History of Linguistics 2002: Selected Papers from the Ninth International Conference on the History of the Language Sciences, 27-30 August 2002, Sao Paulo ... in the History of the Language Sciences)

History of Linguistics 2002 AMSTERDAM STUDIES IN THE THEORY AND HISTORY OF LINGUISTIC SCIENCE General Editor E. F. KO...

Author: Eduardo Guimaraes | Diana Luz Pessoa De Barros

135 downloads 1107 Views 1MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

History of Linguistics 2002

AMSTERDAM STUDIES IN THE THEORY AND HISTORY OF LINGUISTIC SCIENCE General Editor E. F. KONRAD KOERNER Zentrum für Allgemeine Sprachwissenschaft, Typologie und Universalienforschung Schützenstrasse 18/11, D-10117 Berlin Series III – STUDIES IN THE HISTORY OF THE LANGUAGE SCIENCES

Advisory Editorial Board Cristina Altman (São Paulo); Lia Formigari (Rome) Gerda Haßler (Potsdam); John E. Joseph (Edinburgh) Barbara Kaltz (Aix-en-Provence); Douglas A. Kibbee (Urbana, Ill.) Hans-Josef Niederehe (Trier); Emilio Ridruejo (Valladolid) Otto Zwartjes (Amsterdam)

Volume 110

Eduardo Guimarães and Diana Luz Pessoa de Barros (eds.) History of Linguistics 2002 Selected papers from the Ninth International Conference on the History of the Language Sciences, 27–30 August 2002, São Paulo – Campinas

History of Linguistics 2002 Selected papers from the Ninth International Conference on the History of the Language Sciences, 27–30 August 2002, São Paulo – Campinas

Edited by Eduardo Guimarães Universidade Estadual de Campinas

Diana Luz Pessoa De Barros Universidade de São Paulo

JOHN BENJAMINS PUBLISHING COMPANY AMSTERDAM/PHILADELPHIA

4-

The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences — Permanence of Paper for Printed Library Materials, ANSI Z39.48–1984.

Library of Congress Cataloging-in-Publication Data International Conference on the History of the Language Sciences (9th : 2002 : São Paulo-Campinas) History of linguistics 2002 : selected papers from the Ninth International Conference on the History of the Language Sciences, 27-30 August 2002, Sao Paulo-Campinas / edited by Eduardo Guimarães, Diana Luz Pessoa De Barros. p. cm. -- (Amsterdam studies in the theory and history of linguistic science. Series III, Studies in the history of the language sciences, ISSN 0304-0720; v. 110) Includes bibliographical references and index. 1. Linguistics--History--Congresses. I. Guimarães, Eduardo, 1948- II. Barros, Diana Luz Pessoa de. III. Title. P61.I57 2002 410.9--dc22 2007015636 ISBN 978 90 272 4601 1 (alk. paper) © 2007 – John Benjamins B.V. No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publisher. John Benjamins Publishing Co. • P.O.Box 75577 • 1070 AN Amsterdam • The Netherlands John Benjamins North America • P.O.Box 27519 • Philadelphia, PA 19118-0519 • USA

v

CONTENTS Part I FROM THE LATIN GRAMMARIANS TO THE IDÉOLOGUES Some Problems in Transferring the Latin Model to the First French Grammars Verbal voice, impersonal verbs and the –rais form ............................................ 3 Bernard Colombat Un exemple du transfert du modèle latin aux premières grammaires du français:l’analyse des temps du passé ......................................................... 17 Jean-Marie Fournier Linguistic Ideas and the Discourse about Languages in early Brazilian History ............................................................................................... 27 Bethania Mariani Grammaire générale et grammaire particulière: Les Méthodes de Claude Irson ................................................................. 41 Simone Delesalle & Francine Mazière Texts of Reference and Serial Texts in the Constitution of a Notional Paradigm: The example of the french idéologues .............................................................. 63 Gerda Haßler Part II LINGUISTICS IN THE 19TH AND 20TH CENTURIES The Brazilian Hyperlanguage Mark in the Traditional Grammar of the 19th Century ........................................................................................... 75 Marli Quadros Leite Revista Ilustrada: un document sur le langage des Noirs à la fin du XIXe siècle .................................................................................................. 87 Margarida Maria Taddoni Petter The “Arrested Evolution”: Notion, theories, myth? ........................................ 93 Ekaterina Velmezova

vi

CONTENTS

On the Defense of von Kempelen as the Predecessor of Experimental Phonetics and Speech Synthesis Research ...................................................... 101 Plínio A. Barbosa Positivism and Neo-positivism in Linguistics and Language Philosophy ...... 107 Paul Laurendeau From Universal Languages to Intermediary Languages in Machine Translation The work of the Cambridge Language Research Unit (1955-1970) .............. 123 Jacqueline Léon The Presence of Absence in Saussure’s Linguistic Theory .............................. 133 T. Craig Christy La création des cours de lettres au Brésil et les premières orientations de la recherche linguistique universitaire ........................................................ 141 José Luiz Fiorin The Portuguese Language in the Institutionalization of Linguistics .............. 149 Suzy Lagazzi-Rodrigues Humor and Language Acquisition: Anecdotal data and their route in the history of language acquisition studies .......................................................... 157 Rosa Attié Figueira Part III PLENARY PAPERS Reflexions sur l’experience grecque du langage ............................................. 171 Maria Helena de Moura Neves Johann Jacob Reiske (1716-1774): Leading Force in the Establishment of Oriental and Classical Scholarship in Germany ........................................ 183 Kurt R. Jankowsky The Context and Sense of Humboldt’s Statement that Language ‘ist kein werk (ergon), sondern eine tätigkeit (energeia)’ ............................... 197 Hans Aarsleff On the Notion of Structure and Structuralism in Brazil .................................. 207 Eni P. Orlandi Les avancées de notre discipline ..................................................................... 223 Sylvain Auroux

CONTENTS

vii

INDEX OF NAMES ..................................................................................... 235 INDEX RERUM .......................................................................................... 239

viii

ix

FOREWORD The Ninth International Conference on the History of Language Sciences (ICHoLS IX), held from August 27 to 30, 2002, in Brazil, was an initiative of the Campinas State University (Unicamp) and the University of São Paulo (USP) Linguistic Departments. The interest in holding this Conference in Brazil was due to the existence of research on the History of Linguistic Ideas in Brazil since the 1980s. Altogether, one hundred and six regular papers were presented at the meeting, covering a wide variety of themes in the history of language studies. In addition, the conference organizers invited six plenary speakers. Given the wide interest in the history of linguistics in Brazil, in addition to those who presented the result of their investigations, a large number of professors and postgraduate students were in attendance. From the many presentations that were actually made by our Brazilian colleagues, we selected those that we thought represented best current historiographical research activities, especially those related to developments in Brazil. At the same time, given the diversity of the countries of the scholars from abroad and their varying research interests, ICHoLS IX provided a multiple and diversified contact for those interested in the history of linguistic studies, always important in the progress of good research projects, as can be seen in the works now brought together in this volume. The papers presented at the Conference dealt with numerous domains of language study, involving the history of linguistic studies from Antiquity, the Middle Ages, the Renaissance and more recent periods. Many presentations were devoted to the 19th and 20th centuries. Among others, papers were dedicated to the history of concepts and terms, to aspects related to the history of the institutions, to language policies (including questions of prescriptive or normative attitudes). What was interesting about a number of them was the utilization of different points of view and different methodologies, thus providing stimulating discussions following both a good number of regular papers and several of the plenary lectures. This volume brings together a selection of the papers given at ICHoLS IX, organized under three headings. In the first part, papers are presented dealing with studies ranging from the Latin model in post-Renaissance grammars until new scientific propositions at the turn of the 19th century; the second part carries articles devoted to a great variety of subjects; in the third section, are united five plenary presentations ranging from ancient Greek reflections upon language to developments in Brazilian

x

FOREWORD

linguistics beginning with the implantation of structuralist work by Joaquim Mattoso Câmara (1904–1970) in the 1960s. In the concluding contribution, a survey of advances in the history of the language sciences is offered. A glance at the table of contents of this selection, made from the many papers presented and discussed at ICHoLS IX, reflects quite adequately, we are inclined to believe, the diversity of topics that had been offered on the market of linguistic ideas. And those wanting to know more about developments in Brazil in linguistic science and its more recent state may also learn more from the altogether seven contributions by researchers in this country. Despite the roughly chronological organization of the papers in all three sections, it is obvious that a thematic arrangement might also have been possible, such as strictly grammatical subjects, debates about language acquisition and language planning, general linguistic theory and language philosophy, the practical — and technical — analysis of language, and so forth. It is hoped that everyone interested in the history of the language sciences will find something of interest to their own work. At the General Assembly toward the end of the Conference, the University of Illinois at Urbana–Champaign, with Douglas A. Kibbee as its director, was chosen as the venue for ICHoLS X. (In the meantime, this very successful meeting was duly held on 1–5 September 2005.) Before closing, we would like to thank Unicamp and its Institute of Language Studies and the University of São Paulo and its College of Philosophy, Letters and Human Sciences for their all-important logistic support as well as the agencies of FAPESP, CAPES and the CNPq for providing the necessary financial assistance, thus ensuring for the successful realization of this internationally important Conference in Brazil.

January 2007 Eduardo Guimarães, Campinas Diana Luz Pessoa de Barros, São Paulo

1

PART I

FROM THE LATIN GRAMMARIANS TO THE IDÉOLOGUES

2

3

SOME PROBLEMS IN TRANSFERRING THE LATIN MODEL TO THE FIRST FRENCH GRAMMARS VERBAL VOICE, IMPERSONAL VERBS AND THE –RAIS FORM1

BERNARD COLOMBAT Ecole Normale Supérieure Lettres & Sciences Humaines 1. Introduction The grammatical description of Latin has been used for elaborating linguistic descriptions of many languages, especially for European vernaculars. What has been called the “Latin mould” or the “Latin model” was transferred to first descriptions of European and “exotic” languages, whatAuroux (1994) has called “grammaire latine étendue”: we can say “extended Latin grammar”. In this paper, I would like to study the transfer for two categories of French verb: voice and mood, with a special look at two problems: the question of impersonal verbs, which is connected to voice, and conditional forms in -rais. I have selected a few grammars that are representative of the French grammatical tradition (see Bibliography), but I will not do an exhaustive examination. In particular, I have not studied French grammars aimed at an English audience (about these grammars, see Kibbee 1979, 1991). The corpus is: – the first grammar of French: the Donait françois, written in England in 1409, – Palsgrave’s Eclaircissement de la langue française, considered a very important grammar, but with little influence in France (1530), – A Grammatica Latino-Gallica written in Latin, by Sylvius (1531), – French grammars written in Latin for German pupils: Pillot (1550/1561), Garnier (1558), Cauchie (1586), Serreius (1598/1623), – French grammars written in French: Meigret (1550), Estienne (1557), Ramus (1562/ 1572), Bosquet (1586). I particularly would like to stress a few points: – the differences between both structures to study, – specific problems when grammarians wanted to use a Latin description for the French language, – first solutions found by these early grammarians. 1

I am grateful to Douglas Kibbee for advice and help in the translation of this paper.

4

BERNARD COLOMBAT

2. Latin heritage and problems for the French language 2.1 Latin heritage In the past as now, grammarians link grammatical categories to parts of speech. For example, according to Riegel, Pellat & Rioul (1999:244-245), morphological categories linked with French verbs are: person, number, mood, tense, aspect, voice or diathesis. The Latin tradition used the terms accidit, accidunt, accidens, ‘accidents’ of verb: we mention only – to simplify – two authors considered to be the most important by the later tradition, that is Donatus and Priscian. Donatus: seven accidents: qualitas (quality), coniugatio (conjugation), genus (gender), numerus (number), figura (‘figure’, that is composition), tempus (tense), persona (person): and Donatus subdivides qualitas into modus (mood) et forma (verbal suffixation). Priscian: eight accidents: significatio uel genus (meaning or gender), tempus (tense), modus (mood), species (‘species’, that is morphological derivation), figura (that is composition), conjugatio (conjugation) et persona (person) cum numero (with number).

For general outlines, these models are still the same in Latin grammar of the sixteenth century: Donatus’ outline is rather more in Northern Europe (for example by Despauterius), and Priscian’s model is used by Italian grammarians (for example Perotti). Accidentia, ‘accidents’ that interest us here, are gender (or meaning), person (in particular impersonal verbs) and mood.

‘Gender’ For ancient grammarians, there are five distinct genders of verbs: – active: ending in –o and capable of passive transformation (amo, ‘I love’), constructed with an accusative or another case, – passive: ending in –or and capable of active transformation (amor, ‘I am loved’), – neuter: ending in –o, but not suffering passive transformation (eo, ‘I go’), – common: criminor te ‘I accuse you’ and criminor a te ‘I am accused by you’, – deponent: passive form, active meaning; some can be constructed with the accusative (sequor + accusative, ‘I follow’). This classification is complex and combines three criteria, morphological, syntactic, semantic. In fact, there are semantic constraints: for example, according to Priscian (following Apollonius Dyscolus), aro ‘I plough’ is not an active verb, because it is not

PROBLEMS IN TRANSFERING THE LATIN MODEL

5

possible to say aror ‘I am ploughed’, except per figuram, that is as a figure of speech. So, for Priscian, it is a neuter verb. Voice and verbal construction are closely linked, but there are a lot of problems; for example, sequor ‘I follow’ has passive form, but it has an active meaning, it is constructed with an accusative, but cannot be transformed into a passive; uapulo ‘I am hit’ has an active form, but a passive meaning. For us, a same verb can be active or passive, and we have only three categories (active, passive, deponent), but for ancient grammarians, there are five distinct categories. And because the problem is complex, nowadays grammarians of Latin separate voice (for morphological features) and ‘diathesis’ (for semantic analysis). Impersonal verbs In the sixteenth century, grammarians of Latin distinguish two categories of impersonal verbs: – impersonal actiuae uocis ‘with active form’: ex. pluit ‘il pleut’, ‘it is raining’, – impersonal passiuae uocis ‘with passive form’: ex. amatur ‘on aime’, ‘people love’. Moods Ancient Latin grammarians have five moods, indicative, imperative, optative, conjunctive (Donatus) or subjunctive (Priscian), infinitive (infinitiuus for Donat, infinitus for Priscian). Optative, inherited from Greek grammar, is morphologicaly mixed up with the subjunctive, but ancient grammarians (Serbat 1978:265-266) make a distinction by associating the optative with utinam, the subjunctive with the conjunctions cum or si. Moreover, the optative alone can form a complete sentence, but the subjunctive cannot (Serbat 1978:268). The form in -ero (for us today, future perfect of the indicative) is considered by ancient grammarians as a future subjunctive, because it always appears in a subordinate clause (Serbat 1978:271). 2.2 Problems for French ‘Genera uerborum’ vs ‘voix’, voice In French, as in English too, nowadays, the category called genera uerborum (genders of verbs) by ancient grammarians is called ‘voix’ in French, ‘voice’ in English, and in French, we have two or three voices, that is: ‘voix active’, active voice; ‘voix passive’, passive voice; ‘voix pronominale’, pronominal voice. But, since 1975, the new official term has been “tournure pronominale” (see Riegel, Pellat & Rioul 1999:245). Impersonal verb In French, we have:

6

BERNARD COLOMBAT

– an impersonal, as in Latin: il pleut, similar to the Latin impersonal verb with active form pluit, – and an expression with on, similar, for the translation, to the impersonal with passive form in Latin: on aime = amatur. But nowadays, for French speakers, on aime is never felt as impersonal nor as passive, on “renvoyant à une personne ou à un ensemble de personnes d’extension variable, que le locuteur ne peut ou ne veut pas identifier de façon plus précise” (Riegel, Pellat & Rioul 1999:197). Moods We will stress only one problem: the disappearance of the optative and the appearance of a ‘conditional’. Nowadays, neither in French grammar nor in Latin grammar, grammarians recognize an optative mood, but, in French, we have specific forms for a ‘conditional’ (present: -rais: past: -rais + past participle). For a very long time, these forms were considered as a specific mood, but today, most grammarians consider the form in –rais as a tense of the indicative (Riegel, Pellat & Rioul 1999:315316). We can see that the differences between both languages are big enough to allow grammarians to elaborate different, and even conflicting descriptions. In grammars of the XVIth Century, the presentation of features in terms of ‘accidents’ is general. In the Donait françois, a general treatment of ‘accidents’ precedes treatment of parts of speech. Sylvius follows Donatus’ model, but the other grammarians generally follow Priscians’ presentation. A very important distinction for grammarians is the distinction between personal verbs and impersonal verbs: it is usually the first distinction. Most grammarians give specific paradigms for impersonal verbs. Two very important verbs, être (to be) and avoir (to have) are given at the beginning. I guess (but I am not sure) that Pillot is the first grammarian (in the French grammatical tradition) to call these verbs ‘auxiliary verbs’.

3. Verbal voice We will stress particularly general features and selected options. 1. The general distinction is between three ‘genders’ out of the five genders of the Latin tradition: active gender, passive gender, neuter gender. For example, at the end of the treatment of accidents, in the Donait françois:


7

Quantz geners est il des verbes? — Trois. — Qelx? — Le actif, c’est a dire faisant, come ‘je ayme’: le passif, c’est a dire seuffrant, come ‘je suys amé’: le neutre, c’est a dire ne le un ne l’autre clerement, sicome ‘je dois’. (Donait françois:f.318vb)

Sylvius, who wanted to write a Latin-French grammar (the tittle of his book is Grammatica Latino-Gallica), thinks that French has three genders, that is three meanings: active, passive, neuter, “or, as Greek people say, middle (mediam)” (1531:115). He remarks, quite correctly, that these genders are distinguished, not by form (uox), but by meaning (significatio). Actually, French has, unlike Latin, no specific endings for passive. He examines verbs known as neutropassiua: for example exulo, uapulo, and says that “they keep the same meaning in French”: ge sui banni, batu. But he doesn’t notice that these verbs, in French, are normally passive. Meigret, following Priscian, stresses the importance of the opposition between active and passive: “La signification consiste proprement en action ou passion” (1550:H.23.1). He accurately remarks that, although aller and venir are conjugated with être, they are not truly passive. The tripartition into active, passive, neuter is generally well-admitted: we find it by Pillot (1561[1550]:72), R. Estienne (1557:32), who distinguishes three ‘sortes’ [sorts] (sic) of verbs: active, passive, neuter, plus substantive verb, and Bosquet (1586:78). 2. The previous presentation doesn’t allow one to deal with pronominal constructions. On the contrary, Palsgrave’s triple distinction into actyves, meanes, passyves, with his analysis of pronominal verbs as meanes “middle” (ed. Baddeley 2003:345, 348-350, 455-456, 460), allows him to solve this problem (see Stéfanini 1962:23 sq.): The acte of the mean verbes passeth nat from the doar, but retourneth to the doars self agayne, or is done within the parsone of the doars selfe. (Palsgrave 1530, Introductyon of the Authour to his seconde boke, f.C.iiv, ed. Baddeley 2003:(66)/349)

3. Others grammarians keep only two genders of the verb. Garnier says: Genders, moods, tenses, numbers and persons of verbs are exactly the same in Latin and in French, except that French people have not passive verbs. (Garnier 1558:44-45)

It is not false if we consider that the sentence means that “French [unlike Latin] has no specific morphological marks to form the passive”. Ramus (1562) only keeps two sorts of verbs: the active verb can form an entirely passive participle: the neuter cannot:

8

BERNARD COLOMBAT Lê verbê neutrê e’ sêlui ci nê peut former un partisipê entierêment passif, comê Aler, Vênir formêt seulêment Alant, Venant: Car Ale e Vênu, nê son’ point entierêment pasifz, combien c’ilz en eiêt lê semblant. (Ramus 1562:51)

The formulation is more elliptical, but clumsier in the 1572 edition: Le verbe neutre cest celuy qui ne peult former vng participe preterit: comme Rire, Dormir, Mentir, forment seulement Riant, Dormant, Mentant. (Ramus 1572:79)

We don’t know how Ramus would analyse the forms ri, dormi, menti. Serreius in the 1598 edition says that there are two genders, active and neuter. Passiuum non habemus sed circumscribere per verbum substantiuum & praeteriti perfecti vocem cogimur, quae uariatur iuxta casus, genera & numeros. [...] Neutrum verbum est quod vnicum format participium in ant: vt, dormir, dormant: aller, allant: venir, venant. Quae licet formare videantur participia allé, venu: tamen quia accessione substantiui nulla passio significatur, actiua non habentur. Nam cum dicis, Ie suis venu, ie suis allé, participia vera non sunt, sed praeteriti temporis verba, nata ab infinitis aller, venir. (Serreius 1598:61) “We have no passive, but we are forced to express it by circumlocution with the substantive verb and the form of perfect past, which varies according to case, gender and number. [...] Neuter verb is one which forms only a participle in ant, for example dormir, dormant: aller, allant: venir, venant. In fact, although they seem to form participles allé, venu, yet, because no passion is expressed by the adjunction of the substantive verb, they are not considered as active verbs. In fact, when you say Ie suis venu, ie suis allé, there are not real participles, but verbs at the past tense, born from the infinitives aller, venir.”

4. Last position: to exit out of verbal gender Cauchie (1586:f.34r-35v) considers that, except for the verb être, all verbs express an action. Variation of this action (transitive, reciprocal, absolute) allows a more syntactical subdivision of verbs into transitiua, reciproca, absoluta et uariè accepta. But there is no passive verb in French, just a specific use of the verb with circumlocution: In uocem passiuam nostra uerba non mouentur, sed illam periphrasticè explicamus, quod etiam quibusdam temporibus Latini Graecique factitant. (Cauchie 1586:f.35v) “Our verbs are not moved to passive voice [form], but we explain this by circumlocution, what Latin and Greek people often use to do at certain tenses.”

Maupas talks, without other explanation, of ‘disposition’ of verbs: this disposition can be active, reciproquee, ou reflechie, neutre (Maupas 1618:f.98; see Stéfanini 1962:39 sq.). A very important idea is that the same verb can have different dispositions: there are not strict categories. There is another frame which will allow the creation of a syntax for the French verb.


9

4. Impersonal verb 4.1 Dat [1] Donait françois 320va-b: “L’autre maniere [du verbe] est appellé verbe impersonel, pour ce qu’il n’a point de nominatif case du quel il prendroit nombre et personne. Et sçachéz qu’il y a deux paires de verbes, c’est assçavoir actif et passif. Mais le actif aura devvant luy ou aprés un de ces deux seignes ‘l’en’ ou ‘on’, sicome ‘l’en dit’, ‘on dit’: ou ainsi: ‘dit l’en’, ‘dit on’. Et le passif impersonel aura devant luy ou aprés cest seigne ‘il’, sicome ‘il est dit’, ou ainsi ‘est il dit’. Aussi cest seigne ‘il’peut estre actif, sicome ‘Il fait sçavoir que il y a grande joie en paradis’, et ainsi ‘fait’ et ‘a’ sont devant mis pour cest verbe ‘est’.” [2] Palsgrave 1530 (Seconde boke, f.xxxviir; ed. Baddeley 2003:(138)/460): “Verbes imparsonalles be suche as through al theyr tenses have but the thyrde parson singular onely.” [3] Sylvius 1531:113: impersonal is not a mood, but a verb: it is expressed by hom or l’hom: curritur = hom ceurt. [4] Meigret 1550 (H.23.7) remarks that we can express passive by active with ‘réciproque’: le vin se boêt, so that there are ‘indeterminate passive’ verb (‘passifs indéterminés’), equivalent to the construction with an ‘indeterminate subject’ (‘surposé indéterminé’) as on: on boêt le vin. In chapter 7, devoted to person (H.28.6): impersonal verbs “are all of third person and have il or on for their subject” (“sont tous tierces personnes: et ont il ou on pour leur surposé”). [5] Pillot 1561[1550]:72: “Impersonale rursum duplex, Actiuae uocis, cuius nota est Il [...]. Passiuae uocis verbi impersonalis nota est On.” “The impersonal verb has two forms, of active voice, whose mark is Il […]. The mark of impersonal verb of the passive voice is on.” [6] R. Estienne 1557:32: “[Les impersonnels] sont de deux sortes en Latin: les uns finissent en t, pour lesquels expliquer & rendre en François, on prepose il, comme Oportet, il fault [...]. Les autres se terminent en tur. a tels pour les exposer en François, on prepose on: comme, Amatur, on aime.” [7] Garnier 1558:77: “Habent [...] Galli etiam verba impersonalia, eáque duplicia, actiuae vocis scilicet & passiuae, quemadmodum apud Latinos. Actiuis semper praeponimus hanc particulam il, passiuis verò on.” “French people […] have impersonal verbs, and they are of two sorts, that is of active voice and passive

10

BERNARD COLOMBAT

voice, exactly as for Latin people [sic]. Before the active , we always put this particle il, and before the passive, <we put> il.” [8] Ramus 1562:50: the impersonal is conjugated only in the third person singular: Faut, çaut. 1572:78: idem. Syntaxe, 1562:107-108, 1572:170: “Le verbe Latin impersonnel de voix [1562: voes: 1572: voe / voix] actiue est explicque par Il, et de voix passive par, On.” [9] Bosquet 1586:68 [sic, for 72]: “De quantes sortes est l’impersonnel? De deux sortes, à sçauoir Actif (lequel a pour note, & indice-Il) comme il conuient. Et le Passif, ou passionné (quy aussy a sa notte-on) comme on chante.” [10] Cauchie 1586:f.65v: “Impersonalia tertiae personae singu. terminationem sequuntur, sed de certa persona non dicuntur. Sunt autem duplicia, Primitiua quorum nota est il, & Deriuatiua, quorum nota est on.” “Impersonal follow the ending of the third person singular, but they are not used for a specific person. There are two sorts, Primitive, for which the mark is il, and Derived, for which the mark is on.” [11a] Serreius 1598:116: “Duplicia Gallis impersonalia sunt: actiua & passiua. Actiuis praeponimus voculam Il, vt Il faut, Il conuient: passiuis on vel l’on vt on aime, on lit.” “French people have two categories of impersonal verbs, active and passive. Before active impersonal verbs, we put the particle il, as Il faut, Il conuient: before passive we put on or l’on as on aime, on lit.” [11b] Serreius 1623:91: “Impersonalia vel suntActivae significationis quibus in coniungando praefigitur Il [...], uel Passivae significationis quibus praeponitur On uel L’on.” “Impersonal are either of active meaning (in the conjugation of which il is put before the verb) […], or of passive meaning (in which case on or l’on is put before).” [12] Maupas 1618:f.124r: “Nous avons deux natures d’impersonnels, l’une de sens passif, qui est fait de cette syllabe on ou l’on, appliquee devant toute tierce personne singul. de quelque verbe que ce soit [...]. La seconde nature d’impersonnels, est de voix active, au moyen de cette particule il, preposee à plusieurs verbes, non pas à tous universellement comme la devant dite [...].” 4.2. Commentary The question of the impersonal is considered important by all the grammarians, so that some give complete paradigms, for both ‘voices’.


11

The general position is to consider that the form with on, because it corresponds to the Latin passive form in –tur, is of the passive form of the impersonal voice, whereas there is nothing passive in on, at least morphologically speaking. The forme with Il is considered as impersonal of the active form of the impersonal voice, because il pleut corresponds to pluit. Pillot [5] and Garnier [7] have both committed this mistake. Bosquet [9] and Serreius 1598 [11a] don’t use the term uox, but say only ‘active and passive impersonals’. Serreius 1623 [11b] uses the terms Actiuae significationis ‘of active meaning’, Passiuae significationis ‘of passive meaning’, which avoids the unfortunate term uox, but the meaning of on aime is passive only if you have the Latin model in mind. Latin structure is always the reference, and most of these books are written in Latin. Nevertheless, the presentation was exactly the opposite in Donait françois [1]: the form on dit is called active, and the form il est dit is called passive. But il can be active too, as in the example: il fait [faut] savoir), so that criterion is not distinctive. Some grammarians were more skilful: Estienne [6] clearly separates Latin data and French data and avoids using terms actif and passif. These terms are used by Ramus [8], for qualifying voix — it’s probably one of the first occurrences for ‘voix active / passive’ in French —, but only for Latin verbs, not for French verbs. Cauchie [10] says only ‘impersonnels primitif et dérivé’, but he doesn’t justify this terminology: perhaps because the form with on is considered a more recent development than the form with il, or because it is considered more remote from the corresponding Latin form? Maupas’ presentation [12] is interesting: Maupas uses active voice (voix active) for il, but he avoids the term voix for on, and instead uses the term sens (meaning). His formulation shows that, at the beginning of the XVIIth Century, grammarians felt that using the term voix is inappropriate when it would be used for something different from a form. 5. The – rais forms We will take for example the paradigm of verb avoir [to have] whose conjugation is generally the first given in the books we are studying. We will choose books that give clear paradigms, in a spelling that can be understood by a modern linguist. In the sixteenth century, there are many spellings for the form of conditional -rais (roi, -rois, -roy, -roye, and -rée [réè] by Sylvius and -roê by Meigret), but the form of conditional and the form of the future (–rai, or –ray) cannot be confused.

12

BERNARD COLOMBAT

Many grammarians multiply forms and think that many forms are common for several moods. For example, Meigret thinks that “le subjonctif ou conjonctif [est], pour la plus grande part de ses temps, égal ou semblable à l’optatif” (1550: H.24.20), and he gives thirteen possible forms for the subjunctive, as Kibbee (1979:234-235) has noted. Tenses of optative and con- / sub-junctive: habeo, avoir [to have].

Table 1: The situation in Latin, in the tradition inherited from Priscian

Table 2: Sylvius 1531:130-132

Table 3: Pillot 1561:76-82 (1550:f.22r-23v)

Table 4: Estienne 1557:38-40


13

Table 5: Cauchie 1586:f.41r-42r

Table 6: Serreius 1623:55-57 (Serreius 1598 follows Estienne 1557 exactly)

1. When the -rais form is considered as an optative (bold type, underlined) Sylvius [Table 2] isolates well the form g’ha(i)rée, but he does not have the form with the past participle, perhaps because g’ha(i)rée seems to correspond exactly with haberem, but j’aurais eu (that he would have written g’ha(i)rée heu) has no morphological correspondence in Latin. Likewise Cauchie [Table 5] has j’aurois eu as a conjunctive, but not as an optative. Pillot [Table 3] alone (with Garnier, who is not in our tables) doesn’t analyse the rais forms as optative. Estienne [Table 4] et Serreius [Table 6] show the optative meaning of the rais form by introducing them with O que volontiers. The -rais form is analysed by Sylvius, Cauchie and Serreius both as present and imperfect, but Estienne only considers it a present. In fact, our conditional ‘present’ corresponds both with potential and unreal present. Grammarians could have connected with utinam sim, but nobody did. By Cauchie [Table 5], there is a very interesting remark that the –rais form does not inevitably correspond with a wish: “sine optandi forma”, without form of wishing. He gives as examples: “j’auroi tort de lui faire déplaisir sans occasion. Pourquoi auroi-je tant d’envieux que vous dites? je n’ai garde, pauvreté n’apporte point d’envie”, and he gives this explanation: Cum optamus quippiam adiicitur voluntatis adverbium sic j’auroi volontiers, ich wolt gern haben. (Cauchie 1586:f.41r)

14

BERNARD COLOMBAT “When we wish, an adverb of wishing is added, as j’auroi volontiers, ich wolt gern haben.” So, he prepares the analysis of the conditional as not being inevitably connected with a wish.

2. When the -rais form is considered as a subjunctive or a conjunctive (bold type, not underlined) All the grammarians think that the -rais form can be a subjunctive or a conjunctive, and they analyse it as an imperfect introduced by quand [when]. Cauchie alone considers this form as common to the present and to the imperfect: it’s a very good intuition, because this form corresponds to that we call a conditional (potential or unreal), in a quite rare use of quand with a concessive value (quand j’aurais…, that is quand bien même j’aurais..., même si j’avais…, ‘even if I had…’), already attested at the time (Fournier 1998:356). But this use is relatively rare, and we can ask ourselves why it was given as the general example for the form in -rais. The only explanation that I can give is that the French form quand j’aurais seemed to correspond exactly to the Latin form cum haberem, whose use was very frequent, but whose meaning was much broader than that of the French form. 6. Conclusion It was a long road from the Latin tradition to the description of French grammatical categories today. The first French grammars are interesting because they show how difficult the adaptation of the Latin model to another language was. Here we have emphasized three points. 1. It was difficult to give up Latin verbal gender. Most grammarians have trouble analysing the French periphrastic phrase être + past participle as a passive, because, for them, passive must be marked by a specific ending, as -or in Latin. The result is that many of them think that there are only two genders of verbs in French: active and neuter. 2. The invention of ‘voix verbale’ is certainly due to a misunderstanding. But this misunderstanding is double: firstly, because French is without ‘voice’, that is without specific form (uox) to express diathesis: nevertheless, it is this term ‘voix’ which will be used. And secondly, because the use of the term ‘passive’ for the expression on aime was inappropriate, since only Latin amatur is of passive form. 3. It was difficult to analyse forms of the conditional. At the beginning, the -rais form has been analysed only in two very limited uses: firstly, to express wishes, as an optative, secondly, after concessive quand, that allows the use of conditional, but whose frequency is very low. But the French conditional has many other uses.


15

In grammar (as in other fields), the transfer of terminology is not so easy and, in the sixteenth century, there are still much to do to provide a good description of the French language. REFERENCES

Primary sources Bosquet, Jean. 1586. Elemens ou Institutions de la langue françoise. Mons: Charles Michel [Reprint: Genève: Slatkine, 1972.] Cauchie, Antoine Caucius, Antonius. 1586. Grammaticae Gallicae libri tres. Argentinae [Strasbourg]: Bernhardus Iobinus [Reprint: Genève: Slatkine, 1968]. See Demaizière 2001a. Donait françois. See Städtler 1988.128-137 [DonatOxf] and Swiggers 1985. Estienne, Robert. 1557. Traicté de la grammaire françoise. [Genève]: L’Olivier de Rob. Estienne [Reprint: Genève: Slatkine, 1972]. Garnier, Jean. 1558. Institutio Gallicae Linguae, In usum iuuentutis germanicae. Genève: Jean Crispin [Reprint: Genève: Slatkine, 1972]. Maupas, Charles. 1618 [1607]. Grammaire et syntaxe françoise. Orléans: O. Boynard & J. Nyon. Meigret, Louis. 1980 [1550]. Le Traité de la Grammaire française. Paris: C. Wechel [Reprint: Genève: Slatkine, 1972]. See Hausmann 1980. Palsgrave, John. 1530. Lesclarcissement de la langue francoyse. London: Richard Pynson & Johann Haukyns [Reprint: Genève: Slatkine, 1972]. See Baddeley 2003. Pillot, Jean. 1561 [1550]. Gallicae linguae institutio. Paris: E. Groul[l]eau, A. Wechel. See Colombat 2003. Ramus, Petrus. 1562. Gramerê. Paris: A. Wechel [Reprint: Genève: Slatkine, 1972]. . 1572. Grammaire. Paris: A. Wechel [Reprint: Genève: Slatkine, 1972]. See Demaizière 2001b. Serreius, Jo(h)annes. 1598. Grammatica Gallica. Strasbourg: A. Bertram. . 1623. Grammatica Gallica Noua [...] Editio sexta. Strasbourg: héritiers de L. Zezner. Sylvius, Iacobus Ambianus. [Dubois, Jacques]. 1531. Isagùge and Grammatica Latino-Gallica. Paris: R. Estienne. See Demaizière 1998. Secondary bibliography Auroux, Sylvain. 1994. La révolution technologique de la grammatisation. Liège: Mardaga. Baddeley, Susan ed. 2003. John Palsgrave, L’éclaircissement de la langue française (1530), texte anglais original, traduction et notes. Paris: H. Champion. Colombat, Bernard. 1999. La grammaire latine en France à la Renaissance et à l’Âge classique: théories et pédagogie. Grenoble: Ellug. . ed. 2003. Jean Pillot, Institution de la langue française (1561), texte latin original, introduction, traduction et notes. Paris: H. Champion.

16

BERNARD COLOMBAT

Demaizière, Colette ed. 1998. Jacques Dubois (Sylvius), Introduction à la langue française suivie d’une grammaire (1531), texte latin original, traduction et notes. Paris: H. Champion. . 2001a. Antoine Cauchie, Grammaire française (1586), texte latin original, traduction et notes. Paris: H. Champion. . 2001b. Pierre de La Ramée, Grammaire (1572), édition commentée. Paris: H. Champion. Fournier, Nathalie. 1998. Grammaire du français classique. Paris: Belin. Hausmann, Franz Josef .1980. Louis Meigret, humaniste et linguiste. Tübingen: Gunter Narr. . ed. 1980. Meigret, Louis, Le Traité de la Grammaire française (1550). Le Menteur de Lucien. Aux lecteurs (1548). Tübingen: Gunter Narr. Kibbee, Douglas A. 1979. The Establishment of the French Grammatical Tradition, 1530-1580. Thesis (Ph.D). Indiana University. . 1991. For to Speke Frenche Trewely. The French Language in England, 1000-1600, its Status, Description and Instruction. Amsterdam & Philadelphia: John Benjamins (SiHoLS 60). Riegel, Martin, Pellat Jean-Christophe & Rioul, René. 1999. Grammaire méthodique du français. Paris: Presses Universitaires de France. Serbat, Guy. 1978. “Le ‘futur antérieur’ chez les grammairiens latins”. Varron, grammaire antique et stylistique latine ed by J. Collart et al., 263-272. Paris: Les Belles Lettres. Städtler, Thomas. 1988. Zu den Anfängen der französischen Grammatiksprache. Textausgeben und Wortschatzstudien. Tübingen: Niemeyer. Stéfanini, Jean. 1962. La voix pronominale en ancien et en moyen français. Aix en Provence: Publications des Annales de la Faculté des Lettres, Ophrys. Swiggers, Pierre. 1985. “Le Donait françois: la plus ancienne grammaire du français, éd. avec introduction”. Revue des Langues Romanes 89.235-251.

17

UN EXEMPLE DU TRANSFERT DU MODÈLE LATIN AUX PREMIÈRES GRAMMAIRES DU FRANÇAIS L’ANALYSE DES TEMPS DU PASSÉ

JEAN-MARIE FOURNIER Université de Paris III C’est une idée depuis longtemps établie et abondamment illustrée que les premiers grammairiens qui s’efforcent de décrire le français, comme d’autres langues vernaculaires à partir de la Renaissance, procèdent en utilisant les catégories et les notions développées par la tradition plusieurs fois séculaire des grammairiens du latin. La grammatisation des langues du monde s’est opérée d’abord par le moyen d’un véritable transfert du modèle que constitue la tradition latine, et son application à des langues objets différentes. Dans le domaine de la description et de la théorisation du temps verbal, comme en beaucoup d’autres, les auteurs se tournent manifestement vers Priscien, et ils essaient d’appliquer au français le classement des temps, et la terminologie de l’Institutio linguae latinae. L’objet de cet exposé est de montrer comment s’opère ce transfert et quels problèmes émergent du fait du déplacement des notions élaborées pour le latin et de leur application à une langue objet différente. Il s’agit notamment de l’identification des tiroirs verbaux, et de l’émergence d’un problème linguistique dont l’histoire est aussi longue que la tradition grammaticale française elle-même: la question des valeurs respectives du passé simple et du passé composé. Je retracerai les principales étapes de cette histoire jusqu’à la grammaire de Maupas (1607). 1. Le temps chez Priscien L’analyse du temps linguistique qui apparaît dans le texte de Priscien correspond à un classement des formes verbales articulé en genres et espèces, distinguant deux niveaux de catégories. Le premier niveau comprend les trois temps fondamentaux que sont le présent, le prétérit et le futur. Le second niveau identifie 3 autres catégories qui constituent des espèces du prétérit: le prétérit imparfait, le prétérit parfait, et le prétérit plus que parfait. L’ensemble comporte donc au total 5 temps. Les définitions des trois espèces du prétérit tendent à identifier un critère unique, sinon simple, susceptible de distinguer ces trois sous-catégories. Ainsi, le preteritum imperfectum est défini comme le temps “dans lequel une chose commence d’être faite et n’est pas cependant achevée

18

JEAN-MARIE FOURNIER

(perfecta); le preteritum perfectum, signifie un temps “dans lequel la chose est montrée comme achevée”, enfin le preteritum plus quam perfectum signifie un temps “dans lequel une chose est montrée comme achevée depuis longtemps”. Du point de vue terminologique, le terme central est celui de perfectum, à partir duquel sont formés les dérivés et composés imperfectum, et plus quam perfectum. Cette homogénéité terminologique correspond à la tentative d’identifier un critère unique et discriminant, que nous pourrions appeler la ‘perfection d’action’. Notons qu’il ne s’agit pas pour Priscien d’un critère binaire (du type achevé/inachevé), mais qu’il comporte ce qui semble fonctionner ici comme des degrés: ainsi le temps de l’imperfectum est celui de l’inachevé, mais il marque aussi l’action qui commence d’être faite (coepit geri); tandis que la valeur du plus quam perfectum cumule l’achèvement et l’éloignement dans le passé. Dans les deux cas, il est clair que le trait binaire achèvement /inachèvement ne fonctionne pas de façon pure, mais on pourrait dire que Priscien fait en quelque sorte comme si. L’objectif, me semble-t-il, est de conserver entre les différentes catégories de temps un même type de relation. Les catégories doivent être les unes à l’égard des autres dans un rapport de genre à espèce, l’identification des différentes espèces d’un genre reposant sur celle d’une différence spécifique. 2. La question chez les grammairiens français du 16eme siècle Je vais examiner maintenant les systèmes adoptés par les premiers grammairiens qui tentent de décrire le français, en me limitant pour l’essentiel à la question du classement et des critères mis en œuvre dans les définitions de différentes formes de passé. 2.1. Pillot. Institutio linguae gallicae (1550) Le classement de Priscien est pris ici comme un point de départ, par rapport auquel il s’agit de situer le français. Pillot note en effet que si le latin a 5 temps verbaux, le français s’en distingue par le fait qu’il a deux preteritum perfectum. C’est là un point capital présenté d’une façon à peu près identique chez la plupart des auteurs. Tout se passe comme si le français introduisait, dans une classification des temps partiellement identique à celle que présente le latin, une distinction supplémentaire, constituant le preteritum en genre sous lequel se rangent deux espèces. Il y a là l’émergence et la cristallisation d’un problème linguistique dont s’occupera la tradition grammaticale française sous une forme à peu près identique: identifier la différence spécifique qui justifie l’identification de deux espèces au sein du preteritum perfectum, le choix des termes susceptibles de dénommer ces deux catégories étant lui même lié à l’interprétation théorique de cette différence. Pillot propose de les dénommer respectivement, preteritum perfectum indefinitum et preteritum perfectum definitum. La constitution du problème linguistique de la valeur des deux prétérits du français, l’un par rapport à l’autre, de la distribution de leurs emplois, est donc directement liée

UN EXEMPLE DU TRANSFERT DU MODÈLE LATIN

19

au transfert de la classification des temps de Priscien et à son application au français. Son émergence résulte d’une part du type de classement mis en œuvre, et d’autre part de la relation établie entre les formes du latin et leur équivalent, leur traduction en français. En quoi consiste la différence spécifique des prétérits parfait du français telle que la conçoit Pillot ? Le texte comporte l’amorce d’une définition. Le parfait indéfini signifie un temps indéterminé (tempus indeterminatum significat). Le parfait défini en revanche note un temps “plus déterminé” (magis determinatum), et qui n’est pas passé depuis aussi longtemps (non adeo dudum praeteritum). Il s’agit donc, semble-t-il, d’un critère temporel, non aspectuel, qui met en jeu une quantification de l’éloignement dans le passé. Ces définitions sont illustrées dans l’édition de 1550 par des exemples qui apportent plusieurs éléments nouveaux. On peut dire j’ai lu aujourd’hui l’Evangile, je lus hier l’Evangile, mais on ne peut pas dire j’ai lu hier l’Evangile, je lus aujourd’hui l’Evangile. Ces exemples vont en effet un peu au delà de ce qu’énonce explicitement la définition. Ils illustrent l’impossibilité d’établir une corrélation entre le passé simple et l’adverbe aujourd’hui, et entre le passé composé et hier. Autrement dit, il s’agit de la première mention de faits relatifs à ce que la tradition appellera la règle des 24 heures (cf. sur ce point Weinrich 1973; Y. Galet 1997). 2.2. Meigret (1550) On trouve chez Meigret également l’idée selon laquelle l’analyse des temps et leur classification relève d’une procédure de division des catégories, ce qui revient à manifester des rapports d’inclusion entre espèces et genres. Ainsi, le passé, à la différence du présent par nature indivisible, comporte en français plusieurs divisions: – Le prétérit imparfait: “il ne dénote pas un accomplissement ne perfection d’une action ou passion passée mais tant seulement avoir été commencée”. (Meigret 1550: 89) La formule est clairement une sorte de traduction du texte de Priscien. – Un autre prétérit qui dénote une action “un peu plus parfaite, duquel toutefois le temps n’est pas bien déterminé de sorte qu’il dépend de quelque autre comme je vis le Roi lorsqu’il fut couronné.” (Meigret 1550) On retrouve ici les critères énoncés par Pillot en termes de détermination et d’achèvement. Mais la piste suivie par Meigret est différente. La détermination revêt une dimension syntaxique, ou anaphorique. Le passé simple est un temps déterminé parce qu’il n’est pas autonome syntaxiquement et référentiellement; la référence temporelle d’un verbe au passé simple se construit toujours dans une relation avec une autre marque temporelle, verbale comme dans l’exemple, ou autre. – Enfin le système des temps du français comporte une troisième forme, de création française, et qui répond à la nécessité où l’on est parfois de parler du

20

JEAN-MARIE FOURNIER

passé de façon absolue, et “sans suite” (Meigret 1550.). Il s’agit du passé composé. On peut noter que la question de la relation entre le parfait latin et les formes françaises, est posée de façon un peu différente qu’elle ne l’est chez Pillot, puisque Meigret semble souligner ici que le système des temps du français s’écarte du dispositif latin en ce qu’il comporte une forme supplémentaire, sans équivalent dans la langue ancienne. Meigret signale en outre un peu plus loin que le français comporte encore trois autres prétérits; le plus que parfait, une autre forme de prétérit parfait et déterminé, j’ai eu aimé, un autre encore, formé du prétérit parfait indéterminé, j’eus aimé. Tous imposent une corrélation avec une marque du passé, adverbiale, circonstancielle, ou constituée par une autre occurrence verbale. Meigret semble donc un peu plus libre à l’égard du modèle latin transposé. La question des valeurs du passé simple et du passé composé est toutefois bien posée dans les mêmes termes, quoique la réponse apportée fasse intervenir des critères d’un ordre très différent. C’est cette interprétation de la détermination du prétérit défini que reprennent R. Estienne (1557) et A.Cauchie (1586). Chez le premier, la reprise est à peu près littérale. Chez Cauchie, elle est complétée d’observations qui viennent du texte de Pillot. Il décrit le prétérit défini (ou déterminé et circonscrit) comme celui qui ne signifie pas le passé de façon absolue, mais dépend de quelque autre discours ou circonstances temporelles additionnelles. Ainsi, on ne peut pas dire je fis ceci, j’achetai cela, mais on dit en revanche j’ai fait ceci, j’ai acheté cela. C’est là un point déjà signalé par Meigret. En revanche, note Cauchie, le passé simple peut s’employer dans des constructions comme je fis avant hier ceci, j’achetai hier cela (Cauchie 1586: f°38). Ces exemples sont analogues à ceux donnés par Pillot, et reviennent à indiquer, partiellement et implicitement, des contraintes relatives aux corrélations verbo adverbiales normalisées par ce qu’il est convenu d’appeler la règle des 24 heures. Les premières discussions explicites relatives à cette dernière apparaissent à la même époque. La plus ancienne est celle qui figure dans Le Traité de la conformité du langage français avec le grec d’Henri Estienne, de 1569. Il écrit: Nous avons aussi deux prétérits parfaits: desquels il m’a semblé autrefois que l’un se pouvait rapporter au temps que les Grecs appellent aoriste, c’est à dire indéfini et non limité. Car quand nous disons j’ai parlé à lui, et lui ai fait réponse, cela s’entend avoir été fait ce jour là. Mais quand on dit je parlai à lui, et lui ai fait réponse, ceci ne s’entend avoir été fait ce jour même, auquel on raconte ceci, mais au paravant: sans toutefois qu’on puisse juger combien de temps est passé depuis. car soit que j’ai fait réponse le jour de devant seulement, soit qu’il y ait ja cinquante ans passés, ou plus, je dirai, je lui fis réponse, ou Alors, adonc, je fis réponse. Voilà comment par ce prétérit nous ne limitons point l’espace du temps passé. Ce qui autrefois m’a fait pensé que comme j’ai dit il avait accointance avec l’Aoriste grec. (H. Estienne 1569:54)


21

Un peu plus loin, Henri Estienne nuance ce propos en même temps qu’il souligne la valeur normative de la règle en rapportant les erreurs de locuteurs étrangers: Mais depuis, ayant considéré de plus près la nature de cet Aoriste, et pesé les raisons d’une part et d’autre, je me suis douté qu’il y avait quelque autre secret caché sous cet aoriste, quant à son naïf usage. Et confesse que jusques à présent je n’en suis point bien résolu. Or ce qui principalement me garde de prendre quelque résolution est que son usage commun n’est autre que du prétérit parfait. Et qu’ainsi soit, on trouvera souvent de dans les bons auteurs qu’une chose qui aura été dite par le prétérit, sera répétée par l’Aoriste, ou au contraire. (H. Estienne 1569:54) (…) Ce nonobstant je penserais faire tort aux étrangers qui font profession de parler bon français, si je ne les avertissais que c’est ici l’endroit par lequel ils sont le plus aisément découvert, principalement par ceux qui les veulent épier au passage. Car c’est grand cas que de cent à grand peine s’en trouvera il dix qui ne heurtent voire achoppent à cette différence de nos deux prétérits comme à une pierre qui serait au milieu de leur chemin. Et qui plus est sitôt qu’on leur aura donné la main pour se relever, on les y verra retomber. Je le sais pour avoir fréquenté avec plusieurs sortes d’étrangers, gens de bon esprit et de bon jugement, lesquels au demeurant se tenaient si bien clos et couverts en leurs devis, que pour un peu de temps ils pouvaient passer pour Français: mais depuis qu’ils venaient à raconter quelque fait, c’était la pitié. Car d’un homme qui fût venu parlé à eux depuis un demi quart d’heure, voir depuis une minute de temps, ils eussent dit, il vint ici, il parla à moi, je lui dis. Au lieu de il est venu ici, il a parlé à moi, je lui ai dit. (H. Estienne 1569:55)

Bosquet (1586), pour finir avec les auteurs du 16eme siècle, dans ses Elements ou institution de la langue française, introduit des distinctions un peu différentes. Le point de départ semble être encore une fois la classification de Pillot: trois prétérits, dont 2 prétérits parfaits. Le critère retenu est celui de la quantification de l’éloignement dans le passé. Bosquet oppose ainsi de façon binaire un temps qui dénote un événement depuis longtemps passé, et un temps qui dénote un passé proche. On voit bien tout ce que doit cette option à la suggestion initiale de Pillot, qui est au fond ici simplement développée et en quelque sorte régularisée. C’est là un bon exemple de la façon dont procèdent les innovations dans l’histoire des textes grammaticaux. 3. La question dans les grammaires du 17eme siècle Je limiterai dans cette partie le parcours à deux textes, dont l’apport me paraît particulièrement remarquable: il s’agit de L’exact acheminement à la langue française de Jean Masset (1606), et de la Syntaxe française de Charles Maupas (1607). J’ai eu l’occasion de montrer comment, par la suite, ces questions sont traitées dans le contexte théorique de la grammaire générale (Fournier 1991-1994).

22

JEAN-MARIE FOURNIER

3.1. Jean Masset (1606) Il s’agit d’un ouvrage qui occupe un volume assez mince et qui figure en accompagnement du dictionnaire de Jean Nicot. Le classement est d’un type identique à ce que nous avons déjà vu: trois temps simples, et plusieurs espèces incluses dans le passé, constituant ce qui est encore appelé des divisions du passé. Mais ici les formes prises en compte sont plus nombreuses et sont reliées par un réseau de relations très différentes: Le passé comporte en effet quatre espèces de temps: l’imparfait, l’aoriste simple (je courus), le parfait (j’ai couru), et le plus que parfait. La terminologie employée indique bien que l’organisation du système est pensée dans des termes différents. Passé simple et passé composé sont ici sur le même plan. En revanche, l’un et l’autre sont associés à un temps composé: à l’aoriste simple (je courus) correspond un aoriste composé (j’eus courus), et au parfait (j’ai couru) correspond un parfait très parfait (j’ai eu couru). Cependant les définitions sont clairement issues de celles de Meigret: ‘l’aoriste simple dépend “d’un discours précédent, narratif de ce qui se serait passé en ce temps”; “nous usons de l’aoriste simple (…) en oraison bornée par adverbe de temps passé, ou autre circonscription que ce soit de temps par laquelle nous définissons l’action indéfinie, à un certain temps déterminé.” (Masset 1606:11) Ces remarques sont complétées par une variante de la règle des 24 heures dont l’évocation est utile à notre propos: “observez que nous n’usons (de l’aoriste simple) jamais pour une action du jour auquel nous sommes, ni de la nuit précédente, ni même avec les adverbes de temps présent aujourd’hui, maintenant, à cette heure.” (Masset 1606.) Le texte présente donc ce qui apparaît être une synthèse des différentes formes que la définition a prises. Y figure en effet le critère référentiel de l’éloignement de l’événement quantifié à 24 heures qui apparaît dans la règle énoncée par Henri Estienne, et le critère contextuel formulé par Meigret selon lequel le passé simple est un temps non autonome dans le discours, et dont la référence se construit par anaphore avec une autre marque temporelle. Nous y trouvons enfin une réinterprétation de la valeur indéterminée du passé simple (d’où l’usage du terme de temps indéfini) donnée comme la cause de sa dépendance à l’égard d’une détermination contextuelle. 3.2. Charles Maupas (1607) Le principe d’une classification des temps n’apparaît pas de façon nette chez cet auteur. Il semble avoir abandonné l’idée d’une division de la catégorie du passé en différentes espèces. En revanche les définitions sont très développées et très sophistiquées. Je m’arrête rapidement sur trois questions: - celle de l’imparfait:


23

On est très loin de la simple paraphrase/traduction de Priscien qui prévalait jusque là. Il apparaît d’abord que la relation entre l’imparfait et le parfait défini (passé simple) constitue ici un problème. Un problème nouveau, qui n’est pas hérité de la tradition, et dont Maupas justifie en quelque sorte la consistance par les fautes commises par les étrangers dans le choix de l’un ou de l’autre. La différence gist en ce que l’imparfait s’attache à une durée et flux de temps qui se faisait lors dont on parle, et n’était encore parachevé. Le parfait au contraire s’arrête à l’acte fini et parfait, et ce en une fois. Car il concerne la fin, perfection, et final accomplissement de la chose un coup faite, sans avoir égard à la durée ou course du temps pendant lequel elle se faisait, et n’estait encore faite. en un mot, l’imparfait parle du fieri. Et le parfait du factum esse. (Maupas 1607:270)

Sur le plan des innovations théoriques, il a déjà été remarqué (Nathalie Fournier, 1998) que ce que décrit Maupas, c’est l’opposition aspectuelle (sécant/global) de ces deux temps. C’est particulièrement net si l’on considère la définition de l’imparfait: ce temps “ramène et remet l’entendement de l’auditeur à l’instant courant lorsque la chose se faisait et n’avait pas atteint la fin et perfection”. - les prétérits définis et indéfinis Maupas reprend à Masset l’idée que les prétérits marchent par deux avec leur “redoublés”, c’est-à-dire la forme composée, respectivement surcomposée, formée sur le passé simple et le passé composé. On a donc deux séries: les prétérits définis (passé simple et antérieur), et les prétérits indéfinis (passé composé, et surcomposé). Leurs définitions empruntent une part de leur matériel à la tradition. Mais Maupas introduit plusieurs innovations. La plus considérable consiste à réinterpréter le critère référentiel, utilisé comme nous l’avons vu sous une forme ou sous une autre dans les définitions ou à travers l’illustration des exemples depuis Pillot et Henri Estienne. Ce critère, chez Maupas, est conçu non comme un point, le repère des 24 heures institué par la règle, mais comme un intervalle de référence. C’est la position de cet intervalle par rapport à l’instant de la parole qui distingue les deux catégories de prétérit. En effet, les temps définis “infèrent toujours un temps piéça passé, et si bien accompli qu’il n’en reste aucune partie à passer. Et à cette cause requièrent toujours une préfixion et prénotation de temps auquel la chose dont on parle est advenuë, & c’est la raison pour quoi je les appelle définis.” (Maupas 1607:273) Tandis que les indéfinis “signifient bien un acte du tout fait et passé, mais le temps non encore si éloigné qu’il n’en reste encore quelque portion à passer”.[ibid.] Dans les deux cas le temps dont il reste ou non une portion à passer n’est pas celui de l’événement, toujours entièrement passé et accompli, mais celui d’un temps, pourvu d’une certaine étendue, et qui fonctionne ici comme référence. Ces formulations seront reprises partiellement par la plupart des auteurs des 17ème et 18ème siècles. On peut considérer toutefois que leur portée véritable ne sera

24

JEAN-MARIE FOURNIER

comprise que par Girard et Beauzée qui en formalisent le contenu en introduisant la notion de ‘période’ de référence. Elle correspond à une analyse en termes aspecto-énonciatif des valeurs du passé simple et du passé composé: le passé simple représente un événement passé, inclus dans un intervalle de référence d’aspect global, l’instant de la parole étant lui-même en dehors de cet intervalle. Le passé composé, représente un événement passé, inclus dans un intervalle de référence d’aspect sécant, autrement dit qui comprend également l’instant de la parole. Conclusion Le transfert du modèle de Priscien sert de point de départ à l’analyse des temps du français et à leur classement. C’est aussi ce contexte qui favorise, ou même, pourraiton dire provoque l’émergence d’un problème linguistique: l’analyse, et les règles d’emploi des deux parfaits du français. Les pistes suivies pour identifier la différence spécifique de ces deux espèces du prétérits parfaits ont été assez nombreuses. Mais ce qui caractérise la tradition française dans le domaine de l’analyse des temps, c’est la mise en place progressive d’une théorie du temps reposant sur des opérations de repérage, et, corrélativement, sur la valeur aspecto-énonciative d’intervalles de références, ou des événements eux-mêmes. L’essentiel de ces outils est déjà en place dans la grammaire de Maupas. Les grammaires générales développeront et systématiseront ces analyses. RÉFÉRENCES Primary sources Bosquet, Jean. 1586. Elemens ou Institution de la langue françoise. Mons: Charles Michel [Slatkine reprints, Genève, 1972]. Cauchie, Antoine. 1586. Grammaticae gallicae libri tres. Strasbourg: B. Iobinus [Slatkine reprints, Genève, 1968]. Estienne, Henri. 1569. Traité de la conformité du langage français avec le grec. Estienne, Robert. 1557. Traicté de la grâmaire françoise. Paris [Slatkine reprints, Genève, 1972]. Masset, Jean. 1606. Acheminement à la langue française. Inseré dans le Thrésor de la langue françoise tant ancienne que moderne auquel entre autres choses sout les mots de marine, vénerie et faulconnerie ed. by J. Nicot. Paris: David Deuceur. Maupas, Charles. 1607. Grammaire et syntaxe française. Orléans: O. Boynard [Slatkine reprints, Genève, 1973]. Meigret, Louis. 1550. Le Tretté de la Grammere françoeze. Paris: C. Wechel [Slatkine reprints, Genève, 1972].


