Tyranny of Writing: Ideologies of the Written Word 9781474292467, 9781474292436, 9781474292450

Citation preview

Tyranny of Writing

Also available in this series Discourses of Endangerment Language Ideologies and Media Discourse Language and Power Remix Multilingualism The Sociolinguistics of Identity The Language of Newspapers

Tyranny of Writing Ideologies of the Written Word Edited by Constanze Weth and Kasper Juffermans

BLOOMSBURY ACADEMIC Bloomsbury Publishing Plc 50 Bedford Square, London, WC1B 3DP, UK 1385 Broadway, New York, NY 10018, USA BLOOMSBURY, BLOOMSBURY ACADEMIC and the Diana logo are trademarks of Bloomsbury Publishing Plc First published 2018 Paperback edition first published 2019 Copyright © Constanze Weth and Kasper Juffermans, 2018 Constanze Weth and Kasper Juffermans have asserted their right under the Copyright, Designs and Patents Act, 1988, to be identified as Editors of this work. Cover illustration © Martin O’Neill All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. Bloomsbury Publishing Plc does not have any control over, or responsibility for, any third-party websites referred to or in this book. All internet addresses given in this book were correct at the time of going to press. The author and publisher regret any inconvenience caused if addresses have changed or sites have ceased to exist, but can accept no responsibility for any such changes. A catalogue record for this book is available from the British Library. A catalog record for this book is available from the Library of Congress. ISBN: HB: 978-1-4742-9246-7 PB: 978-1-3501-2311-3 ePDF: 978-1-4742-9245-0 ePub: 978-1-4742-9244-3 Series: Series: Advances in Sociolinguistics Typeset by Deanta Global Publishing Services, Chennai, India To find out more about our authors and books visit www.bloomsbury.com and sign up for our newsletters.

Contents Preface List of Contributors Introduction: The Tyranny of Writing in Language and Society Constanze Weth and Kasper Juffermans

vi ix

1

1

Revisiting the ‘Tyranny of Writing’ Florian Coulmas

19

2

How to Write a Birch-Bark Letter: Vernacular Orthography in Medieval Novgorod Daniel Bunčić

31

The End of the Standard Language: The Rise and Fall of a European Language Culture Joop van der Horst

47

The Tyranny of Orthography: Multilingualism and Frenchification at Primary Schools in Late-Nineteenth-Century France Manuela Böhm

63

Ideologies of Language and Literacy in the German Educational Reform Movement at the End of the Long Nineteenth Century Ulrich Mehlem

81

When God is a Linguist: Missionary Orthographies as a Site of Social Differentiation and the Technology of Location Ashraf Abdelhay, Busi Makoni and Sinfree Makoni

97

3 4 5 6

7 8 9

Standard English, Cricket, Nationalism and Tyrannies of Writing in Sri Lanka Harshana Rambukwella

113

Escaping the Tyranny of Writing: West African Regimes of Writing as a Model for Multilingual Literacy Friederike Lüpke

129

Writing Chinese: A Challenge for Cantonese-L1 and South Asian Hongkongers David C. S. Li

149

10 Fangyan and The Linguistic Landscapes of Authenticity: Normativity and Innovativity of Writing in Globalizing China Xuan Wang

165

11 Dialect Authenticity Upside Down: Brabantish Writing Practices of a Black Comedian on Twitter Jos Swanenberg

183

12 Salty Politics and Linguistics in the Balearic Islands: Tracing a Non-Standard Iconization in Metalinguistic Facebook Communities Lucas Duane

197

Index

213

Preface This book is the result of a discussion on writing and tyranny initiated by the two editors in 2013 at the University of Luxembourg. We then had both recently joined the university, as associate professor and postdoc, respectively – and had both worked on literacy in our former academic environments. Coming from Freiburg, Constanze had worked on literacy practices in Moroccan and Occitan speaking families in France and studied orthographic practices of primary school children, primarily from a structural linguistic perspective. And coming from Tilburg and Hamburg, Kasper had studied literacy and multilingualism in The Gambia, West Africa, and digital literacy practices of Chinese-Dutch youth, from a linguistic anthropological perspective. We had much in common, but at least as much to disagree over … and to discover. So we began to (re)read and discuss classic texts in literacy studies. Among the texts we discussed, Florian Coulmas’s Writing in Society (2013) gave us inspiration for an exploratory workshop to widen the scope of our discussions. The opening chapter of that book critically revisited Saussure’s and Bloomfield’s arguments against writing and  the subsequent erasure of literacy as an object of study in the new science of language they founded. We titled that first workshop Beyond the Tyranny of Writing, and invited colleagues from our respective networks to our Walferdange campus in May 2014 to explore uses of the metaphor of tyranny of writing with Florian as our central guest. Drawing on a diverse range of contemporary and historical research contexts in Europe, Africa and Asia, all the workshop contributors tried to make sense of the potential meanings of tyranny of writing following, but certainly also moving beyond Saussure’s initial formulation revisited by Coulmas. Shortly after this workshop, we approached Advances in Sociolinguistics series editor Tommaso Milani with our plan for a collective volume, and asked our contributors to write up their contributions for the workshop. Most workshop participants eventually contributed a chapter to the present book. As we commented on their papers and organized the peer review process, we began to realize that the structure of inviting presentations and chapters, and the process of editing a book inevitably carries the risk to become tyrannical, even if carried out with the most constructive intentions. We thank those among our contributors who pointed this out to us as well as those who recognized our efforts to avoid tyranny in our editorial work. Some contributors used the metaphor productively; others maintained more reservations and, for good reasons, resisted to too freely applying the term in relation to their work. Whether applying the notion of ‘tyranny of writing’ explicitly or not, the individual chapters explore tensions and dynamics between writing and speech, standard and vernacular, center and periphery, normativity and authenticity in and around the written word.

Preface

vii

After receiving the final green light from Bloomsbury about our book project – i.e. after revising and reshaping our initial book proposal following critical but constructive feedback – we invited our participants to Luxembourg again for a second workshop, this time in Belval. We met in October 2016 to discuss advanced paper drafts and to build coherence across the papers. We also invited an additional contribution, to expand the book’s geographical scope (to Sri Lanka) and to make up for the two or three planned chapters that, for various reasons, were withdrawn. Editorial tyranny returned in yet another form when we drafted the introduction – without any doubt the longest and most debated jointly written 15 pages in both of our relatively young careers. In the first place by ourselves: the introduction as it reads now is the result of unrelenting rewriting and unwriting (a euphemism for deleting!) of entire paragraphs and whole pages we drafted independently of, but for, each other. Consequently, what survived Occam’s razor is truly jointly authored. Our contributors also helped to shape and reshape this text, in mostly friendly-toned oral and written suggestions, but oftentimes with compelling arguments we had to obey. The book project also greatly benefited from critical feedback and comments of external referees. If the introduction and the book as a whole are judged to be good, it is because of that collective effort. If not, then it is because of our writing and editing skills alone. This book is entirely written and edited by multilingual writers of English. None of us are typical native speaker and writers of English. Our workshops were pleasantly multilingual, but the final book is in English only, or so at first sight it appears. The chapters’ authors, however, have used different strategies to escape the tyranny of monolingualism and managed to smuggle in, at various places, smaller or larger chunks of German, French, Catalan, Brabantish-Dutch, Novgorodian-Russian, Mandarin, Cantonese, Chinese fangyan, Sinhala, Wolof, Jóola and other languages. The book is also more multilingual than it appears since at least three chapters were written by authors who do not normally write in English. Two chapters are even translations of drafts written in the authors’ respective first academic languages. And many of the reference lists too are a lot more multilingual than usual in contemporary academic publishing. Alstublieft. Finally, we owe a word of thanks to everyone who supported our writing and editing of this book. We are grateful in the first place for the generous working environment the University of Luxembourg provided. We received specific financial support for the two workshops as well as for marginal publication costs, by the Research Priority of Multilingualism and Diversity, the Education, Culture, Cognition and Society (ECCS) research unit, and the Institute for Research on Multilingualism (MLing). Our colleagues at MLing, and beyond, produced a pleasant intellectual climate to work in and participated at various moments, directly or indirectly, in our discussions on tyranny of writing and so helped to shape this book in multiple ways. The project as a whole also greatly benefited from the support of MLing’s excellent secretary Andrea Hake who made the organization of the two workshops as smooth and pleasant as we could possibly have wished.

viii

Preface

The experience to publish with Bloomsbury was from start to finish pleasant and highly professional, and this to the credit of the series’ editor in Johannesburg/ Gothenburg, Bloomsbury’s linguistics team Gurdeep Mattu, Andrew Wardell and Helen Saunders in London, and in the crucial final stage, junior project manager Sweda in Chennai, who skillfully and untyrannically guided us through multiple rounds of proofs. To all of them we say villmools merci to have made us accomplish this book. Constanze Weth and Kasper Juffermans November 2017, Metz (France) and Essen (Belgium)

List of Contributors Ashraf Abdelhay Linguistics and Arabic Lexicography Doha Institute for Graduate Studies Doha, Qatar Manuela Böhm German Department Karlsruhe University of Education Karlsruhe, Germany Daniel Bunčić Slavic Department University of Cologne Cologne, Germany Florian Coulmas Institute of East Asian Studies University of Duisburg-Essen Duisburg, Germany Lucas Duane Institute of English Studies University of Luxembourg Esch-Belval, Luxembourg & Arts and Humanities Department Universitat Oberta de Catalunya Barcelona, Spain Joop van der Horst Faculty of Arts University of Leuven Leuven, Belgium Kasper Juffermans formerly: Institute for Research on Multilingualism University of Luxembourg Esch-Belval, Luxembourg & African Studies Ghent University Ghent, Belgium

David C. S. Li Department of Chinese and Bilingual Studies Hong Kong Polytechnic University Hong Kong Friederike Lüpke Africa Section School of Languages, Cultures and Linguistics SOAS, University of London London, United Kingdom Busi Makoni (née Dube) African Studies Pennsylvania State University Pennsylvania, United States Sinfree Makoni Department of Applied Linguistics Pennsylvania State University Pennsylvania, United States & Department of English Studies University of South Africa Pretoria, South Africa Ulrich Mehlem Faculty of Education Goethe University Frankfurt Frankfurt, Germany Harshana Rambukwella Postgraduate Institute of English Open University of Sri Lanka Colombo, Sri Lanka & School of English The University of Hong Kong Hong Kong

x Jos Swanenberg Department of Culture Studies Tilburg University Tilburg, The Netherlands Xuan Wang School of Modern Languages

List of Contributors Cardiff University Wales, United Kingdom Constanze Weth Institute for Research on Multilingualism University of Luxembourg Esch-Belval, Luxembourg

Introduction: The Tyranny of Writing in Language and Society Constanze Weth and Kasper Juffermans

What could possibly be tyrannical about writing? This book is an attempt to make sense of the written word and its powerful role in society. By using the word ‘tyranny’ we take an explicitly critical stance towards writing, but we also join our voices to numerous essays, books and other texts inspired by this metaphor. This book is inspired above all by Saussure’s argument against writing as an object of linguistic research and what he called la tyrannie de la lettre. He denounced writing as an imperfect, distorted image of speech that obscures our view of language and its structure. In this introduction to The Tyranny of Writing: Ideologies of the Written Word, we discuss the ‘tyranny of writing’ as a critical metaphor for sociolinguistics. The idea of the ‘tyranny of writing’ serves as a heuristic for exploring ideologies of language and literacy in culture and society as well as the tensions and contradictions between the written and the spoken word, linguistic normativity, creativity and authenticity, and centres and peripheries in language practices. This introduction first discusses the unique potential of writing and its risks. It then takes up Florian Coulmas’s (2013) revisiting of Saussure’s and Bloomfield’s arguments against writing. It proceeds to show that writing can be critiqued by and of itself as well as on ideological grounds as a tool used by human beings. We then turn to the critique offered by New Literacy Studies of the ‘great divide’ between literacy and orality, discussing a number of theories that conceive of the differences between written and oral language as continua. The following section probes into the metaphor of tyranny (as oppressive rule) and suggests that ‘tyranny’ is a powerful, if unwieldy, concept to work with. The introduction concludes with laying out the structure of the book and presenting the individual chapters’ contributions to the discussion of the ‘tyranny of writing’ in language and society.

