Mosaics of Knowledge Representing Information in The Roman World Andrew Riggsby Full Chapter PDF

Mosaics of Knowledge: Representing
Information in the Roman World

Andrew Riggsby
Visit to download the full and correct content document:
https://ebookmass.com/product/mosaics-of-knowledge-representing-information-in-th
e-roman-world-andrew-riggsby/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Trade, Commerce, and the State in the Roman World

Andrew Wilson
https://ebookmass.com/product/trade-commerce-and-the-state-in-
the-roman-world-andrew-wilson/
London in the Roman World Dominic Perring
https://ebookmass.com/product/london-in-the-roman-world-dominic-
perring-2/
London in the Roman World Dominic Perring
https://ebookmass.com/product/london-in-the-roman-world-dominic-
perring/
Women and Society in the Roman World: A Sourcebook of

Inscriptions from the Roman West Emily A. Hemelrijk
https://ebookmass.com/product/women-and-society-in-the-roman-
world-a-sourcebook-of-inscriptions-from-the-roman-west-emily-a-
hemelrijk/
New Directions in the Study of Women in the Greco-Roman
World Ronnie Ancona
https://ebookmass.com/product/new-directions-in-the-study-of-
women-in-the-greco-roman-world-ronnie-ancona/
Spectacles in the Roman World: A Sourcebook Siobhán

Mcelduff
https://ebookmass.com/product/spectacles-in-the-roman-world-a-
sourcebook-siobhan-mcelduff/
Coin Hoards and Hoarding in the Roman World Jerome

Mairat
https://ebookmass.com/product/coin-hoards-and-hoarding-in-the-
roman-world-jerome-mairat/
Infancy and Earliest Childhood in the Roman World: ’A

Fragment of Time’ Maureen Carroll
https://ebookmass.com/product/infancy-and-earliest-childhood-in-
the-roman-world-a-fragment-of-time-maureen-carroll/
Greek and Roman Technology: A Sourcebook of Translated

Greek and Roman Texts 2nd Edition Andrew N. Sherwood
https://ebookmass.com/product/greek-and-roman-technology-a-
sourcebook-of-translated-greek-and-roman-texts-2nd-edition-
andrew-n-sherwood/
Mosaics of Knowledge
ii
Classical Culture and Society

Series Editors
Joseph Farrell and Robin Osborne
Emotion, Restraint, and Community in Ancient Rome

Robert A. Kaster
Making Mockery: The Poetics of Ancient Satire

Ralph M. Rosen
Readers and Reading Culture in the High Roman Empire:

A Study in Elite Communities
William A. Johnson
Apollonius of Rhodes and the Spaces of Hellenism

William G. Thalmann
The Captor’s Image: Greek Culture in Roman Ecphrasis

Basil Dufallo
Aratus and the Astronomical Tradition

Emma Gee
Gift and Gain: How Money Transformed Ancient Rome

Neil Coffee
Mosaics of Knowledge: Representing Information in the Roman World

Andrew M. Riggsby
Mosaics of Knowledge
Representing Information in the
Roman World
Andrew M. Riggsby
1
iv
1
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.
Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2019
All rights reserved. No part of this publication may be reproduced, stored in

a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.
You must not circulate this work in any other form

and you must impose this same condition on any acquirer.
CIP data is on file at the Library of Congress

ISBN 978–0–19–063250–2
9 8 7 6 5 4 3 2 1
Printed by Sheridan Books, Inc., United States of America

For Mom and Dad
vi
CONTENTS
List of Figures ix
List of Tables xi
List of Plates xiii
Acknowledgments xv
A Brief Orientation 1
1. Lists 10
Ordered Lists 11
Indexed Lists 15
Tables of Contents 22
Nested Lists 29
2. Tables and Tabular Organization 42

Actual Tables 50
Not Tables 54
Outliers 70
Conclusions 73
3. Weights and Measures 83

How Does Roman Measurement Work? 85
Standards and Standardization 100
Direct Standardization 107
Indirect Standardization 115
Complications 120
Conclusions 125
Chapter Appendix 129
4. Representing Three Dimensions 130

Perspective and the Theory of Space 131
The Corpora 135
Space in the Landscapes 138
Two Comprehensive Examples 147
Conclusions 149
5. Representing Two Dimensions 154

Data Graphics 154
Plans 164
viii
viii { Contents
What Is a “Map”? 172

Ancient Maps 180
Maps as Information Technology 194
Chapter Appendix 201
6. Conclusion 203
Where Are We Now? 203
Going Forward I: Power and Other Topics 210
Going Forward II: An IT Revolution in Late Antiquity? 216
References 223
Index 245
FIGURES
1.1 Supposed theater token from Pompeii 32

1.2 Supposed amphitheater token from Arles 33
1.3 Inscription on a theater seat from Verona; token from the amphitheater
at Frosinone 35
2.1 Organization of status theory 43
2.2 Schematic diagram of Roman centuriation 50
2.3 Military duty roster from Egypt 53
2.4 Military duty roster from Egypt 54
2.5 Victorius, Calculus 63
4.1 Landscape from the columbarium of Villa Doria Pamphilj 142
4.2 Landscape from room 14, Villa A, Oplontis 143
4.3 Landscape from the villa under the Farnesina 145
4.4 Fall of Icarus 153
5.1 Roman portable sundial 156
5.2 Schematic illustration from a land-surveying manual 158
5.3 More naturalistic illustration from a land-surveying manual 158
5.4 Inscription detailing rights to draw water 159
5.5 Inscription showing plans for the funerary complex of Claudia Peloris and
Ti. Claudius Eutychus 166
5.6 “Map” from Dura Europos 173
5.7 Population-adjusted map illustrating the outcome of the 2016
U.S. presidential election 174
5.8 Stylized map illustrating the outcome of the 2016 U.S. presidential
election 175
5.9 Network rendering of places named in Caesar 178
5.10 Schematic, two-dimensional rendering of places named in Caesar 179
5.11 Forma Urbis Romae, detail 182
5.12 Forma Urbis Romae, detail 182
5.13 Scale of FUR implied by individual comparisons with modern measurement
and magnitude of each measurement 183
5.14 Fragment of inscribed map depicting the centuriation at Arausio 187
5.15 Tabula Peutingeriana, detail 192
5.16 Tabula Peutingeriana, detail 192
6.1 Jerome, Chronicle 219
ix
x
TABLES
1.1 Early references to the supposed Arles amphitheater token 34

1.2 Supposed form of reference to a centralized catalog of Roman public
statuary 39
1.3 Data about three American cities, arranged in tabular form 41
2.1 Varro illustrates linked proportions with numbers 52
2.2 Varro uses linked proportions to structure the declension of an
adjective 52
2.3 Modern declension of the phrase hic Marcus 59
3.1 Multiple meanings of symbols used in systems of measurement 92
3.2 Variations in actual weights with respect to presumed standard values 103
3.3 References to measured lots of grain in TPSulp 117
3.4 Standard reference values for several Roman units of measurement 129
5.1 Concordance of Roman building plans 202
xi
xii
PLATES
1 Fasti Amiternini, with color coding

2 Riot in the Amphitheater, Pompeii
3 Landscape from the columbarium of Villa Doria Pamphilj
4 Landscape from the villa under the Farnesina, walkway
5 Landscape from the villa under the Farnesina, cubiculum
6 Landscape from a villa at Boscotrecase
7 Landscape from the villa under the Farnesina, triclinium
8 Fragment of marble map of Rome (“via Anicia” fragment)
xiii
xiv
ACKNOWLEDGMENTS
This project, at least some parts of it, dates back a long time. The seminar alluded to
at the beginning of chapter 2 was offered in the late 1990s, and I suspect that some
of the thoughts here probably first arose before I finished graduate school, while
I was reading Edward Tufte’s books from my mother’s book shelf. I have acquired
an unusually large number of scholarly debts over that time (and unfortunately
have doubtless forgotten others equally important). I got particularly extensive as-
sistance and commentary from Klaus Geus, Paul Keyser, Michael Koortbojian,
Rabun Taylor, and the readers for Oxford University Press. Tony Corbeill, Serafina
Cuomo, Tony Grafton, Joseph Howley, Nate Jones, Stephanie Frampton, John
Clarke, Eric Orlin, Liz Robinson, Philip Stinson, and Tyler Travillian all read
and commented on chapters in draft. I have also gotten other help, particularly
in the form of penetrating questions or advance access to work in progress from
Dorian Borbonus, Alan Cameron, C. Michael Chin, Megan Goldman-Petri, Julia
Hejduk, Alexander Jones, Duncan McRae, Reviel Netz, Carlos Noreña, Laura
Novick, Dan-el Padilla Peralta, Tim Parkin, J.-B. Piggin, Phil Resnik and Jiesi Shi,
Jane Sancinito, Josh Sosin, and a seminar which covered this material (Gabrielle
Bouzigard, Timothy Corcoran, Eli Fleming, Vera Leh, Will Shrout, and Alain
Zamarian). I would also like to thank audiences at Brown, Chicago, Columbia,
Duke, Johns Hopkins, Maryland- Baltimore County, Minnesota, NYU, North
Carolina, Penn, Princeton, Texas Tech, Yale, and the Finnish Institute in Rome for
subjecting various parts of the argument to friendly scrutiny. And, of course, I need
to thank Joe Farrell, the series editor, and Stefan Vranka, the sponsoring editor, for
their interest, encouragement, and assistance in transforming the “project” into an
actual book.
Finally, I would particularly like to signal the role in this project of my ongoing
interaction with two younger scholars. Seth Bernard and Sarah Bond in their dis-
tinct, inimitable ways provided a stream of questions, prods, prompts, and problems
and materials to work with. A project of this scope necessarily relies on the kind-
ness of strangers to have any hope of reaching the necessary breadth, but even be-
yond that the constant presence of these two kept me honest and on my toes.
While I have been working on parts of this project for many years, the core of
the research and writing took place over two academic years, and I am more than
happy to thank the funding entities that made that possible. In 2010–11, I held
the NEH/Roger A. Hornsby Rome Prize at the American Academy in Rome. In
addition to the scholars named earlier, I must thank the Academy for both the
xv
xvi
xvi { Acknowledgments
Fellowship and the atmosphere uniquely hospitable to scholarship. Then in 2013–

14, I was the Stanley Kelley Jr. Visiting Professor for Distinguished Teaching at
Princeton, a position which (perhaps ironically) carries quite modest teaching
responsibilities and which in turn allowed me to take advantage of the remarkable
research resources there. My thanks go to the University, the Classics Department,
and to Andrew Feldherr, who brought it all together.
I also need to thank several institutions which supplied other kinds of intellec-
tual resources. The Soprintendenza Archeologia del Veneto and Dott.ssa Brunella
Bruno, the director of its Nucleo Operativo di Verona, were kind enough to allow
me direct examination of the two bronze map fragments found there (and discussed
in chapter 5). The Bodleian Library in Oxford allowed me to inspect their manu-
script of Jerome’s Chronicle (discussed in the conclusion). Bruce Barker-Benfield,
Senior Assistant Librarian in the Department of Special Collections and Western
Manuscripts there, was particularly generous with his time and expertise in
discussing the manuscript with me, and I hope to be able to publish further fruits of
those discussions in due course. The Bibliothèque municipale d'Avignon provided
images of an extremely rare volume on the antiquities of Arles. Finally, I (as every
academic) must thank the library staff at my home institution, especially Shiela
Winchester, the Classics bibliographer, and the InterLibrary Services Department
for providing (and often finding) an endless supply of research materials. Kristina
Schlegel did all the original drawings masterfully. Andrea Pittard provided assis-
tance with the manuscript and bibliography. Khoa Tran did heroic work with image
permissions. C. Berglie, the copy editor, had to deal with a rat’s nest of references.
It would probably not have been possible for me to write this book a decade
earlier. Modern information technology made it possible for me to gather the kind
of scattered evidence it relies on and to move into several areas that were previ-
ously unfamiliar to me. At the same time, it relied on the serendipity provided by
traditional library shelving, and I fear that in another decade or so it will again be
impossible to write a book of this sort, where the objects of inquiry and sources of
evidence were not givens from the beginning.
My wife, Lisa Sandberg, once again brought her formidable editing skills to bear
to grant this book such readability as it has, despite being subjected to the intermi-
nable process through which I brought the framework together.
My father was a scientist with a strong amateur interest in premodern history,
and was happy that he was able to read this whole book in manuscript before his
death. And even before learning of her career as a computer programmer, one
could spot my mother’s interest in data and design from the way she puts together
a quilt. It is to them that this book is dedicated.
Plate 1 Fasti Amiternini. Color coding indicates categories of information
DeA Picture Library, licensed by Alinari