25

Pillot, Jean. 1550. Gallicae linguae institutio, latine sermone conscripta. Paris: Stephanus Grouleau [Slatkine reprints, Genève, 1972]. Secondary bibliography Colombat, B., éd. Jean Pillot: Gallicae linguae institutio, introduction, traduction et notes par B. Colombat. Champion (à paraître). Fournier, J.-M. 1986. “L’imparfait chez les grammairiens classiques, de Maupas (1607), à Restaut (1730)”. Points de vue sur l’imparfait: 13-30. ed. by Pierre le Goffic. Caen: Centre de publications de l’Université de Caen. . 1991. “L’émergence des catégories aspectuelles dans la grammaire générale de Port-Royal à Beauzée.” Histoire, Epistémologie Langage, 13, II: 113-130. . 1994. La théorie des temps dans la grammaire générale, Thèse pour le doctorat. Université Denis Diderot-Paris 7. Fournier, N. 1998. Gramaire du Français Classique. Paris: Belin. Galet, Y. 1974, “Illustration de la théorie des niveaux d’énonciation”. Langue française, 21, 26-42. . 1977. Les corrélations verbo-adverbiales, fonction du passé simple et du passé composé, et la théorie des niveaux d’énonciation dans la phrase française du XVIIème siècle, Atelier de reproduction des thèses, Université de Lille III, 2 vol. Weinrich, H. 1973. Le temps. Paris: Le seuil.

26

27

LINGUISTIC IDEAS AND THE DISCOURSE ON LANGUAGES IN EARLY BRAZILIAN HISTORY1 BETHANIA MARIANI Universidade Federal Fluminense, CNPq Linguistic ideas that were widespread in the European world, have turned into discourses not only due to the accounts of travelers and missionaries heading for the New World. These ideas can also be inferred from the so-called historical treatises that, written on request in many instances, have carried out the task of building the history of Brazil. Between the 16th and 18th centuries, a set of such texts promoted the gradual elaboration of a historical discourse on the description of the colonization process, allowing for the perception of different meanings attributed to the existing linguistic heterogeneity in the colony. One may also infer from such texts the presence of language policies, established with the arrival of European languages (Portuguese and Latin, as well as Spanish and French). The development of a corpus of knowledge of the land and of indigenous languages, marked by European thought, is responsible for the maintenance of a particular direction of meanings for the languages in the colony. Most of the colonization discourses did not have the intent of primarily discussing linguistic issues, but rather of proclaiming the riches and potentialities of the recently found land as well as describing the nature and customs of its different inhabitants. Such discourses functioned as advertisement for the newly discovered land in order to stimulate its colonization. For the present analysis, a corpus composed of nine chronicle writers of the Brazilian colony was selected. As can be observed from the list below, not all the texts from the 16th and 18th centuries were made public in the century in which they were written. Some circulated in the form of manuscripts, gaining significant importance only after their publication in the 18th and 19th centuries. What is worth noting is the discursive functioning of such writings, constituting a thread in a tapestry of voices that shaped the discourse of colonization. 1) Pero de Magalhães Gândavo, História da província de Santa Cruz; written in 1573 and published in 1576; 1. This study is part of a larger research project on the institutionalization of the Portuguese language in Brazil, and was finished while I was a visiting scholar at Stanford University in 2001 working with Hans Ulrich Gumbrecht, in the Departement of Comparative Literature. Some findings of this research have already been published in Mariani 1998 and 2003.

28

BETHANIA MARIANI

2) Padre Fernão de Cardin, Tratados da terra e gente do Brasil and Narrativa epistolar da missão jesuítica do Padre Cristóvão de Gouveia, both probably written in the late 16th century, first edition in English, from the collection of Samuel Purchas, in 1625; 3) Gabriel Soares de Sousa, Tratado descritivo do Brasil, written in 1587, first complete edition in 1825; 4) Frei Vicente do Salvador, História do Brasil, written in 1627, first complete edition in 1888; 5) Ambrósio Fernandes Brandão, Diálogos das grandezas do Brasil, written in 1618, with partial edition in 1849 and complete edition in 1900; 6) Francisco de Brito Freyre, Nova Lusitania; história da guerra brasílica. Lisbon, printed by Joam Galram, 1675. 7) Sebastião da Rocha Pita, História da América Portuguesa, edited in 1730. 8) Bernardo Pereira de Berredo. Annaes históricos do Estado do Maranhão, written in the first half of the 18th century, first edition in 1749. 9) Frei Gaspar da Madre de Deus, Memórias da capitania de S. Vicente hoje chamada São Paulo do Estado do Brasil, written during the second half of the 18th century, probably in 1797, first edition in 1789. In addition, accounts from travelers were also subject to comparison to the discourse analysis of historians, such as Flecknoe (1649), Rouloux Baro (1651), Edward Barlow (1663), François Froger (1695) and John Nieuhoff (1703). To summarize and systematize the results obtained from the analysis, performed according to the parameters of Discourse Analysis (Pêcheux, 1975; Orlandi, 1990; and Orlandi & Guimarães, 1998), I will present them under different topics. 1. Paraphrases and the image of precariousness of the indigenous languages In the analized discourses of these nine writers there is a reiteration, a recovery of interpretations and comments on what was seen, heard or read. This addition or overlap of discourses takes place in the form of translations, paraphrases, explanations and comparisons concerning the linguistic knowledge being acquired. In the words of Pero Magalhães Gândavo, based on accounts by the Jesuit priest José de Anchieta, we come across one of the most frequently reproduced observations on Brazilian indigenous languages. In the chapter entitled “on the condition and customs of Indians of this land”, one can notice the eurocentric view that projected precarious social and religious conditions, ascribed to the Indians, upon the unknown languages: (...) a lingoa deste gentio toda pella costa he hu’a, careçe de tres letras - S – não se acha nella f, ne l, ne R, cousa digna despanto por ~q assy não tem fê ne lei, nem Rei, & desta maneira viue, sem justiça desordenadamente. (Gândavo 1576:181-183).

LINGUISTIC IDEAS AND THE DISCOURSE ON LANGUAGES

29

[ (...) the language of all the indians along the coast [is one that] lacks three letters – S – there is no f, nor l, nor R, something that causes perplexity because here there is neither faith[FÉ] nor law[LEI], nor King[REI], and, in this way, they live without justice in a disorderly manner. (Translated and adapted from the original text)]

The statement by Gândavo is paraphrased in other works as we can note in the excerpt by Gabriel Soares de Sousa below: (...) [os tupinambás] Têm muita graça quando falam, mormente as mulheres; são mui compendiosas na forma da linguagem, e muito copiosas no seu orar; mas faltam-lhes três letras do ABC, que são F, L R grande ou dobrado, coisa muito para se notar; porque, se não têm F, é porque não têm fé em nenhuma coisa que adorem; nem os nascidos entre os cristãos e doutrinados pelos padres da Companhia têm fé em Deus Nosso Senhor, nem têm verdade, nem lealdade a nenhuma pessoa que lhes faça bem. E se não L na sua pronunciação, é porque não têm lei alguma que guardar, nem preceitos para se governarem; e cada um faz lei a seu modo, e ao som da sua vontade; sem haver entre eles leis com que se governem, nem têm leis uns com os outros. E se não têm esta letra R na sua pronunciação, é porque não têm rei que os reja, e a quem obedeçam, nem obedecem a ninguém, nem ao pai o filho, nem o filho ao pai, e cada um vive ao som da sua vontade. (Gabriel Soares de Sousa 1587: 302) [(...)[theTupinambá indians] They have a special flair when they speak, especially the women; they are very brief and direct in their speech, and truly devoted in their prayers; but three letters are not found in their alphabet, which are F, L and R, long or double, a very significant thing to be noticed because, once there is no F, this is due to lack of faith in anything they adore; not even those born among the Christians and preached by Jesuits of the Company [of Jesus] have faith in God Our Lord, they have no truth, no loyalty to any person who is kind to them. And if there is no L in their pronunciation, this is because they have no laws to be safeguarded, nor principles to govern them; and each one makes use of law in his own individual way, at will; without any rules among them through which they can govern themselves, not even rules among each other. And if there is no letter R in their pronunciation, this is because there is no king to rule them, and to pay obedience to, there is no obedience toward anyone, neither from son to father, neither from father to son, and each one lives according to his own will.]

More than forty years later, after Gândavo and Anchieta had characterized the indigenous languages as precarious or poor languages, Ambrósio F. Brandão takes on the same theme, by means of similar words: “E por isso se diz geralmente que este gentio do Brasil carece, na sua língua, de três letras principais, as quaes são F, L, R”. (Brandão, 1966:266). “And that is why it is generally accepted that the language of the people from Brazil lacks three major letters, which are F, L, R”. This point of view was supported by Friar V. do Salvador: “Mas nem uma palavra pronunciam com f, l ou r, não soh das suas mas nem ainda das nossas, (...) e o peor eh que tambem carecem de fé, de lei e de rei...” (Salvador 1627: 53). “But not

30

BETHANIA MARIANI

in a single word do they pronounce f, l or r, (…) and the worst part is that they also have no faith, no law and no king…”. In the same manner, Richard Flecknoe, in his account, simultaneously questioning the imaginary of that period of time and reproducing ‘anthropological-linguistic’ ideas already disseminated, comments on the Indians and their languages: No tocante à ferocidade desses nativos, não acreditei em tudo o que me disseram. Creio ser verdade que eles devoram uns aos outros, e que na sua língua não constam as palavras deus, lei e rei. Contudo, se fossem de fato ferozes como contam os habitantes locais, certamente não teriam entregado tão mansamente suas terras aos portugueses, nem permitiriam que estes desfrutassem delas sem qualquer transtorno. (Flecknoe in França 1999:40) [As far as the ferocity of these natives is concerned, I did not believe everything I was told. I believe it to be true that they devour one another, and that their language does not include the words god, law and king. Nevertheless, if they were in fact as ferocious as local inhabitants say, they would certainly not have given up their land so peacefully to the Portuguese, nor would they have allowed them to exploit the land without any disturbance.]

2. The construction of the image of the general language: unity and smoothness Without any doubt, Pero de Magalhães Gândavo (1576), Fernão Cardim (1584), Gabriel Soares de Sousa (1587) and Friar Vicente do Salvador (1627) produced the first descriptions of the indigenous languages, promoting the idea of a single language, a “major”, “general” indigenous language, spoken along the Brazilian coast and displaying little variation. To this so-called “Língual Geral” of the coast, the authors contrast the languages of other indigenous nations. Fernão Cardim, in the chapter On Diversity of Nations and Languages, describes the “many and various nations with different languages” found throughout the colony, always stressing which languages are related to one another and which of them are completely different. Yet, one of them, according to the author, “is the major one.” Em toda esta provincia ha muitas e varias nações de differentes linguas, porém huma é a principal que comprehende algumas dez nações de Indios: estes vivem na costa do mar, e em uma grande corda do sertão, porém são todos estes de uma só língua ainda que em algumas palavras discrepão e esta é a que entendem os Portuguezes; é fácil, e elegante, e suave, e copiosa, a dificuldade della está em ter muitas composições; porém dos Portuguezes, quasi todos os que vêm do Reino e estão cá de assento e communicação com os Indios a sabem em breve tempo, e os filhos dos Portuguezes cá nascidos a sabem melhor que os Portuguezes, assim homens como mulheres, principalmente na Capitania de São Vicente, e com estas dez nações de Indios têm os Padres communicação por lhes saberem a lingua, e serem mais domesticos e bem inclinados (Cardim 1584:121)


31

[All over this province there are many and various nations with different languages. One, however, is the major language spoken in some ten nations of Indians: they live along the coastline, and in a large area of the heartland, yet, they all speak a single language although some words do not match and this language is the one the Portuguese understand; it is easy, and elegant, and smooth, and varied, the main difficulty being its many compositions; yet, among the Portuguese, almost all of those coming from the Kingdom who have settled down here and established communication with the Indians, learn it very quickly, and the Portuguese children born in the colony know it better than their parents; male and female alike, mainly in the province of São Vicente, and Priests, once they know their language, communicate with these ten nations of Indians, due to their domestic nature and better inclination]

Among the colonization discourses there is also the reproduction of another line of thought, that places the Tupinambás, Portuguese allies, in opposition to the Tapuias, brave, enemy indians, whose languages, from this standpoint, are neither “easy, elegant nor smooth”. Father Cardim, Gabriel S. de Sousa and Friar Vicente say: Todas estas setenta e seis nações de Tapuyas, que têm as mais dellas differentes linguas, são gente brava, silvestre e indomita, são contrarias quasi todas do gentio que vive na costa do mar, vizinhos dos Portugueses (...) D’estes [tapuias] ha muitos christãos (...) e somente com estes Tapuyas se pode fazer algum fructo; com os demais Tapuyas, não se pode fazer conversão por serem muito andejos e terem muitas e differentes linguas difficultosas. (Cardim, 1584:121) [All these seventy six Tapuya nations, that speak the most different languages, are brave people, wild and savage, have quite different attributes as compared to the people who live on the coast, neighbors of the Portuguese (…) Among these [Tapuyas] there are many Christians (…) and only to these Tapuyas some good can be done; with the rest of the Tapuyas, conversion is not possible since they are true ramblers and have many different and difficult languages.] Descendem estes aimorés de outros gentios a que chamam tapuias (...) e os que destes descenderam, vieram a perder a linguagem e fizeram outra nova que se não entende de nenhuma outra nação do gentio de todo este Estado do Brasil. E são estes aimorés tão selvagens (...) a sua fala é rouca da voz, a qual arrancam da garganta com muita força (...). (Sousa 1587:79) [These Aimorés are descended from other people called Tapuias (...) and these descendedants abandoned the original language and created a new one that no other nation of people all over the State of Brazil can understand. And these Aimorés are so savage(…) their speech is hoarse in voice, which they utter from their throat with great force (…).] Os mais barbaros se chama in genere Tapuhias, dos quaes ha muitas castas de diversos nomes, diversas linguas, e inimigos uns dos outros. (...) comtudo todos falam um mesmo linguage e este aprendem os religiosos que os doutrinam por uma arte de grammatica que compoz o padre Joseph de ancheta, varao santo da ordem da Companhia de Jesus. (Frei V. Do Salvador 1627:52). [The most barbaric are called, in genere, Tapuhias, among whom there are many castes of various names, various languages, and they are enemies among themselves.(…) all speak

32

BETHANIA MARIANI the same language, learned by the clergy men who preach to them by means of an art of grammar developed by Father Joseph de Anchieta, worthy saint from the order of the Company of Jesus.]

These are “linguistic” conceptions not only ratified but also made to circulate in a much more systematic way by the Jesuit missionaries, starting with the studies by José de Anchieta in 1555, that resulted in the publication of a grammar book entitled Arte da Gramática da língua mais usada na costa do Brasil (The Art of the Grammar of the most widely spoken language on the coast of Brazil) (1595). 3. Designations as a form of linguistic colonization The voices of these “historians” echo the voices of administrators and colonists, and their effort to overcome resistances inherent in the Brazilian colony. This has to do, among various activities, with the establishment of commercial relations with the New World. Such relations are intertwined with indigenous languages and the Portuguese language. There is a process of incorporating indigenous words into the Portuguese language, but these words, once incorporated, are inserted in a linguistic universe organized in terms of the European intelligibility. Thus, the indigenous words used to designate new “things from nature” are ultimately linked to the Portuguese language and their meanings are easy to grasp. If, on the one hand, it is not possible to assert that the process of making the unknown New World known among the Portuguese and Europeans is deprived of some prior familiarity with European modes of elaborating knowledge, on the other hand, it is not possible to ignore the influence of this new unknown world over the old European world. On these terms, the act of colonizing implies a linguistic attempt to undo the opacity of the new land, making it transparent to the Portuguese and Europeans with regard to designations, translations and the learning of unknown languages. In this way, it becomes feasible to produce descriptions that start from “similarities”– that is, with explicit attempts to link the meaning that is constructed in the Portuguese America to whatever is already familiar in Europe, to a kind of typology that will allow a new “object” to be fit into a recognizable type, to the body of knowledge already found in the European world which, to some extent, pre-constructs the intelligibility of discourse as a whole. These “similarities” preconfigure actual interpretations, establishing a direction of meaning in relation to which the “differences” are presented as a kind of complementation of a meaning already configured. The act of designating, therefore, is a fundamental step in this process of colonization and domestication of the unknown. It suggests an idea of language transparency based on the presupposition of direct linkage between things


33

and words, between the referent and the term that designates it2. Considering that the use of one or another language is not naïve, designations may render disputes, impositions and existing silences visible. In Gândavo, it is possible to observe some elements of the flora and fauna, described by him, that shape a recurring type of discursive functioning: (...) e se tem pella milhor caça que ha no matto chamão lhes Tatûs são tamanhos como coelhos e te um casco a maneira de lagosta como de cágado mas he repartido em muitas juntas como laminas, pareçe totalmente hu’ cavallo armado, tem hu’ rabo do mesmo casco comprido o foçinho he como de leitaõ, e naõ bota mais fora do casco que a cabeça (Gândavo 1576:160) [(...) and the best game there is in the woods people call them armadillos and they are the size of rabbits and have a shell like that of a lobster, just like the one of a turtle but banded on many joints like blades, they look exactly like armed horses, they have a long, shell-like tail, the nose of a pig, and only the head comes out of the shell.] Hu’a fruita se dá nesta prouinçia do Brasil muito sabrosa e mais prezada de quantas ha na terra. (...) que chamão lhes Ananâzes e depois de maduras tem hu’ cheiro m[ui]to exçellente (Gândavo 1576:165) ~ [There is a certain fruit in this province of Brazil which is very tasty and more appreciated than all the others over the land. (…) that they call Pineapples and when they are ripe they have an excellent smell.] Tambem ha hu’a fruita que lhe chamão banânas, pella lingoa dos indios pacôuas, ha na terra muita abundançia dellas (Gândavo 1576:169) [There is also a fruit that they call bananas, in the language of the Pacôua Indians]

In the fragments above, and in all the others under analysis, the designation of things from Brazil unknown in the European world is usually preceded by statements of the existence of something already known. The statement of existence formulated with the Portuguese verb haver is followed by an adjective clause– que lhe chamão banânas, que chamão-lhes ananâzes – which designates and inserts “the thing from Brazil”– tatu, banânas, anânazes – in the generic category, already known in the European world – game, fruit. Thus, the process of giving transparency to designations such as – tatu, banânas, anânazes – is initially anchored to a field of previous ideas (Pêcheux 1975). The 2. The conception of transparency in language and literality of meanings, central to thinkers of that epoch, is the pioneering notion of the epistemics of the 16th century, according to Foucault (1966). By embracing another discourse standpoint, it must be mentioned that the question of reference deals simultaneously with linguistically based questions (the many possible ways to speak about) and historical ones (considering the restrictions imposed on the processes of signification by not so often visible social and legal designations that, by regulating the forms to speak, prevent one from attributing meaning to anything, from anywhere, at any time). See also Orlandi (1990).

34

BETHANIA MARIANI

designation is then followed by detailed explanations formulated by means of adjectival complementation, comparisons and analogies with what is already known, just as the case of tatu in the first fragment. Designations are always followed by similarities and differences in the form of complements. Therefore, tatus (armadillos) are described by means of similarities with whatever is familiar to Europeans – they are the size of rabbits and have a shell like that of a lobster, just as turtle’s – and later, differences are shown in the form of complements – but banded [the shell] on many joint-like blades. The very same discursive process can be found in the accounts by non-Portuguese travelers. Through this linguistic attempt to produce a pictorial effect, the referents are gradually being constructed in discourse so as to come to life in thought. In other words, the search for a transparency in language is in accordance with the discursive construction of the referent for the designated “thing”. In order to make sense out of words and things, other meanings must be added to these designations. As we clip passages from the accounts related to the theme foods that were not known by Europeans, for example, we can easily observe the discursive process of paraphrases, literal repetitions superposing discourses upon discourses. The excerpts below, written by different travelers, typically exemplify this discursive process: O ananás é, sem dúvida, a melhor e a mais saborosa das frutas que o país produz. Ele cresce como uma alcachofra, suas folhas são espessas e dentadas como as das semprevivas e seu caule e sua casca, escamosos. (...) (Flecknoe, in França 1999:37) [The pineapple is, with no doubt, the best and tastiest of all fruits the country produces. It grows like an artichoke; its leaves are thick and toothed like those of the evergreens and its stalk and bark, scaled. (…)] O ananás cresce como uma alcachofra e é parecido com uma grande pinha; suas folhas são longas, grossas, armadas com pequenos espinhos e em forma de coroa. Essa fruta é a melhor de toda a América. (F. Froger, in França 1999:53) [The pineapple grows like an artichoke and it resembles a large pine; its leaves are long, coarse, covered with little thorns and with the shaped like a crown. This fruit is the best in all of America.]

In the discourse functioning of the two fragments above we can notice: 1) vocabulary repetition – artichoke –, which is 2) point of departure for the repetition of the comparative structure grows like..., 3) synonymic substitution thick leaves instead of coarse leaves and 4) return to a superlative structure the best and tastiest of all fruits and the best one in all of America.. This discourse functioning is the same as the one found in the writings of Portuguese authors, constituting the previously mentioned intertwining of voices in the discourse of the colonization of Brazil and stressing a particular direction of meanings. Indetermination and determination: who designates them?


35

It is important to note that there is another recurring linguistic functioning, which is determination or indetermination in relation to the origin of the designation. There are two alternate discourse mechanisms: the explicitness of the indigenous origin of the designation – in the language of the Pacôua indians – and/or the use of a name without any specification concerning its origin – they call them Pineapples –, this leaves us with a doubt: who calls them pineapples? These alternations can be observed in the following excerpts, taken from the vast number of examples found in the other accounts under analysis: Ha um rio entre Porto-Seguro, e os Ilhéos (...): traz muita copia de rezina que he o mesmo anime, a que os Indios chamão Igtaigcica, e os portuguezes incenso branco, e tem os mesmos effeitos que o incenso.(Cardim 1584:42) [There is a river between Porto-Seguro and the Ilhéos (…): it brings in its own waters a lot of natural resin with arousing effects, which the Indians call Igtaigcica, and the Portuguese, white incense, and it has the same effects of incense.] Há uma casta de mandioca que se diz manipocamirim, e outra que chamam manaibuçu, e há outras castas que chamam taiaçu, e duram estas raízes debaixo da terra sem apodrecerem três, quatro anos. (...) (Gabriel S. Sousa 1587:174) [There is a kind of manioc called manipocamirim, and another kind called manaibuçu, and still there are other kinds called taiaçu, and these roots last, under the ground without spoiling, three, four years…]

From the discursive functioning of the designations above, one can notice that there is a variation between indetermination/determination on the part of the designator: on the one side, indetermination of the grammatical subject, generally expressed in constructions with ‘discendi’ verbs of the type: which they call…, whom they call…, what is said…, called…, which is called (a que chamam.., a quem chamam..., que se diz..., chamados..., que se chama...). On the other side, determination: the Indians call..., the Indians and more people…, that the people call (os índios chamão..., os índios e mais gentes..., que o gentio chama). Such alternations, depicted in the accounts of all the authors, point to the complex network of subject positions, pervasive throughout the colonial period. What is to be understood here as subject position is that the imaginary projection of the socio-historical place from where one speaks as well as the imaginary projection of one’s interlocutor are implied in the formation of this network of imaginary projections. In some fragments, either the relevant subject positions are retrieved from the context or it is possible to clearly recognize the indigenous and Portuguese voices in many designations: some herbs that the Indians call jaborandiba, and the Portuguese, white incense. What calls our attention is that these voices are not always explicitly expressed. In many cases, one may conclude that the absence of the grammatical subject, on the part of the designator indicates an ambiguity in relation to the position of the subject. In other cases, instead of an opposition indians vs the

36

BETHANIA MARIANI

Portuguese, one is confronted with a view of the whole – and to take full advantage of these roots, the indians and more people – where we also find indetermination. After all, what would be the discursive referent for more people? In the gap between the discourse practice that designates, as well as the imaginary projection of one’s interlocutor institutes a direction of meanings, allowing for visibility for things from/in Brazil, and the discourse practice that is designated, deprived of voice, there is a silence marked by the confrontation between meanings that convey power relations current at that time, imposed by the metropolis, and movements of resistance originated in the lands of Brazil. The image of the Indians and mestizos, children born from the union between the Portuguese and the Indians, or from the Portuguese born in the colonized land, is linked and reduced to a set of designations. Once we observe, among foreign travelers, the process of overlapping discourses, it is also possible to discuss a textual disclosure of the discourse positions in the colony which are left undetermined in the written treatises by Portuguese and Brazilian writers. Whatever is undetermined in the Portuguese language is disclosed in the foreign language. The question, however, lies in verifying to which extent this disclosure is actually displaying, erasing or incorporating the differences among the previously mentioned positions of the subject. This is what one may start to observe in the excerpts by Rouloux Baro (1651), Rouloux Baro’s French translator, and John Nieuhoff (1703) transcribed below. Au matin du dixieme les eaux s’étant retirées (...) quelques petites poissons que les sauvages nomment Paramiri, Acaramiri (...) le vingt-vniesme outre ce que i’auoir d’hommes, i’en pris deux dans l’Aldée des Brasiliens, pour nous conduire de-là la riuiere (...) Là nous vuindasmes deux grands arbres pelins de miel sauuage, & chassans nous (...) d’vn gros serpent nommé Cascabilla, d’vn ieune oiseau, appellé Strus, & de deux Tatous (...) Le onzieme sur les dix heures du matin les Brasiliens du bas arriuerent en l’Aldée où nous étions, les habitants de laquelle les receurent auec grands cris de ioye, & propos d’allegresse (In R. Baro 1651:199-208) Remarques du Sieur Morisot sur le voyage de Rouloux Baro (...) Ils sont appellez d’aucuns Tapuias, d’autres Tapoyos: mais comme cette terminaison n’est point Françoise le traducteurs les appelle Tapuies. (...) Il faut dire Cascauela, nom duquel les portugais appellent le serpant nommé des Brasiliens, Boicininga, Boicinininga, Boitinga, & Boiquira, par les Tapuyes Aiugi, par les Hollandois Kaetel Slange. (...) Les Brasiliens appellent cet animal Tatu, & Tatupeda, les Espagnols Armadillo, les Portugais Emcuberro. (In R. Baro 1651:247-262) There are also several sorts of Serpents in Brasil. (...) The serpent of Boicininga, or Boicinininga, likewise called Boiquira, by the Brasilians, is by the Portugueses called Kaskaveda and Tangedor (..) That four-legg’d Creature, by the Brasilians called Ai, by the Portugueses Priguiza, and by the Dutch Luyaert (Lazy-back) from its lazy and slow pace, (...) The four Legg’d Creature, called by the Brasilians, Tatu and Tatupera, by the Spaniards,


37

Armadillo, by the Portugueses, Ecuberto, and by the Dutch, Schilt-Verken, (Shield-Hog) because it is defended with Scales like as with an Armour, resembles in biness and shape our Hogs. (In J. Nieuhoff 1703:15-20)

In the excerpts above, the images created in relation to the positions of the subject in the colony are expressed in a very distinctive way. In the elaboration of the accounts, in order to bring visibility to the translation of indigenous words, other European languages are incorporated. In R. Baroux’s account, there is an opposition between the narrator, conveyed by the first person plural ‘we’, and one ‘they’, successively designated as sauvages, Brasiliens and Tapuias. In fragment #18, the first person singular ‘I’ (Je croy) is opposed to different subjects in different languages: portugais, hollandois, brasiliens and espagnols. In fragment #19, the first person plural, marked in the expression our people is also opposed to different subjects: Brasilians, Portugueses and the Dutch. In this process of incorporation of other languages, there is a return and, simultaneously, a very peculiar reconstruction of the (discursive) functioning that suggests the insertion of an object to be described in a category previously known by the European world. For instance, it becomes crucial to insert the words/names cascabilla, cascauela or kaskaveda in the previous idea d’un gros serpent or in the image that in Brazil there are several sorts of Serpents. Nevertheless, equally important is the indirect reference given to European languages by means of the designation of nationalities. These nationalities-languages function as the generic categories of previous knowledge. There is an implied and previous recognition of the other European languages that promote the configuration of the nations of the subjects that are mentioned. In the accounts under analysis, we have as irrefutable evidence, obvious proof, the fact that the Portuguese, the Dutch and the French have languages that are as much specific as different from each other, and this leads from the reference to nationalities to the necessary translation of unknown terms. These are languages that are already grammatically systematized (Auroux 1997) and used by speakers whose identification is made through their respective national groups (the Portuguese, the Spaniards, the Dutch, the French). In a parallel way, under the words brasiliens or brasilans the homogeneity of the indigenous languages is promoted, and their differences neutralized. In other words, these accounts show the gradual construction of a parallelism premised on the relation “one language-one nation”, product of the European world. However, once shifted to the description of elements of the Brazilian colony, such relationship ultimately erases the linguistic diversity. At this point, it is worthwhile to look up in John Nieuhoff’s work and find out who the “brasilians” are: The Inhabitants of Brasil may at present be divided into Free-born Subjects and Slaves, and these again consist of diverse Nations, both Natives and Foreigners. The Free Inhabitants of Brasil were the Dutch, Portugueses, and Brasilians, the last the Natives of

38

BETHANIA MARIANI the Country. (...) The Natives of Brasil consisted of diverse Nations, distinguished by their Proper Names, to wit; The Tubinambos [sic], tobajaras, Petiguaras and Tapuijas, or Tapuyers, or Tapoyers. The three first use one and the same Language, and differ only in the dialect; but the last are subdiveded into several Nations, differing both in manners and Tongue. (In J. Niehuoff, 1703:15-20)

Brasilians, according to the author, are the natives of the Country, that is to say, the Indians. They are part of the group of Inhabitants of Brasil, a group which, according to this view, is composed of free-born Subjects and Slaves. While specifying who the free inhabitants are, Nieuhoff refers them to the European nationalities and includes on the listing the brasilians, the natives of the country. This task of designating the “Brazilians” (brasileiros), refers back to the division between tupinambás and tapuias and, at the same time, retains the image of a major language opposing itself to a set of different languages from different indigenous nations. In a single act toward designation, there is, simultaneously, the recognition of the indigenous culture and the erasure of diversity. In the same way, with this designation of “Brazilians” the inter-ethnic result of two centuries of colonization is erased as well as the existence of children of the Portuguese (and of the Dutch, the Spaniards and the French) born in the colony. There is a complete loss of specificity regarding a ‘different place’ in gestation throughout history. Concluding, what we see here is an attempt to domesticate the indigenous language in the same way one aims to domesticate the Indian, since for the Portuguese, as well as for other Europeans, it seemed only logical and necessary to have Tapuyas already domesticated, as Bernardo Pereira de Berredo noted during the 18th century, referring to the barbarians, imperfections of nature, savages (Berredo 1905:322). And it is precisely this “domestication” – which is part of the linguistic colonization process that can be verified, after the topics discussed in the present study, not only in the attempt to create meanings in the Portuguese America from a set of European references, but also in the creation of an image of flaw and precariousness of languages and peoples – that sustains colonization. REFERENCES Primary sources Barlow, Edward. 1663 [1999]. Relação de viagem feita em 1695... in: França, Jean Marcel Carvalho. Visões do Rio de Janeiro Colonial – antologia de textos. Rio de Janeiro: José Olympio. Baro, Rouloux. 1651. Relation du voyage au pays des Tapuies (1647). Commencé le troisième Avril 1647. Finy le quatorzième juillet de la même année. Traduit d’Hollandais en François par Pierre Moreav de Paray en Charolois. Paris: A. Courbé,. Berredo, Bernardo Pereira de. 1905. Anais históricos do Estado do Maranhão, em que se dá notícia de seu descobrimento, e tudo o mais que nele tem sucedido


39

desde o ano em que foi descoberto até o de 1718. 3a. Ed. Florença: Tipographia Berbéra,. Brandão, Ambrósio Fernandes. 1966. Diálogos das grandezas do Brasil, de acordo com a edição da Academia Brasileira de Letras, contendo nota preliminar de Afrânio Peixoto. Introdução de J. Capistrano de Abreu. Apresentação de Afranio Coutinho. Recife Cardim, Padre Fernão de. 1584 [1978]. Tratados da Terra e gente do Brasil. Introdução e notas de Batista Caetano, Capistrano de Abreu e Rodolfo Garcia. 2a. Ed. São Paulo: Companhia Editora Nacional. Deus, Frei Gaspar da Madre de. 1975. Memorias para a historia da Capitania de São Vicente. Prefácio de Mario Guimarães Ferri. França, Jean Marcel Carvalho. 1999. Visões do Rio de Janeiro colonial – antologia de textos (1531-1800). Rio de Janeiro: José Olympio. Freyre, Francisco de Brito. 1675. Nova Lusitania; história da guerra brasílica. Lisboa: Officina de Joam Galram. Gândavo, Pero de Magalhães. 1576 [1980]. Historia da provincia de Santa Cruz, a que vulgarmente chamamos Brasil, feita por Pero de Magalhães de Gandavo, dirigida ao muito ilustre Senhor Dom leonis Pereira. Belo Horizonte: Itatiaia. Nieuhoff, John. 1703. Voyages and travels, into Brasil, and the east-indies: containing au exact description of the Dutch Brasil, and divers parts of the East-Indes; their provinces, cities, living creatures, and products: the manners, customs, habits, and religion of the inhabitants: a most particular account of all the remarkable passages that happened during the Author’s stay of nine years in Brasil, especially, in relation to the Revolt of the Portugueses, and the Intestine war carried on there from 1640 to 1649. Translated from the Dutch Original. London: printed for Awnsham and John Churchill at the Black Swan in Pater Noster Row. Pita, Sebastião Rocha. 1850. Historia da America Portuguesa desde o anno de mil e quinhentos do seu descobrimento até o de mil setecentos e vinte e quatro. Segunda edição revista e anotada por J.G. Goes. Lisboa: editor Francisco Arthur da Silva. Salvador, Frei Vicente de. 1627 [1918]. Historia do Brasil, por Frei Vicente do Salvador. Nova edição revista por Capistrano de Abreu. São Paulo. Sousa, Gabriel Soares de. 1587 [1938]. Tratado descritivo do Brasil em 1587. Edição castigada pelo estudo e exame de muitos codices manuscriptos existentes no Brasil, em Portugal, hespanha e França, e accrescentada de alguns commentarios por Francisco Adolpho de Varnhagem. São Paulo. Secondary bibliography Auroux, Sylvain. 1997. (dir.) Histoire des idées linguistiques. Tome 2 Le développement de la grammaire occidentale. Paris: Mardaga. Foucault, Michel. 1971. L’Ordre du discours. Paris: Gallimard. Mariani, Bethania S.C. “L’institutionnalisation de la langue, de la mémoire et de la citoyenneté au Brésil durant le XVIII siècle: le rôle des académies littéraires et de la politique du Marquis de Pombal.”: Auroux, Sylvain, Orlandi, E. P. and Mazière, Francine (Éds.) Langages – l‘hyperlangue brésilienne. Número 130. Paris: Larousse, junho 1998: 84-97.

40 . 2003. “L’État, l’église et la question de la langue parlée au Brésil.” In History of Linguistics 1999 – Selected papers from the eighth International Conference on the History of the Language Sciences, 14 – 19 September 1999, Fontenay – St. Cloud. ed. by Sylvain Auroux, 185-196 (=Studies in the History of the Language Sciences, 99). Amsterdam & Philadelphia: John Benjamins. Orlandi, Eni P. 1990. Terra à vista – discurso do confronto: velho e novo mundo. São Paulo, Campinas: Cortez & Editora da Unicamp. . e Guimarães, Eduardo. “La formation d’un espace de production linguistique. La Grammaire au Brésil.” In: Auroux, Sylvain, Orlandi, E. P. and Mazière, Francine (Éds.) Langages – l‘hyperlangue brésilienne, 130:84-97. Paris: Larousse. Pêcheux, Michel. 1975. Les vérités de la Palice. Paris: Maspero.

41

GRAMMAIRE GÉNÉRALE ET GRAMMAIRE PARTICULIÈRE LES MÉTHODES DE CLAUDE IRSON

SIMONE DELESALLE & FRANCINE MAZIÈRE Université de Paris XIII Université de Paris XIII On veut examiner ici les différences qui apparaissent entre les deux éditions de la Grammaire d’Irson (1656 et 1660), en les rapportant à la parution de la Grammaire Générale et Raisonnée de Port-Royal (1660). Ce qui signifie: considérer le rapport entre développements techniques grammaticaux et luttes religieuses et politiques (Claude Irson semble à cette époque lié au mouvement janséniste); travailler les ajustements nécessaires entre un ouvrage qui traite de la grammaire générale et une grammaire du français; enfin s’interroger sur le va et vient des influences réciproques entre grammaire générale, grammaire particulière et dictionnaire, l’œuvre d’Irson prenant place dans un temps fort de discussion sur la préparation d’un Dictionnaire de l’Académie. 1.

Irson

Ce grammairien est souvent cité, surtout comme arithméticien, mais peu connu. Nous ne sommes pas arrivés à le situer complètement. Il publie sa première grammaire au milieu des années 1650, années porteuses de projets à visée politique, à l’heure où un Académicien comme Pellisson réfléchit au rôle d’un dictionnaire et pense l’importance pour l’Etat d’une “langue commune” avant de s’engager dans la tourmente aux côtés de Fouquet; à l’heure où Port Royal devient inquiétant pour le pouvoir, où se jansénise le clergé de Paris qui refuse la morale jésuite, pétitionne en faveur des Provinciales et ouvre des “écoles de charité” indépendantes de l’autorité du Grand Chantre; où Antoine Arnauld défie les autorités de la Sorbonne et les Jésuites en diffusant en français des analyses de théologie et où Claude Lancelot ose une Méthode d’apprentissage du latin rédigée en français et une Méthode d’apprentissage du grec qui ne passe pas par le latin. Bref, dans une période de changements et d’affrontements, en particulier autour de l’affirmation de la langue française. Cette langue n’avait inspiré que trois grammaires dans la première moitié du siècle: celle de Masset, Acheminement à la langue française, bilingue français-latin, insérée

42

SIMONE DELESALLE & FRANCINE MAZIÈRE

en 1606 dans le Thrésor de la langue française de Nicot, celle de Maupas, Grammaire et syntaxe française, en 1607, écrite seulement en français, et celle de Oudin, en 1632, Grammaire française rapportée au langage du temps, qui comptera 6 rééditions. Les titres montrent assez que l’autonomisation des règles du français contemporain par rapport au latin et à l’état antérieur de la langue est la principale visée de ces ouvrages. Irson publie des méthodes d’apprentissage. – 1656 Nouvelle Méthode pour apprendre facilement les principes et la pureté de la langue française – 1660 Nouvelle Méthode pour apprendre facilement les principes et la pureté de la langue française – 1662 Nouvelle Méthode pour apprendre facilement les principes et la pureté de la langue française(deuxième édition) – 1667 (privilège en 1658) Méthode abrégée et familière pour apprendre en peu de temps à bien lire, à prononcer agréablement et à écrire correctement en français Il faut considérer les dates d’édition. Nous avons volontairement répété trois fois le même titre: la mention “seconde édition”, en 1662, jointe à quelques omissions de l’édition de 1660 dans des fichiers et catalogues de bibliothèques de même que dans des recensions d’ouvrages, avait conduit à considérer 1662 comme la réédition de 1656. Il semblait alors suffisant de lire l’ouvrage de 1656 et d’en signaler la réédition. Or, la Méthode de 1660, rééditée en 16621, est profondément différente de celle de 1656 . Entre temps a été publiée la Grammaire Générale et Raisonnée de Port Royal. C’est cette différence, et cette concomitance, que nous exploiterons dans l’article. Si l’on s’attache au titre lui-même, on note que l’emploi du mot Méthode est nouveau pour le français, toute approche pédagogique se faisant alors par le latin, excepté à l’Oratoire, et à Port Royal. A Port Royal où, précisément, Lancelot s’en sert pour l’apprentissage du latin et du grec à partir du français: – 1644: Nouvelle Méthode pour apprendre facilement et en peu de temps la langue latine. – 1655: Nouvelle Méthode pour apprendre facilement et en peu de temps la langue grecque. Lancelot poursuit, en 1660, avec des Méthodes pour l’italien et l’espagnol, après avoir refusé de préparer une méthode pour le français, projet qui lui paraissait trop difficile à réaliser. La première méthode d’Irson sort donc en 1656. Elle suit immédiatement la méthode grecque de Lancelot qui, exaspérant les Jésuites, servit de prétexte aux premières 1. C’est la version sur laquelle nous avons travaillé car elle existe en microfiches REF

GRAMMAIRE GÉNÉRALE ET GRAMMAIRE PARTICULIÈRE

43

fermetures des Ecoles de Port Royal en mars 1656. Et l’on remarque que les pages de titre des trois ouvrages d’ Irson (y compris l’Abrégé puisque son privilège date de 1658) sont un calque parfait de ceux de Lancelot, à ceci près que dans la Nouvelle Méthode d’Irson apparaissent les principes et la pureté, donc les règles et l’usage, les préceptes et la norme: avec la langue française, nous sommes face à une langue en cours de stabilisation, encore sujette à des doutes, qui s’achemine vers son autonomisation. Les dates de parution, le souci d’aller du plus simple au plus difficile2, la centralité du français, langue-objet et métalangue, situent donc déjà Irson dans l’élaboration par Port-Royal d’une pédagogie raisonnée des langues particulières conduisant à une réflexion sur une grammaire générale. Mais il pourrait ne s’agir que d’une coïncidence de titres et du simple souci d’un auteur de manuel de se placer dans le sillage d’une série prometteuse. En fait, l’examen des textes d’escorte permet d’être encore plus radical en ce qui concerne la parenté de ces textes. A nouveau, un rappel de dates s’impose: c’est en avril 1659, en même temps que pour les Méthodes d’italien et d’espagnol, que sont obtenus les privilèges pour la Grammaire Générale et Raisonnée d’Arnauld et Lancelot, et la Logique d’Arnauld et Nicole. Les deux ouvrages sont donc terminés à cette date, même si la Grammaire n’est publiée qu’en avril 1660 et la Logique en 1662. Or c’est en 1660 qu’Irson rédige, pour la deuxième version de sa Méthode, une préface qui contient non seulement des notions parallèles à celles de port Royal, mais également une citation mot pour mot, et sans marque de discours rapporté, de la GGR ( texte en italique dans les extraits qui suivent). Dans le chapitre premier de la seconde partie de la Grammaire, on lit ceci: Il nous reste à examiner ce qu’elle [la parole] a de spirituel, qui fait l’un des plus grands avantages de l’homme […] et qui est une des plus grandes preuves de la raison […].Ainsi l’on peut définir les mots, des sons distincts et articulés, dont les hommes ont fait des signes pour signifier leurs pensées. C’est pourquoi on ne peut bien comprendre les diverses sortes de significations qui sont enfermées dans les mots, qu’on n’ait bien compris auparavant ce qui se passe dans nos pensées, puisque les mots n’ont été inventés que pour les faire connaître.

Et dans la préface d’Irson (1660): C’est pourquoi nous pouvons assurer que la parole est un des plus grands avantages de l’Homme raisonnable […]. C’est la Raison même qui règle et qui conduit les mouvements de la Parole […]. La grammaire regarde la parole, la parole est une expression sensible de notre pensée, et notre pensée est un mouvement de notre Ame […] C’est dans cette Raison seulement que nous pouvons trouver l’Encyclopédie des Sciences […] Qui ne voit que pour bien parler il faut bien penser, et que l’incertitude de la parole se réduit à la certitude de l’Intelligence. On 2. “J’ai suivi dans ce livre la Méthode de composition en commençant par les choses les plus simples pour parvenir aux plus difficiles et celles qui sont plus composées” Préface de 1656

44

SIMONE DELESALLE & FRANCINE MAZIÈRE ne peut bien comprendre les diverses sortes de significations qui sont enfermées dans les mots, qu’on n’ait bien compris auparavant ce qui se passe dans nos pensées, puisque les mots n’ont été inventés que pour les faire connaître. C’est en ce sens que la Grammaire est universelle et qu’elle suppose la logique, au moins la naturelle, pour la définition de ses termes et pour l’explication de ses règles générales, qui sont infaillibles.

Ce texte en forme de manifeste a d’ailleurs un titre étonnant: “Préface dans laquelle l’excellence de la parole et l’utilité de la Grammaire sont démontrées”. Parler de démonstration, de Grammaire en général et du statut de la parole humaine, c’est se situer, on le voit, bien au-dessus d’une entreprise de manuélisation des règles du français3 En 56, la préface était essentiellement un manifeste en faveur de l’enseignement nécessaire de la grammaire pour “bien parler”. “Je ne nie pas que l’usage ne soit le grand Maître des langues vivantes; mais l’on doit aussi demeurer d’accord que sans la certitude des Règles, l’on ne peut jamais acquérir la perfection d’une langue”. Il reprend ainsi point par point les thèses liminaires de la GGR, jusque dans la précision de ses formules. S’il y a parfois raccourci et simplification, il y a manifestement adhésion à ces thèses. Ce que confirme le contenu du corps de l’ouvrage, qui en contient une exploitation immédiate. A preuve l’ajout, dans la version de 1660, de la connotation dans le traitement du nom adjectif. L’engagement d’ Irson du côté des jansénistes est également sensible dans les épîtres. L’épître de 1656 est adressée à Santeul, un donateur pédagogue de la famille du Santeul qui était poète et qui rédigera l’épitaphe d’Arnauld. Elle reprend à la lettre les préceptes de Saint Cyran mis en œuvre par Lancelot. L’enseignement y est vu comme la tâche de la plus grande charité et le moyen de sauver les enfants de la corruption du monde, mais aussi d’en faire, par un enseignement avisé, des “philosophes”, des “personnes utiles au Public dans toutes sortes d’emplois et de conditions de la vie civile”. Objectif qui n’est pas sans hardiesse puisqu’Irson propose un apprentissage du français à des élèves qui ne connaissent pas le latin et doivent donc assimiler uniquement les Règles de leur propre langue4. Le ton de cette première épître est serein, voire enthousiaste. Ce n’est plus le cas en 1660. La Méthode est adressée, sur un ton presque dramatique, à “Monsieur Gaudin, Primat de l’Église gallicane, Docteur de la Société de Sorbonne et chanoine de Notre Dame de Paris”. Irson demande à Gaudin son “secours” au nom d’une précédente “protection” dont aurait bénéficié la parution de la première grammaire. Et en 1667, la “Méthode abrégée”, dont le privilège est de 58, est précédée d’une épître qui s’adresse ainsi à Monsieur le chantre de l’Eglise de Paris: “Vous n’ignorez pas, Monsieur, 3. En 56, la préface était essentiellement un manifeste en faveur de l’enseignement nécessaire de la grammaire pour “bien parler”. “Je ne nie pas que l’usage ne soit le grand Maître des langues vivantes; mais l’on doit aussi demeurer d’accord que sans la certitude des Règles, l’on ne peut jamais acquérir la perfection d’une langue”. 4. Préface: “L’ordre que j’ai observé dans cet ouvrage n’est fondé que sur la fin que je me suis proposée de soulager ceux qui, pour n’avoir pas appris le latin ni le grec, n’espèrent pas de pouvoir jamais parler correctement soit dans les entretiens ordinaires, soit dans les lettres qu’ils écrivent; d’où vient qu’étant persuadés de leur ignorance, ils se bannissent volontairement des compagnies les plus considérables, de crainte d’y servir de matière de raillerie […]”.


45

que nous avons des gens qui nous traversent dans nos exercices et que nous avons besoin d’appui pour nous opposer à leur entreprise” Il est bien probable que ces traverses ne sont pas apparues en 1667 mais dès 57-58, c’est à dire au moment de la demande du privilège de publication qui est également le moment des attaques contre les Petites Ecoles. Il semble donc qu’on soit en droit de décentrer l’œuvre d’ Irson par rapport à la liste canonique des grammairiens français du 17eme siècle: Maupas/Oudin/Chiflet, pour la situer dans la lignée des travaux de Port-Royal5. On aimerait savoir d’où vient ce personnage, mais l’histoire est muette sur sa formation. Des bibliographies générales en font un bourguignon (Hœfer, Muteau), les archives de Dijon ne le connaissent pas. Les archives de Port-Royal ne signalent ni ce nom ni ce pseudonyme et les études sur les pédagogues jansénistes ne le mentionnent jamais. Il n’est donc pas possible de savoir, en l’état actuel de nos recherches, quelles sont ses fréquentations quand il combat pour Descartes et Port Royal, puis quand il entre, plus tard, au service de Colbert en tant que “Juré teneur de livres” dans une période d’apaisement entre Port Royal et le Roi, pour produire un Traité des changes, qui sera suivi de traités de comptabilité et d’arithmétique6 . Même ses domiciles renseignent peu.Al’époque de la première publication de sa Méthode, il demeure “rue Bourg-l’Abbé à l’Ecole de Charité”, la page de titre l’atteste. On sait que les maîtres et eux seuls ont le droit et le devoir de résidence dans les écoles. Or le dépouillement des archives de la paroisse de Saint Leu/Saint Gilles, aux Archives nationales comme à celles de l’archevêché de Paris, ne livre pas le nom du “maître” de cette école paroissiale7, pourtant directement rémunéré par Santeul, son fondateur, comme le signale la Fabrique. En 1667, nous savons par son Abrégé qu’il demeure dans le Cloître de Saint Jacques de l’Hôpital, au Soleil d’or, rue Saint Denis. Il ne quittera pas le quartier puisque, dans les modèles de lettres figurant dans l’Abrégé, dans ses traités de comptabilité, et encore en 1695 dans son Abrégé de l’Arithmétique pratique et raisonnée, il prend à titre d’exemple sa propre gestion en tant que “négociant” et déclare habiter le Cloître, bien qu’il possédât par ailleurs deux maisons où s’entreposaient ses marchandises. Ce silence autour de la personne d’Irson amène à la limite à se demander s’il ne s’agit pas d’un pseudonyme, d’autant que ce patronyme d’Irson, extrêmement rare (une seule entrée dans le catalogue de la Bibliothèque Nationale) pourrait faire penser à un anagramme; on pourrait par là penser à une opération de prête-nom, le véritable 5. Alors que Chiflet, grammairien jésuite dont les productions ne sont pas attaquées, cite Irson nommément dans une grammaire qui paraît en 59. Irson ne parle ni de cet ouvrage ni de Chiflet dans son tableau des auteurs de 1660, alors qu’il accorde une place très importante aux écrits de Port Royal, traductions comprises, et à Ménage. 6. Ces traités intéressent les historiens des pratiques d’écriture arithmétique, mais eux non plus ne savent qui est Irson. Quant aux rivalités entre maîtres d’école, maîtres arithméticiens et maîtres écrivains, elles ne seront apaisées qu’à la fin du XVIIème siècle. Les deux activités d’Irson ont conduit certains biographes, aussitôts démentis par d’autres, à parler de deux personnages, un père et un fils. 7. Il est vrai qu’il s’agit d’une “école de charité”. Ces écoles sont liées à des paroisses, et sont souvent rivales des écoles de grammaire soumises au Chantre; mais, à la différence des écoles de la “religion prétendue réformée, traitées de “buissonnières”, elles sont tolérées, à condition de ne s’adresser qu’à des pauvres et de ne leur apprendre que des rudiments de lecture et d’écriture. La paroisse est jansénisée et voisine l’abbaye de Saint Magloire qui favorisera à la fin du siècle un grand développement du jansénisme.

46


auteur étant alors Lancelot8. Mais on est obligé en fait de poser l’existence d’un personnage Irson, et ce pour plusieurs raisons: Les traités d’arithmétique qui nous sont parvenus témoignent d’une continuité dans l’exposition pédagogique par tableaux et par modèles concrets (lettres de commandes, comptes du mois). Une même cohérence relie ses factums contre d’autres arithméticiens et l’énumération de ses biens (des épices aux chandeliers et aux fourrures de renard blanc): à travers ces pièces se dessine un personnage qui a le goût du monde et du commerce, de la chicane et de la réussite en affaires. Surtout, son Abrégé de grammaire de 1667 constitue un jalon qui permet de saisir le virage qu’il opère: il quitte la théorie de la grammaire ( “la grammaire dans toute son étendue”) pour se cantonner dorénavant à un enseignement grammatical élémentaire qui prolonge l’apprentissage de la lecture et débouche sur des connaissances pratiques mettant en jeu autant les chiffres que les lettres (modèles de lettres en tout genre, y compris des lettres de change, et modèles de livres de compte). Mais même si on le saisit, son virage est mystérieux, comme l’est ce silence autour de lui, et autour de son renoncement à des recherches réflexives, liées à un groupe à la fois attaqué et prestigieux, et par rapport auquel il existait en tant qu’auteur d’une méthode de français. Dans l’ensemble de sa carrière, son goût de l’enseignement, lui, ne se dément pas, il l’affirme de manuel en manuel, tout en évoluant vers un enseignement plus technique. Dès 1656 il était peut-être un pédagogue hors institution9, mais c’était surtout, à cette époque, un écrivain engagé dans son siècle10. Ses travaux, jusqu’en 1662, prouvent qu’il fut un artisan de la constitution des règles de traitement du vocabulaire d’une langue. 8. Nous réservons la question du rapport à Chaulmer, dont une texte d’éloge est inséré dans l’édition de 56. 9. Gaudin serait intervenu pour “le protéger contre [ses]envieux, dans le dessein qu’il a d’instruire les personnes qui s’adressent à lui”. Il n’était donc pas simplement maître au service de la Fabrique. Etc’est un appel à élèves qu’il lance dans ses préfaces aux traités d’arithmétique: “Avis aux négociants et aux autres gens d’affaires. M. Irson. pour donner une idée parfaite du commerce, enseigne,1/l’Arithmétique, avec ses applications aux Finances et au Négoce de Banque et de Marchandise; à l’Art Militaire et à la géométrie pratique. 2/Les Négociations que la France fait avec les places étrangères[…] 3/La Manière de tenir les Livres par parties Doubles, de faire des Promesses ou Cédules[…] 10. Il faudrait développer également la singularité de la liste des auteurs ayant écrit en français, liste placée en fin d’ouvrage et qui s’étoffe beaucoup entre 1656 et 1660. Les auteurs apparaissent dans un tableau qui suit “l’ordre naturel qu’on peut garder dans la lecture des livres”: grammairiens, critiques et traducteurs, historiens, philosophes (logiciens, moraux, jurisconsultes, politiques, physiciens, mathématiciens), orateurs (en prose et en vers), théologiens. Dans le développement du tableau, Descartes est en tête des philosophes mathématiciens ( avec Roberval, Fermat, Pascal…) lui qui produit des travaux qui ne peuvent être goûtés que si l’on a “l’intelligence des mathématiques”. Dans cette même section de philosophie, l’étendue impressionnante des sujets traités en français est un constat de la victoire en cours de cette langue sur le latin. Parmi les grammairiens et les critiques les auteurs jansénistes sont abondamment cités, tandis que Maupas et Oudin ont disparu et que Chiflet n’apparaît pas. En revanche, de nombreux travaux d’Académiciens sont retenus: Mézeray apparaît et Pellisson, avec son Histoire de l’Académie, reste une référence pour Irson qui connaissait sans doute son texte capital sur “le corps de la langue” comme institution.


