Content uploaded by Joseph Errington
Author content
All content in this area was uploaded by Joseph Errington on Dec 14, 2022
Content may be subject to copyright.
Linguistics in
a Colonial World:
A Story of Language,
Meaning, and Power
Joseph Errington
Linguistics in a Colonial World
Linguistics in
a Colonial World:
A Story of Language,
Meaning, and Power
Joseph Errington
© 2008 by J. Joseph Errington
blackwell publishing
350 Main Street, Malden, MA 02148-5020, USA
9600 Garsington Road, Oxford OX4 2DQ, UK
550 Swanston Street, Carlton, Victoria 3053, Australia
The right of J. Joseph Errington to be identified as the Author of this
Work has been asserted in accordance with the UK Copyright,
Designs, and Patents Act 1988.
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, electronic, mechanical, photocopying, recording or otherwise,
except as permitted by the UK Copyright, Designs, and Patents Act
1988, without the prior permission of the publisher.
First published 2008 by Blackwell Publishing Ltd
1 2008
Library of Congress Cataloging-in-Publication Data
Errington, James Joseph, 1951–
Linguistics in a colonial world : a story of language, meaning, and
power / Joseph Errington.
p. cm.
Includes bibliographical references and index.
ISBN 978-1-4051-0569-9 (hardback : alk. paper) –
ISBN 978-1-4051-0570-5 (pbk. : alk.paper)
1. Linguistics–History. 2. Colonies. I. Title.
P71.E77 2007
410.9–dc22
2007014517
A catalogue record for this title is available from the British Library.
Set in 10.5 on 13 pt Palatino
by SNP Best-set Typesetter Ltd., Hong Kong
Printed and bound in Singapore
by Markono Print Media Pte Ltd
The publisher’s policy is to use permanent paper from mills that
operate a sustainable forestry policy, and which has been
manufactured from pulp processed using acid-free and elementary
chlorine-free practices. Furthermore, the publisher ensures that the text
paper and cover board used have met acceptable environmental
accreditation standards.
For further information on
Blackwell Publishing, visit our website:
www.blackwellpublishing.com
Contents
List of Figures and Table, vi
Preface, vii
Acknowledgments, ix
1 The Linguistic in the Colonial, 1
2 Early Conversions, or, How Spanish Friars Made the Little
Jump, 22
3 Imaging the Linguistic Past, 48
4 Philology’s Evolutions, 70
5 Between Pentecost and Pidgins, 93
6 Colonial Linguists, (Proto)-National Languages, 123
7 Postcolonial Postscript, 149
References, 172
Languages Index, 189
Persons Index, 191
Subject Index, 195
List of Figures and Table
Figure 3.1 The philological image of textual descent. 60
Figure 3.2 Indo-European or proto-Indo-European 63
migrations in time and space.
Figure 4.1 Van Bulck’s linguistic map of the Belgian 86
Congo (reproduced from Van Bulck 1948).
Figure 5.1 Map of the “languages” of Shonaland, 110
southern Rhodesia (reproduced from
Doke 1931).
Figure 5.2 The Swiss mission to the Transvaal (redrawn 114
from Harries 1988).
Figure 5.3 Hulstaert’s linguistic map of the Belgian 118
Congo (reproduced from Hulstaert 1950).
Figure 6.1 Map of Indonesia, the Netherlands East 134
Indies (NEI) (redrawn from Stoler 1985).
Table 3.1 Some Indo-European words for “mother”. 62
Preface
This is a book about European engagements with linguistic diver-
sity in colonial projects around the world. It focuses on the ways
colonial agents made alien ways of speaking into objects of knowl-
edge, so that their speakers could be made subjects of colonial
power. In this way it presents a language-centered survey of the
colonial world from the 16th to the 20th century, showing how the
work of linguistics changed in and with the projects of power it
served.
At the same time, it is a story about the West’s confrontation with
linguistic diversity as a founding conundrum of the human condi-
tion, recasting the old question of what was once called the “confu-
sion of tongues” (in Latin, confusio linguarum): the legacy of the fall
of the Tower of Babel, told of in the book of Genesis, which was
God’s curse on the nations of humankind for their sin of pride.
Two thousand years ago, at the dawn of the Common Era, a
philosopher living in Alexandria – an outpost of the Roman empire
– gave the story of Babel an allegorical reading. This was Philo
Judaeus, a Jewish student of Plato, who regarded his people’s sacred
texts as a kind of narrative shadow of God’s Truth which he aimed
to read on the wall of his worldly cave. Philo pointed out first how
dubious was the story’s literal meaning: it tells of linguistic diver-
sity as a curse aimed at preventing humans from colluding again
as they did at Babel, but even though “[m]en have been separated
into different nations, and have no longer used one language,
nevertheless, land and sea have been repeatedly filled with unspeak-
able evils. For it was not the languages which were the causes of
viii Preface
men’s uniting for evil objects, but the emulation and rivalry of their
souls in wrong-doing” (1854:3). Philo went on to argue that God
did not really create a confusion of tongues, but a division among
them, making “one thing into many parts, as is the case when one
distinguishes a genus into its subordinate species” (1854:41). These
arguments capture two of this book’s broadest themes.
Knowingly or not, willingly or not, colonial linguists carried out
projects of physical and symbolic violence, some of them counting
clearly as what Philo called “unspeakable evils.” But they were also,
knowingly or not, reducing to writing some of the evidence of a
unity underlying linguistic and human diversity, charting proper-
ties shared among these “subordinate species,” as Philo calls them,
of the larger “genus” of language.
Always and everywhere, the work of linguistics in a colonial
world was grounded in this enabling and conflicted condition, and
the contradiction it presented for the work of power and knowl-
edge. In a series of sketches I trace here linguists’ ways of trying to
disentangle languages from the communities and lives in which
they figured, reading for the ways their dealings with language
were enabled and shaped by their ideas about the human condition.
By situating their projects of knowledge in larger projects of power,
I try to show how the work of fixing languages in writing helped
fix speakers in colonial yet “natural” hierarchies.
Acknowledgments
I am a linguist and anthropologist whose work straddles the study
of language as a universal human attribute, and as part of the fabric
of lives people create with each other. To move beyond my own
special concerns with language use and change in Java, I have
charted a course into unfamiliar intellectual, historical, and geo-
graphical territory. If I have managed to avoid writing like a rank
dilettante, it is only because of the generosity of colleagues.
Some who have influenced me most are the least cited here:
Chuck Briggs and Dick Bauman, Sue Gal and Judy Irvine, Jane Hill,
Paul Kroskrity, Michael Silverstein, and Bambi Schieffelin. Kit
Woolard has an inconstant correspondent’s gratitude for her acuity
and patience, and Hal Conklin my thanks not just for his invaluable
example as a colleague, but for help with baybayin, the Philippine
orthography I discuss in chapter 2. For the discussion of Nahuatl
language and history there I am grateful to James Lockhart, more
of whose expertise should probably have found its way into that
same chapter.
My sketch of the political and cultural background of early lin-
guistics in chapters 3 and 4 owes much to Michael Holquist, through
both conversation and his work in progress on the beginnings of
the Prussian system of education; Katya Benes generously discussed
early German philology with me; Sara Pugach kindly allowed me
to read what was at the time some of her unpublished work on Karl
Meinhof. Discussion of Tamil in chapter 4 owes much to the advice
and references at just the right time from my colleague Barney Bate,
for which I am grateful.
Thanks also to David Dwyer for his help in disentangling some
of the complexities of Shona in chapter 5, and Henk Maier for his
help in different times and places to appreciate more fully the impli-
cations of his work on the history of Malay. I am grateful to Nina
Garrett for improving an awkward Dutch translation, and to Shafqat
Hussein for his dusty travails in the Mudd. Conversations with
Doug Whalen and David Harrison helped me to frame difficult
issues of language endangerment discussed in chapter 7. Finally,
I am grateful to three anonymous readers for the care and spirit
of generosity I found in their suggestions and criticisms. I hope
my response gives adequate evidence of my appreciation for
their help.
Finally, loving thanks for support in all times and places to my
mother, Frances Clare Errington. This book is for her.
Joseph Errington
xAcknowledgments
Chapter 1
The Linguistic
in the Colonial
The colonial era ended two generations ago, but colonialism has not
really gone away. Its afterlife has been all too clear in global north–
south inequalities; in bloody politics from Timor to Iraq to Rwanda;
in critical identity politics where former colonial powers now are
homes to former colonial subjects and their children. Many con-
spicuous signs of the colonial past in the globalizing present make
it easy to wonder whether some genuinely new era is here or in
the offing, or whether there has ever been a definitive rupture to
separate us from the colonial epoch.
Some scholars have colonialism on their minds because they
recognize that it might be in our minds, in the guise of durable
categories and ideas which emerged then but still serve now as
common sense for thinking about human diversity and inequality.
These concerns have led many into closer, more critical engage-
ments with textual remains of the colonial past, which they read
with one eye on the present. History, biography, literature, and
other kinds of writing take on new importance from this point of
view, as do maps, censuses, photographs, monuments, and a wide
range of other materials. All of these can be thought of as parts of
a colonial archive: legacies of very different times and places which
differ hugely in their form and content, are scattered over much of
the world, but which all count as traces of the broad projects of
power which accelerated and globalized between the 16th and 20th
centuries.
The colonial archive was once primarily the territory of histori-
ans, but since 1970 or so they have been joined by other readers – in
2The Linguistic in the Colonial
literature and intellectual history, anthropology and political
science, comparative religion and psychoanalysis – who bring their
own concerns and interests. Their agendas differ, as do the ques-
tions they pose and the parts of the archive which interest them,
but some broad similarities can be seen between their strategies
of critical reading. Beyond the archive’s overt informational content
and purposes, they read for broader understandings of the times
and places in which those texts were produced so as to be
useful and meaningful. Developing an imaginative sense of these
texts, and reading them back into their contexts of origin, helps to
make them windows on circumstances in which they could be
useful, marginal, or dangerous; routine or innovative; and so on.
Reading for signs of authors’ times – against the grain of texts,
for what they say that their authors did not always intend – requires
critical strategies for recognizing their partial vision, conflicted
stances, changing perspectives, and mixed motives. When gaps or
“silences” in the textual record are found – excluded facts, over-
simple categories, elided stories, and so on – they can in turn be
read diagnostically, as symptomatic of tensions which animated
broader colonial projects, blurring lines between knowledge and
interest, purpose and effect, between ideology and reality. In these
ways power’s shaping effects on the work of writing can be read
back into what Louise Pratt calls “zones of colonial contact,” and
defines as “space[s] of colonial encounter . . . [as] peoples geograph-
ically and historically separated come into contact with each
other and establish ongoing relations, usually involving conditions
of coercion, radical inequality, and intractable conflict” (Pratt
1992:6)1.
As a practical matter of fact, zones of contact were defined by
lines of human difference bound up with language difference.
Wherever colonial agents stopped short (as some did not) of brutal
violence, they were obliged to find ways of bridging those linguis-
tic gaps with acts of verbal communication, however rudimentary
or inadequate. Talk was one of the lowest common denominators
for colonial dealings when some humans made others targets of
their efforts to persuade or awe, threaten or coerce, and who in turn
resisted or cooperated, retreated or collaborated. Without “lan-
guages of colonial command” even simple ideas, orders, questions,
threats, and arguments could not be communicated across these
The Linguistic in the Colonial 3
lines of colonial difference, which is why colonialists always and
everywhere needed what Bernard Cohn (1996b) calls “command of
colonial languages.”
Given language difference as a fact of life in zones of colonial
contact, in all times and places, it is not surprising that colonialists
produced texts about languages over four centuries, around the
world, or that those texts now represent a significant part of the
colonial archive. It is likewise intuitively obvious that these sorts of
texts – grammars, dictionaries, world lists, and so on – stand out
because of their distinctive modes of organization and content.
They count very obviously as reports on work which made lan-
guages objects of knowledge, so that their speakers could be made
subjects of power.
But these texts are a trial to read because they also appear so
opaque with respect to worlds of talk which they present in partial,
written guises. They offer few clear points of purchase for critical
readings against the empirical grain, back into the zones of colonial
contact from which they emerged and which they so partially rep-
resent. The work of describing languages may require close engage-
ment with complex intimacies of talk, but it results in texts which
stand further from life’s hard edges than many more institutional
or official parts of the colonial archive: censuses or land surveys,
photographs or revenue reports, even broad descriptions of
“customs” or “ways of life.”
This is a conundrum which rests on another. These dictionaries,
grammars, and related texts present enormous amounts of descrip-
tive information about enormously different languages, yet a quick
perusal shows that they also resemble each other in obvious ways.
Resemblances between these texts can be traced to the fact that each
describes an object which falls under a single, common category.
As a practical matter, linguists worked in zones of colonial contact
on the premise that the languages they were describing could be
compared with and presented in the image of others more familiar
to them.
This conundrum of sameness-and-difference is important for
framing larger meanings of the work of linguistics as a special kind
of colonial encounter, and for reading very small bits of this part of
the archive with an eye to its ideological, intellectual, and practical
importance for prosecuting and legitimizing colonial projects.
4The Linguistic in the Colonial
Remembering that dry, minute descriptions emerged from situated
dealings with conditions of human difference/sameness helps one
to read them as part of the work of fixing colonial subjects in and
with categories of colonial otherness. It allows this work to be con-
sidered as a means for adapting and exploiting familiar categories
in ways which enabled power and legitimized authority in un-
familiar tongues.
Colonial Distance and Linguistic Difference
As projects of power, the work of linguistics I call “colonial” here
served the “direct territorial appropriation of another geopolitical
entity, and exploitation of its resources and labor, and systematic
interference in the capacity of the appropriated culture . . . to orga-
nize its dispensations of power” (McClintock 1992:1). Most (not all)
of this was work done by Europeans at considerable distances from
the places where it was planned and authorized.2One influential
approach to colonialism as a world historical phenomenon, then, is
to read complex, scattered events four centuries into patterns of
more and stronger flows of people and goods, capital, and technol-
ogy around the globe. Up to the end of World War II, these political
and economic dynamics gave rise to increasingly integrated geopo-
litical relations between European (and later American) centers of
imperial power and scattered peripheries where that power was
wielded.
From this angle, linguists can be regarded as a small, rather
special group of colonial agents who adapted European letters to
alien ways of talking and, by that means, devised necessary con-
duits for communication across lines of colonial power. However
different the methods they used or objects they described, they
transformed familiar alphabets into visual images of strange speech:
their writing systems, or orthographies, were the common beginning
point for the work of writing grammars, dictionaries, instructional
texts, and so on.
But linguists who worked as technicians, transposing “our”
alphabets onto blooming, buzzing confusions of talk, also served
broader ideological projects. Always and everywhere, their written
images of alien languages demonstrated underlying comparability:
The Linguistic in the Colonial 5
once “their” talk was writable, like ours, resemblances were estab-
lished between them and “us.” This concrete commonality is obvious
but also diffuse because it played out at shifting, slippery intersec-
tions between transient events of talk, what people do, and descrip-
tions of languages which they share and which mark who they are,
with and to each other.
This means that the intellectual work of writing speech was
never entirely distinct from the “ideological” work of devising
images of people in zones of colonial contact.3It means also that
language difference figured in the creation of human hierarchies,
such that colonial subjects could be recognized as human, yet defi -
ciently so. Language difference in this way embodied some of the
most basic “tensions of empire” which developed as colonialists
had to “mark and police boundaries, design systems of punishment
and discipline, [but also] to instill awe as well as a sense of belong-
ing in diverse populations” (Cooper 2005:30). Written images of
languages could embody these tensions, both constructing and
giving a semblance of “naturalness” to other categories of colonial
difference.
One way to read this work, then, is with an eye to the different
hierarchies – religious, philosophical, evolutionary, and so on –
which played into the ways it was done in different contexts, for
different purposes. This leads also to readings of this work not just
as it was written in zones of colonial contact abroad, but as it was
read in Europe’s metropolitan centers, circulating and figuring in
larger ideological and intellectual challenges which colonial expan-
sion continually posed. These were challenges not just to European
understandings of linguistic and human diversity far away in the
colonies, but at home, among colonizing nations.
I develop two different stances to this work of linguists. One is
to regard them as technicians who deployed alphabetic symbols
to “stand for” sounds of speech in unfamiliar tongues, and then
devised descriptions of the meaningful elements those sounds
comprised. The other is to read those texts as the work not of tech-
nicians of literacy but members of literate groups whose work was
enabled and shaped by their social biographies, their broader in-
vestments in larger projects, their membership in certain groups,
their broader beliefs, values, and purposes. From this second point
of view, linguistic descriptions appear as points of convergence
6The Linguistic in the Colonial
between stubborn empirical particulars and diffuse, powerful
habits of thought and action. To read these works without forget-
ting that their authors were never just linguists can make them
speak to the rootedness of that work in broader practices of literacy,
and through those practices in the legitimizing textual traditions of
faith and civilization.
Reducing Speech to Writing
A linguist can still read with profit old grammars, dictionaries, and
similar texts because, as I noted above, they stand together with
each other in the colonial archive, and share an expository logic
which is evident in descriptive details about languages like those I
discuss here, spoken in 17th century Mexico and the Philippines,
18th century India, 19th century Europe, 20th century Africa and
Asia. The organizational coherence of these texts can be understood
in two different ways.
From the empirical, “common sense” point of view, texts resem-
ble each other as do the things they describe. This can be framed in
terms of the guiding principle of modern linguistics: that specific
languages exemplify a panhuman language faculty, a genetic
endowment which grounds distinctively human ways of producing
and understanding verbal communicative behavior. From the more
critical point of view which allows for reading these texts “against
the grain,” resemblances between them are due to the techniques
and strategies used to re-present languages in them. Texts are
similar to each other, by this argument, because each presents such
a partial image of diverse realities of talk. These opposed points of
view, “empirical” and “critical,” help here to develop a productive
tension between readings of linguistic descriptions, one which it is
worthwhile to articulate more fully through this section’s usefully
ambiguous title: “Reducing speech to writing.”
Fifty years ago, linguists I discuss in chapter 7 often used this
phrase to describe their work, and in so doing affirmed that it cap-
tured the empirical reality of properties of language systems. These
are understood to be the abstract, internally organized, conven-
tional linkages which, when embodied in talk, join sound with
meaning. From this point of view, the colonial forebears of modern
The Linguistic in the Colonial 7
linguists were able to capture parts of a reality whose underlying
nature they did not really understand when they thought of it as
the legacy of the curse of Babel, an outgrowth of a natural environ-
ment, an organism which was fated to live and die in historical time,
and so on. Beyond all these conceptions of language, which I discuss
here, their work has value because it keyed to the ontological status
of language systems, which have the thingy-ness of all natural
phenomena.
From this empirical point of view, then, visual symbols map not
just onto sounds of speech as tongues produce and ears hear them;
they represent components of “underlying” language systems
which talk embodies in order to be meaningful. In this sense lin-
guists “reduce” speech to writing by capturing its underlying essen-
tials, as they recur wherever and whenever talk happens. The better
the alignment a linguist creates between sound and meaning, accu-
rately representing them together within a language system, the
more closely he or she cuts at the joints of that language’s nature,
whether it is spoken in 16th century Mexico or 21st century
Indonesia.
A broadly contrary, critical view can also be summarized with
the other sense which can be given to the phrase “reducing speech
to writing.” More philosophical arguments can be made that what
linguists actually do is reduce the reality of language, which is talk,
to their own timeless, airless realm of visual images. This is reduc-
tion as the work of substituting simplified part for complex whole,
work which can seem straightforward and “natural” for linguists,
and other literate people, only thanks to shared habits of thought
and practice in writing. These habits lead them (and others) to
ignore the constant gap which they create between verbal realities
and visual representations.
Ludwig Wittgenstein mounted one such argument by observing
that it is in the nature of humans (including linguists) to think that
they are “tracing the outline of [a] thing’s nature [here, a language]
over and over again,” when they are in fact “merely tracing around
the [literate] frame through which [they] look at it” (Wittgenstein
1953:48). He arrived at this conclusion after he abandoned a philo-
sophical project which had been founded on ideas about language
broadly similar to those outlined above, whose designers had
intended to capture universal properties of all languages with “an
8The Linguistic in the Colonial
exact calculus,” devised and used in the manner of “the sciences
and in mathematics” (1953:25). Critical reflection led Wittgenstein
to reject this project along with one of its basic premises: that the
meaning of talk has the kind of stability which allows it be abstracted
away and captured in written re-presentations.
With a series of simple thought experiments, Wittgenstein tried
to demonstrate, not explain, that philosophers (and linguists) con-
stantly make recourse to what they know, without having been
taught, to describe language as separate from language use. Ab-
stracting out writable elements that are the same in different acts of
speech, they ignore the larger conditions in which those acts occur:
conditions which are shared by those who are present for acts of
speech. Wittgenstein foregrounded the meaningfulness of talk
which depends on what people know, and know others know,
about the times and places where it happens. These extend to
broadly practical senses they can share but not necessarily describe
about “what is going on” during, but also before and after, any
given act of talk. Wittgenstein describes these senses of sharedness
as “language games” which shape talk’s situated meanings, and the
conditions in which humans are with each other as speaking
animals.
In his study and garden, Wittgenstein ruminated on these issues
far from the messy details of exotic languages, or cruel realities of
colonial encounter. But he drew attention to the kinds of habits of
thought and practice which enabled and shaped the work of the
colonial linguists I describe here. A broadly similar critique devel-
oped by Jacques Derrida (1976) can be mentioned here because it
was directed more specifically at foundational work in the modern
science of linguistics, which I discuss in chapter 7. Derrida applied
his deconstructive technique to Saussure’s writings by turning
one of its founding premises on its head: the principle that the
letter is secondary to the reality of speech and voice, which it
represents. Linguists take this as an enabling principle: their
orthographies count as surrogates used to create models of the
primary reality of talk. But, Derrida argues, they proceed on the
opposite premise because, when they assert that writing is second-
ary to speech, they also act as if there can be nothing in writing
which is not already in what it describes. This means that if
The Linguistic in the Colonial 9
orthographies can represent stable patterns of sound and meaning,
then those patterns must be part of the reality to which they are
secondary. In fact, Derrida argues, this is the way linguists “read
back” into speech a stability of meaning which actually exists only
in their descriptions, the “secondary” realm of the letter.4
To make Derrida echo Wittgenstein in this oversimple way helps
to frame very broadly some of the specific issues I deal with here
as they emerged in the work of colonial linguistics. Unlike those
philosophers, I share other linguists’ senses of engagement with
realities which one can work, more or less successfully, to “get
right” in a description. But, as a critical reader of that work, I rec-
ognize not just that it is by its nature highly partial, but also the
result of more than the observation of data and application of
reason. It represents instead the product of language games which
linguists (and other interested parties) adapted and devised in the
face of specific challenges, and in response to practical needs.
So to read the work of colonial linguistics empirically and criti-
cally, I seek out strategies linguists used to arrive at part-for-whole
substitutions of written images for linguistic realities, and the highly
interested, power-laden positions from which they did this work.
For this I need a narrower angle of vision than Wittgenstein’s or
Derrida’s, and more situated ways of tracing the work of devising
linguistic images in complex situations. To see how linguists
extracted legible linguistic sameness from exotic human shared-
ness, I need to consider the ways they did not just devise orthogra-
phies, but adapted practices of literacy.
Practices of Literacy
The most basic way to read linguistic treatises not as records of
facts, but as products of interests and imaginations, is to foreground
the strategies of selection – always needed, and never disinterested
– which went into the production of simple images of linguistic
diversity. Two different strategies of selection were needed to create
two kinds of part-for-whole substitutions, each oriented to a differ-
ent sphere of the social meanings which talk has. One emerges
from the ways differences in speech mark differences in speakers’
10 The Linguistic in the Colonial
identities, biographies, and community membership; the other
emerges from the ways speaking fits into and molds people’s senses
of identity in face-to-face dealings with each other.
Issues of linguistic identity and background can be easily and
oversimply thought of as matters of “dialect” and “accent,” but
these are terms which disguise and straddle complex, overlapping
social categories: region, race, ethnicity, religion, class, gender, age,
and so on. Such linguistic differences, always facts of social life,
were encountered by colonialists as challenges which they dealt
with by selecting some ways of speaking as their objects of descrip-
tion, while ignoring others. The ways they chose to make one way
of speaking stand for many was always shaped by broader factors
and purposes, allowing questions to be posed about what guided
their strategies of selection, whether they knew it or not. What
assumptions, interests, beliefs, and purposes shaped the ways they
devised models of speech which could then be used as models for
speech?
A second dimension of complexity emerges in the ways people’s
sense of themselves and others shifts with the talk which is part
of their condition of being together. “Style” is the convenient,
overeasy label for the molding of people’s intersubjective orienta-
tions as “I” and “you” for each other in talk, perhaps in the presence
of others. In these spaces of sharedness Wittgenstein’s “language
games” are in play, and speech can modulate them in ways
more complex and subtle than can be conveyed with labels like
“formal” and “informal,” “respectful” or “familiar,” “literate” or
“colloquial.”
Both of these kinds of linguistic complexity can be regarded as
secondary from a literacy-centered view of language, each a kind
of “add on” to the real business of talk: communicating information,
expressing intent, and so on. But that complexity is important here
because the step from facts of language difference to ideas about
human deficiency is very short, and so questions of “dialect” and
“style” played simultaneously into power-laden zones of colonial
contact, and the politics of interaction. Strategies of selection, as
linguists devised and used them, played indirectly but sometimes
crucially into constructions of colonial power and authority.
These were engagements, then, which required resources that
were more than technical and intellectual; they were shaped by
The Linguistic in the Colonial 11
broader senses of who people/speakers are to each other which
linguists (among others) brought to their work. These can be called,
as by the political philosopher Charles Taylor, their social imaginar-
ies: the “normal expectations that we have of one another, the kind
of common understanding which enables us to carry out the collec-
tive practices that make up our social life . . . [our] sense of how
things usually go . . . interwoven with an idea of how they ought to
go, of what missteps would invalidate the practice” (Taylor 2002:106,
see also 2004). Though Taylor is concerned primarily with political
culture and social change in modern Europe, his broad notion of
“social imaginary” helps here to deal with the tacit, practical, back-
ground understandings which enabled linguists’ work at the same
time as they were disconfirmed or thrown into doubt by confronta-
tions with radically different ways of being human in zones of
colonial contact.
Taken-for-granted “expectations,” “understandings,” and “col-
lective practices” are of interest here because linguists, as literate
members of societies “back home,” extended them in confronta-
tions with alien languages. To become literate persons they were
drawn into certain institutions, were members of particular groups,
and had investments in the same textual traditions, religious and
secular, which legitimized colonial projects. They shared habits of
thought and belief with others who found what they wrote mean-
ingful and useful for their own purposes, because it partook of a
broader, shared sense of how writing and language use “usually”
and “should” go.
To trace these social imaginaries as enabling and shaping the
work of linguistics in zones of colonial contact, it is useful to fore-
ground linguists’ practices of literacy. “Practice” has become a term
of art used by social scientists to emphasize the hidden power of
habitual dispositions as shapers of much that is “automatic” and
prereflective in human conduct, and so of textures of social life
which lie beyond the purview of conscious, “rational” thinking.5So,
too, for linguists to impose their habits of reading and writing on
alien talk they also had to transpose and transform their literacy-
related values, habits, and beliefs.
Practices of literacy have become a major research concern in a
number of fields in recent years, but in this book I make use of two
broadly opposed views of its nature. One, broadly aligned to the
12 The Linguistic in the Colonial
empirical stance to language outlined above, is the “common
sense” view that literacy is a neutral means for representing char-
acteristics of languages which are independent of it. Literacy then
has the same essential logic and character in all times and places
as a “technology of the intellect.” I explore this idea further in
chapters 2 and 3 as one of the grounds for the work of colonial
linguists. The other view, more critical or “ideological,” as Brian
Street (1984) puts it, frames literacy’s “particular manifestations” as
they are embedded in “activities, events, and ideological constructs”
(Besnier 1995:5). This view of literacy broadly parallels the view of
language I sketched above as always bound up with contexts, and
as something to be understood through particular acts and prod-
ucts, always “interact[ing] with ideologies and institutions to shape
and define the possibilities and life paths of individuals” (Baynham
1995:71).6
Both of these views can help to trace linguists’ understandings
of their work, and the practical and ideological needs they addressed
within larger colonial projects. Both help consider how linguists’
work had meanings and uses which outstripped their own pur-
poses and understandings, taking on lives of their own when, as
texts, they circulated among different readers in different societies,
and served different projects.
Linguistics between Faith and Civilization
The practices of literacy of interest here were bound up with textual
traditions which lent legitimacy to colonialism more generally: of
faith colonialists could seek to spread among pagans, and of civili-
zation they sought to bestow on primitives. From the 16th through
the 20th century, the deep textual roots of traditions of salvation
and enlightenment were constantly invoked to justify brutal
regimes, invasive projects, and pervasive hierarchies. So they figure
in my historical sketch insofar as they were ways of making the
work of linguists useful and authoritative. Whether they believed
they were redressing God’s curse on Babel, achieving minor ver-
sions of the miracle of Pentecost, or spreading the fruits of reason,
linguists could refer their dealings with exotic particulars to tran-
scendent values grounded in traditions and practices of literacy.
The Linguistic in the Colonial 13
Such ideals can only be durable if they are continually reinvented
and adapted across eras and colonial projects, and so also in the
work of different colonial linguists. Under the rubric of faith, then,
I trace the shifting purposes and values of linguistic descriptive
work as it was done by Catholic missionaries in the 17th century
(discussed in chapter 2), Protestants in the 19th century (described
in chapter 5), as well as their postcolonial counterparts (chapter 7).
These linked sketches of practices of literacy also help tie the work
of these linguists at the peripheries of power to the broadest con-
tours of change in Europe, where shifts in these same practices of
literacy were bound up with the character of colonial projects from
before the Reformation, through the industrial age, up to and past
the era of high imperialism.
Faith drew so many to the work of conversion that missionaries
count as the group which has produced the single largest body of
knowledge about linguistic diversity around the world. I discuss
two of the earliest of these projects in chapter 2, with a few of the
dry details which make this kind of writing such a trial to read. But
by drawing on work by friar linguists in two colonial settings –
speakers of Nahuatl in Mexico, and Tagalog in the Philippines – I
try to gauge the capabilities of these men of letters to shake their
ears and minds loose from habits of thought to engage the linguis-
tic face of colonial otherness.
But to the extent any such work is approached with habits of
reading as much “to hand” and “natural” for us as those of the
linguists who wrote them, it is difficult to develop a sense of how
radically different the world of those friars was four centuries ago.
Because it is important to try to bracket those habits of thought, a
good beginning point for this story is the era when the practices of
literacy which were their inheritance first developed. That is why
my story begins with events dating from well before the dawn of
the colonial era, when Charlemagne ruled in God’s name in pre-
modern western Europe.
During the 18th and 19th centuries, many linguists were Protes-
tant missionaries with fundamentally different projects of the spirit
that were enabled by different beliefs, technologies, and forms of
colonial power. Their linguistic work needs to be described not just
in the context of changing forms of faith, but of technology and
especially the first European science of language.
14 The Linguistic in the Colonial
Philology’s Linguistic Images
Linguists in general, and missionaries in particular, partly described
and partly created print-literate forms of colonial languages thanks
to the ideological and intellectual support offered by comparative
philology, a field which developed over the course of the 19th
century. That field was shaped by ideas and values which later
became ideologically and intellectually important for the work of
colonial linguists, which is why I discuss philology’s development
in chapters 3 and 4 as an important dimension of the rise of linguis-
tics, not just at colonial peripheries, but in the colonial world.
Thinking of colonialism as the global integrative dynamic men-
tioned above makes it easy to think that it emanated unidirection-
ally, and more and more forcefully, from metropolitan centers to
scattered peripheries. This is an image which can confirm the view
of civilizational progress held by those who prosecuted and most
benefited from colonial projects. But at least since Edward Said’s
critical reading of European intellectual history, Orientalism (1994),
it has become apparent that this image is itself part and parcel of
that ideology. Said was among the first to search the colonial archive
for evidence of power’s shaping effects on knowledge and intel-
lectual discourse. Reading in the manner of Michel Foucault, but
less eurocentrically,7Said argued that the great archive of European
knowledge about the Orient counted as an indirect but ideologically
crucial response to the cultural and existential anxieties which arose
from intimate engagements with colonial otherness. Reading like
Franz Fanon, but from “above,” not “below,” Said worked to dem-
onstrate that what seemed a stable, “objective” body of European
knowledge of the Orient was in fact the history of engagements
with the anxieties of human diversity, “. . . a continuous interpre-
tation and reinterpretation of their differences from ‘us’ ” (Said
1994:331–332).
The linguistic face of human diversity had always figured in
debates about questions of human origins, and so knowledge of
languages written by linguists abroad could travel back to Europe
and into debates which joined questions of human origins
and diversity with ideologies of European superiority. The role
of linguistics in this intellectual and ideological project is so
The Linguistic in the Colonial 15
important that I devote chapter 3 of this book to two figures whose
writings in the late 18th century shaped the broader meanings and
uses of language difference in the 19th century. I discuss there early
strategies devised for what Said describes as excavating and appro-
priating the “deep” history of colonized peoples, focusing on the
great linguist and colonial officer of India, Sir William “Oriental”
Jones. Though Said knew Jones’ work, he passed over this foun-
dational example of how a language-centered picture could be
developed of linguistic, textual, and civilizational “decay” among
a literate but inferior people. So, too, Said largely passed over
philology’s development as a field which Jones inspired. This is
interesting here because it poses a conundrum for Said’s broader
account of orientalism’s colonial origins.
Chapter 4 discusses the ways that philology, which was to become
a cornerstone in the edifice of Orientalist knowledge, developed in
Prussia at a time when it was the one major European power not
yet invested in colonial projects abroad. Later in life, Said reproached
himself for having passed in silence over the work of German-
speaking intellectuals in Orientalism, noting that during most of the
19th century Prussia had no “protracted sustained national interest
in the Orient” (Said 1994:19, emphasis added). But it is interesting
that German-speaking intellectuals were just those Europeans who
were developing dominant ideas about language, history, and iden-
tity which served their nation-building project at home, rather than
a project of colonial power abroad.
It is worth emphasizing, then, that a European science of lan-
guage helped to legislate national difference in Europe as well as
human inequality in an imperial world. By drawing on recent, intel-
lectual historical work which partakes of Said’s own critical human-
istic spirit, I use this book’s middle chapters to frame the development
of the science of language which, once established as a very
European academic discipline, could be a source of ideological and
intellectual support for colonial linguists abroad.
Colonial Regimes of Language
It is important to show how philological images of the past shaped
the work of colonial linguists in the 19th and early 20th centuries,
16 The Linguistic in the Colonial
which I sketch in chapters 5 and 6. These missionaries sought to
describe languages in order to create literate colonial subjects, which
meant that they transposed practices of literacy from European
nations to “native” communities. To describe the work they did and
their “fields of operation” abroad, I draw on Imagined communities,
Benedict Anderson’s (1991) influential account of print-based lit-
eracy, senses of community, and the grounding of both in senses of
national language and territory. This helps to show how missionary
linguists created written images of languages in and with images
of their speakers, mapping both for colonial regimes of power.
Philological techniques and ideas helped them to appropriate lan-
guages descriptively, before “giving them back” to speakers in
print-literate form. I demonstrate this intimate mode of entry into
local lives with sketches of a few such projects undertaken in
sub-Saharan Africa, where missionary linguists created the unitary
languages they needed as instruments of religious conversion.
Comparing neighboring projects helps to show how different lin-
guistic communities could arise from this work, once colonial lin-
guists made languages into means for “conceptualizing, inscribing,
and interacting [with speakers] on terms not of their own choosing”
(Comaroff and Comaroff 1991:15).
Chapter 6 turns to two quite different projects of fully imperial
linguistics, each beginning at the turn of the 20th century, and each
responding to the political and economic demands of two imperial
structures. In territories of the Congo ruled by Belgium, and the
East Indies ruled by the Netherlands, imperial integration made
linguistic diversity a growing impediment, such that linguists were
delegated to partly describe and partly create non-European lan-
guages of European power: Swahili and Malay were languages of
state in the sense that neither had many native speakers on either
side of the imperial divide.
On one hand, these two projects of colonial linguists stand
out as intellectual and technical accomplishments, and as the
clearest evidence of European capacities to impose unity in the
face of diversity. Swahili and Malay became effective instru-
ments of power only thanks to the expertise of those who fixed
them descriptively, and the capacity of colonial institutions to
transmit those languages, along with broader literacy practices,
among colonial subjects. Those projects now have legacies in
The Linguistic in the Colonial 17
postcolonial nation-states (the Congo, Kenya, Tanzania, and
Indonesia) where Swahili and Malay (now known as Indonesian)
are spoken as national languages.
On the other hand, as Johannes Fabian has shown, these intel-
lectual and technocratic projects were shot through with tensions
of empire which could not be resolved, but only disguised and
displaced. By framing Fabian’s insightful account of the history of
Swahili in the Congo with an eye to the work of imperial linguistics
in the Netherlands East Indies, I show how Malay also figured in
indirect responses to the broader dilemmas of imperial identity, and
so had unintended uses in and consequences for imperial power.
Bringing the linguistics of Swahili and Malay together helps gauge
the broader limits of the power which colonial linguists served.
Though they could describe and impose languages of power, they
could not control their uses: colonial subjects, once “given” those
print-literate languages, pirated them for their own purposes and
projects of resistance.
Making this book fit the broad historical contours of colonial
history brings me, in chapter 7, to the questions posed above about
colonialism’s afterlife in a globalizing present. That is why I
conclude by considering the integrative dynamic of what now
counts as globalization, and conditions of linguistic diversity in
marginal groups’ ongoing engagement with the encroaching
power of postcolonial nations. To read the “flattening” effects of
globalization on language diversity, I draw on earlier chapters to
foreground postcolonial linguists’ continuing involvement with
broader questions of power and identity, obliging those who do
that work to consider new versions of older questions about its
meanings and uses.
Precolonial Prelude: Alcuin’s Literacy
I introduced the paired notions of “practices of literacy” and “social
imaginaries” above as a way of beginning to read linguistic texts
within the lives, social backgrounds, and interests of those who
wrote and read them. The idea of “practice” helps in this way to
foreground the embeddedness of what people do in communities,
political dynamics, and institutional contexts, which is why it is less
18 The Linguistic in the Colonial
useful to define it in abstract terms than to illustrate its application
in particular situations. Recognizing and drawing out the impor-
tance of practices of literacy from their multiple embeddings is best
done, in turn, by identifying conditions in which they conflict,
differ, and change. During one such period, transformations of
religious and secular authority also gave rise to practices of literacy
that served the linguists I describe in chapter 2 six hundred years
later. Their work of conversion in the 16th century in this way was
rooted in textual traditions and practices of literacy of an older
project of faith and power.
Histories of colonialism and linguistics often begin with the
publication in 1492 of Anton de Nebrija’s grammar of Castillian
(Spanish) for two reasons: it was the first printed grammar of a
“living,” spoken language, and its author reportedly commended
it to Queen Isabel with the infamous observation that “[l]anguage
was always the companion of empire. . . . language and empire
began, increased, and flourished together” (Trend 1944:88). No
clearer charter could be given for linguistics as a part of any colonial
enterprise, especially given that Columbus embarked that same
year on his first voyage of discovery at Queen Isabel’s behest.
But this remark is also significant for the ways Nebrija drew on
the past to speak of the future. His humanist contemporaries would
have recognized echoes in his observation of an earlier famous
linguistic description, written in and about Latin – still a language
of power – in 1441 (Lorenzo Valla’s Elegantiae linguae Latinae [1488]).
Among the empires Nebrija may have had in mind with his remark
was that of Charlemagne (742–841 ce), the illiterate emperor who
transformed Christian practices of literacy, and through them the
religion itself.
When Charlemagne was crowned in 800 ce, Latin’s unity as a
language centered more on its textual than its oral forms, because
“literacy” at that time was more than the skill of reproducing speech
sounds in accordance with written letters. To be literate then was
rather to be able to convey with speech the meanings of words
written in characters for people who did not necessarily understand
them unless they were put “in their own terms.”
A Latin text in standard international orthography could be read
aloud to listeners (in church, in law court, in the marketplace or for
The Linguistic in the Colonial 19
literary entertainment) in whatever form and with whatever modifi -
cations were needed to be comprehensible. (Smith 1999:73)
This practice of literacy, fundamentally different from Nebrija’s
and our own, involved no clear line between the activities we call
“reading” and “translating”: pronouncing a text on the one hand,
or recasting its meaning in some other vernacular speech on the
other. Before Charlemagne, “[s]pelling out of the letters that cor-
respond to the sounds of a long-dead language” was less important
than the “transformation of the lines into their own living speech”
(Illich and Sanders 1988:60). These older practices of literacy also
differed from ours in that they required no distinction to be drawn
between the Latin language on one hand, and what we call the
Romance languages (French, Spanish, Italian, Catalan, Rumanian)
on the other.
Charlemagne took steps as God’s servant to purify religious
practice and textual traditions by doing away with these shifting,
plural voicings of the Word. He saw that a new way of reading was
needed so that texts, the doctrine they conveyed, and the religious
practices they supported were fixed and stable. The clergy were to
act as mediators between God and the faithful by transforming
silent letters into audible speech in the same ways, by the same
rules, thus guarding religious truth from the dangers of translation,
and distortion, in any vernacular, “rustic” manner (Illich and
Sanders 1988:61).
Alcuin the Scot, Charlemagne’s servant in this matter, was a
scholar-priest from Northumbria, England, who came to this task
as an outsider; because he did not speak any of the Romance lan-
guages (lingua Romana) natively, he was free of habits of speech
which might skew his work to reduce Latin to uniformity.8(In this
regard his “outsider” position as a linguist anticipated that of many
of the linguists I discuss in this book.) The implications of his work
for the politics of religion were not lost on the Church, as became
clear at the Council of Tours. When Charlemagne urged the gath-
ered bishops to accept his reforms some resisted, appealing to the
Pentecostal miracle; some likely foresaw that the ascendance of
letter over voice, and repetition over translation, would transform
the priesthood. As local practices of literacy and ritual became
delegitimized, priests became progressively distanced from their
20 The Linguistic in the Colonial
audiences, as were texts and doctrine which became opaque to all
but the literate few.
Alcuin the Scot is an important figure for this story about colonial
linguistics in two ways. First, his work demonstrates the power of
practices of literacy, which is thrown into relief by the ways they
change in and with the meanings, uses, and authority of textual
traditions. Second, with his work were promulgated the practices
of literacy which were inherited and adapted by Valla, Nebrija, and
other scholars in Spain, although these only became fully estab-
lished and authoritative as late as 1200 ce and adapted for writing
Castillian speech as late as 1252 (Smith 1999:74).
Perhaps Nebrija’s pronouncement marked his recognition that
power was best served by a “symbolically effective, uniform, impe-
rial, dead language” (Illich and Sanders 1988:59). But it was his
scholarly work on Latin which proved to be of greater practical
importance for the friars who embarked for the New World. They
engaged in projects of empire different from any he might have
imagined, and with languages so different from their own that they
posed challenges to faith, imagination, and intellect alike.
Notes to chapter 1
1 Pratt’s phrase “zones of colonial contact” is especially appropriate in a
book centered on language, since she took it over from “contact linguis-
tics,” the structure-centered study of the dynamics of language
change.
2 That “colonial” power is commonly thought of as being exercised at a
distance can be seen in common usage. Denmark’s brief rule over
islands in the West Indies is more appropriately called “colonial,” for
instance, than its centuries of dominance over nearby Iceland or
Norway, if not more distant Greenland; Native Americans were victims
of projects of violence and genocide as vicious as any in the world, but
these are not usually called “colonial” except by scholars, who will
usually qualify it with the label “internal colonialism,” which is also
applied to English domination of neighboring Ireland. In this respect
the conquest of North America differed from American imperial rule
in the Philippines and Puerto Rico after the Spanish–American war.
Common use also suggests a eurocentric slant to “the colonial:” it seems
a bit awkward as a label for the Moorish occupation of the Iberian
The Linguistic in the Colonial 21
peninsula on the one hand, or Abyssinia’s regime of power over the
neighboring Oromo people, established at the same time and in broadly
the same ways as those of Europe’s powers in the rest of Africa in the
19th century.
3 Broad explorations of these issues from an ideological point of view
can be found in Kroskrity (2000), especially Judith Irvine and Susan
Gal’s essay “Language ideology and linguistic differentiation.”
4 For broadly similar arguments from vantage points still closer to
linguistics, see Roy Harris’s The language makers (1980) and Language,
Saussure, and Wittgenstein: how to play games with words (1988). See also
J. Joseph and T. Taylor’s collection Ideologies of language (1990).
5 Two major figures who developed notions of “practice” are Pierre
Bourdieu in works like Outline of a theory of practice (1977) and Michel
de Certeau’s The practice of everyday life (1984). Bourdieu argues in his
work that linguists’ portrayals of language, besides being inaccurate,
have misled those who have used them as models for social conduct
more generally. Other important work which brings notions of discur-
sive practice to questions of language and power is Grillo 1989.
6 Of the literature on literacy, which has grown tremendously in the last
15 or 20 years, especially in history and education, I mention only
Goody 1986 as a convenient setting out of the core ideas Street critiques
as the “autonomous” model of literacy; Shirley Heath’s (1983) early,
influential ethnography of literacy in the contemporary United States,
which helped to stimulate much ensuing research; and a recent
synthetic review of that work by Collins and Blot (2003).
7 One could add to important work (for instance, Mitchell 1988, Stoler
2002) critiquing Foucault’s (1970) eurocentric focus a commentary on
his account of the idea of language, framed with an eye to the early
work of colonial linguists I discuss in Chapter 2. That is a task beyond
the scope of this book.
8 A much more detailed discussion of Alcuin’s work can be found in
Wright’s (2003) sociophilological study of late Latin.
Chapter 2
Early Conversions, or, How
Spanish Friars Made
the Little Jump
It all goes back to conversion, Father, a most ticklish concept and a
most loving form of destruction.
Louise Erdrich, Last report on the miracles at Little No Horse (2001)
In the 16th century, mendicant friars joined explorers and conquer-
ors in the New World, and from there traveled across the Pacific to
islands on the far side of the world. They did not understand theirs
to be the work of enslaving, murdering, or displacing those they
called in both places indios; it was rather to make them Christians.
Because pagan ways of speaking were their chosen ways of entering
pagan hearts, these friars required of themselves the difficult work,
early and late, of converting living speech to alphabetic writing. I
discuss here some of their engagements with alien languages, parts
of the work of loving destruction which were foundational for
European traditions of language study, and as such are still intel-
ligible four centuries later.
The main goal of this chapter is to show how the work of fixing
pagan speech in writing helped to locate pagan speakers in natural,
spiritual, and political hierarchies. I do this by discussing two of
these projects, carried out for similar purposes but in different zones
of colonial contact. From these I need to draw just a few of the fine-
grained details which friar linguists themselves took so seriously,
exploring their efforts to bracket habits of speech, categories of
knowledge, and so to use and understand alien languages as did
their native speakers. Only in this way could they hope to capture
Early Conversions 23
strange talk’s properties in their written texts, and then appropriate
it for their own purposes in the “New France” of North America,
“New Spain,” Portugese South America, and many other places in
the early colonial era.