The potential of writing In The Decline of the West, Oswald Spengler noted that ‘writing ... implies a complete change in the relations of man’s waking-consciousness, in that it liberates it from the tyranny of the present ... the activity of writing and reading is infinitely more abstract than that of speaking and hearing’ (cited in Goody and Watt 1963: 330). The invention

2

Tyranny of Writing: Ideologies of the Written Word

of writing, at what seem to be independent places and times in history (Mesopotamia 3200 BCE, China 1600 BCE and Mesoamerica 900 BCE), always stood at the cradle of powerful civilizations. It is commonly suggested that writing has been instrumental in the development of complex societies with some degree of urbanization, centralization and stratification. Indeed, it is impossible to imagine modern life without writing. Without falling into essentializing and racialist dichotomies between ‘civilized’ and ‘primitive’ people, it seems fair to say, with Coulmas, that writing is ‘our most consequential invention’ (2013: ix) and has helped in the history of emancipation from the tyranny of nature (Chapter 1). Writing made it possible to keep records, to do more complex calculations, to register information and so on and thereby brought our species to a level of development that would most probably have been impossible without this innovation. In their well-known article Jack Goody and Ian Watt (1963) argue that the consequences of literacy include the possibility of abstract, analytical and logical thinking; a specialization and compartmentalization of knowledge; the possibility of chronological ordering and historical inquiry based on written records; and even the development of political democracy (which required citizens to read the laws). Goody and Watt (1963: 315) maintain that ‘the notion of representing sound by a graphic symbol is itself so stupefying a leap of the imagination that what is remarkable is not so much that it happened relatively late in human history, but rather that it ever happened at all’. Modern writing systems – our alphabets, abdjads, abugidas, syllabaries and character systems – make it possible to write and read anything we can talk about, and more. For Roland Barthes (1977), what distinguishes writing from speech is that writing separates the text from the author. The modern text is ‘made and read in such a way that at all its levels the author is absent’ (145). Once a text is ready to be read, the author, Barthes provocatively claims, ‘enters into his own death’ (142). Barthes’s article criticized the importance contemporary literary and art critics assigned to the intentions of the author or artist. The image of literature in ordinary culture, Barthes complained, ‘is tyrannically centred on the author, his person, his life, his tastes, his passions’ (143). This reduces the work of reading and interpretation to a conforming deciphering, to a discovering of the author’s mind or society beneath the work. Because the temporality of writing and reading are different – any text is read after its writing – there is no fixed meaning in the text, no authority of the author over the reader’s associations with the text. Because the author (‘the instance writing’) is removed from the text when it is read, ‘it is language which speaks, not the author; to write is, through a prerequisite impersonality …, to reach that point where only language acts, “performs”, and not “me”’ (143). The freedom to read a text (or look at a painting, or listen to a piece of music) as it can be perceived rather than as it was conceived constitutes the pleasure of reading and makes writing/reading a liberating, revolutionary activity. Because in writing language can act or perform in the absence of its author, the ability to read and write is a powerful cultural achievement and personal skill. In early history, writing had a divine character in society. As writing was beyond the reach of the masses and restricted to a small class, often priests, ‘writing was a mystery, a secret code in the literal sense of the word’ (Coulmas 1989: 5). On the individual level

Introduction: The Tyranny of Writing in Language and Society

3

as well, writing remains mysterious until the learner is initiated into this sign system. In his biography, Nobel Prize–winning author Elias Canetti (1977) describes how, as a preschooler, he was fascinated by his father reading the newspaper. He tried to understand what his father was so captivated by. First he assumed it was the smell, and then he thought it was the sensation of moving his head alongside the paper. It was only later and with the help of his father that he became aware of the written sign system. After several generations of near-universal schooling in the developed world and considerable progress in achieving Millennium Development Goal 2 (universal primary education) in the developing world, literacy is still very unequally distributed in the world. Some countries boast near-universal literacy skills in their population, while in other countries whole segments of the population are unable to read and write. Of course, even within the same society or community some are evidently more literate than others. Functional illiteracy remains a problem even in the most developed nations (cf. Grotlüschen and Riekmann 2012), and this is not only a result of migration and increasing diversities. However, definitions of basic literacy and its functionality vary across place and time. What counts as literacy in a given community depends on its social, economic, political and religious structures. Some languages are also written more than others. More than anything else, it is writing that effectively distinguishes one language from another and that distinguishes hierarchically between languages and dialects. In Heinz Kloss’s (1967) terms, writing systems or orthographies are key tools for ‘language ausbau’, for innovative language planning or the strategic cultivation and maintenance of difference. The reason why so much time, money and effort has been invested in promoting literacy globally over the last 150 years at least, it seems, is that we – as individuals and social groups – have high expectations of its potential for both societal and personal development. Collectively, we seem to believe rather strongly that literacy empowers and are therefore willing to contest the privilege of the literate elite. The notion of literacy is most explicitly linked with emancipation in the work of the Brazilian educator Paulo Freire (see Freire and Macedo 1987). As much about the world as about the word, Freire’s notion of literacy is rooted in a radical pedagogical agenda and political project that seeks to enable (illiterate) people to participate in the understanding and transformation of their society and to emancipate them from the oppression of the (literate) ruling class. Becoming literate, in Freire’s sense, is more than the mechanical acquisition of reading and writing; it is the acquisition of a critical historical and social consciousness and of a voice that is able to act against oppression and tyranny. In short, critical literacy is about the rewriting or reinvention of history and the future. It is, in Henry Giroux’s (1987: 6) words, ‘an important starting point’ for all of us – not just the poor or subaltern – ‘to reclaim the authorship of (our) own lives’.

The ‘tyranny of writing’ in modern linguistics Despite the massive positive consequences and the potential of literacy, many authors have critiqued writing. Plato already warned against the negative consequences of the

4

Tyranny of Writing: Ideologies of the Written Word

medium of writing for our memory and the pursuit of wisdom. In the Phaedrus, he lets Socrates recount a myth of two Egyptian Gods reflecting on their invention of writing, saying: If men learn this, it will implant forgetfulness in their souls; they will cease to exercise memory because they rely on that which is written, calling things to remembrance no longer from within themselves, but by means of external marks. What you have discovered is a recipe not for memory, but for reminder. And it is no true wisdom that you offer your disciples, but only its semblance, for by telling them of many things without teaching them you will make them seem to know much, while for the most part they know nothing, and as men filled, not with wisdom, but with the conceit of wisdom, they will be a burden to their fellows. (Plato 1952 [278 BCE]: 157; as cited in Goody and Watt 1963: 327)

In what may very well be the first critique of writing, Socrates warns that writing deceives us since it does not create true wisdom in our minds, but only a false wisdom. Writing makes us lazy and relieves us from the effort of memorizing and internalizing knowledge, because we can now access that knowledge outside of ourselves. It appears that in our digital era this critique is more pertinent than ever. In this section we will engage with two more criticisms. We take up Coulmas’s (2013) revisiting of Saussure’s (1916) and Bloomfield’s (1933) arguments against writing. These two fathers of modern linguistics turned away from comparative and historical philology, until then the dominant brand of linguistics, and turned instead towards the study of spoken language as natural human faculty. In order to attain scientific status, the study of language had to abstract from concrete manifestations and variation in language use (parole) and had to focus on the language system (langue). The starting point for both authors was that writing should represent speech and not the other way around. Writing, Ferdinand de Saussure argued, is an imperfect, distorted image of speech and obscures our view of language and its structure. This is what Saussure called the ‘tyranny of writing’ (la tyrannie de la lettre): the fact that linguists until then (i.e. philologists) had studied language only indirectly through writing. His second concern related to the fact that spelling influences and modifies language through ‘pronunciation’. As the wording in the Cours de linguistique générale (CLG) expresses annoyance and despair rather than objective reasoning, we cite Saussure at length, in translation: Grammarians are desperately eager to draw attention to the written form. Psychologically, this is quite understandable, but the consequences are unfortunate. The use acquired by the words ‘pronounce’ and ‘pronunciation’ confirms this abuse and reverse the true relationship obtaining between writing and the language.  … But the tyranny of the written form extends further yet. Its influence on the linguistic community may be strong enough to affect and modify the language itself. That happens only in highly literate communities, where written documents are of considerable importance. In these cases, the written form may give rise to erroneous pronunciations. The phenomenon is strictly pathological. … Probably such misunderstandings will become more and more frequent. More and more

Introduction: The Tyranny of Writing in Language and Society

5

dead letters will be resuscitated in pronunciation. In Paris, one already hears sept femmes (‘seven women’) with the t pronounced. [The French philologist Arsène] Darmesteter foresees the day when even the two final letters of vingt (‘twenty’) will be pronounced: a genuine orthographic monstrosity. (Saussure [1916] 2013: 34–6)

It should be noted in passing that sept in contemporary French is indeed generally pronounced with its final letter. The presumed tyranny of orthography not representing the spoken reality of the language has highly influenced linguistics. The new structural linguistics Saussure founded in Geneva was no longer ‘slavishly subservient to the written language’ (p. 2) but was instead concerned with ‘an amorphous object that is difficult to grasp’ (p. 55), that ‘shapeless and unmanageable mass’ of speech as it reads in another translation. Saussure expressed himself even more critically against writing and French orthography in particular in his recently published Writings in General Linguistics (2006), which contains his own lecture notes and an unfinished book manuscript he was working on at the end of his life. His editors Charles Bally and Albert Sechehaye had in vain searched for these writings before resorting to the course notes they managed to recover from eight of Saussure’s students to compile the CLG. Assumed lost or inexistent, these writings were miraculously recovered from a drawer at the orangery of the Saussure family home in Geneva in 1999, eighty-three years after Ferdinand’s death. In that book, a section entitled ‘Second lecture at the University of Geneva, November 1891’ (Saussure [1891] 2006) describes the principles of continuity and mutability of language. The text begins by observing that the language of today both is and is not the same language as yesterday. It is the same, but different. Saussure compares this with an art exhibition of photographic portraits of the same person in the same identical pose taken over twenty years: Any two photos depict the same person, yet that person is different on every picture. Change is an absolute principle in language: ‘in reality there is never equilibrium, or a permanent, stable point in any language’ (p. 105). Writing, however, produces a false idea of stability and thereby partially inhibits and prohibits variation and change. It is in this context that Saussure writes of the ‘tyranny of the written language’: The ceaseless transformation of language constitutes an absolute principle. A case of an idiom in a static state of rest does not occur. Indeed, the force which creates this movement appears so impossible to bridle or control that languages like ours, whose life has become almost completely artificial, are themselves obliged to give way to it; the tyranny of the written language, this stifling corset of official French, certainly weakens this process, but is incapable of stopping the force completely, and we often little suspect the distance which real language (by which I mean the language of educated conversation) has already covered as it works away ceaselessly underground, under the ‘solidified’ surface, so to speak, of classical French. (Saussure [1891] 2006: 105)

On the other side of the Atlantic, Leonard Bloomfield (1933) presented structural arguments against writing as well. Principally interested in language structure and

6

Tyranny of Writing: Ideologies of the Written Word

sound change, Bloomfield considered writing an ‘external device’ for linguistic analysis, ‘like the use of a phonograph’ (p. 282), and argued that ‘writing is not language, but merely a way of recording language by means of visible marks’ (p. 21). He deplored, however, that ‘the conventions of writing are a poor guide’ for representing phonemes. Observing the constant change in spoken language, he perceived the stability of writing as a ‘handicap’ (p. 21) inasmuch as ‘written records’ are ‘misleading’ (p. 481) and give ‘an imperfect and often distorted picture of past speech’ (p. 293) and thus a false impression of stability. Saussure and Bloomfield not only founded a radically new (structural) linguistics, but also effectively removed writing from the agenda of linguistics. Until today, the priority of spoken over written language is generally accepted in linguistics. The main textbooks and standard reference works do not mention writing at all (Sebba 2007; Coulmas 2013). Nevertheless, Saussure’s and Bloomfield’s arguments for liberating linguistics from the ‘tyranny of writing’ contain some inconsistencies, as Coulmas and others have noted. Robert de Beaugrande for instance, remarked that while Bloomfield found writing to be misleading for the study of spoken language, he still listed phonemes in alphabetical order. Following Bloomfield’s own argument, this ‘ought to be a category mistake’ (Beaugrande 2008: 69). Saussure’s logic seemed to be similarly confused as he complained about the ‘instability’ and ‘deformations’ of French orthography but praised the ‘stability’ of the Greek alphabet, where each graphic sign is neatly mapped onto an acoustic sign and vice versa (Saussure [1916] 2013: 31). A careful reading of Saussure and Bloomfield reveals that they not only are concerned with the different structures of spoken and written language, but also hold truly consistent reservations against the norms of ‘literary’ or ‘cultivated’ language. The separation of both is a delicate matter since the linguistic norm is often fixed in writing. Given the diffusion of literacy in society, speech no longer exists in isolation of writing. In societies that are saturated with literacy, Bloomfield (1927) notes, writing gives rise to images of ‘literate’ and ‘illiterate speech’: The popular explanation of incorrect language is simply the explanation of incorrect writing, taken over, part and parcel, to serve as an explanation of incorrect speech. It is the writing of every word for which a single form is fixed and all others are obviously wrong. (Bloomfield 1927: 433, emphasis in original)

Literate speech, in either the written or spoken medium, involves a long process to acquire and is difficult to master. It is furthermore a risky adventure and requires the highest level of concentration. In everyday language, all speaking, good or bad, is careless; only for a few minutes at a time can one speak ‘carefully’, and when one does so, the result is by no means pleasing. In fatiguing effect and in ungracefulness, ‘careful’ speaking is like walking a chalk-line or a tight-rope. (Bloomfield 1927: 432)

It is in this way that writing can be tyrannical: It influences processes of language change and creates social distinctions between ‘literate’ and ‘illiterate’ speech, correct and incorrect forms, and ignorant and knowledgeable speakers and writers.