47

2. Les deux Méthodes En 1660, Irson n’écrit pas une Grammaire Générale mais une Grammaire du français, qui aménage en fonction de la GGR le texte paru en 1656. C’est parce que la Grammaire ne doit pas être enfermée dans des bornes étroites, c’est parce qu’elle mérite même le nom de science qu’il justifie l’extension de son dessein au delà de la stricte description des parties du discours et de l’étude de la syntaxe: “Je me suis trouvé insensiblement obligé de m’étendre plus que les autres grammairiens en donnant plusieurs traités qu’ils ont cru n’être pas de leur ressort”. Son travail sur le lexique français est ainsi englobé dans cette extension, au même titre que sa liste d’auteurs, ses commentaires sur la Vérité et la Beauté ou ses préceptes d’écriture. Et l’ensemble de la Méthode va bien au-delà du texte de 1656, qui correspondait à une méthode de français, parallèle aux méthodes de latin et de grec précédemment publiées par Lancelot: elle constitue non seulement une illustration des principes de Port-Royal mais aussi une défense contre des ennemis disposés à les anéantir. Cette deuxième Méthode prend donc de la hauteur de toutes les manières, tout en gardant sa spécificité: en tant que grammaire d’une langue, elle traite du lexique, ce que n’a pas à faire la Grammaire Générale (exclusion de son champ de toute la dérivation des mots et, par définition, du lexique français); d’autre part, et pour les mêmes raisons, elle ne suit pas les analyses de Port-Royal concernant le verbe et sa valeur fondamentale de copule affirmative; au contraire, elle s’attache à l’exploration des constructions verbales ( les idiosyncrasies de la syntaxe de régime qui ne sont qu’évoquées dans le chapitre 24 de la GGR). Bref, elle s’inscrit dans une perspective résolument portroyaliste, mais c’est pour étudier une langue, avec ses mots, ses phrases, ses périodes, ses usages et son devenir. En fait, on peut considérer l’ouvrage d’Irson comme un maillon méconnu dans le processus d’analyse qui relie les travaux de Port Royal à l’élaboration du Dictionnaire de l’Académie (Delesalle & Mazière 1998) dans la mesure où, tenant à la fois de la grammaire et du dictionnaire, il donne au lexique une place qui ne lui est pas habituellement reconnue. Du côté de la grammaire, nous avons montré la parenté du travail d’Irson avec les Méthodes de Lancelot; s’y ajoutent les principes et la pureté, arguments pour la promotion de la langue française: deux notions essentielles et complémentaires en tant qu’elles renvoient respectivement à la science et à la norme. Autrement dit, le français a une structure qui lui est propre, et c’est cette structure qu’il faut étudier, en suivant ce qu’il y a de meilleur dans son devenir. Dès la première édition de sa Méthode, Irson essaie de tenir ensemble ces deux points, puisqu’il décrit non seulement les règles de sa morphologie et de sa syntaxe, mais aussi l’usage contemporain de ses mots et de ses phrases. Et dans la seconde édition, l’empreinte de la Grammaire Générale et Raisonnée sur la Méthode se manifeste à travers la transposition à l’identique de certaines phrases et plus généralement dans le souci

48


permanent qu’a Irson d’assurer un va et vient entre une théorie d’ensemble et sa description d’une langue particulière. Du côté du dictionnaire, n’oublions pas qu’au moment de la publication de la Méthode, il n’existe pas encore de grand dictionnaire français monolingue. Ce qui existe, ce sont des Remarques, en particulier les Remarques de Vaugelas, qui étudient un certain nombre de mots et d’expressions en se prononçant sur leur validité. Irson a donc le champ libre et met à profit cette liberté pour travailler les mots de manière plurielle en constituant des listes qui ne sont pas seulement – comme c’est le cas dans la plupart des grammaires – formées d’éléments qu’on n’arrive pas à faire entrer dans les règles instituées. 3. Le rôle des listes et leur évolution 3.1 L’économie des listes Dès l’édition de 1656 ( désormais A), on peut saisir l’originalité de ces listes. Une première liste, Des mots et des phrases qui sont en usage, est suivie immédiatement d’un recensement de quelques noms dont le genre est douteux sous la forme de trois listes: Noms masculins, féminins, de commun genre. Vient ensuite, aussitôt, la liste de quelques mots dont la prononciation est douteuse. Ces trois premières listes sont liées avant tout à la pureté de la langue alors que les deux suivantes tiennent à la fois à la pureté et aux principes: la première, qui se dédouble, énumère les mots commençant par une H aspirée et ceux qui commencent par une H muette, cette différence étant justifiée par l’origine du mot. La seconde est constituée d’homophones non homographes, qu’il s’agit de ne pas confondre lorsqu’on écrit. L’auteur renseigne donc là le lecteur d’une manière précise sur les codes oraux et écrits du français. Enfin, une dernière liste intervient: Les Etymologies ou les origines et les dérivés de quelques mots français. Elle est distincte des autres dans la mesure où elle est une source d’informations qui a deux caractéristiques: 1. elle n’est pas destinée à présenter les fautes que le lecteur pourrait faire en parlant ou en écrivant, mais à l’instruire pour son plaisir en quelque sorte en lui donnant un supplément de connaissance sur la langue qu’il parle (ou qu’il aspire à parler: au XVIIè siècle, beaucoup de lettrés pratiquent le latin d’une part, la langue de leur province de l’autre). 2. le domaine de cette liste est le lexique pris en lui-même, ce qui n’est pas courant dans une grammaire. On y trouve donc l’ examen d’un bon nombre de mots, dont certains sont plus usuels que d’autres; ils sont considérés en ordre alphabétique, mais sans contrainte d’exhaustivité en ce qui concerne la macrostructure et avec la plus


49

grande liberté pour ce qui est de la microstructure. Il s’agit de voir d’où ils viennent, en restant autant que faire se peut dans le domaine français, et l’évolution qu’ils connaissent, à l’époque contemporaine, en raison du jeu des processus de dérivation. Cette dernière liste représente un effort original de systématisation lexicographique. Irson y travaille l’étymologie des mots prise au double sens que l’on trouvera ensuite développé de manière précise dans l’Encyclopédie (l’origine des mots d’une part, la productivité dérivationnelle dans une langue donnée de l’autre), alors que dans les grammaires, l’étymologie – opposée à la syntaxe – recouvre habituellement l’étude des accidents spécifiques de telle ou telle partie du discours. Cette investigation lexicale est d’ordre historique autant que méthodique et touche aussi bien à la pureté de la langue (l’usage est sans cesse questionné) qu’à ses principes (les modes de la dérivation). De plus, loin d’être isolée dans le texte, cette liste rencontre les autres listes, plus grammaticales, par une série d’éléments précis qui permettent de croiser différents points de vue sur tel ou tel mot. Prenons l’exemple de Grâces, traité deux fois, selon deux perspectives différentes. Dans la liste des mots et phrases en usage, on relève: Grâces pour faveurs se doit mettre au pluriel: car on dit gagner les bonnes grâces de quelqu’un, et non pas la bonne grâce. Et dans le Traité des étymologies: Grâce faveur, gratifier, gratuiteent; regracier vieux mot qui signifie remercier et rendre grâces. 3.2 L’évolution des listes entre 56 et 60 L’édition de 1660 (désormais B) fait évoluer l’ensemble de ces listes vers une vision de la langue de plus en plus maîtrisée: les listes ne sont pas exactement les mêmes11; dans celles qui sont communes aux deux éditions, les acceptions des mots sont reliés de mieux en mieux à leur morphologie et à leurs emplois; enfin la liste des étymologies s’enrichit d’un certain nombre de développements d’ordre scientifique et religieux, et affine les champs lexicaux élaborés dans la liste de la première édition. Nous analyserons quelques exemples pris dans chacune des listes. a) La liste des mots et des phrases qui sont en usage Elle n’occupe pas la même place dans les deux éditions. Dans A, c’est dans le chapitre Du style qu’on la trouve sur onze pages. Y figurent des néologismes ou des

11. Ainsi, la liste des homophones non homographes (sain, ceint,saint, etc...) disparaît. Cette disparition correspond au changement de perspective de l’édition de 1660, qui prend de la hauteur et ne s’adresse plus à des lecteurs susceptibles de confondre des formes de ce genre. Cette liste n’était qu’une énumération pure et simple, or Irson désormais explique et justifie de plus en plus.

50


mots récents, tels que Féliciter, dont Irson précise à la fois la construction et le sens: Féliciter quelqu’un d’une charge qu’il a obtenue, c’est s’en réjouir avec lui. On y trouve également des distinctions entre les formes différentes qu’a prises un mot en fonction soit du domaine d’emploi: Courir et courre se disent indifféremment: on doit néanmoins dire courre le cerf, le sanglier, le lièvre; courre fortune, la poste; soit des utilisateurs: Naviger se dit à la Cour, et les matelots disent Naviguer; ainsi que des réticences, voire des refus face à certaines tournures: Son monde pour ses gens ou ses domestiques est une façon de parler très basse. Ou encore: Outre cela se dit et non pas outre ce. Dans B figure une liste équivalente, mais qui comporte 20 pages et qui se situe dans la Syntaxe au chapitre intitulé: Des phrases. On y retrouve des mots qui figuraient dans A mais ils connaissent souvent un traitement plus développé. Ainsi l’article Féliciter, après un début identique, se poursuit ainsi: Ce mot Féliciter, qui est né depuis peu, est fort bien reçu. Comme aussi les suivants: conjoncture, intrépide, insidiateur, insidiatrice, insécurité, et plusieurs autres. Même chose pour Courir et courre. Après se disent indifféremment, Irson ajoute: quoique le dernier soit mieux reçu pour signifier les exercices de la Noblesse; exemple, on dit en terme de chasse: courre le cerf, le sanglier, le lièvre, la poste, courre fortune et non pas courir. Nous reviendrons sur l’importance lexicographique de ces innovations. Par ailleurs, des entrées disparaissent. En général il s’agit de formules mal considérées, telles que Outre ce ou Son monde. Et d’autres apparaissent, tels Mecredy (sic), dont Irson dit qu’il se dit mieux que Mercredy, quoique la raison soit pour ce dernier qui est dérivé de Mercure, ou Insidiateur, et insidiatrice. Ces mots sont en usage, note l’article, qui poursuit ainsi: M. d’Andilly en son admirable traduction de Saint Jean Climaque dit: La vaine gloire est une insidiatrice et une ennemie domestique, qui veut ravir le trésor de vos vertus. Le même auteur se sert aussi d’introductrice, d’exterminatrice, de coronateur. Comme dans le cas de Féliciter, l’auteur part d’une remarque concernant un mot pour aller vers des analyses lexicales plus larges. Ce procédé est amplifié dans le Traité des étymologies que l’on aborde infra. Les développements de cette liste de B renferment beaucoup de commentaires sur la langue qui tendent à accroître les connaissances de lecteurs ayant déjà assimilé la norme, avec quelques ajouts religieux destinés à leur édification. De ce point de vue, l’article Grâce est significatif. Nous avons déjà relevé le texte de A: Grâces pour faveur se doit mettre au pluriel: car on dit gagner les bonnes grâces de quelqu’un, et non pas la bonne grâce. Dans B, Dieu apparaît en même temps que la précision grammaticale: Quand on parle de la grâce de Dieu, on peut dire la Grâce ou les Grâces. Mais quand on parle des hommes, on ne peut dire la bonne grâce; mais il a gagné mes bonnes grâces; il est entré dans mes bonnes grâces.


51

Parfois les différences entre les indications de l’usage apportent des renseignements très précis sur l’évolution de la langue dans les années 1660: on va de A Découverte du Nouveau Monde se dit mieux que Découverture à B: On dit: il a fait la Découverte du Nouveau Monde et non pas la Découverture. Cette précision prend encore plus d’intérêt si l’on compare le traitement de ces mots à celui qu’on en trouve dans l’Essai d’une parfaite grammaire du français de Chiflet, paru en 1659 et dans lequel on peut lire: Découverte ou découverture des Indes sont tous deux bons. Quoi qu’il en soit, qu’il s’agisse d’une édition ou d’une autre, on peut noter que dans cette liste, Irson regroupe sous le chef de l’usage les mots qui appartiennent à différentes parties du discours et qu’il se montre par là original par rapport à un grammairien comme Chiflet, qui se contente de noter des observations de ce genre à la fin de chaque partie du discours. Par exemple, à l’intérieur des pages consacrées aux Observations des verbes, on trouve dans l’Essai d’une parfaite grammaire: Naviguer dites Naviger. Il s’agit là d’une imposition de prononciation et d’orthographe. En ce qui concerne la seule graphie, Chiflet note, dans les Observations des noms: Mercredy dites Mercredi. La comparaison entre les deux ouvrages montre à la fois des zones d’intertextualité (Parricide) et des dissonances éclairantes sur bien des points: pour Mercredy et Naviger, on se reportera au tableau de la page XXX (comparaison avec Vaugelas) C’est en tout cas sur un élément précis de ces listes d’usage que se trouve la seule allusion de Chiflet à la Méthode d’Irson de 1656, dans laquelle on trouve l’article que voici (repris dans les éditions suivantes): Recouvert et recouvré ont deux significations et deux usages différents, quoique du temps de M de Vaugelas on les ait confondus: car recouvert vient de recouvrir et recouvré vient de recouvrer, qui signifie retrouver. Dans les Observations des verbes Chiflet note: J’ai recouvert ce que j’avais perdu; dites, j’ai recouvré, qui vient du verbe recouvrer et non pas recouvrir(…) Ce mauvais usage, provenant de l’ignorance de quelques dames et de quelques courtisans qui ne savaient point de quel mot latin naissait ce verbe recouvrer s’était tellement mis en vogue que M de Vaugelas, en ses Remarques, a soutenu que ce mot était assez autorisé. (…). Chiflet, dont on sait de quelle admiration il entoure Vaugelas, le critique cependant d’avoir cédé sur ce point aux décisions de l’usage passager, en l’occurrence contraire à la raison et dit que Vaugelas a été “payé de sa monnaie” par le texte d’Irson, qui l’a fait rire: A ouïr ce grammairien, diriez- vous- pas qu’il y a quarante ou cinquante ans que M de Vaugelas est mort (…)Voilà une belle leçon pour apprendre à résister au mauvais usage des ignorants, plutôt que de lui tendre les mains, et de l’autoriser par une approbation publique.12 12. Il convient d’indiquer l’embarras de Vaugelas qui disait devoir céder à l’usage (“car l’usage est le roi des langues pour ne pas dire le tyran”), mais uniquement à la Cour, tandis qu’il dirait “recouvré avec les gens de lettres pour satisfaire à la règle et à la raison, et ne pas passer parmi eux pour un homme qui ignorât ce que les enfants savent”.

52

SIMONE DELESALLE &- FRANCINE MAZIÈRE

Ces deux grammairiens sont confrontés, dans une diachronie très courte, à l’examen des variations en référence à l’usage et la raison. Irson a, par rapport à Chiflet, une vision globale de l’usage: il dépasse, on l’a vu, le compartimentage des parties du discours; d’autre part il se situe au-dessus des querelles d’auteurs et de personnes comme le montre l’hommage à Vaugelas de la fin de sa liste: M de Vaugelas a fait des Remarques sur la langue française qui sont très curieuses et pleines d’érudition. On peut le consulter pour avoir une connaissance parfaite de notre langue. (cf.infra tableau de comparaison p.XXX). Cette vision se confirme si l’on examine l’évolution de son traitement des mots d’un genre douteux. b) L’évolution des listes traitant du genre Dans A, ces noms se présentaient sous la forme de trois listes: les noms masculins, les féminins, les noms de commun genre. Alors que B les introduit ainsi: Autre liste comprenant quelques noms dont le genre est douteux et lequel sera déterminé tant par les articles que par les épithètes qui accompagnent les noms suivants. Ainsi, là où on trouvait en 1656 dans la liste des noms masculins espace, exemple, parallèle, période, etc… et à la page suivante les mêmes mots listés dans les noms féminins, avec dans quelques cas une brève indication d’emploi, la liste unique de 1660 va permettre de poser un seul vocable et d’analyser ses acceptions de manière à justifier son inscription dans un genre ou dans un autre. Ce regroupement est évidemment beaucoup plus satisfaisant, de plusieurs points de vue. 1. Des éléments qui sont dans A répétés dans deux listes n’apparaissent qu’une fois dans B et y sont analysés d’une manière qui rend compte de la double assignation de genre que connaît un mot par la prise en compte de son environnement. Ainsi de Parallèle et de Période. Parallèle. Dans A, ce mot figure dans la liste des féminins, et dans celle des masculins avec cette seule précision: quand on s’en sert pour le sens figuré. Tandis que dans B, la liste unique contient cet article: Parallèle: étant pris au sens figuré est masculin, comme le parallèle d’Alexandre et de César; c’est à dire la comparaison de l’un avec l’autre; mais il est du féminin dans sa propre signification; comme lorsque l’on dit une parallèle, on sous-entend ligne, dont toutes les parties sont également distantes de celles d’une autre ligne. Période (A) est, dans la liste des noms masculins, suivi de “quand il sigifie la fin ou la perfection de quelque chose” et, dans la liste des noms féminins, suivi de “pour partie d’une oraison”.


53

Période (B) signifiant la fin ou la perfection de quelque chose est du masculin; exemple, on dit le dernier période de la vie ou de la félicité; mais quand période est pris pour une partie d’oraison, qui exprime un sens achevé, il est du genre féminin, comme on dit fort bien une période quarrée: c’est à dire qui a quatre membres ou quatre parties, période nombreuse, mesurée. On voit clairement à la faveur de ces exemples qu’à la simple séparation d’items identiques, de type homonymique, s’est substituée une exploration de type polysémique avant la lettre, qui fait entrer en ligne de compte aussi bien la différence entre sens propre et sens figuré que les rôles des mots dans les syntagmes. Parfois est prise en compte la différence de domaine d’emploi ( il en est ainsi pour l’utilisation d’espace par les imprimeurs); ou encore, dans le cas d’amour, la différence d’ordre pragmatique entre les utilisateurs du terme. 2. Au contraire, des mots qui dans A n’apparaissent que dans une liste alors qu’ils devraient figurer dans les féminins et les masculins jouissent en B d’une véritable analyse. Par exemple, Amour, en A se trouve uniquement dans les masculins, avec ce commentaire un peu confus: Ce nom est tantôt masculin et tantôt féminin. Quand on le prend en terme de galanterie, il est de l’un et de l’autre genre; comme il est plus élégant en parlant à une dame de lui dire: Madame, l’amour que je vous porte est bien grande. Les poètes s’en servent indifféremment. Mais quand on parle de l’amour que nous sommes obligés de porter à Dieu comme à l’Auteur de tous les biens, il n’est jamais féminin. Ce même mot se retrouve dans la liste unique de B pour être analysé d’une manière plus cohérente: L’amour est presque toujours masculin: Exemple, on dit l’amour divin, l’amour déréglé des créatures, et non pas amour divine, ni déréglée. Bien que la plupart des femmes le fassent féminin, quand elles s’en servent, ou pour exprimer leur tendresse envers leurs enfants, ou pour signifier l’amitié qu’elles portent à quelqu’un. Mais que l’on ait une ou plusieurs listes, on peut noter qu’Irson se montre toujours soucieux de l’usage des mots et de leurs emplois. Ces problèmes du lexique, des frontières entre homonymie et polysémie et du rapport entre sens et morphologie sont traités au début de la Méthode alors que l’Essai de Chiflet les rejette à la fin dans les derniers traités et comme des restes rebelles à toute règle. c) Synthèse de cette double évolution et comparaison avec Vaugelas Pour illustrer non seulement les ajustements entre A et B mais aussi les transformations profondes que connaît la langue à cette époque, nous proposons un tableau qui met en parallèle des “Remarques” de Vaugelas et des “Mots en usage” d’Irson, avec des liens aux listes de genre. La deuxième colonne distingue donc UA (liste d’usage de A), UB (liste d’usage de B) et listes de genre

54


Le choix des entrées a privilégié les inventions dictionnairiques: - la systématisation des introducteurs diatopiques (se dit à) comme pour bétail, naviger, - les indicateurs diastratiques, comme “les exercices de la noblesse” dans courre, en B. - le remplacement de la référence aux “bons auteurs” (Naviger), des jugements de valeur comme “meilleur” (Bétail, Matineux), ou des justifications (Parricide) que l’on trouve dans Vaugelas, par une norme assumée: “est préférable” (Matineux A), “est en usage” (Matineux B), “se dit de” (Parricide) - l’effacement des jugements de salon si nombreux dans Vaugelas: “il serait ridicule de dire” (Matineux), “si quelqu’un disait[…], on se moquerait de lui” (Courre) Notons que, quand la Raison est par trop heurtée, Irson le signale, tout comme Vaugelas. Le débat entre l’usage et la raison est toujours visible (Mecredy). Cette comparaison permet de saisir chez Irson l’efficacité de sa méthode d’exposition et en particulier la systématisation des distinctions de domaines, accompagnée d’une métalangue pour la description des usages diastratiques et diatopiques: se dit à, se dit de seront l’ossature des expositions dans les dictionnaires monolingues de la fin du siècle. Elle permet également, en conjuguant les listes d’usage aux listes de genre, dont on a dit l’importance, d’observer la mise en place des traitements homonymiques (Parallèle, Amour)13.

13. Parmi les accidents du nom, c’est le genre qu’Irson a le plus systématiquement traité, peut-être parce qu’il réservait la question du cas en français au traitement de l’article.


Bétail et bestial To us d e ux s o nt b o ns ma is b é ta il e s t b e a uc o up UB bestail est mieux reçu à la cour que bestial, qui meilleur. Il semble que bestial est plus dans l'usage est plus propre à la campagne que le premier de la campagne et que l'autre est plus de la ville et de la cour UA Naviger se dit à la Cour, et le matelots disent Naviger, naviguer naviguer Tous les gens de mer disent naviguer, mais à la cour UB Naviger se dit à la cour et les matelots se servent on dit naviger et tous les bons auteurs l'écrivent ainsi de naviguer Matineux, matinal, matinier De ces trois, matineux est le meilleur: c'est celui qui est le plus en usage et en parlant et en écrivant, soit UA Matineux est préférable à matinal en prose ou en vers [...] il serait ridicule de dire UB Matineux et matinal sont en usage, dont on rejette l' é t o ile ma t in e u s e o u ma t in a le [ . . . ] il s e r a it matinier insupportable de dire un astre matinier, mais, au féminin... Courir, courre To us d e ux s o nt b o ns ma is o n ne s ' e n s e r t p a s toujours indifféremment; en certains endroits on dit courre, et ce serait très mal parler de dire courir, comme courre le cerf, courre le lièvre, courre la p o s t e . S i q ue lq u' un d is a it c o ur ir le c e r f o n s e moquerait de lui. En d'autres endroits il faut dire courir, comme faire courir le bruit [...] Et en d'autres on peut dire courir et courre, comme courre fortune et courir fortune. Monsieur [...] mais sans doute courre fortune est le plus en usage.

UA courir et courre se disent indifféremment: on doit néanmoins dire courir le cerf, le sanglier, le lièvre, courre fortune, la poste. UB C o ur ir e t c o ur r e s e d is e nt ind iffé r e mme nt, quoique le dernier soit mieux reçu pour signifier les exercices de la noblesse; exemple, on dit en termes de chasse: courre le cerf, le sanglier, le lièvre, la poste, courre fortune et non pas courir.

Amour [résumé] il est masc ou fém mais pas tj indifféremment Amour de Dieu: masc Pour le rest: indifférent Ayant le choix libre, j'userais plutôt du fém Avant, obligatoirement fém. mais "depuis quelques années", le masc se répand. "quoique la plupart [dont les poètes] et particulièrement les femmes le fassent fém.

A liste genre masc Ce nom est tantôt masc et tantôt fém. Quand on le prend en terme de galanterie, il est de l'un et de l'autre [plus élégant au masc, ms les poètes souvent au fém]. Mais quand on parle de l'amour que nous sommes obligués de porter à Dieu.. il n'est jamais fém. B liste de genres douteux L'amour est presque tj masc; exemple on dit l'amour divin, l'amour déréglé des créatures [...] Bien que la plupart des femmes le fassent fém quand elles s'en servent ou pour exprimer leur tendresse envers leurs e nfa nts, o ur signifie r l' a mitié q ue ' e lle s p o rte nt à quelqu'un.

Parallèle Ce mot est masc dans le figuré. Il est vrai que dans le prope...on ne le met guerre tout seul que l'on ne d is e ligne e n mê me te mp s , une . . . , d e ux ligne s parallèles, et alors il est adjectif comme il se voit clairement [discussion sur le changement de cat...]

UB On dit se mettre en parallèle avec une personne A Liste genre masc quand on s'ent sert dans le sens figuré B Liste de genre douteux é ta nt p ris e n un s e ns figuré e s t ma s c c o mme le parallèle d'Alexandre et de C ésar, c'est à dire la comparaison de l'un avec l'autre; mais il est du fém dans sa propre signification, comme lorsqu'on dit une parallèle on sous entend ligne dont toutes les parties sont également distantes de celles d'une autre ligne

55

56


d) Les Traités des étymologies et leur évolution En ce qui concerne les mots en usage et les étymologies, l’édition de 1660 comporte beaucoup d’éléments nouveaux, plus ou moins importants mais qui vont tous dans le sens d’une plus grande clarté et d’une plus grande précision. On passe de 52 à 68 pages et Irson met en place une véritable organisation lexicale. Dans la microstructure, on peut ainsi noter l’ajout de d’où dans l’article Aiguille: Aiguille vient d’aigu d’où aiguillette, aiguillon, aiguillonner. Le mot pris comme entrée s’inscrit ainsi à la fois dans son origine et dans un processus de dérivation. Autre exemple, celui de l’article Prest. (Prêt). Dans la première édition, Irson prend ce mot comme point de départ et se contente de noter deux de ses dérivés: Prester, Prébande. Dans l’édition suivante l’article est rédigé ainsi: Prest, prester, Prébande; parce que Praestare en latin signifie donner, d’où les Prébandes tirent leur nom à cause des distributions que l’on donne aux Prébandés. Se manifeste ici le souci qu’a l’auteur d’être utile à ceux qui ne savent pas le latin; non pas en gommant l’impact de cette dernière langue, mais en explicitant son rôle lorsque c’est pertinent. L’édition B comporte par ailleurs des améliorations significatives en ce qui concerne la macrostructure, par exemple des ajouts d’article par dégroupement homonymique. Ainsi, en A on trouve un article Barbe ainsi constitué: Barbe cheval de Barbarie, d’où ces sortes de chevaux nous viennent. Barbe, barbet, petit chien velu, Barbeau poisson, barbouiller, barbon. En B, on trouve non pas un article mais deux, ce qui est beaucoup plus clair. Au delà d’une recherche de clarté, ce Traité des étymologies, proche d’un dictionnaire par sa présentation, révèle un souci d’organiser le vocabulaire en champs lexicaux, sans préoccupation alphabétique, et au delà de la morphologie.


57

Deux exemples suffiront à le montrer, Bachelier et Bègue. En A, on lit ceci: Bacheliers, que l’on nommait autrefois Baciliers, comme on le peut voir dans les anciens titres, à cause des bâtons que les jeunes Cavaliers portaient à l’Armée: & parce que ces cavaliers étaient jeunes, on appelait un jeune homme bachelier, & une jeune fille bachelette. On a transporté ce nom de l’exercice des Armes à celui des Lettres, & l’on appelle bacheliers ceux qui ont le premier degré de Théologie.

A noter l’explication par le changement de domaine, ainsi que la mention du bâton, objet d’origine. Puisqu’il existe une entrée Bâton le mot Bacheliers devrait figurer dans l’article, ce qui est le cas: Bâton, bâtir, bâcler, fermer la porte en dedans avec des bâtons: Bacheliers ainsi nommés à cause des bâtons qu’ils portaient: Balay, balayer, à cause[…] Mais en B, Irson allonge la fin de l’article Bachelier: On a transporté ce nom de l’exercice des Armes à celui des Lettres, & l’on appelle Maîtres es Arts ceux qui savent les sept arts libéraux. Bacheliers ceux qui ont le premier degré de Théologie. Licenciers ceux qui entrent en Licence. Il ne s’agit donc plus ni de dérivation, ni de transfert de domaine. En donnant la variation lexicale non prédictible Bacheliers/ Maître es Armes, Irson donne également le champ lexical obtenu par régularisation morphologique Bacheliers/ Licenciers. Et en posant cette régularité en synchronie, il libère la langue par rapport à ce qui est origine: En 1660, Bacheliers ne figure plus dans l’entrée Bâton. Quant à l’article Bègue et au champ lexical qui s’y dessine, il subit des changements qui sont certainement dus à des réactions de lecteurs (ou de lectrices). A. Bègue, bégayer, bégayement, &c. L’on remarque que la voix de l’homme ayant plus d’inflexion que celle des animaux, est exprimée diversement. Exemple on dit hucher, crier, bégayer, pour exprimer le cri des enfants; parler convient aux hommes qui usent de la droite raison; radoter aux vieillards; babiller et caqueter aux femmes; cajoler aux courtisans; baragouïner aux étrangers, &c. Le caquetage des femmes, qui va avec leur babil, disparaît quatre ans plus tard, après on ne sait quels reproches: “Exemple crier, bégayer servent proprement à exprimer le cri des enfants14; parler convient mieux à ceux qui raisonnent; radoter aux vieillards; babiller et caqueter aux grands causeurs; cajoler aux courtisans; baragouïner aux étrangers, &c.” Voilà une sociolinguistique qui moleste un peu la langue, puisque “ce grand causeur caquette” est improbable, mais qui est politiquement irréprochable. Par ailleurs, si l’on compare cette liste à celle des mots en usage, le travail d’Irson dans le domaine lexicographique apparaît de plus en plus cohérent d’une édition à l’autre comme le montre le traitement dans B de terroir et terre: 14. Hucher, sans doute vieilli, a disparu aussi.

58


Liste des Mots en usage: Terroir, territoire et terrain ont une signification différente, en ce que le premier signifie une terre qui produit des fruits; le second une juridiction, et le troisième un lieu propre pour la fortification. Traité des Etymologies: Terre, terrier, terroir une terre qui produit des fruits, territoire juridiction d’une terre, terrain, terme de fortification, enterrer, enterrement, déterrer. Dans la liste des mots en usage, les sens des trois mots sont examinés tour à tour, à la manière de ce que l’on trouvera ensuite dans le dictionnaire des synonymes de Girard.Il s’agit là de cette sorte particulière de synonymie entre des mots de même radical qui prennent, en fonction de leur suffixation différente, des sens spécifiques mais voisins. Alors que dans le Traité des étymologies, c’est le mot Terre qui est mis en entrée, en tant que mot primitif, et ses dérivés sont énumérés: terrier, terroir, territoire, terrain, enterrer, enterrement, déterrer. Autrement dit, ici, c’est la dérivation dans son ensemble qui est considérée, sans qu’on se préoccupe du plus ou moins grand éloignement des sens desdits dérivés. En revanche, comme on le constate, des explications, similaires à celles qui se trouvent dans l’autre liste, sont données pour les trois mots à ne pas confondre: terroir, territoire, terrain. e) L’organisation lexicale dans la grammaire Il est donc possible de parler d’une véritable organisation lexicale à l’intérieur de cette grammaire. Irson ne dresse pas de frontière entre règles et listes de mots, parce qu’il vise l’ organisation des sens par la forme. D’où son originalité. Il donne rarement une définition. Le sens d’un mot est traité soit par une remontée vers l’origine, soit par l’énumération des dérivés qu’il produit, soit par les deux. L’étymologie au double sens que ce mot revêt à l’âge classique est donc toujours présente: à la fois un art et une science. Mais l’étymologie -science, étude des processus dérivationnels, l’emporte de beaucoup sur la recherche de l’origine. C’est que l’auteur s’intéresse avant tout au développement de la langue française de son temps et que sa remontée vers l’origine s’arrête, sauf exceptions justifiées, au français. Il est par exemple notable que dans l’article dont l’entrée est Cas, on trouve des cascades et des cadences, mais aucun développement sur les cas grammaticaux. Nous pouvons repérer quelques constantes dans cette organisation: Quand Irson donne l’origine seule, elle peut être montée comme incertaine: Cagnard (canal, canard, chien). Aquitaine (Ville de Aqs, très arrosée ou Aqua). Les développements qui l’accompagnent doivent permettre non de justifier mais d’éclairer le sens du mot: Arrérage vient de arrière parce que c’est un droit qui rappelle les dettes anciennes qui sont demeurées en arrière.


59

Agacer: irriter, vient d’agace Pie, qui criaille toujours. Arlan: Cry de Soldats quand ils veulent piller. Ce mot est pris de la ville d’Arlem, que Frédéric de Tolède traita cruellement. La dérivation peut apparître seule, avec ou sans commentaires adjacents: Bal, baler, baladin, baladine. Nom, nommer, dénommer, renom, renommée, nomination, dénomination, auteur anonyme, c’est-à-dire sans nom ou inconnu. Armes: armer, armure, armoiries, parce qu’on en mettait la figure sur le bouclier, d’où vient que l’on dit il porte pour armes*** à cause qu’on les portait sur le bouclier; armoire, c’est le lieu où on gardait anciennement les armes et aujourd’hui est pris pour celui où on met quelque chose. Mais la plupart du temps, le mot est encadré par son origine et des dérivés. - Soit des dérivés du mot lui-même: Aiguille: vient d’aigu, d’où aiguillette, aiguillon, aiguillonner. Avec, parfois, des données encyclopédiques, surtout lorsqu’il s’agit de noms propres: Bec, Abbaye, vieux mot normand qui signifie ruisseau d’où vient Orbec, Caudebec: Kobec, Ville d’où viennent les chapeaux que l’on appelle Caudebec et Kobec. - Soit des dérivés venant du mot-entrée et des dérivés venant du mot donné comme origine: Brigand: insigne voleur d’où brigander, brigandage, brigantin, vaisseau de pirate dont les bords sont bas, brigantine, habillement, brigand vient de brigue, comme qui dirait ceux d’une même brigue ou brigade: de là vient brigadier, chef d’une brigade. A lire ce Traité, on a l’impression nette qu’Irson a une démarche qui n’est ni systématique, ni purement formelle, mais qu’il utilise les divers procédés à sa disposition pour faire comprendre le plus efficacement possible à tous les lecteurs et particulièrement à ceux qui n’entendent pas le latin, la structure du lexique français. C’est en ce sens qu’il faut comprendre des extensions à son modèle, telles que: - l’explicitation d’expressions couramment utilisées: Anglais: s’entend des créanciers rigoureux à cause des anglais qui exigeaient leur dette avec inhumanité. - la séparation raisonnée des homonymes, fortifiée par la dérivation: Canne, cannule, canon d’artillerie à cause qu’il est droit et creux comme une canne.(….) Canon, règle: d’où canoniser, canonique (…) - des élargissements à des champs purement de sens et non de forme et qui passent souvent par des comparaisons; qu’il s’agisse de métaphore: Carlet, Poisson ainsi nommé à cause de sa figure carrée; comme sole à cause de la ressemblance qu’elle a avec la semelle d’un soulier que l’on appelle en latin solea, ainsi que merlan des perles.

60


ou qu’il s’agisse de métonymie: Bougie, vient de Bugie, ville d’Afrique parce que les marchandises prennent souvent leur nom du lieu d’où elles viennent, comme brie, qui signifiait autrefois de la poix, vient de Brutie, fertile en poix; brésil, bois sec, vient du Brésil en l’Amérique. Par ailleurs, notons que ce Traité, dans son édition de 1660, retient davantage de termes scientifiques et techniques, et davantage de termes religieux et que les commentaires s’étoffent dans ces deux mêmes domaines. Ainsi, en ce qui concerne le domaine religieux, l’ajout de Assomption, Ascension ou de Pape. Dans le domaine scientifique, les explications de Précipité ou d’Aimant. Dernier point, lié à celui-ci: quelle que soit l’édition, ces listes étymologiques insistent sur le sens des préfixes ( appelés à l’époque “prépositions”) en remontant au latin et surtout au grec: il faut donner: “l’explication de la force des prépositions qui ont été examinées dans la grammaire, lesquelles donnent une très grande intelligence des mots dérivés”. ) C’est là un point crucial puisqu’il s’agit de la formation du vocabulaire abstrait qui souvent, en fait, unit science et religion: ainsi d’antéchrist et de paranymphe. 4. Conclusion Le traitement du lexique dans la Méthode d’Irson, et tout particulièrement l’ensemble formé par Les mots et phrases en usage et le Traité des étymologies peut être rapporté, en tant que complémentaire et contemporain, aux analyses de la Grammaire Générale et Raisonnée et, en tant que parallèle et précurseur, à celles de la première édition du Dictionnaire de l’Académie. Lexique et grammaire vont de pair, rappelons-le, dans cette déclaration liminaire du Traité: “Il est en quelque façon nécessaire de connaître les mots primitifs et leurs dérivés pour entendre la force de leur signification et pour savoir le lieu et l’endroit où on les doit placer”. Par là sont liés sens des mots, dérivation et propriétés des parties du discours. Comme on l’a déjà souligné, c’est l’étymologie en tant qu’ étude des processus de l’évolution lexicale qui est ici concernée puisqu’ Irson s’en tient à “l’origine prochaine des mots”, qui “comprend le mot, lequel est pour l’ordinaire le plus simple” et ne signale “l’origine éloignée”, en remontant à d’autres langues, qu’occasionnellement et si cela sert son propos. Pour les curiosités hasardeuses, il renvoie à Ménage15, Borel et Boxorius. Et du côté de la liste des mots et des phrases en usage, on voit bien qu’elle est conçue avant tout en liaison avec l’enchaînement des mots (dans la seconde édition elle fait partie de la syntaxe). Irson l’établit pour remplacer “un grand nombre d’observations”, qui, dit-il “s’entendront d’elles-mêmes par la seule lecture des mots usités et des phrases que cette liste comprend”. D’où son souci, de plus en plus net 15. Cependant Ménage insiste, dans son importante Epître à Dupuy de 1650, sur les fantaisies risibles des étymologistes qui ne connaissent pas, en plus de nombreuses langues mortes et vivantes, un certain nombre de dialectes et l’histoire des langues (bas latin, français ancien)


61

d’une édition à l’autre, de prendre en compte l’usage commun en évitant les invectives comme les effets de mode. Irson a-t-il relayé une diffusion des idées de Port Royal et transmis des techniques vers l’Académie ? Il n’est jamais cité, du moins sous ce nom, si ce n’est, une fois, par Chiflet mais, à cause de l’intertextualité, on aimerait répondre positivement. En revenant sur l’époque, il faut rappeler que Chapelain, qui est très lié à Lancelot, a terminé alors l’élaboration de ce qui sera le plan du Dictionnaire de l’Académie, et que Pellisson réfléchit alors à une fixation souhaitable de la langue par le dictionnaire, “une langue commune”, “l’indifférent de la langue”, qui serait le produit d’un apaisement dans les esprits entre usage et raison. L’Académie a la même évaluation qu’Irson de ce qu’est l’usage admis et, si sa décision d’opérer dans son dictionnaire un regroupement raisonné de familles autour d’un concept primitif est un coup de force, ce coup de force prend appui sur les principes dérivationnels posés par Irson pour la langue française. C’est pourquoi nous aimerions connaître mieux l’auteur de cette œuvre, sa formation, ses fréquentations, son évolution, afin de cerner ce qui nous paraît être un jalon important dans le trajet qui relie l’école de Port-Royal à l’Académie, ou, si l’on préfère, l’exposition des principes de la raison via le langage à celle du lexique de la langue française via la raison. Il y a nous semble-t-il originalité et invention dans l’œuvre de ce grammairien qui a su aller au delà de son domaine strict pour travailler dans une grande liberté les mots de sa propre langue en choisissant autant ses points de départ que l’extension donnée à ses analyses et à ses commentaires. Il mérite d’être reconnu comme un véritable lexicologue au sens où Pierre Larousse entendra ce terme quelque temps plus tard, c’est à dire comme un auteur qui a su, à une époque qui a connu des Remarques délibérément éparses et un Dictionnaire délibérément an-historique, relier le lexique vivant de la langue française à ses Principes.

RÉFÉRENCES

Arnauld, A. & Lancelot, C. 1660. Grammaire générale et raisonnée. Introd. de M. Foucault. Paris: (Rééd. Paris: Republications Paulet, 1969). . Nicole, P. La logique ou l’art de penser. Introd. de L. Marin, 16601683. (rééd. Paris: Flammarion, 1970). Chapelain, Chiflet. 1659. Essai d’une parfaite grammaire du français. Dictionnaire de l’Académie françoise, 1694. 2 vol. Paris: Vve J.-B. Coignard et J.B. Coignard. Girard (Abbé). 1736. Les Synonymes françois, leurs différentes significations et les choix qu’il faut faire pour parler avec justesse. Paris: Vve d’Houry. Irson, C. 1656. Nouvelle Méthode pour apprendre facilement les principes et la pureté de la langue française. . 1660. Nouvelle Méthode pour apprendre facilement les principes et la pureté de la langue française.

62


.1662. Nouvelle Méthode pour apprendre facilement les principes et la pureté de la langue française(deuxième édition). .1667 (privilège en 1658) Méthode abrégée et familière pour apprendre en peu de temps à bien lire, à prononcer agréablememt et à écrire correctememt en francais. Lancelot, C. 1644. Nouvelle Méthode pour apprendre facilement et en peu de temps la langue latine. .1655 Nouvelle Méthode pour apprendre facilement et en peu de temps la langue grecque. Masset, J. Acheminement à la langue française, bilingue français-latin, insérée en 1606 dans le Thrésor de la langue française de Nicot. Maupas, Ch. Grammaire et syntaxe française, en 1607, écrite seulement en français, et celle de Oudin, en 1632, Grammaire française rapportée au langage du temps, Chapelain Ménage. Nicot, J.1606. Thrésor de la langue françoyse tant ancienne que moderne auquel entre autres choses sont les mots de marine, vénerie et faulconnerie. Paris: David Douceur. Oudin, C. 1627. Le Trésor des trois Langues, Espagnole, Française et Italienne. Genève: J. Créspin. Pellisson, Olivet (d’). 1989. Histoire de l’Académie française (avec une introduction, des éclaircissements et des notes par M. C. L. Livet). Genève/Paris: Slatkine Reprints. Vaugelas, (C. Favre de). 1647. Remarques sur la Langue Française. Paris: A. Courbé & Vve Camusat. (Rééd. Champ libre, 1981).

63

TEXTS OF REFERENCE AND SERIAL TEXTS IN THE CONSTITUTION OF A NOTIONAL PARADIGM THE EXAMPLE OF THE FRENCH IDÉOLOGUES

GERDA HAßLER University of Potsdam The linguistic ideas of the idéologues (ideologists) have mainly been studied in relation to theories they took up and modified, as well as from the point of view of the continuation by later linguistic theorists. In this paper we will discuss whether the ideologists had really been representatives of a transition in thought. Did they make possible the explication of the school grammar categories of the 19th and 20th centuries? Were they the starting point and the background for several theoreticians, for example Wilhelm von Humboldt (17671835) and Ferdinand de Saussure (1857-1913) for whom language was above all an instrument of the articulation of thought? 1. Methodology of serial approach For this purpose we propose a serial approach to the texts produced by the ideologists. A series of texts is defined as a quantity of texts, printed or in manuscript form, dealing with the same subject in the same epistemological frame or even without declared methodology, but with the same objective and under comparable conditions. One example of the ideologists’ series of texts are the lectures and debates held at the Ecole Normale. The school opened its doors in the third year after the French Revolution and existed for only a few months in the winter of 1795. The school’s aim was to teach students who were chosen by their fellow citizens and who already had a basic knowledge of the sciences they wanted to reinforce. The Ecole Normale played the role of a kind of theoretical laboratory in the harmonisation of teaching methods (Macherey 1992: 41). Influenced by pragmatic circumstances, the lectures given by the professors and the debates that followed constituted a series. Stenographers wrote everything down in shorthand (Séances 1800-1801 and Débats 1800-1801). The methodological basis of these texts is not coherent. But the courses of lectures given by Roch Ambroise Cucurron Sicard (1742-1822) on general grammar and by Dominique-Joseph Garat (1749-1833) on the analysis of human understanding complement each other in the

64

GERDA HAßLER

analysis of the role of signs for human thinking, and the students tried to solve the same problems. A letter of the Minister of the Interior dated the 20th of Fructidor of the year five requested all teachers of the Écoles Centrales to hand over a copy of the exercise books they had dictated to their pupils. This had given occasion to another series of texts which could be called “manuals of grammar and psychology dominated by the ideologists”. This series was governed by the common objective of propagation of a method appropriate to develop human understanding. In the context of the reinforcement of knowledge, a prize essay competition was announced in 1799 which invited submissions on the topic of the influence of signs on thought. The announcement follows the tradition of European academic concourses of the 18th century, and it provoked a methodological series which shows the differences in adaptation of the ideologists to the new situation. It seems important to distinguish the social and cultural conditions and the constitution of textual series. These conditions are reflected in the manner of quotation and in the integration of a text in a tradition. The rewriting of rules and concepts is processed by successive accumulations, which gradually leads to the elaboration of new concepts. It seems possible to affirm that a text is situated at the intersection between a discourse type and a series of texts, and that the researcher must study both of them to take into account the individuality of a text. What we call reference text is a text which for different reasons has become a typical representative of a series, and it is often considered to be the starting point of a discourse. Following our approach, the ideologists’ reference texts will be studied from the point of view of their reinterpretation in the sense of a paradigm which was being constituted. 2. The ideologists and their reference authors In a paper on the idéologues’ reference authors, one is expected to emphasize the importance of Etienne Bonnot de Condillac (1714-1780). And indeed, we will start with this author and his readoption in a forgotten ideologist’s text which represents the average argumentation line: Condillac qui le premier de tous a porté dans des recherches le flambeau de l’analyse et l’a toujours porté avec circonspection, Condillac s’est écarté de la regle qu’il s’était prescripte, de ne jamais expliquer que ce qui est expliquable, et s’est égaré lui-même en voulant expliquer les causes de la mémoire. (Gattel 1800: 23verso)

This quotation is from the Cours de Grammaire Générale by Claude Marie Gattel (1743–1812), French lexicographer, who was successively a professor of philosophy at the seminaries at Lyon and at Grenoble. He then taught general grammar at the Central School of the Isère and was the head of the Grenoble lycée. While the Cours

TEXTS OF REFERENCE AND SERIAL TEXTS

65

de Grammaire Générale remained a manuscript which is now available at the municipal library of Grenoble, the published works of Gattel are difficult to recognise as written by an ideologist.1 In his Logic, the first part of the Grammaire Générale, he uses Condillac’s ideas to explain analysis, abstraction and human perfectibility, referring to Jean-Jacques Rousseau (1712-1778) for the latter. The second part, the incomplete Grammar, shows that Gattel did not know how to implement the ideologues’ epistemology. It is a traditional grammar on the parts of discourse which explains the elements of the clause to the pupils. Gattel’s example is typical for the average ideologist in the way he deals with the reference texts: referring to Condillac is common in the exposition of the analytic method, but one takes little heed of the method’s coherence and application. It is usual in the ideologists’ discourse that the authors do not attribute much continuity to the theory of Condillac, they even stress their independence from all antecedent authors. This is an expression of a break with the mechanical education which gave more attention to memory and imitation, a break which acknowledges analysis as its motto. Already in his lecture at the École Normale, Sicard called for the development of thinking: Il est tems enfin que l’homme pense, et qu’apprenant l’art de parler et d’écrire, il apprenne sur-tout le grand art de l’analyse qui produit tant de miracles dans la recherche de la vérité. (Séances 1800-1801, vol.1: 122)

In the context of this break in favour of the analytic method, Sicard refers to Condillac and to César Chesneau Dumarsais (1676-1758), but he only gives them credit for initiating a method which must be developed further: Assez de livres élémentaires nous avaient tracés des règles; aucun ne nous avait donné une méthode pour les développer. Tous nous offraient des résultats; Condillac et Dumarsais tout seuls, avaient essayé de faire usage de l’analyse: osons encore faire (Séances 18001801; Troisième Séance. 4 Pluviôse, vol.1: 129)

To continue developing this method, it is necessary to look for a simple language, a language of nature, where we can find the general principles which generate all languages: […] ces grands principes féconds qui jetteront sur les langues modernes une si grande lumière; c’est-là que nous trouverons cette grammaire générale, génératrice de toutes les autres. (Séances 1800-1801; Troisième Séance. 4 Pluviôse, vol.1: 129). 1. Gattel is known as the author of a Dictionnaire universel portatif de la Langue Française avec sa prononciation figurée et l’étymologie de chaque mot (1797), and a Nouveau Dictionnaire Espagnol-Français et Français-Espagnol (1803). Besides this he translated the works of the Marquis de Pombal.

66

GERDA HAßLER

Another idea which circulated in the ideologists’ manuals was the relation between the perfectibility of man and the use of language. For example, François Bernard Mongin (fl. 1803) affirmed that perfectibility comes from reason (raison) and the use of language (parole), without worrying about the relationship between the latter two: Cette perfectibilité de l’homme tient à deux facultés générales et naturelles, dont les développemens plus ou moins rapides donnent toutes les différences qu’on observe d’homme à homme, de peuple à peuple, d’un siècle à un autre: on les a nommées raison et parole. (Mongin 1803: 13)

The Ecole Normale professors’ lectures were certainly written with the intention to produce reference texts, the content of which should influence teaching at all central schools (écoles centrales), and they actually became reference texts in that the debates took place about them and the students used their terminology and their notional instruments. It is important to remark, nevertheless, that there are very few references to other authors in the conferences itself. Thus Garat, the professor of analysis of understanding, gives only the name of Francis Bacon (1561-1626) for the invention of analysis, adding that this method had been continued in England, Germany, and France by brave and wise men (Garat, Séances 1800-1801; Troisième Séance. 4 Pluviôse, vol.1: 144). Furthermore, philosophers had discovered that man thinks because he speaks, and that the improvement of languages would be the best means to improve our understanding: Pour l’exécution d’un pareil ouvrage, il était nécessaire, et de perfectionner tous les instrumens dont on doit se servir: l’attention des philosophes dont je parle se fixa sur les langues. Quel fut leur étonnement! En ne considérant les langues que comme des instrumens nécessaires pour communiquer nos pensées, ils découvrirent qu’elles sont nécessaires encore pour en avoir: ils assurèrent, et ils démontrèrent que pour lier ensemble des idées, que, pour en former des jugemens distincts, il faut les lier elles-mêmes à des signes; qu’en un mot, on ne pense que parce qu’on parle, que parce qu’on fixe et qu’on retient devant son esprit, par la parole, des sensations et des idées qui s’échapperaient et s’évanouiraient de toutes parts, et que l’art de penser avec justesse est inséparable de l’art de parler avec exactitude. (Garat, Séances 1800-1801; vol.1: 146/7)

To find a name for this new method, Garat uses the title of John Locke’s (16321704) Essay concerning Human Understanding. Charles Bonnet (1720-1793) and Condillac are only mentioned to reject the word psychology which had been proposed for methaphysics. Rousseau, who appears under the paraphrase of auteur d’Emile, is mentioned for what he owes to Locke’s work on education (Séances 1800-1801; vol.1: 163). Finally Garat comes to Condillac and praises him for the beauty and the clarity of his works, as well as for his contribution to the perfection of the French language (Séances 1800-1801; vol.1: 165-165). Following Garat, the spirit of simplicity, the work of distribution of knowledge, and the description of a universal system give an


67

important place to this author, but there is nothing in Garat’s lecture which would indicate that he considered him as a school founder. As the texts that were written by the stenographers and published immediately thereafter show, it was common to reproach Condillac and the other founders of the analytical method for having committed errors. These errors consisted in lacking a notional distinction which would be important for the description of human understanding, or in the reductive treatment for which the ideologists reproached Condillac very early and which mainly consists in having reduced all the faculties of the human soul to a transformed sensation (sensation transformée): Ces distinctions paraissent déliées; mais c’est pour ne les avoir pas faites, que la multitude des écrivains est tombée dans des erreurs si grossières, et que les Locke même, et les Condillac, n’ont pu éviter le vague de certaines idées; c’est pour avoir négligé de faire ces distinctions, qu’on a eu, sur l’imagination, des opinions si opposées; qu’on a regardé cette faculté brillante de l’entendement, tantôt comme la folle de la maison, tantôt comme la divinité; [...] (Garat, Séances 1800-1801; vol.1: 17) Condillac a pensé que nous formons les idées physiques sur des modèles que nous présente la nature, et les idées morales sans modèles. Je ne crois pas cette opinion de Condillac très exacte; je la mettrai à votre examen: vous jugerez si nos idées morales, c’està-dire les notions sur les vices et les vertus, n’ont pas de modèle dans nos diverses actions et dans leurs effets. (Garat, Séances 1800-1801; vol.1: 24)

There are other interpretations of Condillac which reduce all the philosophical difficulties which present a coherent sensualist system to a problem of denomination: L’abbé de Condillac fait tout descendre comme moi de l’idée, mais qu’aussitôt qu’il a fait entendre le mot idée, il passe au mot attention, qui, selon lui, est l’idée plus ou moins prononcée, et à son tour l’attention, plus ou moins prononcée, produit toutes les opérations de l’esprit. Il ne peut y avoir deux opinions, deux manières de voir sur cet objet; il ne peut tout au plus y avoir que des mots différens: ainsi l’un dira que la source de toutes les opérations intellectuelles, c’est la sensation, l’autre dira que c’est l’idée, l’autre que c’est l’attention; et tous les trois diront la même chose. (Sicard, Séances 1800-1801; vol.4: 139)

Among the reference authors there are some negative examples, and they are even more negative if their authority becomes inverted in the ideologists’ argumentation and if the disapproved theories are condensed into central notions which constitute a paradigm in contrast to the ideologists’. Thus Garat affirms that the innate ideas (idées innées) were a delirium of the genius inherent to Descartes. Sometimes the students of the Ecole Normale were not satisfied with the reference authors quoted by their professors. For this reason Garat was invited to give an analysis of Georges-Louis Leclerc de Buffon’s (1707-1788) statue. A. Person de Teyssèdre (died 1857) even asked him to sketch a rapid and reasoned

68

GERDA HAßLER

picture of René Descartes’ (1596-1650) opinions, especially of his methodological doubt (doute méthodique) which would lead to the analysis of sensations: Teyssèdre. Citoyen professeur, j’ai lu dans votre programme et dans votre première leçon, avec un grand plaisir, l’éloge que vous avez fait de Condillac, Locke et Bacon. J’ai rendu hommage à ces grands analystes de l’esprit humain. J’ai regretté de n’y point trouver un grand homme, qui a fait une révolution dans les sciences, et sur-tout dans la manière de les étudier. Je veux parler de Descartes. Je crois qu’il eût été intéressant pour l’École Normale de connaître plus particulièrement les opinions, et même les erreurs d’un homme que la patrie reconnaissante d’associer même aux défenseurs de la patrie. Je sais qu’il s’est souvent égaré; je sais qu’il a voulu, trop présomptueux, poser les bornes du monde, et pour ainsi dire, de l’esprit humain. Mais si son imagination l’a entraîné trop loin, son doute méthodique nous conduit à l’analyse des sensations; ce doute me paraît mériter notre reconnaissance; il fait un pas de géant dans la carrière de la vérité. (Débats 1800-1801; vol.1:225/226)

This objection shows that people were aware of the authors who entered the paradigm that was being built, as well of the necessity of exclusion of others. The notional core to which the analysis of sensations (analyse des sensations) certainly belongs is already clearly designed, but one seeks to construct a horizon of retrospection which should be larger and comprise more than empirical philosophy. Besides this, there was fundamental criticism about the grammatical doctrine presented by Sicard. A letter of a docteur H*** quotes authors like Locke, Johne Horne Tooke (1736-1812), Charles De Brosses (1709-1777), John Wilkins (1614-1672), Julius Caesar Scaliger (1448-1558) and James Harris (1709-1780). The originality of Condillac is largely contested, and if the reference text has already been considered as a divulgation of other authors, the question of legitimacy of a system, built on it, has to be brought up: Vous écrivez sur la grammaire, et vous n’avez point, dites-vous, la prétention ridicule de devancer des grammairiens qui vous ont précédé? Des grammairiens, tels que Condillac. Que vous avez peu approfondi un pareil sujet, si vous croyez que l’abbé de Condillac ait fait autre chose que de répéter les opinions de tous ceux qui avaient écrit avant lui sur la grammaire; aucune découverte qui lui appartienne. Rien n’est plus cavalier que la méthode de l’abbé de Condillac, lorsqu’il rencontre des difficultés qu’il ne peut se cacher à soimême, et que trente ans d’études étymologiques ont à peine laissé entrevoir à des hommes qui ont fait l’unique objet de leurs recherches. (Débats 1800-1801; vol.2:148) Sicard’s answer is not only a defence of Condillac and a detailed criticism of Harris’ Hermes, but may well be the clearest description of what the ideologists owe to Condillac’s theory: Condillac a enseigné, formellement, ce que Locke n’osa jamais dire; que c’est dans l’analyse de la pensée qu’il faut chercher l’analyse du langage: et en parlant ainsi, il a rempli le vœu de celui qui n’a vu qu’une grammaire dans l’Essai sur l’entendement humain. [...] Ce que Condillac nous enseigne sur la liaison du langage d’action avec le langage artificiel; sur les rapports de l’un et de l’autre avec la génération de la pensée, si d’autres


69

l’ont imaginé avant lui, n’a-t-il pas toujours l’avantage des vérités éternelles, qu’on pense avoir sçues la veille du jour où l’esprit les reçoit pour la première fois? J’avoue que Condillac n’a rien imaginé; mais on imagine avec lui: et les traits de lumières qui lui échappent, ne manquent jamais d’agrandir l’horizon de ses jeunes lecteurs. (Débats 1800-1801; vol.2:148)

It is possible to assume that this kind of texts constitutes a follow-up series to the exposition of the doctrine in the lectures which already had the objective of dissemination of knowledge on the theory of signs and human understanding. The new series which is constituted by the contributions of the students and the answers of the professors shows the work on terminology as well as the confirmation of what was already incontestable in the notional core of the doctrine. 3. A new way of conceptualisation of linguistic ideas If we want to understand the production of texts at the Ecole Normale, it is important to remember that it was a new form of communication in teaching at this school, creating a new relation between ideas (idées), discourse (discours), and speech (parole). This new relationship was not without consequences for linguistic conceptualisation: Les professeurs aux Ecoles Normales ont pris, avec les Représentans du Peuple et entr’eux, l’engagement de ne point lire ou débiter de mémoire des discours écrits. Ils parleront; leurs idées seront préparées, sans doute; leurs discours ne le seront point. Ni une science ni un art ne peuvent être improvisés; mais la parole, pour en rendre compte, peut l’être: ils ont pensé qu’elle devait l’être; en ce sens, tous improviseront. (Séances 1800-1801; vol.1: III, Avertissement).

