Language and History, Linguistics and Historiography: Interdisciplinary Approaches (Studies in Historical Linguistics) [New ed.] 9783034307611, 3034307616

Citation preview

ISBN 978-3-0343-0761-1

www.peterlang.com

Peter Lang

STUDIES IN HISTORICAL LINGUISTICS VOL. 9

Wim Vandenbussche is Professor of Dutch Linguistics at the Vrije Universiteit Brussel. He teaches courses on Dutch and Germanic language history, as well as on various aspects of sociolinguistics. His research is situated in the domain of historical sociolinguistics, with particular attention to the language situation in Flanders during the eighteenth and nineteenth centuries.

Studies in Historical Linguistics

Steffan Davies is Lecturer in German at the University of Bristol. He studied History and German at the University of Oxford, writing his doctoral thesis on the literary, cultural and historiographical treatment of Albrecht von Wallenstein in the ‘long nineteenth century’.

Language and History, Linguistics and Historiography

Nils Langer is Reader in German Linguistics at the University of Bristol. His primary research interests lie in the area of historical sociolinguistics and he is currently working on language contact in Schleswig-Holstein in the nineteenth century.

Langer, Davies and Vandenbussche (eds)

What are the points of contact between the study of language and the study of history? What are the possibilities for collaboration between linguists and historians, and what prevents it? This volume, the proceedings of an international conference held at the University of Bristol in April 2009, presents twenty-two articles by linguists and historians, exploring the relationship between the fields theoretically, conceptually and in practice. Contributions focus on a variety of European and American languages, in historical periods from the Middle Ages to the present day. Key themes at the intersection of these two disciplines are the standardization and classification of languages, the social and demographic history of medieval and early modern Europe, the study of language and history ‘from below’, and the function of language in modern politics. The value of interdisciplinary collaboration is demonstrated in a wide-ranging set of case studies, on topics including language contact in Northern and Central Europe, the relationship between peninsular and transatlantic Spanish, and new approaches to the recent histories of Nicaragua, Luxembourg and Bulgaria. The volume seeks out the interdependencies between the two fields and asks why exchanges between linguists and historians remain the exception rather than the rule.

Nils Langer, Steffan Davies and Wim Vandenbussche (eds)

Language and History, Linguistics and Historiography Interdisciplinary Approaches

Peter Lang

 

www.peterlang.com

Peter Lang

STUDIES IN HISTORICAL LINGUISTICS VOL. 9

Wim Vandenbussche is Professor of Dutch Linguistics at the Vrije Universiteit Brussel. He teaches courses on Dutch and Germanic language history, as well as on various aspects of sociolinguistics. His research is situated in the domain of historical sociolinguistics, with particular attention to the language situation in Flanders during the eighteenth and nineteenth centuries.

Studies in Historical Linguistics

Steffan Davies is Lecturer in German at the University of Bristol. He studied History and German at the University of Oxford, writing his doctoral thesis on the literary, cultural and historiographical treatment of Albrecht von Wallenstein in the ‘long nineteenth century’.

Language and History, Linguistics and Historiography

Nils Langer is Reader in German Linguistics at the University of Bristol. His primary research interests lie in the area of historical sociolinguistics and he is currently working on language contact in Schleswig-Holstein in the nineteenth century.

Langer, Davies and Vandenbussche (eds)

What are the points of contact between the study of language and the study of history? What are the possibilities for collaboration between linguists and historians, and what prevents it? This volume, the proceedings of an international conference held at the University of Bristol in April 2009, presents twenty-two articles by linguists and historians, exploring the relationship between the fields theoretically, conceptually and in practice. Contributions focus on a variety of European and American languages, in historical periods from the Middle Ages to the present day. Key themes at the intersection of these two disciplines are the standardization and classification of languages, the social and demographic history of medieval and early modern Europe, the study of language and history ‘from below’, and the function of language in modern politics. The value of interdisciplinary collaboration is demonstrated in a wide-ranging set of case studies, on topics including language contact in Northern and Central Europe, the relationship between peninsular and transatlantic Spanish, and new approaches to the recent histories of Nicaragua, Luxembourg and Bulgaria. The volume seeks out the interdependencies between the two fields and asks why exchanges between linguists and historians remain the exception rather than the rule.

Nils Langer, Steffan Davies and Wim Vandenbussche (eds)

Language and History, Linguistics and Historiography Interdisciplinary Approaches

Peter Lang

Language and History, Linguistics and Historiography

Studies in Historical Linguistics Vol. 9 Edited by Dr Graeme Davis, Karl A. Bernhardt & Dr Mark Garner

PETER LANG Oxford • Bern • Berlin • Bruxelles • Frankfurt am Main • New York • Wien

Nils Langer, Steffan Davies and Wim Vandenbussche (eds)

Language and History, Linguistics and Historiography Interdisciplinary Approaches

PETER LANG Oxford • Bern • Berlin • Bruxelles • Frankfurt am Main • New York • Wien

Bibliographic information published by Die Deutsche Nationalbibliothek Die Deutsche Nationalbibliothek lists this publication in the Deutsche National­ bibliografie; detailed bibliographic data is available on the Internet at http://dnb.d-nb.de. A catalogue record for this book is available from the British Library. Library of Congress Cataloguing-in-Publication Data: Language and history, linguistics and historiography : interdisciplinary approaches / Nils Langer, Steffan Davies and Wim Vandenbussche (eds). p. cm. -- (Studies in Historical Linguistics; 9) Includes bibliographical references and index. ISBN 978-3-0343-0761-1 (alk. paper) 1. Historical linguistics. 2. Language and culture. I. Langer, Nils, 1969- II. Davies, Steffan. III. Vandenbussche, Wim. P140.L24 2011 417'.7--dc23 2011037901

ISSN 1661-4704 ISBN 978-3-0343-0761-1

E‐ISBN 978‐3‐0353‐0230‐1

© Peter Lang AG, International Academic Publishers, Bern 2012 Hochfeldstrasse 32, CH-3012 Bern, Switzerland [email protected], www.peterlang.com, www.peterlang.net All rights reserved. All parts of this publication are protected by copyright. Any utilisation outside the strict limits of the copyright law, without the permission of the publisher, is forbidden and liable to prosecution. This applies in particular to reproductions, translations, microfilming, and storage and processing in electronic retrieval systems. Printed in Germany

Contents

List of  Figures

ix

List of  Tables

xi

1



Language and History, Linguistics and Historiography: Theoretical Outlook and Methodological Practices

1

Steffan Davies, Nils Langer and Wim Vandenbussche

Language and History, Linguistics and Historiography: Interdisciplinary Problems and Opportunities

3

Patrick Honeybone

History and Historical Linguistics: Two Types of  Cognitive Reconstruction?

15

Nicholas M. Wolf

History and Linguistics: The Irish Language as a Case Study in an Interdisciplinary Approach to Culture

49

Brian D. Joseph

Historical Linguistics and Sociolinguistics: Strange Bedfellows or Natural Friends?

67

Nicola McLelland

From Humanist History to Linguistic Theory: The Case of  the Germanic Rootword

89

vi

Agnete Nesse

Editorial Practices and Language Choice: ‘Low German Language Monuments’ in Norway

111

2

127

Standardization and Authenticity

Robert Evans

Of ficial Languages: A Brief  Prehistory

129

Tomasz Kamusella

Classifying the Slavic Languages, or the Politics of  Classification

147

José del Valle

Linguistic History and the Development of  Normative Regimes: The Royal Spanish Academy’s Disputed Transatlantic Authority

175

Juan R. Valdez

Colouring Language: Pedro Henríquez Ureña’s Representations of  Spanish and Dominican Identity

193

Laura Villa

‘Because When Governments Speak, They Are Not Always Right’: National Construction and Orthographic Conf licts in Mid-Nineteenth-Century Spain 209 Gijsbert Rutten and Rik Vosters

As Many Norms as There Were Scribes? Language History, Norms and Usage in the Southern Netherlands in the Nineteenth Century 229

vii



Anneleen Vanden Boer

Language and Nation: The Case of  the German-Speaking Minority in Belgium

255

3

269

Demographics and Social Dynamics

Richard Ingham

The Decline of  Bilingual Competence in French in Medieval England: Evidence from the PROME Database

271

Rembert Eufe

Merovingian Coins and Their Inscriptions: A Challenge to Linguists and Historians

293

Remco Knooihuizen

The Use of  Historical Demography for Historical Sociolinguistics: The Case of  Dunkirk

323

4 Language History from Below

341

Judith Nobels and Marijke van der Wal

Linking Words to Writers: Building a Reliable Corpus for Historical Sociolinguistic Research 343 Helmut Graser and B. Ann Tlusty

Sixteenth-Century Street Songs and Language History ‘From Below’ 363

viii

Juan Manuel Hernández-Campoy

Mood Distinction in Late Middle English: The End of  the Inf lectional Subjunctive

389

5

407

Language and Ideology

Lisa Carroll-Davies

Identifying the Enemy: Using a CDA and Corpus Approach to Analyse Sandinista Strategies of  Naming

409

Krassimir Stoyanov

Ritualized Slogan Lexis in the Bulgarian Press during the Times of  Violent Contradiction in Ideologies (1944–1947)

429

Kristine Horner and Melanie Wagner

Remembering World War II and Legitimating Luxembourgish as the National Language: Consensus or Conf lict?

447

Michela Giordano and Federica Falchi

John Stuart Mill and Salvatore Morelli: Language as a Social Tool in Nineteenth-Century Britain and Italy

465

Notes on Contributors

483

Index

489

List of  Figures

Figure 1 The idol Tuisco (Verstegan 1605: 71; Warburg Institute) Figure 2 The frontispiece to Stieler (1691) Figure 3 From the dialogue De dudesche unde de Norman, Ms 1418, Bergenfahrerarchiv, City Archive, Lübeck Figure 4 Front page of  Ms. Gl.kgl.S.2836, Royal Library, Copenhagen Figure 5 A. Schleicher’s genealogical tree of  the Indo-Germanic (Indo-European) languages (1873: diagram after p. 33)

161

Figure 6 Percentage occurrence of  feminine nouns overall, with del/du and al/au

282

Figure 7 Percentage of occurrence of  feminine nouns with de and à, and with combinations ending in -l and those ending in -u

282

Figure 8 NeighborNet diagram showing the similarity of  Fort-Mardyck French to other French varieties

335

Figure 9 Map of  the Dunkirk area, with places mentioned in the text

336

Figure 10 The explicit reference to the writing process in letter 3–1–2008, 129–130, letters-as-loot-corpus Figure 11 Letter by Claeijs Pietersen, letters-as-loot-corpus Figure 12 Letter by Jan Lievensens, letters-as-loot-corpus Figure 13 An example of a signature in a larger handwriting than the body of  the letter, letters-as-loot-corpus Figure 14 An example of a signature in a dif ferent and less experienced hand­ writing than the body of  the letter, letters-as-loot-corpus Figure 15 Jacob van de Velde’s handwriting and signature in letter 3b-1–2008, 131–132, letters-as-loot-corpus

x

List of  Figures

Figure 16 Jacob van de Velde’s handwriting and signature in a receipt kept in the archive of  the Audit Of fice of  Zealand Figure 17 Flow chart to determine the autographical status of  letters Figure 18 The weaver’s song (Stadtarchiv Augsburg, Reichsstadt, Urgichten, 13.09.1591, Peter Müller/Matthäus Sporhan) Figure 19 The tailor’s song (Stadtarchiv Augsburg, Reichsstadt, Urgichten, 28.06.1594, Christof  Halbritter) Figure 20 Percentages of usage of was and were in indicative and subjunctive in singular contexts

399

Figure 21 Percentages of usage of was per singular subject type and mood

399

Figure 22 Percentages of usage of was per plural subject type and mood

401

List of  Tables

Table 1 Consonant alternations in Dutch spelling, compared with German 105 Table 2 Dual classification of  Slavic

154

Table 3 Elaborated dual classification of  Slavic

157

Table 4 Revised triple classification of  Slavic I

157

Table 5 Revised triple classification of  Slavic II

158

Table 6 Revised triple classification of  Slavic III

159

Table 7 Dual-cum-quadruple classification of  Slavic

163

Table 8 Spelling features and the normative framework

242–243

Table 9 Southern (left) and Northern (right) spelling variants per feature (distribution per cent) in a digitized corpus of  handwritten documents from the Southern Netherlands

246

Table 10 Spelling variants per province (distribution per cent)

247

Table 11 Correlation between mother tongue and the sense of af filiation to Belgium

262

Table 12 Frequency distribution for the variables indicating the sense of af filiation to Belgium 262–263 Table 13 Frequency distributions for the language command

264

Table 14 Correlation between language command and the sense of af filiation to Belgium

265

Table 15 Frequencies of masculine and feminine nouns used with masculine preposition + article forms in PROME, 1280–1399 281 Table 16 Frequency of masculine and feminine noun types co-occurring with masculine premodifier forms cel, cest, nul and tut, late thirteenth century/early fourteenth century and 1370s 283–284

xii

List of  Tables

Table 17 Frequencies of del with masculine and feminine nouns, 1280–1324 and 1377–1399 285 Table 18 Locals and immigrants in the Dunkirk marriage register, by gender and sample year

330

Table 19 Distribution of  languages across marriage partners 1647–1697

330

Table 20 Proportions of  French speakers among the population of  Dunkirk, overall and by gender, and the preference for mono-ethnic marriage in the sampled years 332 Table 21 Subject type

393

Table 22 Levelling to was/were in contexts of  Indicative Mood

397

Table 23 Levelling to was/were in contexts of  Subjunctive Mood

398

Table 24 FSLN corpus documents

416

Table 25 Envío corpus contents

417

Part 1

Language and History, Linguistics and Historiography: Theoretical Outlook and Methodological Practices

Steffan Davies, Nils Langer and Wim Vandenbussche

Language and History, Linguistics and Historiography: Interdisciplinary Problems and Opportunities

For the past few years, the notion of interdisciplinarity has been a buzz­ word to be found in any programmatic research outline or grant proposal, understood to be a vital break away from scholarly isolation and too narrow a focus on one’s own methodology and research questions. This pertains to the humanities just as much as to the natural sciences, and it is hardly a new idea. The historian Marc Bloch stated in the 1940s that ‘it is indispensable that the historian possess at least a smattering of all the principal techniques of his trade’ (Bloch 1992: 57). More specifically with regard to the disciplines of linguistics and History, Bloch exclaims: ‘What an absurd illogicality that men who half the time can have access to their subject only through words, are permitted, among other deficiencies, to be ignorant of  the fundamental attainments of  linguistics’ (ibid.).1 Yet in practice, exchange between the two fields has often remained a desideratum rather than actual achievement. The fragmentation of traditional philology into a broad subset of highly specialized branches of linguistics from the 1960s (computational linguistics, feminist linguistics, language acquisition studies, etc.), all of which tended to focus on present-day language use only, meant that the study of  language structure and texts in their historical context lost the prominence it had enjoyed since its emergence as a serious scholarly discipline in the early nineteenth century. The ‘linguistic turn’ much discussed by historians since the 1970s has produced, despite its name, 1

To distinguish between the two meanings of  history, namely History as a scholarly discipline versus history to refer to the past, we use a capital letter for the former. Since no such problem arises to distinguish between language and linguistics, we do not use a capital letter for linguistics.

4

Steffan Davies, Nils Langer and Wim Vandenbussche

little engagement, critical or otherwise, with the discipline of linguistics as most linguists know and practise it, focusing instead on the mediate nature of access to the past, and the ‘literary’ treatment of that past in the present (summary in Fulbrook 2002: 18–24). Exceptions to this mutual isolation can be found on both sides of  the disciplinary fence, of course, but largescale collaboration between our two disciplines has continued to be rare, not least because scholars lack the confidence to step outside their area of expertise and into territories where basic assumptions, technical terms and methodological practices may be alien. In particular, there is a general lack of opportunity to exchange ideas and basic knowledge of core ideas. This book is an attempt to facilitate such an opportunity. It contains studies first presented at a conference at the University of  Bristol in April 2009, generously supported by a Scientific Network Grant from the Arts and Humanities Research Council (AHRC), which explicitly addressed issues of interdisciplinarity, that is, to what extent there is any actual scope for collaboration, common knowledge of ‘tools’, or exchange of ideas between linguists and historians.2 First steps towards a break away from disciplinary isolation had already been accomplished within the field of  historical sociolinguistics over the last ten years. The aim of  key conferences in historical sociolinguistics (Shef field 2001, Bristol 2003, Bristol 2005, and Bruges 2006), edited volumes (Linn & McLelland 2002, Deumert & Vandenbussche 2003, Langer & Davies 2005, Elspaß et al. 2007), and the thinking behind the foundation of  the Historical Sociolinguistics Network (HiSoN, since 2005) has been to achieve a much greater awareness of similarities and dif ferences in the sociological histories of related and neighbouring languages, at first mostly restricted to Germanic languages for practical reasons. Success in bringing separate linguistic traditions and languages together created the impetus for branching out further, to establish how a direct exchange between pure historians and historical sociolinguists may provide further 2

This is an opportunity to give formal thanks to the AHRC, the University of Bristol, the Vrije Universiteit Brussels for their financial support, as well as our student helpers Evi Egger, Catherine Hillier, Stefanie Hundehege, Nadia Taylor, and, in particular, Stef fi John. Special thanks also to Sabine Lehmann, Juliane Fischer, and Patrick Beuge for their help with editing this conference volume.

Language and History, Linguistics and Historiography

5

benefits for our understanding of human societies. Yet the task of fostering interdisciplinarity between linguists and historians is hard. There is much less understanding of what the other discipline is doing, partly because of physical limitations: there are a limited number of  hours in the day, and for many academics, the pressure to publish frequently leaves little time to learn about a whole new scholarly discipline, on the of f-chance that it might enrich their findings. Furthermore, we suspect, scholars feel most comfortable in the areas in which they were trained. In response to such challenges, the authors in this volume have written for a readership not necessarily expert in their discipline. Language is a primary means of access to the past, and the historian’s primary means of expression. The centrality of language as a research tool is most acutely clear to historians who work on countries other than their own. Most countries require their History students to have a good understanding of two foreign languages at least. In the United Kingdom, where this is not the case, Richard Evans, a historian of Germany, warned, upon taking up the Regius Chair of  Modern History at Cambridge, that: ‘The rapid and continuing decline of  language-learning in British schools is in many ways the most important single factor threatening the continuation of the long tradition of British historians’ engagement with the European Continent’ (Evans 2009: 201). But ‘domestic’ historians, too, can ill af ford to neglect foreign languages. Significant aspects of the history of the British Isles, too – requiring knowledge of  Latin, Middle English, French, or the Celtic languages – will remain a closed book to the monoglot reader of modern English; the historian of modern East-Central Europe will sooner rather than later encounter sources in German or Russian. The growth of immigrant communities in twentieth-century Europe may require the historians of  the future to learn a new set of  foreign languages. Historians need to master languages, but, as Robert Evans argued in his inaugural lecture to a Regius Chair of  Modern History, this time at Oxford (Evans 1998), they also need an appreciation of language that goes well beyond what simple words mean. In order to interpret their sources, historians need to understand discourse analytic methods, the impossibility of reading a text without context, and the sociolinguistic dynamics involved when writers use particular linguistic varieties and variants.

6

Steffan Davies, Nils Langer and Wim Vandenbussche

Such a list may, of course, sound like yesterday’s news. E.H. Carr’s What is History?, a staple of  first-year university courses on historical method, pointed out a good half-century ago that historical facts ‘are always refracted through the mind of the recorder’, and noted: ‘The very words which [the historian] uses – words like democracy, empire, war, revolution – have current connotations from which he cannot divorce them’ (Carr 1964: 22, 24–25). Thus an interpretation of historical sources requires a sophisticated understanding of how language works socially – be this in terms of lexical semantics, historical pragmatics, or text linguistics. Providing the source with context is a fundamental, natural historical task (Evans 1997: 103–107). This volume, however, argues for more profound and explicit engagement on the historian’s part with the methods and findings of  linguistics, and demonstrates the potential results of such work. On a pragmatic level, such engagement adds detail and guards against errors. Simple f luency in a living or dead language is arguably insuf ficient to unearth and interpret correctly the linguistic thrusts and nuances contained in a historical text. Where texts are written in a non-standard variety, an understanding of sociolinguistics is crucial for their correct contextualization, since not all non-standard variants are used by the prototypical uneducated speaker. Spelling variation in the seventeenth century carried a dif ferent social association in periods before the existence of a standard language than today; Macha (2004) shows how particular spelling conventions demarcated Catholics and Protestants in a corpus of war correspondence from the sixteenth century, whilst no social stigma was attached to the use of non-standard spelling then. Analysing sixteenth-century spelling variation enables us, following Macha, to say something about the author’s or printer’s deliberate or conventional positioning during the time of  the confessional wars. Without understanding the sociological significance of such linguistic markers, historical textual interpretation would arguably be naïve at best, and skewed or simply wrong at worst. Not only language, but also linguistics, belongs in the historian’s toolbag, even if it is rarely treated thus: History students in Germany are instructed that heraldry, palaeography, numismatics and sigillography are among History’s ‘ancillary sciences’ (die historischen Hilfs­wissenschaften), but there is no mention of  linguistics (e.g. Opgenoorth and Schulz 2001).

Language and History, Linguistics and Historiography

7

For some areas of  History, language and linguistics plays a still more fundamental role. One such area is the ongoing interest in national and ethnic identities and their origins: identities that emerged, as Robert Evans points out, in ‘intimate conjunction with language’ (Evans 1998: 24; for a survey, Joseph 2006). Comparing linguistic evidence from the oldest languages on the British Isles, Trudgill (2009) draws a new and provocative picture of  the extent and nature of  language contact before the Norman Conquest. Given the particular social settings and types of interaction required for significant language-contact phenomena to emerge, Trudgill is able to present historians and archaeologists with a linguistically motivated account of living conditions in Great Britain before 1066. Another example is Ingham (this volume) who uses statistical evidence and language-contact examples to enhance our understanding of immigration and integration in medieval Anglo-Norman England. Greater interest in groups excluded from traditional narratives, and especially in day-to-day lives and horizons, has expanded still further the range of research techniques on which the social historian relies. As John Tosh remarks: ‘There is probably no other field whose primary sources are so varied, so widely dispersed, and so uneven in quality’ (Tosh 2006: 136). Broad and sparse evidence demands a diversity of method: Bloch, the advocate of teaching linguistics to historians, co-founded the inf luential journal Annales d’histoire sociale et économique in 1929 calling for the closer integration of History with other disciplines, notably the social sciences. ‘Linguistics’ in Bloch’s lifetime meant almost exclusively systemic linguistics: sociolinguistics was yet to be born and has trailed behind the historians’ turn to studying ‘history from below’. Elspaß (2005), however, has cemented linguists’ interest in the field, and with Elspaß et al. (2007), there is now a collection of studies focused on unearthing ‘language histories from below’. Linguistics techniques function here not only as a guardian against misapprehension, but in developing historical analysis: comparing common spelling conventions and grammatical phenomena in seventeenthcentury Dutch letters is just as useful as handwriting in the identification of scribes (cf. Nobels and van der Wal, this volume). The ‘social history of  language’ is a final area in which the two disciplines are closely intermeshed: in calling for such a history, Peter Burke remarked in the late 1980s that ‘a

8

Steffan Davies, Nils Langer and Wim Vandenbussche

number of  historians have recently come to recognize the need to study language as a social institution’, and posited ‘a gap between linguistics, sociology (including social anthropology) and history […] which can and should be filled by the social history of language’ (Burke and Porter 1987: 1). Burke’s own work, and projects such as the multi-volume Hanes Cymdeithasol yr Iaith Gymraeg / A Social History of  the Welsh Language (e.g. Jenkins 1997–2000) have responded to the challenge. Similarly, linguists need History. The concentration of many linguistics curricula in European universities on present-day language study (and of sociolinguistics on contemporary language use) has regrettably implied a disregard for historical development, even though the principles and developments in language change are often most insightful for our understanding of  how language ‘works’ as such. As Robert Evans points out, the sociolinguist’s interest in language change is essentially an interest in the past (Evans 1998: 11). Honeybone (this volume) reduces the principal dif ferences between the two disciplines to the following, provocative, core definition: linguistics studies languages, and History studies people. But, he argues, these things are not unconnected: languages live in people, and much historical evidence comes from records that people wrote using their languages. Honeybone identifies areas of  historical linguistics where an understanding of social processes (historical or otherwise) are irrelevant, such as individual sound changes from one set of sounds to another (e.g. the Second Sound Shift in High German, shifting /p, t, k/ to /pf, ts, kx/ and /f, s, x/). Endogenous changes, arising from within linguistic systems themselves, may not be attributable to the social context of the people who innovate them, though as the products of past social activity, they are no less ‘historical’ for it. Exogenous changes, however, can only be explained if we understand the lives of  the speakers involved: one such case is the formation of new dialects or languages where speakers of  ‘old dialects’ mix in areas where no established dialect exists. The emerging new forms and varieties can only be understood, Honeybone claims, if we know the demographics of the speakers, requiring substantial collaboration between linguists and historians. For example, Knooihuizen (this volume) provides one such example of how a detailed demographic examination of marriage registers can inform our understanding of language contact and language loss in seventeenth-century Dunkirk. Linguists writing social histories of 

Language and History, Linguistics and Historiography

9

language use need information from historians as to when a particular text was written, who the likely author was, at what occasions texts were produced, and, very importantly, which texts were not written or did not survive. In multilingual scenarios, dif ferent languages were used but not all were written down; the availability of material resources may have had an ef fect on education, which in turn af fected literacy. The general lack of texts written in Frisian, Low German, or Sønderjysk in eighteenth- and nineteenth- century Denmark gives a distorted view of the thriving actual use of  these languages in the southern part of  the country. Seen this time from the linguist’s perspective, the relationship again functions on three levels: an essential af finity (the common study of past activity), areas of study in which the two disciplines work side by side (such as the social history of  language), and with History functioning as a linguists’ ‘tool’, where historical findings incidental to linguistics bear a direct relevance in informing its work. Two final examples show that the research methods and core wisdoms of  both disciplines are embedded in a historico-linguistic minefield. Key terms such as ‘Germany’ and ‘German’ are known to historians as slippery concepts: the political and cultural entities known as ‘German’ have shifted over centuries, and even at the founding of the Kaiserreich in 1871, there was a deeply-felt mismatch between German ethnic and national identity and the geographical boundaries of the new state. Indeed, the need for historians to learn multiple languages indicates in itself that the concept of the nation, often defined by linguistic community, is shaky. When it comes to the names for languages, however, scholars are much more ready to assume continuity across centuries, even though it is by no means clear where the boundaries between, say, Standard German and Moselle-Franconian were on the one hand, and Luxembourgish and German on the other. The assumption that we can talk about ‘the history of  German’ rests on the identification of German as a separate language – yet the beginnings of any language are always murky and the ‘birth’ of a language can be dated as either the day when a variety was significantly dif ferent from its neighbouring varieties, or the day when the speakers ‘felt’ that their linguistic variety was suf ficiently distinct to be seen as a separate language. The dif ference between ‘fact’ and perception is never clear: even key terms such as ‘language’ or ‘German’ are mired in significant confusion. This confusion cannot be resolved without

10

Steffan Davies, Nils Langer and Wim Vandenbussche

interdisciplinarity since it does not rest on misunderstanding the processes involved, but rather on the complexities and range of situations to which these terms are applied. The danger does not lie in these ‘wobbly’ foundations but rather in the fact that the terms (language, German) are used as real and hard notions in the scholarly discourse without always realizing that terms are historical creations. Kamusella (this volume) presents a similar discussion with regard to the politics of classifying Slavic languages, which in recent years have seen the politically motivated emergence of Czech and Slovak, as well as Serbian and Croatian, as distinct languages, even though their respective linguistic dif ference is no greater than that between major dialects in English or Italian. The same applies to terms which are much more clearly ‘technical’ as used by various disciplines but where the dangers – and the possibilities – lie in their dif ferent understanding by, say, linguists, historians, and literary scholars. The key notion of  ‘politeness’ in the eighteenth century will be understood dif ferently by historians and linguists. For historians, politeness concerns sociability and is a mode of social coordination; literary historians with an interest in the periodical essay genre use the concept as a discursive medium for the development of sociability. In historical sociolinguistics, on the other hand, politeness is identified as a set of attributes that are commercialized and commodified via the identification of good grammar as the way to ‘polite’ conversation and the operationalization of politeness through the medium of humiliative discourse (Fitzmaurice 2010). In turn, both interpretations are dif ferent from the notion of linguistic politeness as a key concept in modern sociolinguistics, which talks about way speakers use the concept of  ‘face’ (after Brown & Levinson 1987) to position their status as speakers in a particular discourse. Dictionary entries on politeness will provide neither the historian nor the linguist with a suf ficiently refined dif ferentiation of  the changes in the word’s meaning and usage over the centuries. Appreciating its range guards us from an over-simplified understanding of the term, and allows us to overcome the anachronism pointed out by Carr. Reinhart Koselleck’s work demonstrates how profitably an appreciation of semantics could be turned to an elucidation of  historical concepts (Koselleck 2004; also Brunner et al. 1972–1997).

Language and History, Linguistics and Historiography

11

There is thus considerable scope for benefits from a better mutual understanding of our conceptual premises, research methods and scholarly outcomes. In particular, the papers in this volume address these general areas of potential common interest: Language History from Below, with a new focus on informal sources such as pauper letters or emigrants’ diaries, which had been largely ignored in linguistics until recently yet which inform, and are informed by, research programmes working on ‘history from below’; Political Language, which investigates discourses relating to political debates and ideas and which demonstrate the importance of  linguistic analyses of political discourse in order to understand how particular terms are used and political ‘realities’ are created and transmitted in a particular community; Language Contact, which focuses on multilingual communities to identify how a particular language is used or suppressed in order to create individual and group identities. Language contact is an important field for understanding the genesis of communities and their feeling of distinctiveness, since language is often one of  the key components that separates a community from others; Historical Semantics, which examines how the meaning of words changes over time, partly, but not exclusively, due to a change in the social environment that require or permit a refocusing of  how words are used; Attitudes to Language, which investigates what people think about language and how this relates to their evaluation of particular varieties as good languages, e.g. standard languages, and bad languages, e.g. particular sociolects and regional languages; Historiography, which – in this context – is concerned in particular with two areas, namely on the one hand, how the description of linguistic phenomena has changed during the last four centuries and, on the other hand, how editorial practices have changed over the years. Practices completely acceptable and sensible for historians can obscure vital information for a linguist wishing to interpret the same source.

The volume thus contains an outline of the possibilities of contact between the two disciplines, using both case studies from cutting-edge research and more general, theoretical discussions on interdisciplinary exchange between language / linguistics on the one hand and History / historiography on the

12

Steffan Davies, Nils Langer and Wim Vandenbussche

other. This volume is intended to serve both as a platform to show where we are, and as a starting point of instruction as to how research in the ‘other’ discipline is conducted and what principal questions and problems concern them. The chapters of  this volume demonstrate how wide-ranging such issues are and how, nonetheless, a common thread in the advancement of our understanding of  human societies in the past and humans – as actors and speakers – can be discerned.

References Brown, Penelope, & Stephen Levinson. 1987. Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Brunner, Otto, Werner Conze & Reinhart Koselleck (eds). 1972–1997. Geschichtliche Grundbegrif fe. Historisches Lexikon zur politisch-sozialen Sprache in Deutschland. 8 vols. Stuttgart: Klett. Burke, Peter, & Roy Porter (eds). 1987. The Social History of  Language. Cambridge: Cambridge University Press. Carr, E.H. 1964. What is History? Harmondsworth: Penguin. Chambers, J.K., & Peter Trudgill. 2005. Dialectology. Cambridge: Cambridge University Press. Deumert, Ana, & Wim Vandenbussche (eds). 2003. Germanic Standardisations. Amsterdam: Benjamins. Elspaß, Stephan, Nils Langer, Joachim Scharloth & Wim Vandenbussche (eds). 2007. Language Histories from Below (1700–2000). Berlin, New York: Walter de Gruyter. Evans, Richard J. 1997. In Defence of  History. London: Granta. Evans, Richard J. 2009. Cosmopolitan Islanders: British Historians and the European Continent. Cambridge: Cambridge University Press. Evans, R.J.W. 1998. The Language of  History and the History of  Language: An Inaugural Lecture Delivered before the University of  Oxford on 11 May 1988. Oxford: Clarendon Press. Fitzmaurice, Susan. 2010. ‘Changes in the Meanings of Politeness in Eighteenth-Century England: Discourse Analysis and Historical Evidence.’ In: Jonathan Culpeper & Dániel Kádár (eds), Historical (Im)politeness. Bern: Peter Lang. 87–115.

Language and History, Linguistics and Historiography

13

Fulbrook, Mary. 2002. Historical Theory. London: Routledge. Jenkins, Geraint H. (ed.). 1997. The Welsh Language before the Industrial Revolution. Cardif f : University of  Wales Press. Jenkins, Geraint H. (ed.). 1998. Language and Community in the Nineteenth Century. Cardif f : University of  Wales Press. Jenkins, Geraint H. (ed.). 2000. The Welsh Language and its Social Domains 1801–1911. Cardif f : University of  Wales Press. Jenkins, Geraint H., & Mari A. Williams (eds). 2000. ‘Let’s Do Our Best for the Ancient Tongue’: The Welsh Language in the Twentieth Century. Cardif f : University of  Wales Press. Joseph, John E. 2006. ‘“The Grammatical Being Called a Nation”: History and the Construction of Political and Linguistic Nationalism.’ In: Love, Nigel (ed.), Language and History: Integrationist Perspectives. London: Routledge. 120–141. Koselleck, Reinhart. 2004. Futures Past: On the Semantics of  Historical Time. Translated by Keith Tribe. New York: Columbia University Press. Langer, Nils, & Winifred Davies (eds). 2005. Linguistic Purism in the Germanic Languages. Berlin, New York: Walter de Gruyter. Linn, Andrew, & Nicola McLelland (eds). 2001. Standard Germanic. Amsterdam: Benjamins. Macha, Jürgen. 2004. ‘Konfession und Sprache. Zur schreibsprachlichen Divergenz um 1600.’ In: Mattheier, Klaus J., & Haruo Nitta (eds), Sprachwandel und Gesellschaftswandel. Wurzeln des heutigen Deutsch. München: iudicium. 161–174. Opgenoorth, Ernst, & Günther Schulz. 2001. Einführung in das Studium der Neueren Geschichte. Paderborn etc: Schöningh. Tosh, John, with Seán Lang. 2006. The Pursuit of  History: Aims, Methods and New Directions in the Study of  Modern History. London: Pearson Longman. Trudgill, Peter. 2009. Investigations in Sociohistorical Linguistics. Stories of  Colonisation and Contact. Cambridge: Cambridge University Press.

Patrick Honeybone

History and Historical Linguistics: Two Types of  Cognitive Reconstruction?

Abstract This article compares principles and practice in history and (structural) historical linguistics. I argue that the disciplines can be both connected and distinguished by the recognition that they engage in acts of cognitive reconstruction. I show that such reconstruction is fundamental to both disciplines, but that they do it dif ferently: historical linguistics reconstructs unconscious entities, while history reconstructs at the conscious level. For these arguments to go through, certain commitments are required from the disciplines’ philosophies: mentalism of the type associated with Chomsky’s linguistics, and idealism of the type associated with Collingwood’s history. Although cognitive reconstruction is important to both areas of study, it is not all they do: to provide the context for my arguments, I also consider a number of other connections and distinctions between the disciplines, in terms of  the questions that they can ask, the evidence available to them, and their relationships to synchrony and diachrony.

1 Introduction The academic disciplines of  history and historical linguistics clearly have some things in common – they both deal with aspects of  the past, after all. There is also much that dif ferentiates them, however, in their aims and methodologies, and in their intellectual context and traditions. In this article I argue that we should recognize one particular point which both connects and distinguishes the disciplines, in ways which are not commonly

16

Patrick Honeybone

discussed (and I consider a few other things which link or dif ferentiate them, too).1 The main point has to do with the very nature of their objects of enquiry, and it will require us to entertain a set of controversial but compelling assumptions about the philosophy of the disciplines. The article can be seen as a contribution to the comparative philosophy of disciplines, and its main aim is to help us better understand what it is that we do when we do historical linguistics and/or history (or, at least, central parts of them) and also in what it is that we don’t. I do not compare everything that historical linguists and historians do here, nor everything that they are interested in. Rather, in seeking to answer the fundamental questions ‘what is the object of study in history?’ and ‘what is the object of study in historical linguistics?’ I argue that certain fundamental aims in the two disciplines allow us to remove the question mark in the title to this piece: history and historical linguistics do engage in two – interestingly connected, but also quite dif ferent – types of cognitive reconstruction. In section 2, I consider a number of basic connections and dif ferences between the two disciplines that are our focus; these are mostly beyond the main topic under consideration here, but they provide a context for the discussion in sections 3, 4 and 5. These latter sections consider historical linguistics and history in the light of  the idea proposed in the article’s title: section 3 focuses on historical linguistics, and section 4 on history. Section 5 concludes.

1

For discussion of these ideas, I would like to thank Nils Langer, Michael and Diana Honeybone, and the audiences at the Workshop on History and Linguistics, Linguistics and Historiography at the University of Bristol in 2008, and at a talk in the Language and History Interdisciplinary Seminar at the University of Oxford in 2009. I would also like to thank two anonymous reviewers for comments on a draft of this chapter. Any errors remaining are, of course, my own.

History and Historical Linguistics

17

2 History and historical linguistics: Parallels and dif ferences One obvious dif ference between historical linguistics and history is that historical linguistics is a subdiscipline, or branch, of a larger area of study: linguistics. ‘History’ is equivalent to ‘linguistics’ as a superordinate disciplinary term, which itself has second-level subfields, such as social history and political history. Certain characteristics of  historical linguistics are inherited because it is a ‘type of  linguistics’, so some of what follows compares aspects of  history with aspects of general linguistics.2 With this in mind, we can recognize a basic dif ference in the two fields’ terminologies: for linguistics, there is a handy distinction between (1a) and (1b), whereas the equivalent is absent for history, as (2a) and (2b) show. (1a) the name of  the discipline: linguistics (1b) the discipline’s object of study: language (2a) the name of  the discipline: history1 (2b) the discipline’s object of study: history2

2

As Lass (1997: 27) has it: ‘The primary constraint on a historical subject is its nonhistorical metasubject. Historical biology is part of biology, and hence constrained by biological knowledge and theory; historical linguistics is a branch of  linguistics, constrained by non-historical linguistic knowledge and theory.’ We will see below (in sections 2.1 and 3) how constraints developed in general linguistics can play a role in historical linguistics. Incidentally, history, as a metasubject, is unusual in not having a historical sub-subject – there is historical sociology, historical geography and historical anthropology, for example – but, perhaps because we need to avoid going round in circles, there is no historical history (there is ‘the history of  history’, but that’s a dif ferent matter). This could be seen as a further dif ference between the two disciplines in question here. On the other hand, historians do have the useful term historiography, to refer to the study of  the writing of  history as an academic discipline. There is no equivalent linguisticography, but this is simply a lexical gap, as the linguistic equivalent has long been studied. If we were to fill the gap, this article might even be described as a contribution to comparative historiography and linguisticography.

18

Patrick Honeybone

It could be claimed that (2b) should be ‘the past’, but there are reasons to reject this: history1 typically requires human agency (or even written records) for its subject matter to count, something considered in some detail in section 4. To be of interest to history1, these acts of human agency need to have occurred in the past, but that is not the same as simply being the past. The term ‘history’ is thus ambiguous (an obvious point, long made by others writing on the nature of  history, such as Carr 1961), and as it is important to be clear about the terms used in a piece such as this, it might sometimes help to have the subscriptal distinction in (2). I use this in the current section, but − rather than cluttering the whole article with subscripts − I adopt the convention in later sections that, if no subscript is given, then ‘history’ should be taken to mean history1 (as has been implicit up till now). It might seem reasonable to expect ‘historical linguistics’ to be an interdisciplinary branch of enquiry existing in the overlap between history1 and linguistics. This is not, however, where much of  historical linguistics lies. Considerable work in historical linguistics deals with autonomously linguistic structural entities (‘autonomous’ in the sense of Newmeyer 1986, for example), such as phonological segments and syntactic categories, and these entities are not subject to the conscious human will that history1 considers. This is not to say that no part of  historical linguistics overlaps with history1 – some parts of it certainly do, and there is an intersecting area of study where precisely the same questions can be asked. It may well be that this is where much of ‘historical sociolinguistics’ lies (the fact that such work is also known as ‘the social history of  language’ – as in Burke 2004 – corroborates this idea). Historical linguistics is thus not a unitary (sub)discipline, because it groups together any aspect of anything which connects language or language use with history2. Questions which need both historical and linguistic investigation include such basic issues as: how does the standardization of languages occur? There are clearly linguistic issues involved in standardization, but we also need to understand the social relations and context which motivated the individuals who enacted the process, and this requires historical methods. Specific questions which exist in the disciplinary overlap might include: what role did the idea that there was a Czech language play in political

History and Historical Linguistics

19

developments in the Central Europe? and what impact did this have on the Slavic dialect(s) spoken there? (see, for example, Törnquist-Plewa 2000, Evans 1998) or what were the demographics of the people involved in the early colonization of  New Zealand? and how did this lead to the formation of  the ‘new-dialect’ of  New Zealand English? (see, for example, Trudgill 2004). Such questions can all be part of historical linguistics, but they are not the kinds of questions considered in this article. I focus here (mostly in section 3) on issues of autonomous linguistic structure which are primarily or only of interest to linguists. With this restriction, we can identify a number of parallels between history1 and (at least the structural part of ) historical linguistics. One seems obvious, so much so that we have noted it already: both are connected with aspects of (actions from, states that existed in) the past, as spelt out in (3). (3a) history1 aims to understand past actions (3b) historical linguistics aims to understand past linguistic systems

There are two reasons why this is not in fact as obvious as it seems: (i) it requires us to reject the postmodern approach to history1, and (ii) it requires us to tease apart the synchronic and diachronic approach to the study of  the past. Postmodernism is a contentious and complex philosophical current in history1 (see, for example, O’Brian 2001 and Munslow 2001), as in many other fields. Some historians (such as Jenkins 1991 and Munslow 2001) avowedly subscribe to it, to argue, contrary to (3a), that we cannot ever rediscover or, therefore, truly understand history2. Postmodern history ties in with the general postmodernist distrust that anything could ever be objectively true, or that we can ever understand the thoughts of anyone other than ourselves, arguing that we are tied to our own contemporary linguistically-determined world-view. This means, the claim goes, that history1 just tells possible tales about history2, and other tales are always just as reasonable (for example, Munslow 2001 writes that ‘[m]y history is just another cultural practice that studies cultural practice’).

20

Patrick Honeybone

The approach that I consider here requires us to reject postmodern doubt, and to assume that we are indeed able to understand history2 in terms of  how it actually was (following, for example, Evans 1998 and O’Brien 2001), just as we can hope to understand the truth about physics and medicine. It is noteworthy that postmodernism has not been inf luential in linguistics (as it is generally conceived) − another dif ference between the two disciplines.3 Perhaps this is because many linguists view linguistics as more of a science than a humanity (and postmodernism has found no place in science), but it may also be that the very existence of  historical semantics and etymology – one branch of linguistics – contradicts or conf licts with postmodern tenets. The argument from etymology is that we are able to recover the meaning of words from the past, with some painstaking work, and hence we can understand what others mean, or meant. In its own way, this point f lags up a further connection between historical linguistics and history1 – the latter relies on the former to some degree, because historians do need to rely on a firm reconstruction of  the meaning of  the texts that they work with if  they are to escape the postmodern current. Thanks to postmodernist arguments, historians are more conscious of the significance of language to historical study (taking a ‘linguistic turn’), but while we know that words can change their meaning – an example of this is given in (11), below – an awareness of  the results of etymology (along with a judicious use of materials such as dictionaries which are contemporary with the texts that are being studied) allows us to reject the postmodern position, as is necessary for the argument to be made below to go through. For (3) to make sense, we also need to tease apart synchrony and diachrony. This distinction is inescapable in linguistics following its description in Saussure (1916), a fundamental text for twentieth-century linguistics. A 3

Rather oddly, some postmodern philosophers claim that they are doing linguistics, or at least are interacting with the concepts that linguistics deals with (e.g. Derrida 1967), or that linguistics was fundamental in the pathway of postmodernist ideas, through the development of structuralism. For a clear explanation of the baf f lement that this causes in linguists, see Dresher (1999), and for a demolition of some of the small amount of work in (applied) linguistics which has been inf luenced by postmodern ideas, see Borsley (2000).

History and Historical Linguistics

21

synchronic approach focuses on a particular system at one point in time, whereas a diachronic approach focuses on the changes that occur between chronologically successive instantiations of one particular system. This is quite straightforward in principle: it was recognized by linguists before Saussure named it, and the same distinction is made in other historical disciplines, even if under dif ferent names (for example, Warkentin 2009 talks of  the ‘horizontal approach’ − that is, synchrony − and the ‘vertical approach’ − diachrony − in historical geography). A naïve view of  historical linguistics and history1 might expect both to be purely diachronic disciplines (for example, Pei 1965 writes ‘[a] term often used as a synonym for historical linguistics is diachronic linguistics’), but for (3a) and (3b) to hold, they must study synchrony: a particular action can only occur at one specific point in history2, and past linguistic systems exist at specific points in history2. In fact, historical linguistics and history1 can and do focus on both diachrony and synchrony: for example, we might synchronically study the society or politics of  the kingdom of  Prussia in the middle of  the nineteenth century, or we might study the diachrony of  the changing organization of  the German states during the nineteenth century; or we might synchronically study the phonology of mid-fifteenth-century Middle English, or the diachrony of  the changes in the tense vowel system of  English from the end of  the Middle English period to the start of  Late Modern English. The synchrony/diachrony distinction will prove important below. We can recognize one further parallel between historical and historical linguistic method, if only because it f lags up one way in which the two disciplines are similar (even if the terminology used in this area hides this). This is pointed out in (4). (4a) the historical linguist uses both direct and indirect evidence (4b) the historian uses both witting and unwitting testimony

The observations in (4) describe two types of sources available in the two disciplines. The distinction in (4a) is analogous to that in (4b), and it is implicit in (4) that both disciplines are limited in similar ways by the evidence that is available to them. If we take historical phonology as an example

22

Patrick Honeybone

of work in historical linguistics, the two parts of (4a) can be explained as follows: in order to do historical phonology, we need to know how past phonological systems worked (which sounds they used, how they were organized, etc.). Direct evidence for this includes overt comments on pronunciation from past orthoepists and spelling reformers (such as Hart 1569, who devised his own phonetic alphabets to better represent English phonology); they needed to explicitly describe how people spoke in the sixteenth century in order to be able to write about what they wanted. Indirect evidence for historical phonology includes the consideration of spellings in documents from the past, and of rhymes and puns from past linguistic states, none of which was intended to provide special evidence about phonology (see, for example, Dobson 1968, Jones 1989 and Lass to appear). Marwick (2001a, b) explains the (4b) distinction used in history1, writing: ‘“Witting” means “deliberate” or “intentional”; “unwitting” means “unaware” or “unintentional”. “Testimony” means “evidence”. Thus, “witting testimony” is the deliberate or intentional message of a document or other source; the “unwitting testimony” is the unintentional evidence (about, for example, the attitudes and values of the author, or about the “culture” to which he/she belongs) that it also contains.’ The parallel in (4) is that both disciplines consider a range of types of sources, which require parallel types of interpretative skill: some aspects of  these sources aim to tell the reader something, and are hidden until research reveals them. 2.1 How can cognitive reconstruction be relevant? In order to show that history and historical linguistics are linked through cognitive reconstruction, we need to establish why we should want to discuss cognitive states at all. The distinctions made in (1) and (2) are worded as if they were very simple; like most things, though, they are not. The claim that cognitive reconstruction is involved in both branches of study derives from positions that have been developed in the disciplines’ philosophies concerning the true nature of (1b) and (2b) − of  their objects of study. They are set out in (5) and (6).

History and Historical Linguistics (5)

in history1: the ‘idealist’ position (e.g. Collingwood 1946) = the aim of  history1 is to rethink the thoughts of people from the past

(6)

in linguistics: the ‘mentalist’ orientation (e.g. Chomsky 1965) = language exists in the minds of speakers

23

The claims in (5) and (6) both link their respective areas of study to the cognitive realm, and will be crucial to the central argument in this article. They are bold and controversial claims, and I explain their bases in sections 3 and 4. In order to do that, one final connection between the two disciplines needs to be recognized: (7)

both historical linguistics and history1 rely on cognitive uniformitarianism

The assumption in (7) is that humans and the human mind have been qualitatively the same throughout history, so that the fundamental nature of the language that is investigated in historical linguistics and the thoughts that are considered in history1 have always been the same. The essential assumption is that both history1 and historical linguistics can only consider the history2 of modern humans and would not seek to consider anything beyond that (thus D’Oro 2006 writes that ‘[i]n ordinary usage history tends to be identified not with natural history but with the history of  human af fairs’). If we assume that the brain has not evolved since the development of modern humans somewhere between 100,000 and 200,000 years ago, we can reasonably adopt a uniformitarian position on the mind: the ways in which our brains allow us to think have not changed, so uniformitarianism is realistic in history1, and as language is uncontroversially enabled in some way by what is in our minds (for Chomsky it is constrained by Universal Grammar which guides the acquisition of  language), uniformitarianism in historical linguistics is reasonable, too. As Lass (1997: 28) shows, by quoting Whitney (1867: 24), a basic, non-cognitive uniformitarianism has been adopted by some in historical linguistics since at least the late nineteenth century, when it was in its prime:

24

Patrick Honeybone The nature and uses of speech […] cannot but have been essentially the same during all periods of its history […] there is no way in which its unknown past can be investigated, except by the careful study of its living present and recorded past, and the extension and application to remote conditions of  laws and principles deduced by that study.

This practically spells out general uniformitarianism, which was first established in geology: the principles governing the world were the same in the past as they are today, and if we establish the principles that govern an object of enquiry by considering how it works in the present (or perhaps also, at Whitney says, in parts of  the past that we understand well), we can restrict our ideas about how that object could have been in the parts of past that we need to understand. For our purposes, uniformitarianism means that (i) we should not allow for any kind of language during history2 which violates the linguistic principles that we have come to understand through our intuition- and corpus-based investigation of contemporary languages, and that (ii) we can assume that human actions and thoughts during history2 were produced on the basis of  the same kinds of  thoughts and impulses that we humans have today, which area available for investigation. If  the claims in (5) and (6) are right, cognitive uniformitarianism is an important methodological guide. Understood this way, both history1 and historical linguistics are interested in things which existed in people’s minds during history2. The claim that both disciplines involve cognitive reconstruction is thus beginning to become clear. To justify it fully, we need to consider what it is that historical linguists and historians do. I turn to this in the next two sections.

History and Historical Linguistics

25

3 What do historical linguists do? In order to make my case, I need to show that there is an aim in (at least part of ) both historical linguistics and history to carry out cognitive reconstruction. The argument is easier to make, and more widely accepted, in historical linguistics. I explain its basis here. Two commitments from the philosophy of the discipline are needed to make the case. These are (i) the fundamental orientation towards mentalism in linguistics, mentioned in (6), and (ii) a commitment from historical linguistics concerning the status of the results of historical linguists’ work: we must be able to give them a realist, and not formulist, interpretation. I explain both of  these brief ly below: the assumption of mentalism immediately, and the requirement of realism in section 3.1. The claim in (6) is that the object of study in linguistics is a mental, or cognitive, object. This is not uncontroversial, but we cannot consider all the arguments that have been had about the nature of  language here; what is important, and is obvious to all, is that humans have knowledge of  language stored in their minds – it is not contentious to say that we need to involve the cognitive realm when we speak. The idea that language is a cognitive entity is often associated with the ‘ideal speaker-hearer’ approach of generativism, one of the mainstream positions in linguistics (although not all aspects of the generative approach need be adopted for my fundamental claim to be accepted). From within this tradition, Chomsky has long argued that we need to recognize a distinction between ‘language’ and ‘the use of  language’. This is essentially the same distinction that Saussure (1916) proposed between ‘langue’ and ‘parole’, so it has long roots. Building on an early (1965) distinction between competence and performance, Chomsky (1986) discusses a distinction between I-Language and E-Language. I-Language is an aspect of knowledge − it is the set of linguistic generalizations that we know in order to be able to speak our language (or languages). For example, we know that in English, the regular plural is formed by adding an alveolar

26

Patrick Honeybone

fricative which takes its voicing from the final segment in the word that it attaches to (so we say [kats] cats but [kabz] cabs), and we know that whphrases like what and how many occur at the start of clauses in questions. Like all types of knowledge, I-language is i-nternal to each i-ndividual. It is the grammar that exists in the mind/brain of speakers. ‘E-language’ refers to the utterances that speakers make when they talk or write. This connects with everything else that the ‘everyday sense’ of language involves, including everything connected to language which is e-xternal to the speaker, such as the communities that they belong to. The standard generative opinion is that linguists who aim to understand such purely (autonomously) linguistic generalizations about phonology, syntax and similar aspects of language should focus on I-language. E-language is subject to inf luence from many dif ferent areas: physical pressures af fecting linguistic performance, social and political pressures af fecting interactions, and historical contingency in terms of what speakers happen to talk about. The claim is that the set of utterances that speakers make is not a well-formed or coherent object, and it is not enough if we hope to discover the generative system that constitutes the language. We cannot wait for occurrences of everything that is possible in a language in corpora of utterances, and they cannot tell us what is impossible in a language, so E-language is not the right object of study if we aim to figure out the generative system that a speaker knows. This is not to say that E-language is uninteresting or unworthy of study: the sociology and history of languages and the interactions and perceptions of their speakers are major objects of study which we need in order to understand the human condition; corpus linguistics has told us a lot about collocation and frequency, but these things are facts about how speakers use their language, not about language itself. Mentalism need not be stripped of all aspects of  linguistic variation, however (as it often is in generative approaches). In order to use their language appropriately, speakers need to know about the types of variation that the members of  their speech community engage in, so this must be part of  their knowledge about language – perhaps not precisely the same kind of  knowledge that we have about purely structural linguistic, but it, too, points in the direction of mentalism for linguistics. The mentalist

History and Historical Linguistics

27

case that ‘language’ exists in the minds of speakers, as I-linguistic states of  knowledge, has been clearly made in mainstream branches of  linguistics. It gives linguists of a structural bent a coherent object of study, and I adopt it here. 3.1 Does historical linguistics involve cognitive reconstruction? We have seen that is it reasonable to claim that language is a cognitive entity. The case that historical linguistics engages in cognitive reconstruction is now not that dif ficult to argue. If  language exists in the mind, and if one of the things that historical linguists do is to try to work out the structure of past synchronic linguistic states, then historical linguistics engages in cognitive reconstruction. In this section, I consider the ways in which historical linguists try to work out the structure of past synchronic linguistic states, and then I turn to issues that some linguists raise concerning the status of  the reconstructions that are done in the discipline. One of the earliest and greatest achievements of historical linguistics was the proof that many currently existing languages are ‘genetically’ related to each other – that they derive in some real sense from the same ancestor proto-language. For example, we know that most of the languages spoken in Europe (and the countries to which they have been colonially exported in the past few centuries), and many spoken in Asia are derived from ProtoIndo-European (PIE). One of the central aims in nineteenth-century linguistics was to reconstruct proto-languages, especially PIE, about whose daughter languages much is known, and work on PIE continues today. The great tool for such work is the comparative method of reconstruction, in which forms for a particular linguistic feature are compared in related languages, to establish correspondences of form, which are then considered together to establish what is the most likely ancestral form from which they all could have derived (for a full description of  the method, see Fox 1995). These forms are taken to be the reconstructed linguistic features of  the proto-language. A classic example is the reconstruction of the PIE stop system. Although most European languages have two series of stops – often transcribed as

28

Patrick Honeybone

/p, t, k/ vs /b, d, g/ – we can be sure that PIE had three series, thanks to the evidence of languages like Sanskrit and Greek, which preserve multiple stop series. Also, we know that one of  the series of  PIE stops became fricatives in Proto-Germanic, so that contemporary English three, with an initial fricative, corresponds with Ancient Greek treis, Latin tres, Old Church Slavonic trije and Old Irish tri, for example, which all have stops (nominative masculine forms for three given here where the number declines in a language). To summarize some of  this, (8) shows in its final column the classic reconstruction of  the three series of  PIE stops (with the coronal stop standing for the full series), which have been reconstructed on the basis of  the oldest attested forms including those given in the other columns (taken from Job 1989). ‘OCS’ stands for Old Church Slavonic, the oldest attested Slavic language, ‘Vedic’ is a form of  Sanskrit, and Gothic represents the Germanic languages, such as English, so it has a fricative (as in three) corresponding to stops in the other languages. The overwhelming preponderance of stops in non-Germanic languages shows that the PIE reconstruction of a stop is on firm ground. (8) Avestan

Vedic

Greek

Latin

Old Irish

Gothic

OCS

PIE

t

t

t

t

t

θ

t

t

d

d

d

d

d

t

d

d

d

dh

th

f

d

d

d

dh

The forms in the last column in (8) are an attempt to reconstruct the stop system of a language for which we have no records. Given that it was argued above that the only coherent locus to situate a language, in its technical sense, is in the minds of its speakers, this means that what is shown here is an attempt at cognitive reconstruction – a reconstruction of something that existed at some synchronic point in the minds of  the speakers of  PIE. Two points should be raised here. Firstly, it is common practice in historical linguistics to indicate forms which have been reconstructed in this way with an asterisk, to contrast them with forms which are attested in written documents and the like. This means that PIE /t/ is commonly

History and Historical Linguistics

29

represented as *t, for example. I do not adopt this practice here because all the forms given in (8) are reconstructed in some way, as discussed below, so the status of  the PIE form is not fundamentally dif ferent from the other forms in (8). Secondly, we should note that the classic reconstruction of  PIE stops has been subject to considerable challenge. It is not in doubt that there were three series of stops, but linguists have argued at length about the precise way in which they should be reconstructed. The best known alternative reconstruction is the ‘Glottalic Theory’ of  Gamkrelidze & Ivanov (1973), which can be seen as contending that rather than voiceless/ plain, voiced and voiced aspirated series, there were voiceless aspirated, ejective and voiced aspirated series. Proposals such as the Glottalic Theory have been made for a number of reasons, in part centring around the idea that the classic reconstruction is typologically highly unlikely, because we do not expect to find languages with a breathy voiced (‘voiced aspirated’) series, like /dh/ if  they do not also have a voiceless aspirated series, like /th/; that is, the uniformitarianism of (7) is applied. Far from casting doubt on the assumption that the final column in (8) is a case of cognitive reconstruction, however, the fact that linguists argue about the best way to reconstruct a language shows that they care about getting the cognitive reconstruction right. The uniformitarian constraints of general linguistics on what is a possible language only apply to a reconstructed language if  the reconstruction is taken to be psychologically real. By PIE, then, we mean the I-linguistic grammar that a speaker of PIE had. In the same way, when historical linguists work to figure out any aspect of an attested language such as Old English (OE), they are trying to reconstruct the mental grammar of speakers of  OE (at a particular point in its development, such as classical West Saxon). It may be that in certain cases we can never be successful in reconstructing complete past I-languages – if evidence in a particular case is scant, for example – but we can still hope to reconstruct at least parts of past languages fully, such as the plosive or fricative system of  the language, or its nominal morphology, and these parts were mental (sub)systems in their own right. To consider an example of synchronic cognitive reconstruction from OE, we know from a range of  types of evidence that OE had essentially only one series of  fricatives, of  the type /f, θ, s/; there were no voiced

30

Patrick Honeybone

fricatives of  the type /v, đ, z/. We also know that OE fricatives were subject to a phonological process of intersonorant voicing. To exemplify with the labial-dentals, this means that the fricatives in fisc ‘fish’ and wulf ‘wolf ’ were voiceless [f ] on the surface, but that those in wulfas ‘wolves’ and drifan ‘drive’ were voiced [v] (as many have shown, such as Hogg 1992b, Lass 1994).4 This takes us beyond comparative reconstruction. However, while evidence in the form of written texts is available for this reconstruction, the issues for the historical linguist are essentially the same. This is why I do not adopt the asterisk to mark ‘reconstructed’ forms in PIE: all historical linguistics deals with reconstructed forms. Historical linguists aim to work out past I-languages, and it is clear that past I-languages, as objects of enquiry, cannot be observed. This is exactly the same for OE (Middle Dutch, nineteenth-century Korean, etc.) as it is for PIE: none of them can be observed. The ultimate object of study for historical phonologists who work with attested languages is not the written forms. Written forms can provide important evidence for historical phonology, but they need to be interpreted with skill and caution (see, for example, Lass to appear, Minkova to appear). Thus, the OE letter is not a phonological (I-linguistic) object, but fisc, wulf, wulfas, drifan, as examples of preserved E-language from the past, can be used as evidence in the reconstruction of  the I-language of the speakers who wrote the Old English texts that survive. In the same way, the comparison of attested forms such as three with an initial fricative (and Gothic nominative neuter þrija, the oldest Germanic form for ‘three’) with the overwhelming majority of other Indo-European languages which have a stop (such as Latin tres and Old Church Slavonic trije), as summarized in (8), can be used as evidence in the reconstruction of a /t/ in the I-language of  the PIE speakers (who happened not to leave us any written records). The OE reconstruction of an underlying /f/, which was realized as [f ] and [v], is a reconstruction − it is not anything that is found 4

In fact, the precise laryngeal characteristics of  the segments involved are subject to debate, like those of  IE, but for dif ferent reasons (see Honeybone 2005). The current article is not the place for that debate, however, and the basic point stands whatever position is taken in laryngeal phonology. Once again, the fact that there is such debate shows that it is important to get the reconstruction right.

History and Historical Linguistics

31

in texts. Past phonological systems need to be reconstructed by (historical) phonologists using evidence from written sources (and from elsewhere, such as the forms found in living descendent dialects, and direct evidence from commentaries on pronunciation, where it is available), and this means that both OE /f/ and PIE /t/ are reconstructed and thus have the same phonological status, meaning that neither (or both) need an asterisk.5 We have dealt with historical phonology in some detail here, but the same conceptual issues arise in the reconstruction of other linguistic levels. I give here much briefer examples of  the reconstruction of past syntactic and semantic states. The same range of indirect and direct evidence is available to do such reconstruction (written records, the comparative method, linguistic commentaries from the past), but comparative reconstruction is less successful in syntax because the systems involved are less straightforwardly compared. A syntactic example: OE is reconstructed as a verb-second (V2) language, similar to contemporary German and Dutch. This means that the finite verb is expected to be the second constituent in main clauses (some syntactic processes can ‘hide’ this order, but it is assumed to be fundamental). Unlike in contemporary English, where the subject typically comes directly before the verb, many dif ferent kinds of constituents could directly precede the verb in OE: subjects, objects, adverbials etc. The sentences in (9), from Orosius, and (10), from the Anglo-Saxon Chronicles (both taken from Kroch & Taylor 1997), show the ef fects of  V2: as is shown in the aligned gloss below the original OE, (9) has the finite verb (marked as

5

In fact, it can be argued that the reconstructed linguistic entities of past languages have the same status as analyses of contemporary I-languages, too. Neither a past linguistic state nor a present one can be directly observed by a linguist. Rather, we need to use whatever E-linguistic evidence is available to us to figure out what the linguistic system is that we are investigating, and the result of linguistic analysis – the model of linguistic knowledge that we develop – is always constructed by the linguist. There is more and better evidence available to those investigating a contemporary language, such as native speaker intuitions, recordings, controlled experiments and massive corpora, but this construction of analyses of contemporary languages is essentially the same kind of activity as the reconstruction of ideal languages: both are building models of  I-languages using the best available evidence.

32

Patrick Honeybone

‘V2’), preceded by its direct object (‘O’) and followed by the subject (‘S’), and (10) has an adverbial (‘A’) where (9) has an object. The reconstruction of the synchronic syntactic patterns which allowed this type of word order need to be figured out by historical syntacticians, and several subtle syntactic subgeneralizations about the precise patterning of V2 constructions in OE have been discovered in a long strand of research on the topic (for example, van Kemenade 1987, Pintzuk 1991, Kroch, Taylor & Ringe 2000). (9) þæt hus

hæfdon Romane

to

ðæm anum tacne

that building had

Romans

with the

O

S

etc.

(10) þær

V2

one

wearþ

se cyning  Bagsecg

ofslægen

there

was

the  king  Bagsecg

slain

A

V2

S

etc.

geworht

feature constructed

A semantic example: as is well known, and as was discussed in section 2, the meaning of words in and from the past needs subtle reconstruction. As an example of  this, (11a) shows one sentence from an English letter from around 1740, from a Dr Rutherforth of Cambridge, which requires careful reconstruction of the meanings of most of its content words in order to be properly understood. This is shown in the translation into contemporary English in (11b). Both text and translation are taken from Honeybone & Honeybone (2010, p.c.); the main cases of semantic reconstruction are identified through underlining, emboldening and italics. (11a) I suppose that natural philosophy and polite literature are the branches of science that you chief ly improve (11b) I assume that the fields of  knowledge that you work in are chief ly science and philology

The meanings of words exist in the minds of the speakers of the languages that they are part of, so the meanings that have been reconstructed in (11b) are things that existed in the minds of a speaker of mid-18th-century English. This is thus, once again, a case of  I-linguistic, cognitive reconstruction.

History and Historical Linguistics

33

In working to understand the linguistic systems and states discussed above, structural historical linguists are (whether they describe it as such or not), aiming to reconstruct past cognitive states. This has led to the writing of largely synchronic historical grammars, such as Campbell (1959), Hogg (1992a) and Braune, Ebbinghaus & Heidermanns (2004). It also expands the database of synchronic linguists, in the quest to understand what is possible or common in language. For example, Dresher & Lahiri (1991) show that the structure of  the metrical foot in the phonology of  Early Germanic languages was an uneven moraic trochee, a type of  foot which had previously been claimed to be unattested (see Kager 1995). In order to make the case that historical linguistics engages in cognitive reconstruction, one final assumption (which has thus far been silently made) needs brief discussion. The ontological status of the entities that historical linguists reconstruct has, in fact, been a subject of some controversy in the literature. Fox (1995: 9), discussing the status of the forms derived through comparative reconstruction, explains the issue thus: Are we entitled to claim for such reconstructions the status of earlier linguistic forms? This controversy can be summed up in a confrontation between two views of reconstruction: the formulist and the realist. The formulist view regards reconstructions merely as formulae which represent the various relationships within the data, while the less cautious realist view assumes that reconstructions can be taken to represent genuine historical forms of a real language, which happen not to have been recorded.

As discussed above, I take all reconstructions – both those done on the basis of the comparative method and those derived from written sources – to have the same status, so at least some of the points considered here are relevant to all aspects of reconstructive historical linguistics (although they have largely been previously considered in relation to the results of comparative reconstruction). Can we really assume that (for example) the phonological system reconstructed for PIE, or OE, or any other past language, should be seen as a ‘real’ I-language? The formulist position sees the reconstructions of  historical linguists as useful fictions which are essentially place holders for the comparisons themselves – they should not be taken to be the real forms of any language. This position has inf luential advocates, for example Meillet (e.g. 1903), but it must be wrong. It is only if we assume that

34

Patrick Honeybone

reconstructed languages are real languages, and should thus be given the same status as the models of I-language that we construct for contemporary languages, that uniformitarianism can place a constraint on reconstruction (as we saw above that it must). Linguistic typology and linguistic theory can only be used in arguments about reconstruction (such as the Glottalic Theory, for example), if we assume that reconstructed languages are the same sort of  thing as the languages that linguists work with to figure out typological and theoretical generalizations in the first place. If we do not, then we have no control on what we can reconstruct. (The need to reject postmodern doubt that we can ever understand the things going on in the minds of others for these claims to hold will be clear.) The realist position is well supported in historical linguistics. The authors of  the two most detailed recent considerations of  historical linguistic method (Fox 1995 and Lass 1997) both tend towards realism in reconstruction. Thus Fox (1995: 13) writes that ‘as realists rightly maintain, to be significant the results [of reconstruction] must be interpretable in historical terms’, and Lass (1997: 274) that ‘protolanguages […] are simply languages like any others, if with poorer information sources’. The formulist position seeks to place the things that are reconstructed outside of the minds of speakers, by claiming that they are not languages at all, but this claim is not consistent with the practice of  historical linguists, nor is it compelling. The aspects of  languages that are reconstructed in historical linguistics must be assumed to represent aspects of past I-languages. We should not claim that languages are always (or ever) fully reconstructable (especially not if we assume that variation is grammatically inherent in language, as it is often dif ficult to reconstruct much of the way in which a language must have varied) but we can reconstruct parts of  the linguistic systems involved (such as the underlying phonological contrasts of  IE, or finite verb placement in OE), and these parts existed in the minds of speakers. Historical linguistics does, then, in the ways described above, seek to engage in cognitive reconstruction.

History and Historical Linguistics

35

3.2 Does historical linguistics only involve cognitive reconstruction? Any acquaintance with work on structural historical linguistics will show that it does not just involve the reconstruction of past linguistic systems. The synchronic aspect of historical linguistics can be underestimated, but it is subservient to diachronic study: much of historical linguistics aims to understand how and why one linguistic state can change into another. Thus, part of  the interest in comparing Gothic þrija with Old Church Slavonic trije certainly is to reconstruct PIE forms, but part of it is to understand the phonological change that occurred in Germanic, by which stops became fricatives – part of  the famous Grimm’s Law. And part of  the interest in working out that OE had a single series of underlying fricatives is to go on to find out how this changed, so that we now have two series. English has lost V2, at least in the basic type of clauses illustrated in (9) and (10), so that the subject must now always precede the verb, and this has been the major focus of work on V2 in the history of  English. Also, of course, the changes that can be seen by comparing (11a) and (11b) have occurred and are of major interest to historical linguists. Synchronic systems are often only reconstructed so that they can be compared with other, temporarily distinct systems and changes can be observed. One central aim of historical linguistics is to work out a theory of linguistic change, addressing such questions as: how does change happen? Can it happen within an individual? Or does it typically happen cross-generationally, when a new generation misacquires the language of their parents? Historical linguistics can try to work out if anything is possible in change, or whether we can formulate a notion of what would be an ‘impossible change’,6 and if so, why they are impossible. These are diachronic questions and particular changes may not be anything to do with I-language – they could be due to population movements, or misperception, or structural ambiguity and

6

Indeed, it seems that we can: a change such as p, t, k > f, θ, x can occur, and did as part of Grimm’s Law, thus explaining the correspondence of non-Germanic /t/ with Germanic /θ/, but the reverse does not ever seem to be attested (see, for example, Kümmel 2007).

36

Patrick Honeybone

the unconscious choice of speakers to adopt a dif ferent linguistic system. While I-linguistic factors may be involved (if change can occur within a grammar), they need not be. Historical linguistics does engage in cognitive reconstruction, but that is not all that it does.

4 What do historians do? Do historians engage in cognitive reconstruction? As was the case in the last section, certain philosophical commitments need to be made in order for this claim to hold. The necessary conceptual position in history is probably proportionately less common than that sketched for historical linguistics above, but it is compelling, is well represented in the literature and is accepted (if maybe tacitly) in certain key intellectual traditions. 4.1 Does history involve cognitive reconstruction? The commitment in terms of  the philosophy of  history that is needed to make my case is an interpretation of the (at least partly) idealist approach to history, which is associated mostly closely with the historian R.G. Collingwood. Idealism focuses on the mental world, leaning towards the notion that a range of things (which might not seem to do so on a ‘common-sense’ approach) actually only exist in the mind. At its most extreme, idealism can argue that no aspect of reality exists independent of our perceptions, but Collingwood was wary of being identified as a total idealist (D’Oro 2006), and we do not need to subscribe to full-blown philosophical idealism to accept Collingwood’s ideas (in the same what that it is not actually necessary to accept full-blown generativism in order to accept the mentalist understanding of language which fits with Chomsky’s ideas). We can retain the term ‘idealism’ to describe the approach, despite these and Collingwood’s own caveats, however, as is conventional in the philosophy of history. The idealist view of history can be understood as cognitivist in this way, in part

History and Historical Linguistics

37

echoing (3a): (i) history aims to understand past actions; (ii) these actions are due to human causality; (iii) human causality is due to the thoughts that the actors have; and (iv) thoughts exist in the cognitive realm. The Idea of History (1946) is Collingwood’s classic work on the philosophy of history (Collini & Williams 2004). It includes his famous definition of  the idea of  history: ‘history should be (a) a science, or an answering of questions; (b) concerned with human actions in the past; (c) pursued by interpretation of evidence; and (d) for the sake of human self-knowledge’ (1946: 10–11). The relevant points here are (b), as mentioned above, and (d), which provides the perspective which leads to the ideas in focus here. Collingwood argued that we can understand ourselves through history because we can develop an understanding of psychology through history, and that this is because history is, practically, psychology. Collingwood wrote about a conception of psychology as a type of historical study, whose aim ‘would be to detect types or patterns of activity, repeated over and over again in history itself ’ (1946: 224, see Connelly & Costall 2000), and this requires us to understand the thoughts of those who performed these activities in the past. (The need to reject postmodern doubt that we can understand the minds of others will again be clear for these claims to hold.) The crucial part of  Collingwood’s argument for our purposes is his emphasis ‘upon history as the interpretation of purposive action and hence upon the primacy of  the need to recover historical agents’ own understanding of  their situation’ (Collini & Williams 2004). This is linked to his conception of psychology, but is not tied to it. Like many others, Collini & Williams (2004) cite Collingwood (1946: 215) as his most famous claim, given here in (12). (12) ‘all history is the history of thought … and therefore all history is the re-enactment of past thought in the historian’s own mind’

Collini & Williams (2004) write further that ‘this was intended not as a prescription of  the method of  historical enquiry, but a description of its success: only insofar as historians can “re-enact” in their own thought what an act meant to its agent can they explain that act.’ This leads us to an understanding of the definition of the object of study in history as actions, as D’Oro (2006) explains:

38

Patrick Honeybone The subject matter of history, understood as a science of the mind, is actions − actions understood not simply as the doings of human beings but of human beings in so far as they are rational. Actions, in the sense in which they constitute the subject matter of historical investigation have an ‘inside’ that events lack. To explain an event all we need to do is to subsume it under a general law that is obtained by inductive generalisation, through the observation of repeated events of  type B following events of  type A. In order to understand an action, by contrast, we need to render it intelligible by reconstructing the thought processes that inform it.

On this approach, the key goal of  history is cognitive re-enactment, as in (12) – the historian should aim to re-enact (that is, to reconstruct) the thought processes that led to the historical actions that history tries to understand. The uniformitarian principle of (7) is important here: we must assume that the mind has not changed in the way that it manipulates or creates thoughts throughout history, for if it had, past thoughts could be unrethinkable. If we take Collingwood at his word (1946: 301), re-enactment should be absolutely literal cognitive reconstruction: the very thought that an agent had in the past can be repeated. […] in its immediacy, as an actual experience of  his own, Plato’s argument must undoubtedly have grown up out of a discussion of some sort, though I do not know what it was, and been closely connected to such a discussion. Yet if  I not only read his argument but understand it, follow it in my own mind re-enacting it with and for myself, the process of argument which I go through is not a process resembling Plato’s, it is actually Plato’s so far as I understand him correctly.

The claim is not that we need to engage in telepathy. Rather, ‘it is possible in principle to re-enact the thoughts of others because thoughts, unlike physiological processes, are not private items unique to the person who has them, but publicly rethinkable propositional contents’ (D’Oro: 2006). D’Oro further explains that this approach has been seen as counterintuitive: since to say that the thought of  the agent and that of  the historian are the same appears to presuppose that there is only one rather than two numerically distinct acts of  thought, that of  the historian and that of  the agent. Collingwood’s point, however, is that, since thought proper is conceptually distinct from the physiological process in which it is instantiated, the criterion of numerical identity that is usually applied to physiological processes is not applicable to thought. Thoughts, in other

History and Historical Linguistics

39

words, are to be distinguished on the basis of purely qualitative criteria, and if there are two people entertaining the (qualitatively) same thought, there is (numerically) only one thought since there is only one propositional content.

If a historian is able to fully reconstruct the thoughts7 that led someone from the past to carry out certain actions, they can be said to understand those actions, and will thus be able to write accurate historical works about those actions. In this sense, historians must engage in cognitive reconstruction. Although the act of writing about this is not reconstruction in itself, the same applies to the act of a historical linguist writing about the PIE consonantal system or OE syntax: the reconstruction must be done first, in order for the work which focuses on it to be written. Part of such work might indeed simply be a record of  the reconstructions: a table of  PIE consonants, or a description of the thinking of a historical agent, but much of it will be ref lection on the result of  the reconstruction, and an attempt to place it in its context. The implications of all this are that, in order to understand the actions of those who agitated for national self-determination in Austria-Hungary, for example, or who persecuted non-Catholics in fifteenth-, sixteenth- and seventeenth-century Spain, or who agreed the Acts of  Union between England and Scotland, we need to rethink the thoughts that led to the actions involved. Most of the evidence for this is found in texts which are contemporary with (or, maybe, later than) the actions involved, and considerable skill is required to interpret these documents (informed by historical linguistics where necessary, as discussed above, particularly around (11)). The approach sketched out here requires the historian to understand the mindset of  the people that they are interested in, to develop a sense of empathy with them (again, this rejects postmodern arguments, such as those of  Jenkins 1991). An example of  the impact of  this approach is considered in M. Honeybone (2008: 16), who is discussing the writing of the history of a well-known witchcraft trial from the 17th century in the Vale of  Belvoir in the English Midlands: 7

We may not need to assume that the very same thought exists in the minds of  the original actor and the historian, as Collingwood argued – if a thought is simply repeated, we may escape the danger of  Collingwood’s mind-body distinction.

40

Patrick Honeybone European scholars have appreciated that the concept of early-modern witchcraft arose as an imaginative response to the widespread myth of  the witches’ sabbath, defined as ‘A midnight meeting of demons, sorcerers and witches, presided over by the Devil, supposed in mediæval times to have been held annually as an orgy or festival.’ [SOED] After extensive historical investigation it is now accepted that no actual ‘sabbath’ ever took place. Equally, however, the notion of the sabbath existed absolutely in the minds of sixteenth- and seventeenth-century men and women. This is a perfect justification for Collingwood’s concept of  history as the process of rethinking the thoughts of past people: ‘wrong ways of  thinking are just as much historical facts as right ones, and, no less than they, determine the situation (always a thought situation) in which the man who shares them is placed’. (Collingwood, 1946: 317)

Reconstructing past thoughts is surely not easy, but it is just as surely the ideal that we should strive for in attempts to understand the past (at least, if we follow Collingwood). As the above point shows, if we are open to all the beliefs of those whose past we consider – ‘false’ beliefs and the general intellectual context, as well as anything we can glean about an individual writing a document – we can hope to come close enough to rethink the thoughts of  those whose actions we are interested in. Other historical artefacts, such as buildings and clothing from the past, can also help in the interpretation of past people’s thoughts, and this has led some to reject the idealist position. As Evans (1997: 92) writes, ‘[m]any sources are not written at all. Getting inside the head of someone who buried treasure in a grave in the fourth century, or made a newsreel in the twentieth, is far from easy.’ This is a reasonable point, of course, but it is reminiscent of the mistaken distinction in historical linguistics between reconstructed forms which are standardly given an asterisk and those that are not, as discussed above. Those forms without asterisks are still reconstructed, even though they derive from written sources. Carr (1961: 16) explains that, in the same way, written sources, while privileged among types of evidence, still require a considerable amount of interpretation. He criticizes the nineteenthcentury ‘fetishism of  facts and documents’, explaining that: The documents were the Ark of  the Covenant in the temple of  facts. The reverent historian approached them with bowed head and spoke of  them in awed tones. If you find it in the documents, it is so. But what, when we get down to it, do these documents – the decrees, the treaties, the rent-rolls, the blue books, the of ficial correspondence, the private letters and diaries – tell us? No document can tell us more

History and Historical Linguistics

41

than what the author of  the document thought – what he thought had happened, what he thought ought to happen or would happen, or perhaps only what he wanted others to think he thought, or even only what he himself thought he thought. None of  this means anything until the historian has got to work on it and deciphered it.

Just as linguistic reconstruction can proceed on the basis of  texts or comparison, historical reconstruction can proceed on the basis of texts or other historical material, but information found in documents still needs processing to form the basis of reconstruction. It all requires careful interpretation in order for us to be able to do the reconstruction. The comparative method and the use of non-written sources simply need more care and provide less good evidence. Although wary of some of  the implications of  Collingwood’s ideas, Carr (1961) is generally sympathetic to the idealist approach, because it shows that history is not just a matter of finding dusty facts in documents. Thus, also citing Collingwood (1946), with an echo of (12), Carr (1961: 22) writes: ‘The past which a historian studies is not a dead past, but a past which in some sense is still living in the present.’ But a past act is dead, i.e. meaningless to the historian, unless he can understand the thought that lay behind it. Hence ‘all history is the history of  thought,’ and ‘history is the re-enactment in the historian’s mind of  the thought whose history he is studying.’ The reconstitution of  the past in the historian’s mind is dependent on empirical evidence. But it is not in itself an empirical process, and cannot consist in a mere recital of  facts. On the contrary, the process of reconstitution governs the selection and interpretation of  the facts: this, indeed, is what makes them historical facts.

Given all of  the above, it is not unreasonable to conclude that historians need to reconstruct past cognitive states and actions (although this may not be a historian’s conscious aim) when they work to understand past actions. It may be that this must sometimes remain an unachievable ideal, in that we can only get close to the original thought, because evidence for it is lacking. The ideal remains, though, that the thoughts that provided the context for the actions that we know of need to be reconstructed, and in principle can be rethought in the mind of  the historian (who can then aim to invoke them in the mind of  the reader of  their work). History does, then, in the ways described above, seek to engage in cognitive reconstruction.

42

Patrick Honeybone

4.2 Does history only involve cognitive reconstruction? Just as there are with historical linguistics, there are some aspects of history which are not themselves attempts at cognitive reconstruction. We have already considered that the actual writing of  history, just like the writing of historical linguistics, may rely on reconstruction, but also involves much more than this. It is also surely the case that much of  the work done by the historian in order to get to the position of  being able to re-enact past thoughts is not reconstruction itself: the work that a historian carries out with documents and artefacts, for example, and conducting the background work which is necessary to ‘do the history’. This is rather dif ferent from the conclusion of section 3.2. It may be that history is essentially a synchronic discipline, to the extent that it does not seek to deal with diachronic change to the same extent that historical linguistics does. Rather than largely focusing on processes of change between one system and another, and the possible types of such changes that can occur, there is more emphasis in history than in historical linguistics on understanding particular times and places. While not all of the conduct of history is an engagement in cognitive reconstruction, more of it is than is the case in historical linguistics.

5 Two types of cognitive reconstruction The case has been made above that the two disciplines under discussion here both engage in cognitive reconstruction. We have seen that it is not all that they do, and it should be remembered that I am focusing on structural historical linguistics in my discussion here, but I have argued that cognitive reconstruction is fundamental to the success of  both. One final point needs to be considered: while the notion of cognitive reconstruction connects the two disciplines, it also distinguishes them, because the two types of cognitive reconstruction may be conceptually similar, but they are also fundamentally dif ferent.

History and Historical Linguistics

43

There are clear conceptual connections between the two types of reconstruction. For example, both rely on uniformitarianism − that the human mind has been the same throughout history2, since we evolved as a species, which means that the nature of  language and of  thought have always been the same. This, in turn, means that we stand a chance of being able to engage in cognitive reconstruction as part of our disciplines. Furthermore, both disciplines have to work with sources which are imperfect ref lections of their true object of study: often written sources, which need careful interpretation. There is a substantial dif ference between the two types of reconstruction, however. This derives from a distinction between the objects of study in the two disciplines, set out here in (13). (13a) language is part of  the unconscious mind (13b) agentive thoughts are part of  the conscious mind

The distinction in (13) is an important one, and it explains many of  the dif ferences between the approaches discussed above. While the distinction between conscious and unconscious mental activity is a complex one – far too complex to be properly considered here – we can assume a basic distinction between those aspects of the cognitive world to which we can, in principle, have direct access, and those to which we cannot. We know an extraordinary number of subtle generalizations about the languages that we speak, but we also know that we do not have direct conscious access to this knowledge – we have intuitions about what is ungrammatical, but we cannot easily explain why. Linguistic rules are part of our unconscious knowledge. Contrary to this, we can ref lect on the thoughts which lead us to perform specific actions, and such agentive thoughts are the subject matter of  history, on the interpretation developed above. One implication of (13) is that historical linguists do not try to reenact a past linguistic state in their own minds, whereas, if we follow the idealist position, this is what historians should try to do. Of course, the reconstruction that historical linguists engage in is a product of their minds, but there is no expectation that the reconstruction should function in their minds as it did in the minds of the original speakers. In history, there

44

Patrick Honeybone

is some sense in which a reconstruction should function in the same way in the historian as it did in the original thinker. In a sense, then, historical linguistics engages in a type of cognitive reconstruction which is less connected to the researcher than is the case in history: historians use the part of  the mind that the object of  the reconstruction resided in to do the reconstruction, whereas historical linguists reconstruct unconscious linguistic knowledge with conscious intent. These dif ferences mean that quite distinct methodological tools and approaches are appropriate for the dif ferent types of reconstruction that the two disciplines engage in. Thus, while structural historical linguists and historians can surely learn from each other, we cannot expect to do things in the same way. We can likely better understand both disciplines if we see the dif ferences as just as important as the similarities. To conclude: on the basis of the philosophical positions defended here, historical linguistics and history do indeed both seek to engage in cognitive reconstruction, although the reconstruction involved is of dif ferent types. It may be that we do not always succeed in our attempts at such reconstruction, or that the reconstruction can only be partial, and it is clear that both disciplines do not only engage in such reconstruction, and sometimes only do it so that they can consider something else, but I hope to have shown here (and to have considered a range of non-obvious but connected issues along the way) that historical linguistics and history are both connected and distinguished by the notion of cognitive reconstruction.

References Borsley, R. 2000. ‘A postmodern critique of linguistics’. Ms, University of Essex. accessed 12 August 2011. Braune, W. 1883. Gothic Grammar with selections for reading and a glossary. Trans. G.H. Balg. New York: Westermann. Braune, W., Ebbinghaus, E., and Heidermanns, F. 2004. Gotische Grammatik, 20th edn. Tübingen: Niemeyer.

History and Historical Linguistics

45

Burke, P. 2004. Languages and Communities in Early Modern Europe. Cambridge: Cambridge University Press. Campbell, A. 1959. Old English Grammar. Oxford: Clarendon Press. Carr, E.H. 1961. What is History? Harmondsworth: Penguin. Chomsky, N. 1965. Aspects of  the Theory of  Syntax. Cambridge, MA: MIT Press. Chomsky, N. 1986. Knowledge of  Language: its Nature, Origin, and Use. New York: Praeger. Collingwood, R.G. 1946. The Idea of  History. Oxford: Clarendon Press. Collini, S., & Williams, B. 2004. ‘Collingwood, Robin George (1889–1943)’. In: Oxford Dictionary of  National Biography. Oxford: Oxford University Press. accessed 12 August 2011. Connelly, J., & Costall, A. 2000. ‘R.G. Collingwood and the Idea of a Historical Psychology.’ Theory & Psychology 10, 147–170. D’Oro, G. 2006. ‘Robin George Collingwood.’ In: Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy, accessed 12 August 2011. de Saussure, Ferdinand. 1916. ‘Cours de linguistique générale’ In: Bally, C., & A. Sechehaye (eds), with A. Riedlinger, Lausanne and Paris: Payot. Trans. W. Baskin, Course in General Linguistics. Glasgow: Fontana/Collins, 1977. Derrida, J. 1967. De la grammatologie. Paris: Editions de Minuit. Dobson, E. 1968. English Pronunciation 1500–1700. 2 vols. 2nd edn. Oxford: Clarendon Press. Dresher, B.E. 1999. ‘Ferdinand, we hardly knew you.’ Glot International 4/6, 9. accessed 12 August 2011. Dresher, B.E., & Lahiri, A. 1991. ‘The Germanic Foot: Metrical coherence in Germanic.’ Linguistic Inquiry 22, 251–286. Evans, R.J. 1997. In Defence of  History. London: Granta. Evans, R.J.W. 1998. The Language of  History and the History of  Language. Inaugural lecture delivered before the University of  Oxford. Oxford: Clarendon Press. Fox, A. 1995. Linguistic Reconstruction: An Introduction to Theory and Method. Oxford: Oxford University Press. Gamkrelidze, T., & Ivanov, V. 1973. ‘Sprachtypologie und die Rekonstruktion der geminindogermanischen Verschlüsse.’ Phonetica 27, 150–156. Hart, J. 1569. An Orthographie: Conteyning the Due Order and Reason, howe to Write or Paint Thimage of  Mannes Voice, Most Like to the Life or Nature. London: William Seres. Hogg, R. 1992a. A Grammar of  Old English. Vol 1: Phonology. Oxford: Blackwell. Hogg, R. 1992b. ‘Phonology and Morphology.’ In: R. Hogg (ed.), The Cambridge History of the English Language, I: The Beginnings to 1066. Cambridge: Cambridge University Press, 67–167.

46

Patrick Honeybone

Honeybone, D., & Honeybone, M. 2010. The Correspondence of the Spalding Gentlemen’s Society 1710–1761. Lincoln: Lincoln Record Society. Honeybone, M. 2008. ‘Practising history: Writing a history book today’. East Midland Historian 16, 12–17. Honeybone, P. 2005. ‘Diachronic evidence in segmental phonology: the case of obstruent laryngeal specifications.’ In: van Oostendorp, M., and van de Weijer, J. (eds), The Internal Organization of  Phonological Segments. Berlin: Mouton de Gruyter, 319–354. Jenkins, K. 1991. Re-thinking History. London: Routledge. Job, M. 1989. ‘Sound change typology and the “Ejective Model”.’ In: Vennemann, T. (ed.), The New Sound of  Indo-European. Berlin: Mouton de Gruyter. Jones, C. 1989. A History of  English Phonology. London: Longman. Kager, R. 1995. ‘The metrical theory of word stress.’ In: Goldsmith. J. (ed.), The Handbook of  Phonological Theory. Oxford: Blackwell, 367–402. Kemenade, A. van 1987. Syntactic Case and Morphological Case in the History of English. Dordrecht: Foris. Kroch, A., & Taylor, A. 1997. ‘Verb movement in Old and Middle English: Dialect variation and language contact.’ In: van Kemenade, A., and Vincent, N. (eds), Parameters of Morphosyntactic Change. Cambridge: Cambridge University Press, 297–325. Kroch, A., Taylor, A., & Ringe, D. 2000. ‘The Middle English verb-second constraint: a case study in language contact and language change’. In: Herring, S., van Reenan, P., and Schosler, L. (eds), Textual Parameters in Older Languages. Amsterdam: John Benjamins, 353–391. Kümmel, M. 2007. Konsonantenwandel: Bausteine zu einer Typologie des Lautwandels und ihre Konsequenzen. Wiesbaden: Reichert. Lass, R. 1994. Old English: A Historical Linguistic Companion. Cambridge: Cambridge University Press. Lass, R. 1997. Language Change and Historical Linguistics. Cambridge: Cambridge University Press. Lass, R. forthcoming. ‘Interpreting alphabetic orthographies and orthographic change.’ In: Honeybone, P., and Salmons, J. (eds), The Oxford Handbook of  Historical Phonology. Oxford: Oxford University Press. Marwick, A. 2001a. ‘The Fundamentals of History’. History in Focus 2. accessed 12 August 2011. Marwick, A. 2001b. The New Nature of  History: Knowledge, Evidence, Language. Basingstoke: Palgrave. Meillet, A. 1903. Introduction à l’étude comparative des langues indo-européennes. Paris: Hachette.

History and Historical Linguistics

47

Minkova, D. forthcoming. ‘Establishing phonemic contrast in written sources: direct and indirect evidence.’ In: Honeybone, P., and Salmons, J. (eds), The Oxford Handbook of  Historical Phonology. Oxford: Oxford University Press. Munslow, A. 2001. Author’s response to O’Brien (2001). History in Focus 2. accessed 12 August 2011. Newmeyer, F. 1986. The Politics of  Linguistics. Chicago: University of  Chicago Press. O’Brien, P. 2001. Book review: ‘An Engagement with Postmodern Foes, Literary Theorists and Friends on the Borders with History.’ History in Focus 2. accessed 12 August 2011. Pei, M. 1965. Invitation to Linguistics: a Basic Introduction to the Science of Language. London: Allen & Unwin. Pintzuk, S. 1991. Phrase structures in competition: Variation and change in Old English word order. PhD thesis, University of  Pennsylvania. The Shorter Oxford English Dictionary on Historical Principles (SOED). 1987. 3rd edn. Oxford: Oxford University Press. Törnquist-Plewa, B. 2000. ‘Contrasting ethnic nationalisms: Eastern central Europe’. In: Barbour, S., and Carmichael, C. (eds), Language and Nationalism in Europe. Oxford: Oxford University Press, 183–220. Trudgill, P. 2004. New-Dialect Formation: The Inevitability of  Colonial Englishes. Edinburgh: Edinburgh University Press. Warkentin, J. 2009 (1985). ‘Historical Geography.’ The Canadian Encyclopedia. Historical Foundation of Canada. accessed 12 August 2011. Whitney, W. 1867. Language and the Study of Language: Twelve Lectures on the Principles of  Linguistic Science. London: Trübner.

Nicholas M. Wolf

History and Linguistics: The Irish Language as a Case Study in an Interdisciplinary Approach to Culture

Abstract The Irish language has been accorded a central place in historical scholarship on modern Ireland as the question of the causes and scope of its decline has become closely intertwined with explanations for the broader cultural transformation that occurred in these same years. Yet prevailing accounts of  the relationship between cultural and linguistic change raise a number of uncertainties about current Irish historiography in this area, prompting an opportunity to investigate how a better disciplinary interaction between history and linguistics might improve scholarly approaches to Ireland’s past. This chapter begins by reviewing the scholarship on this subject in an Irish context before stepping back to consider the larger methodological possibilities and challenges of combining the studies of cultural history and language. Finally, this chapter will sketch out new approaches to Ireland’s linguistic history that take a more integrative approach and are intended to prompt the possibility of their application in the study of other historical and geographic contexts as well.

1 Irish historiography and language Historians of  Ireland have devoted considerable attention to the island’s linguistic past. Undoubtedly, this willingness to focus on language can be attributed to the heightened public visibility of the movement to revive Irish over the past one hundred years, including ongoing debates today in both Northern Ireland and the Republic over how best to reverse its decline. Yet

50

Nicholas M. Wolf

the attention devoted to language as part of Ireland’s larger history cannot simply be attributed to its continued contemporary relevance. On the contrary, historians have long portrayed the decline of Irish – and particularly the precipitous shift toward English of  the nineteenth century in which the proportion of  Irish-speakers declined from approximately one-half of the island’s population in 1800 to less than 15 per cent in 1900 – as a key turning point in the emergence of modern Ireland. The importance accorded to language shift in Ireland’s history was already evident in the first full-length consideration of  the issue, Daniel Corkery’s The Fortunes of the Irish Language (1954). For Corkery, the history of the Irish language followed the same plotlines as the history of the Irish nation itself. Much as the Irish nation had been usurped by British rule before re-emerging at the end of  the nineteenth century with the establishment of a full-f ledged nationalist movement at the popular level, the Irish language had been undermined by a foreign government before being slowly returned to the centre of national life through the ef forts of revivalist organizations like the Gaelic League (founded 1893). In assuming a direct correspondence between nation and language identity, Corkery echoed the formulations of turn-of-the-century language revivalists themselves (e.g. Hyde 1986 [1892]: 153) who saw it as self-evident that a nation – in a typical nineteenth-century Romantic formulation – was the highest manifestation of a people’s culture and that language was the unique marker of that nation’s existence. Since Corkery considered the history of  the nation and the history of the language to be synonymous, his account rarely provided any linguistic details. Rather, Corkery saw it as suf ficient to trace the various political movements to resist or end British rule in Ireland between the seventeenth and twentieth centuries. Whether these historical actors ever explicitly voiced support for the Irish language – and he freely admitted that only with the Gaelic League was a body truly committed to the revival of Irish manifested – was immaterial to Corkery, since any antiBritish action would, in his formulation, represent an implied advocacy of  the modern Irish nation and therefore of  the Irish language. This particular formulation of  language and nation ref lected wider concepts in the general public understandings of Ireland’s history that had solidified by the time of publication of Corkery’s book. Developments in

History and Linguistics

51

Irish history seen as oppositional to the re-emergence of an Irish nation on a culturally distinct basis were elevated to centre stage as explanations for the decline of Irish, regardless of whether this made sense from a sociolinguistic standpoint. Daniel O’Connell (1775–1847), for instance, whose political movements for Catholic emancipation and repeal of  the Act of  Union in the 1820s and 1840s had not acknowledged a need to restore a separate Irish culture, was therefore seen by scholars as committing a particularly dire sin of omission that contributed to language near-death. The Catholic clergy were similarly singled out as primary causes for the decline of  Irish insofar as they had not actively sustained its use despite their leadership roles in local parishes. Finally, the national school system established in 1831 – which, tellingly, also had long been a target of  the language revival movement because it did not, until the 1870s, include Irish in the curriculum in any way – was identified by historians of  Corkery’s generation as the most significant cause for the shift to English. A subtle but powerful reconsideration of these conclusions took place in subsequent years as more archivally intensive, scientifically oriented scholarship emerged in the 1960s. Among this scholarship, the work of  Maureen Wall, Brian Ó Cuív, and Seán de Fréine deserves most attention. Wall’s writing took the most revisionist stance of  the three. The notion that O’Connell, the national schools, and priests had together prompted language shift had, Wall wrote (1969: 81), been an ‘oversimplification’ perpetuated among both academic and general histories ‘in the manner of a catechism answer’. Although Wall provided a slate of evidence challenging earlier assumptions about shift, her most persuasive involved chronology: estimates showed that Irish speaking had already declined considerably before 1800, long before O’Connell’s leadership of a popular movement to secure Catholic emancipation in the 1820s, the introduction of the state schools in 1831, or the re-emergence of a powerful Irish priesthood in those same years. Such findings were echoed in work by Ó Cuív (1969, 1980), who introduced a measure of careful scholarship and mastery of archival sources in his approach evident, for example, in his use of census material and close reading of early documents in Irish. It was de Fréine’s first English-language version of his earlier writings on the subject, The Great Silence (1965), however, that articulated the most

52

Nicholas M. Wolf

in-depth explanation for the decline of  Irish that continues to undergird current historiography. Like Corkery, de Fréine of fered little in the way of sociolinguistic data or analysis. His work instead took as a starting point the investigation of a narrow question: why did Ireland experience a persistent f lood of emigration in the nineteenth and twentieth centuries? The answer, according to de Fréine, could not be found in the usual theories that had been of fered over the years to explain Irish migration, among them economic factors, racially based causes, or religious justifications. De Fréine even cast a critical eye on the role of imperial oppression, arguing that British rule could hardly explain ongoing emigration from postindependence Ireland nor account for discrepancies between Ireland and other nations subject to foreign rule which had not experienced similar mass exodus. Rather, de Fréine proposed that a rupture in Ireland’s culture itself had occurred in the nineteenth century. Culture, in his formulation, is an organic entity made up of communal bonds and shared understandings among all members of a community. Since the ability to pass on these traditions is enabled by language, it was the loss of the Irish language in the nineteenth century that created the abrupt break in Irish cultural continuity and thus the long-lasting climate of national psychological harm and emigration (1965: 79–104). De Fréine’s emphasis on nineteenth-century cultural disjuncture has had a long afterlife in research into other aspects of  Irish history in this period. The concept of cultural loss, for instance, appears in Emmet Larkin’s seminal work positing a Catholic ‘devotional revolution’ in the 1850s and 1860s that enshrined continental European religious practices in place of older indigenous devotions. Larkin (1972: 649) argued that a major factor encouraging the adoption of these religious forms had been the decline of  Irish-speaking culture in the postfamine period that had made Irish men and women into cultural ‘emigrants’ and therefore susceptible to new Catholic orthodoxies. More recent studies of  the decline of  folk practices such as healing by fairy doctors and other forms of popular cultural similarly maintain many of  the assumptions charted by de Fréine. Angela Bourke’s The Burning of Bridget Cleary (1999) portrays the linguistic frontiers between Irish- and English-speakers in nineteenth-century Ireland as directly mapping the break between a religiously unorthodox, orally

History and Linguistics

53

based, pre-modern culture and an Anglicized modern world that emerged sometime in the late 1800s. Notably, this model has also underpinned the latest research into language shift itself, as in Ní Mhóráin’s (1997: 9–20) characterization of  the decline of  Irish as a ‘comhlint chultúrtha idir dhá theanga’ (cultural conf lict between two languages) in which one culture was substituted for another. Despite its longevity, this framework deserves closer scrutiny and revision – especially in light of counterevidence emerging in recent research. Some weaknesses were already evident in The Great Silence itself. If the loss of the Irish language caused the loss of culture, what, according to de Fréine, had sparked language shift? Here, his analysis began to show a circularity as he fell back on political and economic explanations to account for the shift that he had rejected earlier in his study. The Great Silence argues that the failure of  the Jacobite cause in 1745 – the last attempt to restore the Stuarts to the English crown – had cemented the powers of  the Englishspeaking Protestant ascendancy in Ireland. At the same time, the growing strength of the Irish Catholic middle classes in the late eighteenth and early nineteenth centuries prompted a desire to learn English, the language of commerce. In short, Irish declined because English became the language of economic advancement and political dominance, rather than the language of invasion and oppression as it once had been. Yet this embrace of  English-speaking society looks very much like the ‘cultural’ rift that de Fréine had earlier proposed as a result (rather than a cause) of language shift, and indeed he admits that it was ‘the social conditions of  the second half of the eighteenth century’ – among which one would presumably include a new, Anglicized middle-class sensibility – that ‘fostered the change of  language which ensued’ (1965: 141–145). Recent archival investigations have raised further questions. It had been assumed, for example, that the Irish language had become associated with the domestic sphere by the early nineteenth century, allowing the English language to exclusively dominate the new modes of politics emerging by way of widespread newspaper circulation, electioneering, the large outdoor rallies known as ‘monster meetings,’ and other features of  the public sphere. This exclusion of  the Irish language from politics was supposedly a symptom of its larger incompatibility with modernity (e.g.

