs of perception, similar to that proposed for Russian kak by Popova.
18. Marianne Mithun

demonstrate that interrogative pronouns often undergo a certain pathway of development. The first
three stages they list are in (41).

(41) Stages of Grammaticalization: (Heine and Kuteva 2006: 209)

Stage 1 The marker begins in lexical gap questions.

Who came?
Stage 2 The marker is extended to introducing indefinite subordinate clauses.
I don’t know who came.
Stage 3 The marker is extended further to definite subordinate clauses.
You also know who came.

Among European languages, different interrogative pronouns have developed to varying degrees
within and across languages. Developments were apparently stimulated by contact among European
languages, with epicenters in Latin and Slavic, then gradually spread outward geographically. There
is accordingly a high incidence of formal resemblances among interrogative pronouns and
complementizers across European languages, even those that are not related genetically.
This pathway of development could underlie the exclamative constructions that resemble
questions. A starting point would be degree questions: How good is it? The next step would be
extension to embedded questions: I don’t know [how good it is]. These constructions could then be
generalized to include complement constructions like I saw [how good it is], Look at [how good it
is]!, and It is surprising [how good it is]! The final step would be insubordination, the loss of the
matrix verbs from constructions like the last two, though the feature of intensity or unexpectedness
associated with the construction as a whole remained: How good it is! Such a scenario would
explain the recurring similarities between exclamatives and questions in so many European
languages, without the necessity of invoking an abstract synchronic structure.

4 Conclusion

Exclamative constructions have been observed to share certain characteristics: they are expressive
rather than informative, they are subjective, conveying a judgment on the part of the speaker, they
describe a scalable property, and they assert an unexpectedly high degree of that property. They
have attracted special interest among syntacticians because they often share part of their form with
some other construction in the same language. Much previous work has focused on exclamatives
that resemble interrogatives: How cute she is! How cute is she?. The resemblance has generally
been accounted for in terms of a shared abstract syntactic or semantic structure. Differences have
been attributed to a feature embedded in a higher level of structure, under which the exclamative is
embedded.
It has been observed, however, that not all exclamatives that display the key characteristics
(being expressive and subjective, asserting an unexpectedly high degree of a scalable property)
resemble questions. Many resemble the complements of declarative sentences. An alternative
approach that brings together the two types of exclamative constructions, those with links to
interrogatives and those with links to declaratives, explains the similarities in terms of their
development over time. The two types originate in different sources, but converge in the final steps
of their development: a matrix clause expressing surprise or an unexpectedly high degree of some
property described in the complement disappears. The meaning contributed to the construction by
the erstwhile matrix, however, remains a part of the construction as it takes on a conventionalized
syntactically independent use. This, of course, is the process termed by Evans (2007)
‘insubordination’.
 Chapter 14. Fascinating exclamations
19

Abbreviations

AGT grammatical agent

AL alienable possessor
APPL applicative
BEN benefactive
CAUS causative
CISLOC cislocative
CONTR contrastive
DEC decessive
DIM diminutive
DIST distributive
DU dual
DV duplicative
EXCL exclusive
FACT factual
FUT future
HAB habitual aspect
INCH inchoative
INCL inclusive
INDEF indefinite (generic)
INST instrumental
LK linker
LOC locative
MID middle
N neuter
NEG negative
NMLZ nominalizer
NS noun suffix
OPT optative
PAT grammatical patient
PFV perfective aspect
PL plural
POSS possessor
PROG progressive
PRT partitive
Q question
REP repetitive
REV reversive
ST stative aspect
T temporal
TAG tag
TRANSLOC translocative

References

Abels, Klaus and Øystein Alexander Vangsnes (eds.). 2010. Studia Linguistica [Special issue on
exclamatives] 64(1).
20. Marianne Mithun