It is from an already acquired linguistic capacity that children learn grammar and, by this means, they will improve their reasoning: Mais l’on ne peut apprendre la grammaire d’une langue quelconque, même celle de son pays que quand on sait parler et causer. L’enfant qui sait parler et causer, prononce donc sans cesse des jugemens, qui, revêtus de mots, forment des positions et des phrases. (Séances 1800-1801; vol.1: 130)

Oral communication, improvisation, and spontaneity correspond to the objective of the Ecole Normale, i.e. instructing citizens of a republic in which speech has an enormous influence and is even a powerful force (“des citoyens d’une république, où la parole exercera une grande influence et même une puissance” Séances 18001801, vol.1, IV/V). It is no longer the written elaborated text which counts, but oral discourse which maintains traces of spontaneity even if it is written. At the notional level, this novelty can be found in Sicard’s first lecture on grammar:

70

GERDA HAßLER Un très grand nombre d’hommes, destinés à professer les diverses sciences, s’exerceront à ce talent de la parole, avec lequel seul le génie et les lumières des professeurs passent rapidement dans les élèves. Le style a plus que la parole, de cette précision exacte, sans laquelle il n’y a point de vérité; et la parole a, plus que le style, de cette chaleur fécondante, sans laquelle il y a bien peu de vérités. L’organisation de l’enseignement dans les Écoles Normales, fournira peut être les moyens de corriger la parole par le style, et d’animer le style par la parole; et ces deux instrumens de la raison humaine, employés tour-à-tour et perfectionnés l’un par l’autre, seront tous les deux plus propres à perfectionner la raison elle-même. (Séances 1800-1801; vol.1: 12)

Oral and spontaneous discourse are described as parole, while reflected and mostly written language use is called style, refunctionalising an old metalinguistic term. The conditions of communication at the Ecole Normale would allow correcting the parole by the style, and vivifying the style by the parole, both in the service of reason. The parole is considered to be the faculty that marks a borderline between man and animals and that constitutes man. Sicard takes up this Cartesian idea without referring to Descartes, but he gives it an important significance in the ideologists’ sense: the exercise of the language faculty assures human perfectibility. Taking no regard to the sources of anthropological thought, the ideologists refer one to the other as authorities, albeit their notional system was not yet fixed. The same author uses the term parole in a different way, when he speaks about the general possibilities man has to express his thoughts either by written or oral language or by gestures, signs of physiognomy, painted or articulated signs. Incoherence in the use of terms is another consequence of the spontaneity of textual production. There were even contributors who reproached the professors for having thought about their texts with “a clear head” (“à tête reposée”, Débats 1800-1801, vol. 3: 38). Beginning with the first texts, the improvement of the intellectual capacities of man was an important objective. Human capacities being as different as they are, this means that they are not used in an optimal way and that culture plays an important role. One of the students, François-Joseph-Benoni Debrun (1765-1845), asked Sicard after his lecture on abstraction which place he gave to anomalies of language and if he considered them to be irregularities (Séances 1800-1801; vol.4:18). In a later discussion, the same student worried about anomalies in orthography. He demanded a reform which would deliver orthography from the constraints of etymology (Débats 1800-1801; vol.1: 332). 4. Deception and prudence: the idéologie and the limits of the power of signs The announcement of the lectures of the Ecole Normale of 1795 mark a direct line between the Enlightenment and Revolution, and the Revolution could not have been achieved without the progress of Enlightenment (Séances 1800-1801, vol. 1, VII).


71

The ideologists’ paradigm had been constituted with the objective to avoid reductionism of the last works of Condillac, but was implicated into a reductionism of discourse which was based on the relations between fundamental concepts. The idea of vulgarisation of knowledge by a just and simple method was always present. Vous voyez avec un grand plaisir que la Convention s’occupe de ne vouloir excepter du bienfait de l’instruction, aucun individu: elle n’imagine pas que le gouvernement ait besoin de ténèbres; elle ne croit plus l’ignorance nécessaire au bonheur public. Par conséquent, il faut apprendre à lire à tous les républicains français. (Sicard, Séances 1800-1801; vol.1:362)

The moral consequences of a coherent sensationalism induced the ideologists to keep their distance to it. But they were more “ideologist” reasons not to trust too much in the power of signs. The improvement of languages was able to produce better means for the communication of knowledge, but the progress of science depended on the revision of the scientific methods themselves. And finally, the improvement of a language of science does not dispense from its correct and attentive usage. As we have seen, the relations between reference texts and series of texts are of a remarkable functional diversity. Serial texts can contribute to the trivialisation of an argument, to put it into oblivion or to surpass it by another explanation. At the same time, serial texts can refunctionalise an important text and give it the chance to survive in a completely changed scientific context.

REFERENCES

Débats. 1800-1801. Séances des Écoles Normales, recueillies par des sténographes, et revues par les professeurs, 3vols. Nouvelle édition. Débats. Paris: à l’imprimerie du Cercle-Social. An 9 de la République Française. Désirat, Claude & Hordé, Tristan. 1975. “Les écoles normales: une liquidation de la rhétorique? Littérature et grammaire dans les programmes de l’École normale de l’an III”. Littérature, 18: 31-50. Ganault, Joël. 1992. “Idéologie et organisation du savoir à l’Institut national. l’exemple du concours sur l’influence des signes”. Azouvi 1992: 63-81. Gattel, Claude Marie. 1800. Cours de Grammaire Générale. Mr. Gattel an 8, à l’École centrale du Département de l’Isère. Manuscrit. Bibliothèque Municipale de Grenoble. R 12347. 300ff. Macherey, Pierre. 1992. “L’idéologie avant l’idéologie: l’École normale de l’an III”. Azouvi 1992: 41-49. Mongin, François Bernard. 1803. Philosophie élémentaire, ou méthode analytique appliquée aux sciences et aux langues. Par le C. Mongin, Professeur de Grammaire générale, à l’École centrale de la Meurthe. Nancy: Haener et Delahaye. Séances. 1800-1801. Séances des Écoles Normales recueillies par des sténographes et revues par les professeurs, 10 vols. Nouvelle édition. Paris: À l’imprimerie du cercle social, An 9 de la République Française.

72

GERDA HAßLER

73

PART II

LINGUISTICS IN THE 19TH AND 20TH CENTURIES

74

75

THE BRAZILIAN HYPERLANGUAGE MARK IN THE TRADITIONAL GRAMMAR OF THE 19TH CENTURY MARLI QUADROS LEITE Universidade Estadual de São Paulo Our purpose in this paper is to analyze how the traditional grammar reveals certain marks of Brazil’s empirical language. This means that we will focus on the issue of the grammatical rule construction because, if a linguistic phenomenon imposes itself to a grammar by virtue of its use, it is forced to acknowledge it, even by means of remarks loaded with judgments of the type “one should not say…”, “the use of…is vicious”, etc. In fact, it is through these types of comments that one can become somewhat closer to the linguistic reality of a community in the traditional grammar domain. Such indications of the “prescriptive rule deviations” provide real clues on the language practiced by the users of a given synchrony. Our theoretical starting point is the externalist empiricism (Auroux 1998), joined to the linguistic hypothesis instruments and to the conception of hyperlanguage. The corpus is composed of four Brazilian grammar books that have a very important position in the scope of the Brazilian Portuguese language grammaticization history, particularly for renewing the theory and the examples or to introduce comments on the use of the Brazilian Portuguese language. The analyzed grammar books are: Grammatica portugueza, by Julio Ribeiro, whose first edition is dated from 18811; Grammatica descriptiva, dated from 18942, by Maximino Maciel, and finally, Grammatica portugueza – curso superior, by João Ribeiro, dated from 18873. 1. The Brazilian grammar books of the 19th Century Julio Ribeiro is an important grammarian in the scenery of the Brazilian linguistic studies, for having been the Brazilian grammaticization initiator of the Portuguese language, which means that he was the first one to consider the Brazilian hyperlanguage in the scope of grammar, even if under the form of annotations, remarks of familiar, ordinary, rural use and others. The first edition of his grammar book was published in 1881. 1. In this work, we have used the 6th ed., which is a reprint of the 2nd ed. 2. We have used the 10th ed., of 1926. 3. We have used the 21st ed., of 1925, entirely reviewed.

76

MARLI QUADROS LEITE

Effectively, what does interest us is that the author was a real renovator, under two points of view: the work relative to some marks of the language really practiced in Brazil, and the grammatical theory renovation work. On the one hand, he introduced remarks in the grammatical text relative to the use of the Portuguese language; on the other, he renovated the grammatical theory approach, according to the comparatist (Friedrich Diez), historicist and naturalist principles, making use of the ideas of some American philologists (Willian D. Whitney, Essentials of English Grammar, from which the definition of grammar has followed) and European ones (M. Bréal, Mélanges de Mythologie et de Linguistique). On this subject, he understood that the grammar did not make its rules, but limited itself to presenting the language fact in an organized manner. In addition, he introduced another innovation in his doctrine, the grammatical division into lexicology and syntax, also borrowed from foreign authors. Finally, his grammar has become a model for many authors. Ribeiro’s grammar contains many references to the hyperlanguage, relatively to dialects and records, although he does not intend to study or to explain such uses. Thus, our task in this paper is to analyze some examples in which the author has made explicit references to the real use of the language in this reference to dialects and records. First of all, we can observe, in example (1) a reference to the familiar and to the regional language, at the phonetic level: (1) “There is still”: 1) a diminutive in ebre – casebre 2) familiar diminutives, e.g..: of pae, papae, – of thio, titio – of senhor, sôr, sô and even seu – of senhora, sóra, sia (Minas), nha (S.Paulo) – of soror, sôr.” (Ribeiro 1881: 97)

Next, in example (2) a reference is also made at the phonetic level, to the “caipira”[NTRustic] language in São Paulo: (2) There are two more distinct sounds currently banned from use by refined people: dje, tche. The “caipiras” in S. Paulo pronounce djente, djogo.. (Ribeiro 1881: 11)

In Ribeiro’s grammar there are also some references to the general use, which do not bring specifications as to the realization, if Portuguese or Brazilian, as we can observe in examples (3) and (4). Although it is not possible to say if example (3) is particularly related to the Brazilian use, example (4) is undoubtedly a remark on the manner of talking of the Brazilian people, who always preferred to place the atonic pronoun at the beginning of the phrase. We know that such problem of object pronoun placement in the phrase has been the most discussed subject as to the Brazilian use of the Portuguese language. This time the author is in the domain of the morphosyntax.

THE BRAZILIAN HYPERLANGUAGE

77

(3) “Many substantives are more generally employed in the plural; such as: 1) (...) 2) the name of paired objects, ex.: bofes (lungs), bragas (breeches), calças (pants), ceroulas (drawers), tesouras (scissors), ventas (nostrils), etc.

However, one says: grelha (grilles, treva (darkness), refém (hostage), calça (pant), ceroula (man’s drawer), tesoura (scissor), etc. and even in case of some of them, such as pant, man’s drawer, scissor, the use of the singular form is prevailing.” (Ribeiro s.d.:93) (4) The object pronoun, the pronoun in relation to the adverbial objective and to the passive rendered particle shall never begin the sentence: It would be incorrect to say: Me querem lá (They me want there) – Te vejo sempre (You I see always) – Nos parece (Us it seems) – Vos ofereço (You I offer) – Lhe digo (You I tell) – Lhes peço (You I ask) – Se contam coisas feias (Uggly things told are) – Se diz que ele vai (Said it is that he goes), etc. One should say: Querem-me lá (They want me there) – Vejo-te sempre (See you always) etc.” (Ribeiro s.d.: 255)

Example (4) presents a fact of syntax, the basic rule for the atonic pronoun placement, with a sharp remark on its use. Although the author did not refer explicitly to the Brazilian use of such syntax, we do know that such remark is supported by the fact that Brazilian people prefer the initial placement in a phrase to update the atonic pronoun, and this is a feature of the Brazilian use of the Portuguese language. 1.1 A Grammatica Portugueza – curso superior, by João Ribeiro In the grammaticization scenario of the Brazilian Portuguese language, the year 1887 was marked by the remodeling of the preparatory program for the teaching of languages, vernacular and foreign ones. The task was given to professor Fausto Barreto who worked in Colégio Pedro II. According to Maciel (1926:499-508), the new program was important because it emancipated the language teaching from the retrograde doctrines that were used in Brazil up to that time. Soon after the publication of this new program, the preparation of grammar books that were in line with the new guideline was necessary. Thus, some teachers, whose names were already known as masters in the teaching of languages, undertook the charge of writing new grammar books that would be followed in schools. Among these names, we find the one of João Ribeiro, who wrote a grammar book that has innumerable editions and has remained as the textbook for the country secondary schools during many years. The program reform regarded mainly the chance of epistemological orientation: one should abandon the philosophical orientation in order to adopt the naturalism and the experimentation. Thus, the language begins to be conceived as a living organism that is

78

MARLI QUADROS LEITE

born, grows and dies. It is within such a picture that the Brazilian philology has developed at the time João Ribeiro has prepared his grammar book. João Ribeiro, like many other authors, but differently from Julio Ribeiro, sticks to the literary language to make the exposition of his grammar rules. There are, however, some references to the Brazilian use of the Portuguese language, in spite of the fact that these are not sufficient to mark a position on the significant differences between the Brazilian and the European use. João Ribeiro’s concern in characterizing the Brazilian Portuguese language, in relation to the Portugal one can be observed in his work “Prolegomena”. When he presents his conception of grammar and its parts – Phonetics, Morphology and Classes of Words and Syntax -, he explains, as regards the Phonetics, that the “Normal Portuguese prosody is not followed in Brazil, because there is a national prosody in that country, distinct from the European one.” When the author begins his description of sounds and of their pronunciation, he explains the existing differences between the Brazilian and the Portuguese manner of speaking. In relation to the realization of both – a, one central and open and the other one medium (deep), which characterizes the European pronunciation, João Ribeiro discusses the first characteristic of the Brazilian hyperlanguage. This is read in example (5): (5) It is useful to observe that in the Brazilian phonetics certain varieties of timbre escape which are observable in the European Portuguese: the open a only appears in Indian voices (jácá, Párá) and it just distinguishes itself from the atonic a in the Portuguese voices: cása, páta. At school it is usual to teach the deep prosody in mas (conj), para (preposition), a (article distinct from a); such school distinctions do not correspond to the Brazilian pronunciation. (João Ribeiro:19-20)

In this chapter the author also describes the prosodical differences between both pronunciations as to the realization of other vowels and, as regards the diphthongs, he does not forget to mention a very important fact for the differentiation of the Portuguese regional varieties: the updating of the –em segment, which is pronounced [ãe] in Portugal and [ein] in Brazil. This is shown in example (6): (6) Other times the subjunctive i intercalates itself, as in the Brazilian prosody: tem = tein, vem = vein. In Portugal, this case has the sound of the diphthong ãe: também (in Brazil, tambein; in Portugal, tambãe). The Brazilian prosody was the same one as at the time of Camões. (João Ribeiro:22)

The difference in the pronunciation is also explained by the thesis of naturalism, either biological or environmental and also ethnical. The example (7) shows this condition:


79

(7) The prosodic differences between the Portuguese and the Brazilian way of speaking must be referred in great part to the mesological factor in addition to the ethnical one.The mesological action is especially deep in the biological domain. One should not put an exaggerated emphasis to the influence of the climate on the mental work, but it is clear that the cerebellar activity and the functions of the vocal apparatus depend immediately on the physiological state of the organs that live under the continuing environmental action. (João Ribeiro:32)

In example (8) we can see an important remark on the Brazilian Portuguese language: the preference of placing as object a subject pronoun, a characteristic fact of the Brazilian hyperlanguage up to the present days. What we wish to underline is that the author has recognized such fact as ordinary, current in Brazil: (8) In Brazil it is usual: vi ele “(I saw he)”, encontre ele “(Find he)” – archaic forms of speaking as it can be seen from examples of the ante-classical era, in documents of the 12th to the 15th Century, a fact currently acknowledged by the Portuguese philologists themselves. (João Ribeiro:90)

A typical linguistic choice in Brazil is the realization of the gerund for the indication of a durative action, while in Portugal the preference is for the infinitive. João Ribeiro did not fail to focus on that issue and made a very good remark of such fact of the Brazilian hyperlanguage, when he said in the example (9) that: (9) It is excessive to condemn expressions like: un livro contendo orações. In French, the gerund and present participle become mingled in the same way. In the Portuguese language, the function of the present participle (ante, ente) has disappeared, for the benefit of the gerund that replaced it in the ordinary language.

Differential trace between the Brazilian Portuguese and the Portugal one: Está chorando está a chorar = he is crying Ficou escrevendo ficou a escrever =he is writing Both constructions are Portuguese; however, the first syntax is more frequent and preferred in Brazil.” (João Ribeiro:292) Even if João Ribeiro’s grammar is too traditional, there are some passages considered for the comments on the hyperlanguage, as it has already been previously seen, and this evidences that the linguistic reality cannot be ignored, even in the prescriptive grammar books, whose main purpose is, as it is always asserted in the body of each grammar book, to provide the rules of utilization of a language according to the “good usage”, what nowadays concerns the literary written language, without any delimitation of time (or of space, in case the language is spoken in many countries) in relation to the choice of examples.

80

MARLI QUADROS LEITE

1.2 The grammar books of Maximino Maciel 1.2.1 Grammatica Analytica (Analytical Grammar) – 1887 The grammatical work of Maximino Maciel relates, above all, to both grammars: the first one called Grammatica Analytica, dated from the year 1887, and the second one, the Descriptive grammar, of 1984. As the author explains in the introduction, the first grammar was written in 1885, when he was 20 years old. Maybe for this reason, but also because he touched and modified the traditional model, upon its publication, his work was very much criticized. According to Maciel’s own opinion, expressed in the preface of the second edition of Grammatica Descriptiva (Descriptive Grammar), the failure of his first grammar book is attributable to three reasons: 1. the use of modern theories and the rupture with the tradition; 2. the epistemological transition period (from historicism to naturalism); 3. the lack of observation and of experimentation on the language. We may add to these reasons the fact that Grammatica Analytica is not a school book, but a theoretical book in which there existed the intention of “systematizing” the linguistic phenomena, specially at the light of the biological laws. It is interesting to notice that Maciel has created a “special” place where he fits certain Brazilian usages. Such space is the place of forbidden usages, regarding what “one should not say”, the place where the linguistic usages of people who have no social prestige exist. However, we consider important the fact that the author has placed such linguistic usages in the scope of his grammar book, because, this way, we may infer some traces of the linguistic reality of that time. As it could be verified in the precedent passages, there is the statement on the “value” of the mentioned uses. In the summary (10) Maciel exposes his judgment in relation to the usages as such in a very clear way, for example, by the expression “vicious”, “popular use”, “people who have no instruction”. (10) “Brazilianism According to the trends of the Portuguese language , the construction of the phrase spoken in Brazil is called Brazilianism, for ex.: Vi ele “(I saw he)”, guardei ela “(I kept she)”, instead of vi-o (I saw him), guardei-a (I kept her). The use of subjective personal pronouns employed in an objective function currently constitutes a Brazilian construction vice, used even by learned people. Another Brazilianism of popular use is the usage of the pronominal form mim (me), instead of the subjective function eu ( I ) before reduced phrases with the preposition para (to), ex.: “O livro é para mim ler” “(The book is for I to read).” In phrases where the relative pronoun is preceded by a preposition, the uneducated Brazilians place the preposition at the end of the phrase, and the personal pronoun rules, accommodated to the meaning, ex.: O homem que eu estive com ele. O livro


81

que precisei dele (The man who I was with him. The book that I needed it)”. (Maximino Maciel 1887:263) These are the comments about the language practiced in Brazil that Maciel registered in this grammar book. Two facts are clear as to the author’s analysis on the Brazilian hyperlanguage: according to his opinion, the Brazilian variety is incorrect and this occurs because there are factors determined by the society and by the human biology that rule the language evolution, which is a living organism. From now on, we will observe what the author has verified on the Brazilian hyperlanguage and has added to this work of 1894.

1.2.2 Grammatica Descriptiva (Descriptive Grammar) – 1894 In Grammatica Descriptiva, Maciel wanted to get rid of the criticisms that have marked the precedent grammar. In order to avoid them, he affirms in the preface that he has proposed changes in the doctrine, but that all of them were supported by texts of the best authors, those whose philological theories are modern and also internationally acknowledged. In that which concerns the example, on the one hand, the author says he has chosen the best ones in the good Portuguese literature and, on the other hand, that he has created very few examples. In this grammar book, the “remarks” regarding the Brazilian Portuguese language are not numerous, even if they outnumber the precedent one. In the chapter on the Phonology/Phonetics, for example, he did not make more than one remark on the Brazilian realization of the Portuguese language and did not mention differences between both linguistic varieties: the whole description seems to refer to a homogeneous language. The vowel system is not explored, but only partially presented and just the following vowels: the central /a/, the intermediary /e/ and /o/ and the high ones /i/ and /u/, without any reference to the opening, closing and nasalizing or to the several possibilities of their achievements. As we know, the lexicon is the domain in which the novelties are the most evident ones. Thus, the author attempts to characterize the Brazilian Portuguese language mainly by the presence of aboriginal and African words. Therefore, to provide the origins of the proper nouns, Maciel includes, among others, the aboriginal origin, as we verify in the example (11): (11) “Lexicogeny of the proper nouns (1) The personative nouns may derivate from any language. Therefore we have: A) (...) H) Of aboriginal origin, mainly in Brazil: Moema, Coema, Jacy, Aracy, Iery.” 1) Therefore, we designate the appertaining theory to the origin and formation of the personative names. Maciel, 1894, p. 242. Also, in order to explain the general Portuguese lexicon formation, he adds the aboriginal and African term’s heritage specific to the “Brazilian language”. Example (12) reveals this situation:

82

MARLI QUADROS LEITE

(12) Secondary elements. A) (...) H) African ones: batuque, moleque, cangica, samba, lundú, cachaça, vatapá, angú, inhame. I) Aboriginal ones: mandioca, caipira, caroba, pagé, sabiá, (1), goiaba, ingá, pitanga, beiju, cará, mutum, araponga. 1) The African element has spread throughout Northern Brazil and the aboriginal one everywhere, because it reaches almost 6.000 words. This aboriginal lexicon predominates in locative terms and in those appertaining to the plant and animal kingdom, for ex: Niteroy, Andaray, Paraná, Sergipe, Aracajú, Itabaiana, jucá, capim, abacate, caroba, cajú, goiaba, sabiá, onça, jaguar, nadú, maguary.” (Maciel 1894:260) In relation to the syntax, Maciel has made two remarks regarding the Brazilian use of the Portuguese language. The first one, as to the pluralization of the verb haver (there to be) when it is impersonal; the second one, in relation to the employment of the verb ter (to have) instead of haver (there to be). It is interesting to note, in example (13), how the author leans over this point in order to explain and also justify it, saying that it is a form of use of educated people; however, in his first grammar book, this use has been totally rejected. The summary in (13) presents the author’s new position: (13) “In the Portuguese dialect formation, in the prosody of the Portuguese people, the verb haver (there to be), in the 3rd present of indicative appears followed by the ancient adverb hi (ahi), which followed it in the archaic Portuguese. Therefore, whenever we hear ha hi agua, ha hi gente, ex..: ‘If plague was not, all my errors They would not know that ha there was’. {Si peste não fosse, todos meus erros Não conheceriam que ha havia’. (Camões, Não dos Amores)}

In the Brazilian dialect formation, there are two irregular and anomalous phenomena, relative to the usage of the verb haver (there to be): 1°The plurality attributed to it, even by somehow cultured people, who incorrectly use the object as subject, for ex.: ‘Haviam pessoas, houveram festas’, instead of havia... houve... As regards this syntax violation, there is always the resistance to the use of the monosyllabic form ha, because such form is never replaced by its plural hão, since such construction would become aneuphonic. 2º Its replacement by the verb ter (to have), one of the most constant and general phenomena, observable even among cultured people, ex.: ‘Na festa tem (ha) muito povo, ‘ At the party have (there are) many people’ ‘Tinha muita água na rua.’ ‘Have (there was) much water on the street’.


83

Still in the formal language, if not approved, such extremely frequent replacement of the verb haver (there is) by ter (have) constitutes one of the most characteristic features of our way of speaking if compared to that of the Portuguese people, in the familiar relations. However this construction, marked as Brazilianism, was brought by the Portuguese people, who landed here at the colonization time.” (Maciel 1894:397) In the passage designed to the “Syntaxiology”, that is, to the syntax studies, Maciel deals with the dialects of the Portuguese language, which he defines as alterations or deviations from a language in relation to another one, which is the “matrix”. In order to become harmonized with the theory he embraces, the author tries to position the idea of dialect formation in the frame of naturalism, asserting that the dialects are organic and structural transformations that the Portuguese language has suffered in Brazil’s tropical environment. The dialectal difference between the Brazilian and the European Portuguese language, according to the author (Maciel 1894:451) is carried out “thanks to the following organic characteristics: the coexistence of our autochthonous words, the almost definite popular or literary changes of the Portuguese word structure; the semantic alterations, that is, the acquisition of new meanings, truly ours, patriotically ours. Indeed, Maciel admits a strong differentiation between the Portuguese variety and the Brazilian one, and that the language unity can be made by the knowledge of the linguistic tradition. This is what can be understood from example (14): (14) “In relation to the language, sensitive divergences occur comparatively when we and Portuguese people have to speak it, so that if certain Portuguese individuals had to talk to our “sertanejos” [TNRustic Inlanders] they would not understand each other, provided both were illiterates”. (Maciel 1894:453) The author’s requirement is so harsh that he criticizes the cultured Brazilian people who “pronounce certain words in an incorrect and dialectal form”; this means that in Brazil, at this time, those people who wanted (or seemed) to be educated should talk like the Portuguese people. Contradictorily, the author says that maybe the “lexical Brazilianism is amusing, showing the mental elaboration of our people and attesting the aboriginal and African words, the ethnographic factors, inherent to our nationality” (Maciel 1894:445). However, in relation to syntactic facts, Maciel is implacable and condemns them all. In order to clarify the issue, he enumerates them, and this can be verified in: (15) “A) The employment of the pronominal forms ele, ela, and lhe (he, she and to or for him/her/it) as direct objects (…); B) The employment of the pronominal form mim (me) as subject of the infinitive, ruled by the preposition para (to), instead of the subjective form eu ( I ), ex.: (…);

84

MARLI QUADROS LEITE

C) The employment of the preposition em (in), instead of the preposition a (to, at, etc) after the verbs of movement, or when only denoting proximity, ao pé (at the foot of), junto a (close to), ex.: (…); D) Transfer into the relative pronominal preposition the preposition appertaining to the relative one, giving them as complement the pronominal forms ele, ela, eles, elas (he, she, it, they), ex.: (…); E) The incorrect position of the pronominal forms – me, te, se, nos, vos, o, a, os, as, lhe, lhes (me, you, us, them, the, to him/her/it), in disagreement with the observed syntax rules (...); F) The employment of the verb ter (to have), instead of haver (there to be) under impersonal form (...).”(Maciel 1894:455-456) It is interesting to observe that, in relation to this last point, F), the author still makes a significant commentary, which shows very well his ambiguous attitude in face of the Brazilian linguistic variation, because, as we have seen in the example (13), his comment was not so unfavorable to the Brazilian use. This commentary also brings the acknowledgment that the use has the power of changing the language rule. In the following example, the author says: (16) “This substitution of haver (there is) for ter (to have) constitutes one of the most ordinary Brazilianisms which, as it seems to us, will eventually triumph, waiting for the acknowledgement of our writers, because they only use it when transporting the popular phrase to the literature, mentioning it directly, eg.: (...).” (Maciel 1894:457) As to the expression estar com (to be with) instead of the verb ter (to have), Maciel is contradictory, because he approves its employment, even being a Brazilianism like the precedent one. Let us see what he affirms as regards the victory of the Brazilian form of speaking the Portuguese language in summary (17):

(17) “Thus, to the construction – estar com = ter (to be with = to have), such as ‘estou com fome’ (I am hungry), ‘estava com frio’(I was cold), etc, there succeeded the more elegant one, more Brazilian than the one of the verb ter (to have) – tenho fome (I am hungry) teve frio (He was cold)TN. (Maciel 1894:457) The impression that we have in Grammatica Analytica is that Maciel was not a grammarian concerned with the exposure of the language facts. It seems to us that his purpose was to take the philological theories that were fashionable at that time, abroad and in Brazil, and to apply them, with a view to the Portuguese language, but without effectively exploring the phenomena of this language, neither relatively to the literature, nor to the linguistic reality. Maybe for this


85

reason Elia said that Maciel “was more a grammarian in the sense of systematizing the language facts” than a didactic grammarian. And, Elia adds, Maciel’s classifications have been adopted by the secondary course teachers during many years. Final considerations This look over some grammars of the 19th Century shows that, even if the traditional grammar does not aim at describing the effectively realized language, the linguistic reality, in its spoken and written modalities, it registers traces of such reality. Here we could feel that it is not easy for the grammar to deal with a language without making at least a small reference to the linguistic practice, what can be explained by two reasons: first, because the grammarian is a language user too and the knowledge he has about it forces him to make comments about what he disagrees in relation to what is mostly realized; next because the grammarian is a kind of “guardian of the tradition of the language” and, for this reason, he is taken to denounce everything that represents a threat, therefore, he records the uses he considers incorrect, vicious, ordinary, popular, etc, and this allows to trace a profile of the linguistic reality of the hyperlanguage. Thus, the hyperlanguage is observable by two main types of comments: the complaints about the linguistic “bad use” practiced by uneducated people – which are more frequent; and the descriptive comments on the Brazilian way of speaking. These complaints and the comments touch the problems relative to dialects as well as the language registers. Finally, we can say that the Brazilian hyperlanguage is present in this linguistic instrument called traditional grammar and that such fact allows, on the one hand, people to have an idea of the linguistic reality of a time and, on the other hand, researchers to study facts that initially look like deviations and then become normal, according to the traditional perspective.

REFERENCES

Auroux, S. 1994. Remarques sur l’histoire philosophique du concept ‘norme’et sur l’histoire des sciences du langage: éd. par Jean-Michel Kasbarian Genèse (des) normes linguistique(s), 295-301. Aix-en-Provence: Université de Provence. . 1998. La raison, le langage et les normes. Paris: PUF. . 1998. Introduction. In: Langages, 120 – L’hyperlangue brésilienne. Par Sylvain Auroux, Eni Orlandi, Francine Mazière. Paris: Larousse. Barros, D. L. P. de 2000. Conceitos e imagens da norma no português falado no Brasil: o discurso da gramática. Texte inédit. Chevalier, J-C. 1993. La formation de la norme du français à la lumière de l’Histoire de la langue française de F. Brunot. éd. par Jean-Michel Kasbarian et le Centre Dumarsais (Centre des sciences du langage). Genèse de la (des) norme(s) linguistiques. Actes de la 3ème Table ronde de L’Aprodelf, Aix – 18-19 juin.

86

MARLI QUADROS LEITE

Maciel, Maximino. 1887. Grammatica analytica. Rio de Janeiro. . Grammatica descritiva. 1894 [1925]. 21ed. 1894. Rio de Janeiro: Francisco Alves. Orlandi, Eni P. e Guimarães, E. 1998. La formation d’un espace de production linguistique. La grammaire du Brésil. Langages – L’hyperlangue brésilienne, 120. Par Sylvain Auroux, Eni Orlandi, Francine Mazière. Paris: Larousse. Ribeiro, Julio. 1881. Grammatica potugueza. (s.d.) 6ed. São Paulo: N. Falcone & Comp. Ribeiro, João. 1887. Grammatica portugueza – curso superior. (1904) 11ed. Rio de Janeiro: Francisco Alves.

87

REVISTA ILUSTRADA UN DOCUMENT SUR LE LANGAGE DES NOIRS À LA FIN DU XIXe SIÈCLE

MARGARIDA MARIA TADDONI PETTER Universidade Estadual de São Paulo Introduction La presse s’est installée au Brésil au XIXe siècle, après l’arrivée de D.Jean VI et de la famille royale à Rio de Janeiro, en 1808. Vers les années 1850, la presse était clairement séparée: presse académique, presse politique, presse littéraire et presse féminine. La presse humoristique précéda la caricature, qui débuta en 1855, avec des publications régulières. Parmi celles-ci on reconnaît les journaux illustrés, qui disposaient d’une liberté dont la presse politique, plus conservatrice, ne disposait pas. Parmi ces journaux, la Revista Ilustrada (Revue Illustrée) ressort comme la plus notable publication. Cet hebdomadaire fut publié de 1876 à 1895, à Rio de Janeiro, par Angelo Agostini, qui la dirigea jusqu’à 1888. D’après Nelson Werneck Sodré (1966) ce journal est la plus importante publication du genre à l’époque du Second Empire et il représente l’histoire illustrée de cette période. Selon Joaquim Nabuco il était “la bible de l’abolition de ceux qui ne savaient pas lire”, parce que les scènes qui montraient la souffrance des esclaves ont dû certainement être plus convainquantes que les discours des abolitionnistes. La même opinion est partagée par Herman Lima (1963): “il n’y a pas eu, dans le passé de la presse brésilienne, une publication d’une position aussi nette ni de plus haute expression documentaire d’une époque de notre histoire, au point de constituer d’une façon irrefutable l’une des sources les plus sûres et pondérées pour sa connaissance et son analyse”. Monteiro Lobato reconnaissait la grande valeur artistique et politique d’Agostini. Lobato mettait l’accent sur la grande diffusion et circulation de la Revista Ilustrada, lue aussi bien en ville qu’à la campagne avec le même plaisir. Les gens admiraient l’art du dessein d’Agostini, un italien qui arriva au Brésil en 1859 et y resta jusqu’à sa mort en 1910. Il a introduit la caricature au Brésil et a été pour certains le créateur de la bande dessinée (Cagnin 1994). La Revista Ilustrada est un document où le langage visuel se combine avec le langage verbal. La plupart des caricatures sont accompagnées d’un sous-titre ou

88

MARGARIDA MARIA TADDONI PETTER

d’un dialogue entre les personnages qui représentent les acteurs de la société de l’époque. On y voit défiler les habitants du milieu rural et du milieu urbain, les hommes et les femmes, les riches et les pauvres, les Blancs et les Noirs, les Brésiliens et les étrangers. Nous nous proposons de démontrer, à partir de l’analyse des textes des dialogues, que cette revue est un document important aussi bien pour la représentation du langage des Noirs que pour le portugais brésilien de l’époque, dans un genre de texte non littéraire: le journal illustré. Nous développons la suggestion présentée par Alkmim, sur l’importance des charges (caricatures) comme représentation du langage des Noirs au XIXe siècle (2002: 383-402). 1. Langue écrite et langue parlée Nous nous intéressons aux dialogues qui accompanent les caricatures parce que ces dialogues écrits peuvent être un moyen pour accéder à l’oralité de l’époque concernée. Ces dialogues essayaient de reproduire le langage parlée par les différentes couches sociales et par les différents groupes ethniques. Il est vrai qu’il y a le risque de trouver des stéréotypes, mais ceux-ci sont importants pour l’analyse, puisqu’ils nous renseignent sur les représentations de la variation linguistique à l’époque. Il convient de rappeler que les documents sur le langage des Noirs au Brésil sont rares. Il existe beaucoup de travaux sociologiques et historiques sur l’esclavage, les conditions de vie des populations noires, les religions des Noirs, mais presque rien n’a été écrit sur leur langage. Au XVIe et XVIIe siécles il n’y a pas de références à ce langage. Au XIXe s., il y a quelques indications dans les textes des historiens, des voyageurs et des écrivains. Par contre, la pressse – les journaux illustrés surtout – nous offre une opportunité singulière: celle de représenter le langage des noirs dans un contexte de contestation de l’esclavage. Dans ce sens, nous allons examiner des dialogues extraits de trois numéros de la Revista Ilustrada: le premier publié avant l’abolition de l’esclavage, en 1886, nº 425, et les deux autres publiés après l’abolition, en 1888, les numéros 499 et 510. 2. Les dialogues des caricatures (i) 1886 – nº 425 Le dessein nous montre un salon où il y a une jeune fille assise sur un canapé, avec deux monsieurs assis à ses côtés. D’un côté il y a un vieillard, un chapeau à la main droite et portant à la main gauche une boîte de bijoux; de l’autre côté il y a un

REVISTA ILUSTRADA

89

jeune, le chapeau à la main et un mouchoir avec lequel il essaye de cacher des larmes. Au fond de la scène il y a un rideau, derrière lequel se cachent deux filles, des jeunes esclaves. Il y a un titre: A Nação Brasileira entre dois partidos políticos, no dia da eleição ‘La Nation Brésilienne entre deux partis politiques, le jour des élections (le 15 janvier 1886)’ [Les deux partis politiques Lib. = libérateur; Cons.= conservateur] (à esquerda) Cons. – Então, és minha, não é assim? Nação – Que remedio tenho eu senão entregar-me ao Snr.! Eu antes queria o outro, mas elle ficou desempregado e já não me pode sustentar... (no centro) – Lib. – Ah, perfida! (à direita) – Uê! E não é que a Sinhá aceitou o velho! – Chii! Não é tão cedo, agora, que nós fica forro. “(à gauche) Cons. – Alors, tu es à moi, n’est-ce pas Nation – Que puis-je faire sinon me livrer à vous? Jadis j’aurais préféré l’autre, mais il est au chômage à présent et il ne peut plus m’entretenir... (au centre) – Lib. – Ah, perfide! (à droite) – Uê! Et voilá que Sinhá (mademoiselle) a accepté le vieillard! – Chii! Ce ne sera pas de si tôt, maintenant, que nous est [sommes] affranchi[es].”

Malgré l’impact du langage visuel, que malheureusement nous ne pouvons pas reproduire ici, il est évident que le message verbal donne des informations importantes que l’image ne peut pas transmettre: l’identité des personnages et le sujet de leur discours. Les dialogues reflètent les conflits entre deux groupes de personnages: ceux qui décident et ceux qui dépendent des choix des “maîtres”. Leur langage reproduit aussi ces différences sociales: la Nation et les deux messieurs s’expriment dans un niveau de langue standard, très proche du portugais européen, marqué surtout par la place des pronoms clitiques (me). Le langage des servantes ne respecte pas les règles d’accord: nombre et personne, pour le verbe “ficar”- fica, au lieu de ficamos; genre et nombre pour l’adjectif, masculin à la place du féminin, singulier pour le pluriel: forro à la place de forras. Le décor de la scène montre que tout se passe dans un milieu urbain. Après l’abolition (ii) 1888- nº 499 Dans la première page de la Revue: Un Noir déchaussé, la houe à la main, deux hommes blancs à ses côtés, en montrant des billets et posant leurs mains sur les épaules et les bras du Noir.

90


En bas il y a le titre et les commentaires du Noir: A lavoura e os actuaes libertos. “Le labour et les actuels libérés.” – Uê ! Honte tanta lambada p’ra trabaiá, e hoje só dinhero e adulação. Eh! Eh! “– Uê! Hier quantité de coups de fouet pour travailler, aujourd’hui argent et flatteries seulement. Eh!Eh!”

Cette fois-ci la scène se passe en milieu rural où le Noir n’est plus exploité, mais il est flatté et bien payé pour cultiver le champ. Cette caricature pourrait se passer du langage verbal, mais celui-ci est important parce qu’il réitère le langage visuel et ajoute des commentaires de l’ancien esclave sur le changement opéré par l’abolition. L’exesclave continue d’utiliser le portugais d’une façon particulière, comme le montrent ces écarts à la norme: la chute de la nasale finale (honte-)[ontem], l’iéisme – la latérale palate devient [j], la chute de la finale [r] de l’infinitif “trabaiá” [trabalhar] et la réduction de la diphtongue [ej] “dinhero” [dinheiro]. (iii) 1888 – nº 510 Il y a deux pages avec 4 caricatures sous le titre: A festa da Glória e alguns efeitos da lei de 13 de maio – “La fête de la Gloire et quelques effets de la loi du 13 Mai” (la date de l’abolition) 1ère caricature – montre une famille noire bien habillée (père, mère et deux enfants) avec le sous-titre: Pai Zuzé Congo e sua Exma familia “Père Zuzé Congo et son excellentissime famille”

2e caricature – Il y a un Noir bien habillé, assis, les chaussettes et les chaussures à côté, en massant ses pieds. Comme sous-titre il y a le commentaire du personnage: – Liberdade é muito bom, mas cria callos que é o diabo! – “La liberté c’est très bien, mais cela fait des cors......c’est un enfer! “

3e caricature: Deux noirs bien habillés, debout, discutent, l’un d’eux tient les chaussures à la main. Comme sous-titre il y a le suivant commentaire du personnage: – Fiquei com o corpo livre, mas estou com os pés no captiveiro! – Mon corps est devenu libre, mais j’ai (encore) les pieds en captivité!”

4e caricature: Un Noir bien habillé observe que la poche de derrière de son pantalon est vide. Comme sous-titre il y a le commentaire du personnage:

REVISTA ILUSTRADA

91

– Roubado pelas gatunas ! Eis uma sensação que nunca tive antes da lei! “– Volé par des filous! Voilà une sensation que je n’ai jamais eue avant la loi!”

Toutes ces quatre caricatures montrent, avec ironie et humour, les Noirs du milieu urbain comme des personnes des classes moyennes: soit bien adaptés à leur nouvelle situation (caricature 1), soit critiques de cette nouvelle position sociale, remarquant les inconvénients qu’elle apporte au corps et à la poche (caricatures 2,3,4). Leur langage aussi est bien (adapté) au milieu social, il ne s’éloigne pas de la norme. Il n’y a qu’une forme déviante, dans le sous-titre du premier dessein, qui renvoie à la façon de parler des Noirs: ‘Pai’, “père” pour Monsieur, et le nom Zuzé pour José (Joseph, où le [z] prend la place de [Z] et la voyelle [o] en position non accentuée devient [u]. 3. Analyse Etant donné que la Revista Ilustrada fut publiée à Rio de Janeiro, les faits de langue observés dans tous les textes analysés sont limités à leur contexte de production – la ville de Rio de Janeiro. Par ailleurs, pour affirmer qu’il s’agit de la représentation d’une expression particulière à la population noire, les données doivent être mises en rapport avec d’autres représentations de langage non standard de la même époque, comme celles des étrangers et des paysans. Les caricatures publiées avant l’abolition montrent le langage marqué de la population noire, caractérisé par des expressions stéréotypées. Le Blanc parle un portugais plutôt “européanisé”, très conforme à la grammaire portugaise, le Noir présente une expression différente, “africanisée” peut-être. Après l’abolition, les caricatures nous montrent une séparation entre les parlers ruraux et les parlers urbains associés à la situation économique des locuteurs noirs: le travailleur noir à la campagne parle comme un étranger, un Africain; en ville, le Noir, avec assez d’argent, parle comme un Blanc. Cette situation est le reflet du changement de la société brésilienne – un pays rural qui devient un pays urbain. En 1900, 61% de la population vivait en milieu urbain (aujourd’hui 82%, 2000). L’urbanisation a eu comme corollaire linguistique l’effacement d’une variation régionale en faveur d’une variation plutôt sociale, qui opposait des locuteurs de niveaux socioculturels distincts, les couches les plus basses s’intègrant progressivement à la population d’origine rurale. En guise de conclusion Le Noir libre, habitant le milieu rural, tel qu’il est présenté dans la Revista Ilustrada “parle comme un Blanc”. A partir du moment où il obtient sa liberté, et devient un citoyen, s’effacent tous les vestiges de son portugais marqué par le fait d’être africain. Cette “appropriation”/ aquisition rapide du langage standard, non marqué, est exceptionnelle

92


dans la représentation du parler des Noirs dans la littérature en langue portugaise. Déjà au XVIème siècle, Gil Vicent, l’auteur portugais qui a introduit des Noirs dans son théâtre, nous montrait des personnages d’origine africaine incapables de bien parler le portugais. Dans son œuvre Frágoa d’amor (1524) il y a un personnage noir qui a son voeu exaucé – il devient Blanc. Le miracle, cependant n’est pas complet et le Noir se desespère, parce qu’il se rend compte que son expression en portugais n’avait pas “blanchi”, il continuait à parler comme un Africain, avec la même prononciation et la même syntaxe “fautives”. La représentation du portugais marqué des Africains construite par Gil Vicente, le premier modèle de la représentation du langage des Noirs, a été très répandue comme l’image du portugais des Noirs, de leur incapacité innée, viscérale de bien parler le portugais. Les Noirs du milieu urbain des caricatures d’Agostini sont moins marqués que ceux de Gil Vicente. Cependant, ils gardent toujours une difficulté d’adaptation à leur nouvelle condition d’hommes libres, même s’ils parlent comme des Blanc. Il leur manque une adaptation à la vie en société et qui les sépare des Blancs. Du point de vue de leur langage, restent des formes indissociables de leur image: le remplacement des fricatives /Zð/ par /zð/, dans Zuão à la place de João et les expressions Pai (Père), Mãe (Mère) utilisées par les Africains pour s’adresser respectueusement à d’autres Noirs. Les caricatures propagent les stéréotypes construits sur le langage des différents locuteurs/acteurs de la société brésilienne. Malgré l’exagération, elles reflètent une observation attentive de plusieurs caractéristiques phonétiques et syntaxiques du langage de l’époque. Les faits de langue observés – absence d’accord en nombre dans le syntagme nominal et dans le verbe, surtout – se sont conservés jusqu’à aujourd’hui dans les dialectes ruraux et dans les variétés non standard du portugais brésilien.

RÉFÉRENCES

Alkmim, Tania Maria. 2002. Estereótipos lingüísticos: Negros em Charges do séc.XIX. In Alkmim, T. M (org.) Para a História do Português Brasileiro- volume III: Novos estudos. São Paulo: Humanitas/FFLCH/USP. Cagnin, Antonio Luiz. 1994. Diabo Coxo: o primeiro jornal ilustrado de São Paulo. Leitura, 13: 149. Publicação Cultural da Imprensa Oficial do Estado S.S. – IMESP. Lima, Herman. 1963. História da Caricatura no Brasil. 4 volumes. Rio de Janeiro: José Olympio. Sodré, Nelson Werneck. 1966. A História da Imprensa no Brasil. Rio de Janeiro: Civilização Brasileira.

93

THE “ARRESTED EVOLUTION” NOTION, THEORIES, MYTH?

EKATERINA VELMEZOVA Lausanne University (Switzerland) Russian Academy of Sciences Whenever the notion of arrested evolution is mentioned in linguistic works, it is never clearly defined. That is why one should better speak of a simple word-combination than of a precise notion, or of a term. The majority of these works are those by the Russian and Soviet linguist N.Ya.Marr (1864-1934), his colleagues and pupils. Marr, the creator of the “new language theory” (novoe uceË nie ob jazyke), has always been very well-known in the Soviet Union, especially in his life-time. Between 1920 and 1950, his linguistic theory (also known as “marrism”, a common noun derivative from the proper name of its author) was considered as the “official doctrine” in Soviet linguistics and provoked all kinds of reactions, from disdain and mockery to recognition and respect, except indifference. As one of his pupils said, the name of Marr “was if not the most famous, at least the most sensational and scandalous (šumnoe) in the history of linguistics” (Abaev 1960: 90). Yet in 1950, during the public discussion in the Pravda newspaper, it was disapproved by J.Stalin who had decided to participate personally in the linguistic life of the Soviet Union. Subsequently, the immoderate praises for the benefit of Marr were quickly replaced with anathema and numerous charges, particularly with those accusing Marr’s theories of lack of any scientific value. Let us note, however, that Marr has never had either accomplished or noncontradictory theories; he was always changing something, up to the key arguments, in his doctrines, to the horror of his students and pupils (Zvegincev 1990: 9; Rafail 1930: 161). Nevertheless here is a brief summary of his theories, presented by I.MešcaË ninov (1883-1967), one of Marr’s most loyal pupils: 1. all languages present various formulations in the process of the human language development; 2. in their evolution, all languages are going through a number of stages, one after another in a determined succession; 3. these stages are determined by socio-economic features of the corresponding societies, resulting in a particular world vision;

94

EKATERINAVELMEZOVA

4. so constituted, all languages could be characterized with particular features and thus classified as specific systems; 5. the typical features of both stages and systems are unstable. They are constantly changing, as a result of which the whole equilibrium is lost, provoking the languages’ passage from an inferior stage to a superior one (MešcËaninov 1929:138). The majority of works by Marr and his pupils were written in Russia and in the Soviet Union in the first third of the XXth century and were evolutionist, par excellence. That is, they presented the language development as an evolution through several (four, more precisely) stages replacing one another successively. When a new stage is reached, the former one does not disappear completely, but continues to exist without any changes, i.e. it stops in its development – that is why the linguists in the Soviet Union used the expression of arrested evolution. In particular, in the early 1930s Soviet linguists distinguished three stages of language development which could be classified as stopped in their evolution. They spoke of: 1) the first stage languages (Chinese, a number of African languages); 2) the second stage languages (Turkish, Mongolian, the Finno-Ugrian languages); 3) the third stage languages, called “Japhetic” at that time (the Caucasian and Semitic languages, as well as the languages of unknown origin, as Basque). 4) as to the fourth language stage (i.e. the languages which continued to evolve), it included mainly the Indo-European languages (Marr 1928: 405; see also CË ikobava 1985: 16). According to the marrists, each of these stages corresponded to one particular type of social relations which, in their turn, reflected the level of development of the means of production (Marr 1924b: 4). This thesis could be compared with the marxist idea on the interdependency in the development of basis and super-structure. In particular, Marr supposed that the Indo-European languages presented just a specific form of the result of language hybridization (Marr 1924b). In fact, another idea by Marr was that languages developed from a primary multitude towards unity: he turned upside-down the comparativists’ traditional pyramid which represented the opposite direction of language evolution, that is, from unity – the ancestor language, parent language or protolanguage – to variety (see for instance Marr 1926). As to the Indo-European languages, in accordance with Marr they owed their origin to the new forms of social relations, arisen thanks to the discoveries of metallurgy and the use of metals in the production of various kinds of tools (Marr 1924b: 4). One can see in this marrist conception of language evolution by stages a certain number of points of intersection with the linguistic ideas of the Schlegel brothers and of W.von Humboldt. It allowed some historians of linguistics to consider the works by

THE “ARRESTED EVOLUTION”

95

Marr and his pupils as nothing but a weak and faint echo of the German linguists of the XIXth century (Alpatov 1991: 121; Alpatov 1999: 113-115). Is it really so? Even if one particular work or one particular theory seems to resemble certain old ideas, de facto it will always reflect, in one way or another, the epoch of its appearance. That is why the resemblance of this kind, which can be striking at first sight, is very superficial, while a closer analysis of the corresponding theories leads to the discovery of their specifity. In its turn, the latter is very often a mirror of the corresponding epoch, of its dominating paradigms or of its air du temps. In the 1930s, the problem of the speed of language evolution (or, as in our specific case, its zero speed) was not only discussed by as “exotic” linguists as Marr and his colleagues. In his article “Mysli ob indoevropejskoj probleme” (“Thoughts on the IndoEuropean Problem”), written in 1936, Nikolaj Trubetskoy compares the evolutionary speed of two families of languages, the Indo-European and the Altaic languages. According to him, during the so-called prehistoric age, the Indo-European languages, unlike, in particular, the Altaic ones, had failed to pass beyond their inflexional grammatical structure. At the same time, they had been striving for an agglutinative structure, at which the same Altaic languages had finally arrived. That is why Trubetskoy supposes that at present the Indo-European languages continue to develop in the same direction and to change much more quickly than the agglutinative languages, i.e. languages with a relatively stable organization (Trubeckoj 1987: 59). As we can see, in the 1930s the problem of the language evolution speed was discussed in the works of such antagonists as Marr and Trubetskoy. Let us recall that while commenting upon Marr’s theories in his letter to Roman Jakobson, Trubetskoy wrote that they were more closely related to psychiatry than to linguistics (Trubetskoy 1975:74-75). As to Jakobson, answering this letter he called Marr’s ideas “sheer nonsense of a paranoia-patient” (Trubetskoy 1975: 75). And yet the same problem was raised in the works of these two linguists-antagonists. But for Marr the IndoEuropean languages presented the top of language evolution, while for Trubetskoy their inflexional structure was a striking illustration of their backwardness, compared with the development of agglutinative languages. The Schlegel brothers were the first to put forward, in the early XIXth century, a distinction between amorphous languages (called isolating, at present) and agglutinative and inflexional languages. They were inclined to interpret this difference in the grammatical structures of languages as reflecting various stages in the development of human thought. On the other hand, von Humboldt distinguished four stages in language evolution, among which the first three ones corresponded to the isolating, agglutinative and inflexional grammatical structures. According to him, this difference in the structure of languages was related to the stages in the evolution of human thought. Thus the resemblance of Marr’s theories with those of the German linguists is rather small and it seems that nothing repeats itself in the history of ideas.