This chapter deals secondly with this work’s broader meanings
and uses in early debates about the New World’s human-like
creatures, as it emerged from and figured into the Conquest’s polit-
ical and ideological grounding. Against the backdrop of early
Spanish imperial ideology, these descriptions can be read as evi-
dence of what friar linguists themselves recognized to be languages
of humans who were to be saved for God, if not from other
Europeans.
This chapter’s third major purpose is to foreground the simplify-
ing strategies which friar linguists developed to represent complex
worlds of living speech with simple written images. In different
zones of colonial contact, and engagements with speakers and
speech, they encountered different challenges which we can read
into work that served always and everywhere in lovingly destruc-
tive projects of conversion. In this way the reductive force of their
linguistic descriptions can be understood as part of broader appro-
priations of pagan languages in order to supplant pagan ways
of life.
It helps to frame these writings by friar linguists on Nahuatl and
Tagalog opportunistically, as if they were produced in zones of
colonial contact which mirrored each other. In Mexico, they recog-
nized and used Nahuatl as the language of the dominant, civilized
elite in a complex society, which nonetheless lacked anything they
were willing to recognize as Nahuatl practices of literacy. In the
Philippines, friar linguists saw Tagalog as the language of “primi-
tive” people living in simple societies, but at the same time could
not avoid recognizing that they also possessed indigenous practices
of literacy.
To deal with this work in its own terms, with an eye to empirical
particulars, I also opportunistically focus here on one speech sound
which friar linguists were obliged to deal with in their descriptions
of Nahuatl and Tagalog alike. This sound, called in Spanish the
saltillo, or “little jump,” was one they had to learn and write en route
to appropriating the languages in which it figured. But their strate-
gies for doing so differed as did the broader colonial projects in
24 Early Conversions
which their work was situated, and the broader projects of conver-
sion they undertook.
Understanding the “Barbarian”
It might be as challenging for us to imagine the conundrums Euro-
peans confronted after the discovery of the New World as it was
for them to make sense of the distant, human-like creatures which
were encountered there. Politically and religiously fraught ques-
tions about humanity, “theirs” and “ours,” were constantly entan-
gled with European projects of conquest, questions which from the
beginning joined ideas about the nature of humans and the nature
of language.
Early accounts of these early explorations have led two critical
readers to suggest that at the outset of this encounter, Europeans
were unable to categorize their experiences of peoples of the New
World in stable, consistent ways. Tvetzvan Todorov, for instance,
finds in Christopher Columbus’s narrative of his first voyage a kind
of split vision of an alien human condition: he could either “acknowl-
edge [native speech] as a language but refuse to believe it [was]
different” from his own, or he could “acknowledge its difference
but . . . refuse to admit it [was] a language” (Todorov 1984:29).
Other records of early explorers led Stephen Greenblatt to a
similar interpretation of an unstable binary perception of human-
like creatures and the language-like sounds they produced. He sees
the shaping effects of a will to power in shifting perceptions of the
New World’s peoples. Europeans had to “peel away and discard
like rubbish” (Greenblatt 1990:32) what might otherwise be incon-
venient linguistic evidence of the human nature, and so had recourse
to one side or other of a rigid binary distinction. If speech-like
sounds were judged to fall beyond the pale of language, then the
creatures producing them fell beyond the pale of humanity, and
issues of commonality need not enter into any calculation of advan-
tage and obligation. But if those sounds were to be counted as
instances of a human language, however opaque and meaningless,
then their producers could be immediately assimilated not just to
the category of human, but to European interests.
Early Conversions 25
The cruelest example of this refusal of difference in the service
of power was a document called the Requerimiento, which Spanish
Conquistadores carried to the New World. The Requerimiento,
drawn up (in Castilian) at the behest of King Ferdinand of Castille
in 1513, was in the tradition of charters which the Vatican had
begun to issue ten years earlier, requiring soldiers of Christ to
destroy those who refused His love (Moors and Jews on the Iberian
peninsula, later los indios in the New World). In this way the
Requerimiento extended the tradition of the Crusades, and Cas-
tille’s own recent reconquest of Granada, a battle for Christ closer
to home which became a model for Cortez’s expedition to Mexico
in 1519, and Pizarro’s to Peru, where he murdered king Atuahalpa
in 1532.
The Requerimiento is worth discussing here because it was bound
up with practices of literacy which seem bizarre enough now to
require our own imaginative engagement with unfamiliar ideas
about power, writing, and speech. The document was to be read
aloud in the presence of los indios so as to inform them that, through
that act of reading, they were bound either to submit to the king of
Spain and the Christian church, or suffer consequences which the
reader of the document went on to describe. What seems a strange
mixture of duty and cynicism, at least for us, depends on a refusal
of linguistic difference: the enabling presupposition was that (pro-
totypically male) creatures who heard it but did not understand
its import were therefore subhuman, and so could be subjugated
forthwith:
[w]e shall take you and your wives and your children, and shall
make slaves of them, and as such shall sell and dispose of them as
their Highnesses may command; and we shall take away your goods,
and shall do you all the mischief and damage that we can, as to
vassals who do not obey, and refuse to receive their lord, and resist
and contradict him; and we protest that the deaths and losses which
shall accrue from this are your fault, and not that of their Highnesses,
or ours, nor of these cavaliers who come with us. And that we have
said this to you and made this Requisition, we request the notary
here present to give us his testimony in writing, and we ask the rest
who are present that they should be witnesses of this Requisition.
(Helps 1856(1):361)
26 Early Conversions
These readings had meanings reaching far beyond anything Nebrija
imagined when he commented on language and empire going
together.
Denying los indios any presence as speaking and understanding
beings made it easy to dispense with their physical presence for the
reading of the Requerimiento as well:
[i]t was read to trees and empty huts . . . Captains muttered its
theological phrases into their beards on the edge of sleeping Indian
settlements, or even a league away before starting the formal attack . . .
Ship captains would sometimes have the document read from the
deck as they approached an island. (Hanke 1949:33–34)
The Requerimiento can be taken as a kind of early prototype for
linguistic asymmetries of colonial power: the nonintelligibility of
speech provided sufficient grounds for subjugating them because
it was evidence not of their difference, but their deficiency. Given
the undeniable and self-evident meaningfulness of “our” speech,
“their” inability to understand it revealed a more deeply flawed
human condition. The grim absurdity of this logic was clear enough
to some European witnesses; friar Bartolomeo de las Casas, who
witnessed the reading of this document, found himself not knowing
whether to laugh or cry.
In Spain the question of language was central for resolving the
status of los indios so as to legitimize the Conquest. On his return
there to decry to the king cruelties carried out in his name, de las
Casas focused on this disputed category, protesting the willful
refusal of linguistic difference and insisting that this condition of
unintelligibility was at base mutual. Though natives of the New
World might be called barbarians, he argued, that category is not
absolute but relative:
A man is apt to be called barbarous, in comparison with another,
because he is strange in his manner of speech and mispronounces
the language of the other . . . But from this point of view, there is no
man or race which is not barbarous with respect to some other man
or race. . . . Thus, just as we esteemed these peoples of these Indies
barbarous, so they consider us, because of not understanding us. (De
las Casas 1971:166)
Early Conversions 27
“Barbarian,” known to classically trained scholars of the time
from Greek, was a word which was originally onomatopoeic,
reproducing sounds which crude beings used (as the civilized
Greeks saw it) in poor imitation of their own proper speech.
Stammering, meaningless “barbar” sounds were then a sign of
inferiority if not evaluated relativististically, as de las Casas did
by invoking the words of no less a figure than St. Paul: “If I
then know not the meaning of the voice, I shall be to him that
speaketh a barbarian, and he that speaketh will be a barbarian
unto me’ (I Corinthians, 14.11). Behind the authority of the New
Testament, in turn, lay that of the Old: the story of Babel and
the confusion of tongues in Genesis, which explained linguistic
difference as sinful humanity’s common heritage. Los indios, like
other descendants of those who fled the tower’s fall, were inheritors
of the common human curse. This was the original condition
invoked by one friar linguist to describe his purpose: “to restore
in part the common eloquence of which we were deprived by
the arrogance and pride of that building” (quoted in Pagden
1982:181).
To refute de las Casas’ relativistic understanding of “barbarian”
speech and intelligibility, churchmen who had the king’s ear drew
on a different intellectual tradition, giving that word another
meaning. They responded to Ferdinand’s request for opinions on
his right of conquest by arguing that fully human creatures are
distinguished by more than a simple capacity for speech. The fully
human condition is one of sociality which, in the classical tradition
of city-dwelling Greeks, was understood to require a capacity for
linguistic eloquence. Only through such eloquence could humans
be induced to join each other in civilized conditions of hierarchy
and comity. Here the churchmen invoked Cicero’s famous argu-
ment that where eloquence is lacking, bonds of association can only
be forged by brute force among creatures who fall short of a truly
human condition (Pagden 1982:21).
Far away, in Mexico and Peru, the Conquistadores could be in
no doubt that they were in fact destroying civilizations and subju-
gating civilized peoples; as the friar linguists I discuss later saw
very clearly, in these societies people were distinguished by their
capacities for eloquence. But reports about pagan civilizations and
languages traveled poorly and had little purchase on those who
28 Early Conversions
chose to believe otherwise, as friar Bernadino de Minaya discovered
on his return from Peru. He went
begging, to Valladolid, where I visited the cardinal and informed him
[about] the Indians’ language . . . [and] their ability and the right they
had to become Christians. He replied that I was much deceived, for
he understood that the Indians were no more than parrots. (Hanke
1937:84)
It is easy to think that friars who recognized los indios as objects
of Christian love, and witnessed their subjugation in the Conquest,
brought a sense of urgency to their engagements with pagan lan-
guages. Their descriptions could count as massively detailed evi-
dence of capacities for well wrought speech, and so for speakers’
rights both to be converted to the Faith and protected by the king.
In the dedication of his 1560 lexicon of Quechua, the language of
the Incan empire, Friar Domingo de Santo Tomas addressed King
Philip II to point out how “clearly and manifestly . . . false is the
position of which many have sought to convince your Majesty, that
the people of Peru are barbarians.” On the contrary, Quechua was
adorned with the regularity of declension and the other properties
of the noun, and with the moods, tenses, and persons of the verb. In
short, in many respects and ways of expression, it conforms to Latin
and Spanish. The language is civilized and abundant, regular and
ordered by the same rules and precepts as Latin, as is proved by this
grammar. (Quoted in Macormack 1985:448)
A grammar itself counted as direct evidence of a civilized condi-
tion, and could be used to legitimize the work of conversion, giving
pagans knowledge of Christ in words they understood. So argued
a militant French Franciscan in a 1533 letter to Ferdinand’s succes-
sor, Charles V, which chastised those who
would not take the trouble of learning their language, and did not
have the zeal to break that wall to enter their souls and search with
candles for the wonders that God works in their hearts . . . [who
should] now be silent and seal their mouths with bricks and mud.”
(Quoted in Tavárez 2000:22)
Early Conversions 29
Between 1533 and 1620, the kings of Spain (Charles V and after
him Philip II) ordered that los indios be given Castilian, the proper
language of all their subjects. Their decrees were ignored by the
friars because they were impractical, and did not serve their own,
higher purpose. They knew los indios to be children of God who had
not yet been corrupted by the world, and so could only be saved if
they were insulated from the pride and greed of other Europeans.
So they understood that
for the Mexican to become a true Christian he had to break entirely
with his past, except, and this is an important fact, in his language,
because it was clearly understood that to become a true Christian he
did not at all have to become a Spaniard, that it was perfectly allow-
able, and even recommended, that he remain a Mexican. (Ricard
1966:288)
With this observation, Ricard identifies that aspect of the
friar’s work of conversion which might fit least well with his own
influential chronicle of these missionaries’ “spiritual conquest” of
Mexico.
Triangulating on Pagan Languages
The demands of faith and the realities of power both led
[e]very missionary that came to the new world [to think] it his duty
to write at least one grammar (arte de la lengua) during his life. By the
middle of the sixteenth century there were enough grammars and
prayer books written in various dialects to enable the clergymen to
study them in the seminaries. (Campa 1931:549)
These missionaries saw los indios as objects of Christ’s love, and
heard their talk as writable language which they could turn to His
purposes: they could not love what they did not understand, and
so needed to find a fit between alien speech and their own catego-
ries of meaning. Only by appropriating speech with their own prac-
tices of literacy could they give a language back to its speakers as
the medium of Christian discourse, and part of the gift of the Faith.
30 Early Conversions
Missionaries could work in this way to redress what de las Casas
himself acknowledged was los indios’ “lack [of] a literary language
which corresponds to their maternal idiomatic language, as is Latin
to us, and thus [they] do not know how to express what they think”
(de las Casas, quoted in Pagden 1982:130).
The Conquistadores modeled their exploits on the Crusades and
the Reconquista of Spain; trade developed across the Atlantic thanks
to expertise of older Mediterranean mercantile communities (see
Laiou 1998). So too missionaries drew confidence and techniques
from their own textual traditions. Though not practicioners of what
we would call a “linguistic science,” their work was rooted in the
traditions of knowledge of what they called (in Latin) scientia, fol-
lowing Aquinas’ observation that “[w]e all have to learn to interpret
what we see, and this can only be achieved through the use of
books” (Pagden 1982:130).
Nebrija influenced these first friar linguists less through his
famous grammar of Castilian than through his earlier grammar of
Latin, written in Latin, in 1481 (the Introductiones Latinae). His expla-
nation of Latin’s grammatical categories in that book became indis-
pensable for friars’ first attempts to generalize about the New
World’s vastly different languages. They
understood that the traditional Latin grammatical framework needed
to be modified to accommodate them, but they faced a dilemma. On
the one hand, if they forced these languages into the Latin mould
they knew that they would be distorting the facts; but on the other
hand, if they abandoned that framework altogether they would run
the risk of finding themselves in completely uncharted territory.
(Perceval 1999:19)
But the grammatical knowledge they were applying was embed-
ded in broader practices of literacy, and in traditions of meaning
and belief which invested their grammars with values. Vincent
Rafael’s study of early descriptions of Tagalog, which I consider in
more detail later in this chapter, helps to develop a sense of the
hierarchy of literacies and languages which friars brought to their
work in the New World as well.
Rafael suggests that Tagalog could become a descriptive object
only as it came to be located in hierarchical relations to Latin as a
Early Conversions 31
descriptive model, and Castilian as a descriptive instrument. Unlike
Perceval, who emphasizes Latin grammar’s intellectual usefulness
for these friar linguists, Rafael points to the broader meanings of
Latin in textual embodiments of sacred truth (thanks to the practices
of literacy established by Alcuin centuries before). Borrowing from
Benedict Anderson’s observations about social imaginaries in a
world of limited literacy, Rafael suggests that Latin counted for
these friars as a “Truth language,” one which stands apart from
mundane speech and life as a “privileged system[s] of [written]
representation” (Anderson 1991:14). Latin partook, part-for-whole,
of the sacred power of the messages it conveyed; it was not just a
vehicle of meaning but a kind of emanation of the source of meaning,
God’s “superterrestrial order of power” (Anderson 1991:13).
Castilian, for its part, served as a language which could mediate,
textually and ideologically, between text-based sacred Latin and
oral pagan speech. Rafael suggests that in the work of description
Castilian served to deploy a conceptual apparatus or metalanguage
grounded in Latin. More than a (spoken) language of secular power,
Castilian took on legitimacy from its proximity and similarity to the
language of transcendent truth. Rafael reads this authorizing sym-
bology of faith and power from work of friar linguists in the Philip-
pines which I discuss later, but his insight is also suggestive for
considering other zones of contact in which friar linguists dealt
with pagans and their talk. However well they knew, as Perceval
suggests, that they were deforming pagan languages to convert
them to writing, they knew also that these loving deformations
confirmed the fact of pagan languages’ deficiencies in relation to
“the intrinsic superiority of some languages – in this case Latin and
Castilian – over others in the communication of God’s Word”
(Rafael 1993:29).
Nahuatl’s Little Jump
Beyond broad questions of belief, power, and faith lay the
descriptive particulars which consumed much time and required
much attention for these first linguists. I consider here some of the
first, formative parts of the work of description they undertook in
the spirit of Alcuin: establishing regular relations of letters and
32 Early Conversions
voice so that fixed, inspectable written images could come to be
models of and for speech. Of interest here is the intimate zone of
contact where sounds of Nahuatl met friars’ ears before they put
pen to paper to map that language with their own alphabetic
symbols.
Walter Mignolo’s critical readings of some of their work lead him
to conclude that this painstaking work was not just shaped but seri-
ously deformed by habits of thought and practices of literacy, which
friar linguists could not recognize or bracket. For them Latin was
so fundamentally a “universal linguistic system,” he argues, that
they deployed its categories and letters normatively, not descrip-
tively. Latin’s superiority gave
[t]he letter . . . an ontological dimension with a clear priority over the
voice as well as any other writing systems. The classical tradition was
inverted, and the letter . . . had become the voice in itself, while non-
alphabetic writing systems were suppressed. (Mignolo 1995:46)
Mignolo reads in this work a kind of “violence of the letter,” as
Derrida puts it, which is of a piece with the Conquistadore’s sys-
tematic destruction of Nahuatl practices of epigraphic literacy, a
way of “writing without letters” which they neither understood nor
tolerated.1
The more intimate symbolic violence done by these linguists can
be seen, Mignolo argues, in the ways they applied Latin letters to
Nahuatl. The “one letter, one sound” principle led them to perceive
a “lack” of sounds in pagan languages. This reading of difference
as deficiency is evident, Mignolo says, from the use of a “common
and repeated expression in these early texts: esta lengua carece de tales
letras (this language lacks such and such letters)” (1995:46). He
illustrates this bias by quoting from a 1645 grammar of Nahuatl
(Arte de la lengua mexicana) by the Jesuit missionary Horatio Carochi,
who began by observing that “[t]his language is written with the
letters of the Spanish alphabet, although it lacks seven letters, which
are b, d, f, g, r, s, and j” (Carochi 2001[1645]:19). So profound and
natural was the assumption of Latin’s universality, Mignolo sug-
gests, that in the very next sentence Carochi could contradict himself
by urging care on his reader in learning to pronounce the Mexican
language.
Early Conversions 33
But a bit further on in the text Carochi seems less blinkered by
orthographic habit, because he uses the word “letter” more flexibly
and descriptively. He goes on to note that just as Mexican (i.e.,
Nahuatl) lacks “letters” found in Latin, so too it has a “letter” which
Latin lacks:
similar in pronunciation to z and c, but it is pronounced more strongly
and corresponds to the Hebrew letter tsade; in this language it is
written with t plus z, as in Nitzàtzi, I shout; Nimitzno¯ tza, I call
you; but it is a single letter, although written with two. (Carochi
2001[1645]:19)
In this passage Carochi alternates between two senses of the word
“letter” which modern linguists can recognize. It is a rough and
ready metalinguistic vocabulary which is less literal, as it were, than
Mignolo suggests.
Carochi can describe “a single letter” as being “written with two”
without contradicting himself because his first use of “letter” refers
to a recurring pattern or type of sound which he has identified as
common in Nahuatl speech, but which is dissimilar from any which
are ordinarily (and conventionally) associated with Latin orthogra-
phy. His second use of the word “letter” names a symbol which he
stipulates will serve (conventionally) as the visual counterpart of
that sound in his description. Because he alternates between these
two senses – “speech sound” and “visual symbol” – he can stipulate
that “a single letter” (a speech sound) is to be written “with two”
(a combination of characters, what nowadays is called a digraph).
Carochi in fact triangulates on this sound by appealing to his
readers’ presumed knowledge of Hebrew, which happens to be
useful in its own way thanks to spelling conventions which help
him describe acoustic properties of a sound of Nahuatl. (Elsewhere
in this work he draws on Hebrew grammar to describe Nahuatl
affixes, as had his predecessor, del Rincón, in 1595.)
More interesting here, though, is another challenge posed by a
“lack” in Latin orthography for writing a Nahuatl speech sound.
Carochi called this sound the saltillo, or “little jump,” and tran-
scribed it with an apostrophe. (Nowadays it is called in English
“glottal stop,” and can be heard in English interjections written
“uh-oh” or “uh-huh,” using hyphens to mark the point at which
34 Early Conversions
the air stream is briefly halted by closing the vocal cords, or glottis.2)
To represent this sound in this text I use the character from the
International Phonetic Alphabet, whose origins I touch on in chapter
4: the glottal stop will be represented here with the symbol to
discuss friar linguists’ dealings with the saltillo in Nahuatl, and also
in Tagalog.
Nahuatl’s saltillo was of special concern for Carochi, not just as a
fact about speech, but an object of social value for its speakers:
Whether to put a saltillo or a long accent rests on almost impercep-
tible practices, so that not even those who are very expert in this
language can manage to give the reason for the difference. Yet if it
is not observed, it will be a barbarism and a very great impropriety.
(Carochi 2001[1645]:267)
This remark is telling in several ways. First, it suggests a kind of
temporary inversion of linguistic and social hierarchies in a special
zone of colonial contact: here what is barbaric is not the speech of
los indios, but of Spaniards who fail to imitate them adequately.
Second, this comment on usage is grounded in broader concerns
with style of speech and speaker status, reflecting the broader social
hierarchies which I discuss below as having shaped missionaries’
work of spiritual conversion. But before discussing the Nahuatls
whom Carochi judged to be “very expert in this language,” and
whose authority he borrows for his own description, it is useful to
turn first to the ways friar linguists dealt with Tagalog’s saltillo,
which involved them not just with native speech, but native
writing.
Tagalog Sounds and Letters
Early explorers of Luzon and surrounding islands were struck by
the natives’ use of
certain characters [which] serve as letters with which they write
whatever they wish. . . . The women commonly know how to write
with them, and when they write, it is on some tablets made of the
bamboos which they have in those islands, on the bark. In using such
a tablet, which is four fingers wide, they do not write with ink, but
Early Conversions 35
with some scribers with which they cut the surface and bark of the
bamboo, and make the letters. (Quoted in Scott 1994:210)
That writing should exist at all in such remote locations
was as remarkable as the fact that so many people, even women,
mastered it.
These explorers could not know that they had reached a fringe
of the vast area – stretching from Afghanistan to Java, Sri Lanka to
Nepal, and beyond – where South Asian practices of literacy had
been transmitted from the end of the first millennium up through
the 13th or 14th century ce. In chapter 3 I discuss Britain’s colonial
encounters with Sanskrit, one of the oldest parts of this tradition,
but here one of the newest of these scripts is important: one of nine
known to have been in use on the islands of Visaya and Luzon,
called in Tagalog baybayin.
Missionary linguists recognized that baybayin was a technique for
imaging language in a manner similar to alphabetic writing. Both
analogize sequences of sounds in time to linear successions of char-
acters in space. This transposition can be read from the meaning of
the word baybayin, derived from a root (baybay) which, in other
words, serves to speak of ways one follows or traces a line (the
bank of a river, the side of a street, a row or line of people, and
the like).
Babayin, like related orthographies, differs from alphabetic
writing in that each of its 18 “main” characters represents not a
single sound but a syllable, that is, a combination of a consonant
followed by a vowel. This important, basic difference is worth illus-
trating. The alphabet symbol b, for instance, is read as (represents)
one sound except when pronounced as its own name (“the letter
b”), a vowel is added (so, a sound sequence which is the same as
that of the words be and bee). I need to avoid possible confusion
here, and so adapt standard practice among linguists by represent-
ing this vowel in following discussion with the letter i. By this
convention, the “ordinary” spelling of be and bee would be replaced
with bi.
In baybayin, on the other hand, there is a symbol which corre-
sponds to b,, but which is always read with a following vowel. In
its simple, unmodified form, this vowel is like that in the first syllable
of the English word “father,” which I transcribe here alphabetically
36 Early Conversions
as a. So the Tagalog sound represented in baybayin by corre-
sponds to and could be written (transliterated) with letters of
the alphabet as ba. Other syllables which begin with b, but are fol-
lowed by a different vowel, are written in baybayin by adding a
mark to the basic character . So the sound combination I tran-
scribed as bi above would be written in baybayin by adding a mark
above it: .
With two symbols to mark vowels other than a,baybayin’s basic
characters can be used as an elegantly economical way to write 54
syllables, the majority of those found in Tagalog. But Tagalog has
a few relatively uncommon syllables which not only begin but end
(or close) with consonants.3Baybayin has no characters to represent
these speech sounds – n,l, and k– when they appear after a vowel
at the end of a syllable. This means that a Tagalog word which has
closed syllables like bundul (referring to an act of bumping or
ramming violently against something) is written with the same
baybayin characters as is the word bunduk (meaning “mountain,” a
word borrowed into English as “boondocks”). The baybayin version
of these two words, transliterated directly into alphabetic writing,
is in both cases budu.
The friar linguists who first came to the island of Luzon
recognized this as point of deficiency in ways I discuss later, but
did not regard it as a reason to ignore baybayin. The earliest text in
Tagalog, the 1593 Doctrina Christiana en lengua española y tagala, sets
out basic texts and prayers translated into Tagalog, and written
with both Latin and baybayin characters. It can be read, then, as a
kind of primer to help friars learn Tagalog and baybayin together.
Missionaries also saw a way of “improving” baybayin. Thirty years
later a missionary who wrote the Doctrina in a neighboring lan-
guage, Ilokano, introduced a device he saw as remedying this
deficiency:
The reason for putting the text of the Doctrina in Tagalog type . . . has
been to begin the correction of the said Tagalog script, which, as it
is, is so defective and confused (because of not having any method
until now for expressing final consonants – I mean, those without
vowels) that the most learned reader has to stop and ponder over
many words to decide on the pronunciation which the writer
intended. (F. Lopez quoted in Scott 1994:215)
Early Conversions 37
Lopez’s device “improved” baybayin for his “most learned reader”
– non-native speakers/readers, like himself – by aligning it more
closely with Latin orthography. This mark – made in the image,
fittingly enough, of the cross – was placed below a consonant symbol
which was to be read without a following vowel. (In this way he
unknowingly reinvented an orthographic convention common to
related syllabaries used all across Asia, most famous among them
the Sanskrit virama.)
Baybayin is of special interest here, though, for the same reason
it interested these friar linguists: it offered an advantage when it
came to writing the saltillo in Tagalog. Speakers of English can hear
and produce a glottal stop easily between vowels, but these friar
linguists had to learn to hear and utter it in more difficult combina-
tions of Tagalog sounds: at the beginning of one syllable, and imme-
diately after a consonant ending a previous syllable. So for instance
they had to be able to hear and produce words which differed in
the presence or absence of a glottal stop: gab i, for instance, means
“night,” and it is just the use of a saltillo which differentiates it from
gabi, which means “taro.”
Friar linguists were at pains to master these kinds of differences,
and for this reason found that they had special uses for baybayin’s
three other symbols, which each represent a glottal stop followed
by a different vowel. The symbol , for instance, can be transliter-
ated as i. Franciso de San José took advantage of this fact when he
wrote his 1610 grammar of Tagalog, Arte y reglas de la lengua Tagala.
In that work we find him using this baybayin character in an other-
wise alphabetic, Castilian description of Tagalog to make clear dif-
ferences in pronunciation. He did this by first writing words in
baybayin, and then alphabetically. So gab i, the word meaning
“night” noted above, he wrote first as , which would be trans-
literated as ga () followed by i ( ). Then he wrote the second
word as , showing afterwards that this corresponded to ga ( )
and bi ( ).4
This is a tiny teaspoon drawn from the ocean of descriptive detail
which can be a trial to read, but it helps to illustrate two points
about this painstaking work. First, it suggests that friar linguists
were willing and able to adapt their own habits of speech and prac-
tices of literacy in order to capture fine-grained properties of alien
languages which they were, in fact, capable of recognizing. Second,
38 Early Conversions
it demonstrates how missions of faith engendered both a sense of
care and a felt need for mastery over masses of such empirical
details, to be gained by converting talk into inspectable writing.
Even the humble saltillo had to be given its place in their larger
projects of conversion.
Tracing links between these descriptive details and zones of
contact involves broader aspects of the times and places where this
work was done. So I consider next reasons for Carochi’s larger
concern with the “barbarism” of the misplaced saltillo, and then a
question about baybayin: if friar linguists found it useful and
“improvable,” why did it eventually pass out of use among mis-
sionaries and native Tagalog speakers alike?
Civilized Illiterates
I cited Carochi’s cautionary remark about the proper use of the
saltillo earlier to illustrate a kind of political and cultural paradox:
the Nahuatl were illiterate pagans who spoke a civilized language.
This was one aspect of the larger social situation in which Carochi
and other missionaries’ strategies of descriptive selection and sim-
plification served their broader work of conversion.
The Conquistadors realized early on that Mexico’s and Peru’s
peoples were not like the “primitives” Columbus encountered in
the Caribbean. Cortez likened cities he found in Mexico to Seville
and Cordoba, and was as quick as Pizarro in Peru to see that power
was to be had by usurping the native elite. He did this quite literally
by murdering the king of the Mexican Indians, razing his palace,
and building his own capital on its grounds.
In colonial era accounts of precolonical Mexico, the Aztec cult
of human sacrifice often overshadow the sophisticated tributary
systems used by noble elites to rule communities spread across a
wide territory. Drawing members of subordinate, regional chief-
tainships to their royal centers of power was one way dominant
elites could “culturally assimilate . . . [them] to the ruling group . . .
[while] at the same time [they] served as hostage” (Carrasco
1982:35).
This political strategy supported parallel linguistic and social
hierarchies: fluency in Nahuatl was a mark of proximity to, or
Early Conversions 39
membership in, high ranking social circles. Spoken by knowledge-
able elites across more territory and in more situations than any
local, provincial language, Nahuatl (like Quechua in Peru) was
recognized by the Spaniards as what they called the lengua
general of Mexico: a “principal intermediary language”5or, to use a
more recent phrase I discuss in chapter 6, a “language of wider
communication.”
Just as different dialects of Nahuatl were spoken in different
strata of Mexican society, so too different styles of Nahuatl were
used by members of different groups more or less correctly, and
more or less appropriately. Membership in the traditional elite was
indirectly but powerfully marked by mastery of distinctively elabo-
rate, allusive, prestigious genres of “lordly speech” (tecpillatolli),
achieved only through long years of training and memorization.
Fluency in this finely wrought language of poetry, oration, and
ritual was thus a privilege and badge of those at the apex of an
“extremely hierarchical society, in which an unbridgeable gap sep-
arated religious instruction for the rich and that for the poor” (de
Alva 1982:349).
These social and linguistic hierarchies gave friar linguists
strong reasons for focusing on one very specific dialect and style of
Nahuatl to overcome two obstacles to their own purposes. First, to
overcome problems of linguistic diversity in different locales of
their work of conversion, they selected the dialect which was
known in the largest number of places, contexts, and communities.
Nahuatl was the natural candidate for such a lengua general.
Second, they sought a style of Nahuatl which supported the high
seriousness of their sacred purpose, hearing in “lordly speech” an
echo of the authority of the true Faith: “[i]f the friars could usurp
the power of those words, replacing the authority of the Indian
past with that of Christianity, they would gain a significant degree
of control over Indian thought and behavior, with all the social
and political consequences that such control implies” (Burkhart
1989:11).
To convert Nahuatl into a lesser, written likeness of their own
language of authority, missionaries’ work of linguistic description
was done in close parallel with the work of textual translation of
native-language breviaries and confessional guides (Kartunnen
1982:396). This is one part of the backdrop for reading Carochi’s
40 Early Conversions
warnings against misuse of the saltillo, a “barbarism” not just for
those who are “very expert in the language” but in religious dis-
course designed to partake of the authority of “lordly speech.” This
work was descriptive and alphabetical, but also textual and rhetori-
cal, as friars sought to devise genres of speech for addressing audi-
ences of potential converts. But though they understood that reducing
Nahuatl words to writing remade them in the image of Castilian, and
as a surrogate for their own Truth language, they could not fully
recognize that this did not give them ownership of the meanings of
their Nahuatal oratory in the ears of Mexican audiences.
They intended to bring Nahuatl words, and the ideas they could
express, into closer alignment with the reality of God’s world. But
they could not simply remove or displace those words’ meanings
as they already existed in Nahautl contexts, minds, or hearts. Rather,
they produced a distinct new kind of discourse which became one
among others. So, for instance, missionaries needed to explain their
war against the Evil One, the Diablo or demonio, who had to be
named in the Nahuatl language. To this end they took over the
word tlacatecolot with the idea that it would be understood as were
the Spanish words it translated. Missionaries had to ignore or
imagine that they had simply displaced prior meanings of the
Nahuatl language, in Nahuatl communities. In practice the mis-
sionaries did not so much bring awareness of the Evil One to their
new flocks as introduce to them one more resident of the complex
realm of spirits they knew to be populated by tlacatecolot, beings
much more intimately and quite differently involved in Nahuatl
lives than the Fallen Angel.
To designate the Eucharist, missionaries similarly devised a
descriptive phrase with none of the original Latin’s associations
with a deep ritual past: speaking of the “white tortilla” conveyed
no sense of the sacredness of the body of Christ it names, nor did
the phrase which took its place, “divine white tortilla.” Neither
designation could convey the original’s authority. Missionaries
might have believed that with their practices of literacy they could
improve the Nahuatl language along with the lives of their converts
by revealing the real nature of the world. But they produced instead
the distinctive new style David Tavárez (2000:23) describes as “doc-
trinal Nahuatl,” a way of speaking associated with missionaries.
This new style could not “silence indigenous voices, . . . resolve
Early Conversions 41
dialogue into monologue, [and] replace cultural diversity with con-
formity” (Burkart 1989:9). So the friars’ work of conversion was
lovingly destructive in a way they could not really afford to acknowl-
edge: they could appropriate and deform Nahuatl into a written
language of Christianity only by deforming the message of Chris-
tianity in ways they could not fully grasp.
Literate Primitives
I mentioned earlier that in the Philippines, friar linguists like de San
José saw that baybayin was in some ways more useful than Latin, at
least when it came to the problem of writing Tagalog’s saltillo. Given
this close, joined engagement with Tagalog writing and speech,
why did baybayin eventually fall out of use? What in Tagalog life
led missionaries to finally judge it incompatible with their larger
purposes, so that it eventually passed from use? A bit of historical
background helps to frame this question.
Cortez dispatched explorers from Mexico’s west coast in 1528 in
a continuing search for the Spice Islands. As emissaries of an
emperor, Charles V, they carried letters he signed “Caesar Augus-
tus, king of the Spains.” Their voyage set in motion very different
colonial developments in what now counts as island Southeast
Asia, including those islands which were named the Philippines in
1541 in honor of Spain’s heir apparent, Philip II.
At first those islands were a consolation prize and base of opera-
tions for the Spanish as they waited for an opportunity to displace
Portugal from the real prize, the Spice Islands, to the south. A real
irony of colonial history is that the Philippines were still a profitable
possession for the Spanish centuries centuries after Portugal lost
control of the Spice Islands to the Dutch. The Spanish reaped enor-
mous profits from the trans-Pacific trade which met in Manila, the
“pearl of the Orient,” on the island of Luzon. Up through 1815,
silver carried west from the New World was traded there for man-
ufactured goods from China.
Because Spain’s colonial interests became centered on trade and
Manila, their dealings with “natives” beyond the town’s environs
did not become as intense or exploitative as those of their compa-
triots in the New World. Fewer Spanish ventured abroad to the new
42 Early Conversions
colony, and oversight of the native populations and their souls was
left to the Church, represented by a clerisy which by 1722 was more
numerous in Manila than the lay population. Thanks to relatively
weak lines of colonial contact, the Philippines’ natives were spared
many of the devastating diseases, forced labor, and wars of con-
quest which befell los indios in the New World.
On Luzon, the Spanish found little they counted as civilization:
local elites, such as they were, exercised limited authority over few
people and small expanses of territory; explorers found little silver
or gold – always a Spanish fixation – or other forms of wealth aside
from land, which was generally considered a communal resource;
there were no large “temples” and few settlements of any size away
from the coasts or in the inaccessible volcanic interior. Against this
“primitive” backdrop, writing systems like baybayin must have
seemed all the more peculiar.
Beyond any “defects” which friar linguists found in baybayin, two
other reasons can be offered for its demise. On one hand broad
shifts in political power and religious practice emanated from
Europe to the Philippines and elsewhere in the colonial world. On
the other, friars identified limitations not just in baybayin orthogra-
phy, but in Tagalog practices of literacy it served.
In the second half of the 16th century the Church consolidated
its doctrinal foundation in reaction to the Reformation, and deci-
sions taken by the Council of Trent, beginning in 1545, gave new
grounds for caution among missionaries about deviating from reli-
gious orthodoxy (a wariness which might have been augmented by
missionaries’ experiences in Mexico). New concerns arose about the
failure of translation, and the need to “safeguard the key words of
the doctrine from confusion with native beliefs and terminologies”
(Rafael 1993:117). Missionaries had learned that pagan words might
not be easily disentangled from pagan worlds.
For missionaries to introduce more words of Spanish and Latin
into their religious discourse for native converts, they had to intro-
duce foreign speech sounds into native languages. Tagalog could
under those conditions be said to “lack” speech sounds like r,z,f,
v, and f, which are common in many Spanish and Latin words. For
missionaries who wrote these words, and required symbols for
sounds lacking in baybayin, that orthography came not just be
different from but deficient in relation to the alphabet.
Early Conversions 43
Rafael (1993:53) also suggests that beyond baybayin’s orthographic
indeterminacies, sketched above, missionaries resisted its use for a
deeper reason: their inability to tolerate the gaps of meaning which
could be created by the orthographic “gaps” mentioned above.
These indeterminacies, Rafael suggests, marked a willingness to
allow readings which passed “over sense in favor of sensation,” and
in that way ran directly against the grain of the missionaries’ own
“totalizing signifying practices.” In this way the internal logic of
baybayin posed a threat to the practices of literacy which grounded
their sacred traditions of textual transmission.
But the existence of such an aesthetic of indeterminacy would
not account for the fact that missionaries recognized ways to
“improve” baybayin, rather than dispense with it. And it would
seem peculiar that missionaries familiar with Hebrew writing,
which incorporates many more such orthographic indeterminacies,
touched on in chapter 3, would refuse to countenance those in
baybayin which are, in the larger scheme of things, more limited.
One way to recast Rafael’s insight is to consider baybayin ortho-
graphic conventions as they figured in the broader practices of lit-
eracy shared and transmitted among its users. In this respect the
complaint of one missionary quoted earlier – that baybayin is “as
easy to write as it is difficult to read” – is telling because it identifies
a basic asymmetry between practices of writing and reading
baybayin.
What Rafael calls the friars’ “totalizing signifying practices” key
to the one letter/one sound principle is mentioned in chapter 1. This
enables a kind of symmetry between acts of reading and writing:
knowledge of orthographic conventions is necessary and sufficient,
at least in principle, to voice a text. But conventions of spelling need
not require this symmetry, as a few exceptional spellings of English
words help to demonstrate. When the letter aappears in the sequence
of letters b-a-s-s, for instance, it has two possible pronunciations,
rhyming with pace or pass, and can help to convey two different
meanings. Ordinarily collateral information about the context of use
will make this a trivial matter, although it is possible, in a written
sentence like “That’s a big bass,” for that ambiguity to go unre-
solved. Only if reader and writer share knowledge of context and
intention is it a trivial matter to disambiguate such spellings and
meanings.
44 Early Conversions
This example illustrates a general norm by violating it: texts are
read with alphabetic practices of literacy as autonomous bearers of
meaning. As an overriding ideal this is made very clear by Walter
Ong, a modern Jesuit inheritor of the faith of the friar linguists
I have discussed here. He celebrates the alphabet as a special
means for
abstractly analyzing the elusive world of sound into visual equiva-
lents which sets [it] apart from other orthographies still very much
immersed in the non-textual human life-world. (Ong 1977:90)
In chapter 3, I return to the broader meanings and uses of Ong’s
implicit parallel between the alphabet and “other orthographies”
on the one hand, and those whose practices of literacy are less or
more “immersed in the non-textual life-world” on the other.
It is significant in this regard that Europeans who encountered
baybayin in wide use could find nowhere any significant body of
written texts: “[t]hey have neither books nor histories nor do they
write anything of any length but only letters and reminders to one
another . . . [lovers] carry written charms with them” (quoted in
Scott 1994:210). They sought texts which were not just “longer” and
more permanent than “letters” or “reminders,” but which fit Ong’s
description better. “Letters” or “reminders” can be written on the
presumption of knowledge shared by the writer and intended
reader (topic, purpose, relevant history, etc.). A shopping list, for
instance, can be all but indecipherable to someone who did not
write it.6
Where alphabetic practices of literacy were bound to genres of
historical writing and verbal art, those of babayin were not. For
Tagalogs, poetry and literature were celebrated and appreciated in
oral performance:
The noblest literary form was the siday or kandu. This was the most
difficult of all – long, sustained, repetitious, and heavy with meta-
phor and allusion. A single one might take six hours to sing or the
whole night through, or even continued the next night, during which
rapt audiences neither yawned nor nodded, though the frequent
repetition of long lines with only the variation of a few words struck
Spanish listeners as tiresome. (Quoted in Scott 1994:98)
Early Conversions 45
All of this suggests, then, that although missionaries could rec-
ognize baybayin as an orthography which was adequate for writing
Tagalog speech sounds, they found later that it served no practices
of literacy they judged adequate to their purposes. When practices
of literacy appeared not just different but deficient in their situated,
occasion-bound nature, baybayin came to be seen as an attribute of
people who were literate but also primitive, and fell into that class
of things which were to be studied “on the grounds of facilitating
their eradication” (Burkhart 1989:3).
The Work of Loving Destruction
The work Ricard called “spiritual conquest” was done by creating
hierarchical relations between humans with and through their lan-
guages. To appropriate pagan tongues, linguists reformed and
deformed them with practices of literacy which let them remove
words and meanings from native speech and speakers, before
“giving them back” in their own religious discourses, and as symbols
of their authority. These intimate and powerful relations could be
created and sustained as long as natives’ tongues were recognized
as necessary but imperfect vessels of Christ’s message.
But as friar linguists’ early engagements with the saltillo show,
their relations with those languages were fine-grained and empiri-
cal; Christian love drove them to observe, compare, and generalize,
reducing textures of talk as fine and fleeting as the glottal stop to
writing. This ability has always grounded missionaries’ “mastery”
of alien languages, and literacy has always lent authority to projects
of conversion whose effects, as we will see in later chapters, always
outran their intentions.
In Mexico, missionaries recognized hierarchies of language and
class which conformed to their own social imaginaries, and which
helped them to introduce literacy as the supplement needed to
elevate lordly Nahuatl language as a means of spreading the Faith
and converting its speakers. They undertook the work Patricia Seed
(1991:13) calls “conquering language with language” on the presup-
position that that they could bring Nahuatl into alignment with
truth unilaterally, by making it their descriptive object.
46 Early Conversions
Missionaries had difficulty recognizing, then, that they could
only do the work of description and translation on what Dennis
Tedlock (1983:334) calls the “dialogical frontier,” and in larger colo-
nial dynamics of accommodation and resistance, understanding
and misunderstanding. So too the meanings of words shifted and
slipped across colonial, religious, and linguistic divides as Nahuatl
was deformed in writing, and Christian doxa was deformed in mis-
sionary discourse. Over time this dialogical process cumulatively
affected language form and meaning alike so that, as Louise Burkhart
(1989) puts it, the missionizers were themselves missionized.7 Their
founding faith and Walter Ong’s assertion notwithstanding, their
texts were never autonomous, or separable from life-worlds of friars
or their flocks.
In the Philippines, successors to the friar linguists I have dis-
cussed retreated from these early intimate engagements with
Tagalog, and became content to convey Christian doxa in a Spanish
language few Tagalogs understood. Catholic identities became
more firmly grounded in communities of religious practice, and less
in discourse, as gulfs between priestly authority and colonial sub-
jects opened up in ways acidly portrayed by Jose Rizal, an early
Philippine nationalist, in his brilliant 1887 novel Noli me tangere.
This chapter’s epigram is drawn from a parable for the success
and failure of these friar linguists. Erdrich puts this observation
about loving destruction in the mouth of a hero/ine, who speaks to
his/her work as a priest in a community of native North Americans.
Born a woman driven by the love of God, she lovingly destroyed
her own gender to become a priest who could bring faith to others.
This is a more physical image of loving destruction than the work
of conversion I have described here, but it does echo transforma-
tions which Christian doxa had to undergo in Nahuatl and Tagalog
letters, no matter how carefully friar linguists described and under-
stood Nahuatl and Tagalog talk. However closely they listened,
these first linguists could never fully reduce either to writing, or
save the message they brought to destroy pagan ways of life.
Notes to chapter 2
1 In addition to extensive discussion of these traditions in Mignolo,
see also articles in Boone and Mignolo (1994). A convenient
Early Conversions 47
introduction to Mayan epigraphic traditions is given by Coe and Van
Stone (2001).
2 The glottal stop is less obvious but common in casual American English
as an alternant of tin pronunciations of words like “can’t,” “Latin,”
“written,” and others where twould otherwise appear in close proxim-
ity with n.
3 If speakers of Philippine languages adapted scripts from their Buginese
neighbors to the south, then there is a historical explanation for the fact
that baybayin has no symbols for representing consonants when they
close syllables: Buginese has almost no such syllables, and so no need
to transcribe them.
4 Furthermore, he brought Latin letters it into better alignment with
Tagalog speech by stipulating that he would use the letter y – which
he didn’t need for any other Tagalog speech sound – as an equivalent
of i, that is, glottal stop followed by i, represented in baybayin by .
This allowed him to showing in alphabetic writing the difference
between gaby (what I can write here as gab i) and gabi.
5 For discussion of Quechua’s similar position among the Incas see John
Rowe’s (1982) discussion of similar Inca policies and institutions relat-
ing to the cultural unification of the empire.
6 So too Spanish accounts are unanimous in saying that Filipinos did
not use their “alphabet” for record keeping, a practice which by its
nature requires stable, unambiguous letter/sound and sign/meaning
relations. This incompatability between practices of literacy became
especially glaring when baybayin orthography was introduced into the
colonial (and more generally Western) practices of literacy involved
with documents like land transfers. These depend for their usefulness
on readability at a remove, spatial and temporal, from the acts they
record. Two such documents were written in baybayin in the late 16th
century but, as Rafael describes it (1993:49), were readable by an archi-
vist in the mid-20th century only with the aid of alphabetic documents
which had been written by a contemporary Spanish notary.
7 For more on these dynamics see Lockhart 1991, 1992 and Hanks 2000
for discussions of literacy practices among Mayans of the Yucatan
peninsula.
Chapter 3
Imaging the Linguistic Past
Who placed us with eyes between a microscopic and telescopic
world?
Henry David Thoreau, The journal of Henry D. Thoreau (1962(6):133)
By the middle of the 18th century, microscopic accounts of alien
languages like Nahuatl and Tagalog, written by missionaries,
explorers, and others, were circulating back to Europe. Each fine-
grained, highly circumscribed image of distant lives became part of
the growing evidence that linguistic and human diversity far
exceeded that previously imagined in Europe. Projects of explora-
tion and exploitation at far-flung peripheries of the world were
creating severe challenges for received ideas about human nature
and history at home.