54

Nicholas M. Wolf

Ó Loingsigh 1974: 13). But ongoing examinations of  the surviving Irishlanguage manuscripts of  the nineteenth century by Cornelius Buttimer, for example, have shown a sophisticated awareness among Irish-language commentators of world political events, including the War of  American Independence and the Napoleonic Wars (1990a, 1990b, 1992, 1994). The notion that Irish-speakers conceived of a fundamental divide between the Irish-speaker and modern politics looks untenable in light of  these findings. Similar discoveries regarding the use (rather than the prohibition) of  Irish to expand and entrench new religious sensibilities and orthodoxies have also cast doubts on earlier conclusions by Larkin (Ó Ciosáin 2005: 150–151, and Wolf 2008: 86–138). Most seriously, there are fundamental uncertainties about the way culture itself is conceptualized in this historiography. Often, it is portrayed as an external entity independent of human agency – evident in the concept that culture can be ‘possessed’ or ‘lost’ rather than made through ongoing human interaction and contestation. Furthermore as Niall Ó Ciosáin (2005: 139–142) has pointed out, these studies fail to articulate the nature of  the connection between culture and language despite their preeminent reliance on a direct link in order to make their case. De Fréine, for instance, had viewed it as self-evident that language has a fundamental role in perpetuating culture, but never clearly defined the practices, beliefs, or relationships that he believed entailed ‘culture’ or how the Irish language sustained them. At best, he merely insisted (1965: 105–108) that the culture of Irish-speakers not be defined narrowly – that is, by the attributes of  the rural communities in which most nineteenth- and twentieth-century speakers could be found. As for how one language could sustain a particular ‘Irish culture’, but not another, the reader of  The Great Silence is left without direction. Clearly, a better formulation of the terms of debate in this historiographical investigation is needed.

History and Linguistics

55

2 Approaches to language and culture The problematic treatment of  the relationship between culture and language by scholars is hardly confined to de Fréine or to the field of  Irish history. The history of  this subject continues to be underexplored by historians, although there are recent signs that the field is growing (Burke 1987: 1 and 2004: 3). Even when discussing the issue, historians have often treated language as a transparent phenomenon due to the discipline’s lack of a firm command of developments in the field of  linguistics, especially sociolinguistic approaches that aim at studying the phenomenon of  language in the wider context of human society. Consequently, when language appears in historical accounts it is usually as an afterthought – that is, as an element encapsulated by the larger concept of  ‘culture’ but without considering the fundamental dynamics of language itself. Historians rarely, for example, acknowledge the ways in which language encompasses variety (as in dialect, register, or other aspects of a speech repertoire), complexity in application and operation (e.g. conversational structure, bilingualism, code-switching), or a combination of any number of these elements. Nor is there usually a consideration that speakers themselves were aware of  language dif ference, or that language itself might shape the ways speakers conceive of  their social relationships. This oversight may be understandable, however, given the unsettled – even conf licting – treatment of  the language-culture question in the discipline where an answer might naturally be sought to this question: linguistics. On the one hand, there is a firmly established approach to language that sees its features as indexing larger categories in society such as religion, class, gender, and ethnicity. Such studies seek to understand how dif ferences or changes in language might be explained by the status and features of the communities created by these categories. Thus, for example, a recent publication on overgeneralizations of certain gender forms in Dutch among non-native communities considers the possibility of correlating this tendency with immigrant identity (Cornips 2008: 114–120). Lanza and Svendsen (2007: 277–278, 284–296), while acknowledging the multiplicity of identities and the problems of overly mechanical approaches to ethnicity,

56

Nicholas M. Wolf

nevertheless see language and identity as separate variables whose connection they seek to elucidate for multilingual Filipino language networks in Oslo. Conversely, language may be treated as an independent entity that determines cultural identity. Jones (1998: 129–142) investigates the ways in which the growth of a school-learned, standardized dialect of  Breton has in recent decades prompted a wider Breton identity that competes with older allegiances to the local commune and to a broader assimilation to French national identity among the inhabitants of  Brittany. Such findings would appear to provide a suf ficient basis for a historical investigation into language by tracing the transformation of these external social categories. However, this picture is complicated by another a welldeveloped body of research that emphasizes the endogenous nature in which language behaves. One need not confine oneself, say, to the debate over Neogrammarian explanations for language change to find an ongoing conversation over the relative weight of internal and external factors in driving variation. Even among those that see social factors at play in linguistic change, there is disagreement about whether those developments require speakers simply to interact – as in concepts of  koineization, for instance, in which cultural context only needs to provide details about how and when a speech act takes place – or whether speakers themselves enact change directly based on the ideologies of categories of gender, class, or nation. As result, work such as that recently published by Trudgill (2008: 241–254) can make a persuasive case for dialect accommodation of colonial Englishes that explicitly excludes cultural identity as an explanatory factor, while long-standing research into the fundamentally behavioural nature of  language change (e.g. Labov 1994: 13–34) would likewise seem to undermine an oversimplified view of cultural identity as an automatic explanation for any particular linguistic manifestation. My point in citing these studies is not to comment on the merit of  their conclusions, which involve nuanced consideration of many other linguistic questions beyond the issue of the relationship between culture and language. Nor is it possible to trace here the entire trajectory of  linguistic debate over culture and language which, of course, has long been one of  the central features of  the discipline throughout a professional history encompassing Hymes, Ferguson, Whorf, and other scholars of the mid- and

History and Linguistics

57

late twentieth century. Rather, my objective here is to suggest that the historian searches in vain among linguistic case studies for a ready model to use in addressing the language-culture problem. Fortunately, developments common to both the disciplines of history and linguistics in the past three decades have better equipped scholars to address this question as well as find common ground. A solution to a historiographical uncertainty such as that of  the Irish language in nineteenth-century Ireland may in fact require a hybrid examination of the language-culture problem by scholars. Sketching out these evolving approaches to culture helps make clear the possibilities (and even necessity) of combining historical and linguistic techniques in order to untangle the past.

3 Culture as a process By the early 1980s, a number of studies emerging out of cultural anthropology began to place increasing emphasis on the role of  human agency in defining and ordering the relationships that made up society (Ortner 1984: 144–156). In part this was a reversal of  longstanding structuralist views in the social sciences of human behaviour as a by-product of broader economic, political, or social realities – an approach that had begun to appear overly deterministic in its treatment of  the individual or community. Instead, ‘culture’, or ‘the cultural’, was re-defined as the sum of human actions and discourses (i.e. ‘practices’) which together produced (and gave meaning to) collective understandings of categories (or ‘identities’) such as class, gender, or nationality. This rejection of the notion of innate identity brought along with it a number of  linked concepts. It prompted the consideration of community bonds as a function of a ‘process’ or ‘production’ that relied on exclusion (i.e. defining the ‘Other’) as much as inclusion. Furthermore, the treatment of identity as contingent, mutable, and multiple meant a greater emphasis on diachronic study in order to explain change. This contrasted with older anthropological studies that had tended to of fer largely synchronic observations of communities.

58

Nicholas M. Wolf

This shift in anthropology, given this renewed concern with identifying and understanding change, has solidified its impact on the discipline of  history in the past twenty years. The result has been a body of scholarship sharing a number of  key assumptions regarding the need to historicize cultural practices and to understand how those practices are mediated by inequalities in power and inf luence (Sewell 1999: 52–55). Identities, these historians urge, cannot be treated as transparent and unchanging. Not only are they contingent on time and place, but cultural categories themselves must be treated as historically determined. One of  the most active fields emphasizing this need to historicize cultural identity has been the history of nationality and nationalism. Here, scholars such as Hobsbawm and Ranger (1983) and Anderson (2006), among others, have insisted that communities not only constantly negotiate the features of  their own national identity, but that the concept of  ‘nation’ itself as an organizing principle for a collective group is a product of the past 200 years. The same idea can be applied to class or gender, similar categories that must be understood to have a dependent relationship with the society in which they exist. The integration of this approach with the discipline has been liberating in terms of describing how history works. Armed with a greater awareness of  how cultural ‘production’ aims to define identity, historians can trace changing social categories through all levels of society, placing particular emphasis on the capacities and limits of various individuals to control those same categories. This also means that social settings hereto assumed to be nonpolitical – the private sphere, for example, in which certain roles such as gender are defined and enacted – have received significant attention due to an awareness of  how they shape history. The need to explain how these identities are negotiated and constructed demanded a closer look at the role of  language in driving this process, a development that reinforced previous concerns among structuralists and nonstructuralists alike. Described as a ‘linguistic turn’ in interpreting culture, this historiographical revolution significantly expanded the tools brought to bear by historians on dif ficult archival sources even as it raised awareness of  the ways in which language determined the act of writing history itself. The history of events such as the Reformation and the French Revolution, according to scholars like Davis (1975) and

History and Linguistics

59

Ozouf (1988 [1976]), could be reinterpreted by paying closer attention to how human behaviour itself  functioned as a linguistic text, whether in the form of  the rebellious rituals of adolescent male journeymen printers in the case of  Davis, or the quasi-religious French Revolutionary celebrations studied by Ozouf. Others, most notably Hayden White (e.g. 1987: 2–3), drew attention to the ways in which language, through conventions of grammar and syntax, defines dif ferences between narrative and nonnarrative forms of historical reporting – a distinction considered essential to modern western concepts of the past. Ultimately, by focusing on the means by which culture was negotiated (in this case, language as discourse), these approaches made essential contributions to the emerging historiography of  the 1980s and later. Yet even as historians have used the concept of  language to inform their new, anthropology-based approach to culture, their methods continue to define language narrowly. Language, in many of  these studies, remains a metaphorical and a readable ‘text’ in a Saussurian system in which discourse (e.g. expressions printed in newspapers, or phrases used in political speech) acts as a signifier that produces meaning. By contrast, the speech act as a behavioural or ethnographic performance in the sense understood by sociolinguists remains largely unacknowledged by historians, who tend to borrow from literary disciplines far more than linguistics. There are exceptions. A number of  historians addressing the history of nationalism and the emergence of the modern state have specifically addressed the role of language standardization and the existence of minority dialects in these developments. Early work on France, for instance, by Eugen Weber (1976) has been supplemented in recent years by Ford (1993) and Bell (1995) who have concentrated on the many linkages between French Republicanism’s concept of universalism, the emergence of  French national identity, and the enshrinement of an of ficial dialect of French. The case of the multilingual Habsburg empire has received a similarly detailed overview by Robert Evans (2004), while scholarship on the social history of Welsh (e.g. Jenkins 2000) has been unusually active. These studies have been particularly helpful in directing the attention of scholars to the role of the state in shaping multilingual societies, thereby reminding historians of both the power and limits of governments to direct their population’s linguistic makeup.

60

Nicholas M. Wolf

Fortunately, it can be argued that the present identity-based approach to cultural history invites historians to incorporate a more sociolinguistically informed look at language. Certainly, the role of  language as one of many tools in establishing the cultural identity of speakers both consciously and unconsciously – hence the importance of  language to the cultural process – should be self-evident. As a matter of  fact linguists have long been exploring this area, especially as a result of  the same developments in anthropology that had made such a strong impact on historians. There have been, for instance, calls from within the sociolinguistic field for a more critical use of social, economic, or political contexts as an explanation for language features. As Cameron (1990: 81–86) argues, doubt must be cast on the treatment of social structures as independent and determinative inf luences on language; the search by linguists for correlative links between measured variables such as ethnicity, class, or gender and linguistic features assumes, rather than proves, a certain relationship between culture and language. This has prompted linguists to devote more time to understanding how the relationship between language and identity really operates. For Gafaraga (2005: 282–283), extending Cameron’s rejection of  language as simply ref lecting essentialized social identities, language itself  ‘structures society’ – a clear call to take up the notion of cultural identity as a contingent set of categories dependent in part on language itself. These identities, as Irvine and Gal (2000: 74–76) note, have been increasingly considered by linguists since at least the 1970s to encompass not shared, homogenized bonds based on language, but rather a ‘managed’ linguistic diversity. Cultural identities, in this formulation, are not given entities but are constantly calibrated based on assertions of dif ference with an excluded outside group. Indeed, Irvine and Gal go even further in pointing out that a community’s conceptualization of  the features and significance of its own language (as distinct from the language as measured by the linguistic scholar) is a product of  the same cultural process of definition and exclusion responsible for any other category of identity. This emphasis on language as defining identity – rather than cultural identity defining language – adds weight to the explanatory powers of psychology- and behaviour-based explanations for events like language change

History and Linguistics

61

or language shift. Consequences of language contact such as creolization, koineization and dialect levelling can be accounted for quite comfortably within a framework that addresses directly the dynamics of human interaction, the context that Mufwene (2008: 15–26) refers to as the ‘ecology’ of speakers. This is not to suggest that the social realities and circumstances of speakers do not help mediate how (and if ) those interactions take place. However, recognizing identity as an unstable and constantly mediated category makes cultural explanations unsuitable for such large-scale phenomena like language variation. Indeed, if it is accepted that identity only acquires meaning through the definitions and understandings of individuals themselves, it is not possible to conceive of culture as a pre-existing condition capable of determining language. Such a conclusion is discernible in the discomfort expressed by Trudgill (2008: 252–253) concerning ‘identity’ as an explanation for the form of colonial Englishes, and more fully developed in a response by Tuten (2008: 259–260) to Trudgill’s conclusion. Speaking of koineization in medieval Spain, Tuten postulates that formation of a ‘frontier identity’ may be a simultaneous feature of  language change, a much more plausible paradigm for culture than the theories for colonial dialects that Trudgill ef fectively refutes. The virtues of this approach include the ways it opens up possibilities for scholars to focus on the impact of  language variation on identity by liberating culture from serving as a determinant of language. The study of  how speakers perceive their own speech acts, the meanings they attribute to them, and how that meaning is deployed to shape worldviews would thus occupy centre stage in this approach. Irvine and Gal’s concepts of erasure and iconization certainly of fer compelling tools for unpacking this process: the role of language change (or language maintenance) in defining identity would be mediated by the meaning that a community itself attributes to linguistic features. One wonders how the history of emerging working-class solidarity in the early industrial era, for instance, might be rewritten based on an exploration of the communal bonds of shared linguistic features like dialect created by diverse workers increasingly interacting in a large-scale factory setting.

62

Nicholas M. Wolf

4 Proposals for ongoing research To return to the Irish case, two ingredients of any future research agenda in this field are suggested by the above attempt to sketch out a disciplinary overlap between linguistics and history. First, there is a need to take a more behaviourally based approach to understanding language shift in Ireland that integrates many of the observational approaches and questions asked by sociolinguists, especially in the areas of language acquisition, language change (certainly a phenomenon such as accommodation), and language maintenance. The justification for this lies partly in the facts already known about Ireland’s language shift. It is accepted that the adoption of  English and even the abandonment of Irish had already begun in parts of the country as early as the seventeenth century. This process had commenced even earlier in the region around Dublin subject to English rule, the Pale. Such an extended time frame suggests a gradual and long-term transformation in language use, making attempts to explain language shift in terms of specific events such as the introduction of national schools, the Great Famine, the Act of Union (1801), or any other such single occurrence insuf ficient. Indeed, this chronology had long ago been recognized by Wall (1969), although its ramifications for explaining the decline of  the Irish language have yet to become fully entrenched in the scholarly literature. Similarly, the geolinguistic account of  Ireland’s language shift that has already been sketched out by FitzGerald (1984 and 2003) suggests not a replacement of  Irish by English in an uneven patchwork fashion, but in a clear and receding line running roughly north-south and moving westward over at least two centuries. This, too, argues against any island-wide cultural shift in prompting language change, but rather a gradual and persistent behaviourbased transformation that requires specific sociolinguistic investigation into intergeneration transfer and network ties. An advantage of this approach is that it helps support a second necessary investigative feature that arises out of the initial historiographic question of this paper regarding the connection between Irish cultural identity and Irish speaking in the nineteenth century. By removing the need to force language shift to conform to a timeline of cultural change, the historian

History and Linguistics

63

is free to consider a more complex treatment of  Irish culture itself. This approach should incorporate the understanding of cultural identity as a product of language behaviour, with both treated as constructed enterprises based on a number of ingredients: the speech act itself, the speaker’s and attitudes towards that language (including erasure, iconization, etc.), and the inf luence of key institutions and settings (such as religious bodies, the family, the marketplace, or the public sphere) that help define speakers’ views. The result will likely be the discovery of multiple identities constructed by language use, from a broader Irishness defined as the Irishspeaking nation to a local Irishness based on dialect. It is likely, moreover, that plural geographical organizations of cultural identity will not be the only finding when this more open approach is taken. The anti-modern/ modern scheme that scholars often map directly onto the Irish/English linguistic divide should also be considered as only one of several possible views by nineteenth-century speakers of their language. Since such attitudes are not intrinsic, but rather historically contingent, the existence of alternate views of the Irish language as compatible with modernity – however hidden or exceptional – will need to be acknowledged and explored. Obviously, simply adopting this more expansive view of how language ‘worked’ hardly solves the problem of retrieving such detailed sociolinguistic answers from the past. As historical linguists repeatedly acknowledge, the archival record is not always suited to answering such questions. It is far easier for historians to simply argue that a broad economic, political, or social transformation somehow prompted language change than to uncover the behavioural details at the local or individual level that were the actual cause for shift. But this should not discourage scholars from attempting to do so, and in any case the contention of this chapter remains that a shift in approach is needed as much as any return to the archives. Since Corkery first published The Fortunes of  the Irish Language in the 1950s, the methods applied by scholars to this issue have not evolved beyond many of the findings of subsequent revisions in the 1960s. This fact in itself should prompt a return to some of  the fundamental assumptions behind these studies. Fortunately, the centrality of the Irish language to Irish historiography suggests that this issue will remain salient long enough for scholars to revisit this particular aspect of  the history of  Ireland.

64

Nicholas M. Wolf

References Anderson, Benedict. 2006. Imagined Communities: Ref lections on the Origin and Spread of  Nationalism. London and New York: Verso. Bell, David. 1995. ‘Lingua Populi, Lingua Dei: Language, Religion, and the Origins of  the French Revolutionary Nationalism.’ American Historical Review 100, 1403–1437. Bourke, Angela. 1999. The Burning of  Bridget Cleary. New York: Penguin. Burke, Peter. 1987. ‘Introduction.’ In: Peter Burke and Roy Porter (eds), The Social History of  Language. Cambridge: Cambridge University Press, 1–20. Burke, Peter. 2004. Languages and Communities in Early Modern Europe. Cambridge: Cambridge University Press. Buttimer, Cornelius. 1990a. ‘A Cork Gaelic Text on a Napoleonic Campaign.’ Journal of  the Cork Historical and Archaeological Society 95, 107–119. Buttimer, Cornelius. 1990b. ‘An Irish Text on the “War of  Jenkins’ Ear”.’ Celtica 21, 75–98. Buttimer, Cornelius. 1992. ‘A Gaelic Reaction to Robert Emmet’s Rebellion.’ Journal of  the Cork Historical and Archaeological Society 97, 36–53. Buttimer, Cornelius. 1994. ‘Cogadh Sagsana Nuadh Sonn: Reporting the American Revolution.’ Studia Hibernica 28, 63–101. Cameron, Deborah. 1990. ‘Demythologizing Sociolinguistics: Why Language Does Not Ref lect Society.’ In: John E. Joseph and Talbot J. Taylor (eds), Ideologies of  Language. London and New York: Routledge, 79–93. Corkery, Daniel. 1956. The Fortunes of  the Irish Language. Repr. edn. Cork: Mercier Press. Cornips, Leonie. 2008. ‘Loosing Grammatical Gender in Dutch: The Result of Bilingual Acquisition and/or an Act of Identity?’ International Journal of Bilingualism 12, 105–124. Davis, Natalie Zemon. 1975. Society and Culture in Early Modern France. Stanford: Stanford University Press. de Fréine, Seán. 1965. The Great Silence. Dublin: Foilseacháin Náisiúnta Teoranta. Evans, R.J.W. 2004. ‘Language and State-Building: The Case of the Habsburg Monarchy.’ Austrian History Yearbook 35, 1–24. FitzGerald, Garret. 1984. ‘Estimates for Baronies of Minimum Level of Irish-Speaking Amongst Successive Decennial Cohorts: 1771–1781 to 1861–1871.’ In: Proceedings of  the Royal Irish Academy, Section C, 84, 117–155.

History and Linguistics

65

FitzGerald, Garret. 2003. ‘Irish-Speaking in the Pre-Famine Period: A Study Based on the 1911 Census Data for People Born Before 1851 and Still Alive in 1911.’ In: Proceedings of  the Royal Irish Academy, Section C, 103. 191–283. Ford, Caroline. 1993. Creating the Nation in Provincial France. Religion and Political Identity in Brittany. Princeton, NJ: Princeton University Press. Gafaraga, Joseph. 2005. ‘Demythologising Language Alternation Studies: Conversation Structure vs. Social Structure in Bilingual Interaction.’ Journal of Pragmatics 37, 281–300. Hobsbawm, Eric, & Terence Ranger (eds). 1983. The Invention of Tradition. Cambridge: Cambridge University Press. Hyde, Douglas. 1986 [1892]. ‘On the Necessity for De-Anglicising Ireland.’ In: Breandán Ó Conaire (ed.), Language, Lore and Lyrics. Essays and Lectures. Dublin: Irish Academic Press, 153–170. Irvine, Judith T., & Susan Gal. 2000. ‘Language Ideology and Linguistic Dif ferentiation.’ In: Paul V. Kroskity (ed.), Regimes of  Language: Ideologies, Polities, and Identities. Santa Fe (NM): School of  American Research Press and Oxford: James Currey Ltd, 35–83. Jenkins, Geraint H. (ed.). 2000. The Welsh Language and its Social Domains, 1801–1911. Cardif f : University of  Wales Press. Jones, Mari. 1998. ‘Death of a Language, Birth of an Identity: Brittany and the Bretons’. Language Problems and Language Planning 22, 129–142. Labov, William. 1994. Principles of Linguistic Change. Volume 2: Social Factors. Malden, MA: Blackwell Publishers. Lanza, Elizabeth, & Bente Ailin Svendsen. 2007. ‘Tell Me Who Your Friends Are and I Might Be Able to Tell You What Language(s) You Speak: Social Network Analysis, Multilingualism, and Identity’. International Journal of  Bilingualism 11, 275–300. Larkin, Emmet. 1972. ‘The Devotional Revolution in Ireland, 1850–1875.’ American Historical Review 77, 625–652. Mufwene, Salikoko S. 2008. Language Evolution. Contact, Competition, and Change. London: Continuum International Publishing Group. Ní Mhóráin, Brighid. 1997. Thiar sa Mhainistir atá an Ghaolainn Bhreá: Meath na Gaeilge in Uíbh Ráthach. An Daingean: An Sagart. Ó Ciosáin, Niall. 2005. ‘Gaelic Culture and Language Shift.’ In: Laurence M. Geary and Margaret Kelleher (eds), Nineteenth-Century Ireland: A Guide to Recent Research. Dublin: University College Dublin Press, 136–152. Ó Cuív, Brian (ed.). 1969. A View of  the Irish Language. Dublin: Stationery Of fice. Ó Cuív, Brian. 1980. Irish Dialects and Irish-Speaking Districts: Three Lectures by Brian Ó Cuív. Dublin: Dublin Institute for Advanced Studies.

66

Nicholas M. Wolf

Ó Loingsigh, Padraig. 1974. ‘The Irish Language in the Nineteenth Century.’ In: Oideas 14, 5–21. Ortner, Sherry. 1984. ‘Theory in Anthropology Since the Sixties.’ Comparative Studies in Society and History 26, 126–166. Ozouf, Mona. 1988 [1976]. Festivals and the French Revolution. Trans. Alan Sheridan. Cambridge: Harvard University Press. Sewell, William H. 1999. ‘The Concept(s) of  Culture.’ In: Victoria E. Bonnell and Lynn Hunt (eds), Beyond the Cultural Turn. Berkeley: University of California Press. 35–61. Trudgill, Peter. 2008. ‘Colonial Dialect Contact in the History of  European Languages: On the Irrelevance of  Identity to New-Dialect Formation.’ Language in Society 37, 241–254. Tuten, Donald. 2008. ‘Identity Formation and Accommodation: Sequential and Simultaneous Relations.’ Language in Society 37, 259–262. Wall, Maureen. 1969. ‘The Decline of  the Irish Language.’ In: Brian Ó Cuív (ed.), A View of  the Irish Language. Dublin: Stationery Of fice, 81–90. Weber, Eugen. 1976. Peasants into Frenchmen. Stanford: Stanford University Press. White, Hayden. 1987. The Content of  the Form: Narrative Discourse and Historical Representation. Baltimore: Johns Hopkins Press. Wolf, Nicholas. 2008. Language Change and the Evolution of  Religion, Community, and Culture in Ireland, 1800–1900. PhD dissertation. University of WisconsinMadison.

Brian D. Joseph

Historical Linguistics and Sociolinguistics: Strange Bedfellows or Natural Friends?

Abstract Inasmuch as historical linguists and sociolinguists are both interested in dif ferent types of variation – diachronic in the former case and synchronic in the latter case – one might think that they necessarily have much in common. While it is ultimately argued here that such is indeed the case, by way of exploring the relationship between historical linguistics and sociolinguistics, several notions that are stock in trade for most sociolinguists are re-examined from a historical linguist’s perspective. It is suggested, for instance, that ‘change in progress’ is a mirage, and that chain shifts are not as dramatic an event as one might think. At the same time, though, the enterprises are linked by the Uniformitarian Principle, and that construct is extended into general historical investigations through the positing of a ‘social or humanistic uniformitarianism’ that emphasizes what is similar across time in human circumstances and in human reactions to those circumstances.

1 Introduction It should be clear that historical linguistics and sociolinguistics have something to do with one another and presumably something to contribute to one another. Both areas of investigation are interested in variation, diachronic variation in the former case, synchronic variation in the latter case. Moreover, since language is always embedded in some social context, language change necessarily takes place in a given social setting. Such

68

Brian D. Joseph

connections are part of  the rationale for a characterization of  the study of the social dimensions of language change as ‘sociohistorical linguistics’ and they suggest that both social historians and sociohistorical linguists might well be in a position not only to learn from one another but also to learn from the exploration of  the interrelationships of  language and history and of linguistics and historiography. At the same time, though, a comparison just of sociolinguistics with historical linguistics would seem to be a useful way of getting at the potential for ‘value-added’ that a consideration of social factors of fers to historical linguistics. Accordingly, in what follows, I aim to see what points of similarity and dif ference there are between these two commonly paired concerns – note the very terms ‘socio-historical linguistics’ / ‘historical sociolinguistics’, after all – with an ultimate goal of determining if  the coupling of  the two is, as the title suggests, the result of joint membership in a natural class or is instead a forced marriage. In the course of so doing, I re-examine and to some extent debunk, or at least attempt to debunk, a number of concepts that both historical linguistics and sociolinguistics hold dear. In many instances, I pose questions about notions and practices without necessarily of fering answers. In the spirit of  Socrates’ adage about the unexamined life,1 my hope is that asking the right questions is helpful even if clear answers are not of fered. Some of what follows may seem obvious and maybe even trivial to the intended audience of sociolinguists, historical linguists, and social historians, but my intent is in part to call attention here to some shortcuts that practicing socio-historical linguists routinely use. In this way, we can be sure that we are aware of what we are doing when we employ them. I see two important reasons for doing this. First, it is sometimes the case that practitioners can be deluded or deceived in the worst case or just even distracted by their own terminology and their own practices, so that raising questions can be a way of  heightening awareness. Second, there is always a risk that others outside our subfield might adopt (and then alter or misconstrue) our practices without fully understanding why we do what we do, and being explicit about the practices can thus be a safeguard against that. 1

From Plato’s Apology (38a): ‘The unexamined life is not worth living’.

Historical Linguistics and Sociolinguistics

69

To illustrate this last point, we can note that historical linguists often compress sound changes when talking about them. Thus, the development in Latin in which an original s occurring between vowels ultimately turned into an r, known to linguists as ‘rhotacism’, is generally discussed as being simply s > r /V__V, even though there is good reason to believe that it went through a stage of s > z (and then z > r). For one thing, Proto-IndoEuropean *z independently developed into Latin r, as in mergo ‘dive’ from *mezg- (where Sanskrit majj- ‘to dive’ of fers confirmation for assuming *mezg- as underlying Latin merg-). And, in Oscan, a language very closely related to Latin, z is found corresponding to Latin intervocalic r from earlier *s. Thus we talk about s > r in Latin, while recognizing this to be shorthand for s > z > r (maybe even s > z > ž > ř > r). Nonetheless, in an early generative historical linguistic account of  this change (e.g. Kiparsky 1971), it was treated simply as the addition of a rule of s => r /V__V to the grammar, thus taking traditional historical linguists’ practice as defining the issue conclusively. In this way, such a treatment failed to observe Andersen’s (1989) crucial distinction between diachronic correspondence, the static comparison of two possibly widely separated stages of a language, and innovation, the first appearance of a new structure, sound, feature, etc.

2 Some similarities and dif ferences As noted at the outset, a key point of rapprochement between historical linguistics and sociolinguistics is that both disciplines are interested in variation, in the one case in variation across time periods and in the other in variation within a given time period. And relatedly, both are interested in change, but there is a dif ference here: for historical linguists, change is the focus, whereas for sociolinguists, change is just one side of  the overall phenomenon of variation. Even so, there is a further, far more crucial dif ference though, in the way each discipline approaches change, and that is in regard to the relevant timescale. Sociolinguistics typically examines what can be construed as

70

Brian D. Joseph

‘change in progress’ (on which see below), typically based on contemporary usage, while historical linguistics typically examines developments that are over and done with, often, but not always, long past over and done with. Yet, even in this dif ference there is a point of contact, of course, in that the tenets of (contemporary) sociolinguistic investigation and interpretation can be applied to past situations, as done by Teodorsson (1976), for instance, with regard to the role of genderlectal variation in Koine Greek vowel changes. This is in keeping with the Uniformitarian Principle that the processes at work in the past are no dif ferent in kind from those evident today.2 Continuing in the point and counterpoint mode, I note that the available data is also a key point of dif ference. It is especially important to keep in mind here Labov’s (1994: 11) telling characterization of historical linguistics as ‘the art of making the most of bad data’. While I would prefer, following the reasoning given in Janda and Joseph (2003: 14), to say ‘imperfect data’ rather than ‘bad data’, because the data available in a historical account typically is simply not as complete as data collected in a contemporary investigation can be, it must be admitted that the sentiment expressed in Labov’s statement is right on target: we do the best we can in a historical investigation within the limitations imposed by the data.

3 Plus ça change … – the more things change the less things change The points just made lead into what might be the single most important question bearing on possible dif ferences between sociolinguistics and historical linguistics: with regard to changes (putatively) observable now, are there dif ferences in nature/type of change or just in amount or degree? 2

On the history of  the Uniformitarian Principle, first promulgated for geology, and its subsequent application to other historical sciences, including linguistics and for relevant references, see Janda and Joseph (2003).

Historical Linguistics and Sociolinguistics

71

That is to say, do present circumstances make for fundamentally dif ferent types of change now, and is an answer here tied to issues of the availability of data? Relatedly, if the answer is that any dif ferences are just a matter of degree, we can ask how we might quantify rate of change in a meaningful way so as to make it possible to judge and compare the extent of change. It is important to note here that the ‘long (temporal) view’ of historical linguistics teaches that the answer to these questions about the nature of change in dif ferent eras is that the types of change one observes now, that is the mechanisms and processes of change, are essentially the same now as they were in the past. This, again, is what the Uniformitarian Principle asserts. Still, for many, especially lay, observers, that is not the case; thus, there are some who say that the internet, for example, has created a new type of communication and a new type of  human interaction; Paredes & Mário (2007), for instance, claim that ‘the growth of  the Internet and its associated technologies did open space for a new type of  human interaction: virtual, social interaction environments,’ and Yonkers (2008) confidently states that ‘I see this [i.e. Facebook] as an emerging new method of communication in which the rules are still evolving.’ One has to assume that with these new methods of communication and interaction, these observers would expect to eventually see dif ferent types of change emerging. My answer here is to stand by Uniformitarianism and expect that if any dif ferences are to be observed, they will be dif ferences in degree rather than kind. Presumably, time will tell here, but we can glean some insights into this matter from a few small case studies where developments today and those in earlier years are parallel but dif ferent, and the dif ferences are quantitative and not qualitative. For instance, there is a plethora of acronyms in American English today. Interestingly, a good many of  these coinages derive from modern institutions, such as governmental agencies, e.g. IRS (Internal Revenue Service), EPA (Environmental Protection Agency), FAA (Federal Aviation Authority), and OSHA (Occupational Safety and Health Administration); from contemporary professional associations, including sports leagues, and related sponsored activities like conferences, e.g. LSA (Linguistic Society of  America), TESOL (Teachers of  English to Speakers of  Other Languages), CLS ((annual meeting of  the) Chicago Linguistic Society),

72

Brian D. Joseph

NCAA (National Collegiate Athletic Association), and BCS ((college football) Bowl Championship Series), to name a few; and from modern technology, e.g. CPU (central processing unit), LOL (laughing out loud (as a text-messaging abbreviation), among others. The obvious recency of  these coinages makes it appealing to think of  this acronymic coinage as a modern phenomenon. But is it really something new? As it happens, there are documentable instances of acronymic coinage from the past, including ancient times. The most likely source, for instance, of (Latin) elementum is acronymic in nature (see Ernout-Meillet 1939: s.v.), with the word deriving from a fixed sequence ‘L M N’, much like English ‘(learning one’s) ABCs’, a claim that gains some plausibility when one considers that these letters are the first three of the second half of the Canaanite alphabet from which the Latin alphabet ultimately was derived, and thus a likely starting point for some ancient scribal lessons. The ubiquitous English – and now worldwide – word OK is another case in point, if it derives, as Read (1963a, b) has argued,3 from ‘Oll Korrect’, a ‘comical misspelling’ of an initialism ref lecting a jocular pronunciation of ‘all correct’ in east coast American slang of the 1840s that revolved around the use of acronyms and humorous word play. These two historical examples indicate that acronymic coinages have been possible for a long time. It may well be that modernity has provided increased opportunities for this seemingly mundane word-formation process to be realized and to f lourish, but it is hardly a new type of change. Moreover, given that we actually know very little about one-of f  kinds of spontaneous colloquial coinages in ancient times or even in earlier stages of modern English, it is hard to feel confident about saying that elementum and OK are unique sorts of examples for their respective eras. The burden of proof rather should rest with anyone saying that these examples do not indicate a possible word-formation strategy for those periods, one with a possibly broad extent of use. 3

I am not completely convinced that Read is right about OK, but it is the etymology that seems to have gained the greatest acceptance. We must always remember that etymology, in the words of Eric Hamp (1998: 14n.2) is a ‘brittle science’, in the sense that etymologies are generally hypotheses about the earlier form of a word and not established fact. This applies of course to the case of elementum, too.

Historical Linguistics and Sociolinguistics

73

A similar sort of argument can be made with regard to blending as a type of compounding. That is, we find numerous examples in modern times of blended forms, not just the famous deliberate creations like Lewis Carroll’s chortle, coined for comic ef fect, or weathermen’s snizzle (for ‘snowy drizzle’), coined to fill a lexical gap, but rather the considerable number of such forms to be found in names for new products or ideas and in the ‘breathless’ sort of celebrity-based journalism that is intended to attract the attention of readers interested in the latest about some well known personality. That is, the demands of  these venues, all related to marketing ideas or products or newspapers or television shows, seem to call for something ‘fresh’ and ‘new’, and blends appear to fit the bill. Thus in the US today we find product names such as Snausage and Pup-peroni for doggie treats, the verb decorganize for the commercially based new concept of  ‘organizing your home or business with appropriate décor’,4 and labels in the US celebrity media for celebrity couples such as Brangelina (= Brad Pitt + Angelina Jolie) and TomKat (= Tom Cruise + Katie Holmes).5 In this regard one can even cite deliberately humorous coinages such as the following from the Washington Post’s 2009 Mensa Invitational, which asked readers to take any word from the dictionary, alter it by adding, subtracting, or changing one letter, and supply a new definition).

4

5

In a definition that recognizes the commercial side to the word and its use, this verb is defined by The Urban Dictionary accessed 4 May 2010) as: ‘To organize one’s belongings into trendy and obnoxious looking containers. These containers are typically purchased at specialty retail stores and departments such as The Container Store and Target. These containers usually involve the following concepts: large f loral patterns, bright colors, transparent/translucent forms and unique methods of  latching and sealing’. There are at least two other relatively recently emerging ‘social’ domains where blending is rampant now: drink names, e.g. appletini, crantini, mangotini, etc., for mixed drinks served in a martini glass and consisting of vodka (or the like) and apple/cranberry/mango (etc.) juice, giving a pattern so prevalent that -tini could be treated as a morpheme (I thank Marivic Lesho for bringing this example to my attention) and names for hybrid dogs, e.g. labradoodle for a mix of  Labrador retriever and poodle, cockapoo for a mix of cocker spaniel and poodle, etc.

74

Brian D. Joseph

Here is a sampling of the forms that emerged from the competition, each one ultimately a blend, even if created under a particular set of constraints:6 ignoranus ‘a person who’s both stupid and a jerk’ bozone ‘the substance surrounding stupid people that stops bright ideas from penetrating’ Beelzebug ‘Satan in the form of a mosquito that gets into your bedroom at two or three in the morning and cannot be cast out’.

The social context for these creations – commerce, competition, and such – is certainly interesting, but so too is the apparent folk-linguistic view underlying them that words should be compositional in meaning. To return to the issue at hand, however, again we can ask if such blending is really all that new, or if instead we are just witnessing an increase in the documentation of such forms and an increase in their use; thus again, one has to wonder if this is a qualitative change or just a quantitative one. Here the evidence of an apparent ancient composite blend may point towards the same-in-kind-but-dif ferent-in-extent assessment. In particular, the source of English bring, and more generally Germanic *bringan-, is disputed, as the verb in that form seems to be isolated within Indo-European; one etymology (in the American Heritage Dictionary s.v.) takes it to be a blended form of  two roots for ‘carry’, thus *bher-enek- (=> *bhr-enk-). If  this account is right, then it would constitute evidence of a very old blending process and thus suggest that the novelty and abundance we see in today’s blends indicates a shift in degree associated with the demands of modernity rather than a new kind of change per se. Admittedly, what we do not, and probably cannot, know is whether this *bhr-enk-, if properly etymologized here, had the same sort of playful or edgy feel to it for speakers of Proto-Indo-European that some of the abovementioned recent blends in American English have; that is, we really have virtually no way

6

Further evidence of  the prevalence of  these formations comes from an article entitled ‘Lingo-making’ by Mark Leibovich and Grant Barrett in The New York Times’s Week in Review section of  Sunday December 20, 2009. In reviewing some of  the new words and phrases that constituted ‘Buzzwords 2009’, they noted some presumably deliberately coined blends such as aporkalypse for ‘undue worry in response to swine f lu’ and vook for ‘a digital book that includes some video in its text’.

Historical Linguistics and Sociolinguistics

75

of getting at Proto-Indo-European playful word-formation strategies that would correspond among the PIE speech community to the ‘hot news’ sort of blends we see today. That is, a form like *bhr-enk- may of fer a glimpse at the formal mechanisms at the disposal of Proto-Indo-European speakers, but unfortunately says nothing about its af fective value. It is important to note that if the type of change is not really dif ferent in these cases, then in a sense we are looking at a certain sort of stability. This fact must always be kept in mind when claims about change are made. I say ‘a certain sort of stability’ since there are several works recently that make claims about the relative stability of certain types of structures, where ‘stable’ means ‘less prone to change’,7 but skepticism might be in order here. That is, talking about change/stability in language can be likened to talking about change/stability in the weather in Columbus (and elsewhere): ‘if you don’t like the weather, just wait a few minutes!’ That is, if you are willing to wait long enough, everything in language is subject to change – except for the most foundational aspects of Universal grammar, the elements, like the ‘design features’ (Hockett, 1960), without which we cannot say we have a ‘language’. That is, a statement about some part of structure being ‘stable’ must be taken as relative at best, and can only make sense in social terms, i.e. in terms of what a speech community ‘allows’ to persist in their language. As a related point, there is the very interesting set of observations that Harrington et al. (2000/2006) have made on change in the vowel realizations in Queen Elizabeth’s Christmas speeches throughout her reign. In particular, over her fifty years or so on the throne, her vowels have shifted in these speeches to something more in line with younger speakers of  the standard southern British dialect. While these dif ferences would seem to be a clear case of change over the course of an individual’s lifespan, they must be assessed against the fact that we do not really know how the Queen spoke colloquially, in unguarded moments, at other stages in her life. That is, one could argue that what Harrington et al. have measured is changes in the Queen’s linguistic performance in a very circumscribed and 7

Nichols (2003), for instance, while recognizing (p. 284) that ‘nothing in language […] is truly immutable’, argues nonetheless for certain elements showing greater stability, for instance personal pronouns.

76

Brian D. Joseph

controlled set of circumstances, or perhaps even changes in her perception of  how a Queen could or should (verbally) behave in a certain context. We are not in a position, absent such data on the Queen’s colloquial usage in earlier years, to tell if  there has really been any change in her linguistic competence; she may simply be acting (or perhaps even reverting to) more colloquial realizations due to a recognition that expectations of the Queen’s role may have changed over the years. The same kind of objection can be raised to the fine and revealing study of ‘changes’ with the use of as far as to mark topics, as documented by Rickford et al. (1995). They demonstrate that in the materials they examined over a roughly 200-year span into contemporary English, topic-marking As far as NOUN + ‘VERBAL.CODA’ (e.g. As far as Bristol is concerned, I just love the city) shows a significant decrease in use compared to As far as NOUN (without the verbal coda, e.g. As far as Bristol, I just love the city); there are early examples (two occur in Jane Austen’s Emma, 1816) but the increase in recent years is dramatic. Equally dramatic too, however, is the increase in the availability of colloquial materials in recent decades; that is, it is not clear how much of  Jane Austen’s writing, for instance, is suitably colloquial nor do we have a clear idea of what her own colloquial usage was – the couple of examples that are in her works may be the tip of  the early nineteenth-century iceberg, the mass of which is hidden from view. The common thread in these last few examples is that because of an absence of crucial data, claims of change are perhaps overstated, and there may actually be more stability in such cases than change.8 Moreover, change in external circumstances may give certain processes a chance to operate but – very much in Uniformitarian style – the processes themselves are familiar ones at all times. 8

A similar view is expressed by Dennis Baron in the 14 November 2009 posting on his ‘Web of  Language’ blog () entitled ‘Dirty words you can say on television: WTF as the newest cable channel?’. In commenting on whether swearing is more prevalent now than in the past, he says ‘It’s hard to track whether people are swearing more than they used to, since there’s little swearing in the historical record, and it’s just about impossible to know how much people swore in conversation fifty years ago, or in the eighteenth century, or back in Middle Ages, or when humans first emerged from the primordial mists’.

Historical Linguistics and Sociolinguistics

77

4 Yet another key point – a basic foundational issue: What is change? A concept that has figured prominently in the discussion so far is ‘change’, and indeed much in the conclusions drawn here depends on how one defines ‘change’. In particular, is change to be located in the initial structural innovation or rather in the spread of the structural innovation? Linguists take dif ferent stances on this; moreover, the position taken has consequences for various interrelated issues. I survey here some of  those issues that stem from a decision one way or the other on this matter, though, as promised, some are merely posed as questions without there necessarily being answers attached to them. First, if change is defined in terms of  the structural innovation itself, then there is less of a rapprochement between historical linguistics and sociolinguistics. This is because historical linguistics examines completed changes (innovations) both as to their actuation (point of origin of  the innovation), and their complete or incomplete spread (since that is how dialect/language dif ferentiation occurs), whereas sociolinguistics focuses just on the spread of an innovation. Second, we might add contact linguistics as a subfield interested in change, and as closer to some extent to historical linguistics in terms of focus.9 In particular, while contact linguistics examines change brought on by a particular social setting (contact between peoples), and thus might seem to be more aligned with sociolinguistics, studies on language contact have tended to be focused more on actuation (contact as a cause of innovations) than on spread. Third, if change is defined rather in terms of the spread of innovations, then something akin to observation of change is possible (contrary to the traditional view, as in Bloomfield (1933), whereby change was held to be inherently unobservable); this is what sociolinguists mean by talking about examining ‘change in progress’. Admittedly, in another sense, this notion could relate to the timescale issue

9

I thank my colleague (and contact linguist par excellence!) Don Winford for reminding me of  this important point.

78

Brian D. Joseph

mentioned above, with historical linguists looking at completed events at a great time distance and sociolinguists looking at unfolding events in a more finely grained time scale. This leads into a fourth issue or question: The sociolinguistic literature notwithstanding, is there really observable change in progress? Haugen (1958: 777) had an interesting point to make in this regard concerning borrowing (cited in Thomason 1997: 181): ‘Since the actual moment of  borrowing is rarely observable, most conclusions about interference are based on inferences’. If change is defined as the innovation, then Haugen’s point could be made too with regard to change; however, even if change is defined in terms of spread, Haugen’s skepticism is justified since it is not the spread itself that is observed but rather only evidence of the aftermath of spread (in the form of certain distributions in the population at large). This means that talking about vowel fronting in terms, say, like ‘Speaker X shows a high degree of  fronting of [u]’ may technically be inappropriate, for it seems to imply that the movement itself is being witnessed. Rather, what is in fact being viewed is a static result of (someone having engaged in) the process of  fronting; and, that act of  fronting need not even have been undertaken by the speaker who is observed, in that such a speaker may well have simply inherited a word (i.e. learned it from a source such as a parent) with a particular vowel in it that happens to be fronter than the realization inherited (learned) by some other speaker. In some ways, this is rather like the situation commented on in Joseph (2011) regarding use of  the term ‘grammaticalized’. There I suggest that instead of saying that a speaker shows a grammaticalized feature (or structure, or construction, etc.), we might rather say, more neutrally, that the speaker shows a more grammatical feature (structure, etc.); my reason for this terminological suggestion is that talking in terms of something being ‘grammaticalized’, rather than simply stating the static fact of the speaker showing grammatical status for the feature in question, implies that the history of the feature in question and the directionality of its development are known when oftentimes the history is just a matter of speculation. Fifth, as a matter closely related to the consideration of  ‘change-inprogress’, there is yet another way one might interpret this notion that merits some discussion. This other interpretation has to do with the widespread belief  that much of what is seen in language change is gradual. That is, a

Historical Linguistics and Sociolinguistics

79

progression of shifts is often given as a basis for talking about change as gradual, so that the Latin rhotacism change alluded to above in section 1 could be called gradual if it in fact proceeded as s > z > r (or even s > z > ž > ř > r). This characterization is a reasonable one, and common to be sure, but it must be recognized that a language could in principle stop after any of  these steps. That is, progression to the next step is not a necessary outcome, and indeed, if there was any appreciable time between the shifts, perhaps as little as a generation or even less, then clearly the stage reached by a given shift was a viable language state that speakers could learn and use unaltered. This means that in a certain sense, each shift can be treated as a change in and of itself, each one an innovation. Admittedly, from an etymological standpoint, gradual means ‘step-by-step’, but under the interpretation suggested here, instead of using examples like Latin rhotacism to show ‘gradual change’, it might be better to call it ‘(the result of ) cumulative change’, so as to emphasize the temporal sequencing of separate and distinct innovations. In this view too, it might be said that every change, every shift, every innovation, is abrupt, in that each one represents a break from, a deviation from, an existing norm, even if it is a ‘micro-step’, just a small innovation. Sixth, relevant here too is the claim that ‘chain shifts’ represent a special mechanism of change (so Labov 2007); we can understand ‘chain shift’ in a broad sense, referring to any linked sound changes, including cases where sounds A and B move jointly in the direction of sounds C and D respectively, or in a narrow sense, including cases where sound A moving in the direction of sound B is said to trigger a change in B, moving it in the direction of a dif ferent sound C. With a view of  ‘change’ that focuses on innovation rather than spread, this claim needs to be re-examined as to mechanisms and outcome. In particular, if sound A moves in the direction of sound B, then one of  two things can happen: either there is merger (of  A with B) or there is no merger. And, as for the no-merger situation, while it can be the case that there is no merger because A does not move all the way into the realm of  B, another way in which there could be no merger would be B moving somewhere; in such a case, why must that be viewed as having been caused by A’s movement? In such a case, how do we rule out the possibility of  these simply being two independent changes, especially since we know that mergers do happen.

80

Brian D. Joseph

To elaborate on this last point somewhat with a real example, we can consider the sound shifts between Proto-Indo-European (PIE) and the Germanic (Gmc.) languages known collectively as ‘Grimm’s Law’ whereby PIE *dh ultimately gives Gmc. d, and PIE *d ultimately gives Gmc. t. It is possible to conceptualize these changes as chain shifts in that when PIE *dh moved in the direction of  PIE *d on the way to Gmc., PIE *d moved in the direction of *t. This certainly looks like a chain shift when presented this way, with *dh > d triggering the *d > t change. However some other languages, for instance the Balto-Slavic group, underwent a merger of *dh and *d, so that when *dh moved in the direction of *d for Balto-Slavic, the *d stayed put, and a merger occurred. Thus a causal connection between the two Gmc. changes may be hard to maintain. Seventh, there is the related issue, related since chain shifts would seem to involve whole systems of sounds, or at least subsystems within the overall phonological space, of whether there can be ‘system-based’ motivation for change, e.g. change motivated by such considerations as symmetry and balance in the phonological inventory. Here, I would suggest that it is fair to at least wonder whether the system is a reality, and a concern – subconsciously – for speakers as such, or rather is actually just a nicety that is important for linguists only. In other words, are speakers as taken by ‘symmetry’ as linguists are? Eighth, apparent chain shifts seem to involve one change causing another change, but more generally, it can be asked what it means to talk in terms of one ‘event’ causing another? In some instances, temporal adjacency is all that can be demonstrated or perhaps one event is just one contributing factor to an outcome. To take an analogy from sports, if  Bristol City’s Liam Fontaine makes a pass to Nicky Maynard that leads to a goal, he is credited with an assist, and rightly so, but it may be as much Maynard’s ef fort that caused the goal as Fontaine’s pass. Moreover, in a certain sense, everything that precedes X lays the foundation for X, so one could argue that Fontaine’s being born on 7 January 1986 was a causal contributor to the pass he made leading to a Maynard’s goal. And, to continue with a critique of chain shifts, one kind of chain shift – the broad-sense type in which A & B both undergo the same shift, e.g. to C & D – can be reinterpreted to be really just a matter of  how we

Historical Linguistics and Sociolinguistics

81

formulate the relevant events. For instance, mid-vowel raising in northern dialects of Modern Greek af fects both e and o, but their shifting in tandem may be a function of generalization over mid-vowels rather than e going first and then ‘pulling along’ with it the o (in what is sometimes called a ‘drag chain’). In that case, rather than seeing the change(s) as two separate events that are linked by a special type of shifting mechanism, one could simply say that the relevant mechanism is ‘rule generalization’. And, once one takes that step, then it would seem that the real mechanism involved is analogical in nature, specifically a ‘phonetic analogy’ (in the sense of  Vennemann, 1972) that can be schematized as: e is-to i as o is-to X, with X then ‘solved for’ as u (that is, e : i :: o : X, X => u). And, if  that is the case, then the change involved may not be sound change in the strict sense, if it is analogically induced. Alternatively, it may have to be admitted that ana­ logy plays a far greater role in sound change than we might think. In this regard, the interpretation of fered by Anttila (1977) for the Grimm’s Law shifts is noteworthy, as he suggests that they may have had an analogical basis (p : f :: t : θ :: k : x, etc.); see also Hock (2003), where there is further argumentation about sound change as analogy, and Durian (2009, 2012), who pursues a similar line of reasoning concerning some of the vowel shifts evident in contemporary English of  Central Ohio. Ninth, to shift to a dif ferent thorny issue, we can ask how one might go about quantifying change – assuming, that is, that we even can quantify change – in the light of  these dif ferent definitions. In particular, if each ‘micro-innovation’ counts as a change, and if each micro-innovation involves a relatively small shift of some sort, then even fairly straightforward changes that can be characterized as the alteration of a single feature, such as k => x (where only manner of articulation changes from stop to fricative) could represent a large number of minute adjustments, each of which could constitute a separate change, e.g. k first becoming aspirated, then various adjustments to the strength and timing of the aspiration occurring leading to an af fricate-like realization (e.g. as kx or even kx), which is then simplified to x. Tenth, if change is to be located in micro-innovations, then it becomes hard, in taking the long temporal view that many historical linguists do, to legitimately talk about the ‘same’ change over long periods of time. For

82

Brian D. Joseph

instance, repeatedly throughout the history of  Greek, though sometimes at a time distance of several hundred years, one finds the loss of word-final -n occurring. However, is it the ‘same’ change, activated at dif ferent points in time or a separate and distinct change in each era in which it is found? Without a clear reason for linking the events, such as the recurrence in dif ferent time periods of a highly unusual or particularistic constraint on the change, it would seem prudent to treat such cases as involving temporally and ontologically distinct instances of  final n-loss, even at the expense of seeming to miss a generalization (a ‘recurring tendency’ in a given language) over a long time-span; for one thing, once one is talking about time spans of 120 or more years, it must be borne in mind that the speakers, i.e. the members of  the af fected speech community, are an entirely dif ferent set of individuals. For such reasons – the absence of any ‘smoking gun’ types of special conditions repeated across time, and the complete turnover of  the af fected population with each ‘wave’ – I am inclined to distinguish (at least) three waves of n > Ø /__#, i.e. loss of n in word-final position, in the history of  Greek, instead of  talking about a persistent final-n-loss change that stretched over a thousand or more years. Eleventh, once the issue of attempting to quantify rate of change is considered, a related question arises. In particular, except perhaps in cases of intense language contact of  the sort that can lead to language mixing, one has to wonder if it is really possible to talk (as some sociolinguists and historical linguists do) about ‘massive’ or ‘far-reaching’ changes. It would seem that every language at every stage of its development is an amalgam of elements (sounds, morphemes, words, constructions, etc.) that are carried over from previous stages and elements that have been altered in some innovative way. Thus ‘massiveness’ of change often simply means that the analyst has identified a set of elements that show change, without a concern for the concomitantly conservative elements that show no change. This view is especially compelling when one considers that linguists at any one time typically focus on at best a handful of elements that do change when in fact a language contains thousands and thousands of  ‘points’ (sounds, morphemes, words, constructions, meanings, etc.) that could in principle change. Thus, ‘massiveness’ may only be in the eye of the linguist attending to a small subset of changing points in the language overall.

Historical Linguistics and Sociolinguistics

83

Taking this one step further, if this amalgam view is right, then it would seem dif ficult to talk about languages and dialects being ‘conservative’ or ‘innovative’ (in general). It would seem that such an assessment makes sense only as relative to particular features. A useful example to always remember in this regard is the situation with Lithuanian. Lithuanian is often said to be highly ‘conservative’ within Indo-European, but on closer inspection, that is only true with regard to there being a large number of cases within the nominal system (eight, as reconstructed for Proto-Indo-European, and seven in modern Lithuanian, and then only if one ignores the gain of  three cases (illative, allative, and adessive) in Old Lithuanian10 and their subsequent loss by the modern era; that is, conservatism is there but only for the noun system and only for (most of ) the eight original cases.

5 Conclusions Even if  historical linguistics is focused more on change and less on the spread of change, every practicing historical linguist recognizes that understanding spread is one dimension to a full understanding of  the forces of diachrony. Thus overall, historical linguistics and sociolinguistics form a natural class rather than an uneasy forced alliance. Still, it is possible to express some wishes for historical linguists and for sociolinguists in terms of practices and research questions. First, historical linguists could be more interested in stability, even if, as stated above, all we can really talk about is relative stability and not absolute or inherent stability. As noted above in section 3, there is recent literature on this topic, but still for most practicing historical linguists, change is where the action is, so to speak, and not stability. Relatedly, a key question to be addressed is whether change is the default state for language diachronically or whether instead (relative) 10

These cases were secondary formations that came about from the fusing of existing ‘primary’ case forms (genitive, accusative, locative) with postpositions.

84

Brian D. Joseph

stability is the default state. And sociolinguists could be less concerned about ‘change in progress’ since it is often hard to be sure that there really is a change going on; with any variation claimed to betoken change, the question of directionality is an issue (which form is innovative and which is conservative) and so too, as a result, is the question of whether the innovative variant will ultimately take hold or not, as it is not always the case that it does. Admittedly, investigations of change in progress will always be important and revealing and it is not entirely clear what should take the place of interest in the phenomenon; ‘identity sociolinguistics’ is not enough, as there are real issues in determining which states of af fairs constitute generational change and which do not that identity alone does not answer. As for social history and social historians, to return to the theme sounded at the outset, perhaps there is a dif ferent type of uniformitarianism to consider and always keep in mind, another dimension to the adage that ‘plus ça change, plus c’est la même chose’. In particular, the raw material that many historical linguists have to work from are written or orally transmitted texts and thus typically literature or historical accounts from days gone by. Importantly, they generally show humans from thousands of years ago to be much like humans of  today in terms of  the things that they are interested in and worry about. That is, there is a uniformitarianism with regard to the hopes and aspirations of individuals in the distant past and those we observe and feel today. Similarly, day-to-day concerns about one’s welfare, about being nourished and loved, and so on, seem to be as important to individuals in the past as they are to us today. We can refer to this uniformitarianism as a ‘social or humanistic uniformitarianism’, and, drawing on Joseph (1999), we can note a couple of examples that ref lect this state of af fairs, and show stability, and thus uniformity, in what individuals’ lives and interests and concern were. A first example comes from Wendy O’Flaherty, in her 1981 translation of the Rigveda (the oldest Sanskrit text, composed c.1200 BC but with parts much older). In her preface, she comments on the content of  the text as follows, noting that the hymns of  the Rigveda show ‘conf lict within the nuclear family and uneasiness about the mystery of  birth from male and female parents; the preciousness of animals […]; the wish for knowledge,

Historical Linguistics and Sociolinguistics

85

inspiration, long life, and immortality’. In many ways, these seem to be universal human concerns, and such a characterization of content could be applied to what one sees today in the tabloid press (such as the National Enquirer in the US) or in titles on bestseller lists. A second example is Melchert (1991), writing on the last minutes of  life for Hittite king Hattusili (from the second millennium BC). Here are the relevant facts as Melchert lays them out: Hattusili was apparently dictating his last will and testament to a scribe when he suf fered an ultimately fatal or near-fatal episode as he finished the of ficial dictation at the end. He then began ref lecting somewhat incoherently about his impending death, producing ravings which were dutifully copied down and recorded for posterity by the scribe.11 Hattusili ends with an exhortation to a woman he has been calling for: ‘Protect me on your bosom from the earth’, apparently his real last words. Melchert interprets these last words as follows: it is known that the Hittites practiced burial (not cremation) but believed in an afterlife and immortality in divine form for its kings; thus, he writes, ‘Despite […] assurances of happy immortality, however, the dying Hattusili is frightened. He sees only the immediate certainty that he will soon be put down into the cold, dark earth alone, and like many a poor mortal since he finds this a terrifying prospect’. Further, by way of linking modern-day folks with those that preceded them 3,500 years ago, Melchert says, with real eloquence: ‘there seems to be little fundamental dif ference between us and ancient peoples when it comes to facing death. Hattusili’s words speak to us directly across the centuries. His fear is palpable. We not only at once understand but also are moved by his agony and his desperate cry for his loved one’s tender comfort. These emotions are neither Hittite nor Indo-European, neither ancient nor modern, but simply human’. And, finally, to bring this social/humanistic uniformitarianism back to the issue of  language change per se, let me mention these notable lines from Chaucer’s Troilus and Criseyde (II.22–28):

11

This is presumably the closest we will ever come to having a tape recording of spontaneous speech from the second millennium BC!

86

Brian D. Joseph Ye knowe ek that in forme of speche is chaunge Withinne a thousand yeer, and wordes tho That hadden pris, now wonder nyce and straunge Us thinketh hem, and yet thei spake hem so, And spedde as wel in love as men now do; Ek for to wynnen love in sondry ages, In sondry londes, sondry ben usages. [You know too that in form of speech there is change within a thousand years, and words as well that had value now wondrously odd and strange they seem to us, and yet they spoke them in this way, and it served as well in love as men do now; also to win love in various ages, in various lands, various were the usages.]

Thus, not only is change itself something that is pervasive, but so too is the recognition of change and an interest in change.

References Andersen, Henning. 1989. ‘Understanding linguistic innovations.’ In: Leiv Egil Breivik and Ernst Håkon Jahr (eds), Language change: Contributions to the study of its causes. Berlin: de Gruyter, 5–27. Anttila, Raimo. 1977. Analogy. The Hague: Mouton de Gruyter. Bloomfield, Leonard. 1933. Language. New York: Holt, Rinehart, Winston. Durian, David. 2009. ‘Purely a chain shift?: An exploration of  the “Canadian Shift” in the US Midland’. Paper presented at NWAV 38, Ottawa, Canada. Durian, David. 2012. A new perspective on vowel variation throughout the 20th Century in Columbus, OH. PhD dissertation. Ohio State University. Ernout, A., & Antoine Meillet. 1939. Dictionnaire étymologique de la langue latine: histoire des mots. Paris: Klincksieck. Hamp, Eric P. 1998. ‘Some draft principles for classification’. In: Brian D. Joseph and Joseph C. Salmons (eds), Nostratic. Sifting the evidence. Amsterdam: John Benjamins, 13–15.

Historical Linguistics and Sociolinguistics

87

Harrington, J., S. Palethorpe & C. Watson. 2000a, 21 December. Does the Queen speak the Queen’s English? Nature 408, 927–928. Harrington, J., S. Palethorpe & C. Watson. 2000b. ‘Monophthongal vowel changes in received pronunciations: An acoustic analysis of the Queen’s Christmas broadcasts’. Journal of  the International Phonetic Association 30, 63–78. Harrington, Jonathan. 2006. ‘An acoustic analysis of “happy-tensing” in the Queen’s Christmas broadcasts’. Journal of  Phonetics 34.4, 439–457. Haugen, Einar. 1958. ‘Language contact’. Proceedings of  the International Congress of  Linguists 8, 771–785. Hock, Hans Henrich. 2003. ‘Analogical change’. In: Joseph & Janda (eds), 441–460. Hockett, Charles. 1960. ‘The Origin of  Speech’. Scientific American 203, 89–97. Janda, Richard D., and Brian D. Joseph. 2003. ‘On language, change, and language change – or, of history, linguistics, and historical linguistics’. In: Joseph & Janda (eds), 2–180. Joseph, Brian D., & Richard D. Janda (eds). 2003. Handbook of  historical linguistics. Oxford: Blackwell. Joseph, Brian D. 1999. ‘Linguistics for “Everystudent”’. Studies in the Language Sciences 28.2 (Fall 1998), 123–133. Joseph, Brian D. 2011. ‘A critique of grammaticalization’. In: Bernd Heine and Heiko Narrog (eds), Handbook of grammaticalization. Oxford: Oxford University Press. Kiparsky, Paul. 1971. ‘Historical Linguistics’. In: William O. Dingwall (ed.), A survey of linguistic science. College Park: University of Maryland Linguistics Program, 576–642. Labov, William. 1994. Principles of  linguistic change: Internal factors. Oxford: Blackwell. Labov, William. 2007. Transmission and dif fusion. Language 83, 344–387. Melchert, H. Craig. 1991. ‘Death and the Hittite king’. In: Roger Pearson (ed.), Perspectives on Indo-European Language, Culture and Religion. Studies in Honor of  Edgar C. Polomé, Vol. I.182–188 (Journal of Indo-European Studies, Monograph Number 7). McLean, VA: Institute for the Study of  Man, 182–188. Nichols, Johanna. 2003. ‘Diversity and stability in language’. In: Joseph & Janda (eds), 283–310. O’Flaherty, Wendy. 1981. The Rig Veda. An Anthology. New York: Penguin Books. Paredes, Hugo, and Mário Martins, F. 2007. Social Theatres: A Web-Based Regulated Social Interaction Environment (Lecture Notes in Computer Science). Berlin: Springer. Read, Allen Walker. 1963a. The First Stage in the History of  ‘O.K.’. American Speech 38.1, 5–27.

88

Brian D. Joseph

Read, Allen W. 1963b. The Second Stage in the History of  ‘O.K.’. American Speech 38.2, 83–102. Rickford, John R., Thomas A. Wasow, Norma Mendoza-Denton & Juli Espinoza. 1995. Syntactic variation and change in progress: Loss of  the verbal coda in topic-restricting ‘as far as’ constructions. Language 71.1, 102–131. Teodorsson, Sven-Tage. 1979. Phonological variation in Classical Attic and the development of  Koine. Glotta 57, 61–75. Thomason, Sarah G. 1997. On mechanisms of interference. In: Stig Eliasson and Ernst Håkon Jahr (eds), Language and its ecology: Essays in memory of Einar Haugen. Berlin: de Gruyter, 181–207. Vennemann, Theo. 1972. Phonetic analogy and conceptual analogy. In: Theo Vennemann and Terence Wilbur (eds), Schuchardt, the Neogrammarians, and the Transformational Theory of  Phonological Change: Four Essays. Frankfurt: Athenäum, 181–204. Yonkers, V. 2008. New type of communication with facebook? Connecting 2 the World 13 September 2008. accessed 12 August 2011.

Nicola McLelland

From Humanist History to Linguistic Theory: The Case of  the Germanic Rootword1

Abstract This paper traces how the notion of the monosyllabic rootword – a concept that had its origins in sixteenth-century German humanist historiography – became central to German ref lection on the nature and structure of  language. Indexical of  German antiquity and originality in the sixteenth and seventeenth centuries, it came to play a part in eighteenth-century debates on language origin, underlay lexicography into the mid-nineteenth century, and also inf luenced comparative linguists examining Semitic and Sanskrit well into the nineteenth century. In J.G. Schottelius’s (1612–1676) hands, it yielded the basis of present-day word-formation theory, and dif fering beliefs about the integrity of  the rootword led to dif ferences in spelling between German and Dutch that are still apparent today.

1 Historiography and German linguistics Recently historians’ methods have inf luenced sociolinguists and historical linguists, who have followed their historian colleagues to the archives (Elspaß 2005; Elspaß et al., 2007). But it is not just the methods of  history

1

Since this paper was first submitted, Sections 3 and 4 have been integrated in slightly re-arranged and revised form in a longer article on Justus Georgius Schottelius: McLelland (2010).

90

Nicola McLelland

that can inf luence linguistic study. There is no shortage of examples of how historiographers’ readings or (re-)constructions of  the past have shaped linguistic ref lection in Germany. One might note the anxious focus, for a while at least, on possible dif ferences between the language of the GDR and FRG, ref lecting anxieties about the recent history of  the Germany states (Hellmann and Schröder, 2008), the Marxist view of  history that encouraged a particular type of sociolinguistics in the GDR (Stevenson, 2003), or the grand ‘national monument’ of the Deutsches Wörterbuch begun by the Grimm brothers in 1852 (and completed in 1960 (Haß-Zumkehr, 2000)).2 The most notorious examples of how politically motivated readings of history could skew German linguistic study are those of the National Socialist era (Hutton, 1999), tainting figures like Leo Weisgerber (cf. Erlinger 2007 and references there), Jost Trier, and Heinz Kloss (1952) to a greater or lesser degree. Hüllen (2002: 302–332) gives further examples of how the National Socialist reading of the past shaped the work of lesser-known linguistic figures in ways that now seem distasteful. In this article, I consider a case rather more remote in time: sixteenth-century humanist historians’ re-readings of  Tacitus and their idealization of a Germanic past. I shall show how the key notion of the rootword in German first emerged from such humanist readings of ancient Germanic history, became imbued with cultural symbolism, and was ultimately incorporated into mainstream linguistic ref lection.

2 Humanist history and heritage-building The slice of  German historiography that I want to examine here is the German humanist project of creating a shared German heritage in which to take pride. Central to the project was the rediscovery and exploitation of  Tacitus’s account of  Germania by humanist historians of  the sixteenth 2

The overlaps and mutual fertilization between language history and cultural history have been a focus of interest for some time in German studies, as the volume of papers published a decade ago suggests: Sprachgeschichte als Kulturgeschichte, Gardt et al. (ed.) (1999).

From Humanist History to Linguistic Theory

91

century (Mertens 2004). There has been much interest in German studies in how readings of recreations of the notion of Germania have infused German studies over the centuries (cf. Krebs, 2010; Lee & McLelland, 2011). Mertens (2004: 38) commented that historians and philologists need to work together in examining this process of reinventing the past. This paper is a contribution to that discussion. Tacitus’s Germania itself is a short text – about fifteen pages in translation – and as an ethnographic account of  the Germanic tribes, it is highly ambiguous. Passages praising the valour and sexual restraint of the people sit cheek by jowl with accounts of the poor buildings and the lack of agriculture; the potentially touching image of all babes suckled by their own mothers is undercut by the account of  the squalor of  family life, where those same children grow up ‘naked and nasty’ (nudi ac sordidi, Tac. Germ. 20, transl. Thomas Gordon). Yet Tacitus’ Germania formed the basis for the humanist creation of a proud German past. As a measure of its inf luence, Mertens (2004: 61) calculates that there were twenty editions of it in the period 1472–1519, meaning that about 6,000 copies were in circulation over a time span of  fifty years. The first inf luential reading of  Tacitus’s Germania was that of Enea Silvio Piccolomini (1405–1464), whose pamphlet Germania (1457–1458) read Tacitus in such a way as to portray the ancient German tribes as nomadic and primitive, barely distinguishable from brutes, so establishing a discontinuity between that era and the more civilized present. In contrast, Giannantonio Campano’s (1429–1477) reading of  the same text – in a speech intended to motivate the Regensburg Reichstag of 1471 to take up arms against the Turks (though the speech was never delivered, it was published in the 1480s after Campano’s death; Mertens 2004: 72–75) – emphasized both the valour of the Germans and their unity as a people. Even though the entire second half of Tacitus’ Germania deals with the separate tribes, Campano focused on one statement in Germania 1.2 that the Germans were an unmixed people native to the territories where they lived; hence – Campano reasoned – they must never yet have been conquered by invaders. Equally positive was Campano’s interpretation of  the name Germani, which Conrad Celtis (1459–1508) would follow: the Latin word germani (‘brothers’) expressed the sense of  brotherhood amongst the Germanic peoples themselves. From a collection of tribes that had come together, Germania was now viewed by Campano as a ‘natürliche

92

Nicola McLelland

Einheit’ (Mertens 2004: 76–77), a people with a common origin, who only later split up into dif ferent tribes. Conrad Celtis (1459–1508) added to the claims of Germanic valour and indigenousness their impressive geographic expansion. He also praised the uncorrupted simplicity of  the Germanic peoples, by contrast with the customs of his own time. So where Enea had seen the Germanic past as something primitive but happily overcome, Celtis found in the same Tacitus text a Germanic past that was primitive in the positive sense of unspoilt (Mertens 2004: 82).3 Heinrich Bebel (1472–1518) followed suit, emphasizing the original purity of an unmixed, valorous, undefeated people, and these topoi were soon very firmly established. As evidence of  how strong and long-lasting was the tendency to attribute such virtues to the Germanic ancestors, consider Justus-Georgius Schottelius’s (1612–1676) annotation of a German proverb in the list he compiled as part of  his Ausführliche Arbeit ‘Comprehensive Work’ on the German language (1663: 1129). Schottelius was a cultural patriot who clearly shared the agenda of his humanist predecessors to portray the Germanic past positively, and like them, he saw proverbs as an expression of  German culture and character. So in the proverb Penning ist Pfennings Bruder (already found as no. 71 in the 1529 proverb collection of  Johannes Agricola of  Eisleben (1494–1566); cf. Mieder 1982), Schottelius found evidence of  the sense of equality and fairness amongst the ancient Germans: Hieraus erweiset sichs / wie unsere alten Teutschen Collation / Wolleben und Freude gehalten haben / nemlich / daß ein jeder sein Essen hat mitbracht / und zum Getränk haben sie einen Pfennig neben den andern gelegt zu gleicher Zechen / und ist einer nicht höher beschweret worden dan der ander. (Schottelius 1663: 1129)4 3

4

One particular problem for humanist commentators seeking a more positive view of ancient Germania was Tacitus’ reference to the Germans’ human sacrifices to their deities. Conrad Celtis, in his Germania edition of 1500, suggested an emendation to the text: one should read huius for humanis, hence ‘his [of  the God Mercury] sacrifices’, rather than ‘human sacrifices’. Enea’s edition gives the of fending passage in the margin, apparently because of a printer’s error, but many editions following it omitted the problematic passage altogether (Mertens 2004: 74). For fuller discussion of  Schottelius’s treatment of proverbs, see McLelland (2011a, chapter 5).

From Humanist History to Linguistic Theory

93

[This shows how our ancient Germans enjoyed eating together, a good life, and joy, namely that each brought his food, and for drinking they each laid a penny next to another in a row, and not one was burdened [financially] more than another.]

Seventeenth-century art history provides evidence of continued ‘bi-polar’ readings of Tacitus in the Netherlands too. Dutch historiographers initially condemned the barbarous Germanic tribes, but, motivated in part by the Dutch Republic’s need for a founding myth after the Twelve Years’ Truce of 1609,5 they later homed in on certain positive characteristics (Weststeijn 2008: 68, n.86, citing Van der Waal 1952: 172–210).6 Conveniently, Tacitus had singled out the Batavians as the bravest tribe of the Rhineland, and he had also praised their noble simplicity (Weststeijn 2008; 50, 58; Cf. Germania 29). Now, in Germany the 26 engravings illustrating the German Philipp Cluverius’s 3-volume Germania antiqua (1616) were the first systematic illustrations of ancient Germania. Cluverius cited and paraphrased Tacitus on nearly every page, and the engravings illustrated the same points: a warriorlike simplicity (cf. Krebs 2010), including for instance the details already mentioned of a mother nursing her baby (fig. 12; cf. Tac. Germ. 20.1: sua nemque mater uberibus alit …), children playing naked on the ground (fig. p. 158; cf. Tac. Germ. 20.1: In omni domo nudi ac sordidi in hos artus, in haec corpora quae miramur, excrescunt), and animal skins as clothing. In contrast, illustrations by Antonio Tempesta for Batavorum cum romanis bellum (by Otto Vaenius, Antwerp 1612; cf. Westteijn 2008: 50) portrayed the Germanic ancestors as civilized, dressed in classical-style clothing. The Anglo-Dutch antiquarian Richard Verstegan’s (c. 1550–1640) portrayal of  the ancient peoples of Germania had also already been more fanciful. Verstegan (1605) praised the simplicity, masculinity and virtue of the ancient tribes, but also contrived to portray their pagan worship as virtuous, writing that they ‘lived according to the lawe of nature and reason, wanting nothing 5 6

The truce with the Hapsburg empire achieved nominal independence for the Dutch republic in 1609. Dutch historians apparently rehearsed a commonplace of Roman historiography for describing and appropriating foreign civilizations: the simplicity of  barbaric tribes was deemed a positive value which had been lost by the ancients themselves (Van der Waal 1952: 58).

94

Nicola McLelland

but the knowledge of the true God’ (Weststeijn 2008: 51, citing Verstegan 1605: 67). One of  Verstegan’s illustrations showed an idol of  their god Tuisco (Figure 1, from Weststeijn 2008: 53). While Verstegan’s illustration potrays the statue of Tuisco simply clothed in some sort of cloak (cf. Tacitus, Germania 17) and sporting long hair and a beard (Tacitus, Germania 31), the overall impression is one of classical elegance. Significantly, a band of people is seen approaching in the background from the Tower of Babel. Here, the imagery expresses a tour de force already achieved by Annius de Viterbo (Giovanni Nanni, c. 1432–1502) in his Antiquitatum Variarum of 1498 (often known as Antiquities, a forgery of a ‘lost’ history by Berosus, so often called the Pseudo-Berosus), which contrived to combine the Tacitean mention of a Germanic god called Tuisco with a universal world history that chimed with Biblical accounts of the confusion of tongues. According to this account, Tuisco, an additional son of  Noah and the father of  the Germanic race, left Babel with his followers to settle the lands now called Germany (cf. Jones 1999, chapter 1). One could pursue the shifting readings of Germania in the sixteenth and seventeenth centuries in far greater detail, but for my purposes the essential points have been made. Ancient Germania was – after some earlier, more negative readings – portrayed positively as pure, uncorrupted, strong, and original, with a shared and single point of origin at Babel.

3 The rootword as guarantor of  Germanic purity, antiquity and originality It is in the context of positive ‘heritage-building’ (Considine 2007) for Germany and her Germanic neighbours that a key linguistic notion about the German language emerged, the rootword. Tacitus’s Germania had said almost nothing about the language of  the Germanic peoples (though he cited two words, for ‘spear’ and ‘amber’), but Francisus Irenicus

From Humanist History to Linguistic Theory

95

(Franz Friedlieb, 1494 or 1495–1553), in his Germaniae exegesis, described by Mertens (2004: 78) as ‘a kind of  handbook about “Germany” that combines ethnography, history, geography and natural history’ (transl. NMcL), wrote of  the German language that omnis imperatiuus, ac omnia pene germanica uocabula monosyllaba sunt ut brot, susz, disch, der, dem, lisz, morn etc. et pene quicquid monosyllabam excedit, pere­ grinum ac non germanicum est … (Irenicus 1518: lib II cap 31 fo 38v, cited Jellinek, 1898 (1985 rpt.): 60) [every imperative and almost every German word are monosyllabic, such as brot, susz, disch, der, dem, lisz, morn, etc. and almost everything that is longer than a monosyllable, is foreign and not German. (Translation NMcL)]

In this short statement, three crucial features that characterize later discourse about rootwords are already present: 1 2 3

They are monosyllabic; They are identical with the (second person singular) imperative form of  the verb; They are characteristically German.

This created heritage of  the Germanic rootword had a long shadow in later linguistic ref lection, as I shall show below by taking each of the three aspects in turn. 3.1 Monosyllabic rootwords Jellinek (1898: 61) assumed, plausibly, that Irenicus’ belief in monosyllabic rootwords was the result of extrapolating from the southern German varieties of  German, characterized by apocope, i.e. sag, not sage. What is important is that this feature of  the rootword was not, at this early stage, indexical of any other qualities of  German, or indeed of  the Germans. Beatus Rhenanus (1485–1547), at the head of the last section of Book II of  his 1531 Rerum Germanicarium Libri Tres, a section headed Germanornum lingua ‘the language of  the Germans’, repeated Irenicus’ observation: that

96

Nicola McLelland

the German language gaudet vocibus primigeniis monosyllabis ac nonnulla a Prouincia lingua mutuatus est ‘rejoices in originally monosyllabic words and took some from the languages of the provinces’ (cited from the edition by Mundt 2008: 268–69, English translation NMcL).7 The Austrian Johannes Cuspinianus (1473–1529) made the same point, and Georg Henisch (1549–1618) cited both these authorities in the preface to his incomplete German dictionary (Henisch, 1616 [1973], Rössing-Hager, 1984–1985). If it had its origins in a view of the language that was coloured by the local experience of a southern German variety, the notion of the rootword evidently suited north Germanic scholars too, whose languages also exhibited apocoped forms. It soon became a commonplace in linguistic thought in the Low Countries, and in Sweden too, where Georg Stiernhielm and the grammarian Tiällmann (1696) both cited Becanus’s Hermathena (Bk 2, p. 25 of Omnia Antverpiae 1580) (Haapamäki 2002: 55). For it was with the Flemish cultural patriot Goropius Becanus (1518–1572) that the rootword became more than a neutral feature of  the German language, and became indexical of the same Germanic qualities that I identified as typical of cultural patriotic discourse at the end of  Section 2: pure, uncorrupted, strong, and original, with a shared and single point of origin at Babel. Becanus used the notion of the rootword as evidence of the originality, and therefore antiquity and purity, of  the ancestral Germanic language. He believed that the most ancient language on earth would be the simplest language, and that the simplest language would contain the most monosyllabic rootwords. Since the number of short words was (he asserted) higher

7

Beatus Rhenanus marks an important step in the reception of  Tacitean notions of  Germania. He published both the Germania itself, and a commentary (1519), as well as a later work, Rerum Germanicarium Libri Tres, to which I shall return. He rejected Annius de Viterbo’s Pseudo-Berosus as fiction, and at the same time insisted that Tacitus’s Germania should be read as a document of its time. He ‘thus made a substantial step forward in the development of a historical approach to the sources themselves, a step from the philosophical method of textual reconstruction to historical criticism and to the methodical explication of developments’ (Mertens 2004: 93).

From Humanist History to Linguistic Theory

97

in Flemish than in Latin, Greek or Hebrew, Becanus reasoned that his was the oldest language, from which all others were derived. He attempted to demonstrate this by showing that words in other languages were originally composed of Flemish rootwords. For instance, Becanus (1569) argued that the name of the Germanic tribe of Saxons has at its root sagun, a compound made up of the two roots Sac + Gun, where Sac = ‘Sack’, but meaning here ‘causa’ and Gun should be read as in German Gunst, yielding the meaning ‘favoured by the first cause, i.e. favoured by God’ for the name of  the Saxons (Metcalf 1974: 273–274). The German grammarian Schottelius went a step further in using monosyllabicity as an argument for the originality of  German. Agreeing with Becanus’s view that simple things have simple words, so the oldest names are short and simple, he adduced a further powerful argument, that the phylogenetic origin of language would have been similar to the ontogenetic process of language acquisition in children. Just as children first learn to utter monosyllables, so the original words of  language were short and simple (Schottelius, 1663 [1967]: 61, oration 4, ¶32). Hence German, rich in original ancient rootwords, had evidently remained closer and truer to the origin of  language than other languages. Schottelius’s very inf luential grammar of  German (Schottelius 1641, 1651, 1663) with its first book of  ‘orations’ on the German language acted like a kind of funnel through which the notion of the rootword was transmitted to later generations of language scholars. Long after Johann Daniel Longolius (1677–1740) had rejected the notion of monosyllabicity as a criterion for Primitiva (Longolius 1715: 63; cf. Kaltz 2005: 126), Enlightenment debates about the origin of language – now stripped of any national chauvinism – still remained indebted to the notion of  the monosyllabic word as a feature of the first languages. Consider this passage from Herder’s famous prize-winning treatise on the origin of  language (Herder 1772; cf. Kief fer, 1978, Neis, 2003). Herder’s broad argument here is that human language is not perfectly logical, and so is not of divine origin. Einem Verteidiger des übernatürlichen Ursprunges [= Johann Peter Süßmilch (1707– 1767), 1766)] ist’s göttliche Ordnung der Sprache, ‘daß die meisten Stammwörter einsilbig, die Verba meistens zweisilbig sind und also die Sprache nach dem Maße des Gedächtnisses eingeteilt sei’.

98

Nicola McLelland [To a defender of  the supernatural origin [of  language] [= Johann Peter Süßmilch (1707–1767), 1766] it is the divine arrangement of  language, ‘that most rootwords are monosyllabic, the verbs mostly bisyllabic, and that the language is thus divided up according to the measure of  the mind.’ (Translation NMcL)]

Süßmilch had presumably considered verbs to be more complex than monosyllabic (nominal) roots, and so logically subordinate to them in thought. Language apparently mirrored thought so perfectly that it must be of divine origin. Herder disagreed, though not disputing the assumption that monosyllabicity entailed originality. Rather, he argued that since surviving verbs in the oldest languages are bisyllabic, they are evidence that the first, oldest, monosyllabic word have been lost, just as the babbling of children is no longer present in the speech of adults: In den Resten der für die älteste angenommenen Sprache sind die Wurzeln alle zweisilbige Verba; welches ich nun aus dem vorigen sehr gut erklären kann […]. Diese Verba nämlich sind unmittelbar auf die Laute und Interjektionen der tönenden Natur gebauet, die oft noch in ihnen tönen, hie und da auch noch als Interjektionen aufbehalten sind; meistens aber mußten sie, als halbinartikulierte Töne, verlorengehen, da sich die Sprache formte. In den morgenländischen Sprachen fehlen also diese ersten Versuche der stammelnden Zunge; aber daß sie fehlen und nur ihre regelmäßigen Reste in den Verbis tönen, das eben zeigt von der Ursprünglichkeit und – Menschlichkeit der Sprache. Sind diese Stämme Schätze und Abstraktionen aus dem Verstande Gottes oder die ersten Laute des horchenden Ohrs? die ersten Schälle der stammelnden Zunge? Das Menschengeschlecht in seiner Kindheit hat sich ja eben die Sprache geformet, die ein Unmündiger stammlet: es ist das lallende Wörterbuch der Ammenstube – wo bleibt das im Munde der Erwachsnen? [In the remains of  the languages assumed to be oldest, the roots are all bisyllabic verbs; which I can now very well explain […]. For these verbs are built directly on the sounds and interjections of resounding nature, which often still resound in them, and here and there are still preserved as interjections; but mostly they were inevitably lost, as half-inarticulate sounds, when language was formed. In the eastern languages these first attempts of  the stammering tongue are lacking; but the very fact that they are lacking and only their relics resound in the verbs, is exactly what shows the originality and – humanity of  language. Are these roots treasures and abstractions from the mind of  God, or the first sounds of  the hearkening ear? the first sounds of  the stammering tongue? Humankind formed in its infancy the language that a child stammers: it is the babbling dictionary of  the nursery – where does that remain in the mouth of an adult?]

From Humanist History to Linguistic Theory

99

The details of the argument are not my concern here. What matters is that the equation of monosyllabicity with early, primitive language is accepted on both sides of  the origin of  language debate. 3.2 The verbal root and the imperative of  the verb The second feature of  the rootword identified by Irenicus was that it is identical with the second person imperative. Jellinek (1898) summarized the discussion for and against this view. Noted by Henisch (1616), it was championed most vigorously by the German grammarian Schottelius and by the popularizer of  his ideas, Harsdörf fer (1607–1658). Yet for many Germans, the second person imperative wasn’t monosyllabic at all (an uncertainty – mach or mache! – that still persists; cf. McLelland, 2011b). The heated debate about and ultimate fate of such paragogic or apocoped -e’s is a familiar instance of the interactions between social history and language history, as the non-apocoped forms in the language became identified with northern Protestantism in the eighteenth century (Habermann 1997). At any rate, Jellinek noted that despite much opposition from within the Fruitbearing Society (Fruchtbringende Gesellschaft), the Schottelian view dominated. Beyond Germany too, the Danish grammarian Peder Syv (Syv 1979 [1663]: 132) took the rootword to be the imperative, and so did the Swedish grammarian Tiällmann (1696: 212–213), who listed irregular verbs by this root (cf. McLelland, 2011a, Chapter 7; McLelland 2010). Even once it ceased to play a strong role in German grammar,8 the dogma of  the rootword imperative was particularly problematic to later language scholars because it was at odds with what was known about the triliteral and hence bisyllabic verbal roots of  the Semitic languages. As Jellinek (1913: 138) had already implied, it has continued to be assumed (Barbarić 1981: 1195; Takada 1998: 25, n. 98), that Schottelius took the idea of the rootword from Semitic, especially Hebrew grammar, but there

8

There remained, however, a tendency to use the notion of the rootword more loosely as a pedagogical tool, as in the case of Martin Aedler’s (1680) German grammar for English learners (cf. Van der Lubbe 2007: 186).

100

Nicola McLelland

is no evidence that he did more that follow his humanist sources. In fact, later European grammarians of  Hebrew and the Semitic languages struggled mightily to reconcile the monosyllabic verbal root – which it can be shown that they had inherited in a direct line of descent from Schottelius (cf. McLelland, 2011b, chapter 7) – with what they saw before them in the Semitic languages. Wilhelm von Humboldt concluded that the typically bisyllabic roots of the Semitic languages must have emerged from an earlier monosyllabic stage; only in a second stage of  linguistic development did the recognition of a need for grammatical specification lead to the addition of a second syllable for the purposes of inf lection (cf. Rousseau 1984: 300, 320). Franz Bopp (1820) extended the debate about monosyllabic verbal roots to Sanskrit in comparison with the Semitic languages, when he sought to demonstrate that Sanskrit roots were monosyllabic; that they ‘may contain as few letters as are requisite to constitute a monosyllable’ (Bopp 1820: 8, cited Rousseau 1984: 309, 311). Ultimately, then, a humanist observation about a southern variety of  German had a significant impact on the European study of  Semitic and even Sanksrit. 3.3 Rootwords as typically Germanic building blocks for compounding In his obsession with identifying and interpreting the ‘true’ roots of  Germanic proper names like Saxon, Becanus followed the example of  Beatus Rhenanus, who had included much discussion of  German proper names in the commentary to his edition of Germania in 1519 (Mertens 2004: 92). The Protestant humanist pedagogue Philipp Melanchthon followed suit in his editions of  the Germania for school use (1538, 1557). For Melanchthon, the value in etymological study of German proper names lay on the one hand in the reminder of  the inconstancy of all human concerns, and on the other in engendering a love for one’s fatherland (patriae amor, cf. Mertens 2004: 94). So identifying rootwords in German names became the task of any cultural patriot. The Dutchman Simon Stevin (1548–1620) went a step further. Better known in the history of mathematics as someone who popularized the use of decimals, Stevin is important in the history of  linguistic ref lection

From Humanist History to Linguistic Theory

101

for his Vytspraeck vande Weerdicheyt der duytsche tael, his ‘oration on the worth of  the duytsch language’,9 which occupied 24 sides at the front of  his work De Weeghdaet (1586), ‘On the art of weighing’. This oration at the front of  the book served the purpose of justifying the author’s use of  his Dutch vernacular in the work that followed. In his oration, Stevin argued that simple things merit simple (i.e. monosyllabic) words, an idea that he seems to have taken from Becanus, or at the very least shared with him. Crucially for Stevin, such monosyllabic rootwords could then be combined to form new, more complex terms, a facility especially useful for coining technical terms. Stevin demonstrated in a series of tables (1586: 67–79) that Duytsch had 742 eensilbighe woorden inden eersten persoon; daerder de Latinen alleenlick 5 hebben; de Griecken gheen eyghentlicke, maer langhe vercort tot 45 (‘742 monosyllabic words in the first person, while the Latins have only 5 and the Greeks have no monosyllables, but only 45 long words that have been contracted’). The demonstration was developed further in a reworking of this treatise that appeared in 1608 (Stevin 1608: 24–38), where Stevin listed the monosyllabic verb roots as in the 1,586 Uytspraeck, but added 1,428 monosyllabic nouns, adjectives, etc. for Duytsch, 158 for Latin, 220 for Greek, giving totals of 2,170 for Duytsch, 163 for Latin, and 265 for Greek. Since Duytsch had by far the greatest number of these rootwords, Stevin argued that Duytsch was a richer and more capable language than either of  the classical languages, and was particularly well-suited to coining new terms in mathematics (of which wiskonde ‘mathematics’ is a prime example; today Dutch is the only European language that does not use some variant of mathematica, from the Greek, via Latin).

9

It is dif ficult here to translate Duytsch. It means for Stevin something that encompasses Dutch and German at least. Evidence of its wider sense is the anecdote Stevin relates of the German humanist Henricus Glareanus switching from Latin to German during an oration on Suetonius because he found the German words to have the greater emotional appeal (Stevin 1586, in Dijksterhuis 1955: 86–87; Schottelius paraphrases Stevin’s account in the AA 64–65, 4: 42). Here, Stevin uses for German the same word, Duytsch, that he uses for his own vernacular.

102

Nicola McLelland

For Stevin, then, rootwords took on the additionally important feature of indicating both the richness and conciseness (brevitas) of  German, in which it was superior to Greek, Latin and Hebrew. Its brevity was evident in the abundance of monosyllabic words; its richness in their almost infinite capacity for compounding. Want dit moet ghy weten, dat de sprakens goetheyt niet alleen voorderlick en is om de Consten bequaemlick daer duer te leeren, maer oock den Vinders in haer soucking […] wy segghen van Stof f waerheyt, […] en dierghelijcke, daer t’volghende vol af is; welcke woorden de Griecken soo cort, ende by haren yderman soo verstaenlic, oock so eyghentlick haer grondt beteeckenende, noyt en hebben connen segghen, nu niet en connen, noch, dat kennelick ghenouch is, inder eewicheyt niet connen en sullen. want datter niet in en is en cander niet uytghetrocken worden. [For you must know that the excellence of  language is conducive not only to learning the arts well through it, but also to the search of  the inventors […] we talk of  Stof fwaerheyt [‘specific gravity’], […] and the like, in which the following [text] abounds; which words the Greeks never were, are not now, and never to all eternity will be able to say so shortly and so universally intelligibly to everyone of them, and also describing its nature so aptly, as is suf ficiently obvious. For what is not in it [sc. in a language], cannot be extracted from it’. (Stevin 1586: 5–6, in Dijksterhuis 1955: 384–387)]

Similarly, in a passage lifted wholesale by Schottelius (1663: 14, 1: 42), Stevin took pride in the unique ability of  Duytsch to form compounds: want waer wildy spraken halen daermen duer segghen sal, Euestaltwichtich, Recht­ hefwicht, Scheefdaellini, en dierghelijcke daer de Weeghconst vol af is? sy en sijn der niet, de Natuer heeft daer toe aldereyghentlicxt het Duytsch veroirdent. [for where would you find any languages in which one can say Euestaltwichtich, Recht­ hefwicht, Scheefdaellini [‘of equal apparent weight’, ‘vertical lifting weight’, ‘oblique lowering line’] and the like, in which the Art of  Weighing abounds? They do not exist, Nature has specially designed Duytsch for it. (Dijksterhuis 1955: 86–87)]

In the work of the German Schottelius, the monosyllabic rootword found its cultural-patriotic apotheosis. It was an expression of pure, unmixed, ancient Germanness and a feature of original, early languages. And it was the basic unit of language, for Schottelius developed Stevin’s observations about the combinability of  German rootwords into an entire linguistic

From Humanist History to Linguistic Theory

103

theory. That view of  German is eloquently illustrated by the frontispiece to Caspar Stieler’s (1691) dictionary, where the German language appears as a banyan tree sustained by its aerial roots that double as trunks (Figure 2; cf. McLelland 2002 for fuller discussion of this imagery, and McLelland 2011a, chapter 2). Stieler’s dictionary, the first full monolingual dictionary of  German, listed German words by their root, rather than in strict alphabetical order, as the entry following illustrates: Sample entry for Baum, abridged from Stieler (1691: col.113–117) Baum / der / pl. die Beume / arbor, arbos. Baum etiam quod dicitur quodvis lignum aedificationi aptum. […] Beume pf lanzen / arbustare, arbores serere […] Der Baum trägt ein Jahr üms ander / arbor alternat fructus. Fachbaum […] Obstbeume […] Wilde Beume […] Beumlein / das. dim. arbuscula […] Beumen […] Faulbaumen […] Kürbaumen […] Baumlechtig & beumechtig […] Baumelen […] Baumelung […]

The cultural patriotic belief in the rootword as guarantor of  German an­tiquity and purity, as monosyllabic, and as identical with the imperative all faded in the eighteenth century, but Schottelius’s belief in its centrality as the basic unit of  language survived. In the mid-nineteenth century, it was still the structuring principle for the as yet unsurpassed Middle High German dictionary of Benecke (Benecke 1854–1866 [1990]),10 structured just like Stieler’s dictionary by rootword lemmata, and to which a more user-friendly alphabetical index was published by Lexer in the 1870s (Lexer, 1872–1878 [1979]). 10

The first fascicles of the new Middle High German dictionary which will ultimately replace Benecke’s work have been appearing since 2006 (Stuttgart: S. Hirzel, ed. by Kurt Gärtner, Klaus Grubmüller, and Karl Stackmann).

104

Nicola McLelland

Meanwhile, from Steinbach’s (1724) important insight that the root might not be identical with any existing form of  the word ( Jellinek 1898: 71) and then the systematic, synchronic comparison of Semitic and Sanskrit roots by Bopp and others, the idea of  the root as an abstraction gradually emerged, removed from etymological speculations, and, Rousseau claims, ultimately made ‘thinkable’ the idea of  the morpheme (Rousseau 1984: 294). If  Rousseau is right, then we must note the role of  the sixteenthcentury humanist Irenicus, at several removes, in developing the abstract notion of  the morpheme. Furthermore, as any student of  German linguistics knows, Schottelius’s rootword-based account of derivation and compounding as the two basic types of word-formation remains current in 21st-century introductions to German morphology. Naumann (2000: 42) makes the distinction in essentially the same terms as Schottelius: Für die Bildung deutscher Wörter […] hat die Wortbildungsforschung vor allem zwei Grundtypen erarbeitet, den Typ der Komposition oder Zusammensetzung ursprünglich voneinander abhängige Wörter zu einem neuen Wort aus mindestens zwei Teilen und den Typ der Derivation oder Ableitung, d.h. Wortbildung mittels Af fixen, also Präfixen, Infixen und Suf fixen. As regards the formation of  German words, researchers of morphology have postulated two basic types of word formation: composition or the merging of at least two, previously independent words to form one new word and derivation, i.e. word formation by means of af fixes, i.e. prefixes, infixes, and suf fixes.

4 Conclusion: The rootword in other Germanic languages I have already mentioned the inf luence of  the rootword on Danish and Swedish grammarians, to which should be added the striking case of a grammar of Russian that, though never printed, had some inf luence (McLelland 2010; Huterer 2001). One final example from Dutch may suf fice to illustrate the long shadow of  the Germanic root, in linguistic practice as well as in theory, and that is the case of systematic dif ferences between Dutch and

105

From Humanist History to Linguistic Theory

German spelling. In German, Schottelius (AA 191–192) adjudged that we should write Pferd because of  the inf lected form Pferde, even though the -d in Pferd sounds like a -t because of final devoicing (or final fortition). In Schottelius’s language theory, the integrity of  the rootword dictated that the rootword should remain the same throughout all word-forms. The Dutch grammarian Arnold Moonen (1644–1711) was a close adherent of  Schottelius’s linguistic theory, and held with him that the rootwords should remain the ‘same’, but he accepted that this ‘sameness’ could include consonant alternations that ref lected actual pronunciation (Moonen, 1706). He was thus the first of a series of eighteenth-century Dutch grammarians to take issue with the forms that the strict analogical principle of  Schottelius would require, such as wijv–wijven (‘woman, women’, see (Schaars 1988: 92), preferring instead to mark the dif ference in pronunciation. So Moonen accepted alternations of voiced and voiceless consonants in wordforms like geef–geeven (‘give’, first person sg., and infinitive or pl.). Séwel (1708) would follow Moonen, and the defeat of the analogical principle by the phonetic became established. The result is still evident in the contrast between standard Dutch and standard German spelling, where Dutch alternates consonants, and German does not (Schaars 1988: 94; Table 7.3; see Table 1).11 Table 1  Consonant alternations in Dutch spelling, compared with German Dutch

German

(English)

huis, huizen geef, geven

Haus, Häuser gib!, geben

‘house’ ‘give’

To conclude, this study has shown how the rootword – first extrapolated by an early sixteenth-century humanist from southern German as monosyllabic and identical with the imperative, and then repeated by other humanist historians – grew into a defining characteristic of  Germanic language theory. Having become indexical of antiquity and originality, it fed into

11

Compare English wife–wives, hoof–hooves.

106

Nicola McLelland

eighteenth-century debates on language origin. Understood as a verbal root, it inf luenced comparative linguists examining Semitic and Sanskrit well into the nineteenth century. As the building block of derivation and compounding, it inf luenced lexicography into the mid-nineteenth century; it yielded in Schottelius’s work the basis of present-day word-formation theory; and dif fering views as to its integrity led to dif ferences in spelling that are still salient in the Germanic languages today.

References Aedler, Martin [anon.]. 1680. The Hig [sic] Dutch Minerva […]. London: Printed for the author [Facsimile reprint Menston, England: Scolar Press, 1972]. Benecke, Georg Friedrich. 1854–1866 [1990]. Mittelhochdeutsches Wörterbuch. Mit Benutzung des Nachlasses von Georg Friedrich Benecke ausgearbeitet von Wilhelm Müller und Friedrich Zarncke. Nachdruck der Ausgabe Leipzig 1854–1866. 4 vols. and index. Stuttgart: S. Hirzel. Bopp, Franz. 1820 [1974]. ‘Analytical Comparison of  the Sanskrit, Greek, Latin and Teutonic Languages, Shewing the Original Identity of  Their Grammatical Structure’. Annals of  Oriental Literature 1, 1–64. Repr. edn. Amsterdam: E.F.K. Koerner, Benjamins. Considine, John. 2007. Dictionaries in Early Modern Europe. Cambridge: Cambridge University Press. Dijksterhuis, E.J. (ed.). 1955. The Principal Works of Simon Stevin. Vol. I. Amsterdam: C.V. Swets & Zeitlinger. Elspaß, Stephan. 2005. Sprachgeschichte von unten. Untersuchungen zum geschriebenen Alltagsdeutsch im 19. Jahrhundert. Tübingen: Niemeyer. Elspaß, Stephan et al. (eds). 2007. Germanic Language Histories ‘from Below’ (1700– 2000). Berlin: de Gruyter. Erlinger, Hans Dieter. 2007. ‘Zur Geschichte des muttersprachlichen Sprachunterrichts in Deutschland 18.-20. Jahrhundert’. In: Geschichte der Sprachtheorie 6/2. Sprachtheorien der Neuzeit III/2, Tübingen: Narr, 526–558. Gardt, Andreas, Ulrike Haß-Zumkehr & Thomas Roelcke (eds). 1999. Sprachgeschichte als Kulturgeschichte. Berlin: de Gruyter.