Abels, Klaus and Øystein Alexander Vangsnes. 2010. Introduction: Exclamatives in North Germanic and
beyond. Studia Linguistica 64(1): 1–15.
Arutjunova, Nina D. 1988. Tipy jazykovyx značenij: Ocenka. Sobytie. Fakt (Types of linguistic meaning:
Evaluation. Event. Fact). Moscow: Nauka.
Aslanov, Cyril. 2009. Comme / comment du latin au français: Perspectives diachronique, comparatiste et
typologique. Travaux de linguistique 58(1): 19–38.
Austin, John. 1962. How to do things with words. Cambridge, MA: Harvard University Press.
d’Avis, Franz-Josef. 2001. Uber w-Exklamativsätze im Deutschen (On w-exclamatives in German)
Tübingen: Max Niemeyer Verlag.
d’Avis, Franz-Josef. 2002. On the interpretation of wh-clauses in exclamative environments. Theoretical
Linguistics 28: 5–31.
Benincà, Paola. 1995. Tipi di frasi principali. Il tipo esclamativo. Grande grammatica italiana di
consultazione, L. Renzi, G. Salvi and A. Cardinaletti (eds.), 127–152. Bologna: Il Mulino.
Benincà, Paola. 1996. La struttura della frase esclamativa alla luce del dialetto padovano. Italiano e dialetti
nel tempo. Saggi di grammatica per Giulio C. Lepschy, Paola Benincà, Guiglemo Cinque, T. De
Mauro, Nigel Vincent (eds.), 23–43. Rome: Bulzoni .
Bennis, Hans. 1998. Exclamatives! Linguistics in the Netherlands 1998, René van Bezooijen and René
Kager (eds.), 27–40. Amsterdam/Philadelphia: John Benjamins.
Bennis, Hans, Norbert Corver and Marcel den Dikken. 1998. Predication in nominal phrases. Journal of
Comparative Germanic Linguistics 1: 85–117.
Biber, Douglas, Stig Johansson, Geoffrey Leech, Susan Conrad and Edward Finegan. 1999. Longman
grammar of spoken and written English. London: Longman.
Bosque, Ignacio. 1984. Sobre la sintaxis de las oraciones exclamativas (On the syntax of exclamative
sentences). Hispanic Linguistics 1: 283–304.
Castroviejo Miró, Elena. 2006. Wh-exclamatives in Catalan. Ph.D. dissertation, Universitat de Barcelona.
Castroviejo Miró, Elena. 2008. Deconstructing exclamations. Catalan Journal of Linguistics 7: 41–90.
den Dikken, Marcel. 2006. Relators and linkers. Cambridge, MA: MIT Press.
Elliott, Dale. 1971. The grammar of emotive and exclamatory sentences in English. Ph.D. dissertation, The
Ohio State University, Columbus. [Reproduced in Working Papers in Linguistics 8, Computer and
Information Science Research Center, The Ohio State University.]
Elliott, Dale. 1974. Toward a grammar of exclamations. Foundations of Language 11(2): 231–246.
Evans, Nicholas. 2007. Insubordination and its uses. Finiteness: Theoretical and empirical foundations. Irina
Nikolaeva (ed.), 366–431. Oxford: Oxford University Press.
Gérard-Naef, Jocelyn. 1980. L’exclamation en français: la syntaxe des phrases et des expressions
exclamatives. Tübingen: Max Niemeyer.
González Rodriguez, Raquel. 2008. Exclamative wh-phrases as positive polarity items. Catalan Journal of
Linguistics 7: 91–116.
Grimshaw, Jane. 1977. English Wh-questions and the theory of grammar. Ph.D. diss., University of
Massachusetts, Amherst.
Grimshaw, Jane. 1979. Complement selection and the lexicon. Linguistic Inquiry 10(2): 279–326.
Gutiérrez-Rexach, Javier. 1996. The semantics of exclamatives. Syntax at Sunset, UCLA Working Papers in
Linguistics 1996, Edward Garrett and Felicia Lee (eds.), 146–162. Department of Linguistics,
University of California, Los Angeles.
 Chapter 14. Fascinating exclamations
21