Introduction: The Tyranny of Writing in Language and Society

7

A double agency One could very well disagree with Barthes’s reflection on the power of writing and argue in response to Saussure and Bloomfield that writing is only a technology and thus incapable of acting in and of itself. Writing does not produce the effect of power ‘autonomously’ but ‘ideologically’, to borrow from the late Brian Street (1995). It is, of course, human beings that individually and collectively use or abuse writing in making social distinctions, intentionally and consciously or otherwise. As true as that may be, it is of course analogous to the key argument that Second Amendment supporters in the United States naively make when defending the right of the people to keep and bear arms, as in the NRA bumper-sticker slogan ‘Guns don’t kill people, people do’. It is far too simplistic to ascribe agency exclusively to either technology or human beings. It is not guns or people but rather people with guns that kill. Writing similarly acts on people, even if people act with writing. Agency is not an exclusively human affair. Best known for theorizing non-human agency is the philosopher Bruno Latour. A key argument of Latour’s actor-network theory is that scientific discoveries or technological inventions (oftentimes the same thing) are made possible by a collaboration of human and non-human (material) agents. When and if it happens, a discovery or invention only happens when the material world is ready for it. The discovery or invention will then change the material world in ways that go beyond the intentions or even imagination of its discovererinventor. In We Have Never Been Modern, Latour (1991) describes that Robert Boyle’s gas laws are as much to the credit of the air pump he struggled to make as to his genius. Agency is located not in the inventor or the technology but in their networked relation. Just as humans find agency in their capacity to interact with technology, technologies have agency in their capacity to interact with humans. In its technological agency, writing merely represents language visually and thereby necessarily offers a new way of looking at language. In interaction with humans, writing does a lot more than represent speech visually. The technological agency of writing turns out to be not a transcription system and memory record but a tool for solving social problems. Writing did, indeed, solve the problem of human cooperation in large groups and hence substantially influenced reciprocal communication, social norms and their enforcement as well as group identity (Mullins et al. 2013). However, writing did not have the same effect in all societies (see Kulick and Stroud 1990, for a classic example). Writing systems are always embedded in particular historical and social contexts and tied up with specific cultural and social practices. Written codification is associated with correctness, education and other distinctions. These characteristics are, however, not as closely related to writing as such but to a particular view of education: schooled literacy (Cook-Gumperz 1986). This is the ideology of a universalist literacy central to formal education. Moreover, it became the official norm of writing and language use in general. We need only to recall Pierre Bourdieu’s analysis of the relations between institutional expectations and student language: Types of knowledge and ways of speaking and writing are forms of ‘cultural capital’, unequally distributed between social classes and legitimizing the ‘reproduction of social class’.

8

Tyranny of Writing: Ideologies of the Written Word

Schools are mediating institutions between culture and economy, because ‘cultural capital’ – such as abstract reasoning, complex literate language structures and text genre knowledge – is characterized as a ‘school-sanctified symbolic resource’ (Collins 2000: 68). School success is measured in terms of the knowledge of these features. Bourdieu and Passeron (1977) emphasized that the power of schools and universities in defining what cultural capital is depends on the ‘relative autonomy’ of educational institutions. Schools and universities are free to reinterpret the external social requirements according to their institutional heritage norms. The more coherently an institution acts on its heritage norms, the more powerfully it maintains the schoolrelated cultural capital. School-related heritage norms have a profound effect on the ideology of writing. Otherwise, it would be incomprehensible why the general belief in writing still assumes the fixity of a text, the transparency of written language, and the universality of shared, available meaning in written text in contrast to an oral utterance (Olson 1977). The norms of writing, in other words, do not stop at written language. One manifestation of linguistic prescriptivism in oral communication is the practice of correcting non-standard pronunciation, evaluating oral non-standard features as poor and associated with lower social classes (Collins 1996). Non-standard pronunciations are hence as much connected to social identities as are non-standard orthographies (Jaffe 2000; Agha 2003). This ideology – or tyranny – of schooled literacy is the reason why it is so difficult to distinguish writing and codification processes from prescriptive language norms, in oral and written language alike. Our interest here is with the indexicalities of writing in particular social and political contexts and with questioning the circumstances under which writing can be tyrannical.

No dichotomy but a continuum Goody, Olson and other leading literacy researchers were heavily criticized in the 1980s and 1990s by Brian Street (e.g. 1995) and other linguistic anthropologists (e.g. Kulick and Stroud 1990). They were accused of having theorized a ‘great divide’ between literacy and orality, as well as between literate and non-literate societies. The New Literacy Studies as Street baptized his approach, began to ask not how literacy affects people but how people affect and appropriate literacy cross-culturally. Literacy does not operate autonomously, but always ideologically, Street argued, recognizing not a single literacy, but multiple literacies. Street criticized the assumption that literacy in itself – autonomously – has benign effects, that ‘introducing literacy to poor, “illiterate” people, villages, urban youth etc. will have the effect of enhancing their cognitive skills, improving their economic prospects, making them better citizens, regardless of the social and economic conditions that accounted for their “illiteracy” in the first place’ (Street 2003: 77). Street’s alternative ‘ideological model of literacy’ suggests that literacy, rather than a technical and neutral skill, is a social practice, embedded in cultural worlds, local epistemologies and power relations and thus differs from one context to another (see also Papen 2005).

Introduction: The Tyranny of Writing in Language and Society

9

Independently of the New Literacy Studies several authors have studied the conceptual differences between orality and literacy, arguing that orality/literacy should be seen as a continuum defined by its extremes rather than a dichotomy. Corpus studies, for instance, have shown that orality tends to coordinate utterances during conversation, while written genres tend to show a greater degree of grammatical compactness. The increase of explicitness in written texts is hence derived from the ideal-typical social conditions of orality and literacy: interlocutors that are both present in the situation as opposed to a writer and reader who produce and interpret text at different times and places (Biber 1988; Maas 2010). An early and interesting framework that explores the conceptual differences between orality and literacy in terms of a continuum is that of Peter Koch and Wulf Oesterreicher (1985). Fairly well known and influential in German-language scholarship on language, literature and media (see for example the recent collection by Feilke and Hennig 2016), their model is organized around the distinction between ‘conceptually oral’ and ‘conceptually written’ texts and based on the premise that every written text can be spoken just like every spoken word can be written down. The organization of an academic text, for instance, is conceptualized as a monologue, as condensed thoughts shaped through writing and rewriting. Nancy Hornberger’s (2004) continua of biliteracy model also pointed out the need for understanding literacy not as opposed to but intersecting with orality. Biliteracy here refers to any instance of communication that occurs in two (or more) languages in or around writing. And the notion of continuum is meant ‘to break down the binary oppositions’ (e.g. between standard and non-standard language use) and ‘draw attention to the continuity of experiences, skills, practices, and knowledge stretching from one end of any particular continuum to the other’ (p. 156). Hornberger suggests that there is not a single continuum between literacy and orality but a range of continua along various interrelated dimensions. The framework considers power relations in these dimensions and recognizes that one end of the continua tends to be privileged over the other (Weth 2016). As such, the model is intended to serve as a heuristic for understanding (and transforming) language policy and planning. Another attempt to blur the conceptual dichotomy of orality and literacy is that of research focusing on non-formal education (Scribner and Cole 1981) and ‘grassroots literacy’ (Fabian 1992; Blommaert 2008). In their work based on longitudinal research among the Vai in northwestern Liberia, Sylvia Scribner and Michael Cole (1981) attempted to ‘unpackage’ literacy from schooling, showing that assumed cognitive consequences of literacy such as context-independent abstract thought, memorization skills and logical thinking are actually more correlated to schooling and urbanism than literacy as the mere ability to read and write. What matters are the social and cultural conditions under which and the purposes for which a specific type of literacy is acquired. The Vai had three scripts at their disposal – their own Vai syllabary and the Arabic and English alphabets – each with its own history, practices, ways of learning (with or without formal schooling), local functions and understanding of what literacy is. One of the conclusions Scribner and Cole drew is that there is no ‘great divide’ in human modes of thinking beyond historical generalization and that in contemporary societies there is not one unified literacy but a range of different literacies.

10

Tyranny of Writing: Ideologies of the Written Word

Grassroots literacy is writing produced under poor material and infrastructural conditions and at a distance from the institutions of prescriptive elite-linguistic normativity. It is ‘a literacy which works despite an amazingly high degree of indeterminacy and freedom’ in spelling solutions (Fabian 1992: 90). For Jan Blommaert (2008) grassroots literacies are not orthographic but ‘heterographic’, that is, characterized by the absence of a single normative centre with the power to force writers to align with it. Instead, grassroots literacies draw on a multiplicity of near and distant, powerful and weaker centres of authority as well as general creative principles of mapping sound to visual form and are typically the result of self-taught writing or incomplete formal education. These theories of writing were developed before electronic online communication was part of everyday literacy practices. Digital and mobile communication, however, is radically changing written communication and hence also the organization and conception of texts (cf. Deumert 2014). Instant messaging, for instance, is conceptualized as a dialogue, that is, as spontaneous conversational turns similar to informal oral communication. Digital literacy certainly blurs any oral-literate dichotomy further and offers unprecendented opportunities for literacy and multilingualism (cf. Juffermans and Abdelhay 2017). The key notion of the aforementioned models – that of a continuum – emphasizes the fact that dichotomies are purely theoretical, while real instances of language use are situated at some non-finite and non-static point of more than one continuum at a time.