From Humanist History to Linguistic Theory

107

Habermann, Mechthild. 1997. ‘Das sogenannte “Lutherische e”. Zum Streit um einen armen Buchstaben’. Sprachwissenschaft 22, 435–477. Haß-Zumkehr, Ulrike. 2000. ‘Das Deutsche Wörterbuch von Jacob Grimm und Wilhelm Grimm als Nationaldenkmal.’ In: Andreas Gardt (ed.), Nation und Sprache. Die Diskussion ihres Verhältnisses in Geschichte und Gegenwart. Berlin: de Gruyter, 247–272. Hellmann, Manfred, & Schröder, Marianne (eds). 2008. Sprache und Kommunikation in Deutschland Ost und West. Ein Reader zu fünfzig Jahren Forschung. Unter Mitarbeit von Ulla Fix. Mit einem Geleitwort von Wolfgang Thierse. Germanistische Linguistik 192–194. Hildesheim: Olms. Henisch, Georg. 1616 [1973]. Teütsche Sprach und Weißheit: Thesaurus linguae et sapientiae Germanicae, A-G. Augsburg 1616. Repr. edn. Hildesheim. Hüllen, Werner. 2002. Collected Papers on the History of  Linguistic Ideas. Münster: Nodus. Huterer, Andrea. 2001. Die Wortbildungslehr in der Anweisung zur Erlernung der Slavonisch-Rußischen Sprache (1705–1729) von Johann Werner Paus. Munich: Sagner. Hutton, Chris. 1999. Linguistics and the Third Reich. Mother-tongue fascism, race and the science of  language. London, New York: Routledge. Jellinek, Max. 1898 [1985]. Ein Kapitel aus der Geschichte der deutschen Grammatik. In: R. Heinzel and F. Detter (eds), Abhandlungen zur germanischen Philologie. Halle: Niemeyer, 31–111. Jellinek, Max. 1913–1914. Geschichte der neuhochdeutschen Grammatik von den Anfängen bis auf  Adelung. Heidelberg: Carl Winter. Jones, William J. 1999. Images of  Language. German attitudes to European languages from 1500 to 1800. Amsterdam: Benjamins. Kaltz, Barbara. 2005. ‘Zur Herausbildung der Wortbildungslehre in der deutschen Grammatikographie von den Anfängen bis zum Ende des 19. Jahrhunderts.’ In: Schmitter, Peter (ed.), Geschichte der Sprachtheorie. 6/1 Sprachtheorien der Neuzeit III/1. Tübingen: Narr, 105–161. Kief fer, Bruce. 1978. ‘Herder’s treatment of  Süssmilch’s theory of  the origin of  language in the Abhandlung über den Ursprung der Sprache: a re-evaluation’. The Germanic Review 53, 56–105. Krebs, Christopher. 2010. ‘A Dangerous Book: The Reception of Tacitus’ Germania’. In: The Cambridge Companion to Tacitus, ed. A.J. Woodbridge. Cambridge: CUP, 280–299. Lee, Christina, & Nicola McLelland (eds). 2011. Germania Remembered. Tuscon, AZ: ACMRS (Medieval and Renaissance Texts and Studies).

108

Nicola McLelland

Lexer, Matthias. 1872–1878 [1979]. Mittelhochdeutsches Handwörterbuch: zugleich als Supplement und alphabetischer Index zum Mittelhochdeutschen Wörterbuch von Benecke-Müller-Zarncke. Repr. of the Leipzig edition 1872–1878. Stuttgart: Hirzel. Longolius, Johann Daniel. 1715. Einleitung zur gründtlicher Erkäntniß einer ieden, insonderheit aber der Teutschen Sprache. Budissin: Richter. McLelland, Nicola. 2010. ‘Justus-Georg Schottelius and European linguistic thought’. Historiographia Linguistica 37.1, 1–30. McLelland, Nicola. 2011a, 2011. J.G. Schottelius’s Ausführliche Arbeit von der Teutschen Haubtsprache (1663) and its place in early modern European vernacular language study. Oxford: Blackwell (Publications of  the Philological Society). McLelland, Nicola. 2011b. Des guten Gebrauchs Wegzeigere (Schottelius 1663: 10) – du bon usage dans la tradition allemande 1200–2000. In: Wendy Ayres-Bennett (ed.), Le Bon Usage. Paris: ENS. Metcalf, George J. 1974. ‘The Indo-European hypothesis in the sixteenth and seven­ teenth centuries’. In: Hymes, Dell (ed.), Studies in the History of  Linguistics: Traditions and paradigms. London: Bloomington. Mertens, Dieter. 2004. ‘Die Instrumentalisierung der “Germania” des Tacitus durch die deutschen Humanisten. Take-of f Phase der deutschen Tacitus-Rezeption und der Umbau der Geschichtsbilder’. In: Beck, Heinrich, Dieter Geuenich, Heiko Steuer & Dietrich Hakelberg (eds, Zur Geschichte der Gleichsetzung germanischdeutsch, by Johannes Hoops et al. Sprache und Namen, Geschichte und Institutionen. Berlin: de Gruyter, 37–101. Mieder, Wolfgang. 1982. Die Eintstelluing der Grammatiker Schottelius und Gottsched zum Sprichwort. Sprachspiegel 38, 70–75. Moonen, Arnold. 1706. Nederduitsche Spraekkunst. Amsterdam: François Halma. Mundt (ed.). 2008. Beatus Rhenanus. Rerum Germanicarium Libri Tres (1531): Ausgabe, Übersetzung, Studien. Tübingen: Niemeyer. Neis, Cordula. 2003. Anthropologie im Sprachdenken des 18. Jahrhunderts. Die Berliner Preisfrage nach dem Ursprung der Sprache (1771). Berlin: de Gruyter. Rössing-Hager, Monika. 1984–1985. ‘Ansätze zu einer deutschen Sprachgeschichts­ schreibung vom Humanismus bis ins 18. Jahrhundert.’ In: Besch, Werner, Oskar Reichmann & Stefan Sonderegger (eds), HSK Sprachgeschichte. Ein Handbuch zur Geschichte der deutschen Sprache und ihrer Erforschung. (Art.1144). Berlin: de Gruyter, 1564f f. Rousseau, Jean. 1984. ‘La racine arabe et son traitement par les grammairiens européens (1505–1831).’ Bulletin de la Societé de Linguistique de Paris 79, 285–321. Schaars, Frans. 1988. De Nederduitsche Spraekkunst (1706) van Arnold Moonen (1644– 1711). Wijhe: Quarto.

From Humanist History to Linguistic Theory

109

Schottelius, Justus Georg. 1641 [1651]. Teutsche Sprachkunst […]. Braunschweig: Gruber. Rev. edn. Braunschweig: Zilliger. Schottelius, Justus Georg. 1663 [1967]. Ausführliche Arbeit von der teutschen Haubtsprache. Braunschweig: Zilliger. Repr. edn. Wolfgang Hecht. Tübingen: Niemeyer. Séwel, Willem. 1708. Nederduytsche Spraakkonst. Amsterdam: Assuerus Lansvelt. Steinbach, Christoph Ernst. 1724. Kürtze und gründliche Anweisung zur Deutschen Sprache. […]. Rostochii et Parchimi: Apug Georg Ludw. Fritsch. Stevenson, Patrick. 2003. Language and German Disunity. A sociolinguistic history of  East and West in Germany, 1945–2000. Oxford: Oxford University Press. Stevin, Simon. 1586. De Weeghdaet / Beschreven Dver Simon Stevin van Brugghe [contains Vtspraeck vande weerdicheyt der dvytsche tael]. Leiden: Inde Druckerye van Christof fel Plantijn. By François van Raphelinghen. Stieler, Kaspar. 1691 [1968]. Der Teutschen Sprache Stammbaum und Fortwachs […]. Nuremberg: Johann Hof fman. Repr. mit einem Nachwort von Stefan Sonderegger. München: Kösel. Süßmilch, Johann Peter. 1766. Versuch eines Beweises, dass die erste Sprache ihren Ur­sprung nicht vom Menschen, sondern allein vom Schöpfer erhalten habe (1766). Berlin: Buchladen der Realschule. Takada, Hiroyuki. 1998. Grammatik und Sprachwirklichkeit von 1640–1700. Zur Rolle deutscher Grammatiker im schriftsprachlichen Ausgleichsprozeß: Tübingen: Niemeyer. Van der Lubbe, Fredericka. 2007. Martin Aedler and the High Dutch Minerva. The First German Grammar for the English. Duisburger Arbeiten zur Sprach- und Kulturwissenschaft, 68. Frankfurt: Lang. Van der Waal, H. 1952. Die eeuwen vaderlandsche geschied-uitbeelding, 1500–1800: een iconologische studie. The Hague: Martinus Nijhof f. Verstegan, R. 1605. A Restitution of  Decayed Intelligence: In Antiquities. Concerning the Most Noble and Renowmed [sic] English Nation. Antwerp. Weststeijn, Thijs. 2008. ‘The Germanic origins of art: Dutch and English an­tiquity according to Verstegan, Junius and Van Hoogstraten.’ Dutch Crossing 32, 44–70.

Agnete Nesse

Editorial Practices and Language Choice: ‘Low German Language Monuments’ in Norway

Abstract A great deal of documents from Norway in the late Middle Ages and early modern times were written in Middle Low German. Since the middle of the nineteenth century, many Low German documents have been edited, and editorial practices have varied significantly. Some editors have chosen to publish the texts without comments or explanations; others have used combinations of quoting (in the original language) and referencing (in Norwegian) within the same text. There are also editors who have chosen to translate the documents into Norwegian, only brief ly mentioning that the original was written in another language. The editorial practice often arises from the research questions perceived by the editor as relevant for the document in question. Given that what was considered as important for research questions has changed over time, we see that editions that are strongly linked to one specific research tradition can be dif ficult to use successfully today.

1 Introduction When working with historical texts, we are normally aware ‘that texts are produced and reproduced under specific social and institutional conditions’ (McGann 1991: 21). In other words, a given text is not simply the work of an author, but should also be considered as a ref lection of the time and the society in which it was written. It is therefore important, when working with historical texts, to regard spelling, grammar, style, argument and content

112

Agnete Nesse

as related to the social conditions of  the writer – and of  the copiers, who could also be important contributors to a given text (Haugen 1988: 72). Editorial practice must be viewed in the same way. In order to understand and successfully use an edition, it is necessary to analyse the potential research questions of the editor. It is important to recognize the relationship between the edition and the theoretical and ideological background of the editor, in order to avoid misinterpretation of  historical texts. We should ask ourselves a number of questions concerning the current editions we use and the future editions we want to publish: Why is a given text edited at a specific time and in a specific manner? What aspects of the text ref lect the interests of  the time and the society in which it is edited? What choices did the editor make, and why? (McGann 1991: 23). There is an extensive amount of  literature on the subject of editing and literary criticism, written by historians, linguists and literary critics (Fidjestøl et al. 1988, McGann 1991, Hunter 2006). The authors’ dif ferent approaches to the subject are not only concerned with their academic background, but also with the text type itself. Working with printed texts entails a dif ferent kind of challenge from working with handwritten manuscripts (as shown by Hunter 2006), and those who work with several manuscripts of  the same text follow dif ferent methods to those working with just one manuscript. Regarding some of the editors’ choices, I have chosen to examine texts from Norway written in Low German in the sixteenth and seventeenth centuries. The texts were edited over a specific period of  time, from the latter half of the nineteenth to the first half of the twentieth century. None of the texts were originally printed, hence they are all handwritten manuscripts, and with the exception of one text, only one manuscript exists in each case. I will only give a brief analysis of  the dif ferent texts here, as my main focus will be the choices the editors made according to the language of  the editions – in short, to what extent they translated the texts into Norwegian. One might assume that this is essentially an easy task: either they translated the texts or they did not. Yet there is a lot more to such ‘bilingual editing’, and I will discuss five dif ferent strategies.

Editorial Practices and Language Choice

113

2 Transcription only Obviously, one strategy when it comes to editing German ‘language monuments’ is simply to transcribe the handwritten text and publish it. There are dif ferent views among editors as to how many and what kind of alterations can or should be made to a text within this strategy. For example, with regard to dealing with abbreviations: Should abbreviations be expanded, and if so, how? Punctuation marks and capital letters are also often changed or added to make the text more readable, even within academic editions. The Norwegian collection of mediaeval charters, Diplomatarium Norwegicum, serves as an example of  this. The first volume was published in 1847, today twenty-two volumes with around 20,000 texts have been published, both on paper and digitally,1 and the twenty-third volume is on its way. In the preface of the first volume (Lange and Unger 1847: VIII–IX), the editors write that even though they have left all spelling mistakes in the editions, they have expanded abbreviations, because using særskilte dertil stöbte Tegn (‘specially cast symbols’) were expensive and for many readers the use of such symbols made the text dif ficult to read. Furthermore, capital letters are used on all nouns, til Lettelse for Historikeren (‘for the ease of the historian’). The preface states that the edition is for both historians and linguists, but that the two groups should use it somewhat dif ferently. The linguist could achieve a general picture of the linguistic changes in the period (from the eleventh century until 1570), and study dialectal dif ferences. However, for a detailed study of the text, the linguist should, according to the editors, consult the original manuscript (p. IX). In addition, linguists should work only with those texts in the edition that were based on original manuscripts, not with those based on copies. The texts based on copies are meant for historians only (p. X), and are therefore treated dif ferently: mistakes are corrected, the orthography is normalized, and so is the punctuation. The editors’ awareness of  the dif ficulties in providing one edition for both linguists and historians is remarkable, in light of this being one of  the very first academic editions of charters in Norway. 1

.

114

Agnete Nesse

Hunter (2006: 76–80) of fers an interesting discussion about how abbreviations, changed letter-forms and the like should best be presented in print. In his eagerness to modernize old texts within the framework of academic editing, he strongly supports a view that, at least in the eyes of a linguist, seems rather extreme. His claim (p. 85) that ‘Overall, the aim is to produce an edition which does justice to the content of  the manuscript, paying attention to its actual appearance but not fetishizing this’ shows little respect for the linguist’s need to know the actual orthography of  the writer, for example when it comes to abbreviations. However, the question of abbreviations is only mildly relevant to the text I have chosen, since the number of abbreviations in the manuscript is very few. Yet, as we can see from the photograph on the next page and from the copy of  the edition (Nielsen 1877), the abbreviated forms tym ˜ er and Tym ˜ erholt in the manuscript become tymmer and Tymmerholt in the edition, with no marking of  the expansion. The text in question is in the form of a dialogue between a Norwegian, De Norman, and a Hanseatic merchant, De copman, supposedly written in the early sixteenth century in Lübeck. De Norman represents the native inhabitants of  Bergen, and he complains about the behaviour of  the Hanseatic League, represented by De Copman. The merchant, on the other hand, claims that the accusations are false, and that the Hanseatic merchants are in fact following the law. He adds that if this is not the case, they will punish those who break the law. Nielsen’s first edition is followed by a number of comments, but none of the comments are concerned with the lack of translation. Is that because all his potential readers, Norwegian academics in the late 1870s, could read Middle Low German? Or was it because his potential readers were expected to handle linguistic challenges of  this kind, by taking the time needed to find out what the text was all about? From the dialogue De dudesche unde de Norman manuscript2 Nielsen’s edition (1877: 26): 2

The city archive of  Lübeck, Bergenfahrerarchiv, no. 1418.

Editorial Practices and Language Choice

115

De Norman. Item dat de copman suluen In den schof f vart of ft er [sic] volck vmme tymmer Berneholt vnde anderholt vp ores sulues kost vnde nicht enkopen van deme Normanne alsze oldinges plach to wesen Dar is der Normanne bergynge mede vorstort de sick hyr mede plegen to Bargende. De copman. Wy enlatet nenerleye Tymmerholt of fte berne holt houwen of fte halen ydt ensy myt der guder lude wyllen of fte orlof f den de schof f tobehort deme gheue wy dar al vul vor Breckt dar wol an dat is vns vnwitlick De do so vele dar vor alse recht is. [The Norwegian: Further that the merchants themselves go into the woods and obtain timber and firewood at cost price and do not buy it from the Norwegians as it has been customary since olden times. This ruins the livelihood of the Norwegians. The Merchant: We do not permit that timber is chopped or fetched, unless it is with the will or permission of the good people. To those to whom the woods belong, we give what they ask. If  there is anyone who acts against this, it is unknown to us. In that case, he will get what is right (i.e. punishment).]

What Nielsen does comment on, however, is the fact that he chose not to change the language into a standardized version of Middle Low German. This was an important issue of  the time: Scholars busied themselves with what James Milroy (2005: 336) has called ‘retrospective normalisation’, constructing standards like Old Norse, Middle Low German and Middle English.3 Nielsen’s reason for not standardizing was not that he was against it, but that he did not know how to do it:

3

Old Norse was standardized during the 1800s by Peter Andreas Munch and Rudolf  Keyser, amongst others. The reason was mainly romantic-puristic in nature: an aim to achieve the perfection that had once existed. But also pedagogical arguments were important: by standardizing the language, it was thought to be easier to read (Mørck 2006: 27).

116

Agnete Nesse Da jeg ikke tidligere har syslet med Udgiverarbeide af ældre nedertyske Sprogmonumenter, har jeg anseet det for den retteste Fremgangsmaade at levere et bogstavret Aftryk, uden at følge den Normalorthografi, som i den senere Tid er anvendt, f. Ex. i Udgaven af  hanserecesserne. (Nielsen 1877: 34) [As I do not have previous experience in editing older Low German language monuments, I have seen the most correct procedure to make a letter for letter print, without following the standard orthography, that has been used lately, for example in the editions of  the Hanserecesse.]

As a result of this, Nielsen was criticized by the German scholar K. Hegel for having added grammatical mistakes to the text (Nielsen 1892: 1). Yet why did Nielsen choose not to translate this text into Dano-Norwegian? Did he want to distance himself from the Hanseatic League? He interpreted the text as pro-Hanseatic, so editing it at all might have been controversial at a time when the emphasis on the Hanseatic period was negative. However, Nielsen was not one of  the most nationally orientated historians; he belonged to a more urban, European-minded school, so his choice of  text might have been an attempt to oppose the majority of  historians, who concentrated on building a sense of national identity. We could see his editorial practice as a compromise between showing his colleagues that the text existed, and not involving himself with the text by translating it. Was the dialogue pro-Hanseatic? According to Friedrich Bruns, a German historian who worked on the Hanseatic relationship between Bergen and Lübeck from many angles, the text was written by Jacob Dus, the secretary for the Bergen traders in Lübeck from 1512 until 1538 (Bruns 1901: 142). This may well be the case: the secretary for the Bergen traders must have had access to the dif ferent amendments, complaints and other letters concerning the Hanseatic settlement in Bergen. This dialogue text is also clearly a transformation of several political documents dealing with conf licts in Bergen into a didactical text. The most prominent of  these is an amendment issued in both Norwegian and Low German by King Christof fer in 1444 (Taranger 1912: 234–243). The part of De Norman in the dialogue follows this amendment quite closely; the part of De Copman is written as a defence, instead of a counter-attack. This is a choice made

Editorial Practices and Language Choice

117

by Dus; there are several documents where the Hanseatic Merchants complain about the Norwegians in Bergen, so in theory, the dialogue could have been written with De Copman as the plaintif f and De Norman as the defendant.

3 The text with comments Our next strategy is chosen by editors who acknowledge that their readers may have some problems with reading the original language, and therefore include word lists in their editions. In this case, we clearly see how important the target audience is. The first quotation is from an edition published in 1897, and the intended readers appear to be Norwegians who did not necessarily understand Middle Low German. The text itself is a Hanseatic journal (Der Gesellen Boeck) written between 1604 and 1671. Notes were taken each year in May and June, recording the apprentices who had participated in the initiation “games” (gespelet), and how this was celebrated – usually by happily drinking beer in the assembly rooms or in the gardens. Since the manuscript in question has many name lists and many words that are repeated several times, the editors chose to make a word list instead of a full translation: Man har dog ment, til veiledning for det publikum, som faar bogen ihænde, at give en oversættelse af de vigtigste og vanskeligere steder, eller av de steder, hvor en stadig gjenkommende vending første gang findes. (Bendixen and Krohn 1897: 16) [We have, for the guidance for the audience of  this book, given a translation of  the most important and more dif ficult words, or where a frequently repeated phrase comes for the first time.]

The editor makes the judgement for the readers as regards what is important and what is dif ficult within the text: yet, in reality, the readers’ interpretation of  the text often contradicts the editor’s own one.

118

Agnete Nesse

As a contrast to Bendixen and Krohn, we may look at an edition from 1946 of another kind of journal,4 written as an appendix to the Norwegian Germanist Olav Brattegard’s doctoral thesis. This is a so-called Maskopbok, where the people who owned shares in parts of  the same buildings and piers recorded, once a year, what each owner paid for maintenance. Here too, a word list is included, but it is clear that he has a dif ferent group of readers in mind. Eine Auswahl zur Erläuterung einiger mit den Verhältnissen in der hansischen Niederlassung zu Bergen eng verknüpften Wörter. (Brattegard 1946: 78) [A selection to explain some words that are closely connected to the conditions of  the Hanseatic settlement in Bergen.]

Many of  the words connected to the Hanseatic settlement were Norwegian loan words that were part of  the ‘Bergen-German’ language. Examples from Brattegard’s word list are gollfe (modern Norwegian gulv or golv [f loor]) and the phrase kuss mig j Rouven (modern Norwegian kyss meg i ræven [kiss my ass]), which are perfectly understandable for Norwegian readers, but not for German ones. As a result, Brattegard, who wrote and published his thesis in Bergen, explained words of  Norwegian origin to his readers. However, since we know that his aim was to analyse the Low German language in Bergen, it makes sense that he commented only on those words that belonged to this language, i.e. words that were not used in the Low German found in Germany. In other words, he addressed readers who were familiar with Low German of the sixteenth and seventeenth century in Germany. What the editions of  these journals have in common is the fact that the editors have not been interested in the authors, or in the authors’ intention concerning their texts, simply because there were so many authors, or rather scribes. Brattegard (1946: 19–41) recognized sixty-two dif ferent

4

These journals, of which several still remain, written during the period 1529–1936, in Low German, High German, Danish and Norwegian, are given a closer presentation in Nesse (forthcoming) ‘The Neighbours’ books of  Bryggen’.

Editorial Practices and Language Choice

119

writers in the period from 1577 to 1670 in one journal alone. This, of course, of fers great opportunities for variation studies, and Brattegard does give a description of  the orthography of  the dif ferent writers, concentrating on the features that shed light on the transmission from Low to High German (pp. 18–41).

4 The text with translation The strategy of editing a text and then translating the whole text is time consuming, but reader-friendly. In this way, those who can read the language of  the original do so, and those who do not understand the original language can still enjoy the content of the text. Only one of the many German texts written in Norway has been edited in this way,5 namely a Hanseatic journal from the sixteenth and seventeenth centuries, written partly in Low German and partly in High German, and translated into Dano-Norwegian (Bendixen & Krohn 1895). It is probably correct to say that no other text has been used so much for research on the Hanseatic society in Bergen as this one. It seems that the choice to edit both the original language and translation not only makes the text accessible, but it also gives the text an authenticity that makes its use particularly valuable. Hence, owing to its translation, this one small text has provided a more comprehensible (historical) account of  the manuscript, on behalf of a large amount of other texts. Brattegard (1945: 43) is critical of the editorial practice of Bendixen and Krohn, because of  the lack of accuracy in the editions. With regard to linguistic research, Brattegard’s own edition of a similar journal (1946) is a much better choice, given that it is both more accurate and its commentary more thorough.

5

In addition, there are four letters that were translated from Norwegian into Low German in the 1550s, where both versions are edited, three in Nesse 2008 and one in Nesse 2009.

120

Agnete Nesse

5 Translation only Including translation in a discussion of dif ferent editorial strategies may be controversial since, after all, the questions that arise when reading a translation are dif ferent to those that arise from looking at an edition. These questions are not concerned with how the text is presented, in terms of orthography, but rather with how the semantic level is retained. Nevertheless, there may be reasons to include this kind of work, as Hunter (2006: 88) does, if one wants to explore the dif ferent ways a historical text can be presented to modern readers, and what implications the editors’ choices have for research possibilities. One of the Norwegian editors who preferred translation to the original editing of  texts that were not from the Middle Ages was Nicolay Nicolaysen, a contemporary of  Nielsen. However, the two historians belonged to dif ferent schools and this may explain their dif ferent choice of editing. As mentioned earlier, Nielsen belonged to a school interested in urban, European history. Nicolaysen, on the other hand, worked within the national romantic paradigm. Among his many activities was the restoration of old buildings, such as churches – restorations that are today viewed as rather ruthless, given that everything that was added after the time of the original construction was taken away, so that the core building could once again take pride of place (Lidén 2005: 46–47). This can be compared to his ideas of  how historical texts from dif ferent periods should be treated: Anything that was changed after the Old Norse language (or Middle Low German for that matter), was not of linguistic interest, and could be ‘taken away’ – translated into modern Dano-Norwegian, whether it was written in German or Norwegian. This had to do with the process of nation building in Norway, a process that focused on the Middle Ages as the period before the Dano-Norwegian unification ‘ruined’ the political, cultural and linguistic independence of  the country. Since the written language was inf luenced by Danish after the fourteenth century, the nation builders considered it as impure. Nevertheless, Nicolaysen chose to edit several of the texts from the ‘400-year night’ (= the Dano-Norwegian union), and his three-volume work consisting of  texts written in Norway between 1550 and 1760 is of great value, even if his

Editorial Practices and Language Choice

121

editorial practices are considered as questionable by modern linguists and historians. When editing Die Nordische Saw (The Nordic Sow) from the 1580s, a religious and political text that is very critical of both the Hanseatic League and the Dano-Norwegian noblesse, he comments: Dog kan der neppe være tale om nogen anden fremgangsmaade ved udgivelsen, end […] at gjengive bogen i uddrag og paa modersmaalet. Thi dels indeholder den en hel del almindelige og vidtløftige gudelige betragtninger, som ikke ere trykning verd, og dels er sproget forskjelligt i de forskjellige gjenparter, endog i samme afskrift, da det bevæger sig mellem en slags Plattydsk og Højtydsk, hvortil kommer, at retskrivningen er meget unøjagtig. (Nicolaysen 1868: 4) [Still there can hardly be suggestion of any other approach for editing, than […] to reproduce the book in extract and in the mother tongue. Partly, it contains a whole lot of common and high-f lying religious views that are not worth printing, and partly the language is dif ferent in the dif ferent manuscripts, even within the same manuscript, where it moves between a kind of  Low German and High German. In addition, the spelling is very inaccurate.]

Front page of one of  the manuscripts.6 The text under the drawing reads (cf. figure 4): Lucie am .21. Capitt: Es wirden Zeichen geschein ahn der Sohn, Mohn vnd Stirne, Vnd auf f Erden wirtt leutten Bange Sein/: [There will be signs in the sun, moon and stars. And on the earth, the people will be afraid. (Luke 21:25)]

When Nicolaysen wrote mother tongue, he was referring to his own mother tongue, not the author’s. The term modersmålet was used for the DanoNorwegian written language. Neither Danish nor Norwegian were accurate enough, and the term Dano-Norwegian was considered too long. Hence mother tongue was used as a compromise. 6

The royal library of  Copenhagen, Gl.kgl.S.2836, made available to me by Geir Atle Ersland.

122

Agnete Nesse

For researchers today, it might be interesting to investigate the facts that determined Nicolaysen’s choice of editing. For those working with the history of ideas and religion, the combination of  Christianity and astronomy that was so typical of  the latter half of  the sixteenth century would definitely merit further study, since the text is constructed like many religious texts, with metaphors and parables that are then outlined for pedagogical purposes. For those concerned with variation linguistics, the way the author and /or the copiers have mixed Low and High German, and the inaccuracy of spelling, are of particular interest. Ultimately, what did Nicolaysen want to achieve when he edited this text? He wrote that he wanted to show the ‘spirit and tone’ of the text, and include anything of  historical interest. In terms of  historical interest, this mainly involved the fact that the Danish king was a corrupt ruler, who let the Hanseatic League exploit the Norwegian population, both economically and morally. It is clear, therefore, that editing texts from the sixteenth century in the nineteenth century was very much an act of interpretation – in this case, interpreting the past to serve the political aims of the present, i.e. the independence of  Norway. In comparison, a modern edition might emphasize the hatred which the anonymous Lutheran author displays to members of other religious groups. Catholics, Calvinists and Anabaptists are described in much the same way as Muslims are today in many western societies, thereby illustrating that religious dif ferences between the immigrants and the native population need not be great in order for hostility to arise.

6 Combining text and translation on sentence level The last strategy of  bilingual editing that I am going to discuss is in many ways the most intriguing, and also the one that is hardest to analyse. Like translation, this is not editing in the strict sense, but it is nonetheless important to mention in an overview such as this. In this case, the editor splits the text into parts that he chooses to quote directly, and parts that he chooses

Editorial Practices and Language Choice

123

to shorten and translate. This was not an uncommon technique, and my interpretation of this strategy has previously been that is was simply a way to save space, whilst ensuring that those passages that were most important to the editor were quoted directly (Nesse 2002: 143–146). When we look at the following, this may be a plausible explanation: Da i 1660 (29 april) der ehrsame Mikkel von Drebben auss eigener authoritet ohne versammling der nachbarn und gesellen i eldhuset havde spenderet Gesellerne 1 Rdl. og derved handlet imot Gaardsrettens § 20, maatte han bøde 1 Tønde Mel. (Nielsen 1892: 13)

The phrase that is a direct quote is set in italics; the translated and shortened passage is in ordinary font. The original version from the manuscript7 reads as follows: Anno: 1660 denn 29 April: hat der Ersahme Mickel von drebben: auß eigener authoritet, ohne versamblung der nachbarn und gesellen in sohlgarten denn gesellen: in elthause einen schlichten R: spendiret: als uns nachbar nicht anders wißent: undt hat damit erwehnter Mickell von drebben: wieder den 20 Artikkel unßer gartenrechte gehandelt: midt himit an den armen verbrochen Lauth selbige vorberuhrten artikell, i thune mehl und damit alles aufgehobenn. [Anno 1660, on April 29th, the honourable Mickel von Drebben treated the young merchants to 1 Reichstaler (worth of beer) without this being agreed on in the assembly of neighbours, and without the neighbours’ knowledge. In this way, the aforementioned Mickel von Drebben has violated article 20 of our association rules, and must pay to the poor people one barrel of  f lour, and by that it will all be annulled.]

Some of  the direct quotes in German involve individual words of no particular significance concerning the comprehension of the text, such as kersse (candle), malzeit (meal) etc. This means that there must be reasons other than perceived semantic importance regarding the editor’s choice to quote and to translate certain words or phrases. Nielsen may have seen himself as a mediator who could make the text more relevant and interesting to his readers than the original text as a whole would have been. He may have

7

University library of  Bergen, Ms 258.

124

Agnete Nesse

thought, like Nicolaysen when editing ‘The Norwegian Sow’, that the whole text from within these journals was not worth editing. In his edition, he chose excerpts from a large number of manuscripts, in both German and Dano-Norwegian, from the sixteenth century up until his own time. He translated a wide range of extracts from the journals to give his readers an impression of  how the Hanseatic League was organized. He even quoted some parts of the text, to give the readers a historical f lavour of Hanseatic authenticity. Hence, the direct quotes in German may have less to do with language and semantic importance than with style.

7 The editor as interpreter In conclusion, we see editions of  historical texts as voices from the time and the society in which they were made. The question remains, of course, as to whether this way of  looking at editions of  historical texts has any relevance for us today. Discussions of editing today are mostly discussions of digitalization. The possibilities that digital tools have to of fer seem to be limitless, yet the excitement surrounding methodological possibilities can easily overshadow the more theoretical aspects of editing. This becomes evident if we look at some of  the articles and books written in the 1980s on the prospects of  how computers could be used in analysing historical texts (Rindal 1988). Since such articles ref lect very dif ferent ideas from the interests of  those dealing with digital editing today (for example the mediaeval text archive Menota), we see how important the technical possibilities are when formulating research questions. There can be no doubt that digitalization opens up great possibilities when it comes to bilingual editing. By using parallel screens, with one language per screen, and hyperlinks to explain words and phrases, one can make editions that fulfil the needs of  both historians and linguists. Nevertheless, there are still choices to make and research questions to formulate. Consequently, by analysing earlier editions, we will be better prepared to create new ones.

Editorial Practices and Language Choice

125

References Bendixen, B.E., & Krohn, W.D. 1895. ‘Dat Gartenrecht in den Jakobsfjorden vnndt Bellgarden.’ Bergens historiske forenings skrifter 1. Bendixen, B.E., & Krohn, W.D. 1897. ‘Der Gesellen Boeck jm Jakubsforden vnde Belgarden, med indledning.’ Bergens historiske forenings skrifter 3. Brattegard, Olav. 1945. Die Mittelniederdeutsche Geschäftssprache des hansischen Kaufmanns zu Bergen. I. Die Sprache der Blütezeit. Bergen: John Grieg. Brattegard, Olav. 1946. Die Mittelniederdeutsche Geschäftssprache des hansischen Kaufmanns zu Bergen. II Der Ausklang des Niederdeutschen. Bergen: John Grieg. Braunmüller, Kurt, & Diercks, Willy (eds). 1993. Niederdeutsch und die skandinavischen Sprachen I. Heidelberg: C. Winter. Braunmüller, Kurt (ed.). 1995. Niederdeutsch und die skandinavischen Sprachen II. Heidelberg: C. Winter. Bruns, Friedrich. 1901. ‘Norweger und deutsche zu Bergen.’ In: Hansische Geschichts­ blätter. Leipzig: Verlag von Duncker & Hublot, 142–152. Halldórsson, Ólafur. 1988. ‘Tanker om tekstkritiske udgaver.’ In: Fidjestøl, Bjarne, Haugen, Odd Einar & Rindal, Magnus (eds), Tekstkritisk teori og praksis. Eit nordisk symposium. Godøysund 19.-22. mai 1987. Oslo: Novus, 11–23. Hunter, Michael. 2006. Editing Early Modern Texts. An Introduction to Principles and Practice. New York: Palgrave Macmillan. Lange, Christian, & Carl B. Unger. 1847. Diplomatarium Norwegicum, Første Samling, Første Hefte. Christiania: P.T. Mallings Forlagshandel. Lidén, Hans-Emil. 2005. Nicolay Nicolaysen. Et blad av norsk kulturminneverns historie. Oslo: Abstrakt forlag. McGann, Jerome J. 1991. The Textual Condition. Princeton, NJ: Princeton University Press. Menota hand book. Medieval Nordic Text Archive. accessed 12 August 2011. Milroy, James. 2003. ‘Purist ideologies on Historical Descriptions of  English.’ In: Langer, Nils, & Winifried Davies. (eds), Linguistic Purism in the Germanic Languages. Berlin, New York: Walter de Gruyter. 324–342. Mørck, Endre. 2006. Unormaliserte middelaldertekster. Tekstprøver, tekstkommentarer og bakgrunnsstof f. Bergen: Fagbokforlaget. Nesse, Agnete. 2002. Språkkontakt mellom norsk og tysk i hansatidens Bergen. Oslo: Novus.

126

Agnete Nesse

Nesse, Agnete. 2008. ‘Bilingual texts from a bilingual city.’ In: Ersland, Geir Atle & Marco Trebbi (eds), Neue Studien zum Archiv und zur Sprache der Hanseaten. Bergen: Museum Vest. 47–64. Nesse, Agnete. 2009. ‘Flerspråklige kilder fra Bergen.’ In: Folkmålsstudier 47, 109–132. Nesse, Agnete. forthcoming. ‘Four languages, one text type: The neighbours’ books of  Bryggen 1529–1936.’ In: Stenroos, Merja, & Martti Mäkinen (eds), Proceedings from Historical Language and Literacy in the North Sea Area, Stavanger 2009. John Benjamins. Nicolaysen, Nicolay. 1868. Norske Magasin. Skrifter og optegnelser angaaende Norge og forfattede efter reformationen. Vol. 2. Christiania: Johan Dahls forlag. Nielsen, Yngvar. 1877. ‘De dudesche Kopman unde de Norman.’ In: Forhandlinger i Videnskabs-Selskabet i Christiania Aar 1876 8. Christiania: Jacob Dybwad. Nielsen, Yngvar. 1892. ‘Vedtægter og Dokumenter fra det hanseatiske Kontor i Bergen og dettes enkelte Gaarde.’ In: Christiania Videnskabs-Selskabs Forhandlinger for 1892 7. Christiania: Jacob Dybwad. Rindal, Magnus. 1988. ‘Bruk av datamaskin i arbeidet med norrøne tekstar.’ In: Fidjestøl, Bjarne, Haugen, Odd Einar & Magnus Rindal (eds), Tekstkritisk teori og praksis. Eit nordisk symposium. Godøysund 19.-22. mai 1987. Oslo: Novus. 46–62. Taranger, Absalon (ed). 1912. Norges gamle love, anden række 1388–1604, første bind 1388–1447, tekst. Christiania: Grøndahl & søn.

Part 2

Standardization and Authenticity

Robert Evans

Of ficial Languages: A Brief  Prehistory1

Abstract Historians of politics and administration have largely remained distant from the concerns of historical sociolinguistics, whereas practitioners of the latter have tended either to neglect the politics of  language or to use vague and sweeping terms like ‘national’ and above all ‘of ficial’ language. This article examines some cases of  how languages came to be formally recognized and regulated by dif ferent states or governments. It argues that the process was much slower than has often been assumed, precisely because languages which were dominant in social and cultural terms long needed no of ficial mandate, especially when mechanisms of authority were comparatively little developed.

Historians always have to do with language; but they have made slow progress towards any real interdisciplinary dialogue with linguists. So far it has been mainly social and literary aspects of  their common territory which have attracted attention (Burke, 2004). Nowadays the worlds of (shifting) semantics and discourse are also coming into play more fully – as the present collection will confirm. Yet traditional ‘mainstream’ history, especially of  the political-administrative kind, has remained distant. And this distance seems to be mutual, shared by historical sociolinguistics, which either neglects the politics of  language or is given to vague and sweeping usage of terms like ‘national’ and above all ‘of ficial’ language (cf. Haarmann 1975: 92–119, a useful compendium for its day, but superficial). That ref lects an underlying incongruity. Languages nowadays represent a major political issue: the formal and prescribed status of one or 1

This text incorporates some material used in Evans (2006) and Evans (2011).

130

Robert Evans

more of them in a given state can be cause for fierce contention. However, they have only recently become so, and particularly in those states with the most powerful languages. Thus France has experienced centuries of constitutions, currently that of  the Fifth Republic since 1958; but only on 25 June 1992 was a clause added to assert that ‘la langue de la République est le français’. In the US, English enjoys of ficial status in 34 states and territories, but not in the country as a whole.2 The equivalent situation in the UK is perhaps moot, but the rights of English remain basically unwritten, and the same goes for some other comfortably established European languages too, such as Swedish in Sweden. In the USSR, Russian was a ‘national’ (narodny) and an ‘international’ (mezhdunatsionalny), but not a ‘state’ (gosudarstvenny) tongue. The Italian Costituzione of 1947 (art.3) and the German Grundgesetz of 1949 (art.3, iii) have general statements about the equality of citizens in respect of a list of criteria including language, and about rights for minority languages. In Italy further guarantees for specific minority tongues were introduced in 1999, and eventually in 2007 a measure was proposed to confirm the formal status of  Italian itself. In Germany that has not happened – and even the wording of  the Grundgesetz largely repeats the relevant paragraph (art. 113) of the Weimar constitution of 1919.3 Is this state of af fairs more than a curiosity, important mainly for current pressure groups, political scientists and the like? Surely modern ‘national’, standard languages have a clear hegemony anyway? And what about the mass of more contingent language regulations over the centuries, designed to serve the same purpose? That, however, is precisely the problem, and the sense of my title for the present essay: the notion of  ‘of ficial’ language is de facto readily applied these days in much earlier contexts where it is inappropriate. There are real issues here relevant for 2 3

See accessed 30 May 2011, and links. URLs: ; ; ; ; accessed 15 October 2010.

Of ficial Languages

131

sociolinguistic and other kinds of analysis: precisely why of ficial language – in any proper acceptation – has been a rare, often even a non-existent category until lately. A related distinction needs to be made at the outset between, on the one hand, statute law, and in particular modern constitutions, and, on the other, earlier systems, usually monarchical, of administrational directive: decrees, orders, ordinances, regulations, patents, rescripts, edicts … the list of  terms employed is long and (revealingly) confusing. These were products of quite dif ferent and by now unfamiliar forms of executive authority. Not all of  them were necessarily even fully public statements of intent. It is correspondingly dangerous to assume that ‘language legislation’ in pre-constitutional systems belonged to any ‘of ficial’ category at all, in our sense. Equally, what we now identify as significant enactments of that sort were not necessarily seen as such at the time, or through the intervening period. In what follows I shall try to do some preliminary mapping, with a concentration on Great Britain and on central Europe, regions with significantly divergent experience in the field. I shall introduce half a dozen brief case studies, some of  them familiar. Yet their very familiarity can be perilous! These texts are not straightforward exemplars. And we should note areas where the apparent absence of evidence may be equally telling. Besides, many gaps have still to be filled – for whole vast areas, like Scandinavia, we lack readily accessible information. * It’s a wild but serviceable generalization that we find little sign of ‘of ficial’ action to support language across a swathe of ancient and medieval empires and composite monarchies. Prominent examples would be the largely unenforced spread of Latin through the Roman Empire, and its coexistence with a continuing widespread use of Greek at both popular and elite levels. Then, outside the territories under the cultural domination of Byzantium, medieval Latin took on the role which had previously fallen more to Greek, as the language of civilization (Budinszky 1881, Neumann & Untermann 1980, Richter 1995, and cf. Waquet 1998 for the post-medieval development). Unusually early evidence of suasion seems, by contrast, to come precisely from the precocious statehood of countries in western Europe

132

Robert Evans

whose main languages later didn’t need safeguards. I cite three of the most famous cases, two from the British Isles and one from France, although the caveat just entered will need to be borne in mind. In 1366 the English authorities in an already partially conquered Ireland issued the statutes of  Kilkenny. They include decrees whose tenor is clearly dictated by a situation of languages in conf lict. English is set against Gaelic in aggressive fashion, as with condemnation of  the fact that many of  the local settlers, forsaking the English language [etc.], live and govern themselves according to the language [etc.] of  the Irish enemies […] whereby […] the English language [etc.] […] are put in subjection and decayed, and the Irish enemies exalted and raised up, contrary to reason.

It is therefore commanded, among other things, that ‘every Englishman do use the English language’ [etc], and severe punishments are threatened: If any English, or Irish living among the English, use the Irish language among themselves, contrary to this ordinance, and therefore be attainted, his lands and tenements, if  he have any, shall be seized into the hands of  his immediate lord, [or, if not] his body shall be taken of f  by one of  the of ficers of our Lord the King, and committed to the next gaol. (Crowley 2000: 14–16)

But we should bear in mind that all this was part of a much longer and broader text, and had an impact only in the Pale of  English settlement around Dublin; and that the statute itself was actually issued in French (as were most acts of  the Irish parliament till 1472). Moreover, its backdrop lay, as the above extracts imply, in a loss of ground by English locally. That weakness continued in Ireland into and through the sixteenth century. Kilkenny was revoked in its linguistic aspect in 1495, though evidently not because its concerns had been assuaged, since it had to be reimposed a century later. Meanwhile, a series of related enactments were passed to secure or assert English as a tongue requisite for the law, for appropriate forms of schooling, and for grants of urban citizenship (Crowley 2000: 20f f., cf. also Hogan 1927: 15–36 and Rockel 1989: 64–72). Thus the provisions for English in Ireland were still in essence a defensive response.

Of ficial Languages

133

A related instance is the text of  the so-called Act of  Union, issued in 1536, for the incorporation of  Wales into England. Wales, a complex of  lordships and jurisdictions in the west of  Britain, had in fact developed a more sophisticated vernacular legal tradition in the Middle Ages than England. Yet the English language was already taking over in administration there, alongside Latin, by the fifteenth century (Smith 1997, for the historical context, cf. Williams 1997: 253–278). The 1536 measure ref lected that, particularly in its § 20: Also be it enacted by the authority aforesaid, that all justices, commissioners, [etc., etc.] shall proclaim and keep the sessions, courts [etc.] in the English tongue; and all oaths … and af fidavits [etc.] to be given and one in the English tongue; and also that from henceforth no person or persons that use the Welsh speech or language shall have or enjoy any manner, of fice, or fees within this realm of England, Wales or other the King’s dominion, upon pain of forfeiting the same of fice or fees, unless he or they use and exercise the English speech or language. (Bowen 1908: 87)

We should note that this provision related only to Wales. Although there is mention of England too, it is clearly aimed at Welsh people alone. Remember that England under the Tudors still made much use of  law French in its court proceedings, and verdicts even continued to be recorded in Latin till 1732 (Catto 2003). The decision taken in 1536 was connected directly with executive homogeneity, smoothness and ef ficacity, and with judicial reorganization. It confirmed a sociolinguistic process. The ‘Act of  Union’ (Deddf  Uno), with – inter multa alia – its ‘language clause’ (cymal iaith), as much later nationalists would call it, provoked no significant protest in the country at all, and was actively welcomed by the gentry, attracted by the prospect of equality of status and by the opportunities to be vouchsafed them if  they operated in English ( Jenkins and P.R. Roberts in Jenkins 1997: 78f f ). There may have been a culturo-political. agenda. ‘His Highness’ [King Henry VIII], as the Act indicates at another point, ‘minding and intending to reduce [his Welsh subjects] to the perfect order, notice and knowledge of his laws of this his realm’, did undertake ‘utterly to extirp all and singular the sinister Usages and Customs dif fering from the same’ (Bowen 1908: 75). Did such ‘sinister Usages and Customs’ include their Celtic language,

134

Robert Evans

so radically alien to the English eye and ear? If so, however, the agenda was not a consistent one. In 1563, legislation enjoined the production of religious texts in Welsh, building on the claim in the twenty-fourth of the recently promulgated Thirty-nine Elizabethan Articles of faith: ‘It is a thing plainly repugnant to the Word of  God, and the custom of  the Primitive Church, to have Publick Prayer in the Church, or to minister the Sacraments in a tongue not understanded of  the people.’ From that initiative grew the immensely inf luential Welsh Prayer Book and Bible translations, which did so much to sustain the language over the next centuries, for all its subordinate status in worldly af fairs (Williams 1967, Morgan 1988). Exactly contemporary with the legislation for Wales is the best-known of all these directives, the Ordinance of  Villers-Cotterêts, issued in 1539. This lengthy document revised procedures of governance within the French monarchy. Towards the end of it, to cover what we would nowadays call matters of  ‘transparency’, we find articles 110 and 111: nous voullons et ordonnons que [les arrêts] soient faictz et escriptz si clerement qu’il n’y ayt ne puisse avoir aucune ambiguïté ou incertitude [decrees so clear that there is not and cannot be any ambiguity or incertitude] … Et pour ce que telles choses sont souventes fois advenues sur l’intelligence des motz latins contenuz esd arrestz [understanding of the Latin words contained in the decrees] nous voulons que doresnavant tous arrestz … de nos courtz souveraines [all judgments of high courts] (etc.) soient prononcez, enregistrez et delivrez aux parties en langaige maternel françois et non autrement. (Chaurand 1999: 149)

The earlier part of this justification stresses public ignorance of Latin texts, and hence the scope for their abuse. But the larger issue appears at the end, and has given rise to much debate. Is this ‘langaige maternel françois’ to be equated with the langue du roi, the speech of  the Valois court and administration, as long argued – pro and con – by both supporters and detractors of the French state, and by experts like the famous and exhaustive chronicler of  French linguistic history, Ferdinand Brunot? Or did it continue to include dialects (previously specified as ‘vulgaires du pays’ or similar) (Brunot 1906: 30–32)?4 4

Cf. also the important qualifications of  the traditional view in Fioretti (1950) and Trudeau (1983), as well as the more recent discussion in Babel (2007).

Of ficial Languages

135

Whatever its purpose, Villers-Cotterêts certainly in the event encouraged standard forms of French, which were anyway spreading (Febvre 1924). French still needed its Def fences (of  the kind mounted most memorably by Joachim du Bellay); and dialect might still seem a liability for it. Yet clearly it was becoming the language of an increasingly secure culture and of reasonably uniform channels of state power (cf. the useful survey in Schmitt (2000), esp. 682f f.). * In central Europe, later to be the classic linguistic battleground, there were few earlier signs of  trouble. Latin and German coexisted for centuries as public media in the Holy Roman Empire, apparently without any serious confusion or friction (Hattenhauer, 1987). Polish underwent a gradual emancipation, from under a shadow cast by both Latin and German; but there, too, clashes remained limited, and mainly took place at a local, municipal level. And the multiethnicity of  the Dual Commonwealth of  Poland-Lithuania went with a de facto multilingualism which allowed for gradual language shift. The year 1697 saw the final paper confirmation of a sociolinguistic process right across the Rzeczpospolita with the abolition of a residual administrative role for Ruthene in Lithuania and its replacement by Polish.5 Multilingual Hungary and Croatia needed no authoritative assertion of the rights of Latin which for hundreds of years retained, even enhanced, its well-nigh all-embracing functions in public life. The one big exception was Bohemia.6 There already in the high Middle Ages Czech–German discord had a clear linguistic dimension, most famously at the university of  Prague, where in 1409 the decree of  Kuttenberg, alias Kutná Hora, enforced Czech hegemony. The Hussite wars and the establishment of the Utraquist church then confirmed Czech vis-à-vis both Latin and German as the language of the politically dominant estates and towns, as of non-Catholic religion. All this was progressively inscribed in a series of decrees for the diets, law courts, etc., from the later fifteenth century. But it was also accompanied by a defensive strategy for 5 6

Cf. Evans (2006) for further references. For the details of what follows, see Evans (2011) and references there.

136

Robert Evans

this ostensibly beleaguered Slav culture, increasingly so in the sixteenth century (contrast the comparative self-assurance of  Polish-speakers). Hence, alongside much evidence of coexistence in Bohemia, we have some notable private and public statements of  linguistic concern, from the religious leader Jan Blahoslav in 1571 to the oppositional commentators Pavel Ješín and Pavel Stránský at the time of  the revolt against the Habsburgs which broke out in 1618. Above all we have the evidence of a strange and perhaps unique law of 1615, ‘On the Preservation and Cultivation of  the Ancient Czech Language’ [O zachování starožitného jazyka českého a vzdělání jeho]. This bemoans, inter alia, that the extinction of  the Czech language would bring with it also the extinction of  the Czech nation […] introducing all manner of foreign tongues and nations […] more and more foreigners continually enter the country […] many of them unable to string together three words of  Czech […] foreign priests with no Czech are installed and maintained […].

As remedies it enacts that such foreigners should forthwith be obliged to have their children learn Czech from their youth […]. [H]eirs and heiresses […] who can speak the Czech language well shall inherit landed property […] in double measure […]. [N]o foreigner who does not know Czech and cannot intelligibly present his needs in that tongue shall by any means be accepted as a denizen of  the country or as citizen of a town […]. [N]o such foreigner newly accepted into the country nor his children to the third generation shall be preferred to any state or municipal or other of fice, or to any courts of  law [etc., etc.]. (Skála 1865–1870: 355–358)

These weird and unenforceable stipulations were evidently never implemented – and anyway the Thirty Years War soon intervened. In fact the fears underlying this outburst did come to be realized after the defeat of  the Bohemian revolt in 1620. But the decline of  Czech was not accomplished by legal means (the revised land ordinance, or constitution of  the day, merely placed German alongside it as a language of state). Altogether there appears to be little sign of  language ‘legislation’ through the ancien regime Europe of that and the succeeding period. Further examples of active intervention might be found in sixteenth-century

Of ficial Languages

137

Savoy, where promotion of the vernacular ran pari passu with what we have already examined in France, and seventeenth-century Sweden, where the introduction of  Swedish into Skåne, once that province was conquered from the Danes, is cited as a precocious campaign of  linguistic education (although both standard tongues may have been equally remote from the dialect of the indigenous population). Yet such state-led endeavours were typically ad hoc and localized, at best: witness the decrees of the Bourbon kings requiring newly acquired territories to be administered in French (Babel 2007). Only in the eighteenth century did things begin to accelerate. One early and distinctive case was another act of  Bourbon policy: that of  the junior branch which after 1700 won control of Spain by military force. Its tough Decretos de Nueva Planta (1716) enforcing the ‘lengua Castellana’ for courts etc. in defeated Catalonia inaugurated an irregular series of measures which buttressed the legal status of  language in Spain down to our own day.7 Change was, however, mainly associated with reformist absolutisms post-1750, and it remained piecemeal and inconsistent. Emperor Joseph II, for instance, gave support to vernaculars, for the instruction of the people, alongside the ancillary linguistic aspects to his headlong initiatives for centralizing and homogenizing his realms.8 I shall return to him shortly, in the context of  Hungary. * In the history of of ficial languages, as in so much else, the French Revolution played a transformative role, and one which is now better attended to in the literature (cf. Certeau et al. 1975, Renzi 1981, Bell 1995, Bell 2001). It invented the modern written European constitution (alongside the abortive 7

8

Decrees reproduced at [last accessed 30.5.11]. The Spanish constitution of 1931 already prescribed that ‘el castellano es el idioma of ficial de la República. Todo español tiene obligación de saberlo y derecho a usarlo’, a wording substantially repeated in 1978. See Lexikon der roma­ nistischen Linguistik (368–378). For the vicissitudes of  Catalan, see Ferrer i Gironès (1985). I am very grateful to Rodney Sampson (Bristol) for help with Spanish. Much evidence passim in Derek Beales (1987–2009).

138

Robert Evans

Polish equivalent). Some revolutionaries – above all the doctrinaire Abbé Grégoire – pressed for the exclusive claims of  the ‘langue nationale’. This was soon abandoned (and we should bear in mind that the Revolution also temporarily discontinued the Académie Française, which had established criteria for that language). But it presaged the immense post-revolutionary and Napoleonic centralizing pressures for standard French and the whole nineteenth-century normalizing experience described according to the ‘Weber thesis’ as turning ‘peasants’ (and others) into (linguistic) ‘Frenchmen’ (Weber 1976). The impact of  these developments on ‘of ficial’ language policy was slower and more indirect. The French model of  the nation, combined with the enthusiasms of  the Romantic age, tended to revive and assert minority as well as majority cultures on an ethnic base. The resultant f lashpoints lay above all in the linguistic congeries of central Europe. In two important cases this yielded explicitly multilingual stipulations. One was Switzerland, where – it is frequently overlooked – until 1798 all the full cantons had used German, so the issue was not formally raised earlier (cf. Arquint et al. 1982). Even the expanded post-Napoleonic Confederation did not mention language in its constitution. Only in 1848 did the next such document, ref lecting the liberal takeover of  that year, declare that ‘[t]he three chief  languages of  Switzerland, German, French and Italian, are the national languages of  the Confederation’.9 Still nothing further; and even this article stood as number 109, under ‘miscellaneous provisions’, after long sections on political, military, commercial, religious, legal, and institutional matters. The same form of words would pertain until 1938, when Rhaetoroman was added, while the first three tongues gained formal ‘administrative’ status, as Amtssprachen too. Austria, under its own emperors from 1804, had a similar tradition of  German linguistic dominance; but there no language commanded a majority of speakers, and a wide range of other vernaculars existed. The one serious attempt at ‘Germanizing’ proved a short phase during the 1850s, and since the government was then busy promoting a rhetoric of 9

‘Die drei Hauptsprachen der Schweiz, die deutsche, französische und italienische, sind Nationalsprachen des Bundes’: see article 109 in www.verfassungen.de/ch/ verf48-i.htm [last accessed 13.10.2010].

Of ficial Languages

139

multinational harmony, it elicited no clear public statements. Then in the 1860s the Habsburg rulers were forced into granting definitive autonomy to Hungary and concessions to the other nationalities in the rest of  the Monarchy. Hence the Austrian constitutional provision of 1867 to the ef fect that [a]ll the peoples [Volksstämme] of the state have equal rights, and every Volksstamm has the inviolable right to preserve and cultivate its nationality and language. The equality of all customary languages in school, of fice and public life is recognized by the state.10

This brief but portentous statement ef fectively came to ensure a multiplicity of languages of state – and it also went with a parallel guarantee of the freedom of all citizens from any obligation to learn a second language. For all the protests of nationalists, neither then nor later did German become any more of ficial than the other vernaculars; though the stenographic practice of the Reichsrat, Austria’s parliament, of not recording any speeches delivered in those vernaculars, pointed up the continuing practical hegemony of  German in the public life of  the state. The clearest and most detailed example known to me of the making of an of ficial language is the nineteenth-century evolution in the other half of  the Habsburg Monarchy: Hungary. That country saw the longest survival anywhere in Europe of  historic bipolar (ruler and estates), corporative arrangements of rule, and without a measurable language problem despite a range of spoken tongues as great as in Austria. The much-interpreted maxim of  Hungary’s first king, St Stephen, that ‘regnum unius linguae uniusque moris imbecille et fragile est’, was at least ben trovato for this situation, where Latin held its place in administrative functions, and in mediating relations with the largely absentee monarchs, while the vernaculars, some more international and some more local, all operated concomitantly (cf. Evans 2007 for an overview). 10

‘Jeder Volksstamm hat ein unverletzliches Recht auf  Wahrung und Pf lege seiner Nationaltät und Sprache. Die Gleichberechtigung aller landesüblichen Sprachen in Schule, Amt und öf fentlichem Leben wird vom Staate anerkannt’: Bernatzik (1906: 370). See Stourzh (1985) for an authoritative commentary and Fischel (1910) on nineteenth-century Austrian language legislation in general.

140

Robert Evans

Everything quickly began to change, for languages too, with the challenge in the 1780s from a radically reformist ruler, Joseph II, who prescribed German henceforth for all administrative purposes. His measure, although short-lived and futile, yielded a constitutional and then a romantic backlash. This provoked a singularly protracted, concentrated and virulent campaign for legislation in favour of Magyar, the best-established of Hungary’s spoken languages, which delivered a stream of enactments initiated in 1792 and completed by 1844.11 It simultaneously fuelled a secondary backlash from the country’s other nationalities: resistance at the diet mainly from Croat deputies (who regularly protested in Latin!); and beyond the diet a pamphlet war, with supplications and complaints about ‘Magyarization’. What limited ef fective Magyarization there was in fact proceeded organically and peacefully; whereas the public and of ficial measures in its favour contributed to widespread unrest and then in 1848–1849 to a civil war which brought temporary nemesis to the Magyar camp. After the hiatus of  the 1850s, further extended debates took place, in counterpoint with the accomplishment of  Hungarian self-government. They resulted by 1868 in what is described as a ‘nationalities law’ (törvény a nemzetiségekről), in fact the most elaborate language legislation of  the whole period. It starts with an important preamble on what constitutes the ‘of ficial’ sphere: Whereas all citizens of  Hungary, according to the basic principles of  the constitution, form one nation in political terms, the indivisible unitary Hungarian [magyar] nation, of which every citizen of  the country, whatever nationality [nemzetiség] he may belong to, is an equal member; and whereas, further, that equality can only be liable to specific regulation in respect of  the of ficial usage of  the various languages prevailing in the country, and in so far as the unity of the country, the practical scope of government and administration, and the precise dispensation of justice require … [we therefore enact as follows …]12

11 12

Full documentation in Gy. Szekfű (1926) and in a number of brilliant works by Daniel Rapant, of which the first, Rapant (1927–1931) sets the scene in virtuoso fashion. ‘Törvénycikk a nemzetiségi egyenjogúság tárgyában: Minthogy Magyarország összes honpolgárai az alkotmány alapelvei szerint is politikai tekintetben egy nemzetet képeznek, az oszthatlan egységes magyar nemzetet, melynek a hon minden polgára, bármely nemzetiséghez tartozzék is, egyenjogú tagja;

Of ficial Languages

141

Magyar was reasserted as ‘language of state’ (államnyelv); and precise provision made for other vernaculars in (lower levels of ) administration, education, and the judiciary. Minorities qualified for such provision if they formed at least 20 per cent of the population locally; they could draw on translation facilities and interpreters, keep records in their own tongue; etc. Much of  this was merely a paper exercise. The regulations were not properly enforced – a consideration we must always bear in mind when assessing ‘of ficial’ sources. But anyway the non-Magyar leaders rejected the law’s assumptions from the beginning and sought to achieve something much closer to the Austrian solution which we have already encountered.13 In the end their attitudes contributed greatly to the breakdown of historic Hungary after the First World War. * Ironically the Hungarian legislation of 1868 served as a model for some of  the many new enactments after 1918 across central and eastern Europe which at last did place language firmly in the of ficial sphere. Czech, for instance, was now restored, after exactly 300 years, to pre-eminence in the new Czechoslovakia, even if it shared its place with its less well-versed sister-tongue of Slovak (Kučera, 1999).14 In their turn, the post-war settlements in Europe established a yardstick for the formal rights of languages, at least those with state backing, across more and more of the world. Why did of ficial languages have such a long prehistory? A few conclusions may seem to grow out of  the very cursory examination of fered here.

13 14

minthogy továbbá ezen egyenjogúság egyedül az országban divatozó többféle nyelvek hivatalos használatára nézve, és csak annyiban eshetik külön szabályok alá, a men�nyiben ezt az ország egysége, a kormányzat és közigazgatás gyakorlati lehetősége s az igazság pontos kiszolgáltatása szükségessé teszik …’: Magyar törvények, 1865–8 (Pest, 1869), 151 [my translation]. For the debates surrounding this cryptic and unwieldy formulation, see Schlett (2002). For the wider context, cf. Evans (2004). There is no proper study; but many clues in Kemény (1952–1985). The peace treaty spoke of a ‘langue tchèque’; but the constitution identified the language of state as ‘Czechoslovak’, in its two varieties (znění), each of which was de facto of ficial in one of the two halves of the country. For the 1920 laws and their implementation, see Kučera (1999).

142

Robert Evans

It was, of course, the development of administrative organs and the spread of  literacy which first generated the requirements for of ficial prescriptiveness about language. So did the rise of formal institutions needing a (single) medium of communication, particularly diets, parliaments, and the like. Yet the processes were slow-moving, and they did not invalidate surviving modes of more casual and unstudied negotiation. Slow too was the emergence of  language as a mark of prestige and badge of allegiance. That went with the gradually changing role and priorities of multilingual elites, and their interaction with increasingly prominent, and maybe increasingly mobile, monoglot social groups. Moreover, a key underlying factor in rendering language an of ficial concern has repeatedly been a sense of  threat and vulnerability: from the time of  the Kilkenny statutes and the Bohemian patriots of 1615, even to recent reactions to the perceived challenge from English – or franglais – in France, or Spanish in the US. So has my theme been ultimately a kind of non-subject, the more so since some of the most celebrated examples of earlier language enactments may (as we have seen) not bear the weight of all that was subsequently attributed to them? No, because the episodes we have observed were significant ones, ripe to be built upon in later rhetoric. Fuller study of  them would throw up much evidence of the actual workings of both of ficial and unof ficial channels of  linguistic control and direction. In the end they all have to do with power, with its longer-term, subtler manifestations in that public sphere which language always occupies, and thus with political and cultural kinds of interplay between individuals and groups. As such they make a claim on the attention of  historians and linguists alike.

References Arquint, Jachen C. et al. 1982. Die viersprachige Schweiz. Zurich: Benziger. Babel, Rainer. 2007. ‘Sprache und Politik im Frankreich der Frühen Neuzeit. Eine Bestandsaufnahme.’ In: Thomas Nicklas & Matthias Schnettger (eds), Politik und Sprache im frühneuzeitlichen Europa. Mainz: von Zabern, 33–50.

Of ficial Languages

143

Beales, Derek. 1987–2009. Joseph II. Vol. I: In the Shadow of Maria Theresa, 1741–80. Vol. II: Against the World, 1780–90. Cambridge: Cambridge University Press. Bell, David A. 1995. ‘Lingua Populi, Lingua Dei: Language, Religion and the Origins of  French Revolutionary Nationalism.’ American Historical Review 100, 1403–1437. Bell, David A. 2001. The Cult of the Nation in France: Inventing Nationalism, 1680–1800. Cambridge, MA: Harvard University Press. Bernatzik, Edmund. 1906. Die österreichischen Verfassungsgesetze. Leipzig : Hirschfeld. Bowen, Ivor. 1908. (ed.). The Statutes of  Wales. London. Brunot, Ferdinand. 1906. Histoire de la langue française des origines à 1900. Vol. II: Le seizième siècle. Paris: Armand Colin. Budinszky, Alexander. 1881. Die Ausbreitung der lateinischen Sprache über Italien und die Provinzen des römischen Reiches. Berlin: Wilhelm Hertz. Burke, Peter. 2004. Languages and Communities in Early Modern Europe. Cambridge: Cambridge University Press. Catto, Jeremy. 2003. ‘Written English: The Making of  the Language, 1370–1400’. Past & Present 179, 24–59. de Certeau, Michel et al. 1975. Une politique de la langue: la Révolution française et les patois. L’enquête de Grégoire. Paris: Gallimard. Chaurand, Jacques (ed.). 1999. Nouvelle histoire de la langue française. Paris: Éditions du Seuil. Crowley, Tony (ed.) 2000. The Politics of  Language in Ireland, 1366–1922. A Sourcebook. London: Routledge. Evans, Robert J.W. 2004. ‘Language and State-building: The Case of  the Habsburg Monarchy.’ Austrian History Yearbook 35, 1–24. Evans, Robert J.W. 2006. ‘The Politics of Language in Europe, c. 1525–1697.’ Przegląd Historyczny 97, 455–476. Evans, Robert J.W. 2007. ‘The Politics of  Language and the Languages of  Politics: Latin and the Vernaculars in Eighteenth-Century Hungary.’ In: Hamish Scott & Brendan Simms (eds), Cultures of Power in Europe during the Long Eighteenth Century. Cambridge: Cambridge University Press, 200–224. Evans, Robert J.W. 2011. ‘Language and Politics: Bohemia in International Context, 1409–1627.’ In: Eva Doležalová and Jaroslav Pánek (eds), Confession and Nation in the Era of  Reformations. Central Europe in Comparative Perspective. Prague: Institute of  History, 155–182. Febvre, Lucien. 1924. ‘Politique royale ou civilisation française? Remarques sur un problème d’histoire linguistique.’ Revue de Synthèse Historique 38, 37–53.

144

Robert Evans

Ferrer i Gironès, Francesc. 1985. La persecució política de la llengua catalana. Barcelona: Edicions 62. Fioretti, P. 1950. ‘L’ordonnance de Villers-Cotterêts.’ Le Français Moderne 18, 277–288. Fischel, Alfred (ed.). 1910. Das österreichische Sprachenrecht: Eine Quellensammlung. Brünn: Irrgang. Haarmann, Harald. 1975. Soziologie und Politik der Sprachen Europas. Munich: dtv. Hattenhauer, Hans. 1987. Zur Geschichte der deutschen Rechts- und Gesetzessprache. Göttingen: Vandenhoeck & Ruprecht. Hogan, Jeremiah J. 1927. The English Language in Ireland. Dublin: The Educational Company of  Ireland. Jenkins, Geraint H (ed.). 1997. Y Gymraeg yn ei Disgleirdeb: Yr Iaith Gymraeg cyn y Chwyldro Diwydiannol [The Welsh Language before the Industrial Revolution]. Cardif f : Gwasg Prifysgol Cymru. Kemény, Gábor G. (ed.). 1952–1985. I����������������������������������������������� ratok a nemzetiségi kérdés történetéhez Magyarországon a dualizmus korában. 7 vols. Budapest: Tankönyvkiadó. Lexikon der romanistischen Linguistik. 1988–2005. ed. by Günter Holtus et al. 8 vols. Tübingen: Niemeyer. Kučera, Jaroslav. 1999. Minderheit im Nationalstaat: die Sprachenfrage in den tschechisch–deutschen Beziehungen, 1918–38. Munich: Oldenbourg. Morgan, Prys. 1988. Beibl i Gymru. Gwasg Cambria. Neumann, Günter, & Jürgen Untermann. 1980. Die Sprachen im römischen Reich der Kaiserzeit. Cologne: Rheinland Verlag. Rapant, Daniel. 1927–1931. K počiatkom maďarizácie. 2 vols. Bratislava: Filozofická fakulta Univerzity Komenského. Renzi, Lorenzo. 1981. La politica linguistica della Rivoluzione francese: studio sulle origini e la natura del giacobinismo linguistico. Naples: Ligouri. Richter, Michael. 1995. Studies in Medieval Language and Culture. Blackrock/Dublin: Four Courts Press. Rockel, Martin. 1989. Grundzüge einer Geschichte der irischen Sprache. Vienna: Österreichische Akademie der Wissenschaften. Schlett, István (ed.). 2002. A nemzetiségi törvényjavaslat országgyűlési vitája, 1868 Budapest: Kortárs Kiado. Schmitt, Christian. 2000. ‘Nation und Sprache: das Französische’. In: Andreas Gardt (ed.), Nation und Sprache. Die Diskussion ihres Verhältnisses in Geschichte und Gegenwart. Berlin: de Gruyter. 673–745. Skála ze Zhoře, Pavel. 1865–1870. Historie česká, od r. 1602 do r. 1623, ed. Karel Tieftrunk. 5 vols. Prague: Nakl. Kněhkupectví I.L. Kober. Smith, Llinos Beverley. 1997. ‘Yr Iaith Gymraeg cyn 1536.’ In: Jenkins, 15–44.

Of ficial Languages

145

Stourzh, Gerald. 1985. Die Gleichberechtigung der Nationalitäten in der Verfassung und Verwaltung Österreichs, 1848–1918. Vienna: Österreichische Akademie der Wissenschaften. Szekfű, Gyula (ed.). 1926. Iratok a magyar államnyelv kérdésének történetéhez, 1790– 1848. Budapest: Magyar Történelmi Társulat. Trudeau, Danielle. 1983. ‘L’ordonnance de Villers-Cotterêts et la langue française: histoire ou interpretation.’ Bibliothèque d’Humanisme et Renaissance 45, 461–472. Waquet, Françoise. 1998. Le latin ou l’empire d’un signe, XVIe–XXe siècle. Paris: Albin Michel. Weber, Eugen. 1976. Peasants into Frenchmen: The Modernization of  Rural France, 1870–1914. Stanford: Stanford University Press. Williams, Glanmor. 1967. Welsh Reformation Essays. Cardif f : University of  Wales. Williams, Glanmor. 1987. Recovery, Reorientation and Reformation: Wales, c.1415–1642. Oxford: Oxford University Press.

Tomasz Kamusella

Classifying the Slavic Languages, or the Politics of  Classification1

Abstract At present it is often accepted that the Slavic languages are divided into three subfamilies: Western, Eastern and Southern. However, up to the turn of  the twentieth century other classifications had prevailed before this tripartite division became established as the standard model. The taxon of subfamily in the so-called ‘genetic’ classification of  languages is generally applied to groups of interrelated languages that are spoken in a contiguous area (dialect continuum) and that are geographically isolated from other ‘kin’ subfamilies by non-related languages. This, however, is not true of  the Western and Eastern Slavic languages, which are separated from one another not on any areal or linguistic basis, but for political-cum-cultural reasons (such as distinctions among populations professing Catholicism or Orthodox Christianity, use of  the Latin script versus Cyrillic, being part of  the Russian Empire or not). Linguists tend to support this genetic classification (perhaps under the inf luence of political considerations), while other social scientists accept the universal applicability of this standard genetic model because of a mistaken belief  that it is grounded in the findings of  linguistic research.

1

The text gained much from corrections and suggestions of  the two anonymous readers, whom I thank for their help. Another word of  thanks goes to Michael O’Gorman for reading a version of  the text. Obviously, I am responsible for any remaining infelicities.

148

Tomasz Kamusella

1 Introduction In this article I analyse the emergence and configuration of various classificatory schemes of  the Slavic languages, which were devised mainly during the nineteenth century. Next, I trace the way in which one of these schemes, the triple one, became the current standard. I propose that the standard’s distinction between the Western and Eastern Slavic languages is based on extralinguistic factors. Both groups of these languages gradually shade from one into another within the North Slavic dialect continuum, meaning that from the linguistic perspective, they could be usefully classified as a single, Northern, branch of  the Slavic languages. In premodern and early modern Europe legitimizing histories and myths which gave larger, usually politically significant, groups of people rights to their status, to their territory or to other privileges, were grounded in religion or in the mythical past, usually connected to the Roman Empire, heroic Antiquity or the Bible. The search for the lost tribes of  Israel or declaring this or that group descended from Brutus or another Roman aristocrat were accepted instruments of political and scholarly discourse. A change came with the development of  historiography as a discipline which should formulate its theses on the basis of facts, substantiated most often by written records. This rationalistic rigor within the domain of  historiography has not led to the relegation of politically favored myths and legends within public discourse altogether to the realm of  fairy tales. Politically motivated myths pertaining to the past still occupy a large space in the collective imaginations of nations; most notably, they are propagated through schools to future citizens. In the case of  Poland, the sixteenth-century myth of  the separate ethnic, Sarmatian, origins of the nobility in the Kingdom of Poland is no longer seriously advanced. This wholly unsubstantiated myth may have been shed because it clashed with the newer political needs of  the nationalists at the turn of the twentieth century who wished to fashion a Polish nation from all those perceived as the speakers of  the Polish language, extending to include the peasantry. The nobles had tended to perceive their serfs as dif ferent not only in status, but also in origin, by claiming that peasants

Classifying the Slavic Languages, or the Politics of  Classification

149

descended from the Biblical figure Ham, or in a more rational vein, that they were of  Slavic origin, and thus dif ferent from and more vulgar than the nobles who claimed descent from the glories of  Sarmatia. The origins of  the Poles, the Czechs and the Russians are popularly explained in a late thirteenth-century tale of  three brothers, Lech, Czech and Rus, eponyms of the three nations. The legend achieved canonical status in Poland in the late nineteenth century and is still told in Poland to pupils in their early years at elementary school. The tale does not mention any further brothers, representing the Belarusians or the Ukrainians, because when it coalesced, the ethnonym Rus (or rather Rus’), or Ruthenians in English, referred to all of the Slavophone Orthodox population, from which the present-day nations of the Belarusians, the Russians and the Ukrainians later emerged. This story about the three brothers purportedly explains, on the one hand, the common, Slavic origin of  the three nations, and on the other, how they became dif ferentiated into three separate nations. The tale’s focus on ethnolinguistic closeness and dif ference is a ref lection of  the rise of  Pan-Slavism and ethnolinguistic Slavic nationalisms as increasingly potent political ideologies in the second half of the nineteenth century. These were preoccupations of  the times, which gradually erased previously dominant ideas of the divine legitimization of temporal power. In Central Europe this was replaced with ethnolinguistic nationalism, and the entailed idea of  the ethnolinguistically homogenous (or ‘pure’ in nationalist parlance) nation-state. Ethnolinguistically based nationalism could be presented to the population at large as ‘modern’ or ‘progressive,’ not only because it allowed for changing the old order, but also because beginning in the nineteenth century, European scholars made the ‘genetic’ relations between languages an accepted component of scholarly discourse, which later philologists, often without much substantiation, claimed as ‘scientific’. In 1770 János Sajnovics (1733–1785) noticed a genetic link between Hungarian and Saami (Lapp), and sixteen years later was followed by William Jones (1746–1794) who demonstrated the existence of a similar link connecting Sanskrit to Latin. This led to the rise of  historical linguistics, which became a significant tool for both legitimizing and disputing the ethnolinguistic claims of national and pan-national movements. Moreover, Johann Gottfried

150

Tomasz Kamusella

Herder’s (1744–1803) opinions on language, languages, cultures and ethnic groups, culminating in Ideen zur Philosophie der Geschichte der Menschheit (1784–1791, Outlines of a Philosophy of the History of Man, 1800) presented the national movements of  the nineteenth century with a ready-made corpus of apparently incontrovertible ideas and ‘facts’ for furthering and legitimizing their pet political projects. From that moment on, especially in Central Europe in the wake of  the Napoleonic Wars, it was no longer suf ficient for a political project to rely on the long-lasting existence of a given state. The state’s population, in order to qualify as a nation, had to enjoy a ‘scientifically proven’ common ethnolinguistic origin and ideally this should now be bolstered by the adoption of a single standard form of a common national language. Ernest Gellner (1925–1995) memorably defined this ethnolinguistic variety of nationalism as the drive toward attaining congruence between the cultural unit and the political unit, that is, that an ethnolinguistic nation should be housed in its own nation-state, not shared by any other nation (Gellner 1983: 1). This tendency delegitimized the plethora of polities on the territory of  the former Holy Roman Empire and in the Apennine Peninsula; these polities were perceived as being inhabited by German- and Italian-speakers, respectively, and this led to the founding of the Kingdom of Italy (1861) and of the German Empire (1871) as ethnolinguistic nationstates-in-making. Furthermore, the non-ethnolinguistic Austrian Empire, in an ef fort to arrest the fraying of its statehood legitimization which was being increasingly undermined by the rise of ethnolinguistic nationalism, surrendered strategically to this new ideology by transforming itself into Austria-Hungary and facilitating the emergence of variously defined and organized ethnolinguistic autonomies at the sub-state (regional) level.

2 Toward a science of  language Philologists in the second half of the nineteenth century aspired to elevate their preoccupation with texts, grammars, translations and classifications to the status of a science of the same rank of exactness as the natural sciences.

Classifying the Slavic Languages, or the Politics of  Classification

151

The methods of  historical linguists were speculative and only narrowly applicable. An unexpected source of prestigious hard scientific support for their ef forts came from contemporary biology. In chapter fourteen of his On the Origin of Species (1859), Charles Darwin (1809–1882) proposed that his concept of the ‘tree of life,’ intended to illustrate and explicate the process of speciation, could be employed for explaining the rise and dif ferentiation of  languages. Darwin suggested that ‘[i]f we possessed a perfect pedigree of mankind, a genealogical arrangement of the races of man would af ford the best classification of the various languages now spoken throughout the world; and if all extinct languages, and all intermediate and slowly changing dialects, were to be included, such an arrangement would be the only possible one’ (cited in Cavalli-Sforza 1994: 381–382). Having been introduced to Darwin’s theory by his friend, the Prussian naturalist Ernst Haeckel (1834–1919), the linguist August Schleicher (1821–1868) from Saxe-Meiningen wrote Die Darwinsche Theorie und die Sprachwissenschaft (1863, Darwinism Tested by the Science of  Language, 1869). Schleicher made the image of language as an ‘organism’ into one of  the most lasting metaphors in the popular thinking on language. Therefore, most of us now tend unref lectively to believe that languages are ‘born’, ‘grow’ and ‘die’ as though they were species of animals or plants, not phenomena solely dependent on the species of Homo sapiens sapiens. Schleicher also borrowed the Linnaean system of biological classification and used it for classifying these ‘organisms of  languages’ (Sprachorganismen) (Maher 1983; Richards 2004: 13; Schleicher 1869: 13). Schleicher developed Darwin’s tentative and diagram-like tree of life concept (with no actual data on the tree), into a full-f ledged ‘family tree’ (Stammbaum) of  the Indo-European languages. The British theological writer Frederic William Farrar (1831–1903) extended this device and the biologizing approach to language classification in his Families of  Speech (1870). Haeckel in numerous inf luential works adopted as a concrete and uncontested description of reality this initially only heuristically intended device, frequently drawn in the form of a literal tree to classify plants, animals and ‘species [that is, races] of man’ (Alter 1999: 75, 92, 112–121). In this manner, in the twentieth century the image of  ‘family tree’ entered textbooks and the popular imagination as the main tool for classifying, understanding and ordering various social and historical phenomena in

152

Tomasz Kamusella

time and space. This, however, presupposes both the discrete nature of such phenomena and their causal linking in accordance with the linear model of  time. Despite the widespread dissemination of the idea of the genealogical tree, linguists realize that languages cannot be the unambiguously delineated entities of popular belief. Languages are the creation of human beings using them for communication. Languages interact, appear, disappear and shade into one another, ref lecting the spatial and social mobility of human groups and their interactions with one another. The Thuringian pioneer of creole linguistics, Hugo Schuchardt (1842–1927), in his Der Vokalismus des Vulgärlateins (1868, Vocalism of  Vulgar Latin) and the Prussian linguist, Johannes Schmidt (1843–1901), in his Die Verwandtschaftsverhältnsse der indogermanischen Sprachen (1872, The Relationships among the Indo-European Languages) independently proposed the metaphor of  the ‘wave’ (Welle) for classifying languages. This was the beginning of areal linguistics, which researches mutual inf luences between languages spoken side by side in neighboring areas or within a single area. This approach to classification allowed for the grouping of languages into non-genealogical, variously named, ‘linguistic areas,’ ‘linguistic leagues,’ or ‘linguistic unions’ (Sprachbünde). But the fact that terminology for referring to these kinds of linguistic relationships still has not stabilized in English indicates that it remains a specialist field, whose insights have not yet extended into popular knowledge (Busmann 1996: 519; Décsy 1973). The image of  the ‘genealogical tree’ is more attractive to the average language user, due to its anthropomorphic quality (speaking of ‘parent’ and ‘of fspring’ languages), and the prevalent treatment of languages as discrete entities that parallels popular thinking on nations and states as similarly discrete entities. Nationalism became the first globally accepted ideology of statehood legitimization in the wake of decolonization and the breakup into nation-states of  the Soviet Union, the last large and emphatically non-national polity (Kohn 1962). Thus, at the beginning of  the twentyfirst century we tend to believe that humankind is ‘naturally’ divided into nations which should be housed separately in their own nation-states. In Central and Eastern Europe where ethnolinguistic nationalism is dominant,

Classifying the Slavic Languages, or the Politics of  Classification

153

language has been seen as the defining basis of  the nation. This type of nationalism defines the nation as all speakers of a language. This ideological principle reinforces even more the fallacious belief in the discrete character of languages, to the studious disregard of dialects, of multilingualism and of  the fact that languages continuously change through time and interact with other languages, following their speakers’ social, economic, political and cultural choices and interactions (Kamusella 2006). Not surprisingly, the classification of  languages into linguistic areas, which transcends and blurs political boundaries, and which lets one see one’s speech in various degrees of closeness to the standard of one’s national language as well as to national languages in the adjacent nation-states, even if  they are not ‘genetically’ related to one’s language, is an anathema to the national vision of  the world. Hence Schleicher’s admonition that ‘[w]hat some call a language, others term a dialect, and vice versa’ (1869: 46) is conveniently forgotten. In our national world of  today, there is a preference for Leonard Bloomfield’s postulate that dialects are mutually comprehensible while languages are not (1926: 162). The mutual comprehensibility of such languages as Bosnian, Croatian, Montenegrin or Serbian is conveniently overlooked, as, for similar reasons of political interest, is the mutual incomprehensibility of  the dialects of  Chinese. Sometimes even professional linguists (especially in Central and Eastern Europe, where their research is expected to uphold and reconfirm the enshrined tenets of ethnolinguistic nationalism) brush aside such discrepancies between the ideologically determined stance and the evident factual position. They fail to or pretend not to see themselves as part of a political game of nationalism and power struggle to which they contribute with their (ostensibly scientific and value-free) pronouncements and theses on language.

154

Tomasz Kamusella

3 Slavic languages: Toward the tyranny of genealogical classification 3.1 A dual classification The study of Slavic languages as a branch of philology emerged at the turn of the nineteenth century, most obviously represented by the works of the Bohemian (Czech) scholar Josef Dobrovský (or, as he himself spelled it in his signature: Dobrowsky, 1753–1829). He wrote in German but spoke Czech and knew other Slavic languages. Between 1806 and 1822 he developed a dual classification of the Slavic languages. The first group comprised Russian, Old Church Slavonic, Illyrian (used in Bulgaria, Serbia, Bosnia and Dalmatia), Croatian and Windisch (Slovenian). In the second group Dobrovský included: Slovak, Czech, Upper Lusatian (Upper Sorbian), Lower Lusatian (Lower Sorbian) and Polish (Ottův slovník naučný 1901: 448). Table 2  Dual classification of  Slavic Group 1

Group 2

Russian Old Church Slavonic Illyrian (used in Bulgaria, Serbia, Bosnia and Dalmatia) Croatian Windish (Slovenian)

Slovak Czech Upper Lusatian (Upper Sorbian) Lower Lusatian (Lower Sorbian) Polish

He may have distinguished the former group because the Orthodox, Roman Catholic, Greek Catholic (Uniate), and Muslim speakers of the aforementioned vernaculars either used non-Slavic languages or Church Slavonic for of ficial and written purposes. The standardization of Russian was then already under way,2 but the political will to impose it throughout the Rus-

2

Russian as a standard language began to emerge in the second half of the eighteenth century from the mixture of the Moscow dialect and Church Slavonic, or the ‘middle style’ as proposed by Mikhail Lomonosov (1711–1765) in 1755.

Classifying the Slavic Languages, or the Politics of  Classification

155

sian Empire did not appear until after the Polish-Lithuanian nobility’s failed uprising against the Tsar (1830–1831). Russian then replaced Polish as the of ficial language in those Polish-Lithuanian provinces that had been directly incorporated into the empire at the close of the eighteenth century. Furthermore, apart from Russian, all the vernaculars from Dobrovský’s first group were believed to have derived from Old Church Slavonic, while Russian and the languages spoken by Orthodox Slavs functioned in a diglossic symbiosis with Church Slavonic. Dobrovský’s second group of  Slavic languages included those which were spoken exclusively by either Roman Catholics or Protestants, and which had been used in writing since the sixteenth century (with the partial exception of  Slovak, which became more generally perceived as separate from Czech only in the second half of  the nineteenth century). None of  these languages originated from Church Slavonic or was symbiotically connected to it. When Dobrovský put forward his classification, of  the languages in Group 2, only Polish was used in any of ficial capacity, though merely at a substate level. 3.2 A triple classification The ethnically Slovenian linguist in Austrian service, Bartholomäus ( Jernej) Kopitar (1780–1844), and the Russian Slavicist, Aleksandr Vostokov (1781– 1864), disagreed with Dobrovský’s dual classification. Today’s triple classification of the Slavic languages dates back to Vostokov’s work Rassuzhdene o tserkevno-slavianskom iazyke (1820, An Essay on the Church Slavonic Language). He insisted that Russian did not belong to either of Dobrovský’s two groups of  the Slavic languages, but that it constituted a transitional (though in its own right distinctive) bridge between the Western and South-Eastern Slavic languages. He also argued that Russian was closer to the South-Eastern Slavic languages than to the Western Slavic languages. It is likely that the tradition of Church Slavonic literacy and the use of Cyrillic common to the South-Eastern Slavs and to the Russians convinced him of  the existence of a special closeness between these two groups of Slavs and their languages (Ottův slovník naučný 1901: 448): Western Slavic languages – Russian [Eastern Slavic lgs?] – South-Eastern Slavic languages

156

Tomasz Kamusella

3.3 An elaborated dual classification The Czech philologist of  Slovak origin, Pavel Josef  Šafařík (1795–1861, spelled Šáfarik in modern Slovak), in his opus magnum, Slowanské starožitnosti (1837, Slavic Antiquities), supported Dobrovský’s classification. In his scheme Russian, Bulgarian and Illyrian belonged to the SouthEastern branch of the Slavic languages. The Illyrian language was composed from Serbian, Croatian and Slovenian construed as dialects. Lekhian, Czechoslovak and Polabian formed the Western branch. ‘Lekhian’ was the Ruthenian-language3 ethnonym employed for referring to the Poles. It survives to this day in Lithuanian (Lankas), Magyar (Lengyel) and colloquial Ukrainian (Lakh). In the late Middle Ages the Latinized form Lechitae functioned as a synonym for the Poles, and this term was also used in Poland-Lithuania. Through Latin it entered the Polish language as Lechici as a poetic term for the Poles. According to Šafařík, the Lekhian language, or rather sub-branch, included the dialects of Polish, Silesian and Pomeranian. The Czechoslovak sub-branch was composed of the dialects of Czech, Moravian and Slovak. The Polabian sub-branch embraced Sorbian and those Slavic dialects west of the River Oder which had become extinct prior to the nineteenth century. Šafařík’s classification doubled as that of  the Slavic peoples (ethnic groups or nations) in line with the ethnolinguistic equation of  language with nation and state already adopted by German nationalists in the course of the Napoleonic Wars at the beginning of the nineteenth century (Šafařík 1837: 483).

3

Ruthenian (ruski) was a Cyrillic-based vernacular Slavic language used in an of ficial capacity in the Grand Duchy of  Lithuania with its two chancery centers at Vil’na (Vilnius) and Kyiv. It was replaced with Polish in 1699. Today, Belarusian linguists claim the language of  the Vil’na Ruthenian languages as ‘Old Belarusian’ and their Ukrainian counterparts see Ruthenian-language documents produced at Kyiv as written in Old Ukrainian (Anichenka 1969).

157

Classifying the Slavic Languages, or the Politics of  Classification Table 3  Elaborated dual classification of  Slavic 1. South-Eastern Slavic languages Russian

Bulgarian Illyrian – Serbian – Croatian – Slovenian

2. Western Slavic languages Lekhian

Czechoslovak

Polabian

– Polish – Silesian – Pomeranian [Kashubian]

– Czech – Moravian – Slovak

– Sorbian – Polabian (extinct)

3.4 Another triple classification and its variations Frantíšek Palacký (Franz Palacky, 1798–1876) initiated Czech national historiography, writing first in German, and after 1848 also in Czech. Later he became the first leader of the Czech national movement. When still known as Franz Palacky, in 1836 he published the first volume of his Geschichte von Böhmen (History of Bohemia). In this work he divided the Slavs and their languages into three groups. In the first he included Russian and Bulgarian seen as the direct descendents of  Old Church Slavonic – the first written Slavic language. The second, or South-Western, group included Serbian, Croatian and Carinthian (Slovenian). In the third, or North-Western, group Palacký placed Lekhian, which could be interpreted as being composed from Czech, Polish, Slovak and Sorbian. Table 4  Revised triple classification of  Slavic I 1. [Eastern]

2. South-Western

3. North-Western

Old Church Slavonic

Serbian Croatian Carinthian (Slovenian)

Lekhian

Russian Bulgarian

158

Tomasz Kamusella

At the same time Vostokov’s insistence on the special separate position of  Russian among the Slavic languages became popular in Russia. The Russian linguist Mikhail Maksimovich (1804–1873) in his Istoriia drevnei russkoi slovesnosti (1839, History of Early Russian Literature) and Nachatki russkoi filologii (1848, An Introduction to Russian Philology) proposed that Russian constituted a separate branch of the Slavic languages in its own right. The two other branches included the South-Western and North-Western Slavic languages, as in Palacký’s classification. Interestingly, Maksimovich distinguished two sub-branches in the Russian group: the South Russian consisting of Little Russian (or today’s Ukrainian), and the North-Eastern consisting of Great Russian (Russian) and White Russian (Belarusian). At that time Russia’s imperial political ideology was extended into the concept of the Great Russian language that, purportedly, stemmed directly from Old Church Slavonic and comprised Little, or Southern Russian (Ukrainian) and White, or North-Eastern Russian (Belarusian) as its dialects. Table 5  Revised triple classification of  Slavic II 1. Russian [Eastern Slavic languages?] South Russian Little Russian (Ukrainian)

2. South-Western North-Eastern Slavic languages Great Russian (Russian) White Russian (Belarusian)

3. North-Western Slavic languages

In 1843 and 1845 the Russian philologist Izmail Ivanovich Sreznevskii (1812–1880) criticized Šafařík’s dual classification of  the Slavic languages, and gave his own elaborated version of the triple division. In Sreznevskii’s scheme the Eastern group comprised Great Russian and Little Russian. The South-Western included Church Slavonic, Bulgarian, Serbo-Croatian (or Serbian and Croatian as two literary varieties of a single language), Carinthian (Slovenian). In the North-Western group he placed Polish, Polabian, Lusatian (Sorbian), Czech and Slovak (Ottův slovník naučný 1901: 449).

Classifying the Slavic Languages, or the Politics of  Classification

159

Table 6  Revised triple classification of  Slavic III 1. Eastern group

2. South-Western group

3. North-Western group

Great Russian Little Russian

Church Slavonic Bulgarian Serbo-Croatian (or Serbian and Croatian) Carinthian (Slovenian)

Polish Polabian Lusatian (Sorbian) Czech Slovak

3.5 Schleicher: A dual classification with a twist In 1858 August Schleicher forcefully entered the ongoing discussion on the classification of  the Slavic languages (or dialects) with his article ‘Kurzer Abriß der Geschichte der slawischen Sprache’ (A Brief  Outline of  the History of  the Slavic Language). He criticized the then-emerging triple division of  the Slavic languages and argued for the dual one, as proposed by Dobrovský. A qualitative leap occurred in 1863. In his work Die Darwinsche Theorie und die Sprachwissenschaft, Schleicher presented the genealogical tree of  the Indo-Germanic (Indo-European) languages. To the typical ranking of  languages governed by perceived closeness among them, Schleicher added the dimension of  time. The idea of a language as self-contained entity, unambiguously separate from others of the same kind, was projected into the past. The result was a genealogical tree of rapidly branching out languages. In this scheme, at a single moment, from previously common (parent) languages perfectly separate (dif ferent) of fspring languages split in an inexplicable fashion. Darwin treated his highly schematic ‘trees of  life’ as a mental prop for illustrating the idea of speciation that occurs in the course of evolution. In a radical departure from Darwin’s limited conceptual and illustrative use of  the ‘tree’, Schleicher appended meticulous labels to the branches and twigs of  his genealogical tree of  the IndoGermanic (that is, Indo-European) languages, and of fered it as an exact picture and explanation of  how these languages ‘evolved’ through time. According to him, the pedigree of  the Slavic languages commenced with

160

Tomasz Kamusella

the splitting of the Indo-Germanic ur-language into the Slavo-German and Aryan-Greek-Italic-Celtic stock (basis) languages, or Grundsprache. Next the former branched out into the German and Slavo-Latvian (LithuanoSlavic) stock languages. From the Slavo-Latvian, the Baltic (Lithuanian) and Slavic stock languages emerged. Beyond this point Schleicher applied his ‘genetic’ (diachronic) approach to the classification of  the Slavic languages. At first he drew on Dobrovský’s classification when he distinguished the two languages of  West Slavic and South-Eastern Slavic, which he presented as originating from the common Slavic stock language. In the next phase the West Slavic language spawned Sorbian, Polish, Czech and Polabian, the last-named language being already extinct by the nineteenth century. This branch closely corresponded to the second group of Slavic languages in Dobrovský’s classification, to the Western group in Šafařík’s classification, and to the North-Western group in Palacký’s, Maksimovich’s and Sreznevskii’s triple classifications. Schleicher decided that the South-Eastern Slavic language bifurcated into Russian and the South Slavic language. Thanks to this subdivision he merged the insights of the dual and triple classifications of the Slavic languages. Schleicher also bowed to Vostokov’s and other Russian philologists’ insistence that Russian should be accorded a special place, or, best of all, a separate classificatory category in its own right. Instead of  bestowing the same classificatory rank on the remaining Slavic languages, as the proponents of  the triple division did, Schleicher split the South Slavic language into the Serbo-Slovenian and Bulgarian languages, which amounted to the tentative introduction of a quadruple division of  the Slavic languages. Bulgarian, like Russian, constituted a group on its own, whereas Serbo-Slovenian, predictably, gave rise to Serbian and Slovenian. All the final-position languages were adorned with tiny tufts of  hair-like twigs representing the dialects of  these languages (see Figure 5).

Classifying the Slavic Languages, or the Politics of  Classification

161

Figure 5  A. Schleicher’s genealogical tree of  the Indo-Germanic (Indo-European) languages (1873: diagram after p. 33)

Schleicher’s diachronic classification scheme of the Slavic languages gave a priority to the dual division, though at a later (lower) level of classification advocated the triple division, and at an even later (lower) level proposed the quadruple division of these languages. In his 1871 work Laut- und Formenlehre der polabischen Sprache (The Sound and Morphological Structure of  the Polabian Language), Schleicher subdivided the West Slavic languages. According to him, the West Slavic language gave rise to the Lekhian and Czech languages. Then the former split into Polish and the extinct Polabian, while the latter split into Czechoslovak and Sorbian. At the lowest classificatory level of  Schleicher’s scheme it meant a quintuple division of  the Slavic languages (Alter 1999: 75; Ottův slovník naučný 1901: 449; Schleicher 1869: diagram after p. 69).

162

Tomasz Kamusella

3.6 A dual-cum-quadruple classification Schleicher’s genetic tree of  the Slavic languages inspired the Czech linguist Jan Gebauer (1838–1907) to propose his own multilevel scheme for classifying the Slavic languages in 1870. However, he rejected Schleicher’s concept of a genealogical tree with the entailed dimension of  time (linguistic evolution), and returned to the idea of a synchronic multi-level classificatory scheme. As the basic level he chose the dual division of  the Slavic languages, also preferred by Schleicher. This level consisted of  the South-Eastern and Western groups. The latter included Czech, Polish, Sorbian and extinct Polabian, which corresponded to Schleicher’s view before he subdivided this group in 1871. In the case of  the South-Eastern group, Gebauer reduced Schleicher’s further subdivisions into three subgroups of equal rank: Russian, Bulgarian and Serbocroatoslovenian.4 The first subgroup embraced the Great Russian (Russian) and Little Russian (Ukrainian) languages, alongside the White Russian (Belarusian) dialects. In the second subgroup Old Bulgarian (Old Church Slavonic) and Modern Bulgarian were placed, while in the third he placed SerboCroatian and Slovenian. Unlike Schleicher, Gebauer specified the dialects of all the aforementioned languages with the exception of the category of  the ‘White Russian dialects’, which was not subdivided into component dialects. Significantly, Ukrainian’s Carpathian dialect today can be identified with the Rusyn (Lemkian) language; Modern Bulgarian’s Macedonian dialect with the Macedonian language; Czech’s Hungarian-Slovak dialect with the Slovak language; and Polish’s Kashubian and Silesian dialects with the Kashubian language and the Silesian language, respectively. It is also 4

Gebauer’s coinage, ‘Serbocroatoslovenian,’ proved quite inf luential, due to the sustained broad personal and intellectual contacts maintained between the Czechs and the Balkan Slavs within the framework of  Austria-Hungary. Bowing to the normative idea of ethnolinguistic nationalism that a nation should speak a single language, which, in turn, would be the sole of ficial language of the nation’s nation-state, the 1921 constitution of the Kingdom of Serbs, Croats, and Slovenes (after 1929, Yugoslavia) announced Serbocroatoslovenian (later, popularly known as Yugoslavian) as the of ficial language of  this polity.

163

Classifying the Slavic Languages, or the Politics of  Classification

interesting to see that among Slovenian’s dialects, Gebauer distinguished Croato-Slovenian as a transitional dialect between the Slovenian language and Serbo-Croatian. Having skipped one level of subdivision (Russian vs South Slavic) present in Schleicher’s scheme, Gebauer did not incorporate the triple division of  the Slavic languages in his scheme, but adopted Schleicher’s quadruple division (Rieger 1870: 645). Table 7  Dual-cum-quadruple classification of  Slavic

Western Slavic languages

South-Eastern Slavic languages Russian Great Russian (Russian) Little Russian (Ukrainian) White Russian dialects (Belarussian)

Bulgarian Old Bulgarian (Old Church Slavonic) Modern Bulgarian (Bulgarian)

Serbocroatoslovenian Czech Polish Serbo-Croatian Sorbian Slovenian Polabian (extinct)

4 A Criticism of  the Classificatory Schemes The catalogue of dual, triple, quadruple and quintuple classifications of the Slavic languages, of fered by Dobrovský, Šafařík, Palacký, Maksimovich, Sreznevskii, Schleicher and Gebauer, became the basis for further variants ceaselessly worked out by Slavicists. In the late 1870s the Russian Slavicist, Aleksandr Kochubinskii (1845–1907), disputed this method of classifying the Slavic languages in his work, K voprosu o vzaimnykh otnosheniiakh slavianskikh narechii (1877–1878, On the Issue of  Mutual Relationships among the Slavic Dialects). He noticed that depending on which linguistic features (be they syntactical, morphological, phonetic, lexical, or a mixture of them) is considered the site of the defining characteristics of languages,

164

Tomasz Kamusella

an endless chain of dif fering classifications of  the Slavic languages can be generated. The dependence of the outcome of a project of linguistic research on the original choice of analytical classificatory scheme is thus evidently susceptible either to the internal personal prejudices of  the researcher, or to the external demands of political expediency. Even Kochubinskii failed to disentangle himself from the temptations which he criticized. He came to the conclusion that the proper task of Slavicists was to decide which of  the Slavic languages is the oldest. Not surprisingly, he accorded this distinction to his native Russian, arguing that it is closest to the first written Slavic language – Old Church Slavonic. It took Johannes Schmidt’s Zur Geschichte der indogermanischen Voca­ lismus (1875, On the History of  Indo-Germanic Vocalism) to undermine the validity of Schleicher’s model of a genealogical tree. First, like Kochubinskii, Schmidt argued that it does not make sense to come up with isoglosses (based on selected linguistic criteria) to separate Slavic languages from one another. This urge for establishing such isoglosses, in my view, amounts to the projection onto the linguistic reality of the perceived or striven-for clear-cut boundaries among ethnically construed nations, and of the existing political borders. But this linguistic reality is continuous not discrete, so that the Slavic languages (construed as dialect chains) shade into one another. This intuition of  Schmidt’s became the basis for the later development of the concept of dialect continuum,5 which is bounded by other, often sharply (genetically) dif ferent, dialect continua. Second, having dealt with the synchronic plane, he criticized the diachronic dimension of  the genetic tree division of  the Slavic languages. He observed that languages were not born from earlier languages that were extinguished in this process. The observable variety of  languages arose through the internal dif ferentiation of the Slavic dialect continuum 5

The concept of dialect continuum arose in the wake of  the development of creole linguistics. First, in 1914 Schuchardt spoke of  ‘language or speech continuum’ (Sprachkontinuum) and the US linguists, John E Reinecke and Aiko Tokimasa, transplanted this term to English as ‘speech continuum’ in 1934. On this basis, in 1971 another US linguist, David DeCamp, coined the actual term ‘dialect continuum’ (DeCamp 1971; Reincke and Tokimasa 1934; Schuchardt 1914).