2
Plate 2 Riot in the Amphitheater (Pompeii, now in the Museo Archeologico Nazionale, Naples)
© Vanni Archive/Art Resource, New York
Plate 3 Landscape from the columbarium of the Villa Doria Pamphilj (A/XII)
Su concessione del Ministerio dei beni e delle attività culturali e del turismo—Museo Nazionale Romano
4
Plate 4 Landscape from the villa under the Farnesina, walkway F-G (inv. 1233)
Plate 5 Landscape from the villa under the Farnesina, cubiculum D (inv. 1037)
6
Plate 6 Landscape from a villa at Boscotrecase

© Marie-Lan Nguyen /Wikimedia Commons, used by permission
Plate 7 Landscape from the villa under the Farnesina, triclinium C (inv. 1080)
8
Plate 8 Fragment of a marble map of Rome (the “via Anicia” fragment)

Su concessione del Ministerio dei beni e delle attività culturali e del turismo—Museo
Nazionale Romano
A Brief Orientation
This book investigates information technologies in the classical Roman world—

their invention, diffusion, and use, and the interactions among those processes. The
focus is on conceptual developments—e.g., “mapping,” “weighing,” “listing”—
rather than material ones—e.g., “codex,” “abacus.” (Within the area covered, how-
ever, the interaction of concepts with the materiality of their actual uses will be a
recurring theme.) It also focuses principally on “high” technologies rather than, say,
literacy or numeracy in general. Perhaps paradoxically, this will end up setting the
book against most work to date on classical knowledge regimes. Scholarship has
typically dealt with intra-elite and largely discursive phenomena. As a result, we
know a good deal about the intellectual history of antiquity’s formalized disciplines
(e.g., rhetoric, philosophy, law, literature, grammar) and how they competed with
and inflected one another. By contrast, my goal is to uncover an alternative set of
regimes which were generally not theorized in antiquity, but which informed the
practices of daily life, and did so in a broad variety of social locations (even if some
had elite origins). These turn out to include relatively advanced technologies like
complicated lists, tables, and textual illustrations.
While most of the book will be about technologies that were “advanced” in their
time, I want to begin with a brief narrative of the study of a more basic one: literacy.
Until a few decades ago it was a commonly, if not universally, held view that the
ability to read and write was widespread in the Roman world, not different at least
in kind from advanced nations in the modern world. This changed dramatically
with the publication in 1989 of William Harris’ Ancient Literacy. Harris deployed
comparative evidence to argue that such mass literacy could only exist in contexts
that met a number of social and institutional prerequisites—systematic education,
broad economic advantages that flow only to the literate, and the like—then showed
systematically that almost none of this was true anywhere in the classical world. On
this basis he then projected rough literacy rates on the order of 1% to 30% at var-
ious times and places within that world. Much of the response to Harris has been
accepting, if slightly more “optimistic,” at least on a local level. That is, scholars
1
2
2 { Mosaics of Knowledge
have been more willing to see social and institutional supports for literacy, even
if they take different forms than modern ones and even if they only create larger
pockets of literacy rather than universalizing it. But the real differences lie not in
tweaking Harris’ numbers but in adding to his stock of questions and localizing
their answers. There has been an increasing analytic interest in what might be called,
in the plural of the title of Johnson and Parker’s 2009 book, Ancient Literacies.
That is, with the basic quantitative picture already in place, interest has shifted to
more qualitative questions of how reading and writing skills (and, to a lesser extent,
numeracy) were employed by individuals in particular contexts. What kind of infor-
mation, Woolf 2009 asks, would labels on commercial olive oil jars in their highly
stereotyped format have been able to convey to “readers” in the industry, who might
not be fully literate in general terms and might not even be Latin speakers? Or what,
Beard 1991 considers, is the motivating effect of rituals that mediated access to the
divine through writing?
I tell this story, which will already be familiar to many readers, because it has
multiple resonances with the unfamiliar story I will tell in the body of this book.
First, I hope also to make modern states of affairs seem less natural. One reason
(though not, of course, the only one) it was easy to accept a highly literate antiquity
was the ease with which reading comes to individuals today. It does not then feel
like a very strong claim to extend that across an entire society, though in fact it is.
The information technologies discussed in this book—things like numbered lists,
numerical tables, or mechanical weights and measures—offer a similar temptation.
Their use comes so naturally to anyone acculturated in the modern world that we
are likely to take it for granted that they were available in the ancient world as well.
Most insidiously, something merely similar to a modern device can readily be taken
for fully identical.
Additionally, however, I would like to try to borrow some of the qualitative focus
that characterizes much of the response to Ancient Literacy. That is, I ask not just
how often Romans used various technologies but also when, how, and why. Those
contextual questions are probably good historical practice in general, but they
strike me as particularly urgent in this specific area of technology. As it happens,
most of the information technologies discussed here were not huge successes that
spread through the Roman world in the way, say, blown glass or concrete construc-
tion did. But the reason I can talk about them at length is that they were at least
invented, unlike, say, stainess steel or stirrups. One need not (as most readers prob-
ably will not) believe in the necessary “progress” of technology to be puzzled by a
lack of eagerness to adopt what were at least technically useful devices. “Why did
the Romans use tables or scale representations differently than how we do?” for
instance, is interesting because it will likely tell us as much about the Romans in
general as about those tables or plans. Different people use technologies for par-
ticular purposes in particular circumstances. The question, thus, is almost never
whether a particular technology is “good” or “powerful” or “elegant,” or anything
else. The issue is whether particular people (or enough of them) find it worthwhile
A Brief Orientation } 3
to acquire that technology for some particular task before them. In a broad sense,
we could make this point about any adoption of any technology, but I will argue
that Roman cost-benefit calculations in this respect were particularly strict and
particularly local.
While I do not, I hope, adopt a teleological view of technological change,
I should probably also point out that I do not hold a purely culturalist view, ei-
ther (even though the previous paragraph might have been read that way). I don’t
think it is meaningful to describe any technology as “good” or “the best,” only
good or best for some particular end (which might itself be defined by a complex
of mechanical, social, and other aims). However, this does mean that to the extent
that we understand those aims, we can say that some technologies are objectively
better: they have a lower rate of false positives, they create less pollution, they are
cheaper, the hardware is less likely to malfunction, they require fewer (or, if this is
what circumstances demand, more numerous) human workers, they channel rev-
enue to a politically powerful class. My account is also imperfectly culturalist be-
cause I believe that the invention and diffusion of inventions are path-dependent.
Devices do not simply arise whenever and wherever cultural circumstance might
make them desirable. Various material preconditions and contingent discoveries are
required. I don’t claim to have proven the truth of this point of view. Rather, I have
sketched an approach which will stand or fall depending on how well it actually
works throughout the body of the study.
Though quite broad, the scope of this book will be limited in two important
ways. Chronologically, it will extend, at least in principle, from the earliest Roman
times to the year 300 (all dates will be ce unless otherwise noted). Any precise cutoff
is of course arbitrary to some extent, but I have chosen this one for several reasons.
As a practical matter, going substantially later would have expanded the available
evidence too much to be able to handle (the reader will have to decide whether I have
already bitten off more than I can chew). A cut-off around 300 ce also corresponds
to a fairly traditional sense of “classical” (that is, non-Christian) Roman culture.
That would not obviously be relevant in itself, but I suspect that it is connected
to another reason for the cut-off. It appears to me, on the basis of evidence I have
admittedly scrutinized less carefully, that there is an information technology revo-
lution in Late Antiquity. There seem to be significant changes to the technologies
described in most of my chapters (and the invention of at least one important new
one) during roughly the fourth and fifth centuries. I will say a little more about both
the shape and the possible reasons (some Christianity-based) for this revolution in
the last chapter, but the topic seems to me to require separate treatment. The end
date of 300 will not be applied mechanically. First, evidence from later periods can
sometimes be used to cast light on earlier ones. Second, I will be fairly generous in
allowing myself to use evidence (especially inscriptions) that is not definitively dat-
able to within my period.
The other constraint will also be partial, perhaps even more so. Though I have
been speaking (and will continue to do so) of “Roman” information technology, the
4
focus of some of the chapters will be specifically on the Latin-speaking world. That
is hardly an obvious line to draw in a historical rather than, say, a literary context,
so let me say a few words about why a linguistic distinction might be appropriate
to the particular topic and in what ways in which this limitation will and will not
be observed over the course of the book. Many of the technologies discussed here,
especially in those first two chapters, can be seen as specialized forms of literacy.
This seems to me to be the central insight of Jack Goody’s famous 1977 book, The
Domestication of the Savage Mind, and its chapters on the cognitive operations
enabled or encouraged by tables, lists, formulas, and recipes. My subject matter is
deeper and narrower than Goody’s, but I also differ from him in one methodolog-
ical emphasis. He tends to look more at what written technologies can do than what
they actually do. The former was important for opening up a field of inquiry in a
theoretical kind of way, but the historical specificity that I am aiming for seems to
require the latter. Seeing these technologies as an aspect of literacy accounts for a
restriction on the scope of this book. Though I will generally speak of “Rome” and
“Romans,” my focus will be on the Latin world when the devices in question are
used and transmitted by writing, and so we would not necessarily expect them to be
constant across linguistic boundaries.
Two examples might help show how this linguistic division clearly can work in
practice (one is from outside the realm of this study and the other will be treated at
length in c hapter 2). For the former, I offer the following observation on differences
in substance between contracts surviving from Roman Egypt correlated with the
language in which they are written (Alonso 2016, 65):
[T]‌he contracts concluded by Romans in Greek are usually indistinguishable

from those concluded by [non-citizens]. Vain have been most attempts to
identify “Roman” traits in these Greek contracts. Their Latin contracts, in-
stead, which are rather scarce, do adhere to the Roman models both in form
and content, even if occasionally with peregrine accretions.
The contracts in question, even the ones in Greek, were concluded between Roman
citizens. The difference is thus not one of culture in general or even legal culture
but, rather, of language. Alonso suggests, plausibly enough, that the key factor was
reliance on notaries among the less privileged classes to produce these kinds of
documents. The second example has to do with tables (the organizational device,
not the piece of furniture). These are not, as it turns out, particularly common in
either the Latin-or Greek-speaking worlds, though they do appear in certain, very
limited contexts. One of those contexts in the Greek world was the display of as-
tronomical data of various sorts. Yet these do not appear in Latin. Roman authors
either give (some of) the same information in continuous prose or in lists or, in the
extreme case of Vettius Valens (second century), write in Greek themselves.
These examples are not meant to show that Latin and Greek information
technologies are entirely cut off from each other, a claim that would clearly be false.
Rather, I simply wish to illustrate that there are both theoretical and empirical
reasons to suspect that there might be a significant level of separation. Nor do