The ‘tyranny of writing’ in this book The contributions to this book engage with the metaphor of tyranny as a heuristic for ideologies of language and literacy in society, exploring practices and ideologies of writing that exercise, are affected by, or escape from an exercise of power or authority of writing. By using the word ‘tyranny’, we not only revisit Saussure’s critique of writing but also join an old philosophical discussion on power and politics (Xenophon [ca. 365 BCE] 2013; Strauss [1948] 2013) as well as numerous essays, books, blog posts and other text genres inspired by this metaphor (e.g. Easterly 2014, to cite one recent title). Any modern dictionary readily provides a quick entry into the range of meanings occupied by ‘tyranny’. We are informed that ‘tyranny’ refers to a form of government that is absolute and concentrated in one individual (the tyrant) and is oppressive, harsh, unjust, beyond right. Tyranny, in other words, is the opposite (or one of the opposites) of democracy – the form of government in which the state is ruled by commoners or their representatives chosen by lot and/or elections. As Giovanni Giorgini (2004: 1) points out, the invention of democracy around 508 BCE in the city-state of Athens marks the birth of the ideological figure of the tyrant. From that moment onwards, ‘tyranny’ gradually acquired strong pejorative moral connotations. Before that, a tyrant more generally meant an authoritarian sovereign, that is, a one-person government, and could be either good or bad, depending on the person’s character and style of leadership. By including ‘tyranny’ in the title of this book, we view writing as an acting object of tyranny. We have seen with Saussure and Bloomfield that the technology

Introduction: The Tyranny of Writing in Language and Society

11

of writing itself can be frustrating, oppressing or troubling. Their arguments against writing – that it distorts our image of language and interferes with language change – are of course far removed from any political discussion of tyranny. As noted, there are no agents involved in Saussure’s initial formulation. It is the written word itself that rules. Such ‘tyranny of writing’ is not invested with political or ideological interests, but is implicit in the medium of writing itself. Writing here is an index of power and social hierarchies, a carrier of injustices and inequalities in society but also – recalling Freire – a potential vehicle for social emancipation and empowerment. We are well aware that ‘tyranny’ may be an uncomfortable term, too strong or too powerful a metaphor to work with, or not diplomatic enough for some purposes. In writing of the ‘tyranny of writing’, however, we are saying that writing can be like tyranny. The metaphor suggests that what is in power (what is the norm, what defines itself as the centre, etc.) often arrived there by force or by accident, and is rarely carefully chosen or fairly elected. Metaphors typically exaggerate, but they also serve to clarify abstract ideas. It is with these intentions that we suggest ‘tyranny’ as a useful conceptual tool to analyse and critique ideologies of language and literacy in society. The book’s chapters critically examine tyrannies and ideologies of writing in a variety of social domains and contexts. Domains that are investigated in the following chapters include education, tourism, sports, politics, everyday life, comedy, linguistic landscapes and the digital world. The chapters are set in a range of historical and contemporary cultural contexts in Europe, Asia and Africa. We will now introduce the book’s chapters in order of appearance. Chapter 1 provides a further revisiting of Saussure’s critique of writing. In this chapter, Florian Coulmas asks where writing comes into play in linguistics and therefore discusses the question of what writing means for the study of language, from Saussurean, humanistic and comparativist perspectives. Even Saussure, who elaborates on the ‘tyranny of writing’, Coulmas demonstrates, cannot do linguistics without writing. He further shows that the ‘tyranny of writing’ has different meanings in the time of the humanist Nebrija, Saussure and today. He holds that ‘human history is indeed the history of emancipation from the tyranny of nature’. Writing, he argues, is a technology that has been influencing language(s) and communication like any other technology. If writing is one of many external influences on language, he concludes, one may ask why these influences should be called ‘tyrannical’. The following chapters present a journey through history and around the world. We begin our journey at the northeastern edge of Europe in the medieval Novgorod Republic via the European Renaissance and proceed to nineteenth-century France and Germany. The following five chapters take us through colonial and postcolonial Africa and Asia – the Anglo-Egyptian Sudan and contemporary Sri Lanka, the Casamance region of Senegal and Hong Kong as well as Central China. The final two chapters take us to contemporary European contexts: to Brabant in the Netherlands and the Balearic Islands of Spain. Chapter 2 is set in medieval Novgorod in pre-tsarist Russia (Kievan Rus’ eleventh to fifteenth century BCE). Daniel Bunčić presents the biscriptal situation of formal texts produced in monasteries and chancelleries and in private vernacular writing. Both the

12

Tyranny of Writing: Ideologies of the Written Word

writing material and the spelling differ in these texts. The ‘vernacular orthographies’ are written on cheap and available material, birch bark, and seem to answer to an acute need for written communication in everyday life. Bunčić discusses two hypotheses about the origin of the vernacular orthography, interpreting the ‘tyranny of writing’ in two ways. First, he relates to Saussure’s questioning of the correspondences between spellings and pronunciation. He shows that ‘the more inadequately writing represents what it ought to represent, the stronger is the tendency to give it priority over the spoken language’ (Saussure [1916] 2013: 34). Second, he relates ‘tyranny’ politically to the emancipation from the hegemonic centre and its ‘standard’ orthography, drawing parallels between the vernacular orthography in medieval Novogrod and the twentyfirst-century subversive misspelling system of Olbanian on the Russian internet. Chapter 3 is a summary article of Joop van der Horst’s important book, Het Einde van de Standaardtaal, published in Dutch (2007). This book was well received and debated in the Netherlands and Flanders, well beyond a narrow circle of language scholars, and makes an argument that has no parallel in English language scholarship. While many scholars have pointed to the importance of the nineteenth century for the formation of European national standard languages, Van der Horst argues that this process began much earlier with the Renaissance and already began showing its first signs of decline by the end of the nineteenth century. Tracing these changes over a longue durée, Van der Horst argues that standard languages as we know them are coming to an end. ‘Tyranny’ here mainly refers to language purism and the general concern about language correctness. The norms prescribed there are the norms of and for written language, extended to language tout court. Chapter 4 deals with the ‘orthographic tyranny’ of French schooling at the end of the nineteenth century. It analyses the complex interrelationship between alphabetization and Frenchification in multilingual regions of France, specifically the region of Bretagne (Brittany). Manuela Böhm analyses evaluation reports of school inspectors that document the language and orthographic competences of pupils. The inspectors report an urgent need in multilingual regions such as Brittany to improve the pupils’ French language and literacy skills. ‘Tyranny’ here is the ideological construct of national and linguistic unity, imposing French literacy on the Bretonspeaking bilingual population. Written French is the national language and hence the basis of Frenchification. For centuries, French linguistic unity has been recounted as a success story; in this chapter, it is critically reviewed through the lens of writings of multilingual pupils. Chapter 5 presents the ideologies of language and literacy in the German educational reform movement before the First World War. At the time of mass alphabetization, progressive educational philosophies of language emerged that put the child at the centre of learning. These ‘reform pedagogies’ claimed that children should learn what they can discover by themselves and that pedagogy should not curtail the freedom of the child. Ulrich Mehlem presents two leading figures of the movement, Rudolf Hildebrand and Berthold Otto, as well as their impact on the German curriculum and pedagogical theory. The ‘tyranny’ presented in this article is twofold. The reformers considered writing as tyrannical in itself as it represents and (re)produces the standard language and formal grammar teaching as opposed to

Introduction: The Tyranny of Writing in Language and Society

13

the child’s authentic way of speaking. Mehlem’s analysis also reveals a second-order ‘tyranny’ that is exerted not by writing but by the pedagogues themselves as they represent a particular nationalistic view of the German collective soul (Volksgeist). He ultimately argues that neglecting writing in education is itself tyrannical. Chapter 6 presents – like Chapter 5 – a situation in which a group of people discuss and decide on the writing and writing practices of others. It is situated at the height of European colonization of Africa and focuses on a 1928 colonial language conference in the Anglo-Egyptian Sudan – one of many missionary-organized language-planning conferences in the late nineteenth and early twentieth centuries. The aim of that conference was to develop a common writing system for ‘local languages’ for educational purposes. The choice of a unified orthographic system is presented as an ideological act that is nourished by the interests of missionaries, government officials and professional linguists. The ‘tyranny’ of writing in this chapter by Ashraf Abdelhay, Busi Makoni and Sinfree Makoni, is related to the power–knowledge nexus of colonial policy and practice. The authors show how colonial scientific experts carefully erase and misrepresent the ‘local voice’ in this process. The chapter furthermore shows how the planning of a writing system can impact the construction of racial and regional differences. Chapter 7 takes us to South Asia, to the transnational world of cricket and English in postcolonial Sri Lanka. The author, Harshana Rambukwella, illustrates the pervasiveness of monolingual English norms within a multilingual society by looking at two popular and internationally renowned cricket players as public speakers. The two players represent different social classes, educational backgrounds, styles of English speaking and of cricket play. Although both men are national sports heroes in Sri Lanka, their different capacity to speak in English triggers important social distinctions. Rambukwella analyses and contrasts two moments of speaking within the sociolinguistic framework of Sri Lankan English, revealing two levels of tyranny in this system. On the one hand, standard Sri Lankan English symbolizes the legitimate language and continuity with colonial times. On the other hand, the tyranny is embedded in educational policy that denies, in the name of democracy, opportunities to learn standard English despite the fact that it is crucial for social mobility. Chapter 8 is set in the densely multilingual Casamance area of southern Senegal in West Africa. Friederike Lüpke argues that the grassroots literacy practices of informal contexts such as texting and social media are ill-understood by government and development agencies communicating through writing. Multilingual literacy in health information, for instance, tends to essentialize ethnolinguistic identities and reify them into (Roman-script) writing: to each village variety its own written language. While this may be appreciated by the communities in question for symbolic reasons, it ignores actual communicative repertoires and even prevents effective communication. The ‘tyranny of writing’ in this chapter thus refers to the fixity of written standards for each language variety. Lüpke proposes instead to invest in language-independent literacies in adult and community literacy programmes in order to ‘liberate its writers from the tyranny of writing’. In Chapter 9, David C.S. Li introduces the reader to standard written Chinese (SWC). The difficulties of Chinese literacy are compounded for L2 speakers of Mandarin such as Cantonese speakers and, even more so, for migrants from

14

Tyranny of Writing: Ideologies of the Written Word

South and Southeast Asia (India, Pakistan, Nepal, Philippines, Indonesia). In Hong Kong, SWC is crucial for education- and work-related purposes, in addition to Cantonese and English. The chapter discusses some of the difficulties of mastering written Chinese in multilingual Hong Kong. The ‘tyranny of writing’ is inherent here in the Chinese writing system itself but also in the particular curricular design of public education in Hong Kong. Like Chapter 9, Chapter 10 also deals with the challenges of writing Chinese, albeit from a very different angle. Focusing on the severely endangered Tujia language of Enshi in Hubei province in Central China, Xuan Wang shows how the commodification of tradition in the context of heritage tourism leads to the invention of authentic forms of fangyan character writing in the linguistic landscape of a tourist place. The ‘tyranny of writing’ in this example refers to a complex and dynamic sociolinguistic process. The point of departure is the centralized national language policy of standard Chinese (Putonghua) that marginalizes non-standard varieties to the point of erasing them. Another ‘tyranny’ is the imperative of ‘doing authenticity’ in order to meet tourists’ expectations of authenticity. In Chapter 11 we return to Europe. The focus of this chapter is the Brabant dialect in the south of the Netherlands. After introducing some particularities in pronunciation and in the use of Brabantish, Jos Swanenberg presents the case of a young Black comedian, Steven Brunswijk, aka ‘the Braboneger’, who writes in Brabant dialect on Twitter as part of his public comedy performances. The chapter highlights several superimposed tyrannies: the hierarchy between (‘high’) standard and (‘low’) dialect, the purism of small-scale dialect standardization and normalization, and the exclusive (White) authenticity that Brabantish indexes – all of which are challenged by new digital writing and Braboneger’s comedy. The book concludes with Chapter 12, which focuses on language-ideological debates surrounding Spanish and Catalan in the Balearic Islands. Lucas Duane presents an emerging discourse of particularization of Balearic against Catalan organized around Facebook groups lobbying for the emergence of a Balearic variety as a language in its own right. The debate concerns the ‘salty article’ es/sa/es/ses in Balearic vernacular as an emerging differentiation from the standard Catalan forms el/la/els/les. The linguistic difference between the articles in standard Catalan and the Balearic variety, clearly audible in speech, enregisters in writing. In its becoming literary, the salty article turns iconic for the discussion and emergence of linguistic distinction, recognition and authority within a nationalistic ideological debate. This chapter thus shows how the ‘tyranny of writing’ operates at a microlinguistic and metalinguistic level at once. These examples of written language practices and ideologies around the written word demonstrate that writing can indeed be tyrannical. It seems, however, that writing is rarely ‘tyrannical’ by or in itself. Tyranny constructs as much as it is constructed by a particular social and political order. To summarize, there are at least five ways in which writing can be tyrannical: 1. Writing can be tyrannical in its capacity to influence speech. 2. Standard orthography can be tyrannical when it is embedded in political power and enforced in formal education.

Introduction: The Tyranny of Writing in Language and Society

15

3. Writing can be tyrannical – even in spoken language – as a source of prescriptive language norms. 4. Writing can further be tyrannical when it is used as a vehicle for social distinctions, particularization and nationalisms. 5. Neglecting writing in education, finally, can also be tyrannical, as when learners are deprived of the resources they need to participate in their respective societies. In summary, we intend the concept of the ‘tyranny of writing’ to be a critical heuristic for exploring, questioning and redressing ideologies of the written word in language and society.