96

EKATERINAVELMEZOVA

For the German linguists of the XIXth century, the boundaries between the different stages in the evolution of language were exterior to language. Taken at one particular moment, every language could be defined as belonging to one – and only one – certain stage. This idea counted partially for the theories of Russian and Soviet linguists. It was the arrested development of Chinese which they discussed most often. In particular, Esperantist Andrej Andreev said that the Chinese language had stopped in its development before Jesus Christ, because the social life in China had been organized according to the Confucius principles, i.e. rather strict and rigid doctrines resisting to the idea of evolution (Andreev 1930: 36). This theory presupposes that the language follows the development (or, in our case, rather the absence of any development) of the corresponding society. On the other hand, these ties are reciprocal, and if the peculiarities of social life determine the nature of the corresponding language, the opposite is also right. It allowed Marr to affirm: as Chinese had stopped in its development, the Chinese people “stay far behind in the general world evolution” (Marr 1927: 135). Yet if the economic situation in China at that time permitted him to put forward such ideas, other languages and peoples could sometimes cause him problems. For instance, up to the end of his life he could not explain why in Japan, a country with highly developed industrial technologies, an agglutinative language was still spoken, a language which, according to him, had no longer developed for several centuries. At the same time, Marr and his pupils could also draw boundary-lines between the different stages of language evolution inside the languages, unlike the German linguists of the XIXth century, for whom these boundaries were always exterior, with respect to languages. Marr argued that in every language, at every stage of its evolution, one could discern at several levels certain elements of its previous stages, the “strata of different stages”. On the phonetic level, in particular, it was the possibility to trace all the words in all modern languages to the “four primary elements” which, as Marr supposed, had replaced the sign language of gestures and constituted the origin of sound language (sal, yon, ber, roš), originally representing the language of the ruling caste of magicians. “The penetration into the very middle of linguistic stages is practically impossible without the “element analysis”, said one of Marr’s pupils in 1935 (BašindzaË gjan 1935). All words in all languages, according to Marr, are derived from these elements or from their combinations. For instance, it was the roš element which gave birth to such words as krasnyj ‘red’ and russkij ‘Russian’ in Russian, rouge and rot ‘red’ in French in German, correspondingly (Marr 1924a: 115-116). Yet Marr supposed that phonetic traces of ancient language stages could be most easily discovered in the proper names of tribes (“There are no words which are not derived from the names of tribes” [Marr 1926: 207]), which had originally been the gods’ names and become geographic terms. It means that serving as striking illustrations of the previous stages in language evolution, the proper names change much slower that the common nouns.


97

As to the syntactic level, impersonal constructions in modern languages could be considered as traces of the past, of the stopped language evolution. In the early 1930s, Marr analyses the verbal system in Georgian (Marr 1932). In particular, he treats the “impersonal verbs” in the Georgian language as descendants of former personal (from the semantic point of view) constructions. A totem, a genius served as a semantic subject in the corresponding propositions. According to Marr, in the pre-historic age the “personal sense” did not lack in the constructions corresponding, for instance, to the French propositions Il fait froid (‘It is cold’), Il fait chaud (‘It is hot’): it was “a general or collective person”, that of a totem, which had not been differentiated yet (Marr 1932:316). Thus the French construction J’ai froid (‘I am cold’) in Georgian means literally ‘Il me fait froid’ (‘It makes me cold’), where ‘il’ (‘it’) is “a totem, or an implicated genius” (Marr 1932:315). In the same spirit, Marr discusses certain constructions called impersonal in his article of 1931 “Jazyk i myšlenie” (“Language and Thought”) (Marr 1931). He says that in the propositions Il fait chaud or Es ist warm (‘It is hot’) the ‘God of the Sun’ is meant as a semantic subject; in the propositions Il pleut or Es regnet (‘It rains’) it is the question of the ‘God of the Rain’, while in the Russian construction Menja lixoradit (‘I am in a fever’) – that of the ‘God of the Illness’ (Marr 1931:91-92). Though he does not refer to the notion of animism, highly appreciated by the representatives of the British anthropological school (E.Tylor, J.Frazer) at the end of the XIXth century, de facto he discusses it. The theory of animism presupposed that primitive people saw “spirits” and “genii” in all phenomena of nature. So Marr remains true to his convictions and analyses even facts of modern languages from a pre-historic point of view. In time, Marr supposes, there is a semantic evolution: the former subject of action (for instance, the ‘cold’ in the case of the French structure Il fait froid) becomes an object of action, while the former object (‘me’) takes the role of subject (‘I’). So instead of ‘Le froid (il) me fait froid’ (‘The cold [it] makes me cold’) we get ‘J’ai froid’ (‘I am [‘I have’] cold’), at the semantic level. At the same time, the whole verbal construction (probably because of the conservative nature of language and the arrested evolution principle) does not change. In accordance with Marr, the impersonal verbs in modern languages are descendants of the “verbs of action”, while the impersonal structures could be considered as archaisms, “vestiges”, the “traces of the past” in the languages – among many other archaisms, typical of every modern language (Marr 1931: 91-92). After Marr’s death, this style of thinking was taken over by his pupil I.I.MešcËaninov. In particular, comparing the French construction Il fait froid with the proposition having the same sense in the Nemepu language (one of the North American Indians’ languages) Hi-yawtsana, MešcËaninov considers the French structure a “vestige” which had once been a norm for the whole language – in the “pre-historic” age, when the ‘cold’ (in this

98

EKATERINAVELMEZOVA

case) was still thought about as a subject (MešcËaninov 1941). At the same time, MešcËaninov supposed the Nemepu system more archaic than that of French, as, according to him, in Nemepu this verbal construction differs neither in its form nor in its sense from all other structures. It allows him to ascertain that there are no abstract impersonal propositions in this language. Though speaking of the “cold” in the function of subject MešcËaninov does not use the term of “totem” (highly appreciated by Marr), he utilizes the ideas and conceptions of his teacher. The notion of arrested evolution in linguistics had its analogies in the social sciences and in biology, where it concerned the stopped development of an individual. Let us remember the theory of monstrosity. It was in the XIXth century that the scientific explanations of monstrosity were elaborated and the teratology was born at the crossing of comparative anatomy and embryology. Meckel le Jeune explains certain types of monstrosity with stops of development, and Geoffroy Saint-Hilaire replaces the notion of “late” with that of “stopped”, “arrested”. According to him, monstrosity is nothing but a stop of the development of a particular organ at the stage which is passed beyond by the others. That is a survival of a transitional embryonic form. For an organism of one certain species, a today’s monstrosity was a normal state in the past. And if we take a series of species in order to compare them, it could happen that the monstrous forms of certain of them would be quite normal for others (see Canguilhem 1992: 179). The arrested evolution has also some analogies in the social sciences – that is the conception of societies’ late development in the works of Engels. Let us recall an extract from the Family’s Origins, where he says that various societies go through different stages in their social and economic development at different speeds, which is determined, first of all, by their geographic position and the available minerals. For instance, Engels explains the arrested development of the American autochtons with the absence of potentially domestic animals in the Western hemisphere. That is why, according to Engels, the American Indians were far behind the Greeks (Engels 1948: 117). Conclusions The arrested evolution expression appears frequently in the works of the Russian and Soviet linguists in the first third of the XXth century. Despite its certain resemblance with the theories of German linguists of the XIXth century (von Humboldt and the Schlegel brothers), it is the problem of boundaries which remains crucial and differently interpreted in the two cases. If for the German linguists the boundaries between various stages in the development of languages are exterior to the languages, the Soviet linguists saw them inside the languages. We should not forget that the problem of boundaries was very often discussed in Russia and in the Soviet Union during the first third of the XXth century. Boundaries “disappeared” between the languages (this tradition comes from the notion of Sprachmischung by H.Schuchardt [1922]), as well as between the literary


99

genres: literary critic A.N.Veselovskij spoke of the absence of boundaries between the poetic genres in ancient societies (Veselovskij 1939), while M.M. Bakhtine thought that certain works of Dostoevskij could not be described within the framework of the traditional repartition of genres (Bakhtine 1963). The phenomenon of the disappearance of boundaries has touched even biology: proposed in the early 1920s by the Soviet biologist L. Berg, the model of “convergence evolution” presupposed a possibility to mix various vegetal species. On the other hand, Marr’s ideas on the arrested evolution of languages had certain points of intersection with the ideas of the French anthropological school with respect to boundaries between various types of human thought. In particular, according to L.LévyBruhl, logic thought does not completely replace the original forms of thinking (pre-logic thought), which can be proved by a great number of prejudices and superstitions existing even in modern societies: pre-logic does not mean anti-logic (Lévy-Bruhl 1910: 455). This time, it concerned the illustrations of the arrested evolution of human thought. In general, our analysis of the expression of arrested evolution in the works of Russian and Soviet linguists allows us to see that very often, linguistic theories have more in common with the ideas appearing in other disciplines at the same epoch, than with the linguistic conceptions of previous epochs, despite their resemblance, seeming striking at first sight. And if the influence of the dominating ideology on language has never been proved, its influence on the linguistic theories seems quite evident.

REFERENCES

Abaev, V.I. 1960. “N.Ya.Marr (1864-1934). K 25-letiju so dnja smerti”: [N.Ya.Marr (1864-1934). To the 25th Anniversary of his Death]. Voprosy jazykoznanija, N1. Alpatov, V.M. 1991. Istorija odnogo mifa. Marr i marrizm. [The History of One Myth. Marr and Marrism]. Moskow: Nauka. . 1999. “Dissidenty indoevropeizma”: [The Dissidents of IndoEuropeanism]. Alpatov, V.M. Istorija lingvisticËeskix ucËenij. Moskow: Jazyki ruskoj kul’tury. Andreev, A.P. 1930. Jazyk i myšlenie. Opyt issledovanija na base materialisticËeskoj jafeticËeskoj teorii. [Language and Thought. Investigation Based on the Materialist Jafetic Theory]. Moscow: CK SESR, Internacional’naja 39-ja tipografija “Mospoligraf”. Bakhtine, M.M.1963. Problemy poetiki Dostoevskogo. [The Problems of Dostoevsky’s Poetics]. Moscow: Sovetskij pisatel’. BašindzËagjan, L.G. 1935. “ZacËem nuznË y cËetyre elementa?”: [Why do We Need the Four Elements?]. Akademija nauk SSSR akademiku N.Ya.Marru. MoskowLeningrad. Canguilhem, G. 1992. “La monstruosité et le monstrueux”. Canguilhem, G. La connaissance de la vie. Paris: Vrin. CË ikobava, A.S. 1985. “Kogda i kak eto bylo?”: [When and How it was?]. YezeË godnik iberijsko-kavkazskogo jazykoznanija, XII.

100

EKATERINAVELMEZOVA

Engels, F. 1948. L’origine de la famille, de la propriété privée et de l’état. Paris: Alfred Costes. Lévy-Bruhl, L. 1910. Les fonctions mentales dans les sociétés inférieures. Paris: Félix Alcan. Marr, N.Ya. 1924a. “Iz jafeticËeskix pereziË tkov v russkom jazyke”: [On Japhetic Vestiges in the Russian Language]. Marr, N.Ya. Izbrannye raboty, I-V. Vol.V. 1935. Moscow-Leningrad: Gosudarstvennoe social’no-ekonomicËeskoe izdatel’stvo. . 1924b.“Ob jafeticËeskoj teorii”: [On the Japhetic Theory]. Marr, N.Ya. Izbrannye raboty, I-V. Vol. III. 1934. Moscow-Leningrad: Gosudarstvennoe social’no-ekonomicËeskoe izdatel’stvo. . 1926. “O proisxozdË enii jazyka”: [On the Language Origins]. Marr, N.Ya. Izbrannye raboty, I-V. Vol. III. 1934. Moscow-Leningrad: Gosudarstvennoe social’no-ekonomicËeskoe izdatel’stvo. . 1927. “Jazyk”: [The Language]. Marr, N.Ya. Izbrannye raboty, I-V. Vol.II. 1936. Moscow-Leningrad: Gosudarstvennoe social’no-ekonomicËeskoe izdatel’stvo. . 1928. “PocËemu tak trudno stat’ lingvistom-teoretikom”: [Why is it So Difficult to Become Linguist-Teoretician]. Marr, N.Ya. Izbrannye raboty, I-V. Vol.II. 1936. Moscow-Leningrad: Gosudarstvennoe social’no-ekonomicËeskoe izdatel’stvo. . 1931. “Jazyk i myšlenie”: [Language and Thought]. Marr, N.Ya. Izbrannye raboty, I-V. Vol. III. 1934. Moscow-Leningrad. Gosudarstvennoe social’noekonomieskoe izdatel’stvo. . 1932. “BezlicËnye, nedostatocËnye, sušcËestvitel’nye i vspomogatel’nye glagoly”: [Impersonal, Insufficient, Substantive and Auxiliary Verbs]. Marr, N.Ya. Izbrannye raboty, I-V. Vol.II. 1936. Moscow-Leningrad: Gosudarstvennoe social’no-ekonomicËeskoe izdatel’stvo. MešcËaninov, I.I. 1929. Vvedenie v jafetidologiju. [Introduction to the Japhetidology]. Leningrad: Priboj. . 1941. “GlottogonicËeskij process i problema stadial’nosti” [The “Glottogonic” Process and the Problem of Stages]. Izvestija Akademii nauk SSSR. Otdelenie literatury i jazyka. 1941, N3. Rafail, M. 1930. “Marksizm i jafeticËeskaja teorija”: [The marxism and the Japhetic Theory]. Novyj Vostok, 28. Moscow. Schuchardt, H. 1922. “Sprachmischung”: Hugo Schuchardt-Brevier. Ein Vademekum der allgemeinen Sprachwissenschaft. Als Festgabe zum 80. Geburstag des Meisters zusammengestellt und eingeleitet (Leo Spitzer, ed.), Halle: Max Niemeyer. Trubeckoj [Trubetskoy], N.S. 1987. “Mysli ob indoevropejskoj probleme”: [Thoughts on the Indo-European Problem]. Trubeckoj, N.S. Izbrannye trudy po filologii. Moscow: Progress. Trubetskoy, N.S. 1975. Letters and Notes. The Hague: Mouton. Veselovskij, A.N. 1939. Izbrannye stat’i. [Selected Articles]. Leningrad: XudozËestvennaja literatura. Zvegincev, V. A. 1990. “CË to proisxodit v sovetskoj nauke o jazyke?”: [What is Happening in the Soviet Linguistic Science?]. Jazyk i social’noe poznanie. Moscow: Proizvodstvenno-izdatel’skij kombinat VINITI.

101

ON THE DEFENSE OF VON KEMPELEN AS THE PREDECESSOR OF EXPERIMENTAL PHONETICS AND SPEECH SYNTHESIS RESEARCH1 PLÍNIO A. BARBOSA Universidade Estadual de Campinas 1.

Introduction

Several accounts of von Kempelen’s machine, book and legacy are available in the literature, such as those by Ondrejovi (1996), Pompino-Marschall (1991), Dudley & Tarnoczy (1950), as well as the introduction by Brekle & Wildgen (1970) for the facsimile edition of the referred book. This paper recalls the nature of von Kempelen’s contributions and examines the use he made of his sources to build the speaking machine, without forgetting the twenty years of trials before arriving at the final version. By insisting on the role of modeling for acquiring knowledge, we claim that it is indeed the XVIII century – and not the XVII, as Fagyal (2001) sustains – that led to the invention of the (automatic) speaking machine, which determines this century as the “legitimate birthdate of text-to-speech synthesis” (Fagyal 2001:290). 2. A XVIIITH Century model of speech – sound production Wolfgang von Kempelen (Kempelen Farkas in Hungarian) was a lawyer, an engineer, an automata builder and an artist in his time, who built some devices which made the lives of Eastern Europeans easier – one of them was a press machine for Maria Theresia Paradis, a blind musician (Pompino-Marschal 1991; Ondrejovi 1996). In 1769 he starts building his machine, which he will finish only in 1791, even though he had presented previous versions during a European tournée between 1783 and 1785 (see PompinoMarschall 1991 for contemporary reports). The machine and the steps for building it are detailed in the fifth part of Kempelen’s book, Mechanismus der menschlichen Sprache nebst Beschreibung einer sprechenden Maschine where, after an introduction on the nature of language, he continues by giving a 456-page account of language origin 1. This work was partially financed by a research grant (n° 350382/98-0) from CNPq. It is integrated to the Fapesp project number 01/00136-2: “Integrating Continuity and Discreteness in Modeling Phonic and Lexical Knowledge”.

102

PLÍNIO A. BARBOSA

in chapter two, language functioning (including several aspects related to speech production and perception) in chapter three, and pronunciation and spelling of European languages in chapter four. The publication of the 195 copies of his book was considered an important event in his time, with the presence of important personalities (PompinoMarschal 1991; Ondrejovi 1996; Dudley & Tarnoczy 1950).

Figure 1: A picture of von Kempelen’s machine (in the Deutsches Museum of Munich), reproduced from the site with Hartmut Traunmüller’s permission.

In his book, Kempelen synthesizes in a masterly way what is needed to build a model of the speech apparatus: “Zu einer Sprechmaschine braucht man also weiter nichts, dacht ich, als eine Lunge, eine Stimmritze, und einen Mund” (Kempelen 1791:398). He recognizes what today is identified by the three subsystems of speech production: respiratory (for which the lungs are the crucial organ), laryngeal (the vocal folds, which define the glottis, are the main structures), and supra-laryngeal (the mouth and the nasal tract are the relevant units). By the manually-controlled operation of its structures, Von Kempelen’s machine simulates the final stage of the speech production mechanism: the articulation for producing sound. This operation can be described as follows (refer to figure 1). By the cadenced action of the right elbow, the (trained) operator ensures that the bellows (attached to the right end) produce a series of puffs of air. The fingers of the right hand control the levers to yield plosive sounds. Two fingers can also close or leave open the holes visible in Figure 1 to respectively produce oral or nasal sounds. By these manual actions, the machine is able to produce several fricative, plosive and nasal sounds (circa twenty sounds). The vowels were not so intelligible, but their contrast is obtained by shaping, with the left hand, the resonator at the left end of Figure 1 (Dudley & Tarnoczy 1950; Tillmann 1994). 3. Speaking machines’ witnesses and legacy The twenty years spent in building the machine were rewarded by a spectacular technical achievement: the machine in Figure 1 is still operational (Traunmüller 2000

ON THE DEFENSE OF VON KEMPELEN

103

witnesses that he operated it in 1997, and that he was surprised by its good conditions and by the feminine – almost childish – voice of the machine). Another enthusiastic account can be found in the Journal de Sçavans, 1783 edition (p. 629-630), which reports: Mais on voit chez lui [Kempelen] une autre machine qui n’a pas moins de mérite: c’est une machine qui parle & qui articule assez distinctement: Maman, aimez moi, allons à Paris, &c. Nous avons rendu compte du Mémoire qui a remporté le Prix de l’Académie de Pétersbourg en 1780, par M. Kratzenstein, sur la manière d’exprimer les sons des voyelles par des tuyaux d’orgues: mais on n’éttoit pas encore parvenue à imiter l’articulation des consonnes, & cette enterprise de M. de Kempelen annonce un talent également singulier; il est à désirer qu’il publie bientôt les moyens (apud Pompino-Marschall 1991:199)

The von Kempelen speaking machine continued to cause sensation for a century and a half. In the XIX century, the reproductions by the British physicist Sir Charles Wheatstone and by the Swiss mathematician Joseph Faber (Dudley & Tarnoczy 1950, Traunmüller 2000) attracted the attention of several scholars. Faber’s machine, Euphonia, for instance, finished in 1835 and demonstrated in London in 1846, could be operated with a keyboard. A pedal allowed the control of the “laryngeal” tone (see Figure 2) (Flanagan 1972).

Figure 2: Euphonia, reproduced from the http://mambo.ucsc.edu/psl/smus/smus.html with permission of D. Massaro.

Du Moncel reports on the Euphonia On s’est étonné que la machine parlante qui nous est venue, il y a quelques années d’Amérique, et qui a été exihibée au Grand-Hôtel fût d’une extrême complication, alors que le phonographe résolvait le problème d’une manière si simple: c’est que l’une de ces machines ne faisait que reproduire la parole, tandis que l’autre l’émettait, et l’inventeur de cette dernière machine avait dû, dans son mécanisme, mettre à contribution tous les

104

PLÍNIO A. BARBOSA

organes, qui dans notre organisme, concourent à la production de la parole (Moncel 1880 apud Köster 1973).

In the above passage, two methods for reproducing (speech) sound can be identified. The first, realized by the phonograph, records the acoustic signal without reference to articulation. The second, realized by Faber’s machine (and von Kempelen’s), simulates the movement of the speech organs. The two modern methods of doing Speech Synthesis, acoustic (concatenative and parametric) and articulatory, can be recognized in this report. In the XX century, the first electric speech synthesizers were (acoustic and articulatory) models of the speech sound production (see Dudley & Tarnoczy 1950 and Flanagan 1972 for details). 4. Kempelen’s sources and legacy of the mechanismus Fagyal (2001) defends the idea by which the knowledge of musical sound production that can be found in the third volume of Sorel’s La Science universelle (1667 [1641]) or in Mersenne’s Harmonie Universelle is consistent enough to consider both authors as predecessors of the speaking machines, and justify that the XVII century which “led to the invention of the speaking machine”. If the XVII - century episteme was so well established and the simulation of sound in pipes was universally known, why did a skilled engineer such as von Kempelen spend twenty years to build his machine? To shed some light on this matter, it is important to examine his sources. Two of his most important sources for the building of the machine, as well as for becoming acquainted with speech physiology were the works by the French physiologist Denis Dodart (1703) and by Albrecht von Haller (1766), apud Brekle & Wildgen (1970:XIII), both published in the XVIII century. These works can be considered starting points for the building of his machine, since the spectacular performance of the machine can only be attributed to Kempelen’s skills as an automata builder (and as a skilled operator) as well as to a long-standing phase of trials with previous demonstrated versions (see section 3). A detailed description of the machine’s building is given in the fifth chapter of Mechanismus, together with ten figures depicting the parts of the machine, which certainly were fundamental for the subsequent reproductions mentioned above. The previous versions of the machine were responsible for its final achievement, which is clear evidence that theory and modeling were interrelated during the whole process. This trivial relation justifies the fact that his machine was considered unique in his time, without predecessors (Brekle & Wildgen 1970:XVI). Mechanismus, on the other hand, benefited from Kempelen’s experience with the machine. Its importance as a theoretical book was sustained by Ernst Brücke, who considered it one of the best books on physiology and “an der freilich später

ON THE DEFENSE OF VON KEMPELEN

105

mancherlei ergänzt und bisweilen auch gebessert worden ist, die aber so fest begründet war, dass sie den sicherten Unterbau für allen ferneren Forschungen gegeben hat und geben wird.”2 (Brücke 1876:7 apud Brekle & Wildgen 1970:XXI). His book is also considered to have launched the very foundations for a theory of vowel acoustics (Ungeheuer, 1962: 4 apud Brekle & Wildgen 1970:XXI). Also Robert Willis benefited from reading it, as can be attested from his own testimony: “Willis gibt an zwei Stellen seines für die Entwicklung der Phonetik wichtigen Beitrages zu, von Beobachten von Kempelens ausgegangen zu sein.”3 (Brekle & Wildgen 1970:XX). These arguments are solid enough to consider von Kempelen and the XVIII century as the bases for the later speaking machines, and for Experimental Phonetics research. REFERENCES

Brekle, Herbert E. & Wildgen, Wolfgang. 1970. Introduction to Kempelen’s. Mechanismus der menschlichen Sprache: VII-XLV. Stuttgart: Frommann Verlag. Brücke, Ernst. 1876. Grundzüge der Physiologie und Systematik der Sprachlaute für Linguisten und Taubstummenlehrer. Vienna. Dodart, Denis. 1703. “Mémoires sur les causes de la voix de l’homme et de ses différents tons”. Histoire de l’Académie royale des Sciences. Paris: Année 1700. Dudley, Homer & Tarnoczy, T. H. 1950. “The Speaking Machine of Wolfgang von Kempelen 151-166”. J. Acoust. Soc. Am. 22 (2). Fagyal, Zsuzsanna. 2001. “Phonetics and Speaking machines: on the mechanical simulation of human speech in the 17th century”. Historiographia Linguistica. XXVIII (3): 289-330. Flanagan, James L. 1972. “Voices of Men and Machines”. J. Acoust. Soc. Am. 51, 1375-1387. Haller,Albrecht von. 1766. “Elementa physiologiae corporis humani”. Lausanne, 1757-1766. Bd. III: Respiratio, vox. Kempelen, Wolfgang von. 1970. [1791]. Mechanismus der menschlichen Sprache nebst der Beschreibung seiner sprechenden Maschine. Facsimile of the Vienna edition with an introduction by Herbert E. Brekle and Wolfgang Wildgen. Stuttgart: Frommann Verlag. Ondrejovi, Slavomír. 1996. Wolfgang von Kempelen and his “Mechanism of Human Speech”. Available: www.slovakradio.sk/kultura/expstudio/kempe.html. Accessed on June 4th, 2001. Pompino-Marschall, Bernd.1991. “Wolfgang von Kempelen und seine Sprechmaschine”. Forschungsberichte des Instituts für Phonetik und Sprachliche Kommunikation der Universität München. 29, 181-252.

2. Even though later on the book was certainly completed and, from time to time improved, it was so well - founded that it launched the very foundations of later research. 3. Willis confesses in two moments the important contributions of Kempelen’s observations for the development of Phonetics.

106

PLÍNIO A. BARBOSA

Tillmann, H. G. 1994. “Phonetics, early modern, especially instrumental and experimental work”. Asher, R. E., Simpson, J. M. Y. (Eds.) The Encyclopedia of Language and Linguistics, v. 6, 3082-3095. Oxford: Pergamon Press. Traunmüller, Hartmut. 2000. Wolfgang von Kempelen’s and the the subsequent speaking machines. Available: www.ling.su.se/staff/hartmut/kemplne.htm. Accessed on June 4th. 2001. Ungeheuer, Gerold. 1962. Elemente einer akustischen Theorie der Vokalartikulation. Berlin: Springer Verlag.

107

POSITIVISM AND NEO-POSITIVISM IN LINGUISTICS AND LANGUAGE PHILOSOPHY PAUL LAURENDEAU York University Some in their discourse desire rather commendation of wit, in being able to hold all arguments, than of judgement, in discerning what is true; as if it were a praise to know what might be said, and not what should be thought. Bacon, Francis, “Of Discourse”, in The Essays.

1. Positivism, the philosophical fetishization of Science, emerges out of Science, but is not Science For the purpose of the present topic, Linguistics is the description of languages and dialects in a non-normative perspective, whereas Language Philosophy is a specific philosophical framework characterized by the feature of defining language as a determining category of knowledge and/or existence. In that perspective, it is considered that the core of current Language Philosophy is a linguisticist deviation of onto/gnoseological Philosophy. We then assert that such a stand in Language Philosophy is inspired by positivist views, since that stand is a tendency to import the objects and patterns of application of a certain discipline into another. In its fundamental sense, Positivism is an attitude claiming explicitly that the methods and procedures of Natural Sciences are integrally valid in Social Sciences. Fascinated by the power and prestige of Natural Sciences (also called Positive Sciences), Positivism spares them from their own specific epistemology, and becomes a mere scienticism in other spheres, under the form of methodological generalization. In Knowledge and human interest, Habermas explains in what way Positivism is a non-epistemological scienticism: Positivism certainly still expresses a philosophical position with regard to science, for the scientistic self-understanding of the sciences that it articulates does not coincide with science itself. But by making a dogma of the sciences’ belief in themselves, positivism assumes the prohibitive function of protecting scientific inquiry from epistemological selfreflection. Positivism is philosophical only insofar as is necessary for the immunization of the sciences against Philosophy. For methodology by itself does not suffice, it must also prove itself as epistemology or, better, as its legitimate and reliable executor. Positivism stands and falls with the principle of scienti[ci]sm, that is that the meaning of knowledge is defined by what the sciences do and can thus be adequately explicated through the

108

PAUL LAURENDEAU

methodological analysis of scientific procedures. Any epistemology that transcends the framework of methodology as such now succumbs to the same sentence of extravagance and meaninglessness that it once passed on metaphysics. (Habermas 1972: 67)

These developments reached an almost caricatural level in the fetishization of Natural Sciences by Social Sciences. For primitive Positivists (like for instance, Saint Simon, see Marcuse 1996: 332), society was to be treated like nature. Positivism in Sociology lead to a doctrine of society as a “branch of physics”. One can also quote the biologism of Behaviourist Psychology as an example of that phenomenon of crude triumphalist Positivism. In a second phase of development, when Natural Sciences required an increase of the mathematization of their methods and procedures, traditional Positivism (exemplified, say, by the rigid and unilateral promotion of the experimental method) started to lose focus, as mathematization and formalisation became the new fetishes. The new speculative model of Neo-positivism compensated that sharp decline of Positivism as an empirical source of inspiration. Neo-positivism is fundamentally a logic and linguistic theory of knowledge. There is a clear continuity between the general frame of mind of Positivism and the one of Neo-positivism. In fact, when their internal evolution forced Natural Sciences to mobilize the procedures of mathematization and formal abstraction, the positivist sensitivity had to pursue on its path of the imitation of the framework of Natural Sciences within Social Sciences and Philosophy. The hypertrophy of logicism and mathematization, the sacralization of their scientific virtues, in Social Sciences and Philosophy, is nothing other than that second phase of Positivism, namely Neo-positivism. The blatant lack of empirical grounding for Neo-positivism generated the intellectual context that allowed Linguisticism to kick in, in Philosophy and even in certain Natural Sciences. The will to produce a theoretical discipline joined with the undermining of “metaphysics” causes a general escape from all empirical and mental categories. Linguistic categories (and their numerous look-alikes: the “formal” categories, the “syntactic” models, the “languages”) are the only remaining tools for the non-metaphysical theory aimed at. These tools will be thoroughly used in the general framework of Neo-positivism. Let us begin with the most general principles of the neo-positivist school, the so-called Vienna Circle, which are intended to provide us the foundations of an “anti-metaphysical” logic. These principles, as expounded by W.H. Werkmeister, may be summarized as follows. 1. Knowledge is knowledge only in its form; in any cognition, only the form is important, while all the rest is inessential (Moritz Schlick). 2. A proposition has meaning only to the extent that it can be verified (Schlick), and to verify a proposition means only to find out whether or not it accords with the rules established to govern the connections of that proposition in a given language. 3. Knowledge is always empirical, based on that which is given directly (Schlick); moreover, the sense-data of sensation, which lie at the foundation of the edifice erected by this school (the fundamental architecture of which was devised by Ernst Mach), are afforded by protocol propositions – primary, or elementary, propositions that are not debatable.

POSITIVISM AND NEO-POSITIVISM IN LINGUISTICS

109

4. The logical analysis of language demonstrates that all metaphysical propositions are pseudo-propositions wholly devoid of meaning (Carnap). 5. All fields of research are based on a single science: physics (Neurath, Carnap); hence the doctrine of physicalism. 6. The propositions of logic are tautologies (Wittgenstein). Hence: ‘In logic, process and result are equivalent. (Hence the absence of surprise.)’ And: ‘Proof in logic is merely a mechanical expedient to facilitate the recognition of tautologies in complicated cases.’ [Quoted from Wittgenstein Tractacus logico-philosophicus, 6.1261 and 6.1262 – P.L.] 7. Mathematics is a logical method (Wittgenstein); all mathematical concepts can be derived from the fundamental concepts of logic (Carnap). The logic embodied in these principles has been theorized especially by Bertrand Russell, Alfred Tarski, and Carnap, who have given it perhaps its most extensive development. (Della Volpe 1980: 247-248, see also Phillips 1979: 55-64)

As Positivism before it, Neo-positivism spreads, and leads to the formalization and mathematicization of Social Sciences “from outside”. The promotion and generalization of statistics in Social Sciences was the first step of that new development. Fanned by the prestige of Mathematics, Logic and Formal Physics, the influence of Neo-positivism completely reshuffled the epistemology of the Social Sciences. That new “epistemological shift” favoured the emergence of one of the Social Sciences most susceptible to profit from a linguisticist treatment of reality: Linguistics itself, the cherished child of structuralism, the so-called pilot-science of the 1960’s. But, the “shift” that brought Linguistics to the forefront at the peak of its momentum was actually far more superficial than it may seem. It is not certain that the human sciences have really changed their ‘nature’ by changing their name and their methods. The relations that are currently being established between the literary disciplines are proof of that: the systematic mathematicization of a number of disciplines (economics, psychology, sociology); and the ‘application’ of disciplines manifestly more advanced in scientificity to others (the pioneering role of mathematical logic and especially Linguistics, the equally intrusive role of psycho-analysis, etc). Contrary to what has occurred in the natural sciences, in which relations are generally organic, this kind of ‘application’ remains external, instrumental, technical and therefore suspect. The most aberrant contemporary example of the external application of a ‘method’ (which in its ‘universality’ is following fashion) to any object whatsoever is ‘structuralism’. When disciplines are in search of a universal ‘method’, we may wager that they are a little too anxious to demonstrate their scientific credential really to have earned them. True sciences never need to let the world know that they found the key to becoming sciences. (Althusser 1990: 96)

More than rationalization or systematization, what Positivism and Neo-positivism provided to Linguistics (and other Social Sciences) was rather a new set of selflegitimating rhetorical reflexes, which could be summarized in the following motto: Mathematics does it, so let’s do it too. Saussure – to quote one example among many – argues that the shift from diachronic Linguistics to synchronic Linguistics, on which the whole development of Structuralism is based, is grounded, as follows:

110

PAUL LAURENDEAU

An absolute state is defined by lack of change. But since languages are always changing, however minimally, studying a linguistic state amounts in practice to ignoring unimportant changes. Mathematicians do likewise when they ignore very small fractions for certain purposes, such as logarithmic calculation. (Saussure 1983: 100)

Nothing in the internal logic of the object studied is required to legitimate the methodological options displayed here. Its foundations are automatically considered solid and reliable; the minute similarities between it and a Positive Science are spelled out. With Saussure, the 20th century scienticist scholasticism is born in Linguistics. Let us now meet two of its children: Bloomfield and Chomsky. Both have several points in common. One of them is that they are both linguisticist theoricians. As previously mentioned, you are a linguisticist (or a glottocentrist) when you claim that language is a key factor in knowledge and/or existence. You usually tend to do so when your discipline and your inquiry within that discipline attempt at the same time to be non-empirical and non-speculative. The importance of praxis and of generalization is suddenly replaced by the importance of... language. Bloomfield: Language plays a very important part in science. A typical act of science might consist of the following steps: observation, report of observation, statement of hypothesis, calculation, prediction, testing of predictions by further observation. All but the first and last of these are acts of speech. Moreover, the accumulation of scientific results (the “body” of science) consists of records of speech utterance, such as tables of observed data, a repertoire of predictions, and formulas for convenient calculation. The use of language in science is specialized and peculiar. In a brief speech the scientist manages to say things which in ordinary language would require a vast amount of talk. His hearers respond with great accuracy and uniformity. The range and exactitude of scientific prediction exceed any cleverness of everyday life: the scientist’s use of language is strangely effective and powerful. Along with systematic observation, it is this peculiar use of language which distinguishes science from non-scientific behavior.[...] In scientific procedure we mean by deduction the purely verbal part of an act of science which leads from the report of observation and the hypotheses to a prediction. If we replace the report of observation by arbitrarily invented postulates, the discourse makes no pretense to validity in the sphere of handling actions. Deductive discourse of this kind is produced in logic, mathematics, and the methodology of science. It is made to fit some type of observational data, or else it exists for its own sake, in readiness for the emergence of observational data to which it may be applied. Until modern times, Euclidian geometry was viewed as an “a priori” system: the underlying everyday observations about the spatial character of object were viewed as inborn and unquestionable truths. Today the same system, apart from the correction of flaws, is treated as purely verbal discourse of deduction from postulates. It is especially useful because these postulates, by virtue of their historic origin, are such as to make the discourse applicable to the placing of objects, as this placing is observed in the first approximation that is customary


111

in everyday handling. We have learned, however, that astronomical magnitudes make sensible the error in these postulates, and they accordingly demand a different discourse, based upon other postulates which, in turn, will be chosen so as to fit the new observations. (Bloomfield 1965: 1-12)

It seems impossible to abandon a conceptual generalization without becoming linguisticist. And of course, it is even easier to be a linguisticist “scientist” if you are a scienticist linguist. Then, the old Humboldtian idea of the determination of thought by language surfaces easily in the views of such a scienticist linguist. Bloomfield: I shall not presume to enter here upon the epistemology [sic] problems in which linguistic considerations must play a part. Far more of our experience than one generally assumes is shaped by the linguistic habits in which we live. The apparatus of logic, more especially, depends upon the language we speak: the logical forms, in other words, must develop historically with the language. Not only our more abstract concepts, but also those of qualities and actions are due to linguistic forms, or rather, are the subjective phase of linguistic forms, which have been evolved in the course of time. Much of our philosophy, in consequence, moves captive in the plane of its author’s language, which it should, for freedom, transcend, – as it can only through the study of language. (Bloomfield 1983: 324)

Now on the linguisticist stand, Chomsky is, at first sight, in total opposition to Bloomfield’s views: [According to Humboldt:] Speech is an instrument of thought and self-expression. It plays an “immanent” and “constitutive” role in determining the nature of man’s cognitive processes [...]. Although languages have universal properties, attributable to human mentality as such, nevertheless each language provides a “thought world” and a point of view of a unique sort. In attributing such a role in the determination of mental processes to individual languages, Humboldt departs radically from the framework of Cartesian Linguistics, of course, and adopts a point of view that is more typically romantic. Humboldt does remain within the Cartesian framework, however, in so far as he regards language primarily as a means of thought and self-expression rather than as an animal-like functional communication system... (Chomsky 1966: 21)

But is that opposition as fundamental as commonly believed? Humboldt remains a major inspiration to Chomsky. And most of all, the linguisticist dimension of Neopositivism culminates in the fascination for formal language-like models as supreme descriptive and explicative devices. Hence, with Chomsky this conception reached summits previously unattained in a discipline like Linguistics. We propose that the Positivist and Neo-Positivist inspiration is what Bloomfield and Chomsky have in common. We even suggest that the passage from Bloomfield to Chomsky is nothing other than the passage from Positivism to Neo-positivism in the specific development of Linguistics.

112

PAUL LAURENDEAU

2. Bloomfield: observations, trial and error, and the hypertrophy of the empirical datum Positivism is the philosophical stand of the ones who believe that they can rid themselves of philosophical issues. In the direct line of that so-called anti-metaphysical stand of Positivism, Leonard Bloomfield (1887-1949) claims, in broad epistemological terms, that Mentalism is non-scientific: It is the belief of the present writer that the scientific description of the universe, whatever this description may be worth, requires none of the mentalistic terms, because the gaps which these terms are intended to bridge exist only so long as language is left out of account. If language is taken into account, then we can distinguish science from other phases of human activity by agreeing that science shall deal only with events that are accessible in their time and place to any and all observers (strict behaviorism) or only with events that are placed in co-ordinates of time and space (mechanism), or that science shall employ only such initial statements and predictions as lead to definite handling operations (operationalism), or only terms such as are derivable by rigid definition from a set of everyday terms concerning physical happenings (physicalism). These several formulations independently reached by different scientists, all lead to the same delimitation, and this delimitation does not restrict the subject matter of science but rather characterizes its method. It is clear even now, with science still in a very elementary stage, that, under the method thus characterized, science can account in its own way for human behavior – provided, always, that language be considered as a factor and not replaced by the extra-scientific terms of mentalism. (Bloomfield 1965: 13)

That anti-mentalist option is clearly in conformity with the positivist program. Seen the old-fashion way, a “Science” is actions, observations and manipulations leading to empirical statements. The second point of Bloomfield’s axiomatic is that Semantics is nothing other than a mere manifestation of subjectivist mentalism. Bloomfield is in harmony with the views of the character, which he designates here as: the mechanist: Adherents of mentalistic psychology believe that they can avoid the difficulty of defining meanings, because they believe that, prior to the utterance of a linguistic form, there occurs within the speaker a non-physical process, a thought, concept, image, feeling, act of will, or the like, and that the hearer, likewise, upon receiving the sound-waves, goes through an equivalent or correlated mental process. The mentalist, therefore, can define the meaning of a linguistic form as the characteristic mental event which occurs in every speaker and hearer in connection with the utterance or hearing of the linguistic form. The speaker who utters the word apple has had a mental image of an apple, and this word evokes a similar image in a hearer’s mind. For the mentalist, language is the expression of ideas, feelings, or volitions. The mechanist does not accept this solution. He believes that mental images, feelings, and the like are merely popular terms for various bodily movements...[...] Although this difference of opinion plays a decisive part in our views about the fundamentals of language, as of other human activities, and although mentalists lean heavily upon their


113

terminology in all discussion of meaning, the dispute has really very little to do with problems of linguistic meaning. The events which the mentalist designates as mental processes and the mechanist classifies otherwise, affect in every case only one person: every one of us responds to them when they occur within him, but has no way of responding to them when they occur in anyone else. The mental processes or internal bodily processes of other people are known to each one of us only from speech-utterances and other observable action. (Bloomfield 1966: 142-143)

The logical conclusion of all that mechanics is then unavoidable: the grasp of the semantic facet of language is (so far, and possibly forever) external to its scientific description. The solution to the problem of meaning can come only from other Sciences: The meanings of speech-forms could be scientifically defined only if all branches of science, including, especially, psychology and physiology, were close to perfection. Until that time, phonology and, with it, all the semantic phase of language study, rests upon an assumption, the fundamental assumption of Linguistics: we must assume that in every speech-community some utterances are alike in form and meaning.[...] “The statement of meanings is therefore the weak point in language-study, and will remain so until human knowledge advances very far beyond its present state. In practice, we define the meaning of a linguistic form, whenever we can, in terms of some other science. Where this is impossible, we resort to makeshift devices. One is demonstration. (Bloomfield 1966: 78, 140-141)

Clearly, for Bloomfield, Semantics is the “metaphysics” of language. It is subjective, non-empirical, uncertain, “demonstrative”, and fluctuating. A definition in terms of some other science is unavoidable to address it. Therefore, meaning cannot be considered as part of the linguistic datum. This datum is to be investigated in its formal combinations and distributions, in an activity of trial and error blind to the semantic dimension. So the program of Linguistics is then easy to state: like the good “Science” we want to be, let’s stick closely to our empirical datum, our “concrete entities”: the linguistic forms! A workable system of signals, such as language, can contain only a small number of signaling-units, but the things signaled about -in our case, the entire content of the practical world- may be infinitely varied. Accordingly, the signals (linguistic forms, with morphemes as the smallest signals) consist of different combinations of the signaling-units (phonemes), and each such combination is arbitrarily assigned to some feature of the practical world (sememe). The signals can be analyzed, but not the thing signaled about. (Bloomfield 1966: 162)

And there you have it: primitive Positivism in Descriptive Linguistics sticking to the materiality of language to spare itself from the dangers of speculative distortions. Better to destroy the internal logic of the object and turn its study to some absurd set of

114

PAUL LAURENDEAU

manipulations than admit for it a specificity that would deviate from the path of Positive Science. Such was the Bloomfield of 1933, the hegemonic one. Yet that was not the Bloomfield of 1914 (in An Introduction to the Study of Language). Both had in common a joint venture with psychology (actually two opposed schools of psychology from one version of his fundamental treatise to the other) but the epistemological dimension of that specific situation was exaggerated. The real motivation of Bloomfield’s option in Descriptive Linguistics was the positivistic eagerness to be... scientific. Only the champion of Neo-positivism in Descriptive Linguistics will manage to bump the Bloomfieldian hegemony away. 3. Chomsky: formalization as “abstraction”, and the fascination for “explanation” Indeed, Chomsky claims to bring back Cartesianism and Humboldtian mentalism. He is the anti-Bloomfield, and poses as such. The fact that he is rather a mentalist than a behaviourist makes him the most outspoken opponent to primitive Positivism in Linguistics: Behavioral science has been much preoccupied with data and organization of data, and it has even seen itself as a kind of technology of control of behavior. Anti-mentalism in Linguistics and in philosophy of language conforms to this shift of orientation. As I mentioned in my first lecture, I think that one major indirect contribution of modern structural Linguistics results from its success in making explicit the assumption of an anti-mentalistic, thoroughly operational and behaviorist approach to the phenomena of language. By extending this approach to its natural limits, it laid the groundwork for a fairly conclusive demonstration of the inadequacy of any such approach to the problems of mind. More generally, I think that the long range significance of the study of language lies in the fact that in this study it is possible to give a relatively sharp and clear formulation of some of the central questions of psychology and to bring a mass of evidence to bear on them. What is more, the study of language is, for the moment, unique in the combination it affords of richness of data and susceptibility to sharp formulation of basic issues. (Chomsky 1972: 65-66)

This is clear. But despite that very explicit anti-behaviourist stand, the semantic facet of language is (still) external to its scientific description, according to Chomsky. Consequently, Chomsky and Bloomfield ended up working together to build a solid tradition, which gives its specificity to American Structuralism: the marginalization of Semantics. Chomsky: Linguistic theory has two major subdivisions, syntax and semantics. Syntax is the study of linguistic form. Its fundamental notion is “grammatical”, and its primary concern is to


115

determine the grammatical sentences of any given language and to bring to light their underlying formal structure. The goal of syntactic study is to show that the complexity of natural language, which appears superficially to be so formidable, can be analyzed into simple components; that is, that this complexity is the result of repeated application of principles of sentence construction that are in themselves quite simple. Semantics, on the other hand, is concerned with the meaning and reference of linguistic expressions. It is thus the study of how this instrument, whose formal structure and potentialities of expression are the subject of syntactic investigation, is actually put to use in a speech community. Syntax and semantics are distinct fields of investigation. How much each draws from the other is not known, or at least has never been clearly stated. The subject of investigation in the following pages will be syntactic structure, and we shall study it as an independent aspect of linguistic theory. In part, our decision to place no reliance on meaning in systematic developments is motivated by a feeling that the theory of meaning fails to meet certain minimum requirements of objectivity and operational verifiability. We need not enter into this question, however, since a much more important motivation is that semantic notions, if taken seriously, appear to assist in no way in the solution of the problems that we will be investigating. (Chomsky 1975a: 57)

With Syntax as an independent aspect of linguistic theory, and despite the strong opposition formulated against Bloomfield by Chomsky, Semantics does not come back. Why does Chomsky entertain such an inconsistency with his own fundamental neomentalist views? Why is he suddenly so Bloomfieldian, with all these requirements of objectivity and operational verifiability? The answer is that the rejection of Semantics is not consistent with the rest of Chomsky’s stand, neither as a Language Philosopher nor as an epistemologist of Linguistics. The rejection of Semantics is not an internal feature of Chomsky’s vision. It is actually held from outside, as the result of the compulsive imitation of a bundle of Neo-Positivist logicist formalism, emanating from several science-like theoricians, among which the logician Carnap is a main figure. Through these external influences on Chomsky, the heart of the problems ends up being simply that Mathematics have a highly sophisticated syntax, but do not happen to have some Semantics similar to the one specific to Natural Languages. To imitate Mathematics, (and later Computing) being the cardinal priority, the features of Natural Languages that are most mathematic-like are given an automatic “epistemological” priority. The contradiction of Chomsky with himself becomes more obvious when one is given the opportunity to observe that, despite a clear rejection of Semantics in the generative program, the promotion of so-called idealization is present, solidly grounded in the model of Science. Opposition to idealization is simply objection to rationality; it amounts to nothing more than an insistence that we shall not have meaningful intellectual work. Phenomena that are complicated enough to be worth studying generally involve the interaction of several systems. Therefore you must abstract some object of study, you must eliminate those factors which are not pertinent. At least if you want to conduct an investigation which is

116

PAUL LAURENDEAU

not trivial. In the natural sciences this isn’t even discussed, it is self-evident. In the human sciences, people continue to question it. That is unfortunate. When you work within some idealization, perhaps you overlook something which is terribly important. That is a contingency of rational inquiry that has always been understood. One must not be too worried about it. One has to face this problem and try to deal with it, to accommodate oneself to it. It is inevitable. There are no simple criteria that provide the correct idealization, unless it is the criterion of obtaining meaningful results. If you obtain good results, then you have reason to believe that you are not far from a good idealization. If you obtain better results by changing your point of view, then you have improved your idealization. There is a constant interaction between the definition of the domain of research and the discovery of significant principles. To reject idealization is puerile. (Chomsky 1998:57-58)

Clearly opposed to primitive Positivism, the promotion of idealization as the only “rationalist” solution is to be associated with the idea of model, as well as with the will of promoting Explanation over Description in Linguistics. With that “next task” of Linguistics -Explanation- around the corner, the Chomskyan epistemological myth of the progress of the Linguistic Science kicks in. The next task is to explain why the facts are the way they are, facts of the sort we have reviewed, for example. This task of explanation leads to inquiry into the language faculty. A theory of the language faculty is sometimes called universal grammar, adapting a traditional term to a research program somewhat differently conceived. Universal grammar attempts to formulate the principles that enter into the operation of the language faculty. The grammar of a particular language is an account of the state of the language faculty after it has been presented with data of experience; universal grammar is an account of the initial state of the language faculty before any experience. It would include, for example, the principle that rules are structure dependent, that a pronoun must be free in its domain, that there is a subjectobject asymmetry, some of the principles mentioned in the preceding lecture, and so on. Universal grammar provides a genuine explanation of observed phenomena. From its principles we can deduce that the phenomena must be of a certain character, not some different character, given the initial data that the language faculty used to achieved its current state.” (Chomsky 1988: 61-62)

With the formal notion of the UG, the goal is now to constitute a “genuine explanatory theory”... as they did in Physics! Chomsky’s stand is simply that Bloomfield’s Linguistics was empiricist whereas his is scientific... Recall the logic of Descartes’ [sic] argument for the existence of a second substance, res cogitans. Having defined “body” in terms of contact mechanics, he argued that certain phenomena lie beyond its domain, so that some new principle was required; given his metaphysics, a second substance must be postulated. The logic is essentially sound; it is, in fact, much like Newton’s, when he demonstrated the inadequacy of Cartesian contact mechanics for the explanation of the motion of the heavenly bodies so that a new principle, the principle of gravitational attraction, had to be postulated. The crucial difference


117

between the Cartesian and the Newtonian enterprises was that the latter offered a genuine explanatory theory of the behavior of bodies, whereas the Cartesian theory offered no satisfactory account of properties such the creative aspect of language use [sic] that lie beyond mechanical explanation in Descartes’ view. Therefore Newton’s conceptions came to be the “scientific” common sense of later generations of scientists, while Descartes’ fell by the wayside. (Chomsky 1988: 146-147)

Chomsky says nothing other than: the same way my Linguistics is similar to Newton’s Physics, my abstraction is similar to scientific abstraction... and the language structure is a “physical structure”. In what sense is language a physical structure? We do not know for certain, but we believe that there are physical structures of the brain which are the basis for the computations and the representations that we describe in an abstract way. This relationship between unknown physical mechanisms and abstract properties is very common in the history of science. So, for example, in the nineteenth century, chemists constructed abstract diagrams that were supposed to represent a complex molecule with carbon and hydrogen and oxygen attached in some fashion. But that’s a completely abstract representation. For example, the chemist couldn’t say what the particular parts of the diagram referred to in the physical world. In fact, it wasn’t clear whether there were things corresponding to the parts of the diagram. Even now that we know better what carbon is, we recognize that it is something abstract. So, you can’t hit carbon. In fact, it’s a very abstract concept. But the point is that the chemist’s descriptions were part of an explanatory theory. They were part of a theory from which you could predict what would happen if you sent an electric current through some physical object, for example. Now those theories of the chemist are similar to a linguist’s theory of the computations of the brain. In each case the abstract theories pose a further question for the physical scientist. The question is, [sic] find the physical mechanisms that have these properties. In the early part of the twentieth century, physicists began to discover the physical entities that had the properties that had been described by the chemists. In fact, until the early part of the twentieth century, many scientists weren’t convinced that there were even such things as molecules. The thought this was just an abstract idea, an abstract computational idea. In the early part of the twentieth century, evidence accumulated showing that there really are things that have these properties. Now physics could not have developed the structure of the atom and the molecule if nineteenth-century chemistry hadn’t provided the abstract theories. That’s what told the physicists what they should look for. They had to look for things which had the very complicated properties described in the abstract theories. And the brain sciences are in the same state today. They have to ask the linguist or the psychologist what are the abstract structures that human possess for which we have to search for the physical basis.” (Chomsky 1988: 185-186) Let us suppose that we discover a domain of intelligence where human beings excel. If someone has developed a rich explanatory theory in spite of the limitations of available evidence, it is legitimate to ask what the general procedure is that has permitted this

118

PAUL LAURENDEAU

move from experience to knowledge – what is the system of constraints that has made possible such an intellectual leap. The history of science might provide some relevant examples. At certain times, rich scientific theories have been constructed on the basis of limited data, theories that were intelligible to others, consisting of propositions linked in some manner to the nature of human intelligence. Given such cases, we might try to discover the initial constraints that characterizes these theories. (Chomsky 1998: 64-65)

Loud and heavy references to Positive Sciences are made to build credibility to what Chomsky envisions as being a linguistic theory. He is in all shape and form a theorician. That fetishization of theory is a crucial element in Chomsky’s conception of Formal Linguistics. And during the hegemonic period of Generative Grammar, this ideological influence got heavier and heavier on the rest of Descriptive Linguistics. In time, Chomsky has more and more to deny playing Neo-Positivistic “games”, and to explain that all this is... theory! There has been some discussion recently as to whether the linguist “plays mathematical games” or “desc[r]ibes reality” in linguistic analysis of particular languages, where the phrase “playing mathematical games” often appears to refer to the conscious development of a theory of linguistic structure for use in constructing and validating grammars. If by “describing reality” is meant meeting the external conditions of adequacy, then in order to give content and significance to the requirement that the linguist must describe reality, it is necessary to give independent (i.e. outside the particular grammar) characterizations of these conditions e.g., for sentencehood, by constructing informant response tests to determine the degree of acceptability or revocability of sequences. But within whatever bounds can be clearly set independently, the linguist’s goal can only be to construct for each language a simple grammar related to other grammars in such a way as to lead to a revealing general theory of which all are exemplifications. There seems to be no reason to consider the constructs established in pursuit of these goals as being in some sense invalid. If the methods developed with these goals in mind lead to unacceptable results, it is important to show this. But the alternative to ineffective methods is not abandonment of theoretical inquiry. (Chomsky 1975a: 81-82)

Neo-Positivist statements of that nature are the foundations of Chomsky’s spontaneous epistemology. It is obviously more important to Linguistics to “be a Science” and pose as such than to simply do its work, whatever this may be. After almost forty years of this nonsense, can Linguistics simply go back to work before completely collapsing under the weight of its positivist scholasticism? 4. A “logic” for Language Sciences? Let’s not argue about it. Let’s go study it... Captain James T. Kirk, USS Enterprise, A piece of the action, teleplay by David P. Harmon and Gene L. Coon, Paramount, 1968.


119

Social Sciences are seeing more and more clearly that Logical Positivism and its gadgets don’t work. Galvano Della Volpe: The major difficulty in arriving at a proper comprehension and complete critical assessment of the modern formal (or rather, formalized) logic championed by logical positivism lies in its dual character as theory of thought and of language. Once the theory of thought is examined, and its inability to serve as a valid logic, philosophical or otherwise, is demonstrated (for, as we shall see, it leaves the problem of scientific law unresolved), there remains the theory of language, particularly of semiotics, associated most closely with the name of Rudolph Carnap. Excessively abstract and partial, this theory, with its peculiar obsession with ‘correct’ language, or the language of ‘truth’ turns a merely technical language (or a mathematical type) into a dogma and thus fails as a general, truly philosophical theory of semiotics (or semantics). (Della Volpe 1980: 247)

Then what are the other options in our disciplines? First, we have to look into the specificity of our object, into what makes it so different from the object of Positive Science. Bloomfield, in one of his few non-positivist moments, gives us some hint: Language has been developed in the interchange of messages, and every individual who has learned to use language has learned it through such interchange. The individual’s language, consequently, is not his creation, but consists of habits adopted in his impressive intercourse with other members of the community. The result of this is the individual’s inability to use language except in the form in which the community as a whole uses it: he must speak as the others do, or he will not be understood. As a matter of fact, he does not, in normal cases, try to speak otherwise, but unquestioningly follows his and his fellow-speakers’ habits. The change which occurs in language is thus never a conscious alteration by individuals, but an unconscious, gradual change in the habits of the entire community. The motives which cause it are not individual reflective considerations of the result, but new associative tendencies or new conditions of innervations [sic] due to some change in the circumstances of life affecting the community. As we examine more closely the different aspects of language, we shall again and again find the same characteristics: as the individual speaker receives his habits from the community, individual motives do not come into play, but only causes affecting the community as a whole. (Bloomfield 1983: 17)

Language is at the same time social and mind-dependant. It cannot even begin to be sampled as carbon or viruses can be. Consequently, we have to completely review the method of approach of that object, even if that review leads to a quite drastic tabula rasa. Will we have the modesty to admit that we lost ourselves in the forest of external models? Well, Chomsky, in one of the rare non-scienticist moments – where he drops all the Neo-Positivistic flimflam and speaks his heart, has it: As for my own methods of investigation, I do not really have any. The only method of investigation is to look hard at a serious problem and try to get some ideas as to what might be the explanation for it, meanwhile keeping an open mind about all sorts of other possibilities.