I mean to ignore the Greek-speaking world all together. Most mechanically, there
are at least a couple of cases where we have Greek texts that appear to be fairly
direct representations of operations of the Roman state originally conducted in
Latin (e.g., translations of senatorial decrees, labels on weights authorized by a
Roman governor). Though they are rare, I will also record cases where I think di-
rect influence from Greece to Rome can be established (e.g., the introduction of
the chronological table). Occasionally, there will be points at which Greek evi-
dence seems to provide an illuminating comparison, if not of a different sort than
Mesopotamian cuneiform or Renaissance Italian evidence might offer. Finally, the
technologies of c hapters 3 through 5 are less clearly tied to writing and so there is
less call for a distinction. For instance, I look at weights and measures from across
the Roman Empire. Even in those cases, however, the linguistic distinctions can re-
surface. The most “scientific” tradition of ancient (Mediterranean) mapping, for in-
stance, is closely tied, even subordinate to, written geographical texts, and this may
be reflected in broader practice. Greek and Roman practice in textual illustration
overlaps in some respects, but contrasts in others.
In addition to those restrictions of scope I should mention what may seem a
peculiar feature of many of the arguments. This book discusses devices that I pre-
sume will be familiar not only to every reader but also—with the possible exception
of “perspective” in c hapter 4—to their school-aged children. It may then come as
a surprise that in many of the chapters I spend as much time as I do working out
definitions of terms like “list,” “table,” and “map.” I will say a few words here about
why that is. I do not intend to tease out “what we really mean” by these terms in the
manner of an analytic philosopher. Nor do I intend a more historical or philolog-
ical investigation of the possible semantic range of these English words, though for
clarity’s sake I will point out ambiguities that I fear may mislead. Nor do I think
that any of these technologies form a “natural kind” like, arguably, “dinosaur”
or “proton” or “water.” Nor, finally, am I generally trying to reconstruct ancient
Roman conceptual categories. Rather, I am trying to capture features of Roman
practice that line up only approximately with any language’s lexicon. In these cases
I have decided that the easiest way forward is not to try the reader with frequent
neologisms (“category 1 map,” “category 2 map,” etc.) or circumlocutions (“list-
whose-order-conveys-information-to-the-intended-audience” vs. “list-with-indices-
permanently-attached-rather-than-created-on-the-fly”). I have chosen instead to
use ordinary words that make the general area of interest clear, then to stipulate
whatever additional properties are important for my local purpose. I do not defend
these definitions as truer than any other. I intend them as useful, however, if they
help express significant distinctions in Roman practice. For example, are lists of the
types just mentioned restricted to specific use-contexts?
It is also true, however, that there is another, more specific issue at stake, and
it is one that systematically results in more complex definitions. This arises from
the relation of form and function in the technologies. When definitions have been
6
offered in this area by others, they have typically been framed in terms of form.
What, for instance, does a “table of contents” look like? In the last instance, I am
more interested in function instead, but that cannot normally be observed directly.
Fortunately, as a practical matter there are nonarbitrary connections between form
and function, and in many cases the broader context can help us out. Different
ways of organizing data facilitate different operations on those data, and it might
be more or less plausible that anyone would need to consult a document in the
ways that are theoretically possible. Thus I will integrate ideas of function into the
“definitions.” That, however, will mean offering more elaborate specifications of
form and context to stand in as proxies for direct observation of function. This is
not, as I say, a way of giving a truer definition, nor do I suggest that people who use
the words differently in other contexts thereby fall into error. Rather, it is a way of
specifying the precise phenomena that I intend to investigate. I am interested, for in-
stance, in which table-like objects (in a broad, formal sense) serve the functions that
are most typically associated with tables. (Those “associations” will generally be
established by modern comparative and/or experimental research.) Which formally
map-like objects can serve map-related purposes? And so on for other key terms. In
most cases, as it happens, the result is that I end up focusing on a narrower range of
objects than might be expected, but occasionally (as with the case of maps) I will
be arguing on the same basis for taking a broader view than is sometimes taken.
For the reader’s convenience I conclude this introduction with a brief survey of
the individual chapters.
1. Lists. While simple lists are ubiquitous in the Roman world, this
chapter treats only specialized types: tables of contents, alphabetized
lists, indexed lists, and nested lists. Each of these is rare and is largely
or entirely restricted to quite specific contexts. For instance, tables of
contents are used only for miscellanies with little inherent structure,
nested lists for large and continuously expanding public records. (In
passing, I demonstrate that a number of proposed listing schemes,
such as the supposed numbering of theater seats, are mistaken.) These
sophisticated list-forms are used overwhelmingly to confirm or authorize
information rather than provide it in the first instance. Moreover, they
tend to arise in tandem with some physical process of collecting or
sorting data, not merely contemplation of data in the abstract.
2. Tables and Tabular Organization. Tables (in the narrow sense of
matrices with meaningful rows and columns) are also vanishingly rare.
The chapter contrasts a variety of areas in which they might have been
expected but are not in fact found (e.g., grammatical, arithmetical, or
calendrical tables) to the few where they are attested (e.g., centuriation
formae, military duty rosters). Tables tend to arise only in the context of
a combination of circumscribed expert communities of users/producers;
constitution rather than recording of data; contexts in which priority
is given to organization itself over content; and/or physical or enactive

“scaffolding” for cognitive activity. That is, the surprising difficulty of
building a table is eased by what cognitive scientists describe as the “off-
loading” of processing from purely symbolic reasoning to bodily action,
especially interaction with the physical world.
3. Weights and Measures. Serafina Cuomo has shown that at least
some Roman thinkers about measurement harbored universalizing
ambitions—systems of formal equivalences between numerical
quantities that would bypass the particularity of the various objects
being measured. I argue that this regime emphatically failed to
materialize. Measurement remained a (local) process, rather than a
universal “black box” whereby numbers were allowed to stand in for
objects. On inspection, many units of measurement turn out to be
rough approximations, proportions (rather than absolute values), or
not fully realized metonyms (e.g., a literal cup rather than a formally
defined volume which shares that name). Neither state regulation nor
the pressures of the market required adherence to universal standards,
and in fact they would have rendered the notion almost meaningless. In
turn, a wide variety of processes (in commerce, construction, recipes) are
designed to be locally rather than globally predictable.
4. Representing Three Dimensions. Roman painting, it is rightly agreed,
did not use single-point or other formal perspectival systems. It has
been influentially argued, then, that they lacked coherent spatial
representation altogether. However, examination of three related sets
of “sacro-idyllic” representations show that they deploy a variety of
spatializing tactics instead of a single system; for example, rough scaling
and recession, encoding of distance by overlapping and differential
shading, cuing by narrative content and by radically perspectival view of
certain kinds of objects, and the physical height/depth of plaster relief.
The combination of these features creates a network of connections at
a distance which gives three-dimensional structure to representations
even if they cannot be precisely or uniquely reconstructed. They are in
this more topological sense “spatial,” even if not narrowly Euclidean.
The fact that the sets of tactics overlap but are not the same in the
various cases suggests that spatiality is a goal in itself rather than an
epiphenomenon of other generic features.
5. Representing Two Dimensions. The chapter focuses on maps, but
also examines (as background phenomena) data graphics, textual
illustrations, and building plans. I consider issues such as reproducibility,
social status, and representational minimalism. The notions of spatial
representation developed in the previous chapter, combined with some
of the specific features of the secondary technologies listed earlier, allow
us to set expectations for what an ancient “map” should have looked like
8
and thus to collect evidence for their limited (but indisputable) patterns
of development and use. (The vast majority of the objects discussed in
this chapter operate in the titular two dimensions, but there are a few
exceptions.)
6. Conclusion. The first section attempts to tie together the content of the
main chapters, particularly in terms of the themes listed immediately
following. Then I sketch out several areas of related interest that I think
are ripe for further inquiry in light of the conclusions of this book.
My subject matter is large and complex, so it should not be surprising that there
is no monocausal explanation for the state of Roman information technology. As
a result none of the following themes plays a role in every chapter. Still, all recur
frequently, and the ensemble, I hope, binds the work together:
• Tolerance and context-specificity. Roman information systems (e.g.,

volumetric measurement, records of land survey, representations of three
dimensions) show in general a remarkable tolerance for error and/or
omission. The particular limits are a function of their use in the context
of preconceived target-tasks for which the information will be “good
enough” and without regard for other possible uses or users. Those tasks,
in turn, are themselves shaped so that poor-quality information suffices to
complete them.
• Recording vs. constitution. Representations can, of course, be used to
record facts which have some kind of prior existence (e.g., the list of
distances in an itinerary). But they can also be used to bring facts into
being (e.g., you cannot serve on a jury until your name appears in the
praetor’s album; eligible soldiers are not in fact discharged until the
relevant paperwork is completed). Romans are more comfortable in the
latter mode than in the former. Moreover, the understanding of many
cases hinges on recognition that what appears to be recording in fact
hides constitution. So, for instance, Roman tables of contents serve
poorly in a pure reference function, but make much more sense if they are
read to create segmentation of the main text.
• Text and paratext. Even in their most literary moments, Romans preferred
imagining texts (at least potentially) as speech acts. This makes many
informational devices (tables of contents, section numeration, tables,
illustrations) problematic, insofar as they are inherently paratextual
(and doubly so for items that cannot be linearized—e.g., the last two
examples). Problems of textual status discourage integration of formal
data records with discursive texts. Conversely, the most sophisticated
representations (fully realized tables, 3-D spatial representations) tend
to occur as stand-alone objects, rather than being embedded in larger
projects.
• Shared knowledge. These technologies require that producers and

consumers of individual representations (e.g., individual tables, maps,
etc.) share at least some knowledge of the technology itself. This is
particularly important to notice in cases where the general outline is
common knowledge (use of weights and measures), but crucial details
(objective values of those measures) are only knowable in extremely local
contexts. In practice, the required knowledge seems to have been restricted
to relatively small and isolated (if not always face-to-face) communities, a
limitation which is self-reinforcing.
As a consequence of all these interacting factors, Romans lived in a balkanized

informational world. Persons in different “locations”—whether those are geo-
graphical, social, or occupational—would have had access to quite different infor-
mational resources, and the overall situation is thus not controlled by the needs of
any particular class or group. Rather, there is a “tessellation” of contributions by
intellectuals, bureaucrats, and tradesmen (hence the “mosaic” of the title).
10
1}
Lists
Perhaps the most basic information technology made possible by writing is the list.
Lists were ubiquitous in the Roman world, and I do not intend anything like a full
account of them here. The scope of this chapter will be restricted in two ways. First,
in keeping with the idea of “high” technology, I will be talking not about simple
lists but about versions that have been augmented in various ways: ordered lists,
indexed lists, tables of contents, and nested lists. Second, even within most of those
categories, I will not attempt an exhaustive survey of either themes or materials but,
rather, make selected points.
When I say that lists were “made possible by writing,” a reasonable objection
might immediately be raised. Human beings have of course always talked about one
thing, then another, then another. That is not just a matter of, say, the catalogs that
figure prominently in the earliest poetry of Homer and Hesiod; it is the linear nature
of human speech. But I want to insist that there is nonetheless something distinc-
tive about writing down sequences of information. Here, I follow Goody’s classic
discussion of lists, tables, and formulas as specialized technologies of literacy. He
argues that writing transforms this oral discourse, giving the proper written list a
variety of distinctive features:
The list relies on discontinuity rather than continuity; it depends on physical