References Agha, Asif (2003), The Social Life of Cultural Value, Language & Communication 23: 231–73. Barthes, Roland (1977), ‘The Death of the Author’, in Image, Music, Text (essays selected and translated by Stephen Heath), London: Fontana Press. Beaugrande, Robert de (2008), A Friendly Grammar of English, Free of copyright: www.beaugrande.com. Biber, Douglas (1988), Variation Across Speech and Writing, Cambridge: Cambridge University Press. Blommaert, Jan (2008), Grassroots Literacy: Writing, Identity and Voice in Central Africa, London: Routledge. Bloomfield, Leonard (1927), Literate and Illiterate Speech, American Speech 2: 432–9. Bloomfield, Leonard (1933), Language, New York: Henry Holt and Co. Bourdieu, Pierre, and Jean-Claude Passeron (1977), Reproduction in Education, Society, and Culture, Beverly Hills: Sage. Canetti, Elias (1977), Die gerettete Zunge: Geschichte einer Jugend, München: Hanser Verlag. Collins, James (1996), ‘Socialization to Text: Structure and Contradiction in Schooled Literacy’, in Natural Histories of Discourse, edited by Michael Silverstein and Greg Urban, 203–28, Chicago: University Press of Chicago. Collins, James (2000), Bernstein, Bourdieu and the New Literacy Studies, Linguistics and Education 11: 65–78. Cook-Gumperz, Jenny (ed.) (1986), The Social Construction of Literacy, Berkeley : University of California Press. Coulmas, Florian (2013), Writing and Society: An Introduction, Cambridge: Cambridge University Press. Deumert, Ana (2014), Sociolinguistics and Mobile Communication, Edinburgh: Edinburgh University Press. Easterly, William (2014), The Tyranny of Experts: Economists, Dictators, and the Forgotten Rights of the Poor, New York: Basic Books. Fabian, Johannes (1992), ‘Keep Listening: Ethnography and Reading’, in The Ethnography of Reading, edited by Jonathan Boyarin, 80–97, Berkeley : University of California Press. Feilke, Helmuth, and Mathilde Hennig (2016), Zur Karriere von Nähe und Distanz: Rezeption und Diskussion des Koch-Oesterreicher-Modells, Berlin: de Gruyter Mouton. Freire, Paolo, and Donaldo Macedo (1987), Literacy: Reading the Word and the World, Westport, CT: Bergin & Garvey.

16

Tyranny of Writing: Ideologies of the Written Word

Giorgini, Giovanni (2004), The Place of the Tyrant in Machiavelli’s Political Thought and the Literary Genre of the Prince, The Italian Academy for Advanced Studies at Columbia University Lunch Seminar, 18 February. Giroux, Henry A. (1987), ‘Introduction: Literacy and the Pedagogy of Political Empowerment’, in Literacy: Reading the Word and the World, edited by Paolo Freire and Donaldo Macedo, Westport, CT: Bergin & Garvey. Goody, Jack, and Ian Watt (1963), The Consequences of Literacy, Comparative Studies in Society and History 5: 304–45. Hornberger, Nancy H. (2004), The Continua of Biliteracy and the Bilingual Educator: Educational Linguistics in Practice, Bilingual Education and Bilingualism 7: 155–71. Jaffe, Alexandra (2000), Introduction: Non-standard Orthography and Non-standard Speech, Journal of Sociolinguistics 4: 497–513. Juffermans, Kasper, and Ashraf K. Abdelhay (2017), ‘Literacy and Multilingualism in Africa’, in Encyclopedia of Language and Education: Literacies and Language Education, edited by Brian V. Street and Stephen May, New York: Springer. Kloss, Heinz (1967), Abstand Languages and Ausbau Languages, Anthropological Linguistics 9: 29–41. Koch, Peter, and Wulf Oesterreicher (1985), Sprache der Nähe – Sprache der Distanz: Mündlichkeit und Schriftlichkeit im Spannungsfeld von Sprachtheorie und Sprachgeschichte, Romanistisches Jahrbuch 3: 15–43. Kulick, Don, and Christopher Stroud (1990), Christianity, Cargo and Ideas of Self: Patterns of Literacy in a Papua New Guinean village, Man 25: 286–304. Latour, Bruno (1991), We Have Never Been Modern, Cambridge, MA: Harvard University Press. Maas, Uts (2010), Literate und Orat. Gundbegriffe der Analyse geschriebener und gesprochener Sprache, Grazer Linguistische Studien 73: 5–20. Mullins, Daniel A., Harvey Whitehouse, and Quentin D. Atkinson (2013), The Role of Writing and Recordkeeping in the Cultural Evolution of Human Cooperation, Journal of Economic Behavior and Organization 90: 141–51. Olson, David R. (1977), From Utterance to Text: The Bias of Language in Speech and Writing, Harvard Educational Review 47: 257–81. Papen, Uta (2005), Adult Literacy as Social Practice: More than Skills, London: Routledge. Plato (1952 [278 BCE]), ‘The Superiority of the Spoken Word: Myth of the Invention of Writing’, in Plato’s Phaedrus, edited by Reginald Hackforth, 156–64, Cambridge: Cambridge University Press. Saussure, Ferdinand de ([1891] 2006), ‘Second Lecture at the University of Geneva, November 1891’, in Ferdinand de Saussure: Writings in General Linguistics, edited by Simon Bouquet, Rudolf Engler, Carol Sanders and Matthew Pires, 104–9, Oxford: Oxford University Press. Saussure, Ferdinand de ([1916] 2013), Course in General Linguistics, edited and translated by Roy Harris, London: Bloomfield. Scribner, Sylvia, and Michael Cole (1981), The Psychology of Literacy, Cambridge: Harvard University Press. Sebba, Mark (2007), Spelling and Society: The Culture and Politics of Orthography around the World, Cambridge: Cambridge University Press. Strauss, Leo ([1948] 2013), ‘On Tyranny’, in On Tyranny: Corrected and Expanded Edition, Including the Strauss-Kojève Correspondence, edited by Victor Gourevitch and Michael S. Roth, 22–131, Chicago: The University of Chicago Press. Street, Brian V. (1995), Social Literacies: Critical Approaches to Literacy in Development, Ethnography and Education, London: Longman.

Introduction: The Tyranny of Writing in Language and Society

17

Street, Brian V. (2003), ‘What is “new” in New Literacy Studies? Critical Approaches to Literacy in Theory and Practice’, Current Issues in Comparative Education 5: 77–91. van der Horst, Joop (2007), Het einde van de standaardtaal. Een wisseling van Europese taalcultuur, Amsterdam: Meulenhoff. Weth, Constanze (2016), ‘Bilinguisme et bilittéracie’, in L’Education bilingue en France: Politiques linguistiques, modèles et pratiques, edited by Jürgen Erfurt and Christine Hélot, 565–77, Limoges: Lambert-Lucas. Xenophon ([ca. 365 BCE] 2013), ‘Hiero, or Tyrannicus’, edited and translated by Leo Strauss (1948), in On Tyranny: Corrected and Expanded Edition, Including the StraussKojève Correspondence, edited by Victor Gourevitch and Michael S. Roth, 1–21, Chicago: The University of Chicago Press.

1

Revisiting the ‘Tyranny of Writing’ Florian Coulmas

‘We generally learn about language only through writing.’ (Ferdinand de Saussure)

The study of language, as the study of any subject, is dependent on writing because the scientific enterprise is. The scientific world view assumes that the things and events that constitute the universe are understandable. Another fundamental assumption is that knowledge accumulates and progresses, that is, we know more now than people knew in Aristotle’s time. In the absence of writing people are not ignorant, but for science as we understand it, writing is indispensable. It enables scientific insights to be given permanence, separating message from messenger, text from author, judgement from judge, sentence from speaker. And it allows us to critically assess, take issue with, and build on the knowledge of our forebears. This chapter discusses the question of what writing means for the study of language, taking as its point of departure Ferdinand de Saussure’s critique of spelling conventions and its consequences for the evolution of modern linguistics. As in other scientific disciplines, in linguistics, too, writing is a major tool. However, what distinguishes the role of writing in linguistics from other fields of scholarship is that it relates to the object of investigation in complex ways concerning both the scientific analysis of language and the social conditions of its use. In literate society it is imperative to understand what the ‘tyranny of writing’ meant for the study of language when Saussure first used this term a century ago, and what it means today.

Introduction At the outset of chapter VI of the Course in General Linguistics Saussure remarks that ‘we generally learn about language only through writing’ (1959: 23). Rather than being a factual observation, this is the introduction of a critique of linguistics which, as he saw it, suffered from ‘the tyranny of writing’ or la tyrannie de la lettre. Wade Baskin’s rendition of Saussure’s term in the English translation of his unmatched Course in

20

Tyranny of Writing: Ideologies of the Written Word

General Linguistics as the tyranny of writing has a slightly different meaning. ‘La lettre’ is a noun defined in the Larousse dictionary as chacun des signes graphiques don l’ensemble constitue un alphabet et qui, seul ou en combinaison avec d’autres, correspondent à un son de la langue, [Each of the graphical signs which taken together constitute an alphabet corresponds, alone or in combination with others, to a sound of the language.]

whereas writing is defined in the Oxford English Dictionary as 1. the activity of writing, 2. a sequence of letters, words or symbols marked on a surface, and 3. written work. Hence, when it comes to ‘tyranny’, we have an object, in French, and an activity and variously defined objects, in English. Rather than dwell on this apparent difference, however, I want to examine what exactly Saussure considered tyrannical about writing and letters. To begin with, let me say that Saussure is hard to argue with. Over a century’s time his ideas about language have lost nothing of their sagacity; his observations are keen and well informed; and his theory is an utterly convincing integrated whole. It is hardly an exaggeration to say that everyone doing linguistics today is indebted to his wisdom.

What Saussure said That said, let us re-examine chapter VI of the Course titled ‘Graphic representation of language’. At issue here are difficult questions concerning the ontological status of language and writing that are at the heart of our understanding of what linguistic theory should be about. The very title of the chapter I just quoted is indicative of the direction of Saussure’s thinking: There is language, and there is representation. As Saussure sees it, there are two distinct systems of signs, language and writing, and the latter’s sole purpose is to represent the former. Hence, he argues that ‘the linguistic object is not both the written and the spoken forms of words; the spoken forms alone constitute the object’ (Saussure 1959: 24). Saussure’s purpose, we must not forget, was to lay the foundations of a new theory of language that was at variance with traditional philological studies. ‘The first linguists’, he criticized, ‘confused language and writing, just as the humanists had done before them’ (1959: 24). Worse, they recognized an influence of writing on language. He writes: ‘Take the notion that an idiom changes more rapidly when writing does not exist. Nothing could be further from the truth’ (1959: 24), but continues right there: ‘Writing may retard the process of change under certain conditions.’ And on the next page asks: ‘How is the influence of writing to be explained?’ If no influence existed, there was not only no need to explain it, but also every attempt to do so would be

Revisiting the ‘Tyranny of Writing’

21

bound to fail. Saussure further speaks of ‘the undeserved importance of writing’ which makes ‘people forget that they learn to speak before they learn to write, and the natural sequence is reversed’ (Saussure 1959: 25). The key term here is ‘natural sequence’. As Saussure defines it, language is ‘deposited in the brain of each individual’ (1959: 23). Clearly, whatever it is that is deposited there, it cannot be letters, but it cannot be sounds either which in his diction belong to parole rather than langue. It can only be the abstract units of la langue which could be given expression by means of either phonic or graphic signs. Actually, Saussure admits as much when he discusses the Chinese writing system which according to him is to the Chinese ‘a second language’ (Saussure 1959: 25). By nonetheless insisting that ‘the linguistic object is not both the written and the spoken forms of words [but] the spoken form alone’ (ibid.: 24) he demonstrates that his thinking is firmly grounded in the alphabetic, not to say alphabetocentric, tradition of Western scholarship. He takes it for granted, as the Greek and Latin alphabets suggest, that the sole purpose of writing is to map speech and that, accordingly, letters are but substitutes of sounds. Had he also taken non-alphabetic writing systems into consideration, he might have admitted the possibility that writing, rather than being a mere substitute, is a supplement of speech that opens up new modes of human communication that speech – or a mere surrogate thereof – cannot achieve. Language is a defining characteristic of humanity, a natural faculty that cannot be subject to the influence of an artefact, which writing indubitably is. Yet, by asking ‘how the influence of writing is to be explained’, Saussure implicitly concedes that there is an influence. Rather than denying that writing has an effect on language, he objects to its ‘undeserved importance’ (ibid.). ‘Undeserved’ could mean one of two things or, perhaps, both: undeserved in scholarship or undeserved in actual fact, that is in the evolution of language. Some observations Saussure cites to bolster his argument suggest that he means undeserved not just on the level of inquiry, but on the level of actual facts, too. In any event, his argument is based on what he considers the natural order of things, about which his predecessors had different ideas. He explicitly mentions the humanists; so let us revisit one of them.