Classifying the Slavic Languages, or the Politics of  Classification

165

due to historical and political processes such as state formation, warfare, population movements, invasions and so on. Schmidt’s and Schuchardt’s insights underlay the areal (wave, Welle) theory of representing relationships between languages. Faced with the immense success of  the graphic representations of  Schleicher’s genealogical tree (Stammbaum), Schmidt devised a circle diagram to illustrate his ideas on how languages are related to and inf luence one another. The circle was divided into triangular pieces, whose apexes met in the middle of the circle as in a pie. Each piece of such a ‘linguistic pie’ represented a language construed as a segment of a dialect continuum prior the emergence of a standard language. Applied to the Slavic languages this circular diagram placed languages side by side in the order Sorbian – Polish – Russian – Bulgarian – Serbo-Croatian – Slovenian – Czech – Sorbian (returning to the starting point) (Ottův slovník naučný 1901: 449–450). Although the wave classification theory emphasized the gradual change within the dialect continuum, Schmidt’s circle diagram partially defied this logic by introducing boundaries among the ‘dialectal languages’. The Polish linguist Jan Baudouin de Courtenay (1845–1929) in his book, Übersicht der slavischen Sprachenwelt (1884, An Overview of the World of the Slavic Languages), repeated that linguists should not aspire either to establishing or to identifying clearly delineated borders that would unambiguously separate Slavic languages from one another. Dialect territories popularly associated with standard languages are indistinguishable parts of  the Slavic dialect continuum and gradually shade from one into another. Like Schmidt, Baudoin de Courtenay also criticized Schleicher’s concept of the common (stock) languages of (Proto-)Slavic, Western Slavic and South-Eastern Slavic from which all the existing Slavic languages were supposed to have branched out. Such languages had not existed. The rise and persistence of  the concept of such ur-languages is a function of the entrenched dominance of  the genealogical tree model for visualizing the classification of species, languages and human groups (Ottův slovník naučný 1901: 450). The attraction of  the genealogical tree method of classification was both graphical and political. By presenting elaborate genealogical trees of ethnic groups, animals, plants and languages as ‘scientific’, the depiction of such images added to the scientific prestige and potential success of books

166

Tomasz Kamusella

and articles in which they were used. The potential to represent certain languages as older than others on such trees could be mobilized for political ends which were not intrinsically linguistic. One of  the main tools for justifying the continued existence of an existing state, or the establishment of a new (or renewed) state for a group of people construed as a nation was provided by ancient pedigree. Those nations and states which were successfully presented as ‘older’ acquired a ‘higher right’ to existence than their rivals perceived as ‘younger’. The Württemberg philosopher Georg W.F. Hegel (1770–1831) set this dif ferentiation in stone, terming the older states or nations as ‘nations with history’ and the younger as ‘nations without history.’ Since the 1840s this distinction had became popular equally among both nationalist and Marxian thinkers (Herod 1976). The graphic representation of  the Stammbaum perfectly suited the propaganda aim of ‘proving’ that one nation or state was older than another. This device allowed for conf lating and projecting into the distant past a contemporary language, ethnicity and statehood. Given the extraordinary political utility of  the genealogical tree method of classification, it was adopted as the dominant paradigm, and the ‘wave’ model of  language classification remained secluded as a minority voice within scholarly discourse, despite its explicatory power. There are two clear reasons for the public success of  the ‘tree’ model and the relative failure of  the ‘wave’ model. First, the proponents of the wave-based areal method of classification did not develop an appealing graphical representation for it. Second, this method of classification went against the nationalist logic of the radical normative isomorphism (tight spatial and ideological overlapping) of state, nation and language. In the age of nationalism that commenced in the nineteenth century and has lasted to this day, this condemned the wave theory to obscurity.

Classifying the Slavic Languages, or the Politics of  Classification

167

5 Today’s consensus: Triple classification In the twentieth century the family tree classification and presentation of  the Slavic languages was the conventional wisdom, and it is still so in the twenty-first century. The triple classification of these languages also became part of the scholarly norm in the interwar period, and as such it has thrived unchallenged to this day. Two recent English-language academic overviews of the Slavic languages follow this pattern with minor variations. In 1980, Alexander Schenker and Edward Stankiewicz in their The Slavic Literary Languages: Formation and Development included Bulgarian, Macedonian, Serbo-Croatian and Slovenian in the Southern branch. Belarusian, Russian and Ukrainian were placed in the Eastern branch; the Western branch included Czech, Kashubian, Polish, Slovak and Sorbian. Church Slavonic, the survey of which opened the book, was accorded a separate place outside the branches of the triple classification (Schenker and Stankiewicz 1980). Thirteen years later, the editors of The Slavonic Languages decided to add the liturgical language of Old Church Slavonic to the Southern branch and to add the extinct Polabian to the Western branch. They also gave priority to the West Slavic languages over the East Slavic ones presented at the end of  this work (Comrie and Corbett 1993). Scholars still produce variations on the genetic triple division of  the Slavic languages. For instance, in 1992, the Polish Slavicist, Ewa Siatkowska, presented one in her Rodzina języków zachodniosłowiańskich (The Family of  the Western Slavic Languages). According to her the Proto-Slavic language was the precursor of the Proto-Rus’ language, which gave rise to the Eastern group of  the Slavic languages. With time the Proto-Ruthenian (Proto-Russian) language emerged from Proto-Rus’, and, later, split into White Ruthenian (Belarusian), Great Ruthenian (Russian) and Little Ruthenian (Ukrainian). Interestingly, Siatkowska proposed that Bulgarian, Serbo-Croatian and Slovenian emerged directly from Proto-Slavic, but form a Southern group of  Slavic languages due to the geographical proximity maintained between the users of all three languages. Further, in Siatkowska’s scheme, Bulgarian gave rise to Macedonian, and Old Church

168

Tomasz Kamusella

Slavonic was made into an equidistant relative (though whether predecessor or side-branch is unclear) of  Bulgarian and Macedonian. Lastly, the languages of  the Western group (Slovak, Czech, Upper Sorbian, Lower Sorbian, Polish and the extinct Polabian) are portrayed as emerging from a common but nameless proto-language. Perhaps she feared that naming such a proto-language would necessitate giving precedence to one of  the Western Slavic languages over the others, as in the case of Proto-Russian and Great Russian to the detriment of Belarusian and Ukrainian (Siatkowska 1992: 34). Such academic variations on the theme of  the triple genetic division of  the Slavic languages are not ref lected in general textbooks or encyclopedias. The standard triple division remains the dominant model. It molds the imagination of new generations of  Slavicists and linguists, and has already become the accepted ‘truth’ among intellectuals worldwide. Through endless reiteration and reproduction in mass education and mass communication, the standard triple classification of  the Slavic languages overshadows and condemns to oblivion its own ambiguous origin and the rival classifications that had persisted into the three first decades of  the twentieth century. It seems that this triple division, having initially been advanced as a tentative consensus reached in the milieu of  Russian and Central European Slavophone philologists, is in danger of congealing into an unchallengeable dogma-like consensus. Unfortunately, this state of af fairs discourages further scholarly probing into the issue.

6 Conclusion One can paraphrase Schleicher’s opinion: ‘many glossologists [linguists] speak of the Slavonic dialects, others of the Slavonic languages’ (1869: 46) by proposing that most Slavicists speak now of  the triple division of  the Slavic languages, while in the past others advanced alternative classification models, especially dual models. Languages and dialects are largely

Classifying the Slavic Languages, or the Politics of  Classification

169

subjectively defined entities, existing in the eye of the beholder. However, although the distinction between them is less linguistic but rather imagined, it is nevertheless real enough, enshrined to this end in constitutions and legislation, regulated through of ficially approved grammars and dictionaries, and imparted to the population at large through popular education and mass media. The same is true of the genealogical tree-type classifications of genetically closely related languages. In their field work on dialects, linguists working in this paradigm determine the so-called ‘bundles of isoglosses’ that allow them to buttress one or another ethnolinguistically defined political project with an appropriate ‘linguistic boundary’. Such a boundary between languages or groups of languages lends itself, in turn, to a desired classificatory scheme. In this manner the error of anachronism is committed, as findings of dialectal research are usually transposed onto standardized national languages. These languages are a product of recent times, commonly not older than one or two centuries, while dialects are much older, being the very medium of oral (that is, natural) communication between humans. Thus, the origin of a dialect (obviously, without denying constant change, which any form of spoken language undergoes from generation to generation) can be dated to the time a stable social group employing it had come into being, be it in a village or town in the case of agricultural societies or in extended family clans, when it comes to nomads. Hence, the popular subsuming of dialects under the roof of languages (Dachsprache) is an anachronism. This tendency is widely accepted as appropriately scientific, because it agrees with the political needs of nation states, whose legitimacy is at least to some degree contingent on the idea that language equates to nation, which, in turn, equates to an ethnolinguistically defined nation-state for such a nation. (I term this underlying principle of ethnolinguistic nationalism the ‘normative isomorphism of  language, nation and nation state’.) On a broader plane, bundles of dialects ascribed in this way to standard languages are subsumed under the larger classificatory categories of  language branches. Some even believe that in the past such branches used to be singular parent (stock) languages from which contemporary (of fspring) languages emerged. However, this model leaves out of account the pervasive phenomenon of unwritten spoken dialects

170

Tomasz Kamusella

which had sustained the bulk of human communication throughout Central Europe until the rise of genuine full literacy in the first two decades after World War II, when language committed to writing began to exert an unprecedented mass inf luence, ef fectively leveling dialectal dif ferences. What are we then to make of  the established tradition that distinguishes between the branches of West and East Slavic languages as separate from each other? In the latter case, the Rus’ (‘Old Russian’) language can be proposed as the common progenitor of today’s Belarusian, Russian and Ukrainian. However in the case of the West Slavic languages, linguists are either silent on identifying a parent language of this group or indecisively put it on diagrams without naming it. First, there was in the past no polity which had contained all the territories inhabited by West Slavs. Second, proposing a clearly delineated and unambiguously named progenitor of  the West Slavic languages would mean, in ethnolinguistically determined politics, conceding to a single West Slavic nation the right to dominate the others. That is how, since the mid-nineteenth century, Russian tsars and then the Kremlin have repeatedly employed the politicized claim of identifying the Rus’ language with (Great) Russian in order to ‘prove’ that Belarusian and Ukrainian are ‘mere dialects (narechia)’ of the Russian language, and hence, Belarusians and Ukrainians are ‘ethnographic branches’ of  the (Great) Russian nation. Another problem that challenges the now already conventional separation of  the West Slavic branch of  languages from the Eastern branch is the impossibility of distinguishing on a purely linguistic basis which of  the ‘border or transitional dialects’ belong to the West Slavic or East Slavic languages. There is no geographical discontinuity in the form of a sea or a vast swath of  land inhabited by non-Slavic speakers that would separate West Slavs from East Slavs, unlike in the case of  South Slavs who are separated from both the other groups by the territories inhabited by Finno-Ugric-speakers (Hungarians) and Romance-speakers (Romanians and Moldovans). The undeniable linguistic continuity between West and East Slavic languages obliges the consideration of the question of whether the basis of  the supposed linguistic separation of  the West and East Slavic languages is extralinguistic, or, as best, perilinguistic. The extralinguistic factors may

Classifying the Slavic Languages, or the Politics of  Classification

171

be religious, historic or political. An overwhelming majority of West Slavs profess either Roman Catholicism or Protestantism, while most East Slavs profess either Orthodox Christianity or Greek (Uniate) Catholicism. Looking at the West-East Slavic divide from a political vantage, one can readily notice that the lands inhabited by East Slavic-speakers were historically included within the boundaries of  Rus’. The Russian Empire, as Muscovy renamed itself in 1721, grew from the policy of  ‘gathering the lands of  Rus’. Not surprisingly as a result, the perceived linguistic boundary between West and East Slavs largely happens to coincide with nineteenth-century Russia’s western frontier. The minor exceptions to the rule were St Petersburg’s historically non-Rus’an autonomous Congress Kingdom of Poland, and Vienna’s historically Rus’an eastern half of  Galicia. This discrepancy disappeared after the westward expansion of  the Soviet Union in the wake of  World War II. Then all the historically Rus’an lands, thus seen as East Slavic, found themselves in one polity, and today are shared between Belarus, Russia and Ukraine. The perilinguistic division between West and East Slavs is that between the Latin and Cyrillic alphabets.6 All West Slavic languages are written in Latin characters, while East Slavic ones are written in Cyrillic. This ref lects an old religious division, as in Central Europe Western Christianity (Roman Catholicism and later Protestantism) was liturgically and culturally connected to the Latin language with its specific script, while Slavic Eastern Christianity (Orthodoxy, and later Greek Catholicism or Uniatism) had and retains a similar relationship with Cyrillic-based Church Slavonic. However, the now apparently clear-cut writing system division between East and West Slavs is of recent making. Until the 1850s some Little Ruthenian (Ukrainian) publications appeared in Latin characters in Galicia, while Belarusian-language book production was bi-scriptural, in the Latin script and Cyrillic, until after World War I.

6

I refer to writing as ‘perilinguistic,’ due to its close relationship with language. Technically this technology is as ‘extralinguistic’ as political fiat, but scripts map languages at various linguistic levels, and have been in symbiotic relationships with some of  them for five millennia.

172

Tomasz Kamusella

Bearing these facts in mind, it is clear that the division between West and East Slavs is religious, historic and political in its nature, not linguistic. Unfortunately, this division, extralinguistic and perilinguistic in its nature, doubles as the linguistic division between the Eastern and Western branches of  the Slavic languages. I am sure it would be equally beneficial for linguistics and politics to decouple, or at least demythologize, this spurious complex. Then linguistics would not be demeaned as a ‘scientific argument’ for propping up political projects, and, conversely, unspoken political axioms of national correctness would not have to be observed in pursuing linguistic research. But one can ask if all that means that there is no possibility of classifying East and West Slavic languages on any linguistic ground. That would be going one step too far. A solution to this dilemma is to cease to be dazzled by the deceptive allure of  the genealogical tree model of classification and instead to introduce into the standard model some elements of  the areal classification model of dialect continuum. Should we agree to adopt this proposal, it emerges clearly that there are two Slavic dialect continua – Northern and Southern (Chambers and Trudgill 1998: 6; Crystal 1987: 25).

References Alter, Stephan G. 1999. Darwinism and the Linguistic Image: Language, Race, and Natural Theology in the Nineteenth Century. Baltimore: The Johns Hopkins University Press. Anichenka, U.V. 1969. Belaruska-ukrainskiia pis’mova-mounyia suviazi. Minsk: Navuka i tekhnika. Bloomfield, Leonard. 1926. ‘A Set of  Postulates for the Science of  Language’. Language, 153–164. Busmann, Hadumod, ed. 1996. Routledge Dictionary of  Language and Linguistics. London: Routledge. Cavalli-Sforza, L. Luca, Menozzi, Paolo, & Piazza, Alberto. 1994. The History and Geography of  Human Genes. Princeton, NJ: Princeton University Press.

Classifying the Slavic Languages, or the Politics of  Classification

173

Chambers’s Encyclopaedia: A Dictionary of Universal Knowledge (vol 9). 1908. London: William & Robert Chambers and Philadelphia: J.B. Lippincott. Chambers, J.K., & Trudgill, Peter. 1998. Dialectology. Cambridge: Cambridge University Press. Comrie, Bernard, & Corbett, Greville (eds). 1993. The Slavonic Languages. London: Routledge. Crystal, David. 1987. The Cambridge Encyclopedia of  Language. Cambridge: Cambridge University Press. DeCamp, David. 1971. ‘Toward a Generative Analysis of a Post-Creole Speech Continuum.’ In: Hymes, Dell (ed.), Pidginization and Creolization of  Language. Cambridge: Cambridge University Press, 349–370. Décsy, Gyula. 1973. Die linguistische Struktur Europas. Vergangenheit – Gegenwart – Zukunft. Wiesbaden: Otto Harrassowitz. Gellner, Ernest. 1983. Nations and Nationalism. Oxford: Blackwell. Haardt, W., & Gustawicz, B. 1909. Kozenna Atlas szkolny. Vienna: Edward Hölzl Geographisches Institut. Herod, Charles C. 1976. The Nation in the History of  Marxian Thought: The Concept of  Nations with History and Nations without History. The Hague: Martinus Nijhof f. Kamusella, Tomasz. 2006. ‘The Isomorphism of  Language, Nation, and State: The Case of  Central Europe.’ In: Burszta, W. et al. (eds), Nationalisms Across the Globe: An Overview of  Nationalisms of  State-Endowed and Stateless Nations (Vol 2: The World). Poznań: Wyższa Szkoła Nauk Humanistycznych i Dziennikarstwa. 57–92. Kochubinskii, Aleksandr. 1877–1878. K voprosu o vzaimnykh otnosheniiakh slavianskikh narechii. Odessa: Ulrikh i Shults. Kohn, Hans. 1962. The Age of Nationalism: The First Era of Global History. New York: Harper & Row. Lehr-Spławiński, Tadeusz (ed.). 1949. Chrestomatia słowiańska (Vol I: Teksty południowo-słowiańskie, Vol II: Teksty zachodnio-słowiańskie, Vol III: Teksty wschodnio-słowiańskie). Kraków: Wydawnictwo Studium Słowiańskiego Uniwersytetu Jagiellońskiego and Ossolineum. Maher, J. Peter, & Koerner, Konrad (eds). 1983. Linguistics and Evolutionary Theory: Three Essays by August Schleicher, Ernst Haeckel, and Wilhelm Bleek. Amsterdam: John Benjamins. Meyers Konversations-Lexikon 1889. Vol. 14. Leipzig: Verlag des Bibliographischen Instituts. Ottův slovník naučný. 1901. Vol. 22. Praha: J. Otto.

174

Tomasz Kamusella

Reinecke, John E., & Aiko Tokimasa. 1934. ‘The English Dialect of Hawaii.’ American Speech, 9, 48–58, 122–130. Richards, Robert J. 2004. The Linguistic Creation of Man: Charles Darwin, August Schleicher, Ernst Haeckel, and the Missing Link in Nineteenth-Century Evolution Theory. accessed 2 April 2009. Rieger, František. 1870. Slovník naučný (vol 8). Praha: I.L. Kober. Šafařík, Pavel Josef. 1837. Slowanské starožitnosti, Praha: Tiskem Jana Spurného. Schenker, Alexander M., & Stankiewicz, Edward (eds). 1980. The Slavic Literary Languages: Formation and Development. New Haven, CT: Yale Concilium on International and Area Studies. Schleicher, August. 1858. ‘Kurzer Abriß der Geschichte der slawischen Sprache.’ Beiträge zur vergleichenden Sprachforschung 1, 1–27. Schleicher, August. 1869. Darwinism Tested by the Science of Language. London: John Camden Hotten. Schleicher, August. 1873. Die Darwinsche Theorie und die Sprachwissenschaft. Of fenes Sendschreiben an Herrn Dr. Ernst Häckel. 2nd edn. Weimar: Böhlau. Schuchardt, Hugo. 1914. Die Sprache der Saramakkaneger in Surinam. Amsterdam: Johannes Muller. Siatkowska, Ewa. 1992. Rodzina języków zachodniosłowiańskich. Warszawa: PWN.

José del Valle

Linguistic History and the Development of  Normative Regimes: The Royal Spanish Academy’s Disputed Transatlantic Authority

Abstract The Royal Spanish Academy (RAE) was created in 1713 to preserve the purity of  the language in Spain and the territories of the Empire. However, once Spain’s American colonies became independent in the early nineteenth century, the Academy’s authority was challenged and declarations of  linguistic independence proliferated. The RAE’s 1870 initiative to create associated language academies and thus retain control over the language was only moderately successful: a strong perception remained that the Spaniards would not share linguistic power. In 1950, Mexican president Miguel Alemán invited the RAE and all associated Academies to meet in Mexico and coordinate ef forts; all attended but the Spaniards. Incensed by this rejection, a member of  the Mexican delegation called for linguistic independence; subordination to the RAE should end and a truly democratic association of  Academies should be created. His proposal was soundly defeated. In this article, I discuss the debates surrounding the RAE’s absence as discursive sites where the nature of  the pan-Hispanic community was being worked out.

1 Introduction On 14 June 1950, Mexican president Miguel Alemán (1902–1983) launched an original and ambitious cultural initiative that would end up having a profound impact – through a series of  both expected and unexpected

176

José del Valle

turns – on the institutional framework that supports the standardization of  Spanish:1 he charged the Mexican Academy of  the Spanish Language with organizing a conference whose goal would be to protect the nature and integrity of  Spanish. All language academies, from Spain, Latin America, and the Philippines, were invited.2

2 The Royal Spanish Academy The history of  these institutions begins, of course, much earlier, in 1713, when the Royal Spanish Academy (Real Academia Española: RAE) was created in Spain under the protection of the first Bourbon King, Felipe V. Modelled after the Académie Française – created in 1635 – and Italy’s Academia della Crusca – created in 1582 – it embraced as its most pressing responsibility the production of a dictionary that, in addition to stabilizing Spanish and safeguarding its purity, would also stand as a proud display

1

2

My understanding of standardization draws from two dif ferent but related categories. I rely, first, on Einar Haugen’s definition of language planning as ‘the activity of preparing a normative orthography, grammar, and dictionary for the guidance of writers and speakers in a non-homogeneous speech community. In this practical application of linguistic knowledge we are proceeding beyond descriptive linguistics into an area where judgment must be exercised in the form of choices among available linguistic forms’ (Haugen 1972: 133). Secondly, I rely on Milroy and Milroy’s (1999) take; in particular on their recognition of standardization both as process and ideology. Thus, I see the history of Spanish not as the development of a highly standardized language (not as a single-minded march towards the standard, as James Milroy has ironically put it) but as a deeply ideological process through which Spanish is associated with various forms and given social meanings at dif ferent points in history. In this understanding, the historicity of language is to be found not only in its deployment along a chronological axis but also – and especially – in the dynamic nature of its relation to cultural, political, and social contexts (see also Blommaert 1999). In this article, ‘Latin America’ refers only to the Spanish-speaking countries in the continent.

Linguistic History and the Development of  Normative Regimes

177

of  the country’s cultural stature. For the Academy’s founding fathers, the dictionary – its first edition was published in six volumes between 1726 and 1739 – was a matter of national pride (Álvarez de Miranda 1995: 272).3 As the institution, slowly but surely, consolidated its status within Spain – broadening its scope to include orthography (first published in 1741) and grammar (first published in 1771) and strengthening its ties to the State apparatus4 – it also began to concern itself with its wavering reputation overseas, especially after the wars that, during the second decade of  the nineteenth century, led to the collapse of the Spanish Empire and the emergence of new Spanish-speaking independent nations.

3 The origins of panhispanism Obviously, the collapse of  the Spanish Empire and the independence of  the American colonies resulted in broader projects of emancipation aimed at the construction of cultural systems autonomous – at least, relatively autonomous – from that of  the former metropolis (Rama 1982). In this context, the intellectual leaders of  the independence movement as well as the most prominent figures of  the subsequent generations tended to turn their eyes away from Spain, and more often than not, France and the Anglo-Saxon world came to be preferred political and cultural beacons for the emerging Latin American countries. However, in the process of developing new national – properly American – identities, language 3

4

For a lengthy and of ficial history of the Royal Spanish Academy see Zamora Vicente 2000. A good article-length overview of the institution’s history is Álvarez de Miranda 1995. For an in-depth study of the linguistic climate in eighteenth century Spain see Lázaro Carreter 1949. The RAE’s grammar ‘was declared by Carlos III in 1780 the of ficial textbook for teaching Spanish in schools’ (Álvarez de Miranda 1995: 276). In the nineteenth century, the Academy’s orthography became of ficial in schools in 1844 (see Laura Villa’s article in this volume) and its grammar became again mandatory in 1854.

178

José del Valle

was a strategic resource, and the predominantly white creole elites that had led the independence movement were Spanish-speaking. Like North Americans vis-à-vis the British, Latin American elites had to reconcile the embrace and cultivation of  the colonial language – which was their own – with the cultural and linguistic autonomy that they had struggled so vigorously to attain. Esteban Echeverría (1805–1851), for example, a prominent Argentinian intellectual, accepted the language, Spanish, without reservations but claimed the legitimate right of Spanish-speaking Americans to control its elaboration on their side of the Atlantic: ‘The only legacy that Americans could accept, and do gladly accept from Spain because it is truly precious, is the language; but they accept it on condition that it be improved, that it be progressively transformed, in other words, on condition that it become emancipated’ (quoted in Alfón 2008: 52).5 Similarly, Juan Bautista Alberdi (1810–1884), of  the same generation, linking language and thought and embracing France’s intellectual tradition, predicted divergent linguistic developments on both sides of the Atlantic: ‘If language is just one aspect of  thinking, ours demands intimate harmony with our American way of thinking, a thousand times more congenial to the rapid and direct movement of French thought than to the interminable contours of Spanish thought’ (quoted in Alfón 2008: 53). Domingo Faustino Sarmiento (1811–1888), a firm and salient voice among those of  his generation (he was Argentina’s president between 1868 and 1874), also af firmed Latin America’s linguistic independence and aimed many of  his rhetorical darts at what he felt was a powerless and useless Royal Spanish Academy: ‘To expect a powerless academy – which has no authority even in Spain itself, lacks prestige, and is benumbed by the knowledge of its own impotence – to give us rules that will in the end be useless to us is an humiliation unworthy of nations that have attained the rank of same’ (Sarmiento 1843: 29). Ángel Rosenblat, who studied the linguistic ideas of this Argentinean generation (Rosenblat 1960), summarized their position stating that ‘they were of one mind in their cultural and linguistic anti-Spanish attitude, which at times became 5

Throughout the article, all translations from Spanish are mine.

Linguistic History and the Development of  Normative Regimes

179

plain Hispanophobia; in their fervent and neophyte enthusiasm for French literature and thought; […] in their af firmation of [North] America as a source of inspiration; […] in their rejection of any form of academic or academistic tutelage; in their af firmation of the language’s freedom, so that it might go forward with the new ideas’ (Rosenblat 1960: 557). In Spain, the cultural, economic, and political establishment did not remain completely passive after the colonial mishap. Through private and at times public initiatives, a slow but steady mobilization of institutional resources began to unfold in the hope that Spain could somehow recycle to its advantage the transatlantic space left void by the dismantlement of  the Empire. These ef forts to engage in post-colonial community-building with Latin America have come to be known variously as hispanismo, hispanoamericanismo, or panhispanismo (Pike 1971, Sepúlveda 2005). While this cultural trend was mostly discursive, it also materialized in the form of a number of cultural initiatives that included congresses and symposia (such as the Ibero-American congresses of 1892 and 1900 organized by the Unión Íbero-Americana) as well as journals (such as La Ilustración Ibérica, La Revista Española de Ambos Mundos, and La Ilustración Española y Americana) (Fogelquist 1968). Panhispanismo was grounded in the belief that a common Spanish culture embodied in the Spanish language existed on both sides of the Atlantic and was the basis for an economically and politically operative entity, for a true hispanofonía. In this use of the term, hispanofonía does not simply refer to a group of nations linked by a shared communicative code; it is rather, following Benedict Anderson’s (1983) notion, an ‘imagined community’ grounded in a common language, itself imagined, that ties together in an emotional bond those who feel they possess it and those who have a sense of  loyalty to it (Del Valle 2006 and 2008). It is, therefore, according to Susan Gal and Kathryn Woolard’s definition of the term (2001), a language ideology, a historically situated conception of Spanish as an enactment of a collective order in which Spain performs a central role.

180

José del Valle

4 The creation of associated academies As a result of concerns regarding the possibility of safeguarding linguistic unity and in keeping with broader policies aimed at maintaining Spain’s post-colonial preeminence in Latin America, in 1870, the Academy’s director, Mariano Roca de Togores (1812–1889), Marquis of Molins, appointed a special committee charged with exploring the creation of a series of associated academies throughout the Spanish-speaking world. In a matter of weeks, the committee issued a report stating that, while political ties had been severed by the wars of independence, the common language continued to be the anchor of a single shared intellectual system. They identified the language as the common fatherland and recognized, for the first time, that the majority of  Spanish speakers was not to be found in Spain but in the American continent. They further suggested that the active intellectual life of  the new nations and their frequent contact with other countries and their languages – more frequent than with Spain itself – should be a source of concern, since it might lead the language to follow dif ferent paths in dif ferent places. In view of these circumstances and following the committee’s recommendation, the decision was made to create a system of associated academies that would secure the preservation of  the language and support the Spanish institution’s authority in the former colonies.6 It is crucial to point out that, in designing this new consortium, the Royal Academy retained control over the dictionary, grammar, and orthography; imposed its statutes and regulations; and retained the right to confirm all new members of  the associated academies.7

6

7

The first new academies were founded in the 1870s – the first one in Colombia in 1871 – but the full process would turn out to be more cumbersome than expected: many in fact did not take of f until the third decade of the twentieth century or even later. Honduras, for example, did not have an academy until 1949. The report – which includes the statutes that would regulate the relationship between the Royal Academy and the associated academies – was written by Fermín de la Puente Apezechea and published in the Royal Academy’s Memorias (De la Puente Apezechea 1873). The process is also discussed in Zamora Vicente (1999: 347–349).

Linguistic History and the Development of  Normative Regimes

181

5 The Academies’ Congress of 1951 Given the Spaniards’ determination to safeguard their authority and selfappointed role as leaders of  the standardization process, it seems intriguing, to say the least – certainly fascinating from a geopolitical perspective on language policy – that it was the Mexican and not the Spanish institution that first summoned all academies to the defense of  the common language. The fact is that on 14 June 1950, José Rubén Romero (1890–1952), Mexican writer and academician, publicly declared on behalf of president Alemán – whom he served as special advisor – the Mexican government’s commitment to organizing and fully subsidizing this landmark event. He predicted: a beautiful spectacle […] [that will bring together] all academies without exception […] without consideration for relationships among governments […] caring only for the common interest and mutual sympathy that springs from language, the spiritual blood of races. (Comisión Permanente 1952: 11)

An organizing committee was appointed on the 27th and, a few weeks later, dispatched across the Atlantic. Its members landed in Madrid on 13 October and were received at the airport by a generous cohort of  Spanish academicians. On the 18th, at a special session of  the Royal Academy (broadcast live to Spain and Latin America through Radio Nacional de España), the Mexican delegation issued the of ficial formal invitation and asked the head of the Spanish institution, distinguished philologist Ramón Menéndez Pidal (1869–1968), to of ficiate as chairman of  the upcoming conference.8 In his response, Pidal enthusiastically accepted the generous invitation and, with an eloquent speech, reminded his fellow academicians of  their immense responsibility: 8

Prior to the trip, the organizing committee had developed a set of rules and regulations. These (article 2) established that the conference’s president would be the director of  the Royal Spanish Academy or any Spanish academician that he might wish to appoint as his replacement (Comisión Permanente 1952: 17).

182

José del Valle One does not have to be a prophet to predict and expect with all confidence that this meeting will initiate a crucial period in which the Academies’ mutual relation will be closer and more fruitful in the serious task that they face: maintaining our common language as one of  the most powerful bonds created by Humanity. Although this language community may be strained, it is essentially unbreakable […]. Securing the superior, prosperous, and ef fective unity of our language or letting it decay – not into complete fragmentation but through excessive valorization of  the dialectal dif ferences that are normal in all languages – is in the hands of  those who speak it and in the inspiration of  those who write in it; but also – and this appeals to our sense of responsibility – in the conscience and the minds of those of us who belong to these bodies that will now meet in Mexico. (‘Información Académica’ 1950: 480–481)

6 The polemic before the conference By early 1951, everything seemed to be on track and the conference promised to be a celebration of pan-Hispanic harmony and commitment to the joint defense of the common language under Spain’s tutelage and Mexico’s sponsorship. The program had been designed and approved by the Royal Academy.9 But suddenly, on 26 February, the Mexican Academy received a telegram: ‘SURGIDA DIFICULTAD INSUPERABLE / UNSURMOUNTABLE PROBLEM ARISEN’. It was signed by Julio Casares (1877–1964), the Royal Academy’s secretary, and followed by a letter of the same date confirming, with barely a few words more, the Spanish institution’s withdrawal from the conference. The circumstances surrounding the agency’s apparent change of heart would only be aired six weeks later, on 7 April, through a public statement made by Spain’s Education Minister: Upon receiving the invitation from the president of  the Republic of  Mexico, the Royal Spanish Academy stated that, for patriotic reasons, it demanded, as a necessary moral pre-condition of its participation, that the Mexican government publicly state 9

The program was organized around four major topics: the unity and protection of  Spanish, grammar issues, lexical issues, and interacademic cooperation (Comisión Permanente 1952: 21–24).

Linguistic History and the Development of  Normative Regimes

183

that it had ended its relations with the red government10 and suspended recognition of  the so-called Spanish diplomatic representation in Mexico. Since the Mexican government did not comply with this requirement – which in the present circumstances our national dignity deems indispensible – the Royal Spanish Academy has decided not to attend. (Pagano 1951: 253)

The Spanish government’s attitude triggered, as one might expect, a heated reaction throughout Spanish America and most intensely in Mexico. The daily press filled its pages with articles reporting on the episode and expressing indignation at what was felt to be Spain’s slap in the face to the Mexican people. However, in spite of the commotion, the conference organizers were determined to minimize the impact of the crisis. The opening ceremonies proceeded with apparent normality and with nothing but a mild passing reference to the Spaniards’ absence: We wanted a Hispanism that was fitting and logical, an integral Hispanism led by the Motherland. But since that has been impossible to achieve, the only thing that can be done is what we are trying to do: a self-governed Hispanism […] Provisionally self-governed, it should be understood; because neither the Mexican Academy nor the other academies from this hemisphere nor that of Malaysia have thought for one second of disregarding the authority of  the Royal Spanish Academy. (Nemesio García Naranjo, Mexican academician, on April 23rd, while proposing, at the opening ceremonies, a toast to the success of  the conference, quoted in Guzmán 1971: 1375)

10

In 1936, during Spain’s Second Republic (1931–1939), the democratically elected leftist government of  the Frente Popular came under siege as a result of an attempted military coup. Resistance by forces loyal to the Republic as well as by significant sectors of Spanish society was initially successful and a three-year civil war ensued that ended with the insurgents’ victory in 1939. While Spain’s political regime became a right-wing military dictatorship led by General Francisco Franco, a Republican Spanish government was established in exile. It was referred to by Francoists as ‘the Red government’. Initially Franco’s Spain was excluded from the United Nations, but beginning in late 1950 diplomats succeeded at breaking the isolation and various countries began to formalize relations with Spain. Mexico, which had been the first to recognize the legitimacy of  Spain’s Republican government in exile, refused to establish diplomatic relations with Franco’s government (in fact, it would only send an of ficial diplomatic mission to Spain in 1977, two years after Franco’s death).

184

José del Valle

7 The polemic during the conference However, the undercurrent of tension and the symbolic power of the event were too high to be contained. During the conference’s first plenary session, on 27 April, Mexican writer and academician Martín Luis Guzmán (1887–1976), with a passionate but carefully crafted speech, forced his fellow academicians to face the problem and take a stand. The Spanish Academy’s absence, he maintained, was, first, an insult to Mexico and all other Spanishspeaking countries; second, a f lagrant violation of  the statutes that, since 1870, had regulated the relationship among the academies;11 and, third, an irresponsible act that jeopardized the Academies’ mission of protecting the nature and integrity of the language. If the problem created by the Spanish Academy were not remedied, he asked, ‘what hope can we have that our indispensible cooperation will prevent the danger that already looms on the horizon: the disintegration of  Spanish?’ (Guzmán 1971: 1377). We would risk, he went on, ‘because of Spain and Mexico today or any other two tomorrow, that practical and cooperative unity […] be rendered impossible; that this dangerous fragmentation become chronic’ (Guzmán 1971: 1380). He argued forcefully against the notion that linguistic unity could only be safeguarded through the current hierarchical model in which the Spanish American and Philippine academies were treated as simple subsidiaries. He claimed that, in fact, the opposite was true: real unity could only be achieved through a dignified agreement among equals that imposed on no one a humiliating oath reminiscent of  feudal relations and colonial ties: [T]he unity that you are trying to defend does not exist. But the one that we should be advancing has already begun to emerge; the one that would spring from an honorable agreement between equals, between peers; not one on the basis of an oath of service, a humiliating pledge after the existence of  feudal lords ended with the end of  feudalism and after our status as colonies ended with the end of  the Spanish empire. (Guzmán 1971: 1383)

11

Article 11 of  the 1870 statutes stated: ‘Since the purpose for which the Associated Academies are created is purely literary, their association with the Spanish Academy must be isolated from any political objective and, consequently, independent from the actions and relations among the respective governments’ (Zamora Vicente 1999: 363).

Linguistic History and the Development of  Normative Regimes

185

Guzmán maintained that, through its actions, the Spanish Academy had exposed the weakness of  the system and, most importantly, surrendered its credentials to act as the cultural leader of  the Spanish-speaking world. He concluded by proposing that the Mexico Conference adopt the following resolution: 1  We recommend that the American and Philippine academies, af filiates of  the Royal Spanish Academy, renounce their association with the latter […] and fully assume the autonomy from which they should not abdicate and their full inalienable personality […] 3  We also recommend that […] as soon as they reconstitute themselves as autonomous entities, steps be taken to summon representatives from each and all of  them, as well as the Royal Spanish Academy, to meet in Mexico or any other capital city in this continent in order to create, now as equals, a clear, egalitarian, and fruitful association. (Guzmán 1971: 1387)

The discussion that ensued was spirited and, at the end, left two motions on the table: some endorsed the outright rejection of  Guzmán’s proposal while others suggested that it be sent to a special committee for further consideration. The Philippines abstained, four delegations – Guatemala, Panama, Paraguay and Uruguay – voted to allow further discussion, and a wide majority of  thirteen voted to kill the initiative.12 While the debate triggered by Guzmán’s speech was undoubtedly the biggest controversy of  the conference, it was not the only one. New sparks re-ignited the polemic several times as other academicians put forth initiatives that, as in the first debate, were seen by others as threats to linguistic unity. Among those proposals were, for example, the creation, as a joint ef fort by all academies, of a new grand dictionary of Spanish and the production – again, with the consent and participation of all – of a new grammar that would help unify the teaching of  the language throughout the continent (Comisión Permanente 1952: 387–990 and 407–416).

12

A detailed description of the debate and vote can be found in Comisión Permanente (1952: 380–383).

186

José del Valle

8 Metalinguistic discourses and linguistic history This necessarily cursory review of the events shows that the Mexico conference ignited an explosion of metalinguistic discourses in which the nature and unity of Spanish as well as the proper organization of the institutions charged with supporting it were central themes. The conference provides us, therefore, with a wealth of precious materials crucial to the reconstruction and interpretation of Spain’s and Spanish America’s recent sociolinguistic history. These materials are, of course, dif ferent in nature from the type of data on which conventional approaches to historical linguistics usually rely. In contrast with the centrality of  language structure to the interests of  historical pragmatics or socio-historical linguistics – and I select these two because of  their explicit concern with context – metalinguistic discourses are central to a dif ferent kind of  historical take on language, one that assumes the sociopolitical origins and ef fects of linguistic phenomena and hopes to unwrap the sociopolitical entanglements of  the normative regimes in relation to which linguistic practices are deployed.13 By taking this perspective, we search for the ideological character of the linguistic representations that emerged in Mexico. We see them as producing, reproducing, and even challenging a normative regime; as historically situated conceptions of  language in general and of  Spanish in particular, as enactments of some sort of collective order, and as traces of the cultural, political and social interests involved in disputes over the production and reproduction of a normative regime in which Spanish and its varieties are assigned dif ferential values.14 13

14

With the label ‘normative regime’ I refer to a system of implicit and explicit rules and representations that attach dif ferent social values to dif ferent linguistic forms within a definable, socially structured human community. Context – social conditions and power relations – is crucial in the determination of  how a normative regime comes to be as it is and how it changes – or does not change – over time. To a great extent, this approach is consistent with the premises and objectives described in Blommaert 1999. The view of  linguistic ideologies that I embrace is presented in detail in Woolard 1998, Kroskrity 2000, and Gal and Woolard 2001.

Linguistic History and the Development of  Normative Regimes

187

9 Spanish and the Academies: A disputed normative regime We saw, for example, that the conference was organized on the premise that the nature and integrity of the language needed protection – this was repeatedly stated during the preparatory stages and opening ceremonies as well as in the course of the conference itself. The fear of linguistic deterioration was not by any means a new concern – and it was not, of course, an exclusively Spanish apprehension. It was a historically situated manifestation of one particular form of verbal hygiene (Cameron 1995), a discourse on language closely related to what James and Lesley Milroy have termed the complaint tradition (1999: 24–46). Throughout the nineteenth century, greater political protagonism by the working classes and increased social mobility were considerable sources of anxiety for the bourgeoisie in general and for the lettered elite in particular (González Stephan 1995, Ramos 1993). Similarly, the greater cultural prominence of countries such as France – which, as I indicated above, had become beacons of progress for Spain’s former colonies – was a disconcerting reminder of Spain’s relatively low status in Europe’s cultural hierarchies. These social and cultural concerns made their way to language, which became a discursive and political site where those anxieties were worked out. The undesirable presence of dialectal and foreign forms in textual genres thought to be traditionally safe from such invasions – such as the appearance of dialectal forms and Gallicisms in Spanish and Spanish American literature – was the principal trigger of  fears of  linguistic deterioration.15 15

The prologue to the Venezuelan Andrés Bello’s 1847 grammar of Spanish for Spanishspeaking Americans of fers a good example of  how these fragmentation anxieties played themselves out. Similarly, the controversy between Colombian philologist Rufino José Cuervo and Spanish writer and diplomat Juan Valera – which developed between 1898 and 1905 – was grounded in the Colombian’s prediction that, due to the proliferation of dialectal forms and foreign loans in literature and educated speech, Spanish would eventually split into a multiplicity of dif ferent languages (see Del Valle 2002).

188

José del Valle

In the middle of the twentieth century, as we saw, in the representations of language that emerged around and within the Mexican conference, similar concerns were voiced about the purity of the language. But the context, the circumstances invoked as sources of linguistic deterioration, had taken a new shape: on the one hand, the mass media had now replaced dialectal writers as the main disruption to the internal sociolinguistic hierarchy of  Spanish and, on the other, Anglo-Saxon culture – especially the United States’ post-war prowess and cultural sway – had now replaced France as the main threat to the cultural integrity of  Hispanic civilization. In keeping with the conference’s objectives, the need to protect the purity of Spanish and the strategies to do so were, of course, preferred objects of discussion – eight out of the thirty-one policy resolutions approved had an explicitly protectionist orientation. But, as we saw from the outset, the discourse of  language endangerment went beyond the standard forms of purism. In fact, the need to protect Spanish was linked to fears of  fragmentation that haunted academicians just as they had haunted Spanish and Spanish American men of letters at least since the middle of the nineteenth century. In those days, these fragmentation anxieties had been deeply entangled with the crises produced by the fall of  the Spanish empire and the nation-building projects undertaken not only by the former colonies but by Spain itself. Still, the focus of  the discourses associated with those fears was predominantly linguistic: if dialectal forms were to percolate to the speech of  the educated in each Spanish-speaking country – the argument went – the language would soon meet the same destiny as Latin and evolve into a number of related but independent tongues. These arguments, however, had practically no echo in 1951. In fact, one can legitimately conclude that, by this date, although the language of  fragmentation was still in play, the threat of actual linguistic divergence was no longer a serious concern – we might recall at this point Menéndez Pidal’s words (‘Although this language community may be strained, it is essentially unbreakable’). What we witness instead is a fractal projection of disintegration anxieties from language itself  to language academies. In this new reincarnation of  the fragmentation discourse, the nature and unity of the pan-Hispanic linguistic field is threatened not by the possible divergent evolution of linguistic forms but by alternative – and contradictory – conceptualizations of  the body politic of  the language.

Linguistic History and the Development of  Normative Regimes

189

On one hand, Guzmán’s opponents believed the current institutional arrangement to be the adequate framework for defending the nature and unity of  the language. In keeping with the spirit of panhispanismo, these Spanish American academicians accepted a language community built under Spain’s tutelage. There was to be no questioning of the Royal Academy’s authority, no challenging of its dictionary and its grammar’s value as the only tools of standardization. Martín Luis Guzmán, however, stood against the traditional model of interacademic relations and the type of panHispanic community that it mirrored. As for all academicians attending the conference, for Guzmán, Spanish was in need of protection as the linguistic, cultural, and political inf luence of Mexico’s powerful northern neighbour loomed in the horizon – a very literal and not at all distant horizon for Mexicans. But a unified institutional approach to this defensive strategy was only possible if the institutions involved renounced a relationship that the Royal Academy’s absence had revealed as tarnished by the imprint of  bygone colonial hierarchies. The competent defense of  Spanish, claimed Guzmán, had to be grounded in an institutional arrangement in which all Spanish-speaking nations converged as equals.

10 Conclusion To conclude, although the Mexico conference had little immediate impact on the nature of standard Spanish, the debates over the standardization process triggered by Spain’s absence and Guzmán’s proud and lucid voice would in fact end up having a significant ef fect not only on how the Academies relate to each other but also, and most prominently, on how the Academies conceive the norm. We will leave that for future articles, hoping, however, to have convincingly argued for a political and historical approach to language that illuminates, in our case, not only Spanish’s history and present shape but also the complex regimes of linguistic normativity within which it has been inscribed.

190

José del Valle

References Alfón, Fernando. 2008. ‘Los orígenes de las querellas sobre la lengua en Argentina.’ In: González, Horacio (comp.). Beligerancia de los idiomas: un siglo y medio de discusión sobre la lengua latinoamericana. Buenos Aires: Colihue. 43–77. Álvarez de Miranda, Pedro. 1995. ‘La Real Academia Española’. In: Seco, Manuel, & Gregorio Salvador (coords.), La lengua española, hoy. Madrid: Fundación Juan March. 269–279. Anderson, Benedict. 1983. Imagined Communities: Ref lections on the Origin and Spread of  Nationalism. London & New York: Verso. Bello, Andrés. 1981. [1847]. Gramática de la lengua castellana destinada al uso de los americanos. In: Obras completas. Vol. 4. Caracas: La Casa de Bello. 1–382. Blommaert, Jan (ed.). 1999. Language ideological debates. Berlin: Mouton de Gruyter. Cameron, Deborah. 1995. Verbal hygiene. London: Routledge. Comisión Permanente del Congreso de Academias de la Lengua Española. 1952. Memoria del Primer Congreso de Academias de la Lengua Española, celebrado en México del 23 de Abril al 6 de Mayo de 1951. México: Academia Mexicana de la Lengua. Cuervo, Rufino J. 1899. ‘Prólogo.’ In: Soto y Calvo, Francisco, Nastasio. Chartres: Durand. vii–x. De la Puente Apezechea, Fermín. 1873. ‘Academias americanas correspondientes á la Española.’ In: Memorias de la Academia Española 4. Madrid: Rivadeneyra, 274–289. Del Valle, José. 2002. ‘Historical Linguistics and Cultural History: The Polemic between Rufino José Cuervo and Juan Valera.’ In: Del Valle, José, & Luis GabrielStheeman (eds), The Battle over Spanish between 1800 and 2000: Language ideologies and Hispanic intellectuals. London: Routledge. 64–77. Del Valle, José. 2006. ‘US Latinos, la hispanofonía and the language ideologies of high modernity’. In: Mar-Molinero, Clare & Miranda Stewart (eds), Globalization and language in the Spanish-speaking world: Macro and micro perspectives. London: Palgrave Macmillan, 27–46. Del Valle, José. 2008. ‘The Pan-Hispanic community and the conceptual structure of  linguistic nationalism.’ International Multilingual Research Journal 2, 5–26. Del Valle, José, & Luis Gabriel-Stheeman (eds). 2002. The Battle over Spanish between 1800 and 2000: Language ideologies and Hispanic intellectuals. London: Routledge.

Linguistic History and the Development of  Normative Regimes

191

Gal, Susan, & Kathryn A. Woolard. 2001. ‘Constructing languages and publics: Authority and representation.’ In: Gal, Susan, & Kathryn A. Woolard (eds), Languages and publics: The making of authority. Manchester: St Jerome. 1–12. González Stephan, B. 1995. ‘Las disciplinas escriturarias de la patria: constituciones, gramáticas y manuales.’ Estudios 3/5, 19–46. Guzmán, Martín Luis. 1971. Obras completas. México: Compañía General de Editores. Haugen, Einar. 1972. The Ecology of  Language. Stanford, CA: Stanford University Press. ‘Información Académica.’ 1950. Boletín de la Real Academia Española XXX, 480–481. Kroskrity, Paul V. 2000. ‘Regimenting Languages: Language Ideological Perspectives.’ In: Kroskrity, Paul V. (ed.), Regimes of language: Ideologies, polities, and identities. Santa Fe, NM: School of  American Research Press. 1–34. Lázaro Carreter, Fernando. 1949. Las ideas lingüísticas en España durante el siglo XVIII. Madrid: Espejo. Lázaro Carreter, Fernando. 1980. Estudios de lingüística. Barcelona: Crítica. Menéndez Pidal, Ramón. 1918. ‘La lengua española.’ In: Hispania 1. 1–14. Milroy, James, and Lesley Milroy. 1999. Authority in Language: Investigating Standard English. 3rd edn. London: Routledge. Pagano, José León. 1951. ‘Primer Congreso de Academias de la Lengua Española. Informe del Señor académico don José León Pagano.’ Boletín de la Academia Argentina de Letras XX/76, 249–262. Pike, Fredrik. 1971. Hispanismo, 1898–1936. Notre Dame, IN: University of  Notre Dame Press. Rama, Carlos M. 1982. Historia de las relaciones culturales entre España y la América Latina. Siglo xix. México: Fondo de Cultura Económica. Ramos, Julio 1993. ‘El don de la lengua.’ Casa de las Américas 34, 13–25. Rosenblat, Ángel. 1960. ‘Las generaciones argentinas del siglo xix ante el problema de la lengua.’ Revista de la Universidad de Buenos Aires (Quinta época) 4/4, 539–584. Sarmiento, Domingo Faustino. 1949 [1843]. ‘Memoria sobre ortografía americana.’ In: Obras completas, vol. 4. Buenos Aires: Luz del Día, 1–50. Sepúlveda, Isidro. 2005. El sueño de la Madre Patria. Hispanoamericanismo y nacionalismo. Madrid: Marcial Pons. Woolard, Kathryn A. 1998. ‘Introduction: language ideology as a field of inquiry.’ In: Schief felin, Bambi B., Kathryn A. Woolard, & Paul V. Kroskrity (eds), Language ideologies: Practice and theory. New York: Oxford University Press, 3–47. Zamora Vicente, Alonso. 1999. Historia de la Real Academia Española. Madrid: Espasa.

Juan R. Valdez

Colouring Language: Pedro Henríquez Ureña’s Representations of  Spanish and Dominican Identity1

Abstract The Latin American linguist and philologist Pedro Henríquez Ureña raised significant and inf luential questions about the origin and development of  Spanish in the New World. However, most assessments of  his work tend to ignore the fact that these questions were part of  fundamental ideological debates. Henríquez Ureña’s work on Dominican Spanish emerged in a political climate in which an instrumentalist approach to language and history became the standard practice for Dominican intellectuals. This practice authorized images of  Spanish as representative of  the Dominican national character and a specific type of  Dominican Spanish that had to be protected from pernicious external forces. Dominican Hispanism strongly reemerged in this context as a political and cultural ideology based on the concept of the Dominican national identity as pure, white and Spanish. In Henríquez Ureña’s studies of Dominican Spanish, linguistic forms become linked to some of these racial and cultural categories. My analysis (inspired by critical linguistic historiography and language-ideological research) takes advantage of several methodological tools available for interdisciplinary research in order to fully understand the sociohistorical meanings and implications of  Henríquez Ureña’s linguistic texts.

1

I thank the organizers for the opportunity to participate at the 2009 Bristol conference and my colleagues José Del Valle and Laura Villa for their insightful comments, questions, and feedback. Segments of this analysis appeared in a previous publication (Valdez 2009); it is part of a wider research project whose complete results appear in my book, Tracing Dominican identity: the writings of  Pedro Henríquez Ureña (2011).

194

Juan R. Valdez

1 Introduction During defining moments, language has figured prominently as a subject of history in Spain and Latin America. For example, in order to unify the diverse regions of Iberia under one kingdom, King Alfonso X (1221–1284) gave priority to the standardization of written Castilian, an ef fort that led to the first orthography of any Romance language (circa 1270). Antonio de Nebrija (1441–1522), in the prologue to his Castilian Grammar (1492), addressed to Queen Isabel of Spain (1451–1504), advanced the conclusion ‘that language always accompanies empire’, thus persuading the Spanish Crown to pay particular attention to the relationship between language and power. When the Spanish empire had begun to disintegrate a few centuries later, the Venezuelan-born Andrés Bello (1781–1865), concerned with the uncertain future of  the newly independent Latin American nations, proposed his Grammar, intended for Latin Americans, as a device that could help check the dynamics of cultural, dialectal, and imperial fragmentation. Finally, after Spain lost its last colonies and entered a cultural crisis, the Spanish intellectual Ramón Menéndez Pidal (1869–1968) produced Orígenes (1926) in which he chronicles the historical conditions that in his view initially allowed the Castilian form of  Spanish to dominate. In the Spanish-speaking world, at dif ferent times and in dif ferent contexts, language historians have selected and prioritized only those sociolinguistic facts which give centre stage to some specific images and constructions of  the Spanish language. Yet, rarely do they ref lect upon the specific historical complications that had to be overcome or erased in the establishment of a particular linguistic hegemony. It is up to the linguistic historiographer to answer questions such as why and how certain images or visions of  the language f lourished while others conspicuously disappeared or withered away. This task can only be accomplished by stepping outside the boundaries of autonomous linguistic research. Pedro Henríquez Ureña (1884–1946), a key figure in Hispanic linguistics and philology, and his work have been the objects of numerous studies. However, to this day his work has not received the critical attention

Colouring Language

195

it deserves (Sarlo 2000). This lacuna is even more obvious in the studies related to his linguistic production. Among Henríquez Ureña’s most important linguistic work, we find first Observaciones sobre el Español en América (1921), El supuesto andalucismo de América (1925) and La gramática castellana (1931), co-authored with Amado Alonso. A linguistic-historiographical approach reveals that his texts abound with some curious ideas that were prevalent in his professional circle and which belonged to what Koerner (1995: 17) calls the ‘intellectual climate’. An intellectual climate is defined by the professional conditions of production, the period’s intellectual currents, and the immediate socioeconomic and political situation. Henríquez Ureña’s seemingly contradictory representations of Latin American Spanish (‘polygenesism’) and Dominican Spanish (‘archaicism’) underscore notions of language, race, and nation that emerged from specific sociopolitical circumstances and struggles. In his classic text, El español en Santo Domingo (1941), Henríquez Ureña advanced the idea that what characterizes the Dominican variety of Spanish is its vocabulary and syntax, the most archaic in the continent. At the same time, the author consistently argued for the absence of  African substrate elements in the speech and culture of  his country. A language-ideological analysis2 reveals that Henríquez Ureña’s characterization of  Dominican Spanish emerges in a specific context, in the debate and struggle for the construction of postcolonial Santo Domingo. Building on the definition of  language ideology, of fered by José del Valle (in this volume), as ‘a historically situated conception of  Spanish as an enactment of a collective order in which Spain performs a central role’, the purpose of my study is twofold: a) to demonstrate the ways in which these sociopolitical phenomena surface in the construction of Henríquez Ureña’s ‘archaic’ Dominican Spanish; and b) to show how his language representation is intricately connected to historical problems, i.e. the emergence and development of  the Dominican state.

2

My study is consistent with this particular line of research defined in Crowley (1996), del Valle and Gabriel-Stheeman (2002), Irvine and Gal (2000), Joseph and Taylor (1990), and Woolard (1998).

196

Juan R. Valdez

2 Language and history in the Dominican Republic Although Spanish is undoubtedly dominant, in the Dominican Republic there are speakers of  languages other than Spanish and a significant number of people whose sociolinguistic competence is better characterized as multilingual, multidialectal, or complex repertoire. Contrary to the claims of one analyst (Núñez 2003), it is the complex speech communities and vibrant contact zones which linguistically define this little half-island nation. For example, in the Samaná Province we find speakers of  English, descendants of  African-American ex-slaves, invited by Haitian president Jean-Pierre Boyer (1776–1850) in 1824, and descendants of English-speaking West Indians who migrated to the Dominican Republic in the late nineteenth and early twentieth centuries. We also find Dominicans who speak a variety of Haitian Creole, locally identified as ‘Patois’, and a large number of  Haitian immigrants and their Dominican-born descendants who are bilingual to dif ferent degrees. In sharp contrast to the culture of monoglossia3 and monoethnic ideologies which have historically dominated the development of national identity in the Dominican Republic, this multidialectal and multilingual repertoire has been glossed over, minimized or erased in almost all accounts of Dominican society and history. Manuel Núñez (2003 and 2001), a linguist and member of  the Dominican Academy of  the Spanish Language, who has become the intellectual spokesperson for ‘anti-Haitianism’, provided a recent example of this monoglossic culture. In justifying his commitment to support and protect the Spanish language, he declared language as the first line of defence in harnessing the Dominican Republic’s identity, 3

According to del Valle (2000: 119–120), the culture of monoglossia, in which the dominant discourse on language is grounded, consists of  the principle of  focused grammar or the assumption that what linguistically characterizes individuals as well as a community is possession of a well defined and relatively stable grammar and the principle of convergence, or the assumption that people’s linguistic behavior tends to become homogeneous over time through pressure from the dominant norm of  the community.

Colouring Language

197

and that the country was not bilingual. He wrote: ‘only the Hispanic component is truly Dominican culture’ (Núñez 2001: 234). Dominicans are not alone in this form of erasure: modern ideologies of the nation advance powerful homogenizing tendencies. However, cultural historian Odalís Pérez (2002) documented the development and adoption by the State of a specifically Dominican nationalist ideology that demands, first, the Hispanization and whitening of Dominican society, second, the adoption of  the dominant elites’ cultural behaviour and narratives; and, third, the erasure of cultural features associated with the popular classes. Composed of various ideas and proposals that centred on the historical and cultural bonds between Spain and the Spanish-speaking countries in America, the subject of  Dominican Hispanism merits a fuller discussion, but here a summary will suf fice: one of  the branches of  Hispanism, advanced by both Spanish and Latin American conservative and liberal politicians and intellectuals, defends the formation of a transatlantic Spanish community or identity that justifies Spain’s cultural presence, commercial interests, and foreign policy in Latin America.4 These particular Hispanists proposed a shared (idealized) identity of the Peninsular and American people, based on a cultural community comprised of  historical, linguistic, philosophical, and religious elements that underlie the constitutions of  the respective states and transcend the political divisions that followed the wars of independence. It is this particular vision of the Pan-Hispanic community that underlies the Dominican brand, amplified to accommodate the specific anti-Haitian ideology that emerged during the foundation of  the Dominican state. Elaborated over a long period of social, economic, and political struggles, this of ficial ideology and its related policies were grounded in an af firmation of Dominicanness as the opposite of all black things Haitians symbolized: within this identity framework, Dominicans are, by definition, everything that Haitians are not. Ties with Haiti are denied, especially any tie that binds both parts of  the island through an African heritage. The particular circumstances surrounding the birth of  the Dominican state 4

See del Valle (in this volume) and Sepúlveda (2005).

198

Juan R. Valdez

shed light on the development and evolution of this ideology. In founding their nation, a group of  leading Dominican political figures sought independence from the political jurisdiction of  Haiti, which, having gained its independence in 1804, retained control of Santo Domingo until 1844. This dif ficult process of independence from Haiti and the various military attempts of  Haitian leaders between 1844 and 1855 to restore their rule in the Spanish-speaking half of  the Hispaniola contributed to a nationbuilding ideology that produced a powerful element of self-identification in opposition to Haitians. Already in 1822, years before independence, José Núñez de Cáceres (1772–1846), a local political leader, stated while handing the keys to the city of Santo Domingo over to Jean Pierre Boyer that the Haitian plan for the political unification of the entire island would inevitably fail. According to Rodríguez Demorizi (1975: 17), there and then Núñez de Cáceres declared that the Spanish language alone ‘was a separation wall, as natural and insurmountable as were the Alps or the Pyrenees’.5 Since that early stage and throughout the many phases in the Dominican Republic’s process of nation-building, language and race have been deployed as strategic discursive sites where Dominicanness and Dominicans’ linguistic behaviour become racialized. Historically, many of  the authors that contributed to the dominant ethnopolitical discourse have represented the Dominican nation through the lens of race and language.6 Consistently, they envision the Dominican Republic as a homogeneous community without contradictions or adulterations. Along with race, language was and remains the crucial element in the construction of  this identity (Bailey 2002 and Toribio 2005). The link between language and national identity is a recurrent theme in nationalist discourses which frequently appears in public discussions of Dominican identity. Time and again, questions concerning language are raised in order to address and, in some cases, ignore larger social issues and problems.

5 6

All translations in this chapter are mine. See San Miguel (2005).

Colouring Language

199

3 Colouring Dominican Spanish Henríquez Ureña’s El Español en Santo Domingo (ESD, from here on), for many years the most comprehensive work on Dominican Spanish, is still a central landmark in any investigation of Dominican and Caribbean dialectology. In this book, Henríquez Ureña advanced the thesis that the vocabulary and syntax of  Dominican Spanish is ‘the most archaic’ in the American continent. In chapter three, the author postulated the argument that the archaic nature of the lexicon is unmistakably demonstrated by the widespread use in Dominican speech of expressions that became obsolete in other Spanish varieties. Dominican Spanish, Henríquez Ureña af firmed, remains faithful to its Spanish roots. At the end of chapter four, he posed the following question: are the numerous parallels between the contemporary Dominican lexicon and the Castilian lexicon of the sixteenth and seventeenth centuries suf ficient to support the claim that Dominican Spanish is a derivative of  the Castilian Spanish that was spoken in seventeenth-century Salamanca? No, answered Henríquez Ureña (1941: 54). However, this comparison became his point of departure and is implied throughout the book. His description of the lexicon in chapter five contains words such as: aguaitar; bregar; despacharse; dilatarse; guayar; prieto; privar en; tíguere; tostón. The author explained that words like, for example, aguaitar [‘to watch; to look’], were commonly used in the Dominican Republic, after falling out of use in other Spanish-speaking regions where educated people no longer used them. Another example is dizque [‘apparently’] which, according to Henríquez Ureña (1941: 56), was no longer used in most of  Spain or Argentina. He claimed that, at the time he was writing ESD, even the Spanish Royal Academy, or rather the lexicographers working on its dictionary, were unfamiliar with many of these Dominican words (1941: 90). For his description of the lexicon he selected a series of words documented in Dominican Spanish, many of them certainly archaic, and he attributed to these words the meaning given to them by the original Spanish colonizers. Henríquez Ureña’s descriptive strategy consisted of connecting typically

200

Juan R. Valdez

Dominican words to their Spanish Golden Age origin. In and of itself, this lexicographic procedure can be perceived as ideologically neutral, yet closer attention and further scrutiny soon shows Henríquez Ureña linking the archaic lexicon to Dominicans’ loyalty to the Hispanic cultural traits and by extension to the purported whiteness of  Dominicans. A careful reading of chapters five and six of ESD reveals the discursive construction of a socio-historic image that links the Dominican lexicon to the original Spanish settlers and their linguistic past. In fact, in order to make this colonial connection stronger, Henríquez Ureña advised that ESD should be read along with the supplementary text La cultura y las letras coloniales en Santo Domingo (1936), a text whose cultural and historical facts about Spanish colonialization would help explain and solidify the linguistic ones. We find further evidence of the link between linguistic description and extra-linguistic cultural and racial categories in Henríquez Ureña’s treatment of certain aspects of the phonology of Dominican Spanish. The phonological phenomenon in which both postvocalic /r/ and /l/ are pronounced as a glide, thus, realizing words such as comer ‘to eat’ and porque ‘because’ as [ko.méi] and [pói.ke] has been documented as characteristic of a variety from the northern part of the country known as cibaeño. In describing this peculiar and divergent pronunciation, he highlighted the fact that the phenomenon is also found in rural areas of Puerto Rico where, he emphasized, the people are generally white (Henríquez Ureña 1941: 49). In chapter twelve, the book provides another example of  how Henríquez Ureña used linguistic dif ferences to convey and mediate a social and ethnolinguistic identity that is, above all, white. He described a phonological process in which the intervocalic consonant [r] is pronounced as [l]: words such as cerebro ‘brain’ are pronounced as [se.lé.bro]. Earlier scholars thought they had discovered a possible African origin for this feature. Esteban Pichardo, for example, the Dominican-born Cuban lexicographer, had attributed it to the bozales (African-born second language learners of Spanish). Henríquez Ureña, after stating that the phenomenon was generally rare, responded to claims of its African origins by noting that it is never found among blacks that had been born on the island. These, he insisted ‘speak like white natives’ (Henríquez Ureña 1941: 168). Presenting what he thought was insurmountable evidence against the African origin

Colouring Language

201

of  this feature, he stated that the phenomenon was also found in Spain where African inf luence was dif ficult, if not impossible, to explain. To summarize, first Henríquez Ureña minimizes the scope of  this phonological phenomenon, second he located it within an exclusively Hispanic linguistic tradition, and third he af firmed the Hispanization and whitening of  the black Dominican through the acquisition of white speech. This process of strictly associating sociocultural traits with linguistic phenomena is known by linguistic anthropologists as ‘iconization’.7 In Pedro Henríquez Ureña’s linguistic texts as well as in the work of numerous scholars, we find the iconization of  Dominican Spanish representing Hispanicness as the only relevant component of  Dominican culture.

4 Erasing the black In addition to establishing the iconic relationship between linguistic forms, hispanicity and whiteness among Dominicans, Henríquez Ureña took on the task of contesting the existence of an African cultural and linguistic heritage. In a polemic against the Neogrammarian philologist Wilhelm Meyer-Lübke’s claim of the existence of a ‘black’ dialect of Spanish in the Dominican Republic, Henríquez Ureña (1919: 430) wrote: The black race has never predominated [in the Dominican Republic] and the Castilian language remains pure. There has never been, nor is there now, a black dialect in the Republic. To the contrary, Santo Domingo belongs to that segment of America where the language remains closer to its Castilian roots.

7

Irvine and Gal (2000: 37) defined ‘iconization’ as semiotic and discursive processes by which some aspect of or feature in the linguistic repertoire of a given community is perceived as an iconic representation of its members, as if a linguistic feature could inherently ref lect the nature of a social group. They also defined the concept of ‘erasure’ as the process that ‘renders some people or activities invisible in order to simplify the sociolinguistic field’.

202

Juan R. Valdez

For the observer acquainted with Dominican society and history, this is indeed a surprising minimization if not an all-out erasure of Africanness as a component of Dominican identity. The same minimization emerged in ESD when Henríquez Ureña (1941: 130) took on two of his contemporaries, the Cuban Fernando Ortíz and the Dominican Carlos Larrazábal Blanco, questioning their theories of African inf luence in Cuba and the Dominican Republic. Citing demographic figures from early colonial chroniclers such as Bartolomé de Las Casas (1484–1566) and Gonzalo Fernández de Oviedo (1478–1557), Henríquez Ureña (1941: 132–133) maintained that: In Santo Domingo African inf luence is very scarce […], the fact is that there was no systematic importation of slaves during the first half of  the sixteenth century […]. In the nineteenth century, there was even less […] and from the beginning a good number of the slaves did not arrive directly from Africa: It is a fact that they arrived from Spain where they had been bought from the Portuguese; they were already Hispanized. Interestingly enough; the first slaves were not all black; white slaves were also brought to America. Anyway, what characterizes the population of African descent in Santo Domingo is its complete Hispanization.

The image that emerges from this sociodemographic description is of a complete colonization of the eastern part of the Hispaniola by the Spaniards who successfully restricted the arrival of African slaves and prevented their proliferation. In reference to the arrival of migrant workers from Haiti and the English-speaking West Indies in the twentieth century, a phenomenon that was forcing some to question prevailing theories of race in Dominican society, Henríquez Ureña (1941: 133) wrote: This invasion is quickly blackening the country. It is estimated that there are more than 2,000 immigrants; the total population barely reaches a million and a half. Francisco Eugenio Moscoso Puello’s interesting novel, Cañas y bueyes, portrays aspects of this invasion […] Today the recent Caribbean fad of negro poetry that f lourishes in Cuba and Puerto Rico with the works of  Luís Pales Matos, Ramón Guirao, José Zacarías Tallet, Alejo Carpentier, Nicolás Guillén, Tomás Blanco, Emilio Ballagas, Marcelino Arozarena, and Vicente Gómez Kemp has reached Santo Domingo. The Negros in [the Dominican] Manuel Cabral’s Doce poemas negros (1935) are primarily Haitians or cocolos from the English-speaking West Indies, because the ones that are native to Santo Domingo have less colorful customs […] Arturo Pellerano Castro’s old and wonderful Criollas are really criollas ‘natives’; they speak of white and light-skinned women [italics in the original].

Colouring Language

203

These passages reveal the insistent presence, in Pedro Henríquez Ureña’s vision of  the Dominican Republic, of  the dominant ideology of national identity in which blackness is ef fectively erased. Whatever blackness exists in the Spanish-speaking part of  the island, as the last quote indicates, is foreign: primarily related to Haitians or cocolos from the English-speaking West Indies.

5 Dominican politics of identity in the twentieth century Language and history inevitably converge in scenarios where people (linguists included) attempt to draw or erase political and cultural boundaries. Nineteenth century nation-building discourse, anti-Haitian rhetoric, and the anti-North American sentiment that followed the US occupation of  the country all converged into the Dominican nationalism of  the 1920s when Henríquez Ureña was especially intellectually active in the US and Mexico and methodically outlined his ideas on the subject of Dominican Spanish. The brand of  Dominican nationalism that emerged during this particular period in the country’s history set forth a series of measures that Dominican leaders attempted to implement during, first, the Horacio Vasquez (1860–1936) presidency (1924–1930). However, the establishment and promotion of anti-Haitian and xenophobic attitudes became more vociferous and violent during the dictatorship of  Rafael L. Trujillo (1891–1961). Trujillo and his government launched a complete reevaluation and reconstitution of the Dominican nation. Trujillo, his acolytes, and government of ficials became alarmed by the number of Haitian immigrants in the country, particularly the Haitians that had been long established on the Dominican side of the border. The Trujillo government (1930–1961) initiated projects to ‘Dominicanize’ the border and force out as many Haitians as possible. Not content with the pace of deportation and the failure of some of the policies, in 1937, Trujillo ordered the military to slaughter thousands of Haitians by a river that divides the two countries in the northwest, aptly named El Masacre ‘The Massacre’. In order to distinguish dark-skin

204

Juan R. Valdez

Dominicans from Haitians, a shibboleth test was applied, in which only those that could pronounce the non-lateral simple vibrant [r] in the word ‘perejil’ [parsley] instead of  the uvular [R] would be spared from getting shot or hacked to death. This event drew serious international criticism which led many Dominican historians and intellectuals to of fer alternative accounts of this shameful episode and some of  the reasons that led to it. This period saw the consolidation of politically conservative forces and what Dominican historian Valentina Peguero (2004: 3) calls ‘the militarization of Dominican culture’. The worst elements in the military, the bourgeois bureaucracy, the Jesuit wing of  the Catholic Church, and one of  the most powerful and ruthless dictatorships in Latin America’s history converged in the exploitation and domination of  the Dominican masses and the legitimization of absolute despotic rule. The era of  Trujillo constituted the most reactionary period in the country’s history which Dominicans still struggle to overcome today and strongly conditioned the country’s intellectual and cultural production. It is in this specific political climate that Dominican historians and philologists instituted the instrumentalist and selective approach to language and history as the standard practice. As Dominican author Manuel Matos Moquete (1986: 30) argues, this practice authorized images of  the language as representative of  lo dominicano ‘Dominicanness’ and ‘pure speech’. It was decreed that Dominicans were responsible for protecting their speech and nationality from Anglicisms, Gallicisms (or rather Haitian creolisms), barbarisms, and ‘deterioration’ at the hands of  the lower classes. These sociolinguistic concerns coincided with the active policies of increased European immigration and the eradication of Haitians pursued by the Trujillo government, which invited white Europeans to settle in the Dominican Republic under the most favourable conditions. In this context Dominican Hispanism strongly reemerged as a political and cultural ideology based on the concept of the Dominican national identity as pure, white, Catholic, and Spanish. While not all Dominicans accepted this self-image and neocolonial status, this ideology became, and still is, dominant.

Colouring Language

205

6 Conclusion Pedro Henríquez Ureña’s linguistic analyses of  Dominican Spanish go beyond the confines of plain and simple dialectology. When read in light of  the intellectual and political context of its emergence, his linguistic description ultimately reveals the moving forces of  history. In Henríquez Ureña’s studies of  Spanish in Santo Domingo, linguistic forms are iconically mapped onto racial and cultural categories and, in their interaction, an image of the nation emerges, a linguistic ideology that ref lects the white, Hispanic identity that has dominated the discursive construction of  the Dominican Republic since its historical emergence as an independent nation. While his knowledge of  the linguistic reality of  the island was unprecedented and his familiarity with the Spanish language and its history admirable, his texts reveal a complex interaction between the coherent dialectological description of  Dominican Spanish and the ideologically driven contemporary discourses of national identity in which whitening the race was a central theme. Ultimately, in order to accomplish some of  his objectives in the representation of Dominican Spanish and society, he was unable to disengage from the neocolonial regimes of truth previously established by some of  his predecessors and the State apparatus. As Del Valle (2007) and Valdez (2011) have argued, the possibility of analysing such complex semiotic and ideological language phenomena emerges from approaches that go beyond the formal and material dimension of language and examine the culturally and politically situated nature of visions of  language, linguistic dif ferences, and communication. Understanding the manner in which linguists’ descriptions and representations of  language are related to historically salient ideologies in dif ferent eras and contexts requires an interdisciplinary approach between linguistics and historical studies. A historically situated approach to linguistic phenomena implies crossing the autonomous boundaries that representatives of each discipline sometimes zealously guard.

206

Juan R. Valdez

References Alonso, Amado and Pedro Henríquez Ureña. 1938. Gramática Castellana I & II. Buenos Aires: Losada. Bailey, Benjamin. 2002. Language, race and negotiation of identity: A study of Dominican Americans. New York: LFB Scholarly Publishing. Crowley, Tony. 1996. Language in history: Theories and texts. London: Routledge. Danesi, Marcel 1977. ‘The case for andalucismo re-examined.’ Hispanic Review 45, 181–193. Del Valle, José. 2007. ‘Glotopolítica, ideología y discurso: categorías para el estudio del estatus simbólico del español.’ In: del Valle, José (ed.), La lengua, ¿patria común? Ideas e ideologías del español. Madrid: Iberomamericana, Vervuert. Del Valle, José. 2000. ‘Monoglossic policies for a heteroglossic culture: misinterpreted multilingualism in Modern Galicia.’ Language and Communication 20, 105–132. Del Valle, José, & Luis Gabriel-Stheeman (eds). 2002. The Battle over Spanish between 1800 and 2000: Language ideologies and Hispanic intellectuals. London, New York: Routledge. Henríquez Ureña, Pedro. 1919. ‘La lengua en Santo Domingo (rectificación al romanista Meyer-Lübke).’ In: Obras completas (tomo IV). Santo Domingo: Editora Universal, 49–52. Henríquez Ureña, Pedro. 1975 [1941]. El español en Santo Domingo. Santo Domingo: Taller. Henríquez Ureña, Pedro. 1988 [1936]. ‘La cultura y las letras coloniales en Santo Domingo.’ In: Obra Dominicana. Santo Domingo: Sociedad Dominicana de Bibliófilos Inc. Henríquez Ureña, Pedro. 2003 [1925]. ‘El supuesto andalucismo de América.’ In: Obras completas (tomo IV). Santo Domingo: Editora Universal, 97–104. Henríquez Ureña, Pedro. 2003 [1921]. ‘Observaciones sobre el español en América’. In: Obras completas (tomo IV). Santo Domingo: Editora Universal, 61–85. Irvine, Judith, & Susan Gal. 2000. ‘Language Ideology and Linguistic Dif ferentiation.’ In: Kroskrity, Paul (ed.), Regimes of  language: Ideologies, polities and identities. Santa Fe, NM: School of  American Research Press, 35–84. Joseph, John, & Talbot Taylor (eds). 1990. Ideologies of  language. London: Routledge. Koerner, Konrad. 1995. Professing linguistic historiography. Amsterdam: John Benjamins.

Colouring Language

207

Matos Moquete, Manuel. 1986. La cultura de la lengua. Santo Domingo: Biblioteca Nacional. Menéndez Pidal, Ramón. 1964 [1926]. Orígenes del español: estado lingüístico de la península ibérica hasta el siglo XI. Madrid: Espasa-Calpe. Núñez, Manuel. 2003. ‘La lengua española, compañera de la nación dominicana: discurso de ingreso.’ Boletín de la Academia Dominicana de la Lengua 17, 137–195. Núñez, Manuel. 2001. El ocaso de la nación Dominicana. Santo Domingo: Alfa y Omega. Peguero, Valentina. 2004. The militarization of culture in the Dominican Republic: From captains general to General Trujillo. Lincoln: University of  Nebraska Press. Pérez, Odalís. 2002. La Ideología rota: el derrumbe del pensamiento pseudonacionalista dominicano. Santo Domingo: Editora Manatí. San Miguel, Pedro L. 2005. The imagined island: History, identity and utopia in Hispaniola. San Juan, Santo Domingo: Isla Negra, La Trinitaria. Sarlo, Beatriz. 2000. ‘Pedro Henríquez Ureña: lectura de una problemática.’ In: Abellán, José Luis and Ana Maria Barrenechea (eds), Pedro Henríquez Ureña: ensayos. Madrid: Allca XX, 880–889. Sepúlveda, Isidro. 2005. El sueño de la Madre Patria. Hispanoamericanismo y nacionalismo. Madrid: Marcial Pons. Toribio, Jacqueline. 2005. ‘Linguistic displays of identity among Dominicans in national and diasporic settlements.’ In: Davies, C, and J. Brutt-Grif f ler (eds), English and ethnicity. New York: Palgrave, 135–155. Valdez, Juan R. 2011. Tracing Dominican identity: the writings of  Pedro Henríquez Ureña. New York: Palgrave. Valdez, Juan R. 2009. ‘The iconization of  Dominican Spanish in Pedro Henríquez Ureña’s linguistic texts.’ Spanish in Context 6(2), 176–198. Woolard, Kathryn. 1998. ‘Language ideology as a field of inquiry.’ In: Schief felin, Bambi B., & Kathryn Woolard (eds), Language ideologies: Practice and theory. New York: Oxford University Press, 3–47.

Laura Villa

‘Because When Governments Speak, They Are Not Always Right’: National Construction and Ortho­ graphic Conf licts in Mid-Nineteenth-Century Spain

Abstract This paper analyses the polemics surrounding the discursive legitimation and political institutionalization of dif ferent spelling systems circulating in mid-nineteenth-century Spain. In 1843, teachers associated with Madrid’s Literary and Scientific Academy of  Primary Education developed a simplified orthography and began to implement it in schools. In response to this independent initiative, Queen Isabel II signed a Royal Decree in 1844 that mandated the exclusive use of Royal Spanish Academy’s orthography in Spain’s primary education. The Literary and Scientific Academy contested the imposition and took actions to oppose its implementation, by organizing meetings and publishing essays to defend both the simplified orthography and the legitimacy of  the institution. My study examines of ficial documents, textbooks, pamphlets, minutes, etc., which provide us with a record of this linguistic ideological debate that reveals a struggle over authority and power within the linguistic and educational markets.1

1

The reconstruction of  this salient episode in the linguistic history of  Spain is the result of documentary research carried out in Spain’s National Library and in the Library and Archive of the Royal Spanish Academy thanks to the financial support of the Program for Cultural Cooperation between Spanish Ministry of Culture and United States Universities, managed by the University of  Minnesota.

210

Laura Villa

1 Introduction In the mid-nineteenth century, the orthographic and grammatical norms of  the Royal Spanish Academy (henceforth RAE)2 were made of ficial. They were to be taught and used in the national school system which, at the time, was being developed in the broader context of the construction of a liberal Spanish nation (Álvarez Junco 2001; Burdiel 1998; Puelles Benítez 1999, 2004). In a larger project I look at the historical conditions that resulted in this of ficialization and the consequent promotion of the linguistic norms proposed by the RAE. The main objectives of  this project are, on the one hand, to describe the processes of selection and codification of Spanish as Spain’s national language, and the implementation of the resultant standard variety through the public school system; and on the other, to analyse the increasing relevance of  the RAE as the legitimate institution in charge of  the standardization of Spanish, as well as the resistant voices that challenged the constitution of  the RAE as Spain’s linguistic authority. The study is framed by an ef fort to understand linguistic polemics within the specific historical and cultural context of its production. This entails adopting a linguistic, political and historical approach that looks at the of ficialization of  Spanish norms – as well as the discourses surrounding its implementation in the emerging public school system – as part of a broader language planning process deeply involved in tensions over nation-building and the power of  the State. These tensions clearly appeared in the debate that surrounded the of ficialization of  the RAE’s orthography in 1844 and its imposition as the legitimate spelling norm in Spain’s education system, which is the focus of  this paper. The nineteenth century witnessed a number of similar orthographic debates over languages such as Czech,3 2 3

This language institution was created in 1713 after the Académie Française in order to ‘fix the voices and vocabularies of the Castilian language with propriety, elegance, and purity’ (). Czech’s spelling reforms in the nineteenth-century where consistent with a project of  linguistic and national revival. The reforms, guided by the Matice česká, a selfappointed linguistic institution in charge of  Czech corpus planning, found the opposition of important Czech linguists (Bermel 2007: 95–105).

‘Because When Governments Speak, They Are Not Always Right’

211

Dutch,4 German,5 Spanish,6 and Swedish, Danish and Norwegian,7 to name just a few. These orthographic controversies can be understood as language ideological debates because, one the one hand, in controversies over the spelling system ‘language is central as a topic, a motif, a target’ and ‘language ideologies are being articulated, formed, amended, enforced’ (Blommaert 1999: 1); and on the other, these orthographic debates ‘are part of more general sociolinguistic processes, […] sociopolitical developments, conf licts and struggles’ (2). This model allows us to emphasize the significance of the social and political contexts, as well as the ‘historical horizon of relationships of power, forms of discrimination, social engineering, [and] nation building’ (2), in which these linguistic controversies emerged. The analysis of  the polemics surrounding the legitimation of competing spelling systems in mid-nineteenth-century Spain will show that Spain’s orthographic debate ref lected larger socio-political controversies in the educational and national contexts. The standard language, codified 4

5

6

7

Rutten & Vosters in this volume study orthographic practices and discourses on orthography in nineteenth-century Flanders. I would like to thank Rik Vosters (VU Brussels) for a fruitful exchange of ideas and materials on the parallels, dif ferences and theoretical approaches to our objects of study. German’s lack of consensus resulted in the coexistence of several orthographic schools in dif ferent regions until the nineteenth century. The first attempt of orthographic standardization took place in 1876, right after German political unification ( Johnson 2005: 19–22). In addition to the controversial of ficialization in Spain that I analyse, there were simultaneous orthographic debates in other Spanish-speaking countries, such as Chile, Colombia and Mexico, where the cultural separation from the linguistic authority of  Spain – the former metropolis – plays a central role. The Chilean orthographic reform – planned by Andrés Bello (1781–1865) and approved, also in 1844, amidst an intense public debate – is central to my wider project for its many connections with Spain’s debate: besides a number of biographical links among the Chilean and Spanish orthographic agents, both spelling disputes shared a crucial tie to intense nation-building processes in which the education system was seen as a central instrument of  the State to the promotion of  the of ficial orthography (Contreras 1993; Narvaja de Arnoux 2008; Rosenblat 1951). The Scandinavian Orthographic Conference in Stockholm in 1869 recommended some reforms to be implemented in these three languages (Dalen 2005: 1406; Torp 2005: 1427–1431). I would like to thank Olle Josephson (Stockholm) for directing my attention to these Scandinavian orthographic debates.

212

Laura Villa

by and for the school apparatus, constitutes a special linguistic practice in which socio-political antagonisms are produced and reproduced (Balibar & Laporte 1976: 21). These were projected onto the discursive representations constructed by the actors involved in the standardization processes and related educational debates. In consequence, my analysis hopes to unveil the non-linguistic dimensions of the debate over the of ficialization of the RAE’s spelling system, by examining the linguistic ideologies proposed, exploited and reformulated in favor of and in opposition to of ficialization. I rely on the concept of linguistic ideologies as an analytical tool because of its potential, first, to explain ‘the tie of cultural conceptions to social power’ (Woolard 1998: 10), and second, to integrate the domains of language and politics, emphasizing the socio-political and economic interests embedded in linguistic matters (Kroskrity 1999: 2–3). In my analysis I focus on the discursive strategies of formulation, exploitation and reformulation of  linguistic representations, that is ‘systems of ideas that articulate notions of language with specific cultural, political and/or social formations’ (Del Valle 2007: 22). These linguistic representations are ‘related to practices and ideas that naturalize and normalize a particular socio-political order and legitimate a certain type of  knowledge that supports the exercise of power and authority’ (Del Valle 2007: 22). The of ficialization of  the RAE’s spelling system was followed by an intense debate between a group of primary school teachers that proposed and defended a simplified orthography in order to universalize literacy, and governmental agencies that sanctioned and promoted a traditional spelling norm with the aim of centralizing and retaining control over language and education. When read against the broader political and historical context, the discursive confrontation between these two groups reveals the ef forts made by the participants in the debate to naturalize dif ferent conceptualizations of the nation and the national education system through dif ferent strategies of representation of  the self and the other.

‘Because When Governments Speak, They Are Not Always Right’

213

2 The orthographic debate in mid-nineteenth-century Spain In 1843, teachers associated with Madrid’s Literary and Scientific Academy of Primary Education – an independent organization – developed a simplified spelling system and began to implement it in schools. In response to this independent initiative, the government, through the Council for Public Instruction – an advisory body created the same year to oversee the public education system – asked Queen Isabel II (1830–1904; reigned 1833–1868) to ban the reformed system from schools and to of ficialize the RAE’s orthographic norms. The Queen agreed and in 1844 signed a Royal Decree that mandated the exclusive use of  the RAE’s orthography in Spain’s primary instruction. The Literary and Scientific Academy of  Primary Education contested the imposition and took actions to oppose its implementation by moving the debate to the public sphere. They organized meetings and published essays to defend both the simplified orthography and the legitimacy of the institution. For my study I have examined of ficial documents, textbooks, pamphlets, minutes, etc. which provide us with a record of this linguistic ideological debate and reveal a struggle over authority and power within Spain’s linguistic and educational markets. The teachers of the Literary and Scientific Academy defended a ‘radical reform’ – their own words – that would produce a phonological spelling system ‘completely based on the principle that, since writing is the faithful representation of  the written word, its signs ought to be fixed and precisely determined’ (Academia 1844: 4a).8 Their orthographic ideal lay in the exact correspondence between sounds and phonemes and, in consequence, they defended the following reforms: first, the simplification of pairs of  letters that represented the same sound, such as, , , y , writing cinze instead of quinze, jente instead of gente and

8

All translations are mine. The version respecting the original orthography reads, in Spanish: ‘reforma radical y basada toda sobre el principio de que, siendo la escritura la representacion fiel de la palabra, deben ser fijos sus signos y esactamente determinados’.

214

Laura Villa

Lola i Juan instead of Lola y Juan; second, the substitution of simple letters and for digraphs and and of letter for before and

; third, the replacement of for the combination in intervocalic position and for in pre-consonantal position; fourth, the elimination of silent letters, such as or in the combinations y , writing onbre rather than hombre and giso rather than guiso; finally, the modification of  the names of some letters, in order to achieve a regularity in the designation of  the sounds, for instance, the letter would be renamed ‘gue’ (pronounced /γe/) instead of  the traditional ‘ge’ (/χe/), would be ‘que’ (/ke/) instead of  ‘ce’ (/θe/).9 In contrast, the RAE privileged an orthographic system based on the traditional criteria of etymology and usage, since ‘the Orthography of all languages is founded on two principles, namely, the origin of the voices and the alterations introduced in many of them by usage, which is the supreme linguistic arbiter when it becomes general and uniform’ (RAE 1845: ii).10 The disparities of  these proposals – a radical phonology-based reform versus an orthography founded on etymology and usage – are rooted in dif ferent traditions that were part of the linguistic culture of the nineteenth century: the ideas of  the Enlightenment versus traditional ideas derived from the study of  Latin (Mourelle de Lema 2002).11 For the members of  the teachers’ Academy, the rationality of a logical form of writing would be the main source of  legitimacy for their proposal. And they did in fact claim the following:

9

10 11

Pradel y Alarcón (1845: 7–9) of fers us a contrast between the traditional orthography defended by the RAE and the radical simplification of the primary school teachers. The former would write ‘el hombre que bebe mucho vino no se robustece’, the latter ‘el onbre ce bebe muho bino no se řobusteze’. ‘la Ortografía de todos los idiomas se funda en dos principios; á saber, el orígen de las voces y la alteracion que en muchas de ellas ha introducido el uso, que es el árbitro supremo de las lenguas cuando llega á hacerse general y uniforme’. The discussion on the principles that should drive the spelling system was by no means a new one. The history of the Spanish shows in fact a number of controversies over the spelling system as well as attempts to simplify Spanish orthography from the fifteenth century to the present day (Esteve Serrano 1982).

‘Because When Governments Speak, They Are Not Always Right’

215

where reason is in play there is no resorting to usage, where reason is in play there is no resorting to tradition, where reason is in play there is no other authority, there is no Academy [RAE], no Council for Public Instruction, no government. (Academia 1844: 28a)12

It is necessary to point out the fact that, in spite of their discrepancies, both teachers and government banked on the promotion of one orthographic norm through the education system, as a means of solving the problem posed by the coexistence of dif ferent spelling norms. This alignment with linguistic unity is consistent with a monoglossic ideology that, relying on the romantic correspondence between language and nation, sees linguistic homogeneity as an indispensable requirement for, and the symbolic realization of, national unity.13 In this sense both proposals negatively portrayed the coexistence of several orthographic varieties as a source of disorder and confusion, and discursively represented themselves as the solution to orthographic chaos. For instance, the Royal Decree that imposed the of ficialization of the RAE’s orthography explained the necessity of  the intervention because ‘from frequent alterations may result confusion and mistakes’ (Villalaín Benito 1997: 99–100).14 More passionately, a member of  the Literary and Scientific Academy, in a pamphlet entitled No more objections and obstacles to the instruction of the people!, stated: ‘So it is clearly seen how unreasonably they label us as orthographic anarchists, when in reality our project hopes to do away precisely with anarchy and not with order’ (Macias 1846: 27).15 12 13

14 15

‘donde hay razon no hay uso, donde hay razon no hay antigüedad, donde hay razon no hay autoridad, no hay Academia, no hay consejo de instruccion pública, no hay gobierno’. Del Valle (2000; 2006) explains that this ideology is grounded on the following assumptions: ‘that what linguistically characterizes an individual as well as a community is possession of a well defined and relatively stable grammar’ (2006: 119) and ‘that people’s linguistic behavior tends to become homogeneous over time through pressure from the dominant norm of  the community’ (120). ‘de frecuentes alteraciones puede resultar confusión y equivocaciones’. ‘Véase pues claramente con cuan poca razon se nos acusa de anarquistas ortográficos, cuando en la realidad á extinguir la anarquía y no el buen órden […] es á lo que camina nuestra empresa’.

216

Laura Villa

Both sides of  the debate have also in common the discursive representation of  their orthography as a good for the nation. Thus, the Royal Decree justified the imposition because ‘all nations always proceed with extreme caution in such a delicate matter, preferring the advantages of a fixed and uniform orthography understood by all’ (Villalaín Benito 1997: 99–100).16 For their part, the teachers’ association publicly defended that the simplified orthography is ‘illustrious, useful and beneficial not only for the good of  the children, but for the good of  the nation in general’ (Academia 1844: 38b).17 The previous description of the discourses for or against the orthographic simplification leads us to two important conclusions. On the one hand, both teachers and government agreed on the necessity of unifying the orthography ‘for the good of the nation’, relying on nationalist arguments in order to defend the unification of  the linguistic market. On the other hand, each actor involved in this controversy represented its proposal as having some value outside the market, presenting the spelling systems as more rational or more natural and, therefore, concealing the extra-linguistic motivation of the debate: a struggle to position themselves as the legitimate authority in a changing educational market. Earlier valuable studies of  this important episode in the history of  the standardization of  Spanish have followed this logic in their analysis: they described the linguistic characteristics of each proposal, linked the of ficialization to the growing power of  the central State, or evaluated the benefits of the simplified orthography in terms of the spread of education. Among the most relevant work on the of ficialization of the RAE’s orthography are Esteve Serrano (1982) and Martínez de Sousa (1991), who brief ly describe the episode in their histories of the Spanish orthographic reform, Martínez Alcalde (2010), who also mentions the of ficialization of  the RAE’s orthography in her study of the standardization of Spanish spelling, 16 17

‘Todas las naciones proceden siempre con suma circunspección en tan delicado punto, prefiriendo las ventajas de una ortografía fija, uniforme y comprendida por todos’. ‘eminente, útil y provechosa no solo al bien de la niñez, sino de la nacion en general’.

‘Because When Governments Speak, They Are Not Always Right’

217

Rosenblat (1951), a key study to understand the of ficialization as tied to the project of construction of a liberal Spanish nation, and Quilis Merín (2008), who links the orthographic simplification to the universalization of literacy intended by liberal political movements, in her examination of  the orthographic ideas in Dominguez’s Dictionario nacional. Despite the great interest of  these studies, they focus exclusively on the simplification carried out by the Literary and Scientific Academy before the of ficialization of  the RAE’s orthography, thus overlooking the teachers’ reaction to the enforcement of the Royal Decree. These works also failed to take into account the socio-political nature of  the debate. An in-depth analysis of primary source and original documentation will reveal an extra-linguistic dimension in the discourses displayed by the participants in this orthographic controversy.

3 The non-linguistic nature of  the orthographic debate Interestingly enough, and significantly for the outcome of  the debate, a great discursive ef fort was oriented to legitimate not the advisability of the orthography itself, but rather the authority of  the institutions that raised their voices in this linguistic-educational matter. The Literary and Scientific Academy put forward a pedagogical argument to defend their proposal: a simplified orthography, they claimed, would be learned faster and ‘written by everyone with more propriety and fewer mistakes’ (Academia 1844: 14b).18 The teachers grounded their legitimacy to intervene in linguistic matters in their experience educating both primary school children and adult workers, both men and women. One member of the educational association explained that the orthographic reform is a response to ‘a thousand and one troubles and inconveniences that only one dedicated to teaching

18

Emphasis in the original Spanish: ‘se escribe por todos con mas propiedad y menos equivocaciones’.

218

Laura Villa

can duly perceive’ (Academia 1844: 12a-b).19 Another teacher denounced the RAE’s orthographic manual imposed as the of ficial textbook in primary schools pointing out that the RAE’s reference book could by no means exceed ‘the many compendia that have been written by experienced teachers that touch and feel the dif ficulties caused by the anomalies in several letters of our alphabet’ (Hernando 1845: 3–4).20 If the Literary and Scientific Academy relied on pedagogical and professional arguments to defend the legitimacy of both their institution and their orthographic proposal, the government, in line with their project of a centralized public education system, used openly political arguments, represented the of ficialization of Spanish orthography as a national necessity and used its power to impose it. The governmental Council for Public Instruction turned to the Crown, the highest institution of political authority and symbolic power, and to the State apparatus, making political leaders and the Committees of Primary Instruction responsible for implementing the Royal Decree. With this of ficialization of  the orthography, linguistic matters became ‘a concern of the State’ (Rosenblat 1951: cxxiv): in Spain’s language policy and linguistic discourse, language unity turned into an instrument to maintain and promote the national consensus. After the Royal Decree, the orthography became a matter of national interest because it symbolically represented the nation, and its of ficialization came to be considered ‘an issue of extreme significance due to the serious damage that can be done by the misunderstanding caused by an impure orthography in important documents’ (Villalaín Benito 1997: 99–100).21 The Spanish language was therefore declared a political issue which, due to its centrality to the nation, should be managed by the Crown, the central government and the institutions tied to them, such as the Council for Public Instruction or the RAE. 19

‘mil fatigas é incomodidades, que solo puede apreciar debidamente el que está dedicado á la enseñanza’. 20 ‘lo mucho que hay escrito por profesores práticos en diferentes compendios […] los que tocan y palpan las dificultades que producen en la enseñanza las anomalías que contienen varias letras de nuestro alfabeto’. 21 ‘asunto […] de suma trascendencia por los graves perjuicios que puede acarrear en documentos importantes la equivocada inteligencia de lo escrito por efecto de una ortografía adulterada’.

‘Because When Governments Speak, They Are Not Always Right’

219

The public debate ref lected the political nature of  the orthographic controversy and the discourses displayed both in favor of and against the proposed reform were riddled with political representations of  the opponent: the teachers were labeled as linguistic anarchists and antipatriotic; they, in turn, represented the central government as absolutist and despotic. For instance, they accused the government of excessive authoritarianism: primary school teachers were forced to comply with the Royal Decree under threat of suspension. In a heated public meeting, organized by the Literary and Scientific Academy to discuss the advantages of  the phonological orthography, some teachers called for action against the ‘despotic agents’: ‘Let’s join forces, now is the time, to face our worries and even to confront power, since it rides roughshod over our rights and of fend against our honor’ (Academia 1844: 17a).22 One teacher present in that meeting even made the government responsible for the uprising of the people since ‘when governments speak, they are not always right […] and unfortunately at many times they impose restrictive precepts to good things […] whence their discredit as well as the frequent revolutions that we are going through spring up’ (Academia 1844: 30b).23 The linguistic ideological debate that surrounded the of ficialization of the Spanish orthography confirms that, as Woolard states: ‘[t]he definition of what is and what is not literacy is never a purely technical but always a political matter’ (Woolard 1998: 23). The discursive representation of  the self and the other, in this sense, ref lected a broader socio-political struggle. The exchange of accusations, such as the Literary and Scientific Academy’s ardent insistence on the government’s tyranny and despotism, or the government’s fervent proclamation of  the teachers’ whim, ignorance and irresponsibility, went beyond the linguistic terrain. What was at stake, I argue, was control over the educational market, that is, the governance of public and private educational spaces, institutions and legal competence.

22 23

‘Unámonos; que aun es hora, para hacer frente á la preocupacion, y aun al poder, cuando holle nuestros derechos ó ultraje nuestro honor’. ‘que cuando los Gobiernos hablan, no siempre tienen razon […] y que muchas veces por desgracia ponen preceptos restrictivos á lo bueno y […] de lo cual nace su descre­ dito y las frecuentes revoluciones por las cuales vamos atravesando’.

220

Laura Villa

I rely on Bourdieu’s (1991) work and terminology to emphasize the field tensions operating within those educational spaces as well as the struggles over authority within the market. Bourdieu’s model also accounts for the dif ferent value acquired, based on social conditions, by diverse linguistic practices (dif ferent spelling systems, for instance). Moreover I chose the term ‘educational market’ for its relation to the linguistic and labor markets, since the linguistic practices unfavorably valued in the first one could restrict access to the others. For instance, after the of ficialization of the orthography, teachers would be severely evaluated on this subject. Those who failed the exam would lose the authorization to practice their profession. The steps taken by the central government in order to promote a public national education brought about the rearrangement of  the educational market, a redistribution of power and, consequently, a struggle to manage the market, intensified by the severe centralization that the public education was undergoing. The Literary and Scientific Academy, an independent association which ‘hoped to gain the monopoly of  the Court’s public education’ (Gómez R. de Castro 1983: 50), fought to retain power over the educational space not monopolized by the Catholic Church. The government was also trying to control strategic markets in order to spread the national consensus. They realized that, according to the prominent intellectual Antonio Gil de Zárate (1796–1861), first director of  the Council for Public Instruction, also a member of  the RAE: ‘education is a matter of power: the one who teaches dominates; given that to teach is to form men, men adapted to the viewpoint of the one who indoctrinates’ (Gil de Zárate 1855: 117).24

24 ‘la cuestión de la enseñanza es una cuestión de poder: el que enseña, domina; puesto que enseñar es formar hombres, y hombres amoldados a las miras del que los adoctrina’.