placement, on location; it can be read in different directions, both sideways
and downwards, up and down, as well as left and right; it has a clear cut be-
ginning and a precise end. (1977.81)
These physical features in turn give rise to various practical consequences such as
increased abstraction, increased attention to categorization, and an invitation to al-
ternative arrangement or processing of the items listed. Not every written list takes
advantage of all, or even any, of these opportunities, and—as often—the distinc-
tion could be framed as a spectrum rather than as a binary opposition. Still, Goody
has identified a distinctive set of written forms, and convincingly connected them to
important differences in use and power. Some of these advantages derive from liter-
10 ally having the kinds of written forms specified; others could perhaps attach to an
Lists } 11
oral list within a more broadly literate culture. My interest lies in texts that are not
merely sequential but which also exhibit at least some of Goody’s “written” charac-
teristics, whatever other specific additions (e.g., indexing, ordering) are the focus of
each individual section of the chapter. (This same set of observations about writing
down will be important to the next chapter, as well.)
Ordered Lists
I begin with what I will call “ordered lists.” Of course, all lists have an order; that
might even be considered their principal definitional feature (even if, as I mentioned
earlier, readers do not have to respect that order). What I mean here is much more
specific than that. “Ordered lists” will be shorthand for ones in which the order of
the text itself is meant to inform the reader. This can happen in two rather different
ways. On the one hand, a list can be organized by reference to properties of the
words (or other individual entries) themselves (“word order”). Typically this helps
a reader find out whether an item is on the list at all (sometimes to be used as an
index). On the other hand, a list can be ordered in terms of some feature of the real-
world entities that are referred to by its entries (“topic order”). This can in theory
be used in the same way, but more often it is meant to convey something about that
feature—for example, “what is the fourth most populous city in Italy?”
In principle, word order could appeal to the length of the words or some other
feature, but in practice it normally means alphabetical order. Unlike the topics of
the rest of this chapter and the next, alphabetization in the ancient world has al-
ready been the subject of an extensive and valuable basic study. In addition to var-
ious specific points which I will note individually, the basic narrative of the next
two paragraphs will rely heavily on Daly 1967. With this framework in place it will
be easy to move quickly to my own analytical remarks and in particular to some
observations that will foreshadow many of the important themes of the rest of
the book.
Latin lists in deliberate alphabetical order appear to go back at least as far as a
joke in a play by Plautus—that is, the earliest extended Latin texts we have today
(c. 200 bce)—and continue through to the end of our period. (There are somewhat
earlier Greek examples, and Plautus may well be adapting one such in his text.) The
earliest examples, and perhaps the bulk of all those of our period, are alphabetized
only with respect to the first letter of each word. Occasionally we see alphabetiza-
tion through the second or later letters. (Full alphabetization appears occasionally
in Greek, but seemingly not in Latin until after our period.) There is not, however,
an overall trend over time toward more detailed arrangement.
There seems to be something ordinary or matter of fact about Roman use of
alphabetization. The joke in Plautus’ play does not call explicit attention to the
ordering. Rather, a character starts reeling off a list of (otherwise unremark-
able) names in alphabetical order. The humor seems to lie in having the audience
12
recognize on the basis of form that the list threatens to extend to the end of
the alphabet (Asin. 864–6). It is unclear what fraction of the audience could
have gotten the joke—we do not really know the fraction of the population that
would have had the low requisite level of literacy—but the fact that Plautus took
the trouble to make a somewhat elaborate joke anyway suggests that many would
have. It’s also true that authors who use alphabetical order either do not bother
to explain what they are doing or use only a summary phrase to do so (“ar-
ranged by letters,” “in the order of letters,” etc.). This is in contrast to the use of
some other devices we will see later. Alphabetization is also a common feature
of school exercises, so it is not hard to see why it might be familiar to the general
(literate) public.1
At the same time, as Daly has noted, it appears to be used principally in contexts
which are anything but general or ordinary. He points out a number of schol-
arly uses; I would add only the case of Flavius Caper (Verb. Dub. 107–12K)—no
later than Daly’s earliest example—whose brief list of morphologically “doubtful
words” is transparently in alphabetical order. Daly also notes administrative uses.
He finds their number less than impressive, and that is a fair judgment, but I would
note that subsequent scholarship (not focusing on alphabetization as such) has ex-
tended the roster of examples at least a bit.2 It is still not common in either realm,
particularly in “outward-facing” texts—that is, those aimed at a reading public dif-
ferent from the producer(s).
I am inclined to explain the gap between high accessibility and low actual use
by another phenomenon that Daly 1967.52 had already noted: alphabetization
was inelegant. The contexts in which it typically appears (education, handbooks,
internal state documents) are all relentlessly utilitarian. The two texts of some pre-
tension which nonetheless use alphabetical order resort to it only in residual cases;
it is how “the other” items in various categories are listed once more substantive
categorization has been exhausted (Var. RR 1.1.9; Plin. HN 3.46, 26.164, 37.138).
And at one point Pliny even apologizes as he begins a shorter group by noting
that he is only alphabetizing peoples “who do not warrant closer attention.”3 Daly
notes in this connection that the agricultural writer Columella, in borrowing two
alphabetized lists of authors from Varro, re-sorts one by birthplace and appears to
invert two persons in the other simply to avoid alphabetical order.4 There is also a
well-known passage in the Aeneid in which Italian heroes are listed in alphabetical
1
Cribiore 1996.161–7. Frampton 2018.62–70.
2
Nicolet 1991.135, 173–7; Salway 2012.200–2. Tim Parkin points out to me that the alphabetical list
of centenarians from Italy’s regio VIII preserved in Phlegon, Makrobioi (taken with Pliny the Elder’s
access to a similar data set; HN 7.162–4) could reflect some government document ordered on the
same principle, though (as he also points out) alphabetization by praenomina seems very odd in such a
context.
3
Plin. HN 3.130, quos scrupulosius dicere non attineat.
4
Perhaps also relevant here is the frequently made observation that Latin authors who receive a list
in Greek alphabetical order often fail to rearrange it to maintain that order in Latin.
Lists } 13
order except for the intrusion of Messapus between Caeculus and Clausus. O’Hara
1989, however, has shown that Vergil tells Messapus’ story in such a way as to as-
similate him to another well-known hero, Cycnus. Cycnus’ name would be in cor-
rect first-letter position here. Vergil, too, seems to be going out of his way to avoid
alphabetization.
There is also an exception which may prove the rule. The Augustan-era scholar
Verrius Flaccus wrote an extensive lexicon, perhaps closer to a modern encyclo-
pedia than a dictionary. Our preserved versions have gone through some kind of ed-
iting at the hands of Festus (perhaps second century; very fragmentarily preserved),
and large parts survive only via the subsequent epitome of Paul the Deacon (eighth
century), but it seems likely that the basic structure is Verrius’. The work is organ-
ized by first-letter alphabetization of the head-words, and within each letter group
there is a first section that is further alphabetized and a second one which is grouped
more or less thematically. Glinister (2007.23–4, 29–32) has made a strong argu-
ment that alphabetization is not a pragmatic choice here. The work has no cross-
references, and many discussions are attached to lemmata that are not necessarily
obvious. The innocent reader would find it very hard to track down a particular bit
of information, and even the expert might be daunted. Verrius, Glinister argues,
uses alphabetization’s blandness to assert intellectual authority, the authority to
distribute knowledge according to a system that is superficially objective, but in fact
quite idiosyncratic.
There are several patterns here that we will see throughout the book. First, al-
phabetization demonstrably exists in the Roman world, but its use and diffusion is
perhaps surprisingly limited. This quantitative phenomenon also has a qualitative
parallel. Romans do develop more “sophisticated” or powerful or (to put it less
prejudicially) intensive alphabetization, but the more complex forms do little or
nothing to displace the simpler ones. One should never cite technological progress
as an explanation for anything, but from a modern perspective it is probably useful
to point out that progress does not work even as a description here. In particular,
in the variation between one-letter, two-letter, and so on alphabetization, we see
a Roman tendency not to apply any more technology than is required by a par-
ticular situation. Second, the development is generally sparse, but also clustered.
Alphabetization is relatively common in a few contexts and unknown in others
where it could have been used. Additionally, I have mentioned an apparently aes-
thetic objection to alphabetization, and we will see other cases of resistance to for-
malism in data organization.
The last pattern is slightly more complicated. Small 1997.63– 5 has
hypothesized that the rarity of alphabetization is due (in part) to the instability
at the time of the notion of the alphabet itself. Following her, Glinister 2007.22
has gone so far as to say alphabetization “is not an obviously useful means of
arranging data in a world where ‘alphabetical order’ may fluctuate, where there
appears to be no concept of the alphabet as a distinct entity.” Subsequent work
by Frampton 2018.67–70 has shown that the premise is false—the alphabet was
14
very much reified by the period we are talking about—but in any case, the evi-
dence cited here for the ordinariness of alphabetization tells against such a deep
cognitive explanation of the weakness of the technology.5 Cognitive factors
will be significant throughout the book, but will rarely give simple, categorical
answers to our questions. Here we should favor shallower and perhaps more di-
verse explanations.
The other approach to ordered lists is by sorting the topics, not the words. In
principle, the potential variety of topic orders is infinite; one could imagine, for
instance, lists based on the date, location, or size of the items listed. There are
many such lists surviving, at least at first glance, but it is often difficult to know
that the objective order is meant to have (or in fact had) any informational value
for the reader rather than being an artifact of composition.6 For instance, the list
of canonical authors recommended by the rhetorician Quintilian (late first cen-
tury) is at least roughly in chronological order within each genre (10.1.46–131), but
there are no absolute dates or even (for the most part) indications of gaps in time.7
Moreover, chronology applies only within genres, and those generic categories are
in turn subordinated to a division between Greek and Latin texts. Chronology may
have helped Quintilian pick an order, but it is so far down the organizational hier-
archy that it could only help a reader find an author that she is nearly on top of
already. Or take the lists of cities on the pocket sundials recently studied by Talbert
2017.147–52, which instruct the user how to calibrate the timepiece for use at the
local latitude. Many show rough clumping of nearby places, and some may reflect
itineraries, but do not appear to follow these tendencies precisely, which means that
the order is not in fact predictable. Nor is it clear what value the sundial reader
would get even from knowing a correct order. All she really needs to know is her
current position. For an author, though, ordering even a fairly simple list might
make sense as a check for completeness.
In the two cases just mentioned, there is good reason to think of the lists as not
being ordered in my restricted sense, but more often there is little clear reason to
guess one way or the other. However, I can suggest two sets of circumstances in
which order is more likely than not to be important. On the one hand, we have a
number of kinds of practical records such as financial accounts, rosters of mili-
tary units (which are arranged by rank, then by year of enlistment), and various
5
More plausible is the notion (Glinister 2007.22–3) that the syllable was more prominent in the
Roman mind, and so may have made limited alphabetization more natural (cf. Frampton 2018.63–4 on
the status of syllables). However, the order of syllables seems still to depend on the order of the compo-
nent letters.
6
Cf. Riggsby 2007.96–8 for structure in HN that seems to be a residue of Pliny’s methods of compo-
sition, likely invisible to his original readers, who had neither the modern paratextual apparatus we do
nor (I suspect) professional interest in reverse-engineering Pliny’s working methods.
7
It might fairly be objected that what we have in Quintilian’s text is not really a list, since it fails
in most respects to show the Goody features described a few pages ago. I imagine that Quintilian was
working from an outline in the form of a much rawer list. If the end product was not really a list itself,
that just suggests all the more strongly that the ordering was meant for the author not the readers.
Lists } 15
government registers. The registers record, for instance, births in the order received
and note both the date and the position of the record within the sequential register.
We will see all these document types again soon, but let me note here one feature
they share (which itself will also be important later). In all of them the chrono-
logical order seems to be generated automatically, as the records in question are
compiled over time. That is, for instance, a small business owner might list pay for
his employees for a given day, then repeat the process on the same piece of papyrus
the next day and the next, and so on.8 The fact that the individual records already
in order in these cases are tagged with the date in question suggests that the order
of composition was deliberately being harnessed to provide the user with chrono-
logical information.
On a grander scale, the various state annals seem to have been constructed in the
same way, with various kind of information (e.g., lists of prodigies) recorded in order
at least yearly.9 I separate these out because they point, at least by association and
perhaps also causally, to the second set of contexts in which pointedly ordered lists
are used. These are works of what might broadly be called scholarship. Most obvi-
ously there is Rome’s entire tradition of annalistic history, but we might also look at
Cicero’s Brutus, which writes the history of Roman oratory mainly through a series of
capsule biographies arranged in generations of orators. Unlike Quintilian, Cicero ex-
plicitly points to his chronological arrangement, and he uses it to make a point about
what he sees as progress over time. In a less literary vein, we can point to the lists
of officeholders, military triumphs, and calendrical information Romans carved into
stone and which go under the cover term Fasti.10 We do not generally know who com-
posed these, but the one exception (Fasti inscribed at Palestrina and composed by the
first-century bce scholar Verrius Flaccus11) is likely representative. The people who use
topic-ordered lists, then, look like roughly the same people as who use alphabetical
lists. They are scholars and bureaucrats who frequently deal with information as such
(without necessarily intervening in the management of whatever reality the data de-
scribe) and who interact intensively with each other more than speaking to a broader
public.
Indexed Lists
“Indexed lists” might include in principle any whose members are matched one to
one with a set of indices—numbers, letters, colors, and so on—to help find material
8
For a specific example, see chapter 2, note 46.
9
For the Annales Maximi, see Cic. de Or. 2.52 and DServ. Aen. 1.373, with Frier 1979. For other
magisterial (and thus annual) records, see the references cited at Riggsby 2006.134. Most or all of these
will have been internal rather than published records (Riggsby 2006.140, 149–50).
10
On Fasti, see further at chapter 2 “Appearance: Fasti”.
11
Suet. G&R 17.
16
within the list. As with the ordered lists, however, this type is dominated by a single
sub-type: the numbered list.12
The fact that labels are almost always based on numerical order creates a poten-
tial evidentiary problem when we do not have the entire list in hand. If we have a
passing mention of, say, the fifth item on the list, does that mean that the whole list
was fitted with numerical references (and so is a truly indexed list) or did someone
just happen to count forward from the beginning on a particular occasion?13 This
distinction will often not be a real problem, but it will be significant in at least a few
cases that will come up in what follows. Thus I want to stipulate that for the rest of
this section I mean only lists with built-in indices, not ones generated on the fly and
partially by a later user.
Compared with most of the other technologies discussed in this book, numbered
lists are fairly common. Here are some contexts in which they are used (most of
these will be discussed at more length at various points in the book):
• sections of archives
• segments of buildings
• administrative regions of Rome and of Italy
• military units of various sizes
• milestones along roads
• lots
• units of time
• segments of texts
In what follows I want to argue for the existence of two limitations on num-
bered lists. One limitation is (like the items in the list) essentially a matter of genre
or use-context. The other is more functional, a way that numbering of lists is not
exploited, even when it exists.
12
Two (or three) exceptions: the Roman market (nundinae) cycle divided an eight-day “week” into
days labeled A–H; a work by Casellius Vindex (cited by Charisius 254.18, 312.24B) was arranged not
in books 1, 2, etc., but A, B, though here the letters appear to be literal; each book discusses words
that begin with the letter in question. Years were often identified by the names of (at Rome) the year’s
consuls or (elsewhere) various other “eponymous” magistrates; Julius Obsequens’ book of prodigies is
essentially a list structured in this way. But the unpredictability of these names, however, means that
they cannot be used as an index in the same way as numbers or letters.
13
There are also at least a few cases that turn out to be neither. Roman legions were conventionally
designated by numbers. At various points in the early empire there were two fifth legions, two sixth,
two tenth, and (it appears) three third legions (generally distinguished by further epithets); the numbers
are clearly not generated by counting, but because they are often repeated, they do not really work as
indices at all. Also, probably, the tokens known today as spintriae. These feature large numerals on one
side and erotic scenes or imperial portraits on the other. It has been hypothesized that they correspond
in various ways to a catalog of sexual positions or the like, but Campana 2009 makes a strong case
for regarding them as game tokens, with the numbers perhaps representing the value of the individual
piece. The same is likely the case for similar tokens which match numerals on one side with counting
gestures on the other (Williams and Williams 1995.590–1).
Lists } 17
The last item on the foregoing list was “segments of text.” (I mean here indi-
vidual texts; archives will come up later.) By the time we have longer surviving lit-
erary texts in Latin, they are divided by their authors into “books”—that is to say,
units that will notionally fit on a single papyrus roll—which are then numbered.14
(I use “literary” here in a broad sense common in the Classics—texts which were
meant to be copied and circulated to an indefinite audience, not just those of some
artistic ambition.) Given this nearly universal convention, it is then striking that
smaller units of Latin texts are seldom if ever numbered. In principle, several such
units were available. Today we refer to bits of Latin poetry by line number and (in
the case of collections of short poems) the number of a given poem within its book.
There seems to be no evidence for either practice in antiquity, whether directly in
manuscripts or in external reference (say, in commentaries). This is despite the fact
that lines were often counted in individual manuscript copies, apparently as part of
the pricing process.15
Prose works were often thought of as being divided into capita, a word
often translated as “chapters,” but perhaps closer to English “sections” or even
“paragraphs” (there is a related, perhaps original, sense of “high points”).16 Not
only do we have frequent reference to the general idea, but there is even some man-
uscript evidence for the marking out of such sections with ekthesis (reverse inden-
tation) or even the letter k (for kaput, as the word was often spelled).17 Crucially,
however, none of these divisions appears to come with reference numbers, not even
in prose texts explicitly divided into units by rubrics. Latin also has a word pagina
which can mean the same thing as the English word we derive from it: page. Before
the rise of the codex, however, it typically refers to a field within a larger support
medium—say, a column of writing on an inscription or roll of papyrus. In no sense,
however, are paginae used in our period as the basis for standard numeration of lit-
erary texts (we will see later that the case of archives is somewhat different).18
The lack of numerical references (other than book numbers) in commentary is
particularly striking.19 The genre was a flourishing one already in antiquity, with
subject texts including Vergil, Cicero, Terence, Homer, and others. In its modern
form, the commentary is typically organized by numeration first (line numbers
for verse, section/chapter numbers for prose), then by rubric within any given nu-
merical range. Ancient commentaries are generally organized by rubric alone. In
14
Moatti 1997.223.
15
Birt 1882.159–209; Hall 1913.9, 13.
16
On the terminology, see Butler 2009.16–7, 20.
17
Butler 2009.10–5, 18–9, 21–3.
18
Grafton 1997.30 cites references to page numbers in an apparently standardized edition of certain
legal texts, but this dates from the fifth century. Shafer 2017 has argued that certain effects in Vergil’s
Georgics presuppose that certain words and phrases will fall at predictable points on the “page,” though
for poetry this would only have required an early edition with a set number of lines to the page.
19
For possible (but still extremely scanty) Greek examples, see Barnes 2015.348–50, Mansfeld and
Runia 2009.199.
18
continuous reading, there is little difference, but if you pick up a literary text in
the middle, it is much harder to find your place in an ancient commentary than
in a modern one. For comparison, it may be useful to look at one very partial
exception to the usual ancient practice. In the manuscripts of Asconius’ histor-
ical commentaries on several speeches of Cicero (first century), we have frequent
(though not systematic) allusion to the position of his lemmata within Cicero’s text:
around line 640 (10.10C)