Antonio de Nebrija’s idea of writing Antonio de Nebrija (1441–1522) was a letrado, a ‘Renaissance man of letters’, and letters were what he considered important. In the fateful year of 1492 he famously published the first grammar of a European vernacular ever, the Gramática de la lengua Castellana, where he declares: Among all the things that human beings discovered through experience, or that were shown to us by divine revelation in order to polish and embellish human life, nothing was more necessary, nor benefited us more, than the invention of letters. Such letters, which by a common consent and the silent conspiracy of all nations have been accepted, have been invented – according to those who wrote about antiquity – by the Assyrians; with the exception of [Gellius], who attributed the invention of letters to Mercury in Egypt. (Nebrija 1492, I, ch. 2: 14; quoted from Mignolo 1992: 188)

22

Tyranny of Writing: Ideologies of the Written Word

Regarding the spreading of letters Nebrija speaks of ‘a common consent and the silent conspiracy of all nations’. What kind of ‘conspiracy’ that might be is a question to which we will return later. Suffice it at this point to notice that Nebrija puts himself in the classical, that is, Aristotelian tradition, describing the letter as a ‘trace or figure to represent the voice’. Letters were an important invention, beneficial to embellish human life, but ontologically that which they represent is more important or essential to human nature. So far, Nebrija’s understanding of letters seems to follow Aristotle’s (1938: 115) notion of letters as a secondary system of signs: ‘Words spoken are symbols of affections or impressions of the soul; written words are symbols of words spoken.’ However, Nebrija is concerned with letters not just as images of spoken sounds, but as models as well. In his Castilian orthography Reglas de ortografía en lengua castellana (1517) he complains that ‘these days no one writes our language purely, due to the lack of some letters which we pronounce but do not write, and others, on the contrary, which we write but do not pronounce’ (Mignolo 1992: 190). This is not as it should be, for a close match between letters and sounds, according to Nebrija, is the very raison d’être of letters. He was not guilty as charged by Saussure; he did not confuse language and writing, but his idea about the relationship between the two was different. Like Saussure, Nebrija distinguished between letters and sounds and detected discrepancies between them, but while Saussure complains that ‘visual images lead to wrong pronunciations  … mistakes [that] are really pathological’ (Saussure 1959: 31), Nebrija recognized in grammar (in the sense of letters) a means not just ‘to polish and embellish human life’ but to preserve the meaning of the divine word, thanks to the equivalency between the spoken word and the letters representing it: So that those words which God first made known through Moses and other prophets and hagiographers … would not be erased by the long passage of time, they were entrusted and commended to grammar, the preserve of letters. (Nebrija 1503, De vi ac potestate litterarum, quoted from Rojinsky 2010: 117)

Nebrija welcomes the stability that language is afforded by letters. They were ‘a trace or figure by means of which the voice is represented’ (Reglas de ortografía en la lengua castellana, Mignolo 1992: 189), which in his view meant representing an ideal state of the language, a state that should be preserved, because it embodied the word of God. Rather than condemning the influence that writing might exert on language, Nebrija sees in it the embodiment of language as it should be. The observed discrepancy between writing and pronunciation should be corrected by adjusting the pronunciation rather than the spelling. According to the second principle of his orthography, we should ‘write as we pronounce and to pronounce as we write’, for otherwise the letters ‘would have been invented in vain’. This is a step away from the Aristotelian conception of letters as mere secondary symbols. His daring project of a vernacular grammar was intended to raise Castilian on a level with the languages of scripture. Grammar, the art of letters, as he put it in the prologue to his Castilian grammar, was a means of transforming the unruly vernacular into an artefact, protecting it against decay. Although invented in order to represent voice, letters under his hands thus acquire

Revisiting the ‘Tyranny of Writing’

23

ontological independence and are attributed ‘a clear priority over the voice’, as Mignolo (1992: 189) has convincingly argued.

God, nature and society It is with this notion that Saussure takes issue, because he considers it pathological that ‘some Parisians already pronounce the t in sept femmes’ (1959: 32). ‘Mispronunciations due to spelling’, he complains, ‘will probably appear more frequently as time goes on’ (1959: 31). Pathology is a deviation from a healthy normal condition of a natural organism. Saussure’s example is very instructive, for during the one hundred years that separate us from his observation the pronunciation of sept with a final voiceless stop has become the norm, however pathological or healthy. The question that arises is whether what Saussure concludes from his observation is compelling – namely the conclusion that ‘the pronunciation of a word is determined, not by its spelling, but by its history’ (1959: 31). If so, should we then ignore phonic deformations ‘that do not stem from its natural functioning’ such as sept? If it is true that in French the t got pronounced as a result of writing, why should this be considered ‘tyranny’? Because, Saussure argues, ‘by imposing itself upon the masses, spelling influences and modifies language’ (ibid.). However, he continues, ‘this happens only in highly literate languages where written texts play an important role’ (1959: 31). Yet, he insists that data such as ‘sept femmes’ should be set aside because ‘they are teratological cases’ (1959: 32). Teratology is the scientific study of congenital abnormalities and abnormal formations. This terminology is a bit surprising. It is a metaphor akin to Herbert Spencer’s evolutionary conception of society that was very influential in the late nineteenth century. It should not be taken, however, to indicate that Saussure conceived of linguistics as a natural science. He did not. He rejected the conception of language as a natural organism to which the ‘first linguists’, that is, the comparativists, subscribed just as vigorously as the humanists’ idea that letters were meant to guide pronunciation. We have three distinct positions here as to what the task of linguists should be. The humanists, some of them at least, saw grammar as the art of letters invented as a means to guarantee that the meaning of God’s utterances would not be disfigured in the course of time. The grammarian’s task was to honour the authority of God by preserving a close correspondence between sound and letter. The enlightened comparativists dispensed with God as the supreme arbiter of right and wrong in matters of writing, turning language into an organism governed by natural laws. In the spirit of his time, Saussure introduced yet another position, replacing nature by society. He agrees with his predecessors in historical–comparative linguistics in as much as he considers language as an integrated whole that determines individual speakers’ speech but is beyond individual intervention. He parts company with them in regard to the formative forces of language which he famously characterized as a ‘social product’ rather than a natural one. Where does writing come into play? Writing, according to him, is external to language, but so is phonē, voice. Language is an abstract system that as a matter of principle is independent of any material manifestation, but he recognizes that, because

24

Tyranny of Writing: Ideologies of the Written Word

Table 1.1 What linguistics is all about: Three points of view Humanists

Comparativists

Saussure

Reference point

God

Nature

Society

Individual speakers are guided by:

Ignorance

Forces of nature

Social dynamics

The linguist’s task is:

to establish and preserve the original meaning of the divine word.

to reconstruct the natural laws that determine linguistic change.

to discover the conventions adopted by a social body to permit individuals to exercise the faculty of speech.

it is a social fact, it cannot exist in individual brains alone without any communicable manifestation. Thus the embodiment of language as speech sounds comes in again through the back door. Writing, however, Saussure sees in the Aristotelian tradition as a derived sign system only, is designed to represent speech, a function it fulfils more or less faithfully. He fails to appreciate the fact that once used by a large part of the population, written language acquires a life of its own and that, accordingly, the logic of (oral) language and the logic of writing follow different rules. This has often been said and criticized, for instance by Roy Harris (2000) and myself (Coulmas 2003), among others. It is not only the genealogy of writing that is grossly misrepresented if it is described as an instrument to record speech. For all we know, no writing system was planned in order to record oral sounds from the beginning. Much rather, the linguistic interpretation of graphic signs that had been devised as a recording device, for bookkeeping, for example, came after the fact. There are some well-founded doubts about the ‘natural sequence’ whose reversal Saussure criticizes. Where do we let the history of writing begin? Is it a continuous evolution, or should we assume discontinuities and posit its ‘true’ beginning once a notation was at hand that could be employed to represent speech sounds? In the history of writing this would be a rather arbitrary demarcation. The second aspect that warrants some rethinking has to do with Saussure’s conception of language. He speaks of a social fact that is to be understood in terms of its internal logic and must be kept apart from external phenomena such as foreign words, writing, language spread, among others. His analogy with chess (Saussure 1959: 22f) is illuminating. Many things can happen on a chessboard, but the logic of the game is fixed once and for all. In this sense Saussure’s theory of language is static.

Nature and culture Much like some evolutionary scientists believe that the human species has stopped evolving, Saussure’s conception of language does not allow for the possibility that it might evolve. The moves and configurations on a chessboard are variable, but the totality of possibilities does not change. A chessboard with more than 64 fields or more

Revisiting the ‘Tyranny of Writing’

25

than 32 pieces would be a different game. We could of course accept Saussure’s analogy as a working hypothesis, for arguably what linguists should explain is language as a fully developed system, not its embryonic stages or its decaying remains. What does this imply for the analysis of social facts? As defined by Durkheim, social facts ‘reside in the society itself that produces them and not in its parts – namely its members. In this sense therefore they lie outside the consciousness of individuals’ (Durkheim 1982: 39). Although social facts are functioning outside human introspection, they can be changed, and they do change. Take such a fundamental social institution as the family. The study of kinship shows that virtually nothing is immutable there. The other important point is Saussure’s above-quoted remark that ‘spelling influences and modifies language … only in highly literate languages where written texts play an important role’ (1959: 31). What does this imply? The relatively recent appearance of writing has often been cited as the principal reason for ignoring it in the study of language. Humanity took to writing about 5,000 years ago, admittedly a short period of time when matched against the appearance of homo sapiens, said to have reached ‘anatomical modernity about 200,000 years ago and … full behavioral modernity around 50,000 years ago’ (McHenry 2009: 265). However, consider this. Two thousand years ago, there were about 300 million humans on the planet. It took 1,800 years then for the world population to triple to one billion. In the next 300 years it increased sevenfold, approaching exponential growth. During the past century, the world population grew exponentially, and it grew much older. For example, the average probability of a present-day 70-year-old Japanese dying is the same as a 30-year-old hunter-gatherer 100,000 years ago. The main part of this progress was made in the course of the twentieth century. Hygiene, medicine and technological progress have changed human life profoundly in a very short time. That is, the interaction between genes and human intervention in the environment has had unpredictable consequences. My question then is: What does this mean with regard to language? That languages change is uncontroversial, but does language change? And could not a technology that is 5,000 years old exert an effect on the population that uses it? I do not profess to have answers to these questions, but I think we shouldn’t shy away from asking them, just because we feel uncomfortable talking about the evolution of our species, unless the discussion is about the distant past. According to some evolutionary biologists (e.g. O’Neil 2013), our DNA today differs from that of our forebears around the time of the end of the last ice age – that’s just 10,000 years ago. Civilization, so the argument goes, made a difference. People learnt to live in cities where among many other things contagious diseases were much more devastating, speeding up the process of natural selection, that is, favouring the transmission of the genes of individuals lucky enough to survive epidemics. The point at issue here is the environmentally induced selection pressure, that is, the potential impact on our nature by external changes in the environment. Should students of human biology ignore them on the grounds that they obscure human nature? Similarly, we must ask whether writing ought to be excluded from the realm of linguistics proper because it is external/artificial. It seems to me that the question is settled. Just as modern humans manipulate nature including their own, they manipulate their language in

26

Tyranny of Writing: Ideologies of the Written Word

many ways, of which those associated with writing are the easiest to raise to the level of conscious reflection. In this connection it should also be noted that literacy rates have been rising continuously. Currently, 84.1 per cent of adults and 89 per cent of youth are literate, according to UNESCO, which implies that vastly more people than in the past are influenced in their language behaviour by writing. Writing should, therefore, be considered one of several external, that is, social and artificial, influences on language. If these influences deserve to be called ‘tyrannical’, what about vaccination, not to mention foetal intervention, and organ transplantation?! There are, of course, people who reject these procedures, but this is because of ideological reasons rather than because these procedures obscure our insight into human nature. Human history is the history of emancipation from the tyranny of nature. The first fire that was lit, the first stone axe that was flaked and the first shelter that was built helped early humans to withstand the elements and defend themselves against predators. Ever since, our relationship with the environment has been technologymediated. Saussure’s argument for the exclusion of writing from the study of language rests on the untenable assumption that it is possible to separate external, artificial forces such as the borrowing of foreign words and writing from the natural ‘constant forces in the life of language’ (1959: 22). What is more, if language is a social product, then how can loan words, spelling pronunciations, etc. reasonably be excluded from linguistic analysis? All of these phenomena are social products. The pristine homogeneous language community that lives in complete isolation without any contact with the outside world is an illusion rather than an abstraction. As we have seen earlier, Saussure concedes that ‘in highly literate languages’ ‘spelling influences and modifies language’ (1959: 31). This admission clearly undermines his argument and is enough to assign writing a place in linguistics where all aspects of the influence of writing on the social institution of language are studied. Notice in passing that expelling writing from the realm of linguistics would reduce its object of investigation to the present. About earlier historical stages of languages and languages no longer transmitted from one generation to the next we would know little if it wasn’t for writing, as Saussure would be the first to admit. Harris (1987: 43) has made this point: If writing only obscures the view on language, Saussure would be ‘obliged to conclude that in the case of dead languages a study of la langue (“the social product stored in the brain”) would be impossible in principle.’ That, Saussure would not have wanted to accept; but that nonetheless he argued that only the spoken forms of words ‘rather than both the written and the spoken forms of words are the linguistic object’ (1959: 23) is inconsistent.