‘Because When Governments Speak, They Are Not Always Right’

221

4 Linguistic standardization and the nation-building project The of ficialization of the RAE’s linguistic norms, a milestone in the standardization of Spanish, occurred throughout Queen Isabel II’s reign. During those decades the central government engaged in the construction of a liberal monarchic Spanish nation, projected since the Cadiz Constitution (1812) and slowed down during Fernando VII’s (1784–1833) absolutist rule (1814–1820 and 1823–1833).25 The legislative machinery of  Queen Isabel II was mobilized at the service of  the nation-building project, conceived during her reign from a moderate and traditionalist perspective in keeping with the conservative character of the majority of the governments formed since 1843. As part of  the development of  State power and institutions, a national centralist school system was developed in the mid-century.26 For example, the education system became fully regulated, through educational plans and legislative initiatives – such as the ‘plan Rivas’ in 1836, the ‘ley Someruelos’ for primary education in 1838 or the ‘plan Pidal’ for secondary and university instruction. The legal constitution of  Spain’s instruction culminated in the first comprehensive education law, the ‘ley Moyano’, approved in 1857 – that in its article 88 ratified the special status of the Royal Academy’s linguistic norms. The central government also took steps to create institutions, such as the abovementioned Council for Public Instruction, in charge of controlling public education since 1843. The processes of  State formation typically create the conditions for ‘the constitution of a unified linguistic market, dominated by the of ficial language’ which is ‘bound up with the state’ (Bourdieu 1991: 45). 25

I follow Álvarez Junco (2001), Burdiel (1998), and Carr (1969) for a historical contextualization of Queen Isabel II’s reign, as well as for a detailed description of political movements and nation-building processes during her sovereignty. 26 Puelles Benítez’s works (1999, 2004, and 2005) provide a precise account of  the significance of the public school system within the politics oriented to the centralization and growth of State power. The author concludes that a conservative turn in the educational project and a lack of investment in public instruction resulted in the failure of  the universalization of education in nineteenth-century Spain.

222

Laura Villa

Language seems to be an essential element in the promotion of the national feeling of  loyalty and solidarity to the national community that imagines itself as unitary thanks, in part, to the standard language spread through the print, the literature and the press (Anderson 1983). Education appears as a central mechanism in the spread of national consensus and the homogenization of the national community and, therefore, in the standardization process ‘that leads to the construction, legitimation and imposition of an of ficial language, the educational system plays a decisive role’ (Bourdieu 1991: 48). Earlier research on the rise of public school systems in the nineteenth century has shown that the State’s interest in public instruction is necessarily tied to the larger project of national construction, both in its most material dimension and in the formation of collective national ideologies that would legitimate State power (Green 1990: 77). The processes of  linguistic standardization in mid-nineteenth-century Spain are also embedded in nation-building processes: the selection, promotion and implementation of  the of ficial variety – and all related language policies – contributed to the consolidation of  State power in several ways. On the one hand, the standard language made it possible to spread a national school system that would disseminate public instruction and create citizens capable of exercising the rights and fulfilling the duties imposed by the State. On the other, the of ficial variety facilitated the consolidation of the State’s bureaucracy and legislation, by of fering the homogenization required to the viability of a central administration. And finally, the dissemination of (the idea of ) a language shared by all members of  the national community promoted a loyalty attachment to the nation, strengthening its symbolic representation. In this fight over the control and management of  the education, the RAE began to position itself as the institution of  linguistic and cultural reference, a position that this academy would eventually maintain until the present day. The privileged political situation that its most renowned members enjoyed at the particular time covered by my study contributed to the constitution of  the RAE as the language authority, as the legitimate prescriptive voice that dictates the norms of  linguistic correction. For instance, the director of  the RAE at the time of  the of ficialization, Francisco Martínez de la Rosa (1787–1862), was also the president of  the

‘Because When Governments Speak, They Are Not Always Right’

223

inf luential Ateneo Científico y Literario de Madrid – an exclusive private cultural club where the most significant intellectuals contributed to Spain’s scientific and cultural discussion since the beginning of  the nineteenth century – and occupied throughout his life several political and governmental positions, such as representative in the Cortes de Cadiz, president of  the Conservative party, Prime Minister, or president of  the Congress. The inf luence of  the RAE in educational of fices was also crucial: Gil de Zárate, who joined the cultural institution in 1841, held dif ferent of fices within the developing public school system, such as general director of  Public Instruction, or director of the Council for Public Instruction. He also played a central role in the redaction of the ley Moyano and his Manual de literatura española, composed between 1842 and 1844, contributed to the early institutionalization of Spain’s literary canon. The Council for Public Instruction, that alerted Queen Isabel II to the abuses of school teachers in orthographic matters, counted among their members with four academicians at the time of its creation: Manuel José de la Quintana (1772–1857), Eugenio de Tapia (1776–1860), Martín Fernández Navarrete (1765–1844) and RAE’s perpetual secretary Juan Nicasio Gallego (1777–1853) (Ceprián Nieto 1991: 437–439). The exploitation of  the royal status of  the institution and, in particular, of  the connections of fered by the political of fices held by its members would determine their ability not only to intervene in the State educational institution but also to access the public spaces of intellectual discussion.

5 Conclusion To conclude, my analysis understands language as a socio-cultural construction and aims at studying linguistic polemics within their specific historical and cultural context. I have presented an example of this historically situated approach to language, analyzing the linguistic ideological debate surrounding the of ficialization of  the RAE’s spelling system. The

224

Laura Villa

orthographic debate was certainly one of the most intense linguistic controversies in Spain during the nineteenth century: it reached the public sphere and transcended both the academic realm and the linguistic field. It was not, however, an isolated event. It was in fact embedded in the process of linguistic standardization of Spanish which, in the mid-nineteenth century, played a central role in the representation of and exclusion from the national community. The standard language, likewise, became an essential object, necessarily shared by the members of the national community, and therefore it was promoted through the development of ‘a free, mandatory state education system’ that was felt to be a requirement for building the nation-state, as well as a mechanism, together with the anthem, the f lag, the army and national history, to spread the national consensus (Álvarez Junco 2001: 545). The linguistic ideological debate between governmental agencies and the Literary and Scientific Academy ref lected an intense struggle to gain or retain authority over educational matters in the rearrangement of  the market brought about by the centralization of instruction. As previously discussed, the battle over control of  the linguistic and educational markets bent to the central government and, in particular, to the RAE, whose orthography and grammar were ratified, by the celebrated ley Moyano, as the of ficial textbooks both for primary and secondary schools. The centrality of  the orthographic episode derives from the fact that it shows that the rise of  the Spanish orthographic system that is still in use today, responded not to a strictly linguistic process but rather to the complex interaction between linguistic practices and social and political institutions. Our approach aims at bringing to the forefront the sociopolitical dimension of  the history of  Spanish, by attending to those salient moments that ref lect the dynamic intersection between linguistic, cultural, social and political issues and help us to understand language policies and linguistic authority in contemporary Spain.

‘Because When Governments Speak, They Are Not Always Right’

225

References Academia Literaria i Zientifica de Instruczion Primaria. 1844. Sesion Publica celebrada el dia 3 de octubre de 1844, en el salon del instituto español. Por la Academia de profesores de primera educación, para demostrar las ventajas que ofrece la reforma de ortografía adoptada y publicada por la misma Academia. Dedicada a los profesores y amantes de la educación. Madrid: Imprenta de Dª. Francisca Estevan. Álvarez Junco, José. 2001. Mater dolorosa. La idea de España en el siglo XIX. Madrid: Taurus. Anderson, Benedict. 1983. Imagined Communities. London: Verso. Balibar, Renée, & Dominique Laporte. 1976. Burguesía y lengua. Barcelona: Avance. Bermel, Neil. 2007. Linguistic Authority, Language Ideology, and Metaphor. The Czech Orthographic Wars. Berlin: Mouton de Gruyter. Blommaert, Jan. 1999. Language Ideological Debates. Berlin: Mouton de Gruyter. Bourdieu, Pierre. 1991. Language and Symbolic Power. Cambridge: Harvard Univ. Press. Burdiel, Isabel. 1998. La política en el reinado de Isabel II. Madrid: Marcial Pons. Carr, Raymond. 1969. España 1808–1939. Barcelona: Ariel. Ceprián Nieto, Bernardo. 1991. Del Consejo de Instrucción Pública al Consejo Escolar del Estado: origen y evolución (1836–1986). Madrid: UNED. Contreras, Lidia. 1993. Historia de las ideas ortográficas en Chile. Santiago de Chile: Editorial Universitaria. Dalen, Arnold. 2005. ‘Sources of Written and Oral Language in the 19th Century.’ In: Bandle, Oscar (ed.), The Nordic Languages. An International Handbook of the History of  the North Germanic Languages. Berlin: Walter de Gruyter, 1406–1419. Del Valle, José. 2000. ‘Monoglossic Policies for a Heteroglossic Culture: Misinterpreted Multilingualism in Modern Galicia.’ Language and Communication 20, 105–132. Del Valle, José. 2006. ‘US Latinos, la hispanofonía and the Language Ideologies of  High Modernity.’ In: Mar-Molinero, Clare, & Miranda Stewart (eds), Globalization and Language in the Spanish-Speaking World: Macro and Micro Perspectives. New York: Palgrave MacMillan, 27–46. Del Valle, José. 2007. ‘Glotopolítica, ideología y discurso: un marco conceptual para el estudio de la difusión del español.’ In: Del Valle, José (ed.), La lengua, ¿patria común? Frankfurt/Madrid: Vervuert/Iberoamericana, 13–29.

226

Laura Villa

Esteve Serrano, Abraham. 1982. Estudios de teoría ortográfica del español. Murcia: Universidad de Murcia. Gil de Zárate, Antonio. 1855. De la Instrucción pública en España. Madrid: Impr. del Colegio de Sordomudos. Gómez R. de Castro, Federico. 1983. ‘La resistencia a las innovaciones. Informe de la Academia de Profesores de Primera Educación.’ Historia de la educación 2, 49–53. Green, Andy. 1990. Education and the State Formation: The Rise of Education Systems in England, France and the USA. Basingstoke: Macmillan Press. Hernando, Victoriano. 1845. Impugnación razonada en contra del prontuario de ortografia castellana que de Real Orden ha compuesto la Academia de la lengua española, con arreglo a su ultimo diccionario, para uso de las escuelas publicas. Madrid: Imprenta de D. Victoriano Hernando. Johnson, Sally. 2005. Spelling Trouble? Language, Ideology and the Reform of German Orthography. Clevedon: Multilingual Matters. Kroskrity, Paul V. 1999. Regimes of Language: Ideologies, Polities, and Identities. Santa Fe, NM: School of  American Research Press. Macias, Felipe Antonio. 1846. ¡No mas trabas ni obstáculos ä la instruccion del pueblo! ¡Abajo! (Entre las clases sin pretension de eruditas). Ortografía irracional. Debate lógico, sobre las diferentes anomalias de la ortografía castellana, y sobre la conveniencia ë inconveniencia de su proyectada reforma. Para servir de aditamento crítico, al Manual completo de instruccion primaria elemental y superior de D. Joaquin Avendaño. Bilbao: Imprenta y litografía de Delmas é hijo. Martínez Alcalde, María José. 2010. La fijación ortográfica: norma y argumento historiográfico. Bern: Peter Lang. Martínez de Sousa, José. 1991. Reforma de la ortografía española. Estudio y Pautas. Madrid: Visor. Mourelle de Lema, Manuel. 2002. La teoría lingüística de la España del siglo XIX. Madrid: Grugalma Ediciones. Narvaja de Arnoux, Elvira. 2008. Los discursos sobre la nación y el lenguaje en la formación del Estado (Chile 1842–1862). Buenos Aires: Santiago Arcos. Pradel y Alarcón, Francisco. 1845. Juicio crítico acerca del prontuario de Ortografía de la lengua castellana dispuesto de real órden para el uso de las escuelas públicas por la Real Academia Española, y sobre su impugnacion por los que prestenden la reforma del alfabeto. Madrid: Imprenta del Colegio de Sordo-Mudos y Ciegos. Puelles Benítez, Manuel. 1999. Educación e ideología en la España contemporánea. Madrid: Tecnos. Puelles Benítez, Manuel. 2004. Estado y educación en la España liberal (1809–1857): un sistema educativo nacional frustrado. Barcelona: Pomares-Corredor.

‘Because When Governments Speak, They Are Not Always Right’

227

Puelles Benítez, Manuel. 2005. ‘El Consejo Escolar del Estado: antecedentes, evolución y situación actual.’ In: Revista del Consejo Escolar del Estado: La participación de la comunidad educativa en el debate de la LOE 44, 44–49. accessed 18 August 2011. RAE. 1844. Prontuario de Ortografía de la lengua castellana dispuesto de Real Órden para el uso de las escuelas públicas por la Real Academia Española con arreglo al sistema adoptado en la novena edición de su diccionario. Madrid: Impr. Nacional. RAE. 1999. Ortografía de la lengua española. Madrid: Espasa. . Rosenblat, Ángel. 1951. ‘Las ideas ortográficas de Bello.’ In: Bello, Andrés, Obras completas. V. Caracas: Ministerio de Educación, 9–138. Torp, Arne. 2005. ‘The Nordic Language in the 19th Century I: Phonology and Orthography.’ In: Bandle, Oscar (ed.), The Nordic Languages. An International Handbook of  the History of  the North Germanic Languages. Berlin: Walter de Gruyter, 1425–1436. Villalaín Benito, José. 1997. Manuales escolares en España. Legislación 1812–1939. Madrid: UNED. Woolard, Kathryn. 1998. ‘Introduction: Language Ideologies as a Field of Inquiry.’ In: Schief felin, Bambi, Kathryn Woolard & Paul V. Kroskrity, Language Ideologies: Practice and Theory. New York: Oxford University Press, 3–50.

Gijsbert Rutten and Rik Vosters

As Many Norms as There Were Scribes? Language History, Norms and Usage in the Southern Netherlands in the Nineteenth Century

Abstract Contrary to the fixed language norms in the Northern part of  the Dutch language area, the Southern provinces are commonly said to have known as many orthographical and linguistic norms as there were scribes. According to our analysis of a large number of orthographies and grammars, however, there seems to have been a vivid normative tradition, from which a shared body of linguistic norms can be deduced. In comparing these norms with a corpus of early nineteenth-century manuscripts, we argue against the traditional view of orthographical chaos in the South and even suggest a considerable and increasing dif fusion of  the Northern language norms.

1 Introduction The present chapter discusses the linguistic situation in the Southern Netherlands1 in the eighteenth and nineteenth centuries from a historicalsociolinguistic perspective. Whereas a lot of research has been done on 1

Nowadays, roughly speaking, the Dutch-speaking part of Belgium. For stylistic reasons, we use ‘Flanders’ along with ‘the South’, ‘the Southern Netherlands’, etc. We are of course aware of  the possible anachronism in using ‘Flanders’ to refer to the Southern Netherlands as a whole, that is, the entire Dutch-speaking part of Belgium.

230

Gijsbert Rutten and Rik Vosters

the history of  Dutch linguistics in the Northern Netherlands, the South has gained far less attention, and in so far as historical and/or linguistic research has been done, language ideological myths often seem to have been involved. Two such myths are distinguished in this article: the manynorms myth (section 3.1) and the orthographical-chaos myth (section 3.2). The first refers to the idea that eighteenth and early nineteenth-century grammatical texts from Flanders display a wide variety of norms and do not constitute a coherent normative tradition at all. The latter refers to the idea that actual writing practices in this period are characterized by orthographical chaos. Both myths are reassessed and rejected. First, a vast and coherent Southern normative tradition is identified (section 4.1). Secondly, orthographical practices appear to have been fairly regulated (section 4.2). The latter conclusion is drawn on the basis of a corpus research using judicial and administrative documents from the 1820s. In general, we argue that any interpretation of  the historical-linguistic situation should be founded upon empirical studies of  that situation. The discussion and demythologizing of  the two myths is preceded by an introduction to the historical-sociolinguistic context (section 2.1), which brief ly addresses the political circumstances, and on which the language ideological background of  the myths is based (section 2.2).

On the other hand, Willems (1824) already used the term Vlaemsch (‘Flemish’) in this general sense, and whenever we refer to the areas historically belonging to the Duchy of Flanders, we will indicate this by referring to the actual provinces concerned (East Flanders and West Flanders).

As Many Norms as There Were Scribes?

231

2 History and linguistic history 2.1 Historical-sociolinguistic context Following the final defeat of Napoleon Bonaparte in 1815, the Low Countries were united after more than two centuries of separation. The origin of  this split between Southern and Northern Netherlands must be dated over two centuries earlier, with the revolt against Spain at the end of  the sixteenth century. The North became the independent Republic of  the Seven United Provinces and began its so-called Golden Age. The South remained under Spanish (Habsburg) control, and from 1714 onward under Austrian (still Habsburg) rule. After the French invasion in the 1790s, the South was incorporated into France, while the North eventually became a vassal kingdom under Napoleon’s brother Louis Napoleon. After the fall of  Napoleon, the European superpowers decided to create an enlarged Dutch buf fer state to the north of France. This led to the United Kingdom of the Netherlands (UKN) in 1815, which united the present-day Dutch, Belgian and Luxembourgish territories under the reign of  King William I. The UKN only existed for about fifteen years, when the Belgian Revolution ended it, with the birth of the Kingdom of Belgium in 1830. These fifteen years together, however, are traditionally considered to have been crucial for the future of  the Dutch language in Flanders: Without this brief family reunion, Belgium would probably have become a francophone nation. […] The pioneers of the Flemish movement, those who fought for the Dutchification of Flanders after 1830, were all educated during the United Kingdom of  the Netherlands. (De Vries et al. 1993: 117; our translation)

From the sixteenth century onwards, language standardization took place in the Dutch language area, especially in the Northern Netherlands in and around the province of Holland and its main city Amsterdam. A preliminary written standard was created in the seventeenth century (macro and micro selection, first codification), maintained in the eighteenth century (further micro selection and codification), and transformed into

232

Gijsbert Rutten and Rik Vosters

nineteenth-century written Standard Dutch (codification and elaboration) (Van der Wal 1995, Van der Sijs 2004, Van der Wal & Van Bree 2008). The nineteenth century soon bore two major results of  this standardization: an of ficial orthography (Siegenbeek 1804) and an of ficial grammar (Weiland 1805), endorsed by the state and to be used in the administrative and educational domains. Although this received view of  the development of (Standard) Dutch is teleological, too general and unjustifiably disregards language variation (Van der Wal 2006), it does provide us with a clear view of  the development of written Dutch in the Northern Netherlands, which eventually led to present-day Standard Dutch. In this view, the strong eighteenth-century normative tradition plays a crucial rule, as it is held responsible for guiding the preliminary seventeenth-century written standard of  Holland into the nineteenth century (Rutten 2009a). The reverse of  this preoccupation with language standardization in the North is that insight into Southern Dutch of  the period is largely still lacking. Linguists have mainly focused on the North, and some even assume that, in a history of  Standard Dutch, Southern Dutch has had no role to play after the sixteenth century. Consider Van der Sijs (2004: 53; our translation): Because of  the political circumstances, the Southern Low Countries did not contribute to the standard language any more after the fall of Antwerp in 1585. For this reason, this book will essentially not deal with Southern Dutch from the seventeenth century onwards.

Parallel writing traditions and emerging standardization in the South have not received much attention. At the same time, it is acknowledged that some normative linguistic works were published in the Southern Netherlands during the eighteenth century, but these are not considered to constitute a normative tradition; on the contrary, they are criticized for their internal dif ferences in language norms (Smeyers 1959, cf. section 3.1). Furthermore, although we know a fair amount about the status of  the language at the time, especially with regard to the opposition of  Dutch and French (e.g. Smeyers 1959, De Ridder 1979, De Groof 2004), almost no research has been done into the actual form of  Dutch in the Southern Netherlands.

As Many Norms as There Were Scribes?

233

2.2 The image of  the eighteenth century in the South For our research into the linguistic situation in the South during the UKN (1815–1830), a look into the Southern Dutch linguistic situation of  the eighteenth and early nineteenth centuries is indispensable. The image provided by historians in general of especially the eighteenth century has been quite negative for a long period of  time (Hanou 2004, De Vries 2004). Illustrative of this is the image of the Southern Netherlands in the middle of  the eighteenth century drawn up by Elias (1963: 106; our translation): The intellectual life in the entire Southern Netherlands […] around 1750 of fers us a view of the most barren landscape one can imagine. There was simply nothing. There was the most complete silence in the deepest intellectual poverty.

Linguistically, this historical image of the eighteenth century is paralleled by the ‘myth of eighteenth-century language decay’ (Van der Horst 2004): the idea that Southern Dutch, as opposed to Northern Dutch, did not show standardization but dialectization, a regression to locally defined varieties. In this vein, Wils (1958: 527–528; our translation) contrasts Northern uniformity with Southern diversity: By the end of the seventeenth century in the North, the colorful diversity in writing slowly yielded to a uniform written language, based on the good usage of  the classic authors […] The language in the South had undergone a dif ferent development from the seventeenth century onwards, [and] tended to regress to its purely local character.2

The myth of eighteenth-century language decay can be traced back to contemporary comments on the state of  the language. Especially during the early years of the UKN, those advocating a closer connection between Southern Dutch and Northern Dutch, for political reasons, had good cause

2

Although there is no doubt that Northern practice was not as uniform as claimed here – Siegenbeek’s (1804) spelling and Weiland’s (1805) grammar were only obligatory in the administration and education – we will have to leave this additional ‘myth of  Northern uniformity’ for another occasion; cf. Vosters et al. (2010).

234

Gijsbert Rutten and Rik Vosters

to uphold the image of eighteenth-century Flanders as an intellectual wasteland (cf. Elias 1963). Consider Jan Frans Willems’ (1793–1846) comments on the eighteenth-century linguistic situation (1819: 34–35): [F]lemish spelling has not been fixed to the level of a general Flemish standard by anyone up to the present. […] [E]ach schoolteacher in the Southern provinces […] considers himself qualified to teach the children whatever language rules his whim might have dictated him. Anarchy is a serious evil, both in spelling and in politics.3

Willems advocated political and linguistic unity with the North. The political context induced commentators such as him to describe the recent past as negatively as possible. A similar ref lex can be witnessed in contemporary histories of the Southern Netherlands and Belgium in which the past functions as a long and dif ficult period of ‘slavery’, that is, of foreign rule, which only ended during the UKN, and especially from 1830 onwards (Peeters 2003). Van der Horst (2004: 73; our translation) also explains the myth of eighteenth-century language decay by referring to its rhetorical function in nineteenth-century linguistic debates: ‘By emphasizing that the South had no tradition of its own, no basis, no language culture, nothing, they strengthened their argument in favor of a closer connection to Northern Dutch’. Building on this myth of eighteenth-century language decay as proposed by Van der Horst (2004), we distinguish two (mis-)conceptions in the traditional literature, two ‘sub-myths’: first, the many-norms myth, as opposed to the vivid normative tradition of  the North; and secondly, the orthographical-chaos myth, as opposed to the assumed fixed written language in the Northern Netherlands.

3

‘[D]e Vlaemsche spelling [is], tot heden toe, nog door niemand op vaste gronden van algemeenen Vlaemschen aerd gebracht is. […] [E]lke schoolmeester, in de Zuidelyke Provincien, […] acht zich bevoegd om den kinderen alzulke taelwetten voorteschryven, als hem door het hoofd zyn gewaeid. Anarchie is een erg kwaed, zoowel in de spelling, als in de regering’.

As Many Norms as There Were Scribes?

235

3 The language myths 3.1 The many-norms myth We already mentioned J.F. Willems’s strong claims about the state of Southern Dutch at the end of the eighteenth and start of the nineteenth century. Concerning the body of normative publications in the South, he furthermore explains how he spent many years going through all the books dealing with language and spelling, only to af firm what most already believed to be the case: ‘there are no Flemish orthographies or grammars of any lasting authority’,4 let alone a fully f ledged normative tradition. To further prove his point, Willems discusses several grammars and orthographies which had been available since the early eighteenth century,5 emphasizing inconsistencies and divergent opinions on specific orthographical points. Concluding that Flemish grammarians all adhere to dif ferent norms, Willems’s solution is simple: Flemings should point their gaze northwards, and at least partially adapt to the Northern Dutch orthographical norms (Vosters 2009). Many twentieth-century historians echo Willems’s claims. Sluys (1912: 53; our translation) speaks about ‘the greatest possible confusion’ in normative publications, with every author adhering to a dif ferent spelling system. Concerning the work of Des Roches (1735/1740–1787), no doubt the most authoritative of  the eighteenth-century Southern grammarians, he even concludes ‘[n]either his grammar nor his orthography were followed by anyone’ (Sluys 1912: 53; our translation). De Vos (1939: 50–52; our translation) follows suit, using phrases such as ‘mind-numbing drudgery’ to describe most of the eighteenth-century normative works. He also lashes out at Des Roches, who supposedly wanted to promote the Antwerp dialect to the level of a literary language. 4 5

‘Er bestaen […] geene vlaemsche Spel- en Spraekkunsten van doorgaende gezag’. For the second half of  the 18th century, he discusses P.B. (1757), the second edition of  Verpoorten (1759), Des Roches [1761], Van Belleghem & Waterschoot [1773], Ballieu (21792), a fifth print of  Ter Bruggen (11817, 51822), and several other minor works and/or reprints. He also mentions Vaelande [=Van Daele] (1805/1806) and Behaegel (1817), but does not discuss their work in more detail.

236

Gijsbert Rutten and Rik Vosters

Even balanced accounts such as Smeyers (1959: 112), who should certainly be praised for calling attention to the eighteenth-century codifiers and their grammars and orthographies, clearly states that none of the pre1815 grammarians ever strove for a uniform spelling, and that they all had dif ferent linguistic opinions depending on whichever dialect they spoke. After discussing a significant number of normative texts from the South, Smeyers concludes that most grammarians did nothing to contribute to a way out of ‘the maze of orthographical lawlessness’, and that the only thing bringing them together was their obsession with purism and fighting of f  loan words (Smeyers 1959: 127–128; our translation). In sum, the idea that the South lacked a proper grammatical tradition and that every grammarian constructed his own idiosyncratic spelling system is defended by Willems (1824), and taken over by historians dealing with the issue in the twentieth century.6 3.2 The orthographical-chaos myth The second myth under discussion is built around the idea of orthographical chaos in Flanders up until the UKN – not just the lack of a normative tradition, but spelling chaos in actual writing practice. ‘Try to read a hundred dif ferent […] books,’ the grammarian P.B. claims in 1757, ‘and you will find a hundred dif ferent spellings’ (P.B. 1757: 3; our translation).7 Half a century later, his much younger colleague Pieter Behaegel (1817: 250; our translation) repeats this lamentation, suggesting that there are ‘almost as many ways of spelling, as there are people who worked on improving the spelling’.8

6

7 8

The fact that Sluys (1912), De Vos (1939) and Smeyers (1959) all base themselves on the claims of J.F. Willems can already be demonstrated by the choice of publications under discussion, which De Vos supplements with the anonymous Inleyding from Dendermonde (1785), and Smeyers completes with Van Boterdael (21776, 31785). ‘[W]ant leést honderd verscheyde schriften, zelfs boeken, gy zult honderd verscheyde spellingen vinden’. ‘Men ziet in onze landstreéken bynae zoo veel wyzen van spellen; als er verscheydene persoónen zyn, die zich op het verbeteren der spelling toegelegd hebben’.

As Many Norms as There Were Scribes?

237

Especially during the period of the UKN, this idea of linguistic decay during the preceding Austrian and French rule becomes commonplace, and the image persists well into the twentieth century. De Vos (1939: 45) mentions that decades of foreign control prevented the Dutch language in the South from manifesting itself as a civilized language of culture, which made it shrivel to the level of patois. Deneckere (1954: 326) takes this still one step further, denying the existence of supraregional written varieties by claiming that even administrative documents were hardly intelligible from one town to another. Also in the 1950s, but repeating these claims more recently,9 Wils describes how, under the French rule, ‘Southern Dutch withered and weakened to such an extent that all contact with Northern Dutch threatened to be lost’ (Wils 1956: 529; our translation). Along the same lines as Deneckere (1954), Wils (1956: 530; our translation) is also very clear in reporting that ‘Flemish dialects and spellings were still being used in school books, in courts of justice and in notary deeds, and in the administration’ during the early years of  the UKN. Maybe the best summary of  this viewpoint comes from Suf feleers (1979: 19; our translation): ‘As opposed to the relative uniformity in the written language of the North, absolute chaos ruled the South’.

4 Revisiting the norm-myth 4.1 The many-norms myth revisited Do the eighteenth and early nineteenth centuries indeed display normative chaos in the South? Was there no real normative tradition but, at the most, a relatively small collection of grammatical works characterized by normative variation, as even Smeyers (1959) implied? It should be noted beforehand that spelling dif ferences between authors of grammars and orthographies

9

Cf. Wils (2003: 33): ‘Voor een hele generatie werd het onmogelijk een Nederlandstalige cultuur te verwerven, zodat de taal verschrompelde en verarmde’.

238

Gijsbert Rutten and Rik Vosters

in itself, as well as an interest in purist activities, do not prove the absence of a vivid normative tradition. On the contrary, both characteristics also apply to the seventeenth- and eighteenth-century linguistic situation in the North, which is commonly described as heavily prescriptive (Knol 1977: 70–71). A steady stream of publications containing diverging opinions only proves the existence of a linguistic debate, not its absence. Moreover, on closer examination there appears to have been a vivid normative tradition in the South, and the norms proposed are not at all based upon local dialect features. After two early works from the first half of the century,10 it is precisely from 1750 onwards – when Elias (1963) envisioned an intellectual wasteland – that several linguistic publications have come down to us. In the 1750s and 1760s, three Antwerp grammarians laid the foundation of the Southern normative tradition of the later eighteenth and early nineteenth centuries: Jan Domien Verpoorten (1706–1773), P.B. (?-?) and Jan Des Roches. In 1752, the schoolteacher Verpoorten published the first edition of  his Woôrden-schat, oft letterkonst, ‘Vocabulary, or grammar’.11 The greater part of the book consists of lists of loan words, mainly of French or Latin origin, with Dutch equivalents. Verpoorten also brief ly discusses some spelling issues. Verpoorten’s ‘new manner of writing’, as he proudly calls it, among other things has to do with getting rid of ‘superf luous’ consonants in consonant clusters representing only one sound. We summarize: [k] which is commonly spelled in auslaut and which should be spelled e.g. ik ‘I’ instead of ick; [γ] which is commonly spelled in anlaut and which should be spelled e.g. geven ‘give’ instead of gheven.

These kind of spelling proposals are not in any way related to the dialect of  Antwerp. Instead, these are innovations already put forward in the seventeenth- and eighteenth-century normative tradition in the North, as well

10 11

E.C.P. (1713), see Dibbets (2003); Stéven (1714), see Rutten (2009c). The discussion of Verpoorten (1752, 1759) and Bincken (1757) is based upon Rutten (2009b).

As Many Norms as There Were Scribes?

239

as in the scarce early eighteenth-century grammars from the South (E.C.P. 1713, Stéven 1714). Verpoorten is just linking up with and re-implementing common orthographical innovations. Verpoorten’s few pages on orthography are rather basic and mainly contain examples. Similar orthographical proposals are put forward five years later by P.B. in his Fondamenten ofte grond-regels der Neder-duytsche spel-konst, ‘Foundations or basic rules of  Dutch orthography’ (1757). P.B. criticizes schoolteachers who only present examples of correct spellings without explaining the rules, but he does not mention Verpoorten. In 1759, however, Verpoorten publishes the second edition of  his Woôrdenschat, in which he extensively elaborates on all kinds of grammatical rules, orthographical as well as morphological, without mentioning P.B., and at the same time constantly referring to the Northern normative tradition. An appeal to Northern norms apparently strengthened one’s proposals. It seems that P.B. and Verpoorten were competitors, linguistically as well as commercially on the schoolbook market. They take part in an implicit yet lively linguistic discussion that rapidly changed from fairly basic orthographical and lexical (purist) matters into a broader linguistic approach. This broader approach is further developed by another Antwerp schoolteacher. In 1761, Des Roches published the Nieuwe Nederduytsche spraek-konst, ‘New Dutch grammar’. Contrary to his predecessors, Des Roches does not limit himself  to spelling, loanwords and some morphological issues, but writes a full grammar of  Dutch, based on the Northern tradition, texts from the Latin and French grammatical traditions, as well as his Southern colleagues P.B. and Verpoorten, albeit without mentioning them (Rutten 2009d). Des Roches’ grammar is the first Southern grammar for decades and counts as one of the most important contributions to the codification of  Dutch in the South throughout the eighteenth century. These three Antwerp grammarians of the 1750s and 1760s were aware of and reacted to each other’s works. They proposed similar grammatical rules and presumably taught these rules in their classes. Furthermore, for our research concerning the period of the UKN, it is important to remark that this Southern normative tradition survived into the later eighteenth and early nineteenth centuries as well. Verpoorten published a third edition in 1767. The books by P.B. and Des Roches were printed over and

240

Gijsbert Rutten and Rik Vosters

over again, well into the nineteenth century, and especially of  P.B., many (partial) pirate editions appeared. In the last decades of  the eighteenth century and at the beginning of  the nineteenth century, the normative tradition was not only continued but intensified as well: dozens of works were published in the whole of the Southern Netherlands, which are always concerned with orthography and pronunciation, but often also with other grammatical features.12 The potential spread of  these works was wide, as we know that most primary schools in the later eighteenth century owned a grammar, along with reading matters and a catechism (Put 1990: 202). Besides, nine out of ten elementary schools for boys in the city of Leuven of fered orthography as a separate subject according to a 1795 survey (Put 1990: 208). However, the fact that there was a vivid normative tradition in itself does not imply that it was coherent as well. Therefore, we tried to distill language norms from this vast body of normative works. We selected several recurrent features and made an inventory of the prescribed use in the grammars, a selection of which will be shown below. The choice of  these features depended on their importance in the nineteenth-century spelling debates, in which the two most important spelling options for every feature were divided into a typically ‘Southern’ and a typically ‘Northern’ variant (Bormans 1841; see also Vosters 2009). The features are the following: (1) dotted or undotted [ei], e.g. wijn or wyn ‘wine’; (2) the second element in the diphthongs [ei] and [œy], either or , e.g. klein or kleyn ‘small’, and bruin or bruyn, ‘brown’; (3) vowel lengthening, either by adding an or by doubling the original vowel, e.g. zwaard or zwaerd ‘sword’ (with [a:]), zuur or zuer ‘sour’ (with [y]); (4) the form of the definite and indefinite article in the nominative singular masculine form: spelled with or without a final , e.g. de man or den man ‘the man’, een man or eenen man ‘a man’; (5) the so-called superf luous letters: or in anlaut, or in auslaut, e.g. ik or ick ‘I’, and gheven or geven ‘give’.

12

E.g. Van Belleghem & Waterschoot [1773], Janssens (1775), Van Boterdael (1785), Ballieu (21792), Van Aerschot (1807), De Neckere (1815), Henckel (1815), Behaegel (1817), Gyselynck (1819). See also Table 1 and the references.

As Many Norms as There Were Scribes?

241

We examined these features in a selection of normative works from the South. The results can be seen in Table 8. The features are displayed horizontally. Vertically, the names of  the authors are given, along with the date of  the used edition and its origin. The bottom line gives the of ficial Northern spelling rule of Siegenbeek (1804). When the prescribed use in one of the Southern grammars coincides with this Northern rule, the feature is shown in boldface. Note that and are generally accepted and that every grammarian in North and South rejects and – these features cannot be considered to be either typically Northern or Southern. As can be deduced from the table, most Southern grammars before 1815, including the early eighteenth-century works as well as the Antwerp grammarians of  the 1750s and 1760s dif fer from the Northern 1804 rule on nearly all of the selected features. More importantly, it is clear that there is almost complete general agreement in the South. In other words, there was a vast Southern normative tradition in which spelling choices were to a great extent identical. However, in the period of the UKN both writing traditions collide. The Southern tradition still continues for a while, but the Northern norms are gradually being brought to the fore, sometimes only as alternatives (e.g. Cannaert 1823), often as the only norm. In sum, we have argued that a distinct Southern orthographical tradition existed, as can be seen from the implicit debates among grammar writers, and from the coherent normative framework put forward by dif ferent grammars and orthographies published especially in the second half of the eighteenth century.

1757 1761 1792 1807

P.B.

Des Roches

Ballieu

Van Aerschot

Antwerp

1714 1773 1775 1785 1792 1805

Stéven

Van Belleghem & W

Janssens

Anon. [Inleyding]

Van Boterdael

Vaelande

-Flanders

y

y

y

y

y

y

ij

y

y

y

y

y

y/ij

Author

Anon. [Grond-regels]

Origin

Brabant/

1817

Year

y

y/ij

-y

-y/-i

-y

-y

-y

-i

-y

-y

-y

-y

-y

-y

-y

-y

-y/-i

During the United Kingdom of the Netherlands

1713

East/West- E.C.P.

1752

Verpoorten

Brabant/

Year

Author

Origin

Before the United Kingdom of the Netherlands

V+e

V+e/V+V

V+e

V+e

-n

-n/-ø

-n

-n

-n

-n

V+e V+e

-n

-n

-n

-n

-n

-n

-n

-n

-n/-ø

V+e

V+e

V+V

V+e

V+e

V+e

V+e

V+e

V+e/V+V

Table 8  Spelling features and the normative framework

g-

gh-/g-

g-

g-

g-

g-

g-

g-

g-

g-

g-

g-

g-

g-

gh-/g-

-k

-ck/-k

-k

-k

-k

-k

-k

-k

-k

-k

-k

-k

-k

-k

-ck/-k

242 Gijsbert Rutten and Rik Vosters

Ter Bruggen

1822

y

-y

V+e

-n

g-

-k

Zilgens

1824

y

-y

V+e

-n

g-

-k

Willems

1824

y

-i

V+e

-ø

g-

-k

East/West- De Neckere

1815

y

-y

V+e

-n

g-

-k

Flanders

Henckel

1815

ij

-i

V+e

-n

g-

-k

Gyselynck

1819

y

-y

V+e

-n

g-

-k

Cannaert

1823

y/ij

-y-i

V+e/V+V

-n/-ø

g-

-k

Moke

1823

ij

-i

V+V

-ø

g-

-k

Behaegel

1824

y

-y

V+V

-n

g-

-k

De Simpel

1827

ij

-i

V+V

-ø

g-

-k

Limburg

Anon. [Beginselen]

1819

ij

-i

V+V

-ø

g-

-k

In French

Van der Pijl

1815

ij

-i

V+V

-ø

g-

-k

Meijer

1820

ij

-i

V+V

-ø

g-

-k

As Many Norms as There Were Scribes?

Antwerp

Official norm Origin

Author

Year

y/ij

-y/-i

V+e/V+V

-n/-ø

gh-/g-

-ck/-k

North

Siegenbeek

1804

ij

-i

V+V

-ø

g-

-k

243

244

Gijsbert Rutten and Rik Vosters

4.2 The orthographical-chaos myth revisited The final section of this article will assess the ‘orthographical-chaos myth’. Did scribes in dif ferent towns write in their own dialects, and how chaotic were the actual spelling practices? To put the earlier claims to the test, we used a digitized collection of  handwritten documents, which are being transcribed at the Centre for Linguistics of  the Vrije Universiteit Brussel as part of ongoing research.13 The corpus contains formal and less formal texts from the judicial and administrative domain, including a roughly equal number of: (1) (2) (3)

police reports, drawn up at the local level by police constables, rangers, or other members of  the municipal authorities; interrogation reports, written down by district-level scribes and signed by the juge d’instruction in question; indictments, issued by the professional scribes of one of  the high courts.

Among these three text types, which already range from the very local to the supraregional level, all five Southern provinces are represented, with an equal amount of material per region coming from a main city and dif ferent peripheral towns or villages. A test version of the full corpus was used, containing a total of 61,912 words (excluding editorial and linguistic markup). The material allows us to compare writing practices in dif ferent regions, and furthermore has a built-in diachronic dimension, with texts from approximately 1823 and 1829. Both years have been chosen because of  their sociohistorical importance. In January 1823, language laws came into practice that made the use of  Dutch compulsory in most of  the government administration and judicature in the Dutch-speaking Southern provinces. For the majority of  the departments which were operating in French before the Dutchification policy took place, this means that the documents under investigation are among the first of  their kind to be 13

The actual source material was gathered and compiled into a digital image database by Isabel Rotthier and the Royal Academy for Dutch Language and Literature (kantl; see Rotthier 2007). For more background on the research project as a whole, see Vosters & Vandenbussche (2008) and Vosters & Vandenbussche (2009).

As Many Norms as There Were Scribes?

245

written in Dutch since before the French rule of 1794–1814. The manuscripts give us an interesting overview of  the Dutch language in Flanders during the early years of  the UKN. This allows us to compare them with 1829, at the end of  the Kingdom, and to see if any changes occurred after those years of political union with the North. The investigation of  this corpus material consisted of electronic searches for the orthographical variables discussed earlier: how often do traditional Southern forms such as or occur, in comparison with the and forms codified by the Siegenbeek norm in the North? It must be emphasized at this point, however, that we refer to orthographical variants as being either Southern or Northern based on the findings of our norms research. ‘Southern’, then, means that a spelling variant coincides with the prescribed usage in the large majority of  the pre-1815 Southern normative publications, while ‘Northern’ is used to indicate that a variant corresponds to the prescribed use in the of ficialized orthography of  Siegenbeek (1804).14 Clearly, this approach is reductionist in enforcing a strict dichotomy, and the presented results should only be interpreted as a first and tentative indication of the spread and success of both orthographical traditions, rather than capturing all possible variation, at least as far as the selected features and the employed sources go. We do not mean to suggest that there is any direct causality between linguistic forms being prescribed in normative publications and their use in actual written language.

14

Note that the so-called superf luous letters and were also investigated, even though they seem to have disappeared from the Southern normative tradition in the eighteenth century already. The results for these searches are not taken up in the tables below, as they represent values below the 1% mark. These orthographical variants had made their way out of grammars and out of usage by the time of  the UKN.

246

Gijsbert Rutten and Rik Vosters Table 9  Southern (left) and Northern (right) spelling variants per feature (distribution per cent) in a digitized corpus of  handwritten documents from the Southern Netherlands15 dotting of [ei]

diphthongs

vowel lengthening

y

ij

-y

-i

V+e

V+V

Total

66%

34%

21%

79%

21%

79%

1823

74%

26%

34%

66%

33%

67%

1829

55%

45%

3%

97%

4%

96%

article N Sg M

Total

-n

-ø

Southern

Northern

Total

17%

83%

38%

62%

1823

25%

75%

48%

52%

1829

10%

90%

24%

76%

A first observation while searching through the corpus is that the situation was far from chaotic. As can be deduced from Table 9, scribes seem to have been f lexible in adopting the Northern norm: overall, 62 per cent of  the total number of tokens was Northern. For three features (diphthongs with -y, long vowels in -e and masculine den), the Flemish variant is quite rare, being used in less than a quarter of all cases. In the case of  the dotting of  the [ei], the Southern variant stands stronger, which might be due to the minimal orthographical dif ference between and in handwriting. More surprising is that a similar pattern holds true for the situation in 1823 already. Rather than what might be expected from observations in the literature (i.e. a disparate but Southern preference), slightly over half 15

For the sake of clarity, we only shown the results per cent. The actual number of attestions is high in all cases (a total of 10,295 analyzed tokens, 3,917 of which are Southern and 6,378 are Northern). The full numbers can be sent upon request. Proper names, place names, uncertain transcriptions, and stretches of  text in a foreign language were excluded from the present analyses.

247

As Many Norms as There Were Scribes?

of all the attested forms correspond to the Northern Siegenbeek norm (52 per cent). Although the Flemish is still very generally used, the other features occur in a Northern form in over two-thirds of  the cases. This is highly remarkable, considering the claims about the spelling chaos and strong dialectal uses in written documents in Flanders, and considering that the investigated documents are among the first of  their kind to be written in Dutch for all of  the South. Furthermore, when we analyze the data diachronically, a clear movement away from the Southern variants can be observed. Three quarters of all features in 1829 are Northern, and the Southern spellings for diphthongs with a , for long vowels with an , and for den as the masculine form of  the article have almost completely disappeared. In general, there is an increase of about 20 to 30 per cent in Northern forms for each of the individual features between 1823 and 1829. Table 10  Spelling variants per province (distribution per cent)16

1823

Total Southern forms

Total Northern forms

1829

Total Southern forms

Total Northern forms

Antwerp

36%

64%

Antwerp

44%

56%

Brabant

71%

29%

Brabant

33%

67%

Limburg

32%

68%

Limburg

21%

79%

East Fl.

68%

32%

East Fl.

30%

70%

West Fl.

62%

38%

West Fl.

33%

67%

The text distribution per province also allows us to look into regional variation, as shown in Table 10. In 1823, we still observe rather strong regional dif ferences, with Brabant, East Flanders and West Flanders showing a clear preference for the Southern forms, and the Northern forms prevailing in 16

The indictment section of the corpus has been excluded from these regional analyses, as these documents all originate from one of  two supraregional high court of fices (Brussels or Liège). This still leaves the total number of analyzed tokens at 5,499.

248

Gijsbert Rutten and Rik Vosters

Limburg and Antwerp. However, when we look at the situation for 1829, we cannot only discern the general trend towards the Siegenbeek norm as discussed above, but it is also remarkable that the regionally divergent patterns make way for a more uniform preference for the Northern variants. In 1829, the Northern forms are used in the majority of cases everywhere, and the figures lie much closer together in general. Great shifts can be observed in those provinces which were still mainly using Southern spellings in 1823, with a doubled number of  Northern forms in Brabant and East Flanders.17 To come back to the myth of spelling chaos in the Southern Low Countries, our corpus research allows us to draw several conclusions. The orthographical landscape in general is not at all distinctly Southern, and even when the Dutchification policy had just come into practice in 1823, the Siegenbeek variants were already widely spread. Apart from this far from chaotic spelling situation, we might add that none of the documents we transcribed and examined showed any sign of transliterated dialect (cf. Wils 1956). The number of  Northern forms furthermore casts doubts on the claim that administrative documents were hardly intelligible from one town to another (cf. Deneckere 1954). The steady spread of  the Northern spelling norms in a mere six years between the measuring points furthermore emphasizes the importance of  the political union between North and South for the increasing convergence in writing practices during the nineteenth century. These results supplement the earlier findings of Vanhecke (2007), who investigated orthographical shifts in town council reports of seven Flemish municipalities between 1795 and 1900. She concluded that 1823 marked the start of a gradual shift from Southern practice to more Northern Siegenbeek variants, and concludes that the dif ferent orthographical shifts under changing political circumstances, often without a change of  hand, are evidence of the remarkable linguistic competence of the scribes. Additionally, her large-scale investigation of language choice (Dutch/French) in 131 Southern chancery administrations shows that there is a uninterrupted 17

We did observe a slight decrease in Northern forms for the province of  Antwerp, which calls for further research.

As Many Norms as There Were Scribes?

249

tradition of Dutch usage all throughout the nineteenth century, which in itself already challenges the idea of the language ‘withering and weakening’ under foreign inf luence (cf. Wils 1956).

5 Conclusion In this chapter we discussed the dif ferences between language myths prevailing in the historical and historical-linguistic literature, and the actual linguistic situation in Flanders during the eighteenth and early nineteenth century. Focusing on orthography, we first explored the surprisingly large body of normative publications available for the South in the decades preceding the UKN. There seemed to have been a clear and coherent tradition of  Southern writing practices and orthographical prescriptions. The publications analyzed were not based on dialectal uses, and often maintained an awareness of  the tradition of good usage in the Northern Netherlands. During the brief union of  North and South between 1815 and 1830, grammarians and schoolteachers from Flanders played their own part in the new political scenario. A number of them switched to what had then become the of ficial spelling norm in the North (Siegenbeek 1804), while others continued to propagate Southern writing practices. Building on the importance of  the UKN, we then proceeded to investigate actual writing practices in a corpus of  handwritten documents from the 1820s. We already witnessed a preference for Northern spellings in the first documents written after the 1823 Dutchification, which led us to believe that the linguistic situation at the start of the UKN could hardly be described as ‘absolute chaos’. Moreover, we observed the firm spread of the Northern variants, which revealed how the way people wrote was directly or indirectly inf luenced by the contemporary socio-historical circumstances. In sum, the period of the UKN has been shown to be an interesting case study at the intersection of both Northern and Southern writing traditions, where the foundations of the later linguistic convergence of Northern and Southern Dutch developed.

250

Gijsbert Rutten and Rik Vosters

References Ballieu, J. 1792 [1772]. Néderduytsche spel- en spraek-konst. Antwerp: J.E. Parys. Behaegel, Pieter. 1817. Nederduytsche Spraekkunst. Eerste boekdeel. Bruges: Wed. De Moor en Zoon. Bormans, Jean Henri. 1841. Verslag over de verhandelingen ingekomen bij het staetsbestuer van België ten gevolge der taelkundige prijsvraeg. Ghent: Annoot-Braeckman. De Groof, Jetje 2004. Taalpolitiek en taalplanning in Vlaanderen in de lange negentiende eeuw. Unpublished doctoral dissertation. Brussels: Vrije Universiteit Brussel. De Neckere, Philips Jaques. 1815. Bewerp van Vlaemsche spelling, gevolgd van eenen oogopslag op de nederlandsche dichtkunst. Ypres: J.B. Smaelen-Moerman. De Ridder, Paul. 1999. ‘The use of languages in Brussels before 1794.’ In: Blockmans, Wim, Marc Boone, & Thérèse de Hemptinne (eds), Secretum Scriptorum. Liber alumnorum Walter Prevenier. Leuven: Garant, 145–164. De Vos, Hendrik. 1939. Moedertaalonderwijs in de Nederlanden. Een historisch-kritisch overzicht van de methoden bij de studie van de moedertaal in het middelbaar onderwijs sedert het begin van de 19e eeuw. Turnhout: Van Mierlo-Proost. De Vries, Jan W., Roland Willemyns & Peter Burger. 1993. Het verhaal van een taal. Negen eeuwen Nederlands. Amsterdam: Prometheus. De Vries, Marleen. 2004. ‘Efemeriden, seniliteiten, republiekjes der letteren. De geheime fantasieën over literaire genootschappen op een rijtje’. Spiegel der Letteren 46, 241–254. Deneckere, Marcel. 1954. Histoire de la langue française dans les Flandres (1770–1823). Ghent: Romanica Gandensia. Des Roches, Jan. [1761] (2007). Nieuwe Nederduytsche Spraek-konst. Derden Druk, oversien en verbetert doór den Autheur. Antwerp: Grangé. Reprint by J.M. van der Horst & Jozef  Smeyers. Amsterdam: Stichting Neerlandistiek VU / Nodus. Dibbets, Geert. 2003. ‘Een nieuw spoor van de Port-Royalgrammatica in Nederland’. In: Dibbets, Geert, Taal kundig geregeld. Amsterdam: Stichting Neerlandistiek VU / Nodus, 129–156. E.C.P. [De Witte, Egidius]. 1713. Ontwerp van eene Nederduytsche Spraek-konst. [Meenen]: Theodorus vanden Eynden. Elias, Hendrik Jozef. 1963. Geschiedenis van de Vlaamse gedachte. De grondslagen van de nieuwe tijd (1780–1830). Antwerp: De Nederlandsche Boekhandel. Gyselynck, C.L. 1819. Nieuwe grond-beginselen der Vlaemsche tael, geschikt na de beste Nederduytsche taelmeesters. Ghent: J. Begyn.

As Many Norms as There Were Scribes?

251

[Grond-regels]. 1817. Grond-regels der Nederduytsche spel-konst. Mechelen: P.J. Hanicq. Hanou, André. ‘“De Verlichting brandt.” Een kwart eeuw verlichte letteren in Noord en Zuid (1670–1815)’. Spiegel der Letteren 46, 215–225. Henckel, Frans Lodewijk N. 1815. Nieuwe Vlaemsche spraek-konst. Ghent: P.F. de Goesin-Verhaege. [Inleyding]. 1785. Inleyding tot de grondregels der Vlaemsche spraek- en spelkonste. Dendermonde: Wed. J. Decaju. Janssens, Balduinus. 1775. Verbeterde Vlaemsche spraek- en spel-konste. Bruges: Joseph De Busscher. Knol, Jan. 1977. ‘De grammatica in de achttiende eeuw’. In: Bakker, Dirk Miente & Geert Dibbets (eds), Geschiedenis van de Nederlandse taalkunde. Den Bosch: Malmberg, 65–112. [P.B.]. 1757. Fondamenten ofte Grond-Regels der Neder-Duytsche Spel-Konst. Antwerp: Hubertus Bincken. Peeters, Evert. 2003. Het labyrint van het verleden. Natie, vrijheid en geweld in de Belgische geschiedschrijving (1787–1850). Leuven: Universitaire Pers. Put, Eddy. 1990. De cleijne schoolen. Het volksonderwijs in het hertogdom Brabant tussen Katholieke Reformatie en Verlichting (eind 16de eeuw – 1795). Leuven: Universitaire Pers. Rotthier, Isabel. 2007. ‘“In the picture”. Een bronnencorpus / beeldbank van juridische teksten uit de periode van het Verenigd Koninkrijk der Nederlanden’. In: Handelingen van de Koninklijke Zuid-Nederlandse Maatschappij voor Taal- en Letter­ kunde en Geschiedenis LX (2006). 131–149. Rutten, Gijsbert. 2009a. ‘Grammar to the people. The Dutch language and the public sphere in the 18th century with special reference to Kornelis van der Palm’. Beiträge zur Geschichte der Sprachwissenschaft 19, 55–86. Rutten, Gijsbert. 2009b. ‘Lezen en schrijven volgens Verpoorten en Bincken. Onderwijsvernieuwingen in Antwerp in de achttiende eeuw’. E-Meesterwerk 2. accessed 18 August 2011. Rutten, Gijsbert. 2009c. ‘Over Andries Stéven, wie het Voorschrift-boek schreef.’ In: Boogaart, Ronny, Josien Lalleman, Marijke Mooijaart & Marijke van der Wal (eds), Woorden wisselen. Voor Ariane van Santen bij haar afscheid van de Leidse universiteit. Leiden: Stichting Neerlandistiek Leiden, 301–312. Rutten, Gijsbert. 2009d. ‘De bronnen van Des Roches. Jan Des Roches’ Nieuwe Nederduytsche spraek-konst (1761) en de taalgeschiedenis van de achttiende eeuw’. Tijdschrift voor Nederlandse Taal- en Letterkunde 125 (4), 362–384.

252

Gijsbert Rutten and Rik Vosters

Siegenbeek, Matthijs. 1804. Verhandeling over de Nederduitsche spelling, ter bevordering van eenparigheid in dezelve. Amsterdam: Allart. Sluys, A.L.I. 1912. Geschiedenis van het onderwijs in de drie graden in België tijdens de Fransche overheersching en onder de regeering van Willem I. Ghent: Koninklijke Vlaamsche Academie voor Taal- en Letterkunde / Sif fer. Smeyers, Jozef. 1959. Vlaams taal- en volksbewustzijn in het Zuidnederlands geestesleven van de 18de eeuw. Ghent: Secretarie der Academie. Stéven, Andries. 1714 [1784]. Nieuwen Néderlandschen Voorschrift-boek. Ypres: Moerman. Suf feleers, Tony. 1979. Taalverzorging in Vlaanderen. Een opiniegeschiedenis. Bruges: Orion / Gottmer. Ter Bruggen, Joannes Abraham. 1822 [1817]. Nederduytsche spraek-konst ten gebruyke der schoólen. Antwerp: J.S. Schoesetters. Vaelande van Ieper [Van Daele, François Donatien]. 1805/1806. Tyd-Verdryf. Ondersoek op de Néder-duytsche spraek-konst. [Ypres]: [De Varver]. Van Aerschot, M. 1807. Nieuwe Nederduytsche spraek- en spel-konst. Turnhout: J.H. Le Tellier. Van Belleghem, P.J. & Daniël Waterschoot. [1773]. Deure oft Ingang tot de Nederduytsche Taele. Bruges: Van Praet. Van Boterdael, L. 1785 [1776]. Gemaklyke wyze om óp korten tyd grooten voortgang te doen in de Nederduytsche spel-konst. Bruges: Joseph Bogaert. Van der Horst, Joop. 2004. ‘Schreef  J.B.C. Verlooy echt zo gebrekkig? Het 19de / 20ste-eeuwse beeld van de 18de eeuw getoetst’. Verslagen en Mededelingen van de Koninklijke Academie voor Nederlandse Taal- en Letterkunde 114 (1), 71–82. Van der Sijs, Nicoline. 2004. Taal als mensenwerk. Het ontstaan van het ABN. The Hague: Sdu. Van der Wal, Marijke. 1995. De moedertaal centraal. Standaardisatie-aspecten in de Nederlanden omstreeks 1650. The Hague: Sdu. Van der Wal, Marijke. 2006. Onvoltooid verleden tijd. Witte vlekken in de taalgeschiedenis. Amsterdam: Koninklijke Nederlandse Akademie van Wetenschappen. Van der Wal, Marijke & Cor van Bree. 2008. Geschiedenis van het Nederlands. 5th edn. Houten: Spectrum. Vanhecke, Eline. 2007. Stedelijke kanselarijtaal in Vlaanderen in de negentiende eeuw. Unpublished doctoral dissertation. Brussels: Vrije Universiteit Brussel. Verpoorten, Jan Domien. 1752. Woorden-schat oft letter-konst. Antwerp: A.J. du Caju. Verpoorten, Jan Domien. 1759. Woorden-schat oft letter-konst. 2nd edn. Antwerp: Berbie.

As Many Norms as There Were Scribes?

253

Vosters, Rik. 2009. ‘Integrationisten en particularisten? Taalstrijd in Vlaanderen tijdens het Verenigd Koninkrijk der Nederlanden (1815–1830)’. Handelingen van de Koninklijke Zuid-Nederlandse Maatschappij voor Taal- en Letterkunde en Geschiedenis LXII (2008), 41–58. Vosters, Rik, Gijsbert Rusten, V Marijke van der Wal. 2010. ‘Mythes op de pijubank. Naar een herwaardering van de taalsituatie in de Nederlanden in de achttiende en negtutiende eeuw’. Verslagen en Mededelingen van de Koninklijke Academie voor Nederlandse Taal- en Letterliunde, 120, 93–112. Vosters, Rik, & Wim Vandenbussche. 2008. ‘Wijzer worden over Willem. Taalgebruik in Vlaanderen ten tijde van het Verenigd Koninkrijk der Nederlanden (1814–1830)’. Internationale Neerlandistiek 46 (3), 2–22. Vosters, Rik, & Wim Vandenbussche. 2009. ‘Nieuw onderzoek naar taalbeleid en taalvariatie in de Zuidelijke Nederlanden ten tijde van Willem I’. In: Backus, Ad, Merel Keijzer, Ineke Vedder & Bert Weltens (eds), Artikelen van de Zesde Anéla-conferentie. Delft: Eburon, 389–395. Weiland, Petrus. 1805. Nederduitsche Spraakkunst. Amsterdam: Allart. Willems, Jan Frans. 1824. Over de Hollandsche en Vlaemsche schryfwyzen van het Nederduitsch. Antwerpen: Wed. Schoesetters. Wils, Lode. 1956. ‘Vlaams en Hollands in het Verenigd Koninkrijk’. In: Dietsche Warande en Belfort, 527–536. Wils, Lode. 2003 [2001]. Waarom Vlaanderen Nederlands spreekt. Leuven: Davidsfonds.

Anneleen Vanden Boer

Language and Nation: The Case of  the German-Speaking Minority in Belgium

Abstract This paper investigates the use of  language as a tool for nation-building, based on a study of  the German-speaking minority in Belgium, in particular with regard to the correlation between knowledge of a language and the sense of af filiation to the nationstate for German-speaking Belgians. There are two opposing discourses asserting on the one hand that German-speaking Belgians ought to be ‘Frenchified’ or even ‘degermanized’ to become better Belgians and, more recently, on the other hand that the only ‘loyal Belgians’ are those who are native speakers of German. My findings appear to confirm the ‘stereotypes’ that German-speaking Belgians are royalists and patriots. Also, the national identity of  this minority is no longer supported by knowledge of one specific language. A good overall knowledge of all of ficial languages, however, is a good predictor of  the strength of  Belgian identity.

1 Introduction The focus of  this paper concerns the relationship between language command and sense of af filiation to the nation-state, based on a study of  the German-speaking minority in Belgium. There are two interesting and opposing points of view on this issue, which can be traced to two dif ferent moments in Belgian history. The earlier of these two discourses asserts that German-speaking Belgians ought to be ‘Frenchified’ or even ‘degermanized’ to become better Belgians. More recent opinion, on the other hand, states

256

Anneleen Vanden Boer

that the only ‘loyal Belgians’ are those who are native speakers of German. Based on empirical evidence collected in 2008, I will investigate the correlation between knowledge of a language and the sense of af filiation to the nation-state for German-speaking Belgians. I will begin by giving an overview of the geographical and historical background of the (of ficially) recognized German-speaking minority in Belgium. This will be followed by a discussion of  the theory relating to the notion of  language as a tool for nation-building in this specific context. I will then proceed to explain the methodology of my own survey, reporting and interpreting the results before attempting to draw conclusions from my findings.

2 Geographical and historical context Belgium has ten million inhabitants, roughly six million of whom speak Dutch, four million French and seventy thousand German. Dutch, French and German are the of ficial national languages in Belgium. The Germanspeaking Belgians are recognized as a national linguistic minority and enjoy special language rights. The German-speaking part of Belgium consists of  the areas surrounding three cities: Eupen, Sankt Vith and Malmedy. This region has been home to the German-speaking minority and belonged to Belgium since 1920. Although not itself  the focus of my research, I would also like to draw attention to three additional areas where German is spoken. These are situated around the municipalities of  Welkenraedt (north-east of Eupen), Bého (south of Sankt Vith) and Arlon (in the south of Belgium at the border with the Grand Duchy of Luxembourg). All three were initially German-speaking areas that have been part of Belgium since its foundation in 1830. Over the course of  time they have become almost completely Frenchified. The German-speaking inhabitants in these areas are not recognized as a linguistic minority.

Language and Nation

257

The territory of  the municipalities concentrated around Eupen and Sankt Vith, by contrast, has formally been recognized as a German-speaking area. This means that German is the of ficial administrative language in this region. Politically, the municipalities around Eupen and Sankt Vith constitute part of  the German-speaking Community. This community is a legal entity of  the Belgian federation with autonomy in personal and cultural af fairs. By contrast, the municipalities Malmedy and Waimes are situated in the French-speaking area, where the of ficial language is French, although the German-speaking minority living in these areas enjoys language facilities.1Malmedy and Waimes are not part of the German speaking community, but of  the French community (cf. Vanden Boer, 2009). Subjected to a change in nationality four times since the Congress of  Vienna in 1815, each of  these areas has borne witness to an eventful past. Before being conquered by French forces at the end of the eighteenth century and annexed by the French empire, Eupen and Sankt Vith belonged to the Austrian Netherlands, to the Duchy of  Limburg and the Duchy of  Luxembourg respectively. Malmedy formed a sovereign abbey territory together with Stavelot. After the defeat of Napoleon in 1815 both duchies were divided during the Congress of  Vienna. No attention was paid to their socio-cultural characteristics, as economic interests prevailed, and the duchies were thus divided between the United Netherlands and Prussia. Eupen, Sankt Vith and Malmedy became part of  Prussia. Following the First World War, Belgium was recognized as a victim of war by the Treaty of Versailles in 1920, and was granted permission to annex the municipalities around Eupen, Sankt Vith and Malmedy. Twenty years later, however, the areas were incorporated into Germany again during the course of  the Second World War. By the end of  World War Two the area was liberated, returning once more to the control of  Belgium, and continues to be governed by the country today (cf. Vanden Boer, 2008).

1

Language facilities give a citizen with a mother tongue other than the language of ficially spoken in the language area the right to use his own language in his contacts with of ficial authorities.

258

Anneleen Vanden Boer

3 Theoretical outline The central topic of this paper is the theory of language as a tool for nationbuilding, an issue which has been characterized by two opposing discourses with regard to the position of the German-speaking minority in Belgium. I will provide a brief contextual overview of the foundations for these two distinct viewpoints on the subject, before discussing my own data. View 1: ‘German-speaking Belgians should be Frenchified to become better Belgians.’ Following the annexation of the region in the immediate aftermath of the First World War, the areas surrounding Eupen, Sankt Vith and Malmedy were ruled as a colony by the Belgian government. Governor Baltia, who had previously been posted in the Congo, an African colony also belonging to Belgium, was put in charge of  the region and governed from 1920 to 1925. The aim was to integrate the newly acquired areas into Belgium quickly. As most of the elite spoke only French, this lead to a drastic wave of  ‘Frenchification’, with the French language being used in Belgium as a tool for nation building during this time. French had been implemented by the Belgian government in its African colony, and consequently this policy was also applied to the German-speaking area. French was the language of good Belgians. For Baltia the Belgification of  EupenMalmedy implied Frenchification. ([translation A.V.B.] Wenselaers 2008: 63) It was Baltia’s duty to adapt the former Prussian territory […] to the Belgian structures. An important measure was the introduction of  French […]. ([translation A.V.B.] Darquennes 2005: 6)

After twenty years of  Belgian sovereignty, the areas around Eupen, Sankt Vith and Malmedy were annexed by Germany during the Second World War. The war had filled the Belgian population with aversion to Germany and the German language, and so when the German-speaking areas were liberated at the end of the Second World War, the cry for ‘Frenchification’ of  these areas was immense. The inhabitants of  these regions had fought

Language and Nation

259

with the German enemy as they had also done during the First World War. It made no dif ference to the Belgians that this time Germany had forced them to do so. After the war the Belgian government did not know what to do with the German speaking inhabitants. The attempts in the 1920 and 1930 to Belgify the population did not succeed. It was considered to eliminate the German language and culture immediately this time, starting with a radical denazification. ([translation A.V.B.] Wenselaers 2008: 105)

A remarkable trend emerged in the areas around Montzen, Bého and Arlon (see above). These, originally German-speaking areas began to voluntarily ‘Frenchify’ themselves in order to distinguish themselves from the ‘German enemy’. This tendency was already noticeable after the First World War but increased after the Second World War. The Prussian invasion in 1914 evoked a strong belgicist ref lex in the Old-Belgian areas, which was ref lected mostly in the language use. The dif ference between the own German dialect and the standard German was emphasized and French was used more frequently. The use of  French became a sign of patriotic unionism. ([translation A.V.B.] Verdoodt 1968: 9)

View 2: ‘German-speaking Belgians are the last Belgians standing.’ The parliamentary year 2007–2008 marked the beginning of a political crisis in Belgium. Since its foundation Belgium has united two big language communities: one Dutch-speaking (Flemings) and one French-speaking (Walloons). The collaboration between them often creates problems. From the 1960s to the year 2001, several state reforms have redefined the Belgian state model. In 1993 this resulted in the creation of a federal state with widespread autonomy for Flanders and Wallonia. The Belgian federal structure still could not reconcile a stable balance between Flemings and Walloons, however, and the reforms continued after that. The tension reached a climax after the federal elections in June 2007 when it became impossible to form a government as a result of the disagreements between Flemings and Walloons over the aims of  the new political manifesto, which was to include yet another state reform. The future of Belgium and the option of splitting

260

Anneleen Vanden Boer

up the country was publically debated. The German-speaking community was repeatedly overlooked in this discussion, and as a result, assumed the role of the underdog. For this reason it is often said that German-speaking Belgians are the last true Belgians standing: The Eastern Belgians are unknown to many Belgians. They are viewed as monarchists, Catholics or the best protected minority in the world. Or even as the last Belgians. ([translation A.V.B.] Wenselaers 2008: 165)

4 Method I will use data from a written survey I conducted in September and October 2008 among a cross-section of 977 non-German-speaking Belgians (NGB) and German-speaking Belgians (GB) between 20 and 74 years old, with the aim of revealing the current relationship between citizenship and language command within the German-speaking minority of present-day Belgian society. The survey has been designed specifically to enable the correlation of  the scores relating to the variables ‘language command’ and ‘mother tongue’ with the scores relating to the variables ‘royalism’, ‘patriotism’ and ‘separatism’. The results will be analysed using the SPSS software and based on frequency tables and the Kendall tau b test for correlation.2 Where the variable ‘language command’ or ‘mother tongue’ is used, it is important to note that the scores for language knowledge are based on self-assessment by the respondent. Bias as a result of over- or underestimation of  linguistic knowledge on the part of  the respondent is, of course, possible. In some cases I will use an overall score for ‘language command’. This score is composed of separate scores for command of  Dutch and/or 2

Correlations with a rho-value (ρ) higher than 0.05 are not significant. Correlations with a rho-value (ρ) smaller than 0.05 are significant. Correlations with a rho-value (ρ) smaller than 0.01 are highly significant.

Language and Nation

261

French and/or German. For each respondent the mother tongue has been not been considered in this overall score. When I use the label ‘foreign national language 1’ (FNL1) this refers to the foreign national language that is most likely to be the best known in a specific language group (e.g. French is the FNL1 of  the GB). The label ‘foreign national language 2’ (FNL2) is the foreign national language that is more likely to be known, only after a certain level of the FNL1 has been reached. The label ‘Feeling of af filiation to Belgium’ refers to an overall score, composed by the individual scores on: – degree of royalism (Do you think it is desirable that Belgium has a king?) – degree of unionism (Do you think Belgium should be split up into two independent   states, Flanders and Wallonia?) – degree of patriotism (To what extent do you consider yourself  Belgian?)

These individual responses will be evaluated using likert items. This is a bipolar scaling method commonly used in questionnaires to measure the extent to which a respondent agrees or disagrees with a given statement, and to allow a degree of dif ferentiation in the answers.

5 Results 5.1 A loyal Belgian is likely to speak German Let us first take a look at the correlation between the feeling of af filiation with Belgium and the mother tongue of  the respondents, and answer the question: Is there a specific language group in Belgium that feels significantly more af filiated to Belgium than the others? These results are shown in Table 11. Highly significant relations are indicated in bold.

262

Anneleen Vanden Boer

Table 11  Correlation between mother tongue and the sense of af filiation to Belgium GB + NGB

Feeling of af filiation to Belgium

French mother tongue

0.1386

Dutch mother tongue

< 0.0001 INVERSE CORRELATION

German mother tongue

< 0.0001

There is no significant relation as far as French-speaking Belgians are concerned. In Dutch and German-speaking Belgians, by contrast, we notice a high significance, but for opposing reasons within each group. If Dutch is the mother tongue there is a high probability that one will feel less af filiated to Belgium. Speaking German by contrast, manifests itself additionally in higher likelihood of feeling more af filiated to Belgium. This result also appears in the frequency tables of my questionnaire (Table 12). These show the results of  GB and NGB. I dif ferentiate between GB from Malmedy and Waimes, situated in the French-speaking area (GB MAWA) and the GB from the residents of  the German-speaking Community, situated in the German speaking area (GB GC). Table 12  Frequency distribution for the variables indicating the sense of af filiation to Belgium Frequency agrees with the presence of a king in Belgium NGB

GB MAWA

GB GC

Strongly disagree

18.3

8.8

6.9

Disagree

9.2

4.4

4.4

Neutral

17.9

8.8

18.6

Agree

13.3

14.7

10.1

Strongly agree

41.3

63.2

60.1

%

263

Language and Nation Frequency agrees with the separation of  Belgium in Flanders and Wallonia NGB

GB MAWA

GB GC

Strongly disagree

54.4

82.1

80.3

Disagree

14.1

6.0

7.1

Neutral

7.1

4.5

4.4

Agree

8.3

1.5

3.9

Strongly agree

16.2

6.0

4.4

NGB

GB MAWA

GB GC

Very low

5.9

4.4

4.8

Low

6.3

1.5

3.5

Quite low

11.7

10.3

11.1

Quite high

25.5

20.6

17.9

High

28.5

20.6

28.1

Very high

22.2

42.6

34.6

%

Frequency degree of patriotism %

In contrast to NGB we notice a distinctly more positive answer pattern for GB. In particular, the GB living in the Malmedy-Waimesarea are shown to be the most loyal Belgians. 5.2 German-speaking Belgians are the most multilingual Before studying the relation between language command and Belgian identity, we should further investigate the level of language command of Belgian citizens (Table 13). In this paper we will only consider the language command of  the Belgian national languages Dutch, French and German.

264

Anneleen Vanden Boer Table 13  Frequency distributions for the language command

Frequency degree of  knowledge of  Dutch NGB % DU FR None at all n.a. 7.2 Very poor n.a. 20.5 Poor n.a. 24.1 Suf ficient n.a. 25.3 Good n.a. 9.6 Very good n.a. 9.6 Native n.a. 3.6 Frequency degree of  knowledge of  French NGB % DU FR None at all 2.6 n.a. Very poor 3.2 n.a. Poor 9.0 n.a. Suf ficient 32.9 n.a. Good 31.6 n.a. Very good 18.1 n.a. Native 2.6 n.a. Frequency degree of  knowledge of  German NGB % DU FR None at all 9.9 51.9 Very poor 13.2 14.8 Poor 20.4 16.0 Suf ficient 36.2 11.1 Good 17.8 2.5 Very good 2.0 3.7 Native 0.7 0.0 Note. n.a.: not applicable DU: native Dutch speakers FR: native French speakers

GB MAWA

GB GC

14.1 18.8 15.6 32.8 12.5 4.7 1.6

16.8 10.6 18.7 30.2 16.5 6.7 0.5

GB MAWA

GB GC

0.0 0.0 0.0 4.4 17.6 47.1 30.9

1.4 1.7 4.8 17.3 37.9 32.4 4.6

GB MAWA

GB GC

n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a.

265

Language and Nation

The frequency tables concerning the degree of  knowledge of  Dutch and French among GB clearly show that almost every GB has at least suf ficient linguistic command of French. GB MAWA emerge as having an exceptionally good knowledge of  French. One GB out of two has suf ficient command or better of  Dutch. For Dutch there is no big dif ference between GB MAWA and GB GC. The knowledge of the German language among NGB in general is quite bad, with French-speaking Belgians in particular scoring very poorly; eight out of ten French-speaking Belgians have no or only poor command of  German. The German language level of  Dutchspeaking Belgians is comparable to the Dutch language level of GB, which is poor to suf ficient. The language knowledge of  French among Dutchspeaking Belgians is not as good as that of  GB, but still better than the language level of Dutch among French-speaking Belgians. Overall the GB can be defined as the most multilingual Belgians, whereas French speaking Belgians are the least multilingual. 5.3 Multilingualism supports the Belgian identity of German-speaking Belgians Having successfully analysed knowledge of the national languages, we can now investigate the relationship between this knowledge and the feeling of af filiation to Belgium. I will do this by measuring the significance of the relation between both variables. The significant relations are underlined and high significance is indicated in bold (see Table 14). Table 14  Correlation between language command and the sense of af filiation to Belgium GB

Degree of royalism

Degree of unionism

Degree of patriotism

Sense of af filiation to Belgium

Command of  French

0.375

0.718

0.015

0.086

Command of  Dutch

0.709

0.462

0.002

0.026

Overall level of  language knowledge

0.457

0.856

0.003

0.037

266

Anneleen Vanden Boer

To study this correlation, I will consider both the composite variable of  ‘sense of af filiation to Belgium’ (in the last column) and the component parts of that variable. If we consider the composite variable ‘sense of af filiation to Belgium’ and its relation to language knowledge, we observe that, with the exception of the relationship with command of French, which is almost significant, the sense of af filiation to Belgium is likely to increase when the level of  language knowledge increases. Next we will also consider the component parts separately. Here we notice that the degree of  both royalism and unionism have no significant relation to language knowledge. In contrast, the correlation between patriotism and language knowledge turns out to be highly significant. In order to find out whether this result is characteristic for all Belgians, I conducted the same test for Dutch- and French-speaking Belgians. In both groups no significant relationship between language command and Belgian identity could be found. Therefore we cannot assume that multilingualism is, in itself, a mark of  Belgian identity.

6 Conclusion The results of my empirical research appear to confirm some ‘stereotypes’ that appear in the academic literature on the subject: German-speaking Belgians are royalists and patriots. In this investigation I have demonstrated that Belgians with German as their mother tongue are likely to be more loyal with regard to Belgium than other Belgians. After the two World Wars it was impossible to think that the most loyal Belgians would speak German. To create ‘good Belgians’ a radical Frenchification of the German-speaking Belgians seemed the only way. Fifty years later, the social demographic has obviously changed, and the national identity of this minority is no longer supported by knowledge of one specific language. A good overall knowledge of all of ficial languages, however, is a good predictor of  the strength of  Belgian identity.

Language and Nation

267

Not only ‘Belgian-ness’, but also multilingualism, is a defining part of  the identity of  German-speaking Belgians, many of whom could even be additionally defined as bilingual French-German. This multilingualism and the strong sense of, particularly patriotic, af filiation to Belgium, seem concurrent in enhancing the mutual intensity of one another. It would have been interesting to discover that multilingualism is an intrinsic part of Belgian identity on a more national level too, but it does not seem to be significant for the Dutch and French linguistic groups. Neither multilingualism, nor mastery of a specific language, can be described as components of a more general Belgian identity.

References Darquennes, Jeroen. 2006. ‘Duits als autochtone taal in België: toekomstige onderzoeksdesiderata.’ In: Handelingen van de Koninklijke Zuid-Nederlandse Maatschappij voor Taal- en Letterkunde. Gent KZM, 93–109. Vanden Boer, Anneleen. 2008. ‘Die Deutschsprachigen in Belgien.’ Der Sprachdienst 6/08, 245–252. Vanden Boer, Anneleen. 2009. ‘Die Position der deutschsprachigen Minderheit im belgischen gemeinschaftlichen Rahmen – erste Befunde.’ Muttersprache, 13–32. Verdoodt, Albert. 1968. Zweisprachige Nachbarn: die deutschen Hochsprach- und Mundartgruppen in Ost-Belgien, dem Elsass, Ost-Lothringen und Luxemburg. Wien: Braumüller. Wenselaers, Selm. 2008. De laatste Belgen, een geschiedenis van de Oostkantons. Antwerp: Meulenhof f/Manteau.

Part 3

Demographics and Social Dynamics

Richard Ingham

The Decline of  Bilingual Competence in French in Medieval England: Evidence from the PROME Database

Abstract The maintenance and demise of  Anglo-Norman is examined using petitions in the Parliament Rolls of Medieval England (PROME) electronic database. Noun gender is shown to have been largely error-free in contexts involving preposition + article combinations around 1300, but to have become massively error-prone from about 1370 on. The change is argued to have been related to the disappearance of French as a vehicle language for education in first schools attended by clerks, around the time of  the Black Death. Conventional accounts of  later Anglo-Norman as an instructed foreign language are shown to be inadequate to deal with the phonologically driven patterning of other gender marking errors which, it is concluded, were an ef fect of spoken transmission.

1 Introduction In the later medieval period, vernacular languages in many regions of Europe began to challenge the hegemony of  Latin over the written word for purposes of record-keeping. In England the principal beneficiary of the rising use of  the vernacular as a written medium between about 1250 and 1370 was not English but French, in the insular variety which will be referred to here as Anglo-Norman, following the convention set by the Anglo-Norman

272

Richard Ingham

Dictionary (Rothwell et al. 2005).1 If standard textbook accounts are to be believed, this is in some ways surprising. To quote Singh (2005: 109): ‘As England moved into the thirteenth century, the tide turned increasingly in favour of  English and proficiency in French waned.’ Yet far from declining, the use of French in written documents spread across genres in the later thirteenth and fourteenth centuries, in a way reminiscent of how a standard language is created (Haugen 1966). Indeed it looked to some in the earlier fourteenth century that French was threatening to become a kind of of ficial language.2 The far greater use made of  French in charters, wills, chronicles, correspondence and so on at this time is easy to observe, and would certainly have been very noticeable to contemporaries: for a while, at least, French became virtually another language of record alongside Latin, which it had not been in the first two centuries after the Conquest. It is not easy to understand what enabled the survival of  linguistic competence in French that permitted this expansion in the functional range of  Anglo-Norman. Previous approaches to this question have proposed solutions that for one reason or another are unconvincing. Authors dubbed by Kibbee (1991) ‘French maximisers’, such as Legge (1980), maintained that the use of insular French was widespread and that it was acquired as a spoken vernacular until a late date; ‘French minimisers’, represented by Rothwell (1976, 1978), on the other hand, have claimed that French was taught in England as an instructed second language in medieval schools by this time, and was ‘no more a vernacular than was Latin’. French ‘maximisers’ have been discredited by a number of scholars (see e.g. Berndt 1972, Rothwell 1976), and the critique of them is typically

1

2

An alternative nomenclature would be to reserve the term ‘Anglo-Norman’ for the earlier, largely literary, period lasting until the middle third of the thirteenth century (Pope 1934), and use a term such as ‘Anglo-French’ for the later, mainly non-literary period. For reasons of institutional conventions, as well as in the interest of avoiding the dif ficult issue of what linguistic changes marked the transition from one period to the next, we have not done this, however. A concern which may have lain behind the accusation directed by Edward II at Philip of France at the beginning of Hundred Years’ War, that the French wished to replace English by the French language.

The Decline of  Bilingual Competence in French in Medieval England

273

rehearsed in textbooks such as Lass (1987). The weaknesses in the arguments advanced by French ‘minimisers’ are less widely appreciated, but are nevertheless serious, and can be summarized as follows. First, if Anglo-Norman was an instructed language taught in a classroom, it is surprising that its spelling was so irregular. Second, French was still used in English grammar schools as a vehicle language through which to teach Latin in the first half of  the fourteenth century, according to contemporary witnesses, the chroniclers Higden and Holcote (Orme 1973). This presupposes that when pupils started Latin they already had some French; teaching one unknown language through another unknown language is unlikely to have been educationally successful. So French cannot have been first encountered in grammar school, which pupils began at the age of seven. Third, pedagogical grammar treatises written around 1250–1300 and discussed by Rothwell (1976) and Hunt (1985) teach Latin tenses through French, not the other way round, so cannot be used as evidence of the instructed nature of later Anglo-Norman. The same applies to word lists put forward by Rothwell (1976) as pedagogical materials for the teaching of  French. Fourth, the so-called ‘Treatise’ (a title created and adopted by modern editors of  the poem) written by Walter de Bibbesworth in the later thirteenth century is the only item clearly in the category of a text for language learners in the relevant period, but it was intended to support language learning in an aristocratic home setting. It tells us nothing about school contexts, and how less well-of f scholars who went on to become clerks acquired their French. In short, the notion that French had by 1300 become a language of  formal instruction in England rests on no empirical foundation. The problem we are left with, however, is to understand how AngloNorman was transmitted across three centuries following the Norman Conquest, the majority of the time with no, or almost no, remaining native speakers.3 If it was neither acquired as a mother tongue in the home nor taught in schools, in what context was it acquired? Present-day societies tell us that there are other options beyond this simple mother tongue vs. instructed foreign language dichotomy. In par3

The royal family is often thought to have been an exception here.

274

Richard Ingham

ticular, a second language may be learnt quasi-naturalistically in childhood, as for instance in immigrant communities in a country where a dif ferent public language from the home language is spoken. In second language acquisition in such contexts, age at initial exposure has repeatedly been shown to be a key determinant of native-like ability (Oyama 1976, Patkowski 1980, Johnson & Newport 1989). Although English would have been the first language acquired by users of  later Anglo-Norman, they could in principle have achieved native-like ability in that language variety if  they were exposed to it conversationally while young children. In this article the hypothesis will be adopted that those who were professionally involved with creating the texts which have come down to us, and which constitute our evidence base for the study of  Anglo-Norman, acquired their knowledge of  this variety in early to middle childhood from its use in spoken interaction, not as an instructed classroom subject. The contexts in which this spoken interaction took place were high-status households and the preparatory schools run by the church, known as song schools, to which future clerks were sent before entering grammar school (Orme 1973). If exposure to language took place in such settings, the grammar of  Anglo-Norman could have been reliably acquired by young children, even though their pronunciation would have been heavily inf luenced by the ambient English-inf luenced phonology of Anglo-Norman. This scenario is more fully developed by Ingham (2010). In this article we explore one aspect of the linguistic predictions made by the hypothesis just outlined: the marking of noun gender agreement. Noun gender is a domain well suited to the research issue, since numerous studies of second language (henceforth L2) acquisition have established a clear distinction between the essentially target-like attainment of natur­ alistic child L2 learners and the error-prone performance of instructed adult learners (see e.g. Franceschina 2005 and references therein). Even advanced instructed learners make fairly frequent gender marking errors. Reportedly, they tend to be most accurate on gender marking on articles and somewhat more error-prone on other noun modifiers. These findings, though obtained in a modern context, give some idea of what to expect, had AN been an instructed L2.

The Decline of  Bilingual Competence in French in Medieval England

275

Problems with noun gender agreement have indeed been observed in AN. Bolland (1914) commenting adversely on the standard of  French in later AN, drew particular attention to gender agreement errors. Tanquerey (ed.) (1916: lv) found gender errors with determiners that were ‘dus a l’ignorance’ as he put it, as well as failures of adjective-noun gender agreement that he took to be either spelling mistakes or due to English inf luence. Kibbee (1996: 10) also saw gender errors as a factor ref lecting English inf luence on AN. However, it is not entirely clear how English inf luence would have operated in this domain. Since Old English noun gender distinctions had largely disappeared by Early Middle English, there was no question of the gender of an English noun af fecting that of an AN noun. Rather, such inf luence would presumably have involved neutralized gender marking,4 either by using a single form to replace the two members of a gender marking opposition, or by using the two forms indiscriminately. Indiscriminate usage is indeed observed, as in the following examples from about 1300 showing gender clashes between modifier and noun: (1)a

en cel prison

Bolland (ed.) (1914: 37)

in this (MASC) prison (FEM)

(1)b

sicum en acun rivere

Nichols (ed.) (1865: 404)

as in some (MASC) river (FEM)

(2)a

de ceste trepas

Bolland (ed.) (1914: 12)

of  this (FEM) trespass (MASC)

(2)b

en le haute chemin vers Donestaple

Tanquerey (ed.) (1916: 50)

in the high (FEM) way (MASC) towards Dunstable

However, it is important to recognize that the AN gender agreement errors noted by earlier authors strongly tended to involve a final . A major inf luence on Anglo-Norman phonology was the loss of  the final weak ‘e’ 4

As happened in Cappadocian Greek, under the inf luence of  Turkish, which lacks grammatical gender (Karatseareas 2009).

276

Richard Ingham

sound in Middle English, already observable by the late twelfth century (Hogg & Denison 2006: 66), among numerous ef fects of English phonology on Anglo-Norman (Duf fell 2005). As noted by Pope (1934: 438), later Anglo-Norman spelling showed the ef fect of  this tendency by the omission of final . So apparent gender errors such as (1) – (2) may not have been a matter of imperfect grammar, but rather the erosion of particular phonological means of marking gender distinctions.5 In order to determine whether noun gender in AN was impaired independently of phonology, specifically the loss of weak ‘e’, Ingham (2010) therefore analysed AN texts for gender agreement on pairs of premodifier forms having very distinct phonology, the possessive determiners mon/ma, and son/sa. These forms, in addition to their dif fering vowel qualities, were distinguished by the nasal consonant in the masculine, which is absent in the feminine. Had they too shown gender agreement errors,6 this would have furnished good evidence for imperfect grammar learning, not just phonological inf luence from English. Ingham (in press) searched the petitions in the electronic version of the Parliament Rolls of Medieval England (Given-Wilson et al. 2005), henceforth referred to as ‘PROME’, covering the years 1310–1399. Errors with son and sa were found to be extremely rare events indeed until the last few decades of  the fourteenth century, averaging just one in twelve parliamentary sessions. Thereafter, however, they became much more common, with 34 errors occurring in the 29 sessions between 1371 and 1399. Until then grammatical gender properties of nouns must have very reliably known to insular users of  French, despite their English-inf luenced phonology. One might suppose that the presence of any errors at all is a sign of non-nativeness in language learning, but this would be over-strong. Continental French texts very occasionally show non-conventional use of noun gender, and in modern studies of noun

5 6

This concept is already familiar in medieval French dialect phonology in the case of  Picard, which neutralised the distinction between le/la, me/ma, etc. (Pope 1934: 488). Determined with reference to a standard Old French Dictionary, Greimas (1992).

The Decline of  Bilingual Competence in French in Medieval England

277

gender acquisition native speaker errors have also been reported.7 The same would apply to insular French too, until the second half of the fourteenth century. So the results were consistent with imperfect grammar learning by AN users only from the 1370s onwards.

2 The study The claim made by Ingham (2010) that AN users continued to observe noun gender with possessive determiners to a level of accuracy comparable to that of native speakers is acknowledged to be controversial: previous commentators on Anglo-Norman are near-unanimous that its users, by the fourteenth century, were not grammatically proficient. Even in terms of the hypothesis to be adopted here, that of quasi-naturalistic childhood learning of  AN, it is open to objection. If only possessive determiners were reliably marked for gender agreement in the language they heard, could naturalistic learners have acquired gender properties of nouns from such contexts, which were relatively uncommon compared with the high frequency of article and premodifier forms? This study therefore aims to identify other contexts, in addition to possessive ones, in which learners of  Anglo-Norman could have acquired noun gender. The most frequent opportunities for learners to have heard gender marking agreement were provided by definite articles; indefinite noun phrases in Old French did not yet need an indefinite article. As noted by Tanquerey and others, later AN writers are known to have used le with feminine nouns. An initial examination of  PROME indicated that such errors are absent from the petitions presented up to about 1310, suggesting that in this genre at least definite articles resisted gender confusion better

7

A study reported in Franceschina (2005) show Spanish L1 speakers making gender errors on about 1 per cent of occasions (tokens).

278

Richard Ingham

than did weak ‘e’ contexts.8 But from 1315 onwards, clear gender errors with definite articles occur. Between 1315 and 1330, they included the following feminine nouns given le : ley, rente, garde, copie, communalté, tour (‘tower’), feste, discomfiture, journee, and masculine nouns given la: roialme, castel, service, évesché.9 However, the opposition le: la is arguably unstable since the masculine form involves the weak ‘e’ vowel, so there could be doubt over whether such errors are really a grammatical or a phonological phenomenon. If  the latter, they would not provide evidence for the loss of noun gender as such. The most fruitful context for investigation seemed to be the observance of gender agreement when a definite article combined with a preceding preposition, especially à and de. As in Modern French, the Old French masculine article form contracted on to the preposition as au and du, alternatively spelt al and del. The feminine counterparts à la and de la also required knowledge of noun class membership, of course, but here it was felt that the growing confusion between le and la observed from 1315 onwards, as mentioned above, would have complicated the issue. Accordingly the data analysis was restricted to the masculine preposition plus article combinations, al/au and del/du, and to see if  they occurred only with a masculine nouns. Three exclusionary categories were determined. Since grammatical gender was at issue, animate nouns, where natural gender usually determines grammatical behaviour, were not used. Also excluded were cases where the target variable is followed by an element beginning with a vowel or with . An expression such as del annee might have been written in this form to represent the elision of  the /a/ of a feminine definite article, as in Modern French; in this case the use of del preceding a feminine noun 8

9

Le was regularly observed with feminine nouns beginning with a vowel, e.g. le almone, le esglise, le isle, le universeté regularly take le rather than la in the Edward I petitions. This should be seen as a phonologically conditioned sign of the elision of the vowel, or of its reduction to weak ‘e’, rather than a gender error, since feminine nouns beginning with a consonant did not show this. No gender variation is reported for these items in continental Old French by ToblerLommatzsch (1956–).

The Decline of  Bilingual Competence in French in Medieval England

279

should not count as a gender agreement error. Note that exclusion was determined on the basis of the word following the article, which was most often the noun, but could be a premodifier of the noun, e.g. avandite in the expression del avaundite dette. Finally, cases were observed where du was written for de e.g. du son doun bien, en fraunchise com hors du fraunchise, or au is written for à. These were not contexts for a definite article, so they were excluded. In sum, therefore, the data to be analysed consisted of cases where a definite article is grammatically expected, preceding an inanimate noun, where the noun or the first premodifier preceding it and following the article had a consonant as its initial phoneme. In scoring frequencies of  target items in a corpus, there is a choice to be made between recording the type-frequency or token-frequency of the item or combination targeted. The justification for doing the latter – the approach usually followed in the L2 learning studies cited above – is that in principle each instance is a new opportunity to display correct knowledge or to make an error. However, the often repetitive discourse of  the petitions in PROME made this an unsatisfactory procedure: the results would have been f looded with thousands of correct uses of very commonly used nouns, giving a percentage of errors that risked understating the incidence of errors AN users made in more ordinary discourse. Accordingly, we scored noun types used in the target combinations. This gave a much higher profile to any errors that were found. A single error with a noun occurring erroneously in that context would be counted, even though the rest of the time its gender agreement properties were target-like. It was felt that this procedure would pick up the tendency of an instructed L2 learner to make occasional careless mistakes, rather than burying them amongst huge numbers of correct instances. The main data collection concerned the preposition + article data analysis, as described above. However, the performance of AN users in the weak ‘e’ contexts was also assessed, as a comparison. Four frequent modifiers cel, cest, nul, and tout were selected for this purpose, and the gender of noun types co-occurring with the masculine form was scored. Both analyses are reported in the next section. The same database used by Ingham (2010) was searched, except that this time the petitions to parliament in the time of Edward I (1272–1307)

280

Richard Ingham

were included. Unlike other data files, the petitions for this period cannot be generally linked to specific dates, only to the reign as a whole. It was felt that this was not a serious problem, and was outweighed by the need to give as great a time depth as possible to the analysis of gender marking. The study went up to the last parliament in the reign of  Richard II (1377–1399). Although there were still petitions in French up to around 1430, by the end of the fourteenth century, grammatical marking involving prepositions and articles in AN was blatantly failing to respect the grammar of French, cf. the uses of a le and du in the following extract from the first parliament of  Henry IV: (3) L’evesqe d’Excestre… fist arester le dit Thomas count de Warrewyk, et luy amesner a le tour du Loundres October 1399 ‘The Bishop of  Exeter… had the said Th. Earl of  Warwick arrested and taken to the Tower of  London’

The issue was when this systematic breakdown in the syntax of gender took place.

3 Results Table 15 shows frequencies of masculine and feminine nouns co-occurring with the four preposition + article forms, in each 30-year period of  the PROME database between 1280, the presumed date of the earliest Edward I petitions in French, and 1399.10

10

In fact the latest parliament held by Richard II for which petitions are extant was in 1397.

281

The Decline of  Bilingual Competence in French in Medieval England Table 15  Frequencies of masculine and feminine nouns used with masculine preposition + article forms in PROME, 1280–1399 del

du

Years

MASC.

FEM.

TOTAL

MASC

FEM

TOTAL

1280–1309

42

1

43

54

1

55

1310–1339

34

5

39

60

4

64

1340–1369

26

2

28

59

7

66

1370–1399

86

94

180

70

54

124

MASC.

FEM.

TOTAL

MASC

FEM

TOTAL

1290–1309

23

1

24

24

2

26

1310–1339

24

0

24

46

4

50

al

au

1340–1369

22

3

25

35

7

42

1370–1399

56

58

114

49

33

82

These figures reveal a very clear change across the time period. The ‘Fem.’ columns, which show the frequencies of errors, display a low level of occurrences in almost every category until 1370–1399. Then errors become approximately as frequent as correct uses. The percentage occurrence of  feminine nouns for all four preposition + article types combined is represented in Figure 6, which displays an unmistakeable ‘broken stick’ trend, usually indicative of a change in the status of  the language trait under investigation. Were some preposition + article combinations contributing more than others to this trend? It is conceivable, for example, that one preposition changed its distribution more than the other, or that forms ending in were used with more respect for gender than those ending in . These possibilities are examined in Figure 7. Separate graph lines are shown for those cases where the combination ended in and , for cases where the preposition was de, and for those where it was à.

282

Richard Ingham

Figure 6  Percentage occurrence of  feminine nouns overall, with del/du and al/au

Figure 7  Percentage of occurrence of  feminine nouns with de and à, and with combinations ending in -l and those ending in -u

283

The Decline of  Bilingual Competence in French in Medieval England

It can be seen that the trend was virtually identical, regardless of the preposition and regardless of whether the contracted form ended in -l or in -u. With all prep + article combinations, there was a shallow increase to around 10 per cent in the percentage of feminine nouns used with masculine contracted forms between the later thirteenth century and the middle of  the fourteenth century. Thereafter, they all sharply increased, to a level of around 50 per cent of cases: in 1370–1399 all contracted forms were thus being used indiscriminately, regardless of noun gender. Whether this means that noun gender as such had now been completely lost in AN is a question to which we return below. In any case, it is clear from these results that the 1370s constituted a major turning point in the observance of noun gender marking: prior to then, errors were rather rare. A way of defending the instructed L2 position might be to take these results to show that at first French was being accurately taught, but that in the last decades of the century the ef fectiveness of instruction declined. Although this is not a hypothesis we find at all plausible, it is worth addressing. If it were correct, we would surely expect a similar trend with gender agreement errors involving weak ‘e’, since they too would surely have been subject to the vigilance of  language instructors. A subset of  the data was therefore searched to find out if the rate of weak ‘e’ errors increased in the late fourteenth century in line with errors on proposition + article combinations. The Edward I petitions, from around 1300, were compared with petitions to Parliament in the 1370s. Specifically the masculine premodifier forms cel, cest, nul and tut, on which errors had been previously noted, were targeted. Scoring by noun type was done on the same basis as for Table 15, and results are shown in Table 16. Table 16  Frequency of masculine and feminine noun types co-occurring with masculine premodifier forms cel, cest, nul and tut, late thirteenth century/ early fourteenth century and 1370s cel MASC.

cest FEM.

TOTAL

MASC

FEM

TOTAL

1280–1307

20

8

28

6

6

12

1370s

10

6

16

15

10

25

284

Richard Ingham nul MASC.

tout FEM.

TOTAL

MASC

FEM

TOTAL

1290–1307

16

8

24

13

4

17

1370s

17

10

27

18

8

26

Around the turn of the thirteenth century, cel, cest, nul and tout co-occurred with a total of 55 masculine and 28 feminine nouns, the latter counting as errors and constituting 33.7 per cent of instances. In the 1370s the corresponding figure was 60 masculine and 34 feminine nouns (36.2 per cent). It seems no great change took place in the abilities of  AN users regarding gender marking with weak ‘e’ forms in the intervening 70 years. Errors where weak ‘e’ was involved were already common at the beginning of the period studied, unlike the very low incidence of errors at that time with preposition + article forms. The constant error rate in the former and the sudden rise in the latter would be very hard to explain in terms of some notional change in the standard of instructed learning. In the remainder of  this study, therefore, we shall concentrate on the hypothesis that later AN was acquired naturalistically, at least until around the mid fourteenth century. However, there is still a methodological issue that needs to be addressed. As we saw in Table 15, the number of contexts for appropriate use of del, al etc. for the three decades 1370–1399 is much greater than in the earlier 30-year periods, by a factor of 4 or 5 to 1, in the case of all items except du. This is because the number of petitions in French grew in the later fourteenth century, as did their length in many cases. There may then be a concern that the dif ference in number of contexts biased the outcome by making it more likely to get gender errors in the later period, since more opportunities inevitably arise in a larger body of data. If so, this would invalidate the conclusion that the key factor in the incidence of gender errors was competence in AN in an earlier versus a later time-period. To address the problem, one would ideally take a much larger number of contexts from an earlier period, and see whether gender errors were still negligible. Within the confines of  the PROME petitions, this solution is unavailable, so instead it was determined to lump together the Edward I

285

The Decline of  Bilingual Competence in French in Medieval England

and Edward II petitions, covering c. 1280–1324, and compare them with those of  Richard II (1377–1399). In this way an earlier and a later body of data would be created which would now be considerably closer in terms of sample size, as can be seen from the fact that the form del achieved 662 hits in the Edward I + Edward II petitions, and 1067 hits in the Richard II petitions. The figures still show a numerical advantage for the later period, but it is a far less disproportionate one than the early and late overall contexts in Table 15. Using these two analysis periods, frequencies of masculine and feminine noun types occurring with del were tabulated, as in Table 17. Table 17  Frequencies of del with masculine and feminine nouns, 1280–1324 and 1377–1399 Masculine

Feminine

Edward I and II petitions

50

1

Richard II petitions

81

71

In line with their roughly 60 per cent larger sample size for del, the Richard II petitions produced rather over half as many masculine noun types again as the Edward I and II petitions. But the frequency of  feminine nouns rose from 1 to 71, dwarfing the increase in sample size. The greatly increased frequency of  late period gender errors on preposition + article combinations in the main data in Table 15 does not seem, therefore, to be merely an artefact of sampling. The much greater error frequencies from the 1370s on seem to have been a genuinely new development.

4 Discussion Two main conclusions seem warranted on the basis of  the findings presented above. One is that in the early fourteenth century the writers of  these petitions showed a highly accurate knowledge of noun gender.