a little later (11.1)
around line 800 (11.8)
around the middle (11.19)
around the middle (12.7)
****
around the third part (23.21)
around 160 lines from the last (53.5)
This seems to be unique among commentaries. In fact, it has even been suggested
that these headings were not part of the original text but, rather, were introduced by
a later reader for his own benefit, then absorbed into the manuscript tradition. I will
return to the reasons for this proposal in a minute. Whatever their source, these
references obviously do not refer to standardized numerical labels. They are vague,
they vary choice of units (lines, “parts,” none at all), and they count in whichever
direction is locally convenient.
The same approach seems to be visible in a text which is not literary even in the
broad sense used here but also which bears on the more general issues of both this
section and the last section of this chapter. In the early second century, a man named
Vesbinus erected an inscription at Caere which recorded acts of the municipal gov-
ernment and subsequent correspondence with an imperial official authorizing him
to set up a shrine. His inscription quotes extensively and with citations from the
records (commentarius) of the municipality. The sections with the reference (as they
are usually translated) are as follows:
in which [commentarius] had been written this which is written below: In the
consulship of L. Publilius Celsus and C. Clodius Cripinus [i.e., 113], on the
Ides of April, in the [local] dictatorship of M. Pontius Celsus and aedile-
ship of C. Suetonius Claudianus, the commentarius of the municipality of
the people of Caere. From there (inde) page 27, sixth chapter. [provisional
approval of the project subject to higher authority] From there page two, first
chapter. [Letter to the relevant imperial official] From there page eight, first
chapter. [Reply from that official] (CIL 11.3614.6–18)
In all three cases, the reference is to page and chapter. Birt 1882.158 takes the page
and chapter numerations as running independently, but in parallel, so pages 2 and
8 would fall within c hapter 1, while page 26 lay somewhere in c hapter 6. This hy-
pothesis ignores the fact that the three documents—the initial decree calling for
Lists } 19
(inter alia) the approval from the curator, the letter making this request, and the
curator’s reply—are clearly in chronological order, as they would have been in the
commentarius which is a running record.20 Birt’s scheme makes the initial decree
the last of the three to appear in the commentarius. In fact, the same objection still
applies even if we take pages to be individually broken down into chapters. I sug-
gest that the page references have wrongly been translated as “page 27/2/8.” They
should instead be read as “the 27th/next/eighth page” from the previous citation
(an equally possible reading of the Latin). That is, inde (“from there”) doesn’t refer
generally (and thus redundantly) to “from the commentary” but, more specifically,
“from the last page mentioned.” The locations of the second and third documents
are given by counting forward from the first and second documents, respectively.
The first instance appears to count from a more public reference point marked out
by the list of eponymous officials (perhaps some important date in the civic year).
Note also that this material is placed after “what is written below” and so looks to
be itself a quotation from the record. Moreover, the first date quoted (13 April) is
very early if we take it to be the time of the original council decree, since the letter
they ordered dispatched was not sent until 13 August (line 15). Rather, it is a ref-
erence point considerably before that decree, as represented by the twenty-six-page
gap. It does not look as if this record had fixed numbers attached to its paginae, just
ad hoc counting. In principle the capita could have been numbered, but that seems
much less likely if there was no indexing at the higher level. This is made possible by
the fact that there was, seemingly, only a single exemplar of the text.
All that said, I know of two circumstances in which texts do have numbered
subsections: one in a particular text and the other an entire category. First, in the
mid-first century, Scribonius Largus wrote a collection of recipes for 271 medicines.
The individual recipes are numbered, and the opening table of contents uses this
numeration to direct the reader from symptoms to their proper remedy. I will say
more about why this might be at the end of this section, after the other example
and some of its implications. Second, though arguably not literary texts even in
the broad sense I have offered, Roman statutes were often divided into numbered
capita.21 For instance, Cicero takes for granted that he can refer to these capita by
number even when speaking of a legislative proposal not yet passed:
Why does it matter in chapter 3 you require the ratification of officials by the
passage of a law by the curiae, when in chapter 4 the elected are given the
same powers even without such a law. (Leg. agr. 2.29)
20
Riggsby 2006.138.
21
The tablets on which a few statutes are inscribed seem to be numbered, but this is almost certain
to be for local use in placing the tablets, not a general reference system. The Flavian municipal law does
not even show consistent formatting at lower levels, so it is unlikely the system assumed uniform carving
(González 1986). In the enabling law for Gallia Cisalpina (CIL 11.1146) the (apparent) tablet numbers
do not respect breaks in the chapter numbering. Cf. Butler 2009.16–21.
20
Similar references appear sporadically in inscriptions and in the later juristic lit-
erature, and there is plausible reason to think that such numeration had at least
appeared centuries earlier as illustrated by the three numbered capita of the lex
Aquilia.22
Note, however, that in the Cicero passage I have just quoted, the reference to
chapter numbers is not strictly necessary, since the relevant material from the cited
source is paraphrased anyway. Looking up the proposal Cicero refers to could con-
firm what he says, but that appears to offer authority rather than information. This
illustrates the second, functional limitation I alluded to earlier. There is a strong
Roman norm against what I’ll call “obligatory cross-reference.” That is, they do not
like to force readers to look up a second text to be able to understand what they are
reading in the first. Here is a modern example of an obligatory cross-reference of
the sort I contend the Romans avoid:
AMERICAN DOCUMENTATION OF VESSEL REQUIRED. If 18