Can we do linguistics without writing? But what about languages that have never been written, including dialects and substandard varieties that have caught the linguists’ attention? Here a different question arises: Can these languages be studied without recourse to writing? It is a matter of interest (also of historical interest) that Saussure considers what possible

Revisiting the ‘Tyranny of Writing’

27

alternative there could be. He writes: ‘In Vienna and Paris samples of all languages are being recorded. Even so, recorded specimens could be made available to others only through writing’ (ibid.). Although nowadays it is much easier to reproduce speech recordings and make them available to others, this is essentially unchanged. We cannot begin analysing linguistic data unless we set pen to paper and work on a transcription. In contradistinction to some natural sciences, linguistics constitutes its object of investigation. The relationship between observable facts and units of analysis is very indirect. In the field of language, there are no given observable objects to start out with, because speech communities are not uniform, and no two people speak exactly alike. In fact, no one speaker speaks exactly the same on different occasions. Linguists, therefore, have to deal with an ‘unmanageable mass’, as Saussure called it; and to get a grip on it, they make assumptions about it and impose upon it structures, whose plausibility can be assessed, both internally as being more or less consistent (free of contradictions, redundancies and ad hoc rationalizations), and externally on the basis of speakers’ judgements about similarities and differences. In the background of linguistic analysis the Tekhnê Gramatikê in the sense of the art of letters therefore looms large. Both speakers whose judgements are relied upon for organizing the ‘unmanageable mass’ and linguists are, as Saussure readily admits, influenced in their speech behaviour and in their perception of language by writing. For this reason alone, a proper understanding of writing is essential for the study of language. Languages cannot be dissected like plants and animals and human cadavers. Rather than starting out from observable objects, knowledge generation in the field of language proceeds from models that impose structures on the object of study, or, to put it differently, that constitute the units of investigation. The most commonly used model is the International Phonetic Alphabet (IPA) which, true to its name, is an alphabet. Since it was first designed in the 1880s, it has been refined, to be sure, but it is still an offshoot of the Latin alphabet. Aronoff (1992) and Faber (1992) have convincingly shown that phonemic segmentation is an outflow of alphabetic writing. The underlying principle of dividing the continuous flow of speech into discrete units is still the same. The visualization of speech that is produced by means of the IPA is a model of speech rather than its faithful image. Tone languages such as Chinese and many African languages provide further evidence for that. There are several options for notating tones in phonetic descriptions of languages where tone is distinctive that include numerals, a set of diacritic and tone letters. They all have in common the fact that they submit pitch accents and contour tones to the segmental structure of the IPA, although tones are not segments that precede or follow consonant and vowel segments. The imagery of a tonal language suggested by an augmented IPA transcription is as misleading and obscuring of linguistic facts as Saussure thought the t was in sept femmes. The Chinese writing system engenders a different kind of model that favours the perception of syllables. By means of a procedure known as Fanqi (反切) or ‘reverse cutting’ it led to sophisticated analyses of syllables in terms of onset and rhyme by matching characters that share initial and terminal sounds with each other. For instance, the fanqie for 東 is 德紅. The initial of 德 (d/e) is d, and the final of 紅 (h/ong) is ong. Accordingly, 東 is to be pronounced dong ‘east’.

28

Tyranny of Writing: Ideologies of the Written Word

The Devanagari script is also focused on the syllable as a unit; each akshara contains a neutral vowel and is modified with a diacritic for syllables containing a different vowel. These are just some other examples of how writing systems function as models of language. It is an illusion to assume that the IPA is much closer to the reality of speech than other writing systems.

Conclusions To conclude, Saussure was certainly right to point out that discrepancies between sound and letter can be misleading and, therefore, most written language material should not be regarded as reliable data at face value. He was wrong to argue that these problems could be solved, once and for all, by means of a ‘truly phonological system of writing [… where there is] one symbol for each element of the spoken chain’ (1959: 33). And he was wrong to argue that the linguistic object is not both the written and the spoken forms of words, but the spoken forms alone (1959: 23f). I have argued against this position because writing is crucially important for linguistic analysis on three levels. First, on the methodological level we have to understand the properties of the tool we work with. What are the raw data of linguistic research? How are they processed for analysis? Why use the IPA rather than Hangul? In scientific inquiry the effects that research tools have on the outcome of the investigation is a well-known phenomenon. No matter which transcription system we use, it is a tool whose properties we must understand. Second, on the theoretical level we have to be able to answer the question whether the structures we analyse are derived from or imposed onto the object of our investigation. What is a phoneme, a word, a sentence? In the absence of writing, these questions are not easily answered. As a matter of fact, common definitions of these notions are heavily influenced by writing. Third, on the object level we have to come to grips with the influence that writing may exert on a language, for instance by comparing written and unwritten languages; by comparing the linguistic output of speakers who communicate much in writing with that of others who do not. In this connection, phenomena such as the following are to be investigated: language standardization, diglossia, Ausbau, spectrum of varieties, spelling pronunciation and text-mediated linguistic borrowing, among others. This is an ample field, and if it is not for linguists to plough, I would not know for whom, pace Saussure to whom we can still turn, once again, by way of ending where we began, for ‘we generally learn about language only through writing.’

References Aristotle (1938), Peri Hermeneias (De interpretatione), translated by H. P. C. Fook, London: Loeb Classical Library.

Revisiting the ‘Tyranny of Writing’

29

Aronoff, Mark (1992), ‘Segmentalism in Linguistics: The Alphabetic Basis of Phonological Theory’, in The Linguistics of Literacy, edited by Pamela Downing, Susan D. Lima and Michael Noonan, 71–82, Amsterdam: John Benjamins. Coulmas, Florian (2003), Writing Systems: An Introduction to their Linguistic Analysis. Cambridge: Cambridge University Press. Durkheim, Emile (1982), The Rules of Sociological Method, translated by W. D. Halls, London: Macmillan. Faber, Alice (1992), ‘Phonemic Segmentation as Epiphenomenon: Evidence from the History of Alphabetic Writing’, in The Linguistics of Literacy, edited by Pamela Downing, Susan D. Lima and Michael Noonan, 111–35, Amsterdam: John Benjamins. Harris, Roy (1987), Reading Saussure, London: Duckworth. Harris, Roy (2000), Rethinking Writing, London: The Athlone Press. Mignolo, Walter D. (1992), ‘Nebrija in the New World: The Question of the Letter, the Colonization of American Languages, and the Discontinuity of the Classical Tradition’, L’Homme 32, 185–207. McHenry, H.M. (2009), ‘Human Evolution’, in Evolution: The First Four Billion Years, edited by Michael Ruse and Joseph Travis, Cambridge, MA: The Belknap Press of Harvard University Press. Nebrija, Antonio de ([1492] 2017), Gramática de la lengua Castellana, Barcelona: Red Ediciones. O’Neil, Dennis (2013), Evolution of Modern Humans, Website: http://anthro.palomar.edu/ homo2/default.htm (accessed 08/22/2017) Rojinsky, David (2010), Companion to Empire: A Genealogy of the Written Word in Spain and New Spain, c. 550-1550. Amsterdam and New York: Rodopi. Saussure, Ferdinand de (1916), ‘Cours de linguistique générale’, edited by Charles Bally and Albert Sechehaye, with the collaboration of Albert Riedlinger, Lausanne and Paris: Payot; English translation by Wade Baskin (1959), Course in General Linguistics, New York: Fontana.

2

How to Write a Birch-Bark Letter: Vernacular Orthography in Medieval Novgorod Daniel Bunčić

Rus’ and Novgorod Since the ninth century the predecessors of the modern Russians, Ukrainians, Belarusians and Rusyns lived in a common polity called Rus’ with its centre in Kiev, which had been founded by Varangians (Vikings from eastern Sweden) to secure the Dnieper waterway from the Baltic to Constantinople (cf. Primary Chronicle, sub anno 6370 [AD 862], Cross and Sherbowitz-Wetzor 1953: 59–60). However, the lingua franca of the polity was East Slavic, and the thin Varangian upper class gradually Slavicized as well (their names are Germanic at the beginning – Rjurik = Hrørekr, Askol’d = Haskuldr, Ol’ga = Helga, Igor’ = Ingvarr, Oleg = Helgi, Glěb = Guðleifr (cf. Vasmer 1953–8) – but later Slavic: Svjatoslav, Jaropolk, Izjaslav, Svjatopolk, Jaroslav, Mstislav, Vsevolod, Vjačeslav, etc.). After Grand Prince Volodiměr (or Vladimir) had been baptized in 988 and had turned Rus’ into a Christian country, the Common East Slavic language (Russian drevnerusskij jazyk ‘Old Russian language’, Ukrainian davn’oukrajins’ka mova ‘Old Ukrainian language’, Belarusian staražytnabelaruskaja mova ‘Old Belarusian language’) came to be written in Cyrillic letters, and it was normalized on the basis of the chancellery language of Kiev. Throughout the Middle Ages there was a diglossic situation with Common East Slavic as Low variety (L) and Church Slavonic as High variety (H) (Worth 1978; Uspenskij 1987). Consequently, the orthography of Common East Slavic was also based on Church Slavonic. This orthography was used all over Rus’, including the chancelleries and monasteries of Novgorod. However, Novgorod had a special status within Rus’. It was the centre of by far the largest principality, the Novgorodskaja zemlja ‘Land of Novgorod’, and from 1136 to 1478 of the so-called Novgorod Republic (Novgorodskaja respublika), which constituted roughly the northern half of the traditionally Slavic territories that later became Muscovy and then Russia, reaching from Pskov and its surroundings to the

32

Tyranny of Writing: Ideologies of the Written Word

White Sea and to the northern Urals. Novgorod maintained close trade ties across the Baltic Sea, for which there were two foreign trading posts in the town: Gotenhof ‘Yard of the Goths’ (Russian Gotskij dvor ‘Gothic Yard’, named after the Swedish island of Gotland) and Peterhof ‘Peter’s Yard’ (Russian Petrovo podvor’e ‘Peter’s Yard’ or Nemeckij dvor ‘German Yard’). From the thirteenth century on Novgorod even was one of the four kontors of the Hanseatic League. Consequently, Novgorod had great autonomy within Rus’ and was not affected by the ‘Tatar yoke’ (the Mongol invasion of and suzerainty over large parts of Rus’, 1237–1480). After a long period of conflict, Novgorod was finally conquered by Muscovy in 1478, which ended its autonomy. A specific feature of medieval Novgorod is a special kind of archaeological find: people in Novgorod used strips of birch bark to write mundane messages. Since 1951, 1,087 such birch-bark documents, which were written between the mid-eleventh and the mid-fifteenth centuries, have been found in Novgorod (see, for example, Gippius 2004; Janin, Zaliznjak and Gippius 2015; Gippius and Zaliznjak 2016). In the context of this book, the most interesting fact about the birch-bark letters is that many of them were written in a different orthography than religious books, chronicles and other texts written on parchment in monasteries and chancelleries (see Bunčić 2016 for an exhaustive treatment of the correlation between the two orthographies). In this chapter I argue that this special orthography of the Novgorod birch-bark letters, though originally resulted from incomplete spelling instruction, constituted a systematic counter-norm that was actively promoted against the tyranny of the standard orthography on the basis of associations with different social values. To this end, I first present the two orthographies, then describe another situation of an in-group spelling that arose in the 1990s for comparison, discuss the origin and spread of the special orthography of the birch-bark letters, and finally demonstrate the role that the tyranny of writing plays for the establishment of this orthography.