286

Richard Ingham

The other is that in the last part of the fourteenth century they either lost that knowledge or at least ceased to access it when producing preposition + article combinations. If  these claims are to be properly informative, an account is needed of why the observed change took place when it did: what was upholding competence in insular French up to that time, and why did it then give way? The contrast between the frequent agreement errors in weak ‘e’ contexts (Table 16), and their avoidance with preposition + article combinations in the same time-frame (Table 15), makes it dif ficult to appeal to explicit instruction as a way of accounting for the maintenance of Anglo-Norman in the early fourteenth century. If  the people who wrote the documents analysed were being taught French as a foreign language, some non-trivial level of gender assignment errors would be expected across various contexts: the gross imbalance discovered between preposition + article contexts and weak ‘e’ contexts would have no reason to arise. If  AN was still being learned naturalistically from spoken interaction, however, the outcome observed is readily explained. Child learners would have heard gender marking consistently realized in the preposition + article forms, but not with the weak ‘e’ forms. They would have registered the gender properties of nouns on the basis of their distribution in input – young children’s abilities in this domain are well documented (see e.g. Clark 1985, Granfeldt 2005). AN users’ largely successful performance up to the mid-fourteenth century on possessive determiners, reported by Ingham (2010), had the same ontology. We do not advocate a wholesale reversion to the ‘maximising’ position adopted by Legge (1980), who made two logically distinct claims, one that the use of  French was widespread in later medieval English society, and the other that it was a genuine spoken language. The first claim was undoubtedly overstated. Only those with access to a school education, or to life in aristocratic households, would have been able to benefit from childhood exposure to French. Consequently, speakers who acquired a naturalistic competence in French would have made up only a very small proportion of  the population as a whole. To this extent the ‘minimising’ perspective taken by Berndt, Rothwell, Lass and others is valid. However, for the people using it that does not mean that AN was reduced to being

The Decline of  Bilingual Competence in French in Medieval England

287

a poorly known foreign language. It is our view that spoken transmission in childhood meant that it was maintained within these social circles as a natural language variety, just as Legge envisaged it, until the conditions for its transmission collapsed. The findings on gender agreement in this study and in Ingham (in press) converge on a similar time for when that process took place. By the 1370s, gender marking errors outside weak ‘e’ contexts were becoming common. Why should this have happened just then? To attribute it to English inf luence, as if the latter would have waited until that point to intervene, is unconvincing: AN users had mostly been bilingual for around two centuries (Short 1980). In our view, a crucial point is that at that time the generation of clerks growing up after the Black Death would have been entering the profession. According to Trevisa, around that time the use of  French as a medium language in grammar schools was abandoned. Kibbee (1991) has suggested that the Black Death, whose incidence was particularly high in towns and religious communities where AN was used, was related to the demise of  AN. We would add that it was the role of priests to look after song schools, so the admission of novices to the clergy having little knowledge of French would have spelt the end for the transmission of AN via that route. Pupils would then have arrived at their grammar school without the basic knowledge of the language gained from spoken interaction that their predecessors would have enjoyed prior to the Black Death. On the hypothesis adopted here, clerks growing to maturity around 1370, and writing the documents surviving from that time, would have been the first generation not to have experienced childhood naturalistic exposure to French in song schools. A dif ference between the outcome of the present enquiry and that of  Ingham (2010) must nevertheless be acknowledged. The latter found that from 1310 to 1369 gender agreement errors were negligible. However, as shown in Table 15, the incidence of gender agreement errors with preposition + article forms between 1310 and 1369 stood at over 30 errors (not far short of 10 per cent of  the 338 contexts), which was surely too high to consider as negligible. Some loss of the ability to mark noun gender on preposition + noun combinations evidently started to take place before 1370.

288

Richard Ingham

Why should a somewhat higher rate of agreement marking errors have occurred in the preposition + article data than with the possessive determiners analysed by Ingham (in press) during the decades preceding 1370? If we look at the problem from a language learner’s perspective, two sorts of information must be acquired in order to mark gender accurately. The gender of the noun must be known, and specific agreeing forms, such as al, must identified as part of  the gender agreement system. These are logically separate, as we see from the existence of  the so-called ‘epicene’ (i.e. invariable) adjective forms such as grant, which in Old French did not show gender agreement; the use of grant as a gender-neutral form obviously did not mean the speaker or writer was unaware of noun gender as such. Likewise, the use of al or del with a feminine noun may not have denoted lack of knowledge of gender assignment: the contracted form itself might no longer have been accurately identified as restricted to masculine contexts, rather than behaving as an invariable form. In the long run, of course, as previously gender-marked forms were neutralized, the fewer opportunities there would be for noun gender distinctions to be realized on systematically contrasting forms, and eventually the noun gender system would collapse. This gradual process of  losing gender marking context by context seems to have been at work in the later stages of  AN: first, by the end of  the thirteenth century gender contrasts involving weak ‘e’ on noun modifiers were no longer being reliably made, then as the fourteenth century went on errors arose with preposition + article combinations, and finally in the last decades of  the century possessive determiners become error-prone. The process followed this particular sequence because it was originally driven by phonology. Gender agreement with modifiers such as cel and cest was distinguished only by a weak ‘e’ syllable, which was liable to phonological elision, so this was the most vulnerable context. Le and la before a consonant formed a more distinctive contrast as the syllable also contained the initial consonant segment /l/ and so could not simply be elided.11 Accurate agreement marking in PROME was found to have lasted 11

In modern French fast speech, the /l/ segment can assimilate to a vowel at the end of  the preceding word, e.g. /tylmeisi/ ‘Tu le mets ici’. However, in medieval French and especially in AN, the retention of word-final consonants would have militated against such a process.

The Decline of  Bilingual Competence in French in Medieval England

289

somewhat longer here. Still, the opposition involved weak ‘e’ and to that extent was fragile. Le increasingly became used with feminine nouns, so al and del with a feminine noun could arise if  the weak ‘e’ of  le preceding a feminine noun was elided, as was prone to happen. This stage could ensue only once the le/la contrast started to erode, which was not yet the case in the late thirteenth century in PROME, but happened in the first half of  the fourteenth century; we then see the association of del and al with masculines starting to be undermined. In possessive determiners, on the other hand, which were unaf fected by the attrition of weak ‘e’, gender marking remained highly accurate, only starting to show frequent errors from the 1370s onwards. The observed sequence of gender agreement breakdown thus followed a phonologically motivated series of steps, from the most vulnerable (weak ‘e’ in modifiers), to the least vulnerable (possessives), via preposition + article combinations as an indirectly vulnerable category. It originated as a phonological process but eventually came to af fect gender marking as a grammatical property. In modern advanced instructed L2 learners, a common finding, already noted, is that gender agreement is most successfully achieved with articles, rather than other types of modifier. The PROME data do not show this pattern. Agreement with articles, both per se and in preposition + article combinations, starts to vacillate before agreement with possessive determiners does. The nature and sequencing of  the error types observed here show a strong ef fect of phonological conditioning, i.e. a natural process of  language transmission at work which was reliant on spoken communication rather than formal instruction.

5 Conclusions In this study an empirical investigation has been carried out of the accuracy of the language used in approximately the final hundred years of the career of  AN. The petitions to Parliament in the electronic version of  PROME were analysed in order to establish the writers’ level of accuracy on gender

290

Richard Ingham

marking, known to be a good discriminant of native-like ability in a language. The main focus was on the accuracy of gender agreement with preposition + article combinations. For most of the period studied, those who wrote the petitions in PROME showed overwhelmingly correct use of the relevant forms depending on the gender of the noun. The key period in which change in the status of AN became observable was the decade of  the 1370s, when errors in preposition + article forms suddenly ballooned. The results of  this study are consistent with those of  Ingham (2010) on gender agreement with possessive determiners in that the main change took place in the 1370s. Until then, gender errors in Anglo-Norman were largely restricted to contexts where phonological attrition had eroded the means of making gender marking contrasts. Even in the earliest files, from around 1300, they were quite numerous in contexts where phonological neutralization of gender marking in AN is known to have been present. But preposition + article forms, the domain studied here, maintained accurate gender agreement until much later. The sequence of contexts in which gender errors arose seems to bear witness to a transmission scenario up to the mid fourteenth century in which spoken input formed the source of  learners’ knowledge of  the language, not formal instruction. At this distance in time it is clearly impossible to be certain about the precise nature and extent of  the contexts in which this took place. By analysing the textual record, however, it becomes possible to advance some tentative ideas about the way this took place. The PROME database is particularly advantageous here, as the texts it makes available are based on manuscripts roughly contemporaneous with their composition, and are within the same genre, written by the same sort of people on the same sort of  topics. Consequently, the changes observed in the series of petitions in the behaviour of  the linguistic variables selected can be reliably imputed to the evolution of  the language over time, rather than to extraneous complicating and confounding factors which so often bedevil diachronic research. Our findings suggest that Anglo-Norman lost its status as a spoken natural language among educated professionals after the Black Death, rather than following the loss of Normandy to the English crown in 1204, a landmark regularly mentioned by textbook authors seeking politico-historical

The Decline of  Bilingual Competence in French in Medieval England

291

explanations for linguistic developments. The expansion of  the functions of insular French in the later thirteenth century and fourteenth century was not the work of clerks struggling to write an artificial language they no longer properly controlled, the scenario retailed in many mainstream treatments. It took place because those who were responsible for leaving the documentary records of medieval England were still competent in the living language they had acquired in childhood, which was now achieving the higher status of a second language of record, probably with the active encouragement of the royal government (Lusignan 2004). The extensive use of insular French as a written vernacular at this time is only a paradox, then, if viewed through the fallacious assumption of recent decades that it was no more than a rather poorly learnt written code. Its superficial dif ferences from continental French, though real, have for too long inhibited a proper understanding of  how its survival, and also its eventual demise, related to the socio-historical context of the late medieval period in England, when its matrix of  transmission was destroyed by the impact of  the Black Death.

References Berndt, R. 1972. ‘The period of  the final decline of  French in Medieval England (14th and early 15th centuries).’ Zeitschrift für Anglistik und Amerikanistik, 20, 341–369. Bolland, W. (ed.). 1914. Select Bills in Eyre, AD 1292–1333. Vol. 30. London: Selden Society. Clanchy, M. 1993. From memory to written record. England 1066–1307. Cambridge, MA: Wiley-Blackwell. Duf fell, M. 2005. ‘Some Phonological Features of Insular French: A Reconstruction’ In: R. Wright & P. Ricketts (eds), Studies on Ibero-Romance Linguistics Dedicated to Ralph Penny. Newark, NJ: Juan de la Cuesta, 103–125. Franceschina, F. 2005. Fossilized Second Language Grammars: The acquisition of grammatical gender. Amsterdam: Benjamins. Granfeldt, J. 2005. ‘The development of gender attribution and gender agreement in French: a comparison of  bilingual first and second language learners.’ In:

292

Richard Ingham

J.-M. Dewaele (ed.), French as a foreign language: Multidisciplinary approaches. Clevedon: Multilingual Matters, 164–190. Greimas, A. 1992. Dictionnaire de l’ancien français. Paris: Larousse. Hunt, T. 1985. ‘Anecdota Anglo-Normannica.’ The Yearbook of  English Studies 15, 1–17. Ingham, R. 2010. The transmission of later Anglo-Norman: some syntactic evidence. In: R. Ingham (ed.), The Anglo-Norman language and its contexts. Woodbridge: Boydell. Johnson, J. & E. Newport. 1989. ‘Critical period ef fects in second language learning: the ef fect of maturational state on the acquisition of  English as a second language.’ Cognitive Psychology 21, 60–99. Kibbee, D. 1991. For to Speke Frenche Trewly: The French Language in England, 1000– 1600. Amsterdam: J. Benjamins. Kibbee, D. 1996. ‘Emigrant languages and acculturation: the case of Anglo-French.’ In: H. Nielsen and L. Schoesler (eds), The origins and development of emigrant languages. RASK supplement No. 6. Denmark: Odense University Press, 1–20. Lass, R. 2006. ‘Phonology and morphology.’ In: R. Hogg & D. Denison (eds), A History of  the English Language, 43–108. Legge, M.-D. 1980. ‘Anglo-Norman as a spoken language.’ Anglo-Norman Studies 2, 108–117. Lusignan, S. 2004. La langue du roi. Paris: Presses Universitaires de France. Nichols, F. (ed.). 1865. Britton: The French Text Carefully Revised with an English Translation, Introduction and Notes. 2 vols. Oxford: Clarendon Press. Orme, N. 1973. English Schools in the Middle Ages. London: Macmillan. Oyama, S. 1976. ‘A sensitive period for the acquisition of a nonnative phonological system.’ Journal of  Psycholinguistic Research 5:3, 261–283. The Parliament Rolls of Medieval England, 1275–1504 [CD], gen. ed. C. Given-Wilson et al., Scholarly editions, 2005. Patkowski, M. 1980. ‘The sensitive period for the acquisition of syntax in a second language.’ Language Learning 30, 449–472. Pope, M. 1934. From Latin to Modern French, with especial consideration of Anglo-Norman: phonology and morphology. Manchester: Manchester University Press. Rothwell, W. 1976. ‘The role of French in thirteenth-century England.’ Bulletin of the John Rylands University Library of  Manchester 58, 445–466. Short, I. 1980. ‘Bilingualism in Anglo-Norman England.’ Romance Philology 33, 467–479. Singh, I. 2005.The History of  English: A Student’s. Guide. London: Hodder Arnold. Tanquerey, F. (ed.). 1916. Recueil de lettres anglo-françaises (1265–1399). Paris: Champion. Rothwell, W., Gregory, S. Trotter, D. 2005. Anglo-Norman Dictionary: revised edition, A–C; D–E, 2 vols. London: MHRA. Wilson, R. 1943. ‘English and French in England 1100–1300.’ History 28, 37–60.

Rembert Eufe

Merovingian Coins and Their Inscriptions: A Challenge to Linguists and Historians

Abstract The present article gives a report on a joint project of  historians and linguists who are engaged in examining a collection of 478 Merovingian coins conserved in the Bode-Museum in Berlin. The coins minted in the Frankish kingdoms from the late sixth until the early eighth centuries are of a special type: they show a large variety of dif ferent personal and place names. The anthroponyms of  the so-called monneyers are of  Gallo-Latin and Germanic origin, inviting conclusions on the relations and the mixing between Gallo-Roman and Germanic population groups and their language uses. In contrast, among the toponyms those with Gaulish origin largely prevail. It is considered how both linguists and historians can use this source to gain new insights.

1 Introduction From 2006 to 2009, historians and specialists in German and Romance linguistics from the universities of Berlin, Paderborn and Regensburg carried out a research project focused on a collection of 478 Merovingian coins conserved in the Münzkabinett of the Bode-Museum in Berlin. Their aim has been to compile a printed and an electronic catalogue of  this second largest collection of  Merovingian coins in the world.1 1

Greule, Jarnut, Kluge & Selig (in preparation) and the digital version IKMK, already available on the museum website.

294

Rembert Eufe

Why are linguists involved in a project like this? Normally, coins display just the name of a king or an emperor, perhaps with a short inscription in his praise or something similar. We do have coins of  this type from the Merovingian period, but mostly from the earliest part (see below). From the late sixth until the early eighth centuries, the coins were dif ferent: they came to show a large variety of dif ferent personal and place names, whose number has been estimated at 1,2002 and 600 respectively for the 8,000–10,000 coins alone that we know of at present (Felder 2003: 22). Given these figures, the identification of the names and the explanation of  their structure is an ambitious but also very fruitful task.

2 The value of  the Merovingian coins as a source It should be recalled that the Merovingian period has left us only very few sources of  linguistic information – above all lives of saints and chronicles (posing many unanswerable questions concerning their origin and transmission, as it is typical for texts conveyed by medieval manuscripts). As for the approximately 200 known Merovingian charters, two thirds of them prove to be falsifications, following the recent edition by Kölzer (2001: XII). Only the authenticity of the thirty-eight original charters conserved as originals (not as copies) remains beyond any doubt. These charters contain 12,000 words – whereas one of  the most important chronicles, the ten books of  histories written by the bishop Gregory of  Tours (Buchner 1964) at the end of  the sixth century, contain 120,000 words! An advantage of coins is that they are often easier to localize, especially when place names occur on them, as is the case on many of our coins (other­ wise the find spot and the style and motif of  the illustrations are of  help). In addition, they can be easily dated, e.g. on the basis of the names of kings mentioned on the coins, where the dates of their reigns are known, but also in the light of  the dif ferent coin types, in particular silver or gold coins. 2

Kluge (2007: 83) even gives the number of 2,000 personal names.

Merovingian Coins and Their Inscriptions

295

However, it must be emphasized that not all the coins of fer fairly readable inscriptions. Sometimes the f lan was too small for the die, so that only the lower part of  the legend appears. In other cases, the letters are blurred because they were coined with a die worn of f by long previous use. Furthermore, the coins raise considerable philological problems, not only due to abbreviations, but also due to inaccurate engraving (note that the coins are very small, being only the size of a thumbnail), including the commutation of letters. Even the direction of the legends cannot be taken for granted, as is shown in an example mentioned by Felder (2003: 25): according to him, the toponym +COSRVBET on a Merovingian coin is actually a ghost-name. When read in the opposite direction and with a G in the form of an S, it appears as TEBVRGO, to which [STRA]+ can be added which then reads STRATEBURGO C[IVITATE], Strasbourg.3 It has to be underlined that the value of  the coins as a linguistic source is very dif ferent for scholars of Romance and Germanic languages. Because the oldest preserved Franconian texts date from the ninth century, the attested names of  Germanic origin are important as remnants of older lexical material for Germanic linguists. But the linguistic environment, to which the legends belong as written sources, is primarily a Latin and Pre-Romance one; not surprisingly, the Latin of the inscriptions does not correspond to the standards of  Classical Latin, but exhibits the graphophonetic and morphological characteristics of  Merovingian Latin which is of particular interest to Romance linguists. An adequate interpretation of  the data has to account for the context of use and production in terms of a recontextualization,4 i.e. reconstruction of  the sociolinguistic and pragmatico-textual conditions, crucially based on the findings of  historians.

3 4

On other coins, the city appears with its Gaulic name ARGENTORATE (see below). Selig / Eufe (in preparation) and the considerations in Greule, Jarnut, Kluge & Selig (in preparation).

296

Rembert Eufe

3 The Berlin collection The 478 Merovingian coins conserved at the Bode-Museum can be assigned to three consecutive periods. The first period, ranging from 500–585, is the period of so-called ‘pseudo-imperial’ coins, emulations of Roman coinages, showing more or less botched or feigned imitations of the model legends. Eighty-five coins belong to this period; however twelve of them cannot be considered as ‘pseudo-imperial’ but ‘royal’ coins with the names of Frankish sovereigns. The second period, running from approx. 585–675, is characterized by ‘typical Merovingian’ gold coinages with frequent mentions of monneyers’ names. In the Bode-Museum collection, a majority of 268 coins belong to this period, including thirty-one royal coins. During the third period from 675–750, the coinage changed in material, switching to silver. One hundred and eighteen coins of  the collection stem from this period. To complicate matters, besides two bronze coins even five counterfeits are present, produced mostly in the nineteenth century.

4 Interdisciplinary aspects The central issue of this chapter is the usefulness of collaboration between historians and linguists and their dif ferent perspectives; in this light, research on the coins with their legends consisting mainly of names opens the question on the importance of onomastic research for linguists and historians. In linguistics, onomastics certainly has not been very popular in recent decades. One reason might be that it represents an area of study within linguistics which has the strongest appeal for the amateur researcher, which makes the value of many investigations hard to evaluate. Furthermore, in many respects (above all from a semantic perspective) names cannot be seen as being fully part of  the language system. In the perspective of  linguistic approaches which insist on explaining language solely in terms of its internal structure, the strong interdisciplinarity of

Merovingian Coins and Their Inscriptions

297

onomastics represents a further disadvantage: for it is usually impossible to focus on a name without also looking for information on its bearer, be it a person (his or her dates of birth and death, origin, relatives etc.), a settlement (its foundation, size, geographical position, population etc.) or something else. Of course, linguists remain primarily interested in names as manifestations of historical languages and should investigate them with regard to other aspects of the languages in question. Moreover, they try to draw conclusions on the linguistic behaviour of those who bear or use the names. For obvious reasons, historians follow the opposite perspective: they focus on the persons and institutions behind the names that is the latter are interesting whenever they provide further information on the former (cf. Geuenich / Hawicks 2006: 82). Some gained researchers (including current project members) working on the coins and their legends can draw on their experience within the context of a very large older project, Nomen et gens. It resulted in a large database of several thousands of name forms, collected in well-known early medieval sources.5 Among other things, this resource should help to answer the question whether or to what extent names of this period allow us to conclude to which ethnic groups (dealing mainly with Germanic groups) their bearers belonged and thus whether typical naming practices for certain ethnic groups existed (Patzold 2008: 139).

5 The anthroponomastic uses of  the Merovingian period The Merovingian rule brought about the rearrangement of  the ‘anthroponomastic landscape’ of Gaul, which consisted mainly of Latin and Greek names (formed in the classical period, but very often in advanced imperial

5

Cf. the list of already searched texts, mainly of hagiographic type, in Patzold (2006: 148–151). For the problems in finding a system for arranging the onomastic data which meets the standards of  both historians and linguists see Geuenich / Hawicks 2008.

298

Rembert Eufe

times, with or without Christian implications)6 but also comprised some Semitic and Celtic ones (be they loans or calques and cover names, often hard to determine with certainty).7 These names were to a very large extent supplanted by the Germanic names of the Franks. As a survey of episcopal names appearing in council records from the provinces in Gaul8 suggests, Germanic names are in the majority from the seventh century onwards.9 It has to be kept in mind that bishops were initially drawn from the ancient Senatorial class, the traditional Roman upper class with privileged access to Latin education. The rate of Latin names is even lower among the names of  the persons close to the king mentioned in the Merovingian charters (15 per cent, for both religious and civil dignitaries; Bergmann 1997: 105). Unsurprisingly, the Germanic inf luence was stronger in the North, as is shown by two important sources from the early ninth century, the polyptych of  St-Germain des Prés (Paris) with more than 90 per cent Germanic names (Hägermann 1997: 111–112), contrasting with 38.5 per cent of  Germanic names in the polyptych of  St Victor (Marseille, 814; Bergh 1941: 184). The dif fusion of  Germanic names is remarkable because it contrasts with the linguistic Latinization / Romancification of the Germanophone groups in most parts of the Frankish kingdom. In general, their predominance in the Galloromance area lasted until the twelfth century, when saints’ names of Latin and Greek origin started to prevail (such as Jean and Pierre, which became the most frequent male names in the late Middle Ages). 6 7

8 9

For general information on the Latin and Greek name systems see Kajanto (1963, 1965, 1973), Solin (1995, 2003a, 2003b), Solin & Salomies (1994). Coşkun / Zeidler (2003: 10) suspect the name of  the fourth-century Bordelaise Ausonius (attested also as a monetary name in the Berlin collection) to be a Gaulish name (perhaps derived from Auso or Ausios) disguised in Latin. In fact, a considerable number of  the Latin and Greek names on the coins of  the Berlin collection appear concentrated in Gaul, indicating a distinct regional anthroponomastic profile, ascribable at least partly to Gaulish inf luence (cf. Eufe (in preparation)). Note that we refer here just to the provinces of  Gaul, comprising the former Roman Germania inferior and Germania superior up to the river Rhine. In fact, no Merovingian coins are known to have been minted east of it. Kajanto (1973, 390–392) counts 21.8 per cent of Germanic names in the period from 511–600, which increased to 59.7 per cent between 600–695. For further statistics concerning the spread of  Germanic names in Gaul cf. Bergh (1941: 185).

Merovingian Coins and Their Inscriptions

299

The names of  Germanic origin are of  three dif ferent types: – bithematic names (as in other Indo-European languages, such as Ancient Greek, but not in Latin), composed of  two elements, e.g. Adalbertus, derived from athala ‘noble’ + berhta ‘brilliant’. They represent dif ferent types of compounds, mainly of a determinative morpho-semantic structure (with a first element modifying the meaning of  the second, e.g. Theoda-rich / Diet-rich ‘people’ + ‘powerful, rich’ > ‘powerful (second element) among the people’) but also copulative (e.g. Bern-ulf   ‘bear’+‘wolf ’) and Bahuvrīhi-composita (e.g. Wolfhelm ‘the one with the helmet decorated with the wolf ’). Note that by this time names without any semantic relation between their two components were created (Greule 1996: 1184), especially by combining elements of  the names of  the bearer’s parents. – monothematic names, often to be regarded as short forms of  bithematic names, e.g. Dodo and Dedo, which our scholars of German have traced back to bithematic names with the first element theud-a ‘people’. These names are supposed to have represented more ‘colloquial’ forms (Greule 1996: 1185).10 Interestingly, in some cases monneyers’ names seem to appear on some coins in a bithematic form, on others in a monothematic one, e. g. Alligisels and Alloni on coins from Angers (Breillat 1935: 24; although we could not find any examples on the Bode-Museum coins). – finally, derivations by means of various suf fixes of  Germanic, Latin or mixed origin can occur. In this respect, the Germanic names of our collection do not show great diversity, only some derivations by means of -īna-, -līna- and -linus / -lenus (a typical Merovingian suf fix of mixed origin, see below).

10

Sometimes, the monothematic short names are attested in the sources, as in the case of ‘Berthranno sive Bettone’ (abl. / oblique case in -one), mentioned in the will of a Merovingian bishop (Weidemann 1986: 28).

300

Rembert Eufe

6 The names of  the monneyers According to Felder (2003: 26), who examined the names on the 5,000 Merovingian coins conserved in the Cabinet des Médailles at the National Library in Paris, some 75 per cent of the monneyers’ names are of Germanic origin. Our research on the coins of  the Berlin collection shows a similar tendency: leaving aside the pseudoimperial coins of the first period, displaying mainly names of kings and Roman emperors and only two monneyers named Maurentius11 (Lyon) and Dedo (Alsheim), we find 109 names of monneyers from the second period12 from 585–675, among them 68 ≈ 62.39 per cent13 of clear Germanic origin, like Adelbertus, Aegulfus, Bertechramnus, Charegisilus, Filaharius, Launomundus, Madelinus and Anglo, Babo, Dedo, Suno and others. To these could be added six other names (≈ 5.5 per cent) that show Germanic bases with an attached diminutive suf fix -lenus / -linus quite common in Merovingian personal names (Abolinus, Agilinus, Audolenus, Gaudelinus, Leobolenus and Theudelenus). Its frequent occurrence whith , abnormal for long -ī- which is in general conserved in Romance (in contrast to short -ĭ-), does not support the explanation as an exclusive outcome of the Germanic diminutive suf fix *-līna-z. It rather suggests a fusion of  the latter with the popular element *wini-z ‘friend, beloved, spouse’ (with short -ĭ-) and extraction of  the suf fix from names 11

12 13

Due to space constraints, we restrict ourselves to some recapitulary remarks based on the more detailed expositions in Eufe (in preparation) and Greule, Jarnut, Kluge & Selig (in preparation, with comments on the structure and origin of  the single names), where the legends are also reproduced in forms more close to what is visible on the coins and more precise datations are given. For simplicity’s sake, we mention the names here only in an adjusted form, which is based to a certain extent on our interpretation of the legends that are sometimes actually quite dif ficult to read and reproduce exactly (on this aspect cf. Eufe in preparation). See Greule, Jarnut, Kluge & Selig for more precise datations of  the single coins and the names on them. Percentages have been calculated without regard to unidentifiable names like AOMIO MON (ca. 585–675) or INTE MONETA (ca. 620–640) from the total account.

Merovingian Coins and Their Inscriptions

301

like the aforementioned Madelinus -v-, resulting in forms like CAV[Λ]-LOΠ Γ40 and also + CΛVELONO – VICTVI (three cases against four with , next to three others where the relevant part of the name is abbreviated or illegible).41 Moreover we do have a large number of case endings in comparable syntactic positions. It is important to recall that Old French (as Old Occitan) displays a two-case system.42 The endings of the personal names on the coins testify to the genesis of this system, representing for the second declension (o-declension) a previous stage with a 3-case system. Names like Maurinus or many Germanic names, almost always of a bithematic structure like Sigimundus,43 appear often with -us in the function of a Casus rectus (28 occurrences in the second period).44 In addition, there are some attestations 40 The legend is very and almost illegible, but the can still be clearly discerned. 41 For an overview of phonetical and phonological changes attested on Merovingian coins, cf. Chambon / Greub 2000, who even find evidence for the emerging dia­ topic subdivision of  Gallo-Romance (Chambon / Greub 2000: 167, 171, 174–175). Following Selig / Eufe (in preparation), the sparse occurrences of  the features in question should be rediscussed in a broader picture of  the formation of  GalloRomance. 42 This is actually astonishing because in many other respects, French appears as one of  the Romance languages which has developed the most distinctly from Latin – like Romanian, which until the present day has a two-case system (although dif ferent in structure). 43 An exception is monothematic Badus which is regarded as a ref lex of Germ. *baduor *badwa- ‘fight’ (cf. the comment in Greule, Jarnut, Kluge & Selig). 44 In the following, we only indicate the figures that we could ascertain for the second period. For the third period, the data from the Berlin collection are not suf ficient.

308

Rembert Eufe

of  the genitive case in place names, like SCI PETRI. But for the majority, these names appear with -o which might represent a Casus obliquus in expansion of use (36 occ.). Unfortunately, Merovingian Latin shows oscillations and blendings of  the Latin second and third declensions (use of  the types Gallus, gen. Galli and Gallo, gen. Gallonis, with occasional fusion in Gallus, gen. Gallonis and similar), which means that in cases like PRIMANO MONETΛ V and AEGVLFO MCN it cannot be entirely discarded that they are actually intended as nominative forms of Primano, gen. Primanonis (comparable to Petro, gen. Petronis) or Aegulfo, gen. Aegulfonis, respectively.45 The third declension is common with monothematic Germanic names like Betto or Anglo, but also with bithematic names like Chagnomiris, which appear with -o resp. -is / -es as a Casus rectus (9 occ.) and -one / -oni resp. -e / -i (Casus obliquus, 13 occ.). A genitive occurs in the place name IOHANNIS PORTO. And why should historians care about this project? Of course the personal names are indicative of  the contact between Gallo-Latin and Germanic-speaking groups. Furthermore, the appearance of the place names on the coins gives precious proofs to historians for the existence of  these places during the Merovingian period. In consequence, everyone who writes the medieval history of a certain town or region in the former provinces of  Gaul should control if  the place(s) he deals with are mentioned on the coins. It must be stressed that for the localization of the place names, a lot of work has still to be done: In most of the older numismatic publications no toponomastic literature has been consulted, the places on the coins are identified on the basis of rather vague phonetic similarities with toponyms on present-day French maps, at most mentioning the existence of remnants from Gallo-Roman antiquity – which actually are very frequent all over former Gaul. Only recently, in Chambon 2001, a first ef fort has been made to examine the localizations current in the numismatic literature on serious toponomastic and philological grounds, according to which only 3/4, 3/5 and 1/5 of  the localizations in Limousin, Arvernie and Rouergue, 45 On this question cf. Pitz (2009), who regards these oscillations rather as a continuation of  Latin name morphology than as a result of substratum inf luence exerced by Germanic noun inf lection.

Merovingian Coins and Their Inscriptions

309

respectively, can be maintained (Chambon 2001: 348–349). Actually, the identification of the places which the place names in the sources refer to is a very crucial problem for any historical research, relying not only on coins, but also on other sources like manuscripts.46 It is hardly ever accorded the attention it deserves when preparing editions and preparing indices for them, which require a good deal of philological, linguistic and historical background knowledge and meticulous verification of sources, editions and reference guides.47 But how can research on the perception and the division of space or the concept of border in the Middle Ages, for example, arrive at any valid conclusions if  the places they rely on were most likely located elsewhere than is assumed?48

9 Continuity of place names, continuity of places? But there is much more for historians about Merovingian coins. They can be discussed in the context of  the old question of whether the end of  the Roman Empire represents a sudden decline, the ‘crash of civilization’ caused by the barbarian invasions as it has often been regarded as of  humanism onwards.49 Or should we better think of it as a ‘surface event’ with scarce impact on a continuous f low of  transformations? 46 The importance of  toponomastic research has already been stressed by Bloch 1956, 8–14. 47 Cf. the general remarks in Chambon 1997 and 2009. 48 A very striking example is Pitz 2007 on the places names of a mansus indominicatus in Lorraine, in the urbarium of the abbey of Prüm (Rhineland-Palatinate, Germany) as its possession. The author re-examines the identifications of the place names, correcting many of  the hitherto proposed identifications and observing that the corresponding places are not dispersed, but situated in fact close to each other, forming a compact and clearly confined manor. 49 On the prerequisites and implications of this question, cf. Fehr (2010: 42–69). Note that Ward-Perkins in his ‘schematic reasoning’ (Fehr 2010: 62) relies on the disappearance of  Roman bronze small coins (Ward-Perkins 2005: 110–117).

310

Rembert Eufe

As shown above, the personal names on the coins do testify to the anthroponomastic reorientation towards names of  Germanic origin after the Frankish takeover, while this does not hold for the place names. To a great extent, these are of Gaulish and Latin origin, showing an impressive toponomastic continuity which deserves our attention as such, in accordance with the fact that place names represent an important form of collective memory and tradition. Interestingly, place names developed dif ferently in England: The Anglo-Saxon dominance brought about a ‘vast renaming’ (Gelling 1995: 787) of places so that the place names on the Merovingian coins confirm by and large the view of  Wickham (2009: 200): The real contrast inside the ex-Roman provinces was not between societies that had been invaded or conquered and the others, but between the Continent and Britain: in the former, the basic Roman political and social structures survived (though they were in most places ramshackle and underfunded).

Although the etymological distribution of the place names appears indicative of a settlement history of  late-antiquity Gaul, further investigations are of course needed in order to make clear that the preservation of antique place names corresponded to a continuity of institutions and social life, not just continuity of ruins. In this respect, coins are significant by their very existence, testifying to a certain economic or at least financial activity. It has already been mentioned that a large number of 37 (against 45 other) place names of  the Berlin coins refers to main places of  former Roman civitates – note that the ‘cellular civitas-structure inherited from the Roman period’ was maintained in ‘perhaps four-fifths of Roman Gaul’, in any case ‘[…] no region of  Frankish Gaul exhibits the total breakdown of the civitas-framework which occurred just across the Channel’ (Loseby 2006: 93). But do the old regional capitals turn out to be actual centres of Mero­ vingian coinage, and if so, did it include all of  them? Judging only by the number of monneyers known to have worked there, the following places stand out with ten to twenty-five dif ferent names: Angers, Clermont-

Merovingian Coins and Their Inscriptions

311

Ferrand, Orléans, Bourges, Bordeaux, Limoges, Maastricht, Marsal, Metz,50 Paris, Poitiers, Rennes, Rouen, Rodez, Toul, Toulouse, Trizay, Vienne, Quentovic, exceeded by Chalon-sur-Saône with even thirty names (Pol 2002: 353).51 In addition, other important mints were at work in Reims, Verdun, Dinant, Namur, Treves, Cologne, Autun, Mâcon, Troyes, Lyon, Banassac, Brioude, and, albeit to a lesser extent, also Tournai (Dahmen in preparation). Now, most of  these places are indeed well known through archeological and historical research as important administrative, religious and economic capitals of  Roman civitates.52 But those who were not definitely deserve a comment: Dinant, Namur, Dorestad, Maastricht and Quentovic, together with Huy, ef fectively emerged as new centres in the heartland of  the Franks and north of it. They were situated in or near ancient Roman places, but these had at best the form of small vici or castra. Furthermore, continuity of  habitation was ‘fairly rare in the area between the Meuse and the North Sea’ (Verhulst 1999: 5), traceable almost exclusively in Maastricht, Tournai (a former civitas, however), Boulogne and Huy, of which only the latter gives prove of  functional continuity in the form of artisanal activities (Verhulst 1999: 21). Especially the medieval emporia of  Dorestad and Quentovic indicate the importance of  harbours as places of  Merovingian coinage, also af firmed by Marseille with its very large production of coins. Furthermore, we have to mention Mâcon, originally a castrum with a port on the Saône. However, in this case Christianity must be named as a decisive factor of transformation: The Merovingian king Childebert I is said to have installed the cult of  Saint 50 Halsall (1996: 61) perceives for Metz a civilizational hiatus in the 5th–6th centuries, calling it a ‘town without urban life’ – in any case, if ever traces of urban life subsisted, it seem to have been in civitas main places like this rather than anywhere else. 51 Moreover, coins from Chalon-sur-Saône and Lyon are preserved with two monneyers’ names on one coin. Of course we can judge only from the coins that are known to us, ignoring the original output of coinage. Future large coin finds may of course change the picture. 52 In Carolingian times, then, Pippin again limited coinage to the main places of  the civitates, some new centres in the north (Quentovic, Dorestad, Dinant, Namur, Huy and Maastricht) and some large suburban abbeys (Saint-Martin de Tours, Saint-Pierre de Sens, Fixot 1980: 527).

312

Rembert Eufe

Vincent of Saragossa in the cathedral of Mâcon in 543 ( Jeanton 1934: 15). Similarly, Brioude, maybe an old Roman vicus, owes its importance to the veneration of  Saint Julian, already beginning in imperial times and later stimulated by Visigothic and Frankish rulers. Of course, there would be a lot to say on the imprint of  Christianity and the Catholic Church on the landscape of  Gaul;53 as it has certainly been much more important than that left by the Germanic rulers. However, the latter exerted an indirect inf luence inasmuch as they relied on the first, giving their own impact by granting the construction of churches and monasteries or even founding them themselves and promoting the veneration of certain saints, like Remigius of Reims, occurring in S-ANTI REMI VICO. In fact, the coins bear testimony of  these general tendencies, with pieces showing legends like ECL + // ID / EPS (maybe from Senlis (Département Oise)), + SCI PETRI // ME[DI]OLΛNO M[O]N (assigned to Saintes (Département Charente-Maritime)), originally Mediolanum Santonum and including since the sixth century a cathedral dedicated to St Peter in its walls) or + SCO FILBER // + GEMEDICO CAV or COL (from Jumièges (Département Seine-Maritime), where Saint Philibert had founded an important monastery in the seventh century).54 The occurrence of ecclesiastical institutions as places of coinage is certainly astonishing and thus calls for further ref lections on the function of  the monetary system.

10 Places names and monetary system The large number of place names on the coins leads us to assume that the Frankish kings did not enjoy a monopoly over coinage (Grierson / Blackburn 1986: 97–98), this usually being a very important instrument of power. Among other things, the number of monasteries dif fered strongly between the civitas capitals! In the Viennensis province, e.g. in ten out of fourteen capitals no monasteries have been detected so far, whereas Marseille is likely to have had eight monasteries, Arles five, Vienne assuredly eight, but maybe even thirteen (Biarne 2002: 125). 54 Cf. the comments on these legends in Greule, Jarnut, Kluge & Selig (in preparation). 53

Merovingian Coins and Their Inscriptions

313

Now it is precisely the designations of ecclesiastical institutions, together with other indications like racio fisci, racio domini, scola Regis or in palacio, which raise doubts whether we really deal here with a coin system in the usual sense. They rather point to a special and extremely particularistic tax system, with the monneyers as of ficers responsible for the collection and registration of taxes in public places and their immediate transformation into coins whose purity they had to attest.55 This would not only fit in with the high value of  the Merovingian gold tremisses, worth approx. £ 50 (whereas a Carolingian denarius corresponds to only £12), so that they could have been only ‘somewhat clumsy aids to exchange’, better serving as ‘convenient ways of hoarding wealth’ (Wickham 2009: 227). But it can also explain why many sea and river ports appear among the most important Merovingian mints, places where extensive customs revenues must have been charged56 – like in Marsal and some other nearby places in the south of  Metz whose importance was related to the extensive production and commerce of salt, also known for antiquity (Stahl 1982, Pol 2001). On these grounds, a study of thirty-four surely identified places mentioned on Merovingian coins from around Limoges has shown that four of them were castra and fourteen are attested as Gallo-Roman or Merovingian vici. In addition, they were all equipped with a parish church and functioned (or still do) as centres of parishes, dedicated to saints popular in Roman and Merovingian times, which further suggests that they were ‘sites of 55

56

This hypothesis is often corroborated by a passage from the Vita of  St Eloy (edited in Krusch 1902: 680–681, cited e.g. by Boyer 2007: 152–153), which reports how a monetarius is about to melt and purify the tax revenues collected on a royal manor, not knowing that it has just been conferred to the saint, who makes him stop his work. Cf. the remarks by Strothmann and Dahmen in Greule, Jarnut, Kluge & Selig (in preparation) – note that this episode documents also the exemption of clergymen and landowners from tax allowances, leaving the revenues at their own disposal. In this light, the introduction of silver coins in the second half of  the seventh century, probably on the initiative of  the mayor Ebroin, appears as a tax reform. Accordingly, Verhulst (1999: 32) supposes that the important numbers of monneyers in Maastricht, Huy, Namur and Dinant ‘could have had a political rather than an economic significance in relation to the position of power of  the Austrasian nobility and in particular of  the Pippinids in the Meuse Valley and their expansion in a northerly direction along this river to Dorestat […]’.

314

Rembert Eufe

evident public nature’ (Boyer 2007: 150). Of  the remaining sixteen place names, one refers to a monastery, all the others are toponyms derived with -acum, indicating old Gallo-Roman rural manors likely to have developed similarly into settlements of public interest. In general, these observations are in line with what we have found out for the place names of  the Berlin collection.

11 New places? However, it is worth noting that on the coins there is evidence for Mero­ vingian institutions which have left some, albeit small, traces on the soil of  Gaul: One of the coins from the Bode-Museum is among those minted in palacio ‘in the royal palace’. At least in one case, Palaiseau (Département Essonne), a settlement is known to have grown around a Merovingian palace.57 Moreover, the Paris collection comprises pieces coined in MALLO ARLAVIS, MALLO MATIRIACO and MALLO CAMPIONE (Felder 2003: 472, 466, 473), all localized in Belgica prima and referring to the thingstead where the Germanic government assemblies met. While the first of these places still seems to await localization, the other two are identified as Mairy (Meurthe-et-Moselle, Buchmüller-Pfaf f 1990: 315–316, Haubrichs 1985: 30) and Champ, one of the two hamlets forming present-day Champneuville (Meuse, Haubrichs 1985: 12, 28). Note that even in these cases, the Latin part of the place name is continued, although place names which are traced back to the Frankish designation for the thingstead do exist, such as Les Maus (near Arras, Pas-de-Calais), Le Malet (near Laon, Aisne) and others (Schmidt-Wiegand 1973: 493).

57

This does not call into question that in many towns, on the contrary, Merovingian palaces were established in older Roman public buildings. Pfalzel (Treves) grew out of a fortified Roman villa of  the 4th century.

Merovingian Coins and Their Inscriptions

315

Acting on the assumption of  the Merovingian coinage as a form of  taxation, two places with ‑(i)acum-names yet deserve a comment, because they appear among the places disposing of important mints listed above: At first sight, the role of Banassac (department Lozère) as a centre of coinage from which a great number of coins are preserved comes as a surprise. It is sometimes associated with the existence of a f lourishing pottery industry in Roman times in the third century. In any case, a considerable portion of royal coins in the name of  Charibert II, king of  Aquitaine (629–632), indicates that it owed its importance to one of the numerous subdivisions of  the Merovingian among its various kings. Little is known of Trizay-sur-le-Lay (department Vendée), regarded as the present-day of fspring of TIDIRICIACO. As already said, it is derived from a Germanic name, like only few other places of Merovingian coinage named after the -(i)acum-type. Their exiguous number contradicts earlier views of  the monneyers acting on behalf of rich landlords,58 as the latter should have borne predominantly Germanic anthroponyms from which their manors’ names would have been derived.59 TIDIRICIACO as a form of Teuderic-iacum is derived from Theuderich, a name also born by Merovingian kings – as well as TEODEBERCIACO, identified as Thiverzay in the same département and derived from Theudebert. These two names might indicate royal Merovingian manors (especially the first with a large number of monneyers who seem to have been occupied with collecting taxes) in a region where actually few Gallo-Roman settlements are to be found (Gauthier 1999: 146). This would fit partly with the division of  Gallo-Romance: On the Atlantic coast, the present-day border between the French and the Occitan area recedes Southwards, across the Loire almost down to the Gironde. But further research would be needed that goes into detail on the settlement history of Pays de la Loire and Poitou-Charentes. 58 59

On these approaches cf. Grierson-Blackburn 1986: 99–100. Buchmüller-Pfaf f (1990: 10) comes across a ‘considerable number’ of -(i)acumtoponyms derived from Germanic anthroponyms in Lorraine which she regards to their majority as ‘(early) Merovingian formations’. Of course the Germanic base of such a toponym does not necessarily indicate a Germanic settlement (BuchmüllerPfaf f 1990: 754). But inversely, Frankish landlords can be expected to bear mostly names of  Germanic origin.

316

Rembert Eufe

12 Conclusion As we have seen, Merovingian coins present considerable challenges that are best met in a joint ef fort. The work of numismatists will always be needed with their expertise in assessing datings, comparing iconography, analysing coin finds etc. Such information on preserved Merovingian coins in museums and private collections should be united in data bases which facilitate access for historians and philologists to this type of source. Historians should find out about the Merovingian tax system and the institutions linked with it, not only for getting more precise judgements on the coins, but also for gaining further insights into the economic and political developments between the deposition of  the last West-Roman emperor and the Carolingians’ coming into power. Moreover, it turns out fruitful to collaborate with archaeologists who can provide data of the settlement history, including seemingly ‘small’ places like those mentioned on many Merovingian coins but not in other sources. And linguists should carry on with research on place names and re-examine existing identifications with attention to historical phonetics and phonology. In summary, the need for close collaboration between historians and linguists appears evident in our case; their cooperation is particularly recommendable in order to value the coins as a source and open up new perspectives on a period for which only very few other sources of information are extant.

References Belfort, Auguste de. 1892–1895. Description générale des monnaies Mérovingiennes par ordre alphabétique des ateliers. 5 vols. Paris: Société Française de numismatique. Bergh, Åke. 1941. Études d’anthroponymie provençale. Les noms du Polyptyque de Wadalde (a. 814). Göteborg, Elanders Boktryckeri Aktiebolag. Bergmann, Werner. 1997. ‘Personennamen und Gruppenzugehörigkeit nach dem Zeugnis der merowingischen Königsurkunden.’ In: Geuenich et al. (eds), Nomen et gens. Zur historischen Aussagekraft frühmittelalterlicher Personennamen. Berlin: De Gruyter, 94–105.

Merovingian Coins and Their Inscriptions

317

Bloch, Marc (1956): Les caractères originaux de l’histoire rurale française, vol. 2: Supplément établi par R. Dauvergne d’après les travaux de l’auteur. Paris: Colin. Biarne, Jacques. 2002. ‘L’inf luence des monastères sur le paysage urbain en Occident (IVe–VIIe siècles).’ In: Beaujard, Brigitte (ed.), La naissance de la ville chrétienne. Mélanges en hommage à Nancy Gauthier. Tours: Maison des Sciences de l’Homme Villes et Territoires, 123–134. Boyer, Jean François. 2007. ‘À propos des triens mérovingiens: Approche du Système de collecte et de traitement de la recette fiscale en Limousin aux VIe–VIIe siècles.’ Annales du Midi 119, 141–157. Breillat, Pierre. 1935. ‘La langue des inscriptions monétaires mérovingiennes.’ In: Positions des thèses soutenues par les élèves de la promotion de 1935. Paris: Presses Univ. de France, 15–24. Buchmüller-Pfaf f, Monika. 1990. Siedlungsnamen zwischen Spätantike und frühem Mittelalter. Die -(i)acum-Namen der römischen Provinz Belgica Prima. Tübingen: Niemeyer. Buchner, Rudolf (ed.). 1964. Zehn Bücher Geschichten. Darmstadt: Wiss. Buchges. Chambon, Jean-Pierre. 1997. ‘Sur une technique de la linguistique historique: l’identification des noms de lieux, en particulier dans les texts du passé (avec des exemples concernant l’Auvergne et ses marges)’. Lalies. Actes des sessions de linguistique et de literature 17, 55–100. Chambon, Jean-Pierre. 2001. ‘L’identification des noms d’ateliers monétaires méro­ vingiens (Arvernie et entours): point de vue de linguiste.’ Revue numismatique 157, 47–405. Chambon, Jean-Pierre. 2009. ‘À quelles conditions une onomastique occitane estelle possible?’ In: Latry, Guy (ed.), La voix occitane. Actes du VIIIe congrès de l’Association Internationale d’Études Occitanes. Pessac: Presses Univ. de Bordeaux, 1019–1036. Chambon, Jean-Pierre, & Yves Greub. 2000. ‘Données nouvelles pour la linguistique galloromane: les légendes monétaires mérovingiennes.’ Bulletin de la Société de Linguistique de Paris 95, 147–182. Coşkun, Altay, & Jürgen Zeidler. 2003. ‘“Cover Names” and Nomenclature in Late Roman Gaul. The Evidence of  the Bordelaise Poet Ausonius.’ In: Keats-Rohan, Katharine (ed.), Unit for Prosopographical Research (UPR). Oxford: Linacre College. . Dahmen, Karsten. 2011. ‘Zum Münzwesen des Merowingerreiches.’ In: Greule et al. Manuscript in preparation. Deroy, Louis, & Mulon, Marianne. 1993. Dictionnaire de noms de lieux. Paris: Le Robert. Eufe, Rembert. ‘Die Personennamen auf den merowingischen Monetarmünzen als Spiegel der romanisch-germanischen Sprachsynthese im Frankenreich.’ In: Hau­

318

Rembert Eufe

brichs, Wolfgang et al. (eds), Kulturelle Integration und Personennamen im frühen und hohen Mittelalter. Manuscript in preparation. Fehr, Hubert. 2010. Germanen und Romanen im Merowingerreich. Frühgeschichtliche Archäologie zwischen Wissenschaft und Zeitgeschehen. Berlin: De Gruyter. Felder, Egon. 2003. Die Personennamen auf den merowingischen Münzen der Bibliothèque Nationale de France. München: Verlag der Bayerischen Akad. der Wiss. Fixot, Michel. 1980. ‘Une image idéale, une réalité dif ficile: les villes du VIIe au IXe siècle.’ In: Duby, Michel (ed.), Histoire de la France urbaine, vol. 1: La ville antique des origines au IXe siècle, 497–563. Gamillscheg, Ernst. 1970. Romania germanica. Sprach- und Siedlungsgeschichte der Germanen auf dem Boden des alten Römerreichs. vol. 1: Zu den ältesten Berührungen zwischen Römern und Germanen. Die Franken. 2nd edn. Berlin: De Gruyter. Gauthier, Pierre. 1999. ‘Les antroponymes dans les toponymes de la Vienne.’ In: Kremer, Dieter (ed.), Onomastik. Akten des 18. Internationalen Kongresses für Namenfor­schung, vol. IV: Personennamen und Ortsnamen. Tübingen: Niemeyer, 142–153. Gelling, Margareth. 1995. ‘Place-Names in England’. In: Eichler, Ernst (ed.), Namenforschung. Ein internationales Handbuch zur Onomastik. Vol. 3. Berlin, New York: De Gruyter, 786–792. Gendron, Stephane. 2008. L’origine des noms de lieux en France. Essai de toponymie. 2nd edn. Paris: Errance. Geuenich, Dieter, & Hawicks, Heike. 2006. ‘Probleme der Lemmatisierung frühmittel­ alterlicher Personennamen im interdisziplinären Projekt “Nomen et gens”.’ In: Brylla, Eva & Mats Wahlberg (eds), Proceedings of  the 21st International Congress of Onomastic Sciences. Vol. 3. Uppsala: Institutet för sprak och folkminnen, 81–90. Grierson, Philip, & Mark Blackburn. 1986. Medieval European coinage, vol. 1: The early Middle Ages (5th-10th centuries). Cambridge: Cambridge University Press. Greule, Albrecht. 1996. ‘Morphologie und Wortbildung der Vornamen: Germanisch.’ In: Eichler, Ernst (ed.): Namenforschung. Ein internationales Handbuch zur Onomastik, Vol. 2, Berlin: De Gruyter, 1182–1187. Greule, Albrecht, Jörg Jarnut, Bernd Kluge & Maria Selig (eds). 2011. Die mero­ wingischen Monetarmünzen. Historische, numismatische und philologische Unt­ er­s­­­­­uchungen. Grundlage des Bestandes im Münzkabinett der Staatlichen Museen zu Berlin. Manuscript in preparation. Hägermann, Dieter. 1997. ‘Die Namengebung in den Unterschichten der Karolinger­ zeit.’ In: Geuenich, Dieter et al. (eds.): Nomen et gens. Zur historischen Aussage­kraft frühmittelalterlicher Personennamen, Berlin: de Gruyter, 106–115.

Merovingian Coins and Their Inscriptions

319

Hägermann, Dieter, Wolfgang Haubrichs & Jörg Jarnut (eds). 2004. Akkulturation. Probleme einer germanisch-romanischen Kultursynthese in Spätantike und frühem Mittelalter. Berlin: De Gruyter. Halsall, Guy. 1996. ‘Towns, societies and ideas: The not-so-strange case of late Roman and early Merovingian Metz.’ In: Neill Christie & Simon T. Loseby (eds), Towns in Transition. Urban Evolution in Late Antiquity and the Early Middle Ages. Aldershot: Scolar, 235–261. Haubrichs, Wolfgang. 1985. ‘Die Urkunde Pippins des Mittleren und Plectruds für St-Vanne in Verdun (702). Toponomastische und besitzgeschichtliche Über­ legungen zum frühen Besitz der Pippiniden-Arnulfinger und zum Königsgut im Verdunois (avec résumé français).’ Francia 13, 1–46. Haubrichs, Wolfgang. 1997. ‘Galloromanische Kontinuität zwischen unterer Saar und Mosel. Problematik und Chancen einer Auswertung der toponymischen Zeugnisse aus germanistischer Perspektive.’ In: Holtus, Günter et al. (eds), Italica et Romanica. Festschrift für Max Pfister zum 65. Geburtstag. Vol. 3. Tübingen: Niemeyer, 211–237. Haubrichs, Wolfgang. 2004. ‘Romano-germanische Hybridnamen des frühen Mittel­ alters nördlich der Alpen.’ In: Hägermann, Dieter, et al. (eds), Akkulturation. Probleme einer germanisch-romanischen Kultursynthese in Spätantike und frühem Mittelalter. Berlin: De Gruyter, 179–204. Haubrichs, Wolfgang. 2008. ‘Hybridität und Integration. Vom Siegeszug und Untergang des germanischen Personennamensystems in der Romania.’ In: Dahmen, Wolfgang et al. (eds), Zur Bedeutung der Namenkunde für die Romanistik. Roma­ nistisches Kolloquium XXII. Tübingen: Narr, 87–140. IKMK. Interaktiver Katalog des Münzkabinetts der Staatlichen Museen zu Berlin. accessed 18 August 2011. Jeanton, Gabriel. 1934. Le vieux Mâcon. Histoire des rues et des quartiers de la ville. Mâcon: Renaudier. Kajanto, Iiro. 1963. Onomastic Studies in the Early Christian Inscriptions of Rome and Carthage. Helsinki: Institutum Romanum Finlandiae. Kajanto, Iiro. 1965. The Latin Cognomina. Helsinki: Soc. Scientiarum Fennica. Kajanto, Iiro. 1973. ‘The Disappearance of Classical Nomenclature in the Merovingian Period.’ In: Due, Otto Steen et al. (eds), Classica et mediaevalia Francisco Blatt septuagenario dedicata. Kopenhagen: Gyldendal, 383–395. Kluge, Bernd. 2007. Numismatik des Mittelalters, vol. 1: Handbuch und Thesaurus Nummorum Medii Aevi. Wien: Verlag der Österreichischen Akademie der Wissenschaften. Kölzer, Theo (ed.). 2001. Die Urkunden der Merowinger. Diplomata regum Francorum e stirpe Merovingica. Hannover: Hahn.

320

Rembert Eufe

Krusch, Bruno (ed). 1902. ‘Vita Eligii Episcopi Noviomagensis.’ In: Krusch, Bruno (ed.), Monumenta Germaniae historica. Scriptores rerum merovingicarum, vol. IV: Passiones Vitaeque Sanctorum aevi Merovingici. Hannover: Hahn, 634–761. Loseby, Simon T. 2006. ‘Decline and Change in the Cities of  Late Antique Gaul.’ In: Krause, Jens-Uwe & Christian Witschel (eds), Die Stadt in der Spätantike – Niedergang oder Wandel? Stuttgart: Steiner, 67–104. Patzold, Stef fen. 2008. ‘Probleme der ethnischen Identifikation. Der Beitrag der Geschichtswissenschaft zu einem interdisziplinären Personennamenlexikon am Beispiel der Namen des Frankenreichs (Hagiographie/Historiographie)’. In: Brylla, Eva & Mats Wahlberg (eds), Proceedings of the 21st International Congress of Onomastic Sciences. Uppsala 19–24 August 2002, Vol. 4, Uppsala: Språk- och Folkminnesinstitutet, 135–151. Pitz, Martina. 2007. ‘La représentation de l’espace durant le haut Moyen-Âge: éléments pour une analyse onomastique du Polyptyque de Prüm.’ In: Herbin, Jean-Charles & Michel Tamine (eds), Espace représenté, espace dénommé. Géographie, cartographie, toponymie. Valenciennes: Presses Univ. de Valenciennes, 289–304. Pitz, Martina. 2009. ‘Le type Charles-Charlon ou Eve-Evain en ancient français – vraiment un ref let du superstrat germanique?’ Nouvelle Revue d’Onomastique 51, 175–200. Pohl, Walter. 2004. ‘Vom Nutzen des Germanenbegrif fes zwischen Antike und Mittel­ alter: eine forschungsgeschichtliche Perspektive’. In: Hägermann et al., 18–34. Pol, Arend. 2001. ‘Marsal.’ In: Hoops, Johannes (ed.): Reallexikon der Germanischen Altertumskunde. Vol. 19. 2nd edn. Berlin: De Gruyter, 345–346. Pol, Arend. 2002. ‘Münzmeister. Merowingerzeit’. In: Hoops, Johannes (ed.), Real­ lexikon der Germanischen Altertumskunde. Vol. 20. 2nd edn. Berlin: De Gruyter, 353–355. Prou, Maurice. 1910. ‘Notes sur le latin des monnaies mérovingiennes.’ In: Mélanges de philologie romane et d’histoire littéraire of ferts à Maurice Wilmotte. Paris: Champion, 523–540. Prou, Maurice. 1995 [1892]. Les monnaies mérovingiennes. Catalogue des monnaies françaises de la Bibliothèque Nationale. Nîmes: Lacour. Schmidt-Wiegand, Ruth. 1973. ‘Sali. Die Malbergischen Glossen der Lex Salica und die Ausbreitung der Franken.’ In: Petri, Franz (ed.), Siedlung, Sprache und Be­völkerungsstruktur im Frankenreich, 490–530. Selig, Maria, & Rembert Eufe. In preparation ‘Les monnaies mérovingiennes et leur analyse linguistique.’ In: Biville, Frédérique (ed.), Actes du VIIème Colloque International sur le Latin Vulgaire. Solin, Heikki 1995. ‘Namen im Alten Rom.’ In: Eichler, Ernst (ed.), Namen­forschung. Ein internationales Handbuch zur Onomastik. Vol. 2. Berlin: De Gruyter, 1041–1048.

Merovingian Coins and Their Inscriptions

321

Solin, Heikki. 2003a. Die griechischen Personennamen in Rom. Ein Namenbuch. 2nd rev. edn. 3 vols. Berlin: De Gruyter. Solin, Heikki. 2003b. ‘����������������������������������������������������������� Le trasformazioni dei nomi personali tra antichità e medioevo.’ In: Rubeis, Flavia De & Walter Pohl (eds), Le scritture dai monasteri. Atti del IIo seminario internazionale di studio ‘I monasteri nell’alto medioevo.’ Rome: Institutum Romanum Finlandiae, 15–45. Solin, Heikki, & Olli Salomies. 1994. Repertorium nominum gentilium et cognominum Latinorum. 2nd edn. Hildesheim: Soc. Scientiarum. Stahl, Alan M. 1982. The Merovingian coinage of the region of Metz, Louvain-La-Neuve: Institut supérieur d’archéologie et d’histoire de l’art. Stotz, Peter. 2000. Handbuch zur lateinischen Sprache des Mittelalters, vol. 2: Bedeutungswandel und Wortbildung. München: Beck. Strothmann, Jürgen. 2011. ‘Merowingischen Monetarmünzen und die Gallia im 7. Jahrhundert. Münzorte und Monetarnamen.’ In: Greule et al. Manuscript in preparation. Verhulst, Adriaan. 1999. The Rise of  Cities in North-West Europe. Cambridge: Cambridge University Press. Ward-Perkins, Bryan. 2005. The fall of Rome and the end of civilization. Oxford: Oxford University Press. Weidemann, Margarete. 1986. Das Testament des Bischofs Berthramn von Le Mans vom 27. März 616; Untersuchungen zu Besitz und Geschichte einer fränkischen Familie im 6. und 7. Jahrhundert. Mainz: Verl. d. Röm.-German. Zentralmuseums. Wickham, Chris. 2009. The inheritance of Rome. A history of Europe from 400 to 1000. London: Allen Lane.

Remco Knooihuizen

The Use of  Historical Demography for Historical Sociolinguistics: The Case of  Dunkirk1

Abstract This chapter combines historical demographic and linguistic research to come to an understanding of  language shift in seventeenth-century Northern France. The article details the Francophone immigration to the town of  Dunkirk around and after its annexation to France in 1662, focusing in particular on the integration of migrants into the local community through intermarriage. The author then reports on a quantitative dialectological study of  the French of  Dunkirk in comparison to other French varieties. This study shows similarities in particular with varieties from the migrants’ places of origin, lending credence to the suggestion that Francophone migration played an important role in the language shift. Moreover, this research signals the importance of an integrated social and linguistic perspective in the writing of  language histories.

1

This article would not have been possible without the statistical help of Dan Dediu with the demographic data from Section 4, and the support of  Paul Heggarty and Warren Maguire in dealing with the linguistic data from Section 5. I would further like to thank April McMahon, Graeme Trousdale, Wilson McLeod, Peter Trudgill, Adam Fox, and the audience at the Bristol conference for their useful comments. This research forms part of a doctoral project supported by funding from the Arts & Humanities Research Council.

324

Remco Knooihuizen

1 Introduction The study of sociolinguistics involves correlating linguistic variation in a community with varying social characteristics of the people in whose linguistic data we found the variation. This means that sociolinguists should want to know who their informants are, and how they behave. Even in modern sociolinguistic research, this is dif ficult enough: we are interested in people’s most natural behaviour, linguistic or otherwise, but by observing and recording that behaviour, we introduce outside factors that compromise the naturalness of the situation – Labov’s Observer’s Paradox. In addition to that, there are so many social and attitudinal factors that could inf luence language use, that it seems almost impossible to obtain a complete record of  them. Historical sociolinguistics faces problems of a slightly dif ferent nature. The historical linguistic record may not be inf luenced by the presence of a sociolinguist’s recording device, but we often have no idea what the true circumstances of  the linguistic data were either. The speakers, or writers, may even be more abstract entities than the individuals from modern-day sociolinguistics – and either way, our informants are dead. Suf fice it to say that historical sociolinguistics faces a ‘bad data’ problem (Labov 1994: 11), both with regard to the linguistic production itself and with regard to the social background of that production. In order to make the best use of such bad data, we need to apply dif ferent methodologies and draw on data from other fields, such as social history, to complement our research. In this chapter, I explore the links between historical sociolinguistics and historical demography, in order to see how both fields can inform each other. Data for the chapter is drawn from a study of language shift in Early Modern France. I begin the chapter with some introductory remarks about minority languages and language shift in Europe in the Early Modern period (Section 2), before giving a brief introduction of the case study in Section 3. The historical demographic evidence is discussed in Section 4, followed by historical linguistic evidence in Section 5. I finish with some conclusions (Section 6) about both the case study and the research methods used.

The Use of  Historical Demography for Historical Sociolinguistics

325

2 Minority languages and demographic evidence Minority language groups are usually defined with reference to a larger group, most commonly the population of a nation-state. By this reasoning, minority languages have only existed since the emergence of nationstates in the Early Modern period (c. 1500–1800). This point is made, for example, in Wright (2004: 24). The Early Modern period is the period of  the rise of vernacular languages at the expense of  Latin, but also of  the rise of certain vernacular languages at the expense of others (see, e.g. Burke 2004, Baggioni 1997). Several minority languages were considerably worse of f with these Early Modern developments. A number of  them died out, such as Old Prussian, Orkney and Shetland Norn, and Cornish, while others were restricted in terms of domains of use, geographical area, and number of speakers. Examples of  this last category are Welsh, Frisian and Sorbian. Often, restrictive language policies by governments and churches, proscribing the use of minority languages, are blamed for the demise of  these languages. There are however a number of problems with such an account. Policies were often aimed predominantly at streamlining government communication, not at changing the language of  the minority language speaker’s home. For example, there is great debate about the specific variety of  language prescribed in the French Decree of  Villers-Cotterêts (1539) (see Martel 2001 for an overview), but the decree’s restriction to the judicial domain is undisputed. Moreover, there is little evidence of the policies being enforced, and there are few ideas about how they could have been enforced. Most problematically, the language policy theory cannot explain why rapid language shift occurred in the absence of such policies (Orkney and Shetland Norn, see Knooihuizen 2005), or why minority languages were maintained to this day despite such restrictive policies, as is the case with Sorbian and most minority languages in France. (See also Evans, this volume, on the propagation of of ficial languages in this period.) Perhaps a better explanation for the language shift in all these cases is the migration of majority-language speakers to the (often recently annexed) minority-language area. A comparative survey of cases across Early Modern

326

Remco Knooihuizen

Europe suggests that such migration may be a necessary and suf ficient condition for language shift; social network theory of fers a possible mechanism for the language shift. A further indication that migration may be a factor in language shift is that some of  the anti-minority language (or rather, pro-majority language) legislation is explicitly aimed at causing a language shift through the organized settlement of the minority-language area: this is the case in some areas of  Lusatia (the Friderizianische Kolonisation, see Kunze 1999: 9), the Scottish Highlands, and Ulster. In order to investigate how migration may have inf luenced language shift, I will chart the demographic change caused by migration and the interactions between immigrants and locals in Dunkirk, a town in a minoritylanguage area in Northern France.

3 Case study: Dunkirk In the mid-seventeenth century, Dunkirk was a small town of approximately 5,000 inhabitants. The town and the surrounding ‘Westhoek’ area of Flanders were politically part of the Spanish Netherlands. The population were predominantly Dutch speakers, although due to the proximity of  the Dutch-French language border, French ‘was never a truly foreign language’ in the area (Coornaert 1970: 136). During the Franco-Spanish war, Dunkirk was taken by French forces, ceded to Cromwell’s England (1659), and finally sold back to France after the Restoration (1662). The surrounding countryside also came under French control (Coornaert 1970: 148–150; Cabantous 1983: 58–59). The annexation of  Dunkirk was followed by a migration of  French speakers to the town (Lottin & Guignet 2006: 210; Coornaert 1970: 171) and a spectacular growth to approximately 14,000 in 1706 (Lambin 1980: 163–164; Cabantous 1983: 89). These included administrative and military personnel, but also builders involved in the fortification of the town, works that were repeated several times when peace treaties required the fortifications to be taken down (Cabantous 1983: 73; Lottin & Guignet 2006: 210).

The Use of  Historical Demography for Historical Sociolinguistics

327

The questions I aim to answer in this article are threefold. Where did the French-speaking immigrants come from? How much did the immigrants interact with the local population? And what were the linguistic ef fects of  these interactions?

4 Demographic developments in seventeenth-century Dunkirk 4.1 Data I investigated the interaction between the Dutch-speaking and Frenchspeaking population groups in Dunkirk by analysing marriage registers from the town, as marriage is obvious evidence of interaction. I collected data at ten-year intervals from 1647 to 1697, a period spanning both sides of  the annexation. In total, the marriage registers for these six years contain just over 1,000 entries; the uneven distribution over the years shows the growth of  the town in this period, especially in the years after the annexation. The entries in the marriage register contain the names of both spouses, of a varying number of witnesses, and of  the minister who registered the marriage. By describing the spouses as juvenis ‘youth’ or viduus/vidua ‘widow(er)’, they also contain information on people’s previous marriages. The registers for 1647 and 1657 also mention the spouses’ parish of origin, while the 1697 register contains spouses’ and witnesses’ signatures or a declaration of illiteracy. The data collected from these marriage registers can be seen as fairly representative of Dunkirk society as a whole. There are no indications that the data is biased towards higher classes, for example. As the data is taken from Catholic marriage registers, it does not show the small Protestant community in the town, estimated to be 7 per cent of  the population (Coornaert 1970: 117). The Protestants would mostly be Dutch or English speakers; the few English speakers in this Catholic data bear names like Catriona MacDonald and Malachi Donnely.

328

Remco Knooihuizen

4.2 Methods The dif ferent demographic characteristics of the individuals from the marriage registers were correlated with patterns of migration and marriage, looking for significant interactions. Because the demographic aspect of interest, language allegiance, was not directly available from the registers, I used two proxy variables: names and parish of origin. The link between a person’s name and their language of preference is not a direct one, especially in a long-standing contact situation as on the Germanic-Romance linguistic border. There is, however, a tradition of using onomastic evidence for language allegiance, also in this area (e.g. Poulain & Foulon 1981). Despite not being a fully reliable indicator, names are the best indicator available to us. Similarly, a person’s parish of origin is not connected directly to their language allegiance. Societal language preferences do not necessarily say anything about individual language preferences: we may be dealing with a Dutch speaker from a majority French-speaking town. Furthermore, the earliest reliable data on societal language preference dates from the mid-19th century (Pée & Blancquaert 1946), and in the French Flemish situation of a ‘language shift from within’ (Willemyns 1997), two centuries can mean a world of dif ference. However, the correlation between the name-based and the location-based language allegiance of a person is highly significant (p = 5.176 × 10–8), suggesting that these methods of determining language allegiance are reasonably reliable. Further, I used people’s ability to sign (in 1697 only) as a proxy for literacy or education. Again, these are very tenuous links indeed, and this type of evidence of fers only a rough indication of people having had very basic schooling. As problems with this method af fect dif ferent population groups equally (Cressy 1977: 142), however, any patterns emerging from this evidence will still represent actual dif ferences. There is a wide range of  f luency in signatures, which was collapsed into a binary variable following the guidelines from Grevet (1991: 44–45). Impressionistically, there appears to be a division between the italic hands of presumed French speakers and the secretary hands of presumed Dutch speakers, but this distinction has not been looked into in more detail.

The Use of  Historical Demography for Historical Sociolinguistics

329

4.3 Results 1: Migration patterns Migration patterns could only be investigated for those years when the entries in the marriage register mentioned the spouses’ parish of origin, i.e. 1647 and 1657. Unfortunately, this is before the annexation by France, and no clues to migration patterns immediately post-annexation are available. The pre-annexation data will supply a baseline for migration, and other types of data will have to be used as evidence for continuity or change in these patterns after 1662. (Marriage registers from the late eighteenth century did include this information again, and have been analysed in Cabantous 1983: 93–94; these however have no bearing on the situation in the mid-to-late seventeenth century.) A striking first observation is that the marriage registers contain roughly equal numbers of people from Dunkirk and from elsewhere (Table 18). This high proportion of immigrants is indicative not of the growth of the town – this was minimal before the annexation – but of the fact that mortality in Early Modern urban areas was so high that continued immigration was necessary only to maintain the population size (Sortor 2005: 165). There was no gender imbalance in the immigration in 1647, but in 1657 there were significantly more male than female immigrants due to a drop in the female immigration rate. The reasons for this drop are unknown, but may be found in the ongoing war in the area at the time. The parishes of origin could be identified for 168 immigrants. Of these, 122 (73 per cent) were from parishes where Dutch was the majority language, and 45 (27 per cent) were from majority French-speaking parishes. One immigrant was originally from Lübeck, Germany. An analysis of  the distances between Dunkirk and the parishes of origin shows that this was a short-distance migration. The median distance overall was 31.7 km; for obvious reasons – the location of  the language border some 30 km south of  Dunkirk – the median distance travelled by Dutch-speaking migrants (23.1 km) is smaller than that by French-speaking migrants (57.9 km). As most parishes supplied only one migrant to Dunkirk, there is no evidence of so-called ‘chain migration’. This may however be an ef fect of  the small, intermittent sample.

330

Remco Knooihuizen Table 18  Locals and immigrants in the Dunkirk marriage register, by gender and sample year

Men Women

Year

Locals

Immigrants

1647

23 (29.5%)

55 (70.5%)

1657

22 (38.6%)

35 (61.4%)

1647

36 (40.9%)

52 (59.1%)

1657

37 (58.7%)

26 (41.3%)

These findings on migration before the annexation of Dunkirk form a baseline from which to analyse the developments after 1662. The two years for which migration data are available show no dif ference in the areas migrants originated from, but changing patterns in the population after the annexation might give more information about dif ferent migration patterns. 4.4 Results 2: Inter-ethnic marriage patterns Data for all sampled years between 1647 and 1697 was used to look at patterns of ethnic intermarriage in Dunkirk. All individuals in the sample were assigned a language-based ethnicity on the basis of  their name. The data comprise just over 1,000 marriages; for both genders, the ratio of  Dutchto French-language background is approximately 3:2. The distribution of marriage patterns for all years together is shown in Table 19. Ignoring marriages involving a partner with a language background other than Dutch or French, the pattern shows a clear preference for in-group marriage (p  /au/), or, even more important, the wide-spread unrounding of  front vowels. Rounding before /l/ occurs only once (außerwöhlten), and so does velarization of  MHG /â/ > /ô/ (ohn, a form that has become integrated into modern German). ‘Morphological’ spelling (Voeste 2008: 15–17), an orthographic principle that has shaped the modern standard decisively, is already well represented. ärmlein, Kräntzlein, Wängelein (cf. Arm, Krantz, Wange); but behelt (cf. behalten)

Sixteenth-Century Street Songs and Language History ‘From Below’

379

‘Syllabic’ spelling (Voeste 2008: 13–15), the insertion of , normally as a marker of vowel length, in some cases occurs as in the modern standard; in others it does not. Ehr, sehr, Bůhle, etc. erst

(= NHG); (= NHG);

můht ‘Mut’ war ‘wahr’, wol ‘wohl’

(≠ NHG) (≠ NHG)

The can also be used to enhance the weight of small words (Vergewichtung, Voeste 2005: 12) with lengthened, but originally short, vowels. jhr, jhren (MHG /i/) ‘ihr, ihren’ (= NHG) ~ jr, irem ‘ihr, ihrem’ (≠ NHG)

In this respect the text only partly corresponds with modern German, but one should bear in mind that insertion or omission of remains the most unsystematic feature of  German spelling up to this very day. Initial , whose role as vowel quantifier is disputed, occurs solely in combination with inherited long vowels or diphthongs. Thon, thůt, theür, etc. (MHG /ō/, /uo/, /iu/) vs. tag, Tugendt (MHG /a/, /u/)

4.1.2 Vowels in unstressed syllables The vocalism of  the unstressed syllables does not dif fer much from the modern standard. Although the East Swabian dialect has a strong tendency towards oppressing the schwa sound [ə] in medial position (‘syncope’) as well as in final position (‘apocope’), syncope is rare. Other than in the few modern German cases (bleiben, gleich), it can be found only twice in prefixes (gwinnen, gwiß; NHG gewinnen, gewiss). Apocope is more frequent, the shortened forms actually appearing about twice as often as the full ones, but here we have to allow for metrical restraints. As a poetic licence apocope has remained well established in modern German. Ehr, farb, Lieb, Nam; hett, möcht, schenk; ohn, etc. (NHG Ehr’, Farb’, etc.) vs. Bůhle, ende, Künste, liebe; Erste, klare, newe, Schöne; jre (= NHG)

380

Helmut Graser and B. Ann Tlusty

Suf fixes with vowels other than [ə] are almost completely in accordance with the modern standard, one of  the rare exceptions being the typically Swabian diminutive ‑le. ‑heit, ‑in (= NHG); zeügknüs (NHG ‑nis) ‑lein (= NHG); 1x ‑le: Schenklen (plural, ~ Schenklein)

4.1.3 Consonants The consonants display the typical supraregional features of late sixteenth century German printing. Geminate consonants are not random specimens of what has been called Early NHG consonant accumulation, but in fact follow certain distributive rules. They can ref lect ‘old’ geminates (alle, küssen), represent ‘new’ morphological spellings (schaf ft, will, sinn, etc.), more distinctly mark the right word boundary (auf f, inn) or just enhance the weight of short words (of ft). None of  these principles are observed without exceptions, however (cf. in ~ inn, vnd ~ vnnd, etc.), which suggests that the printers welcomed variation in the number of letters as a means to fill their columns evenly. Certain consonant clusters follow similar lines, for example which is frequently used to underline the right word or morpheme boundary. Neydt, Mundt, Tugendt, endtlich, etc.; nyemandt ~ nyemand, freündtlich ~ freündlich

The cluster occurs only once (zeügknüs). The and spellings can hardly still be interpreted as expressions of  final fortization. Consonant assimilation, often seen as an indicator of proximity to spoken language, does occur, but only on a very limited scale, and in some cases undoubtedly resulting from metre and rhyme. Adeliche (MHG adellîche); nit ~ nicht; deim (NHG deinem), eim (NHG einem); frischt (NHG frischst)

The cluster /mb/ is not yet assimilated to /mm/ or /m/, although epenthesis of /b/ after /m/ does occur.

Sixteenth-Century Street Songs and Language History ‘From Below’

381

frumb (NHG fromm), vmb‑ (NHG um‑); Reichthumb (MHG rîchtuom)

Metathesis of /t/ and /s/, which occurs once in solst (instead of solts ‘should she’) is evidently a misprint. How detached the language of the printed song is from the oral substrate is underscored by the fact that ref lexes of  the so-called CentralGerman consonant weakening are limited to merely two lexemes with instead of  MHG and NHG (Dantz; roden, rodes). All other initial and intervocalic consonants are completely in accordance with the modern German standard, including and , where no trace of  the typical Upper German af fricate spellings or is found. On the whole, the spelling of  the printed song may still be well distanced from today’s standard, but its governing principles appear quite ‘modern’ and are no longer primarily linked to regional orality. 4.2 The handwritten songs With the handwritten songs things are quite dif ferent. For reasons of space, it is impossible to present the variable and sometimes very colourful spellings of the songs in the same detailed manner as was possible with reference to more regulated language of  the printed song. Only the major features can be presented here in order to demonstrate the essential dif ferences between professional printing and non-professional writing in Augsburg at the end of  the sixteenth century. We will begin with those traits of  the handwritten texts that can best be explained as ref lexes of local or regional oral speech. 4.2.1 Vowels in stressed syllables In the printed song we noticed an almost complete merger of  ‘new’ and ‘old’ diphthongs, distancing the print from local vernacular. In the weaver’s and tailor’s songs (henceforth WS, TS) the opposite holds true, the spellings for the ‘new’ diphthong beginning with , the spellings for the ‘old’ diphthong beginning with . Exceptions, for instance in the indefinite article, can be explained prosodically and grammatically.

382

Helmut Graser and B. Ann Tlusty MHG /î/ MHG /ei/

ein etc. (indefinite article)

TS, WS TS WS TS, WS

As mentioned above, Upper German dialects do not show monoph­ thongization of  falling diphthongs. Our texts show no sign of  this phen­ omenon, either. MHG /uo/, for example, is represented by in TS and by in WS. Primary (East) Swabian dialect features, especially the typically Swabian diphthongs /au/ and /ui/, which were absent in the printed song, are accounted for in the handwritten songs. glaun, bestaun (MHG gelân, bestân) luigts (MHG liuget sie)

WS TS

The same holds true for unrounding of front vowels (MHG /ü/, /üe/, /ö/, /eu/, etc.), which was carefully avoided in the printed text, but is dominant in the handwritten songs. Neyß, kreitlers, hipsch, thir, mecht, die (MHG tüeje), schenst (NHG neues, Kräutlers, hübsch …), etc. glick, girtell, knepf, schenes, leytt (NHG Glück, Gürtel, Knöpfe …), etc.; fred (MHG vröude); rhyme beütt: hochzeitt

TS WS

One of  the songs even presents examples for hypercorrect rounding, the counterpart of  this development (bütterin, faüll [MHG bitteren, veile] WS). Another typical dialect feature is the development of  MHG /â/ > /ô/ which appears frequently in the tailor’s song. lon, ston, gon, hon ~ hants (MHG lân, stân, gân, hân …), etc.

TS

4.2.2 Vowels in unstressed syllables As regards the vowels of the unstressed syllables, forms close to spoken language can be found everywhere. The diminutive suf fix appears in full, but also in shortened forms (liedlein, medlin, kreitlermedli TS), and syncope, apocope, as well as diverse clitic forms must have given the handwritten songs, when performed, a definite local character.

383

Sixteenth-Century Street Songs and Language History ‘From Below’ pricht, grinckh, verglognes (MHG berichtet, gerinc, vergelogenes) verlof fens, truckh (MHG verloufenes, drücke) etc.

WS

gf fal; gselln, gspil; zkurtz (MHG gevalle; gesellen, gespil; ze kurz) kert, dhofart; hab; duets (MHG gehœret, die hôchvart, habe; tuot sie), etc.

TS

Some of  these features were certainly employed in order to cope with the stanza structure, but for an East Swabian public even such extremely shortened forms as pricht (MHG berichtet) or verglognes (MHG vergelogenes) would have sounded very familiar. 4.2.3 Consonants Lenization of initial /t/, especially in the verb ‘tun’ (‘to do’), which frequently occurs with instead of MHG and NHG /t/, certainly indicates proximity to local speech. dantz, dref ft; don (MHG tuon), det (~ thuest, thet), etc. dust (~ thuest), dondt

TS WS

Addition of to or especially in medial and final position, originally indicating af fricative pronunciation, occurs throughout. dunckht, denckhst, schickh, etc. schreckhen, tranckh, kranckh, etc.

TS WS

The k-af fricate is a traditional Upper German dialect feature that would certainly still have been heard in spoken sixteenth-century East Swabia, although it was gradually becoming obsolete. This may have been one reason why the compositor of  the printed song shunned it altogether. Various kinds of assimilations and related variations signal the closeness of  the handwritten songs to spoken language. fünstu (MHG vündest du), wúrstu (MHG würdest du); verthrumm (MHG vertrumbe); samempst (MHG samenest); nit, nix (MHG nihtes); reckhemempt (MHG -hemede); etc.

WS

384

Helmut Graser and B. Ann Tlusty

The form reckhemempt could be regarded as hypercorrect haplography. The negation particle nix instead of nichts has remained an indicator of colloquial orality up to this very day. 4.3 Handwritten versus printed styles Based on these observations, the two handwritten songs clearly represent an older type of writing, which is much closer to spoken forms of the language than the more progressive supraregional style of  the printed song. But even the obviously non‑professional writers of  these street songs certainly had no intention of simply duplicating in writing the phonetics of  their spoken vernacular. Rather, they wrote (or at least tried to) as they had learned to write, which means that they were at least partly inf luenced by schoolbooks and anything else they saw in print. They may even have held Valentin Schönigk’s little song brochure in their own hands. So we should not be surprised to find traits of autonomous orthography in their texts as well, for example graphotactic and syllabic spellings comparable to those we saw in the printed text. In the handwritten songs, however, these features are employed in a far less systematic way. This may be demonstrated by one final set of  forms taken from both TS and WS. Although there is no trace of morphological umlaut spellings as in the printed song, spellings with doubled consonants in cases like schaf ft (infinitive schaf fen), voll ‘full’ (inf lected form voller) suggest that a morphologic principle – the same consonant structure in every single form of  the inf lection paradigm – was at work here. But this principle cannot account for all cases, because also occurs invariably in the suf fix ‑schaf ft (MHG, NHG ‑schaft) as well as after a diphthong in teif fell (MHG tiuvel). In words like schlaf fen (MHG slâfen) the originally long vowel may have been shortened (Moser 1929: 80), but in others like haf fen (MHG haven) the originally short vowel should already have been lengthened, so that the NHG principle of double consonants indicating shortness of the preceding vowel does not yet apply. Indeed, usage of double consonants is excessive and seems rather arbitrary. There is a trend towards graphotactic marking of  the right word boundary, even in unstressed syllables (auf f, lauf f; spindell,

Sixteenth-Century Street Songs and Language History ‘From Below’

385

teif fell ~ teufel). Several graphotactic clusters like (dref ft, noturf ft, etc.) or (nachtt, gemachtt, etc.) seem to be almost obligatory at least in the tailor’s song, but overall the use of double consonants appears less regulated than it was in the print. The tailor’s song even has in morpheme‑initial position (gf fal, gf felt; MHG gevalle, gevelt). For , , , and there are similar findings.

5 Conclusion Our street songs, then, clearly represent a type of written language that was essentially linked to the local vernacular, especially in the stressed vowel system. But these non-professional writers did not just pursue techniques that would result in a one-to-one rendering of segmental phonology. Even for them spelling had already to some degree become autonomous, allowing a wide range of variation. Certain trends towards a more reader-friendly presentation of the texts, for instance the more distinct marking of the right word boundary, cannot be overlooked. But on the whole, the spelling is still far from what one could call an ‘orthography’ in the strict sense of the word. The features we observed support the conclusion that the writers of these verses were well practised in their popular art, and very likely had attended some kind of school, although they did not share the skill level of professional printers. Some inf luence ‘from above’ – that is, style gleaned from more standardized, printed texts – is nevertheless evident. This holds true of the content of the songs as well as of their formal style, as the songs play with motifs that parallel those in the printed versions (pride and sin, spurned love, etc.), albeit on the level of satire. At the same time, the song texts clearly belong in the realm of oral culture, having been used as a basis for singing on a public street. They thus provide us with an exciting opportunity to explore the links between written and oral communication. Text production in the city at the end of the sixteenth century occurred in a series of  layers that operated in discourse with one another. What

386

Helmut Graser and B. Ann Tlusty

appeared in handwriting or print was also inf luenced by oral forms. It would also be interesting to compare the writing of our lay writers with texts which were produced by professional scribes, for instance with the records of the trials which led to the preservation of these street songs. Space does not permit such an experiment here, but our impression is that the writing of  the professional court scribes – especially in the vowel system – is also still strongly linked to the regional vernacular and, in this respect, is similar to what one finds in the songs. At the same time, the use of double consonants or consonant clusters, for example, seems to be much more regulated and thus closer to the printed song discussed above. There are also a number of other text genres produced below the stratum of the printers and educated scribes that lend themselves to this kind of analysis. These include pasquils, letters, fraudulent documents, and various kinds of petitions and witness statements that were written by less practiced hands, all of which display various degrees of distance from the language of professional scribes (Graser 2000, 2011; Graser & Tlusty 2009). But even this is not the lowest layer. Ordinary people caught up in the day-to-day bureaucracy of city life also produced a variety of simple notes, among them, for example, passes which were required for city residents attending social events in surrounding villages in order to prevent being arrested for evading the city’s high alcohol taxes (Graser 2011: 29–32). Some of  these writers were barely literate. Similar passes were required from Jews who wished to enter the city to conduct business (Graser 2011: 33f.). These and other unexplored text genres promise to provide further insight into usage of the written word by simple folk, even below that of the slanderous songs considered here. Examining such documents through the lens of  both social history and linguistic analysis is certain to enhance our knowledge of  language usage in the early modern city, ultimately resulting in a true history of  language ‘from below’ for this period as well.

Sixteenth-Century Street Songs and Language History ‘From Below’

387

References Bickerton, Derek. 1971. ‘Inherent Variability and variable rules.’ Foundations of Language 7, 457–492. Creasman, Allyson. 2002. ‘Policing the Word: The Control of Print and Public Expression in Early Modern Augsburg, 1520–1648.’ Unpublished doctoral dissertation, University of  Virginia. Elspaß, Stephan. 2005. Sprachgeschichte von unten. Untersuchungen zum geschriebenen Alltagsdeutsch im 19. Jahrhundert. Tübingen: Niemeyer. Graser, Helmut. 2000. ‘Augsburg und die deutsche Sprachgeschichte.’ In: Funk, Edith et al. (eds), Bausteine zur Sprachgeschichte. Referate der 13. Arbeitstagung zur Alemannischen Dialektologie. Heidelberg: Winter, 99–120. Graser, Helmut. 2011. ‘Quellen vom unteren Rand der Schriftlichkeit – die Stimme der einfachen Leute in der Stadt der Frühen Neuzeit?’ In: Elspaß, Stephan & Michaela Negele (eds), Sprachvariation und Sprachwandel in der Stadt der Frühen Neuzeit. Heidelberg: Winter, 7–48. Graser, Helmut, & B. Ann Tlusty. 2009. ‘Layers of Literacy: Non-Professional Versus Professional Writing in a Sixteenth-Century case of  Fraud.’ In: Barnes, Robin B. & Marjorie E. Plummer (eds), Ideas and Cultural Margins in Early Modern Germany: Essays in Honor of  H.C. Erik Midelfort. Farnham: Ashgate, 31–47. Hof fmann, Ludger. 1997. ‘Graphotaktik.’ In: Zifonun, Gisela et al. (eds), Grammatik der deutschen Sprache. Vol. 1. Berlin, New York: De Gruyter, 263–268. Ingram, Martin. 1988. ‘Ridings, Rough Music and Mocking Rhymes in Early Modern England.’ In: Reay, Barry (ed.), Popular Culture in Seventeenth-Century England. London: Routledge, 166–197. Ingram, Martin. 2004. ‘Charivari and Shame Punishments: Folk Justice and State Justice in Early Modern England.’ In: Roodenburg, Herman & Pieter Spierenburg (eds), Social Control in Europe. Columbus: Ohio State University, 288–308. Koch, Peter & Wulf  Oesterreicher. 1985. ‘Sprache der Nähe – Sprache der Distanz. Mündlichkeit und Schriftlichkeit im Spannungsfeld von Sprachtheorie und Sprachgebrauch.’ Romanistisches Jahrbuch 36, 15–43. Künast, Hans-Jörg. 1997. ‘Getruckt zu Augspurg’. Buchdruck und Buchhandel in Augsburg zwischen 1468 und 1555. Tübingen: Niemeyer. Nehlsen, Eberhard. 2008–2009. Berliner Liedf lugschriften. Katalog der bis 1650 erschienenen Drucke der Staatsbibliothek zu Berlin Preußischer Kulturbesitz. 3 vols. Baden-Baden: Koerner.

388

Helmut Graser and B. Ann Tlusty

Nießeler, Martin. 1984. Augsburger Schulen im Wandel der Zeit. Ein Beitrag zur Geschichte des Augsburger Schulwesens. Augsburg: Mühlberger. Reifsnyder, Kristen L. 2003. ‘Vernacular Versus Emerging Standard: An Examination of Dialect Usage in Early Modern Augsburg (1500–1650).’ Unpublished doctoral dissertation, University of  Wisconsin-Madison. Roeck, Bernd. 1989. Eine Stadt in Krieg und Frieden: Studien zur Geschichte der Reichsstadt Augsburg zwischen Kalenderstreit und Parität. Göttingen: Vandenhoeck & Ruprecht. Roper, Lyndal. 1998. ‘Tokens of Af fection: The Meanings of Love in Sixteenth-Century Germany.’ In: Eichberger, Dagmar & Charles Zika (eds), Dürer and his Culture. Cambridge: Cambridge University Press, 143–163. Stopp, Hugo. 1978. ‘Verbreitung und Zentren des Buchdrucks auf  hochdeutschem Sprachgebiet im 16. und 17. Jahrhundert. Fakten und Daten zum “organischen Werdegang der Entwicklungsgeschichte der neuhochdeutschen Schriftsprache”’. Sprachwissenschaft 3, 237–261. Stewart, William A. 1965. ‘Urban Negro Speech: Sociolinguistic Factors Af fecting English Teaching.’ In: Shuy, R.W. et al. (eds), Social Dialects and Language Learning. Illinois: Institute of  Technology, 10–18. Stuart, Kathy. 1999. Defiled Trades and Social Outcasts. Honor and Ritual Pollution in Early Modern Germany. Cambridge: University Press. Taylor, Brian. 1991. ‘Einleitung.’ In: Horst Brunner et al. (eds), Die Schulordnung und das Gemerkbuch der Augsburger Meistersinger. Tübingen: Niemeyer, 3–13. Tlusty, B. Ann. 2008. ‘Rumor, Fear, and Male Civic Duty during a Confessional Crisis.’ In: Hendrix, Scott & Susan Karant-Nunn (eds), Masculinity in the Reformation Era. Kirksville: Truman State University 140–163. Voeste, Anja. 2008. Orthographie und Innovation. Die Segmentierung des Wortes im 16. Jahrhundert. Hildesheim, Zürich, New York: Olms.

Juan Manuel Hernández-Campoy

Mood Distinction in Late Middle English: The End of  the Inf lectional Subjunctive

Abstract During the Middle English period, the synthetic expression of  tense, person and number was reduced and even led to the gradual loss of certain moods, like the subjunctive. If  the ef fects of analogy were felt in the ME verbal system, the evolution from a synthetic to an analytic system was reinforced and strengthened during the Early Modern English period. The distinctive ending for the subjunctive for both present and preterite and both singular and plural number (-e) had remained as an archaic feature in some southern dialects, while northern and Midland varieties had completely erased the inf lectional morphemes of the subjunctive in the late fifteenth century. Eventually, in the course of  the EModE period the distinctive subjunctive endings also disappeared. Given the paradigmatic existence of mood dif ferentiation still in Middle English, the aim of  the present work is the study of  the presence of subjunctive mood in a corpus (The Paston Letters) that belongs to the very end of  this period, when it was in the process of extinction. The detection and quantification of the usage of the preterite subjunctive forms of  the verb to be allow us to analyse the evolution from mood distinction to mood neutralization during the transition from Middle to Early Modern English in terms of heterogeneity and variability, as a symptom of the process of encroachment of the was-form on the expected were-contexts, and the subsequent confusion, linguistic insecurity and hypercorrection among speakers.

390

Juan Manuel Hernández-Campoy

1 Introduction During the Middle English (ME) period (c. 1100–1500), the levelling of  final unstressed vowels to /-ə/ or -Ø and the unstable quality of inf lectional /-n/ af fected the conjugation of  ME verbs: the synthetic expression of  tense, person and number was reduced and even led to the gradual loss of certain moods, like the subjunctive, eventually. Furthermore, the merging of  final unstressed vowels into /-e/ was a source of ambiguity, which was compensatorily repaired with the development of analytic resources: a) the consolidation of periphrastic or compound tenses, and b) the auxiliarization of anomalous and preterite-present verbs (see Nevalainen 2006a, b). Therefore, if the distinction between indicative and subjunctive moods was morphologically clear-cut in Old English, in Middle English this distinction was af fected by the levelling of  final unstressed syllables to /-ə/ and Ø, which led to the complete erasing of  the subjunctive at the end of  the period. As a result, it was gradually replaced by constructions involving the use of  the infinitive, or by periphrastic patterns containing a modal auxiliary (< OE preterite-present verbs) + infinitive. In this way, the typical subjunctive expression of wishes, commands, emotion, possibility, judgement, necessity, unfulfilled and unreal states or statements contrary to fact at present appear in analogical constructions where dependent clauses follow a conjunction – so that the subjunctive mood is sometimes referred to as the conjunctive mood. If the ef fects of analogy were felt in the ME verbal system, the evolution from a synthetic to an analytic system was reinforced and strengthened during the Early Modern English (EModE) period (c. 1500–1900). This means that prepositions and a relatively fixed word order were used as basic markers of the function of words in the sentence, while case endings were almost completely ousted. The distinctive ending for the subjunctive for both present and preterite and both singular and plural number (-e) had remained as an archaic feature in some southern dialects, while northern and Midland varieties had completely erased the inf lectional morphemes of the subjunctive in the late fifteenth century. Eventually, in the course of 

Mood Distinction in Late Middle English

391

the EModE period the distinctive subjunctive endings also disappeared, although this remnant of  the ME paradigm still survived in subordinate clauses. The definite demise of  the subjunctive mood reinforced the use of modal verbs in combination with an infinitive. In current English, most of the functions of the old subjunctive have been assumed by auxiliary verbs like may and should, and the subjunctive survives only in very limited situations, such as formulaic expressions (God help him, be that as it may, come what may, and suf fice it to say), that-clauses (I insist that she do the job properly), in concession and purpose clauses (Even though he be opposed to the plan, we must try to implement it; They are rewriting the proposal so that it does not contradict new zoning laws), in some conditional clauses (Whether he be opposed to the plan or not, we must seek his opinion), or in a few other constructions expressing hypothetical conditions (If  I were rich …, If  he were rich …, If  they were rich …; I wish you were here) (see Harsh 1968; Khlebnikova 1976; James 1986; Övergaard 1995; or Peters 1998).

2 Objectives The diachronic evolution of  languages consists of  the gradual accumulation of synchronic stages. Any distinction between periods is arbitrary and normally a practical convenience for pedagogical reasons, since they are not as homogeneous and concrete as they are usually shown and they do not have clear boundaries stating the end of one and the beginning of another. Likewise, the distinction between Middle English and Early Modern English itself is arbitrary, since the latter preserves some variations associated to the former due to the lineal-continuous process of development of  the English language. Given the paradigmatic existence of mood dif ferentiation still in Middle English, the aim of  the present work is the study of  the presence of subjunctive mood in a corpus (The Paston Letters) that belongs to the

392

Juan Manuel Hernández-Campoy

very end of this period, when it was in the process of extinction. The detection and quantification of  the usage of  the preterite subjunctive forms of  the verb to be allow us to analyse the evolution from mood distinction to mood neutralization during the transition from Middle to Early Modern English in terms of  heterogeneity and variability, as a symptom of  the process of encroachment of  the was form on the expected were contexts, and the subsequent confusion, linguistic insecurity and hypercorrection consequences among speakers of  the late medieval English community.

3 Methodology 3.1 Linguistic variables The study of mood distinction was carried out through the analysis of the usage of past forms of  the verb to be, whose subjunctive forms in Middle English used to be were for singular and were(n) for plural in the South and Midlands of medieval England. The experience with cases of variability in spelling in previous studies of the Paston family (see Hernández-Campoy & Conde-Silvestre 1999 and Conde-Silvestre & Hernández-Campoy 2004) made us consider also other possible orthographic forms for both was, such as waz, wos, woz, wos, wus, or wuz, that were not present in the corpus, and were, such as wer, werre, weere, weer, war or ware, which were unevenly found in some members of  the family. We also had to be cautious with mispellings such as were instead of  the conjunction or the relative adverb where, occasionally. Given that the use of were in singular contexts of the Middle English paradigm still signalled subjunctive, the detection of any mood distinction was carried out through the quantitative and qualitative analysis of  the process of  levelling to was/were both in indicative and subjunctive contexts, as part of a larger project on medieval and contemporary was/ were-levelling in East Anglian English. Each token of past be form from the Paston Letters collection was analysed and coded for subject type:

393

Mood Distinction in Late Middle English

1st, 2nd, 3rd person singular and plural, 3rd person singular and plural Noun Phrase, and 3rd person singular and plural Existential, and Polarity (see Table 21). Two distinctions were specified: a) levelling to was/were in contexts of  Indicative Mood, and b) levelling to was/were in contexts of  Subjunctive Mood. Table 21  Subject type SINGULAR SUBJECT First I was I were

PLURAL SUBJECT First We was We were Second You was You were

Third NP The farm was The farm were

Third NP The farms was The farms were

Third Pronoun It was It were

Third Pronoun They was They were

Third Existential There was a farm There were a farm

Third Existential There was farms There were farms

Diachronic examples obtained from the corpus are: • 1st person sg was: ‘for at þat tyme I was in hope þat the Bysshop of f  Wynchester sholde have payd it, …’ ( John Paston II, to Margaret Paston or John Paston III 1472, 11, 22). • 1st person sg were: ‘… y be sette in good and sad rewle, for I were loth to lese hym; for I trust to haue …’ (Margaret Paston, draft letter probably addressed to James Gloys 1473, 01 18?). • 2nd person sg was: none • 2nd person sg were: ‘I wold, sauyng yowre dy[s]plesure, that ye were delyueryd of hym, …’ ( John Paston II, letter sent to John Paston I, 1441?).

394

Juan Manuel Hernández-Campoy

• 3rd person sg pronoun was: ‘by whiche wordes commvnely it was vndyrstandyn þat þe forgeers and makers of þe seyd billes jmagyned to þe seyd William, …’ (William Paston I, Memorandum to Arbitrators 1426–1427). • 3rd person sg pronoun were: ‘It were well doo þat ye wer a lytell sewrere of f yowr pardon than ye be’ ( John Paston II, letter sent to John Paston III 1471, 09, 15). • 3rd person sg Noun Phrase was: ‘Gef freie axid Palmere why the rentte was notte axid in myn husbonddis tyme, …’ (Agnes Paston, to Edmond Paston I 1445, 02, 04). • 3rd person sg Noun Phrase were: ‘and at large fro Lancasterother vn-tylle the mater were duly examynd’ ( John Paston I, draft letter addressed to Richard Southwell 1452, 07, 20?). • 3rd person sg Existential was: ‘he seyth feythefully ther was no swyche thyng desyiryd of hym, …’ (Margaret Paston, letter sent to John Paston I 1461, 11, 02). • 3rd person pl Existential was: none • 3rd person sg Existential were: ‘… that there were such a won in that countray that oght you gode wyll, …’ (Margaret Paston, letter sent to John Paston I 1465, 05, 13). • 3rd person pl Existential were: ‘and also þer were more specyal termys in hys þan wern in oderys’ (Margaret Paston, letter sent to John Paston I 1464, 05, 06). • 1st person pl pronoun was: none. • 1st person pl pronoun were: ‘But or we had thus moche of  his confessyon we were resonably well jn credens wyth hym’ (William Paston II, to John Paston I 1460, 01, 28). • 2nd person pl pronoun was: none. • 2nd person pl pronoun were: ‘… as 3e were servyd at Gressam wyth-in ryt3 schort tym’ (Margaret Paston, letter sent to John Paston I 1449, 04, 02). • 3rd person pl pronoun was: • 3rd person pl pronoun were: ‘… and that they were well assuryd men’ ( John Paston II, letter sent to John Paston III 1468, 11, 09).