U.S.C. Section 1082 is repealed, the affirmative defenses provided by Section
47.09(b) apply only if the vessel is documented under the laws of the United
States. (TxPC §47.10)
This means nothing without consulting other sections of both the Texas penal code
and the U.S. federal code. This is entirely normal legislative drafting. Here is a
Roman example, which I suggest is extraordinary:
if there is no judgment within the time which is foreseen by chapter 12 of the

recent lex Iulia on civil trials or the decrees of the Senate pertaining to this
chapter. (lex Irnitana 91)
There are several points to be made about the Roman example. First, it is the
only obligatory cross-reference to a numbered chapter in the surviving remains
of Roman statute law. And I have not yet found another example in the ancient
legal literature more generally. Again, a reference to provide authority is common
enough, but not ones that must be tracked down to figure out what the immediate
passage is saying. Second, the very text in which this reference appears would itself
be difficult to refer to in the same way. It is an instance of the so-called Flavian mu-
nicipal law. The latter is a template for town charters in late first century ce Spain.23
We have fragmentary instances of laws based on the same template from other
towns in which the chapters are numbered. At Irni, however, the only place where
this quoted passage is preserved, the chapters of the law are identified by rubric, not
by number.24 In fact, as a general matter, the chapters of inscribed laws are more
often marked off by summary rubrics, by spacing, or by nothing at all rather than
22
Gaius 3.210, 215, 217; Rotondi 1966[1912].241–2.
23
González 1986.
24
González 1986.148.
Another random document with
no related content on Scribd:
against him. By way of revenge, they could think of nothing better
than to accuse him of being himself guilty of the very offences
against which he had set his face.
“They are trying to make out that I am the owner of ships
which I use for the purpose of conveying this false paper-
money to Brittany,” he writes to Lady Atkyns. “They have
stated this to the Government. Fortunately, my whole conduct
and reputation, and all that I have done to destroy this
shameful traffic, serve to show the improbability of such
accusations.”
But, in spite of all his energy and determination, Cormier’s
enemies were too strong for him. It was in vain that he demonstrated
his good faith. Calumny had done its work.
The British Government had decided, in concert with the Comte
d’Artois, to send an important mission to the Netherlands, with a
view, doubtless, to establishing relations with the Stadtholder, whose
position was becoming critical owing to the sequel to the Revolution.
The man to be entrusted with this mission would have to be some
one who had given proof of his qualifications. Cormier seemed cut
out for the post, and he stood in readiness for it, enjoying the
prospect of thus getting into touch again with France, and of perhaps
being able to serve the interests that were so dear to him. But he
had reckoned without his foes. Their efforts were redoubled, and in
the course of November Cormier learnt that another had been
entrusted with the mission. His anger and disappointment can be
imagined. He decided that, in spite of all, he would leave England
and betake himself to Holland on private business. Doubtless he
imagined also that it would be an advantage to be near the French
frontier, and that he would be the better able to follow the course of
events at the Temple. It was a risky step to take, for there was
nothing to guarantee his complete security in the Netherlands.
However that might be, his decision was taken, and on November
25, 1794, Baron d’Auerweck wrote to Lady Atkyns to acquaint her
with the news of Cormier’s departure, conveying to her at the same
time many apologies for his having himself neglected to write to her
to take farewell. During the months that follow the “little baron”
replaces the Breton magistrate as principal correspondent of Lady
Atkyns.
It is a strange personality that stands revealed in these letters of
Baron d’Auerweck. Keen and resourceful, the baron lays himself out
to exploit to the utmost the valuable friendship of the English lady,
thus bequeathed to him, as it were, by Cormier. Trained by Peltier,
d’Auerweck seems to have modelled himself upon his master, and to
have become in his turn the accomplished publicist, plausible, fluent,
supple, with a gift of raillery and sarcasm, together with a turn for
philosophy. Lady Atkyns, though not unappreciative of his copious
epistles, shows clearly that she estimates him at his real value, and
is careful not to take him too much into her confidence. It must be
enough for him to know that there is still reason to hope that the
Dauphin may be saved. D’Auerweck himself is not in a position to
give her much information in return. His letters consist rather of a
bright and lively commentary upon the political situation and the
course of events generally in France.
Upon Cormier’s decision to leave England the Baron expresses
himself in downright language, and makes it a text for a disquisition
upon his elder’s character.
“Cormier’s departure has disturbed me a good deal,” he
writes to Lady Atkyns, “the more so that, with a little
prudence, he could have spared himself this unpleasantness,
and might have succeeded in getting what he wanted. A man
who has passed his whole life in the magistracy ought, at the
age of fifty-six, to know something about men, but Cormier
has never got further than the A B C of such knowledge. I
have had some rather hot disputes with him over his rash
confidence, his purposeless explosions, his sudden
friendships that ended in ruptures, thus increasing the number
of his enemies.... But we both of us felt the parting. I must do
him the justice of admitting that there is a lot of kindness and
sympathy in his character. I think he has the same feeling of
friendship for me that I have for him. It is my wish to serve him
whenever the opportunity may arrive.”
By an unfortunate coincidence, the political situation in Holland
was undergoing a disquieting change at the moment of Cormier’s
arrival. Until then England had exercised a decisive influence there,
both by reason of the presence of her army and through counsels of
the Stadtholder. But in the autumn of 1794 a popular feeling in favour
of the Revolution began to make itself felt, fanned by the hostility
aroused against the undisciplined English troops, with their looting
and pillaging, and intensified by an unlooked-for piece of news: the
French, led by Pichegru, had crossed the frontier and were
advancing by long marches, and seizing all the places they passed
through on their way. In a few weeks the power of the Stadtholder
would have gone! Though clothed in rags, the soldiers of the
National Convention were welcomed with transports of delight.
Never did troops show such discipline, it should be added.
But Pichegru was not alone. Beside him marched representatives
of the Convention, eager to institute in the United Provinces the
principles of the Revolution and to establish the guarantees of order
and security inseparable therefrom.
Therein lay the danger for those who, like Cormier, were to be
found in flagrante delicto of emigration. On November 8, 1794, an
order came from the Committee of Public Safety to the
representatives with the army, commanding them to seize the
Stadtholder, together with his wife and children, as well as to arrest
immediately all émigrés who might fall into their hands.
Knowledge of this important decree had not come to London on
December 15, for on this date we find d’Auerweck writing to Lady
Atkyns that he has had news of Cormier, “who is now at La Haye in
good health and spirits.”
The extreme cold which prevailed this year contributed in a
remarkable degree, as is well known, to the success of Pichegru’s
operations in Holland. Shut in by the ice, the powerless fleet was
obliged to surrender to the French cavalry—a memorable incident in
the military annals of the Republic. The famous dams, which were to
be opened and to flood the country and submerge the French,
became useless by reason of the frost. In short, Pichegru triumphed
throughout. He made his entry into Amsterdam on January 10, 1795,
and eight days later the Stadtholder embarked for England. The
Dutch Republic had come into being.
Cormier’s fate throughout this period must have been a matter for
anxiety to Lady Atkyns, but the absence of anything in the shape of
definite news from Paris as to the state of things at the Temple
continued to be to her a source of far greater disquietude. The vague
assurance as to the Dauphin’s well-being, which d’Auerweck
transmitted to her from time to time, counted for nothing, as she
knew herself to be better informed as to what had been under way.
What had been happening? A third letter, addressed by Laurent to
his correspondent, under date of March 3, 1795, enlightens us a
little:—
“Our little mute has now been smuggled away into the
palace of the Temple and well concealed. There he will
remain, and if need be can be passed off as the Dauphin. The
triumph is altogether yours, general. You can now be quite at
ease in your mind—send me your orders and I shall carry
them out. Lasne will take my place now as soon as he likes.
The best and safest steps have been taken to ensure the
Dauphin’s safety. Consequently I shall be able to get to you in
a few days, and shall be able to tell you all further details
orally.”
These lines herald a momentous alteration in the régime of the
prison. First of all, there is the question of Laurent’s leaving it.
Presumably his presence is no longer needed there. This suggests
that success is assured. And Lasne—how is it that his name makes
its appearance here for the first time? We shall find him declaring in
1834 that his service in the Temple began in Fructidor year II., that is
to say, between August 18 and September 16, 1794.[75] In that case
Laurent would have had him as his colleague for several months
already! The Temple documents preserved in the National Archives,
and examined fifteen years later, establish the fact that Lasne did
not, indeed, enter upon his duties until March 31, 1795, thus bearing
out the accuracy of Laurent’s statement.
We see, then, that the little mute has been transferred to the
palace of the Temple—that is to say, into one of the many empty
suites in the great maze of buildings that surrounded the Tower. Here
he has been, or perhaps will soon be, joined by the Dauphin himself,
for means of retreat from this labyrinth of buildings are infinitely
greater than from the fourth storey of the Tower.
To replace the mute, another substitute has been found, a
scrofulous boy who may be expected soon to die. All barriers to the
Dauphin’s escape will thus be removed. So much we gather from
Laurent, and all his statements are borne out by documents which
have been left by Royalist agents.
This second substitution effected, Laurent was able to quit his post
with an easy mind, and we find that he did actually leave the Temple
on March 29, 1795. His successor, Lasne, arrives two days later.
Gomin, who perhaps knows part of the truth through Laurent (and,
moreover, his rôle is more especially to attend to Marie Thérèse), is
careful not to confide in him, knowing well the risk he would run by
so doing. Lasne finds in the prison a boy who is evidently very ill, in
great suffering, whose death is soon to be expected. What would be
the use of asking questions? It is enough for him to attend to the
child as best he may during the few weeks of life that still remain to
him.
Spring had passed and June had arrived before Lady Atkyns was
again to see the familiar handwriting, rounded and minute, of her
friend the Breton magistrate. The letter bore the postmark of
Hamburg. What was Cormier doing on the banks of the Elbe? He
would seem to have had some perilous adventures. Probably he had
been arrested as an émigré and had escaped the guillotine by some
happy chance. However that may be, the news he had to tell of
events in France came as a great relief to his correspondent.
“We have been better served, my dear friend, than we
ourselves arranged. Our agents have not kept to our plan, but
they have done wisely.... But we must have patience. Things
are in such a condition at present that they can be neither
hastened nor delayed. A false move might have very bad
results.”
Within a week of the arrival of this letter, an announcement, that
came to many as a surprise, found its way round London. It was
officially reported that the Dauphin had died in prison on June 8,
1795. Had not Cormier’s assurances come in time to buoy her up, so
categorical a statement might well have given Lady Atkyns a severe
shock. She knew now, however, that it could not be of her boy that
there was question.
Some weeks pass in silence, and Lady Atkyns, impatient for news,
urges the “little baron” to set out for Hamburg. He starts in the first
week of July, but is delayed at Ocfordnese, whence he writes to her
on the 16th. At last he reaches his destination, but means of
communication are so uncertain that several more weeks elapse
before she hears anything further. September finds d’Auerweck
returning to London with a letter from Cormier to Lady Atkyns. In
October, again unable to curb her anxiety, she had just decided to
send d’Auerweck to Paris, when, to her deep grief and dismay, she
learnt suddenly from Cormier that everything had gone wrong—that
“they had all been deceived, shamefully deceived.” The child that
had died on June 8 was, indeed, the second substitute, and the
Dauphin had undoubtedly escaped, but others had got possession of
him, and the boy handed over to Lady Atkyns’ agents was the young
mute.
“Yes,” he writes, “we have been taken in totally and
completely. That is quite certain. But how have they managed
to do it? And did we take every step that could be taken to
make this impossible? These are matters you will want me to
go into in detail, and I shall not fail to do so; but I must wait
until I have time to trace the sequence of events from a diary
day by day for a year past. The entries for the first two months
are missing for the present—the least interesting period
certainly, since down to that time, and for several months
afterwards, only the project of carrying off the Dauphin was
being kept in view, the project which had to be abandoned
afterwards in favour of another which seemed simpler and
more feasible, as well as less perilous.”
Cormier’s long letter left Lady Atkyns completely in the dark as to
what exactly had happened. They had been tricked somehow—that
was all she knew.
To us, as to her, the names of most of the many participants in this
mysterious intrigue remain unknown. Laurent went off to San
Domingo in the following year, where he died on August 22, 1807.
Gomin, to some extent his accomplice in the matter of the
substitution, followed Marie Antoinette’s daughter to Austria, and
was careful to keep what he knew to himself. As for our three
friends, Cormier, Frotté, and d’Auerweck, we shall learn presently
the reasons for their silence.
The one person who has tried to clear up the obscurity of these
happenings inside the Temple is the wife of the bootmaker, Antoine
Simon, the Dauphin’s first warder. Considerations of space prevent
us from entering here upon any detailed examination of her
evidence, but we must not pass it by without a word. Mme. Simon,
after her husband’s death during the Reign of Terror—he was
guillotined in Thermidor—withdrew to the asylum for incurables in
the Rue de Sèvres, where she was to spend the remainder of her
existence. Here she was heard on many occasions to assert that she
was convinced the Dauphin was alive, having seen him carried off
when she and her husband were leaving the Temple, on the evening
of January 19, 1794. If this were true, it would result that that child
looked after by Laurent was not the Dauphin at all! This does not fit
in with the version that we have put together from Laurent’s own
letters and the various other documents which we have been able to
examine. But even if it were true, the poignant question would still
call for an answer—what became of the young Dauphin after his
escape? Into whose hands did he fall?
FOOTNOTES:
[75] His deposition at the Richemont trial.—Provins.
CHAPTER VI
THE FRIENDS OF LADY ATKYNS
What was the Chevalier de Frotté doing all this time? What steps
was he taking towards the realization of what he had called so often
the goal of his life, and towards the execution of the promises he had
made with so much ardour and enthusiasm?
Transported with joy on hearing that the British Government at last
contemplated listening to his projects and sending him to Normandy,
Frotté, when leaving London, betook himself with four comrades-in-
arms to Jersey—the great rendezvous at that time for the insurgents
engaged in dangerous enterprises on the Continent, and seeking to
find landing-places on the French coast.
It was the middle of winter—snow was falling heavily, and there
were strong winds. Several weeks passed, during which the patience
of our émigrés was severely taxed. Nothing was more difficult than to
effect a landing in Normandy under such conditions. Apart from the
difficulty of finding a vessel to make the crossing, it was necessary to
choose some spot where they might succeed in escaping the
vigilance of the troops stationed all along the cliffs, whose forts
presented a formidable barrier. In short, Frotté and his friends found
themselves confronted with serious obstacles.
On January 11, 1795, they were observed to leave Guernsey in a
small sailing-vessel manned by English sailors, taking with them
three émigrés who were to act as guides. What happened to them?
No one knows exactly. Certain it is merely that the boat returned
rudderless and disabled, with Frotté and his four companions.
According to their own account, they took a wrong direction in the
dark, and sailed along the coast in the midst of rocks. Their guides
landed first, and disappeared from sight under a hail of bullets, and it
was with great difficulty that they themselves had been able to get
back to Guernsey.
At the beginning of February they made another effort, and
succeeded in landing near Saint-Brieuc. Frotté at once made his way
inland to join the insurgents, but ill fortune followed him. He had not
been a fortnight in the country when he learned, to his surprise, that
the Chouans under Cormatin had just concluded a truce to prepare
the way for peace. His feelings may be imagined. To have waited so
long for this! So much for his hopes and castles in the air! But there
was no help for it. On February 17, 1795, the treaty of Tannaye was
concluded, and a month later Frotté, who had kept moving about
over La Vendée and Normandy unceasingly to survey the ground,
established himself at Rennes, where he assisted at the conference
of La Mabilais, which was to confirm the truce already agreed to.
Marie-Pierre-Louis, Count de Frotté, 1766-1800.
(After a portrait belonging to the Marquis de Frotté.)
If the turn taken by events had led him off temporarily in a different