This chapter is about some important, early engagements with
larger questions raised by this new knowledge of linguistic diver-
sity, and new, telescopic views of the human past which began to
develop in response. First I summarize speculative ideas of the
German philosopher Johann Herder, and discoveries of the British
philologist and colonial official William “Oriental” Jones, which
were important in different ways for language-centered images of
the deep human past in a 19th century science of language. Then I
discuss some of the assumptions and purposes which informed
these authors’ two projects, showing how they gave new life to
older social imaginaries of language and human history. It is impor-
tant here to foreground overlapping images of the histories of
Imaging the Linguistic Past 49
texts and languages which helped make the science of language
a way of legislating human differences in a rapidly colonializing
world.
Herder’s early, famous essay on language origins and diversity
is too speculative to count as a foundation for any science, but it
framed questions of human origins and identity in ideologically
powerful ways. He offered compelling answers to pressing ques-
tions about language and human nature, and intellectual resources
for two different kinds of projects: asserting colonial authority
abroad, and creating national communities at home.
William “Oriental” Jones is a founding figure in linguistics thanks
to studies in India, which led him to recognize resemblances between
Sanskrit on one hand, and Greek, Latin, and languages of modern
Europe on the other. But Jones’ studies as a “philologer” were in
service to his primary duties as a judge working for the British East
India Company in Bengal. His joined projects of (linguistic) knowl-
edge and (colonial) authority made him famous as “Oriental” Jones,
an emblematic figure in the intellectual tradition which came to be
known as orientalism.
Important here are the ways Herder and Jones blurred lines
between what now seem very different kinds of histories: of written
texts on one hand and language systems on the other. These overlaps
helped give new life to old biblical images of the past, renewing
what Trautmann (1997) calls “Mosaic ethnology.” New telescopic
visions of language change and diversity helped to “reconstitute,
redeploy, and redistribute old religious patterns of human history
and destiny” (Said 1994:121). To show how these two new images
of the linguistic past could legitimize European authority abroad, I
conclude this chapter with a sketch of two different versions of
Javanese history, allowing otherwise different stories about its lit-
erature and language to prove, in correspondingly different ways,
the inferiority of Javanese in a colonial present.
Herder’s Story of Language Origins
Johann Herder (1744–1803) stepped onto the European intellectual
stage as Immanuel Kant’s greatest student, and became a famous
public intellectual whose voluminous writings helped to “open
50 Imaging the Linguistic Past
philosophy out onto the world.” Most famous among his writings,
though, is his early essay submitted for a contest sponsored by the
Berlin Academy of Science in 1769. The academy’s question, posed
in French, was whether “men, left to their natural faculties, [are] in
a position to invent language, and by what means do they, by them-
selves, accomplish that invention.” Herder’s victory in this contest
marked this expatriate school teacher’s rapid ascent into the elite of
German-speaking society, a figure who would cast a shadow over
generations of political philosophers, educators, philologists, and
others.
The prize question was hardly novel. It had been the subject of
longstanding debate, particularly among leading French thinkers
whose positions had become widely known in European intellec-
tual circles. Hans Aarsleff, a historian of linguistics, argues that
there was little new in Herder’s essay (Aarsleff 1982). But besides
considerably simplifying Herder’s position and tracing his intel-
lectual debts to influential French philosophers of the time, Aarsleff
is concerned more with ideas speaking to ideas than an essayist
addressing a particular audience. More important here are ques-
tions about this essay’s influence and rhetorical force for its pre-
dominantly German readership.
Considering the sources of this essay’s influence helps to read it
as an answer to one question, about language, in order to address
another, about identity and history. The more immediate question,
posed by global linguistic and human diversity, had implications
for Europeans’ understandings of their place in a widening world.
The other question, less direct and more local, was driven by a
political and cultural crisis confronting Germans (or speakers of
German): an encroaching French state and civilization. France’s
intellectual and political ascendance over the Holy Roman Empire
– where Germans lived as subjects of a fractured nobility, on patch-
works of landholdings – was evident in the very fact that the Berlin
Academy posed this question in French. I discuss these political and
cultural issues in the next chapter, but mention them here as an
aspect of the context in which Herder aimed to address his readers
in a forceful, somewhat colloquial German, not just as a philosopher
but a German writing for other Germans.
It is convenient here to summarize Herder’s argument as having
two parts. The first is about the coalescence of languages out of
Imaging the Linguistic Past 51
organic relations between humans and their environments; this is
a story which denies divine intervention, and insists instead on the
naturalness of linguistic sharedness and expressive authenticity.
This became a foundational theme for romanticist philosophy and
nationalism in following decades, a part of what Terry Eagleton
calls Herder’s “conscious assault on the universalism of the Enlight-
enment” (2000:12). For Eagleton, this also grounded Herder’s protest
against European colonial projects on behalf of those from “all
the quarters of the globe who have not lived and perished for
the dubious honour of having their posterity made happy by a
speciously superior European culture” (2000:12).
The second part of Herder’s essay, less well-known, extends this
story of origins, arguing that languages have developmental trajec-
tories which explain intrinsic inequalities between them and their
speakers. This is important as a kind of prototype for the much
more elaborate, influential accounts of linguistic and human history
which developed over the next century, allowing telescopic con-
structions of global human difference to be based on specious com-
parisons of microscopic linguistic facts. Herder showed how to
make “facts” of language history into proofs of human inequality;
in this way he also demonstrated the darker side of linguistic and
cultural relativism, as romanticism contributed to the sense of
otherness of colonial subjects whose humanity could be counted as
incommensurable with that of their masters.1
Herder begins his essay with a paradox: how is it that the
human organism has physical senses which are so much weaker
than those of many other species in the world, yet human commu-
nities – as explorers had amply demonstrated – are able to survive
and thrive in so many more environments around the world than
other species? The diversity of what we now call ecological niches
in which human communities have places must be due, Herder
argued, to some other distinctive human capacity, which became
central to his explanation of language origins. The term he used for
this capacity he borrowed from his teacher, Immanuel Kant. Beson-
nenheit might be translated into English as “reflection,” but Herder
describes it as a kind of synthetic, intuitive sense behind the
physical senses, a capacity constantly engaged with and shaping
experience as it flows through the senses from the world into
bodies. Besonnenheit is what allows humans to abstract out or fix
52 Imaging the Linguistic Past
some elements of that flow, and so mediates between natural
environments in which the body’s senses are located, and internal
experiences of those environments.
After arguing (none too plausibly) that hearing is privileged
among the senses, Herder proceeds to trace the emergence of lan-
guage from dynamic relations between Besonnenheit and auditory
experience in particular locales. Thanks to their ability to bind small
parts of the flow of auditory experience in time, humans can recog-
nize that they resemble each other, and so also are bound to experi-
ences of the environment. Once some patterns in the flow of auditory
experience are recognized – the sound of a sheep bleating is
Herder’s example – they can be brought into reflective relation with
experience of things in the world which produce them, and so
understood to represent those things (“the bleater”). In this way the
first words emerge, and with them language, from experience of
the world. By appealing only to natural capacities of human organ-
isms, Herder is able to explain language origins without any appeal
to divine intervention.
Linguists, who nowadays refer to this as the “bow-wow” theory
of language origins, distance themselves from what they regard as
rank speculation. But Herder broadened his argument to frame
language as both human and natural because it related “external
and internal to one another reciprocally . . . There is no expression
of the physical which does not immediately appear as ‘spiritual,’ as
the symbol of some psychic process of reworking” (Pross quoted
in Zammito 2002:324). By focusing on senses, environments, and
experience, Herder downplayed the conventionality of language –
its character as a kind of sameness of knowledge distributed among
speakers – to foreground its experiential and expressive immediacy,
grounding modes of social sharedness. Language joins bodies to each
other in and through shared environments. A “place is sensed
[when] senses are placed,” and language helps “places make sense
[when] senses make place” (Feld 1996:91).
Herder’s story allows no firm line between nature and culture,
or individual and community. Thought and expression combine in
words and the “ocean of sensations” from which they emerge. This
profoundly plural vision of “human nature” is grounded in what
Charles Taylor calls, in his insightful discussion of Herder, the “lin-
guistic dimension” of life: the “space of attention, of distance from
Imaging the Linguistic Past 53
the immediate instinctual significance of things” (Taylor 1995:88).
As languages differ, so do “linguistic dimensions” of life.
One attractive point of Herder’s essay was that it accounted for
language origins and diversity in the same way, offering an alterna-
tive to the story of Babel. If Besonnenheit is a part of human nature,
then linguistic diversity emerges as a natural consequence of expe-
rience of the diversity of environments in which it has operated.
Languages and communities are natural outgrowths of diverse
locales, which is one reason Herder was deeply suspicious of
colonialism: it was a literally unnatural outside force or power
which could only destroy any locale, language, and community, a
“Trojan Horse . . . that seek[s] to subvert this naturally plural world”
(Pagden 1995:144).
In his essay’s second part, Herder shifts attention from language
origins to language history, and develops a strategy of linguistic
comparison to corroborate a broader argument based on linguistic
“facts.” This early framing of what would later become dominant
organic images of linguistic and human difference keyed to the
notion that languages, like all living things, partake of the cycle
of birth, maturation, and death. Once this cycle is in motion, it sus-
tains and is sustained in language and culture across generations
of shorter lived human organisms, such that human collectives
develop in one or another direction, and achieve a greater or lesser
apex of maturity. Each language has a developmental dynamic of
its own, like any living thing: “it sprouts, it blossoms, it flowers,
and it withers – so it is with language” (Menze and Jenges
1992:104).
These organic trajectories, Herder argues, are evident in specific
properties of exotic languages which, in his judgment, have moved
past their highest stage of development into states of decadence or
dormancy. Here his philosophical, telescopic point of view is joined
to a presentation of microscopic details drawn from various linguis-
tic descriptions. These provide what counted for him, and the lin-
guists who followed him as ample evidence that languages around
the world are either immature or moribund. They provide detailed
evidence, in his view, of the rudimentary character of what he calls
“unpolished” languages: Chinese, Siamese, the language of the
Lapps, Quechua (spoken in the Andes), as well as that “outstand-
ingly imperfect” language, Arabic (Herder 1966:154). That these
54 Imaging the Linguistic Past
languages exist is significant for Herder as important evidence
against any argument for the divine origin of human languages: no
divine being, he notes, would ever have created such imperfect
things.
This “empirical” part of Herder’s argument develops as demon-
stration, rather than speculation, through a wide series of parallel
contrasts joining language structure, language use, and human
communities. Words and word roots of unpolished languages, he
argues, conform to sensual experience (i.e., are onomatopoeic) more
than those of polished languages; unpolished languages show an
“unnecessary abundance” of synonyms, unlike polished languages,
but have less developed grammars; unpolished languages belong
to communities of speakers who have (feminine) genres of song and
poetry, unlike those whose polished languages make possible (mas-
culine) genres of prose and philosophy.
Read with hindsight, these analogies seem at best to be evidence
of ignorance or naiveté about facts and meanings of linguistic diver-
sity. But in Herder’s time and place the success of this argument
testifies to the power of these language-based ideas about human
difference, such that they could shape the 19th century science of
language I describe in chapter 4.
Herder has a special place in his essay for literacy as the crucial
point dividing languages (and peoples) who count as “unpolished”
and “polished.” This is because literacy is both an outcome and a
sign of the workings of Besonnenheit which in turn becomes an
instrument of Besonnenheit. So different literacies can be compared
on the same universal metric of “polish,” and Herder could estab-
lish the importance of Hebrew, a language which he studied
throughout his life, in human development. Hebrew never crossed
the crucial threshold to alphabetic writing, he argues, because its
letters “are exclusively consonants [whereas] precisely those ele-
ments of the word on which everything depends, the self-sounding
vowels, were originally not written at all” (Herder 1966:93). Why
was Hebrew’s development truncated? Because, in Herder’s judg-
ment, of the quality of spoken Hebrew’s vowels: “they could not be
written. Their pronunciation was so alive and finely articulated,
their breath so spiritual and etherlike that it evaporated and
eluded containment in letters. It was only with the Greeks that these
Imaging the Linguistic Past 55
living aspirations were pinned down in formal vowels” (Herder
1966:93).
Perhaps Herder had not read about Nahuatl as described by friar
linguists in Mexico, or Tagalog in the Philippines; if he had perhaps
he would not have argued that “[t]he more alive a language is the
less too is it writeable” (Herder 1966:93). But as it happens he set
out a position echoed much later by Walter Ong, whose reflections
on alphabetic writing I mentioned in chapter 2. Herder and Ong
both developed analogical differences between alphabetic and
nonalphabetic orthographies on one hand, and mental abilities,
arrayed along a developmental trajectory, on the other. Because
they make language difference symptomatic of deeper differences
between humans, they require only a short step from a relativist
celebration of difference to a legitimizing explanation of inequality.
Herder establishes a fundamental asymmetry between speakers of
more polished languages (like Herder, and us) who can know and
those who, as speakers of unpolished languages, are fated to be
known.
Within Europe, this pluralistic vision inspired nationalist ideolo-
gies which I discuss in chapter 4, as it progressively forced “old
sacred languages – Latin, Greek, and Hebrew . . . to mingle on equal
ontological footing with a motley plebeian crowd of vernacular
rivals” (Anderson 1991:70). But in the larger world, it combined
with equally powerful images of the advancement of reason to
explain the gulf between modern and primitive humans. It is impor-
tant here that Herder did this by turning language on itself to
establish his own position as inheritor and exponent of the civiliza-
tional dynamic. This double distancing of the modern from the
primitive can be seen by locating his argument in a broader, mature
definition of “civilization” which was formulated in 1930, at the
height of the imperial age:
[linguistic diversity] is . . . a fact and . . . [Herder] studies it as such;
he exercises his mind on facts; and when he discovers the general
laws which govern the development and life of the world, even those
laws are simply facts which he observes. And then the knowledge of
external facts develops in [him and] us ideas which dominate these
facts . . . as a spectator he is subject to facts; as an actor he remains
56 Imaging the Linguistic Past
master in imposing upon them a more regular and a purer form.
(Febvre 1930:247)
Herder’s self-positioning was implicit in his linguistic image of the
past, and later in the first science of language he inspired.
Imaging the Oriental Past
About the time Herder wrote his essay on language origins, William
“Oriental” Jones’ studies of the Sanskrit language were coming to
the attention of educated Europeans. So important were Jones’ dis-
coveries that his Presidential address at the first meeting of the
Asiatic Society of Bengal, in 1786, is taken by many as foundational
for linguistics. So it is common for histories of the field to include
the following quotation:
[T]he Sanscrit language, whatever be its antiquity, is of a wonderful
structure; more perfect than the Greek, more copious than the Latin,
and more exquisitely refined than either; yet bearing to both of them
a stronger affinity . . . than could possibly have been produced by
accident; so strong indeed that no philologer could examine them all
three without believing them to have sprung from some common
source, which, perhaps, no longer exists; there is a similar reason,
though not quite so forcible, for supposing that both the Gothick and
Celtick, though blended with a different idiom, had the same origin
with the Sanscrit; and the old Persian might be added to the same
family. (Quoted in Trautmann 1997:38)
Hindsight makes it is easy to find in this passage what Thomas
Trautmann (1997:39) calls a “precocious modernity:” an ability to
recognize the implications of new facts when take together with
others long known, and to reason from them all to a sweeping new
vision of the past.2
But as founding figure in the larger colonial enterprise of power
and knowledge, critiqued by Edward Said in his book Orientalism
(1994), “Oriental” Jones can also be regarded as having accom-
plished this feat because of the transparent relations between his
need for linguistic knowledge and political interests. Jones came to
Imaging the Linguistic Past 57
the study of Sanskrit in the first place because he judged it necessary
for his work as an officer of the East Indies Company responsible
for administering justice to “the natives.”
Said passed over Jones’ linguistic work in Orientalism in favor of
French studies of Semitic languages, but he surely knew that Jones’
work answered his broad description of a “Western style for dom-
inating, restructuring, and having an authority” (Said 1994:3). Jones
answers well in this respect, at least, to Said’s description of the
orientalist as
a hero rescuing the Orient from the obscurity, alienation, and strange-
ness which he himself had properly distinguished . . . having trans-
ported the Orient into modernity, the Orientalist could celebrate his
method, and his position, as that of a secular creator, a man who
made new worlds as God had once made the old. (Said 1994:121)
But Jones’ “discoveries” and methods were not entirely new.
Europeans, including other Britons, had been engaged with exotic
Indian languages and literatures long before he studied them, and
had produced grammars, dictionaries, and translations of the New
Testament for speakers of the Tamil language, spoken further south
on the subcontinent. Other Europeans had encountered Sanskrit
long before Jones, noted the same correspondences with European
languages which caught his attention, and speculated in the same
vein about some ancient original source.3
Jones did his linguistic work in large part by extending traditions
of etymological study already well known in Europe, which were
being exported to other borders of European expansion. In Russia
and the United States – just freed from one colonial project and
embarking on its own conquests – word lists were circulating to
gather information about aboriginal languages. These projects were
inspired by Gottfried Leibniz, who had observed in the 17th century
that comparison between languages – “the most ancient monu-
ments of peoples” – could provide clues to the common origin of
nations.
So Jones’ discoveries had tremendous impact not just because
they were novel, but because he had readers in the intellectual
circles of Europe who were already “seething with curiosity” about
languages of the Orient (Schwab 1984:23). He gave them texts which
58 Imaging the Linguistic Past
could be read for a fine-grained but telescopic view of historically
distant objects, albeit within a highly delimited field of vision.
Reading at a distance, they could extend the shared social imaginar-
ies and habits of thought which made Jones’ methods and conclu-
sions plausible for them. They were ready to recognize, for one
thing, that scholarly expertise which had been developed in the
study of the history of texts could be transposed naturally to the
study of the history of languages. This simple but important point
has been discussed by a historian of linguistics, Henry Hoenigswald
(1974), who pointed out how it was these two joined areas became
subjects of different fields of study in the 19th century. Drawing on
Hoenigswald’s observations here, I foreground the ways Jones’
image of “deep” linguistic and human history gained plausibility
and power because from it extended ideas and images which now
count as conceptually distinct.
Jones took to India a considerable reputation as a “philologer,”
that is, a student of exotic textual traditions who at age 33 “knew”
a dozen languages: not just Latin, Greek, and Hebrew, but the more
exotic Arabic, Turkish, and Persian, this last especially important
as a language of power in India. Philologers had no need to be able
to speak these languages, so long as they were versed in practices
of literacy which allowed them to engage with texts written in those
languages.
In 1783, after Jones left the scholarly life for the law, and traveled
to Fort William in Calcutta, he took an appointment as judge for
the British East India Company. He discovered, though, an unex-
pected but real need for his expertise as a philologer when con-
sultations with native experts – “professors of the Hindu law,” he
called them – showed them to be insufficiently knowledgeable
about their own textual traditions. In his view they neither under-
stood nor applied their legal traditions consistently and correctly
and so obliged Jones to study and master that tradition himself in
order to gain direct access to original authoritative texts, and avoid
“reliance . . . on [native] opinions or interpretation . . . examin[ing]
their authorities and quotations, and detect[ing] their errors and
misrepresentations” (quoted in Aarsleff 1983:119). Jones the philol-
oger came to the aid of Jones the judge to make it an “impossibility
for the Mohammedan or Hindu lawyers to impose upon us with
erroneous opinions.”
Imaging the Linguistic Past 59
This work of textual excavation required Jones to obtain multiple
versions of texts which, once gathered, could be compared with
each other in great detail. Each had value, then, not in itself, but as
one of a group, an instance of a category, and a more or less imper-
fect replica of the ultimate object of knowledge: the original, author-
itative, but absent ur-text from which each was “descended.” Here
is a broad description of the goals and methods of this kind of
reconstruction:
The aim of this method is to construct a family-tree of MSS., called a
stemma. At the top stands the archetype, the MS. from which all
extant MSS. are derived. . . . If it can be shown that the surviving
MSS. are traceable to several lost ancestors, intervening between the
archetype and them, then these intermediate stages can be indicated
with Greek letters, e.g., αand β. . . The value of the stemma is that
it can show the genetic relation of MSS. And hence which are closer
to what we are seeking. (Robson 1988:14)
I draw here from Stuart Robson’s introduction to Dutch tradi-
tions of study of Javanese literature, which I discuss later, to fore-
ground the basic images and metaphors which went into the work
of reconstructing texts. Relations of “descent” between texts can be
charted on “family trees” which link existing texts to their “ances-
tors” through “genetic relations.” These biological metaphors
combine and support a linguistic image of the past like that in
Figure 3.1.
In this scheme, texts are objects (MS1, MS2, etc.) located at dif-
ferent distances in historical time (tn+1,tn+2, etc.) from a common,
archetypal point of origin (tn). To work out links between them is
to deduce the properties of their common point of origin by com-
paring traits to establish the intervening acts of writing/copying
which produced them. This way of reading makes texts’ internal
contours and coherence less important than their isolable traits,
which can be identified and compared as individual facts. There is
a circular logic at work here: relations between physical entities can
be established by postulating that there exists some shared but
absent ur-text; but that ur-text can only ever be known by studying
its putative descendants.
Figure 3.1 also presents an image of history as a series of punctu-
ated events, rather than cumulative development: texts result from
60 Imaging the Linguistic Past
events of writing in different times (and maybe spaces) which need
to be brought into analytic proximity. The overall contour that con-
nects these points, in turn, is a progressive distancing and falling
away from a perfect, originary moment, because acts of copying
introduce cumulatively more deviations.
Said argues that this kind of textual appropriation was ideologi-
cally important because it blurs the difference between senses of
textual history on one hand, and social history on the other, Sub-
stituting the first for the second, as part for whole, helped to
“convert . . . Indian forms of knowledge into European objects”
(Cohn 1996b:21). Bernard Cohn, a critical historian of colonial India,
suggests that Jones was able in this way to bring together two
modalities of knowledge and power: as judge he operated in an
investigative modality, gathering facts and reaching conclusions; as
a philologer he operated in a historiographic modality, saving texts
from the corrupting effects of time and larger processes of civiliza-
tional decay.
Said also recognized the self-authorizing grounds of civiliza-
tional advance in these methods. The application of reason could
not only provide natives with their own languages in writing, as
discussed in chapter 2, but could also give them back their textual
legacy in a new era of knowledge and reason.
tn+1
MS (tn)
tn+2
MS1 MS2
βα
MS3 MS4
Figure 3.1 The philological image of textual descent.
Imaging the Linguistic Past 61
From Text to Language
Philologers worked to reconstruct original texts by studying their
descendants as ensembles of instances of language use – words,
roots, and grammatical elements – in the European tradition of
etymological speculation. This was the study reputed to have been
called by Voltaire “a science in which vowels count for nothing, and
consonants for very little.” Jones devoted considerable attention to
the histories of particular words, but the logic of comparison and
generalization which he demonstrated could be transposed to other
dimensions of language, which I discuss in chapter 4.
To illustrate his etymological approach, I can sketch a single
group of words: patterns of letters and meaning which can be
drawn out of texts, and compared with each other to allow infer-
ences about larger aggregates or ensembles of facts: the entire lan-
guages in which those texts (and perhaps indefinitely many others)
were written. So words for “mother” can be abstracted from texts
written thousands of miles and centuries apart, and listed together
as Table 3.1.
Compared across texts and languages, such elements license
broader inferences about where and when languages were spoken,
and so about geographical and historical relations between groups
or communities of speakers. The comparison of languages in this
way allows much greater historical and geographical depth, which
can be illustrated here by anticipating my sketch in chapter 4 of the
comparative science of language which explored that “deep history.”
The culmination of this image of the linguistic past, devised in the
1860s, is the so-called Indo-European language tree (Stammbaum in
German) shown in Figure 3.2.
I present this image here to emphasize its parallels and differ-
ences with the image of textual descent in Figure 3.1, which are easy
to take for granted because both are grounded in biological meta-
phors of descent. Languages, like texts, are commonly spoken of as
being related to each other like kin, as “parent” forms, which may
no longer be extant, with “daughter” forms.
This is a convenient way of framing relations in time because it
makes it easy to overlook differences between the dynamics and
events which go into relations of “descent” mapped with these two
62 Imaging the Linguistic Past
figures. Figure 3.1 displays nodes which branch at points in the lives
of individuals, copyists who produced different instances of texts;
Figure 3.2 displays nodes which branch at points in the histories of
communities as they moved in different directions or otherwise
became separated. Texts are products of human agents; “daughter”
languages emerge from “ancestor” languages by a kind of partho-
genesis or fission in which individuals count as an anonymous,
homogeneous substrate.
Because relations of descent among languages involve move-
ments in space, and continuity between generations of speakers, the
Indo-European tree in Figure 3.2 can be read as an image not just
Table 3.1 Some Indo-European words
for “mother”.
Old Indic *ma
¯tár-
Avestian ma
¯tar-
Armenian mair
Albanian motrë
Latin ma
¯ter
Umbric matrer
Lettisch mate
Lithuanian motera
Germanic *mo
¯thar
Dutch moeter
Old Church Slavonic mati
Old English mo
¯dor
Tocharian ma
¯car
Greek me
¯te
¯r
Low German moder
Old High German muoter
Old Saxon môdar
Old Norse móthorner
Swedish, Danish moder
Old Teutonic mo
¯dar
Asterisks are used by philologists to mark forms of words not as they have been
encountered in texts, but that have instead been reconstructed by comparative
analysis of forms found in texts. This convention was introduced by August
Schleicher, and is discussed in chapter 4.
Figure 3.2 Indo-European or proto-Indo-European migrations in time and space. Redrawn from Crystal 1997. Reprinted
by permission of Cambridge University Press.
GOIDELIC
Scots Gaelic
Manx Gaelic
Irish Gaelic
BRYTHONIC
Cumbrian
Welsh
Cornish
Breton
(Insular)
CELTIC
(Continental)
Celtiberian Gaulish Galatian
Icelandic
Faeroese
Norwegian
Swedish
Danish
GERMANIC
North
East
Gothic
English
Frisian
West
Flemish
Dutch
Afrikaans
German
Yiddish
West
Lekhitic
Czech
Slovak
Sorbian
French
Occitan
Catalan
Spanish
Portuguese
Rhaetian
ITALIC
(Latin)
Sardinian
Italian
Romanian
ALBANIAN
GREEK ANATOLIAN
Ossetic
Kurdish
Persian
Baluchi
ARMENIAN
IRANIAN
Tadzhik
Pashto
Panjabi
Lahnda
Sindhi
Maldivian
Sinhalese
Pahari
Dardic
INDO-IRANIAN
TOCHARIAN
INDO-
ARYAN
(Sanskrit)
Northwest
West and
southwest
Gujarati
Marathi
Konkani
PROTO-
INDO-EUROPEAN
East
Belorussian
Russian
Ukrainain
Midland
Rajasthani
Bihari
Hindi/Urdu
SLAVIC
South
Bulgarian
Macedonian
Serbo-Croat(ian)
Slovene
BALTIC
Latvian
Lithuanian BALTO-
SLAVIC
East
Assamese
Bengali
Oriya
64 Imaging the Linguistic Past
of relations of “descent” in time, but of the movement of languages/
speakers in space (roughly speaking) from east to west (i.e., “right”
to “left”) and north to south (i.e., “top” to “bottom”).
Jones’ textual studies were predicated on the assumption that
texts, once produced, could not somehow come back together or
“merge,” as might indeed happen should some copyist compare
two older texts – in the manner of a philologer – to produce one not
exactly like either. So too mappings of linguistic descent depend on
the assumption that communities of speakers, once having become
separate, remain so, such that their ways of speech do not influence
each other.
“Oriental” Jones’ broad vision of the deep linguistic past can also
be seen as being grounded in another authoritative resource for
thinking about deep history: the Old Testament. Jones was a man of
reason but also faith, who saw in India’s textual traditions a resource
for arriving at a clearer understanding of biblical narratives of the
origins of the world and humanity, and correcting errors introduced
into copied and recopied versions of the original, true story of
Genesis. Thomas Trautmann has pointed out that Protestant faith
grounded Jones’ work, and led him to emulate his hero Isaac Newton,
who similarly sought to correct biblical chronology with his astro-
nomical expertise, fine-tuning calculations of the exact date of the
creation of the world proposed by Bishop Usher.4So, too, Jones saw
in exotic texts and linguistic diversity a key to the location of the
tower of Babel, and a way of recovering the language spoken when
Adam and Eve were cast out from the garden of Eden.
Trautmann describes Jones’ view of the biblical past as being
grounded in a “Mosaic ethnology,” and an image of the human past
grounded in lines of patrilineal descent: lists of names of men con-
nected by a series of “begats.” This is a kind of descent which is
called unilineal, because it allows for a reckoning of human related-
ness across generations through just one parental line. By naming
links between fathers/sons, but not mothers with children of either
gender, patrilineal descent makes for a bracketing of parts of the
biological givens of human “descent.” Familiar images of the deep
past, transposed to previously unknown circumstances, helped to
create a new framework of knowledge when “language” came to
name a group of objects which were similarly located and related
in time and space.
Imaging the Linguistic Past 65
Two Stories about Sanskrit and Javanese
Texts, grammars, and genealogies can all be represented with
the broad historical image I have outlined here without discussing
any particular colonial project. To show its importance as a resource
for legitimizing colonial authority, I can conclude this chapter by
comparing two different ways this image figured in the study of
the literature and grammar of a single language, Javanese. These
accounts are worth sketching to show that this image helped to
create pictures of that language’s past which were in important
ways contradictory, but which nonetheless both helped to
explain Javanese speakers’ inferiority to the Dutch (among other
Europeans) who came to rule them in a 19th century colonial
regime.
I noted in chapter 2 that the Sanskrit language which Jones “dis-
covered” was in its own way as well traveled as Latin, but not as a
language of power imposed by conquerors on their subjects (Pollock
2000). As Sanskrit traveled with traders, religious professionals, and
literati it was assimilated into local languages and societies. In much
of Asia, Europeans encountered orthographies they came to recog-
nize as similar to Sanskrit, and literary traditions eight hundred or
a thousand years old. Long periods of time had allowed for dynamic
interplay between Indic literature and local languages, producing
the “vernacular” literary cultures, as Pollock calls them, which
Europeans discovered and studied.
Java offered clear evidence of extensive influence from India in
the deep past: Hindu (Sivite) and Buddhist temples, and an exten-
sive body of literature written in what was once called the Kawi
language, and is now more frequently called Old Javanese. Dutch
colonialists paid little attention to this literature until the beginning
of the 19th century, when their interests shifted from trade with the
Spice Islands to control of the kingdoms of inland Java, whose
ruling nobility had inherited Kawi texts as a legacy from their pre-
Islamic past.
The British occupied Java during the Napoleonic Wars, and so
between 1812 and 1814 Kawi came to the attention of governor
Stamford Raffles (who owes his fame more to his founding of
Singapore). His compendious History of Java (1979[1817]) drew the
66 Imaging the Linguistic Past
attention of European readers to this ancient Javanese language,
including Wilhelm von Humboldt, a founding figure in German
philology, whom I discuss in the next chapter.
Von Humboldt took steps to acquire unpublished Kawi texts
because he regarded that language as a test case for the broad ideas
about language, culture, and history which were then developing
in German intellectual circles – which included Johann Herder. Von
Humboldt saw in Kawi texts the possibility of establishing whether
(as Herder argued) languages are in fact organically bound to envi-
ronments and communities of speakers, and so “naturally” resistant
to influences from “outside.” Kawi was interesting in this regard
because it was originally a “primitive” language, ordinary Javanese,
which had come under the influence of polished, written Sanskrit.
Kawi literature was where “Indian influence was most deeply and
penetratingly at work upon Malayan (i.e., Javanese) culture” and
so, von Humboldt believed, should bear clearest evidence of “the
most intimate intertwining of Indian and indigenous culture . . . on
the island that possessed the earliest and most numerous Indone-
sian settlements” (von Humboldt 1988[1836]:20).
Von Humboldt explored this question in what became a sprawl-
ing, famous, but now largely unread four-volume opus: On the Kawi
language. In this work von Humboldt sifted lexical material (as did
Jones) for evidence of “genetic” relations, discovering them between
Javanese and languages spoken as far north as Luzon, as far east as
the South Sea islands, and Madagascar to the west. (In this way he
charted the largest dimensions of another of the world’s major lan-
guage families, now called Austronesian.)
But von Humboldt’s focus was Kawi grammar, which led him to
study the oldest available Javanese version of the Sanskrit epic
Bharata Yuddha. He appreciated this text aesthetically, as a “literary
monument” from the past, but sought to read it diagnostically for
the data that would speak to larger questions about language and
human history. His basic conclusion was that “however many San-
skrit words it may have incorporated, [Kawi] does not cease on that
account to be a Malayan tongue.” In fact, “it robbed the incompa-
rably nobler Sanscrit of its own form, to force it into the local one”
(von Humboldt 1988[1836]:33).
Von Humboldt developed a language-centered story about a
primitive people who were incapable of fully incorporating the
Imaging the Linguistic Past 67
benefits of an “outside” civilization, one with a larger moral for
those concerned with contemporary questions about civilizational
expansion and colonial power. If Java had never been home to more
than a kind of civilization manqué, then Europeans there were
dealing now with modern inheritors of a lesser tradition whose
ancestors had fallen away from a civilization they never really made
their own. In this way they differed from Europeans, von Humboldt
noted, whose ancestors had taken up the language of Rome’s im-
perial project with a new vigor and liveliness.
Another, very different, text-centered picture of Java’s past
coalesced in another body of work by Dutch scholars. By 1830, the
Dutch had supremacy in Java after defeating Prince Dipanegara,
whose revolution in the name of Islam gave them one more reason
to fear and distrust that religion: its monotheistic theology strength-
ened native resistance to Christianity, and had the potential to
transform docile peasants into revolutionary fanatics. As they con-
solidated control of the interior of Java, and increased their contact
with its noble houses, the Dutch became more aware of the rich
body of literary Javanese texts, and began the work of textual com-
parison and reconstruction I noted above. Over the decades,
scholars produced critical editions and translations of a range of
works of Javanese literature written as early as 1000 ce, and as
recently as the 18th century.
Taken as a whole, and simplifying greatly, the overarching his-
torical narrative which came to be told about this body of work can
be seen as resonating with that developed by Jones and other British
scholars for India. It told of the decay of an ancient Sanskritic tradi-
tion that fell away from its original, pure beauty. The commonest
periodizing of this literature – Old, Middle, and Modern Javanese
– implicitly makes Old Javanese a reference point for evaluating the
fidelity of later texts to the original, “correct” Sanskrit vocabulary,
quotations, and glosses of verses. This narrative also fits Said’s
profile for the orientalist scholar as bringer of European reason to
recuperate and restore traditions for their native inheritors.
In Java, at least, this way of elevating an older, more “authentic”
past had broader political significance in an Islamic present. Islam
in this narrative was a latecomer to the island, arriving no earlier
than the 15th century, and putting a “thin varnish of Mohammad-
anism over a base of Buddhism and Saivism” (Florida 1995:188).
68 Imaging the Linguistic Past
The textual scholars who recovered that “base” for the Javanese
could in this way help save them from the “gullibility and igno-
rance . . . [which made them] easy prey for those who, under mask
of religion (i.e., Islam), hide sinister intents” (Florida 1995:188).
In this way, then, grammar-centered and text-centered versions
of Javanese history could differ in substance but legitimize both
European superiority in the world, and Dutch rule in Java. Von
Humboldt compared lexical and grammatical elements to under-
stand “large-scale” historical connections between peoples, whereas
Dutch philologists compared texts within a single historical tradi-
tion to deduce links of “descent” (and decay) between them. Where
von Humboldt demonstrated the imperviousness of a primitive
language to the benefits of its civilized superior, the Dutch traced
the decay of an ideal language rooted in that same civilized lan-
guage, and showed how that decay was being hastened if not
caused by the incursions of a false religion. Two language-centered
images of the past, guided and shaped by social imaginaries and
scholarly procedures, helped develop multiple understandings
of Europeans’ place, obligations, and privileges in a colonizing
world.
The Linguistic Past in the Colonial Present
The linguistic images of the past I have sketched here came into full
maturity in the science of language, comparative philology, which
I discuss in the next chapter. There I trace images of linguistic
descent-as-reproduction, which in turn were exported to the colo-
nialized world as an important aid for Europeans who made lan-
guages and language difference go hand in hand “with the creation
and reification of social groups . . . [on a] social and political map”
(Cohn 1996b:22).
I have focused here on comparative philology’s prehistory to
show how microscopic, recondite linguistic details could figure in
broader questions about global human diversity, and telescopic
views of alien, colonialized peoples. William Jones may have been
constructing a “rational defense of the Bible out of the materials
collected by Orientalist scholarship” (Trautmann 1997:42), but the
historical contours of his defense could be transposed and adapted
Imaging the Linguistic Past 69
to many other contexts and purposes. Shared social imaginaries
helped make the story that Jones and other philologists dev-
eloped the best sort, which, as Winston Smith notes in 1984,we
already know.
Herder was a speculative philosopher but no linguist, and his
framing of linguistic diversity was more speculative than empirical,
and explicitly nonbiblical. But his was a story of destinations pre-
given with origins, and carried its own legitimizing moral for
Europe’s nations in the world they were beginning to dominate. He
showed Germans, and other Europeans, that with their own lan-
guages they inherited their futures, each what the linguist von
Humboldt called an “emanation of the spirit, no work of nations
but a gift fallen to them by their inner destiny” (von Humboldt
1988[1836]:24).
Notwithstanding differences between their images of the linguis-
tic past, Herder and Jones together helped to frame a scholarly and
colonial project which would “transmute polyglot agonies of Babel
into a cult of transcendent European erudition” (Herzfeld 1987:31).
From their painstaking microscopic studies emerged a telescopic
view of linguistic and human diversity as “a marvelous, almost
symphonic whole whose progress and formations, again as a whole,
could be studied exclusively as a concerted and secular historical
experience, not as an exemplification of the divine” (Said 1995:21).
Notes to chapter 3
1 For further discussion of this issue from an anthropological point of
view, see Johannes Fabian’s Time and the other (1983).
2 For extensive discussion of scholars who prepared the ground for Jones’
famous work on one hand, and Herder’s influential philosophy on the
other, see Aarsleff 1983 and Trautmann 2006.
3 See for instance Metcalf 1974 and Gulya 1974.
4 Bishop Ussher had calculated that the world was created in 4004 bc,
and that God caused the flood of Noah in 2349 bc.
Chapter 4
Philology’s Evolutions
[T]his crowd has thrown itself into the science of language: here, in
an endless region of tillable land, freshly opened up, where presently
even the most mediocre gifts can be employed with profit and a
certain emptiness is already even considered as a positive talent, with
the newness and insecurity of the methods and the continuous danger
of fantastical errors.
Friedrich Nietzsche, On the future of our educational institutions
(2004[1872]:71)
Nietzsche may have had more respect for leading figures in the
“science of language” than members of the crowd whose imagina-
tions, as he observed in the passage above, the new field had seized.
By the time he made this remark, in a lecture on Prussia’s educa-
tional institutions, that science – Sprachwissenschaft, or comparative
philology – had come fully into its own due to new discoveries and
insights into “deep” historical relations among languages and
humankind. He might have been skeptical rather of comparative
philology’s ascendance over his own field, text-centered, classical
philology (Alterthumwissenschaft), thanks to “the crowd” that
misperceived their field’s importance and misused its key terms in
increasingly broadened, literalized ways. Through repetition, revi-
sion, and the “dull regimental routine” of normal science, terms like
“organic” and “linguistic organism” became more and literal, losing
what Nietzsche would call their “sensuous force” as metaphors.
Perhaps Nietzsche had comparative philology on his mind when
Philology’s Evolutions 71
he wrote a year or so later about people’s inability to deal with
reality without making use of a “movable host of metaphors,
metonymies, and anthropomorphisms” (1979[1873]:77).
Certainly core terms of the new science of language were trans-
formed in sense and significance when comparative philologists
spoke as public intellectuals about questions of national culture,
politics, and history, and as they created the first academic disci-
pline which had “evolution as its very core” (Hobsbawm 1964:337).
So it is important here to trace the development of philological
notions of “evolution,” and show how this science produced what
Stephen Alter calls a “master metaphor” of a range of disciplines:
“comparative mythology, comparative jurisprudence . . . [and]
Indo-European studies in general” (Alter 1999:109), and eventually
Darwinian natural history. In this way, finally, comparative philol-
ogy provided evolutionary support for ideologies of European
empire.
This chapter is centered on two ways that philological images of
language were influential beyond the academy. The first is the
broadly organic view of history which helped to explain Europe’s
superiority in a colonial present, naturalize its ongoing civiliza-
tional advancement, and frame linguistic difference as human
inequality in a colonial world. The second centers on philology as
a very German science which made the past into a resource for
nationalist ideologies in an industrializing Europe, nowhere more
importantly than in Germans’ confrontations with a political and
cultural crisis of identity quite close to home.
By focusing on the progressive loss of these organic metaphors’
“sensuous force,” I frame a sketch which must be somewhat repeti-
tious, bypassing abstruse details which distinguished and marked
the field’s progress. It may seem specious to focus more on broad
pronouncements than painstaking empirical work, especially when
connections between the two are not always clear. The tenuousness
of some of these broad claims based on narrow descriptions was
not lost on some of these German philologists’ clear-eyed contem-
poraries, including the German-trained American William Dwight
Whitney. In his 1873 review of an essay on language origins by one
German philologist, he commented acidly on the tendency of
Germans to present “a minimum of valuable truth wrapped up in
a maximum of sounding phraseology” (Whitney 1873:292). But
72 Philology’s Evolutions
what seems otherwise to be a mixing of empirical wheat and rhe-
torical chaff is important here for understanding how philologists
used linguistic diversity to confirm the place of Germany in the
world, and the superiority of what is now called “The West” over
“The Rest”. In this way their work helped provide a linguistic image
of what Nietzsche called the “sum of human relations,” metaphors
of evolution which became “intensified, transferred, and embel-
lished” in broader projects of colonialism on one hand, and nation-
alism on the other.1
The Brothers Schlegel
Before 1806, Liah Greenfeld (1992:278) has argued, no idea of nation-
alism existed among German speakers. But by 1815 it had already
come of age among members of a “peculiar class of educated
commoners, professional intellectuals” who were afflicted by an
“oppressive sense of status-inconsistency” after the demise of the
Holy Roman Empire. Greenfield identifies 1806 as a formative year
in this development because it was then that Napoleon humiliated
an overconfident Prussian army at Jena, forced the king to cede
almost half his realm to the French, and threw what were then still
called “the Germanies” into crisis.
Johann Herder, a member of Greenfeld’s group of proto-nation-
alists, had already been engaged with this very crisis of identity
when he wrote his speculative account of language origins dis-
cussed in chapter 3. His celebration of local, expressive authenticity
inspired others who were likewise grappling with the political and
cultural crisis brought on by Prussia’s defeat, including the brothers
Friedrich and August von Schlegel. They were to broaden Herder’s
ideas in their own work of comparison between languages and
literatures.
Friedrich, born in 1772, might be called a visionary rather than a
founder of philology thanks to his 1808 book The language and
wisdom of the Indians. This helped promote the flow of new knowl-
edge about India to German via Paris, intellectual crossroads of
Europe, where Friedrich studied Sanskrit with one Alexander
Hamilton, a retired army officer in the East India Company who
later became Britain’s first Sanskrit professor.
Philology’s Evolutions 73
Friedrich von Schlegel sought to develop a systematic under-
standing of historical relations between Europe and Asia by
paying more attention to the grammatical elements of languages
than the lexical material which William Jones made the object of his
etymological speculations. Schlegel recognized grammar to be a
more coherent domain for study, its elements being far fewer than
indefinitely large vocabularies, and recurring more frequently
as “obligatory” components of language use. Like his brother
August and his successors, he approached these elements with
a founding typological distinction between languages whose
grammars counted as “organic” or “mechanical.” This distinction
was supported, in his view, in two different ways: by differences
in the structures of languages, and the ways they changed
over time.
For Friedrich, “organic” languages – Sanskrit, Persian, and those
of Europe – shared a common grammatical technique he called
“Flexion”: their words were obligatorily comprised of roots and
additional elements marking number, tense, gender, etc. On the
other hand “mechanical” languages – Chinese, Coptic, Basque,
Amerindian languages, and Arabic (Koerner 1990:250) – had more
primitive grammatical elements, called “Affixa” by Friedrich, which
attached more loosely to word roots.
The superiority of organic languages endowed with “flexion,”
Friedrich argued, is evident in their suitability for comparative and
historical analysis. They are not passive in the face of external forces
like mechanical languages, which possess no internal coherence or
capacity to maintain their continuity over time. Because grammars
of mechanical languages have an accidental character and artificial
complexity, he argued, it is “nearly impossible to trace them to a
common ancestor” (Koerner 1990:250). Because organic languages
actively engage with historical forces, reshaping themselves rather
than passively suffering change, they are more than mere objects of
historical forces, and so can be studied in and with history. In this
way Friedrich gave Herder’s broad organismic vision of language
empirical purchase for questions about historical relatedness of lan-
guages, proposing that grammatical categories offered a means for
comparing languages “in a similar way as comparative anatomy
has illuminated the higher natural history” (Schlegel 1808, quoted
in Koerner 1990:243, my translation).
74 Philology’s Evolutions
August von Schlegel, Friedrich’s brother, extended and particu-
larized the organic image further in his studies of literary history,
which did not yet count as a field distinct from that of grammar. In
1818 he gave a series of lectures on the diversity and development
of genres of verbal art using a distinction much like his brother’s to
distinguish two kinds of literary form. A text has mechanical form
“when, through external force, it is imparted to any material as an
accidental addition without reference to its quality. . . . Organical
form . . . is innate; it unfolds itself from within, and acquires its
determination contemporaneously with the perfect development of
the germ” (Schlegel 1965[1818]:340).
This “lesson of history” offered a broader moral, which August
drew by echoing Herder. These facts of literature demonstrated
how a true poetic spirit develops “according to laws derivable from
its own essence” (Schegel 1965[1818]:340), working in accord with
(its own) nature rather than laboring under “the authority of the
ancients” (1846:339). These remarks allowed August to introduce
comments on a more immediate problem into his discussion of
Provençale literature: the Francophile Prussian nobility which, like
Herder, he criticized for turning away, unnaturally, from its own
language, culture, and identity. August concluded his lectures with
a veiled reference to these “persons of higher ranks, [who] by their
predilection for foreign manners . . . have long alienated themselves
from the body of the people” (Schegel 1965[1818]:529).
By embedding the study of grammar and literature in history,
the Schlegel brothers helped invest “language” with larger political
and cultural meanings. At the same time they lent empirical weight
to Herder’s speculative ideas about language origin and change,
while proclaiming that Germans must recognize their language and
literature’s organic vitality if they were not to risk what August
called “disappearing altogether from the list of independent
nations” (Schegel 1965[1818]:529).2
Wilhelm von Humboldt
The year after his lectures on Provençale literature, August von
Schegel accepted the chair of Sanskrit at the University of Bonn,
where he further elaborated his brother’s typological approach to
Philology’s Evolutions 75
grammar. He owed this appointment, and much intellectual inspi-
ration, to a friend and fellow scholar: Wilhelm von Humboldt,
whose studies of Kawi language I discussed briefly in chapter 3.