120

PAUL LAURENDEAU

Well, that is not a method. It is just being reasonable, and so far as I know, that is the only way to deal with any problem, whether it is a problem in your work as a quantum physicist or whatever. There are certain fields like psychology where people do carry out extensive study of methods of investigation. There are other fields like physics where you do not study methods of investigation. So at MIT the physics department does not have a course in experimental methods, but many psychology departments spend a lot of time on what they call methodology. Well, there is a lesson there, but I won’t draw it. (Chomsky 1988: 190)

Here Chomsky is closer to Althusser than anywhere else. And, we have to admit it. In our teaching as in our research, our hypertrophy of “methodology” proves more than anything else that we still have a lot to do in terms of genuine method. Physics is supposedly free minded about all this. Why aren’t we? Because, we imitate Science, instead of simply doing it. Let us not argue about what our object should be, or which method would be the best to borrow from the Sciences. Let us simply study our object and pull our method out of its own specific logic.

REFERENCES

Althusser, L. 1990. Philosophy and the Spontaneous Philosophy of the Scientists – and other essays. London & New York: Verso. Bloomfield, L. 1965. Linguistic Aspects of Science, International Encyclopedia of Unified Science, Volume 1, number 4. University of Chicago Press & University of Toronto Press. . 1966. Language. New York: Holt, Rinehart & Winston. . 1983. An Introduction to the Study of Language. Amsterdam/ Philadelphia: John Benjamins Publishing Company, Classics in PsychoLinguistics. Chomsky, N. 1966. Cartesian Linguistics – A Chapter in the history of Rationalist Thought. New York & London: Harper & Row Publishers, Studies in Language. . 1972. Language and Mind. New York: Harcourt Brace Jovanovich. . 1975a. The Logical Structure of Linguistic Theory. New York & London: Plenum Press. . 1975b. Reflections on Language. Pantheon Books. . 1986. Knowledge of Language – Its Nature, Origin, and Use. Praeger, Coll. Convergence. . 1988. Language and Problems of Knowledge. Cambridge Massachusetts: The MIT Press. . 1995. The Minimalist Program. Cambridge Massachusetts: The MIT Press. 1998. On Language. New York: New Press. Della Volpe, G. 1980. Logic as a Positive Science. London: NLB. Habermas, J. 1972. Knowledge and Human Interests. Boston: Beacon Press. . 1989. On the Logic of the Social Sciences. Cambridge: The MIT Press.


121

Marcuse, H. 1996. Reason and Revolution: Hegel and the Rise of Social Theory. Atlantic Highland, N.J.: Humanities Press International, Humanities Paperback Library. Phillips, D.L. 1979. Wittgenstein and Scientific Knowledge – A Sociological Perspective. London: The MacMillan Press. Saussure, F. de. 1983. Course in general Linguistics. London: Duckworth. Wittgenstein, L. 1981. Tractacus logico-philosophicus. London: Routledge & Kegan Paul.

122

123

FROM UNIVERSAL LANGUAGES TO INTERMEDIARY LANGUAGES IN MACHINE TRANSLATION THE WORK OF THE CAMBRIDGE LANGUAGE RESEARCH UNIT (1955-1970)

JACQUELINE LÉON Université Paris 7, CNRS Introduction The Cambridge Language Research Unit (CLRU), founded in 1955 to start experiments in Machine Translation (MT), gathered many different and remarkable personalities: Margaret Masterman (1910-1986), the director of the group and a Wittgenstein’s pupil; R.H. Richens (who died in 1984), a biologist specialist of plant genetics; and linguists such as Martin Kay and MAK Halliday; computer scientists, among them Yorick Wilks who became one of the first researchers on Natural Language Understanding. The originality of the CLRU is that it is the only MT group, besides the Russians, to develop a method of Machine Translation using intermediary language. The most striking aspect of their work is that the construction of this intermediary language stems directly from the 17th-century universal language schemes. 1) The universal language Nude, conceived by Richens, is widely inspired by Dalgarno’s Ars Signorum (1661) and Wilkins’ Essay (1668)1. 2) The CLRU also makes use of the Thesaurus published in 1852 by Roget, one of Wilkins’ continuators (Salmon 1979a). It may seem strange that the universal language issue was considered seriously by scientists two centuries later whereas these schemes, at their time, came to almost nothing. To enlighten this point one may assume that the issue of the feasibility of MT in the 1950s raised questions comparable with those raised by philosophers in the 17th century2. In my paper, I will address these issues by examining the various versions of MT methods using intermediary languages proposed by the British group in the 1950’s. I will try to explain how the achievement of a practical task, the automatization of translation, as well as the implementation of a specific conception of word meaning, modified the notion of universal language itself. 1. George Dalgarno (1626-1687); John Wilkins (1614-1672). 2. “Philosopher” also means “scientist” in the 17th century.

124 1.

JACQUELINE LÉON

Historical and intellectual context

Although the two periods present some analogies, it is not relevant here to compare the historical, economic and intellectual context of the apparition of universal languages in Great Britain in 17th century with the 1950s context of apparition of MT. However it is interesting to note that, in both cases, universal languages and intermediary language schemes were anchored in a strong social demand for interlingual means of communication3. In the 1950s, facing with the internationalization of science and the politico-military requirements of the cold war, multilingual communication technologies were greatly needed. Within a frame of unprecedented technological development as electronical computers, MT had to play a leading part in responding this demand. The main point here is to see if the issues raised by the scientists on the feasibility of MT in the 1950s can be compared with those raised by the authors of Universal language schemes. Weaver’s conceptions are very enlightening on this point4: The imperfection of languages was a recurrent issue for conceptors of universal languages. It is the same for ambiguity and polysemy in MT which is one of the most complicated problems to solve by machines. Besides, in Weaver’s viewpoint, the connection between translation and cryptography led “very naturally” to the idea that translation makes deep use of language invariants. Hence the translation between two languages cannot be done word by word but only by using a universal language “the real but as yet undiscovered universal language” (Weaver 1949:23); such a project requiring considerable work on the logical structure of languages. 2. Nude: from universal language to intermediary language Nude is the first project of intermediary language devised by the CLRU. Although it may look rough and based upon naive conceptions of meaning and translation, it is worthy of interest because it is the first time that semantic information and primitives were used in natural language processing5. Besides it raised interesting questions about language representation. Like all the members of the group, Richens shared the idea of pre-eminence of semantic analysis upon syntax. This idea was very original among MT pioneers who, for most of them, thought morphology and syntax were dominant in the process of MT6. 3. On the context of apparition of universal language schemes in 17th century in Britain, see Cram (1985) and Salmon (1979c, 1992). 4. Thanks to his Memorandum “Translation” which was widely distributed among scientists in 1949, Warren Weaver (1894-1978) promoted MT in Great Britain and in the USA. 5. There are also technical issues for the use of intermediary languages in MT. Because they constitute semantic representations common to every language, they require far less algorithms and dictionaries than transfert methods which necessitate two algorithms for each pair of source and target languages. 6. Note that syntax, excluding any other linguistic area, will be at the heart of Computational Linguistics, after the ALPAC disaster and the wreckage of MT in 1966.

FROM UNIVERSAL LANGUAGES TO INTERMEDIARY LANGUAGES

125

His experience in word by word translation led Richens to introduce semantic information to solve ambiguities. To use semantic information in procedures, Richens proposed to build an interlingua where structural distinctive features of source languages are suppressed. Interlingua is devised as a “semantic net of naked ideas”, hence its name Nude. For Richens, semantic networks is what remains invariant during the translation process. Nude is conceived as an algebraic language; it comprises about fifty elements, each of them denoting a basic (naked) idea, such as plurality, plant or negation, represented by a letter. Here are nineteen elements (out of fifty) which were used for the translation from Japanese to English of the sentence “the percentage of matured capsules and the number of grains of seeds of one capsule are different according to the time of hybridizing”: B c C f H I L M n N

becoming, change straight, plane causation, influence possibility, potentiality pertain in, inside living, alive much, more, great near, adjacent, together contact, adhere, attach

p P Q S T u x X z

plant plurality, group, number hard, firm same, equal time, period, duration elongate textile part, component negation, opposite, contrary

Nude has a syntax. A word is regarded as a relation with either 0 adjunction; or one adjunction [.]: for example an adjective or a transitive verb expect a noun as an adjunct; or 2-adjunctions [:] transitive verbs7. Apostrophes and quotes are used as brackets within a word. During the translation process, the source text is divided into chunks, the minimal sense units. The result of the translation of chunks into Nude is called a formula. Here is the formula of “one seed”: .Pz = one Xp’CL= seed Xp’CL.Pz = one seed

Richens had probably read Wilkins’ Essay. As a network of semantic primitives represented by letters, Richens’ interlingua is very close of a universal characteristics8. However Richens’ componential representation of word meaning is closer to Dalgarno’s than to Wilkins’. Remember that Wilkins’ Characteristics is devised as a 7. Richens’ syntax can be viewed as of prefiguration of case grammar: a transitive verb is marked to expect a subject and an object. 8. Moreover, just like Wilkins, he raises the issue of verbal particle compositionality (Cram 1994).

126

JACQUELINE LÉON

hierarchical classification of concepts based on Aristotelian categories9. Conversely, instead of trying to codify the contents of universe as Wilkins, Dalgarno’s purpose was to distinguish the different semantic components of each concept, to give a sign to each component and to form the name of objects and concepts by combining the signs of all their components10. 3. The epistemological status of intermediary language Richens’ interlingua was widely discussed by the Cambridge group. One of the problems raised by Nude was its lack of empirical foundation in natural languages. As Wittgenstein’s pupil, Margaret Masterman could not consider Nude primitives as universal concepts. Besides she was impervious to any cognitive hypothesis according to which primitives could be the elements of a language of thought – such as Fodor’s Mentalese created a few years later (1975). For Masterman an intermediary language could not be a universal language. She agreed with Wittgenstein that the logic unit for studying language should not be word or proposition but word context, namely word use. In Masterman (1954:209) she defined use and usage in the following way: “the Use of a word is its whole field of meaning, its total “spread”. Its usages, or main meanings in its most frequently found contexts, together make up its Use”. Because of its structure, based on the classification of words according to a set of contexts, Masterman chose thesaurus organization to create a new intermediary language, “a thesauric interlingua”. For Margaret Masterman (1959:34) “the fundamental hypothesis about human communication which lies behind thesaurus making is that, although the set of possible uses of words in a language is infinite, the number of primary extra-linguistic situations which we can distinguish sufficiently to talk to one another is finite. Given the complexity of the known universe it might be the case that we refer to a fresh extra-linguistic situation every time we create a new use of a word. In fact we do not; we pile up synonyms, to rerefer, from various and differing new aspects, to the stock of basic extralinguistic situations which we already have.” The consequences for MT are important. Communication and translation depend on the fact that two people and two cultures, however much they differ, can share a stock of extra-linguistic contexts. 9. In his Essay Part 4 “A Real Character and a Philosophical Language”, Wilkins gives the translation of the prayer the Our Father in fifty-one languages. In particular bread (p.454) is translated into his characteritics as saðba, where sa denotes the genus of Oeconomical provisions, b the first difference, and a the second species (bread); such a decomposition reflects Wilkins’ hierarchical classification of concepts. 10. This difference of opinions pulled the two philosophers apart for ever whereas they had worked together on a common scheme of universal language for a long time (Salmon 1979b).


127

This is how Masterman defined language universality. The idea of an intermediary language refers to a stock of extra-linguistic contexts, which can be represented by a thesaurus11. 4. Roget’s Thesaurus Roget’s Thesaurus, in spite of its drawbacks such as incoherence and non-systematicity, was chosen by the CLRU to build an interlingua combining a thesaurus with Nude. Peter Mark Roget (1779-1869) quotes Wilkins and is considered as one of his continuators. As Wilkins, he was a philosopher and the secretary of the Royal Society. What is common between Wilkins’ Essay and the Thesaurus is the classification of words based on concepts. However Roget took care not to build a universal language scheme. His purpose was essentially pedagogical as is indicated by the title of his book. Roget’s Thesaurus, taking Wilkins’ Essay as a model, is divided into two parts: a thematic thesaurus and an alphabetical index. The thematic part comprises six primitive classes (abstract relations, space, matter, intellect, volition, emotion) divided themselves into sections then into heads. Heads are followed by a list of words connected semantically. A word can appear in several lists under different heads or classes. To build an interlingua from Roget’s thesaurus, it is necessary to have a set of coherent heads. These (archi)heads will be provided by Nude primitives. As heads can belong to several archiheads, they are classified according to a multiple hierarchy, and not only a tree organization, as in Roget’s. To formalize such a thesaurus, the CRLU members chose the Lattice theory. 5. Thesauric Nude For Masterman, the thesauric interlingua is not an algebraic universal language where elements are represented by letters. Actually the archiheads are English words. As basic semantic categories “archiheads must be below the meaning-line”. They are not words which could exist in any language. But they must be sufficiently like words which can be handled in any language. Archehead TRUE! must be like true; or at least TRUE! must be more like true than it is like please. To Masterman the interlingua should be a genuine language, able to cope with problems of meaning such as metaphors. It is worth mentioning that at a certain point the CLRU 11. The CLRU was also influenced by the contextualists of the London school, namely by John Ruppert Firth (1890-1960). Firth only attended to the first meeting of the CLRU in 1955 but was not very interested in MT himself. However MAK Halliday, one of his most famous pupils, was an active member of the CRLU from 1955 untill the beginning of the 1960s.

128

JACQUELINE LÉON

members considered using Basic English or Esperanto as intermediary languages. Anyway what is at stake in new Nude is more the representation of natural language meaning than the universal representation of knowledge. Here are fourteen Nude primitives (out of fifty) (Masterman 1959:62) N UDE ELEMEN T

APRO XIMATIN G AREA O F MEAN IN G

EXAMPLE

1

BAN G !

Sudden action

Bang: think (idea)

2

DO N E

C ompleted action

(done: change): folk (banquet)

3

WILL

Deliberate Intention

For: (will:do) (try)

4

MUC H

A lot of

Have/ (much: (count: (part: where))) (long)

5

FO R

Motive, Because

For: (will:do) (try)

6

C AUSE

C ausative actions

C ause/ (have/sign) (say)

12

IN

Be situated In, or having the Property of Being able to C ontain Something

In: thing (container)

13

HAVE

Pertain "of"

C ause/ (nothave/life) (kill)

... 41

... SIGN

... Symbol (any sort)

... C ause/ (have/sign) (speak)

45

GRAIN

Pattern (artistc, thought)

Think: (stuff:grain) (chemistry)

46

HO W

Mode, quality, adjective

(think/same): how.

47

WHEN

Time

C ount: (part:when) (unit of time)

48

WHERE

Space

C hange/where (move)

NOT

C auses all N ude elements to mean their opposites.

00

Syntax is nearly the same as in Richens’ version. [: ] connects one element and its adjunct; [/] is a verbal connector between subjet and verb or verb and object. For example to speak is in Nude cause/(have/sign). Brackets replace apostrophes and quotes and unit primitives in pairs. As in Richens the process is recursive. speak he says cause / (have /sign)

speaker man/(cause / (have /sign)) man:(cause/ (have /sign))

The issues raised by thesauric Nude were meaning abstraction and category attribution. Thus the CLRU members had to find means to extract primitives from texts experimentally. From their point of view the only justification of meaning abstraction is


129

of pratical order. They do not believe in universal knowledge representation. What is taken from universal language tradition is the empirical tradition. Just as British universal language schemes were always anchored in technological developments and social demand, such as stenography cryptography, logarithms, printing characters, language planning, multilingual communication (Maat & Cram 2000), the CRLU aimed at devising MT and information processing systems. These practical options had great impact on the role of language formalization which was one of the main topics discussed by the group. The CRLU, boosted by MT objectives, implemented a conception of language formalization which was based on reflections upon context and meaning, independently and in competition with Bar-Hillel’s12 and Chomsky’s logico-mathematics hypotheses (Léon 2000). For the CLRU language must be considered as a whole, and mathematically formalizable only in a second step. Whereas for Bar-Hillel it is the opposite: language is considered as mathematically formalizable a priori; and it is the task for the researcher to discover how natural languages can be adapted to formalization. 6. Wilks and templates I will conclude this paper by mentioning the works of one of the youngest members of the CLRU, Yorik Wilks, who continued the work on Nude in the USA in the late 1960s within the very new domain of Natural Language Understanding. Wilks modified Nude in order to resolve semantic ambiguities in texts. He radicalized the CLRU conception of ambiguities which should be defined with reference to dictionaries, which is the common view of MT experimenters, but within a text. He was then led to develop what he called “preferential semantics”: for a given text, a specific meaning is chosen preferably over another, so that no definitive choice should be made. To solve ambiguities, he devised a semantic representation system for texts using “templates” which captures the “gist” of text message. Templates are pattern-matching formulas representing the meaning of a clause. These formulas are very close to Richens’ but instead of encoding the various meanings of a word, they encode the meaning representation of a clause. Here are some of the fifty-three primitives used by Wilkes to build formulas, of which forty-five are thesauric Nude’s archiheads: BE BEAST CAN

FORCE FROM GRAIN

MAN MAY MORE

12. The Essays on and in Machine Translation by the Cambridge Language Research Unit, where the thesaurus method was presented, were dedicated to Bar-Hillel in response to a first version (1959) of his critical record on MT (1960).

130

JACQUELINE LÉON CAUSE CHANGE COUNT

HAVE HOW IN

MUCH MUST ONE

Here is the formula representing the meaning of “colourless”: (COLOURLESS (((( ( (WHERE SPREAD)(SENSE SIGN) )NOT HAVE) KIND) (COLOURLESS AS NOT HAVING THE PROPERTY OF COLOUR))))

A formula is a pair, the first part is the head COLOURLESS, the second part is a new pair which represents the translation of the word into primitives (((WHERE SPREAD)(SENSE SIGN) )NOT HAVE) KIND) and its definition in natural language (COLOURLESS AS NOT HAVING THE PROPERTY OF COLOUR). The formula in that sense-pair can be explained as follows: ‘colourless’ is a sort; a sort indicating that something does not possess some property; the property is an abstract sensorial property of a certain sort; that certain sort has to do with spatial extension. Thus the meaning of the whole word-sense is ‘a sort that lacks an abstract, sensory, spatial property’, and it is not difficult to see that this is what (in right-left order) the formula conveys. (Wilks 1972:107)

Coded in LISP which was already the programming language of Artificial Intelligence, the formulas are recursive and dynamic. The first element of the list, the head, is a function, the rest of the list is an argument. Thus this meaning representation is also a computer procedure. Through pattern-matching procedure, word representations are compared with word representation in the text. If the same primitives are in the same clause, they help solving ambiguity by offering a preferred meaning. Wilks’ works had the merit of introducing semantic primitives, conceived within MT research, into the new field of non-referential semantics and artificial intelligence. He chose a far more radical semantic position than Katz and Fodor (1963) since his analysis units are not grammatically correct sentences but texts. His works within the CLRU allowed him to work on semantics preferentially while syntax was dominant. This research on word and text meaning was strongly anchored in the British tradition, which was pratically based. Besides the CLRU works fit in with the 20th-century British contextual tradition which was empirically based.

REFERENCES

Primary sources Bar-Hillel, Yehoshua. 1960. “The present Status of Automatic Translation of Languages”. Advances in Computers. vol.1, 91-141. F.C. Alt ed. N.Y., London: Academic Press.


131

Roget, Peter Mark. 1852. Thesaurus of English words and phrases classified and arranged so as to facilitate the expression of ideas and assist in literary composition, [2ème édition, 1853]. London: Longman. Firth, John Ruppert. 1957. Papers in linguistics 1934-1951. Oxford University Press. Masterman Margaret, 1954. “Words”. Proceedings of the Aristotelian Society: 209232. . 1959. “What is a thesaurus?” Essays on and in Machine Translation, by the Cambridge Language Research Unit. rapport non publié, ML90. Katz, Jerrold J. & Fodor, Jerry A. 1963. “The structure of a semantic theory”. Language vol. 39, n°63: 170-210. Richens, R.H. 1955. “A general programme for mechanical translation between any two languages via an algebraic interlingua” [archives du CLRU, ML5]. Weaver, Warren. [1949]. 1955, “Translation”. in Machine Translation of Languages, 14 essays, (W.N.Locke and A.D. Booth, eds.), MIT et John Wiley:15-23. Wilkins, John. 1668. An Essay towards a real character and a philosophical language. London: S. Gellibrand and J. Martin. Wilks, Yorick. 1968. “On line Semantic Analysis of English Texts”. Mechanical Translation, vol 11, n°3-4:59-72 . 1972. Grammar Meaning and the machine analysis of language. London: Routledge and Kegan Paul. Secondary bibliography

Cram, David. 1985. “Universal Language Scheme in 17th century Britain”. Histoire Epistémologie Langage vol 7-2: 35-44. . 1994. “Collection and Classification: Universal Language Schemes and the Development of Seventeenth –Century Lexicography”, Anglistentag 1993 Eichstätt Proceedings, 59-69. edited by Günther Blaicher and Brigitte Glaser. Tubingen: Max Niemeyer Verlag. Léon, Jacqueline. 2000. “Traduction automatique et formalisation du langage. les tentatives du Cambridge Language Research Unit (1955-1960).”: The History of Linguistics and Grammatical Praxis, 369-394. (eds. P.Desmet, L.Jooken, P.Schmitter, P.Swiggers) Louvain / Paris: Peeters. Maat, Jaap. & Cram, David. 2000. “Universal Language Schemes: the 17th Century”: History of the Language Sciences, An International Handbook on the Evolution of the study of Language from the Beginnings to the Present 1030-1042. Éd. by Sylvain Auroux. E. F.K. Koerner Hans-Josef Niederehe Kees Versteegh, Volume 1. Berlin, New York: Walter de Gruyter. Salmon, Vivian. 1979a. “John Wilkins’ Essay (1668): Critics and Continuators”. The study of language: 17th century England: 97-126. Amsterdam: John Benjamins. . 1979b. “The evolution of Dalgarno’s “Ars signorum’”.The study of language: 17th century England: 157-175. Amsterdam: John Benjamins. . 1979c. “Language-planning in 17th century England; its context and aims”, The study of language: 17th century England: 129-156. Amsterdam: John Benjamins. . 1992. “Caractéristiques et langues universelles”. Histoire des Idées Linguistiques t.II: 407-423. Liège: Mardaga.

132

133

THE PRESENCE OF ABSENCE IN SAUSSURE’S LINGUISTIC THEORY T. CRAIG CHRISTY University of North Alabama The category of absence figures explicitly in Saussure’s well-known distinction between associative and syntagmatic relations, and, consequently, necessarily intersects and informs the allied, and dependent, concepts of value and identity. Syntagmatic relations result from the ways in which linguistic units influence each other by dint of being co-present, in linear sequence, at any given instant. This co-presence is manifest in the varyingly complex syntagms store-housed in langue, and in the phrases and sentences of discourse. Associative relations, by contrast, result from the ways in which linguistic units influence each other without being co-present, in linear sequence, at any given instant. Associative relations hold units together only in the brain, as mnemonic groups, outside discourse, indeed outside any set syntagmatic structure. As recorded in the Cours, “Syntagmatic relations hold in praesentia. They hold between two or more terms co-present in a sequence. Associative relations, on the contrary, hold in absentia. They hold between terms constituting a mnemonic group” (Saussure 1916:122 [171]).1 The semantic value2 of any given term derives, in turn, from the indissoluble nexus of the syntagmatic and associative relations in which that term figures. Accordingly, the concept of identity – technically, synchronic identity – conventionally denominated in terms of presence, must, for the linguistic term, be co-denominated in terms of absence, since the identity of the linguistic term, at all levels of analysis, relates either to what is, or is not, materially manifest in that term, or to both. This is Saussure’s concept of functional identity, which he illustrates with his famous example of the ‘identity’ of the 8:45 Geneva-to-Paris train, a seeming sameness which obtains despite the absence of material trappings otherwise requisite to establish identity. The identity in value associated with phonetically distinct Latin genitive case endings is likewise adduced to demonstrate that identity can be grounded in associations of any kind, not just those with a material basis of comparison. “The mechanism of a language,” 1. Eleven passages from the Cours are cited from Roy Harris’s 1983 translation. The pagination of the French editions from 1922 onwards is given in brackets. 2. In a chapter dedicated to “La valeur sémantique”, Bouquet (1997:311-345) addresses complexities associated with this term.

134

T. CRAIG CHRISTY

we are told, “turns entirely on identities and differences. The latter are merely counterparts of the former” (1916:107 [151]). Thus sounds3, signifiers and signifieds are alike defined negatively, in terms of what they are not, in terms of difference and opposition. The resultant identity of the categories ‘identity’ and ‘difference’ underwrites the claims that “In the language itself, there are only differences”: and “In a sign, what matters more than any idea or sound associated with it is what other signs surround it” (Saussure 1916:118 [166]), that is, what is absent in the sign itself. This differential approach to identity, to definitions delimited in terms of what is not, what is absent, suffuses the Cours, and has been cited as attesting to Saussure’s “...condemnation of nomenclaturism” (Harris 1987:11; cf. also Saussure 1916:65 [97ff]).4 It would seem, however, that it attests more importantly to the centrality of absence within Saussure’s linguistic theory. The concluding remarks to the chapter somewhat paradoxically entitled “Concrete Entities of a Language” echo a sense of near frustration, if not vexation, at this state of affairs: Like chess “...a language has the character of a system based entirely on the contrasts between its concrete units. One cannot dispense with identifying them, nor move a step without having recourse to them. And yet delimiting them is such a tricky problem that one is led to ask whether they are really there” (Saussure 1916:105 [149]). The identity of a unit is, then, a function of all that it is not, of all that is outside the unit, together with which it comprises an equilibrium of contrasts. Its status within this equilibrium constitutes its value. In other words, its identity is its value. In the words of the Cours, “...in semiological systems, such as languages, where the elements keep one another in a state of equilibrium in accordance with fixed rules, the notions of identity and value merge” (Saussure 1916:109 [154]). Absence, it seems, has a large presence in Saussure’s linguistic theory, as laid out in the Cours. It is not, however, only in this exposition of general principles that we find Saussure focusing on the absent. On the contrary, a pronounced attentiveness to the category of absence in fact characterizes virtually all Saussure’s work, both prior to and concurrent with the teachings recorded in the Cours. Take, for example, his brilliant reconstruction, in 1878, of what, at the time, were utterly unattested Proto-IndoEuropean sounds – now referred to as ‘laryngeals’– a feat later duly hailed as a crowning victory of the comparative method (Saussure 1879). Here the focus was on materially absent sounds, the onetime presence of which was nevertheless registered in the form of extant phonetic irregularities affecting the quality and quantity of vowels. Reconstructing materially absent sounds was, of course, a commonplace and cornerstone of the comparative method: sounds absent in some or most of a set of related languages could, on the basis of their presence in one or 3. “Speech sounds are first and foremost entities which are contrastive, relative and negative” (Saussure 1916:117 [164]). 4. In Saussure’s conception, value, not meaning, is the heart of any semiological system. While meaning is dependent on value, value is dependent on absence..

ABSENCE IN SAUSSURE’S LINGUISTIC THEORY

135

more of those languages, be reconstructed for the proto-language. Saussure’s approach here represents a radical departure from that method precisely because the imputed sounds in question were in fact absent in all the descendant languages. Yet the discovery, in 1928, fifteen years after Saussure’s death, that the Hittite archives actually contained written signs for these otherwise unattested sounds confirmed both his approach and his results. With this reconstruction we have, I believe, a first clear sign that Saussure was committed to, if not obsessed with, detecting the codes running the language program in absentia. Saussure’s sustained involvement, between 1895 and 1898, with his Geneva colleague, Théodore Flournoy, in trying to make sense of the bizarre glossolalic productions of a medium, known under the pseudonym ‘Mlle Hélène Smith’, offers yet another intriguing case in point.5 What especially fascinated Saussure in this medium’s so-called ‘Sanskritoid’ pronouncements was that the sound f was conspicuously and consistently absent. This could, he reasoned, be taken as a sign that her garbled utterances were still somehow authentic, given that Sanskrit does in fact lack f, and that the probability would be extremely low that someone faking this language would commit precisely this seemingly authenticating omission. Nevertheless, the sheer fact that there was no evidence that Mlle Smith had ever been to India or otherwise been exposed to Sanskrit should have pointed Saussure to the alternate interpretation which Flournoy, and Saussure’s fellow Sanskritist, Victor Henry (1901), ultimately proved applicable: namely, that the absence of f was symbolically crucial lest the French language program running in the background of her Sanskritoid be unmasked, a conclusion Saussure himself approximated when he observed that “...the only thing that does not come to her mind is... to pronounce... in the French words which remain the theme or the substratum of what she is going to say; and the rule her mind obeys is that these familiar words must each be replaced by an exotic-sounding substitute.... it is essential only, and above all, that these sounds not resemble French...” (cited from Todorov 1982:260). This whole experience seems to have had several lasting influences on Saussure. It alerted him on the one hand to the limitations of linguistic science as a source for explaining complexities of human discourse phenomena, yet, on the other, it must have made him intensely aware of what he subsequently set out to capture in his linguistic theory, namely the thoroughgoing systematicity of language. After all, Mlle Smith’s seemingly automatic language reflected the workings of a system so structured on integrated relational values that what were homophones or figures of speech in French regularly surfaced in her Sanskritoid permutations, either as one-to-one ‘translations’ or as fantastically convoluted circumlocutions predicated on all manner of improbable linkages. But there were linkages, and these defined the ways in which encodings regularly unfolded, just as Saussure was later to explicate in the teachings of the Cours. And precisely because 5. Flournoy (1900) chronicles this curious case. See also Christy (1999) and (1999a) for a detailed discussion.

136

T. CRAIG CHRISTY

there were linkages, never mind how fantastical, the role of chance in linguistic encodings was held in check, just as arbitrariness, as explicated in the Cours, reined in by systemic motivation. Linguistic productions, then, are either fully intentional, or else constrained in the subconscious by systemic strictures, or, in some sense, both, an ambiguity captured in Saussure’s use of the phrase “...the rule her mind obeys...” in characterizing Mlle. Smith’s remarkably regular, though seemingly unrehearsed and automatic Sanskritoid. Just as systemic motivation separates radical arbitrariness from radical iconicity, so the uncanny regularity of this automatic language left little doubt that it was system-driven, springing from a language mechanism situated, in some sense, between chance and choice. Between 1906 and 1909 Saussure was to devote an inordinate amount of intense study to what seemed to him yet another manifestation of an absent presence in language. I refer here to his analysis of what he called ‘anagrams’. In poetic and prose works of classical Latin and Greek authors, Saussure became intrigued with seeming regularities in vowel and consonant repetitions. These repetitions occurred with such mathematical precision that Saussure was convinced they could only have been deliberately produced.6 Moreover, they appeared to be more than a means for achieving a sort of sonic equilibrium: these repeated sounds, he believed, were a device for reconstituting, over the course of a line, a theme word or proper name implicated in the text. In other words, the text was being to some extent driven by an overarching, otherwise absent, antecedent fact. The net effect of the anagrams was, then, to keep the absent present, and this strategy crisscrosses and challenges the principles of the Cours. Sound sequences in the anagrams, for instance, have a curiously dual character: at one and the same time they are part of syntagmatic relations, as linear sequences, as well as part of associative relations, as constituents of the echo of the evoked name or theme word. However, unlike the associative relations defined in the Cours as being strictly in the mind, arising from mnemonic groups without any fixed co-spatial structuring, the associative relations steering anagrammatic encodings in fact involve spatially co-present sounds. Since the guiding name or theme word, the fractionalized subtext, is reconstituted discontinuously over the course of the text, does this, paradoxically, require that we differentiate between contiguous and non-contiguous syntagmatic relations? Or between associative relations without, versus with spatial co-presence? All such theoretical complications and conundra of course disappear at once if the echos of the anagrams 6. In the discussion of analogy, Saussure observed: “Early Latin thus possessed a high degree of awareness of the constituent parts of a word (stems, suffixes, etc.) And of their fitting together. It is probable that in our modern languages it is not felt so acutely. But German probably has it more than French” (Saussure 1916: 166 [230]). These remarks should be factored in to the discussion of possibly creating a typology of languages based on the ratio of intrinsicallyarbitrary to relatively-motivated elements each manifests (cf. Saussure 1916: 131 [182-183]. Saussure’s colleague and former teacher, Michel Bréal, advanced a similar proposal to establish a linguistic typology based on what languages leave unexpressed. See Christy (2000) and Christy (2002) for discussion of Bréal’s ‘latent ideas’.


137

are simply ascribed to chance. But this is precisely what Saussure was not willing to do. On the one hand, the phenomenon seemed simply too regular, too measured to be mere statistical static; on the other, his conviction that parole reflected individuals’ deliberate choices, as opposed to the trappings of tradition located in langue, effectively contraindicated any alternative reading. Never mind that Saussure was unable to find even a shred of evidence attesting to any tradition of anagram encoding; never mind that Mlle Smith’s Sanskritoid ramblings turned out to be the product of a subconscious working in overdrive, the encodings of a protracted dream, as it were, marching through the measured paces of tropic space.7 Thus, as in the medium’s case, so with the anagrams Saussure fails to recognize that not only conscious, but also subconscious regularizing forces are informing the message manifest in parole. This, it appears, is the issue Saussure has on his mind in the discussion of syntagmatic relations. Here we read: “The characteristic of speech is freedom of combination: so the first question to ask is whether all syntagmas are equally free” (Saussure 1916:122 [172], emphasis mine). He follows this question with the clarification that fixed sequences, whether idiomatic expressions or paradigmatic components, belong to langue, not parole: “To the language, and not to speech, must be attributed all types of syntagmas constructed on regular patterns” (Saussure 1916:123 [173]). Saussure appears to be ruling out the possibility that subconscious patterning extends beyond the syntagmas of fixed lexical, paradigmatic and syntactic strings which he locates in langue, a possibility which would address the crux of the anagram phenomena, since they would then not absolutely have to be a matter of deliberate, conscious choice. The whole issue would seem, once again, to boil down to the matter of identity and value, not to mention arbitrariness versus relative motivation. After all, the anagrams can be seen as a ‘combinatory type’ every bit as legitimate as the prefix-stem-suffix sequences or the other “...sentences and groups of words based upon regular models” (Saussure 1916:123 [173]) given as examples. Even more telling is the claim that these combinatory types archived in langue are based “...on specific examples heard and remembered” (Saussure 1916:123 [173]). Hearing and remembering are, of course, the whole basis and point of anagrammatic encoding, and there is no reason to assume that the subconscious mind is not as attentive to phonic resonances as it is to grammatical patterning. Clearly, something is absent in Saussure’s model, and this is evident in the evasive conclusion to the discussion of syntagmatic relations: “Where syntagmas are concerned... one must recognise the fact that there is no clear boundary separating the language, as confirmed by communal usage, from speech, marked by freedom of the individual. In many cases it is difficult to assign a combination of units to one or the other. Many combinations are

7. As Todorov points out, “The mechanisms that Henry sets forth are familiar to any specialist in etymology (or, more generally, in rhetoric, since the mechanisms of etymological derivation are... only a projection of the tropic matrix onto history)” (1982:262).

138

T. CRAIG CHRISTY

the product of both, in proportions which cannot be accurately measured” (Saussure 1916: 123 [173]). What is absent is acknowledgment of a level of systemic patterning at work in language, a level somewhere between the legacy syntagmas of langue and the deliberately selected syntagmas of parole. What should be emphasized, however, is that such a level of patterning can be accommodated in Saussure’s model. In the discussion of syntagmatic interdependences, for example, we are told that “Almost all linguistic units depend either on what precedes or follows in the spoken sequence, or else on the successive parts of which they are themselves composed” (Saussure 1916: 126 [176]). In addition to this fuzzy and indeterminate “almost all” qualification, we are reminded of the power of associative groups to evoke “... not just one form but a whole latent system...” (Saussure 1916: 128 [179]). The absent name recuperated piecemeal over the course of an anagram-infested text flirts, to be sure, with the standard notion of linearity, what with its discontinuous distribution, and yet its constituent sounds and syllables precede and follow each other all the same, in the course of evoking the latent signified. The category of absence is central to Saussure’s linguistic theory –more specifically to his semiological theory— precisely because value, the reflection of différence, of the presence-absence tension, is the first principle of this theory. Moreover, since “…in semiological systems, such as languages, where the elements keep one another in a state of equilibrium in accordance with fixed rules, the notions of identity and value merge” (Saussure 1916: 109 [154]: emphasis mine), absence also underwrites the concept of linguistic identity: the identity of a given element is defined as being precisely all that it is not. ‘Equilibrium’ is the key to semiological systems, where everything hinges on sustained symmetries. In Saussure’s reconstruction of Indo-European vowels, the phantom laryngeals not only accounted for quality/quantity irregularities in the vowel system, but also restored symmetry to the system of Indo-European roots by replacing asymmetrical vowelconsonant or consonant-vowel sequences with uniformly symmetrical consonantvowel-consonant sequences. In the case of Mlle Smith’s Sanskritoid, the materially absent French language program, humming in the mental background, was the source of otherwise inexplicable inter-language symmetries of homophones and figures of speech. Clearly, systematicity and regularity are anything but alien to subconscious linguistic processing. Even in the case of Saussure’s anagrams a concern for symmetry looms large, with a certain conceptual equilibrium being achieved through the piecemeal phonic recapitulation of an otherwise absent, or at least non copresent, theme word or name. In a word, the category of absence is integral to the differential concepts of value and identity, the mechanism of all semiological systems. In all his teachings Saussure was in fact consistent in maintaining emphatically that the linguistic sign must be denominated negatively, for the simple reason that it exists only in that it contrasts with all that it is not, with all that is absent in itself. Intrinsically


139

and necessarily incomplete, sustained and renewed through the ever-fluctuating ways in which it takes on context-appropriate complements, the sign is necessarily and inescapably suspended between the two poles of absolute arbitrariness – absolute absence of apparent systemic motivation – and relative motivation. In recognizing the category of absence it becomes clear that what otherwise seem unrelated, if not unreconcilable, components of Saussure’s research are, in fact, informed by an overarching conceptual common denominator –absence. As Harris insightfully maintains, “Most of what remains ‘unread’ in Saussure is not in any way obscure or difficult to read. …it lies there rather like some hidden object in a puzzle picture, which remains ‘invisible’ until we look at the picture in a certain way. Once the object is ‘seen’, it becomes clear that it was visible all the time” (1987:xv-xvi).

REFERENCES

Bouquet, Simon. 1997. Introduction à la lecture de Saussure. Paris: Payot. Christy, Craig. 2002. “From Articulation to Comprehension: Steinthal and the Dynamics of Linguistic Intangibles”: Chaim H. Steinthal. Sprachwissenschaftler und Philosoph im 19. Jahrhundert (Studies in European Judaism, 4) ed. by H. Wiedebach and A. Winkelmann, 3-16. Leiden: Brill. Steinthal. Wirken und Rezeption (Studies in European Judaism, Volume IV), ed. by G. Veltri, H. Wiedebach and A. Winkelmann. Leiden: Brill. Flournoy, Théodore. 1900. Des Indes à la planète Mars. Reissued 1983, Marina Yaguello and Mireille Cifali (eds.). Paris: Seuil. Harris, Roy. 1987 Reading Saussure: A Critical Commentary on the Cours de linguistique générale. La Salle, Illinois: Open Court. Henry, Victor. 1901 Le langage martien. Paris: Maisonneuve. Saussure, Ferdinand de 1879. Mémoire sur le système primitif des voyelles dans les langues indo-européennes. Hildesheim: Georg Olms Verlagsbuchhandlung, 1968. Reprografischer Nachdruck der Ausgabe Leipzig 1879. . 1916. Cours de linguistique générale. Edited by Charles Bally and Albert Sechehaye with the collaboration of Albert Riedlinger Translated by Roy Harris 1983. La Salle, Illinois: Open Court. Todorov, Tzvetan. 1982. Theories of the Symbol. Transl. by Catherine Porter of Théories du symbole 1977. Ithaca, New York: Cornell Univ. Press.

140

141

LA CRÉATION DES COURS DE LETTRES AU BRÉSIL ET LES PREMIÈRES ORIENTATIONS DE LA RECHERCHE LINGUISTIQUE UNIVERSITAIRE JOSÉ LUIZ FIORIN Universidade Estadual de São Paulo Dans cette étude, je me limiterai à l’étude des orientations de la recherche au cours de Lettres de l’Université de São Paulo, durant la période 1934/1962, à savoir la période qui va de la création du premier cours supérieur de Lettres au Brésil jusqu’à l’année où a eu lieu une grande réforme des cours de Lettres brésiliens, ce qui allait avoir une profonde répercussion sur la recherche linguistique réalisée au Brésil. La recherche linguistique à l’université brésilienne a vu le jour avec la création des cours de Lettres. Ceux-ci n’apparaissent au Brésil que dans les années 30 du XXe siècle à la suite des projets de création des Facultés de Philosophie. Bien que l’existence d’une formation supérieure en langues et littératures eût déjà été revendiquée auparavant, et qu’une expérience à la Faculté de Philosophie São Bento, à São Paulo, au monastère du même nom, eût été tentée en 1908, puis une autre avec la création d’une institution libre appelée Faculté Pauliste de Lettres et Philosophie, laquelle a fonctionné dans la ville de São Paulo de 1931 à 1934 (Anuário1952a:170 )1, les premiers cours de Lettres au Brésil surgissent dans les années 30:1934, à la Faculté de Philosophie, Sciences et Lettres de l’Université de São Paulo; en 1935 à l’Université du District Fédéral; en 1939 à la Faculté Nationale de Philosophie de l’Université du Brésil et à l’Université de Minas Gerais. La création des Facultés de Philosophie a lieu quand s’ouvrent les premières universités brésiliennes dans la décennie des années 30 du XXe siècle. Des raisons nombreuses et complexes expliquent la création tardive de l’université brésilienne. Durant la période coloniale, la métropole détenait, en plus du monopole commercial, etc., le monopole de la formation supérieure, qui se réalisait à l’Université de Coimbra. En 1808, la famille royale portugaise se réfugie au Brésil quand les troupes françaises, commandées par le général Junot, envahissent le Portugal, allié de l’Angleterre. En 1815, le Pays est élevé au titre de Royaume Uni de Portugal et d’Algarve et, en 1822, l’Indépendance est proclamée et Pedro I, fils du roi de Portugal devient empereur du 1. Quand dans ce travail figure une citation comportant seulement la date et la page, sans indication d’auteur, il s’agit d’une citation de l’Annuaire de la Faculté de Philosophie, Sciences et Lettres.

142

JOSÉ LUIZ FIORIN

Brésil. Le transfert au Brésil du siège du pouvoir métropolitain et l’émergence de l’état national créent le besoin de la fonction de l’enseignement supérieur, destiné d’une part à former des fonctionnaires pour l’Etat et d’autre part des spécialistes de la production de biens symboliques pour la consommation des classes dominantes. Pendant la période impériale, les écoles créées à l’époque royale furent maintenues avec les adaptations nécessaires et d’autres furent créées à leur tour au cours de ces périodes royale et impériale, le modèle adopté était celui de grandes écoles destinées à former des cadres nécessaires pour la réalisation d’activités bien définies. Pour cette raison, on arrive à la période républicaine avec un enseignement réduit en nombre d’établissements et d’étudiants. La République brésilienne naît sous le signe du positivisme. Les adeptes de cette doctrine prenaient position contre la création d’universités, car, pour eux, il s’agissait d’une institution réactionnaire car, dans le monde entier, elle était devenue rétrograde (Mendes 1882:70). A partir des années 20 du siècle dernier, avec les conceptions nouvelles concernant la nécessité de rattraper le retard brésilien, l’idée de créer des universités au Brésil prend vigueur. La création de l’Université de São Paulo doit être entendue dans le cadre de la défaite pauliste dans ce qu’on a appelé la Révolution de 32. Un groupe de l’élite industrielle et agraire pauliste, à la tête duquel était Julio Mesquita Filho, commence à défendre l’idée de créer une université d’état. L’Université de São Paulo se destinait à créer une nouvelle élite, pouvant assurer les commandes du pays, pour surmonter le retard national. Pour cette raison, cette Université devait être publique et laïque. Elle devait disposer d’une autonomie académique et professionnelle. Le noyau en serait la Faculté de Philosophie, Sciences et Lettres, où seraient enseignées toutes les disciplines de base. Les professeurs de la nouvelle Faculté travailleraient en temps intégral, et se consacreraient aux activité de recherche et d’enseignement. L’union de la recherche et de l’enseignement constituait la base de la conception de la nouvelle université. Les travaux pratiques seraient réservés aux écoles professionnelles. (Schwartzman 2001:164-160; Campos 1954:73-99). L’Université de São Paulo a été créée par le décret 6.283, du 25 janvier 1934. Par ce même décret, était créée la Faculté de Philosophie divisée en trois sections: celle de Philosophie, celle de Sciences et celle de Lettres.. La section de Sciences était subdivisée en six sous-sections: Sciences Mathématiques, Sciences Physiques, Sciences Chimiques, Sciences Naturelles, Géographie et Histoire et Sciences Sociales et Politiques. La section de Lettres était subdivisée en deux sous-sections: Lettres Classiques et Portugais et Langues Etrangères.2 Lorsque on avait pensé à fonder l’Université de São Paulo, ce qu’on voulait, c’était former une nouvelle élite pour le pays, éduquée sur le modèle des pays les plus avancés du 2. Le projet de l’USP (Université de São Paulo) n’a pas remporté un succès total. De fortes tensions entre les anciennes écoles supérieures et la Faculté de Philosophie obligèrent à une série de compromis.

LA CRÉATION DES COURS DE LETTRES AU BRÉSIL

143

monde. On avait alors prévu de recruter tous les professeurs en Europe. Avec la vague montante du nazi-fascisme en Allemagne et en Italie, il fut décidé que dans ces pays ne seraient recrutés que des professeurs de Sciences Physiques et Biologiques, tandis que les professeurs des Sciences Humaines viendraient de France, considérée comme une alternative libérale au fascisme (Duarte 1976:70; Mesquita 1969:192). Ce projet a été réalisé dans ses grandes lignes. A l’Université de São Paulo ont enseigné des professeurs tels que Roger Bastide, Fernand Braudel, Giuseppe Ungaretti, Claude Lévi-Strauss, etc. Au commencement, la section de Lettres était organisée en deux cours: Lettres Classiques et Portugais et Lettres Etrangères. Le premier comprenait les chaires suivantes: Philologie Grecque et Latine; Philologie Portugaise; Littérature Luso-Brésilienne; Littérature Grecque et Littérature Latine; le second comprenait les chaires de Langue et Littérature Française et de Langue et Littérature Italienne (1937:1). C’est en 1940 seulement qu’ont commencé à fonctionner les chaires de Langue et Littérature Espagnole, Langue et Littérature Anglaise et Langue et Littérature Allemande. La chaire de Langue Tupi-Guarani, dans les débuts de l’existence de la Faculté, n’appartenait pas à la Section de Lettres, mais à celle de Géographie et Histoire. Il y avait une chaire d’Ethnographie Brésilienne et Langue Tupi-Guarani. En 1939, la Faculté dut s’adapter à la norme de la Faculté Nationale de Philosophie, créée à Rio de Janeiro, par le décret Fédéral n° 1.190, du 4 avril 1939. Dans la section de Lettres, sont constitués les Cours de Lettres Classiques, Lettres Néolatines et Lettres Anglo-Germaniques, organisation qui perdurera jusqu’en 1962, quand, sur la base de l’avis 283/62 du Conseiller Valnir Chagas, du CFE (Conseil Fédéral de l’Education), approuvé le 19 octobre 1962, tous les Cours de Lettres sont réorganisés dans le pays. Les chaires chargées de ces trois nouveaux cours étaient celles de Langue et Littérature Latine, Langue et Littérature Grecque, Philologie et Langue Portugaise, Littérature Portugaise, Littérature Brésilienne, Philologie Romane, Langue et Littérature Française, Langue et Littérature Italienne, Langue Espagnole et Littérature Espagnole et HispanoAméricaine, Langue Anglaise et Littérature Anglaise et Anglo-Américaine, Langue et Littérature Allemande. 1.

Evolution de l’orientation des chaires au long des années

En Philologie et Langue Portugaise, quoiqu’il soit reconnu qu’au Brésil on parle une variante linguistique différente de celle du Portugal, si bien que le premier professeur de la chaire, le portugais Rebelo Gonçalves, parle de la confection d’une grammaire lusobrésilienne de la langue portugaise (1937:197), le premier programme réserve une place réduite aux caractéristiques du portugais du Brésil (1937:289-291). L’idée d’une langue commune dont la norme était la variante européenne imprègne tout le programme. L’orientation des programmes pour l’étude de la langue était avant tout historique. Outre la grammaire historique, on insiste, en étudiant l’histoire de la langue, sur la langue

144

JOSÉ LUIZ FIORIN

littéraire, principalement sur l’étude des auteurs qui ont été considérés comme des modèles de perfection linguistique, Camões, Vieira et Bernardes (1937:289). Les commentaires de textes pour l’étude des phases de la langue, se concentrent sur des textes de la phase archaïque du portugais (1937:291). Si l’orientation était l’orientation historique, la Géographie Linguistique, en accord avec la tradition portugaise d’études linguistiques, commence à être prise en considération (1937:289; 1937a:281). Ainsi, la recherche linguistique réalisée à la chaire de Philologie et Langue Portugaise aux premiers temps de l’USP suit les courants dominants de la Linguistique Historique, mais commence à s’intéresser à la Géographie Linguistique, à ses méthodes et ses objectifs. Le 22 avril 1940 Francisco da Silveira Bueno est nommé à la chaire de Philologie et Langue Portugaise. Quoique cette chaire conserve son orientation essentiellement philologique et historique, d’autres influences commencent à se faire sentir: d’une part il publie, en 1950, son Traité de sémantique générale, qui révèle les influences de sémanticistes français et allemands de l’époque; d’autre part, en 1951, démarre l’installation à la Faculté d’un Cabinet de Phonétique Expérimentale (1952:217). Pour Silveira Bueno, la philologie est une discipline qui a pour fonction de coordonner et d’agréger les connaissances des différentes disciplines, car elle pour objet particulier la connaissance parfaite et complète de la vie intellectuelle d’un peuple a travers tous ses monuments littéraires (1953:84). La philologie Portugaise doit être(…) l’étude de la civilisation, de l’esprit, de toute la vie intellectuelle du peuple lusitain, par l’intermédiaire des monuments que nous ont légués ses générations passées (1953:85).

Les textes anciens sont la raison d’être des études philologiques. Pour établir les textes, il est nécessaire de connaître à fond la langue du temps où ceux-ci ont été écrits. La conception de philologie qui oriente les travaux de la chaire de Philologie et Langue Portugaise est la conception classique de cette branche de la connaissance, qui se fonde sur le concept de culture en tant que l’ensemble des connaissances intellectuelles d’un peuple donné. Selon cette conception, la culture est comprise comme une réalisation des classes dominantes. Les chaires de Langue et Littérature Grecque et de Langue et Littérature Latine avaient une orientation nettement littéraire. Les études linguistiques se destinaient seulement à permettre aux élèves d’avoir accès aux textes dans l’original. Un cours supérieur de Lettres Classiques devait étudier, d’une part l’histoire de la langue; d’autre part sa littérature (1937:187). De la sorte, pendant longtemps, les études linguistiques, en Lettres Classiques, se destinaient à étudier la phonétique, la morphologie, la syntaxe et la lexicologie historiques. La recherche linguistique avait ainsi une orientation nettement historique. Les programmes de Littérature visaient, d’une part, à donner une vision panoramique de la littérature étudiée; d’autre part, à amener l’élève à faire des explications de textes (1952:244; 1952a:257; 1953:501). D’un autre côté, les premiers professeurs de Lettres Classiques insistèrent beaucoup


145

sur l’élaboration d’instruments de travail: grammaires, dictionnaires, et surtout des éditions de textes classiques, avec des traductions et des commentaires en portugais (1937:190; 1954:270). De par le propre objet dont elles traitaient, dans les chaires de Philologie Romane et de Glottologie Classique, l’orientation était aussi historico-comparative. La Glottologie Classique s’occupait de ce que nous appellerions aujourd’hui Linguistique IndoEuropéenne, avec une importance toute particulière attribuée à la comparaison du latin et du grec. ( 1952:244-246; 1953:439-440; 1954:279). Toutefois, c’est à la chaire de Philologie Romane qu’a commencé l’étude de la linguistique moderne. Le Professeur Teodoro Henrique Maurer Junior, titulaire de la chaire, avait été boursier, au cours de l’année scolaire 1945-1946 de la Fondation Rockefeller, à l’Université de Yale, où il étudia le sanscrit, le hittite, la phonétique et la linguistique générale, et fut alors l’élève de Bloomfield. Il était membre de la Linguistic Society of America et fut collaborateur de la revue Langage (1952:180-181). C’est lui qui, à São Paulo, commença à enseigner et à répandre les théories de Saussure, Bloomfield, etc. Les cours de Langues Etrangères, durant la période considérée, étaient beaucoup plus tournés vers le monde de la réflexion poétique que vers la description linguistique. Toutes les chaires suivaient la même orientation: la connaissance de la langue était pratique pour que l’élève pût lire les textes littéraires dans l’original; il était important de connaître l’histoire de la langue, pour une étude plus précise des textes littéraires des différentes époques; la connaissance de la littérature était la finalité ultime d’un cours supérieur de Lettres (1937:198-206, 294; 1938:364; 1952:222, 226, 239; 1952a:239, 245, 251252; 1953:467, 469, 473, 481, 490, 491; 1954:255). Le premier professeur titulaire de la chaire de Tupi-Guarani, dans un rapport de 1934-1935, disait que l’étude de cette langue ne pourrait s’orienter dès le début avec l’assurance qui serait souhaitable parce qu’elle ne disposait pas de bibliographie obéissant aux exigences d’un cours académique. Pour cette raison, le premier travail de la chaire consisterait à préparer l’ébauche d’une grammaire de la langue, à partir de la collecte d’éléments dans des œuvres anciennes et modernes (1937:140-141). Cette ébauche devrait recueillir, “avec simplicité et méthode, les particularités idiomatiques découlant des lois et des faits d’ordre général” (1937:141). Le professeur s’interroge si la Faculté doit donner un cours de caractère pratique, c’est-à-dire s’il faut enseigner aux élèves la pratique de la langue et en même temps, il se demande quelle langue doit être enseignée, celle qui se parle aujourd’hui au Paraguay ou le tupi-guarani amazonien, celui qui est appelé nheengatu. Pour lui, un cours supérieur ne comporte pas d’enseignement aux finalités pratiques, où l’on s’occuperait d’apprendre à parler telle ou telle langue, tel ou tel dialecte. D’un autre côté, de la même manière que le grec qu’il faut étudier est le grec classique, il faut étudier ce qu’on pourrait appeler le tupi classique, qui, bien qu’il puisse être compris par les pratiquants du tupi-guarani d’aujourd’hui, en diverge par de nombreux aspects. La chaire de Tupi-Guarani doit être située au même niveau où se situent celles de Grec

146

JOSÉ LUIZ FIORIN

et de Latin. Pour cette raison, on ne s’occupera pas de la partie pratique de la langue, “de son étude dans le champ limité des aires géographiques où elle est employée aujourd’hui”, de “sa situation actuelle face aux influences de plus en plus fortes qu’elle reçoit des peuples dominants, mais exclusivement de ses aspects génériques, classiques et fondamentaux, seuls capables de fournir une vision panoramique de sa structure et de son caractère” (1937:144-145). La recherche linguistique, faite dans un Institut de Philologie devrait avoir une orientation historique et comparée: il serait nécessaire d’étudier l’évolution des faits linguistiques du tupiguarani et de comparer le tupi avec d’autres langues américaines, pour établir les familles linguistiques, ou, selon les propres mots du professeur de la chaire, établir “l’interdépendance avec d’autres langues américaines” (1937:145). Comme on le voit, l’orientation initiale de la chaire de Tupi-Guarani était une tentative de donner à cette langue le statut des langues classiques et des autres langues, le portugais, le français et l’italien. La recherche linguistique avait une orientation historico-comparative et la langue que les élèves devaient étudier, était celle qui a été désignée comme le tupi classique. Le tupi-guarani était passible d’une étude philologique telle qu’elle était pratiquée dans d’autres langues (1953:562). L’orientation de la première chaire de langue indigène créée dans une école supérieure était bien distante de l’orientation qui a prévalu au cours des décennies suivantes. En effet, plus tard, les chercheurs allaient s’attacher à décrire les différentes langues parlées par les indigènes sur le territoire brésilien. Conclusions 1) Dans les Langues Etrangères, ce qui était l’objectif de la recherche, c’était la littérature; l’enseignement de la langue était le moyen d’amener les élèves à lire les textes originaux. Même les cours d’histoire de la langue avaient pour finalité de préparer l’élève à la lecture de textes produits à d’autres étapes de la langue. Par conséquent, au cours de la période qui va de 1934 à 1962, à part de rares exceptions, il n’a pas été réalisé de recherches linguistiques dans le domaine des langues étrangères. 2) Dans les Lettres Classiques, la littérature était également l’objet privilégié de la recherche. Au cours de la période considérée, en plus des études sur des auteurs et des œuvres, la priorité a été donnée à la production d’instruments pour le travail littéraire, à savoir, des traductions en portugais des auteurs greco-latins, avec une introduction, des notes et des commentaires. Les cours de langue suivaient l’orientation historico-comparative et quelques travaux ont été produits selon cette orientation théorique. Toutefois, le gros des travaux de recherche portaient sur des thèmes littéraires.