direction, his mind never abandoned the secret purpose which had
brought him to France. Nevertheless, a change, at first
imperceptible, but afterwards obvious enough, was coming over him.
The reader will not have forgotten the way in which a feeling of
antagonism had grown up between Cormier and the Chevalier. The
ill-will cherished by the latter for his quondam friend had not
disappeared. On the contrary, the belief that Lady Atkyns was
keeping him deliberately at arm’s length had intensified the jealousy.
The result was inevitable. Chagrined at being thus left on one side,
and at being supplanted, as he felt, in his fair lady’s affections, he
soon began to devote himself entirely to his new rôle as a Chouan
leader, and ceased to interest himself any longer in the drama of the
Temple. In truth, he was not without pretexts for this semi-desertion
of the cause.
On March 16, anxious to explain himself to Lady Atkyns, he writes
to tell her just how he is feeling on the subject. He would have her
realize that there is no longer any ground for hopes as to the
Dauphin’s safety. When in touch with the representatives of the
Convention who took part in the conference at La Mabilais, he had
taken one of them aside, it seems, and questioned him frankly as to
whether the Republican Government would consent to listen to any
proposal regarding the young Prince, and whether he, Frotté, would
be allowed to write to the Temple. The member of the Convention
made reply, after taking a day to consider the matter and to consult
his colleagues, that what Frotté suggested was out of the question.
“Your devotion,” he said, “would be fruitless, for under
Robespierre the unhappy boy was so demoralized, mentally
and physically, that he is now almost an imbecile, and can’t
live much longer. Therefore you may as well dismiss any such
idea from your head—you can form no notion of the hopeless
condition the poor little creature has sunk into.”
These lines, reflecting the view then current among the official
representatives of the Convention, stand out strikingly when we
recall the situation at the Temple in this very month of March, 1795,
and the absolute order given to Frotté not to allow the child to be
seen. They tally at all points with what we know of the substitution
that had been effected. To this substitution, indeed, Frotté himself
proceeds to make an explicit allusion towards the end of the letter.
“Perhaps the Convention is anxious,” he writes, “to bring
about the death of the child whom they have substituted for
the young King, so that they may be able to make people
believe that the latter is not really the King at all.”
As for himself, he has made up his mind. He will make no further
efforts for the deliverance of the Dauphin.
On April 25, 1795, the La Mabilais Treaty was signed, and Frotté,
who refused to subscribe to it, went off again to Normandy, confident
of seeing the struggle recommence, and impatient to set going a
new insurrection. Had he received any reply from Lady Atkyns to his
outspoken missive? Assuredly not. If she gave any credence to his
statements at the time, they must soon have passed out of her
memory, for, thanks to Cormier, June found her quite confident again
of the success of their plans. Not knowing, therefore, what to say to
her old admirer—Cormier having forbidden her to tell him the names
of their agents—she determined to keep silent.
Shortly afterwards, on the day after June 8, the report of the
Dauphin’s death reached Normandy. The proclamation of the Comte
de Provence—for how many weeks must he not have been waiting
impatiently for it to be made—as successor to the throne of France
in his nephew’s place was read to the insurgents. Frotté, who for
some time already had been responding to the advances made to
him by the pretendant, now formally placed his sword at the service
of the new King.
What would have prevented him from taking this step? Would a
personal interview with Lady Atkyns have had this effect? Perhaps;
but devoted now to his new mission, passing from fight to fight,
Frotté was no longer his own master.
Nevertheless, at the end of 1795, some feeling of remorse, or else
the desire to renew his old place in the goodwill of Lady Atkyns, who
had twice asked him to write and tell her about himself, moved Frotté
to take pen in hand once again. He had been engaged in fighting for
several months, concerting surprises and ambuscades, always on
the qui vive. He had twice narrowly escaped capture by the enemy.
In spite of this he managed to keep up an interesting
correspondence with his companions operating more to the south
and to the west, in La Vendée and in Le Bocage, and with the chiefs
of his party in London, who supplied the sinews of war, as well as
with Louis XVIII. himself, in whose cause he had sworn to shed the
last drop of his blood. There is no reason to be astonished at finding
our “Général des Chouans” expressing himself thus, or at the
changed attitude adopted by him, dictated by circumstances and the
new situation in which he has now found himself. Here is how he
seeks to disabuse Lady Atkyns of the hope to which she is still
clinging:—
“No, dear lady, I shall not forget my devotion to you before I
forget my allegiance to the blood of my kings. I have broken
faith in no way, but, unfortunately, I have none but untoward
news to give you. I have been grieved to find that we have
been deceived most completely. For nearly a month after
landing I was in the dark, but at last I got to the bottom of the
affair. I was not able to get to see the unfortunate child who
was born to rule over us. He was not saved. The regicides—
regicides twice over—having first, like the monsters they are,
allowed him to languish in his prison, brought about his end
there. He never left it. Just reflect how we have all been
duped. I don’t know how it is that without having ever received
my letters you are still labouring under this delusion. Nothing
remains for you but to weep for our treasure and to punish the
miscreants who are responsible for his death. Madame alone
remains, and it is almost certain that she will be sent to the
Emperor, if this has not been done already.”
These lines but confirmed what Frotté had written in the preceding
March, after his talk with the representative of the Convention. The
news of the Dauphin’s death having been proclaimed shortly after
that, there had been no longer any difficulty in persuading the
Chevalier to take up arms in the service of the Comte de Provence.
He discloses himself the change that has come over his sentiments.
“How is it,” he writes to Lady Atkyns, “that you are still
under the delusion, when all France has resounded with the
story of the misfortunes of our young, unhappy King? The
whole of Europe has now recognized His Royal Highness, his
uncle, as King of France.... The rights of blood have given me
another master, and I owe him equally my zeal and the
service of my arm, happy in having got a number of gallant
Royalists together. I have the honour of being in command of
those fighting in Normandy. That is my position, madame. You
will readily understand how I have suffered over the terrible
destiny of my young King, and nothing intensifies my sorrow
so much as the thought of the sadness you yourself will feel
when you learn the truth. But moderate your grief, my friend.
You owe yourself to the sister not less than to the brother.”
And to enforce this advice, Frotté recalls to her the memory of the
Queen, which should serve, he thinks, to remove all scruples.
“Remember the commands of your august friend, and you
will be able to bear up under your misfortunes. You will keep
up your spirits for the sake of Madame. You will live for her
and for your friends, to whom, moreover, you should do more
justice. Adieu, my unhappy friend. Accept the homage of a
true Royalist, who will never cease to be devoted to you, who
will never cease either to deplore this deception of which we
have been victims. Adieu.”
Was this farewell, taken in so nonchalant a fashion, to denote a
final sundering of two hearts united by so many memories in
common? It would appear so. Lady Atkyns was so strong in her
convictions that the only effect of such words would be to make her
feel that all was over between her and the Chevalier. Later, when he
made an effort to renew relations with her and asked her to return
the letters he had written to her, she would seem to have refused
point blank, from what she wrote to a confidant.
He must, however, have got hold of some portion of their
correspondence, for on his return to his château of Couterne, this
indefatigable penman, in the scant leisure left him by his military
duties, filled several note-books with reminiscences and political
reflections tending to justify his conduct. In one of these note-books,
which have been carefully preserved, he transcribed fragments of his
letters to his friend—fragments carefully selected in such a way as
not to implicate him in the affair of the Temple, once the death of the
Dauphin had been announced. Had he lived, he would doubtless
have learned what had really happened, as set forth in the
documents we have been studying; but his days were numbered.
His end is well known: how, having fallen into an ambush, he and
six of his companions were shot by Napoleon’s orders, in despite of
a safe-conduct with which he was furnished, on February 18, 1800,
at Verneuil. If in the course of these five years he did learn the full
truth about the Dauphin, he doubtless abstained from any reference
to it out of regard for the King. He carried his private convictions in
silence to the grave.
The news of his death was received with emotion in London.
Peltier, who had had good opportunities for forming an opinion of
him, gave out a cry of horror. “This act,” he wrote in his gazette,
“covers Bonaparte for ever with shame and infamy.”
The small circle of Lady Atkyns’ London friends lost thus one of its
members. Meanwhile, Lady Atkyns had been making the
acquaintance of a French woman who had been living in England for
some years, and whose feelings corresponded to a remarkable
degree with her own. This lady had found a warm welcome at
Richmond, near London, on her arrival as an émigré from France.
Pale, thin, anxious-looking, the victim of a sombre sorrow which
almost disfigured her face, Louise de Chatillon, Princesse de
Tarente, wife of the Duc de la Tremoïlle, had escaped death in a
marvellous way. A follower of Marie-Antoinette, from whom she had
been separated only by force, she had been arrested on the day
after August 10 as having been the friend of the Princesse de
Lamballe. Shut up in the sinister prison of l’Abbaye, she had felt that
death was close at hand. From her dungeon she could see the men
of September at their work and hear the cries of agony given forth by
their victims. At last, after ten days of imprisonment, she was
liberated, thanks to an unexpected intervention, and in the month of
September, 1792, she succeeded in finding a ship to take her to
England.
Hers was a strikingly original personality, and it is not without a
feeling of surprise that one studies the portrait of her which
accompanies the recent work, Souvenirs de la Princesse de Tarente.
The drama in which she had taken part, and the bloody spectacle of
which she had been a witness, seem to have left their mark on her
countenance, with its aspect of embittered sadness. Her eyes give
out a look of fierceness. Save for the thin hair partially covering her
forehead, there is almost nothing feminine in her face. Seeing her for
the first time, Lady Atkyns must have received an impression for
which she was unprepared. They took to each other, however, very
quickly, having a bond in common in their memories of the Queen.
Both had come under the charm of Marie-Antoinette, their devotion
to whom was ardent and sincere. The Queen was their one great
topic of conversation. Few of their letters lack some allusion to her.
Knowledge of Lady Atkyns’ devotion to the Royal House of
France, of the sacrifices she had made, was widespread in the world
of English society, and the Princess, having heard of her, was
anxious to meet the woman, who, more fortunate than herself, had
been able to afford some balm to the sufferings, to prevent which
she would so willingly have given her life. The Duke of Queensberry
brought about a meeting between the two ladies. What passed
between them on this occasion? What questions did they exchange
in their eager anxiety to learn something new about the Queen?
Doubtless the most eager inquiries came from the Princess, and
bore upon the achievements of Lady Atkyns, her visit to the
Conciergerie, her talks with the illustrious prisoner. For weeks
afterwards there was an interchange of letters between the two, in
which is clearly disclosed the state of affectionate anxiety of the
Princess’s mind. They address each other already by their Christian
names, Louise and Charlotte. Lady Atkyns shows Mme. de Tarente
the few souvenirs of the Queen she still possessed, the last lines the
Queen wrote to her. It is touching to note, in reading their
correspondence, how every day is to them an anniversary of some
event in the life of the Queen, full of sweet or anguishing memories.
“How sad I was yesterday!” writes the Princess. “It was the
anniversary of a terrible day, when the Queen escaped
assassination only by betaking herself to the King’s
apartments in the middle of the night. Why did she escape?
To know you—but for that the Almighty would surely have
been kind enough to her to have let her fall a victim then.”
For all the affection which surrounds her, Mme. de Tarente
constantly bemoans her solitude.
“I am in the midst of the world,” she writes, “yet all alone.
Yesterday I longed so to talk of that which filled my poor heart,
but there was none who would have understood me. So I kept
my trouble to myself. I was like one of those figures you wind
up which go for a time and then stop again. I kept falling to
pieces and pulling myself together again. Ah, how sad life is!”
In the summer of 1797 the Princess came to a momentous
decision. The Emperor and Empress of Russia, whom she had
known formerly at the French Court, having heard of her trials and of
the not very enviable condition in which she was living, pressed her
to come to Russia, where she would be cordially greeted. After long
hesitation she decided to accept, but it was not without genuine
heartburnings that she separated from her English friends, from her
Charlotte most of all. She left London at the end of July, and arrived
at St. Petersburg a fortnight later. Very soon afterwards she wrote
Lady Atkyns an account of the journey and of her first impressions of
her new surroundings.
The Emperor and Empress received her in their Peterhof palace
with the utmost consideration. Appointed at once a lady-in-waiting on
the Empress, she found herself in enjoyment of many privileges
attached to this post. The house in which she was to live had been
prepared for her specially by the Emperor’s command. Finally, she
was decorated with the Order of St. Catherine, and the Empress on
her fête day presented her with her portrait. Different indeed is her
position from what it had been at Richmond.
“I never drive out without four horses, and even this is my
own doing, for I ought not, as a lady-in-waiting, to have less
than six. They tell me I shall be obliged to get myself made
the uniform of the Order of St. Catherine, and that would cost
me 1200 roubles, that is, 150 louis.”
But the very marked favour met with by the Princess could not but
disquiet some of the courtiers at the Palace. Within a week of her
arrival, one of the ladies in attendance upon the Empress, Mme. de
Nelidoff, at the instigation of Prince Alexandre Kourakine, hastened
to represent Mme. de Tarente’s conduct and the unusual honour that
had been shown her under the most unfavourable light to her
Majesty the Empress; and her jealousy thus aroused (so one of
Mme. de Tarente’s friends tells the tale), she had no difficulty in
settling matters with her husband, and when the Princess next
entered the imperial presence, the Emperor neither spoke to her nor
looked at her.
The snub was patent, but the Princess seems to have taken it
nonchalantly enough. The friendly welcome accorded to her by St.
Petersburg society, the kindness and affection she met with from the
Golowine family, in whose house she soon installed herself, there to
remain until her death, enabled her speedily to forget the intrigue of
her enemies at the Court. The incident is barely alluded to in her
letters to Lady Atkyns, which continue to be taken up chiefly with
reminiscences of their beloved Queen.
Towards the end of 1798 the two friends are sundered by Lady
Atkyns’ decision to return to France, impelled by the desire to be
near those who had played so important a rôle in her life, and to
meet again those friends who had co-operated in her work—perhaps
also to meet and question those who might be in a position to
enlighten her regarding the fate of the Dauphin. This decision she
communicates to the Princess, who opposes it strongly, warning her
against the imprudence she is about to commit. Lady Atkyns
persists, and the Princess at last loses patience. “I have so often
combated your mad idea,” she writes nobly, “that I don’t wish to say
anything more on the subject.”
In the spring of 1814 the news came to St. Petersburg of the
defeat of the armies of Napoleon and the accession of Louis XVIII.
Immediately large numbers of exiles, who were but waiting for this,
made haste back to France. Mme. de Tarente contemplated being of
their number, but before she could even make arrangements for the
journey, death came to her on January 22, 1814.
Hamburg, where our friends Mme. Cormier and the “little Baron”
took refuge in 1795, was already a powerful city, rich by reason of its
commerce, and its governing body, conscious of its strength, were
not the less jealous of its independence. Its unique position, in the
midst of the other German states, the neutrality to which it clung and
which it was determined should be respected, sufficed to prevent it
hitherto from looking askance at the ever-growing triumphs of the
armies of the French Republic, and the Convention, too much taken
up with its own frontiers, had done nothing to threaten the
independence of the Hanseatic town.
This fact did not escape the émigrés, who were finding it more and
more difficult to evade the rigorous look-out of the Revolutionary
Government, and soon Hamburg was filled with nobles,
ecclesiastics, Chouans, conspirators, Royalist agents, just as
London had been some years earlier. Safe from surprises, and in
constant communication with England, Germany, and Italy, this world
of wanderers had discovered an ideal haven in which to hatch all
their divers plots. Clubs were started by them, called after celebrated
men. Rivarol was the centre of one set, noted for its intellectual
stamp and its verve and wit. The publications also that saw the light
in Hamburg enjoyed a wide liberty, and this it was that opened the
eyes of the Republican Government to the state of things.
On September 28, 1795, there arrived at Hamburg, Citoyen
Charles-Frédéric Reinhard, official representative of the Convention,