Mood Distinction in Late Middle English

395

In order to facilitate the detection and quantification of the dif ferent possibilities for each variable in each possible morpho-syntactic combination, we used the Concordance Package MonoConc Pro (ver. 2.0, Build 228, by Michael Barlow). The statistical analysis of the results obtained was carried out using VARBRUL. Typically expressing condition, hypothesis, contingency, possibility, wishes, commands, emotion, judgment, necessity, and statements contrary to fact at present in subordinate clauses, cases of subjunctive with past be forms were found with sentence connectors such as as son(e) as, till/tyll, in cas(e), if/yf, before that, as ever, in asmuch as, so that, for (if ), after that, as, so, that soone aftyr that they, whether, I pray + clause, I wish/wold + past, etc. 3.2 Informants: The Paston family The Pastons are the best documented gentry family of late medieval England. They take their name from a Norfolk village about twenty miles northeast of  Norwich. Written evidence on this family mentions Clement Paston as founder of  the dynasty. The initial social position of  this Norfolk family was not originally as high as it became later in the century. The family fortunes improved with William Paston I (1378–1444). He was the only son of  Clement, who, after school, was trained as a lawyer in the Inns of  Court in London and gained a good local reputation: he acted as counsel for the city of  Norwich from 1412, and in 1415 became steward to the duke of Norfolk, beginning a successful career at the royal courts. There he married Agnes Berry in 1420, and became Justice of  the Common Bench in 1429. The members of  this family evolved from the middle-high position of  the professional lawyer William Paston I to the higher one attained by John Paston II, who had become a member of  the court nobility when knighted in the 1460s. This could be taken as a clue to their upward social mobility.1

1

Further information on the Paston family can be found in Bennett (1990), Richmond (1990, 1996), Barber (1993), Gies & Gies (1998), or Bergs (2005).

396

Juan Manuel Hernández-Campoy

3.3 Period: The Wars of  the Roses The Pastons lived in the England of  the Wars of  the Roses (1455–1487). They found themselves embroiled in dif ferent struggles, such as the royalist allegiance during the dynastic civil wars – which severely penalized the family descendants – and the Siege of  Caister – triggered by the disputes with the Duke of  Norfolk and the Duke of  Suf folk about the ownership, at that time held by John Paston III. 3.4 Instrument: The Paston Letters For the present study, the body of letters used was taken from the Middle English Collection of the Internet electronic edition of the Paston Letters (First Part) from the Virginia University Electronic Text Center (http:// lib.virginia.edu/digital/collections/). The Paston Letters is the name given to a collection of 422 authored documents (letters and notes) written by fifteen members belonging to dif ferent generations of this Norfolk family mainly during the fifteenth century (from 1425 to 1496), with roughly 246,353 words. The historical and philological interest of these documents is outstanding, not only because they of fer data on the political and domestic history of fifteenth-century England, but also because they were composed at a crucial period in the development of the English language. Additionally, they all belong to the same dialect area under research (East Anglia). Previous studies such as Hernández-Campoy & Conde-Silvestre (1999) and Conde-Silvestre & Hernández-Campoy (2002, 2004) have shown that the dif ferent members of the Paston family had an uneven adoption of  the incipient Standard English written norm depending on their personal circumstances: the individual rate of adoption of this written standard was associated with their social and geographical mobility, as some of  them quickly rose in the social scale and/or had travelling experience.

397

Mood Distinction in Late Middle English

4 Results and Analysis The results obtained for the use of was and were in the Paston letters collection per mood context are shown in Tables 22 and 23. Table 22  Levelling to was/were in contexts of  Indicative Mood Singular subject First I was I were

Number of  tokens 106/110 4/110

% 96% 4%

Plural subject

Number of  tokens

%

First We was We were

0/17 17/17

0% 100%

Second You was You were

0/28 28/28

0% 100%

Third NP The farm was The farm were

531/545 14/545

97% 3%

Third NP The farms was The farms were

4/145 141/145

3% 97%

Third pronoun It was It were

278/309 31/309

90% 10%

Third pronoun They was They were

0/51 51/51

0% 100%

Third existential There was a farm There were a farm

47/50 3/50

94% 6%

Third existential There was farms 0/15 There were farms 15/15

0% 100%

Total was:

966/1270

% was:

76%

Total were:

304/1270

% were:

24%

Levelled was:

4/256

% levelled was:

1.56%

Levelled were: 52/1014

% levelled were: 5%

398

Juan Manuel Hernández-Campoy Table 23  Levelling to was/were in contexts of  Subjunctive Mood

Singular subject First I was I were

Number % of  tokens 9/33 24/33

27% 73%

Plural subject

Number % of  tokens

First We was We were

0/4 4/4

0% 100%

Second You was You were

0/36 36/36

0% 100%

Third NP The farm was The farm were

20/111 91/111

18% 82%

Third NP The farms was The farms were

0/61 61/61

0% 100%

Third pronoun It was It were

24/166 142/166

14% 86%

Third pronoun They was They were

0/23 23/23

0% 100%

Third existential There was a farm There were a farm

0/8 8/8

0% 100%

Third existential There was farms 0/2 There were farms 2/2

0% 100%

Total was:

53/444

% was:

12%

Total were:

391/444

% were:

88%

Levelled was:

0/126

% levelled was:

0%

Levelled were: 265/318

% levelled were: 83%

First, we examined the indicative-subjunctive distinction in the examples of  the corpus. We also had to consider those contexts where the application of  the subjunctive would make a dif ference to the realization of the verb. The VARBRUL analysis revealed a ‘knockout’ where there is no variation for verbal subject, specifically for most plural forms (you, we, they and plural existentials), which categorically were realized as were, and thus allowed us to exclude them. As for plurals it would not make any dif ference, we looked only at 1st singular, 3rd person singular NP, 3rd singular PRO and singular existentials (see Tables 22–23). The dif ference between indicative and subjunctive is highly significant (at p < 0.001) with was being used 95 per cent (966/1014) of the time in indicatives and only 17 per cent (53/318) in subjunctives (see Figure 20). However, the use of were in the same contexts

Mood Distinction in Late Middle English

399

is radically the opposite: 5 per cent (52/1014) with indicatives and 83 per cent (265/318) with subjunctives.

Figure 20  Percentages of usage of was and were in indicative and subjunctive in singular contexts

The fact that was is used in subjunctives (27 per cent) and were in indicatives (4 per cent) in singular contexts is revealing in itself since, for example, the form were was still most often associated with subjunctive contexts.

Figure 21  Percentages of usage of was per singular subject type and mood

400

Juan Manuel Hernández-Campoy

Individually, from the point of view of the dif ferent subject type contexts, was was used in indicative mood singular in more than 90 per cent of instances: 1st p. sg 3rd p. p. PRO (he/she/it) 3rd p. NP sg. 3rd p. Existential singular

96% 90% 97% 94%

which is similarly significant at p < .001. Contrarily, with only the subjunctive, there are also some dif ferences in the presence of was in singular contexts (see Figure 21): its presence goes down dramatically to no more than 27 per cent, though variably, which also means that the form were was still most often associated with subjunctive contexts: 1st p. sg 3rd p. p. PRO (he/she/it) 3rd p. NP sg. 3rd p. Existential singular

27% 14% 18% 0%

but this fact is not significant (p = 0.117), very probably because of  the small number of  tokens. In plural contexts of subject type, according to Table 23 (see also Figure 22), there is some levelling to was (1.56 per cent) specifically in the case of 3rd person plural NP (3 per cent), which might mean that the Northern Subject Rule2 is weakly in place here in these data, though with

2

There are some contexts in which these levelling contructions are consistently – though geographically – constrained: in the cases of Existential sentences, full Noun Phrases (Northern Subject Rule), and third person plural Pronouns (Southern Subject Rule). The first constraint refers to levelling to was in existentials with plural nouns, as in There was loads of things to do. The second constraint has to do with the Northern Subject Rule and af fects the levelling to was after full Noun Phrases (The cats was purring) or non-adjacent pronouns – when the subject and verb are separated by a clause –, rather than after adjacent pronouns (They were purring). Nevertheless, the Northern Subject Rule is overturned in East Anglia: levelling to was occurs after third person plural Pronouns (They was purring) but not so much after third plural

Mood Distinction in Late Middle English

401

very few tokens (4/145). This is also contrained by mood, as it only occurs in indicative (Table 22) but not in subjunctive cases (Table 23). However, there are no instances of the Southern Subject Rule here yet, which might mean that this is a change that happened at later stages of  the history of  East Anglian English.

Figure 22  Percentages of usage of was per plural subject type and mood

The four cases where the NSR appears are: • ‘… and by þe seales of þe same vj arbitrores and þe seyd Walter and William was af fermed and ensealed and left in þe handes of þe seyd Ser Bryan, saueliche to be kept in pleyn …’ (William Paston I: Memorandum to Arbitrators, 1426–1427). • ‘… by whose reule and commaundement þe seyd arbitrement and award was mad in þe fourme aforn seyd’ (William Paston I: Memorandum to Arbitrators, 1426–1427).

noun phrases (The cats was purring). Given the robustness of this pattern across East Anglia and in both past be and present tense verb forms, it is also known as East Anglian Subject Rule, EASR.

402

Juan Manuel Hernández-Campoy

• ‘I sende yow the namis of the men that kaste down the pittis that was Gynnis Close wretyn in a bille closid in this lettre’ (Agnes Paston: to Edmond Paston I, 1415, 02, 04). • ‘Item, that syche othyr wryghtyngys and stuf f as was in my kasket be in your kepyng, and þat no body look my wryghtyngys’ ( John Paston III: to Margaret Paston, 1471, 04, 30). But what is the Northern Subject Rule doing in the land of  the Southern Subject Rule? Language and dialect contact appear as the most plausible explanation for this phenomenon. The Northern Subject Rule is historically associated with northern English dialects and the dialects of  Scotland. Nevertheless, in our analysis there is some evidence that, over time, it came to operate over a much wider geographical area. Evidence for the presence of  the Northern Subject Rule in historical varieties of  London English, particularly with regard to past be, is sparse, although the historical inf lux of migrants from the north of  England to London lends some weight to the possible spread of the Northern Subject Rule to the capital in the late fifteenth and sixteenth centuries (see Schendl 1996; Wright 2002: 244; Nevalainen 2006b: 356). According to Schendl (1996), the speech of migrating northern traders was an important focus for the transmission of  this linguistic phenomenon to Londoners. In fact, early evidence of constraints associated with the Northern Subject Rule was found by Bailey, Maynor & Cukor-Avila (1989: 20) in the Cely Letters written between 1472 and 1488 by a family of wool merchants resident in London (see also Conde-Silvestre & Hernández-Campoy 2004). Although Tagliamonte (2002, 2006) claims that the association of was with plural NPs appears to have spread to the south in the sixteenth and seventeenth centuries, as we have seen, the Northern Subject Rule is already present in our results, from documents dating back to the end of  the fifteenth century in a very embryonic and incipient form.

Mood Distinction in Late Middle English

403

5 Conclusion Our informants and their family epistolary correspondence from 1425 to 1496 are immersed in the period of transition from Middle to Early Modern English and the transition from mood distinction to mood neutralization. The shift from a synthetic to an analytic system was gradually taking place in the late medieval local communities. The detection and quantification of  the usage of  the preterite subjunctive forms of  the verb to be have allowed us to analyse the evolution from mood distinction to mood neutralization during the transition from Middle to Early Modern English in terms of heterogeneity and variability. The patterns of  heterogeneity and variability found in the use of  the past Be forms was and were in the Paston family during the seventy-one years at the end of  the Middle English period may lead us to conclude that it is a symptom of  the process of encroachment of  the was-form onto the expected were-contexts and the subsequent confusion, linguistic insecurity and hypercorrection consequences amongst speakers. We know that in language variation and change situations, while all change involves variability, not all instances of variability involve change. That is, linguistically speaking, change goes through a number of stages in the transition from a categorical use of one variant to its categorical replacement by another (see also Bailey 1973). X → X/Y → Y/X → Y

During this process of replacement of an older form with a new one within this transitional period of f luctuation, such as the blurring of the subjunctive mood in English, intermediate stages of instability and variability are normal part of  the shift: ‘Speakers do not suddenly adopt a new form as a categorical replacement for an older form, whether the form involves a gradual, imperceptible change in the phonetic value of a vowel within a continuum of phonetic space or an abrupt, readily perceptible change involving the metathesis of consonants or the linear realignment of constituents within a syntactic phrase. Instead, there is a period of variation and

404

Juan Manuel Hernández-Campoy

coexistence between new and old forms in the process of change’ (Wolfram & Schilling-Estes 2003: 715). And this is the case with these heterogeneous results in the use of past be forms. In fact, as shown in Hernández-Campoy (2008), the standard ideology was present in speakers such as William Paston II in his description of  the English grammar and the omission of  Mood distinctions, despite using subjunctive 16 per cent of the time in his writing as a covert prestige example. The crucial complementary combination of  Language and History, Linguistics and Historiography, based on paradigmatic complementarity, would provide us with a greater refinement and precision in analysis, as well as signifying a step forward in the development of our understanding of language change and dif fusion. It will in turn contribute, therefore, to the location of  language in and through time, social and geographical space, and, finally, constitute points of convergence in the improvement of linguistic theory, and, ultimately, of our understanding of its nature and structure as a human faculty.

References Bailey, Charles-James. 1973. Variation and Linguistic Theory. Arlington: CAL. Bailey, Guy, Natalie Maynor & Patricia Cukor-Avila. 1989. ‘Variation in subject-verb concord in Early Modern English.’ Language variation and change 1, 285–300. Barber, Richard. (ed.). 1993. The Pastons. A Family in the Wars of  the Roses. Woodbridge: The Boydell Press. Bennet, Henry S. 1990/1995. The Pastons and their England: Studies in an Age of  Transition. Cambridge: Cambridge University Press. Bergs, Alexander. 2005. Social Networks and Historical Sociolinguistics: Studies in Morphosyntactic Variation in the Paston Letters (1421–1503). Berlin: Mouton de Gruyter. Conde-Silvestre, Juan Camilo, & Juan Manuel Hernández-Campoy. 2002. ‘Modern Geolinguistic Tenets and the Difusión of Linguistic Innovations in Late Middle English’. Studia Anglica Posnaniensia 38, 147–177.

Mood Distinction in Late Middle English

405

Conde-Silvestre, Juan Camilo, & Juan Manuel Hernández-Campoy. 2004. ‘A Socio­ linguistic Approach to the Difusión of  Chancery Writen Practices in Late Fifteenth Century Private Correspondence.’ Neuphilologische Mitteilungen 105 (2), 133–152. Gies, Frances, & Joseph Gies. 1998. A Medieval Family: The Pastons of Fifteenth-Century England. New York: HarperCollins Publications. Harsh, W. 1968. The subjunctive in English. Alabama: University of  Alabama Press. Hernández-Campoy, Juan Manuel. 2008. ‘Overt and Covert Prestige in Late Middle English: A Case Study in East Anglia.’ Folia Linguistica Historica 29, 1–26. Hernández-Campoy, Juan Manuel & Juan Camilo Conde-Silvestre. 1999. ‘The Social Dif fusion of Linguistic Innovations in 15th century England: Chancery Spellings in Private Correspondence.’ Cuadernos de Filología Inglesa 8, 251–274. James, F. 1986. Semantics of  the English subjunctive. Vancouver: University of  British Columbia Press. Khlebnikova, Irina B. 1976. The conjunctive mood in English. The Hague: Mouton de Gruyter. Nevalainen, Terttu. 2006a. An Introduction to Early Modern English. Edinburgh: Edinburgh University Press/ New York: Oxford University Press. Nevalainen, Terttu. 2006b. ‘Synchronic and diachronic variation.’ In: K. Brown (ed.), The Encyclopedia of  Language and Linguistics. Vol. 12. 2nd edn. Oxford: Elsevier, 356–362. Övergaard, G. 1995. The mandative subjunctive in American and British English in the 20th century. Uppsala: University of  Uppsala. Peters, Pamela H. 1998. ‘The survival of the subjunctive: evidence of its use in Australia and elsewhere.’ English World-Wide 19 (2), 87–102. Richmond, Colin. 1990/2002. The Paston Family in the Fifteenth Century (vol. 1): The First Phase. Cambridge: Cambridge University Press. Richmond, Colin. 1996. The Paston Family in the Fifteenth Century (vol. 2): Fastolf ’s Will. Cambridge: Cambridge University Press. Schendl, H. 1996. ‘The 3rd plural present indicative in Early Modern English – variation and linguistic contact.’ In: D. Britton (ed.), English Historical Linguistics 1994. Amsterdam: John Benjamins, 143–160. Tagliamonte, Sali. 2002. ‘Comparative sociolinguistics.’ In: J.K. Chambers, P. Trudgill & N. Schilling-Estes (eds), The Handbook of  Language Variation and Change. Oxford: Blackwell, 729–763. Tagliamonte, Sali. 2006. ‘Historical Change in Synchronic Perspective: The Legacy of  British Dialects.’ In: A. van Kemenade & B. Los (eds), The Handbook of  the History of  English. Oxford: Blackwell, 477–506.

406

Juan Manuel Hernández-Campoy

Wolfram, Walt, & Natalie Schilling-Estes. 2003. ‘Dialectology and Linguistic Dif fusion.’ In: Brian D. Joseph & Richard D. Janda (eds), The Handbook of  Historical Linguistics. Oxford: Blackwell, 713–735. Wright, Laura. (ed.). 2000. The Development of Standard English, 1300–1800. Theories, Descriptions, Conf licts. Cambridge: Cambridge University Press.

Part 5

Language and Ideology

Lisa Carroll-Davis

Identifying the Enemy: Using a CDA and Corpus Approach to Analyse Sandinista Strategies of  Naming

Abstract The Sandinista government, upon gaining power in Nicaragua, faced the task of constructing a new national identity over and against that of  the former regime. During the campaign’s first four years, the increasing intensity of the counter-revolutionary war dictated that much of the identity formation appeared as an opposition to the enemy. I show, through a combination of corpus analysis and critical discourse analysis of an original 125,000-word corpus of  the Sandinista print propaganda, that the Contra enemy was situated in a delegitimizing frame. This frame allowed the Sandinistas to establish themselves and their revolution as the only legitimate heir to power in the country. What can be seen then in this situation is the use of  language as a political means to garner hegemonic support through the construction of illegitimacy, rather than through direct imposition and presentation of its own power and dominance.

1 Introduction John B. Thompson (1984: 2) states that ‘language is not simply a structure which can be employed for communication or entertainment, but a sociohistorical phenomenon which is embroiled in human conf lict’. For this reason, discourse is a necessary part of any research on political actions and ideology. Employing language as a substitute for physical aggression has created an area of  language use which is replete with ideological and political posturing. One can create a sense of  legitimacy or illegitimacy, authority or submission, through carefully crafted speech. Within a group,

410

Lisa Carroll-Davis

the modes of communicating these perceptions on particular issues become part of  the group-speak through exposure and use over time. Patterns are established which convey power or control through inexplicit and at times unconscious means. Naming is one of the aspects where patterning is evident. Biases are encoded in the label assigned to a person or group through the choices about ‘what to emphasize and what to overlook’ (Spender 1980: 164) through recourse to the dominant group’s available, pre-existing patterns. With these ideas in mind, the purpose of  this article is to examine how social conf lict is produced and managed within a particular political discourse and answer the following questions: • What means might be used to construct the conception of an opposition force within a dominant discourse? • How can naming and labelling such out-groups promote the dominant group’s ideology? These questions are answered by looking at the ways the Sandinistas in Nicaragua framed their opposition from 1980 to 1983 through naming strategies in their propaganda publications. As was stated above, assigning a name is a process which typically relies on patterns. These patterns provide a means of categorization for the receivers of a particular discourse, which has particular importance when considering language use in the political realm. Paul Chilton recasts the Sapir-Whorf  hypothesis to ref lect the political impact of  language saying it is ‘the use of a […] discourse, especially in a repeatable, institutionalized form, that governs the way people think’ (2004: 26). This group patterning function allows the field of discourse studies to probe how discursive structures and other linguistic elements are combined in order to constrain the outcome to the one desired by the political agents involved in their production. The interplay of discourses and social construction situates ‘discourse analysis as a powerful method for studying social phenomena’ (Phillips & Hardy 2002: 5). Such studies have applicability, particularly in the modern context, as a way of interpreting, understanding and even using political language more ef fectively. For Norman Fairclough, discourses are both socially constituted and socially constituting, allowing elements to become common-sense, the ‘taken for granted’ knowledge of social practice (Fairclough 1995: 76).

Identifying the Enemy

411

By crafting the common sense of a nation, those in power create hegemony for their ideology. Thus for the present study, discourse theory, particularly that of critical discourse analysis (CDA), is employed as it is the best positioned to examine the social or political consequences of the use of particular discursive elements. Following a discussion of  the relevance of  the theory of naming in political discourse, this article will present a combined CDA/corpus analysis of the Sandinista naming strategies used in its political propaganda. I argue that this analysis shows a strategy of delegitimization through the application of a negative label and the denial of any one name for the opposition.

2 Theoretical considerations: Naming as a function of political discourse In political speech, the construction of an out-group contrasting with the in-group is necessary, for without a comparison, there is no basis for dif ferentiation among groups. Delimiting the other creates boundaries of what constitutes the in-group, creating the us while the them is defined (Smith 1998: 29). Us/them distinctions are particularly salient in discussions on identity. Separating groups entails highlighting the dif ference between them as well as identifying the similarities within a group. Dan Smith (1998: 18) views the creation of collective identity as a ‘two-way process’, where recognizing in-group likeness creates and emphasizes out-group dif ference. Framing the other is an attempt to present the out-group in a specific way to a target audience so that the perception of  that group is retained and recalled when the audience members think about it (Mintz & Redd 2003: 193). The labelling of a group relies on pre-existing categories in a group’s ideology which supply the emotional or ideological meaning intended by the namers. As is indicated by Dale Spender (1980: 164), ‘new names, then, have their origins in the perspectives of  those doing the naming rather than in the object or event that is being named, and that perspective is the product of  the prefigured patterns of  language and thought’.

412

Lisa Carroll-Davis

Assigning a name to a group allows for its objectification and thus directs the discourse surrounding it. Attempts to legitimize or delegitimize a group during a war or conf lict often revolve around the labels used to name the enemy or allies. According to Michael Bhatia (2005: 9), merely having the power to assign a name to a group, which is appropriated into the discourse on that unit as a whole, denotes authority on the part of the namer. Similarly, Spender (1980: 165) posits that the power to name entails the power to ‘inf luence reality’. This echoes Chilton’s rewording of  the Sapir-Whorf  hypothesis previously mentioned. The ability to determine the terms which people use to conceive of and categorize the world gives one the power to constitute what that world looks like. In the Nicaraguan case, the naming of the opposition added to the Sandinistas’ authority by discrediting the remaining Somoza supporters, and establishing the of ficial discourse surrounding those supporters. It also allowed them to establish themselves as the authority and legitimate power in contrast to the outgroup of  former power holders. As Jacques van Doorn notes (1975: 103), giving names with negative connotations, such as ‘terrorist’ or ‘subversive elements’, is a means of ‘denying the legality of their opponents and emphasizing the need to maintain law and order’. The Sandinistas of  the 1980s were the reincarnation of a Nicaraguan fight for freedom begun sixty years earlier.1 Augusto C. Sandino (1895– 1934) gathered a small group of men to fight for the expulsion of  the US Marines from Nicaragua in the 1920s and 1930s. Sandino was assassinated by General Anastasio Somoza García (1896–1956), commander of  the Nicaraguan National Guard, in 1934. Somoza, with the backing of  the USA government, used the presidency as a platform to create a dictatorship, which became a dynasty as both of his sons succeeded him in power. The Sandinistas of  the FSLN (Sandinista Front for National Liberation) took up Sandino’s fight, three decades later, against both US interference and the Somozas’ authoritarian rule. Upon defeating Anastasio SomozaDebayle and coming to power in Nicaragua in 1979, the Sandinista party

1

For an insightful discussion of the Sandinistas in the context Nicaraguan historical context, see Hazel Smith’s (1993) Nicaragua: Self-determination and Survival.

Identifying the Enemy

413

began printing numerous propaganda pamphlets and weekly newspapers aimed at educating the masses in Sandinista thought. State speeches also contained ideological elements and their transcripts were frequently printed for distribution to the people. The existence of such material allows for the study of  the development, creation and re-creation of a Sandinista national discourse. In Nicaragua during Sandino’s era and also that of the FSLN, naming from the USA acted as a means of discrediting the fight for national liberation by both groups of  Sandinistas. Sandino and his original peasant army in the 1920s were titled ‘bandits’ and ‘blanket thieves’ whereas the Sandinistas of the FSLN were 60 years later branded ‘communists’ by the USA and Somoza (Schroeder 2005: 69, 70). In one example, FSLN leader Daniel Ortega (1981: 5) acknowledges this labelling of  Sandino and also of a regional Nicaraguan label assigned to the current Sandinistas saying, ‘Somocism called Sandino a bandit. But the people and the sandinistas were not of fended by that term. And Sandino was the divine bandit! We say this because history is repeating itself and those same people who called Sandino a bandit, now are calling us turbas [a mob of agitators and dissidents]. And here are those turbas, those divine turbas of  the Revolution!’2 The term turba carries a derogatory meaning of  ‘an of fensive mob’ in Nicaraguan Spanish, but was seized upon by the FSLN of ficials as a positive label for their devoted, yet militant, supporters in the 1980s (Arellano Oviedo 2007: 353). As seen in this passage, the Sandinistas both past and present embrace their titles as rebels. The following paragraph in the document explains that the people are not embarrassed over the name-calling by the opposition because the people of  Nicaragua ‘have a revolutionary consciousness’ (Ortega 1981: 5). This sentiment illustrates the perspective advanced by Michael Schroeder (2005: 71), that in order to counter a label assigned by a dominant group, in this case the USA and the opposition, one

2

‘… el somocismo llamó bandolero a Sandino. Pero el pueblo y los sandinistas no se afrentaron de ese término ¡y Sandino fue el bandolero divino! Decimos esto porque la historia se repite y los mismos que llamaron a Sandino bandolero, ahora nos llaman turbas. Y aquí están esas turbas, las turbas divinas de la Revolución’.

414

Lisa Carroll-Davis

must challenge the narrative created around the name through a rigorous appeal to history. The FSLN repeatedly called on the historical foundations of the key characteristics embodied by the Sandinistas and the pueblo. With this background, the revolutionary consciousness would steel the people against imperialist, discursive attacks. Perhaps having learned from Somoza and the USA, the FSLN directorate also began to use naming as a means of delegitimizing those whom the USA called ‘freedom fighters’, the Nicaraguan opposition to the revolution. Using a familiar pattern for labelling the opposition illustrates Spender’s claim stated above that naming draws on past meanings and patterns to provide the name with validity and acceptability.

3 Method and design: A corpus-assisted critical discourse analysis The research method used for this paper is a combination of elements of  CDA and corpus linguistics. CDA, as described by Ruth Wodak, is based on context and looks at ‘the immediate language or text, the intertextual […] relationship between […] discourses, the extralinguistic social/sociological variables […] of a specific “context of situation”’, and ‘the broader sociopolitical and historical contexts’ (Wodak 2001a: 67). Wodak’s discoursehistorical branch of  CDA attempts to create ‘conceptual frameworks for political discourse’ (Meyer 2001: 22) and aims at incorporating empirical research within the theoretical bounds of the discipline (Kendall 2007: 14). The discourse historical approach takes into consideration the historical and social context of a certain political discourse. The contextualization of discourse is imperative in this approach as one of  the assumed precepts is that discourse is socially constituted and constitutive, and as such is embedded in societal structures (Wodak & Meyer 2009: 6). This suggests that to ‘[…] understand discourses and their ef fects, we must also understand the context in which they arise’ (Phillips & Hardy 2002: 4). Thus, the

Identifying the Enemy

415

discourse historical approach requires the researcher to not only consider the social situation of the discourse under investigation, but also the historical context in which it emerged. What this approach also allows for and implies is a constant moving from the text to the interpretation and back again, allowing the text to inf luence the interpretations and using those interpretations to discover other aspects within the text. Drawing on work by Paul Baker (2006), the aspects that corpus linguistics analyses, which I combine with CDA in this study, include collocations and keyness. Confronting a major criticism of  CDA, that it does not present any reproducible results, applying a corpus approach to discourse analysis provides the research with a quantitative element. By supplying a measure of verification, the incorporation of a reproducible, quantitative aspect serves to decrease concerns about researcher bias in CDA analyses, acting as a stop-gap against ‘over- or under- interpretation’ (Baker et al. 2008: 298). An additional benefit of using a corpus approach is the ability to expand the amount of data that can be analysed by using computer programs. Gerlinda Mautner stresses that the increased data, coupled with the computation of  frequencies and statistical tests, not only provides a measure of empiricism, but also deepens the analysis that can be performed on the material (Mautner 2009: 123). Through collocation and keyness analysis using the program WordSmith developed by Mike Smith (Smith online), a strategy of delegitimization through the absence of naming emerges in the corpus of  Sandinista propaganda. The thirteen documents which I chose for the creation of a corpus for this study comprise various themes and topics within the FSLN’s discourse. The criteria for the selection of these publications were their publication by the Sandinista Department of Propaganda and their length and presentation as pamphlets. All of the pamphlets were published by the Sandinista government between 1980 and 1983. Some of the documents were created originally for text publication and distribution as a pamphlet or article, whilst others were originally speeches which were transcribed after the fact and put into distribution. Also included in the corpus are compilations of articles and notices from government publications, which were brought together thematically and published in propaganda pamphlets to provide the public with a comprehensive treatment of a single topic. Covered in

416

Lisa Carroll-Davis

this corpus are themes of national history, membership in the FSLN, continuous revolution, threats from the opposition, economic and agrarian improvements, national unity, and the position of  the masses in the new government. Dif ferent individuals authored the documents, yet all were members of  the FSLN holding some of ficial position within the government. Specific authorship is attributed to five of the ruling members of the FSLN, including two of the founding members, Tomas Borge and Daniel Ortega. Four of  the documents are presented as merely coming from the Department of  Propaganda and Political Education of  the FSLN, with no single author. Ranging in length from 701 words to 17,118 words, the documents together total 81,678 words (see Table 24). Table 24  FSLN corpus documents Author

Title

Word count

Year

B. Arce Castaño

La Revolución Nicaragüense

1727

1980

D. Ortega

Las Reglas del Juego las Dicta el Pueblo

6441

1981

DEPEP

Nicaragua Bandera de la Paz

9477

1983

Dirección Nacional Ayer Unidad Nacional, Hoy Unidad Nacional 763

1980

FSLN

Acta de Compromiso del Pueblo Nicaragüense

701

1980

FSLN

Métodos de Planificación en la Dirección del Trabajo de Masas

2858

1980

FSLN

Nicaragua Golpea a la CIA

17118

1983

FSLN

Propaganda de la Producción

10979

1980

R. Morales Aviles

La Dominación Imperialista en Nicaragua

15060

1982

R. Morales Aviles, O. Turicos, & C. Fonseca

Qué Es un Sandinista?

2091

1980

T. Borge

La Bandera Azul y Blanco Defendida por la Bandera Roja y Negra

5938

1981

T. Borge

Nuestro Niños Lo Van a Entender

2390

1980

V. Tirado

La Historia Dio la Razón a Carlos Fonseca

6135

1981

417

Identifying the Enemy

The corpus which I created for a comparison is taken from the online archives of the journal Envío which began printing in 1981.3 Envío originated as a means of supplying ‘critical support’ (Envío online) to the Sandinista government. The journal, which is published by the Jesuit University of  Central America in Managua, does not currently, and did not originally, have any formal ties to the revolutionary government. Though several newspapers, magazines and journals were censored during the FSLN regime, Envío never had its content challenged, changed or suppressed by the government (Envío online). Given that the journal was sympathetic to the FSLN and yet treated it with a critical eye, Envío provides a good comparison with the of ficial government discourse. In order to provide a relevant comparison with the FSLN corpus, the articles I chose contain similar themes and topics within a similar timeframe to the government documents. Ranging from 1981 to the beginning of 1984, the articles cover discussions of the opposition, problems with the revolution, issues of religion and government, Nicaragua’s history, and the position of  the masses under the new government. These ten articles total 42,676 words with the shortest being 1,639 words and the longest 6,938 (see Table 25). Table 25  Envío corpus contents Title

Words

Year

Los Contrarrevolucionarios: Paladines de la Libertad?

4179

1983

Discusión de Problemas Nacionales

2650

1981

Ejército del Pueblo y para la Defensa

4690

1983

La Guerra Silenciosa contra Nicaragua

3701

1982

Jalapa Es Nicaragua: Testimonios de una Nueva Consciencia

6938

1984

La Ley de Medios de Comunicación y el Diario

3265

1982

Sandino: Cronología del Heroe Nacional

4989

1984

Segundo Aniversario de la Revolución

1639

1981

La Invasión Ya No Es Silenciosa

6507

1983

La Lucha Ideológica en las Iglesias Evangelicas Nicaragüenses

4118

1982

3

The Envío journal archives from 1981 to the present are available at .

418

Lisa Carroll-Davis

The corpora were analysed using the WordSmith tools version 5.0 computer program in conjunction with methods from Wodak et al. (1999) and other branches of  CDA. In using WordSmith, the WordList feature was employed to highlight frequently occurring elements by displaying the number of occurrences of a specific token and the number of individual texts in which that token appeared. The Concord feature of  the program then was used to create a keyword in context concordance of each instance of a token’s use. Within Concord it then is possible to examine the collocations, or co-occurring words, associated with the token under inspection. The program also allows for a variety of statistical calculations relating to the collocations. The statistical test chosen for the analysis presented here is the t-score, which indicates the likelihood that two words would occur together at random, given their individual distribution across the corpus. According to Church and Mercer, a t-score of 1.65 indicates a significant relationship between two collocates, with 95 per cent certainty that the pair did not occur through random chance (Church & Mercer 1993: 19–20). Additionally, it is possible with WordSmith to perform comparisons between two corpora, creating a keyword list comparing the frequencies of words on both wordlists. The product is a list which indicates the likelihood that words occurring in the comparison corpus were intentionally chosen by the author (Baker 2006: 125–126). The KeyWord feature in WordSmith produces a list by comparing the frequency of each word in the two corpora against each other. The result is a list that indicates words that are either unusually frequent, or infrequent, in one corpus in comparison with the other (Smith online). The outcome of  this combined study revealed that the Sandinistas crafted their propaganda materials to portray the opposition in a delegitmizing manner, particularly when compared to the manner in which the opposition was labelled in the Envío materials.

Identifying the Enemy

419

4 Analysis: Naming the Sandinista opposition Using the functions of  WordSmith discussed above, my analysis revealed that the issue of naming the opposition to the Sandinista revolution was an area which warranted further consideration in light of  the historical and cultural situation. Through the KeyWords function in WordSmith, a disparity between the FSLN and Envío corpora emerged. Despite the numerous references in the FSLN documents to fighting to protect and defend the Revolution against opposition, the Envío corpus used the term contrarrevolucionario statistically more often than the FSLN corpus. With significant results in KeyWords beginning with a keyness score of 0.000001, comparing the Envío corpus to the FSLN corpus presents a keyness score of 0.0000000004 for the term, indicating a greater than 99 per cent certainty that the result was not a random occurrence (Baker 2006: 126). Taking this as a point of departure, an examination of  the use of all of  the lemmas of contrarrevolucionario (accounting for masculine, feminine, plural and abbreviated variants) shows that in the Envío corpus there are 106 total instances of the word. The same examination of the FSLN corpus produces a total of only fifty-three instances. Given that the FSLN corpus is nearly twice the size of the Envío corpus, this result is significant in its own right. Whilst the FSLN is engaged in a fight for its own legitimacy it references the contrarrevolucionarios, the exiles against whom the Revolution is fighting, a mere fifty-three times. Further investigation into the use of  the term reveals that of  those occurrences all but thirteen are used as adjectives describing actions or people. Within the thirteen in which the term is a noun, labelling a particular person or group, six are found in reported speech which does not come directly from the FSLN directorate or the government. By contrast, thirty-five of  the 106 uses of  the term in the Envío corpus are used as a noun labelling the opposition to the FSLN. The questions this information raises are why did the FSLN intentionally avoid this specific title and what names did they use to identify and label the opposition?

420

Lisa Carroll-Davis

Contrarrevolución, a counter-revolution, in one sense co-opts the monopoly on the revolution which the FSLN claims. The use of  the possessive pronoun nuestra with revolución creates unity among the population under the new government. With revolución appearing 253 times in the FSLN corpus, the t-score associating it with nuestra at 6.473 points to a strong relationship between the collocates. In turn, this relationship indicates the likelihood that the intent in creating the text was to juxtapose these two tokens frequently, forty-one times throughout the corpus and occurring in each of  the thirteen documents. In context, the collocation nuestra revolución presents a perspective of  ‘our revolution’ as one that is grounded in the common history and which represents all those who are patriots of  the nation. Here one sees the exclusionary ef fect of  the we/us element; those who are ideologically in line with the governmental organizations are included as part of the supposed whole, whilst those ideologically opposed to it are positioned outside of the whole and therefore not part of  the country. In the following example, those who are against the revolution are depicted as outside the collectivizing of  the first person plural, even though the people referred to are exiled Nicaraguans: ‘They are the same ones who go around working for the isolation of our revolution outside of our country’4 (Ortega 1981: 7). The revolution is presented as belonging to all those who remain inside the country, regardless of membership in the FSLN. Allowing the opposition to be seen as a form of revolution creates an appearance of  legitimacy since the Sandinista government was itself  founded on revolutionary ideals. Also, contra, as it means ‘against’, would give resonance with those inside Nicaragua who, while not necessarily militant, opposed the FSLN government, thus providing a support base within the nation for the opposition fighters. Given that one of the opposition’s goals was to establish a rival government in a Nicaraguan border province, support throughout Nicaragua would allow such a government to claim legitimacy.

4

‘Son los mismos que andan haciendo el labor de aislamiento en contra de nuestra Revolución en el exterior’.

Identifying the Enemy

421

As a prefix contra typically implies a negative connotation; however, it was a name embraced by the opposition. The United States government likened the Contras to ‘freedom fighters’ and the American ‘founding fathers’ (Schroeder 2005: 70), casting the name in a positive light. So, for the FSLN, naming the opposition by a variety of names which would rob the movement of  legitimacy was a strategy that strengthened their claim to power and their exhortation to fighting the resistance. Thus while the Envío authors, whose express intention was to address a more international audience (Envío online), elected to employ a name internationally recognized and also accepted by the resistance themselves, the Sandinista authors chose to use other names which supported their ideology and their own claims to power and legitimacy. This illustrates how the FSLN leadership purposely chose labels and terms to construct a discourse that reinforced their position. The discourse was constructed to reinforce their position within the country as the governing party. Particularly during its nascent period, the Sandinistas relied on patterns employed by Sandino himself, who called those who supported the US Marines in the 1920s and 1930s vendepatrias or traitors. This recourse to prior patterns can also be seen in the Sandinista leadership’s use of  the term contrarrevolucionario. With only thirteen occasions of contrarrevolucionario as a noun, the FSLN significantly more often chose to label the Contras using dif ferent terms. Three nouns in association with various adjectives depicting enmity are used fifty-eight times to identify the opposition. One such noun is the term bandas. Fifteen out of  the sixteen occasions where bandas appears in the corpus refer to the exiled opposition. These ‘bands’ were described as counterrevolutionary, armed, and Somocists. They also were described as villainous and assailants (Borge 1980: 15). Another term employed in naming the enemy was elemento. This word and its morphological variants occur in the corpus fifty-one times, nineteen of  them indicating the anti-Sandinista opposition. Both terms are used in the following quote from FSLN leader Tomas Borge: ‘But there are elements in that region linked to Somocism, linked to assaulting

422

Lisa Carroll-Davis

bands’5 (Borge 1980: 15–16). Similar to the ‘bands’, the ‘elements’ are armed, counterrevolutionary supporters of Somocism. However, the most telling name applied to the opposition is guardia. The Somozas’ military force was the Guardia Nacional (National Guard), which was used to oppress and control the citizens of  Nicaragua as the dynasty enforced their control over the country. Originally created in 1926 by the US Marines stationed in Nicaragua during Sandino’s fighting, the Guardia Nacional came to be seen by the Nicaraguans as a symbol of  Yankee imperialist interference and also of  the ruthless reign of  the Somozas. The Sandinistas as well as others who joined them in the fight against Somoza and his army were particularly targeted by the Guardia, which meted out brutality in an ef fort to dissuade others from joining the cause. Given the notorious reputation the Guardia members had among the people in Nicaragua, this label was likely more ef fective than that of contrarrevolutionario in framing the opposition as lawless rebels seeking only to regain the power they once enjoyed. Twenty of the thirty-six uses of guardia and its variants by the FSLN refer to post-revolutionary opposition fighters. The other thirteen are references to the Somozas’ former army. In one instance the opposition is called ‘former members of  the National Guard’ (DEPEP 1983: 34), and on nine other occasions they are called exguardias somocistas. A t-score of 2.998 for the collocation guardias somocistas emphasizes the strong connection between these two terms. Three times los guardias is used with somocistas, leaving the other ten uses of  the term with an implied connection to Somoza. In the following example we see the connection to the prior regime without direct reference to it: ‘[The businessmen] go as far as to even defend the criminal guards who are captured and those criminal guards who are conspiring from outside the country’6 (Ortega 1981: 7). These guards are not of any of ficial regime, but are criminal, and supported by the same citizens who supported the Somozas, the businessmen. This connection linked 5 6

‘Pero hay elementos en esa región ligados al somocismo, ligados a bandas de asltantes …’. ‘…[los empresarios] llegan incluso a defender a los guardias criminales que están presos y a los guardias criminales que están conspirando desde el exterior’.

Identifying the Enemy

423

the opposition with criminality, abuse, corruption, and a return to the old regime, something which only a small number of Nicaraguans supported. Beyond tying the violent, repressive past to the present counterrevolutionary violence, using labels like banda, elemento, and guardia rather than contrarrevolutionario denied the opposition legitimacy in its fight. Significantly, the primary means of denying the opposition legitimacy appears to be through the denial of an identity for them within the Sandinista literature. The FSLN leadership achieved this by not assigning them one name in particular. Via the use of several dif ferent terms, no one specific identity was conferred to the counterrevolutionary groups. In the Sandinista propaganda materials, this strategy obscured the opposition and fragmented their identity. Thus, one can find in the Sandinista discourse on their opposition the impact that framing and naming the other can have on the construction of an out-group.

5 Conclusion A corpus approach to the Sandinista propaganda materials using WordSmith along with a CDA evaluation of  the results has brought out an important tool in the propagandist’s arsenal, that of naming and framing the enemy. Without a comparison between the FSLN and Envío documents, the stark contrast in use would not have become apparent. These results show that the name contra was not widespread in use among the Sandinista leaders in their propaganda during the regime’s first three years in power. Further study is needed to determine if this trend in naming the opposition continued throughout the ten years of  Sandinista rule, or if  the tide of  legitimacy from abroad prevailed upon the Sandinista leaders, causing them to adopt the name’s use as well. In my view, this initial study of the Sandinista regime’s use of political discourse does, however, show a willingness on the part of  the Sandinistas to discredit their enemy in the same way they were discredited themselves by the USA and the Somozas.

424

Lisa Carroll-Davis

The Sandinistas’ intentional use of  labelling displayed their own grasp of  the important role political positioning plays in building hegemony. The historical setting, where Sandino and his original fighters were primarily delegitimized within the country and internationally through naming, such as calling them bandits or blanket thieves, provided the FSLN with a pattern which was familiar to the Nicaraguan population. Exploiting this pattern, the Sandinistas used the category of  the former military power, the National Guard, and the negative associations it carried as the basis of its label. This ef fectively connected two patterns in Nicaraguan political discourse, which provided the label for the opposition with a stronger foundation and also allowed it to carry a more significant bias towards the FSLN’s position. More generally, this article has shown that the strategy of assigning an opponent a name which has a negative frame of reference is ef fective in associating the opponent with that group. The related strategy of using varied labels and references when discussing the opposition can serve to deny that entity an identity within the dominant discourse. What both strategies provide is a means for the dominant party of patterning that encodes their ideology in the accepted, taken-for-granted discourse. Through their control of the discourse, the power-holders were shown as directing the of ficial perception of  the opposition. This study provides an initial look at the means through which CDA and corpus studies can be applied in tandem to the examination of discourses of political conf lict, and the importance of labelling within a discourse, which will perhaps lead to additional means of understanding and analysing the discourse of conf lict.

References Primary Literature Arce Castaño, Bayardo. 1980. La Revolución Nicaragüense. Managua: Secretaria Nacional de Propaganda y Educación Política. Borge, Tomás. 1980. Nuestro Niños Lo Van a Entender. Managua: Silvio Mayorga.

Identifying the Enemy

425

Borge, Tomás. 1981a. ‘La Bandera Azul y Blanco Defendida por la Bandera Roja y Negra.’ In: Daniel Ortega, Las Reglas del Juego las Dicta el Pueblo. La Bandera Azul y Blanca Defendida por la Bandera Roja y Negra Hoy Ondea Segura de sí Misma. La Historia Dio la Razón a Carlos Fonseca. Managua: Departamento de Propaganda y Educación Política del FSLN, 23–40. Borge, Tomas. 1981b. Los Primeros Pasos. Mexico City: Siglo Veintiuno. DEPEP (Departamento de Propaganda y Educación Política del FSLN). 1983. Nicaragua: Bandera de la Paz. Managua: Secretaria Nacional de Propaganda y Educación Política. Dirección Nacional. 1980. ‘Ayer Unidad Nacional para lograr el Triunfo Hoy Unidad Nacional para mantener la Victoria.’ In: Poder Sandinista. Managua: Secretaria Nacional de Propaganda y Educación Política. Envío. 1981a. ‘En el Segundo Aniversario de la Revolución.’ Revista Envío 3. accessed 20 May 2010. Envío. 1981b. ‘Foro de Discusión de Problemas Nacionales: una Posibilidad más de Entendimiento.’ Revista Envío 2. accessed 20 May 2010. Envío. 1982a. ‘La Guerra Silenciosa contra Nicaragua: Estrategia del Terror.’ Revista Envío 17. accessed 20 May 2010. Envío. 1982b. ‘La Lucha Ideológica en las Iglesias Evangélicas Nicaragüenses.’ Revista Envío 15. accessed 20 May 2010. Envío. 1983a. ‘Los Contrarrevolucionarios: Paladines de la Libertad?’ Revista Envío 29. accessed 20 May 2010. Envío. 1983b. ‘Un Ejército del Pueblo y para la Defensa.’ Revista Envío 28. accessed 20 May 2010. Envío. 1983c. ‘La Invasión Ya No Es Silenciosa.’ Revista Envío 22. accessed 20 May 2010. Envío. 1984a. ‘Jalapa Es Nicaragua: Testimonios del Nacimiento de una Nueva Consciencia Popular.’ Revista Envío 32. accessed 20 May 2010. Envío. 1984b. ‘Sandino: Cronología del Héroe Nacional de Nicaragua.’ Revista Envío 33. accessed 20 May 2010. Envío. n.d.. ‘A Brief History of Envío Magazine.’ Revista Envío. accessed 20 May 2010. FSLN. 1980a. Acta de Compromiso del Pueblo Nicaragüense. Managua: Secretaria Nacional de Propaganda y Educación Política. FSLN. 1980b. Métodos de Planificación en la Dirección del Trabajo de Masas. Managua: Secretaria Nacional de Propaganda y Educación Política. FSLN. 1980c. Propaganda de la Producción. Managua: Secretaria Nacional de Propaganda y Educación Política.

426

Lisa Carroll-Davis

FSLN. 1983. Nicaragua Golpea a la CIA. Managua: Secretaria Nacional de Propaganda y Educación Política. Morales Aviles, Ricardo. 1982. La Dominación Imperialista en Nicaragua. Managua: Secretaria Nacional de Propaganda y Educación Política. Morales Aviles, Ricardo, Oscar Turicos & Carlos Fonseca. 1980. Qué Es un Sandinista. Managua: Secretaria Nacional de Propaganda y Educación Política. Ortega, Daniel. 1981. ‘Las Reglas del Juego las Dicta el Pueblo.’ In: Daniel Ortega. Las Reglas del Juego las Dicta el Pueblo. La Bandera Azul y Blanca Defendida por la Bandera Roja y Negra Hoy Ondea Segura de sí Misma. La Historia Dio la Razón a Carlos Fonseca. Managua: Departamento de Propaganda y Educación Política del FSLN, 1–22.

Secondary Literature Arellano Oviedo, Francisco. 2007. Diccionario del Español de Nicaragua. Managua: PAVSA. Baker, Paul. 2006. Using Corpora in Discourse Analysis. New York: Continuum. Baker, Paul, et al. 2008. ‘A Useful Methodological Synergy? Combining Critical Discourse Analysis and Corpus Linguistics to Examine Discourses of Refugees and Asylum Seekers in the UK Press.’ Discourse & Society 19(3), 273–306. Baracco, Luciano. 2005. Nicaragua: The Imagining of a Nation – from Nineteenth Century Liberals to Twentieth Century Sandinistas. New York: Algora. Bhatia, Michael V. 2005. ‘Fighting Words: Naming Terrorists, Bandits, Rebels and Other Violent Actors.’ Third World Quarterly 26(1), 5–22. Chilton, Paul. 2004. Analysing Political Discourse: Theory and Practice. London: Routledge. Church, Kenneth, & Robert Mercer. 1993. ‘Introduction to the Special Issue on Computational Linguistics Using Large Corpora.’ Computational Linguistics 19(1), 1–24. Doorn, Jacques van. 1975. The Soldier and Social Change: Comparative Studies in the History and Sociology of  the Military. Beverly Hills, CA: Sage. Fairclough, Norman. 1995. Critical Discourse Analysis: The Critical Study of Language. London: Longman. Kendall, Gavin. 2007. ‘What Is Critical Discourse Analysis? Ruth Wodak in Conversation with Gavin Kendall.’ Forum for Qualitative Social Research 8(3), article 29. accessed 20 May 2010.

Identifying the Enemy

427

Mautner, Gerlinda. 2009. ‘Checks and Balances: How Corpus Linguistics Can Contribute to CDA.’ In: Ruth Wodak and Michael Meyer (eds), Methods of Critical Discourse Analysis. 2nd edn. London: Sage, 122–143. Meyer, Michael. 2001. ‘Between Theory, Method, and Politics: Positioning of  the Approaches to CDA.’ In: Ruth Wodak and Michael Meyer (eds), Methods of  Critical Discourse Analysis. London: Sage, 14–31. Mintz, Alex, & Steven B. Redd. 2003. ‘Framing Ef fects in International Relations.’ Synthese 135, 193–213. Phillips, Nelson, & Cynthia Hardy. 2002. Discourse Analysis: Investigating Processes of  Social Construction. Sage University Papers Series on Qualitative Research Methods, 50. Thousand Oaks, CA: Sage. Schroeder, Michael J. 2005. ‘Bandits and Blanket Thieves, Communists and Terrorists: The Politics of  Naming Sandinistas in Nicaragua, 1927–36 and 1979–90.’ Third World Quarterly 26(1), 67–86. Smith, Dan. 1998. ‘Language and Discourse in Conf lict and Conf lict Resolution.’ In Sue Wright (ed.), Language and Conf lict: A Neglected Relationship. Philadelphia& Clevedon: Multilingual Matters, 18–42. Smith, Hazel. 1993. Nicaragua: Self-Determination and Survival. London: Pluto. Smith, Mike. 2008 (online). WordSmith Tools Online Manual: Version 5.0. accessed 20 May 2010. Spender, Dale. 1980. Man Made Language. London: Routledge. Thompson, John B. 1984. Studies in the Theory of  Ideology. Oxford: Blackwell. Wodak, Ruth. 2001a. ‘The Discourse-Historical Approach.’ In: Ruth Wodak and Michael Meyer (eds), Methods of  Critical Discourse Analysis. London: Sage, 63–94. Wodak, Ruth. 2001b. ‘What CDA Is about – A Summary of its History, Important Concepts and its Developments.’ In: Ruth Wodak and Michael Meyer (eds), Methods of  Critical Discourse Analysis. London: Sage, 1–13. Wodak, Ruth, & Michael Meyer (eds). 2001. Methods of  Critical Discourse Analysis. London: Sage. Wodak, Ruth, & Michael Meyer. 2009. ‘Critical Discourse Analysis: History, Agenda, Theory and Methodology.’ In: Ruth Wodak and Michael Meyer (eds), Methods of  Critical Discourse Analysis. 2nd edn. London: Sage, 1–33. Wodak, Ruth, et al. 1999. Discursive Construction of National Identity. Angelika Hirsch & Richard Mitten (trans.). Edinburgh: Edinburgh University Press.

Krassimir Stoyanov

Ritualized Slogan Lexis in the Bulgarian Press during the Times of  Violent Contradiction in Ideologies (1944–1947)

Abstract This chapter deals with slogans published in the Bulgarian press during the period 1944–1947. This was the period after World War II when the Fatherland Front coalition was formally in power, but in practice, it was the Bulgarian Workers’ Party, later named the Bulgarian Communist Party, that ruled the country. It analyses the language of typical slogans used to implement the political agenda of the totalitarian government. A characteristic feature of these texts was the fact that they were largely inf luenced by similar texts in political speech practices in Soviet Russia. This chapter also identifies a number of semantic fields of ritualized lexis, the social nature and the ritualized realization of such lexis within the context of the socio-historical situation. The research takes a comparative approach to the aforementioned linguistic phenomena. It compares the linguistic features of ritualized texts in the Bulgarian press to the pattern used by the Soviet party organs – the Pravda and Izvestiya newspapers. Finally, some conclusions are of fered about the language policy of  Bulgaria’s ruling party at the time.

This chapter analyses the linguistic phenomena that transformed the Bulgarian vernacular in the wake of the major political and socio-economic changes occurring in Bulgaria after the end of  World War II. The focus is on a specific historic period beginning in September 1944. Up until that point, Bulgaria had been Germany’s ally in the war, but by September Soviet troops were approaching the Bulgarian frontier. The USSR declared war on Bulgaria on 5 September 1944, and the Red Army marched into the country receiving a huge welcome. Before this time, the Bulgarian people

430

Krassimir Stoyanov

had been hard pressed by both domestic and international circumstances, the pressure from the powerful Nazi military being the most inf luential among them. The pro-German Bulgarian government had already declared war on the USA and England, although the arrival of  Soviet troops was received as ‘liberation’ by the greater part of  the Bulgarian population. This was well demonstrated by the fact that the Bulgarian army put up no resistance. The Russians showed respect for the Bulgarian military and civil law. The government of the Fatherland Front took power, doing away with the fascist dictatorship. The Bulgarian army was completely transformed. The new troops were reinforced by over 10,000 anti-fascist guerrilla groups, who were former political and concentration camp prisoners. Bulgaria then declared war on Germany, joined the third Ukrainian front, and sent an army of 180,000 soldiers to liberate large territories of southern Yugoslavia, Hungary, and Austria victoriously marching as far west as the foothills of  the Alps, where they met the Eighth English Regiment. During just this final stage of  the war Bulgaria suf fered significant losses – over 32,000 of ficers and soldiers were killed in the line of duty and financial losses came to about 133 billion leva (about EUR 25 billion in today’s currency). This was part of  Bulgaria’s contribution to the anti-fascist war and the success of  the progressive coalition, which fought against Nazism. Beginning in September 1944, Bulgaria had already been freed from fascist institutions and had opened politically to democratic change; however, the country was facing rather complicated circumstances. Under the conditions of complete economic and political deterioration the Bulgarian Workers’ Party1 controlled the key positions of power and strictly adhered to Soviet practices. The severe winter threatened to starve the nation. The political struggles and ef forts to sustain Bulgaria’s place on Europe’s political map and its status as a sovereign state were yet to come. The new government claimed to belong to the people. Through a systematic series of administrative rulings, this coalition government took care of  the food, clothes, education, health and security of the entire nation. All the reserves

1

БРП – [BWP] Bulgarian Workers’ Party, after 1947 named БКП – [BCP] Bulgarian Communist Party.

Ritualized Slogan Lexis in the Bulgarian Press

431

were used to provide food for a population of six and a half million people, and to support the army to fulfil ‘its democratic duty’ beyond the country’s frontiers. Simultaneously an opposition movement was beginning to arise, which remained rather aloof from politics until the end of its existence in the autumn of 1947. This movement was largely sidelined as the majority felt it would take the country back to its reactionary pre-war state. Many young adults volunteered or were recruited by the army and they tried to find fulfilment from within the new social order while chasing the trendy ideas of a new Europe. The Workers’ Party – the dominating force of  the Fatherland Front – managed to construct a new social paradigm attuned to the progressive ideas popular in Europe at the time. The popular daily press was a powerful tool in this process. The press’s ef fectiveness in using propaganda was well known in traditional Bulgarian journalism – from the time of the monarchy’s nationalistic disposition through to the end of fascism in 1944, as well as after its fall. At this point, the Workers’ Party exercised full control over the country and quickly organized the nation’s social and cultural life. In this respect, if we turn to the strategic perspective of social events, we find an explicit the, systematic presentation of an expressive, contemporary ‘contrast to the dark past’, thus making a new and very attractive type of social ritualism. The meaning of  the ritual can be thought of as a particular kind of activity central to the collective way of  life. In this light, an array of concepts like custom, rite of passage, votive of fering, ceremony etc. are used as synonyms or as similar in meaning. We do not accept these concepts as identical, although in Bulgarian and Russian literature the meaning of the concepts ritual and rite is often the same. Moreover, the term ritual has been rejected as being foreign to the Bulgarian language. For more clarity, we must highlight the circumstances that distinguish a ritual from a rite, a custom, and even more so a ceremony, which involves activities with more procedural description. Although both rite and ritual convey a sense of irrationality (Bayburin 1993) in relation to the activity being carried out, each places a dif ferent degree of social significance on the procedure at hand. A rite is a mystical traditional practice, resulting from a mythological perception of  the world’s reiteration of overcoming crises; whereas a custom refers to traditional practices aiming at consolidating the collective

432

Krassimir Stoyanov

memory, which is accepted as sacred in transitional and critical moments encountered in the daily routines of a society. A ritual on the other hand, marks significantly more valuable phenomena and events in the social hierarchy. These occasions are related to the crucial transitions and transformations in the existence of significant social communities, which include social organizations such as political parties and institutions; groups of ethnic-, economic- or ethics-related individuals such as secret societies, religious groups, etc. Through performing rituals these groups verify the actual value of  their existence at a particular point in time.2 It is important to note that a ‘rational explanation’ of ritual procedures is a matter of cultural viewpoint. As it has been pointed out, ‘If  “non-rationality” is made a criterion of ritual, it must be remembered that the judge of what is rational is the observer, not the actor’ (International Encyclopedia of  the Social Sciences, 1968). Explained in this manner the act of ritual becomes a traditional form of demonstrating power, social strength and vitality. A ritual is carried out in a particular environment involving a set timeline as well as a fixed succession of activities, words, and symbols, which are all strictly performed according to rules. In the context of a socially significant event, ritual elements such as time, actions, words, colours, and music as well as various objects, all function as symbols of  the expressed, designed and planned, desired and expected vitality in the realization of a society. In anthropology and ethnology there is the concept of ritually organized communities and these societies have their own corresponding culture. We will argue that the totalitarian society that existed in post-war Bulgaria is a clear example of a state formation of a ritual kind. If we look into the aforementioned associations and conventionalities related to prestige in this particular social context (Bulgaria during conditions of severe ideological conf lict throughout the years of  transition), we can single out the following four characteristics: 2

Eliade points out that repetition through ritual always brings back sacred time. If a person goes back to the mythic event, which has been reconstructed through ritual, he/she does not go back in historic time but leaves their profane time to go to sacredness (Eliade 2008: 26).

Ritualized Slogan Lexis in the Bulgarian Press

433

1 Gathering masses of people. The tendency towards gathering masses of people in one place existed not only in Bulgaria, but across Europe, and even on the other side of the Atlantic. This tendency exploited the symbols of democracy, freedom, and anti-fascism. These symbols of  the new times were adopted by the ruling coalition led by the Communist Party, which ef fectively used them to create a catalogue of ritual practices, carried out on an amazingly large scale. 2 Nationwide participation. Each stratum of  the society was involved. 3 The idea of imminent economic prosperity for all was widely proclaimed in connection with the dramatic victory over Nazi fascism and the exploitation of  the power seized by the coalition. 4 Supremacy of  the Russian, Soviet and Slavic ideals of strength, glory, and greatness on a vast historic scale. By adopting such symbolism, and implementing it in routine practices, the new government suf focated all other political factions, depriving the opposition of opportunities for ritual practices, celebrations, and social gatherings. The new ritual symbolism was also fashioned as prestigious by utilizing the lexis and phraseology of  the new public speaking. As previously mentioned, words play a main role in ritual. Words used in ritual are ‘free from the concreteness of everyday communication’ (Bayburin 1993: 209). In this case, the words’ usage reveals ‘a special organisation of the verbal text’, in which ‘there is no spontaneous speech’, and in which ‘not mere utterances, but formulas’ are expressed. So, the verbal text becomes an integral part of  the ritual only when it is not spontaneous and it is antecedently organized. Therefore, it can be argued that the defining textual features of a slogan as part of a ritual are the particular structure and lack of spontaneity, in contrast to the emotional stimulation of repetition and chanting. Therefore, these practices assert the claim that words have power. The power of words is expressed through various procedures and verbal behaviour such as exorcism, casting hexes and spells, swearing oaths and even passwords. We will cite an example in a popular newspaper at the time called Narodna armiya [Army of  the People] from 5 March 1945. The newspaper published an article under the headline ‘The Exarch cast

434

Krassimir Stoyanov

a curse over Slavic foes’: ‘His Beatitude Exarch Stefan gave a speech […] raising his hand in a gesture of anathema the Exarch uttered, “Cursed be the one who would raise his hand against the Slavic commonwealth!” All the people who were present were standing up and loudly cursing the Slavic foes in unison.’ Of course, the opposite behaviour – non-speaking, or symbolic silence – is also part of ritual and has its rationale. For example, Enyovden (24 June) is a holiday on the Bulgarian folk calendar when a woman can tell your fortune by using f lowers and herbs that have been left overnight in a menche (a small metal vessel) filled with spring water. The water is brought from a well by young unmarried women who must keep complete silence while they are carrying the water to the place where the ritual is to be performed, so as not to spoil the supernatural presence. The water they have brought is called malchana voda (silent water). Our aim is not to fully describe the various rituals of communist propaganda during the initial stages of the Party’s coming to power. We will only deal with one particular element – slogans – as an example manifested in communist verbal behaviour. The rituals of totalitarian publicity required large-scale locations for carrying out the activities, which systematically integrated colour, music, and verbal expressions. It was common practice that important events in the transition to the new state were celebrated through ritual, which repeated the traditions and emphasized the vitality of  the future social prospects. These characteristics required providing a special setting: representative publicity – a gathering of  thousands of people (a meeting, a party congress); a huge hall (stage, square); imposing solemn music; colour symbols in bright red; portraits – enormous faces of important political figures (deceased or living); impressive speaking; authoritarian order and control over the situation, despite the countless masses of people. The solemn speeches on the stage were ritually started and ended with slogans. For the purpose of this paper we have extracted a corpus of 117 slogan phrases from the major newspapers of  this particular historic period –

Ritualized Slogan Lexis in the Bulgarian Press

435

Работническо дело3 [Rabotnichesko delo], Народна армия4 [Narodna armiya] and Отечествен фронт5 [Otechestven front]. The variety of slogans can be divided into the following groups: 1

Chanting – repeating emotive words usually only by the audience: БКП!БКП!СЛАВА!СЛАВА!КПСС!КПСС – БКП! [BCP! BCP! Glory! Glory! CPSS! CPSS – BCP!]6

2

Short phrases – phrases that were most often used in the first years after 1944: СМЪРТ НА ФАШИЗМА – СВОБОДА НА НАРОДА! [Death for fascism – freedom for the people!] ВСИЧКО ЗА ФРОНТА – ВСИЧКО ЗА ПОБЕДАТА! [Everything for the Front – everything for the Victory!] ВЕЧНА ДРУЖБА! [Friendship forever!]7 НИКОГА ПОВЕЧЕ ВОЙНА! [No more war!] ВЕЧНА СЛАВА НА ПАДНАЛИТЕ ПАРТИЗАНИ И ПАРТИЗАНКИ ОТ ОБЛАСТТА! [Eternal glory for the fallen guerrilla men and women from this region!]

• • • • •

3 3.1

Exclamations with May …!: Да живее …! [May … live!]. This is the most common way to begin a slogan during the period described: ДА ЖИВЕЕ ЧЕЛНИЯТ ОТРЯД НА РАБОТНИЧЕСКАТА КЛАСА И НЕГОВИЯТ ЦК!8 [May the forefront party of  the working class and its CC live!]; ДА ЖИВЕЯТ СВОБОДОЛЮБИВИТЕ НАРОДИ! [May the libertyloving peoples live!]

3

Работническо дело [Rabotnitcesko delo] – an of ficial daily of the ruling BCP until November 1989. Народна армия [Narodna armija] – The People’s Army daily from September 1944. Отечествен фронт [Otetcestven front] – The Fatherland Front daily 1944–1989. БКП – [BCP] Bulgarian Communist Party; КПCC – [CPSS] Communist Party of  the Soviet Union. It expresses the idea of eternal friendship with the USSR. ЦК – [CC] Central Committee of  The Bulgarian Communist Party.

4 5 6 7 8

436

Krassimir Stoyanov

3.2 3.3 3.4

Да крепне …! [May … grow stronger!]: ДА ЖИВЕЕ И КРЕПНЕ ЕДИНСТВОТО МЕЖДУ НАРОД И АРМИЯ! [May the unity between nation and army grow stronger and stronger!] Да пребъде …! [May … live forever!] ДА ЖИВЕЕ И ПРЕБЪДЕ ДЕЛОТО НА ОФ!9 [May the work of  the FF live forever!] etc. In order to make these exclamations more emphatic, combinations of two such exclamations were often used e.g. May … live and grow stronger! Да бъде …! [May it be …!]; Да бъдем …! [May we be …!]: ДА БЪДЕМ СМЕЛИ ПРЕД ХИТЛЕРИСТКИТЕ БАНДИТИ! [May we be fearless of  Hitler’s gangs!]

4 4.1 4.2

Consolidating appeals: Нека да …! This construction is also translated with [May …!], but in Bulgarian Нека …! [Neka …!] expresses an emphatic appeal e.g. НЕКА ДА НЕ ГУБИМ НИТО МИНУТА! МОМЕНТЪТ Е СЪДБОНОСЕН! [May we not waste a single minute! It’s a moment of destiny!] It is a matter of  fact that in slogan texts during this period this construction was hardly ever used. It was characteristic of the slogans of  the country’s next period of  ‘socialist construction’. Да + призивен глагол [Da + ‘appeal verb’ (Levin 1993)] – a widely used appeal structure: ДА НАПРЕГНЕМ ВСИЧКИ СИЛИ, ДА МОБИЛИЗИРАМЕ ВСИЧКИ СРЕДСТВА ЗА РАЗГРОМА НА ОБЩИЯ ВРАГ НА ВСИЧКИ БАЛКАНСКИ НАРОДИ, НА ЦЯЛОТО СВО БОДОЛЮБИВО ЧОВЕЧЕСТВО – ГЕРМАНСКИТЕ РАЗ БОЙНИЦИ! [Let us make all ef forts, let us mobilize all our forces to defeat our common foe of all the Balkan Peoples, of all liberty-loving mankind – the German bandits!]

9

ОФ – Отечествен фронт [Otechestven front] – Fatherland Front: A union of a number of anti-fascist parties with the BCP at the forefront, which formally took power as soon as the fascist regime fell in September 1944.

Ritualized Slogan Lexis in the Bulgarian Press

437

4.3 За + именна конструкция [Za + Noun Phrase]: • ЗА ПРИЯТЕЛСКИ ОТНОШЕНИЯ С АМЕРИКА, АНГЛИЯ И ФРАНЦИЯ! [For friendly relations with America, England and France!]; • НАПРЕД ЗА ОКОНЧАТЕЛНО ОЧИСТВАНЕ НА БАЛКАНИТЕ ОТ ГЕРМАНСКИТЕ ВОЙСКИ, ЗА ПЪЛНИЯ РАЗГРОМ НА ХИТЛЕРИСТКА ГЕРМАНИЯ! [On for the complete riddance of German troops from the Balkans, for the complete defeat of  Nazi Germany!] 4.4 На + именна конструкция [Na + Noun Phrase]: НА БОЙ ЗА ОКОНЧАТЕЛЕН РАЗГРОМ НА ХИТЛЕРИСТКА ГЕРМА НИЯ! [Let’s fight for the complete defeat of  Hitlerite Germany!] 5



Brandishing constructions. Such constructions were slogans containing the pre-prepositional nominal element death: • СМЪРТ НА ФАШИЗМА! СМЪРТ НА ПРЕДАТЕЛИТЕ! [Death to fascism! Death to traitors!] • СМЪРТ НА НАРОДНИТЕ МЪЧИТЕЛИ! [Death for the people’s torturers!] etc. In the years of  the People’s Court the usual slogans were: • НЯМА МИЛОСТ ЗА НАРОДНИТЕ ИЗЕДНИЦИ! [No mercy for the people’s blood-suckers!] • НИКАКВА ПОЩАДА ЗА ХИТЛЕРИСТКИТЕ ПАЛАЧИ! [Don’t spare Hitler’s executioners!] • БЕСИЛКИ ЗА НАРОДНИТЕ ПАЛАЧИ! [Gallows for the people’s executioners!] • БЕСИЛКИ ЗА ВСИЧКИ, КОИТО КЛАХА НАРОДА! [Gallows for all those who slaughtered the people!] In this way, death is emphasized by eliminating virtues like mercy and life-sparing, and approval is given for the means of causing death, a gallows in this case. The construction of emotional reprehensible expressiveness strikes a balance by contrasting the meaning of the first word of the phrase to the meaning of the last one: (NO) mercy – blood-sucker; (NO)mercy – foe; (NO)mercy – executioners; defeat – victory. This lexical group contains a number of verb

438

Krassimir Stoyanov

variants with the meaning of causing death e.g. the appeals: Let us kill + object!; Let us destroy + object!; Let us smash + object!; Let us shatter + object!; Let us break X’s head! e.g.: • ДА ДОУБИЕМ ФАШИСТКИЯ ЗВЯР В НЕГОВАТА БЪРЛОГА! [Let us finish killing the fascist beast in its den!] • ЗА ОКОНЧАТЕЛНО ОЧИСТВАНЕ НА БАЛКАНИТЕ ОТ ГЕРМАНО-ФАШИСТКИТЕ ОКУПАТОРИ, ЗА ДОУБИВАНЕ НА ФАШИСТКИЯ ЗВЯР! [For a complete eradication of  the German fascist occupiers of  the Balkans, for complete killing of  the fascist beast!] Despite the great variety within this thematic and structural group, it cannot be compared with the next group of slogan texts. 6

Glorifying texts, which refer to an idea, an organization, a political party or a person. Slogans of  this kind proliferated over the next few years of totalitarian communist authority and personality cult (1946– 1954). The most significant events10 that were objects of glorification in slogans at the beginning of  this period: • taking power by The Fatherland Front • winning the victory over Nazi Germany • the general elections on 18 November 1945

We will address some lexical aspects of slogans as a ritual manifestation of  these three political projects. Two major processes emerge: 1

Militarizing everyday lexis: a beast = an enemy; a den = Headquarters; purification = a defeat, a complete victory.

10

In their organized mass perception the socializing and integrating function of ritual is fulfilled – to prepare a person for a collective life, to meet collective needs and strengthen the unity of  the community, the solidarity of all the members as a condition for its conversation in the long term (according to Durkheim’s classification, 1912, following Bayburin 1993: 31).

Ritualized Slogan Lexis in the Bulgarian Press

2

• • • •

•

• • • •

439

Military lexis permeates everyday lexis: a front = a field of activity, a new initiative in civil life; a struggle = a strategy for the fulfilment of an economic project; a weapon = a means for day-to-day propagandistic activities, etc. Regarding the first event – overthrowing the fascist government and establishing the coalitionist Fatherland Front (FF) – the following common slogans were found in the press: ДА ЖИВЕЕ ОФ! [May the FF live!] ДА ЖИВЕЕ ВЕЧНО ДЕЛОТО НА ОФ! [May the work of  the FF live forever!] ДА ЖИВЕЕ И ПРЕБЪДЕ ЕДИНСТВОТО НА ОФ! [May the unity of  the FF live and grow stronger!] ОТЕЧЕСТВЕНИЯТ ФРОНТ Е ПОЛИТИЧЕСКИ ОРГАНИЗАТОР НА НАРОДНИТЕ МАСИ, СЪТРУДНИК И КОНТРОЛЬОР НА УПРАВЛЕНИЕТО [The Fatherland Front is a political organizer of  the masses, a collaborator, and an overseer of  the State’s rulers!] ДА ЖИВЕЕ НОВА ОТЕЧЕСТВЕНО-ФРОНТОВСКА БЪЛГАРИЯ!�������������������������������������������������������� [May the new Fatherland Front’s Bulgaria live!] Chronologically, the second event was the victory over German fascism. The following slogan turns out to be a peculiar transition between the first and the second group: АКО НЯМАШЕ ОФ – НЯМАШЕ ПОБЕДА НАД НЕМСКИЯ ВРАГ! [If it hadn’t been for The Fatherland Front – the victory over the German foe would have been impossible!] НА БОЙ ЗА ОКОНЧАТЕЛНИЯ РАЗГРОМ НА ХИТЛЕРИСТКА ГЕРМАНИЯ! [To fight for the complete defeat of Hitler’s Germany!] ДА ЖИВЕЕ НАШАТА НЕПОБЕДИМА СЪЮЗНИЦА ЧЕРВЕНАТА АРМИЯ! [May our invincible ally The Liberator Red Army live!] ДА ЖИВЕЕ СЛАВНАТА БЪЛГАРСКА ВОЙСКА! [May the glorious Bulgarian army live!]

440

Krassimir Stoyanov

In this group of slogans there is a greater variety. The slogan texts usually refer to the idea of a Slavic unity and accession to European democracy. For example: • ЗА НАШАТА СВОБОДА И НЕЗАВИСИМОСТ! ЗА НАШЕТО СВЕТЛО БЪДЕЩЕ! ЗА БЪДЕЩЕТО НА ЦЯЛОТО ЧОВЕЧЕСТВО – НАПРЕД, СЛАВЯНИ! НАПРЕД КЪМ БЕРЛИН! [For our liberty and independence! For our bright future! For the future of mankind – Forward Slavic peoples! On to Berlin!] • ДА ЖИВЕЕ БОЙНОТО ЕДИНСТВО НА СЛАВЯНСКИТЕ НАРОДИ В БОРБАТА СРЕЩУ ВЕКОВНИЯ ВРАГ НА СЛАВЯНИТЕ! [May the military unity of  Slavic peoples in the struggle against the age-old foe of  Slavs live!] • ЕДИНСТВОТО НА СЛАВЯНИТЕ – НАЙ-ДОБРАТА ГАРАНЦИЯ СРЕЩУ ТЕВТОНСКАТА ОПАСНОСТ! ДА ЖИВЕЕ СЪВЕТСКИЯ СЪЮЗ – ДВА ПЪТИ НАШ ОСВОБОДИТЕЛ, ГРАНИТНАТА ОПОРА ЗА СВОБОДАТА И НЕЗАВИСИМОСТТА НА СЛАВЯНСКИТЕ НАРОДИ! [The unity of  Slavs – the best warrant against the Teutonic danger! May The Soviet Union live – twice our liberator, a tower of strength for the liberty and independence of  Slavic peoples] • ДА ЖИВЕЕ БОЙНОТО ЕДИНСТВО НА ВСИЧКИ СЛАВЯНИ, НАЧЕЛО С ВЕЛИКИЯ ИМ ОСВОБОДИТЕЛ – СЪВЕТСКИЯ СЪЮЗ! [May the military unity of all Slavs live, led by their great liberator – The Soviet Union!] The third event – the general elections – was also embedded in a ritual system, verbalized in the following slogans: • ДА ЖИВЕЕ ЧЕЛНИЯТ ОТРЯД НА РАБОТНИЧЕСКАТА КЛАСА И НЕГОВИЯТ ЦК! [May the front line team of the Workers’ Party with its Central Committee live forever!] • НАПРЕД, КЪМ СЛАВНИ ПОБЕДИ В ЗАЩИТА НА ВЪРХОВНИ НАРОДНИ ИНТЕРЕСИ! ДА ЖИВЕЕ СВОБОДНА, НЕЗАВИСИМА, ДЕМОКРАТИЧНА И МОЩНА БЪЛГАРИЯ! [On to new glorious victories in defence of supreme national interests! May a liberated, independent, democratic, and strong Bulgaria live!].

Ritualized Slogan Lexis in the Bulgarian Press

441

A group of core lexical items emerge as a compulsory part of of ficial political procedures, which functioned as rituals in times of ideological conf lict between 1944 and 1947, thus becoming ritualized. After 1948, this lexis was organized around new topics, related to the totalitarian party and the cult of personality i.e. the concept of  the Fatherland Front was left out of  this ritualized lexical set and new items were added like party, BCP, CPSS and revolution. The following lexical collocations emerge as ritualized: The Fatherland Front collocates with rule, government, Bulgaria; country collocates with great, free, democratic; unity collocates with military, indestructible, brotherly; friendship collocates with great, among Slavic peoples, Bulgarian-Soviet; The Red Army: invincible, heroic, glorious, liberator; fighter: great, ingenious, unf linching; leader: people’s, ingenious, dignified, great; people: Slavic, Bulgarian, dignified; peoples: liberty-loving (the collocation liberty-loving peoples was first encountered in a reply-telegram to Stalin from December 1944), heroic, united, Slavic, Balkan; liberty and independence were frequently used as a binominal; army collocates with glorious, Bulgarian, heroic, brave, our; Bulgaria: new, free, democratic, independent, powerful, strong; the Germans: fascists, oppressors, foe, mankind’s foe, the Slavs’ foe, bandits, Teutonic barbarians, Hitler’s bandits; The Soviet Union: great, liberator, twice liberator, invincible; victory: great, final, glorious, definitive, heroic.

The ritualization of this kind of language turned lexis into a sacred, operative structure for the political scenario. This lexis became part of a special stylistic layer. It was uttered with a special sound and intonation, accompanied by a ritualized standing ovation.11 Newspaper slogans were assigned a special place in the spatial design of  the papers. It was not infrequent that these verbal structures were placed in a distinct box at the top or the bottom of  the page, sometimes in the middle. It was not unusual to encounter slogans with such ritualized lexis printed in red over the text of  the newspaper articles.

11

See the description of the scene of articulating the text in the example cited above from The Narodnaarmiya Daily.

442

Krassimir Stoyanov

Slogan ritualization implied placing the slogan in spatial and temporal positions in relation to the main text. Slogan corners were designed for special occasions such as anniversaries, jubilees, parties, other celebrations, and congresses. This topic was dealt with in a separate paper where such slogan structures were termed cascading clusters. They contained a succession of slogan phrases, arranged in a hierarchical order according to the growing significance of  the item, which was the object of acclamation. First came the event (usually a congress), then the leader, the party, the particular communist country, the USSR and the communist nations at the end. Such clusters often consisted of up to twelve phrases. Making use of ritualized slogan lexis attached particular values to events. One of  the first actions taken by the new totalitarian government was taking possession of  the ritualism that had been newly established. Moreover, in times of ideological conf lict the new government deprived Bulgarian political opposition groups of socially significant rituals as part of the process of isolating them from social life. In this way, the Bulgarian political opposition parties were unable to develop new or creative prospects for society. We contend that this is the reason why there were no slogans in the opposition papers. Rare exceptions were slogans calling for workers’ solidarity on 1 May; however, they were in harmony with the powerful propagandistic choir of  the of ficial press rather than being individual. The ritualism of  the slogans of  the ruling party further stressed the fact that there was a tradition, according to which all the papers had to publish special corners with the slogans of  the day. Apart from the structural and thematic characteristics of  the slogans presented so far, it is quite logical to ask the question: Who creates them? In our search for an answer there are two guiding points: Firstly, we have to take into account the fact that these are political texts, i.e. texts that are always the result of a particular ideology at a particular historical moment. Secondly, the pozivni (important written instructions from political parties) preserved in the archives of  the period (especially during the first days after the political change that took place on 9 September 1944) clearly and unequivocally reveal the patterns of verbal rhetoric used to gradually create slogans in the political life of  Bulgaria during the period in question.

Ritualized Slogan Lexis in the Bulgarian Press

443

No pozivna, telegram, report or instruction dating from 10 to 13 September 1944, and bearing the signature of a political functionary, ends without a slogan or a group of slogans (which later grow into ritualized ‘cascading clusters’).12 For instance, the FF Committee of  ‘Geo Milev’ District13 in the village of  Slatina, writing about the establishment of  the people’s new democratic power, sent this report to the city committee, ‘Today, on 9 September 1944 the FF Committee with its military groups at about 3.30 seized the 4th Police Station – Tsarigradsko shosse […]. From the district and the village we sent two imposing groups of demonstrators into town with posters, f lags and slogans’. ‘The first question, among many others, in this document is: ‘1. What should we do to the fascist of ficers, who did not allow the soldiers to take part in today’s demonstration under the pretext that they were tired?’ (State Archives 1964: 83). In the text of another document, Okrazhno Nr5 (a set of urgent guidelines to the whole district) of  the CC of the BCP to the party organizations in regard to the consolidation of  the democratic power of  the people, the participation in the war against Germany etc. of 12 September 1944, nearly all the collocations of the ritualized slogan lexis are encountered: ‘The fascist beast is hard-hit, but it is not completely murdered’; ‘the German-fascist gangs’; ‘the Balkan peoples’; ‘for friendly relations with the peoples of the USSR, England and America’ etc. Among the 21 tasks listed in the document, one is of particular importance: ‘Undertaking broad-based political work: organizing gatherings, conferences, meetings as well as holding manifestations and demonstrations for clarification and popularization of  FF slogans’. The document concludes with slogans that become patterns (Documents of  the Institute of  History of  BCP 1965: 338–345). The use of similar slogans by trade unionists and political leaders in other non-totalitarian countries was indicative of  the fact that such slogans were extremely popular at the time. For example, Новое время

12 13

See The Patriotic War of Bulgaria 1944–1945. Documents and materials. Sofia, 1978; Kostov, 1964; Dimitrov, 1969; also Proceedings of  the Institute of  History of  the BCP, Vol 23, Sofia, 1975. An area of  Sofia city.

444

Krassimir Stoyanov

[Novae Vremya]14 published an article by Jack Tunner (Chair of  the British Machine-building Trade Unions), on the occasion of the forthcoming trade union world conference in London in February, 1945, in which he concluded, ‘The representatives of  the three major allies have the responsible task to fulfil the important aim of  the international conference and to carry this principle […] like a torch of  the global rescue from the greed and aggressiveness of capitalism in all its forms in the future until economic and political freedom, and global justice for the world’s peoples is established!’ (NV 1944). This is a slogan par excellence due to the intensity of  the appeal and the stylistic emphasis on universal concepts at the time: a responsible task, an important aim, peoples’ torch, future, political freedom, global justice, people of the world. Another example is Will Walter’s article. Will Walter was chair of  the British Miners’ Trade Union. He ended his speech in a ritual manner, ‘May the unity between English and Soviet trade unions live! May the international unity of  trade unions live!’ (NV 1944). During this period the new political power in Bulgaria was consolidated through a new ritual processionality, which was used to transform the social mentality. Generally speaking, the new ritualism that was being created was represented as contrasting to the dark past – the years of  the monarchist regime that up until September 1944, had tied the Bulgarian state to Nazi Germany. As noted earlier, ritual marked significant events that were related to the crucial transition of  Bulgarian society during the period 1944–1947 under the conditions of ideological conf lict. Words, being the most expressive elements of the publicly performed verbal symbolism, were used as an instrument of power. This instrument was particularly expressive and active in the ritual type of culture, which was the case with the culture of totalitarianism. At this particular historic moment, the grandest social procedures were developed in all of Bulgaria’s contemporary history.

14

‘Новое время’ [Novae vremya] – New Times Magazine: A Soviet magazine for international politics, which was initially called ‘Народът и войната’ [The People and The War] and greatly inf luenced political life in East-European countries.

Ritualized Slogan Lexis in the Bulgarian Press

445

These ritual procedures were aimed at making a whole lexical layer salient in the national consciousness through invocation (both af firmative and condemning), chanting and glorifying exaltedness. This lexical layer served the purpose of a nearly sacred political and social expressiveness. This expressiveness had a special spatial and time structure in the ritual. The thus ritualized lexis can be defined as tub-thumping language in terms of its originally intended function. Its original function was to meet the needs of the tub-thumping behaviour of  the crowds in a state of ideological ecstasy. The aforementioned expressiveness was systematically translated from the oral into the written journalistic text, so it was transformed into a social pattern of  thinking and civil behaviour. In thematic and formal grammatical terms the lexis that was ritualized in this particular socio-historical situation in Bulgaria can be divided into three groups: • Denotations: names of social objects and phenomena, factors of change – names of political associations, state and power institutions, events related to them; • Verb constructions expressing existentialism – approving or suppressive, destructive; • Adjectives related to the topical political evaluation. These days, rhetorical slogans are used on a global scale. In Bulgaria, which is going through new transitions, this kind of ritual rhetoric has been brought into disrepute. Using slogans in political rallies is considered rather out of date, as it is associated with a past era – times of  totalitarianism, and a demonstrative and manipulative use of public language.

446

Krassimir Stoyanov

References Bayburin = Байбурин, Алберт. 1993. Ритуал в традиционной культуре. Наука, Санкт-Петербург, [Ritual in Traditional Culture, Nauka, St Petersburg, 1993]. Dimitrov = Димитров, Георги. 1969. �������������������������������������� Установяване и укрепване на народнодемократичната власт. Септември 1944 – май 1945. Сборник документи. Издателство на Българската академия на науките, Институт за история при БАН, София [Establishing and Consolidating People’s Democratic Power. September 1944 – May 1945, Collection of Documents, The Bulgarian Academy of  Sciences Press, Institute for History, Sofia, 1969]. Eliade = Елиаде, Мирча. 2008. The Sacred in the Secular World. In: сп. Философски алтернативи, София, бр. 5–6, стр. 3–12 [The Philosophical Alternatives Journal, Sofia, Nr 5–6, pp 3–12]. International Encyclopedia of  the Social Sciences. 1968. The McMillan Company & The Free Press. David L. Sills – Editor. Vol.13, pp 521. Kostov = Костов, Трайчо. 1964. Избрани статии, доклади, речи. Издателство на Българската комунистическа партия, София. [Selected Papers, Presentations, Speeches, The Bulgarian Communist Party Press, Sofia]. Levin, Beth. 1993. English Verb Classes and Alternations: A Preliminary Investigation. Chicago: The University of  Chicago Press. NV 1944 – Novae Vremya, New Times Magazine, Vol. 24, 15 December 1944. Proceedings of  the Institute of  History of  the BCP, Vol. 14, Sofia, 1965. Proceedings of  the Institute of  History of  the BCP, Vol. 23, Sofia, 1975. Proceedings of  the State Archives, Vol. 8, Sofia, 1964. Отечествената война на България 1944–1945. Документи и материали. Военно издателство, София, 1978 [The Patriotic War of  Bulgaria 1944–1945. Documents and materials. Voenno izdatelstvo, Sofia, 1978].

Kristine Horner and Melanie Wagner

Remembering World War II and Legitimating Luxembourgish as the National Language: Consensus or Conf lict?1

Abstract This chapter critically engages with attempts to portray the Nazi-German occupation (1940–1944) of  Luxembourg as the key turning point underpinning the shift in the categorization of  Luxembourgish from a ‘dialect’ to the ‘national language’, and it shows how the collective memory of World War II is being harnessed as a mechanism to validate contemporary language policies. To explore the disjuncture between narratives of consensus and contradictions experienced by social actors, examples from two sets of data are discussed: a) media representations linking the elevated status of  Luxembourgish to events during World War II and b) ethnographic interviews with a World War II letter writer focusing on language choice. By considering the interrelationship between the concepts of remembering and legitimating – as they are used in social history and sociolinguistics respectively – we show that complementary paradigms inform research in both disciplines and that narratives about language and history are intertwined.

1

We would like to thank Jean-Jacques Weber for constructive comments on this paper. We remain fully responsible for all opinions expressed and any remaining inaccuracies.

448

Kristine Horner and Melanie Wagner

1 Introduction With a geographical size of 2,586 square kilometres and a population of 502,066 (Statec 2010), the Grand Duchy of  Luxembourg is the second smallest member-state of  the European Union (EU). The language situation in Luxembourg is frequently referred to as triglossic in reference to the three languages recognized by the 1984 language law: Luxembourgish, French and German. The spoken/written distinction has been pivotal to understanding long-standing norms and patterns of  language use in Luxembourg, with most spoken communication taking place in Luxembourgish and written functions carried out primarily in standard French and/or German. Luxembourgish language varieties are Germanic and bear similarities to Moselle Franconian varieties spoken in parts of  Germany, Belgium and France; this relationship provides the rationale underpinning the decision for basic literacy skills to be taught via standard German in state schools. French is introduced as a subject in the second year of primary school, becomes a full subject in the third year and gradually replaces German as the main medium of instruction, particularly in the framework of the lycée classique or college preparatory secondary school. Based on the Education Act of 1843, state schooling institutions and practices have perpetuated elite bilingualism, or the valorization of standard written German plus French (Davis 1994). The addition of Luxembourgish to the curriculum as part of the Education Act of 1912 is often regarded as marking the of ficial recognition of a trilingual paradigm (Spizzo 1995). In relation to this point, parallel identification strategies with Luxembourgish and trilingualism – discursively linking two interrelated forms of  linguistic identification with national group membership – are anchored in key texts published in the 1910s (Horner 2007). However, it is the Nazi-German occupation of Luxembourg (1940–1944) that is frequently portrayed as the pivotal historical moment solidifying the recognition of  Luxembourgish as the national language. Despite the prominence of this dominant discourse, it must be stressed that the active promotion of Luxembourgish only began to accelerate from the 1970s onwards, culminating in the of ficial recognition of Luxembourgish

Remembering World War II and Legitimating Luxembourgish

449

as the national language in 1984 (Horner & Weber 2008: 71–76).2 In line with Fernand Braudel’s (1995: xxxviii) description of  history as ‘a science of  the present, and of a present which is ambiguous, at that’, this chapter explores discursive moves to construct consensus about historical events and lays bare tensions bound up with certain aspects of these attempts. We will analyse excerpts from two sets of data constituting a range of discourses: 1) media representations linking the rise of Luxembourgish to events during World War II and 2) ethnographic interviews with a World War II letter writer focusing on his language choices. First, we will discuss key symbols in relation to the discursive processes of collective memory.

2 Remembering World War II: Key symbols and national histories Social historical studies on memory have been shaped in large part by two paradigms: social psychology with the focus on the ways that social actors experience memory in relation to mourning, and political science with the focus on the institutionalization and politics of memory linked to nationbuilding. According to Ashplant et al. (2000: 7), there exists ‘a tendency to construe these two paradigms as if they were unrelated alternatives’, and linkages between bottom-up and top-down dynamics of memory are often left unexplored. Bridging this divide is the work of Jan Assmann (1995), who builds on Halbwachs’ (1950) conception of memory as inherently social or collective, and points to the intersecting processes of communicative and cultural memory. In the former case memory is constructed via forms of everyday interaction that enable shared experience, and in the latter case the construction of memory entails the (often state-sanctioned) institutional objectification of select manifestations of shared experience. 2

For example, many street signs were changed from German to French during the immediate post-World War II period, whereas bilingual French-Luxembourgish signs were gradually put up from the 1980s onwards.

450

Kristine Horner and Melanie Wagner

In relation to Assmann’s discussion of memory, it is also useful to consider Ortner’s account of  key symbols, which she describes as follows (1973: 1339): I am not assuming that there is only one key symbol to every culture; cultures are of course a product of the interplay of many basic orientations, some quite conf licting. But all of them will be expressed somewhere in the public system, because the public symbol system is ultimately the only source from which the natives themselves discover, rediscover, and transform their own culture, generation after generation.

Ortner describes these symbols as existing on a continuum, with summarizing symbols located at one end and elaborating symbols at the other. A summarizing symbol is perceived as embodying a matrix of interrelated values and is sometimes regarded as a sacred object, for example a f lag or a monument. Elaborating symbols consist of a combination of root metaphors (thought) and key scenarios (action); Ortner underlines the dif ficulty of drawing the line between thought and action. Because discourse provides a window to what otherwise would be unobservable phenomena, we only have access to the key scenarios as they are manifested in discursive (or other behavioural) patterns. Discourse constitutes a vehicle for remembering and representing historical events, thereby connecting the past and present as well as personal experiences and the processes of cultural memory. It is the selecting and foregrounding of certain events that shape the formation of patterns. Authoritative entextualization of  these patterns in key sites enables them to be transformed into key symbols (cf. Blommaert 2005, Gee 2005, Silverstein & Urban 1996), which are often appropriated as building blocks for historical narratives of the nation and underpin ideologies of nationalism conf lating nation and state (Grillo 1980). In his analysis of  the construction of  American national identity, Ricento (2003) discusses the ways in which key symbols are combined with key rhetorical strategies: forging ‘us’ versus ‘them’ boundaries, nationbuilding transformation scenarios and the perpetuation of nationhood. These strategies can be drawn upon to project an image of consensus and homogeneity and are not dissimilar from the processes described by Ortner from an anthropological perspective.

Remembering World War II and Legitimating Luxembourgish

451

Elaborating symbols are linked to cultural patterns, which in turn can be mapped out onto key scenarios, for example the elaborating symbol of  the gudde Lëtzebuerger ‘good Luxembourger’ who engaged in acts of resistance vis-à-vis Nazi-Germany during World War II, thereby leading the country from defeat to victory. Similar to White’s (1999) study on the memorial to the bombing of  Pearl Harbor, this chapter focuses on connections between ‘the “out there” of collective representation and the “in here” of personal cognition and emotion’ (1999: 505). We consider events during the Nazi-German occupation (1940–1944) that are often the focus of print media representations and remembered in everyday conversation among members of  the elderly generations. These entextualized historical narratives proliferate in multiple sites and underpin the present-day promotion of  Luxembourgish as the national language.

3 Media representations: Constructing consensus about Luxembourgish World War II is frequently thematized in the Luxembourgish print media, especially in the Luxemburger Wort (LW), which is the dominant newspaper on the national market. Media representations of events during World War II are often discursively linked to a broad range of other topics, in particular the promotion and legitimation of  the national language. Unsurprisingly, the circulation of  this kind of media discourse peaks on anniversaries marking milestone dates during World War II. Here, we will focus on the two events that function as a focal point of  World War II memory and are routinely represented as the most noteworthy forms of organized protest against Nazi-German occupation: the census of 10 October 1941 and the strikes on 31 August 1942.3 3

See Horner (2004) for a detailed discussion of the role of the print media in relation to Luxembourgish nation-building, with particular attention to discursive strategies linking language and nationhood.

452

Kristine Horner and Melanie Wagner

The following excerpt is a press release in the LW written by the president of  Actioun Lëtzebuergesch (AL), the non-profit organization whose goal it is fir alles anzetrieden, wat lëtzebuergesch as, apaart fir d’Sprooch ‘to promote the cause of everything that is Luxembourgish, especially the language’ (AL 1978: 63). The thirtieth anniversary celebration of the association was scheduled to coincide directly with the date marking the sixtieth anniversary of the Nazi-German census carried out in parts of Luxembourg on 10 October 1941, frequently referred to as dräimol Lëtzebuergesch:4 (1) Et war dem Comité vun der AL en Uleies, den 30. Gebuertsdag zesummefalen ze loosse mat dem 60. Joresdag vum Referendum vum 10. October 1941, wi de Gauleiter Gustav Simon de Lëtzebuerger mat engem Trick an d’Fieder wollt diktéieren, hir ‘Jetzige Staatszugehörigkeit’, hir ‘Muttersprache’ an hir ‘Volkszugehörigkeit’ wieren Däitsch. Mee Gott sei Dank hat de Gauleiter sech ferrem geschnidden: ganz massiv, anscheinend zu 97 Prozent, hunn d’Lëtzebuerger dräimol mat ‘lëtzebuergesch’ geäntwert an domat den Däitschen eng bei den Deckel ginn, sou dass de sou genannte Referendum mol net ausgezielt ginn ass. (Roby Zenner in Luxemburger Wort, 9 October 2001: 10) [It was a concern of the committee of AL to have the thirtieth birthday coincide with the sixtieth anniversary of  the referendum of 10 October 1941, when the Gauleiter Gustav Simon wanted to dictate to the Luxembourgers through a trick of the pen that their Jetzige Staatszugehörigkeit [current citizenship], their Muttersprache [mother tongue] and their Volkszugehörigkeit [ethnicity] were German. But thank God the Gauleiter was seriously mistaken: quite overwhelmingly, apparently with 97 per cent, the Luxembourgers answered with ‘Luxembourgish’ three times and therewith gave the Germans such a hard time that the so-called referendum could not be completed. (translation K.H.)]

4

Gauleiter Simon initiated a census of the Luxembourgish population which, among other things, included questions on Jetzige Staatsangehörigkeit ‘current nationality’, Muttersprache ‘mother tongue’, and Volkszugehörigkeit ‘ethnicity’. Many people answered these three questions with Lëtzebuergesch, thus symbolically resisting incorporation into the Third Reich, though the exact numbers of people who responded to the census and percentage of people who answered in the above way are matters of  historical debate. It is also worth noting that the annual Journée de la Commémoration nationale falls on the Sunday closest to the date of  the 1941 census.

Remembering World War II and Legitimating Luxembourgish

453

The ‘birth’ of AL is connected to the historic and symbolic event of dräimol Lëtzebuergesch, which often is regarded as the first collective manifestation of Luxembourgers asserting their will to independence. This discursive link may be viewed as an attempt to establish the historical continuity of  the development of Luxembourgish from dialect to language, and to connect it with the transition from occupation to independence. References to key historical events also serve to legitimate the activities of AL and to generate consensus as to what the organization represents. Luxembourgish resistance movements often are directly linked to the various strikes that took place on 31 August 1942 – normally referred to collectively as the Generalstreik ‘general strike’ – in protest against the forced conscription of  Luxembourgish citizens (born between 1920 and 1924) into the Nazi-German army. The sixtieth anniversary of the Generalstreik was remembered in multiple ways, for example the Musée d’Histoire de la Ville de Luxembourg organized an exhibit on World War II that took place from May to November 2002 (Musée d’Histoire de la Ville de Luxembourg 2002). Moreover, a great deal of media discourse focused on events related to the sixtieth anniversary of this series of related strikes. On 31 August 2002, the Luxemburger Wort (LW) appeared with a reprint of  the original front page of 31 August 1942, where the draft was of ficially announced. That edition of  the LW included a number of articles about the Generalstreik written by various people, including a two-page piece by the renowned Luxembourgish historian, Gilbert Trausch. In an editorial thematizing the sixtieth anniversary of the Generalstreik, editor in chief of  the Luxemburger Wort, Léon Zeches, weaves together a sequence of heroic events to obtain dramatic ef fect. He personifies the two nation-states, Luxembourg and Nazi-Germany, and maps them out onto the Biblical story of  David and Goliath, highlighting the uneven match between the two nation-states: (2) Das schutzlose kleine Luxemburger Land wird 1940 überfallen, die elementaren Freiheiten werden abgeschaf ft, die Presse wird gleichgeschaltet und zu Hitlers Propagandazwecken missbraucht, eine Volksbefragung soll den Luxemburgern das falsche Geständnis der deutschen Volkszugehörigkeit und Muttersprache entlocken, wird aber bereits im Vorfeld der of fiziellen Resultate für den Gauleiter und damit für Hitler und sein Nazideutschland zur Abfuhr und Blamage ohnegleichen, die

454

Kristine Horner and Melanie Wagner Rache folgt durch die Ankündigung der Zwangsrekrutierung mehrerer Jahrgänge der Luxemburger männlichen Jugend, Luxemburg bäumt sich wieder gegen den Goliath auf und streikt vom Süden bis in den Norden. (Léon Zeches in Luxemburger Wort, 31 August 2002: 3) [In 1940 the defenceless little country of Luxembourg is invaded; basic liberties are abolished; the press is brought into line and misused for Hitler’s propaganda aims; a census is meant to get the Luxembourgers to falsely admit to German citizenship and mother tongue, but even before of ficial results are out it turns into a major defeat and loss of  face for the Gauleiter and hence also for Hitler and his Nazi-Germany; revenge follows with the announcement of forced recruitment of several age-classes of young Luxembourgish men; Luxembourg rises up once more against Goliath and goes on strike, from the South to the North (translation K.H.)]

In addition to the aforementioned examples of media discourse, World War II memory is frequently evoked in AL’s monthly column in the LW entitled eng Klack fir eis Sprooch (the metaphorical bell ringing for ‘our language’). In excerpt 3, Lex Roth, the author of  the column, refers to the 1941 census and 1942 strikes and attributes the victory over Nazi-Germany to the headstrong and persistent Luxembourgish national character, as well as the presupposed Luxembourgish belief in the right to freedom and independence. In line with AL’s goals to promote the Luxembourgish language, Roth positions Luxembourgish as a key weapon in the battle against the enemy, who is denoted by the terms d’Preise and d’Nazi-Preise: (3) ‘Huel engem Vollek seng Geschicht; huel him seng Sprooch…da kanns du et an de Sak stiechen’, esou huet de Schrëftsteller Milan Kundera dee Gedanke vu Sprooch a Geschicht ausgesponn. Just dat wollten d’Preise vru 60 Jor mat äis maachen! […] Dat waren eis Elteren, eis Grousselteren, déi deemools séier bekäppt haten, wat d’NaziPreise mat äis wëlles haten! ‘Deemools’, dat war am Oktober 1941, an et gong géint déi verhuele ‘Personenstandsaufnahme’ mat deem éischten Akt vu Resistenz vun eisem klenge Vollek. Op deem huet zugudderlescht dee verbruete Wëll vun eisem Mini-Staat opgebaut, grad ewéi op dem Streik vun 1942 deem Lëtzebuerg seng Existenz als eegent, souveränt a respektéiert Land verdankt. (Actioun Lëtzebuergesh in Luxemburger Wort, 29 April 2000, eng KLACK fir eis Sprooch 60: 24) [‘Take away a people’s history, take away their language … then you can oppress them’, is how author Milan Kundera elaborates these thoughts about language and history. That is exactly what the d’Preise [Germans] wanted to do with us 60 years ago! […] It

Remembering World War II and Legitimating Luxembourgish

455

was our parents, our grandparents, who quickly understood what the d’Nazi-Preise [Nazi-Germans] wanted to do with us back then! ‘Back then’, that was in October 1941, and it was the first act of resistance by our small people, directed against this really horrible ‘census’. With that the stubborn will of our mini-state was built up, as well as with the strike of 1942, to which Luxembourg owes its existence as independent, sovereign and respected country. (translation K.H.)]