Later in the 19th century, von Humboldt came to be known as the
“cofounder of the new linguistics” (Benfey 1869:279), but he had
made another place for himself in Prussian history through his
work as architect of an entirely new system of national education.
King Wilhelm, galvanized by his disastrous defeat at Jena, swept
away the fragmented, backward-looking noble class ruling a patch-
work of 1,800 tiny territories. He abolished serfdom and permitted
non-nobles to own land; he removed the guild’s stranglehold on
crafts and professions; he instituted new forms of local self-govern-
ment. Beyond all of these moves, he saw the need for a new class
of governing officials, and an entirely new set of institutions to
educate its members. He summoned Wilhelm von Humboldt
to design them.
A wealthy noble, steeped in the study of Latin and Greek, von
Humboldt had spent much of his life in close contact with intel-
lectuals and language scholars in western Europe, not Prussia. But
he answered his king’s call, returning to Berlin in 1809 to help
devise an educational response to his country’s crisis. He did this
work as a child of the Enlightenment, but also as a scholar deeply
imbued with Herder’s sense of language as “the outer appearance
of the spirit of a people” (von Humboldt 1988[1836]:46). During his
one year in service to the king he achieved at least the reputation
of having designed an entirely new, language-centered system of
Prussian education (Holquist n.d.:31) in which those seeking to
become teachers were required to devote years of study to Latin
and Greek.
Philology had a central place in von Humboldt’s new universi-
ties, and his efforts to create “a German national identification with
the classical past centered around ancient Greece as . . . a cultural
ideal to which German education aspired” (Benes 2001:175). As
German scholars and institutions became pre-eminent in the field,
they were able to demonstrate historical parallels between the pol-
itics of Prussian resistance to the French in contemporary Europe,
and the politics of a politically fragmented Greece which had been
dominated by Rome, but was still united by a superior language
and culture.
76 Philology’s Evolutions
Today von Humboldt is known to linguists much less for his
studies of Kawi, discussed in chapter 3, than for a grammatical
typology he developed in the introduction to his voluminous body
of work. That introduction, still in print – in translation with the
title On language: the diversity of human language structure and its
influence on the mental development of mankind (von Humboldt
1988[1836]) – extended the Schlegels’ line of thought by distinguish-
ing not two but four types of grammar. Languages whose words
consist of single elements, like Chinese, von Humboldt counted as
“isolating” languages, “with no grammatical structure” and so in
his view (and August’s) as essentially flawed as a means for intel-
lectual expression (Schlegel 1965[1818]:14). Other languages, many
spoken in North America, combine words to create more complex
words, with grammars von Humboldt called “agglutinating.” A
third group of languages, represented by Basque, which von Hum-
boldt himself had studied, have grammars he called “incorporat-
ing” or “polysynthetic” to describe the ways they fuse words into
inseparable wholes which were equivalent to entire sentences.
In the fourth category von Humboldt, like August von Schlegel,
placed the “first rank” of languages: those that have inflecting
grammars. Words in languages like Greek, Latin, and Sanskrit
(but not Javanese, as we saw in chapter 3) incorporate obligatory
modifying grammatical elements which count as signs of “a liv-
ing principle of development and increase.” These languages
alone have “fecund and abundant vegetation” (von Humboldt
1988[1836]:86).
This global comparative perspective which von Humboldt
brought to bear on his reading of Kawi would, he thought, help to
answer “the difficult question” of whether “languages are able or
not to gradually change in nature, to pass from the first class to the
second, and from the second to the third. If it were possible to
respond to these questions with facts as certain evidence, a mass of
problems relating to the origins of civilization would be thereby
resolved” (von Humboldt 1988[1836]:85, fn.7).
Sanskrit and other inflected languages could serve as reference
points, von Humboldt explained, because in them “the mental
cultivation of mankind has evolved most happily in the longest
sequence of advances. We can therefore regard them as a fixed point
of comparison for all the rest” (von Humboldt 1988[1836]:216).
Philology’s Evolutions 77
Javanese grammar “suffers . . . from the Chinese lack of inflection
. . . [but] does not, like Chinese reject grammatical formation with
scornful resignation” (von Humboldt 1988[1836]:191). So the organ-
ism of Javanese failed to move beyond its initial, “formative” period,
or to achieve the kind of “elaboration” (Ausbuildung) which would
lift it above the rank of derivative languages and civilization.
These joined questions about grammar and history were a con-
stant element of philological efforts to make the past speak to the
present, and to single out one group of languages and peoples from
all others. Philological images of civilizational history, running
from Greece to Germany, seized the imagination of some Germans
so powerfully during this time that they ventured into the Caucasus
mountains in search of “their” people’s original locale and lan-
guage. By 1823 Germans were studying there under sponsorship
from the Russian czar, who had strategic interests of his own in the
area. They contributed to the conjectural history of Indo-Germanic
peoples who were understood to have migrated from the Caucasus
as one of the 13 “main tribes” (Stammvolker) who survived the bib-
lical flood (Benes 2001, 2004). It was this image of the past, regular-
ized and elaborated by the philologists I discuss below, which was
popularized late in the 19th century by the German philologist Max
Müller as the story of the Aryans, and later was appropriated by
the Nazis in the 20th century. I discuss this image of the past later
as a template for the stories linguists devised about other peoples
as colonial subjects, and for purposes of colonial regimes.
Franz Bopp
Von Humboldt and the Schlegel brothers’ sweeping vision now
seems unscientific because of the ways it blurs lines between spheres
of study: speech and text, grammar and history, language and
culture. Science involves procedures and questions which produce
a tighter empirical focus, and insulates data from extraneous con-
siderations. In this respect, another breakthrough to the science of
language can be found in the work of Franz Bopp, von Humboldt’s
own tutor in Sanskrit, who published in 1816 On the conjugation
system of Sanskrit in comparison with those of Greek, Latin, Persian and
German languages.
78 Philology’s Evolutions
Bopp, unlike von Humboldt, came to ancient texts not as “literary
monuments” – expressions of the genius of a people, or windows
on distant cultures – but as repositories of examples of usage. He
found there the specific data he could abstract to develop a rigor-
ous, comparative metric for grammatical study. He developed a
more abstract notion of linguistic form, isolating part-for-whole
relations between roots and grammatical elements as the core
of “the true organism of a language” (quoted in Perceval 1987:7).
It was this organism which, under close study, would allow for
new understandings of “physical and mechanical laws” of opera-
tion, and “the origin of the forms which indicate grammatical
relations.”
Philology in this way became truly anatomical because texts
were used “to describe the organism of a language for its own
sake,” as Bopp wrote in 1829 to von Humboldt (quoted in Davies
1987:89). When a language was understood to be a “natural body,”
its properties could be traced to the ways they “form themselves
according to definite laws, develop carrying in themselves an inter-
nal life principle, and gradually die” (quoted in Davies 1987[1836]:84).
New rigor allowed narrower criteria for evaluating languages
through their grammatical properties, as Bopp argued, by dis-
tinguishing languages that have grammar (Grammatik) and so an
organism (Organismus) from those that do not. By this measure,
Chinese counted (again) as lowest among the world languages,
and the Semitic languages a bit higher thanks to their disyllabic,
triconsonantal roots. But both types of language were in turn
inferior to others whose monosyllabic roots are capable of com-
pounding, “acquir[ing] their organism, their grammar, almost
only in this way” (quoted in Davies 1987:85). These are languages
of the family now called Indo-European, which are unmatched
in the “beautiful joining of these complements into a harmonic
whole with the appearance of an organic body” (quoted in Davies
1987:85).
With this new analysis of form, Bopp joined grammar to history
in another way: organic languages are dynamic, autonomous, and
change in accord with their internal developmental logics. Other
languages’ “variously disfigured and mutilated forms” are evidence
of their lack of such principles of internal development, and their
passivity in the face of historical forces which change them.
Philology’s Evolutions 79
Jakob Grimm
At the same time Bopp was developing his rigorous approach to
linguistic diversity, Jakob Grimm was publishing discoveries about
a narrower range of languages in a shallower, predominantly
Germanic period of history. But this work brought him major sta-
ture as a philologist and public intellectual. Born in 1785 in French-
occupied Prussia, into marginally middle-class conditions, Jakob
made his way in the world with his younger brother as a scholar
whose different projects served a single end: recapturing and resus-
citating the true heritage of German-speaking peoples. The brothers
Grimm heard Herder’s call for cultural renewal very clearly, as can
be seen in the title of a journal they established in their youth: Old
German forests (Altdeutsche Walder), devoted to “customs, laws, and
norms that bound German people to each other” (Seitz 1984:48,
quoted in Zipes 1988:45). This title echoes that which Herder chose
for four volumes of his own early nationalist writings, Critical forests
(1769).
The Grimms undertook a massive dictionary project, now far less
famous than their collection of folk stories which remained incom-
plete at their deaths. But in the course of his work on that project,
Jakob came to recognize that there were systematic patterns of
sound difference between words of modern German, older German,
and a range of related languages. From these patterns he inferred
that they were all linked to each other by an extended historical
process of sound shift (Lautverschiebung). In his compendious
grammar of German he set these patterns out as the basis of what
came to be known as Grimm’s law – “If non-specialists know any-
thing about historical linguistics, it is Grimm’s law. The history of
views on the consonant shift is virtually a history of linguistic
theory until 1875” (Lehmann 1967:46).
Grimm demonstrated a general, systematic correspondence
between subsets of the consonants found in western European lan-
guages, as they appeared in certain positions in cognate words. He
noticed for instance that words in Gothic texts which were spelled
with an fhad Greek counterparts spelled with the equivalent of p,
while their Old High German counterparts were words spelled
with b. These patterns were general because they extended across a
80 Philology’s Evolutions
wide range of words, allowing Grimm to demonstrate part-
for-whole relations between three trios of sounds: one pronounced
with the lips and/or teeth (p/b/f), a second with the tongue and
teeth (t/d/th), and a third in the back of the tongue and mouth
(k/g/ch). These patterns were systematic because the only way to
account for these correspondences was to recognize that each was
nested in a larger whole, each trio linked to the others: p/t/kunder-
went similar changes, over and against b/d/gand f/th/ch.3
This was a discovery about language which brought “deep”
history close to home, and so had immediate purchase on the liter-
ate German-speaking public. Grimm had shown how everyday
speech bound them organically, through speech, to their ancestors
in a kind of intimate but unconscious project, working itself through
between generations and across centuries. Right under their noses,
the German language’s historical unity was being created and recre-
ated by speakers every day.
Grimm’s life work drew inspiration from Herder’s philosophy,
but he avoided pronouncing on the question of language origins
until very late in his career. In 1851, he finally brought comparative
philology, the new science of language, to bear on that old problem.
After King Friedrich Wilhelm II invited him to join and address the
Royal Academy of Sciences in Berlin – his progressive nationalist
politics had cost him a job at the University of Gottingen – Grimm
took up the problem of language origins by demonstrating (again)
how organic languages are bound up with the broader develop-
mental dynamics of human history. He presented his law of sound
change as empirical evidence that “perfected inflections for
grammar” arise from processes of development, and as part of the
joined ascent of language and humanity. This path led for Grimm,
as for Bopp, from the polysyllabic words of an “artless,” “sensory,”
“naïve,” “garrulous period of use” to monosyllabic words which
were products of grammatical simplification and superior instru-
ments of “reason” (Grimm 1984[1851]:20). Grimm in this way
located himself, his audience, and their society in the civilizational
dynamic I sketched in chapter 3: the work of uncovering these
developmental laws demonstrated the capacities of reason they
enabled. Grimm showed how, as inheritors and witnesses to this
development, modern Europeans were becoming ever freer from
the limitations of linguistic form.
Philology’s Evolutions 81
Through these developments the field of comparative philology
came into its own as the first science of language as scholars rea-
soned against the grain of history to excavate original, pure, gram-
matical forms (Kiparsky 1974:333). In 1827 Bopp could already
summarize the common substance of the philological science in a
review of Jakob Grimm’s (1822) grammar of German. Science had
demonstrated, he asserted, that languages are
organic bodies of nature, forming according to certain laws, develop-
ing through an inherent life principle, dying gradually as, not under-
standing themselves anymore, they cast off their once meaningful,
now superficial mass of parts or forms, or mutilate them or abuse
them for purposes for which they were not originally suited. (Bopp
1827:251 quoted in Schlapp 2004:377)
Schleicher’s Evolutionary Tree
One last step in the literalization of organic metaphors is important
here as a kind of quantum leap made by August Schleicher (1821–
68). He had trained in text-centered classical philology, but after
shifting to grammar-centered comparative philology he gained a
considerable reputation with new strategies for organizing the
enormous masses of data assembled by others. His first major pub-
lication in 1848 augmented Bopp’s work with new evidence about
the history of Indo-European words and grammatical units, build-
ing on the distinction between isolating, agglutinating, and inflect-
ing languages set out above. Schleicher used comparative procedures
like those sketched in chapter 3 to reconstruct languages which are
now extinct (like absent “ancestor” texts) but whose properties can
be deduced from those of their “descendants.”
Schleicher in this way developed an account of what a linguist
of the next generation (discussed in chapter 6) called “that lofty
structure, the Arian ‘ursprache’ ” (Jespersen 1894:4). Although
Schleicher was not the first philologist to use the idiom of kinship
to describe relations between languages – an English student of
sound change had already done so (Alter 1999:10) – he gave that
metaphor weight and geohistorical scope with the image of the
Indo-European tree (a version of which is shown in Figure 3.2).
82 Philology’s Evolutions
Schleicher was to have a broader role in intellectual history,
though, when a zoologist who knew of his fondness for horticulture
and botany gave him a German translation of Darwin’s Origin of
species. After reading it Schleicher wrote an open letter to his young
friend, Ernst Haeckel, to assert that comparative philology not only
corroborated Darwin’s argument, but anticipated it. Philologists, he
asserted, had already laid down the very laws Darwin sought to
prove: “[t]he rules now, which Darwin lays down with regard
to the species of animals and plants, are equally applicable to the
organisms of languages.” So, he agrees, “species stand to genus as
daughters of one stock” (Schleicher 1983[1863]:33).
Schleicher in this way saw that comparative philologists had
without knowing it discovered and developed “methods largely the
same as that of the other natural sciences.” It turned out that lan-
guages really were organisms subject to laws of evolution: some
“[s]pecies and genera of speech disappear, and . . . others extend
themselves at the expense of the dead. . . . Not a word of Darwin’s
need be changed here if we wish to apply this reasoning to lan-
guages” (Schleicher 1983[1863]:62–64).
This kind of evolutionism has rightly and long been in disrepute
among linguists, who now all share William Dwight Whitney’s
skepticism about such pronouncements. But Schleicher’s reputation
was such that his letter became valuable rhetorical ammunition in
England, where Darwin’s supporters were seeking to convince a
wider audience (and so also seeking the upper hand in debates
with another well-known philologist, Max Müller, touched on
below).
The fundamental flaws in this analogy, which I noted in chapter
3, did not rob it of its rhetorical and ideological power. In this way
comparative philology helped to disguise the difference between
two different senses and images of “evolution,” which were so
important for imperial ideologies that they are worth distinguishing
here along lines developed by Joseph Fracchia and Robert Lewontin
in their critical review of the concept.
Fracchia and Lewontin (1999) distinguish two historical dynam-
ics which have been called “evolutionary.” The one they call “trans-
formational” is directional and articulates in ordered stages, each
depending on those preceding, and so incorporating them. This
broad transformational vision of change served, from the philoso-
Philology’s Evolutions 83
phy of Herder to the philology of Schleicher, to develop a teleo-
logical vision of languages and so of cultures, moving from state to
state in accordance with their internal “natures.”
The other kind of evolution Fracchia and Lewontin call “varia-
tional.” This was the focus of Darwin’s account of evolution, which
served to explain facts of difference among members of a given
species. Where comparative philology explained variation (or diver-
sity) in languages as a kind of byproduct or residue of their differ-
ing rates and trajectories of transformational evolution, Darwin
focused on mechanisms of sexual reproduction.4Though he had
available no strong understanding of the mechanisms of genetic
transmission, Darwin recognized that reproduction was a key point
of interaction between species and environments: through re-
production the organism proposes, and through selection the envi-
ronment disposes. Selection, then, is the name for outcomes of
interaction between environments and genetically determined char-
acteristics of organisms (Lewontin 2000). No directional dynamic is
needed or implied in this mechanism.
Schleicher’s parthogenetic image of linguistic “reproduction”, as
I showed in chapter 3, made it possible for his Indo-European “tree”
to blur this basic conceptual difference, and so also differences
between Darwinian (variational) and romanticist (transformational)
views of history. Once philology had literalized metaphors of
descent, language and linguistic change could be made to stand,
part for whole, for communities of humans and their histories, but
for natural history as well. By helping to make social Darwinism a
natural fact, comparative philology demonstrated the underlying
dynamics of civilizational progress, from industrialization to im-
perialism, in the world at large.5
Recovering Origins
This quick review of comparative philology’s development may
seem conceptually and geographically distant from work done by
linguists in zones of colonial contact which were developing so
rapidly around the world at this time, just as philologists’ academic
agendas seem quite detached from Europe’s colonial projects. But
the work of linguistics in the larger colonial world was in fact
84 Philology’s Evolutions
guided by these philological images of the past, and required tech-
niques that were developed and legitimized by the academy.
Chapter 5 sketches this interplay in the work of colonial linguists
who took their cues from philological images of the past in order
to devise strategies of descriptive selection and simplification. Here
it is useful to show first how philological images of the past played
into the ways colonialists explained what they discovered so as to
legitimize their own presence.
Well before philology came of age, the past had served as a
resource for explaining linguistic complexities in the present, and
for simplifying those complexities in writing. John Gilchrist, for
instance, a colleague of William Jones in Bengal, deduced that the
babble of speech confronting him descended from the original lan-
guage of an “Indian arcadia” which had been corrupted by repeated
waves of invaders, Arab and Persian, whom he likened to Norman
invaders of Saxon England. He sought with his descriptions to
capture a written image of that original language, and this became
a model for the British language of command in northern India,
called Hindustani (Cohn 1996b:37).
Colonialists of a later era used philological images of the past I
have discussed here to devise broadly similar histories. They regu-
larly had recourse to stories of war, conquest, and forced displace-
ment as explanatory factors for linguistic diversity, preferring these
to processes of emigration, trade, or cultural exchange (like that
which carried South Asian writing and literature so far from home).
The most famous and compelling of such images was developed by
Max Müller, a German philologist whose popular writings as a
professor in England helped make the story of the Aryans widely
known. This involved two waves of Aryan migration/conquest out
of the Caucasus, through Europe and South Asia. In the colonial era
these two branches of a single family were reunited when the British
came to India. Thomas Trautmann (1997) describes the meanings
of this story as it played into early 19th century British debates on
colonial policy, and its uses to throw still developing racial ideolo-
gies of empire into doubt.
By the beginning of the 20th century, though, it was common
knowledge that the British had followed in the footsteps of an
earlier wave of Aryan newcomers. The Oxford history of India, a
standard textbook for candidates in the Indian Civil Service and
Philology’s Evolutions 85
higher education, described how “[f]rom the Vedic hymns it has
been possible to piece together a reasonably coherent picture of the
Aryan invaders on their first impact with the black, noseless (flat
nosed) dasyus who comprised their native opponents and subjects”
(Smith 1919:2).
Images of invasion – often from the north, as in South Asia – were
a leitmotif of conjectural colonial histories, a series of “just so”
stories to account for conditions of social and linguistic complexity
which Europeans encountered and themselves intended to domi-
nate. Highly encapsulated versions of a few such histories that were
devised for regions of sub-Saharan Africa help to illustrate how this
image could be transposed, and provide background for the work
of colonial linguists discussed in chapter 5.
In the Senegambia region of Africa’s eastern coast, for instance,
French explorers and military forces encountered neighboring com-
munities where three different languages were spoken. According
to Irvine and Gal (2000), Fula was spoken over a wide area and was
strongly associated with Islamic orthodoxy; Wolof was spoken pre-
dominantly in northern coastal regions by Muslims; and Sereer was
the language of non-Muslims living in a more restricted area to the
south. Evidence now suggests that this diversity emerged from
political dynamics which contributed to Wolof’s value as a lan-
guage of high level political relations, and so as an advantageous
second language for native speakers of Sereer.
For the French, though, this condition of asymmetric bilingual-
ism was evidence that the Wolof had conquered the Sereer, while
the Fula, for their part, were superior, lighter skinned migrants
from regions in Upper Egypt to the north. They were thought to
possess higher intelligence, a superior religion, and a more elabo-
rately hierarchical society. So too it followed that their superior
culture and language had a kind of trickle down influence upon the
Wolof, as Wolof had on the “simple” Sereer.6
To the west, in central Africa, similar histories helped account for
the enormous linguistic diversity Europeans encountered in the
Congo. One library researcher in Europe, Gaston Van Bulck (1903–
66), pieced together the linguistic map partly reproduced in Figure
4.1, a mosaic of scores of languages scattered across tiny bits of ter-
ritory. The basic typological demarcation on this map – represented
here by a heavy black line running roughly east and west – divides
Figure 4.1 Van Bulck’s linguistic map of the Belgian Congo (reproduced from Van Bulck 1948).
BELGIAN
CONGO
ANGOLA
Philology’s Evolutions 87
languages of two different families: Bantu to the south (which I
discuss later) and Nilo-Saharan to the north. Van Bulck inferred
from the literature then available, produced by linguists working
in the region, that there were numerous tiny, moribund language
“enclaves” on either side of this line. This scanty evidence sufficed
for him to deduce that these were remnants of intertribal conflicts
which he often described “in terms of an extended battle metaphor,
reflecting his view on today’s languages as winners in the survival
of the fittest” (Van der Velde 1999:480).
I return to the Congo’s linguistic past in chapter 5; here it is
enough to see how Van Bulck’s new-yet-old story of invasion and
conquest could be plausible because it partook of philological
images and historical narratives which had been used to concep-
tualize linguistic diversity elsewhere, beginning with the Indo-
European languages.
Van Bulck relied heavily on work by William Bleek, a profes-
sional philologist who, quite unusually for his time, actually
engaged speakers of languages he studied in their own communi-
ties. In 1862 he had published a comparative grammar of the South
African languages in which he introduced the label “Bantu,” still
used today. (It was Bleek whose reflections on language origins
elicited Whitney’s sarcastic observation on the “sounding phraseol-
ogy” of German writing quoted at the beginning of this chapter.)
Bleek’s most famous successor, Carl Meinhof, had developed this
comparative enterprise into an independent academic field called
Afrikanistik, breaking with older philology by claiming that he had
the distinct advantage of studying languages as they were spoken.
Eliciting data of speech directly from native speakers, he claimed,
brought philology finally to “the life of language,” and not just its
“shadowy fragment[s]” in the texts studied so intensively by his
Indo-Europeanist colleagues.
But from Meinhof’s new science came the familiar story which
Van Bulck in turn adopted and specified for the Congo. Africa was
home, Meinhof argued, to three races: the Sudanic (true Negroes),
Hamitic (light-skinned conquerors), and Bantu (a mixture of the
two). In these names lies the kernel of Meinhof’s story: Hamites
entered Africa as the Aryans had India, “a superior race with the
most noble physical and psychic characteristics” (quoted in Pugach
2001:41). They were closely related to Semitic speakers who had
88 Philology’s Evolutions
come to sub-Saharan Africa from the Middle East or Central Asia,
migrating south until violent encounters with dark-skinned Negroes
gave way to contact, intermarriage, and “the languages and com-
munities we still call Bantu” (quoted in Pugach 2001:60).
The “Hamitic hypothesis” did not survive scrutiny for long, in
part because it depended on the blurring of differences between
race and language in empirically implausible ways. What is impor-
tant here, though, are the intellectual traditions and colonial inter-
ests which gave Meinhof’s account plausibility and usefulness. It
framed the precolonial past to fit the colonial present, especially
during Prussia’s aggressive imperial African campaign in the 1870s,
and survived up through Afrikaner apartheid policies in postcolo-
nial South Africa (Pugach 2004).
As men of science and faith, Meinhof (Lutheran) and Van Bulck
(Catholic) counted as scholars in the academy, not researchers in
the field. Once Meinhof’s “school” of African linguistics in Hamburg
became a center for research with native speakers who were
imported from Africa, it also became an important destination for
Europeans (mostly German, many missionaries) embarking for or
returning from Africa. That Meinhof was an armchair scholar in no
way diminished his reputation among these visitors, including
those who had grown up in Africa and had acquired native fluency
in languages they came to Meinhof to study. Such was the author-
ity of philological science that he could hold out the prospect of
helping them really understand what they already knew.
From Nationalism to Colonialism and Back Again
This foreshortened sketch has focused more on dubious heuristics
than substantive discoveries because I am concerned with com-
parative philology as a means for making linguistic diversity useful
in a world of nation-states and empires. As organic metaphors
became literal, images of the linguistic past took on the kind
of facticity which allowed them to be transposed between places
and issues: kinship and texts, language and race, grammar and
history.
I have also emphasized the political and cultural contexts of these
formative images as they were devised, applied, and elaborated.
Philology’s Evolutions 89
Herder, the Schlegels, von Humboldt, the Grimms, and others
saw their work as parts of a nationalist project, shared with
other German-speaking peoples, to realize an autonomous nation.
Applied locally, images of language history fitted and corroborated
their shared sense of national destiny in Europe, and later, as an
imperial power, the world at large. Applied globally, these same
images helped to chart the long historical trajectory in which im-
perialism emerged as a natural continuation of the civilizational
dynamic. When Europe’s nation-states could be understood to have
achieved historical autonomy as active, collective agents, the condi-
tion of modernity (Dirks 1990) they had come to inhabit could be
extended to other places, and bestowed on other peoples. By focus-
ing on comparative philology’s evolution in Europe before turning
to work of linguists abroad, I have emphasized the ways this science
of language emerged within the “unitary field of analysis” (Cohn
1996a:4) of colonial history.
This sketch also suggests an important critique of Said’s account
of orientalism, which focused on work done by French philologists
at a time when Prussia was too occupied with its own survival to
embark on imperial projects abroad. The result was a split between
Germans, who were the primary driving force behind philology’s
evolution, and the agents of imperial power who found that field
ideologically useful, as Said suggests, for legitimizing projects of
conquest. One symptom of this division of labor is apparent in the
career of Max Müller: when the xenophobic English sought out an
academic expert on India, the jewel in their imperial crown, they
finally and grudgingly employed a scholar whose native country
had no comparable presence in the colonialized world. (See van der
Veer 2001 for further discussion.)
This transposability of linguistic images of the past to political
and cultural projects in the present – for autonomy (as in Germany)
and domination (as in many different colonies) – is important
enough to deserve one more illustration from a situation which
combined both conditions. I mentioned in chapter 3 that European
missionaries had long been familiar with Tamil, a language spoken
in southern India, before William Jones “discovered” Sanskrit in
Bengal. The earliest descriptions of Tamil by Europeans date from
1310, well before the discovery of the New World, and by the time
German Protestant missionaries arrived in the early 17th century
90 Philology’s Evolutions
considerable attention had been devoted to the language. Nonethe-
less, the vast majority of Tamil speakers recognized the religious
and political superiority of members of the Brahman class who, as
the story of the Aryans explained, were superior by virtue of their
genealogical and cultural inheritance: the original Hindu religious
tradition and its Truth language, Sanskrit.
After Jones published his findings on relations between lan-
guages of Europe and those of northern India, another British civil
servant working in Madras, Francis Ellis, published his own find-
ings on relations between languages spoken to the south.7These
became better known in 1856 when Robert Caldwell, a Scottish
missionary, published a comparative grammar of these so-called
Dravidian languages, demonstrating philologically that Tamil and
neighboring languages stood together as a group over and against
languages spoken to the north, including Sanskrit. He went on to
speculate about the origins of a Tamil Dravidian culture, suggesting
that a pure Tamil language could be traced back to the time of
Christ, and that Aryan Brahman colonists had only arrived, with
their religion and language, much later.
By the end of the century, Tamils who had been exposed to
British educational institutions and ideas assumed a new, politically
motivated stance to this picture of the past. It gave them grounds
for rejecting the notion that the “Aryan” cultural tradition was older
than, different from, and superior to their own, and fueled a debate
about the facts and meanings of Tamil and Dravidian linguistic
autonomy. This debate engaged with the status of the small, domi-
nant, Brahman fraction of Tamil society, who were distinguished
as the direct inheritors of Sanskritic knowledge (Irschik 1969).
Harold Schiffman (1996) has drawn on work by Sumathi Ramas-
wamy (1993) to sketch this as a philological debate between “Com-
pensatory Classicists,” who argued from philological evidence that
Tamil in the past had been on a par with Sanskrit, and the “Con-
testatory Classicists,” who argued that Tamil’s superiority to San-
skrit had only been obscured by invaders from the north. But both
parties agreed on a central point: an originary, “pure” Tamil lan-
guage could be recovered even after it had been sullied by repre-
sentatives of an outside power. Images of language purity thus
began to figure in longstanding contests over the proper genres of
liturgical speech (Appadurai 1981; Schiffman 1996) in an ethno-
Philology’s Evolutions 91
national separatist movement opposed not to the power of British
colonialists (at first), but other colonized ethnic groups to the
north.
Even sketched so briefly, this final example helps to show that
although William Dwight Whitney’s complaints about “sounding
phraseology” distorting “valuable [philological] truth” were entirely
legitimate, they did not speak to the ideological power of the
metaphors which that phraseology repeated and expanded across
conceptual spheres. With a broad sense of the development of phil-
ological images of evolution, we can consider their meanings and
uses in particular zones of colonial contact, where linguists under-
took the formidable challenge of reducing the speech of colonial
subjects to writing.
Notes to chapter 4
1 Interested readers could recast this brief review of the development of
philology in terms used by Pierre Bourdieu (1993) to account for the
dynamics in a wide range of literary and academic “fields,” but that is
beyond the scope of this chapter.
2 I must ignore here questions of class which, as Greenfeld indicates,
were crucial for Herder and other romantics’ self-positioning in relation
to “the people” (das Volk). Discussion of this important issue can be
found in Bauman and Briggs 2003.
3 Grimm’s “discovery” was in fact anticipated by the Danish philolo-
gist Rasmus Rask, and counted as a “law” in no more than a rhetorical
sense.
4 Adaptations of evolutionary theory to the study of language change
and contact are far from dead; see for instance Mufwene 2001.
5 Darwin himself crossed this line, in the other direction, in The descent
of man (1936b[1871]), to distinguish not just between animals and men,
but “modern” men and “savages” of the “lower races.” Nowadays
sociobiologists have more sophisticated versions of transformational
understandings of history as “evolution.” They assume that some
natural (genetically determined) properties of humans confer on them
relatively broader adaptive abilities across a range of natural environ-
ments, and so have broadly directional shaping effects on cultural
transmission, which turns out to be a subform of natural selection. But
this idea was already at the heart of Herder’s account of language
92 Philology’s Evolutions
origins which I sketched in chapter 3, which began with the riddle
of the human species’ relatively weak senses, but strong adaptive
capacities.
6 For extended, insightful discussion of a wider range of French scholar-
ship of African languages, see Irvine 1993, 1995, and 2001.
7 For discussion of the development of Ellis’ work in scholarly inter-
change with scholarly speakers of these languages, see Trautmann
2006.
Chapter 5
Between Pentecost
and Pidgins
Each village has its own language; take this then to pray to our
Father!
And the Lord will understand our poor and needy word
Dida hymn of the Harrist Church, Ivory Coast (quoted in Krabill
1995:253)
To make him talk pidgin is to fasten him to the effigy of him, to snare
him, to imprison him, the eternal victim of an essence, of an appear-
ance for which he is not responsible.
Frantz Fanon, Black skin, white masks (1967:35)
In this chapter I discuss work of linguists who helped to make the
19th century not just an age of empire but of missions, as “the rather
idiosyncratic concern of a handful of Moravians” in 1780 became
by 1840 “almost the very raison d’être of the all the mainline
churches as understood by their more lively and enthusiastic mem-
bership” (Hastings 1994:245).1From the enormous body of work
about languages spoken all over the missionized world I draw here
on a few examples, mostly from sub-Saharan Africa, to discuss the
purposes, social imaginaries, and practices of literacy which shaped
their content (sometimes inaccurate), uses (sometimes conflicted),
and long-term effects (sometimes unforeseen and unintended).
The work of missionaries has attracted increasing interest among
historians, anthropologists, and literary scholars who recognize that
missionaries were agents of some of the most intimate forms of
94 Between Pentecost and Pidgins
colonial power. Reading critically in the archives has offered insights
into the gentler, more pervasive micropolitics of missionary author-
ity, and the kinds of conflictedness which enabled it. One common
dilemma facing missionaries arose from their double engagement
with colonial subjects as well as other colonial agents, institutions,
and interests. They might believe that they were “men [sic] of God
and ambassadors for Christ [who] have nothing to do with trade in
any way whatever”(quoted in Hastings 1994:284), but they needed
support of fellow citizens of Europe’s colonial powers. They might
have regarded the brutal use of power as means to God’s real end,
which they served “following at a distance, in the rear of victorious
armies, to plant her stations,” as a minister preached to the London
Missionary Society in 1819 (quoted in Thorne 1999:38). But mis-
sionaries were obliged to accept the authority and agendas of pow-
erful colonial institutions, particularly when their work of conversion
involved educational projects which helped create hierarchies of
people and language alike.
Broadly similar callings, goals, and circumstances make it easy
to overlook differences between missionaries’ social backgrounds
which had shaping effects on their work. Some missionaries were
marginal or self-marginalized in their “home” countries; others
embraced the double authority of their nation and their religion.
Some were highly educated; others had only limited literacy in their
own languages. Some were profoundly transformed by years spent
with “their” natives; others were loyal soldiers of Christ among
pagans they neither understood nor trusted. Each had what can be
called, using James Clifford’s word for the work of anthropology,
a preterrain, comprised of “all those places you have to go through
and be in relation with just to get to your village or to that place of
work you will call your field” (Clifford 1992:100).
Faith, status, and nationality are useful categories for considering
such differences between missionaries’ stances and strategies, and
their more or less successful work as linguists (describing lan-
guages), teachers (of literacy), and preachers (to pagans). Reading
their linguistic work with an eye to the times and places they
produced it, as well the languages they described, helps to get a
sense of how they developed the strategies of selection which
allowed simple written images to substitute for complex worlds
of speech. This way of reading also helps to recognize how that
Between Pentecost and Pidgins 95
work was shaped by the institutions and ideologies which made
it possible, legitimate, and useful in larger regimes of colonial
power.
This chapter’s two epigraphs help to emphasize religious belief
as a basic grounding for all of this work, in the 19th century as in
the 16th. The first is an exhortation translated from a hymn sung in
the Dida language in Harrist churches on the Ivory Coast. It calls
on God’s children to worship in their own languages, evoking a
Pentecostal sense that those words are needy not because they are
primitive or uncivilized, but because they are human.2The second
epigraph is Franz Fanon’s protest against the psychopathology
afflicting those who cannot avoid having a “pidgin” version of a
“real” language affixed to their senses of self. Fanon’s protest can
be read as warning more broadly against severe dislocating effects
of linguistic descriptions which might have had an aura of colonial
authority, but failed to present colonial subjects with recognizable,
written images of their own speech. Such a linguistic effigy of speech
could be imposed by colonial regimes as a model for language, and
have far-reaching effects on senses of linguistic sharedness and
community under colonial regimes.
Faith and Literacy, Faith in Literacy
Differences between Protestant and Catholic strategies of conver-
sion broadly reflected their faiths, and shaped their engagements
with pagan languages. Like the friar linguists discussed in chapter
2, Catholic missionaries in the 19th century were concerned mainly
to induct their converts into a global community of ritual and belief.
They understood that their calling was to mediate between the
Latinate body of textual Truth and new converts in “vocalized
exchange between God and the priest, and between the priest and
the hearer” (Peterson 2004:122). An understanding of the gist of
faith sufficed, in their view, for natives to participate in the ortho-
doxies of Catholic religious and social practice. Catholic missionar-
ies who regarded linguistic difference as a barrier to conversion to
be overcome had little reason to invest much time in describing
or learning their converts’ languages. Instead they sought out lan-
guages which were widely spoken, natively or not, across groups,
96 Between Pentecost and Pidgins
and which were relatively prestigious among different communi-
ties (like Nahuatl, the lengua general of Mexico). Some used the
languages of colonial regimes to Christianize and civilize their con-
verts, which I discuss this in chapter 6.
Protestants understood that biblical truth could only come
directly to individuals through personal knowledge of that text, and
so that they were called to provide translations which could produce
internal conversions of mind, imagination, and soul. For them, the
gift of faith went with the gift of literacy, and so the ability to “read
the Word through their individual, cognitive work” (Peterson
2004:122). For native languages to become bridges to faith, then,
Protestant missionaries engaged local speechways as means for
“transform[ing] interior worlds” (Meyer 1999:38) through practices
of literacy. From the earliest, “idiosyncratic Moravians” of the late
18th century3 up to missionary work going on today, touched on in
chapter 7, faith and literacy were constantly joined in the work of
Protestant missionaries. Richard Lepsius, a philologist who lent his
expertise to this global endeavor, explained that only when “the
Word of God is read by the people themselves, and where a whole
people are made susceptible of the spirit of Christianity by the dis-
tribution of the Bible and of Christian school-books can a rapid, a
deep and lasting work be hoped for” (Lepsius 1855:6).
So it is “not surprising,” as William Samarin observes, “that
Protestant missionaries would have had more interest in the native
languages and that Protestants would have been even more com-
mitted to learning them than Catholics” (1984:436). This meant,
though, that Protestant missionaries were committed to a double
paradox of translation and conversion. All missionaries relied on
faith to draw lines between what in pagan lives could be preserved
and what had to be banished. But to acquire the languages they
needed for that work, they had to engage with the lives of those
languages’ speakers in their entirety, what one American mission-
ary more recently referred to in the jargon of social science as “the
totality of a social system.” To refuse full engagement with a com-
munity would produce linguistic knowledge as a broken, pidgin
idiom “suitable for dealings only with individuals who are periph-
eral in the community at large” (Smalley 1958 paraphrased by
Beidelman 1982:17). To avoid learning and using the kind of broken,
Between Pentecost and Pidgins 97
pidgin language Fanon criticizes, missionaries needed sustained
contact with the full range of non-Christian speechways and
lifeways.4
This was seen clearly by one of the colonial era’s most distin-
guished linguists. Hermann Neubronner van der Tuuk was
employed in the 1850s by a Dutch missionary organization (the
Nederlands Zendeling Genootschap, or NZG, which I discuss
below) to translate Christian texts into the Batak language, spoken
by peoples of the interior of the island of Sumatra. Van der Tuuk
was no devout Christian, but had no qualms about the work of
christianizing Bataks who could then serve as buffers between
Dutch settlements in the southern parts of Sumatra, and anticolo-
nial Islamic peoples to the north and west.
In 1867, with the benefit of hindsight, van der Tuuk drew a moral
from his work with the Batak in a letter to a friend written with a
“pen dipped in bile”: “[a]nyone who learns a language for the
purpose of translating the Bible into it is nothing but a villain, and
therefore I have more contempt for myself than for anyone else”
(1962:109). Van der Tuuk acknowledged gaps of meaning between
his translations and the Batak language, due to his partial knowl-
edge of the language’s uses and speakers. This ignorance was
required of him by his employers, who regarded any close engage-
ment with pagans as possibly tainting their missionaries’ faith. “To
learn a language well,” van der Tuuk continued, “one must become
familiar with its community (volk), and this for some nations (natiën)
is not possible except by considering their religion. And it is exactly
this which would count as a deadly sin for a society that lives by
bigotry”5(1962:109). The bitterness of van der Tuuk’s complaint
against his own “nation” (the Netherlands) and employers arose
from his professional frustration with his descriptive “effigy” of the
Batak language. He surely knew also that less talented, observant,
or committed linguists were unlikely to even recognize, let alone
resolve, the dilemma he came to see as a founding condition for all
their work.
The second common dilemma was more pressing for some mis-
sionaries than others: the uses and effects of their linguistic work
always outran their own purposes. This problem was most acute
for those who adhered to the romanticist, organic vision of the
98 Between Pentecost and Pidgins
earliest German missionaries, and distinguished categorically
between the work of christianizing pagans and civilizing primi-
tives. They sought to provide the gift of faith without creating an
avenue of entry for secular institutions or ideas. This vision of lan-
guage, locale, and identity was encapsulated in the ideal of a national
Church (Volkskirche) which was to be “planted in the soil of heathen
nations [so] that [they] become . . . naturalized there as a domestic
growth” (Meyer 1999:34). One influential Swiss missiologist
appealed to this ideal of a truly native Christianity in the 1870s with
a warning to missionaries against any sense of “cultural superiority
and . . . national egoism” (Warneck 1901:104). Their lack of flexibil-
ity would lead to a “lack of pedagogic skill in dealing with those
who are the objects of missions,” and so to the “birth of a feeling of
inferiority” among them.
This split vision, which sought to avoid “confounding . . . Chris-
tianisation with Europeanisation or Americanisation,” reflects a
tension which ran throughout the work of missionaries under colo-
nial regimes, including the linguistic descriptions they devised and
practices of literacy they taught. Beyond the ideological difficulties
involved in maintaining this distinction, missionaries faced practi-
cal difficulties wherever they did not exercise exclusive oversight
over every aspect of their converts’ lives, but instead were located
with those converts in broader networks of colonial power and
hierarchies. Then their linguistic descriptive work served not just
to teach literacy, but to incorporate those they converted as subjects
of colonial regimes.
A useful, somewhat extreme, example of this merging of the
work of faith/literacy and projects of power can be drawn from
New Zealand’s early colonial history, where missionaries provided
a pretext and means for incorporating their new flocks into an
emerging British empire. Missionaries established a school there in
1815 to begin teaching an improvised system of Maori writing
which was later improved on by one of the missionaries who
returned to England with two Maori chiefs in 1820. They consulted
with a Cambridge professor of Arabic to devise a grammar and
vocabulary of “the language of New Zealand” which they took back
to the islands as the template for locally printed materials. By 1830,
English observers happily reported, written materials had spread
widely enough that Maoris were acquiring literacy “with a degree
Between Pentecost and Pidgins 99
of decorum and regularity which would have reflected credit on
a school of the same kind in England” (quoted in McKenzie
1985:14).
What now seems a delusional analogy between English and
Maori societies counts as evidence of Protestant doxa’s shaping
power on those who perceived the Maori as untainted innocents,
and candidates for salvation ready to “naturally” take over faith
and practices of literacy together. Another observer predicted three
years later that “[t]he day is not far distant, when the people gener-
ally will be able to read for themselves, in their own tongue, the
wonderful works of God” (quoted in McKenzie 1985:18). This shows
how easy it was for outsiders to find what they “wanted to find,
[and] report what [they] knew their London [missionary] commit-
tee wished to hear” (McKenzie 1985:16). (This “committee” was the
London Missionary Society, discussed below.)
This same interested, partial vision made possible the now infa-
mous ritual which ushered the Maori into the colonial era in 1840:
the signing of the Treaty of Waitangi. The soon-to-be governor of
New Zealand assembled Maori chiefs at Waitangi to sign a docu-
ment which officially and legally ceded their authority to the British.
First proclaimed aloud in English, and then in a Maori translation,
it was dubious in the first place because it incorporated a large
number of words (actually, just the sounds of those words) from
English into Maori. When these unintelligible sounds had been
uttered in the presence of ears which could not grasp their meaning,
or the import of the act of reading itself, the gathered chiefs made
their marks at the bottom of the paper, contributing to a ritual of
literacy whose context and effects they could not understand.
As long as the British could regard these chiefs as “literate,” in
some sense of the word, their acts could be regarded as binding,
even if the cultural gap glossed over in this way was glaring enough
to lead one witness – the printer of Maori literacy materials – to ask
the new governor if the chiefs understood what they were signing.
The governor responded that they should trust in the advice of
“their” missionaries, and expressed the hope that no “reaction”
would follow should they come to apprehend later what they did
not understand at the moment. This bland observation reflected a
thin sense of legitimacy, based on an implausable refusal of linguis-
tic difference, not unlike that which legitimized readings of the
100 Between Pentecost and Pidgins
Requerimiento by the Conquistadores in the New World (discussed
in chapter 2). There is no record, though, of any witness at Waitangi
who was left not knowing whether to laugh or cry, like Bartolomeo
de las Casas in Mexico.
The treaty of Waitangi provides a clear example of the ways
literacy allowed regimes of power to be bound up with the work
of faith. But missionaries elsewhere found themselves to be accom-
plices of other kinds of power, as happened to linguists working on
another island at about the same time: Madagascar.
The Scholarizing Project
The vagaries of politics in Europe, along with Madagascar’s loca-
tion off the east coast of Africa, led to a situation in which mission-
aries worked more than 60 years before Madagascar was formally
colonized by France in 1895, very late in the imperial game. In his
insightful description of linguistic work by missionaries there, Louis
Raison-Jourde traces far-reaching effects of the practices of literacy
missionaries used to describe the Imerina language, and so chris-
tianize and “scholarize” its speakers.
Madagascar had for centuries been a meeting point for traders
between Asia and Africa, and was divided into small kingdoms
stratified into classes of nobles, commoners, and slaves. By 1642, the
French had managed to establish only small footholds on the coast,
and not until a century and a half later were the British finally able
to gain an alliance with the king of the inland realm, Radama I
(ruled 1810–28), through the time-honored technique of entering a
local struggle for power, and tipping the balance in favor of their
future ally.
The British helped Radama expand his domain to the entire
island, and induced him to end the lucrative slave trade, at least
officially. The king also allowed missionaries from the London Mis-
sionary Society to join their French counterparts in opening schools
in different parts of Madagascar. These missionaries, working under
Radama’s close supervision, were of different faiths (Protestant and
Catholic), nationalities (British and French), and ethnicities (Welsh
and English), all factors which exacerbated their “natural” competi-
tion for spheres of influence.
Between Pentecost and Pidgins 101
So arose one instance of a common problem for colonial linguists:
missionaries of different backgrounds working in different locales
produced different literacies for similar speech and, by teaching
those literacies, produced different senses of linguistic identity. On
Madagascar, different descriptions of Imerina, also known as
Malgache, created such a situation which Radama took into his
own hands. In 1823 he convened a meeting of pupils whom mis-
sionaries had taught literacy and Christianity in their different
schools. As one pupil told it,
[t]he king told us: “Write the name Rakoto,” so we twelve wrote in
different ways: some wrote Roccotoo or Racwootoo, or Raquootwoo,
Then Radama said: “unite your two groups, unify your ways of
writing so that Malgache [i.e., Imerina] writing conforms to our lan-
guage. If your ways of writing are not identical, it is as if my realm
isn’t just mine but that of many masters.” (Quoted in Raison-Jourde
1977:644, my translation)
Though illiterate, Radama was not insensitive to the political
and symbolic import of a plurality of written images of Imerina:
a language divided in this way would be in the hands of the “many
masters” of literacies which they were teaching away from his
own center of power. So he moved unilaterally to create a propri-
etary relation with the language by requiring that a single literacy
be propagated uniformly across his territory and among his
subjects.