Gutiérrez-Rexach, Javier. 2001. Spanish exclamatives and the interpretation of the left periphery. Selected
papers from going Romance 99, Johan Rooryck, Yves de Hulst and Jan Schroten (eds.), 167–194.
Amsterdam and Philadelphia: John Benjamins.
Gutierrez-Rexach, Javier. 2008. Spanish root exclamatives at the syntax/semantics interface. Catalan
Journal of Linguistics 7: 117–133.
Heine, Bernd and Tania Kuteva. 2006. The changing languages of Europe. Oxford: Oxford University Press.
Hopper, Paul and Elizabeth Traugott. 2003. Grammaticalization (2nd edition). Cambridge: Cambridge
University Press.
Huddleston, Rodney. 1993. Remarks on the construction You won’t believe who Ed has married. Lingua 91:
175–184.
Kiparsky, Paul and Carol Kiparsky. 1970. Fact. Progress in linguistics, Manfred Bierwisch and Karl Erich
Heidolph (eds.), 143–173. The Hague: Mouton.
Kobozeva, Irina M. 1988. Otricanie v predloženijax s glagolami vosprijatija, mnenija i znanija (Negation in
sentences with verbs of perception, opinion, and knowledge). Logičeskij analiz jazyka: Znanie i
mnenie (The Logical analysis of language: Knowledge and opinion, 82–98. Moscow: Nauka.
Kobozeva, Irina M. and Dar’ja Popova. 2013. Faktory vybora iz’’jasnitel’nyx sojuzov ‘kak’, ‘čto’, ‘čtoby’
(opyt tipologičeski orientirovannogo formal’nogo analiza). (Factors in choices among the
complementizers kak, čto, čtoby [an attempt at a typologically-oriented formal analysis].) Vestnik
Moskovskogo Universiteta. Serija 9. Filologija. Issue 1, 21–34.
König, Ekkehard and Peter Siemund. 2007. Speech act distinctions in grammar. Language typology and
syntactic description, Vol. 1, Clause structure (2nd edition), Timothy Shopen (ed.), 276–324.
Cambridge: Cambridge University Press.
Lahiri, Utpal. 1991. Embedded interrogatives and predicates that embed them, Ph.D. diss., MIT.
Langacker, Ronald. ms. An analysis of English questions.
Mayol, Laia. 2008. Catalan “Déu n’hi do” and conventional implicatures in exclamatives. Catalan Journal
of Linguistics 7: 135–156.
Marandin, Jean-Marie. 2008. The exclamative clause type in French. Proceedings of the HPSG08
Conference, NTCT, Keihanna, Japan, Stefan Müller (ed.). CSL Publications. http://cslo-
publications.stanford.edu.
Michaelis, Laura. 2001. Exclamative constructions. Language typology and language universals, vol. 2,
Martin Haspelmath, Ekkehard König, Wulf Oesterreicher and Wolfgang Raible (eds.), 1038–1050.
Berlin: de Gruyter.
Michaelis, Laura and Knud Lambrecht. 1996a. The exclamative sentence type in English. Conceptual
structure, discourse and language, Adele E. Goldberg (ed.), 375–89. Stanford, CA: CSLI.
Michaelis, Laura and Knud Lambrecht. 1996b. Towards a construction-based theory of language function:
The case of nominal extraposition. Language 72(2): 215–47.
Näf, Anton. 1996. Die w-Exklamativsätze im Deutschen – zugleich ein Plädoyer für eine Rehabilitierung der
Empirie in der Sprachwissenschaft (W-exclamative sentences in German, together with a plea for a
rehabilitation of empiricism in linguistics). Zeitschrift fur Germanistische Linguistik 24(1): 135–152.
Noonan, Michael. 2007. Complementation. Language typology and syntactic description, vol. 2, Complex
constructons (2nd edition), Timothy Shopen (ed.), 52–150.
Padučeva, Elena. 1986. O referencii jazykovyx vyraženij s nepredmetnym značeniem (On reference in
linguistic expressions with abstract meaning). Naučno-texničeskaja informacija (Scientific and
Technical Information). Ser. 2. N 1.23–31.
22. Marianne Mithun