147

3) La recherche linguistique proprement dite est demeurée confinée aux chaires de Philologie et Langue Portugaise et Philologie Romane et Glottologie Classique. Dans la première, il y avait d’une part le travail philologique, qui était considéré comme le plus important. Cependant, comme le travail philologique portait, d’abord, sur les textes archaïques du portugais, il était important de réaliser des travaux sur l’évolution du portugais. Ainsi, la recherche diachronique occupe une place importante parmi les travaux réalisés. D’autre part, suivant une tradition portugaise, représentée par des noms comme Leite de Vasconcelos, et Carolina Michelis de Vasconcelos, quelques travaux ont été réalisés suivant la ligne de la Géographie Linguistique. 4) La chaire de Linguistique Romane et de Glottologie Classique ont une orientation historico-comparative. C’est là néanmoins, grâce à la formation de son professeur titulaire, qu’ont commencé à se diffuser les idées des fondateurs de la Linguistique Moderne. Ce sera la base de la formation de toute une génération de linguistes qui sont aujourd’hui en activité dans différentes universités brésiliennes. 5) L’orientation philologique et historico-comparative était si forte que même la chaire de Tupi-Guarani l’imprime à ses études et cherche à étudier un stade de la langue, celui qui a été décrit par Anchieta, Montoya et Figueira, et qui est désigné comme tupi classique. On ne réalise pas d’études descriptives du tupi-guarani, mais on s’efforce de donner à l’étude de cette langue un statut identique à celui du grec et du latin. 6) Comme on le remarque, l’Université de São Paulo suit l’orientation pédagogique et les lignes de recherche de la moyenne des universités européennes où ont été recrutés ses premiers professeurs. RÉFÉRENCES Primary sources Anuário da Faculdade de Filosofia, Ciências e Letras 1934-1935, 1937. São Paulo, Revista dos Tribunaes. Anuário da Faculdade de Filosofia, Ciências e Letras 1936, 1937a. São Paulo, Faculdade de Filosofia, Ciências e Letras. Anuário da Faculdade de Filosofia, Ciências e Letras 1937-1938, 1938. São Paulo, Faculdade de Filosofia, Ciências e Letras. Anuário da Faculdade de Filosofia, Ciências e Letras 1939-1949, 1953. São Paulo, Faculdade de Filosofia, Ciências e Letras, v. I e II. Anuário da Faculdade de Filosofia, Ciências e Letras 1950, 1952. São Paulo, Faculdade de Filosofia, Ciências e Letras. Anuário da Faculdade de Filosofia, Ciências e Letras 1951, 1952a. São Paulo, Faculdade de Filosofia, Ciências e Letras. Anuário da Faculdade de Filosofia, Ciências e Letras 1952, 1954. São Paulo, Faculdade de Filosofia, Ciências e Letras.

148

JOSÉ LUIZ FIORIN

Secondary bibliography Campos, Ernesto de Souza. 1954. História da Universidade de São Paulo. São Paulo: USP. Cunha, Luiz Antônio. 1986. A universidade temporã: da colônia à era Vargas. 2 ed. Rio de Janeiro: Francisco Alves. Duarte, Paulo. 1976. Memórias. Selva obscura, vol. 3. São Paulo: Hucitec. Mendes, Raimundo Teixeira. 1882. A universidade. Rio de Janeiro: Centro Positivista Brasileiro. Mesquita Filho, Júlio de. 1969. Política e cultura. São Paulo: Martins. Schwartzman, Simon. 2001. Um espaço para a ciência: a formação da comunidade científica no Brasil. Brasília: Ministério da Ciência e Tecnologia, Centro de Estudos Estratégicos.

149

THE PORTUGUESE LANGUAGE IN THE INSTITUTIONALIZATION OF LINGUISTICS SUZY LAGAZZI-RODRIGUES Universidade Estadual de Campinas The movements to grammatize the Portuguese language and the legitimation of Linguistics in Brazil as a scientific and institutional process are a fact. Princípios de Lingüística Geral (Principles of General Linguistics) and Estrutura da Língua Portuguesa (Structure of the Portuguese Language) by Joaquim Mattoso Câmara Jr. were used to analyze and demonstrate this fact, and the scientific, institutional and legitimated aspects of Linguistics are evident. The discursive memory shows the path of scientific and institutional legitimation in the studies of the Portuguese language in Brazil, as will be demonstrated in this article. 1. Setting the Point When we think of Linguistics in Brazil, we cannot forget the great author and researcher Mattoso Câmara, whose work provided us with a scientific description of grammar. Guimarães (2000) declares that “Said Ali was the one who not only recognized the distinction between two kinds of descriptive grammars (one prescriptive and the other scientific), but he also put the scientific grammar to work”. On the other hand, Guimarães points out that, although Said Ali inscribed the referred scientific descriptive grammar into Brazilian history, he did not actually complete it. We can say that this position is made clear in Estrutura da Língua Portuguesa (Structure of the Portuguese Language) by Mattoso Câmara in the 1970. We will observe that the establishment of scientific descriptive grammar will be accomplished by the linguist as opposed to the prescriptive grammarian. Related to this, Orlandi (2000) states that, at the time when the changing of the authorship from the grammarian to the linguist occurs, “the question posed by the language also changes from a political and intellectual perspective to an explicitly scientific one.” As Orlandi points out, in spite of authorship effacing in grammar, the position of the author still prevails but now as a school imposition of a sort of grammatical linguistic standard, which “manifests effects on the knowledge of the language which develops parallel to linguistic knowledge and its metalanguage.” The author goes on with the argument that each one of the disciplines presents a peculiar answer to the unity x diversity relationship

150

SUZY LAGAZZI-RODRIGUES

concerning language. Such an interchange occurs in a place where contradiction is manifest in the school system, which is the place for teaching language. “On the one hand, legitimation through Linguistics and on the other, grammar through terminology. Philology goes between the two of them - through comments - by interweaving the complex articulation between knowledge and language practice in building up the relationship unity x diversity.” Therefore, we observe that linguistic knowledge supported by metalanguage is represented within a science-based process which will be legitimated by the school as an institution. Generally speaking, the school makes the difference between grammar rules and the scientific description of language stable as its operation becomes more and more structured by the teaching of the language grammar. School is the place for the language teacher. He works under the standardization of NGB (Brazilian Grammatical Nomenclature) according to the position of a grammarian. On the other hand, the linguist takes over a scientific perspective of thinking about the language. University language courses represent remarkable places concerning this redistribution of knowledge. Guimarães (1996) defines four periods concerning the development of the Brazilian grammar of the Portuguese language. There are debates on language issues between Brazilian and Portuguese people during the first and second periods, aside from the foundation of the ‘Academia Brasileira de Letras’ (Brazilian Academy of Letters) and the publishing of several important grammars. The third period starts at the end of the thirties, with the introduction of University Language courses. The 1943 Orthographic Agreement regulations also take place in this period as well as the debate to define the name to be given to the language in use in Brazil. Guimarães mentions a series of works brought forth in this period: History of the Portuguese Language (1952) by Serafim da Silva Neto, The Historical Formation of the Portuguese Language (1955) by Silveira Bueno, the descriptions of the Portuguese language by Mattoso Câmara, published in the sixties and republished in 1970 as a book - Structure of the Portuguese Language - and some studies on native languages by the same author, the famous NGB (Brazilian Grammatical Nomenclature) in 1958, Atlas prévio dos falares baianos (1963-65) by Nelson Rossi, A Policy of the Language (1965) and The Portuguese Language and Brazilian Reality (1968) by Celso Cunha, and Grammar of Contemporary Portuguese (1970), also by Celso Cunha. We observe that the specific works of the third period show in their very titles the prevailing thought concerning language studies in those days: ‘Língua Portuguesa’ (the Portuguese Language). As an object, it was established as the national language, clearly defined from the grammarian position. It also follows an institutional process that grows bigger and bigger. Guimarães also states that the foundation of University Language courses opens up a ‘place for research on language issues’ with enquiries that show concern with both literary standard as well as with teaching. From 1965 onward, Linguistics as a discipline is introduced and is compulsory in University Language courses. According to the author, this fact marks the fourth period of the grammatization of the Portuguese language in Brazil.

PORTUGUESE IN THE INSTITUTIONALIZATION OF LINGUISTICS

151

At first sight we might say that the period of Brazilian debates on grammatization of the Portuguese language, when research on language issues is brought into University language courses, represents an interesting gap which leads to the introduction of Linguistics as a compulsory discipline of the curricula of the referred courses. Language issues which mark this gap also contribute to the building of a stable basis for standardized grammar, besides laying solid foundations for the development of Linguistics in Brazil. I restate that the scientific practice process takes place in an institutionalized context. This point is crucial to understand the process in analysis. Both science and institution become superimposed. Orlandi and Guimarães (1998) state that although there is not a straightforward extension between nineteenth-century studies on language in Brazil and the way Linguistics developed later in the country “there is already a sort of work that arises out of Linguistics, such as it was developed here. Above all, there is a work of institutionalization of the relationship between Brazilians and the Portuguese language. Parallel to this, a representation of knowledge (such as schools, grammar, manuals, and literature) is constituted in our society. I imagine that science and institution cross at a two-way road”. Both authors argue that “writing, knowledge of language and identity with a national language are conclusive for the institutional form of our society and politics”. Guimarães and Orlandi go on stating that both “the explanation of the relationship subject x language by school as an institution and the publishing of grammars represent the scientific practice process” that took place at that time. “It is a conclusive time for both, the constitution of the historical form of Brazilians and the institutionalization of Linguistics as a discipline”. It is important to point out that, in this complexity of relations, the connection between science and institution is legitimated. Science has to be legitimated as an institution. How does this process take place with the discipline of Linguistics in Brazil?

2. Taking the matter into consideration When we think of Linguistics as a discipline, Mattoso Câmara Jr.’s works, Princípios de Lingüística Geral (Principles of General Linguistics) and História e Estrutura da Língua Portuguesa (History and Structure of the Portuguese Language), are reference texts. Published in 1941, the first one marks the birth of Linguistics in Brazil according to Silvio Elia. The author writes about this fact in the 1977 edition of the referred work: PORTUGUESE IN THE INSTITUTIONALIZATION OF LINGUISTICS the birth of Linguistics in Brazil and perhaps in the world of the Portuguese language.

152


On the introductory note to the second edition, in 1954, Mattoso stresses that he tried not to change the book’s plan: the effort not to modify the book’s plan and to tie it into the subject that focalized it from the beginning. The subtitle – “How to Introduce Superior Studies of the Portuguese Language” – explains this intentional limitation of ambit. (Mattoso Câmara 1977)

According to the title and subtitle we can observe that General Linguistics establishes the approach for the studies of the Portuguese language. On the other hand, the Portuguese language gives details of the thought on general Linguistics. The movement from general Linguistics to the specificity of the Portuguese language is a distinguishing trait of this work by Mattoso and it is also relevant for us to understand the process of the institutionalization of Linguistics. It is characteristic of the structuralist perspective, which supports the thought of Mattoso’s work as we can observe below. The general table of contents of Princípios de Lingüística Geral (Principles of General Linguistics) provides us with an overall view of the work developed. We count twenty chapters that run as follows: I. Content and Scope of linguistics; II. Phonemes or units of phonation; III. Syllable and word phonetics; IV. Units of the Language; V. The functions of morphemes; VI. The grammatical categories; VII. The categories of gender; VIII. The category of aspect; IX. The verbal voices; X. Types of words; XI. The phrase and its structure; XII. Modality of the phrase; XIII. Classification of the languages for descriptive purposes; XIV. The classifications of Schleicher and Sapir; XV. Concepts of the linguistic evolution; XVI. Linguistic causes of the evolution; XVII. Aspects of the phonetic evolution; XVIII. Phonetic Laws; XIX. Borrowing and its concept; XX. Social and linguistic aspects of Borrowing

Taking the chapters as a whole, we can observe the structuralist principles that set the boundaries for thought on language. From the phoneme to the sentence, structuralism settles the border lines of analysis. We can also note that the name ‘Portuguese language’ is not mentioned in the chapter names. This fact leads us to raise questions on how the relation between Linguistics and the Portuguese language takes place. But when we take the chapters into consideration, we immediately encounter examples in Portuguese for all the general concepts accepted and explained by Mattoso, based upon great authors formed and legitimated by Linguistics as a science. In the second section of Chapter I, we find the first reference to the Portuguese language. Mattoso then makes the comparison between the ‘Portuguese system’ and the ‘Portuguese language,’ which is essential for the structural perspective he adopts: Each community of men is served by a system of language, the essential property of which is to be representative. Therefore, in the Portuguese system, or Portuguese language, the phenomena that impress our senses are interpreted, and REPRESENTED, - either consubstantiated in a


153

“being”, not rarely merely conventional, as death (substantive names), or as «processes» in development (verbs) […] are divided into classes by a given criterion (such as that of nominal gender), and are given a common or different denomination according to, for example, the way in which they are interpreted on the basis of certain constitutive elements; A, in which we pick apples, B, in which we collect firewood, C, in the shade where we find protection, such as a tree, or to the contrary, on the base of other elements, an abstraction of the general groups is made and called A apple tree, B eucalyptus, C jequitibá tree. In this way, the space in which we live […]. Mattoso Câmara Jr. 1977:17)

The word ‘assim’ (‘so’) is meaningful to the relationship established between the representative attribute of the language systems and the fulfillment of such representation with terms of the Portuguese language. Due to the fact that representativity of language is general, examples can be formulated in Portuguese. Formulation causes the effect of direct consequence between attribute and language in such a way that the Portuguese language can only provide examples. Below we can see some other passages by Mattoso: The current descriptive linguistics tends to consider these groups of opposition on the basis of a “binarity”, that is, a conjugated binary, or having two members. When a linguistic trace distinguishes one of them and opposes the other, we have the GRAMMATICAL MARK OF THE OPPOSITION. […] Even when we come upon three or more members, there is an implicit rule, a hierarchical organization that may conduct us to understand successive binary groups: […] For this reason, our set of seven vowels consists of a complex and decomposable structure in binary oppositions: […] (Mattoso Câmara 1977:43) This led Brugmann to a theoretical schematization of the diverse types of aspects possible for the existence of a verbal conjugation, highlighting: 1) the PUNCTUAL aspect, or momentaneous […]; 2) DURATIVE […]; 3) PERMANSIVE […]. In Portuguese, some of these aspects are rooted in the meaning of certain verbs; this can be seen in the proposal of to leave and to arrive, which are essentially verbs – the first inceptive and the second, cessative. Analogically, to fall is punctual; so, in the infantile folklore song, we have it repeated to indicate a slow fall (durative aspect): “cai, cai, balão na rua do sabão” [“fall, fall, balloon in the street of soap …”]. Others are characterized by suffixes. This is the case of the frequentative aspect, with a diminutive notion added, implicit in the reach of the “itar” suffix, in the manner already used in Latin, where this was the essence of certain derived verbs, in “are”, related to the radical of the supine of a simple verb […]. In this way, we have the aspect functioning on a lexical level, by the change of semanteme, or on the level of derivation vocabulary, in which an affix imprints a particular value on a semanteme without affecting its inherent meaning (cf. §59).

154


However, even on the purely grammatical level of the conjunction, the aspect-category functions subsidially in Portuguese. The resource consists of the use of composed or periphrastical conjugations, in which an auxiliary verb […]. So, the CURSIVE ASPECT is formulated in the present, past, future and in any mode by means of the auxiliary verb estar [to be] conjugated with a gerund (estou cantando [I am singing], estava cantando [I was singing], estive cantando [I had been singing], estarei cantando [I will be singing], etc.) […]. In other and innumerous languages, these are the notions – whether subsidiary or adventitial in Portuguese – that constitute the dorsal spine of verbal conjugation. In the African languages Bantu and Sudanese, Maurice Delafosse deducts, analogally, the existence of three central aspects […] (Mattoso Câmara 1977: 142-144).

In the second passage, as in the first one, the Portuguese language is in the place of a conclusive fulfillment, in this case, the binary opposition. In the third passage, we can observe the question of the conceptual aspect being discussed according to its functioning on several levels of the language system of both the Portuguese language as well as “outras e inúmeras línguas” (other and uncountable languages). Taking this instance, we can perceive that it is essential to have a structural thought concerning languages, specifically the Portuguese language. In a general overview, I would say that these language formulations make up the evidence of languages as systems and for the history of linguistic ideas in Brazil. The fact that Portuguese language as a system is constructed through such formulations is what really matters. On that account, in the opposite direction, we find the manifestation of linguistic thought in our country. Legitimation of linguistic thought is built in Brazil at the same time as theoretical concepts are manifest in the Portuguese language, supporting structuralist perspectives. It is extremely relevant when we think about the Brazilian authorship process which is part of the manifestation of the linguistic thought in Brazil, where we find references to the author himself among several quotations by Mattoso Câmara: Understood as a game of physionomy, it is undeniable that the gesture of the arms and even the whole body ineluctably accompanies the vocal enunciation and a “specifying function of the word” is integrated in it (Kainz, 1943, II, 498), but only, as I said before, «in the manner of a musical background that accompanies the words of a song» (Mattoso Câmara 1977: 19).

When we examine the whole list of the referred works, we observe that besides Mattoso Câmara we also find Capistrano de Abreu, Said Ali, Antenor Nascentes, Theodoro Sampaio, Serafim Silva Neto e Sousa da Silveira. These are Brazilian authors who make up a group of legitimated and legitimating authors concerning the linguistic thought.


155

Now we will examine the work História e Estrutura da Língua Portuguesa. This text was elaborated between 1963 and 1965 and published in English in 1972 as part of the set of books The History and Structure of Languages. This text is different from Princípios de Lingüística Geral in which the Portuguese language is formulated according to a mechanism of representation and instances related to the general theoretical concepts taken from authors that form a long list of referred books. In this case, the Portuguese language is not parallel to several other systems, but examples from other languages are brought out to elucidate theoretical questions. The main point in História e Estrutura da Língua Portuguesa (The History and Structure of the Portuguese Language) as well as in Estrutura da Língua Portuguesa (Structure of the Portuguese Language) (1970) is that the Portuguese language is taken and described according to a system of its own. The description must begin with the unity of the fundamental features and, primordially, support itself in it. The differences enter secondarily and scale up in importance in the general linguistic structure. Therefore, they arrive at a comprehension of the Portuguese language that will be a constant directive for the study we are going to begin. (Mattoso Câmara 1976:9)

When Mattoso starts Chapter II named ‘Fonologia’ (Phonology) in part I ‘Prosódia’ (Prosody), he defines the subtitle ‘O acento em Português’ (The Accent in Portuguese) and begins his considerations as follows: It is best to begin the study of Portuguese Phonology by its accents, or prosodies, due to their fundamental importance in all of the phonological system. The Portuguese accent is extensive but not violent. It is much stronger in Portugal than in Brazil with a great contrast between the tonic syllable, which is not found in Brazil. […] (Mattoso Câmara 1976:33)

It is remarkable that the comparisons with the Latin system presented throughout the entire book are used to contextualize specific statements on the Portuguese system, as we can observe in the passages below: The Portuguese accent is free within the limits comprehended between the last and the antepenultimate syllable of the word (an acute, grave and proparoxitone, respectively). […] In Latin there were the limits between the next to last and antepenultimate syllable, […] (Mattoso Câmara 1976:33-34) Therefore, according to the nature of the predicate – verb or name – a Portuguese phrase, as occurs with the Latin phrase, is respectively – verbal or nominal. (Mattoso Câmara 1976:233)

156


We now arrive at the most important radical variation for the total comprehension of the structure of Portuguese verbs: that which occurs from an opposition between the forms of imperfect and perfect that disappeared from the great majority of the verbal radicals in consequence of the elimination of mark -u- of the latin perfectum. The small number of verbs that preserve the distinction in Portuguese are essentially divided in three types: The first presents an opposition of themes, not exactly radicals. It is constituted by two verbs: […] (Mattoso Câmara 1976:156)

Princípios de Lingüística Geral (Principles of General Linguistics) brought about the manifestation of the Portuguese language as a system. In its turn, História e Estrutura da Língua Portuguesa consolidates the Portuguese language as the place for linguistic description, as previously stated by Guimarães. The place of the analyst is brought forth after the beginning of the Portuguese language description. The position of the linguist as an analyst is consolidated through the Portuguese language description, considering the area of Linguistics in Brazil. The meaning of the Portuguese language as a manifestation of a structural place for the description of the analysis consolidated, with the work of Mattoso Câmara being essential for both, the institutional and scientific legitimation of Linguistics in Brazil as well as for the acknowledgement of linguistic studies. In an attempt to draw an overview, I would say that the Portuguese language means the scientific institution itself because it makes Linguistics and linguistic studies possible in Brazil. It is a question of legitimation in the scientific area which necessarily takes place through the institutionalization of the Portuguese language in the same scientific area. REFERENCES

Mattoso Câmara Jr, J. 1976. História e Estrutura da Língua Portuguesa. Rio de Janeiro: Padrão. . 1977. Princípios de Lingüística Geral. Rio de Janeiro: Padrão. Guimarães, E. 1996. “Sinopse dos Estudos do Português no Brasil: A Gramatização Brasileira”. In: Guimarães,E. e Orlandi,E. (orgs). Língua e Cidadania: O Português no Brasil. Campinas: Pontes. . 2000. “Entre o Estilístico e o Gramatical. Mattoso Câmara na História da Lingüística no Brasil”. Apresentado no “Congresso Internacional 500 anos de Língua Portuguesa no Brasil”. Évora: Portugal. Orlandi, E. 2000. “Métalangage et Grammatisation au Brésil: Le Rapport Grammaire/ Philologie/Linguistique”. In: Orbis Supplementa. Louvain: Peeters. . & Guimarães, E. 1998. “La Formation d’un Espace de Prodution Linguistique. La Grammaire au Brésil”. In: Langages, 130. Paris: Larousse.

157

HUMOR AND LANGUAGE ACQUISITION ANECDOTAL DATAAND THEIR ROUTE IN THE HISTORY OF LANGUAGEACQUISITION STUDIES

ROSA ATTIÉ FIGUEIRA Universidade Estadual de Campinas It is well known that, in addition to the material recorded in video and audiotape sessions, researchers in the area of Language Acquisition frequently analyze segments which were registered by members of children’s families. Such linguistic occurrences, at a certain point, called attention for their unusual nature and were taken by the child’s interlocutors as a kind of “linguistic deed” or as a funny linguistic novelty. Generally, this linguistic registered novelty interests the researcher and is added to other collected data in order to identify particular linguistic phenomena which might be involved in that material. One of the objectives of this paper is to examine how representative this type of data is within the history of the area of Language Acquisition, from its birth to the present. This aim not only includes, but brings out the oldest contributions made by the so-called diarists. Many of the studies carried out by Sully (1896) and Jespersen (1922), are, as we all know, important contributions to the area and, as such, were incorporated to more recent studies. One could say that, whereas for us, it is very easy to recognize, admit and deal with anecdotal1 data, it is not that easy to define the exact meaning of the adjective “anecdotal”, appositively to data or even to some specific occurrences. Therefore, I propose some research in dictionaries focusing on the item anecdote, from which anecdotal derives. In O Novo Dicionário Aurélio, I found the following entry for anecdote: (a)1. Relato sucinto de um fato jocoso ou curioso.2 Particularidade engraçada de figura histórica ou lendária. For the English anecdote, The new shorter Oxford English dictionary acknowledges basically the same senses, but in reverse order from than that in Aurélio: 1. I speak here as one of the members of the Language Acquisition Project, proposed and coordinated by Prof. Cláudia de Lemos (1982, 1992) at IEL, Unicamp. This project celebrated its 25th anniversary in 2002, when a seminar was held in honor of Cláudia Lemos. 2. Some of these data were analyzed in Figueira (2001a). A little before that, in 1999, they were the theme of a conference delivered at the opening of Enapol I, at USP.

158

ROSA ATTIÉ FIGUEIRA

(b)1. secret or hitherto unpublished details of history: 2. A narrative of an amusing or striking incident (orig. an item of gossip) (...) In French, Le nouveau petit Larousse. Dictionnaire de langue française also lists the two senses above for the item anecdote, but, it adds the acknowledgement of a third sense, which actually is held as the least important. (c) particularité historique, petit fait curieux dont le récit peut éclairer le dessous des choses, la psychologie des hommes. Récit d’un fait curieux ou pitoresque, historiette. Détail ou aspect sécondaire, sans généralisation et sans portée. Without settling on one specific sense, it is possible – from the very (anecdotal) data chosen to be examined in this paper – to show that the researcher’s decision to incorporate such data to his/her material, situates him/herself between the two basic senses listed in the dictionary: anecdotal meaning “funny, entertaining, picturesque, laughter-inducing,” and anecdotal meaning “unique, typical of the subject’s personal history.” As a result, instead of avoiding this duplicity, by choosing one of the uses of the word, I will deliberately take advantage of it, because embracing this slackening in meaning seems to be a providential and convenient prospect. In other words, it serves the purpose of making evident a fact to be underscored in the very dynamics of anecdotal data: many of the private or episodic facts, sometimes considered as secondary or peripheral with respect to a person’s history (and here I am referring to the child as a subject constituted by language) are eventually the target of laughter, being included in the family collection of anecdotes, i.e., they become a type of production ready to circulate in the domestic environment, and, when and if directed to a language researcher will turn into a precious finding, part of his/her corpora, as qualified as any other (or maybe even more qualified) to compose the explanation chart of the native language acquisition phenomenon. Some discoveries circulating among us, attributed to small children, will serve as point of departure. For example, after hearing her mother, who was attentively looking at some plants in the garden, say: Veja, a primavera está chegando!, the child remarked, between surprised and confused: – Não sabia que tinha uma prima chamada Vera! (a)

Similar to this utterance, which involves a divergent segmentation of the sound chain (primavera à prima Vera), there is one registered in Bonnet & Tamine-Gardes (1984: 51), which is worth transcribing: (Delphine 3;4 a cueilli des fleurs) X. Est-ce que tu sais comment s’appellent ces fleurs...ça s’appelle des primevères. D. Non, c’est des primes jaunes. (b)

HUMOR AND LANGUAGE ACQUISITION

159

In the dialogue above, Delphine rejects the term primevères and replaces it with primes jaunes - a sequence that, as the one in the previous dialogue, obliges you to carefully analyze how the mother’s speech is captured by the child. What is remarkable in this child’s utterance is the appearance of a feature which, in adult language is apparently not necessary or even pertinent - the color of the plant (jaune/yellow), which at the moment of her apprehension of the world and language, appears to be necessary and pertinent. (In fact, it is not for any other reason that this example fits in with creations related to referential motivation.) Consequently, the final segment vères, which, in adult speech, is by no means a morpheme, becomes one – an occurrence clearly fostered by homonymy (the similarity between the signifiers vères and vert3) showed in a datum which, unfortunately, is poorly explored by the authors, who do not seem to have been able to grasp the semiotic complexity of the child’s reply. This fact is not negligible, so much so that they had, at their conceptual and descriptive disposal, the notion of autonimy. Another anonymous finding, taken from what could be conceived as folklore concerning children’s utterances, comes from a child who, on opening the door for a gentleman and hearing that he was a tax collector, announced him to her mother as follows: (batem à porta; uma criança de três anos e meio vai atender; um homem se apresenta como sendo “o cobrador de impostos”; a criança vira-se para a mãe e diz) - Mãe, taí o impostor. ( c )

This example also has equivalents in the English and French literature of Language Acquisition, related to innumerable cases of new agentive items, triggering reactions from listeners ranging from light laughter to peals of laughter. Let us consider, for example, limacier, poubellier, creations which designate respectively, the one who kills snails (from the French limace), and the one who collects the garbage (from the French poubelle). The former was coined by a boy not identified by name, about whom Sully reports that, in order to dignify his task of getting rid of the snails in the garden, called himself a “limacier;” the latter, produced by Valérie, Aimard’s subject, who delights us with this and other countless innovations, of which “poubellier” is only an example.4 3. Homophony offers undeniable room for the play of words, which, in the case of vert, was registered as having been produced by more than one French child. Valérie’s beautiful example is taken from Aimard’s work: (en coloriand , V. prend un crayon de couleur). V. Je prends ce vert, Pi un verre où on boit, Pi des vers dans la terre, Pi des verts crayons. (Aimard 1995: 37). In another speech, it is herbette which makes room for a resemantization, whose effect is undoubtedly funny. Once again, the datum below is Valérie’s: (V. entend) M. Alors, je prends une petite herbette... V. Un petit air bête!!! - (Aimard 1975:33 – Observation: I replaced Moi for M, the initial for mother). 4. Observe that the example only acquires the status of vocabulary innovation in French, where the person who takes care of the garbage (poubelle) has a different designation. On the other hand, in Portuguese, in this case, the agentive derived item is formed from lixo: lixeiro, via the suffix –eiro. Below, in order to provide a wider variety of examples, are other creations for designating agents, collected from Bonnet & Tamine-Gardes: arrangeur: celui qui arrange (Guillaume 3;7); cremeur: mangeur de crême (Edmond: 3;8); tambourier: celui qui joue du tambour ( Charles 5;6); gymnasteur: celui qui fait de la gymnastique (Lev 5;11). (Bonnet & Tamine-Gardes 1984:95).

160


If we refer to the corpora on English acquisition, we will find two great examples, within the same category of agentives: typewriter e rainer. The former, quoted by Karmillof-Smith in the epigraph to the chapter “The child as a linguist,” from the book “Beyond modularity,” belongs to 3-year-old Yara. Y. What’s that? M. A typewriter. Y. No, you’re the typewriter, that’s a typewrite. ( d )

Taking the word as bearer of the agentive suffix – er, applicable to humans, the girl refutes or denies its application to the machine and addresses the suffix to the agent, the one who does the job, in a kind of utterance that has the typical structure of reply. Yara talks about people and things, but, indirectly, she talks about language, too. The second occurrence with the suffix -er is a finding of Sully’s, and refers to a boy (3;7.7) whose initial is C.. Hearing his father say that nobody could walk in the rain, he replied: C. The rainer can. ( e )

Asked by his father who that person was, the boy replied with a statement which had definition traits: A man who lives in the forest. Both utterances (d) and (e) are similar to (c) for triggering a process of word formation in language, whereas the example from Portuguese – impostor - shows, fortuitously, an extra ingredient, and maybe because of that, it is superior in its capability of making people laugh. Indeed, in the case of impostor, the new word (an agentive) coincides with a sign currently used in Portuguese, which conveys, however, one more meaning (according to the dictionary (cf. Houaiss): “person that deceives with false looks; trickster”). This fact makes the child’s speech an undoubtedly funny circumstance to the adult – unique, in fact, in its nature, for the child him/herself does not realize the potential for laughter that his/ her innovation entails.5 Against the background of these initial considerations, one can ask, “What is the participation of data of this kind in research on the acquisition of Portuguese as a native language?” One element of the answer has already been discussed in a recent paper I have written on the unusual marking of gender/sex, carried out with data of Diaries of Brazilian children (Figueira 2001b). As I try to demonstrate, the heuristic power of this type of material is clear in some realms of Language Acquisition such as gender, domain in which unusual changes in marks of masculine and feminine can be seen as reflecting the subject’s sexual identity and, as such, part of the process of the constitution of subjectivity, which testifies to the presence of the subject in language. Now, I proceed 5. For a characterization of the different condition of the child and adult in view of verbal games of a witty nature, see Figueira (2001a).


161

my studies on anecdotal data, with the backing of the existing material concerning the acquisition of English and French, which directly leads to the reexamination of two authors: Sully and Jespersen. As a psychologist, interested in children’s overall behavior, including language, Sully gave a warm welcome to this mode of register. His most important work, Studies of Childhood, was published in 1895. In 2000, the second edition came out with a long introduction written by Susan Sugarman and a biographical note written by Elizabeth Valentine. More than one century after the first publication, one can say that Sully’s work can still fill in the reader on information about the child’s world, evoking – as Sugarman points out in her introduction to the second edition – in a convincing manner and agreeable style, the top categories of human experience: play and imagination, thought and concepts, language, emotional life, and so on. How did Sully work? Right in the introduction, addressed to the new readers of Studies of childhood, Sugarman offers us the answer, making use of a passage written by the author himself. Professor Sully will be greatly obliged if parents or teachers of young children can supply him with facts bearing on the characteristics of the childish mind. What he especially desires is first-hand observations carried out on children during the first five or six years of life. In another part, she writes: Compiled from anecdotes Sully collected from both published sources and informants he solicited, and crafted into flowing and engaged prose, the books journeys through major categories of human experience that remain of interest today: play and imagination, thought and concepts, language, emotional life (specifically fear), morality and discipline, aesthetic sensibility and drawing. The book provides a natural history of development in each of these areas beginning in infancy and extending through early childhood, though it concentrates upon the toddler years. (Sully 1895: VII; my italics)

Within the prospect offered by historical research concerning the source “Diary”, I am especially interested in thoroughly surveying the material in the chapter of Studies of childhood dedicated to language: “The little linguist.” There, one can find examples of children of several nationalities (English, Italian, French), collected by lay persons, adults who, not less sensitive to linguistic facts than researchers themselves, made their children’s Diaries a rich and fertile depository of the most important phenomena of childhood: language emergence. This chapter makes available, in this fashion, a precious set of data to whoever wants to follow the kind of analysis put into effect in a period preceding the official birth of the area of Language Acquisition. From the wide range of language aspects observed by the author: sounds, invented words and constructions, I have selected the mastery of words, to proceed in the direction heralded by the data shown above in this paper. I will take up the innovations or neologisms (word inventions), looking at the contributions of yesterday (19th century and the first half of the 20th century) from today’s point of view.

162


Accordingly, and still with Sully, I have registered some examples of the naming of individual objects by the child, a phenomenon that would be better described as cases of “overextensions:” bow-wow, used not only for a specific dog, but for other dogs, paintings of dogs and, not seldom, for other things resembling dogs. As described by the author, such a phenomenon allows itself to be lightly touched by metaphorical extension. Coming to words which we call names we find that the child will often extend a recognition-sign from one object to a second, and to our thinking widely dissimilar objects through the discovery of some analogy. Such extension, moving rather along poetic lines than those of our logic classifications, is apt, as we have seen, to wear a quaint metaphorical aspect. A star, for example, looked at, I suppose, as a small bright spot, was called by one child an eye. The child M called the opal globe of a lighted lamp a “moon.” “Pin” was extended by another child to a crumb just picked up, a fly, and a caterpillar, and seemed to mean something little to be taken between the fingers. (...) Taine speaks of a child of one year who after first applying the word “fafer” (from “chemin de fer”) to railway engines went on to transfer it to a steaming coffee-pot and everything that hissed or smoked or made a noise. In these last illustrations we have plainly a rudimentary process of classification. Any point of likeness, provided it is of sufficient interest to strike the attention, may thus serve as a bias of childish classification. (Sully 1895: 162-3) In addition to this comment, Sully also leaves his contribution in another field well exploited by the literature in the area: the use of a lexical item for its opposite or correlative: yesterday for to-morrow, too little for too much (from Preyer’s boy), borrow for lend and learn for teach. At this point, my interest in the reading redoubled, reencountering, on Sully’s list, many examples analogous to or even the same as many found in the acquisition of Portuguese: the so-called “verbal changes” (ensinar for aprender, emprestar for devolver, etc 6). Let us see what he says: Such words as lend, teach, call up first the pictorial idea of an action in which two persons are seen to be concerned. But the exact nature of the relation, and the difference in its aspect as we start from the one or the other term, are not perceived. (Sully 1895: 165)

The excerpts above belong to the chapter “The little linguist,” but it is in another chapter that Sully presents what he calls “Excerpts of a father’s dairy,” where a comment on the funny/humorous effect, that the child’s creations produce on the adult who grasps it directly from the little ones’ mouths is not missing. We, thus, have the opportunity to get in touch not only with the linguistic fact but with the reaction it produces in the child’s interlocutor. There is little to note in the way of verbal invention. (...) His father asked him whether his toy-horse was tired, whereupon he answered: “No, I make him untired.” This leads off the 6. See in Figueira (1977), examples of similar “changes” registered in the corpus of Anamaria, between 2.8 and 3.2 years old.


163

writer to an abstruse logical discussion of “negative terms”, and how it comes about that we do not all of us talk in C’s fashion and say “untired,” “unfatigued.” Another quaint invention was the use of “think” as a noun. It was funny, writes the father, to hear him rejecting his sister’s statements by the contemptuous formula: “That’s only your thinks.” (The italics are mine.)

From Sully, I will go over to Jespersen, the Danish linguist, whose taste for data “collected live” is praised by Mattoso Câmara Jr. (1972). From his Language: it’s nature, development and origin (1922, 1st edition), I chose the example below, collected from the speech of 7-year-old Antoinette. In reply to somebody who addressed her a compliment: Eh, bien, je te félicite!

the girl said: Eh, bien, je ne te fais pas licite!

episode whose anecdotal outcome does not go unnoticed by observers of the child’s speech, be them researchers or lay persons. But, surely, it is among investigators of the area called Language Acquisition, officially born in the late (19)50s, that findings such as this one (and others, exposed more than one century ago), arouse greater interest, since they contribute to shed light on the relationship of the child with his/her native tongue. When it concerns vocabulary innovations, it would be suitable to say that, from my very first writings, my attention was strongly directed to them, starting from the acquisition of verbs with the prefix “des” (Figueira 1999). This is what makes me have a high estimate and admiration for the pearl utterance above (untired), capable of adding to other data of English acquisition from the study of M. Bowerman (1982)7: unhate, unclothes, unopen..., collected from the speech of her daughters, Eva and Christy. These innovations fit in side by side with other similar creations I have registered from the speech of two children learning to speak Portuguese as their native tongue: desabre, desfecha, dessai, desmuda, desmurcha. It can be stated that an equivalent attitude is found among other current researchers. Eve Clark, for example, in her book The lexicon in acquisition was not indifferent to the contributions of yesterday, giving heed not only to Sully’s, but Guillaume’s, Stern & Stern’s contributions, all of them forerunners of the studies of language acquisition, based on Diary data. In connection with this, let us look at the following chart, taken from the chapter Words for undoing actions, where creations of reverse verbs with the prefix dé-, in French, are listed. Its source? Several authors, from the beginning of the 20th 7. It is not inappropriate to note here that the majority of Bowerman’s findings (collected from his two daughters Christy and Eva) come from what seems to be the Diary of the two girls.

164


century, namely: Vinson (1915-6), Grégoire (1947) up to Aimard (1975) and François (1977) – a constellation of scholars who, taken as a whole, clearly reflect the value Clark attributed to the (precious) contribution of, on one hand, the diarists and, on the other hand, some current investigators. As a matter of fact, en passant, one can say that the frequency with which Diary data appear in the studies of lexical innovations is worth mentioning8 and shows the prestige of this source among current investigators: 1. Typical early innovative uses of French dé- for reversal (1) LA (2; O) déconstruire ‘to un-build’ (from construire ‘to build’ /adult démolir ‘to demolish, knock down’) (2) EA (2;6) dégrandir ‘to un-grow big’ (from grandir ‘to grow, enlarge’ /adult rétrécir ‘to shrink’) (3) GP (3;0) débagager ‘to un-baggage’ (from GP’s innovative bagager la voiture ‘to baggage the car’, meaning ‘to go on holiday’; said here of unloading the car, meaning ‘to come back from the holiday’) (4) V A (3; 3) désendormir ‘to un-fall asleep’ (from endormir ‘to fall asleep’/ adult réveiller ‘to wake up’) (5) V A (3; S) démonter ‘to un-climb (up)’ (from monter ‘to climb (up)’ / adult descendre ‘to climb down’) (6) V A (3; 5) déchauffer ‘to un-warm up’ (from chauffer ‘to warm’/ adult refroidir ‘to cool down’) (7) SF (3;6) défroidir ‘to un-cool’ (from froidir ‘to cool, make cold’/ adult rechauffer ‘to warm up (again)’) (8) SF (3; 6) débimer ‘to un-spoil’ (from abîmer ‘to spoil’, meaning ‘to make new again’) (9) V A (3; 6) décoincer ‘to un-wedge’ (from coincer ‘to wedge, stick’) (10) EV (3;6) déprocher ‘to un-approach’ (from approcher ‘to approach’/adult éloigner ‘to go away, depart’) (11) V A (3;9) détourner ‘to un-turn’ (from tourner ‘to turn around’ meaning ‘to go back in the opposite direction’) (12) EV (3; 10) débâtir ‘to un-build’ (from bâtir /adult démolir ‘to knock down, demolish’) (13) V A (4;0) désattacher ‘to un-attach’ (from attacher ‘to attach’ / adult détacher ‘to detach’) 8. The reason for this is quite clear. The neologism is obvious as a difference and offers itself to the common observer as the emergence of something new, even though it is built – after all – from processes known in the language. To remember Saussure, one should read what the master says about in-décorable (1916/1971:193).


165

(14) V A (4; I) désécarter ‘to un-spread out’ (from écarter ‘to spread out’; of a hand of cards) (15) V A (4: I) déprisonner ‘to un-prison’ (from prison ‘prison’ / adult libérer ‘to free’) (16) V A (4; 1) développer ‘to un-wrap up’ (from envelopper ‘to wrap (up)’ /adult ouvrir ‘to open, unwrap’) (17) EV (4;2) décorder ‘to un-arrange’ (from accorder ‘to arrange, organize’/ adult déranger ‘to disarrange’) (18) VA (4:3) désescabeauter ‘to un-stool’ (from escabeau ‘stool’; for adult descendre ‘to get down ) (19) CG (4;3) dégrasser ‘to un-fal’ (from gras ‘fat’/adult dégraisser ‘to remove fat (from meal)’) (20) VA (4;6) désorteiller ‘to un-toe’ (from orteil ‘big toe’, meaning ‘to cut toe-nails’) (21) LC (5;3) désorer ‘to un-like’ (from adorer ‘to like, adore’ / adult détester ‘to dislike’) (22) PM (6;0) dessoufler ‘to un-blow’ (from souffler ‘to blow’ / adult dégonfler ‘to deflate’) Source: Aimard 1975, Cohen 1969, Decroly 1932, François 1977, Grégoire 1947, MéressePolaert 1969, Vinson 1915-16. (Taken from Clark 1993: 235) Among the researchers that contributed to the creation of the chart above, Paule Aimard undoubtedly deserves special recognition. Referring to another work, Quand l’enfant parle du langage, by Bonnet & Tamine-Gardes, one can find, in the part designed for verbs with the prefix dé- (1984: 95), a list of examples, whose source massively points to a subject: Valérie. Who is Valérie? Paule Aimard’s main subject. The assured presence in this work, as well as in that of several other experts in the acquisition of French, of such a comprehensive and exuberant register, capable of spreading itself throughout the work of, not only Aimard, but other researchers, (which is certainly not a fortuitous coincidence), leads us to inquire about its source (or sources). Indeed, it is known that Valérie’s speech was extensively and carefully observed by Paule Aimard: in fact, she is her daughter. (It is not necessary to conceal this fact; on the contrary, it is even more interesting to mention this condition from the start, for we know that many of the old and new observers of the child’s speech, started out at home.) Well, then, it is with Valérie’s data that Aimard fills dozens of pages of her Les jeux de mots chez l’enfant e Les bébés de l’humour, to mention only two books by this author, an enthusiastic expert on the child’s speech, who, as I see

166


it, shows a very deep intimacy with data collected live and a refined taste for revealing what they can mean as a linguistic accomplishment rather than, one could say, a linguistic fact.9 Accordingly, the collection of data by the author along Valérie’s childhood was not unjustifiable, and such data go beyond her work and eventually compose a well articulated set of examples of anecdotal data, source tapped by several other authors, be they French, or not. Transcending the work of Aimard herself, they incorporate the work of others: Bonnet & Tamine-Gardes (1984), Quand l’enfant parle du langage; Tamine-Gardes (1982), “La comprehension des métaphores chez l’enfant”; Tamine-Gardes (1988-89), “Le jeune enfant et les jeux de langage”; Clark (1993), The lexicon in acquisition, who use it when seeking elements to structure their own work. In conclusion, the contents of my paper, made known in this (concise) presentation of anecdotal data, opens reassessment opportunities for such a methodological route, putting me in the position of defending, vis-à-vis the approaches to data in the area of Language Acquisition, the adoption of a double methodological procedure: recordings and diary registers. In doing so, one eventually takes up and values the work of those that, driven both by parental and professional interest, have preceded us by a few decades, bequeathing us rich and profuse material of children’s spontaneous speech. REFERENCES

Aimard, P. 1975. Les jeux de mots de l’enfant. Villeurbaine: Simep Editions. . 1988. Les bébés de l’humour. Liège: Mardaga. . 1995. “Premières parlettes”. G. Cahen, Le plaisir des mots. Paris: Ed. Autrement. 28-39. Authier-Revuz, J. 1995. Ces mots qui ne vont pas de soi. Boucles réflexives et noncoïncidences du dire. Paris: Larousse. Bonnet, C. & Tamine-Gardes, J. 1984. Quand l’enfant parle du langage. Connaissance et conscience du langage chez l’enfant. Bruxelle: Mardaga. Bowerman, M. 1982. Reorganizational processes in lexical and syntactic development. In E. Manner & L. R. Gleitman (eds), 320-346. Language acquisition: the state of the art. Cambridge: Cambridge University Press. Buarque de Holanda, A. 1986. Novo dicionário Aurélio da língua portuguesa. Rio de Janeiro: Nova Fronteira. 9. I am interested in contrasting the terms fact [fato], accomplishment / occurrence [feito] and deed [ façanha]. Although they are used in a close sense, one can, from my point of view, establish a difference with respect to the position of the investigator face to face the child’s speech. Considering the object as a fact implies destroying the lingusitsic production of his/ her entourage, applying, to the registered sequence, the formal instruments normally required within lingustic science. Considering it as an accomplishment or occurence implies going beyond and examining what this phenomenon represents in the suject’s history, and this very often leads to acknowledging the utterance as a deed in which language itself is at issue.


167

Clark, E. 1993. The lexicon in acquisition. Cambridge: Cambridge University Press. De Lemos, C. 1982. “Sobre aquisição de linguagem e seu dilema (pecado) original”. Boletim da Abralin 3. 97-136. De Lemos, C.1992. “Procesos metaforicos y metonímicos como mecanismos de cambio”. Substratum, 1, 121-135. Barcelona: Meldar. Figueira, R. A. 1977. “Áreas de dificuldade na aquisição do léxico”. Anais do II Encontro Nacional de Lingüística: 352-386. PUC-RJ. . 1984. “On the development of the expression of causativity: a syntactic hypothesis”. Journal of child language, 11: 107-127 Cambridge: Cambridge University Press. . 1996. “O erro como dado de eleição nos estudos de aquisição da linguagem”. In M. F. Castro (ed.), O método e o dado nos estudos da linguagem: 55-86. Campinas: Editora de Unicamp. . 1997. “Children’s riddles: what do they tell us about change in language acquisition?” Cadernos de estudos linguísticos 33: 15-26. 1999. “A Aquisição dos verbos prefixados por des-”. PaLavra 5: 190211. Rio de Janeiro: PUC. . 2000. “L’Acquisition du paradigme verbal du portugais. Les multiples directions des fautes”. CALAP 20: 45-64. . 2001a. “Dados anedóticos : Quando a fala da criança provoca o riso... Humor e aquisição da linguagem”. Línguas e instrumentos lingüísticos 6: 27-61. Campinas: Pontes. . 2001b. “Marcas insólitas da aquisição do gênero. Evidência do fato autonímico na língua e no discurso”. Lingüística, 13. ALFAL. . 2001c. “La Propriété réflexive du langage dans le parler de l’enfant: quelques aspects pragmatiques et discursifs”. Pragmatics in 2000. Selected Papers from the 7th International Pragmatics Conference: 207-221. IPrA. Antwerp, Belgium. . 2003. “La Propriété réflexive du langage: quelques manifestations du fait autonymique dans l’acquisition du langage”. Parler des mots. Le fait autonymique en discours. J. Authier-Revuz, M. Doury e S. Reboul-Touré (eds). Paris: Presses Sorbonne Nouvelle. Jespersen, O. 1964. It’s nature, development and origin. N.Y.: WW Norton & Company Inc. Karmiloff-Smith, A.1995. Beyond modularity. Londres & Cambridge: MIT Press. Larousse. 1969. Le Nouveau Petit Larousse. Paris: Larousse. Mattoso Câmara Jr, J. 1972. Dispersos. Rio de Janeiro: Fundação Getúlio Vargas. Oxford. New Shorter Oxford English Dictionary. Saussure, F. 1916/1971. Curso de lingüística geral. São Paulo: Cultrix. Sully, J. 2000. Studies of childhood. London/N.Y.: Free Association Books. 2nd. ed. (1st. ed. 1895). Tamine-Gardes, J. 1988-89. “Le jeune enfant et les jeux de langage”. SIGMA, 12-13: 255-273.

168

169

PART III

PLENARY PAPERS

170

171

REFLEXIONS SUR L’EXPERIENCE GRECQUE DU LANGAGE MARIA HELENA DE MOURA NEVES Universidade Estadual Paulista, CNPq Introduction Pour parler de l’expérience grecque du langage, nous devons certainement remonter à la poésie, et, pour traiter ce thème, nous devons faire abstraction de l’expérience que nous vivons dans le monde moderne, et ce n’est pas tâche facile. Aujourd’hui, nous lisons, nous sentons, nous apprécions Homère, Hésiode, les lyriques, les auteurs tragiques et, selon l’habitude analytique de notre modernité, nous ne voyons que ce qui est là, dans les mots des poèmes. Nous oublions qu’ils ont vécu grâce à un type de circulation qui n’est pas celui de la parole écrite. Récitée dans les palais, dans les réunions, dans les festivals, ou chantée au son d’instruments, la poésie promouvait par elle même la cohésion sociale du monde grec, ce qui a une importance grande, mais exige beaucoup d’effort pour se le représenter. C’est justement par son caractère oral, que la poésie grecque assume ce caractère de poíesis (terme lié au verbe grec poiéo, “faire”). La poésie instaure et crée les choses. Elle enveloppe non seulement son créateur mais aussi les chanteurs, et, également les auditeurs. C’est en tant que vécue intimement par l’auditeur qu’elle est poíesis. En réalité, ce que les grecs pensaient – comme on le voit plus tard dans la propre critique de Platon à la poésie – c’est ce qu’ils écoutaient des poètes1. Voilà pourquoi les grecs ont appelé la parole (épos) poésie. Les épe récités dans les festivals constituaient toute la configuration des choses. C’était par l’intermédiaire de la voix de la poésie, que les grecs vivaient leurs expériences et constituaient leur culture. Celle-ci résulte donc d’une longue et riche expérience vécue du langage2. Nous devons encore une fois prendre nos distances. Aujourd’hui, nous divisons en trois: homme/monde/langage, et nous avons l’homme avec sa place dans la réalité. Si nous proposons une étude théorique du langage, immédiatement nous viennent à la pensée des notions telles que signe, communication, représentation de la réalité, objectivité – et, en contre-partie, des notions vagues, considérées comme non scientifiques, que nous écartons du domaine linguistique. 1. Rappelons que c’ est ainsi bercés que les grecs s’acheminent vers la philosophie. 2. Nous savons que la philosophie a tué la tragédie, a relativisé le langage, mais sans annuler le poids de cette expérience vécue.

172

MARIA HELENA DE MOURA NEVES

Quel lien unit cette notion de langue, et toutes les questions théoriques qui l’appuient, avec l’expérience de vie du langage? Est-ce que nous avons les conditions pour sentir cette problématique? Non, certainement, car nous sommes un produit post-rhétoricien et post-philosophique. Nous éprouvons des difficultés à comprendre – comme les grecs le comprenaient – la parole comme création des choses. Le langage n’est plus une expérience qui constitue notre vie, mais une expérience qui lui est extérieure, au point que nous mettons en équation ses problèmes et nous nous disposons à les résoudre. Ce n’est plus comme expérience vécue que les questions linguistiques sont importantes pour nous, mais comme objet d’analyse, comme objet théorique. Nous concevons la réalité comme quelque chose de stable; d’autre part il y a l’effort constant de l’homme pour établir une relation avec les choses. Nous concevons donc une réalité extérieure à nous, que nous dominons plus ou moins bien, grâce à une culture accumulée au cours du temps. Et l’un des moyens grâce auquel nous dominons cette réalité extérieure, c’est justement le langage, le langage qui dénomme et qui dit les choses3. Différemment de ce qui se produisait chez Homère – où la poesie représente le langage en train de se faire – nous considérons le dire comme une activité qui contraste et même entre en conflit avec le faire. Nous établissons une hiérarchie – comme postplatoniciens que nous sommes – entre le dire et le faire, dans laquelle le faire tient la place d’honneur. Revenons donc au commencement de ce que nous connaissons de l’expérience grecque du langage. 1. L’expérience intuitive du langage et les premières réflexions sur l’activité linguistique 1.1 La poésie 1.1.1 Homère Chez Homère on ne remarque pas encore, une réflexion fixée sur le langage, quoiqu’il ne ressort pas de l’épopée cette conception selon laquelle le langage est une force réelle, matérielle, cette conception primitive d’un lien entre langage et réalité corporelle. Le contenu de l’épopée ce sont donc les actions humaines, (qui s’amplifient dans les actions divines). Mais parler (travailler avec la langue) et agir (travailler avec les membres) s’associent. Les actions s’articulent entre le dire et le faire; le genre épique lui-même reflète cette unité de la narrative avec le drame (du verbe dráo, “agir”): l’auteur des actions parle. Tant en ce qui concerne les héros qu’en ce qui concerne les dieux, le dire et le faire sont en parallélisme constant, parallélisme qui constitue l’essentiel à propos des dieux et des hommes. 3. Rappelons que le grec n’avait pas le terme réalité. Il avait le mot être, et c’était sur celuici qu’il jouait.

REFLEXIONS SUR L’EXPERIENCE GRECQUE DU LANGAGE

173

Qu’est-ce qui met en mouvement le héros de l’épopée homérique? Ce qu’il souhaite atteindre c’est l’excellence dans le conseil (par la parole) et dans la guerre (par l’action). L’idéal de l’éducateur apparaît chez Homère, dans les paroles de Phénix à Achille: Je devais t’apprendre à être en même temps un bon diseur d’avis, un bon faiseur d’exploits. (Iliade IX, v.443).