Radama’s successor, Ranavalona, helped British missionaries
carry out a dictionary project not unlike that being undertaken at
the same time in Germany by the Grimm brothers, discussed in
chapter 4. The missionaries were dealing with two languages, not
one, but also had an advantage over the Grimms: hundreds of liter-
ate native speakers worked as assistants under them on the king’s
orders, fostering a “scientific project, in the best tradition of the
English academies or German university of the 18th and 19th
centuries . . . [but] taken up by the Merina state with a national
significance” (Raison-Jourde 1977:645, my translation).
The writing of this dictionary involved practices of literacy which
required a particular kind of “meaning” to create word-to-word
relations between the elements of English and Imerina. Isolating
102 Between Pentecost and Pidgins
and juxtaposing elements of alien and familiar languages in written
lists required that there exist a kind of lowest common denominator
of meaning. Only by assuming that a unitary “semantic field”
underlay both languages could missionaries bridge enormous gaps
between social imaginaries.
Like Radama before him, King Ranavalona had a political inter-
est in this project: it promoted unity among members of the new
class of literate speakers being produced by missionary schools. The
dictionary contributed to the same authoritative, literate image of
“the” Imerina language his predecessor had sought to create as
a representation of centralized political power (Raison-Jourde
1977:645). Literate subjects could serve royal power but also threaten
it, a possibility which fueled the king’s suspicions of the missionar-
ies’ ulterior motives. He forced the missionaries to stop that work
in 1835, even though they were allowed to continue their dictionary
project.
In 1895 the French assumed control over a vastly changed Merina
society dominated by scholarized, christianized elites. As in New
Zealand, but more slowly, missionary linguists’ work of education
and faith helped to make their converts subjects of an “outside”
colonial power. But in both places, as all over the colonial world of
the 19th century, this work was done not just by individuals, but
by institutions in the “home countries” of Europe which supported
them. Most obviously, missionaries relied on the infrastructures
which produced texts in the languages they described and trans-
lated for new Christians to read. Secondly, missionaries unversed
in the difficulties of linguistic analysis needed broadly intellectual
support for their work of reducing speech to writing. Some com-
parative philologists in Europe brought their science to bear on this
challenge for the work of faith, trying to help missionaries avoid
orthographic confusions of tongues, and enlisting them in a metro-
politan project of global civilization and knowledge.
From Letters to Orthography
Though the Africanist Carl Meinhof was one of the first professional
linguists in Europe to study the actualities of exotic speech, text-
Between Pentecost and Pidgins 103
centered philologists were always interested in the work of reduc-
ing speech to writing. At the same time “Oriental” Jones was
studying Sanskrit texts, for instance, he was also reading work by
his fellow colonialist William Halhed on the “Hindoo” language of
Bengal and Upper India. Halhed, like Jones and Gilchrist, inferred
from Sanskrit elements he found in Hindu speech that it was a
decadent version of an older, literate language. He understood also
that his job was to “cultivate” the language, as Bernard Cohn puts
it, by recovering authentic forms of the original language from
“fallen, broken, or corrupt” speech. In the same volume of Asiatic
Researches which presented his famous address about Sanskrit’s
deep linguistic past, Jones discussed Halhed’s grammar in an article
entitled “On the orthography of Asiatic words in Roman letters”
(Jones 1788).
Jones identified two basic challenges faced by Europeans and
commended Halhed for overcoming the first: defectively applying
“the same letter to several different sounds, and . . . different letters
to the same sound” (Jones 1788:7).6Jones acknowledged the imper-
fection of European alphabets in this respect, and their inferiority
to those of scholars who created invariant, one-to-one correspon-
dences. The second challenge, beyond the power of any individual
to solve, arose from inconsistencies between different conventions
established by different authors. Jones bemoaned liberties taken by
writers who each devised “a method of notation peculiar to himself,
but none has yet appeared in the form of a complete system, so that
each original sound may be rendered invariably by one appropriate
symbol, conformably to the natural order of articulation, and with
a due regard to the primitive power of the Roman alphabet” (Jones
1788:1–2).
Taming the “primitive power of the Roman alphabet” was more
than a matter of aligning its symbols with sounds of speech pro-
duced phonetically or, as Jones put it, a “natural order of articula-
tion.” The physiology of sound production had become an object
of scientific study in Europe but, as Jones demonstrated in this
article, the creating of a uniform scientific orthography required
institutional coordination between all linguists studying all
languages.
As the case of Imerina shows, no such uniformity had been
achieved 50 years later even on one relatively small island. But over
104 Between Pentecost and Pidgins
this same period of time, eminent figures in the science of com-
parative philology were taking up the challenge Jones had identi-
fied. Richard Lepsius, whom I quoted above on the needs of
missionary linguists, was a faithful Lutheran (like Carl Meinhof)
whose fame came from discoveries in Egypt and Nubia during an
1842 expedition sponsored by Fredrick William the Fourth of
Prussia. Surveying the burgeoning body of linguistic descriptive
work by missionaries around the colonial world, he recognized, as
had Jones, that
a diversity of signs for one and [the] same sound in different lan-
guages . . . [is a problem which] has become so great that the transla-
tor of Oriental works, the Tourist, the Geographer and Chartographer,
the Naturalist, the Ethnographer, the Historian . . . and above all
others the Linguist, who studies and compares languages, find them-
selves entangled in an intolerable confusion of orthographic systems
and signs, from which each individual finds it impossible to extricate
himself. (Lepsius 1855:3)
Lepsius took on the task of overcoming this Babel-like “confu-
sion” by publishing what he envisioned as a standard alphabet for
reducing unwritten languages and foreign graphic systems to a uniform
orthography in European letters.
Phrases like “graphic system” and “orthography” reflect Lepsius’
concern to identify empirical “imperfections” in received European
practices of literacy, none of which “could claim in its present state
to be used as a standard system.” Lepsius promoted his own orthog-
raphy as being grounded not just in modern science, but ancient
wisdom which that science had uncovered. His discussion of what
Jones called the “natural order of articulation” drew on knowledge
in Sanskritic texts which his colleagues had discovered, “physiolog-
ical and linguistic views more accurate than those of any other
people:”
[t]hese grammarians [of ancient Sanskrit] penetrated so deeply into
the relations of sounds in their own language . . . that no language
and no alphabet are better suited to serve . . . as a starting point for
the construction of a universal linguistic alphabet than that of ancient
India. (Lepsius 1855:15)
Between Pentecost and Pidgins 105
In this way Lepsius brought comparative philology full circle:
advances in his own modern science were validated in the ancient
texts which were now that science’s object of interest and source of
guidance.
Lepsius also recognized that for his orthography to be not just
valid but widely used, it was a matter of practical importance
that it be accepted by those who would oversee its use. This was a
transnational collection of missionary groups – German, French,
American, English, and Swiss – whose endorsements he sought and
gained, and whose names are listed on the first pages of his orthog-
raphy’s first edition. Thanks to these endorsements, too, Lepsius’
alphabet became better known than another devised by Max Müller
about the same time, even if Müller had a higher popular profile
for reasons noted in chapter 4.
Müller’s “missionary alphabet” was also designed to overcome
the “defects peculiar to each” of Europe’s national literacies, recog-
nizing that “it would be wrong to smuggle any one of these im-
perfect systems into those languages . . . which have not yet
been reduced to alphabetical writing” (Müller 1854a:xviii). Müller
seems to have devoted less time to presenting his alphabet to mis-
sionary organizations than Lepsius, perhaps because he was also
trying to attract the attention of Britain’s diplomatic and military
elite. This is evident in another work presenting this orthography,
published the same year under the portentous title Suggestions for
the assistance of officers in learning the languages of the seat of war in the
East (i.e., the Crimea). In this work’s introduction he noted the
obstacle posed to Britain’s imperial power by the fact that “[a] man-
of-war is built in less time than an Oriental scholar can be launched
ready to converse with natives, and capable of producing supplies,
gathering information, translating proclamations, writing circu-
lars . . . and, finally, of wording the conditions of a treaty of peace”
(Müller 1854b:ix).
Language, Territory, and Identity
The idea of a single, coordinated system for writing all the
world’s languages can be taken as a kind of secular counterpart to
the vision of global Christianity that missionaries shared with each
106 Between Pentecost and Pidgins
other, and offered their converts. Viewed from a distance, as toilers
in the vineyards of the Lord, differences between them can be dis-
counted, at least when considering their work as linguists. But the
work of missionaries in Madagascar demonstrates that in fact they
cooperated and competed with each other according to nationality,
social status, and faith. Whether or not they served the same God,
they were members of different denominations, citizens of different
nations, occupants of different social niches, products of different
educational systems. These factors all shaped their social imaginar-
ies, their practices of literacy, and so their work of spiritual and
linguistic conversion.
Virtually all missionaries abroad required material support from
organizations “back home.” The London Missionary Society (LMS),
for instance, had dispatched missionaries to New Zealand and
Madagascar, among many other places. It was founded in 1795 with
financial support largely from the nonaristocratic middle class
emerging in England’s industrializing, urbanizing centers. Although
it became the most liberal missionary group in Britain over the first
part of the 19th century, it was firmly in the hands, Susan Thorne
(1999) notes, of an educated middle class whose assumed superior-
ity was duly recognized by the largely lower status missionaries
they sent abroad. Similarly, the Dutch Missionary Society (NZG),
which supported van der Tuuk’s study of Batak in Sumatra, was
“securely in the hands of a social and cultural elite which was
supported by an emerging bourgeoisie” (van Rooden 1996:81). This
was an organization van der Tuuk himself acidly described as
depending on speculation on the pockets of “pious cheese-
buyers . . . ” “a pack of saints who did not care a straw for study”
(van der Tuuk 1962:109).
Such organizations were founded and run by socially and reli-
giously conservative men, even if economic and social resources
came to them thanks to their new places in rapidly industrializing
societies. Brigit Meyer (1999) points out that founders of Bremen’s
Norddeutsche Missiongesellschaft, for instance, earned profits in a
new economy, but were profoundly distrustful of empirical-
rational thought. They, like the heads of other such organizations,
chose as their missionaries people less fortunate if no less devout
than themselves: pious, rudimentarily educated, not far removed
from the farm, local market, or craftshop.
Between Pentecost and Pidgins 107
The same groups which were sending missionaries abroad were
doing similar work at home among those who were bearing the
brunt of Europe’s urbanizing, industrializing dynamics. Evangeliz-
ing and educating efforts were putting literacy and faith on offer to
the new urban poor, creating another project to integrate (and sub-
ordinate) those who were otherwise marginal in the global advance
of Christian civilization. Just as evangelists were working to spread
literacy in countries like England, so too the expanding state-
sponsored education system of France was engineering the trans-
formation of peasants into literate citizens. In Europe as abroad,
literacy was being propagated as the touchstone for new identities,
standing “at the entrance of the modern world as dragons guard
the gateway of a temple” (Weber 1976:452).
Whether or not they knew and used Lepsius’ orthography, many
missionaries embarked with neither broad education nor specific
training in the work of linguistic description. They were obliged to
do the work of writing and translating exotic languages by relying
heavily on their own “common sense” ideas about languages gener-
ally, and practices of literacy in particular. I can sketch these by
considering the ways they dealt with problems like those encoun-
tered by the friar linguists discussed in chapter 2: devising the
strategies of selection to reduce complex differences in speech –
between dialects of different groups, and styles used in different
contexts – to simple alphabetic images.
National Languages, Communities, and Territories
Missionaries’ practices of literacy and social imaginaries can be
thought of as grounded in their broadest, common condition as
literate citizens of 19th century nation-states. This makes it helpful
to review briefly Benedict Anderson’s (1991) influential, language-
centered historical account of the emergence of nationally imagined
communities. Nationally imagined communities became possible in
Europe, Anderson argues, after print technology made it economi-
cally feasible to produce large numbers of texts for large numbers
of potential readers. As early capitalists, printers sold texts, like
other commodities, by maximizing their markets. They needed
print-literate images of what people could recognize as “their own”
108 Between Pentecost and Pidgins
speech, despite the ways it differed from place to place, or class to
class. In this way print literacy became infrastructural for a sense
of linguistic sharedness, both an aspect of national identities and an
instrument of state institutions.
Anderson’s language-centered account of nationalism draws on
political and cultural changes I discussed in chapters 3 and 4 as part
of comparative philology’s rapid development over the 19th
century. Philological images of deep human history made it pos-
sible for unitary national languages to be understood as both trans-
historical and everyday realities, binding the present to the past,
and the personal to the collective. Thanks to Herder and those who
followed him, languages counted as evidence of how individuals
were bound to society, as were their biographies to history; lan-
guage became immediate evidence for the claims of groups on
individuals: “my” language is already there for me before I am
born, and for my descendants after I die.
Most important here are the European “cultures of standardiza-
tion,” to use Michael Silverstein’s (1996) phrase for largely unspo-
ken but powerful understandings of national languages as unified,
coherent, and “pure.” These cultures, grounded in literate images,
authoritative institutions, and normative ideals, allow features of
nonstandard speech to be regarded not just as differences of usage
but marks of personal deficiency, often slotted into categories of
race, class, gender, region, and others. Thanks to cultures of stan-
dardization, “accent” and “dialect” count as names for attributes of
speech and speakers who are relatively marginal or inferior in
broader social hierarchies.
Cultures of standardization thus substitute the “standard” part
for diverse varieties, creating relations of “internal translation”
between them. In this way a seemingly unitary national language
can be thought of, as Etienne Balibar (1991:351) points out, as being
not just a natural attribute of its speakers, but the territory they
occupy. Like Anderson and others, Balibar links the integrity of
modern nation-states to a shared sense (if not the reality) of social
and linguistic homogeneity within a demarcated expanse of terri-
tory. In this way cultures of standardization, and the images of a
“pure” language they support, enable a broader sense of the shared-
ness of ways of speaking and places of speakers.
Between Pentecost and Pidgins 109
Missionaries, as literate citizens of modern nation-states, partook
of such social imaginaries and cultures of standardization, which
made literacy much more than a resource for devising scientific
orthographies. It was easy for them to understand that languages
counted as natural bonds between speakers and territories, and to
read complexities of speech as secondary to the underlying reality
of a unitary language. If languages’ forms and attributes were
disguised by blooming, buzzing confusions of speech, they could
nonetheless be re-created in written and print-literate images. Cul-
tures of standardization could shape the work of description by
allowing language-centered ideas about society and identity to be
projected onto potential converts; in this way practices of print lit-
eracy could transform communities to fit European understandings
of how languages were located on the territories given to missionar-
ies as their “fields of operation.”
Here I sketch just two of the many such projects of missionaries
around the colonial world who partly described and partly created
languages and communities of converts. They were carried out in
neighboring parts of subequatorial Africa, beginning in the late
19th century, and each is useful here as a kind of political and
linguistic mirror-image of the other. One was done in southern
Rhodesia (now Zimbabwe) where missionaries of different faiths
and nationalities transformed an area of relative linguistic homo-
geneity into a region where speakers of distinct languages had
distinct colonial identities. In the Transvaal, to the southeast, mis-
sionaries who encountered an area of massive linguistic diversity
worked, with faith and philological images of the past, to impose
an image of linguistic unity, and with it new religious and cultural
identities.
Many from one: dialect and language in Shonaland
Zimbabwe is linguistically homogeneous in comparison with many
regions and nations of Africa. Of 13 languages spoken there, Shona
has by far the most native speakers (10.6 million in a population of
about 12.6 million). Before 1890, speech variation in this region
formed what linguists call a “dialect chain.” As one traveled from
one locale to the next, one would encounter shifts or variation in
Figure 5.1 Map of the “languages” of Shonaland, southern Rhodesia (reproduced from Doke 1931).
Between Pentecost and Pidgins 111
the ways people spoke – pronunciation, grammar, word choice –
which would not necessarily prevent members of neighboring com-
munities from understanding each other. As one traveled, though,
more such differences would be encountered, until eventually
mutual intelligibility of speech would be lost. Before the coming of
British colonialists to this region, these kinds of language difference
had little importance as markers of social difference in comparison
with attributes of kinship (affiliation with the Karanga or Manyika
lineages of chiefly authority) and region (Korekore to the north, or
the Zezuru highlands).
Emissaries of five different missionary groups came to this region
in 1890: Anglicans from England, Catholics of two orders, Dutch
Reformed Church missionaries, and American Methodists. Each
was given its own “field of operation,” demarcated by the colonial
government which oversaw their work, and in each field were
established churches and schools where the “local” language was
taught with its own system of writing. Methodists, Anglicans, and
Marianhill fathers, working in the east, devised a common writing
system for what they called “chiManyika,” but which “created
rather than merely reflected” (Ranger 1989:127) the language of the
“Manyika nation.” Parallel projects in the other fields of operation
produced other groups of Christian, colonial subjects who were
literate in similarly compartmentalized languages, as mapped in
Figure 5.1.
Two generations later the result of this work was described by a
professional linguist, C. M. Doke, as follows:
let us suppose England to be a heathen country. Four distinct Mis-
sionary Societies commence work, one among the Cockneys, one
among the University class, one in Yorkshire, and one in Devonshire.
Each produces a translation into the “local” vernacular, each further
uses a different orthography and some split up their words into their
component parts. What an enormous difference there would be
between the four literary efforts; they would not be mutually under-
stood. (Doke 1931:3–4)
Doke attributed this unfortunate situation to the amateurish work
of untrained missionary linguists. It fell to him, a professional, to
oversee the elimination of idiosyncratic uses of letters (in some
112 Between Pentecost and Pidgins
cases as many as four) for a single speech sound. With a mandate
from the government of Rhodesia, and financial support from the
Carnegie Foundation, Doke resolved this Babel-like confusion in a
meeting of representatives of the missionary groups to devise what
was known as the Union Orthography.
In effect, the commission compromised on a writing system
which corresponded to the speech of no single group or region.
Herbert Chimhundu (1992), a historian of colonial Rhodesia, points
out that the commission favored features of Zezuru, and second-
arily Karanga and Manyika, as the “peak” dialects most strongly
associated with Christianity, literacy, and colonial power. Ndau
was less used, Korekore and Kalanga not at all.
Zezuru became the basic model for the Shona language in part
because it was spoken natively in the environs of Salisbury (now
Harare), the colonial (now national) center.7“Shona” was, to be
sure, not a name known to its speakers, but “it [was] essential,”
Doke observed, “to use a definite name as a label for the whole
cluster of groups. The fact that the people themselves do not
acknowledge this name is really immaterial” (Doke 1931:3).
The Kalanga “language,” for its part, was reclassified as a
“dialect” of Ndebele, which was in fact an entirely separate lan-
guage from a structural point of view, and not mutually intelligible
with any “dialect” of Shona. Perhaps because the Kalanga region
had no representatives at the committee meeting, though, its lan-
guage and people were reallocated to the London Missionary
Society’s neighboring jurisdiction in Ndebele land.
When Doke came on the scene, then, no sense of Shona identity
or language existed among speakers of these “dialects” (Ranger
1989:142), some of whom joined their missionaries to actively resist
the reforms he instituted. They were defending not just habits of
reading and writing, but internalized senses of sharedness of lit-
eracy and speech, faith and community. Beyond this, they were also
resisting larger political and economic forces which were eroding
the privileged social positions of scholarized elites by integrating
the colony’s different regions. Top-down linguistic “improvements”
served the colonial state at a time when newly mobile migrant
laborers left villages and agrarian economies behind for urban
centers. They needed a language of wider contact in a modernizing
colonial society, and it fell to the Doke Commission to “assemble
Between Pentecost and Pidgins 113
dialects” of Shona by the work of “internal translation” into a
common standard.8
Before long, though, the spread of Shona had its own naturaliz-
ing impact on speakers’ senses of “ethnic” identity. “If in the 1930s
no one in Makoni would have described themselves as ‘Shona’, by
the late 1950s, when the nationalist movement came to the district,
very many people thought in such terms” (Ranger 1989:143). Shona
had by then become the symbol and instrument of an ethnic idea
in modernizing Rhodesia. So, too, the new linguistic reality imposed
by colonial policy was giving rise to a sense of distinctness among
the Kalanga whose speech, once placed in the same territorial group
as the Ndebele, had begun to change in ways which have now
created a situation where its speakers cannot read the Shona Bible
(Grimes 1984:302).
It is important to see that the establishment of “Unified Shona”
did not return this region to some precolonial state of linguistic
homogeneity which misguided missionaries had temporarily dis-
rupted. Instead, it installed a new homogeneity conforming to the
political, economic, and territorial logic of the colonial regime which
oversaw those missionaries’ fields of operations. When the logic of
political control and economic advantage dictated the erosion of
demarcations which the state had previously supported, language
differences began to be effaced among members of what came to
count as the larger community of “Shona.” This new ethnolinguis-
tic sharedness made language a central item in what Vail calls,
referring to 19th century Africa more widely, the colonial “cultural
package” (Vail 1989:11).
One from many: inventing origins in the Transvaal
Missionaries who sought to convert natives without destroying
native cultures needed to develop place-linked images of linguistic
purity, which were important intellectual resources in situations of
great linguistic diversity. In one such region, the northern Trans-
vaal, well-educated Swiss missionaries deployed philological
images of the past like those sketched in chapter 4 to grapple with
enormous complexity. South and east of Shonaland, on the borders
of modern South Africa, Zimbabwe, and Mozambique, they encoun-
tered speakers of entirely different languages whom they were
114 Between Pentecost and Pidgins
obliged, by the call of faith and the politics of missionary work, to
convert to a unitary written language. So they partly described and
partly created a print-literate language, Thonga, along with a spec-
ulative history of its speakers (Figure 5.2).
Paul Berthoud and Ernest Creux came to this area in 1872,
sponsored by the Paris Missionary Society (PMS), to proselytize
O
l
i
f
a
n
t
s
R
i
v
e
r
SOUTHERN RHODESIA
(ZIMBABWE)
Levubu River
MOZAMBIQUE
Limpopo River
Louis
Trichardt
LOW
N
VELD
Pietersburg
NKUNA
MODJADJI
SEKUKUNI
SWAZI
SWAZILAND
Lourenco
Maputo
Marques
Indian
Ocean
Lydenburg
Z
o
u
t
p
a
n
s
b
e
r
g
M
t
s
.
N
k
o
m
a
t
i
R
i
v
e
r
Chiefdoms
Major Swiss
mission stations
NKUNA
0 100 km
Towns
Lydenburg (old)
Pietersburg (new)
T
e
m
b
e
R
i
v
e
r
Figure 5.2 The Swiss mission to the Transvaal. Redrawn from Harries
1988. Reprinted by permission of Oxford University Press.
Between Pentecost and Pidgins 115
using the Sotho language they had learned during their earlier
work among neighboring peoples. When they discovered that
none of their prospective flock knew Sotho, they were – as good
Protestants – reluctant to impose it as what would be an elitist
mission language. Instead they undertook to learn “the local lan-
guage,” not knowing how many such languages were spoken in
the region they took as their own. Historical records suggest
that the linguistic diversity they encountered had developed
recently when people moved inland from the coast of southeast
Africa to escape the depredations of the Portugese. Living in
scattered villages, sharing no allegiance to a common chief, and
speaking no common tongue, they had little sense of community
with each other or peoples already in the region.
The Swiss missionaries recognized early on the facts of linguistic
diversity, but had strong economic and political reasons for using
a single language in their work. Print materials were expensive,
prohibitively so if they had to be produced in more than one lan-
guage; linguistic diversity could weaken claims that theirs was a
unitary field of operation, especially in the eyes of German mis-
sionaries who had already established jurisdiction over neighbor-
ing chiefdoms. The Swiss needed to operate as if there were
comparable unity, linguistic and otherwise, among people in the
territory they sought to claim.
To recover the unity they knew must be there, the missionaries
focused on terms used by long-term residents for newcomers,
whom they called Gwamba and Thonga. Though both words identi-
fied them as “outsiders,” that is “not natives,” each counted as
evidence for Berthoud and Creux of underlying commonalities
among them. The challenge, then, was to recover the linguistic
grounds of that commonality and in so doing explain the historical
forces which had eroded it so heavily. It fell to Paul Berthoud to
partly describe and partly invent written images of the language he
sometimes called Gwamba, and sometimes Thonga. This work was
taken over by his brother Henri in 1882, who “systematized” the
language in eight territorial divisions, although these “dialects”
were, unfortunately, not mutually intelligible.
To reconstruct their underlying unity, Henri developed a philo-
logical image of linguistic descent like those discussed in chapters
3 and 4, together with a conjectural, Darwinistic history of ancient
116 Between Pentecost and Pidgins
unity and subsequent dissolution. This theory, already widely rec-
ognized in educated European circles, led him to regard the Thonga
people as inhabiting an early stage of social evolution like that of
prefeudal Europe. He deduced “that foreign language forms,
brought to the coastal areas by invaders in the fifteenth and six-
teenth centuries, had mixed to varying degrees with the indigenous
Thonga language” (Harries 1988:39).
With hindsight, the Berthoud brothers can be seen to have worked
against the grain of historical and linguistic reality to create written
Thonga and its history. But they succeeded in making them social
facts among their literate converts. Once they had some intellectual
purchase on local conditions of linguistic cultural difference, they
could teach Thonga literacy and so also the Thonga language itself.
By this process, over time, print literacy produced a language which
a later missionary called “one of the most trustworthy and complete
manifestations of [the Thonga nation’s] mind,” and “the oldest
element in the life of the tribe . . . the great bond which bound
the Thonga clans together in past centuries” (quoted in Harries
1988:39).
Philological images of the past also gave missionaries a propri-
etary relation to Thonga. It was theirs, as Edward Said might argue,
because they created (or “systematized”) it, but also because they
monopolized its literate forms and uses. As Henri Berthoud put it,
“their” print-based Thonga stood over and against the messy diver-
sity of talk as a language does to its dialects, which “would be
forced ipso facto into the position of patois destined to disappear
with time” (quoted in Harries 1995:166).
A more complete rendition of the history of Thonga/Ronga
would trace the results of regional conflicts among missionaries,
centered on different written versions of the language, following
the Berthoud brothers’ pioneering work. Here it is enough to note
that these conflicts centered on a common issue: which written
images would represent language and identity in which regions.
These developments gave rise to a community described by an
American anthropologist in 1971 as “the Ronga tribe,” which had
“a delimited territory, a common language, common political struc-
ture, cultural unity; and an awareness of themselves as a distinct
group” (Binford quoted in Harries 1995:171).
Between Pentecost and Pidgins 117
Missionaries’ Positions
I have provided these simple sketches to show how missionaries
were guided by European cultures of standardization as they
devised linguistic descriptions with an eye to transmitting print-
literate practices. In this way their linguistic work presupposed and
helped to create unitary “languages” like Shona or Thonga/Ronga,
but also broader part-for-whole relations reaching beyond writing
and speech to language, identity, and territory. Philological images
of the past were important, then, for more than reducing diversities
of speech to writing: they helped missionaries conform to the exi-
gencies of colonial power with “just so” stories which marginalized
and eroded conditions of diversity in a colonial present.
Thus far I have largely passed over the different effects that mis-
sionaries’ backgrounds and lives, or preterrains, could have on their
work as linguists. These could be important, and real, as can be
shown with one last example: a linguistic controversy among Cath-
olic missionaries about the Congo. This can be read as being due
partly to differences in their social biographies, senses of linguistic
identity, and attitudes to colonial power. One is the Jesuit linguist
Gaston Van Bulck whose 1948 linguistic map I presented in Figure
4.1 as a visual image for his Darwinian history of invasions and
displacements of the weak by the strong. The other is Gustaaf
Hulstaert, a linguist whose very different map of the same region
(Figure 5.3) tells a very different story.
Hulstaert’s map is strikingly simple and homogeneous in com-
parison with the one drawn up by Van Bulck, who distinguished
seven distinct Mongo languages where Hulstaert maps one. Where
Van Bulck portrayed scattered linguistic “enclaves,” Hulstaert drew
a picture of unfragmented linguistic sameness.
This simplicity was no reflection of ignorance. Van Bulck did his
work in European libraries, but Hulstaert studied Mongo languages
in the Congo from 1925, when his work for the Missionaries of the
Sacred Heart eventually made him a “towering figure among
missionary linguists” (Fabian 1986:81). These two pictures reflect
disagreements less about linguistic facts than their meanings and
relevance for Belgium’s colonial agenda. Van Bulck’s map is
Figure 5.3 Hulstaert’s linguistic map of the Belgian Congo (reproduced from Hulstaert 1950).
Between Pentecost and Pidgins 119
grounded not just in a history of survival of the fittest, but the
legitimacy of work by another superior power (Belgium) to impose
civilizational authority on a disunited region. Because (as Schleicher
showed) it is entirely natural for weak languages to die, this was
an entirely natural result of Belgium’s colonial project to educate
(or scholarize) its subjects in more civilized but non-European
languages of wider communication: Lingala and Swahili, which I
discuss in the next chapter.
Hulstaert’s map, on the other hand, reflects what Michael
Meeuwis (1999) calls its author’s “ideology of the natural.”
Hulstaert knew quite well that diverse dialects were spoken in this
region, but saw in that diversity no evidence of the relative “weak-
ness” and “strength” of languages. Minor differences, in his opinion,
were viewed properly against the larger backdrop of natural, lin-
guistic sharedness among the region’s peoples. Though he acknowl-
edged that a merging of some dialect clusters was appropriate,
Hulstaert insisted that these languages represented the core of their
speakers’ identities, echoing Herder’s philosophy of linguistic
autonomy and authenticity. Hulstaert’s map in this way counted as
an argument against any colonial language of wider communica-
tion, which could only be an “anti-popular influence,” “artificial
and European,” and a “rudimentary passe-partout language” which
would “caus[e the native] civilization to regress in many degrees
. . . [after] an incalculable number of years of cultural evolution”
(Hulstaert and DeBoeck, quoted in Meeuwis 1999:405–406).
The roots of Hulstaert’s stance can be found not in his Catholic
religion or Belgian nationality, but other parts of his preterrain. He
was Flemish, from a region which was both linguistically distinct
and politically and economically marginal in relation to Belgium’s
dominant, urban, Francophone population. He shared with many
other Flemish speakers a strong sense of ethnic nationalism and of
his native language as a crucial resource for resisting oppression.
Hulstaert’s life experience in this way resonated strongly with the
conditions which animated romanticist philosophy and compara-
tive philology, discussed in chapters 3 and 4, and with his brief
against colonial encroachment on Congolese languages. Hulstaert’s
map can be read, then, as more than a rebuttal of claims by a fellow
linguist. It reflects his cross-cultural and cross-linguistic identifica-
tion with speakers of these “minority” languages, as he projected a
120 Between Pentecost and Pidgins
sense of his own position among Europeans onto conditions on the
other side of the colonial divide. (For discussion of other instances
of such identification, see Steinmetz 2003.)
Conclusion
Whether or not they were talented, or inspired by the miracle of
Pentecost, missionaries could only produce linguistic descriptions
much simpler than the complex worlds of talk they found in their
fields of operation. Because the work of linguistic conversion was
founded in the work of selection, missionaries ran the risk of pro-
ducing print-literate images which might count as what Franz
Fanon called linguistic “effigies.” This work was enabled, and con-
strained, by practices of print literacy and social imaginaries which
extended beyond the dictates of faith and demands of power. Cul-
tures of standardization lent missionaries’ linguistic work authority
and meaning not just in their eyes, or those of their converts, but in
broader colonial regimes of power. Because they engaged with such
an immediate and basic aspect of life, missionary linguists did some
of the most intimate colonial work of “conceptualizing, inscribing,
and interacting [with colonial subjects] on terms not of their own
choosing” (Comaroff and Comaroff 1991:15).
By focusing here on the strategies and practices of this work, and
some of its cumulative effects under colonial regimes, I have left
to one side important and complex issues of how this work was
received, resisted, and used by those whose speech the missionaries
so partially described. Targets of conversion may not have rejected
new versions of their languages, but by the same token they did not
passively accept them, as missionaries to “the Shona” discovered.
The capacity to devise print-literate images or practices of literacy
sketched here did not translate into full control of the ways they
were “transmitted” to colonial subjects: not just taught but learned,
not just imposed but assimilated in ways missionaries did not nec-
essarily recognize or condone. 9
It has been useful here to present this work of reducing speech
to writing as if it happened just from “above,” not “below,” and to
consider in chapter 6 this more complex dynamic as it developed
in two other projects of colonial linguistics and power. There I chart
Between Pentecost and Pidgins 121
work which served projects of power as well as resistance, as lin-
guists partly described and partly created languages which bridged
the transition between colonial and postcolonial eras.
Notes to chapter 5
1 Readers may be reminded by this chapter’s title of remarks which close
George Steiner’s After Babel (1975:470). It should be clear, though, that
it serves here to name a very different range of concerns.
2 The biography of this hymn also offered a hopeful example for mis-
sionaries. The African evangelist Prophet William Wade Harris, who
took his mission of conversion to tens of thousands of West Africans
early in the 20th century, offers a native example of what many foreign
missionaries sought to achieve. Harris was a Liberian who spoke
English natively and, after a visitation by the angel Gabriel, began to
preach in pidgin English, but with the Standard Authorized English
Bible tucked under his arm. In this way he created Christian communi-
ties which are still vibrant today.
3 For the influence of this same strand of German mysticism on early
nationalists, including Herder, see Greenfeld 1992.
4 A paradigmatic example of this dilemma has been very well described
by James Clifford (1982) in his book about Maurice Leenhardt, a French
missionary who spent years in Melanesia.
5 “Om een taal goed to leren moet men met het volk familjaar omgaan,
en dit is bij sommige natiën niet anders mogelijk dan door hun gods-
diesnt ann to nemen. En juist dit zou een genootschap dat van bigot-
terie leeft, als een doodzonde aanwitten.”
6 The exception was Halhed’s use of /oo/ for the “long vowel” at the
ends of works like Hindu, which was too sedimented in English prac-
tices of literacy, it seemed, for even him to escape.
7 The representation or missionary groups and regions/dialects at the
meeting was as follows:
Zezuru Roman Catholic/Jesuits
Manyika Roman Catholic, American Methodist, Anglican
Karanga Dutch Reformed
Kalanga London Missionary
Ndebele London Missionary
Ndau American Methodist
Korekore None
122 Between Pentecost and Pidgins
8 Doke’s work in fact was part of a much larger initiative begun in 1928
by the International Institute of African Languages and Cultures (1930),
which devised a comprehensive phonetic alphabet for more than a
dozen sub-Saharan African languages, including Shona.
9 Excellent sources for such work in sub-Saharan Africa are Stumpf’s
(1979) account of the politics of linguistics in Cameroon under three
different colonial regimes, and Derek Peterson’s Creative writing (2004),
which describes the intellectual and political dynamics set in motion
by missionaries and their linguistic work among the Gikuyu of colonial
Kenya. See also Blommaert 2005 for a semiotically sophisticated account
of this work.
Chapter 6
Colonial Linguists,
(Proto)-National Languages
A language never spreads like a liquid, nor even like a disease or a
rumor.
Johannes Fabian, Language and colonial power (1986:8)
The linguistic projects I sketch in this chapter might be called impe-
rial, rather than colonial, to emphasize the ways they emerged and
matured within the political and economic dynamics some called
the New Imperialism. They differ from those discussed in chapter
5 most obviously in that they focused on languages which came to
serve as languages of European power but were not native to Euro-
peans: Swahili in the Belgian Congo, and Malay in the Netherlands
East Indies (or NEI). Because both languages became central instru-
ments of rule, linguists who engaged with them were also involved
with the broader intellectual and ideological challenges of creating
and legitimizing imperial power.
One ex-colonial officer, Leonard Woolf (husband of Virginia),
used the phrase New Imperialism at the turn of the 20th century to
describe the ways Europe’s nations and the United States were
imposing the will of their states “upon the world outside the State
for the economic purposes of the world within the State” (Woolf
1919:16). Competition between imperial powers was shaped by a
global economy as projects of power abroad came into closer align-
ment with national agendas and the dynamics of international
commodity markets. Colonial subjects became increasingly impor-
tant consumers of goods and targets of exploitation for labor on
124 Colonial Linguists, (Proto)-National Languages
plantations, in mines, and so on. Technologies of communication
and information created tighter administrative lines between
peripheries and centers of empire, which needed more intimate
forms of oversight over colonial subjects.
Imperial dynamics in this way produced another demand: colo-
nial subjects who, after being suitably scholarized, could function
as surrogates for Europeans in the infrastructures of colonial exploi-
tation. These were new kinds of colonial subjects, members of a
new dominant – yet still dominated – class, the kind of subaltern
elite which made it possible for 60,000 Britons rule 300 million sub-
jects in South Asia, and 30,000 Dutch to oversee some 70 million
“natives” (in Dutch, inlanders) in the NEI. In 1916 the Belgian
Ministry of Colonies similarly began to pressure functionaries in
the Congo to begin creating a class of “black clerks and craftsmen”
to replace “lower-echelon white personnel who were very expen-
sive and produced mediocre results” (Fabian 1986:50).
As the New Imperialism gave rise to new institutional hierar-
chies among colonial subjects, it also created hierarchies of lan-
guages, which had new places and effects on senses of social
identity. In the British and French empires, colonial subjects were
commonly educated in their masters’ national, native languages.
But in the Belgian Congo and the NEI, other hierarchies were created
as Europeans and non-Europeans alike learned non-native, non-
European languages of power as what the French called a “vehicu-
lar language” (langue véhiculaire), the Germans a “language of unity”
(Einheitsprache), and the English a “language of wider communica-
tion” (a phrase I discuss below). Such languages came to the fore
in the late colonial era in response to requirements of power much
like those served by Nahuatl in 16th century Mexico.
Swahili was the language of wider communication (or LWC)
that Van Bulck saw as a civilizing force which would drive out
weaker languages (discussed in chapter 4), and which Hulstaert
regarded with deep suspicion for the same reason (discussed in
chapter 5). Because Johannes Fabian likewise recognized its impor-
tance as an object of imperial knowledge and instrument of imperial
rule, his account of Swahili’s development, Language and colonial
power, is also a groundbreaking account of the contradictions and
dilemmas of empire in the Congo. Drawing on his critical reading
of this work of imperial linguistics, I present here similar work by
Colonial Linguists, (Proto)-National Languages 125
Dutch linguists which had Malay as its object, tracing conundrums
of language and human difference which Dutch imperialists were
no more able to avoid, resolve, or fully stabilize than their Belgian
counterparts.
First I describe the philological images of the past and strategies
of selection which linguists in the NEI, like their counterparts else-
where, used to partly describe and partly create an imperial Malay.
Unlike colonial linguists in many other places, though, they were
not just technicians of literacy who reduced an unruly diversity to
a unitary written image. They operated also as technocrats of edu-
cation, extending their custodial relation beyond that written image
to secular projects of power, rather than religious conversion.
The second important issue here is how their technical and tech-
nocratic work shaped categories of linguistic and social difference
which came into play when the languages used in zones of colonial
contact were spoken non-natively on both sides of the imperial
divide. In this respect, social histories and forms of use of both
Swahili and Malay embodied what Cooper and Stoler (1997) have
called the “tensions of empire.” Each language was designed as a
way to absorb colonialized speakers into a project of civilization,
but also to impose on them projects of power and exploitation; each
served to “project outward [European] ways of understanding the
world, [but also] to demarcate colonizer from colonized, civilized
from primitive, core from periphery” (Cooper 2005:4). Tracing this
tension as it played out in the politics of Malay helps to chart that
language’s development into an instrument for protonational resis-
tance to the imperial power it originally served.
This leads to the third major concern in this chapter: effects of
work by colonial linguists that outran their intent, which neither
they nor other imperial officials could fully control or recognize.
Colonial subjects pirated “their” languages for purposes of their
own, showing how teaching a language is a bit like providing infor-
mation or money: once given, the giver loses control of the ways
they are used. As empires’ custodial relations to Malay and Swahili
were eroded by communities of subaltern speakers, those languages
could come to embody newer, increasingly public versions of
protonational, anti-imperial identities.
With these three broad issues in mind, I chart Malay’s development
in the space between imperial institutions and interests on one
126 Colonial Linguists, (Proto)-National Languages
hand, and the dynamics of community solidarity and conflict
on the other. Its shifting uses as a communicative conduit, and
meanings as a symbol of identity, can be articulated with Fabian’s
observation about Swahili, quoted at the head of this chapter. I
foreground “language spread” as a label which is not just oversimple,
but because its simplicity makes it ideologically useful for imaginaries
of empire. In this way, I suggest, these two projects of imperial
linguistics were founded on and helped make “natural” a range of
metaphors of social and linguistic change –osmotic flow, quasi-
mechanical replication, biological reproduction, etc. – which are
still in use to describe and disguise the workings of power through
language.
Imperial Ideologies of Language Difference
As the New Imperialism created tighter relations between metro-
politan centers and peripheries of imperial power, language poli-
cies abroad were indirectly shaped by ideologies of language and
national identity in Europe. This interplay between national ideol-
ogy and imperial policy can be read from controversies which
emerged when Prussia began the work of overtaking its imperial
neighbors, not long after the science of language I sketched in
chapter 4 reached the apex of its prestige and influence.
As a latecomer to the club of colonial powers, Prussia’s imperial
ambitions were formally ratified by the other participants at the
Berlin West Africa Conference in 1885. Zeal for colonial projects ran
strong in several sectors of Prussian society, including those which
had long supported religious missions. Friedrich Fabri, a minister
and administrator of missions to Africa, argued in an influential
1859 book that Africans were fit colonial subjects because they had
inherited the curse placed on their forebear, Ham, for his part in
building of the tower of Babel. Twenty years later he argued that
Prussia’s troubled economy could be helped by forming economic
colonies (Handelskolonien) in Samoa, New Guinea, North Borneo,
Formosa, Madagascar, and Central Africa (Fabri 1998[1879]).
Identity-linked ideologies of German language and nation
discussed in chapter 4, could only be transposed with difficulty to
colonial zones of contact, and communication across lines of racial
Colonial Linguists, (Proto)-National Languages 127
difference. On the one hand Germans believed, Mazrui suggests,
“that no African was good enough to speak the German language”
(1975:142) thanks to their “ideology of the natural,” to recall
Meeuwis’ phrase from chapter 5. But this created a practical dilemma
for colonial agents who needed communicative conduits across
colonial territory, and within colonial institutions.
German colonial authorities at first unanimously [took] the view that
the learning of the German language by the African population had
to be carried out as quickly as possible as the most expedient way of
achieving education in the spirit of “Germanism” of the imperialist
type and of making the colonies a safe field for exploitation by
German monopoly capital.” (Mehnert 1973:384)
But nativist senses of linguistic identity caused many to balk at
the prospect of “giving” German to their native subjects. The
German Evangelist Missionary Societies called on Prussia in 1897
to avoid creating a “demanding, easily inflamed and educated
proletariat due to the spread of European languages” (quoted in
Mehnert 1973:388). Seven years later members of this same commit-
tee reiterated that knowledge of German was
a danger for the colony since it leads to the development of a con-
ceited, demanding, and easily dissatisfied race . . . [I]n possession of
the language of the Europeans, [the colonized] feel tempted to con-
sider themselves as equal to them. . . . With the understanding of the
language of the foreigner, his personal authority also vanishes for
ever. (Quoted in Mehnert 1973:391)
The prospect of a native yet German-speaking community was wor-
risome enough to Carl Meinhof, the eminent Africanist I discussed
in chapter 5, that he presented to the Colonial Conference of 1905
a strategic negative example. The spread of English in South Africa,
he argued, was causing the “rising of several tribes [which before]
could not be prepared on account of the language frontiers and
tribal differences” (quoted in Mehnert 1973:389).
Inheritors of Herder’s tradition may have feared the dislocating,
disordering effects of their language on the natural state of their
subjects, but the French and English had no such qualms. They
128 Colonial Linguists, (Proto)-National Languages
“spread” their languages as symbols of civilization, and as instru-
ments of rule. The French regarded no African as being of any
worth until he or she spoke French, Ali Mazrui suggests, because
the spread of their language abroad (la Francophonie) extended the
same assimilationist policies being pursued at home, transforming
peasants into French-speaking citizens of the Republic.1Colonial
subjects in the empire were that part of “a white man’s bur-
den . . . whose first conquests were to be right at home” (Weber
1976:73).
British imperial language policy coalesced formally and famously
with the victory of the so-called Anglicists over the orientalists in
the 1835 debate on the future of India. (See Trautmann 1997 for
further discussion.) The colonial regime, as Thomas Macaulay wrote
in his (in)famous “Minute on Indian education,” was instructed to
make English a means for “uplifting” the Indian people, creating
an Anglophone elite which would be “Indian in blood and colour,
but English in taste, in opinions, in morals, and in intellect
(Macaulay 1972[1835]:249).2
Belgium and the Netherlands stand apart from Europe’s other
imperial powers in two obvious ways. Both were small countries
which were really interstitial on the political landscape, positions
indirectly reflected by the fact that neither possessed a distinctive,
unitary national language of its own. Flemish was spoken “natively”
in Belgium, but only by members of the politically and economi-
cally marginal ethnic group of which Hulstaert was a member.
Dutch, which I discuss below, was spoken as a range of dialects
which had strong resemblances to dialects of German on the
Netherlands’ eastern border.
When these two tiny nations came into possession of enormous,
linguistically diverse, imperial territories much larger than their
own countries, they elected to withhold access to “their” languages
from almost all their colonial subjects. Adopting the stance urged
on the Germans by Meinhof, they made language difference a cor-
relate of “social distance . . . in order to know who was the ruler and
who was the ruled” (Mazrui and Kazungu quoted in Mkangi
1985:334). From an ideological point of view, nativeness of knowl-
edge made speech a constant, embodied mark of the collective
difference of colonial subjects from themselves. Because talk was
self-evidently exemplary of identity, minor differences in accent
Colonial Linguists, (Proto)-National Languages 129
and intonation could be “natural” and naturalizing marks of differ-
ence and deficiency.
From a technocratic point of view, it seems that the ideological
was ascendant over the practical, because European languages
were already institutionally grounded in the print-literate practices
which Europeans needed to develop for Swahili and Malay to
serve as efficient means of communication, and as “second”
languages learned and spoken in the absence of native, “exem-
plary” speakers. To discuss the larger political and cultural
implications of this important circumstance, it is useful to think
of Swahili and Malay not as having been spoken non-natively (like
French by an African or English by an Indian) but un-natively:
in the absence of native-speaking reference points. This label
helps to examine the importance of institutions and practices of
literacy which made it possible for Malay and Swahili to be
described and used as un-native, non-European languages of wider
communication.
Languages of Wider Communication
To frame the issues confronting imperial technocrats in the linguis-
tically diverse Congo and NEI, it is useful to draw on more recent
discussion of challenges faced by their postcolonial Cold War suc-
cessors as “language problems of developing nations.” (See, for
instance, Fishman et al. 1968.) Before and after the end of the im-
perial era, a common solution was to “superpose,” or introduce
from above, literate forms of European languages of power, through
the work of suitably scholarized members of otherwise linguisti-
cally distinct populations. I noted in chapter 5 that during the impe-
rial era this educational work fell to missionaries in some colonies,
whether they wished to do it or not. This was a strong reason for
colonial regimes to require that missionaries be from the home
nation, and native speakers of the national language.