This use of the term Preis (literally, Prussian) dates back to the time when the Prussian garrison was installed in Luxembourg city (1815–1867) and this label has historically been used to construct ‘us’ versus ‘them’ group boundaries between Luxembourgers and their neighbours to the East. Diverging from the usage of  this term in Germany, Preis is used as a homogenizing label for Germans (the Other) in Luxembourg, and it is highly negatively charged when used in connection with or even alluding to events during World War II.5 As is the case in excerpts 1 and 2, a collective national past is constructed in excerpt 3, in which gudd Lëtzebuerger are discursively linked to acts of resistance against Nazi-German oppression; even the Zwangs­ rekrutierten ‘forced recruits’, who were drafted into the Nazi-German army, are held up as national heroes and distanced from the Germans, with the use of the label Preise underlining the distinctive boundary between national groups (i.e. between ‘them’ and ‘us’). The direct linkage between the Luxembourgish language and nation in the Luxemburger Wort works towards constructing consensus about the importance of the Luxembourgish language. The above excerpts illustrate strategies drawn upon to reinforce the status of  Luxembourgish as the national language by entextualizing its symbolic value in World War II memory. As will be shown in the following section, this discourse resonates with the experiences of individuals who were alive during World War II, including those who were forcibly conscripted into the Nazi-German army during the occupation of  Luxembourg.

5

Luxemburger Wörterbuch 1950–1977: (Däitsch) Preiss (pej.).

456

Kristine Horner and Melanie Wagner

4 Ethnographic interviews: Experiencing contradictions about Luxembourgish The data to be analysed here are part of a larger project on private literacy in Luxembourg during World War II and belong to a medium that thus far has been more or less ignored in Luxembourg, namely letters sent through the army postal service at that time. In addition to the analysis of the corpus of  letters, sociolinguistic interviews were conducted with the surviving writer, Roger,6 to establish a sociolinguistic profile and to gain further insight into the linguistic value system of the writer as well as his motivations for the linguistic choices made. These interviews, conducted in November and December 2007, were semi-structured and included mainly open-ended questions. During the interviews, which lasted around two and a half hours in total, the informant was asked about his literacy practices and language choices during World War II and today. Questions were asked about the informant’s family background, the learning and teaching of Luxembourgish at the time when the informant went to school, his language values and attitudes towards Luxembourgish and German and also about his attitudes towards writing in Luxembourgish.7 An analysis of the interview data reveals the psychological importance of  the Luxembourgish language as it clearly had a great deal of emotional value for Roger; for him it was the language of  the home, the language used for oral communication and for private written communication. He explains the language values attached to German and Luxembourgish in the following excerpt, in which he recounts his experiences as a Zwangsrekrutierter: (4) ech war dräi Joer dat éischt Joer do hat ech nach vill Lëtzebuerger ëm mech, am Arbeitsdienst do ware mir méi Lëtzebuerger wéi Preisen, an en halleft Joer an der Wehrmacht och nach, ma vun do waars de mutterseelenaleng ënner de Preisen, vun

6 7

The identity of  the informant has been anonymized. See Wagner & Davies (2009) for a detailed discussion of  this study, including the metalinguistic analysis of  letters written during the World War II period and interviews with the letter writers.

Remembering World War II and Legitimating Luxembourgish

457

do un ass näischt anescht geschwat gi wéi, an elo ech ka soen, ech hunn herno zum Schluss Preisesch gedreemt, well dat déi däitsch Sprooch war déi s de ëmmer ëm dech has, mee schreiwen an zemol heem schreiwen op Däitsch, dat war eng Kujenad do bass de der virkomm wéi e Friemen deen e Bittgesuch géing schreiwen oder ech wees net wat. Nee nee nee d’Sprooch war schon dat wat eis iwwer Wassser gehal huet an esou. (Interview with Roger, Diekirch, 12 April 2007) [I was [away] for three years and in the first year, I still had many Luxembourgers around me, during Arbeitsdienst [work duty in Germany] there were more of us Luxembourgers than Preisen [Germans], and for half a year in the Wehrmacht [German armed forces] too, but from then onwards you were all alone among the Preisen [Germans], from then onwards nothing other [than German] was spoken, and now I can say, in the end I dreamt in Preisesch [German], as it was the German language that you always had around you, but to write, and especially to write home in German was painful, you felt like a stranger who was writing a petition: no the language [Luxembourgish] was really what kept our heads above water. (translation M.W.)]

Here the Luxembourgish language is considered the language of closeness, which is the variety in which writers express emotions, feelings and worries, and in which they write to their loved ones, including family and friends (Koch & Oesterreicher 1985). It becomes clear that German had become part of  Roger’s everyday life, as he was always surrounded by Germans and the German language and as he even began to dream in German. On the other hand, Luxembourgish was the language that he related to home, the language he felt comfortable in and the language that he perceived as giving him strength. Similar to excerpt 3 from the Luxemburger Wort in the previous section, Roger refers to the Germans as Preisen (literally, Prussians) and to the German language as Preisesch, déi däitsch Sprooch and Däitsch. When German is spoken by Germans or when he seems to lack control over the language he uses, i.e. in his dreams, he refers to it as Preisesch, but when he uses it consciously, as for writing home, he refers to it as Däitsch or déi däitsch Sprooch, as if  to dif ferentiate between himself and the others, namely the Germans, and to distance himself  from them and the language of  the ‘Other’. Comparing the metalinguistic comments made by the informant throughout the interview to the comments in the letter corpus, one comes to realize that his positive emotional memory of the use of Luxembourgish

458

Kristine Horner and Melanie Wagner

during World War II is based on the way that it marks intimacy and ingroup solidarity, and this seems to have remained the case given the continued value attached to the language. But his strong association with the Luxembourgish language on the emotional level is not used to argue in favour of standardization. In fact the interview suggests that there exist potentially conf licting issues with regard to standardization. Even though the informant welcomes the fact that Luxembourgish has an of ficially recognized orthography, he still does not consider it much of an issue that many people make ‘mistakes’ in written Luxembourgish and he explains this by the fact that the older generations did not all learn how to write it at school: (5) et ass gutt datt mir elo eng Orthographie hunn, elo wann ech eppes schreiwen muss ech laang kalennere fir all d’Feeler ze verbesseren a wann ech der well 100 verbessert hunn, da sinn der nach dran […] nujee do ass et och et brauch ee sech netz e genéiere wann een e Feeler mecht, et huet een et net an der Schoul geléiert, an zumol mir Al. (Interview with Roger, Diekirch, 12 April 2007) [it is good that we have an orthography now, now when I have to write something I have to spend a lot of  time to correct all the mistakes and even if  I have corrected 100 of  them, there are still many more there […] well one need not feel uncomfortable if one makes a mistake, it wasn’t taught in school, especially not to us older generations. (translation M.W.)]

Of significance is the comment that Luxembourgish ‘now’ has an orthography. Over the course of the twentieth century there have been four orthographies of  Luxembourgish: 1914, 1946, 1975 and 1999. However, some Luxembourgish speakers only regard the most recent reform of 1999 as of ficial due to the shift in legal status from an Arrêté ministériel ‘Ministerial decree’ (1946 and 1975) to a règlement grand-ducal ‘Grand-Ducal bill’ (1999). Not all speakers are aware of  the latest reform; of  those who are, some seek to carefully adhere to the reformed spelling conventions, whereas others disregard them for various reasons (Newton 2002: 187). In line with the 1912 Education Act, Luxembourgish continues to be taught as a subject for one hour a week in primary school, and since 1946 it has also been taught in the first year of secondary school, with the goal being to teach students that d’Lëtzebuerger Land eng eege Sprooch a Literatur huet

Remembering World War II and Legitimating Luxembourgish

459

‘the country of Luxembourg has its own language and literature’ (MENFP 2007–2008). Materials for teaching Luxembourgish as a subject during the first year of secondary school and related learning outcomes only began to take a more concrete shape in the 1970s. In relation to the fact that Luxembourgish is still not used for a wide range of written functions by a large segment of its speakers, it is not fully bound up with the ritualized processes of standardization (cf. Deumert 2003). In fact, the Ministry of  Education conceptualizes the instruction of  Luxembourgish as a subject – taught to people who use it as a home language – as being incompatible with the standard language ideology (cf. Milroy & Milroy 1999): (6) Dës Aféierung soll de Kanner hëllefen, d’Lëtzebuerger Sprooch besser ze verstoen, se besser liesen ze kënnen an och vläicht hiirt Schreiwen ze verbesseren. D’Schreifweis duerf awer bei deene villen Dialekter, déi d’Kanner an de verschiddene Schoule schwätzen, net zu enger Diskriminatioun vun eenzelne Sproochgewunnéchte féieren a soll dofir och zimlech wéineg bei der Note zielen. (Ministère de l’Education nationale et de la Formation professionnelle. 2007–2008. Horaires et programmes. http://content. myschool.lu/ sites/horaires/2007–2008/html/es/branche/branche_LUXEM. html) [This introduction [to spelling] is meant to help the children understand the Luxembourgish language better, to be able to read it better and also, maybe to improve their writing. Spelling must not, however, lead to the discrimination of  the individual language practices linked to the many dialects that children in various schools speak and therefore should only count rather minimally towards the grade. (translation K.H.)]

This declaration of tolerance towards linguistic variation potentially stands in tension with the recent organization of  training programs for teachers of  Luxembourgish as a foreign language, together with a qualitative and quantitative increase in the production of pedagogical materials and dictionaries, as well as on-going debates about what constitutes ‘good’ and ‘bad’ Luxembourgish and more generally about ‘what to do with Luxembourgish?’ (Horner 2005). Whether or not Luxembourgish should be widely propagated as a written medium has become a greater point of national controversy over the past decade, especially the suggestion that this be made compulsory in the framework of  the educational system.

460

Kristine Horner and Melanie Wagner

Unlike the excerpts from the print media in section 3, the interview data in this section points to the controversial issue of standardizing Luxembourgish, rather than focusing solely on its in-group value, which is underpinned by the collective memory of World War II. Concerns about the future of Luxembourgish do appear in the Luxembourgish print media, although there is a tendency for media discourse to focus on Luxembourgish as potentially in danger rather than on controversies as to whether it should be standardized or not (Horner 2004).

5 Legitimating Luxembourgish as the national language: Consensus or conf lict? Luxembourgish was for the first time of ficially recognized as the national language and, in theory, as an administrative language in the language law of 1984. However, the law simultaneously reinforced the status quo by designating French and/or German as legal, judicial and administrative languages, precisely the state of af fairs prior to the ratification of  the law even if it was de facto rather than de jure policy. As Spolsky (2004: 26) points out, the use of  the terms ‘national language’ and ‘of ficial language’ is embedded in localized practices that must be considered when interpreting what they represent. The ratification of  the 1984 law signals a shift towards explicit language policy, thus imparting on Luxembourgish a higher position in the hierarchy of  languages. With Luxembourgish in position number one as the ‘national language’ as stated in Article 1 and no explicitly designated ‘of ficial language’ in the text of  the law, the wording provides a f lexible springboard for language ideological debates (Blommaert 1999), which frequently revolve around the status and use of  Luxembourgish.8 8

See Hof fmann (1987) for critical ref lections on the implications of the 1984 language law and its potentially divisive impact.

Remembering World War II and Legitimating Luxembourgish

461

Although it is not dif ficult to maintain a relative image of consensus regarding the symbolic value of (spoken) Luxembourgish among its speakers, the propagation of  Luxembourgish as a written language has proven a more controversial yet (seemingly) necessary step towards the international recognition of  Luxembourgish as a legitimate language of  the state (Bourdieu 1991). There is no consensus on what should be done with Luxembourgish in terms of standardization, yet pressure to establish normative awareness and related sociolinguistic practices is increasing. In particular, language ideological debates related to the increased levels of migration and shifting demographics intersect with those concerning language requirements and citizenship legislation (Horner 2009). The move to introduce formal language testing procedures for candidates applying for legal citizenship status has raised issues in terms of what shape the tests should take. The question of ‘what to do with Luxembourgish?’ lies at the heart of these debates as well as those about language-in-education policy in contemporary Luxembourg (Horner & Weber 2010). Just as the discursive processes of remembering World War II depend highly on elaborating symbols and key scenarios, so too do those of legitimating the national language. In these processes, the shift from occupation to independence is linked to the shift in categorization of  Luxembourgish from dialect to language. At the same time, we have seen how the contemporary legitimation of  Luxembourgish has opened up a Pandora’s box of issues concerning the sociolinguistic processes of standardization. Looking at bottom-up perspectives allows us to understand contentious issues that are not normally foregrounded in top-down discourses. Hence, the combination of multiple perspectives is essential for research on language policy as well as social historical studies on memory. Taking a critical approach to the foregrounding of  World War II in discourses legitimating the national language provides us with a more holistic viewpoint on the inextricable links between Luxembourgish social history and current language ideological debates.

462

Kristine Horner and Melanie Wagner

References Actioun Lëtzebuergesch (AL). 1978. ‘Statutten.’ Eis Sprooch 9, 63. Ashplant, T.G., Graham Dawson & Michael Roper. 2000. ‘The Politics of War Memory and Commemoration: Contexts, Structures and Dynamics.’ In: Ashplant, T.G., Graham Dawson & Michael Roper (eds), The Politics of War Memory and Commemoration. London: Routledge, 3–85. Assmann, Jan. 1995. ‘Collective Memory and Cultural Identity.’ New German Critique 65, 125–133. Blommaert, Jan. 1999. ‘The Debate is Open.’ In: Blommaert, Jan (ed.), Language Ideological Debates. Berlin: Mouton de Gruyter, 1–38. Blommaert, Jan. 2005. Discourse: A Critical Introduction. Cambridge: Cambridge University Press. Bourdieu, Pierre. 1991. Language and Symbolic Power. Cambridge: Harvard University Press. Braudel, Fernand. 1995. A History of  Civilizations. New York: Penguin. Davis, Kathryn Anne. 1994. Language Planning in Multilingual Contexts: Policies, Communities, and Schools in Luxembourg. Amsterdam: John Benjamins. Deumert, Ana. 2003. ‘Standard Languages as Civic Rituals – Theory and Examples.’ Sociolinguistica 17, 31–51. Gee, James Paul. 2005. An Introduction to Discourse Analysis: Theory and Method. London: Sage. Grillo, Ralph. 1980. ‘Introduction’. In: Grillo, Ralph (ed.), ‘Nation’ and ‘State’ in Europe: Anthropological Perspectives. London: Academic Press, 1–30. Halbwachs, Maurice. 1950. La mémoire collective. Paris: Presses Universitaires de France. Hof fmann, Fernand. 1987. ‘Letzebuergesch: Mundart und Nationalsprache: Sprachenpolitische und sprachensoziologische Überlegungen zum luxemburgischen Triglossie-Problem und zum Sprachgesetz von 1984.’ In: Brücher, Wolfgang & Peter Robert Franke (eds), Probleme von Grenzregionen: Das Beispiel SAAR-LOR-LUXRaum. Saarbrücken: SDV Saarbrücker Druckerei und Verlag, 49–65. Horner, Kristine. 2004. Negotiating the Language-Identity Link: Media Discourse and Nation-Building in Luxembourg. Unpublished doctoral dissertation, State University of  New York at Buf falo. Ann Arbor: UMI. Horner, Kristine. 2005. ‘Reimagining the Nation: Discourses of Language Purism in Luxembourg.’ In: Langer, Nils & Winifred V. Davies (eds), Linguistic Purism in the Germanic Languages. Berlin: Walter de Gruyter. 166–185.

Remembering World War II and Legitimating Luxembourgish

463

Horner, Kristine. 2007. ‘Language and Luxembourgish National Identity: Ideologies of Hybridity and Purity in the Past and Present.’ In: Elspaß, Stephan, Nils Langer, Joachim Scharloth & Wim Vandenbussche (eds), Germanic Language Histories from Below (1700–2000). Berlin: Walter de Gruyter, 363–378. Horner, Kristine. 2009. ‘Regimenting Language, Mobility and Citizenship in Luxembourg.’ In: Extra, Guus, Massimiliano Spotti & Piet van Avermaet (eds), Language Testing, Migration and Citizenship: Cross-National Perspectives on Integration Regimes. London: Continuum Press, 148–166. Horner, Kristine, & Jean-Jacques Weber. 2008. ‘The Language Situation in Luxembourg.’ Current Issues in Language Planning 9/1, 69–128. Horner, Kristine, & Jean-Jacques Weber. 2010. ‘Small Languages, Education and Citizenship: The Paradoxical Case of  Luxembourgish.’ International Journal of  the Sociology of  Language 205, 179–192. Luxemburger Wort. Actioun Lëtzebuergesch. 10. Mee 1940–2000, 29 April 2000, eng KLACK fir eis Sprooch 60: 24. Luxemburger Wort. Zenner, Roby. 30 Joer ‘Actioun Lëtzebuergesch – Eis Sprooch’ a 60 Joer Referendum, 9 October 2001, 10. Luxemburger Wort. Zeches, Léon. Auf  Amnesie baut man keine Zukunft, 31 August 2002, 3. Koch, Peter & Wulf  Oesterreicher. 1985. ‘Sprache der Nähe – Sprache der Distanz.’ Mündlichkeit und Schriftlichkeit im Spannungsfeld von Sprachtheorie und Sprachgeschichte.’ Romanistisches Jahrbuch 36, 15–43. Milroy, James & Lesley Milroy. 1999. Authority in Language: Investigating Standard English. London: Routledge. Ministère de l’Education nationale et de la formation professionnelle (MENFP). 2007–2008. Horaires et programmes. accessed 28 April 2011. Musée d’Histoire de la Ville de Luxembourg (ed.). 2002. … et wor alles net esou einfach: Questions sur le Luxembourg et la Deuxième Guerre. Luxembourg: Publications scientifiques du Musée d’Histoire de la Ville de Luxembourg. Newton, Gerald. 2002. ‘The Standardization of  Luxembourgish.’ In: Linn, Andrew R. & Nicola McLelland (eds), Standardization: Studies from the Germanic Languages. Amsterdam: John Benjamins, 179–190. Ortner, Sherry B. 1973. ‘On Key Symbols.’ American Anthropologist 75, 1338–1346. Ricento, Thomas. 2003. ‘The Discursive Construction of  Americanism.’ Discourse and Society 14/5, 611–637.

464

Kristine Horner and Melanie Wagner

Silverstein, Michael & Greg Urban. 1996. ‘The Natural History of Discourse.’ In: Silverstein, Michael & Greg Urban (eds), Natural Histories of Discourse. Chicago: Chicago University Press, 1–17. Spizzo, Daniel. 1995. La nation luxembourgeoise: Genèse et structure d’une identité. Paris: L’Harmattan. Spolsky, Bernard. 2004. Language Policy. Cambridge: Cambridge University Press. Statec. 2010. Le portail des statistiques: Grand-Duché de Luxembourg. accessed 28 April 2011. Wagner, Melanie & Winifred V. Davies. 2009. ‘The Role of World War II in the Development of  Luxembourgish as a National Language.’ In: Horner, Kristine (ed.), Special issue of  Language Problems and Language Planning 33/2, 112–131. White, Geof frey M. 1999. ‘Emotional Remembering: The Pragmatics of  National Memory.’ Ethos 27/4, 505–529.

Michela Giordano and Federica Falchi

John Stuart Mill and Salvatore Morelli: Language as a Social Tool in Nineteenth-Century Britain and Italy

Abstract A comparative analysis of  the process of women’s involvement in social, political and institutional contexts in Italy and in Britain calls to mind two nineteenth-century thinkers, Salvatore Morelli and John Stuart Mill. Although dif ferent in origin and education, they were inspired by the same desire to see women obtain the status of citizens already granted to men. An investigation of the two thinkers’ parliamentary addresses on the introduction of new laws to defend women’s rights will aim to pinpoint the similarities and diversities in their theoretical instruments and linguistic features. Despite the equally important political and social value of their individual battles to promote women’s social and political equality with men, the British and Italian parliaments and societies responded dif ferently to these pioneers of  feminine emancipation.

1 Introduction John Stuart Mill (1806–1873) and Salvatore Morelli (1824–1880) were two thinkers with very dif ferent cultural backgrounds who both actively worked for the recognition of women’s rights in Great Britain and Italy during the nineteenth century. However, the two men and their ideas did not gain the same fame or fortune because their respective national parliaments reacted in a diametrically opposite manner to these pioneers of women’s emancipation. Mill was a renowned intellectual and a Member of  Parliament respected by his colleagues and even by those who did not share

466

Michela Giordano and Federica Falchi

his ideas and theories. On the other hand, Morelli was scorned by both fellow Parliamentarians and the press, and even by some women involved in the emancipation process, who accused him of  thwarting their ef forts with his ‘revolutionary’ demands. The present study, carried out jointly by a historian and a linguist, analyses Mill and Morelli’s parliamentary speeches on women’s voting rights. Their contributions are investigated from the point of view of  the linguistic structures used to convey their messages, taking account of  the diverse importance and respect they held in their home countries, their dif ferent cultural and educational backgrounds, and above all, the dif ferent historical contexts and social and political settings in which these two men lived and worked. Our analysis of two parliamentary speeches that comprise the data for the present work explores the several dimensions of ‘discourse as social practice’ devised by Wodak (2001: 63–94). Discourse is considered as a way of signifying a specific domain of social practice: discourse and situational, institutional and social setting inf luence one another, making historical and ideological context a crucial notion in the analysis. Applying several discursive practices, language can both uphold and reproduce the social or political status quo, and at the same time, can contribute to its transformation, thereby inf luencing socially shared knowledge, attitudes, ideologies, and even historical processes. In this chapter, an introduction to the historical and philosophical background is followed by a linguistic analysis of  the two speeches. The excerpts in Italian, those by Morelli and those about him and his works, have been translated by the authors.

2 Historical background and political thought Our motivation for choosing to compare these two figures lies in the common principle that inspired both writers, originally formulated by the French philosopher Charles Fourier (1772–1837):

John Stuart Mill and Salvatore Morelli

467

Social progress and historic changes occur by virtue of the progress of women toward liberty, and decadence of  the social order occurs as the result of a decrease in the liberty of women. Other events inf luence these political changes, but there is no cause that produces social progress or decline as rapidly as change in the condition of women.1

While both Morelli and Mill took interest in various aspects of the societies in which they lived, they dedicated much of their intellectual and physical energy to the removal of what they saw as the greatest obstacle to liberty and democracy; namely, the exclusion of half of the population from public life. A factor common to both was the conviction that the question of  female emancipation was an integral part of  the rights of  humanity as a whole. In his key work, Morelli states that the rights of women are the rights of all human beings, ‘la causa della donna è la causa dell’umanità’ (Morelli 1869),2 while Mill not only writes ‘Men are so much accustomed to think of women only as women, that they forget to think of them as human’ (12 January 1871, in Robson 1963–1991, vol. XXIX: 37) but in an amendment to article 4 of  the Representation of  the People Bill in the House of  Commons in 1867 he also proposes that the word person should substitute man, in order to include women in the electorate. Both writers formed part of a group of intellectuals engaged in the debate on democracy, who admired the works of  Fourier, Saint Simon (1760–1825) and Owen (1771–1858), writers who firmly believed in democratic principles and who asserted that men and women should participate actively and equally in public and social life with no discrimination. These intellectuals came from countries widely diverse in terms of  their history, institutions, culture and economies, yet they worked together towards the goal of achieving parity between men and women. Besides Mill and 1

2

Charles Fourier, ‘Degradation of  Women in Civilization’, Théorie des Quatre Mouvements et des Destinées Générales (The Theory of the Four Movements and of the General Destinies), 3rd edn. Originally published in 1808, this edn 1841–1848, accessed 18 August 2011. Salvatore Morelli, La donna e la scienza, edited by Anna Maria Colaci, Pensa Multimedia, Lecce, 2008, is the reprint of the third edition of Morelli’s text of 1869 La donna e la scienza o la soluzione del problema sociale.

468

Michela Giordano and Federica Falchi

Morelli, other members of this group included Legouvé (1807–1903) and Leroux (1797–1871) in France, the Englishmen Bentham (1748–1832) and Thompson (1775–1833), and, in Italy, Mazzini (1805–1872) and Garibaldi (1807–1882). The Englishman and the Italian came from highly contrasting educational and social backgrounds. Indeed, if we consider that Morelli’s intellectual development took place during the twelve years he spent in prison, their qualitative dif ferences and the chronology of their lives is fairly striking. While Mill’s political career lasted only one legislature, from 1865 to 1868, Morelli’s actually continued for four legislatures, from 1867 to 1880, notwithstanding the backwardness of Italy in general and in particular of  the southern region where he was from. During his years in Parliament, Morelli tirelessly presented Bills to promote the cause of women’s rights, encouraged by the commitment of  bright and courageous women such as Cristina di Belgioioso (1808–1871), Jessie White Mario (1832–1906), Anita Garibaldi (1820–1849), and enlivened by the intellectual support of  Mazzini, by whom he was inspired. Mill was aware of Morelli’s political activity and spoke highly of him after Morelli informed him by letter that he was drawing up a parliamentary Bill demanding full legal recognition for the status of women. Morelli’s Bill never even made it into Parliament in 1867 because his fellow deputies claimed that it was not even worthy of discussion and would not admit it.3 Although Mill had not yet received any drafts of these Bills, he immediately expressed solidarity in a letter to Morelli (1 September 1867): I shall read them with great interest, because the issues involved here, and which you refer to in your letter, THE FREEDOM OF CONSCIENCE, THE FREEDOM OF EDUCATION, AND THE EQUAL LEGAL STATUS OF WOMEN are matters that I hold dear more than any other social or political question. (Morelli 1869) 3

Professor Estival, who wrote the Introduction to the 1869 edition of  La donna e la scienza, underlines that none of Morelli’s colleagues, even those belonging to his political party, endorsed his proposal, leaving him alone: ‘Morelli fu solo in Parlamento nel sostenere la sua idea eminentemente civilizzatrice!!!’ (Morelli was alone in furthering his own eminently civilizing idea in Parliament!!!), La donna e la scienza, 1869.

John Stuart Mill and Salvatore Morelli

469

Both men produced important texts, The Subjection of  Women and La donna e la scienza, refuting all the entrenched prejudices on the intellectual capabilities of women, and proposing a series of reforms required not only for the full recognition of women’s participation in society, but also necessary for the progress of  humankind. The definitive versions of both works appeared in the same year, 1869, and Morelli received a copy of  Mill’s work in French directly from the author (he cites it in La donna e la scienza). Although both works dealt with the same theme and had much in common, they by no means had the same publishing success. Firstly, the two men’s reputations in intellectual circles were not comparable; most significantly perhaps, Mill’s style and form of writing was more f lowing and accessible, while Morelli’s prose is considered ‘prosa enfatica e pesante’ (Conti Odorisio 1992: 42), wordy and plodding. Mill was also slightly more moderate than Morelli and tried to be strategic and realistic in his planning; for example, he took the decision to publish his work, written originally in the early years of  the 1860s, at a time when he knew there would be greater likelihood and feasibility that suf frage would be extended, following his numerous addresses to Parliament. Morelli, on the other hand, presented drafts of  Bills that were in retrospect ahead of their time, and today give us the impression he was an early pioneer. Yet this was the very reason why his actions were actually damaging to the cause, as many emancipationists claimed. One of  the main themes common to both writers’ works was the conviction that the traditional structure of  the family was backward and continued to perpetrate and perpetuate a patriarchal model of society where the rights of women were not recognized. Women lived in subordination within the family sphere and union in marriage was often motivated by convenience rather than sentiment. Indeed, Morelli firmly believed that the family as such went beyond any legal bond and that, even following a divorce, the family nonetheless continued to exist. Both men campaigned actively in favour of divorce and against legalized state prostitution, which denied basic rights of freedom to women and placed no restrictions or responsibility on men. In addition, both fought for the abolition of laws that prevented women from managing their own wealth autonomously. Both Mill and Morelli strongly believed that women

470

Michela Giordano and Federica Falchi

should have the right to vote, and were convinced that this was the first reform needed because as Mill said in a letter to Florence Nightingale on 31 December 1867: I think, firstly, that political power is the only security against every form of oppression; secondly, that at the present day in England it would be easier to attain political rights for such women as have the same claims as enfranchised men, than to obtain any other considerable reform in the position of women; thirdly, I see no danger of party spirit running high between men and women and no possibility of its making things worse than they are if it did. (Mill 1867, in Robson 1963–1991, vol. XVI: 333)

Morelli was most vehement in attacking those men (and also women emancipationists) who contended that women were too immature to be given the right to vote: It is the f lawed argument of all despots. When urged to recognise the liberty and other basic rights of people, they exclaim: they are not mature enough! But I say, precisely because they wish to cultivate their maturity, we need to sanction this desire by fully according them such rights. (Morelli 1869)

In addition to proposing legislative reforms, Morelli encouraged women to fight for their own rights: ‘Dear Ladies, the world belongs to those who take hold of it – if you want your rights to be legally recognised, you have to demand them with force’ (Morelli 1869). Regarding the issue of education, there were some divergences between the two thinkers. Mill maintained that there should be free choice in education and teaching while Morelli argued strongly in favour of  lay teaching and against what he saw as excessive church interference in schools. Morelli went as far as to suggest that primary schools should be gradually abolished and that women should be assigned the task of educating good citizens. Mill, on the other hand, did not think it appropriate that the education of children be entrusted to mothers alone, because it was a task to be performed by teachers assigned this specific duty. Where this was not possible, mothers could be given this responsibility, but not because they had any particular teaching vocation as such. As already mentioned, one dif ference between the two authors was in their respective styles of writing: Morelli’s style was verbose and Mill’s

John Stuart Mill and Salvatore Morelli

471

more f luid and readable. Yet another dif ference was in their definitions of specific characteristics of the nature of women, which Mill claimed could not be described objectively, but were inf luenced by environment and education (Pichetto 1996: 50). Although Morelli refuted the traditionally accepted model for women, he apparently contradicted himself by insisting that women needed to be educated precisely so that they could fulfil this ‘traditional’ function of educators of  the young. Neither author disputed the idea that there should be a clear division of labour in the house, which was probably due to their chauvinistic cultural conditioning. Morelli was very probably inf luenced by the ideas of the Genoese philosopher Mazzini, who, in his work ‘Doveri dell’uomo’ explains that ‘like two distinct branches springing from the same trunk, man and woman are varieties springing from the common basis – Humanity’ (‘come due rami che muovono distinti da uno stesso tronco, l’uomo e la donna muovono, varietà, da una base comune, che è l’umanità’) (Mazzini 1860 in Mazzini 1875: 71). This seems to be an early reference to a dif ference of function but not of nature between men and women. This contrast led Morelli to make the uncharitable assertion in 1876 that Mill’s interest in women’s rights was limited to campaigning for the right to vote, ignoring other aspects. According to Morelli for the great Englishman the ultimate truth of  law was a guiding principle and he called for the law to assign to women what rightly belonged to them. As for my own thinking, I have always recognised and acknowledged woman’s essential role in generating mankind, in educating and accompanying him from the cradle to the grave; I have thus not only always argued for the necessity of recognising women’s rights but have likewise asserted that modern society has supreme need to educate and enlighten the minds of women, thereby giving them the pedagogical tools for self development and for the advancement of  humanity. (Morelli 1876, cited in Conti Odorisio 1990: 90–91)

Notwithstanding their evident divergences, both men were without doubt sincere and courageous supporters of  female emancipation in both the political and social arena, and dedicated a notable part of  their intellectual and parliamentary life to this cause. They also lent their support to the emancipationist cause at an international level, and were involved in

472

Michela Giordano and Federica Falchi

numerous related initiatives, such as the battle for legalizing divorce, for the legal right to establish paternity, without which the burden of looking after illegitimate children fell entirely on the mother; access to all kinds of employment; the right to administer personal wealth, the abolition of state prostitution and of course voting rights in political and administrative elections.

3 Theoretical framework and data The general theoretical framework within which the parliamentary speeches have been analysed is that of  Critical Discourse Analysis or CDA (Fairclough; Wodak; Van Dijk), and in particular the discourse-historical model devised by Wodak in Wodak and Meyer (2001) and the analysis of parliamentary debates as suggested by Van Dijk (1992, 2000). Despite some dif ferences between the various theoretical approaches within the abovementioned framework, a common tenet is the centrality of  language as a means of social construction, a form of social practice. Discourse is inf luenced by society and it can itself have a major impact on society: discursive practices produce and reproduce unequal power relations between social classes, genders, ethnic minorities, and peoples. Some of  the characteristics of  the discourse-historical model adapt perfectly to the kind of analysis which is intended here. The approach is interdisciplinary, both historical and linguistic; multiple genres can be studied and the historical context is integrated in the interpretation of discourses and texts. This approach calls for investigation of  the discourses of discrimination involving racist, ethnic, class or gender bias or prejudices (Wodak 2001: 69–70). Applying it to our study, the general discourse or macro-topic would be female emancipation or discrimination against women which allows for many sub-topics such as women’s political representation, suf frage, family, women and children’s education, and women’s work.

John Stuart Mill and Salvatore Morelli

473

Discourse is framed into dif ferent fields of action as segments of ‘societal reality’. Within the ambit of political discourse lie dif ferent fields of action, such as legislation, parties and political advertising and propaganda, and so on. The field of action in this case is that of  ‘law-making political procedures’, developed through dif ferent discursive genres such as laws, bills, amendments, regulations, and the speeches and contributions of  MPs. A genre is characterized as the conventionalized and schematically fixed use of  language associated with a particular activity (Wodak 2001: 66) and a way of acting and interacting linguistically, which structure texts in specific ways (Fairclough 2003: 17). Parliamentary speeches and contributions of MPs present many structures at dif ferent levels. According to Van Dijk (2000: 85–103), relevant discourse categories and identifying properties of parliamentary debates include semantic structures (topics, local coherence, implicitness, vagueness, intertextuality), lexical features, clause and sentence syntax, interactional strategies (phonology, intonation, nonverbal interaction), rhetoric, and global structures (schemata, argumentation structures and topoi). Wodak argues that the various forms of discrimination can be discussed by means of argumentation strategies or topoi of which she provides a list (Wodak 2001: 73–77) against which Mill’s and Morelli’s speeches are analysed here. The present study focuses on a) the topoi, themes or motifs used to argue in favour of women’s suf frage and emancipation or against its opponents, and b) the lexicalization (Van Dijk 2000: 95) of  the conceptual meaning of ‘woman’ i.e. the choice of words used to identify women and to describe their actions and properties, a choice that implies evaluations. The data consists of  Mill’s speech ‘The admission of women to the electoral suf frage’ delivered on 20 May 1867, the introductory speech to the amendment to substitute the word person for the word man in Clause Four of  the Reform Bill which dealt with the qualification of electors by occupation; and Morelli’s speech ‘Riforme legislative proposte al Parlamento italiano dal deputato Salvatore Morelli, per assicurare con nuove guarentigie giuridiche la sorte dei fanciulli e delle donne’, delivered on 18 June 1875, on the presentation and reading of nine Bills which summed up Morelli’s modern and progressive ideas regarding voting rights for women, family law, marriage, the possibility to divorce, and children’s education.

474

Michela Giordano and Federica Falchi

Morelli was not f lustered by his fellow deputies’ negative response and stonewalling attitude towards him and his previous four-article proposal of 1867 which, in some aspects, constitutes the basis of  his 1875 speech.

4 Analysis and discussion Parliamentary speeches are typically argumentative. MPs put forward their proposals or motions for debate, and following the introductory speech and reading of the proposed bills, they have to be ready to defend their ideas and to refute the opposing opinions and counter-arguments. The traditional abstract categories of argumentation, such as premises (or grounds) and conclusions, warrants and backings (Toulmin 1958; Fairclough 2003) can be frequently found in parliamentary speeches and explained in terms of semantic and schematic superstructural categories, organized hierarchically and producing the global ordering of  the argumentative discourse (Van Dijk 1992: 245–246). Nevertheless, Fairclough himself (2003: 82) states that an analysis based on these categories only is very abstract, because it represents the logical structure of the main arguments but not the texturing of  the arguments, which is not the way in which they are developed in the text. As explained by Wodak (2001: 74), within argumentation theory, topoi (i.e. cultural stereotypes) or loci communes can be described as parts of argumentation which belong to the premises; they are the contentrelated warrants which connect the arguments with the conclusions or claims. In other words, they justify the transition from the argument to the conclusion. As Van Dijk explains, they represent the contents and arguments adduced to argue for or against an issue: they stand for the ‘common sense reasoning typical for specific issues’ (2000: 97). The analysis of  the two parliamentary speeches will be carried out against the background of some of  the topoi listed by Wodak (2001: 74) and in particular usefulness and advantage, uselessness and disadvantage, danger and threat, law and right, justice, and reality.

John Stuart Mill and Salvatore Morelli

475

John Stuart Mill was a successful speaker in Parliament, a talented orator able to use the best selection of words and a very formal and elegant style, both clear and coherent at the same time. He was perfectly aware of  the possibilities of fered by language, linguistic expressions and logical argumentation. He was described as having a ‘thin, fragile, awkward frame’ with a ‘nervous twitching of the lips and eyes’, a ‘weak voice and a sort of stammer’, often interrupted and coughed and laughed into silence (McCarthy 1872). Nevertheless he managed to express his ideas and thoughts openly and clearly, becoming a famous writer and philosopher. On the other hand, Morelli was not only laughed at and interrupted, but also not even allowed to speak: he was not held in such high consideration as his counterpart in England. His parliamentary speeches, in particular the one in 1875, were often interrupted by laughter or ironical comments such as ‘Bravo!’ Even if Morelli’s language is considered old-fashioned today because of some of  the structures used which are strikingly formal and outdated, the present paper will argue that those structures and the lexis are loaded with meaning and implications, charged with associative significance. Both speeches’ main force lies in their overall structure and in their pattern of argumentation. The English philosopher structures his explanations by anticipating what the objections to his proposal may be and answers the possible questions that might arise. According to Mill, the issue of  the extension of suf frage is a simple question, ‘whether there is any adequate justification for continuing to exclude an entire half of  the community, not only from admission but from the likelihood of being ever embraced within the pale of the Constitution’. This can be considered the key claim of  his argument, exploited through a rhetorical question that clearly anticipates his conclusions, that there is no adequate justification for excluding women from suf frage, and that ‘we ought not deny to them, what we are conceding to everybody else’. Mill begins by discussing the tame character of the political and juridical action involved. The extension of the suf frage would not entail high risk, harm or loss for party or class feelings, and would have only the ‘smallest tendency’ to ‘disturb the balance of political power’. The topos of danger or threat is based on the following condition: if a political action or decision bears specific dangerous, threatening consequences, one should not perform it, or one should do something against them (Wodak 2001: 75).

476

Michela Giordano and Federica Falchi

Morelli poses a question to his fellow deputies: ‘Temete forse che le mie proposte disturbino l’ordine della famiglia, spostandone la missione o conducendo la donna sulla via della dissolutezza?’ (‘Are you afraid that my proposals will disturb family order and lead women to debauchery?’). He himself provides the answer by contending how opponents might advance this argument, but not his listeners! He thus uses irony to af firm that their intellect cannot (or should not) possibly perceive his proposals as threatening or dangerous. Mill poses a similar question: Or is it feared that if  they were admitted to the suf frage they would revolutionize the State, would deprive us of any of our valued institutions, or that we should have worse laws, or be in any way whatever worse governed, through the ef fect of  their suf frages?

This is followed by his consideration ‘No one, Sir, believes anything of the kind’, indirectly, and ironically implying that nobody could really oppose his proposal by reason of an unjustified and irrational fear. The resistance to the ideas and concepts Morelli had been trying to present in Parliament in the previous few years had been a ‘tempesta incessante di ridicole opposizioni’ (‘a never-ending stream of ridiculous opposition’), which led him to compare himself to Galileo, condemned for heresy following his scientific discoveries in the seventeenth century when society was not ready for his innovative theories. However immodest and arrogant he may have been considered for making such a comparison, another strong similarity between the two politicians’ ideas can be found here: the suggestion that a scientific method should be used when dealing with ethical and social matters, not leaving anything to tradition and customs. Science and reason must be employed to help discredit the ‘anacronismo omicida, antieconomico ed immorale che si chiama sistema di repressione’ referring to an Italian society characterized by repressive ‘homicidal, anti-economic and immoral anachronisms’. Here the topos of threat, or menace caused by innovative and progressive ideas is merged with the topos of reality. Instead of enumerating the practical arguments in favour of the extension of suf frage to women, Mill asks ‘what are the practical objections?’ and through an explicit ‘protagonist-antagonist’ organization of his speech

John Stuart Mill and Salvatore Morelli

477

(Fairclough 2003: 82), the fears and concerns of the antagonists of women’s suf frage (other MPs, but also British society and its conventions) are placated by the counter-arguments of  the protagonist. Mill opines that no sensible or realistic objections can exist; instead, the issue of women’s rights evokes a ‘strange feeling’ due to the novelty, or rather to the apparent novelty of  the issue, since, as he asserts, ‘it is a very old proposal’. The topos of reality (Wodak 2001: 76) can be paraphrased saying that ‘because reality is as it is, a specific action should be performed’. Customs and habits cannot be the judge any more. Reason must be the judge. Social, economic and political realities have changed, therefore the law must also be changed. Morelli basically has the same feeling about the conventions followed by the firm and resolute legislators, conventions which go against nature and reason, habits which no longer fit, and which consequently damage and impede human progress: ‘[…] niente più nuoce al progresso umano della resistenza che i legislatori oppongono alla ragione per rispetto ad un convenzionalismo diametralmente opposto ai fini supremi della natura’. He opens his speech referring back to his proposals of 1867 on women’s emancipation and tries to find an explanation for its rebuttal at the time. He hypothesizes that the fear and distrust brought about by ‘new things’ i.e. ‘la dif fidenza che suscitano le cose nuove’, may have been the cause of  the refusal. On that occasion, he had asked for new guarantees (‘guarentigie’) to be granted to Italian women and children. Later in his speech, Mill states that there is an important branch of expediency which is called justice; and justice requires that nobody can or should ‘capriciously and without cause’ withhold from one or deny to anyone what is given to another. The topos of justice (Wodak 2001: 75) is based on the principle of  ‘equal rights for all’. If persons, actions or situations are equal in specific respects, they should be treated or dealt with in the same way. The topos of justice is employed by Morelli as well: ‘E se vi propongo che le donne sieno ammesse all’esercizio elettorale, la mia esigenza è suf fragata da un sentimento di giustizia …’ (‘And if  I suggest to you that women must be granted voting rights, I am motivated by a strong sense of justice …’), since a similar proposal had already been supported in the British Parliament by 152 votes. In his previous four-article proposal

478

Michela Giordano and Federica Falchi

of 18 June 1867 when he tried to put forward the ‘disegno di legge’, or bill (the reading of which was not allowed) for the legal and juridical reintegration of women, he had proposed the recognition of equal opportunities and civil and political rights for women. Since women and men have the same (mental or intellectual) abilities, justice demanded the same civil and political rights for both. According to Mill, the possible objections to women’s suf frage include the following: politics is not women’s business, women do not desire suf frage, they are suf ficiently represented by their male relatives, and they already have enough power. He outlines his opposition to these four statements, stressing their futility and senselessness, denying their objective truth and advancing real and veritable facts. These objections imply that his proposal has no practical value, and is ef fectively worthless and useless. Thus, utilizing the topos of uselessness and disadvantage (Wodak 2001: 74), this time Mill states that ‘[t]he ordinary occupations of most women are, and are likely to remain, principally domestic’ and that, even if  he averts an obscure feeling ‘as if women had no right to care about anything, except how they may be the most useful and devoted servants of some man’, he ironically states once again his conviction that no member of  the House of  Commons could harbour such mean feelings towards women. The counter-argument which Mill opposes to this objection is based on the topos of advantage and usefulness, which af firms that ‘if an action under a specific relevant point of view will be useful, then one should perform it’ (Wodak 2001: 74). The political revolution is preceded by a ‘silent domestic revolution’ and men and women are each other’s companion. Some may believe that women do not need direct power, because they already have so much indirect power: Mill explains how that power, being indirect, is in fact irresponsible and so making it a ‘responsible power’ by awakening in women their ‘political point of  honour’ ‘with the acknowledged right to a voice’, would ensure ‘a sense of corresponding duty’. Morelli as well makes use of  the topos of advantage and usefulness when he asserts that ‘Il Parlamento Italiano ha l’obbligo di organizzare il nuovo potere facendo della donna con la scienza e la libertà, la forza illuminatrice ed attiva che prepara nel seno della famiglia, mercè l’istruzione ed il lavoro, la moralità ed il benessere economico, da quali emerge durevole e senza scosse il vero progresso sociale e politico della nazione.

John Stuart Mill and Salvatore Morelli

479

[the Italian Parliament ought to provide all women with an education and liberty, so that they can become an active and inspiring strength, able to prepare, within the family, the morality and economic welfare which form the roots of  lasting social and political progress.]

The topos of advantage and usefulness had already been used in Morelli’s first proposal in 1867, in particular in Article 4, dealing with maternity as a national value, which embodies one of  his deepest and most fascinating insights. Italian women very diligently participate in the improvement and progress of  the human race, they are ‘più diligenti al miglioramento della razza umana’, because they provide their home country with beautiful and healthy and robust children, bringing them up to become heroes and accomplished thinkers, ‘eroi, pensatori e produttori distinti’. Using language loaded with sentiment and emotion, Morelli puts forward the then innovative concept that the precious and invaluable service of maternity which women of fer to their home country should be recognized with the award of  honours, voting rights, and the granting of access to prominent public positions or of fices and pensions still denied to the female sex. Advocating the need for and validity of the guarantees he is asking for women, for example those protecting maternity and those which represent ‘mezzi di svolgimento per raggiungere il fine supremo del miglioramento della famiglia e della società’, i.e. the means to reach the ultimate goal of  the improvement of  family and society, in the conclusion of  his speech Morelli employs the topos of  law and right which can be paraphrased as follows: ‘if a law or an otherwise codified norm prescribes or forbids a specific politico-administrative action, the action has to be performed or omitted’ (Wodak 2001: 76). On a strictly local level, the texts were scrutinized for the conceptual meanings of  ‘woman’, i.e. the words used to identify women and their actions. According to Van Dijk (2001: 103) local meanings are not only the results of  the selection made by speakers in their mental models of events or their more general, socially shared beliefs, but also denote information that most directly inf luences the opinions and attitudes of recipients, and hence will have social consequences. In Mill’s speech, women are not only schoolmistresses or governesses or heads of families, not even just the ‘petted children of society’. They are the ‘most useful and devoted servants of man’, ‘an entire half of the community’.

480

Michela Giordano and Federica Falchi

Women are those who are ‘principally charged with the moral education of the future generations’, those ‘whose chief daily business is the judicious laying-out of money’. A woman is a ‘man’s habitual society’, ‘his chief associate’, his ‘most confidential friend’, his ‘most trusted adviser’, his ‘nearest companion’ and ‘a stimulus and a support to his highest aspirations’. According to Morelli, a woman is the noblest being, ‘essere nobilissimo’. Even if often ‘madri schiave’, mothers treated as servants, they are those who can lead human education and spirit to ‘verità e virtù’ or to ‘pregiudizio e vizio’, either to truth and virtue or prejudice and vice. Woman is the primary force of  the process towards enlightened civilization, ‘la prima forza del nuovo incivilimento’. If she is not allowed to study and take part in social and political life, she can become the unwitting and inadvertent cause of our malaise, ‘inconscia causa ef ficiente dei mali che ci contristano’. The saying ‘Della donna nelle dita è il telaio della vita’ used in the speech explains how life and the loom of life lie in women’s fingers, in women’s hands. Using rhyme here, Morelli gives evidence of his linguistic and rhetorical awareness in delivering his speeches. For both thinkers, women are not child-bearers, but those who prepare them for social and political life, who lead to success those same men who are denying them the rights to vote.

5 Conclusions A common interest and active participation in the battle for democracy, fought by a handful of courageous intellectuals in Europe, brought together these two men of vastly dif ferent backgrounds and education. Although Mill was a man of greater learning and famed in international circles, it is undoubtedly the case that Morelli greatly contributed to the issue of women’s emancipation, even though he did not possess Mill’s cultural and intellectual weight and fame acquired for his numerous and heterogeneous works. If a comparison between the two could seem ungenerous in the first place, due to this wide gap in their quantitative and most of all qualitative

John Stuart Mill and Salvatore Morelli

481

written output, Morelli’s commitment and long-range perspective within the specific issue of women’s rights cannot be denied. A comparison not only of  the deep insights but also of  the superficial linguistic forms used to convey those contents, is therefore not only possible but appropriate and even enlightening. Mill was undeniably a more accomplished thinker and speaker due to his formal and elegant written style, and his f lowing, accessible, clear and coherent way of delivering speeches. He was fully aware of the possibilities of fered by linguistic expression and logical argumentation. On the other hand, Morelli, often forgotten and disregarded by both his countrymen and historians, had to cope with scarce interest in his thoughts, and his old-fashioned language and too formal and outdated structures contributed to the opinion that his style was heavy, prolix and plodding at times, especially in his written works. Surprisingly, our linguistic analysis of  his spoken, or written-to-be-spoken words, showed that he was just as aware as his colleague overseas of the importance of language as a social tool to put forward proposals for a change, and was able to employ and make conscious and careful use of the power of argumentation techniques, lexical meanings and topoi inextricably linked to the issue of women’s emancipation.

References Conti Odorisio, Ginevra. 1990. Salvatore Morelli: politica e questione femminile. Roma: L’Ed. Conti Odorisio, Ginevra. 1992. Salvatore Morelli (1824–1880): emancipazionismo e democrazia nell’Ottocento europeo. Napoli: Edizioni Scientifiche Italiane. Fairclough, Norman. 2003. Analysing Discourse. Textual analysis for social research. London: Routledge. Mazzini, Giuseppe. 1875. Doveri dell’uomo. Roma: Tip. Alle Terme Diocleziane. McCarthy, Justin. 1872. ‘John Stuart Mill as a speaker in Parliament. A contemporary view’, In: McCarthy, Justin, ‘The Liberal Triumvirate of England.’ Modern Leaders: Being a Series of Biographical Sketches. New York: Sheldon & Company. 106–115. accessed 18 August 2011.

482

Michela Giordano and Federica Falchi

Morelli, Salvatore. 2008 [1869]. La donna e la scienza. ed. by Anna Maria Colaci Lecce: Pensa Multimedia Editore. [Originally published as La donna e la scienza o la soluzione del problema sociale]. Pichetto, Maria Teresa. 1996. John Stuart Mill. Un classico del pensiero liberal-democratico. Milano: FrancoAngeli. Robson, John M. (ed.). 1963–1991. Collected Works of John Stuart Mill. Toronto: University of Toronto Press, London: Routledge and Kegan Paul. 33 vols. accessed 18 August 2011. Toulmin, Stephen E. 1958. The Uses of  Argument. Oxford: Oxford University Press. Van Dijk, Teun. A. 1992. ‘Racism and argumentation: Race Riot Rhetoric in Tabloid Editorials.’ In: van Eemeren, Frans H., et al. (eds), Argumentation illuminated. Dordrecht: Foris, 243–257. Van Dijk, Teun A. 2000. ‘On the analysis of parliamentary debates on immigration.’ In: Reisigl, Martin, & Ruth Wodak (eds), The semiotics of racism. Approaches to critical discourse analysis. Vienna: Passagen Verlag, 85–103. Van Dijk, Teun A. 2001. ‘Multidisciplinary CDA: A plea for diversity’. In: Wodak, Ruth, & Michael Meyer (eds), Methods of  Critical Discourse Analysis. London: Sage Publications, 95–120. Wodak, Ruth. 2001. ‘The discourse-historical approach’. In: Wodak, Ruth & Michael Meyer (eds), Methods of Critical Discourse Analysis. London: Sage Publications, 63–94.

Notes on Contributors

Lisa Carroll-Davis is studying for a PhD in Spanish at the University of  Southampton, focusing on political discourse in Nicaragua during the 1980s. She was a researcher for the LINEE project under the sixth framework of  the European Commission, where she examined the discursive relationship between nationalism and tourism materials in Cyprus. She is currently a Spanish Teaching Assistant at the University of  Southampton. Steffan Davies is a Lecturer in German at the University of  Bristol. He studied History and German at the University of  Oxford, writing his doctoral thesis on the literary, cultural and historiographical treatment of  Albrecht von Wallenstein in the ‘long nineteenth century’. José del Valle is Professor of  Hispanic Linguistics at The Graduate Center – CUNY. His research specialisms include the history of Spanish between 1800–2000, in particular with regard to the language politics of  Spanish nationalism and Panhispanism. He is currently coordinating and editing a new multi-authored book on the political history of  Spanish. Rembert Eufe is a postdoctoral researcher and Lecturer in Romance Linguistics at the University of Regensburg. He obtained his PhD in 2004 at the Ludwig-Maximilians-Universität München with a study of the use and status of Venetian during the Republic of Venice. He is currently working on personal and place names on Merovingian coins. Robert Evans was educated at Cambridge ( Jesus College) and became a lecturer at Oxford (and fellow of Brasenose College) from 1969. Since 1997 he has been Regius professor of History (and fellow of Oriel College). His main publications focus on the history of the Habsburg Monarchy and his current research embraces aspects of  the Habsburg lands as a whole from the 16th to the 20th century, with a special focus on modern Hungary; as well as the study of  Welsh history in comparative context.

484

Notes on Contributors

Federica Falchi obtained a PhD in Political Thought and Gender Studies at the University of  Rome 3 in 2006. She currently holds an academic research scholarship from the Sardinian Regional Government. Michela Giordano took an MA in Linguistics at California State University, Long Beach, and is a researcher and Lecturer in English Language and Translation at the Faculty of Political Science, University of Cagliari. Her research interests include applied linguistics and discourse and genre analysis, especially in the legal and political fields. Helmut Graser studied English and German at Bonn and Nottingham. Until his retirement in 2007 he held a position as Akademischer Direktor at the University of Augsburg. He has published mainly in the field of Early New High German linguistics. His current research focuses on the history of  literacy and language from below in early modern Germany. Juan Manuel Hernández-Campoy (BA and PhD Murcia) is Professor of  Sociolinguistics at the University of  Murcia where he teaches and researches English sociolinguistics and the history and varieties of English. He has published widely in these areas and is the co-editor of the Blackwell Handbook of  Historical Sociolinguistics. Patrick Honeybone studied English, German and General Linguistics at the University of  Newcastle upon Tyne, where he obtained a PhD in 2002 on Germanic phonology. He has worked at Edge Hill University and the University of  Edinburgh, where he is currently a Senior Lecturer in the department of Linguistics and English Language. His main research interests lie in historical and theoretical phonology and in the investigation of varieties of  English from the North of  England. Kristine Horner is Director of the Centre for Luxembourg Studies and Senior Lecturer in the Department of Germanic Studies at the University of Shef field. She has published widely on language politics and ideologies in Luxembourg and Luxembourgish-speaking communities.

Notes on Contributors

485

Richard Ingham studied at Oxford and London and is currently a Reader in English at Birmingham City University. He teaches language acquisition, English grammar and the linguistic history of English, and researches topics in English language acquisition and change. His current research focus is on language in medieval England, especially English-French bilingualism. Brian D. Joseph received his PhD in Linguistics from Harvard University in 1978, working on historical linguistics and Indo-European linguistics, and writing his dissertation on syntactic change in Medieval and Modern Greek. After a postdoctoral fellowship year at the University of  Alberta, he took a position at The Ohio State University, where he has been ever since, becoming the Kenneth E. Naylor Professor of South Slavic Linguistics in 1997 and Distinguished University Professor in 2003. He continues to teach and research in historical linguistics and especially Greek, both in its Indo-European context and its Balkan language-contact context. He is co-editor of  the Blackwell Handbook of  Historical Linguistics. Tomasz Kamusella is a Lecturer in Modern History at the University of  St Andrews. He received an interdisciplinary education in the Czech Republic, Poland and South Africa and has taught at the University of  Opole, Poland, Trinity College, Dublin and the Cracow University of  Economics. The focus of his research is primarily on the history of language politics and nation-building in Central Europe. Remco Knooihuizen has a background in Scandinavian linguistics and medieval studies and holds a PhD in English Linguistics from the University of Edinburgh (2009). His doctoral research concerned the sociolinguistics of minority languages and language shift in the early modern period. He is currently a Lecturer in English Linguistics at the University of Groningen and is continuing his research into the (historical) sociolinguistics of  language contact and small language communities. Nils Langer studied English and German Linguistics at the University of  Newcastle upon Tyne where in 2000 he obtained his PhD with a study of prescriptive grammarians from the seventeenth century. He has

486

Notes on Contributors

lectured at the University of  Bristol since then, where he is a Reader in German Linguistics. His primary research interests lie in the area of  historical sociolinguistics and he is currently working on language contact in Schleswig-Holstein in the nineteenth century. Nicola McLelland is an Associate Professor in German at the University of Nottingham, where she specializes in linguistics and in the history of  linguistic ideas. She has published on topics from thirteenth-century German literature to twenty-first-century spelling reform, but has a particular interest in the history of German grammars. She is an editor of the journal Language and History. Agnete Nesse studied Scandinavian and German Linguistics at the Universities of  Bergen and Tromsø and obtained her PhD in 2002 with a study of  the contact between Norwegian and Low German in Bergen during the Hanseatic period. She has worked at the university colleges of  Bodø and Bergen and is currently a post-doctoral researcher at the University of  Bergen, studying the destandardisation of radio language in Norway after 1968. Judith Nobels is a PhD-student at Leiden University where she is a member of the project ‘Letters as Loot’. She is preparing a dissertation on the historical sociolinguistics of seventeenth-century Dutch and is working with a corpus of personal letters. Gijsbert Rutten studied Dutch and Philosophy at the Radboud University Nijmegen where, in 2006, he obtained his PhD with a study of Dutch linguistics in the eighteenth century. He currently works as a postdoctoral researcher at Leiden University and at the Vrije Universiteit Brussel. His main research interests are historical sociolinguistics of  Dutch and diachronic construction grammar. Krassimir Stoyanov obtained his PhD at the St Kliment Ohridski University of  Sofia with an analysis of social changes and newspaper language since 1989. In 2002 he completed his habilitation, ‘The Newspaper

Notes on Contributors

487

and Foreign Language Teaching’. His research interests lie in the language of the print media and contemporary language education. He is currently an Associate Professor at the Department for Language Teaching and International Students at the University of Sofia. He is researching the language policy of  Bulgarian newspapers after the end of  World War II. B. Ann Tlusty is Professor and Chair of History at Bucknell University in Pennsylvania. A specialist in early modern Germany, she has published books on the culture of drink and on the martial ethic, and numerous articles on gendered behaviours including drinking, gambling, violence, military culture and civic defence. Her current research (with Helmut Graser) focuses on the social history of  literacy and language from below. Juan R. Valdez earned his PhD in Hispanic Linguistics at The Graduate Center of  The City University of  New York in 2008. His primary area of research includes critical linguistic historiography and language representations and he has recently published on the linguistic writings of Pedro Henríquez Ureña. He is currently a Visiting Scholar at Stanford University. Marijke van der Wal is Professor of  the History of  Dutch (Chair Merweborgh Foundation) at the University of  Leiden. Her publications cover the fields of  both historical linguistics and the historiography of  linguistics. Her current research is focussed on ego-documents and the language history from below. In particular she is the director of Brieven als Buit (‘Letters as Loot’) which explores a large corpus of  Dutch 17th- and 18th-century private letters, kept in the National Archives (Kew, UK), in order to contribute to a non-standard view of  the history of  Dutch. Anneleen Vanden Boer obtained her PhD at the Katholieke Universiteit Leuven and the Vrije Universiteit Brussel with a study of  language conf lict in German-speaking Belgium in 2011. She has lectured at the Katholieke Universiteit Brussel and the Hogeschool-Universiteit Brussel and researches language contact and language conf lict in European indigenous language minority settings. She currently works on language-in-education policy in the Euregio.

488

Notes on Contributors

Wim Vandenbussche is a Professor of  Dutch Linguistics at the Vrije Universiteit Brussel. He teaches courses on Dutch and Germanic language history, as well as on various aspects of sociolinguistics. His research is situated in the domain of historical sociolinguistics, with particular attention to the language situation in Flanders during the eighteenth and nineteenth centuries. Laura Villa received her PhD in Hispanic Linguistics from The Graduate Center of  the City University of  New York in 2010. Her dissertation explores the connections between the language standardization and nationbuilding processes in Spain and Latin America in mid-nineteenth century. Her scholarly research has focused on current and historical language policy in Spain and Latin America. Rik Vosters currently works as a researcher at the Vrije Universiteit Brussel, where he is finishing his PhD in historical sociolinguistics. His main area of interest is the sociolinguistic situation of eighteenth- and nineteenth-century Dutch in Flanders, focusing mainly on the tension between language use, language norms and language planning. Melanie Wagner studied in Lancaster and Aberystwyth where she obtained her PhD in German Sociolinguistics. She is currently a postdoctoral researcher at the Department of  Luxembourgish Linguistics and Literatures at the University of  Luxembourg and conducts research on sociolinguistic topics such as language choice, norm awareness, language attitudes and language ideologies. Nicholas M. Wolf completed a PhD in British and Irish history at the University of  Wisconsin-Madison, and currently holds an appointment at Virginia Commonwealth University. He specializes in the social and cultural history of eighteenth- and nineteenth-century Ireland with a focus on the Irish language. He has also published more broadly on subjects such as Irish religious history, and is currently working on a close study of  the Irish-speaking community between 1790 and 1870.

Index

abbeys  311 abbreviation  72, 113–14, 295 academies  176, 180–9 Academia della Crusca  176 Académie Française  176 Dominican Academy of  the Spanish Language  196 Real Academia Española (RAE)  175–9, 182–3, 189, 210–24 Mexican Academy of  the Spanish Language  176, 182–3 accommodation  62 acronyms  71, 73 Act(s) of  Union  39, 51, 62 Actioun Lëtzebuergesch  452–4 adjective  101, 419, 421, 445 administration (language of )  222, 232, 237, 244, 257, 304–5, 353, 374, 460 adverbial  31–2 Aedler, Martin  99 Africa(n)  195–7, 200–2, 258, 346 Alberdi, Juan Bautista  178 Alemán, Miguel  175, 181 Alfonso X  194 Alonso, Amando  195 Alps  198, 430 America(n)  177–8, 199, 374, 437, 443, 450 American English  71, 74 War of  Independence  346 Amsterdam  231, 347, 354 Anabaptists  122 analogy  81, 105, 390 ancien régime  348

Angers  304, 310 anglicism  204 Anglo-Dutch Wars  345–6 Anglo-Norman  271–6, 283–90 Anglo-Saxon  177, 188, 310 Annales d’histoire sociale et économique  7 Annius de Viterbo  94–6 anthropology  57–60, 432 anthroponyms  305–15 Antwerp  235–48 apocope  379, 382 Aquitaine  315 archaeologists  7, 316 archaism  195, 199–200, 390 areal linguistics  152 Argentina  178, 199 argumentation theory  474 Arles  312 Arlon  256, 259 army  430–1, 453–7 Aryan  16 Asia  27 assimilation  383 Ateneo Científico y Literario de Madrid  223 Atlantic  178–9, 315 Atlas Linguistique de la France  333 Atlas Linguistique et Ethnographique  333 Augsburg  364–81 Austen, Jane  76 Austria(n)  155, 231, 237, 430 Austria-Hungary  39, 162 Austrian Netherlands  257 authority  364, 368, 371, 373, 409, 412 autograph  348–58

490 Babel  94, 96 bad data  70, 324, 338 Balkan  162, 436–8, 443 Baltic  160 banking  374 barbarism  204 Basseterre  350 Batavian  93 Baumgartner, Philip  364–9, 372 Bavarian Swabia  374 Beauchamp, Thomas, Earl of  Warwick  280 Bebel, Heinrich  92 Becanus, Goropius  96–7, 100–1 Behaegel, Pieter  236 Bého  256, 259 Belarus  158–71 Belgioioso, Cristina di  468 Belgium  229, 231, 234, 255–67, 448 unionism  261, 266 Bello, Andrés  187, 194, 211 Benecke, Georg Friedrich  103 Bentham, Jeremy  468 Bergen  114–19 Berlin  440 Bode-Museum  296, 304 Bernaers, Elisabeth  349–50 Berosus  94 Berry, Agnes  395 Bible  94, 134, 148–9, 453 bilingualism  55, 271, 448 bilingual editing  112, 122 bishop  298, 301–4 Black Death  287, 290–1 Blanco, Carlos Larrazábal  202 Blaringhem  334, 336 blending  73–5 Bloch, Marc  3, 7 Bloomfield, Leonard  77, 153 Bohemia  135–6, 142, 154 Bopp, Franz  100 Bordeaux  307, 311

Index Borge, Tomas  416, 421 borrowing  78 Bosnian  153 Bourdieu, Pierre  220, 222, 461 Bourges  311 Boyer, Jean Pierre  196, 198 Brabant  247–8 Brioude  312 Britain  5, 7, 50, 52, 178, 310, 465 historians  5 Parliament  467, 477–8 Brittany  56 Brutus  148 Bulgaria  156–68, 429–44 Old Bulgarian  162 bureaucracy  204 Burke, Peter  7–8, 18 Cáceres, Josén Núñez de  198 Cadiz, Cortes de  223 calque  298 Calvinists  122 Cambridge  5, 32 Campano, Giannantonio  91 Canaanite  73 capital letters  113 Carinthian  158–9 Carlos III  177 Carolingian  311, 316 Carpathian  162 Carr, Edward H.  6, 10, 18, 40, 41 Carribean  345 Carroll, Lewis  73 Castilian  194–201 catechism  240 Catholic  6, 51–2, 122, 154–5, 171, 204, 220, 260, 312, 327, 371–2 Greek (Uniate) Catholicism  171 Celtic  5, 16, 298 Celtis, Conrad  91–2 census  51 Central Europe  149, 153, 170

Index century, eighteenth  9–10, 32, 53, 99, 103–6, 155, 229–49, 329, 344–5 century, fifteenth  21, 39, 214, 374, 390, 396, 402 century, fourteenth  120, 272–7, 280, 283–91 century, fourth  40, 298 century, nineteenth  9, 21, 40, 50–7, 62–3, 76, 103, 106, 148–50, 154–6, 160, 166, 170, 177, 187–8, 196, 202–3, 209–14, 222–4, 229–40, 249, 296, 328, 345, 354 century, ninth  295, 302 century, seventeenth  6–8, 39–40, 62, 93–4, 112, 118–19, 199, 231–2, 238, 326–7, 332, 338, 344, 347–8, 352–5, 359, 374, 376, 402, 476 century, seventh  298 century, sixteenth  6, 22, 39–40, 90, 94, 104–5, 112, 114, 118–24, 148, 199–202, 231, 354, 363, 374–7, 380–5, 402 century, sixth  294, 312 century, thirteenth  149, 272, 284, 289, 291 century, twelfth  276, 298 century, twentieth  40, 52, 54, 57, 151, 167–8, 188, 196, 202, 235–7 century, twenty-first  167 chain migration  329 chain shift  79–80 Chalon-sur-Saône  301, 311 change  56–8, 61–2, 67, 70–1, 73–85 actuation  77 endogenous/exogenous  8 gradual  79 sound change  8, 69, 81 chant  435, 445 Charibert II, King of  Aquitaine  315 charters  113, 294 Chaucer, Geof frey  85 Childebert I, King of  Paris  311 children  98, 216, 274, 287, 291, 344, 374, 470–9

491 Chile  211 Chinese  153 Chomsky, Noam  23–5, 36 Christianity  298, 311–12 Christof fer, King of  Bavaria  116 chronicles  272, 294 Anglo-Saxon Chronicles  31 church  331, 470 Church Slavonic  154–9 citizenship  461 class (social)  55–61, 344, 346–7, 349, 395 classification  147–67 clergy  350–6 Clermont-Ferrand  304, 310 Cluverius, Philipp  93 code-switching  55 codification  231–2 cognitive reconstruction  16–44 coins  6, 293–314, 316 Collingwood, R.G.  23, 36–41 collocation  415–22, 441, 443 colloquial forms  76, 373 Cologne  311 Colombia; Colombian  179, 187, 211 colonies  56, 61, 152, 177–8, 188, 258 commerce  53 communism  90, 166, 413, 433–5, 438, 442 community  9, 11, 36, 52–61, 179, 186, 197–8, 257, 324, 392, 403, 438 imagined  179 linguistic  9, 182, 189 comparative method  30–41 compounds  73, 102–6, 299 conf lict  53, 409–10, 447, 460 Congo  258 consonant  39, 380–6 fricative  29 plosive  29 voiced/voiceless  29, 105 weakening  381 contact linguistics  77, 328 Contras (Nicaragua)  421 Coquelles  334, 336

492 Cornish  325 corpus  26, 346–7, 354–9, 409, 414–19, 424, 434, 457 Council for Public Instruction (Spain)  213, 221–3 Courtenay, Jan Baudouin de  165 creole  61, 152 Critical Discourse Analysis  5, 409–24, 472 Croatian  10, 153–9 Cromwell, Oliver  326 Cuba  200–2 Cuervo, Rufino José  187 culture  52–63, 177, 197, 369 cultural history  60 cultural symbolism  90 see also oral culture curse  76, 434 Cuspinianus, Johannes  96 custom  431 Cyrillic  155, 171 Czech  10, 18, 149–68, 210 Daniel, Hans  368, 372 Darwin, Charles  151, 159 de Ruyter, Michiel  346 demography  323–4, 328, 333–9 Denmark; Danish  9, 104, 118, 121, 211 Dano-Norwegian  116, 120–4 derivation  104, 106, 299 Derrida, Jacques  20 Des Roches, Jan  235, 239 deterioration  204 diachronic  20–1, 35, 42, 57, 69, 83, 160–1, 164 dialect  5, 8–10, 31, 55, 59, 61, 63, 77, 81, 83, 113, 153, 160, 162, 168–9, 187–8, 233, 235–6, 244, 259, 333–8, 382, 453, 461 contact  402 continuum  164–5 dialectology  199, 204 levelling  61

Index diaries  40, 343 dictatorship  183, 204, 412 dictionary  96, 103, 169, 176–7, 180, 189 digitalization  124 diminutive  300, 380, 382 Dinant  311, 313 diphthongization  378 discipline (academic); interdisciplinarity  3–5, 7–12, 15–18, 20–5, 27, 42–4, 55–9, 62, 69, 129, 148, 205, 296, 344, 359, 414, 447, 472 discourse  212, 216, 409–12, 424, 450, 455, 466 discourse analysis see CDA divine origin  97–8 Dobrovský, Josef  154, 155, 156, 159–63 doctors  352, 356 Dominican Spanish; Dominican Republic  195–204 identity  193 nationalism  203 Dorestat  313 drag chain  81 Dublin  62 Dunkirk  8, 323–37 Dunkirk French  338 Dus, Jacob  116 Dutch  7, 31, 101, 104–5, 211, 231–49, 261–6, 326–58 grammarians  105 Republic  93 Southern Dutch  235 E-Language  25–6 East Anglia  400–1 English  392 East Anglian Subject Rule  401 East-Central Europe  5 East Swabian  377–9, 383 Eastern Europe  152–3 Echeverría, Esteban  178 ecology  61 edition  11, 112, 116–20, 124, 309

493

Index education see school Edward I  278–85 Edward II  272, 285 ego documents  343–4 Eisleben, Johannes Agricola  92 ejective  29 elaborating symbols  450–1 elite  131, 142, 178, 187, 197, 258, 448 emigration  11, 52, 374–5 encoding  348–9, 358 enemy  409, 421, 423 England  39, 272, 291, 310, 345, 396, 430, 437, 443, 468, 470, 475 English  10, 21, 25, 28, 31–2, 35, 50–3, 61–3, 73, 75–6, 81, 149, 196, 203, 271–5, 286–7, 327, 396, 402–4, 444 Early Modern English  390–2, 403 Middle English  5, 21, 115, 275–6, 389–92, 396, 403 Northern Subject Rule  400, 402 Old English  29–39, 275, 390 Southern Subject Rule  401–2 Enlightenment  97, 214 Étaples  337 ethnicity  7, 9, 55, 60, 150, 297, 330–1, 338 ethnolinguistic  149–50, 152, 169–70, 200 ethnography  91, 432 etymology  20, 214, 301, 303, 310 etymon  305 Eupen  256–8 Europe  27, 148, 271, 433 European Union  448 face (pragmatics)  10 Facebook  71 fairy doctors  52 family  396 tree  151, 166–7, 334 Fampoux  334, 336 Farrar, Frederic William  151 fascism  430–43 see also National Socialism

Felipe V, King of  Spain  176 female emancipation  467, 472 female writers  347 Fermín de la Puente Apezechea  180 Fernando VII, King of  Spain  221 Filipino  56 Finno-Ugric  170 Flanders  96–7, 211, 228–47, 234, 245, 249, 259–63, 326, 332 East/West Flanders  247–8 Fontaine, Liam  80 formulaic expression  391 Fort-Mardyck  333–7 Fourier, Charles  466–7 France, French  5, 56, 59, 177–9, 231–2, 237–9, 244–5, 248, 257–9, 261–7, 271–87, 303, 306–8, 324–37, 437, 448, 460, 468–9 empire  257 French-Luxembourgish  449 Insular French  286, 291 Medieval French  288 Old French  276–8, 288, 307 Revolution  58–9 Franco, Francisco  183 Franconia  295 Moselle Franconian  9, 448 Francoprovençal  303 Franks; Frankish  296, 298, 310–14 Frankish Gaul  310 freedom fighters  414, 421 Frente Popular  183 Frisian  9, 325 Fruchtbringende Gesellschaft  99 FSLN (Sandinista Front for National Liberation)  412–24 Gaelic League  50 Galicia  171 Galilei, Galileo  476 Gallego, Juan Nicasio  223 Gallicism  187, 204 Gallo-Roman  301, 305, 308, 313–15

494 García, Anastasio Somoza  412 Garibaldi, Anita  468 Gaul  298, 304–6, 306, 310, 312, 314 Gebauer, Jan  162–3 gender (noun)  274–8, 283, 285, 288 agreement  279, 286–7, 289, 290 gender  55–60, 70, 329–32, 349 generativism  25–6, 36, 69 genitive  308 geology  24 German; Germany  5, 8–10, 21, 31, 90, 93–7, 103, 105, 113, 123–4, 150, 154–7, 160, 211, 255–67, 293, 374–5, 379–80, 429, 430–57, 460 Early New High German  374, 380 High German  8, 118–19, 121–2, 377 Low German  9, 111, 117–18, 122 Middle High German  103, 377–8, 381–4 Middle Low German  114–15, 117, 120 New High German  377–84 Old High German  301 Second Sound Shift  8 Southern German  96, 105 spelling  105 Upper German  378, 381–3 Germania  91, 93–4 Germanic  4, 28, 30, 33, 35, 80, 90–1, 93–94, 97, 100, 159, 295–315 Germanic-Romance border  328 North Germanic  96 Proto-Germanic  28 rootword  90, 94–105 Gironde  315 Glottalic Theory  29, 34 Golden Age  231, 374 Gothic  28, 30, 35 government  184, 215–21, 259, 291, 415– 20, 430, 441–2 grammar(s)  26, 33, 59, 69, 78, 150, 169, 176–80, 182, 185–9, 194, 210, 224, 232, 235–40, 273–8, 280, 289, 404 grammarians  97–104, 235–38, 249

Index Greek  16, 28, 70, 81–2, 97, 101–2, 297–9, 301 Cappadocian Greek  275 Gregorian calendar  371 Gregory of  Tours  294 Grimm, Jacob and Wilhelm  90 Grimm’s Law  35, 80–1 Guatemala  185 Guzmán, Martín Luis  184–5, 189 Habsburg  59, 231 Haeckel, Ernst  151 hagiography  297 Haiti  196–8, 202–4 Haitian Creole  196, 204 Halbritter, Christoph  368, 372 handwriting  7, 349–7, 351, 368, 377–8, 386 handwritten documents  112–13, 246, 249, 376–84 see also scribes Hanseatic League  114, 116–17, 119, 121–4 Harsdörf fer, Georg Philipp  99 Haugen, Einar  176 Haverskerque  334, 336 Hebrew  97–102 Hegel, Georg W.F.  166 Henisch, Georg  96, 99 Henry IV, King of  England  280 heraldry  6 Herder, Johann Gottfried  97–8, 149 heritage  94, 201 Hispanic  183, 197, 201–2 Panhispanism  177, 179, 182, 189, 197 history; historical of art  93 artefacts  40 ‘from below’  7, 11, 344, 373–4, 386 geography  21 phonology  21–2, 31 pragmatics  6 semantics  11, 20 social  7, 17–18 sociolinguistics  18

495

Index historiography  17, 52–4, 59, 62–3, 93–4 Hitler, Adolf  436–41 Hobsbawm, Eric  58 Holland see Netherlands Holy Roman Empire  135, 150, 374 Reichstag  91 Honduras  179 Humboldt, Wilhelm von  100 Hundred Years War  272 Hungarian  149, 156, 162, 170, 430 Huy  305, 313 hypercorrection  384, 392, 403 I-language  17, 25–36 iconization  201 idealism  23, 36, 40–1 ideal speaker-hearer  25 identity  50, 56–61, 63, 84, 177, 202, 204, 263, 266, 267, 411, 423 collective  411 cultural  62–3 see also national identity ideology  149, 152, 158, 176, 197–8, 203–4, 211, 213, 219, 223, 230, 409–11, 413, 421, 429, 441–2, 444, 450, 460–1, 466 language ideology  179, 195, 212, 459 illiteracy  331, 347 Illyrian  154, 156–7 immigration  5, 7, 55, 196, 202–4, 274, 327, 329, 331, 333, 337–8 imperative  94, 99, 103, 105 imperfect grammar  277 imperialism  414, 422 India  345 Indo-European; Indo-Germanic  30, 34, 83, 85, 159–60, 299 infinitive  390–1 inscriptions  293 institution  177, 219, 312–16, 432, 445, 467

insults  367 intellectual  204, 223, 465, 467, 480 interdisciplinarity see discipline internet  71 interrogation reports  244 Ireland  49–63, 50, 52–3, 57, 62–3, 326 Old Irish  28 see also Gaelic Irenicus, Francisus  94–5, 99, 104 irony  476 isogloss  164, 169 Israel  148 Italy, Italian  10, 150, 465–8, 477, 479 italic  123, 328 Jacobite  53 Jesuit University of  Central America  417 Jews  386 Jones, William  149 journal  117–18, 124, 417 journalism  73 Jumièges  302 Kashubian  157, 162, 167 Keyser, Rudolf  115 Kloss, Heinz  90 Kochubinskii, Even  163 koineization  56, 61 Kopitar, Bartholomäus  155 Kundera, Milan  454 Kyiv  156 Labov, William  56, 70, 79, 324 language acquisition  97, 274 contact  7–8, 11, 77, 82 decay; decline; deterioration  187–8, 233–4, 271, 309 first language  274 history from below  7, 11, 344, 363, 373–4, 386

496 legal language  374, 460 legislation; policy  131–6, 139–41, 169, 244, 256, 325–6, 448, 460–1 myths about  170–2, 230, 233–7, 244, 248–9 of ficial language; of ficialization  59, 129–42, 154, 155, 162, 169, 177, 201, 212–13, 215–24, 232, 241, 245, 249, 256–7, 266, 272, 325, 458, 460 origins  97, 99, 106 of proximity  343, 346, 375 revival  50–1, 210 second language  274, 279, 283, 289 shift  50–1, 53, 61–2, 79, 324–8, 332, 338 Las Casas, Bartolomé de  202 Latin  5, 28, 30, 69, 72, 79, 97, 101–2, 149, 156, 171, 214, 238–9, 295, 297–310, 314, 325, 364, 374 Pre-Latin  304 Latin America  176–81, 194–5, 204 wars of independence  180 Legouvé, Ernest  468 Lekhian  156–7, 161 lenization  383 Leroux, Pierre  468 letter  32, 40, 343–59, 364–5, 374–5, 396, 456 Leuven  240 levelling  390 Lexer, Matthias  103 lexicography  106 lexis  429, 441–5, 475 lexical features  473 lexical gap  73 lexical semantics  6 lexicalization  473 military  439 Liège  247 Lievensens, Jan  350, 357 Limburg  248, 257

Index Limoges  311, 313 ‘linguistic turn’  3, 20, 58 literacy  328, 331, 347, 448, 456 literary form; literary language  235, 373 Lithuania(n)  83, 156 loan words  118, 236, 238–9, 298 Loire  302, 315 Lomonosov, Mikhail  154 London  395, 402, 444 Longolius, Johann Daniel  97 Losch, Jonas  368, 372–3 Low Countries see Netherlands Lübeck  114, 116, 329 Lusatia(n)  154, 158–9, 326 Luxembourg; Luxembourgish  9, 231, 257, 447–61 Luxemburger Wort  453–7 Lyon  311 Maastricht  311, 313 Macedonian  162, 167–8 Mâcon  311–12 Magyar see Hungarian Maksimovich, Mikahil  158, 160, 163 Malaysia  183 Malmedy  256–8, 262–3 Managua  417 Mario, Jessie White  468 marriage  327–32, 367, 469, 473 register  8, 327–9, 338, 354–5 Marsal  311 Marseille  311–12 Marxism see communism mass media  169, 188 Mazzini, Giuseppe (1805–72)  468, 471 media discourse  451, 453–4, 460 Meistersinger  370 Melanchthon, Phillipp  100 memory  310, 432, 449–51, 454–7, 460–1 Menéndez, Pidal  188 mentalism  23, 25–6

Index metalinguistic comments  456–7 discourse  186 metathesis  381 metre  33, 366, 380 Metz  311 Meurs, Hendrik  352 Meuse  311, 313 Mexico  175–6, 182–9, 203, 211 Middle Ages  7, 40, 61, 76, 113, 120, 124, 131, 133, 135, 156, 271–92, 294, 297–8, 303, 305, 308, 309, 311, 389–406 Middleburg  354, 355 migration  52, 202, 325, 326, 328, 329, 330, 337, 402, 461 military  198, 203, 204, 439 Mill, John Stuart  465–81 minority  59, 255–60, 325–6, 332, 331 Moldovans  170 monarchy  431 monastery  302, 312, 314 monoglossia  196, 215 monosyllabic  95–105 Montenegrin  153 Montzen  259 mood  389–401, 403 Moonen, Arnold  105 Moravian  156–7 Morelli, Salvatore  465–81 morpheme  104, 390 morphology  104, 239, 307, 378, 380 mother tongue  121 Müller, Georg  371–3 Müller, Peter  366 multilingualism  8–9, 11, 56, 59, 153, 196, 263, 265, 266, 267 Munch, Peter Andreas  115 Muslim  122, 154 n-loss  82

497 name  294, 296–308, 328, 337, 412–13, 421, 409–23 place names  246, 294, 304–14, 316 proper names  100, 246 Namur  311, 313 Napoleon Bonaparte  231, 257 Napoleon, Louis  231 Napoleonic Wars  54, 150, 156 narrative  59 nation  9, 50, 58, 149–53, 166, 169, 177, 195–8, 204, 210, 216–22, 255, 411, 413, 416, 420, 449–51, 454–5 community  222–4 nation-building  120, 198, 203, 221, 256–8, 451 nation-state  152–3, 169, 224, 255, 325, 453 national identity  7, 9, 59, 116, 196, 203, 450–1 national language  138, 150, 153, 169, 210, 256, 261–5, 447–61 National Enquirer  85 National Schools (Ireland)  51, 62 National Socialism  90, 430–7, 448–55 denazification  259 nationalism  58–9, 149–50, 152–3, 157, 166, 197, 450 nationality  57–8, 257 Navarrete, Martín Fernández  223 Nazism see National Socialism Nebrija, Antonio de  194 negation  384 neogrammarian  56, 201 Netherlands  93, 96, 229–34, 240, 246, 248–9, 344, 345, 347, 348 network  334, 368, 372–3 New Zealand  19 newspapers  53, 59, 73, 413, 433–4, 441, 451 Nicaragua  410–24 Nicaraguan Spanish  413

498 Nightingale, Florence  470 Nord-Pas de Calais  333 Norfolk  395–6 norm  115, 175–6, 186, 189, 210, 213, 221, 229–41, 245–6, 248–9, 448, 461 Norman Conquest  7, 273 Normandy  290 North Sea  311 Norway; Norwegian  111, 113–22, 211 Norwich  395 noun  83, 101, 279–89, 308, 390, 419, 421 see also gender numismatics see coins object (syntax)  31–2, 438 Occitan  303 Old Occitan  307 O’Connell, Daniel  51 Old Norse  115, 120 onomastics  296–7, 308–10, 314, 338 oral communication; culture; transmission  52–3, 84, 169, 373, 375, 381, 384–6, 445, 456 Orkney  325 Orléans  31 Orosius  31 Ortega, Daniel  413, 416, 420, 422 Orthodox Christianity  154–5, 171 orthography see spelling Ortíz, Fernando  202 Oscan  69 Oslo  56 Oviedo, Gonzalo Fernández de  202 Owen, Robert  467 Oxford  5 Palacký, Frantíšek (Franz)  157–8, 160, 163 palaeography  6 pamphlet  91, 140, 213, 215, 413, 415 Panama  185 Paraguay  185

Index Paris  300, 304, 311, 337, 348 parliament  132, 139, 142, 259, 279–80, 283, 289, 465, 468–79 Parliament Rolls of  Medieval England  276–85, 289–90 Pas-de-Calais  305 Paston family; Paston Letters  391–404 patois  196, 237 patricians  364, 377 patriotism  259–61, 266 Pearl Harbor  451 petitions  280, 283, 285, 289–90, 386 Philip of  France  272 Philippines  176, 185 philology  3, 91, 154, 156, 158, 194, 201, 295, 365 phonetics; phonology  18, 30–4, 80–1, 200–1, 214, 274–6, 278, 289, 306–7, 316, 333–4, 385, 403, 473 Picardy  334, 337 Piccolomini, Enea Silvio  91–2 Pichardo, Esteban  200 Pidal, Ramón Menén dez  181, 194 Pieters, Maaike  350 Pietersen, Claejs  350, 357 Plato  38, 68 Poitiers  311 Polabian  156–62 Poland; Polish  148–9, 154–65, 171 Police Ordinance  370 police reports  244 political discourse  11, 59, 410–11, 414, 424, 473 history  17 parties  432, 438 songs  372 Pomeranian  156–7 Portugese  202 postal service  456 postmodernism  19–20, 34, 37, 39 postvocalic /l/  200

Index postvocalic /r/  200 prescription; prescriptiveness  37, 142, 222, 238, 245, 249 press  222, 429, 466 priests  51, 287 print; printers  59, 112, 114, 222, 374, 376–7, 380, 384–6, 451, 460 prison; prisoners  370, 430, 468 private sphere  58 privateers  345 pronunciation  105, 240, 274, 383 propaganda  166, 373, 410–15, 418, 423, 431, 434, 439, 453–4, 473 Protestant  6, 53, 99–100, 155, 171, 327, 371–2 Reformation  58, 375 proto-languages  27–8, 167–8 Proto Indo-European  27–39, 69, 74–5, 80 Prussia  21, 257–9, 455 Old Prussian  325 psychology  37 public sphere  63, 348 Puerto Rico  200, 202 pun  22 punctuation  113 pure language  96, 102–3, 115, 176, 188, 236, 238 Pyrenees  198 Queen Elizabeth II of  England  75–6 Queen Isabel I of  Spain  194 Queen Isabel II of  Spain  213, 221 Quentovic  305, 311 Quintana, Manuel José de la  223 race  195, 198, 200, 204 realism  33–4 reconstruction  15, 25, 30, 40–4 Red Army  429, 439 Regensburg  91

499 region (linguistic)  11, 347, 381, 386 register  55 Reims  311 relative adverb  392 religion  52, 54–5, 59, 63, 121–2, 134–5, 138, 148, 171–2, 197, 287, 298, 306, 311, 367, 371, 417, 432 Remigius of  Reims  312 Rennes  311 restoration  120, 326 Rhenanus, Beatus  95, 100 rhetoric  138, 142, 178, 203, 234, 365, 373, 442, 445, 450, 473, 475, 480 Rhine  298 Rhineland  93 rhotacism  69, 79 rhyme  22, 366, 380 rhythm  367 Richard II  280, 285 right to vote see suf frage Rigveda  84 ritual  59, 429, 431–4, 438, 440–3, 445, 459 Roches, Jan Des  238 Rodez  311 Roman  154, 296, 298, 310–15 Roman Empire  148, 309–10 Romance languages  170, 293, 295, 303, 307 Romanian  170, 307 Romero, José Rubén  181 rootword  90–105 Rosa, Francisco Martínez de la  222 Rosenblatt, Ángel  178, 217–18 Roth, Lex  454 Rotterdam  354 Rouen  311 Rousseau, Jean Jacques  104 Rus’  149, 170 Proto-Rus’  167 Russia  171

500 Russian  5, 104, 149, 154–70, 431, 433 Great Russian  168 Proto-Russian  167–8 Rusyn  162 Ruthenian  149, 156 Saami  149 Šafařík, Pavel Josef  156, 160, 163 Sailer, Balthasar  366–7 sailors  337, 346, 355 St Eloy  303, 313 St Germain des Prés (Paris)  298 St Julian  312 St Kitts  350, 357 St Peter  312 St Petersburg  171 St Philibert  312 St Victor  298 St Vincent of  Saragossa  311 Sajnovics, János  149 salesmen  352, 355–6 Samaná Province  196 Sancto Filiberto  302 Sandino, Augusto C.  412 Sandinista  409–24 Sankt Vith  256–8 Sanskrit  28, 84, 100, 104, 106, 149 Santo Domingo  195–204 Sapir-Whorf  hypothesis  410, 412 Sarmatia  148–9 Sarmiento, Domingo Faustino  178 Saussure, Ferdinand de  20–1, 25 Saxon  97, 100 West Saxon  29 Schleicher, August  151, 153, 159–63, 165, 168 Schmidt, Johannes  152, 164–5 Schönigk, Valentin  376–7, 384 school  51, 56, 100, 149, 168–9, 177, 182, 210–13, 215, 218–24, 232, 272, 274, 286, 298, 328, 353, 374, 430, 448, 456, 458–9, 468–79

Index books  237, 384 grammar schools  273, 274, 287 legislation  221, 448, 458 teachers  213–20, 234, 238–9, 249, 456, 470 Schottelius, Justus Georgius  92, 99, 100–6 Schuchardt, Hugo  152, 164, 165 Scotland  39, 326, 402 Scots (language)  332 scribes  7, 85, 229, 244, 246, 348–9, 358, 386 semantic  10, 31–2, 120, 123, 378, 473 Semitic  99–100, 104, 106, 298, 301 Serbian  10, 153, 156–60 Serbo-Croatian  158–67 Serbo-Slovenian  160 Shetland  325, 332 shibboleth test  204 sigillography  6 Silesian  157, 162 Simon, Gustav  452 Sint-Christof fel  357 Slatina  443 Slav; Slavic  10, 18, 147–8, 154–64, 167, 172, 433–4, 440–1 Eastern Slavic  148 Pan-Slavism  149 Proto-Slavic  167 Western Slavic  148, 155, 161 slaves  196, 202, 234, 344 Slavonic  28, 168 Old Church Slavonic  30, 35, 154–5, 158, 162–4 slogan  429, 433–4, 438, 442–5 Slovak  10, 154–9, 162, 167–8 Slovenian  154–67 social anthropology  8 social network  326, 365 social psychology  449 sociolinguistics  7–8, 67, 77–8, 83–4, 90 Socrates  68

Index soldiers  355, 430 Somoza-Debayle, Anastasio  412–23 Sønderjysk  9 Sorbian  157–67, 325 Lower/Upper Sorbian  154, 168 Soviet Union  152, 171, 429–30, 433, 435, 440–1, 442–4 Spain; Spanish  39, 61, 176–231 speech act  56, 59, 61, 63 speech community  75, 82 spelling  7, 22, 106, 113–14, 119, 122, 176–7, 180, 209–10, 213, 216–20, 224, 230–40, 245–9, 273, 377–8, 384–5, 392, 458–9 norms; system  105, 212–15, 223–4, 241, 244, 248 simplification  212, 216–17 syllabic  379, 384 variation  6 Speyer  305 spoken language  286, 380, 383–4 dialects  169 Sporham, Matthäus  366 Sprachbünde  152 Sreznevskii, Izmail Ivanovich  158, 160, 163 stability  83–4 Staiger, Georg  366 Staiger, Johannes  366 standardization  6, 11, 18, 59, 153–4, 176, 181, 189, 194, 211–12, 216, 221–2, 224, 231–4, 272, 458–61 statistics  7, 298, 331, 395, 415, 418–19 Stavelot  257 Stefan I, Exarch of  Bulgaria  434 Steinbach, Christoph Ernst  104 Stevin, Simon  100, 102 Stieler, Casper  103 Stiernhielm, Georg  96 Stockholm  211 stop  27, 81 Strasbourg  295, 305 structuralism  20, 42, 44

501 subject (syntax)  31–2 subjunctive  389–93, 398–404 subordinate clauses  391 suf fix  380, 382 Suf folk  396 suf frage  466, 469–72, 475–8, 480 Süßmilch, Johann Peter  97–8 Swabian  366, 378, 380, 382 swearing see curse Sweden; Swedish  96, 104, 211 syllable  379, 381, 384 synchrony  20–1, 35, 42, 162 syncope  379, 382 syntax  18, 26, 31–2, 39, 59, 195, 199, 473 system linguistics  7, 403 Syv, Pedder  99 Tacitus  90–4 Tapia, Eugenio de  223 Tatinghem  334, 336 tax  313–16, 386 television  73 tenses present  390, 401 preterite  390, 392, 403 terrorist  412 testimony  21–2, 365, 386 textbook  218, 224 thingstead  314 Thiverzay  305 Thompson, William  468 Tiällmann, Nils  96, 99 Togores, Mariano Roca de  180 toponomastics see onomastics totalitarianism  432, 445 Toulouse  311 Tournai  311 Tours  301 Tower of  Babel  94 translation  84, 112, 114, 116–17, 119–20, 122–4, 134, 141, 150 Trausch, Gilbert  453

502 Treves  303, 311 Trier, Jost  90 trilingualism  448 Trizay-sur-le Lay  305, 311 Troilus and Criseyde  85 Troyes  311 Trudgill, Peter  7 Trujillo, Rafael L.  203–4 Tubersent  334–7 Tuisco  94 Tunner, Jack  444 Turbas  413 Turkish  275 Ukraine  149, 156–71 Ulm  366, 368, 372 uniformitarianism  23–4, 29, 34, 38, 43, 70–1, 76, 84 United Kingdom of  the Netherlands (UKN)  231–7, 241, 245, 249, 257 United Nations  183 United States of  America  188, 203, 414, 421, 423, 430 see also America US Marines  412, 421–2 universal grammar  23, 75 Upper Lusatian  154 Ureña, Pedro Henríquez  193, 194, 195, 199, 200, 201, 202, 203, 204 Uruguay  185 V2 (verb second)  31–2, 35 Vale of  Belvoir  39 Valera, Juan  187 vandalism  370 Vasquez, Horacio  203 Velde, Jacob van de  355, 357 Venezuelan  187, 194 verb  31–2, 34–5, 73–4, 76, 95, 98–101, 106, 378, 390–2, 398, 400–1, 403–4, 436–7, 440–2, 445 auxiliary  390 modal  390–1

Index verbal hygiene  187 Verdun  311 vernacular  101, 271, 325, 364, 374, 384–5 Verpoorten, Jan Domien  238–9 verse  366–7 Verstegan, Richard  93–4 Vieille-Égliese  334, 336 Vienna  171, 311–12 Congress of  Vienna  257 Villers-Cotterêts, Decree of  325 Vil’na/Vilnius  156 Vis-en-Marais  305 Visigothic  312 Vlaemsch  230 Vlissingen  354 vocabulary  195, 199, 238 Vostokov, Aleksandr  155, 158, 160 vowel  377–81, 384–6, 390 front  78, 382 length  379 raising  81 Waimes  257, 262–3 Wallonia  259, 261, 263 Walter de Bibbesworth  273 Walter, Will  444 War of  American Independence  54 Wars of  the Roses  396 Weisgerber, Leo  90 Welkenraedt  256 Welsh  8, 59, 133–4, 325 West Indies  196, 202–3 Whorf, Benjamin  56 see also SapirWhorf  hypothesis Wij bij Duurstede  305 Willems, Jan Frans  234–6 William I  231 witchcraft  39–40 women  330, 344, 347, 352, 367, 369, 466–80 emancipation; rights of  465, 468, 471, 473, 477, 480–1

503

Index word-formation  73, 75, 106 word lists  117, 118, 273 WordSmith  415, 418–19, 423 World War I  171, 258–9 World War II  170–1, 258–9, 429, 447–61 written language  30, 121, 231–2, 291, 350, 385, 456, 461

Württemberg  166 Yugoslavia  162, 430 Zárate, Antonio Gil de  220, 223 Zealand  355 Zeches, Léon  453–4

Figure 1  The idol Tuisco (Verstegan 1605: 71; Warburg Institute)

Figure 2  The frontispiece to Stieler (1691)

Figure 3  From the dialogue De dudesche unde de Norman, Ms 1418, Bergenfahrerarchiv, City Archive, Lübeck

Figure 4  Front page of  Ms. Gl.kgl.S.2836, Royal Library, Copenhagen

Figure 10  The explicit reference to the writing process in letter 3–1–2008, 129–130, letters-as-loot-corpus

Figure 11  Letter by Claeijs Pietersen, letters-as-loot-corpus

Figure 12  Letter by Jan Lievensens, letters-as-loot-corpus

Figure 13  An example of a signature in a larger handwriting than the body of  the letter, letters-as-loot-corpus

Figure 14  An example of a signature in a dif ferent and less experienced hand­writing than the body of  the letter, letters-as-loot-corpus

Figure 15  Jacob van de Velde’s handwriting and signature in letter 3b-1–2008, 131–132, letters-as-loot-corpus

Figure 16  Jacob van de Velde’s handwriting and signature in a receipt kept in the archive of  the Audit Of fice of  Zealand

Start

Y

(1)  Content expl. points to autographical?

N

(2)  Content expl. points to nonautographical? N

D Look for signatures or samples! N C Maybe profes. writer

Y

(4)  Experienced hand, low status?

Y

(3)  Same hand in letter by other sender?

N

(6)  Writer is salesman, captain, helmsman or wife

N

Y (5)  hand letter = hand signature? N

Y A  Autographical

B  Non-autographical

Figure 17  Flow chart to determine the autographical status of  letters

Y

Figure 18  The weaver’s song (Stadtarchiv Augsburg, Reichsstadt, Urgichten, 13.09.1591, Peter Müller/ Matthäus Sporhan)

Figure 19  The tailor’s song (Stadtarchiv Augsburg, Reichsstadt, Urgichten, 28.06.1594, Christof  Halbritter)

Studies in Historical Linguistics Edited by Dr Graeme Davis, Faculty of Arts, The Open University, UK, Karl A. Bernhardt, English Language Consultant with Trinity College London and for the London Chamber of Commerce and Industry International Qualification, and Dr Mark Garner, School of Language and Literature, University of Aberdeen, UK. Studies in Historical Linguistics brings together work which utilises the comparative method of language study. Topics include the examination of language change over time, the gen­etic classification of language, lexicography, dialectology and ety­ mology. Pronunciation, lexis, morphology and syntax are examined within the framework of historical linguistics. Both synchronic and diachronic approaches are used so that language is examined both at one time and across time. Historical Linguistics is still a young area of academic study, but it has its foundations in one of the oldest – philology. This series recognises both the seminal importance of philology, and the recent development through the conceptual framework provided by linguistic science. Vol. 1 Graeme Davis: Comparative Syntax of Old English and Old Icelandic. Linguistic, Literary and Historical Implications. 190 pages. ISBN 978-3-03910-270-9 Vol. 2 Graeme Davis (ed.): Dictionary of Surrey English. A New Edition of A Glossary of Surrey Words by Granville Leveson Gower with an Introduction by Graeme Davis. 186 pages. ISBN 978-3-03911-081-0

Vol. 3 Kim Schulte: Prepositional Infinitives in Romance. A Usage-Based Approach to Syntactic Change. 409 pages. ISBN 978-3-03911-327-9 Vols 4-5 Forthcoming. Vol. 6 Cynthia Lloyd: Semantics and Word Formation. The Semantic Development of Five French Suffixes in Middle English. 313 pages. ISBN 978-3-03911-910-3 Vol. 7 Tomonori Matsushita, A.V.C. Schmidt and David Wallace (eds): From Beowulf to Caxton. Studies in Medieval Languages and Literature, Texts and Manuscripts. 377 pages. ISBN 978-3-0343-0104-6 Vol. 8 Tadao Shimomiya: Alliteration in the Poetic Edda. 197 pages. ISBN 978-3-0343-0151-0 Vol. 9 Nils Langer, Steffan Davies and Wim Vandenbussche (eds): Language and History, Linguistics and Historiography. Interdisciplinary Approaches. 525 pages. ISBN 978-3-0343-0761-1