Padučeva, Elena. 2004. Dinamičeskie modeli v semantike leksiki (Dynamic models in lexical semantics).
Moscow: Jazyki slavjanskoj kul’tury.
Popova, Daria. 2011a. Contrasts among three Russian complementizers ‘kak’, ‘chto’ and ‘chtoby’. Final
paper for Barbara Partee’s speckurs Formal Semantics and Semantic Universals, Moscow State
University, June 2011.
Popova, Daria. 2011b. ‘Kak’/‘chto’ alternation in Russian. Final paper for Beth Levin’s course Lexical
Semantics, Stanford, December 2011.
Postma, Gert. 1995. Exclamative wh and verb-second. Linguistics in the Netherlands 1996, Crit Cremers and
Marcel den Dikken (eds.), 207–220. Amsterdam/Philadelphia: John Benjamins.
Potts, Christopher and Florian Schwarz. 2008. Exclamatives and heightened emotion: Extracting pragmatic
generalizations from large corpora (manuscript). University of Massachusetts, Amherst.
Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech and Jan Svartvik. 1985. A comprehensive grammar of
the English language. Harlow: Longman.
Radford, Andrew. 1982. The syntax of WH-exclamatives in Italian. Studies in the Romance Verb, Nigel
Vincent and Martin Harris (eds.), 185–204. London: Croom Helm.
Rett, Jessica. 2011. Exclamatives, degrees and speech acts. Linguistics and Philosophy 34: 411–442.
Rizzi, Luigi. 1997. The fine structure of the left periphery. Elements of grammar: Handbook in generative
syntax, Liliana Haegeman (ed.), 281–337. Dordrecht: Kluwer.
Rosengren, Inger. 1992. Zur Grammatik und Pragmatik der Exklamation. (On the grammar and pragmatics
of exclamations.) Satz und Illokation, Inger Rosengren (ed.), 263–304. Tübingen: Niemeyer.
Rosengren, Inger. 1997. Expressive sentence types – A contradiction in terms. The case of exclamation.
Modality in Germanic languages. Historical and comparative perspectives, Toril Swan and Olaf
Jansen Westvik (eds.), 151–184 Berlin: Mouton de Gruyter.
Ross, John Robert. 1970. On declarative sentences. Readings in English transformational grammar,
Roderick Jacobs and Peter Rosenbaum (eds.), 223–77. Waltham, MA: Ginn and Co.
Sadock, Jerrold. 1970. Whimperatives. Studies presented to Robert B. Lees by his students. Jerrold Sadock
and Anthony Vanek (eds.), 223–38. Edmonton, Alberta: Linguistic Research Inc.
Sadock, Jerrold and Arnold Zwicky. 1985. Speech act distinctions in syntax. Language typology and
syntactic description, vol. 1, Clause structure, Timothy Shopen (ed.), 155–196.
Villalba, Xavier. 2001. The right edge of exclamative sentences in Catalan. Catalan Working Papers in
Linguistics 10: 119–135.
Villalba, Xavier. 2002. Dos tipus d’oracions exclamatives en català. Revista de llengua i literatura, Els
Marges (ed.), 83–93.
Villalba, Xavier. 2003. An exceptional exclamative sentence type in Romance. Lingua 113(8): 713–745.
Villalba, Xavier. 2004. Exclamatives and negation. Technical Research Report GGT-2004-02. Bellaterra:
Grup de Gramàtica Teòrica, Universitat Autònoma de Barcelona.
Villalba, Xavier (ed.). 2008. Exclamatives at the interface. Special issue, Catalan Journal of Linguistics 7.
Villalba, Xavier and Anna Bartra-Kaufmann. 2010. Predicate focus fronting in the Spanish determiner
phrase. Lingua 120(4): 819–849.
Zaliznjak, Anna. 1990. O ponjatii “fakt” v lingvističeskoj semantike (On the notion of ‘fact’ in linguistic
semantics). Logičeskij analiz jazyka: protivorečivost’ i anomalnost’ teksta (The logical analysis of
language: Contradictoriness and abnormality in texts), 21–32. Moscow: Nauka.
Zanuttini, Raffaella and Paul Portner. 2000a. The characterization of exclamative clauses in Paduan.
Language 76: 123–132.
 Chapter 14. Fascinating exclamations
23

Zanuttini, Raffaella and Paul Portner. 2000b. The force of negation in wh exclamatives and interrogatives.
Negation and polarity: Syntactic and semantic perspectives, Laurence Horn and Yasuhiko Kato (eds.),
201–239. Oxford: Oxford University Press.
Zanuttini, Raffaella and Paul Portner. 2003. Exclamative clauses: At the syntax-semantics interface.
Language 79(1): 39–81.