Postcolonial framings of these issues center on “development,”
but during the imperial era civilizational images were elaborated
by linguists with an older, universalist teleology of progress and
modernity. This can be read from the work of the well-known lin-
guist Otto Jespersen, a Dane with no national investment in any
130 Colonial Linguists, (Proto)-National Languages
imperial project, but a strong intellectual investment in Darwinian
history.
Jespersen was born in 1860, the year after Darwin’s Origin of
species was published, a man of his times steeped in social Darwin-
ism before publishing his first philological treatise, Progress in lan-
guage (1894). In this he showed his adeptness in the time-honored
academic tradition of honoring, critiquing, and then appropriating
ideas (and prestige) from a major intellectual ancestor, in his case
August Schleicher. Jespersen used the evolutionary doctrine of
uniformitarianism to develop more rigorous analogies between lin-
guistic and natural history than Schleicher did. If forces of change
operate uniformly over time in the domain of language as well as
nature, Jespersen argued, then deep linguistic history can be exca-
vated with the guidance of more accessible, “shallower” stages of
history. In this way Jespersen bracketed the idea that the present
state of a language depends crucially on its formative conditions of
origin, and focused instead on its susceptibility to universal forces
of change.
Jespersen also criticized Schleicher’s lack of “a rational basis for
determining the relative value or merit of different languages.”
Invoking their common ancestor, Wilhelm von Humboldt, Jespersen
observed that “language means speaking, and that speaking means
action on the part of a human being to make himself understood
by somebody else,” and so concluded “that language ranks highest
which goes farthest in the art of accomplishing much with little
means, or, in other words, which is able to express the greatest
amount of meaning with the simplest mechanism” (Jespersen
1894:13).
Jespersen’s utilitarian metric for gauging progress in language
was innovative in that he centered on a much narrower, more
abstract sense of “meaning” than had Herder, von Humboldt, or
many others attuned to expressive authenticity. Jespersen, like
many of his intellectual contemporaries, accounted for “meaning”
as a matter of how elements of languages serve to “pick out” and
characterize things spoken of in the world. When speech is a matter
of identifying things with acts of reference, and characterizing them
with acts of predication, then its efficiency can be calculated as a
ratio between properties of linguistic form (“mechanism”) and
content (“amount of meaning”).
Colonial Linguists, (Proto)-National Languages 131
This jump from a romanticist to a positivist view of language
allowed Jespersen to compare languages and their stages of devel-
opment, as he showed with one of Schleicher’s own examples:
the historical development of modern English had from Gothic
habaidêdeima:
The English form is preferable, on the principle that any one who has
to choose between walking one mile or four miles will, other things
being equal, prefer the shorter cut. . . . The English form saves a con-
siderable amount of brain work to all English-speaking people, and
especially to every child learning the language. (Jespersen 1894:
19–20)
With vague aesthetic considerations set aside, Jespersen turned
the question of linguistic development on its head: “[t]he so-called
full and rich forms of the ancient languages are not a beauty but a
deformity” (Jespersen 1894:14). This becomes clear in the linguistic
present and “analytic structures” of modern European languages,
which mark their “unimpeachable superiority over the earlier stages
of the same languages.”
Jespersen continued to demonstrate their superiority in Language;
its nature, development and origin, published in 1922, when he was
one of the best known linguists in the world. Taking up the peren-
nial question of “lines of development” of speech, he drew on
languages of “barbarous races” which are rich in words for par-
ticulars, but not for broader categories which subsume them. Know-
ingly or not, he echoed Herder’s comments on Arabic 140 years
earlier, with examples from the language of the aboriginals of Tas-
mania – already exterminated, along with their language – which
offered no general word for “tree” but plentiful names for specific
varieties of tree (gum-tree, wattle-tree, etc.) (Jespersen 1964[1922]:429).
Jespersen gave a new accent to old observations and what was by
then common knowledge: primitiveness was evident in languages
which prevented speakers from seeing the abstract woods for the
concrete trees.3
Jespersen’s vision of “progress” in language paralleled work
among philosophers of language who were modeling languages as
“referential systems sensitive to nature and blind to society” (Gellner
1994:51). (This was the project rejected by Wittgenstein which I
132 Colonial Linguists, (Proto)-National Languages
noted in chapter 1.) Together, linguists and philosophers could
justify a utilitarian ideal of language uniformly distributed across
space, groups, and institutions, and a maximally efficient conduit
for information – identifying and characterizing things – indepen-
dently from other kinds of shared knowledge. Messages are more
autonomous when the conventions of languages are established
independently of communities and locales.
There is an ideological connection between this conception of
language and the workings of empire which is important enough
to be discussed briefly with an eye to the work of Ernest Gellner, a
philosopher of language and social change. Gellner argues that
modern nations need to “spread” languages through educational
infrastructures and practices of print literacy. Such work must be
coordinated by a centralizing state, he further asserts, so that people
who would otherwise never know each other all undergo parallel
processes of what he calls “exo-socialization [and] exo-education.”
This makes it possible for them to communicate despite differences
between the “local intimate units” of which they are members
(Gellner 1983:38).
Gellner framed this argument to account for the rise of modern
nations, but it fits well the logic of work by colonial regimes to
“spread” the languages of interest here. Malay and Swahili had to
be made into what Europe’s languages already were, what Gellner
calls “school-mediated, academy-supervised idiom[s] codified for
the requirements of reasonably precise bureaucratic and techno-
logical communication” (Gellner 1983:57). Only then could they
count as what Jespersen would have recognized as products and
instruments of genuine progress.
I emphasize these as background assumptions for the work
of “spreading” a language of imperial power, and making
plausible, part-for-whole linkages between forms of language on
one hand, and conditions of modernity on the other. There is an
important break with romanticist tradition here, because these
imperial projects were designed, “top down,” to distribute linguis-
tic knowledge uniformly across individuals, as if the work of pro-
ducing linguistic sameness need have no effects on senses of social
sharedness. This partialness of vision can be traced in the history of
Dutch imperial efforts to describe and “spread” Malay as “their”
language of wider communication.
Colonial Linguists, (Proto)-National Languages 133
From Pidgin to Language: Malay People,
Malay Speakers
From a technocrat’s point of view, Swahili and Malay might seem
poor candidates for languages of wider communication because
they require more “initial inputs” than European languages. Euro-
peans who encountered these ways of speech without any suitable
practices of literacy – although both languages were in fact written
by native speakers with Arabic-derived scripts – needed alphabets
suitable for institutionally print-literate practices. Only then could
Malay “spread” across colonial territory and otherwise diverse
populations.
Swahili presented challenges to the Belgians, Fabian argues,
which can be read from the colonial archive of linguistic work to
reduce it to alphabetic writing. Drawing on the philological tradi-
tion I discussed in chapters 4 and 5, linguists created an image of
what they assumed to have been a “pure” Swahili which originated,
conveniently for their own purposes, outside their domain: among
native speakers living as far away as the east coast of Africa and on
the island of Zanzibar. This philological displacement of the lan-
guage’s origin helped to cement the Belgians’ custodial relation to
Swahili, cutting it off from what would otherwise be its dangerous
associations with Islam.
At the same time, linguists used strategies of selection, much like
missionaries discussed in chapter 5, to bypass messy pluralities of
talk in Swahili in different colonial communities. Unitary images of
“their” Swahili helped marginalize other versions which Europeans
never used or understood, but which were nonetheless intimate
parts of the lives of some of their subjects.4
Parallel work partly described and partly created Malay as a
print-literate language of wider communication (LWC) in the NEI,
which covered as much of the globe as the United States, and was
home to millions of speakers of hundreds of native languages. By
1930, Malay was fully officialized, even though it was spoken
natively by a tiny fraction of the colonial population, under the
name bahasa Melayu in that language. “Malaysian” (bahasa Malaysia)
now refers to both native and official dialects of so-called indige-
nous peoples of the nation-state of Malaysia (i.e., non-Chinese,
BORNEO
(Kalimantan)
0 100 200 300 miles
INDIA
CHINA
Philippines
Area of Map
SINGAPORE
MALAYSIA
Riau Islands
JAKARTA
(Batavia) JAVA
SUMATRA
I
N
D
O
N
E
S
I
A
S
t
r
a
i
t
o
f
M
a
l
a
c
c
a
Figure 6.1 Map of Indonesia, the Netherlands East Indies (NEI) (redrawn from Stoler 1985).
Colonial Linguists, (Proto)-National Languages 135
non-Indian, bumiputra).5“Indonesian” (bahasa Indonesia) now names
the official language of the nation-state which emerged from much
of the territory of the former NEI, still spoken un-natively by most
citizens.
In the 1600s, Portuguese explorers and traders had encountered
bahasa Melayu as a “trade language,” “market language,” or non-
native LWC needed to do business as far east as the coast of India
and as far north, by some accounts, as Japan. In the Straits of
Malacca they found people who wrote Malay with an Arabic-based
script, and spoke it as one of the “cultured languages of the
world . . . [the] language of educated people, from the flooding
Indus to China and Japan, and in most of the Eastern islands, much
like Latin in our Europe” (Jean-Baptiste Tavernier 1605 quoted in
Werndly 1736:xxxvii).
Bahasa Melayu was bound to no sense of ethnic or national iden-
tity among these speakers, whose primary allegiances were to Islam
and the kingdom of Melaka. The Portuguese defeat of that kingdom
caused a diaspora of “Malays” to trading ports on coasts of the
many neighboring islands in the early 16th century (Maier 1993:45).
So by the time the Swiss linguist Werndly wrote his Malay grammar
in 1736, Malay was spoken in ports which had become important
for the Dutch, who had forced the Portuguese out of the area and
monopolized the lucrative spice trade.
Extensive experience had taught the Dutch that there were two
basically different kinds of Malay:
High Malay, which is spoken among the high-ranking persons at
the Courts and is used in matters pertaining to the Mohammaden
religion; and Low Malay or Pasar, the Market Malay spoken as the
everyday language in the community. (Valentijn 1724–1726 II-
1:244)
Malay’s doubleness was to become a recurring theme in a
centuries-long controversy among traders, military men, and mis-
sionaries: what language would best serve their purposes – Dutch,
High Malay, or some variety of Low Malay (see Hoffman 1979)?
Colonial expansion only complicated this question as the Dutch
created new zones of colonial contact in pursuit of new interests
and projects.
136 Colonial Linguists, (Proto)-National Languages
From these colonial adventures also arose a Creole colonial com-
munity of native speakers of yet another variety of Malay. In the
16th century, Dutch operations were centered in the trading entrepot
called Batavia on Java’s northwest coast (now Jakarta, Indonesia’s
capital). The dominant segment of this polyglot Creole community
had developed earlier when Portuguese traders and “native”
women had “creole” or “mestizo” children. These children, and
later those of Dutch men and their native or mestizo wives and
concubines, grew up as native speakers of Portuguese and Malay
dialects. One traveler noted late in the 16th century that “when they
reach adulthood they can scarcely speak a word of Dutch decently,
far less maintain a rational conversation without mixing many
words of pidgin Portuguese in it” (N. De Graaff quoted in Taylor
1983:43).
By the 18th century, though, Dutch colonial interests had shifted
to the fertile lands of central Java, and required intensive, sustained
contact with the Javanese elite I mentioned in chapter 3. These were
“natives” distinguished not only by their Kawi literary tradition,
but a complex system of linguistic politesse in their spoken
language.6
The nuanced styles of Javanese speech, keyed to finely dif-
ferentiated ranks and statuses, created problematic interactional
politics in zones of colonial contact which Europeans elsewhere in
the world might have recognized. On one hand, foreigners who
incorrectly used those styles could offend just those powerful native
elites to whom the most deference was owed, and whose support
was most needed. On the other hand, ignorance of those niceties
could be taken as license by native inferiors to show the Dutch less
respect than was their due. As William Gilchrist, the British student
of Hindustani I mentioned in chapter 4, observed: “[t]he insult of
the use of a familiar form by the servant to the sahib was not just a
personal insult but had a much greater consequence of the loss of
dignity for his country and nation” (quoted in Cohn 1996b:43).
To avoid both these interactional pitfalls, and the need to teach
Dutch to the Javanese, the colonialists had recourse to Malay, such
as they knew it, to communicate across the colonial divide. From
the early 18th to the late 19th century, the Dutch satisfied them-
selves with ad hoc, improvised communication in Malay, preferring
it, as one linguist acidly noted, in ways “inversely proportional to
Colonial Linguists, (Proto)-National Languages 137
their knowledge of the language” (Heinrich Kern quoted in
Groeneboer 1998:142). These improvised modes of un-native speech
came to have a name which reflected its home in zones of colonial
contact: “service Malay” (dienstMaleisch).
Imperial malaise
Fabian’s careful study revealed a broad contrast between two types
of Swahili communication which are unequally described in the
colonial archive. “Vertical” communication occurred in “highly
structured and hierarchical interaction” (Fabian 1986:108), while
“horizontal” communication went on within subaltern communi-
ties, usually beyond colonial oversight. Un-native service Malay,
used but looked down on by Dutch and Javanese alike, emerged in
contexts of “vertical communication” as a barely adequate means
for doing business across constantly patrolled borders of colonial
difference. But as colonial society developed along with increas-
ingly differentiated zones of contact, a plurality of needs for
“horizontal communication” gave rise to the plurality of Malays,
reflected in the 30 or so labels which were in use by the end
of the 19th century: “barracks Malay,” “market Malay,” “service
Malay,” etc.
These Malays were basically similar in grammar and lexicon, but
they differed depending on the zones of contact in which they
served. A “loose network of differences and tensions” came to be
mirrored in the ways these Malays were spoken “in the shadow of
a center . . . [which was] strong enough to impose some kind of
political and economic unity but too weak to impose a distinct cul-
tural hegemony” (Maier 1993:46).
Over the course of the 19th century, a burgeoning mercantile
economy and urban society also gave rise to print-literate Malay,
used in newspapers written in the Roman alphabet for a heteroge-
neous readership. Newspapers were written in a distinctly “low”
Malay because “high” varieties of the language – associated in any
event with Arabic practices of literacy – differed too much from the
speech of prospective readers to be understood easily, including
those in the important ethnic Chinese market (Adam 1995:50).
This new Malay – vernacular yet literate, used widely in imperial
society, yet not under the control of imperial authorities – posed an
138 Colonial Linguists, (Proto)-National Languages
indirect but real threat to the NEI government. This undomesti-
cated, quasi-public language arose from dynamics which profited
the empire but which it could not fully govern. Beyond this was
another social problem with a linguistic face: a linguistically hybrid
“mestizo” or Creole community whose members were called Indo-
Europeans, or Indos for short. Some Indos were “pure blood” chil-
dren of Europeans born in the Indies or outside the Netherlands,
but living in the NEI and owing it allegiance as a “second father-
land” (Stoler 2002:106).
More problematic were interracial Indo descendants of (mostly)
Dutch men, who were numerous enough that by 1924 some 10,000
had joined an official group called the Indo-Europeesch Verbond.
Racially interstitial Indos were not regarded as a political threat to
the colonial regime; they “could neither enlist a popular constitu-
ency nor dissociate from [a] strong identification with the
European-born Dutch elite” (Stoler 2002:107). But just as their bodies
were undeniable physical evidence of hybridity, which was prob-
lematic for a racial ideology of empire, so too their distinctive ways
of speaking native Malay set them off as socially hybrid. It posi-
tioned Indos well, in fact, for important roles in the burgeoning
public press, thanks to “their knowledge of Low Malay or Javanese
and their reputed familiarity with Indonesian and with other
Foreign Orientals” (Adam 1995:44).
By the 1890s “the Indo problem” was so acute that the colonial
regime took steps to address their troublesome linguistic hybridity
with educational programs designed to educate young Indos into
correct use of “their” Dutch language. But these programs neither
removed telling traces of Malayness from their Dutch speech – at
least as far as the Dutch from the Netherlands, the totok, were
concerned – nor induced them to renounce their own Malay, called
Petjo. Distinguished by a heavy admixture of Dutch, Petjo was the
linguistic emblem of a hybrid identity, and for the imperial regime
what one linguist called “the one great enemy of the social-
economic development of the Indo population group” (Fokker
1891:83 quoted in Hoffman 1979:84).
This proliferation of varieties of Malay – serving imperial inter-
ests and threatening imperial authority – gave impetus to the project
of colonial linguistics which was designed to finally establish
governmental authority in this sphere of social life. The work of
Colonial Linguists, (Proto)-National Languages 139
“describing” Malay in the NEI, like Swahili in the Congo, counted
in part as a response to social tensions that the language of Malay
embodied, and the challenge it represented to colonial racial
hierarchies.
Imperial Malays
Over most of the 19th century, Dutch imperialists, like their Belgian
counterparts, managed without a “public register or form of dis-
course” (Fabian 1986:139) that also counted as “theirs.” As early as
the 1820s, some had recognized their need for institutional purchase
on Malay, and sought to develop a “self-consciously proprietorial
attitude of didactic Dutch scholarship towards the [Malay] lan-
guage” (Hoffman 1979:78). Three generations later, though, Dutch
linguists were still calling for a Malay which could be “an identical
idiom . . . understood and spoken everywhere” (Fokker 1891 quoted
in Hoffman 1979:84–85).
In 1855 the colonial government commissioned a Malay–Dutch
dictionary from the linguist H. von de Wall, which he based on
research in the Riau islands, in the straits of Malacca, and on the
Malay peninsula. The work he left unfinished at his death passed
to H. N. van der Tuuk, the linguist whose rueful remarks on the
work of translating the Batak language I quoted in chapter 4. He
completed von de Wall’s work in 1877, but not without bitter com-
plaints about the deficiencies resulting from his predecessor’s
reliance on spoken Malay, and his lack of reference to literary
sources.
Van der Tuuk, the Indo son of a Dutch official and Eurasian wife,
had traveled to Holland to study law but threw it over for philol-
ogy. Though he became well known in his own time for ground-
breaking studies of a range of languages in the NEI, he brought a
lifelong concern for Malay, philological expertise, and an acid
writing style to the problem of what he called the “babble” lan-
guage (brabbeltaal) of Low Malay. This “demeaned officials when
they spoke it, and could not elevate natives because it did not speak
to their hearts in their language” (Grijns 1996:369).
Van der Tuuk criticized his colleague’s dictionary as a philologist
who took Bopp and Grimm as his “heroes in the field” (Teeuw 1971:
xviii). He grounded his own image of a “general, cultured Malay”
140 Colonial Linguists, (Proto)-National Languages
(in Dutch, algemeen beschaafd Maleis) in the philological notion that
“every language [is] more or less a ruin, in which the plan of the
architect cannot be discovered, until one has learned to supply from
other works by the same hand what is missing in order to grasp the
original design” (Grijns 1996:359). Van der Tuuk may have been
unusual as a philologist who sought out intimate contact with
speakers of the languages he studied, but rooted his descriptions in
the philological, text-centered methods of reconstruction described
in chapter 3. So fundamental were written materials for van der
Tuuk’s methods that if texts were not available he would induce
literate natives to commit some of their language to writing for him
to study (Teeuw 1971:xx–xxxiv).
Van der Tuuk was not the linguist who finally determined the
forms of “general, cultured Malay,” but his reputation and tren-
chant opinions did help induce the government to deal once and
for all with “the Malay question.” Charles van Ophuijsen – another
child of the Indies, trained in philology and several Asian languages
in the Netherlands – was commissioned in 1908 to devise a Malay
orthography and grammar. He followed in von de Wall’s footsteps
to Sumatra and the Riau islands, but also worked “abroad” – which
is to say beyond Dutch imperial territory – in the Malay peninsula
and Singapore.
In 1910 van Ophuijsen’s Malay grammar was published in
Holland to educate candidates for colonial officialdom. In the intro-
duction, he explained why the Malay of the Riau islands was
the “best”:
In our country [the Netherlands] a variety of dialects are found which
show considerable variation; their influence cannot be denied, even
in speech of civilized people. Nonetheless we speak of one Dutch
language, and regard it as the language of the Dutch people as a
whole, even if it originally was only a dialect which has come to
overshadow other regional ways of speaking, because it came to be
means of literary and other kinds of writing.
So it is also with Malay.
Among the various dialects, Malays – and they are after all the
only qualified judges – give priority to that dialect spoken in Johor,
a part of Malacca, in the Riau Lingga Archipelago. . . . This so-called
Riau or Johor Malay, in which the greater part of the literature is
written will be dealt with in this work. (Van Ophuijsen 1910:2)
Colonial Linguists, (Proto)-National Languages 141
Van Ophuijsen’s parallel between standard Dutch (called in
Dutch Algemeen Beschaavd Nederlands) and “his” Malay is interesting
because, according to Bruce Donaldson (1983:108), standard Dutch
was never natively spoken in any region of the Netherlands. It
seems to have emerged, rather, from a process rather like that used
by Doke to produce the “Shona language” of Rhodesia, discussed
in chapter 5. It was the product of a committee charged by state
authority to devise a uniform orthography for a related group of
dialects. Early 17th century controversy over Calvinist doctrine
induced leading figures in the northern part of the region to com-
mission a new translation of the Bible, which would bear none of
the distinctive traits of speech heard to the south. But this required
accommodating differences between the dozen or so dialects spoken
in the north. The committee undertook this work by making conces-
sions here and there to “assemble” those dialects in a text-based
image of linguistic unity which reflected all partly, but none
entirely.
Whether or not van Ophuijsen knew this history, he did his own
work with recourse to philological images of pure texts and pure
forms. Like the missionary linguists I described in chapter 5, philol-
ogy licensed his part-for-whole substitutions between letter and
voice, simple past and complex present, formal styles and living
diversities of talk. But van Ophuijsen’s written Malay was part of
a print-centered, state-sponsored project enabled by an “extraordi-
nary symbiosis of scholarship with the metropolitan politics of a
colonizing state” (Hoffman 1973:22).
Imperial institutions and ideology strengthened the NEI state’s
custodial relation to Malay enough that some colonial subjects
began to call it, in one or another of its varieties, bahasa Belanda
(literally, “Dutch language”) rather than bahasa Melayu (Maier
1993:57). This label reflects the success of government efforts to
establish Malay as a symbol of its power, and an instrument of its
institutions. A “long-term project to homogenize and unify Malay”
(Maier 1993:55) was feasible precisely because it was artificial, or
“stiff” (Teeuw 1973:120), thanks to the ways it could be “spread”
along with literacy. The state publishing house (Balai Pustaka)
supplied reading materials in “good” Malay (as well as major ethnic
languages) for “a new class of potential readers, with different
living and reading habits, with different expectations with regard
142 Colonial Linguists, (Proto)-National Languages
to books, based on their school experiences” (Teeuw 1973:112).
“These language officers” (called in Dutch taalambtenaren) practiced
a “unique combination . . . of . . . pure [linguistic] science and
applied science oriented to social needs” (Teeuw 1973:112) as they
oversaw Balai Pustaka publications of fiction, poetry, and criticism
which muted Islamic and ethnic Chinese themes in favor of visions
of modernity, progress, and humanism.
Diglossic hierarchies
Van Ophuijsen reduced Malay to alphabetic writing by appropriat-
ing and fixing what Fabian, describing the colonial linguistics of
Swahili, called “chains of hierarchical relations” (1986:4). In both
colonial projects, Latin orthographies came into ascendance over
Arabic scripts, representing the authority of colonial regimes over
politically dangerous Islam. But neither print-based language could
gain purchase in a colonial society at large unless it came to be
recognized as an authoritative reference point, standing apart from
and “above” a wide range of dialects.
From the apex and center of imperial power, “general, cultured
Malay” could be viewed as part of the work of linguistic progress
Jespersen spelled out with his refined social Darwinism. The science
of language promoted “progress” in Malay as a uniform medium
for communicating information (or “content”) in efficient, context-
free “forms.” At once like and inferior to European languages, it
was ideologically grounded in what Philip Abrams (1988[1977]:58)
calls a state idea, and institutionally grounded in a state system. Its
written images made the empire’s modernity concrete and legible
in “a palpable nexus of practice and institutional structure.” In these
two ways Malay, like Swahili, served in what Fabian calls “profes-
sional routines and ideological stereotypes” which insulated simple,
literate images from “the subversive effects of close human inter-
action” (Fabian 1986:131).
But the view was very different from other places in colonial
society. At the margins of colonial power and oversight – on the
street and in the market, on the dock and in the neighborhood –
Malay continued to be a messy plurality of ways of talking. What
Fabian calls the “subversive effects of . . . human interaction”
Colonial Linguists, (Proto)-National Languages 143
allowed Malay vernaculars to coalesce in “horizontal” communica-
tion beyond state oversight.
So the work of colonial linguistics in the NEI had as one side
effect a broad linguistic asymmetry between print-literate and
“vernacular” Malays. This kind of durable, hierarchical relation,
reproduced over time and across generations, has come to be called
“diglossic” in the field of sociolinguistics. In the 1950s, Charles
Ferguson’s studies of Arabic led him to propose diglossia as a
descriptive term for societies which use two varieties of a single
language, “low” and “high” (Ferguson 1959). Since then the term
has been broadened to refer to similar relations existing between
distinct languages: Alcuin’s Europe can be called diglossic because
unitary Latin, the Truth language of Christianity, stood over and
against a plurality of local vernaculars; the “spread” of Nahuatl’s
“lordly speech” in Mexico could likewise be classified as the “super-
position” of the literate “high” form, once friars appropriated it
for their own purposes. Here I can call Malay in the colonial NEI
broadly diglossic, in the sense of the word set out by Ralph Fasold
(1984:53).
This kind of diglossia exists when, in addition to some variety
of language learned first and in informal situations, another rela-
tively highly valued variety of language is learned not in the
home but later, more consciously, through formal education, and
is used in situations perceived as more formal and guarded. The
“high” variety is in one way or other associated with institutions
of authority (religious, political, economic, etc.) and, quite often,
with practices of literacy which shape its transmission. So colonial
regimes that imposed non-native, literacy-related languages on
colonial subjects sought to create diglossic situations by trans-
mitting “high” language varieties to segments of the colonized
population.
Within dominant segments of society, then, High Malay could
be understood, thanks to the work of philologists, as the “better,”
“real” language and could stand, part-for-whole, for the quotidian
realities of Malay which included everyday, “low” talk in society
at large. But these print-based, school-supported notions of
“correctness” lacked purchase in diverse communities of colonial
subjects, where communicative practice, social sharedness, and
144 Colonial Linguists, (Proto)-National Languages
appropriateness made for alternative, quasi-oppositional ways of
speaking “the regime’s” language.
Pramoedya Ananta Toer, Indonesia’s leading man of letters,
described this situation around the same time Ferguson was devel-
oping his ideas about diglossia. In his 1963 account of the “prehis-
tory” of Indonesian, he charted the development of two Malays
which he termed not “high” and “low,” but “school language” and
“working language.”
School language [basa sekolah] . . . [was] a means of communication
between the colonial government and the people . . . useful only for
those who were educated to become civil servants . . . fertilizing the
seeds of the bureaucrat. . . . Working language [basa kerdja] . . . [was
a rubric for the Malay which could] be used at any time . . . develop-
ing on its own, growing steadily, practical and available whenever
needed, spreading more widely and dynamically than school lan-
guage. (Pramoedya Ananta Toer 1963)
This language emerged spontaneously among colonial subjects in
response to their own needs and, indirectly, to imperial forces –
urbanization, trade, commercially based print literacy, etc.
Pramoedya recognized that “school language” had power and
value, but only for “vertical communication” across lines of hierar-
chy within colonial institutions and with colonial subjects. He also
recognized how this broad institutional sphere had been identified,
part-for-whole, with print-based practices of literacy. Beyond this,
he pointed out that Malay was being spoken in communities and
contexts which the state could marginalize but not erase, where
forms and norms of “good” Malay were not really important. A
Marxist who knew a dialectic when he saw one, Pramoedya framed
this as a complementary, conflictual dynamic which would be
resolved through the two languages’ synthesis in the nationalist
revolution which would be led, in part, by those he called “civil
servants,” speakers of the imperial regime’s own Malay.
Subaltern languages and their elites
”High” Malay could only serve the NEI by being “projected
outward,” as Cooper (2005:4) puts it, to a scholarized segment of
Colonial Linguists, (Proto)-National Languages 145
colonial society. At the same time Malay marked their interstitial
position between other natives and the Dutch, members of this
group developed into the subaltern elite which could pirate that
language for projects of proto-national identity building and anti-
colonial resistance.
By the beginning of the 20th century, knowledge of Malay was
being “spread” by a network of schools which, like those in the
Congo, were established to produce a class of native colonial func-
tionaries. “Highly rationalized, tightly centralized, structurally
analogous to the state bureaucracy itself,” (Anderson 1991:121)
these schools were scattered across the vast expanse of colonial ter-
ritory, all transmitting the state’s unified, print-literate Malay as the
key competence for colonial officials.
Viewed from “above,” and with an eye to its institutional purpose,
this system produced not just High Malay-speaking candidates for
government service, but diglossic situations on colonial territory.
Viewed from “below,” though, and with an eye to the persons who
used Malay with each other in those institutions, this system also
created groups of natives who shared not only a language but
membership in a radically new kind of community.
In Imagined communities, Benedict Anderson (1991) describes the
coalescence of such groups in the postcolonial dynamic he calls the
“last wave of nationalism.” He foregrounds imperial Malay’s mean-
ingfulness as a commonality among a new group of functionaries
whose professions made them migrants, “tender pilgrims” as
Anderson calls them, whose careers took them from place to place
within the empire. Their paths crossed and ran parallel as each
made an “inward, upward” journey through the colonial hierarchy
centered on Batavia.
“School Malay” was a constant in these journeys and among
those who made them, whose native languages often differed. As
part of their shared experience, Anderson suggests, Malay helped
them to imagine a society located on the same expanse of territory
which they covered in their collective travels, where that language
was spoken by more people. Projecting outward and into the future
their sense of linguistic sharedness, they helped make Malay focal
for the proto-nationalist movement which was formally and pub-
licly declared in 1928 with the slogan “One island, one language,
one nationality.”
146 Colonial Linguists, (Proto)-National Languages
In a way, this appropriation of High Malay mirrors the work by
an earlier generation of linguists. They had worked in a text-
centered philological mode to describe and partly create literate
Malay, and then subaltern elites appropriated print-based “school
language” into their lives as a language of experience, one which
embodied and helped resolve dilemmas of identity. Perhaps it was
the notion that philologists were promoting progress through lan-
guage which made it easy for the NEI’s Dutch rulers to take an easy,
custodial attitude to “their” surrogate language of power, not grasp-
ing until it was too late that what they had appropriated in one way
could be appropriated from them in another.
Why Languages don’t Spread
I have sketched Malay’s colonial history with an eye to Fabian’s
description of Swahili, trying to demonstrate the importance of his
observation at the head of this chapter. Fabian drew this moral from
a project of imperial linguistics in which Belgian functionaries also
failed to see that the work of linguists had effects and uses which
outran their intentions. This led Fabian to identify social positions
and interests which make it convenient and plausible to think of
such social and linguistic dynamics as a matter of languages “spread-
ing” across territory. I have touched on Darwinian visions as one
element of such views: “stronger,” more “progressive” languages
“spread” by displacing their less efficient, “weaker” counterparts.
Biological metaphors of contagion are plausible partly because they
resonate with Herder’s organismic vision, be they microbial, as
Fabian suggests, or botanical, as in Peter Mühlhaüsler’s warnings
(1996) about languages which “spread” like weeds, from locale to
locale, displacing other languages from their niches.7
Jespersen’s story of progress in language, together with Gellner’s
account of national modernity, have helped here to foreground
another kind of “language spread.” Framing language as a conven-
tional instrument used to talk about the world helps to make
language spread a matter of distributing knowledge of those conven-
tions among individuals separated on some expanse of territory. I
have suggested, along lines set out by Fabian, that colonialists were
able to ignore the fact that the creation of linguistic homogeneity
Colonial Linguists, (Proto)-National Languages 147
could not be isolated from social dynamics, or linguistic sameness
from social sharedness. Perhaps the Belgians, like the Dutch, could
not imagine that “their” language (Swahili) could ground opposi-
tional stances and identities, or that it could be anything but “arti-
ficial,” “secondary,” and “safe” just because it had no native
speakers or place.8
The dynamics I have sketched here have carried over into the
postcolonial era, as languages of wider communication are “spread-
ing” at an accelerated rate. Schleicher, Jespersen, and some of their
modern day successors would see this as the ongoing victory of
stronger languages over weaker. But present-day linguists now
regard these developments as pressing intellectual challenges and
social problems, whether or not they recognize them as legacies of
the colonial era. So in the last chapter I turn to the implications of
that past for a science of language which is increasingly obliged to
engage not just with languages, but speakers and communities, in
a globalizing world.
Notes to chapter 6
1 Relations between ideology and colonial practice were in fact more
complex, as Alice Conklin (1997) shows in her discussion of France’s
West African empire.
2 This debate about language policy involved no dispute about language
value; it was commonly known among all involved that even if a
“native” language were “elevated” to the status of a colonial adminis-
trative language, it would never for that reason, or in that use, be
comparable with English. This shared understanding grounded a
debate on policy on the shared awareness of British superiority and
imperial legitimacy (see Pennycook 1998, Powell 2002).
3 For an overview of general attitudes to the primitive among Jespersen’s
colleagues in the field of ethnology, see Henson 1974.
4 For reasons of brevity I do not discuss here the ways Fabian’s two-sided
approach to this body of work can be applied to the work of colonial
linguistics of Swahili in neighboring British Kenya and German (later
British) Tanganyika (now Tanzania).
5 But not, it should be noted, “aboriginal” speakers of the unrelated
Senoic languages living in remote areas of the peninsula’s highlands.
148 Colonial Linguists, (Proto)-National Languages
6 For further discussion see Errington 1985, 1988, and sources discussed
there.
7 Another metaphor which is less useful here, at least, is developed by
Jean-Louis Calvet (1974) in his digestive description of “spread” of
languages in sub-Saharan languages as a process of “glottophagie,” in
which they “consume” others.
8 This enduring sense of languages having locales can be read from a
silence or gap in the work of colonial linguistics I have not discussed
here: a broad disregard for languages which colonial contexts engen-
dered, and which had no “places” of their own. These are so-called
creole or “pidgin” languages like that which Fanon made the object of
his protest cited in chapter 5. The longstanding marginality of these
languages in linguistics has been subject to an extended intellectual,
ideological, and theoretical critique by Michel Degraff as “the fallacy
of creole exceptionalism; . . . a set of beliefs, widespread among both
linguists and nonlinguists, that Creole languages form an exceptional
class on phylogenetic and/or typological grounds . . . with nonlinguis-
tic implications” (Degraff 2005:533).
Chapter 7
Postcolonial Postscript
Do we need foreign linguists? Unfortunately, yes.
Demetrio Cojtí, at the 1989 Maya Linguistics Workshop
(quoted in England 1995:140)
But as we mock these genteel fumblers of a previous era, we should
prepare ourselves for the jeers of a later century. How come we never
think of that? We believe in evolution at least in the sense of evolu-
tion culminating in us. We forget that this entails evolution beyond
our solipsistic selves.
Julian Barnes, The lemon table (2004:105)
Contemporary linguists deal with data and models very different
from those I have discussed here. Their field’s central agenda, cen-
tered on neurocognitive reality, is to deduce a small number of
abstract parameters, or “atoms of language” (Baker 2001). By
explaining how these parameters combine according to the “mind’s
hidden rules of grammar,” they aim to explain also what count as
superficial phenomena of linguistic diversity. New strategies for
discovering these universal parameters make an ocean of descrip-
tive details of difference less interesting, and the work of engaging
diversity in “the field” a peripheral activity which one committed
fieldworker calls, tongue in cheek, “butterfly collecting” (Everett
2004:141).
But the “withering of fieldwork” among academic linguists
over the last part of the 20th century has not ended traditions of
150 Postcolonial Postscript
linguistic work I have discussed in this book. In this chapter I con-
sider two contemporary engagements with linguistic and human
difference – one obviously rooted in the colonial past, the other less
so – which are reshaping older interests and values, and which
require revised practices and different habits of thought. The main
question here is whether and how such projects differ from those
of an earlier era, and whether linguists are justified in regarding
their field as having left its colonial roots far behind.
I discuss first linguists who have carried over from the colonial
past the double work of conversion sketched in chapters 2 and 5:
missionaries still seeking to transform speakers of exotic languages
in the formerly colonialized world into literate Christians. Once
again I frame a few thin historical slices of this work against a
broader historical and political backdrop. They can be made to
reflect more broadly on the Summer Institute of Linguistics (SIL),
an American organization which now supports missionary linguists
with new methods, goals, and statuses.
My sketch of the SIL’s presence in Guatemala is meant to fore-
ground new versions of ambivalences which have always emerged
in the language-centered work of faith, but which seem clearest
when they are articulated by literate native speakers like Demetrio
Cojtí, “dean of Maya public intellectuals” (Warren 1998:74), quoted
above. He speaks of a “field of operation” where missionary lin-
guists have been joined by their secular counterparts, and about a
politics of identity emerging from the contested senses of “owner-
ship” of language which linguists help to create and shape.
This chapter’s second main topic is a global trend which now has
a claim on the attention of all linguists: social transformations which
are causing thousands of languages to “die” or at least become
“endangered.” The intellectual and political significances of this
impending loss are now redefining linguists’ goals, and the strate-
gies they use to describe languages. A glance at the practices of
literacy linguists are introducing in order to “save languages” shows
also the transformative effects their work can have in what Talal
Asad described in 1986 as “monolithic, ineluctable, and titanic
struggle[s] between strong and weak languages.” Viewed against
the backdrop of translocal, integrative dynamics of “globalization,”
linguists can be seen as engaged not just in such “local struggles”
but with the larger trajectories of change reviewed in previous
Postcolonial Postscript 151
chapters. Now those trajectories are making linguistic diversity into
a scarce resource, and so giving linguists reasons to rethink their
field’s “evolution” in a fast changing world.
Linguistics in a Postcolonial World
The word “postcolonial” has multiple senses because, as a negative
label, it identifies things in relation to an era which is no more; so
the meanings it can express in a phrase like “postcolonial linguis-
tics” vary as do the enterprises I have sketched in this book. One
such postcolonial linguistics is ongoing in nation-states that have
inherited colonial borders and infrastructures, where linguistically
diverse citizenries recognize the enduring importance of European
languages of wider communication. That these are colonial legacies
can only partly be disguised by designations like “official language,”
language of state,” and so on. Such labels mute but do not erase
durable hierarchies and relations of dependency they help to repro-
duce, especially between new national elites who control those
European languages of power, and others whose competences
are restricted to “national,” “regional,” “ethnic,” “tribal” languages,
and so on.
According to the logic of “progress in language” outlined in
chapter 6, these postcolonial confusions of tongues are both causes
and consequences of the backwardness of these “developing”
nation-states. A broadly evolutionistic point of view still allows
such diversity to be regarded as a barrier to efficient communication
which can be overcome by state-centered “language engineering,”
projects which can answer to Raison-Jourde’s description of “schol-
arization” in 19th century Madagascar, or Ernest Gellner’s descrip-
tion of the “exoeducation” and “exosocialization” of citizens of
modern nation-states. Where such transformative efforts fail, people
of the developing world would seem to rely on European languages
as “perpetual consumers of modernity. Europe and the Americas,
the only true subjects of history, have thought out on our behalf
not only the script of colonial enlightenment and exploitation, but
also that of our anticolonial resistance and postcolonial misery”
(Chatterjee 1993:5).
152 Postcolonial Postscript
Chatterjee’s much broader, ironic story of modernity helps here
to consider how a “postcolonial linguistics” might figure in the
shaping of postcolonial national ideologies, and whether it might
have a place in forming new linguistic identities like those dis-
cussed in chapters 5 and 6. I suggested there that linguists devised
print-literate representations of spoken language by projecting their
own social imaginary of national identity and territory, imposing
it from above, through language, along with colonial hierarchies.
The results of that work are still evident in the ethnopolitics of
some developing nations, even if they have become attenuated in
others.1
Part of Chatterjee’s argument about the specter of postcolonial
nationalism is, in fact, that some “ethnic languages” had already
emerged as languages of modern nationalisms during the colonial
era, and as more than lesser likenesses of their European and North
American counterparts. By suggesting that British linguists inad-
vertently produced resources for ethnically based anticolonial resis-
tance in Bengal, he suggests also the possibility of similar projects
in other postcolonial nations.2He argues that “the Bengali lan-
guage” could be appropriated by a native, literate, bilingual elite
after British linguists developed the orthographic substrate for its
print-literate forms. As these developed and circulated, educated
Bengalis came to invest “their” language with senses of sharedness
and community “outside the purview of the state and the European
missionaries” (Chatterjee 1993:7). This became the linguistic ground
for a new social formation and an “inner domain of cultural iden-
tity, from which the colonial intruder had to be kept out; language
therefore became a zone over which the nation first had to declare
its sovereignty and then had to transform in order to make it ade-
quate for the modern world” (Chatterjee 1993:7).
Chatterjee’s account begins where the work of colonial linguists
ended, but is suggestive here for considering unintended uses and
effects of work by other linguists on other “ethnic” languages.
Where social hierarchies continue to be reproduced in conditions
of language difference, he suggests, the results of the work of lin-
guistics might play into analogous projects of autonomy. With this
colonial era scenario in mind, I turn to work on the language
of people who confront a postcolonial regime grounded in an
ethnopolitical hierarchy established by its colonial predecessor.
Postcolonial Postscript 153
The New Missionary Linguistics
Most missionaries these days speak English, and most with an
American accent. The American Baptist International Mission
Board, for instance, currently oversees 5,000 missionaries world-
wide, operates with a 282 million dollar budget, and claims a
church membership abroad of 7.4 million. Of more interest here,
though, is the Summer Institute of Linguistics (SIL). Based at the
University of Texas at Arlington, the SIL has trained and supported
5,300 missionary linguists in the study of some 1,000 languages
around the world. It is also responsible for coordinating and
organizing the single largest, most widely cited compendium of
knowledge of global linguistic diversity: Ethnolo ue, produced
in print form (Grimes 1984) and also available at www.
ethnologue.com.
The SIL’s self-described mission is “facilitating language-based
development . . . [among] the peoples of the world through research,
translation, and literacy” (www.sil.org, accessed February 6, 2007).
This public statement passes over the spiritual mission the SIL has
taken over from its Protestant colonial forebears, described in
chapter 5: the SIL aims to foster literacy as a way of bringing Chris-
tianity’s transformative power to people as they read the gospel in
their native tongues.
This duality of mission suits the postcolonial conditions in which
the SIL works, which require that two separate constituencies be
addressed. Like other missionary societies before it, the SIL depends
for financial support on believers “at home,” to whom an “evan-
gelical mission” can be presented independently of the “strange
procedure” (Stoll 1982:4) of technical linguistic description. Unlike
colonial era organizations, though, the SIL needs support (or at
least tolerance) from officials and apparatuses of postcolonial states
which govern the people they seek to convert. To this constituency
the SIL can present itself as an agency of civilization, working to
give literacy to backward peoples in support of “developing”
nation-states’ own modernizing agendas. This basic duality, and
the conflicts it can create, can be sketched along with the history of
SIL work among speakers of Mayan languages in the highlands
of Guatemala.
154 Postcolonial Postscript
The founder of the SIL, Cameron Townsend (1896–1982), had his
formative experiences as a young evangelical in Guatemala in the
1920s proselytizing and selling Spanish-language Bibles. Guatemala
had by then been independent from Spain for a century, but a series
of dictatorships, coups, and stretches of military rule had done little
to elevate the status of the Mayan Indians whom Townsend aimed
to convert. Then, as now, Mayans comprised a majority of the
country’s population – currently 5.4 million, or 60 percent – but
were politically and economically marginalized by a dominant,
Catholic, Spanish-speaking population.
Successors to the Spanish rulers understood that they were
also assuming the role of civilizers of the Mayans, as mandated by
law in 1824: the Mayan languages were to be rendered extinct
because they were “imperfect” and “insufficient for enlightening
the people or perfecting their civilization” (quoted in Richards
1989:97). But Townsend heard a different calling in Guatemala’s
highlands.
According to the SIL’s own history (Wallis and Bennett 1959:viii),
he began to study the local language, Kaqchikel, after hearing a
Mayan’s “plaintive plea:” “Why hasn’t God learned our language?”
By 1931 he had responded with a grammar of Kaqchikel, and a
translation of the New Testament. But instead of presenting the first
copy of his new translation to this or any other speaker of
Kaqchikel, Townsend gave it to the Guatemala’s strong man of the
moment, Jorge Ubico. Ubico already knew that he and other wealthy
Guatemalan landowners had “overlooked a gold mine” (Stoll
1982:42) in the potential pool of Mayan plantation labor, and saw
Townsend’s linguistic work as a means to that end. Literacy in
Mayan languages, Townsend explained, was just the first step
towards “Indians acquir[ing] the more prestigious, advantageous
[Spanish] tongue, whereupon parents would raise their children as
Spanish speakers” (quoted in Stoll 1982:37).
From the outset, then, Townsend foregrounded the civilizing
effects and social uses of his linguistic work, in effect promoting
“progress through language” by working to induct literate Mayans
into the Spanish language, the dominant society, and a national
political economy, albeit as subordinated workers on land expropri-
ated from their ancestors. Ubico for his part invited Townsend to
Postcolonial Postscript 155
continue his work with the neighboring Q′eqchi′Mayans, who were
creating trouble on plantations recently established on lands where
they had lived.
But Townsend’s wider ambitions led him back to the United
States in order to train Christian soldiers in techniques of linguistic
description who would continue the work of conversion in what
was still a colonial world. With a close eye on developments in the
American academy, he devised training methods by drawing
on “descriptive” or “structural” linguistics, a new discipline
whose development marked a sea shift away from comparative
philology.
To sketch the break between structural linguistics and compara-
tive philology, it is useful to mention influential arguments made
by Ferdinand de Saussure, an eminent Swiss comparative philolo-
gist, in lectures he gave just before World War I. While Otto
Jespersen was elaborating images of language evolution I sketched
in chapter 5, Saussure argued against what he viewed as “absurd
notions, prejudices, mirages, and fictions [which] have sprung up”
(1966:7) in the older field of philology. Point for point, Saussure
denied comparative philologists’ claims and assumptions about lin-
guistic change, granting them interest at most from what he called
a “psychological viewpoint.”3
Saussure argued that elements of language are not organically
bound to meanings, uses, or users, but are rather entirely conven-
tional, abstract, and virtual. They are not natural organisms but
conventional symbols. He regarded speech as a source of data about
the recurring patterns which define a language’s properties, along
lines I set out in chapter 1. In this way Saussure located linguists’
objects of descriptions not in territories, or history, but in human
minds as sociopsychological (or cognitive, as one might say now)
entities. When languages are framed as self-contained systems
which are not bound up with any evolutionary force, then change
in language counts as the effect of socially disruptive forces which
throw those systems into disequilibrium.
Saussure’s groundbreaking ideas resonated with approaches to
the study of North American languages which were emerging in
the United States about the same time. Thanks to a range of devel-
opments over the next 30 years, on both sides of the Atlantic,
156 Postcolonial Postscript
linguists began what one envious anthropologist called their migra-
tion across “the border between the social sciences” and the “exact
and natural sciences” (Levi-Strauss 1967:68).