Mosaics of Knowledge Representing Information in The Roman World Andrew Riggsby Full Chapter PDF

Uploaded by

Document Information

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

Mosaics of Knowledge Representing Information in The Roman World Andrew Riggsby Full Chapter PDF

Uploaded by

Copyright:

Available Formats

Mosaics of Knowledge: Representing

Information in the Roman World

Trade, Commerce, and the State in the Roman World

London in the Roman World Dominic Perring

London in the Roman World Dominic Perring

Women and Society in the Roman World: A Sourcebook of

Spectacles in the Roman World: A Sourcebook Siobhán

Coin Hoards and Hoarding in the Roman World Jerome

Infancy and Earliest Childhood in the Roman World: ’A

Greek and Roman Technology: A Sourcebook of Translated

Classical Culture and Society

Emotion, Restraint, and Community in Ancient Rome

Making Mockery: The Poetics of Ancient Satire

Readers and Reading Culture in the High Roman Empire:

Apollonius of Rhodes and the Spaces of Hellenism

The Captor’s Image: Greek Culture in Roman Ecphrasis

Aratus and the Astronomical Tradition

Gift and Gain: How Money Transformed Ancient Rome

Mosaics of Knowledge: Representing Information in the Roman World

Published in the United States of America by Oxford University Press

© Oxford University Press 2019

All rights reserved. No part of this publication may be reproduced, stored in

You must not circulate this work in any other form

CIP data is on file at the Library of Congress

Printed by Sheridan Books, Inc., United States of America

2. Tables and Tabular Organization 42

3. Weights and Measures 83

4. Representing Three Dimensions 130

5. Representing Two Dimensions 154

What Is a “Map”? 172

1.1 Supposed theater token from Pompeii 32

1.1 Early references to the supposed Arles amphitheater token 34

1 Fasti Amiternini, with color coding

Fellowship and the atmosphere uniquely hospitable to scholarship. Then in 2013–​

Plate 6 Landscape from a villa at Boscotrecase

Plate 8 Fragment of a marble map of Rome (the “via Anicia” fragment)

This book investigates information technologies in the classical Roman world—​

[T]‌he contracts concluded by Romans in Greek are usually indistinguishable

reasons to suspect that there might be a significant level of separation. Nor do

is given to organization itself over content; and/​or physical or enactive

• Tolerance and context-​specificity. Roman information systems (e.g.,

• Shared knowledge. These technologies require that producers and

As a consequence of all these interacting factors, Romans lived in a balkanized

The list relies on discontinuity rather than continuity; it depends on physical

around line 640 (10.10C)

AMERICAN DOCUMENTATION OF VESSEL REQUIRED. If 18

if there is no judgment within the time which is foreseen by ­chapter 12 of the

You might also like

Fellowship and the atmosphere uniquely hospitable to scholarship. Then in 2013–

This book investigates information technologies in the classical Roman world—

is given to organization itself over content; and/or physical or enactive

• Tolerance and context-specificity. Roman information systems (e.g.,

if there is no judgment within the time which is foreseen by chapter 12 of the