The two orthographies When the first birch-bark letters were found, it was assumed that the writers of the letters could not spell properly because of their lack of schooling, but later it was understood that the deviations from the standard spelling constituted a system of its own (cf. Žukovskaja 1959: 100–8, Zaliznjak 1986: 93, 104). Zaliznjak (2004: 21–2) calls the spelling that the traditional texts from Novgorod share with the texts from Kiev and other places of Rus’ standartnaja orfografija ‘standard orthography’, and the deviating system used in many birch-bark documents bytovaja orfografija ‘everyday orthography’ – I will call it vernacular orthography.

How to Write a Birch-Bark Letter

33

The difference between these two spelling systems can be seen in Figures 2.1 and 2.2. Figure 2.1 (from Dietze 1971: 51r) shows a typical text written in standard orthography, the First Novgorod Chronicle, which records important events for posterity: Въ лѣⷮ ·҂ꙅ҃·ѱ҃ · иде кн҃ ꙁь ꙗрослаⷡ пльсковоу на петровъ дн҃ ь · а новъгородьци въмале · а самъ седе на пльскове · а дворъ свои пославъ съ пльсковїци воеватъ · и шьдъше въꙁѧша городъ медвежю головоу · и пожьгоша · и придоша сторовї (Dietze 1971: 51r)

In the year 6700 [= AD 1192]. Prince Yaroslav went on St. Peter’s Day to Pskov with a few men of Novgorod; and himself took his seat at Pskov, having sent his court with the men of Pskov to make war; and having gone they took the town of Medvež’ja Golova, burned it, and came back well. (Michell and Forbes 1914: 35)

Figure 2.1 From the Novgorod First Chronicle (2nd half of 13th century.)

34

Tyranny of Writing: Ideologies of the Written Word

Figure 2.2 (tracing (proris’) by Arcixovskij and Borkovskij 1958, from , 18 April 2017; license: obligatory attribution) shows a typical business letter on birch bark, spelt according to the vernacular norm and dealing with rather more mundane affairs (in all the following quotations of birch-bark documents the letters that attest to the vernacular orthography are underlined; for a transformation into standard spelling, 〈о〉 would have to be replaced by 〈ъ〉, 〈е〉 by 〈ь〉, 〈ъ〉 by 〈о〉 and 〈ь〉 by 〈е〉 or 〈ѣ〉): ⁘ ѿ есифа · к ънфиму · чтъ пришле ѿ маркь к тобѣ людии ѡл ькса · или къ жене мъѥи · ѿвѣцаи ѥму · такъ · какъ ѥси дъкънчалъ марке съ мнъю · мнѣ выѥхати на петръво дн҃ е · к тобѣ и росмътрити сьла своѥгъ · тъбѣ ръже свъѧ снѧти

From Esif to Onfim. If Oleksa sends people from Mark to you or my wife, answer him [i.e. Mark] as follows: ‘What deal did you make with me? I have to ride to you on St. Peter’s Day and inspect my hamlet, and you have to harvest your rye.’

The main difference between the spelling systems lies in the use of the letters 〈о〉, 〈ъ〉, 〈е〉 and 〈ь〉 (and sometimes, as in this document, also 〈ѣ〉; see Zaliznjak 1986: §5–24 for a detailed account of the subsystems of the vernacular orthography). In the standard orthography each of these letters directly corresponds to a phonological value: 〈о〉 to /o/, 〈ъ〉 to /∅/ (zero), 〈е〉 to /e/, 〈ь〉 to /ʲ∅/ (the palatalization of the preceding consonant) and 〈ѣ〉 to /ě/ (an [e]-like phoneme with unclear pronunciation, which is traditionally called jat’). In contrast to this, the vernacular orthography treats 〈о〉 and 〈ъ〉 as well as 〈е〉 and 〈ь〉 (and partially also 〈ѣ〉) as allographs, that is, they are either used in free variation, or one is replaced by the other (frequently 〈ъ〉 is not used at all, being completely replaced by 〈о〉). This results in a much less direct representation of the pronunciation in the vernacular orthography: Here the grapheme 〈о ~  ъ〉 represents either /o/ or zero, and the grapheme 〈е ~  ь (~  ѣ)〉 can stand for any of the two (or three) sound values /e/ or /ʲ∅/ (or /ě/). Figure 2.3 shows in direct comparison how the expression na Petrov den’ ‘on St. Peter’s day [29 June]’ (which goes back to Common Slavic *na Petrovъ dьnь) is spelt in the two documents shown in Figures 2.1 and 2.2: In the First Novgorod Chronicle, /ʲe/ is represented by

Figure 2.2 Novgorod birch-bark letter №142 (beginning of 14th c.)

How to Write a Birch-Bark Letter

35

Figure 2.3 na Petrov d(e)n’ ‘On St. Peter’s day’ 〈е〉, /o/ by 〈о〉, /∅/ by 〈ъ〉, and /ʲ∅/ by 〈ь〉. In the same three words from the document on birch bark, /o/ is spelt as 〈ъ〉, /∅/ as 〈о〉, and both /ʲe/ and /ʲ∅/ are represented by 〈е〉. Note that in this respect the standard orthography is more ‘phonetic’ than the rather abstract vernacular spelling (cf. Zaliznjak 1986: 109).

A modern parallel The phenomenon of the birch-bark letters has sometimes been compared to modern ways of electronic communication, for instance by Andrej Zaliznjak, who compares birch-bark letters to SMS text messages: Сейчас действительно общество вернулось к обмену информации немножко похожему на обмен берестяными грамотами в древности; неожиданно, после всех гораздо более развитых форм коммуникации до нашего времени, тексты и по длине, и по типу содержания и часто по повторяемости сообщений очень похожи. (Kapica 2009: 1m35s)

Now society has really returned to an exchange of information which is a bit similar to the exchange of birch-bark letters in ancient times; unexpectedly, after all the much more sophisticated forms of communication before our time, the texts are very similar with respect to length, type of content and often also recurrence of messages.

While the circumstances are of course not identical, the mundane content of both birch-bark letters and text messages is caused by the higher availability of birch bark in contrast to parchment and by the higher speed with which text messages are written and transmitted. The briefness of both kinds of communication is necessitated by the space restrictions of a piece of birch bark and by the size of an SMS message of 140 bytes (which results in a maximum text length of 160 characters in the standard 7-bit GSM character set, 140 characters in a custom 8-bit character set or 70 characters in 16-bit Unicode, which is needed to reliably display languages like Russian, Arabic or Japanese, for which therefore Latin transcriptions are often used; cf. Birzer 2004). However, a more striking parallel to the orthography of the Novgorod birch-bark letters is the orthography of the so-called Padonki. The Padonki (from Russian podonki ‘scum’) were a subculture on the Russian internet (Runet) in the late 1990s and early 2000s. They used their own orthography for Russian, which constituted the main element of their in-group slang (which of course also included specific words and idioms); this slang is often called olbanskij ‘Olbanian’

36

Tyranny of Writing: Ideologies of the Written Word

(e.g. Krongauz 2013; Scharlaj 2016; Burkhart and Schmidt 2009). The orthography was characterized by the deliberate misspelling of words in such a way that the spelling (usually) still reflected the correct pronunciation: ●

●

●

●

●

Russian has vowel reduction in unstressed vowels, which results in stressed /a/ and /o/ merging to [ʌ ~ ə] in unstressed position after a non-palatalized consonant and in stressed /a/, /o/, /e/ and /i/ merging to [ɪ] in unstressed position after a palatalized consonant. Consequently, Standard Russian 〈албанский (albanskij)〉 ‘Albanian’ is spelt 〈олбанский (olbanskij)〉, retaining the pronunciation [ʌɫˈbaˑnskʲɪj], 〈подонки (podonki)〉 [pʌˈdɔˑnkʲɪ] ‘scum’ is re-spelt 〈падонки (padonki)〉, 〈министерство (ministerstvo)〉 [mʲɪnʲɪˈsʲtʲɛˑrstvə] ‘ministry’ can turn into 〈менестерства (menesterstva)〉, etc. Russian obstruents are devoiced at the end of words, and vowel clusters exhibit regressive assimilation of voice. Consequently, this neutralized voice contrast is exploited for the slang orthography: 〈все (vse)〉 [ˈfsʲɛˑ] ‘all’ can turn into 〈фсе (fse)〉, 〈привет (privet)〉 [pɾʲɪˈvʲɛˑt] ‘hello’ into 〈превед (preved)〉, etc. The vowel letters 〈я (â/ja)〉, 〈ё (ë/jo)〉 and 〈ю (û/ju)〉 can be decomposed into 〈й (j)〉 + 〈а (a)〉, 〈о (o)〉 and 〈у (u)〉, for example, 〈яд (âd)〉 [ˈjaˑt] ‘poison’ into 〈йад (jad)〉, or – as a spelling reflecting all three rules given so far – 〈язык (âzyk)〉 [jɪˈzɨˑk] ‘language’ into 〈йезыг (jezyg)〉, etc. Geminate consonant letters are often pronounced as single consonant sounds, and the adjectival genitive singular masculine/neuter endings 〈ого (ogo)〉 and 〈его (ego)〉 are pronounced with a /v/ in the place of 〈г (g)〉. Therefore, for example, 〈русского языка (russkogo jazyka)〉 [ˈruˑskəvə jɪzɨˈkaˑ] ‘of the Russian language’ can turn into 〈рускава йезыка (ruskava jezyka)〉. The slang spelling has a preference for the low-frequency letter 〈ф (f)〉 and for geminates, which results in spellings like 〈аффтар (afftar)〉 for 〈автор (avtor)〉 [ˈaˑftər] ‘author’ (which is also reminiscent of outdated Latin transcriptions of Russian names like 〈Prokofieff〉 for 〈Прокофьев (Prokof ’ev)〉, 〈Rachmaninoff〉 for 〈Рахманинов (Raxmaninov)〉, etc.) or 〈хочецца (xočecca)〉 for 〈хочется (xočetsja)〉 [ˈxɔˑtʃ͡ ʲɪts͡ ːə] ‘would like to’.

There are several more ways in which words can be deliberately misspelt that are regularly exploited in this slang orthography. All these are options that can but need not be followed when writing in the ‘Olbanian’ orthography. Consequently, this orthography allows for great variation, and authors sometimes use alternative spellings of the same word within the same text. However, its particular characteristics make this orthography recognizably different from unintentional misspellings by uneducated people or children. Obviously, this is an instance of heterography (andersschreiben) as outlined by Schuster and Tophinke (2012), which was created deliberately as an in-group code. In the terminology introduced by Coseriu (1952), the members of this subculture deliberately use the possibilities of the orthographic sistema (according to which, e.g., [ʌ] can be spelt as 〈а (a)〉 or 〈о (o)〉 but not 〈ж (ž)〉) to deviate from the orthographic norma (which determines for each word form whether it has to be spelt with 〈а (a)〉 or 〈о (o)〉). Consequently, the target of the Padonki is the tyranny of the orthographic norm, not of the writing system or of writing as such.

How to Write a Birch-Bark Letter

37

Just as with the vernacular orthography of ancient Novgorod, there is a direct conditional relationship between the medium (birch bark, electronic visual display) and the form (vernacular orthography, ‘Olbanian’). Keeping this parallel in mind, let us now return to Novgorod.

Origin and spread of the Novgorod vernacular orthography As Zaliznjak (1986: 100–9; 2002: 607–11) has shown, the vernacular orthography was used in the Novgorod birch-bark letters since the middle of the eleventh century, but it became more widespread in the course of the twelfth century. Throughout the thirteenth century it was virtually unchallenged: В средний период (кон. XII – рубеж XIII и XIV вв.) … системы со смешением ъ, ь и о, е не просто преобладают – они являются нормой для берестяной письменности. (Zaliznjak 1986: 103; bold type added)

In the middle period (from the end of the twelfth to the turn of the fourteenth century) … the systems with a confusion of 〈ъ〉, 〈ь〉 with 〈о〉, 〈е〉 do not just predominate – they are the norm for birch-bark literacy.

The fourteenth century then saw a gradual decline of the vernacular orthography, but even among the very last birch-bark letters from the fifteenth century there are still a few using the vernacular spelling (cf. Figure 2.4 after Zaliznjak 2004: 25). Unfortunately, we know very little about the people who first used the new spelling system. Nonetheless, Zaliznjak (1986: 103–5) has provided a possible explanation of how it might have evolved. In order to understand this explanation, we need some knowledge about East Slavic historical phonology. Originally, each of the following five letters had represented a distinct vowel phoneme: 〈о〉 stood for Common Slavic o (