Townsend adapted these new approaches and ideas to train mis-
sionaries at what he called Camp Wycliffe, and later renamed the
SIL. He saw in this new science the prospect that a set of rigorous
procedures, properly learned and applied, could allow any mission-
ary to reduce any language to writing. (This was the dream of what
professional linguists came to call, tongue in cheek, “cookbook lin-
guistics.”) Townsend was in contact, for instance, with Edward
Sapir, a major figure among the new linguists whose insights into
the “psychological reality” of sound types, or phonemes,4he incor-
porated into his own “psychophonemic” methods for reducing
speech to writing. Though he declined Sapir’s invitation to study
with him at Yale, one of his own protégés, Kenneth Pike, began a
career there which was illustrious not just because he assumed the
presidency of the SIL from 1942 to 1979, but because of academic
publications like Phonemics: a technique for reducing languages to
writing (Pike 1947).5
The new linguistics was also rhetorically useful, because it
allowed Townsend to claim the mantle of scientist, and ideologi-
cally useful because it provided a guide for developing intimate yet
asymmetrical relations with native speakers of the languages mis-
sionaries sought to describe. For them, natives counted as sources
of the data which, under analysis, let them understand a language
better than its speakers. By analyzing an abstract, underlying
system, linguists were able to establish their own more profound
understanding of how “sounds [of a language] are automatically
and unconsciously organized by the native into structural units . . .
[in] varieties a trained foreigner might detect but which a native
speaker may be unaware of” (Pike 1947:57).
So, too, scientific neutrality and objectivity made linguists not
just exponents of civilization and reason, but the bringers of
“descriptive linguistics to country after country for the announced
purpose of foreign national unity” (Stoll 1982:63). These dual pur-
poses were given sharper institutional shape in 1942, when
Townsend created another organization, the Wycliffe Bible Transla-
tors (WBT), which operated much in the manner of missionary
organizations discussed in chapter 5 to support the publishing of
Postcolonial Postscript 157
Christian texts in languages that SIL workers studied. The WBT and
SIL shared a board of directors, but the semblance of an institutional
division of labor lent plausibility to both kinds of work, and reflected
more clearly the double task of describing languages before using
them to translate holy writ. This linking of intellectual and textual
traditions gave rise to a “spectacle of dedication [faith] and technol-
ogy [science] which invites endless speculation” (Stoll 1982:6) about
missionaries’ purposes and motives.
A Newer Linguistics and a Newer Politics
During the Cold War, SIL linguists were able to gain long-term
access to regions and communities of the postcolonial world which
were off limits to other Americans and Europeans. Across much of
what came to be called the Third World, the SIL enlarged its fields
of operation, often in parallel with expanding American military,
political, and economic interests. Ample evidence of SIL workers’
complicity with these larger agendas of power and capital, espe-
cially in the New World, has been gathered by David Stoll in Fishers
of men or founders of empire? (1982) and Gerard Colby and Charlotte
Dennett in their aptly titled Thy will be done (1995). Of interest here,
though, are more local, ambiguous dimensions of what always and
everywhere justified the SIL’s presence: the work of reducing speech
to writing.
Townsend dispatched some of his students to study Mayan lan-
guages in Guatemala, where they were supported by an agency
established in Guatemala’s Ministry of Education: the Instituto
Indigenista Nacional (IIN), which promoted “literacy [in Spanish]
and castillianization” (López 1989:31 quoted in French 2003:488).
Together, the IIN and SIL convened the First Linguistic Conference
(Primer Congreso de Lingüistica) in 1949 to coordinate their strate-
gies for dealing with another instance of the challenge I discussed
in chapter 4: different linguists in different regions, working on dif-
ferent languages, were unfortunately producing “multiple forms of
graphic representation for the indigenous languages of the country”
(IIN 1950:5 quoted in French 2003:488).
In fact the SIL and IIN aimed to devise one orthography which
could serve two incompatible goals. On one side were the metrics
158 Postcolonial Postscript
of empirical accuracy; on the other, the demands of “progress” for
uniform ways of writing diverse Mayan languages, so that speakers
could be quickly assimilated to the Spanish-speaking, Guatemalan
citizenry. The second of these criteria carried the day. Following
Townsend’s strategy, conference participants devised a unified
orthography for Mayan languages in the image of Spanish spelling.
So, for instance, the single digraph qu was chosen as a symbol for
a range of significantly different speech sounds in various Mayan
languages, some only vaguely resembling the sound it represents
in Spanish writing. So the name of one Mayan language was spelled
Quiché, even though that word’s first sound has a pronunciation
and sound qualities quite distinct from that which qu represents in
Spanish.6
The conference’s agenda was complicated also by a small but
troublesome constituency of newly literate Mayans, educated
largely by the SIL, who by 1945 had begun to devise their own
orthographies of their languages. In this way they not only sought
more accurate orthographies for their peoples’ speech, but practices
of literacy which could mark an “inner domain of cultural identity,”
as Chatterjee calls it. By rejecting Spanish images for Mayan speech
they rejected, part-for-whole, the assimilationist policies the SIL and
IIN so transparently supported.
Struggle ensued between “strong” and “weak” languages as
these Mayans tried to develop infrastructures of literacy which mir-
rored and opposed those of dominant outsiders. In the early 1940s,
an association of Mayan teachers and an Academy of Mayan lan-
guages were founded, one of their members also the designer of a
new orthography. Adrián Chavéz could claim for it not just authen-
ticity but superior accuracy: he devised symbols for distinct sounds
which SIL linguists had either failed to recognize, or elected to
ignore as too inconveniently different from Spanish. So, for instance,
Chavéz used k′, not qu, to represent the first sound in the name of
his native language, K′iche′, and further differentiated Mayan from
Spanish writing by adapting characters from precolonial Mayan
systems of writing. These had been largely destroyed by the
Conquistadores as the work of the devil, but counted for him as
“genuinely indigenous symbols” which represented “the marvel-
ous beauty of the old culture” (quoted in French 2003:487).
Postcolonial Postscript 159
Chavéz’s orthography may have been more accurate than the
SIL’s, but without a political and institutional infrastructure it could
not “spread” to members of other communities of speakers of dif-
ferent Mayan languages. This situation began to change, though, in
the 1970s, along with the local ethnopolitics of language, and
guiding ideas of academic linguistics. This second development
came to Guatemala with secular linguists guided not by religious
calling but by a “political subjectivity” (Warren 1998:x): witnesses
to the civil rights and antiwar movement in the United States and
a brutal counterinsurgency movement being waged by the govern-
ment against Mayans.
In 1971 a new nongovernmental organization (NGO) was founded
as a venue for linguistic projects which were both intellectually and
politically motivated. The Proyecto Lingüistico Francisco Marro-
quín (PLFM) – begun by a Briton and a few Americans, one a
research linguist – offered to train Mayans in the field to a level of
proficiency comparable with that of a Master’s degree. But by this
time a paradigm shift in American linguistics had given rise to new
research strategies.
In the 1960s Noam Chomsky reframed the conundrum of linguis-
tic unity and diversity by focusing on the genetic capacities of every
human to learn any language. He posed a simple, compelling ques-
tion: how is it that any normal human child is able to learn a lan-
guage – that is, becomes capable of producing and understanding
indefinitely large numbers of sentences in that language – after
comparatively limited, fragmentary exposure to its use? This ques-
tion shifted the field’s explanatory focus to the study of grammati-
cal phenomena for clues to the nature of a species-wide language
faculty, manifested in each individual’s linguistic competence.
Framed in this way, linguistic research requires as data not
just instances of native speech, but intuitions native speakers have
about speech, based on internalized knowledge which allows them
to distinguish just those combinations of words which count as
“acceptable” or “well formed” in their language. For the study of
English, for instance, it became significant that speakers recognize
that “It is easy to please John” is acceptable, but “It is eager to please
John” is not. This is evidence of underlying “rules” of English which
speakers internalize as part of their competence.
160 Postcolonial Postscript
Because “acceptability judgments” must be elicited, and not
just observed, linguists needed new techniques and ways of
engaging speakers, posing questions which only native speakers
could answer. They were no longer Pike’s unknowing producers of
data, but possessors of knowledge which made them privileged
interlocutors. In principle, at least, the best analysis can be done by
a native speaker, which is one reason linguists in North America
and European increasingly turned to the study of their own
languages, contributing to what Everett called the “withering of
field work.”
Chomsky’s new paradigm, and new ways of engaging speakers
in “the field,” made it increasingly advantageous for linguists to
engage closely and collegially with speakers of Mayan languages,
and by 1976 the PLFM was fully in the hands of Mayan “native
speakers/analysts,” who brought together “expert knowledge, sci-
entific analysis, and Maya professionalization” (French 2003:493).
Working in the tradition of Adrián Chavéz, they began describing
Mayan languages with scientific techniques, and for an inner domain
of Mayan cultural identity, doing genuine and rich postcolonial
linguistics. As lines blurred between linguist and native speaker,
literate and illiterate, “civilized” and “uncivilized,” Mayans were
able to criticize work by SIL linguists on both social and empirical
grounds, pointing to inconsistencies and empirical gaps which
threw their privileged status as scientists into doubt. Already by
1977, SIL workers were shifting alignments away from the metrics
of science to the values of progress, arguing that the paramount
need was for literacy among “ethnic minorities” who were to even-
tually become Spanish speakers.
Only from this point of view could the SIL criticize Mayans’
linguistic work, not as inaccurate but impractical, threatening to
“hinder their own people in making appropriate gains in the mestizo
world” (Henne 1991:5). This criticism effectively presupposed a
class hierarchy within Mayan communities corresponding to that
which had existed previously between Mayans and “outsiders”
who had dealings with them. These native linguists counted as
“elites,” as Henne put it, who had usurped the SIL’s place to pit
their own agenda against one which was ostensibly more accept-
able among less educated but more practically oriented Mayan
“educators” whom the SIL counted as partners and allies.
Postcolonial Postscript 161
With new ethnic violence perpetrated by the Guatemalan state
on Mayans in the 1980s, the PLFM’s work took on new urgency in
a language-centered politics of recognition and a project of linguis-
tic self-determination. Its goal was recognition not just for the
authenticity of all Mayan identities, but also their internally plural
ethnolinguistic character, described in the charter of another NGO
of native linguists as grounding “[t]he Mayan nation[’s] . . . own
values that constitute a great human richness. Among those stron-
gest values that are found are the twenty Mayan languages spoken
today” (Oxlajuuj Keej Maya’ Ajtz’iib quoted in translation in French
2003:495).
As part of a postcolonial, liberationist dynamic, the nativist
linguistic project can be seen as enlisting practices of literacy in
struggles between “weak” languages and a “strong” one. But those
practices could not be adapted without politically and culturally
difficult effects, sketched in previous chapters: even when they
focused on particular languages, native linguists had to devise strat-
egies of selection to resolve differences and create gaps between
unitary letters and diversities of speech. Part of the conflictedness
Cojtí expressed in this chapter’s first epigraph can be traced to this
dimension of Mayans’ literacy-centered engagements with their
own languages.
Cojtí made this remark with an eye to enduring inequalities and
difference in interests between speakers of Mayan languages on one
hand, and foreign academics on the other. He insisted that foreign
scientists had no license to avoid politics in their particular zones
of postcolonial contact:
It is difficult, above all in Guatemala, where Ladino colonialism
reigns . . . for linguists to define themselves as neutral or apolitical,
since they work on languages that are sentenced to death and offi -
cially demoted. In this country, the linguist who works on Mayan
languages only has two options: either active complicity in the pre-
vailing colonialism and linguistic assimilationism, or activism in
favor of a new linguistic order in which equality in the rights of all
language is made concrete. (Cojtí 1990:19 quoted in translation in
England 1995:139)
Even foreign linguists who recognize these obligations cannot shed
their roles as “[o]utsiders . . . [who are ] both resented and viewed
as useful sources of information” (England 1995:141).
162 Postcolonial Postscript
But native linguists can also have trouble finding middle ground
among fellow speakers of languages outside their own expert
circles. They are not always immune from suspicion that they are
prescribing rather than describing a language, and an identity which
goes with it. Nikte′Sis Iboy, for instance, is a linguist who studied
her native Achi language and described it on structural grounds as
a dialect of K′iche′, the larger Mayan language mentioned above.
This earned her accusations from some other Achi speakers that she
was trying (like SIL linguists before her) to “destroy their [Achi]
identity” (England 2003:739). Any response based simply on “facts”
about these languages would fail to engage that criticism in its own
terms, leaving her, like other linguists in other situations, unable to
control the meanings and uses of her work.
Indigenist Identities and Languages
These politics of linguistics emerge when linguists seek to help
speakers remake their “weak” languages in the literate image of
encroaching “strong” ones. But they can only produce relatively
simple images of linguistic complexity which circulate not just as
descriptions, but templates for practices of literacy with meanings
and uses which can shift among speakers and contexts.
What Cojtí described as the “death sentence” placed on Mayan
languages is a dramatic instance of larger patterns of social and
linguistic change around the world. Though they may not be threat-
ened by an oppressive state, other communities and their languages
are being affected by flows of people, capital, and information
which commonly go by the name “globalization.” Some see this as
just the newest phase of the integrative dynamic I have traced in
the work of colonial linguists, which will continue to homogenize
the world’s peoples and the ways they speak. Others see instead a
“massive human-made extinction crisis of languages and cultures”
(www.terralingua.org, accessed 6/9/06).
In The origins of Indigenism: human rights and the politics of identity,
Ronald Niezen (2003) foregrounds a paradox of globalization: the
same forces which threaten marginal communities and languages
are also creating communication channels, contexts, and global
audiences which members of those communities can use to claim
Postcolonial Postscript 163
rights to local autonomy. When members of otherwise scattered
communities recognize their common predicament, they can devise
common responses to similar encroaching forces (political, eco-
nomic, and environmental). Over the last 20 years, transnational
forums have developed where indigenous peoples from different
conditions have been able to “defiantly enter the public sphere” and
where “leaders from Asian, northern Europe, Africa, the Americas,
and the South Pacific . . . meet . . . to discuss the development of
human rights standards for indigenous peoples” (Niezen 2003:
3–4).
Claims for local autonomy need to be issued to global audiences
with vocabularies and rhetorics which all parties can understand,
if not always in the same ways. So in these contexts, issues of
“language” have been invoked with different accents by different
speakers, for different audiences. The organistic tradition of thought
about language I discussed in chapters 3 and 4 is still a strong and
well-traveled means for associating language with place, identity,
and history, and for asserting rights to territories through language
and “natural” genealogical links between speakers and their distant
ancestral pasts. These have the potential to trump historically “shal-
lower” claims of “outsiders” and “newcomers,” and so to contrib-
ute to what Amy Muehlebach (2001) calls “place-making strategies”
of indigenous peoples.
Images of such organic links between people and environments
via language are being mobilized in support of many claims to
indigenous rights. In venues like the Indigenous Peoples’ and
Globalization seminar of July 2001, for instance, the Anishinaabe
activist Winona Laduke from Minnesota said:
[the] teachings of our people concerning our relationships to the land
are deeply embedded in our language. . . . Without our languages,
we are simply wandering – philosophically, spiritually, economi-
cally. To preserve our languages we need to protect our lands and
our historic practices. (Mander and Tauli-Corpuz n.d.:19)
Jeanette Armstrong, of the Okanagans of British Columbia, echoed
Laduke at that meeting by offering a single definition of the Okan-
agan words meaning “our place on the land” and “our language:”
“We think of our language as the language of the land. This means
164 Postcolonial Postscript
that the land has taught us our language” (Mander and Tauli-
Corpuz n.d.:31).7
The broader cultural grounds for understanding meaningful, real
linkages between languages and locales can differ, and do not nec-
essarily translate as easily as statements which affirm that they
exist. But this need not detract from their rhetorical effectiveness if
they resonate with other understandings of languages as ways of
“being-in-the-world” (Muehlebach 2001:416). Claims to local rights
and languages have to fit such understandings if they are to circu-
late in translocal networks. Certainly the new rhetorics of local
identity would be recognizable to Herder, as would descriptions of
indigenist struggles as parts of a battle against the “third extinction
crisis” (Maffi 1999:21) which, after loss of biodiversity and tradi-
tional cultures, extends to language.
At a 1992 UN Conference on Environment and Development
vocabularies of indigenous languages figured centrally in discus-
sion of them as “repositories of vast accumulations of traditional
knowledge and experience” (Brundtland 1987:114). Only with a
sense of language-in-the-world like that I have discussed here – its
situatedness in natural environments, the purity of its origins, and
so on – is it intuitively plausible that the “death” of a language
marks an “extinction of experience” (Nabhan and St. Antoine 1993).
Now this old vision of language has new implications for linguis-
tics. It promotes an indigenist linguistics centered not just on the
work of reducing of speech to writing, or contributing to the archive
of linguistic knowledge, but also and most crucially nurturing a
language in its native habitat:
[T]here is a very close parallel between [ex situ] language preserva-
tion and ex situ conservation in biology: while both serve an impor-
tant function, in both cases the ecological context is ignored. Just as
seed banks cannot preserve a plant’s biological ecology, ex situ
linguistic documentation can not preserve a language’s linguistic
ecology. (Skutnabb-Kangas 1999:46)
But unlike the work of linguists discussed in chapters 5 and 6, who
understood that they were engaged with impure remnants of “pure”
original languages, these situations require that the present (or near
past) be taken as a point of purity which needs to be guarded
Postcolonial Postscript 165
against the hybridizing effects of future contact with “outside” lan-
guages. Images of linguistic and cultural purity can themselves
have different uses and effects in different contexts, as is demon-
strated by Renée Sylvain’s review of claims to indigenous sover-
eignty made on behalf of the San people of South Africa: “[t]he more
essentialized the ‘cultural’ [and linguistic] features become, the
more they are seen as contrary to the historically transitory features
of political economy” (Sylvain 2002:1076).
Linguists working in these postcolonial “zones of contact” now
find that they need to gauge the collateral effects of their work in
fine-grained ways: what “borrowed” words, grammatical construc-
tions, or idioms, for instance, should find a place in their descrip-
tions? What in an orthography they devise might help, hinder, or
skew “indigenous” literacies in different varieties of a language,
sustaining some better than others in ongoing struggles with
“strong” languages? The social importance of such narrow, empir-
ical questions might be much clearer in the postcolonial than the
colonial era, now that linguists recognize that they do their work in
zones of contact populated by multiple voices and interests, and are
not just describing speech but promoting practices of literacy.
Unintended Consequences
Linguists share no party line or agenda for dealing with language
death and endangerment. The late and distinguished Peter Lade-
foged, among the field’s foremost connoisseurs of linguistic diver-
sity, went on record in 1992 with a call to his colleagues to recognize
that speakers of “endangered languages” can decide, in their own
lives and on their own terms, which of several languages is best for
them and their children. On the other hand Steven Pinker, well-
known social Darwinist and sociobiologist, asserts that “[e]very
time a language dies, we lose thousands of unique insights, meta-
phors, and other acts of genius” (Endangered Language Fund n.d.).
Here he draws on ideas and sentiments from the relativist tradition
of linguistics which I have sketched in this book, and which he has
stridently attacked in his own work (Pinker 1995).8
Linguists can motivate their studies of endangered languages
locally or globally. Local views are bound up with the romanticist
166 Postcolonial Postscript
past, privileging uniqueness as intrinsically valuable. Global views,
on the other hand, recognize the value of any one language’s loca-
tion in the full spectrum of human linguistic diversity, and as a
potential source of evidence for the discovery of universal attributes
of language. This position can be framed with Mark Baker’s meta-
phor of the “atoms of language,” noted above. Just as Mendeleev
could only conceive of the periodic table after isolating and compar-
ing properties of many different kinds of elements, so too linguists
can only hope to grasp underlying principles or parameters of lan-
guage by isolating and comparing properties of many different
kinds of languages. If languages “die out” before they have been
studied, linguists can not be confident that they have fully grasped
their underlying unity.9
From this latter position, languages have a different status
as facts, or “data.” It brings with it more concern for sufficiently
complete records of phenomena of speech than the prospect
that the language in question will “die” in the future. But this
dispassionately framed position has its own political implications
for zones of postcolonial contact between linguists and speakers.
As Steven Pinker’s comment quoted above illustrates, this global
view promotes what Jane Hill (2002) has called the “hyperbolic
valorization” of languages as having value transcending any
speaker or community of speakers. When languages count as part
of the universal “property” of mankind, it is “we” (not they) who
lose, as Pinker puts it, if one dies. This is, in effect, a claim to lan-
guages that some speakers refuse to accept, as in the Hopi and
Cupeño communities of the southwest United States which Hill
describes.
So challenges and conundrums confront those who continue the
work of linguistics in a postcolonial world. A linguist may not rec-
ognize speakers’ rights of “ownership” of languages he or she seeks
to describe, but must recognize that he or she cannot “own” that
work. One last example is worth using to show how the lesson of
unintended consequences was learned by linguists working to
“save” an endangered language in the 1970s, well before globaliza-
tion’s effects on linguistic diversity became a widely discussed
issue.
Kenneth Rehg, a student of Micronesian languages, spoken on
tiny islands flung across the face of the western Pacific ocean, has
Postcolonial Postscript 167
recently described his work with colleagues at the University of
Hawai’i when they saw that a foreign presence (mostly American)
was affecting the ways the local languages were spoken; English
was “spreading” among their native speakers. Recognizing the
“threat” posed to these languages, they assumed (“perhaps arro-
gantly” as one put it) that they had “not only a role, but a respon-
sibility to help preserve” them (Topping 2003:524 quoted in Rehg
2004:498).
Driven by what Topping later called a “messianic complex,”
these linguists initiated the Pacific Languages Development Project
(PLDP) which eventually resulted in the writing of grammars of
seven Micronesian languages, and dictionaries of five. These were
first steps in the project’s efforts to teach literacy in these languages
in the islands’ schools. I focus here on the language of Pohnpeian,
spoken on Pohnpei, a roughly circular island just 13 miles across
which is now home to about 30,000 people, roughly one-fourth of
them foreigners. Pohnpeian is interesting because it presented these
linguists with the fewest “dialect problems,” and already had what
was “probably the closest to . . . a widely accepted standard spelling
system” (Rehg 2004:509).
There was only one “dialect problem,” really an accentual
difference in the pronunciation of a few vowels. Speakers of
Pohnpeian’s northern dialect pronounced some words with a vowel
ε(as in “pet”), which speakers of the southern (or Kitti) dialect
pronounced with the vowel ɔ(as in “caught”).10 So the Pohnpeian
word which can be translated as “spouse” is pronounced in the
north as werek, and in the south as w
ɔ
r
ɔ
k. But some words are pro-
nounced identically in the two dialects, using the εvowel (for
instance, the word which means “run aground”, s
ε
r).
Twenty years previously, the linguist Paul Garvin devised what
he called a “cross-dialect grapheme” to capture this correspon-
dence. He made one symbol stand for both pronunciations, a solu-
tion which he reported was “enthusiastically received by both
[southern] and [northern] dialect speakers, as the only way in which
acceptance by both dialect communities could be assured” (Garvin
1954:121 quoted in Rehg 2004:508). Twenty years later, though,
linguists who were seeking to introduce different practices of lit-
eracy found that Garvin’s orthography was inadequate, but were
hard put to devise a better solution.
168 Postcolonial Postscript
Garvin saw that when a speaker of Pohnpeian was reading,
they would “naturally” pronounce the “cross-dialect grapheme” in
accordance with their dialect, northern (as
ε
) or southern (as
ɔ
). But
when speakers set out to write Pohnpeian, they could not be sure
whether they should write a given word with the “cross-dialect”
symbol or not, since a speaker of the northern dialect could not
know (unless familiar with the southern dialect) whether it was
pronounced with the
ε
vowel or not. To resolve this problem, the
PLDP convened a committee which decided to use a symbol cor-
responding to the northern accent. (There were more northern
dialect speakers on the PLDP committee, and their southern-
speaking counterparts acceded to their wish.)
What might seem a minor matter of spelling to outsiders turned
out to be an issue in southern Pohnpei. Residents there asserted
their dialect’s difference, and ensured its preservation in writing,
by stipulating in the district constitution that local pronunciation
be reflected in all official documents (Rehg 2004:509). This counted
not just as a collective, official response to an encroaching “outside”
power, but a reflex of a more fundamental sense of linguistic iden-
tity that Pohnpeians had only partly come to share:
Before the advent of the new orthography, the [southern] dialect was
considered to be of the same status as the Northern dialect. Tradition-
ally, Pohnpeians did not judge people’s speech on the basis of where
they were from; rather, they evaluated it on the basis of social appro-
priateness and eloquence. What is at issue here, then, is the common
controversy that arises in relation to issues about who owns the lan-
guage. (Rehg 2004:509)
The work of linguists in postcolonial Pohnpei in this way had
“unintended consequences” extending beyond the teaching of
letters for speech sounds to the learning of a new sense of owner-
ship. Like their forebears, these linguists brought literacy’s indirect
but powerful shaping effects to bear on Pohnpei, creating new
senses of the value of speech, and of speakers as social beings.
Considering this recent work against a historical backdrop helps
to recognize that linguists cannot avoid shaping the zones of post-
colonial contact where they engage with endangered languages. It
also shows how the work of linguists in a postcolonial world may
Postcolonial Postscript 169
be scientific, but is never insulated from the worlds of those who
speak what they study.
The Truth, and Something Besides the Truth
I have sketched the progress of linguistics in a globalizing world by
emphasizing the preoccupations and prejudices of an older, genteel
philology which were abandoned as questions about a panhuman
“language faculty” have come into ascendance. But I have also
argued by example that when linguists enter and create zones of
postcolonial contact with speakers of endangered languages, they
must navigate newer versions of older conundrums without much
guidance from their descriptive metrics. When linguists describe
their fields of operation as scenes not just of “contact” but “colli-
sion,” “competition,” or “conflict” (Joseph et al. 2003), they acknowl-
edge that there are issues and requirements in their work which fall
beyond the purview of their science.
So the science of language needs to seek purchase on these new
yet old challenges which linguists confront “in the field.” Otherwise
the progressive flattening of linguistic diversity in a globalizing
world will make the field’s investment in the universal properties
of “Language,” rather than in individual languages, seem increas-
ingly solipsistic. As both a theoretical and practical matter, linguists
are increasingly obliged to think through their work’s entanglement
with contexts, projects, and ideas from the earlier era.
Perhaps linguists can recognize more fully their work’s situated,
provisional character. Perhaps essentialized images of languages,
either as natural organisms or neurocognitive systems, may serve
better when describing a language spoken by people concerned less
with its purity than its social meanings and uses. Recognizing dif-
ference and conflict within communities can be recognized as a fact
of linguistic life as much as social life.
Daniel Everett, an accomplished field linguist who has studied
exotic languages spoken along the Amazon river, makes a parallel
point by invoking the philosopher William James as a guide for his
broadly pragmatic, “coherent” fieldwork. The linguist’s obligation,
James gives him license to argue, is less to “truth” than coherence,
such that “whatever I say . . . about one aspect of my object should
170 Postcolonial Postscript
cohere with other statements I have made about the object, and the
sum of my experience with the object” (Everett 2004:142). This
might be a new and not just a different way of developing a more
holistic linguistics if, as Everett suggests, it opens up a language’s
status as an “object” to revision in the course of research, rather
than being assumed to dictate its predetermined directions. Coher-
ence depends on engaging dimensions of linguistic experience
which may come to be known only through research, which is to
say in zones of contact with people who are not just producers of
data, but interlocutors.
Such a program would run against the grain of linguistics in the
current universalist paradigm, as well as that of linguistics during
the colonial era I have discussed in this book. First, it would have
to take seriously contrasting and variable modes of speech and
expressiveness, just the kinds of complexities which colonial lin-
guists stripped away when they reduced “native” speech to writing,
and reduced languages to their own purposes. Second, this work
would require linguists to recognize, with Demetrio Cojtí, that
power differences are always in the zones of contact which they
create in and for their work. These are differences which may be
impossible to eradicate but, once recognized, may help to develop
broadened agendas and strategies: if those differences become more
explicit for all concerned, counting as topics in and not just as
enabling conditions for the work of linguistic description, perhaps
the field can find a different place in a postcolonial world.
Notes to chapter 7
1 One of the clearest examples of such a situation is South Africa, as the
work of colonial linguists survives in the present configuration of
“ethnicities.” For discussion see de Klerk 2002.
2 This question bears very obviously on the politics of national lan-
guages for Swahili in Kenya and Tanzania, and for Indonesian in
Indonesia. But the political and cultural pressures mediated by these
languages came to the fore as they came into the hands of government
officials – technocrats of education – rather than linguists in the nar-
rower sense of the word I have been using here. Discussion of these
languages by historians, political scientists, anthropologists, and
Postcolonial Postscript 171
others is extensive. Blommaert 1999 provides a good discussion of the
complex politics of Swahili in Tanzania; Sneddon’s 2003 state-centered
history of Indonesian can be read together with more politically
oriented reviews by Anderson (1990), Errington (1998), and Vikor
(1988).
3 For evidence, however, of the influence of philological cum racialist
theories of language history on Saussure’s thought, see the discussion
of his and his brother’s interaction with Adolphe Pictet in articles by
John Joseph (2003 and 2000, respectively).
4 Sapir’s famous article on this concept is called “The psychological
reality of the phoneme” (1949b); his much more general book, Lan-
guage (1949a), covers a wider range of issues and shows more clearly
his intellectual debt to the romantic tradition I discussed in earlier
chapters.
5 But Townsend’s awareness of the embeddedness of language in cul-
tures and contexts is evident from his care to include in his curriculum
the study of “customs, psychology, superstitions, vices, economic and
cultural status” of peoples to be missionized (Wallis and Bennett
1959:46).
6 In Castilian spelling qu represents what linguists call a velar stop,
whereas the first sound in the name of this Mayan language, now
more commonly written k′, is a velar ejective.
7 For a sensitive ethnographic account of another way of linking land
and language among a North American people, see Basso 1990.
8 Another linguist who has warned his colleagues against their preju-
dices in favor of endangered languages is Salikoko Mufwene (2003).
His and Ladefoged’s considered arguments benefit from comparison
with less informed celebrations of language death, some measured,
like Malik 2000, and others straightforwardly ethnocentric, like John
Miller’s (2002) editorial in the Wall Street Journal.
9 A newer biological metaphor for this process is developed by a student
of aboriginal languages, Robert Dixon, in his book The rise and fall of
languages (1997), borrowing Stephen Gould’s punctuated equilibrium
model of speciation. Conversely, ecologists – for example, William
Sutherland (2003) – see the problems of endangered languages and
species on a global basis as broadly parallel, statistically relatable
processes. One need only understand languages to be differentiable
members of a species, or species of a genus, instead of part of the fabric
of shared experience in communities of speakers.
10 I use symbols of the International Phonetic Alphabet here.
References
Aarsleff, H. 1982. The tradition of Condillac: the problem of the origin of
language in the eighteenth century and the debate in the Berlin Academy
before Herder. In From Locke to Saussure: essays on the study of language
and intellectual history. Minneapolis: University of Minnesota Press. 146–
209.
Aarsleff, H. 1983. Sir William Jones and the New Philology. In The study of
language in England, 1780–1860. Minneapolis: University of Minnesota
Press. 115–161.
Abrams, P. 1988 [1977]. Notes on the difficult of studying the state. Journal
of historical sociology 1(1):58–88.
Adam, A. 1995. The vernacular press and the emergence of the modern Indonesia
consciousness (1855–1913). Ithaca, NY: Cornell University Press.
Alter, S. G. 1999. Darwinism and the linguistic image: language, race, and
natural theology in the nineteenth century. Baltimore: Johns Hopkins Uni-
versity Press.
de Alva, K. 1982. Spiritual conflict and accommodation in New Spain:
towards a typology of Aztec responses to Christianity. In Collier, G.,
Rosaldo, R. and Wirth, J. (eds) The Inca and Aztec states 1400–1800. New
York: Academic Press. 345–366.
Anderson, B. 1990. Language and power: exploring political cultures in Indone-
sia. Ithaca, NY: Cornell University Press.
Anderson, B. 1991. Imagined communities, 2nd edn. New York: Verso.
Appadurai, A. 1981. Worship and conflict under colonial rule: a South Indian
case. New York: Cambridge University Press.
Asad, T. 1986. The concept of cultural translation in British social anthro-
pology. In Clifford, J. and Marcus, G. (eds) Writing culture: the poetics and
politics of ethnography. Berkeley: University of California Press. 141–164.
Baker, M. 2001. The atoms of language: the mind’s hidden rules of grammar.
New York: Basic Books.
References 173
Balibar, R. 1991. The nation form: history and ideology. In Balibar, E. and
Wallerstein, I. Race, nation, class: ambiguous identities. London: Verso.
86–106.
Barnes, J. 2004. The revival. In Barnes, J. The lemon table. Knopf: New York.
97–116.
Basso, K. H. 1990. Speaking with names: language and landscape among
the Western Apache. In Western Apache language and culture – essays in
linguistic anthropology. Tucson: University of Arizona Press. 175–182.
Bauman, R. and Briggs, C. L. 2003. Language, poetry, and Volk in
eighteenth-century Germany: Johann Gottfried Herder’s construction of
tradition. In Voices of modernity – language ideologies and the politics of
inequality. New York: Cambridge University Press.
Baynham, M. 1995. Literacy practices: investigating literacy in social contexts.
London: Longman.
Beidelman, T. O. 1982. Colonial evangelism: a socio-historical study of an East
African mission at the grassroots. Bloomington: Indiana University Press.
Benes, K. 2001. German linguistic nationhood, 1806–1866: philology, cul-
tural translation, and historical identity in preunification Germany.
Ph.D. dissertation, University of Washington.
Benes, K. 2004. Comparative linguistics as ethnology: in search of Indo-
Germans in Central Asia, 1770–1830. Comparative Studies of South Asia,
Africa, and the Middle East 24(2):117–132.
Benfey, T. 1869. Geschichte der sprachwissenschaft und orientalischen philologie
in Deutschland. [History of linguists and oriental philology in Germany.]
Munich: Cotta.
Besnier, N. 1995. Literacy, emotion and authority: reading and writing on a
Polynesian atoll. New York: Cambridge University Press.
Blommaert, J. 1999. State ideology and language in Tanzania. Köln: Rüdiger
Köppe Verlag.
Blommaert, J. 2005. From fieldnotes to grammar: artefactual ideologies and the
production of languages in Africa. Universiteit Gent-Vakgroep Afrikaanse
Talen en Culturen Research Report No. 6. Gent: Academic Press.
Boone, E. and Mignolo, W. (eds) 1994. Writing without words: alternative
literacies in Mesoamerica and the Andes. Durham: Duke University
Press.
Bopp, F. 1816. Uber das Conjugationssystem der Sanskritsprache in Vergleic-
hung mit jenem der griechischen, lateinischen, persischen und germanischen
Sprache. [On the conjugation system of Sanskrit in comparison with those
of Greek, Latin, Persian and German languages.] Frankfurt am Main:
Andreïsche Buchhandlung.
Bopp, F. 1827. Review of Grimm 1822–1826. Jahrbücher für wissenschaftliche
Kritik. Feb:251–256 and May:725–729.
174 References
Bourdieu, P. 1977. Outline of a theory of practice, trans. R. Nice. New York:
Cambridge University Press.
Bourdieu, P. 1993. The field of cultural production. In Johnson, R. (ed.) The
field of cultural production: essays on art and literature. New York: Columbia
University Press. 29–144.
Brundtland, G. 1987. Our common future. World commission on environ-
ment and development. Oxford: Oxford University Press.
Van Bulck, G. 1948. Les recherches linguistiques au Congo Belge. Mémoires de
l’Institut Royal Colonial Belge, Section des Sciences Morales et Politiques,
vol. 16. Brussels: Imprimeur de l’Académie Royal de Belgique.
Burkhart, L. M. 1989. The slippery earth: Nahua–Christian moral dialogue in
sixteenth century Mexico. Tempe: University of Arizona Press.
Caldwell, R. 1856. A comparative grammar of the Dravidian or south-Indian
family of languages. Madras: University of Madras.
Calvet, J.-L. 1974. Linguistique et colonialisme: petit traité de glottophagie. Paris:
Payot.
Campa, A. 1931. The churchmen and the Indian languages of New Spain.
Hispanic American Historical Review XI:542–550.
Carochi, H. 2001 [1645]. Grammar of the Mexican language with an explanation
of its adverbs, trans. J. Lockhart. Stanford: Stanford University Press.
Carrasco, P. 1982. The political economy of the Aztec and Inca states. In
Collier, G., Rosaldo, R. and Wirth, J. (eds) The Inca and Aztec states
1400–1800. New York: Academic Press. 24–39.
de las Casas, B. 1971. Bartolomé de las Casas; a selection of his writings, trans.
and ed. G. Sanderlin. New York: Knopf.
de Certeau, M. 1984. The practice of everyday life, trans. S. Rendall. Berkeley:
University of California Press.
Chatterjee, P. 1993. The nation and its fragments: colonial and postcolonial
histories. Princeton: Princeton University Press.
Chimhundu, H. 1992. Early missionaries and the ethnolinguistic factor
during the “Invention of Tribalism” in Zimbabwe. Journal of African
History 33:87–109.
Clifford, J. 1982. Person and myth: Maurice Leenhardt in the Melanesian
world. Berkeley: University of California Press.
Clifford, J. 1992. Traveling cultures. In Grossberg, L., Nelson, C. and
Teichler, P. (eds) Cultural studies. New York: Routledge. 96–116.
Coe, M. D. and Van Stone, M. 2001. Reading the Maya Glyphs. New York:
Thames and Hudson.
Cohn, B. 1996a. Introduction. In Colonialism and its forms of knowledge: the
British in India. Princeton: Princeton University Press. 3–15.
Cohn, B. 1996b. The command of language and the language of command.
In Colonialism and its forms of knowledge: the British in India. Princeton:
Princeton University Press. 16–56.
References 175
Colby, G. and Dennett, C. 1995. Thy will be done – the conquest of the Amazon:
Nelson Rockefeller and evangelisms in the age of oil. New York: Harper
Collins.
Collins, J. and Blot, R. 2003. Literacy and literacies: texts, power, and identity.
New York: Cambridge University Press.
Comaroff, J. and Comaroff, J. 1991. Of revelation and revolution. Chicago:
University of Chicago Press.
Conklin, A. 1997. A mission to civilize: the Republican idea of empire
in France and West Africa, 1895–1930. Stanford: Stanford University
Press.
Cooper, F. 2005. Colonialism in question: theory, knowledge, history. Berkeley:
University of California Press.
Cooper, F. and Stoler, A. (eds) 1997. Tensions of empire: colonial cultures in a
bourgeois world. Berkeley: University of California Press.
Darwin, C. 1936a [1859]. Origin of species by means of natural selection; or, The
preservation of favored races in the struggle for life. New York: Modern
Library.
Darwin, C. 1936b [1871]. The descent of man and selection in relation to sex.
New York: Modern Library.
Davies, A. M. 1987. “Organic” and “organism” in Franz Bopp. In
Hoenigswald, H. and L. Wiener (eds) Biological metaphor and cladistic
classification: an interdisciplinary perspective. Philadelphia: University of
Pennsylvania Press. 81–108.
Degraff, M. 2005. Linguistics’ most dangerous myth: the fallacy of Creole
exceptionalism. Language in society 34:533–591.
Derrida, J. 1976. Writing before the Letter. In Of grammatology, trans. G. C.
Spivak. Baltimore: Johns Hopkins University Press. 3–94.
Dirks, N. 1990. History as a sign of the modern. Public culture 2(2):25–
32.
Dixon, R. M. W. 1997. The rise and fall of languages. New York: Cambridge
University Press.
Doke, C. A. 1931. Report on the unification of the Shona dialects. Hertford, UK:
Stephen Austin and Sons, Ltd.
Donaldson, B. C. 1983. Dutch – a linguistic history of Holland and Belgium.
Leiden: Martinus Nijhoff.
Eagleton, T. 2000. The idea of culture. Malden, MA: Blackwell Publishing.
Endangered Language Fund. n.d. Publicity flyer.
England, N. 1995. Linguistics and indigenous languages: Mayan examples.
Journal of Latin American anthropology 1(1):122–149.
England, N. 2003. Mayan language revival and revitalization politics.
American anthropologist 105(4):733–743.
Erdrich, L. 2001. Last report on the miracles at Little No Horse. New York:
Harper Collins.
176 References
Errington, J. 1985. Language and social change in Java: reflexes of modernization
in a traditional royal polity. Athens, OH: Center for International Studies,
Ohio University.
Errington, J. 1988. Structure and style in Javanese: a semiotic view of linguistic
etiquette. Philadelphia: University of Pennsylvania Press.
Errington, J. 1998. Shifting languages: interaction and identity in Javanese
Indonesia. New York: Cambridge University Press.
Everett, D. L. 2004. Coherent fieldwork. In van Sterkenberg, P. (ed.)
Linguistics today – facing a greater challenge. Philadelphia: John Benjamins.
141–162.
Fabian, J. 1983. Time and the other: how anthropology makes its object. New
York: Columbia University Press.
Fabian, J. 1986. Language and colonial power: the appropriation of Swahili in
the former Belgian Congo 1880–1938. Berkeley: University of California
Press.
Fabri, F. 1859. Die Entstehung des Heidenthums und die Aufgabe der
Heidenmission. [The origin of heathenry and the task of missions.] Barmen:
Langewiesche.
Fabri, F. 1998 [1879]. Does Germany need colonies?, trans. E. Breuning and E.
Chamberlain. Lewiston, NY: Edwin Mellen Press.
Fanon, F. 1967. Black skin, white masks, trans. C. L. Markmann. New York:
Grove Press.
Fasold, R. 1984. The sociolinguistics of society. Oxford: Blackwell Publishing.
Febvre, L. 1930. Civilization: evolution of a word and a group of ideas, vol. 2.
Paris: Centre Internationale de Synthèse.
Feld, S. 1996. Waterfalls of song: an acoustemology of place resounding in
Bosavi, Papua New Guinea. In Feld, S. and Basso, K. (eds) Senses of place.
Santa Fe: School of American Research. 91–136.
Ferguson, C. 1959. Diglossia. Word 15:325–340.
Fishman, J., Ferguson, C. and Das Gupta, J. (eds) 1968. Language problems
of developing nations. New York: Wiley and Sons.
Florida, N. K. 1995. Writing the past, inscribing the future: history as prophecy
in colonial Java. Durham, NC: Duke University Press.
Fokker, A. A. 1891. De waarde van het Maleisch als Beschavingsmedium.
[Maly’s worth as a cultural language.] Tijdschrift voor het binnenlandsche
bestuur 5:82–88.
Foucault, M. 1970. The order of things: an archaeology of the human sciences.
New York: Vintage Books.
Fracchia, J. and Lewontin, R. C. 1999. Does culture evolve? History and
theory 38:52–70.
French, B. M. 2003. The politics of Mayan linguistics in Guatemala: native
speakers, expert analysts, and the nation. Pragmatics 13(4):483–498.
References 177
Garvin, P. 1954. Literacy as a problem in language and culture. Georgetown
University Monograph Series on Language and Linguistics No. 7.
Washington, DC: Georgetown University Press. 117–129.
Gellner, E. 1983. Nations and nationalism. Ithaca: Cornell University Press.
Gellner, E. 1994. Culture, constraint, and community. In Anthropology and
politics. London: Blackwell Publishing. 45–61
Goody, J. 1986. The logic of writing and the organization of society. Cambridge,
UK: Cambridge University Press.
Greenblatt, S. 1990. Learning to curse: aspects of linguistic colonialism in
the sixteenth century. In Learning to curse: essays in early modern culture.
Routledge, Chapman and Hall. 16–39.
Greenfeld, L. 1992. Nationalism: five roads to modernity. Harvard: Harvard
University Press.
Grijns, C. D. 1996. Van der Tuuk and the study of Malay. Bijdragen tot de
Taal-, Land- en Volkenkunde 152(3):352–381.
Grillo, R. 1989. Dominant languages: language and hierarchy in Britain and
France. New York: Cambridge University Press.
Grimes, B. (ed.) 1984. Ethnolo ue: languages of the world. Dallas: SIL
International.
Grimm, J. 1822. Deutsche Grammatik. Göttingen: Dieterischsche
Buchhandlung.
Grimm, J. 1984 [1851]. On the origin of language, trans. R. A. Wiley. Leiden:
Brill.
Groeneboer, K. 1998. Gateway to the West: the Dutch language in colonial
Indonesia 1600–1950. Amsterdam: Amsterdam University Press.
Gulya, J. 1974. Some eighteenth century antecedents of nineteenth century
linguistics: the discovery of Finno-Ugrian. In Hymes, D. (ed.) Studies in
the history of linguistics: traditions and paradigms. Bloomington: Indian:
University Press. 258–276.
Hanke, L. 1937. Pope Paul III and the American Indians. Harvard theological
review 30:65–102.
Hanke, L. 1949. The Spanish struggle for justice. Philadelphia: University of
Pennsylvania Press.
Hanks, W. 2000. Genre and textuality. In Intertexts: writings on language,
utterance, and context. Oxford: Rowman and Littlefield. 103–220.
Harries, P. 1988. The roots of ethnicity: discourse and the politics of
language construction in South-east Africa. African affairs 87(346):25–
52.
Harries, P. 1995. Discovering languages: the historical origins of standard
Tsonga in southern Africa. In Mesthrie, R. (ed.) Language and social history:
studies in South African sociolinguistics. Cape Town: David Philip Pub-
lishers. 154–175.
178 References
Harris, R. 1980. The language makers. London: Duckworth.
Harris, R. 1988. Language, Saussure, and Wittgenstein: how to play games with
words. London: Routledge.
Hastings, A. 1994. The Church in Africa,1450–1950. Oxford: Oxford Univer-
sity Press.
Heath, S. 1983. Ways with words: language, life, and work in communities and
classrooms. New York: Cambridge University Press.
Helps, A. 1856–1868. Spanish conquest in America; and its relation to the history
of slavery and to the government of colonies (4 vols). New York: Harper
Brothers.
Henne, M. 1991. Orthographies, language planning, and politics: reflec-
tions of an SIL literacy muse. Notes on literacy 65:1–18.
Henson, H. 1974. British social anthropologists and language. Oxford:
Clarendon.
Herder, J. G. 1769. Kritische Wälder. [Critical forests.] Riga: Hartnoch.
Herder, J. G. 1966 [1787]. Essay on the origin of language. In On the origin
of language, trans. J. Moran. New York: F. Ungar. 85–166.
Herzfeld, M. 1987. Anthropology through the looking-glass: critical ethnography
in the margins of Europe. Cambridge, UK: Cambridge University
Press.
Hill, J. 2002. “Expert rhetorics” in advocacy for endangered languages:
who is listening, and what do they hear? Journal of linguistic anthropology
12(2):119–133.
Hobsbawm, E. 1964. The age of revolution, 1789–1848. New York: Mentor.
Hoenigswald, H. 1974. Fallacies in the history of linguistics: notes on the
appraisal of the nineteeth century. In Hymes, D. Studies in the history of
linguistics: traditions and paradigms. Bloomington: Indiana University
Press. 346–358.
Hoffman, J. 1973. The Malay language as a force for unity in the Indonesian
archipelago. 1815–1900. Nusantara 4:19–35.
Hoffman, J. 1979. A foreign investment. Indonesia 27:65–92.
Holquist, M. n.d. The roots (home) of exile: Germany. mss.
Hulstaert, G. 1950. Carte linguistique du Congo Belge. Mémoires de l’Institut
Royal colonial Belge, Vol. XXXVIII, fasc. 1.
von Humboldt, W. 1988 [1836]. On language: the diversity of human language
structure and its influence on the mental development of mankin, trans. P.
Heath. Cambridge, UK: Cambridge University Press.
Illich, I. and Sanders, B. 1988. The alphabetization of the popular mind. San
Francisco: North Point Press.
International Institute of African Languages and Cultures. 1930.
Practical orthography of African languages. London: Oxford University
Press.
References 179
Irschik, E. F. 1969. Politics and social conflict in South India: the non-Brahman
movement and Tamil separatism, 1916–1929. Berkeley: University of Cali-
fornia Press.
Irvine, J. 1993. Mastering African languages: the politics of linguistics in
nineteenth century Senegal. In Segal, D. and Handler, R. (eds) Nations,
colonies and metropoles. Special issue, Social analysis 33:27–46.
Irvine, J. 1995. The family romance of colonial linguistics: gender and
family in nineteenth century representations of African languages.
Pragmatics 5(2):139–153.
Irvine, J. 2001. Genres of conquest: from literature to science in colonial
African linguistics. In Knoblauch, H. and Kotthoff, H. (eds) Verbal art
across cultures. Tübingen: Gunter Narr. 63–89.
Irvine, J. and Gal, S. 2000. Language ideology and linguistic differentiation.
In Kroskrity, P. (ed.) Regimes of language: ideologies, polities, and identities.
Santa Fe: School of American Research Press. 35–84.
Jespersen, O. 1894. Progress in language with special reference to English.
London: Swan Sonnenschien and Co.
Jespersen, O. 1964 [1922]. Language; its nature, development and origin.
London: G. Allen and Unwin.
Jones, W. 1788. On the orthography of Asiatic words in Roman letters. In
Asiatic researches or, Transactions of the society instituted in Bengal, for
inquiring into the history and antiquities, the arts, sciences, and literature of
Asia I. Calcutta: The Society. 1–56.
Joseph, B. D., DeStefano, J., Jacobs, N. G. and Lehiste, I. 2003. When lan-
guages collide: perspectives on language conflict, language competition, and
language coexistence. Columbus: Ohio State University Press. vii–xii.
Joseph, J. 2000. Language and psychological race: Léopold de Saussure on
French in Indochina. Language and communication 20:29–53.
Joseph, J. 2003. Pictet’s Du Beau (1856) and the crystallization of Saussurean
linguistics. Historiographia linguistica 30(3):365–388.
Joseph, J. and Taylor, T. 1990. Ideologies of language. New York: Routledge.
Kartunnen, F. 1982. Nahuatl literacy. In Collier, G., Rosaldo, R. and Wirth,
J. (eds) The Inca and Aztec states 1400–1800. New York: Academic Press.
395–417.
Kiparsky, P. 1974. From paleogrammarians to neogrammarians. In Hymes,
D. (ed.) Studies in the history of linguistics: traditions and paradigms. Bloom-
ington: Indiana University Press. 331–345.
de Klerk, G. 2002. Mother-tongue education in South Africa: the weight of
history. International journal of the sociology of language 154:29–46.
Koerner, K. F. 1990. Schlegel and comparative linguistics. In de Mauro, T.
and Formigari, L. (eds) Leibniz, Humboldt, and the origins of comparativism.
Amsterdam: John Benjamins. 239–262.
180 References
Krabill, J. R. 1995. The hymnody of the Harrist Church among the Dida of South-
Central Ivory Coast (1913–1949). Frankfurt: Peter Lang.
Kroskrity, P. (ed.) 2000. Regimes of language: ideologies, polities, and identities.
Santa Fe: School of American Research Press.
Ladefoged, P. 1992. Another view of endangered languages. Language
68(4):809–811.
Laiou, A. E. 1998. The many faces of medieval colonization. In Boone, E.
and Cummins T. (eds) Native traditions in the postconquest world. Wash-
ington, DC: Dunbarton Oaks. 13–30.
Lehmann, W. 1967.Introduction to a selection from Deutsche Grammatik.
In A reader in nineteenth century historical Indo-Europan linguistics. Bloom-
ington: Indiana University Press. 46–48.
Lepsius, R. 1855. Standard alphabet for reducing unwritten languages and
foreign graphic systems to uniform orthography in European letters. London:
Seeleys.
Levi-Strauss, C. 1967. Linguistics and anthropology. In Structural anthro-
pology, trans. C. Jacobson and B. Schoepf. New York: Doubleday. 67–80.
Lewontin, R. 2000. The triple helix: gene organism, environment. Harvard:
Harvard University Press.
Lockhart, J. 1991. Nahuas and Spaniards: postconquest Central Mexican history
and philology. Stanford: Stanford University Press.
Lockhart, J. 1992. Nahuas after the conquest: a social and cultural history of the
Indians of Central Mexico, sixteenth through eighteenth centuries. Stanford:
Stanford University Press.
Macaulay, T. B. 1972 [1835]. Minute on Indian education. In Cloive, J. and
Pinney, T. (eds) Selected writings. Chicago: University of Chicago Press.
237–251.
McClintock, A. 1992. The angel of progress: pitfalls of the term “post-
colonialism.” Social text 31/32:84–98.
McKenzie, D. F. 1985. Oral culture, literacy and print in early New Zealand:
the Treaty of Waitangi. Wellington: Victoria University Press.
Macormack, S. 1985. The heart has its reasons: predicaments of missionary
Christianity in early colonial Peru. Hispanic American historical review
65(3):443–466.
Maffi, L. 1999. Language maintenance and revitalization. In Posey, D. (ed.)
Cultural and spiritual values of biodiversity. Nairobi: United Nations Envi-
ronment Programme. 37–44.
Maier, H. M. J. 1993. From heteroglossia to polyglossia: the creation of
Malay and Dutch in the Indies. Indonesia 56:37–65.
Malik, K. 2000. Commentary – let them die. Electronic document, http://
wwww.kenanmalik.com/die.htm, accessed May 21, 2006
References 181
Mander, J. and Tauli-Corpuz, V. n.d. Paradigm wars: indigenous peoples’
resistance to economic globalization. San Franciso: International Forum on
Globalization.
Mazrui, A. A. 1975. The political sociology of the English language: an African
perspective. The Hague: Mouton.
Meeuwis, M. 1999. Flemish nationalism in the Belgian Congo versus
Zairian anti-imperialism: continuity and discontinuity in language ideo-
logical debates. In Blommaert, J. (ed.) Language ideological debates. New
York: Mouton de Gruyter. 381–424.
Mehnert, W. 1973. The language question in the colonial policy of German
imperialism. In African studies – Afrika studien. Berlin: Akademi-Verlag-
Berlin. 383–398.
Menze, E. and Jenges, K. (eds) 1992. Herder’s selected early works, 1764–1767,
trans. E. Menze and M. Palma. Philadelphia: Pennsylvania State Univer-
sity Press.
Metcalf, G. J. 1974. The Indo-European hypothesis in the sixteenth and
seventeenth centuries. In Hymes, D. (ed.) Studies in the history of linguis-
tics: traditions and paradigms. Bloomington: Indiana University Press.
233–257.
Meyer, B. 1999. Translating the devil: religion and modernity among the Ewe in
Ghana. Edinburgh: Edinburgh University Press for the International
African Institute.
Mignolo, W. J. 1995. The darker side of the Renaissance: literacy, territoriality,
and colonization. Ann Arbor: University of Michigan Press.
Miller, J. 2002. How do you say “extinct”?: languages die, the United
Nations is upset about this. Wall Street Journal March 8:W13.
Mitchell, T. 1988. Colonising Egypt. New York: Cambridge University
Press.
Mkangi, K. 1985. The political economy of Kiswahili: a Kenya–Tanzania
comparison. In Maw, J. and Parkin, D. (eds) Swahili language and society.
Beiträge zur Afrikanistik, No. 23. Vienna: Institut für Afrikanistik. 331–
348.
Muehlebach, A. 2001. Making place at the United Nations: indigenous
cultural politics at the U.N. working group on indigenous populations.
Cultural anthropology 16(3):415–448.
Mufwene, S. 2001. The ecology of language evolution. New York: Cambridge
University Press.
Mufwene, S. 2003. Language endangerment: what have pride and prestige
got to do with it? In Joseph, B. (ed.) When languages collide: perspectives
on language conflict, language competition, and language coexistence. Colum-
bus: Ohio State University Press. 324–346.
182 References
Mühlhaüsler, P. 1996. Linguistic ecology: language change and linguistic impe-
rialism in the Pacific region. New York: Routledge.
Müller, M. 1854a. Proposals for a missionary alphabet. London: A. and G.
Spottiswoode.
Müller, M. 1854b. Suggestions for the assistance of officers in learning the lan-
guages of the seat of war in the East. London: Longman, Brown, Green, and
Longmans.
Nabhan, G. and St. Antoine, S. 1993. The loss of floral and faunal story: the
extinction of experience. In Kellert, S. R. and Wilson, E. O. (eds) The
biophilia hypothesis. Washington, DC: Island Press. 229–250.
de Nebrija, A. 1926 [1492] Gramática de la lengua castellana. London: Oxford
University Press.
Nietzsche, F. 1979 [1873]. On truth and lies in a nonmoral sense. In Phi-
losophy and truth. Selections from Nietzsche’s notebooks of the early 1870’s,
trans. D. Breazeale. Atlantic Highlands, NJ: Humanities Press.
Nietzsche, F. 2004 [1872]. Third lecture on the future of our educational
institutions. In On the future of our educational institution, trans. and ed.
M. Grenke. South Bend, IN: St. Augustine’s Press. 61–79.
Niezen, R. 2003. The origins of indigenism: human rights and the politics of
identity. Berkeley: University of California Press.
Ong, W. J. 1977. Interfaces of the word: studies in the evolution of consciousness
and culture. Ithaca: Cornell University Press.
van Ophuijsen, C. A. 1910. Maleische spraakkunst. Leiden: S.C. van
Doesburgh.
Pagden, A. 1982. The fall of natural man – the American Indian and the origins
of comparative ethnology. New York: Cambridge University Press.
Pagden, A. 1995. The effacement of difference: colonialism and the origins
of nationalism in Diderot and Herder. In Prakash, G. (ed.) After colonial-
ism. Princeton: Princeton University Press. 129–53.
Pennycook, A. 1998. English and the discourses of colonialism. London:
Routledge.
Perceval, W. K. 1987. Biological analogy before comparative grammar. In
Hoenigswald, H. and Weiner, L. (eds) Biological metaphor and cladistic
classification: an interdisciplinary perspective. Philadelphia: University of
Pennsylvania Press. 3–38.
Perceval, W. K. 1999. Nebrija’s linguistic oeuvre as a model for missionary
linguists. In Nowak, E. (ed.) Languages different in all their sounds: descrip-
tive approaches to indigenous languages of the Americas 1500 to 1850. Studium
Sprachwissenschaft Beiheft, No. 31. Munster: Nodus Publikationen.
15–30.
Peterson, D. 2004. Creative writing: translation, bookkeeping, and the work of
imagination in colonial Kenya. Portsmouth, NH: Heinemann.
References 183
Philo, of Alexandria. 1854. The works of Philo Judaeus,the contemporary of
Josephus, trans. C. D. Yonge. London: H. G. Bohn. Vol. 2.
Pike, K. 1947. Phonemics: a technique for reducing languages to writing. Ann
Arbor: University of Michigan Press.
Pinker, S. 1995. The language instinct. New York: HarperPerennial.
Pollock, S. 2000. Cosmopolitan and vernacular in history. Public culture
12(3):591–625.
Powell, R. 2002. Language planning and the British Empire: comparing
Pakistan, Malaysia, and Kenya. Current issues in language planning
3(3):205–279.
Pramoedya Ananta Toer. 1963. Setelah abad setelah Abdullah Munsji –
beberapa aspek historik yang digelapkan. [Half a century after Abdullah
Munsji – a few historical aspects which have been hidden.] Minggu
Bintang Timur: Lentera–Lembaran kebudajaan Bintang Timur [Sunday
edition of the Eastern Start, Lantern (culture section)]. August 25, Sep-
tember 22, and October 20.
Pratt, M. L. 1992. Imperial eyes: travel writing and transculturation. New York:
Routledge.
Pugach, S. 2001. Afrikanistik and colonial knowledge: Carl Meinhof, the
missionary impulse and African language and culture studies in
Germany, 1887–1919. Ph.D. dissertation, University of Chicago.
Pugach, S. 2004. Carl Meinhof and the German influence on Nicholas van
Warmel’s ethnological and linguistic writing, 1927–1935. Journal of
Southern African Studies 30(4):825–845.
Rafael, V. L. 1993. Contracting colonialism. Durham, NC: Duke University
Press.
Raffles, T. 1979 [1817]. The history of Java. Oxford: Oxford University Press.
Raison-Jourde, F. 1977. L’échange inégale de la langue: la pénétration
des techniques linguistiques dans une civilisation de l’oral Imerine,
début du XIXe siècle. [The unequal exchange of language: the penetra-
tion of linguistic techniques in an oral civilization.] Annales 32(4):639–
669.
Ramaswamy, S. 1993. En/gendering language: the poetics of Tamil iden-
tity. Comparative studies in society and history 35(4):683–725.
Ranger, T. 1989. Missionaries, migrants and the Manyika: the invention of
ethnicity in Zimbabwe. In Vail, L. (ed.) The creation of tribalism in southern
Africa. Berkeley: University of California Press. 118–150.
Rehg, K. L. 2004. Linguists, literacy, and the law of unintended conse-
quences. Oceanic linguistics 43:498–518.
Ricard, R. 1966. The spiritual conquest of Mexico: an essay on the Apostolate
and the evangelizing methods of the Mendicant Orders in New Spain: 1523–
1572, trans. L. Simpson. Berkeley: University of California Press.
184 References
Richards, J. 1989. Mayan language planning for bilingual education in
Guatemala. International journal of the sociology of language 77:93–115.
del Rincón, A. 1885 [1595] Arte mexicana compuesta por el Padre Anotonia del
Rincon de la compañia de Iesus. Mexico City: Predro Balli.
Rizal, J. 1996 [1887]. Nole me tangere, trans. Ma. Soledad Lacson-Locsin, ed.
R. L. Locsin. Makati City, Philippines: Bookmark.
Robson, S. O. 1988. Principles of Indonesian philology. Providence: Foris
Publications.
van Rooden, P. 1996. Nineteenth-century representations of missionary
conversion. In van der Veer, P. (ed.) Conversion to modernities: the global-
ization of Christianity. London: Routledge. 65–88.
Rowe, J. R. 1982. Inca policies and institutions relating to the cultural
unification of the empire. In Collier, G., Rosaldo, R. and Wirth, J.
(eds) The Inca and Aztec states 1400–1800. New York: Academic Press.
93–118.
Said, E. 1994. Orientalism, 2nd edn. New York: Vintage Books.
Said, E. 1995. Secular interpretation, the geographical element, and the
methodology of imperialism. In Prakash, G. (ed.) After colonialism. Princ-
eton: Princeton University Press. 21–39.
Samarin, W. J. 1984. The linguistic world of field colonialism. Language in
society 13:435–453.
de San José, F. 1832 [1610]. Arte y reglas de la lengua Tagala. [Grammar of
the Tagalog language.] Manila: J. M. Dayot.
Sapir, E. 1949a. Language: an introduction to the study of speech. New York:
Harcourt Brace.
Sapir, E. 1949b. The psychological reality of the phoneme. In Mandelbaum,
D. (ed.) Selected writings of Edward Sapir. Berkeley: University of Califor-
nia Press. 46–60.
de Saussure, F. 1966. Course in general linguistics, trans. W. Baskin, eds C.
Bally and A. Sechehaye. New York: McGraw-Hill.
Schiffman, H. 1996. Linguistic culture and language policy. New York:
Routledge.
Schlapp, C. 2004. The “genius of language” – transformations of a concept
in the history of linguistics. Historigraphia lingusitica XXXI(2/3):367–
388.
von Schlegel, A. 1965. [1818] Course of lectures on dramatic art and literature,
trans. J. Black. London: H. G. Bohn.
von Schlegel, F. 1808. Ueber die Sprache und Weisheit der Indier: Ein Beitrag
zur Begrundung der Alterthumskunde. [The language and wisdom of the
Indians.] Heidelberg: Mohr and Zimmer.
Schleicher, A. 1848–50. Sprachvergleichende Untersuchungen. [Comparative
linguistic reseaches.] Bonn: H. B. König.
References 185
Schleicher, A. 1983 [1863]. The Darwinian theory and the science of lan-
guage. In Koerner, K. (ed.) Linguistics and evolutionary theory: three essays
by August Schleicher, Ernst Haeckel, and Wilhelm Bleek, trans. A. V. W.
Bikkers. Amsterdam: John Benjamins. 1–72.
Schwab, R. 1984. Oriental renaissance, trans. G. Patterson-Black and V.
Reinking. New York: Columbia University Press.
Scott, W. H. 1994. Barangay: sixteenth century Philippine culture and society.
Manila: Ateneo de Manila University Press.
Seed, P. 1991. “Failing to marvel”: Atahualpa’s encounter with the word.
Latin American reseach review 26(1):7–32.
Seitz, G. 1984. Die Bruder Grimm: Leben–Werk–Zeit. Munich: Winkler.
Silverstein, M. 1996. Monoglot “standard” in America: standardization and
metaphors of linguistic hegemony. In Brenneis, D. and Macaulay,
R. K. S. (eds) The matrix of language: contemporary linguistic anthropology.
Boulder: Westview Press. 284–306.
Skutnabb-Kangas, T. 1999. Linguistic diversity and language rights. In
Posey, D. (ed.) Cultural and spiritual values of biodiversity. Nairobi: United
Nations Environment Programme. 46–54.
Smalley, W. A. 1958. Respect and ethnocentrism. Practical anthropology
5:191–194.
Smith, C. C. 1999. The vernacular. In McKitterick, R. (ed.) The New Cambridge
Medieval History, vol. 5. Cambridge: Cambridge University Press. 71–83.
Smith, V. A. 1919. The Oxford history of India, from the earliest times to the end
of 1911, 3rd edn. Oxford: Clarendon Press.
Sneddon, J. 2003. The Indonesia language: its history and role in modern society.
Sydney: University of New South Wales Press.
Steiner, P. 1975. After Babel: aspects of language and translation. Oxford:
Oxford University Press.
Steinmetz, G. 2003. “The devil’s handwriting”: precolonial discourse,
ethnographic acuity and cross-identification in German colonialism.
Comparative studies in society and history 45(1):41–95.
Stoler, A. L. 1985. Capitalism and confrontation in Sumatra’s plantation belt
1870–1979. New Haven: Yale University Press.
Stoler, A. 2002. Carnal knowledge and imperial power: race and the intimate in
colonial rule. Berkeley: University of California Press.
Stoll, D. 1982. Fishers of men or founders of empire? The Wycliffe Bible transla-
tors in Latin America. London: Zed Press.
Street, B. 1984. Literacy in theory and practice. New York: Cambridge Uni-
versity Press.
Stumpf, R. 1979. La politique linguistique au Cameroun de 1884 à 1960: com-
paraison entre les administrations coloniales allemande, francaise et britannique
et du rôle joué par les sociétés missionaires. [Linguistic politics in Cameroon
186 References
1884–1960: a comparison of the German, French, and British colonial
administrations, and the roles of the missionary societies.] Berne: Peter
Lang.
Sutherland, W. 2003. Parallel extinction risk and global distribution of
languages and species. Nature 423(May 15):276–279.
Sylvain, R. 2002. Land, water, and truth: San identity and global indigen-
ism. American anthropologist 104(4):1074–1086.
Tavárez, D. 2000. Naming the Trinity: from ideologies of translation to
dialectics of reception in colonial Nahua texts, 1547–1771. Colonial Latin
American review 9(1):21–47.
Taylor, C. 1995. The importance of Herder. In Philosophical arguments.
Harvard: Harvard University Press. 79–99.
Taylor, C. 2002. Modern social imaginaries. Public culture 14(1):91–124.
Taylor, C. 2004. Modern social imaginaries. Durham, NC: Duke University
Press.
Taylor, J. 1983. The social world of Batavia: European and Eurasian in Dutch
Asia. Madison: University of Wisconsin Press.
Tedlock, D. 1983. The spoken word and the work of interpretation. Philadelphia:
University of Pennsylvania Press.
Teeuw, A. 1971. Introduction. In van der Tuuk, H. A grammar of Toba Batak,
trans. J. Scott-Kemball. KITLV Translation Series No. 13. The Hague:
Martinus Nijhoff. xiii–xxxix.
Teeuw, A. 1973. Pegawai bahasa dan ilmu bahasa, trans. J. M. Polak. [Lan-
guage officers and Indonesian linguistics.] Jakarta: Bhratara Publishers.
Thoreau, H. D. 1962. The journal of Henry D. Thoreau, ed. B. Torrey and
F. Allen. 14 vols. New York: Dover.
Thorne, S. 1999. Congregational missions and the making of an imperial culture
in nineteenth-century England. Stanford: Stanford University Press.
Todorov, T. 1984. The conquest of America: the question of the other, trans. R.
Howard. New York: Harper and Row.
Topping, D. 2003. Saviors of language: who will be the real Messiah?
Oceanic linguistics 42:522–527.
Trautmann, T. R. 1997. Aryans and British India. Berkeley: University of
California Press.
Trautmann, T. R. 2006. Languages and nations: the Dravidian proof in colonial
Madras. Berkeley: University of California Press.
Trend, J. B. 1944. The civilization of Spain. London: Oxford University
Press.
van der Tuuk, H. 1962. De pen in gal gedoopt. Brieven en documenten verzameld
en toegelicht door R. Nieuwenhuys. [With a pen dipped in bile. Letters
and documents assembled with an introduction by R. Nieuwenhuys.]
Amsterdam: G. A. van Oorschot.
References 187
van der Tuuk, H. 1971 [1864, 1867] A grammar of Toba Batak, trans. J.
Scott-Kemball. KITLV Translation Series No. 13. The Hague: Martinus
Nijhoff.
Vail, L. (ed.) 1989. The creation of tribalism in southern Africa. Berkeley: Uni-
versity of California Press.
Valentijn, F. 1724–26. Oude en nieuw Oost-Indiën: vervattende een naauwkeu-
rige en uitvoerige verhandelinghe van Nederlands mogentheyd in die gewesten,
5 vols. [Old and new East Indies: containing a careful and detailed essay
on the Netherlands’ power in the west.] Amsterdam: Van Braam, Onder
de Linden.
Valla, L. 1488 [1441] Elegantiae linguae Latinae. Venice: Bartholomaeus de
Zanis.
van der Veer, P. 2001. Imperial encounters: religion and modernity in India and
Britain. Princeton: Princeton University Press.
Van der Velde, M. 1999. The two language maps of the Belgian Congo.
Annales aequatoria 20:475–489.
Vikor, L. 1988. Perfecting spelling: spelling discussions and reforms in Indonesia
and Malaysia. Dordrecht: Foris Publications.
Wallis, E. and Bennett, M. 1959. Two thousand tongues to go: the story of the
Wycliffe Bible translators. New York: Harper and Bros.
Warneck, G. 1901. Outline of a history of Protestant missions, ed. G. Robson.
New York: Fleming Revell.
Warren, K. 1998. Indigenous movements and their critics: pan-Maya activism
in Guatemala. Princeton: Princeton University Press.
Weber, E. 1976. Peasants into Frenchmen: the modernization of rural France,
1870–1914. Stanford: Stanford University Press.
Werndly, G. H. 1736. Maleische spraakkunst. Uit de eige schriften der Maleiers
opgemaakt. [Malay grammar, from the Malays’ own script.] Amsterdam:
Wetstein.
Whitney, W. D. 1873. Dr. Bleek and the simious theory of language. In
Oriental and linguistic studies: the Veda; the Avesta; the science of language.
New York: Scribner, Armstrong and Co. 292–297.
Wittgenstein, L. 1953. Philosophical investigations, ed. G. E. M. Anscombe.
New York: Macmillan Co.
Woolf, L. 1919. Empire and commerce in Africa: a study in economic imperialism.
London: Allen and Unwin.
Wright, R. 2003. Sociophilological study of late Latin. Turnhout: Brepols.
Zammito, J. H. 2002. Kant, Herder and the birth of anthropology. Chicago:
University of Chicago Press.
Zipes, J. 1988. The brothers Grimm: from enchanted forests to the modern world.
New York: Routledge.
Languages Index
Achi, 162
Amerindian languages, 73
Anishinaabe, 163
Arabic, 53, 58, 73, 80, 98, 131, 133,
135, 137, 142, 143
Austronesian, 66
Bantu, 87, 88
Basque, 73, 76
Batak, 97, 106, 139
Bengali, 152
Buginese, 47
Castilian, 18, 20, 25, 29, 30, 31, 37,
40, 157, 171
See also Spanish
Celtick, 56
Chinese, 53, 73, 75–8
Congolese languages, 119
Coptic, 73
Cupeño, 166
Dida, 93, 95
Dravidian, 90
Dutch language, 128, 132, 135, 136,
138, 140, 141
English, 33, 35, 37, 43, 47, 99, 101,
121, 127, 128, 129, 131, 147,
153, 159, 167
Flemish, 119
French, 19, 50, 74, 119, 124, 128–9,
157–8
Fula, 85
German, 48, 50, 61, 66, 79, 81, 126,
128
Gikuyu, 122
Gothic, 56, 79
Greek, 49, 55, 56, 58, 71, 75–81
Gwamba, 115
Hebrew, 54, 55, 58
Hindustani, 84, 136
Hopi, 166
Ilokano, 36
Imerina, 100–3
Indo-European, 61–2, 71, 77, 78, 81,
83, 87, 138
Indo-Germanic, see
Indo-European
Indonesian, 135, 138, 144
Indonesian, see Malay
Javanese, 65–8, 76, 77, 136–8
Kaqchikel, 154
Karanga, 111–12, 121
Kawi, 65–66, 136
Kíché, 158
Korekore, 112, 121
Latin, 18, 19, 20, 21, 28, 30–3, 36–7,
40–2, 49, 55, 56, 65, 75, 76, 95,
135, 142, 143
Lingala, 119
Malay, 16–17, 66, 125, 129, 132–3,
135–46
Malgache, see Imerina
Manyika, 111–12, 121
Maori, 98–9
Mayan languages, 153–4,
157–62
Micronesian languages, 166, 167
Mongo, 117
Nahuatl, 13, 23, 31–4, 38–41, 45–6,
48, 55, 96, 124, 143
Ndebele, 112–13, 121
Okanagan, 163
Old High German, 79
Oromo, 21
Persian, 56, 58
Petjo, see Malay
Pohnpeian, 167–8
Portuguese, 135–6
Provençale, 74
Qéqchí, 155
Quechua, 28, 39, 48, 53
Quiché, see Kíché
Romance languages, 19
Ronga, 116–17
Sanskrit, 49, 56–7, 65–7
Senoic languages, 147
Sereer, 85
Shona, 109, 112–13, 117, 120, 122, 140
Siamese, 53
Spanish, 18, 19, 23, 25, 28, 32, 40,
42, 46, 154, 157, 158
See also Castilian
Swahili, 16–17, 119, 123–6, 129,
132–3, 137, 139, 142, 146–7, 170
Tagalog, 13, 23, 30, 34–8, 41–2,
45–6, 48, 55
Tamil, 57, 89–90
Thonga, 115–16
Wolof, 85
Zezuru, 111–12, 121
190 Languages Index
Persons Index
Aarsleff, H., 50, 58, 70
Abrams, P., 142
Adam, A., 137, 138
Alcuin, 17, 19, 20, 31, 143
Alter, S., 71, 81
de Alva, K., 39
Anderson, B., 16, 31, 107, 108, 145,
171
Armstrong, J., 163
Asad, T., 150
Baker, M., 149, 166
Balibar, E., 108
Bauman, R., 91
Baynham, M., 12
Beidelman, T., 96
Benes, K., 75, 77
Benfey, T., 75
Bennett, M., 154, 171
Berthoud, P., 114–16
Bleek, W., 87
Blommaert, J., 171
Boone, E., 46
Bopp, F., 77–81
Bourdieu, P., 21, 91
Briggs, C., 91
Van Bulck, G., 85–8, 117, 124
Burkhart, L., 39, 41, 45, 46
Caldwell, R., 90
Calvet, J.-L., 148
Campa, A., 29
Carochi, H., 32–4, 38–9
de las Casas, B., 26, 27, 30, 100
Charlemagne, 18
Chatterjee, P., 151–2, 158
Chavéz, D., 158–60
Chimhundu, H., 112
Chomsky, N., 159–60
Cicero, 27
Clifford, J., 94
Cohn, B., 3, 60, 68, 84, 89, 94, 103,
136
Cojtí, D., 149–50, 161–2, 170
Colby, G., 157
Columbus, 24, 38
Comaroff, John, 16, 120
Comaroff, Jean, 16, 120
Cooper, F., 5, 125, 144
Cortez, 25, 38, 41
Creux, E., 114–15
Darwin, C., 82–3, 91, 130
Davies, A., 78
Degraff, M., 148
Dennett, C., 157
Derrida, J., 8, 32
Dipanegara, 67
Dirks, N., 89
Doke, C., 111–12, 122, 141
Donaldson, B., 141
Eagleton, T., 51
Erdrich, L., 22, 46
Everett, D., 149, 160, 169, 170
Fabian, J., 17, 117, 123–6, 133–7,
139, 142, 146–7
Fabri, F., 126
Fanon, F., 14, 93, 95, 97, 120
Fasold, R., 143
Febvre, L., 56
Feld, S., 52
Ferguson, C., 143–4
Fishman, J., 129
Florida, N., 67
Fokker, A., 138–9
Foucault, M., 21
Fracchia, J., 82–3
Gal, S., 21, 85
Garvin, P., 167–8
Gellner, E., 131–2, 146, 151
Gilchrist, J., 84, 103, 136
Goody, J., 21
Greenblatt, S., 24
Greenfeld, L., 72, 91, 121
Grijns, C., 139–40
Grimm, J., 79–81, 91, 101
Halhed, W., 103, 121
Hanke, L., 26, 28
Harries, P., 116
Harris, W., 121
Henne, M., 160
Herder, J., 48–56, 66, 69–70, 108,
119, 121, 127, 130–1, 146
Herzfeld, M., 70
Hill, J., 166
Hobsbawm, E., 71
Hoenigswald, H., 58
Hoffman, J., 135, 138–9, 141
Holquist, M., 75
Hulstaert, G., 117, 119
von Humboldt, W., 66–9, 77–8,
89–90, 92, 94, 95, 130
Illich, I., 19
Irschik, E., 91
Irvine, J., 85
James, W., 169
Jespersen, J., 81, 129–32, 142, 146–7,
155
Jones, W., 48–9, 56–9, 64–9, 103–4
Kant, I., 49, 51
Karttunen, F., 39
Kern, H., 137
Kiparsky, P., 81
Koerner, K., 73
Krabill, J., 93
Ladefoged, P., 165
Laduke, W., 163
Lehmann, W., 79
Leibniz, G., 57
Lepsius, R., 96, 104–5, 107
Levi-Strauss, C., 156
Lewontin, R., 82, 83
Lopez, F., 36–7
Macaulay, T., 128
McClintock, A., 4
McKenzie, D., 99
Macormack, S., 28
Maffi, L., 164
Maier, H., 135, 137, 141
Mazrui, A., 127, 128
Meeuwis, M., 119
Mehnert, W., 127
192 Persons Index
Meinhof, C., 87–8, 102, 104, 127–8
Meyer, B., 96, 98, 106
Mignolo, W., 32–3, 46–7
de Minaya, B., 28
Mkangi, K., 128
Muehlebach, A., 163–4
Mühlhaüsler, P., 146
Müller, M., 77, 82, 84, 89, 105
de Nebrija, A., 18, 20
Newton, I., 64
Nietzsche, F., 70, 72
Niezen, R., 162–3
Nikté Sis Iboy, 162
Ong, W., 44, 46, 55
van Ophuijsen, C., 140–2
Oxlajuuj Keej Mayá Ajtzíib, 161
Pagden, A., 27, 30, 53
Perceval, W., 30, 78
Peterson, D., 95–6, 122
Pike, K., 156, 160
Pinker, S., 165, 166
Pizarro, 25, 38
Pollock, S., 65
Pramoedya Ananta Toer, 144
Pratt, M., 2, 21
Pross, K., 52
Pugach, S., 87–8
Radama I, 100
Rafael, V., 30–1, 42–3, 48
Raffles, T., 65
Raison-Jourde, F., 100–2, 151
Ramaswamy, S., 90
Ranger, T., 11, 112–13
Ranavalona, 101–2
Rask, R., 91
Rehg, K., 166–8
Ricard, R., 29, 45
Richards, J., 154
del Rincón, A., 33
Rizal, J., 46
Robson, S., 59
van Rooden, P., 106
Said, E., 14, 49, 56–7, 60, 67, 70, 116
Samarin, W., 96
Sanders, B., 19
de San José, F., 37, 41
de Santo Tomas, D., 28
Sapir, E., 156, 171
de Saussure, F., 8, 155, 171
Schiffman, H., 90
Schlapp, C., 81
von Schlegel, A., 72, 74, 76–7, 89
von Schlegel, F., 72–4, 76–7, 89
Schleicher, A., 81–3, 130–1, 147
Schwab, R., 57
Scott, W., 35–6, 44
Seed, P., 45
Seitz, G., 79
Silverstein, M., 108
Skutnabb-Kangas, T., 164
Smalley, W., 96
Smith, C., 19–20
St. Paul, 27
Steiner, G., 121
Steinmetz, G., 120
Stoler, A., 125, 138
Stoll, D., 153–4, 156–7
Street, B., 12
Stumpf, R., 122
Sylvain, R., 165
Tavárez, D., 28, 40
Tavernier, J.-B., 135
Taylor, C., 11, 52–3
Tedlock, D., 46
Teeuw, A., 139–42
Thoreau, H., 48
Thorne, S., 94, 106
Todorov, T., 24
Persons Index 193
Topping, D., 167
Townsend, C., 154–7, 171
Trautmann, T., 49, 56, 64, 68, 70,
128
Trend, J., 18
van der Tuuk, H., 97, 106, 139,
140
Ubico, J., 154
Vail, L., 113
Valla, L., 18, 20
van der Veer, P., 89
van der Velde, M., 87
Vikor, L., 171
Voltaire, 61
von de Wall, H., 139–40
Wallis, E., 154, 171
Warneck, G., 98
Warren, K., 150, 159
Weber, E., 107, 128
Werndly, G., 135
Whitney, W., 71, 82, 87, 91
Wittgenstein, L., 7–10, 131
Woolf, L., 123
194 Persons Index
Subject Index
British colonialism, 56–8, 65, 84–5,
89–90, 98–101, 105, 111, 124,
128, 147, 152, 163
British East India Company, 49, 58,
72
Camp Wycliffe, 156
Carnegie Foundation, 112
Catholicism, 13, 46, 88, 95, 96, 100,
111, 117, 119, 121
Caucasus, 77
Central Africa, 85, 126
Central Asia, 88
Charlemagne, 13, 18, 19
civilization, 6, 12, 14, 15, 23, 27, 38,
42, 50, 55, 60, 67, 68, 71, 76–8,
80, 83, 89, 91, 98, 102, 107, 119,
125, 128–9, 153–4, 156
classical philology, 70, 82
Cold War, 129, 157
community, 10, 16, 51–4, 62, 66,
96–7, 107–9, 112, 116, 126, 134,
142, 152, 166
comparative philology, 14, 68,
70–1, 81–3, 85, 90, 91, 102,
104–5, 108, 119, 139, 155
Congo, 17, 87, 117, 124, 129, 139,
145
Abyssinia, 21
accent, 10, 108, 128
Africa, 6, 16, 22, 93, 100, 109, 113,
115, 126–7, 133
alphabets, 4–5, 11, 22, 32, 35,
40, 44, 48, 54–5, 76, 103–5,
133
Alterthumwissenschaft, 70
anticolonialism, 51, 145, 151–2
arte de la lengua, 29
Aryan, 77, 82, 84–7, 89, 92
Atuahalpa, 25
Babel, 7, 12, 27, 53, 64, 69, 104, 112,
126
Balai Pustaka, 141–2
Bantu, 87, 89, 90
barbarian, 24, 26–7, 34, 131
Batavia, 136, 145
Belgian colonialism, 119, 133, 139,
146–7
Belgian Congo, 119, 123, 124
Belgium, 16, 119, 128
Bengal, 49, 56, 103, 152
Besonnenheit, 51–4
Bible, 49, 64, 77, 90, 96–7, 113, 121,
141, 153, 156
Brahman, 90
196 Subject Index
Conquistadores, 25, 27, 30, 32, 100,
158
consonants, 35–7, 47, 54, 61, 78–80
conversion, 22–4, 28, 29, 34, 38–41,
45, 46, 94–6, 106, 120–1, 125,
150, 155
Council of Tours, 19
Council of Trent, 42
Crusades, 25
culture, 52–3, 65–6, 71, 75–6, 78, 87,
92
cultures of standardization, 108–9,
117, 120
Darwinism, 71, 83, 115, 117, 130,
142, 146, 165
descent, 59, 61–2, 64, 68, 83, 115
descriptive linguistics, 156
dialect, 10, 29, 39, 108–9, 112,
115–16, 119, 128, 133, 136,
140–2, 162, 167–8
dictionaries, 3, 4, 6, 57, 79, 101–2,
139
diglossia, 143–5
Dutch colonialism, 65, 124–5, 128,
132, 135–47
See also Netherlands East Indies
Dutch Missionary Society (NZG),
106
Dutch Reformed Church, 111
education, 75, 94, 102, 107, 125,
151, 160
See also school; scholarization
Egypt, 104
Endangered Language Fund, 165
ethnic languages, 113, 141, 152
ethnology, 49, 64
ethnopolitics, 152, 159, 161, 170
etymology, 57, 61, 73
Europe, 5, 6, 11, 13–15, 22, 57, 69,
71–3, 76, 85–7, 91–2, 94, 100–3,
105, 107, 116, 123–4, 126, 128,
143
evangelization, 107
evolution, 71–2, 82–3, 85, 91, 93,
116, 119, 130, 149, 151, 155
faith, 12–13, 18, 20, 28–9, 31, 38–9,
44–5, 88, 94–100, 102, 106, 107,
109, 112, 114, 120
Formosa, 126
France, 50, 57, 75, 79, 89, 92, 100,
102, 107, 124
Francophonie, 128
French colonialism, 85, 100, 102,
124, 128
German Evangelist Missionary
Societies, 127
Germany, 15, 50, 66, 71, 72,
76–7, 79–83, 86, 89–90, 91,
98, 101, 105, 115, 121, 126–8,
147
globalization, 17, 147, 150, 162–3,
166
glottal stop, 33–4, 37, 45, 47
grammar, 18, 28–33, 37, 54, 57,
65–6, 68, 73, 75–82, 89–90, 92,
98, 104, 111, 135, 137, 140
Granada, 25
Greece, 27, 54, 75–8
Grimm’s law, 79
Guatemala, 150, 153–4, 157, 159,
161
Hamitic hypothesis, 87–8
history, 1, 2, 14–15, 50–1, 58–9, 62,
70–1, 73, 75–6, 78–86, 89–91,
93, 108, 114, 130, 141
Holy Roman Empire, 50, 72
human origins, 14, 49, 64, 69
hybridity, 138, 165
hyperbolic valorization, 166
Subject Index 197
Iberia, 21
Iceland, 21
identity, 1, 10, 15, 17, 50, 98, 101,
109, 112–13, 116–17, 124,
126–8, 135, 138, 145–6, 150,
152, 158, 160, 162–4, 168
ideology of the natural, 119, 127
imagined communities, 107, 145
Imerina, 100–3
imperialism, 49, 67, 123–4, 128–9
India, 6, 15, 58, 60, 65–7, 72, 87, 90,
indigenism, 162, 164
indios, 22, 25–30, 34, 42
Indo-European, 61–2, 71, 78, 82, 85,
89
Indo-Germanic, 78
Indonesia, 7, 17
inflection, 76, 78, 81
Instituto Indigenista Nacional, 157
intelligibility, 27
Islam, 58, 67–8, 85, 97, 133, 135, 142
Ivory Coast, 93
Japan, 135
Java, 35, 49, 59, 65–8, 136, 137
Jews, 25
Johor, 140
Kenya, 17, 122, 170
language communities, 49, 51, 53,
54, 61, 62, 64, 66, 95–7, 107–16
language death, 82, 162, 166
language development, 53, 129,
131
language diversity, 9, 13–15, 16,
48–51, 53, 55, 64, 68–9, 72, 80,
85, 86–9, 90, 109, 113, 115, 126,
149, 151, 153, 159, 165–6, 169
language history, 48–9, 51, 53, 61,
130
language map, 85–6, 117–19
language of wider communication,
39, 112, 119, 124, 129, 132–6,
142–3, 147, 151
language origins, 49–53, 56, 74,
80–1, 83, 113
language rights, 163–4, 166
language spread, 123, 126, 132,
146–7
language systems, 6, 7, 49, 155,
169
language territory, 16, 101, 108,
115–17, 145
language tree, 61, 81, 83
linguistic organism, 53, 70, 73–4,
78, 81–2, 88, 155, 169
literacy, see practices of literacy
literature, 49, 57, 59, 65–8, 74, 75,
86, 89
London Missionary Society, 94,
100, 106, 112
Lutheran, 90
Luzon, 34–5, 41–2, 66
Madagascar, 66, 100–1, 106, 126,
151
Malaya, 139
Manila, 41–2
meaning, 27, 29, 35, 38, 40, 43–4,
46, 97, 99, 101–2, 120, 130
Melaka, 135
Mendeleev, 166
Mexico, 6–7, 13, 23, 25, 27, 29,
38, 39, 41–2, 45, 55, 100, 124,
143
Middle East, 90
missionaries, 13–14, 16, 29–30,
35–46, 48, 90–1, 93–121, 129,
133, 135, 150, 152–3, 156–7
Mohammadanism, see Islam
Moors, 25
Moravians, 93, 96
Muslims, see Islam
198 Subject Index
Napoleonic Wars, 65, 72
national language, 16–17, 108, 124,
126, 128–9, 132, 135, 145,
151–2, 170
nationalism, 15, 51, 55, 57, 69, 71–2,
75–6, 80–1, 88–90, 91, 94, 97,
106–9, 111, 113, 116, 119–20,
126, 132–5, 144, 146
native speakers, 16, 22, 37–8, 88,
122, 129, 150, 156, 159–60,
167
natural history, 73, 130
nature, 48–9, 52–3
Netherlands, 16, 97, 123, 128, 138,
140–1
Netherlands East Indies (NEI),
123–5, 129, 133, 135, 138–9,
141, 143–4, 146
New France, 23
New Guinea, 126
New Imperialism, 123–4, 126
New Spain, 23
New Testament, 27, 57, 154
New World, 20, 22–5, 41–2, 89,
157
New Zealand, 98–9, 102, 106
newspapers, 137
Norddeutsche Missiongesellschaft,
106
North America, 23, 76, 155, 160
North Borneo, 126
Norway, 21
Nubia, 104
Old Testament, 64
organic metaphor, 7, 15, 49, 51, 53,
57, 71, 73, 97–8, 146, 163, 169
orientalism, 15, 57, 91
orientalists, 56, 68, 105, 128
orthographies, 4, 8–9, 18, 33, 35–7,
42–5, 55, 65, 102–5, 107, 111–12,
140–2, 152, 157–9, 165–8
Pacific Languages Development
Project, 167–8
Paris Missionary Society, 114
Pentecost, 12, 93, 95, 120
Peru, 25, 27, 28, 38, 39
Philippines, 6, 13, 20, 23, 31, 41–2,
46
philology, 14, 58, 66, 68, 108–9,
113, 115–17, 125, 130, 133, 137,
139, 140–1, 143, 146, 155, 168–9
pidgin, 93, 95–7, 121, 133, 136, 148
Portugese, 23, 41, 115, 135–6
postcolonialism, 13, 151–3, 157,
160–1, 165–70
practices of literacy, 5–6, 9, 11–13,
16–20, 23, 25, 29–32, 35, 37, 40,
42–5, 54, 58, 93–4, 96, 98–101,
104, 106–7, 109, 112, 116, 120,
125, 129, 132–3, 137, 141,
143–4, 150, 153, 157–62, 165–8
preterrain, 94, 119
primitive, 23, 38, 41, 42, 45, 55, 66,
68, 95, 98
print literacy, 14, 16–17, 107–9, 114,
116–17, 120, 129, 132, 133, 137,
141–3, 145–6, 152
progress, 14, 69, 129–32, 142, 146,
160
progress in language, 130, 146, 151
Protestantism, 13, 64, 89, 95–6, 99,
100, 115
Proyecto Lingüistico Francisco
Marroquín (PLFM) 159–61
Prussia, 70, 72, 74, 76, 80, 90, 91,
104, 126, 127
Puerto Rico, 20
Reformation, 13
Requerimiento, 25–6
Rhodesia, see Zimbabwe
Riau, 139, 140
Russia, 57, 77
Subject Index 199
saltillo, 23, 33–4, 37, 38, 40–1, 45
See also glottal stop
Samoa, 126
scholarization, 100, 112, 124, 129,
144, 151
school, 96, 98–9, 101–2, 111–12,
132, 142–6
Semitic, 57, 78, 87
Senegambia, 85
Singapore, 65, 140
social imaginaries, 11, 17, 31, 45,
48, 58, 68–9, 93, 102, 106–7,
109, 120, 126, 146, 152
sociobiology, 91, 165
sociolinguistics, 143
sound shift, 79
South Africa, 90, 127, 170
South America, 23
South Asia, 84–7, 124
Spain, 41–2, 153–4, 157–8, 160
Spanish colonialism, 25–6, 27, 30,
41–2
Sprachwissenschaft, 70
Sri Lanka, 35
Straits of Malacca, 135, 139
strategies of selection, 9–10, 23,
38–9, 84, 94, 107, 120, 125,
133, 161
structural linguistics, 131, 155
style, 10, 34, 39, 107, 136, 141
subalterns, 124–5, 137, 145–6
Summer Institute of Linguistics
(SIL), 150, 153–60, 162
Swiss, 98, 105, 113, 115
syllable, 35–7, 78
syllabary, 37
Tanzania, 17, 170
Tasmania, 131
texts, 18, 31, 44, 49, 57, 58, 59,
60–2, 64–8, 70, 81, 97, 102–7,
140
translation, 19, 36, 39–40, 42, 46,
57, 67, 96, 97, 99, 101, 103, 108,
111, 113, 141, 153
Transvaal, 109, 113
Truth language, 31, 40, 90, 143
United States, 57, 123, 133
Valladolid, 28
vowels, 35–7, 54, 55, 121, 167
West Indies, 20
word roots, 54, 61, 73, 76, 78
Zanzibar, 133
Zimbabwe, 109, 112, 113
zones of contact, 2–3, 5, 10–11, 20,
22, 23, 31, 38, 83, 91, 125–6,
135, 136, 137, 161, 165–6, 168,
169–70