Le pouvoir est directement lié à l’action, mais à cette action la parole participe. C’est dans ce sens que Zeus apparaît dans une position privilégiée, car sa parole est identifiée avec l’agir. Au début du Chant II de l’Iliade, guidé par la parole persuasive de Zeus, Agamemnon a l’impression de préparer un piège, alors qu’en réalité, c’est lui qui est piégé. L’action de Zeus pour perdre les grecs consiste à parler en rêve à Agamemnon. Guidé par les paroles de Zeus, Agamemnon agit, mais son action consiste aussi à parler. Et son langage trompeur – comme avait été trompeur celui de Zeus, sans que le héros s’en aperçoive – provoque le discours d’Ulysse et celui de Nestor, destinés à conduire l’action des grecs. Zeus veut la fin des grecs: il va l’obtenir par un langage qui est une tromperie. Ce langage est efficace pour Agamemnon, mais à un autre niveau. Dans la mesure où il s’adresse aux anciens, trompé sans le savoir, Agamemnon exécute le dessein de Zeus. Ainsi, Agamemnon a l’illusion de tromper et adopte la technique de Zeus. Ses paroles provoquent l’action (des grecs), mais aussi la parole (dans les discours de Nestor et d’Ulysse). Dans la mesure où elle persuade, dans la mesure où elle suggère, la parole se prolonge dans l’action. La parole a une force persuasive dans l’exacte mesure où elle trompe. L’erreur apparaît comme un des éléments essentiels de l’activité de la parole, pas explicitement toutefois, ce qui représenterait déjà une séparation des plans, celui du faux et celui du vrai, ce qui n’est pas encore le cas. Le seul point de référence est au plan de l’efficacité. 1.1.2 Hésiode Chez Hésiode, il n’y a plus de narrative de faits mais une révélation d’un ordre qui apparaît dans le monde. Le contenu du discours, ce ne sont plus des actions, mais des notions. Il y a clairement une vision du monde qui s’organise en épisodes mythiques. On ne chante pas des hauts-faits, des prouesses, mais on célèbre l’excellence de Zeus (aristéia toû theoû), c’est-à-dire qu’il y a plutôt une qualification de l’agent qu’une narrative d’actions. Zeus est l’áristos (“l’excellent”). C’est à mesure qu’il consolide sa victoire sur l’informe et le terrible que le kósmos émerge. Hésiode fait de la poésie en invoquant les Muses. Filles de la mémoire, elles sont néanmoins “l’oubli des malheurs, la trêve aux soucis.” (Théogonie, v.55). Leurs “hymnes réjouissent le grand cœur de Zeus leur père” (Théogonie, v.37). Cataloguées comme des êtres créés (filles de Zeus et de la Mémoire), Hésiode les traite cependant comme des entités primordiales douées d’un langage poietique (“faiseur”), un langage qui manifeste l’être. Le langage revêt un caractère sacré, ce qui suggère fortement un caractère fondateur, un caractère instituteur du langage.

174


En vérité, les Muses ont une fonction technique bien précise: elles sont les divinités responsables de ce que disent non seulement les poètes mais aussi les rois. Chez Hésiode il y a, déjà, un souci de vérité, ce qui n’existait pas chez Homère. Et la relation entre la divinité et l’homme est changée. Il apparaît bien dans la Théogonie – comme nous l’avons vu apparaître dans l’Iliade – un affrontement qui se construit sur l’erreur: ici aussi s’affrontent Zeus et Prométhée. Mais la relation présente une configuration bien différente. Chez Hésiode, l’erreur se produit surtout à travers des actes. Hésiode ne se réfère pas, dans l’épisode de Prométhée, à l’institution du langage, tandis qu’Homère associe la parole à l’erreur d’Agamemnon. Chez Hésiode, l’intériorité semble en relation avec le silence, alors que chez Homère elle s’associe à la parole. L’action directe est l’élément qui leurre; il n’y a pas de paroles trompeuses, mais Zeus sait qu’il y a ruse, et l’accepte, d’autant plus que grâce à celle-ci il médite des maux pour les mortels. Prométhée a trompé, mais Zeus le savait. Et ce que fait Prométhée, en réalité, c’est ce qu’il fallait faire: les dieux ne doivent pas manger de viande, elle appartient aux hommes car elle est sujette à la putréfaction, et la partie divine est ce qui a une odeur agréable. Cette étiologie est chez Hésiode: Et aussi bien est-ce pourquoi, sur la terre, les fils des hommes brûlent aux Immortels les os nus des victimes sur les autels odorants. (Théogonie, v.556-557)

Zeus assume la ruse de Prométhée pour en forger une autre, et, grâce à celle-ci tracer le destin des hommes. Conclusion: il est impossible de dérober l’intelligence de Zeus. Enfin, chez Hésiode, il n’y a pas simplement la caractérisation du chef des dieux, par la réunion en lui de la plénitude du dire et du faire. Ce qu’on voit c’est la configuration d’une espèce d’épiphanie de l’ordre du monde (kósmos). 1.1.3. Poésie lyrique: Pindare Faisons un saut jusqu’à Pindare, poète lyrique qui élabore l’éloge des hommes vainqueurs. Ce n’est pas une simple association – comme il y en avait une chez Homère – entre le dire et le faire, mais, dès que ce dire a un sens par rapport au faire, on dirait que commence une dissociation qui conduit à l’autonomie du dire. Le lógos est le dit des héros; il a donc leur sens à eux. Il peut arriver que les poètes donnent à des mortels une gloire indue, comme Homère à Ulysse. Pindare affirme que ce que l’on dit d’Ulysse (son lógos) est au-dessus de ses épreuves (son páthos): Cependant, j’imagine que la renommée d’Ulysse a dépassé ses épreuves, grâce au charme d’Homère. (Néméenes VII:20-21).

Il y a une dichotomie en jeu: le lógos (ce que l’on dit du héros) et l’érgon (l’œuvre, l’activité). Et l’activité d’un héros, est, dans le fond, un páthos, c’est ce qu’il vit, ce


175

qu’il ressent, ce qu’il souffre, l’épreuve. Ulysse alors, grâce à Homère (dit Pindare), a une gloire plus grande que celle correspondant à son páthos. C’est une déclaration explicite de la conscience de ce que serait l’idéal guerrier: l’idéal est la gloire du fait et du lógos, ce qui montre que l’exploit (et de là, l’excellence, l’aristeía) sans le lógos n’est rien. Les dispensateurs de la vie éternelle sont les poètes, car, sans la parole, l’action s’efface dans l’oubli. La juste mesure du dire du poète, néanmoins, est dispensée par la divinité. Chez Pindare, pourtant, il n’existe plus cette association entre la belle parole et l’exploit excellent qui existait chez Homère. Il s’établit un désajustement entre la parole et l’action: la parole peut dire plus que l’action, plus que l’exploit. De sorte qu’en plus de pratiquer de belles actions, les mortels doivent rencontrer le poète qui perpétue leur gloire. Au poète, il revient de faire croître ou diminuer les gloires, selon la mesure de l’inspiration divine. La parole complète l’œuvre car c’est par la parole que la gloire s’installe et se défend de la chute dans le néant. La parole du poète inspiré par le dieu est un lógos vivant qui vient au secours des morts. 1.1.4. La tragédie Dans la tragédie, le lógos du héros, ce qu’on dit de lui, s’associe à son nom. Le nom et la perception s’opposent: le nom, avec toutes ses possibilités d’interprétation, est la place de la vérité, contre la perception, qui est l’occasion de l’erreur. La poésie tragique parle toujours d’un nom exact, véridique, qui, néanmoins est un indice de contradiction. Il suffit d’observer les noms des héros tragiques. Dans l’onomastique des héros réside la vérité de leur destin. Lisons ce passage de Prométhée Enchaîné (Eschyle, Sophocle 1960:25), qui dit: Les déités te nomment d’un faux nom, prévoyant Prométhée: il te faut un Prométhée qui te dénoue toi-même de cet engin. (v. 85-87)

Rappelons-nous Oedipe-Roi – Oidypos Týrannos (“le pied enflé” c’est ce que signifie Œdipe), nom qui symbolise la faiblesse, le manque de fermeté dans l’orientation. Dans toute la pièce, on rencontre le jeu des deux concepts: au commencement, il est le roi, le sauveur, le père et le seigneur du peuple; à la fin il est un oedipe, un “pied enflé”. Chez Eschyle (Prométhée Enchaîné), Zeus et Prométhée de nouveau s’affrontent. Ici il n’y a pas de tromperie. Prométhée connaît un secret de Zeus (il sait que Zeus ne peut pas se marier pour ne pas se perdre), et grâce à cela, il est sauvé. Prométhée est un représentant de la parole (rappelons-nous que chez Eschyle, il apparaît comme celui qui, en dérobant le feu à Zeus, donne la parole aux hommes). Zeus est le représentant du pouvoir de l’action. Pour cette raison, à la fin de la trilogie, tous deux s’accordent: Zeus, celui qui a le pouvoir; Prométhée celui qui a la parole (parce qu’il possède le secret de Zeus mais ne le révèle pas).

176


Il n’y a pas d’association entre la tromperie et la parole. Au contraire, Prométhée dit qu’il connaît le mot, mais il le cache (ce qu’il y a, c’est un secret). Comme dénominateur commun, il y a l’idée de cacher quelque chose. 2. Le début de la philosophie Cette époque des premiers auteurs tragiques est aussi celle d’Héraclite et de Parménide, époque où commence à exister une autre façon de considérer le lógos. C’est la philosophie. Nous rencontrons chez Héraclite et Parménide la notion que, s’il y a un désaccord entre le dire et l’agir, s’il y a un lógos qui peut être plus grand que l’action, c’est peut-être qu’il y a un mot qui ne correspond pas à l’action, un mot vide, privé de sens, un dire qui n’est rien. Ce qui importe ce n’est plus simplement le dire ou le faire, ce qui importe c’est le sens de ce dire et de ce faire. Le noûs, l’intériorité, l’intelligence de ce qui se dit et de ce qui se fait s’imposent: Il ne faut ni agir ni parler comme des dormeurs. (Héraclite, Fragm.73).

Il y a une profondeur qui caractérise le lógos: On ne peut trouver les limites de l’âme, quelque chemin qu’on emprunte, tellement elles sont profondément enfoncées. (Héraclite, Fragm. 45).

Ainsi, le schéma dire/faire qui vient depuis Homère évolue. Depuis Homère, vérité et fausseté en accompagnent l’élaboration. Homère ne séparait pas la vérité de l’erreur et Hésiode n’avait fait qu’énoncer, par la voix des Muses divines, l’opposition entre vrai et faux. Dans le schéma d’Héraclite, être un sage c’est dire la vérité et agir selon la nature: La pensée est la plus haute vertu; et la sagesse consiste à dire des choses vraies et à agir selon la nature, en écoutant sa voix. (Héraclite, Fragm. 112).

Le discours philosophique naît pour articuler la totalité des choses. La vérité se place maintenant dans la relation entre le langage et les choses, ce qui implique une dissociation qui permettra de prendre le langage comme objet d’investigation. Alors se séparent le lógos (ce qui se dit) et la léxis (comment on le dit). Le métalangage est institué. Etre théorique est une caractéristique du langage philosophique, qui se forme comme par une espèce de contraste avec le langage poétique. Et, si nous passons à la philosophie, nous voici à la fin de ce bref parcours sur des siècles de vécu intuitif très riche des problèmes du langage, représentés dans le langage poétique des grecs. Les poètes, à des degrés divers, ont eu conscience du caractère de base qu’a le langage. C’est la propre expérience poétique du langage qui rend possible la


177

philosophie, qui vient justement faire le discours critique de la poésie; une expérience poétique riche, intense et profonde qui a équilibré le dire et le faire, qui les a dissociés, qui les a pesés et évalués, qui les a articulés avec l’erreur, la vérité, la fausseté, et qui a découvert l’intériorité (le noûs), une expérience qui a dilacéré l’identité de l’homme en antinomies, qui a créé le tragique et qui nous a laissé, à nous, humains, ce legs de glorification, que nous lisons dans le chœur d’Antigone: Qu’il est de merveilles! mais rien qui soit plus merveilleux que l’homme. (v.332-333)

3.

Le langage comme objet de réflexion théorique

3.1. Introduction Le fait que le langage soit une capacité naturelle de l’homme n’est pas – comme il pourrait le sembler à certains – une découverte de la science, mais constituait déjà l’un des piliers de la théorie aristotélique qui a conduit à une théorie de la signification. 3.2. Platon Avant Aristote nous avons Platon pour qui le nom est, dans une première instance, le lógos de la chose (Cratile), mais, au point d’arrivée des réflexions, c’est le lógos du discours, c’est-à-dire des relations entre les choses (Sophiste), Toute l’argumentation socratique autour de l’établissement du statut du nom suppose qu’il existe avec celui-ci une relation de ressemblance avec la chose nommée. Cette relation, qui est la garantie d’un degré de phýsis (“nature”) du nom, est de ce fait une considération que le nom, d’une certaine manière, dit la chose, et par conséquent, est son lógos. L’instance subséquente, qui est dans le Sophiste, est l’instance proprement logique; Platon ne va déjà plus chercher l’adéquation dans les termes, mais dans l’articulation de ceux-ci, dont l’entrecroisement reflète un accord existant entre les espèces. Dans ce lógos, qui est la communion de genres, il ne convient plus de parler de convention. C’est une instance où l’on quitte la simple dénomination, laquelle exige un certain degré de convention, et où l’on se dirige vers l’articulation des parties de la proposition, révélatrice de l’articulation entre les essences. Cela représente déjà la base d’une théorie de la signification. 3.3. Aristote Chez Aristote, le langage est une capacité biologiquement naturelle et un exercice politique, assurant, par l’intellect de l’être humain, la référence du langage aux choses:

178


il y a un langage qui est le discours de l’être, c’est-à-dire, qui dit vérité et erreur (le lógos) et un langage pratique, d’exercice politique, qui est le discours rhétorique. Comme l’objet de la philosophie n’est pas proprement le langage, lequel n’intéresse que parce que c’est à travers lui que les relations entre les êtres (ousía, gígnomai) se révèlent, dans la philosophie le discours pratique, rhétorique, n’est pris que comme support pour la réflexion focalisée, qui appartient au monde des idées, au domaine du signifié (logique). L’essence de la réflexion d’Aristote sur le langage se trouve dans les œuvres où il se penche sur les catégories (qui comprennent la correspondance entre la structure de la langue et la structure du monde) et sur la proposition (qui révèle la nature, la phýsis). Dans les noms il y a une signification, mais celle-ci est le résultat d’un accord, d’une convention, elle n’est pas naturelle: l’évidence est que les états d’âme sont identiques pour tous les hommes, parce que ce sont des images (homoiómata) des choses, mais les mots, ne ressemblant pas aux choses, ne sont pas les mêmes pour tous: la relation entre les mots et les états d’âme n’est qu’une relation de signification, elle est symbolique, non naturelle, conventionnelle (katà synthéken). Les noms sont des symboles (forme et contenu qui sont à la place des choses, puisque nous ne pouvons pas nous servir des choses dans le langage). Ainsi, en eux-mêmes, les noms ne sont ni vrais ni faux, de même qu’un concept n’est en soi ni vrai ni faux. Ce n’est que dans la liaison des concepts que se révèle ce qui est ou n’est pas, c’est-à-dire la vérité ou la fausseté. Ainsi, dans la proposition, il y a plus qu’une signification, il y a une révélation de ce que les choses sont ou ne sont pas, il y a une “manifestation” (apóphansis). Tout ceci signifie qu’Aristote sépare déjà les deux questions: d’une part la justesse de la dénomination(orthótes), à savoir, la relation entre la forme sonore(le nom) et la chose désignée (qui est conventionnelle, symbolique, et qui par conséquent, comme pour Platon, n’est pas le point focal de la recherche), et, d’autre part, la vérité ou la fausseté de la relation entre les êtres, à savoir, l’affirmation ou la négation, que la proposition (la relation être ou ne pas être) révèle. De ce côté-là, il n’y a rien à chercher, simplement, choisir entre phýsis (“nature”) ou nómos (“convention”), dans la vérification des relations entre le langage et ce qu’il dit. 4. Les stoïciens Mais c’est avec les stoïciens que je vais couronner ce parcours que je fais ici des relations entre le langage et les choses, entre ce qui signifie et ce qui est signifié dans l’histoire de la pensée hellénique et hellénistique. La logique stoïcienne, qui n’est plus une logique de termes, mais une logique d’énoncés (spécifiquement, de prédicats), place les événements (et non les êtres individuels)comme objet de la pensée. L’unité d’examen – l’objet de la logique stoïcienne – est le lektón (“ce qui est dit”, le contenu de l’acte matériel de dire, en particulier la signification d’une assertion à laquelle s’appliquent les catégories du vrai et du faux, enfin, ce qui est signifié quand on parle avec sens).


179

Etre vrai ou faux est une propriété des lektá, non pas de tous, mais, en particulier, de la proposition, qui est un des lektá complets, un des axiómata. Pour les stoïciens le langage est naturel, il est un produit de la nature, ainsi que les représentations de valeur générale qui se créent elles-mêmes chez les hommes, et, de cette façon, sont communes à tous les individus. A partir de là, toute théorie stoïcienne du signe appuie la recherche de la justesse et de la vérité dans le langage, ce qui n’est autre que la recherche de son caractère naturel. Dans une direction théorique, on va vers des spéculations étymologiques. Dans une direction pratique, on part à la poursuite des hellénismes en vue de constituer des paradigmes. Et sur ces deux fronts, théorie et pratique, se pose avec évidence une question qui doit être dominante dans la formation de la discipline grammaticale: la dichotomie analogie/anomalie. Le principe de l’analogie, qui n’est rien d’autre que le reflet de la capacité intellective humaine d’établir des relations et des associations, a toujours été reconnu dans l’activité linguistique. Au fond, c’était aussi ce qui était en jeu dès cette expérience intuitive du langage (avec la relation ambiguë entre vérité et erreur, dont j’ai traité dans la première Partie de cette présentation) et les premières réflexions sur le langage (avec la recherche délibérée de la relation entre langage et vérité, dont j’ai parlé dans la deuxième Partie). Considerations finales: Un angle de vision historique dans l’investigation sur le langage Fixons-nous, maintenant, dans notre temps, dans le domaine temporel de la science linguistique, et, à partir de notre présent, regardons dans la direction inverse, retournant à une histoire de plus de vingt siècles au cours de laquelle la proposition de cette dichotomie analogie/anomalie a accompagné les considérations sur le “phénomène” du langage (en prenant le terme “phénomène” dans son sens étymologique: “quelque chose qui se manifeste”). Ce que nous allons voir, dans l’histoire des études du langage – unissant cette troisième Partie aux première et deuxième Parties – c’est que cette dichotomie, discutée selon des présuppositions et des directions très différentes, s’est toujours révélée comme déterminante pour considérer les relations grammaticales, la grammaire étant vue comme la propre organisation de l’activité de langage. Je vais choisir quatre moments de la pensée linguistique en partant de la modernité de Saussure pour revenir à la tension fondatrice des stoïciens, point où je m’étais arrêtée dans l’examen de la pensée grecque sur le langage que j’ai présenté ici. Mes réflexions seront guidées par quatre centres d’intérêt dans la controverse entre analogie et anomalie:

180


a) la conceptualisation; b) le terrain de l’examen; c) l’attitude d’analyse; d) la finalité de l’invocation du concept. 1) Chez Saussure, en ce qui concerne la conceptualisation, l’analogie s’oppose à l’usage, dans le sens où celui-ci est un mécanisme de création de formes nouvelles. Quant au terrain où la question se situe, il s’agit de la synchronie. C’est une attitude d’analyse non valorative, une simple vérification de symétrie dans la langue. En ce qui concerne la finalité de l’invocation du concept, ce qui est en jeu, c’est l’accentuation de la dichotomie langue/parole. 2) Chez les néogrammairiens, analogie s’oppose également à usage, mais dans le sens où il s’agit d’un mécanisme producteur de groupement de formes divergentes par rapport aux lois de formation primitives. Elle se situe sur le terrain de la grammaire diachronique. L’analyse est également non valorative, c’est une simple vérification de divergence dans la chaîne évolutive. Quant à la finalité de l’invocation du concept, ce que l’on cherche en premier lieu, c’est à expliquer les exceptions aux “lois” phonétiques. 3) Arrivant, dans ce recul, au seuil de l’institution de la discipline grammaticale, chez les grammairiens alexandrins analogie s’oppose de façon dichotomique à anomalie, qui est considérée comme “in-conformité”, “ir-régularité” (mais qui, au fond, est liée aussi à l’usage). Le terrain est la synchronie. Toutefois, l’attitude d’analyse est valorative, étant donné que les formes analogiques sont celles prescrites comme standard. Et quant à la finalité de l’invocation du concept, ce que l’on recherche c’est établir des paradigmes. 4) Chez les stoïciens, finalement – et nous nous trouvons de retour à la philosophie dans cette régression – analogie s’oppose à anomalie – celle-ci étant vue comme “contrariant la nature” (mais détectée dans l’usage). Le terrain est clairement celui de la logique. L’attitude est nettement valorative, puisque les anomalies sont indiquées comme un éloignement du naturel, et par conséquent du vrai. Quant à la finalité de l’invocation du concept, ce que l’on veut prouver, c’est exactement le caractère naturel du langage. Et c’est ici que réside le grand nœud de cette controverse qui a gouverné l’institution de la grammaire en Occident, ce qui a fait que, orientée théoriquement par la philosophie, cette discipline s’est instituée sur tout ce que la réflexion philosophique cherchait justement à “disqualifier”: ce qu’on découvre, c’est que la logique ne gouverne pas le langage et que le langage n’est pas un instrument d’expression de la vérité. D’abord objet de discussion, la relation du mot avec l’objet (la dichotomie phýsei/ nómoi), n’est déjà plus prise au sérieux chez Platon, car Cratile conclut à la fausseté du problème en soi, puis, Aristote démontre irréfutablement le caractère conventionnel des noms. La philosophie classique a établi la prééminence du lógos (de la proposition) sur l’ónoma (le nom) – déjà dans le Sophiste de Platon – pour arriver, chez Aristote à nommer la proposition (l’expression de l’être ou n’être pas) comme apóphansis, comme


181

“révélation”, comme “manifestation”, dans cette conduite typiquement philosophique qui consiste à considérer le langage comme rien de plus qu’un chemin à parcourir- et indispensable pour parvenir à la révélation des choses, le dire étant considéré comme quelque chose de toujours inférieur à ce qui est affirmé ou nié (Platon, Lettre VII). La tension entre analogie et anomalie, qui s’instaure plus tard dans la logique stoïcienne (à la frontière de l’institution de la grammaire), représente le point culminant des réflexions qui, à partir de la pure et simple distinction entre le langage et la réalité ont pu placer le langage comme objet de réflexion, et au service du vrai et du juste. Dans l’esprit de la philosophie, on admettait de partir du langage pour arriver aux choses, à l’autó. Dans une logique de prédicats comme la logique stoïcienne, cette assomption au départ, ne pouvait culminer, que par la considération qu’à une régularité et à une ressemblance entre les objets devrait correspondre une régularité et une ressemblance dans leur expression. Que cherchaient donc les stoïciens dans le langage? La conformité avec la nature, et par conséquent l’analogie. Et qu’ont-ils trouvé? L’anomalie. Il ne s’agit pas, loin de là, d’une fin malheureuse au parcours que nous avons essayé de retrouver dans l’histoire de la pensée grecque sur le langage, car ce qui est évident, c’est que les stoïciens euxmêmes ont découvert qu’on ne peut pas garantir la congruence entre les exigences de la dialectique et les formes d’expression linguistique: le mot, selon son contenu et ses relations, ne correspond pas exactement au concept et aux relations dialectiques de celui-ci. A la régularité universelle ne doit pas correspondre nécessairement la régularité linguistique. Logique et grammaire sont définitivement des domaines entre lesquels il n’y a pas à chercher de symétrie. Si à ce moment-là, l’évolution qui conduisait à ce manque de congruence était vu comme une perversion parce qu’il représentait un éloignement de ce qui était tenu pour naturel – et pour cette raison l’étymologie et le témoignage des anciens étaient tellement valorisés – malgré cela la tension créée dans l’institution de la dichotomie n’en a pas moins été une avance cruciale dans le progrès des réflexions sur le langage, au point que nous pouvons dire que ce sont ces réflexions qui constituent l’assise du principe fondamental: celui qui admet qu’il n’y a pas nécessairement de lógos (la ratio latine, la “logique”, la “rationalité”) dans le lógos (l’oratio latine, l’“expression linguistique”) et que le langage est en soi un objet d’investigation. C’est ainsi que, la science linguistique ayant été instituée comme nous l’avons vu, le propre concept d’analogie a pu ressurgir comme point focal d’investigation, et a pu être mis en confrontation avec l’usage, sans que ce dernier, de manière synchronique aussi bien que diachronique, représente un éloignement du naturel, donc sans que la dichotomie mette en jeu une détermination rigide entre le processus cognitif et la production linguistique – et même sans que l’invocation du concept d’analogie n’embrouille les relations entre langage et “réalité”, après tout ce long et remarquable effort de pensée qui a conduit à la compréhension de ces deux

182


“réalités”, par rapport auxquelles nous menons nécessairement – tous et toujours – nos expériences de vie. RÉFÉRENCES

Eschyle, Sophocle. 1960. Tragiques Grecs. Traduction par Jean Grosjean. Fragments traduits par Raphaël Dreyfus. Introduction et notes par Raphaël Dreyfus. Paris, Librairie Gallimard. Hesiode. 1972. Théogonie Les travaux et les jours Le bouclier. Texte établi et traduit par Paul Mazon. Paris, Les Belles Lettres. Homere. Odyssée. 1955. Traduction, introduction et notes de Jean Bérard. Paris: Gallimard. . 1955. Iliade. Traduction, introduction et notes de Robert Flacelière. Paris: Gallimard. Les Penseurs Grecs Avant Socrate. 1964. De Thalès de Milet à Prodicos. Traduction, introduction et notes par Jean Voilquin. Paris: Guarnier-Flammarion. Pindare. 1967. Néméennes. Texte établi et traduit par Aimé Puech. 4ème. édition revue et corrigée. Paris: Les Belles Lettres. Platon. 1960. Le sophiste. Texte établi et traduit par A. Diès. 5ème. éd. Paris: Les Belles Lettres, 1969. . Lettres. Texte établi et traduit par Joseph Suilhé. 3ème. éd. Paris: Les Belles Lettres.

183

JOHANN JACOB REISKE (1716-1774) LEADING FORCE IN THE ESTABLISHMENT OF ORIENTAL AND CLASSICAL SCHOLARSHIP IN GERMANY

KURT R. JANKOWSKY Georgetown University Oriental studies in Germany – just like the study of Classical languages – have a long tradition within the larger field of Biblical Studies. The Christian church from early on – and long before such studies had begun in Germany after the foundation of the Holy Roman Empire of the German Nation in 962 – had devoted extraordinary resources to the elaborate exploration of the three Holy Languages: Latin, Greek, and Hebrew. For centuries, the study of these languages was pursued within a framework which of necessity took its lead from the needs of Holy Scripture. And it did a great deal of good, first and foremost of course to the three biblical languages themselves, but also to what initially was by no means the main thrust of the research effort: laying the foundation for an independent, self-sufficient study of Oriental languages. Even Classical language study needed a fresh start from what it was like during the 15th and 16th centuries when it served almost exclusively the objectives of the Church. The following assessment of Hugh Lloyd-Jones (Introduction to Wilamowitz 1982: viii) proves the point: “The scholarly interests of the greatest figure of the late Renaissance, [Desiderius] Erasmus [1469–1536], centered upon the Bible and the Christian Fathers.” The teaching of Latin and Greek in the subsequent century and a half, in spite of the devastating effects of the Thirty-Year War (1618–1648), increased significantly and managed to attain a modicum of standards. But much substantive work had to be accomplished before Classical studies could be expected to be launched as an academic discipline in its own right. For a long period of time, direction, objectives, and methodologies were largely chosen at random. The desperately needed guidelines were finally provided by the English scholar Richard Bentley (1662–1742). The sum total of the issues Bentley raised in various programmatic publications (cf., e.g., Bentley 1691 and 1697) amounted to the demand that, in order to do full justice to the writings produced by an ancient scholar or poet, the modern scholar would have to make an all-out effort to establish and reconstruct the entire context for the work in question. Such an endeavor would not only involve acquiring an exquisite amount of knowledge of all aspects of grammar and style, of metrical phenomena and their relationship to grammatical features, as well as historical

184

KURT R. JANKOWSKY

facts pertaining to the period investigated, but would also require the ability, and the courage, to emend and reconstruct by reasoned conjectural work where the text material was suspected to be faulty or was in need of supplementation. Bentley’s venture for the first time did establish Classical scholarship as an independent science, as a discipline which was no longer ancillary to other fields, such as theology or philosophy. Since Bentley’s demands still had to be adopted and implemented by his fellow-researchers in England and in other countries throughout Europe, his envisioned mission was far from being completed by the mere proclamation of his ambitious program. But the first step had now been taken – and a very decisive step at that. Universities in Germany were by no means ready to lend all-out support to the establishment of chairs in Latin and Greek. But at least the tendency was there to rise to the challenge, and it grew steadily. Invaluable help came occasionally from some researchers in Classical studies who produced superb editions of Greek and Latin texts and wrote inspiring methodological treatises without being affiliated to a university, that is, without receiving the appropriate institutional support. One such researcher was Johann Jacob Reiske (1716–1774). His devotion to Classical scholarship is unique in also this respect that, in addition to being rightfully called “one of the first, if not the first, Greek scholar of the 18th century” (Foerster 1889a:131), he is, furthermore, credited with being “truly the founder of Arabic philology” (ibid. –Translations from German here and elsewhere are mine: KRJ). With the unconditional support of his wife Ernestine Christine (1735–1798) whom he married in 1764, but with hardly any other help from the outside world, he accomplished to produce an amazing amount of publications which proved to be of extraordinary merit in the field of Classical as well as Arabic scholarship. Reiske collected accolades from only few of his contemporaries, but their voices count a great deal because they belonged to people of prestigious position in the society of that time. More important, however, are the assessments by his peers of a later time. Rudolf Pfeiffer (1869-1979), for instance, comparing Reiske with Christian Gottlieb Ernesti (1756–1802), Prof. at the University of Leipzig at a time when Reiske attempted in vain to get a professorship there, calls Ernesti “industrious and solid”, but found Reiske an “incomparably greater scholar” (Pfeiffer 1976:172). Ulrich von Wilamowitz is much more outspoken. Referring to the theologian Johann August Ernesti (1707–1781), predecessor to his nephew Christian Gottlieb as chair holder, he finds that “his pedestrian mind was incapable of original, let alone profound, thought”. Contrasting the two, his preference is clearly with Reiske, “the latchet of whose shoes he [Ernesti] was not worthy to unloose” (Wilamowitz 1982: 94). Even an adversary like Johann David Michaelis (1717–1791) refers to Reiske as “a man whose assets of knowledge could perhaps be shared by ten people, and each might be... a great scholar” (cf. Brekle 1992:164). As to Reiske’s achievements in Arabic studies, we should take note of Johann Fück (1894–1974) who attributes to Reiske that Arabic philology at long last came of age.

JOHANN JACOB REISKE (1716-1774)

185

He calls Reiske outright brilliant, being “the first substantial Arabist that Germany had produced” (Fück 1944:192). Reiske is a most unlikely candidate for greatness as a scholar in general and as a Classicist or Arabist in particular, if one looks for the usual criteria which one expects in a candidate most likely to succeed. But, then, he was a highly unusual person, and that pertains above all to his insatiable thirst for digging into manuscripts in Latin, Greek, and Arabic, some of which were unreadable even for those who were acknowledged masters in the field. Reiske entered the University of Leipzig in 1733, at the age of 17. Although his main field of study was theology, he attended classes neither in this nor in any other subject, but devoted, instead, all of his time in independent study to continue learning Greek and Hebrew, as well as to take up a new subject, Arabic. After five years, in 1738, he left Leipzig for Leiden, hoping to increase there and in Amsterdam his knowledge of Arabic with the help of manuscripts available omly there. While he was forced to work many hours for his livelihood, mostly by copying manuscripts and offering private lessons, he took courses in Arabic with Albert Schultens (1686–1750) and in Greek with Tiberius Hemsterhuys (1885–1766). His hopes to secure a university appointment in Arabic were nipped in the bud since Schultens entertained career aspirations for his son and did not want his star pupil to be on the scene as a competitor for the position which he had earmarked for his son. He therefore persuaded Reiske to change fields and study medicine instead of pursuing a degree in Arabic. It was, if nothing else, for Reiske a mere waste of time, even though his doctoral thesis of 1746, Miscellaneae aliquot observationes medicae ex Arabum monumentis, was related to one of his two major fields of interest, Arabic studies. He returned to Leipzig immediately upon graduation, apparently without any prospect for adequate employment. His appointment to an extra-ordinary professorship in Arabic in 1748 brought no relief, since the remuneration was minimal and after some years ceased to be paid altogether. Even his election to principal of the Nicolai School in 1858, while a true life-saver for him, did not substantially improve his economic situation, since the administrative work severely curtailed his time for research. In spite of overwhelming adversities his enormous productivity never abated. In Arabic studies he created an innovative approach in that he focused on evaluating Arabic as a language and culture in its own right, not being motivated primarily by linguistic or theological concerns. His extraordinary status of having mastered two scientific areas, Arabic and Classical studies, and being hailed as “the first Arabist and one of the first, if not the first Greek scholar of the 18th century”(Foerster 1889a:131), is acknowledged by a number of distinguished personalities. Johann Gottfried von Herder (1744–1803), for instance, states (Herder 1785–1792.4:265A): “Our Reiske has become a martyr of his Arabic-Greek enthusiasm... But his scorned learnedness will certainly not come again in a long time.”

186

KURT R. JANKOWSKY

For Theodor Mommsen (1817–1903) Reiske was “the incomparable” (1871: 381). Heinrich L. Fleischer (1801–1888), foremost Arabist at his time, said of a Latin translation which Reiske had prepared at the age of 20, that he doubted there was a youth of the same age today, in 1870, who would be able to produce, even if supported by the best instruction and the most appropriate aids, a better translation (cf. Fück 1944:191192). Already in 1846, Fleischer had dedicated the first volume of his main work Beidhawii Commentarius in Coranum to the memory of Reiske. Carsten Niebuhr (1733–1815), known for his famed Orient expedition of 1764, was full of praise for Reiske, of whom he said: “This scholar has that much advanced in his knowledge of the Arabic language that Germany had perhaps never so far seen anyone like him” (Niebuhr 1772:XXXV). And then there was Gotthold Ephraim Lessing (1729–1781), one of the very few dignitaries of the time, who recognized and widely acknowledged Reiske’s accomplishments and their scholarly importance. He also provided help in many respects, although there were limits to what he could do other than being a good friend, always there to give advice and encouragement. Many instances in the surviving portion of their extensive correspondence prove the point. Unfortunately, Lessing’s plan to write a 3-volume Vita of Reiske, including “a precise listing of each and every paper left by him that is at least in some respect worth the trouble” (cf. Foerster 1889a:138), did not materialize. Ironically, the greater part of Reiske’s misfortune was caused by his difficult personality. He often managed to offend those who wanted to be supportive. Invariably he spoke his mind when it would have been more diplomatic, and even just more polite, to keep his thoughts to himself. But, on the other hand, he had also to endure more than his normal share of injustice, envy, and even causing outright hostility by some of his colleagues and numerous representatives of institutional authorities. His marriage to Ernestine Christine Müller in 1764 proved to be the most beneficial event of his entire life. With his help she had become a competent Classical scholar in her own right, but all through their married life she worked selflessly for the implementation of his projects and for helping him realize his full potential. We will now take a closer look at Reiske’s life and his work from his early education through the stages of his professional and scholarly career ending with his death in 1774. Reiske’s father left the early education of his son entirely to his wife. Anna Christina Reiske suffered for the greater part of her life from a hypochondriac disorder which did not fail to have a pervading effect on her son. In his Vita written in 1770 and published after his death by his wife, he stated without further elaboration: “From childhood I have been sickly, sad, downcast, people-shy, and hypochondriac” (Reiske 1783:5). After initial education at the municipal school of his home town Zörbig near Halle until the age of 10, he was sent to his maternal grandfather in Zöschen, a village near Leipzig. There Christoph Meissner, whom the pastor of the village had employed for his son,


187

taught him within a year’s time that much Greek, “that I could... understand and analyze rather well the four gospels in Greek” (ibid.). The next five years, until 1732, were spent at the Waisenhaus Gymnasium in Halle. It was a mixed bag for the timid child, who suffered severely from the harsh rules in the monastery school and the lack of truly excellent teachers. While he claimed to look back to Halle “always with happiness and gratefulness” (8), he maintained at the same time: “I took along from this school quite a good grounding in Latin, but hardly anything more” (5). Of his university studies in Leipzig, Reiske says: “I must confess, the greater part of those five years was lost time for me” (9). He lists several reasons, among them his desire to keep to himself, his unwillingness to ask anybody for advice, and his aversion to attend classes other than a few in literature. Classes on Classical Greek which he would have loved to attend, were not offered. So he went to work reading Greek writers on his own, but thereby experiencing more frustration due to lack of guidance than achieving satisfying success. A reprieve, mysteriously emerging, brought about a welcome change, yet no lasting relief. In his own words: “But then, a certain, inexpressible, unstoppable desire – I do not know myself how it came about – took hold of my soul, a desire to learn Arabic” (9). He went to work with an all-consuming zeal, devoting to the newly-found task all of his time and energy as well as the financial resources he had at his disposal. “In this I was, just like in everything else that I have ever learned or done, my own teacher” (9-10). By 1736 he had acquired and read, in spite of his limited resources, a great deal of what had been printed at that time in Arabic. This, however, created another, perhaps even more serious problem. When no Arabic manuscripts were obtainable any more in Germany, he decided to go where he would find them, i.e., to Holland, more specifically, to Leiden. For someone without financial means, without any experience as to what such a journey would entail, it was bound to be an extremely dangerous undertaking. Later in life he became aware of the great risks that had been involved: “I have become a martyr of Arabic literature” (11), he admits with some contrition. His Vita contains, however, a revealing statement as to what prompted him to embark on his daring journey into the Arabic language: “I have learned this language with the intention to acquire through it familiarity with the history, oratory, and poetry of the people who in former times have spoken it” (13). Such an attitude put him at odds with the entire purpose of Arab studies as practiced so far: to provide support to “the so-called Holy Philology” (12), i.e., the exploration of Latin, Greek, and Hebrew within the framework of Biblical Studies. Like a journeyman in search for the next master craftsman, he set out on foot in May 1738, traveling first to Hamburg, then by ship to Amsterdam, finally reaching Leiden on June 6. At the two stopovers he visited influential people to whom he had gained access via letters of recommendation. Their help, even monetary support, made it possible for him to stay afloat. In Amsterdam Prof. Jean-Philippe d’Orville, alerted through a letter to the precarious financial situation of the young German scholar, offered him employment as his personal secretary, but to d’Orville’s dismay Reiske declined, pointing out that he

188

KURT R. JANKOWSKY

had come to read Arabic manuscripts, not to make money. Nevertheless, by doing odd jobs for d’Orville and others he had to give up a sizeable portion of his time to earn money for food and shelter. Fortunately for him, proof-reading, editing manuscripts, preparing catalogues in libraries and related tasks kept him somewhat close to the type of work he wanted to do. His chief sponsor in Leiden was Albert Schultens (1689–1750), since 1732 Prof. of Exegesis and Oriental Languages at the University. Reiske owed him a great deal during his 8-year stay in Holland. That their healthy relationship was eventually all but destroyed, had two major reasons. The first one I have already mentioned: Schultens was convinced that Reiske, due to his proven capabilities as a scholar, and not Schultens’ son would get appointed to a professorship in Arabic studies at the University unless an effective plan was designed and implemented to get rid of the inconvenient competitor. This plan entailed to entice Reiske to study medicine rather than to continue with Arabic studies which – so Schultens contended – might leave him unemployed. It is almost unbelievable to believe that Reiske this time did indeed follow the advice, although he had made it a habit to flatly reject truly helpful offers. The second reason was entirely of his own making. He had stated publicly his firm conviction that Schultens’ dealing with Arabic did not amount to anything at all. There was too much speculation, too much arbitrary and empty stuff in his lectures as well as in his writings. Arabic studies should not be treated as part of theology, but as geography, mathematics, physics, and medicine, i.e., as a subject matter in its own right. Needless to say that Schultens’ resolve to remove Reiske from the scene grew only stronger when he became aware of his protégé’s critical remarks. Curiously enough, Schultens personally intervened in behalf of Reiske at the decisive meeting of the promotion committee, and he himself told Reiske afterwards “how little space there actually had been between obtaining and losing the doctorate and how he, and the medical faculty, with great effort had barely been able to rescue me from the danger” (32-33). In his Vita, about 24 years later, he arrived at what one might call a fair assessment of his adventurous 8 years in Holland, stating the conviction that the country and its people had treated him well and had offered him a great deal of worthwhile opportunities, on the personal and professional level, but that he had failed to make use of most of them. He acknowledged to have learned and profited greatly from his stay in both his Arabic and Classical studies. When he arrived in Leipzig at the beginning of July 1746, he knew that making any use of his medical degree would be out of the question. He first spent a few months at his parents’ home in Zörbig, to see “whether there might emerge in the immediate neighborhood a good opportunity to secure an advantageous marriage so as to establish the place for my future life and fortune” (41). Realizing the futility of that hope, he returned to Leipzig. What initially prevented him from obtaining a Magister noster title from the Philosophical Faculty, precondition for the permission to give lectures, was the


189

fact that he already held a higher degree in another faculty, his doctorate in medicine. Finally, with the publication of his Dissertatio de principibus Muhammedanis, dedicated to His Royal Highness, the Crown Prince Frederic Christian, in 1747, he was awarded the title of Professor, but reaped no special benefits from either the philosophical or the medical faculty (43). Obviously, Reiske’s professional life was inextricably intertwined with what he did or did not do on a very personal level. His wife had no illusions as to how often he avoided taking appropriate action, when the display of credible initiative in word and deed could have been to his advantage in getting favorable responses. In his Vita he complained about the slow sale of his books. What a pain it must have been for his loving wife to write down the following revealing comment: “To a certain extent the good man was partly to blame for the bad result in the sale of his books. He neglected far too much to establish and uphold acquaintances with scholars who edited scientific newspapers and journals; and he did not bother much about the book dealers either. Therefore his works remained unknown; for if nobody knows a particular work, its own value alone does not sell it” (68). Similarly, rather than creating job opportunities, he waited for a miracle to happen. In the absence of a vigorous, comprehensive action plan to construct a moneyproducing livelihood, Reiske had to live from hand to mouth. He continued practicing what he had started in Holland: giving private lessions, correcting manuscripts, translating from other languages into German (44), again as before often enough for a mere pittance in remuneration. But he also managed somehow to keep writing scholarly materials, book reviews, small articles for scientific periodicals, and translations of Arabic, Latin or Greek manuscripts. With the years, the number of publications steadily increased, and so did the number of manuscripts for which he could not find a publisher, forcing him to pay for the publication himself as the only alternative. A case in point is his series Animadversiones ad Graecos Auctores. A total of 200 copies were printed at his own expense. Not all of them, he said, were sold, amounting to a loss of about 900 thalers. He considered his Animadversiones “the best piece of writing that I have ever produced. They are flos ingenii mei, the flower of my genius” (70). Publication costs were borne by him repeatedly, when other attempts had failed. He cited as an example his edition of the Anthologiae Graecae, published in 1754: “I did not want to be kept waiting in empty hope any longer. So I dared having it printed at my expense and throw my bread into the water; just like my desire to become useful in my way had tempted me to do many more times later on. I am not put off by the injustice of people, but move along on my path unconcerned, as far as I can, full of confidence that God will always... reward the good and punish the bad” (68). For many years he had worked almost exclusively with Greek writers. He even claimed that he had spent the greater part of his life reading Greek authors. He preferred Greek to Latin because he felt that his initial training received at the Waisenhaus Gymnasium was

190

KURT R. JANKOWSKY

not so much in Classical Latin. “That was a disaster for me. I did acquire from the newer Latinists a certain expertise in Latin so that I could read and write Latin fluently and that even people who were experts in the good Latinity regarded my Latin as beautiful. But the good Latinity of the old genuine authors I have come to know only in my 40th year of life; by then it was too late for me to remedy what I had neglected. Only then I began to get acquainted with Cicero, but it was too late to acquire a taste for him” (7). By 1755 his name had become known to organizations of various academic disciplines such as language, literature, and history. Johann Christoph Gottsched (1700– 1766) invited him to join his Gesellschaft der freien Künste und Wissenschaften (Society for the Liberal Arts and Sciences), and a year later, Christian Wilhelm Walch (1726–1784) offered him membership in the Jenaische Lateinische Gesellschaft, the Latin Society of Jena. Other honorable nominations followed, e.g., inclusion among the honorary members of the Royal Historical Society in Göttingen in 1769. These honors led to contacts with influential people who, once convinced of his expertise, were eager to avail themselves of his skills. After Reiske had successfully completed some challenging tasks for a court official in Dresden which involved the Arabic monetary system, Count Wackerbart, Cabinet Minister in Dresden, put his special knowledge to a serious test. Reiske succeeded in deciphering the inscription on a gold-embedded stone that no one had ever been able to interpret correctly. The ramification of this remarkable achievement was to have a crucial impact on his future life. In 1757, Reiske’s financial plight had reached disastrous proportions. He was forced to accept daily free lunches offered by Prof. Ernesti, not exactly a special friend of his. A most welcome relief emerged when the position of rector of the Nicolai School in Leipzig fell vacant. Reiske applied, driven by the thought: “What if God had here opened to you the door to a better fate?” (78). He was convinced that “God had led me, by an unexpected accident... to that office.... Otherwise I would have perished. I recognize in this God’s almighty hand that can save even from death.” His numerous adversaries immediately attempted to block the appointment, but their efforts were in vain, since, through the intervention of the Cabinet Minister Count Wackerbart, backed by his Royal Highness, the Electoral Prince, Reiske was officially inaugurated as rector in June 1758. By comparison, life now became more bearable for him, and he could devote much more time to his long, inexhaustible list of research projects. If one considers the amazingly large number of scholars with whom Reiske had been in touch throughout his professional career, one could hardly claim that he led an isolated life. But taking into account his propensity for confrontation – whether for invalid or for praiseworthy reasons – the outcome can hardly leave any doubt that longlasting, fruitful, continuously beneficial relationships of his were very few, excepting most notably his marriage to Ernestine Christine Müller. Ernestine Christine was his greatest asset all through their marriage which spanned the last ten years of his life. She had highest regard for him as a person and as a scholar, and he


191

seemd to have his rather erratic personality under perfect control in his relationship with her. She was instrumental in solidifying a number of Reiske’s relationships with colleagues and family friends that had either been disrupted or were not likely to last due to his oftentimes careless attitude towards friends and associates. Restoring the impaired friendship with Gottsched due to Mrs. Reiske’s pleasing personality is a case in point. But there are indications likewise that often enough, where personal and professional decisions were involved, Mrs. Reiske, in spite of all her efforts, was unable to prevent her husband from incurring harm or even outright disaster. When she realized that he felt strongly about a planned course of action, she gave him unconditional support even at great personal sacrifice. On the professional level, she provided any type of help she could, always anxious to make sure that her contributions would fall neatly into the framework of his overall research and publication plans. Furthermore, she prevailed upon him to write up his own life story. This was a remarkable achievement, considering his reluctance to speak freely and openly about himself. The opening remarks in his Vita make that very clear: “I am supposed to write about my life. I have been asked to do it several times, and I have never felt like doing it” (Reiske 1783:1). Nevertheless, his 150-page Curriculum Vitae was written, and it contains a wealth of fascinating details on his personal and professional life. It was published after her husband’s death, together with a number of equally important appendices: (1) Her own account on Reiske, mainly covering the last years of his life, full of praise for his many accomplishments. (2) A listing of Reiske’s published and unpublished works. (3) Several hundred pages of Reiske’s correspondence. To finance this bulky publication project she devised a subscription plan, for which she managed to engage the support of a long list of illustrious personalities from Germany and abroad. Undoubtedly her greatest accomplishment, however, is that she succeeded in completing a large number of unfinished manuscripts and arranging the publication of these as well as of other works which Reiske had completed, but could not get published for lack of either time or money. Last but not least, after her husband’s death in 1774, she did everything in her power to lay the foundation for the comprehensive recognition of his significance for the establishment of Classical and Arabic studies in Germany. First, and most importantly, she prevented the Reiske papers from falling into the hands of false friends who would have utilized them for their own glorification. Second, she saw to it that all scholarly papers finally went to a Danish nobleman, Peter Frederek von Suhm (1728–1798), who would ensure their safe keeping while at the same time continue with the publication process.

192

KURT R. JANKOWSKY

Concluding Remarks Reiske approached classical and Arabic Studies with an open mind, abhoring any preconceived ideas. He was not bound by restrictions that an academic position – which he never held – or an academic tradition might have placed on him. With ease he recognized that Classical studies required a great deal of expansion and intensification. He elaborated on what needed to be done by both theoretical discussions in many of his writings and practical implementations through his research. His example as a protagonist for self-sufficient Classical and Arabic studies had repercussions even at his time, although then only in a limited way due to his peculiar professional circumstances referred to before. In a moving self-assessment he presents an accurate picture: “God has given me gifts; not the best, but also not the worst; and besides those gifts he also endowed me with urge and desire to make use of them to His glory and for the common good.... I would have certainly been able to do much more, had the time been more amenable to my kind of studies and had I received more support and encouragement from my contemporaries. But I have done more than a thousand of others would have done under my circumstances” (Reiske 1783:96). In Arabic studies he was determined from early on to discard what in his opinion proved to be a serious impediment for obtaining the best possible results. For Arabic studies to become a self-sufficient, independent academic discipline, it was imperative for them to be cut loose from their intrinsic relationship to philologia sacra, and cease to be subservient to the primary requirements of the investigation for the benefit of the three Holy Languages, Greek, Latin, and Hebrew. Reiske’s envisioned procedure brought into scope subject matter areas that so far had at best been merely of marginal interest to investigators. Ex oriente lux – light from the east – had previously as its focus anything intrinsically related to Christianity and its origins. Now, for Reiske, the focus shifted to a different dimension of meaning: the oriental world – Reiske preferred the terms Islamic or Moslemic – deserved to be a subject matter in its own right. His point of departure was a thorough investigation of Arabic language texts, and he made sure that the selected texts were restored as closely to their original state as could possibly be achieved by thorough textual criticism. But language was for him not an end in itself. He aimed at exploring Arabic culture, all aspects of it. He discovered the value of Arabic history and was intent on assigning to it its rightful place in world history (cf., e.g., Reiske 1747b, a 274-page treatise published in 1847). For those who knew him well it was no surprise that he proved also to be a numismatic expert, capable of deciphering Arabic inscriptions on coins, which others before him had attempted and failed. The science of numismatics owes a great deal to his pioneering effort. Richard Foerster (1943-1922), biographer of Reiske and editor of a large portion of his correspondence at the end of the 19th century, had this to say about his hero: “The assessment that he belongs to the greatest philologists of all times is today (1897) shared by all evaluators in the field of Greek and Arabic studies, who are truly capable


193

of judging” (Foerster 1897). While Foerster’s claim might be somewhat overstated, I would assert nevertheless that he comes fairly close. What Reiske did achieve in his two academic disciplines was truly unique and is now widely acknowledged as such, even though the adverse circumstances of his professional and personal life prevented him from developing his full potential. Today, in 2002, the riches burried in Reiske’s published and as yet unpublished works and in the huge amount of correspondence still wait for an exhaustive exploration. Once this is accomplished and Reiske’s work is viewed within its appropriate historical context, Reiske’s achievements and Foerster’s assertion might actually be unrestrictedly called identical. REFERENCES Primary sources (1742) Tharaphae Moallakah cum Scholiis Nahas e mss. Leidensibus / Arabice edidit, vertit, illustravit Joann. Jacob Reiske. Lugduni Batavorum: Apud Joannem Luzac. (1746) Miscellaneae aliquot observationes medicae ex Arabum monumentis. Lugduni Bat. (1747a) Dissertatio de principibus Muhammedanis, qui aut ab eruditione, aut ab amore literarum et literatorum claruerunt. Lipsiae. (1747b) Primae lineae historiae regnorum Arabicorum et rerum ab Arabibus medio inter Christum et Muhammedem tempore gestarum: cum tabulis genealogicis tribuum Arabicarum/ Jo. Jac. Reiskii; e libro manuscripto Bibliothecae Gottingensis edidit Ferdinandus Wüstenfeld. Gottingae: Typis Librariae Dieterichianae. (1748a) De Arabum Epocha vetustissima, Sail ol Arem, id est ruptura catarrhactae Marebensis dicta / invitat Ioannes Iacobus Reiske. (Inaugural writing.) Lipsiae: Pouillard. (1748b) Oratio studium Arabicae linguae commendans. (Leipzig: Inaugural address. Printed 1779.) (1754a) Anthologiae Graecae a Constantino Cephala conditae libri tres. Lipsiae: In Bibliopolio Gleditschiano. (1754b) Abu-’l-Fida Ismail Ibn-Ali: Annales moslemici/ lat. ex arab. fecit J. J. Reiske. Lipsia: Gleditsch (1757– 66) Animadversiones ad Graecos Auctores V, Vol. 8, Lipsiae: Gleditsch. (1759) De optimo episcopo. (Gratulationsschrift für Johann August Ernesti.) Leipzig: Langenheim. (1759 – 60) Animadversionum ad Graecos Auctores VI, Vol. 8. Lipsiae: Gleditsch. (1761) Deutsche Übersetzung der Reden aus dem Thucydides, nebst lateinischen Anmerkungen über dessen gesamtes Werk. Leipzig: Löper.

194

KURT R. JANKOWSKY

(1764) De principibus, mahummedanis. Lipsiae: Gleditsch. (1766) Abulfedae Tabula Syriae cum excerpto geographico ex Ibn ol Wardii Geographia et historia naturali (Abulfeda). Lipsiae: Litteris Schoenermarkii. (1770 – 75) Oratorum Graecorum, quorum princeps est Domosthenes, quae supersunt, monumenta ingenii, e bonis libris, a se emendata, materia critica, commentariis integris Hieron. Wolfii, Io. Taylori, Ierem. Marklandi, aliorum, et suis indicibus denique instructa, edidit Ioannes Iacobus Reiske. 12 vols. Lipsiae: Typis W. G. Sommeri. (1776) Opuscula medica ex monimentis Arabum et Ebraeorum. Halae: Apud I. I. Gebauerri vidvam et filium. (1781) “Briefe über das arabische Münzwesen. Mit Anmerkungen und Zusätzen von Joh. Gottfr. Eichhorn.” Repertorium für biblische und morgenländische Litteratur (Leipzig: Weidmann) 9:192-346. (1783) D. Johann Jacob Reiskes von ihm selbst aufgesetzte Lebensbeschreibung. Hrsg. von seiner Frau Ernestine Christine Reiske. Leipzig: Sommer. Secondary bibliography Bachmann, Peter. (b. 1936). 1974. Johann Jacob Reiske: Der Begründer der arabischen Studien in Deutschland. Beirut. Bennholdt-Thomsen, Anke & Alfredo Guzzoni. 1992. Gelehrsamkeit und Leidenschaft: Das Leben der Ernestine Christine Reiske 1735–1798. München: C. H. Beck. Bentley, Richard. (1662–1742).1962[1691]. Epistola ad Joannem Millium. (Repr. from the edition of Alexander Dyce, together with an introduction by George P. Goold.) Toronto: Univ. of Toronto Press. . (1662–1742). 1699[1697]. A Dissertation upon the Epistles of Phalaris. London: Hartley. Brekle, Herbert E. et al., eds. 1992-. “Reiske, Johann Jacob”. Bio-bibliographisches Handbuch zur Sprachwissenschaft des 18. Jahrhunderts: Die Grammatiker, Lexikographen und Sprachtheoretiker des deutschsprachigen Raums mit Beschreibungen ihrer Werke. Band 7. Tübingen: Niemeyer. Eckstein, Friedrich August. (1810–1885). 1871. Nomenclator philologorum. Leipzig: Teubner. [Repr. Hildesheim: Olms, 1966.] Fleischer, Heinrich Leberecht. (1801–1888), ed. 1846–1848. Beidhawii Commentarius in Coranum: ex codd. Parisiensibus, Dresdensibus et Lipsiensibus. 2 vols. Lipsiae: Sumptibus F. C. G. Vogelii. Foerster, Richard. (1843–1922). “Johann Jacob Reiske”. Allgemeine Deutsche Biographie 28.129-140. . 1889b. “Ernestine Christine Reiske”. Allgemeine Deutsche Biographie 28.140-143.


195

. 1897, ed. Johann Jacob Reiske’s Briefe, Leipzig: Hirzel. . 1917, Briefe von J. J. Reiske. Nachtrag. [Abhandlungen der Philologisch-Historischen Klasse der Königl. Sächsischen Gesellschaft der Wissenschaften, 34. Bd., No 4.] Leipzig: B. G. Teubner. Fück, Johann. (1894–1974). 1944. “Die arabischen Studien in Europa vom 12. bis in den Anfang des 19. Jahrhunderts”. Beiträge zur Arabistik, Semitistik und Islamwissenschaft, 85-253. Herausgegeben von Richard Hartmann (1881–1965) und Helmuth Scheel. Leipzig: O. Harrassowitz. Herder, Johann Gottfried von. (1744–1803). 1785–1792. Ideen zur Geschichte der Menschheit. 4 vols. in 2. Riga & Leipzig: Johann Friedrich Hartknoch. Lessing, Karl Gotthelf. (1740–1812). 1789. Gelehrter Briefwechsel zwischen Johann Jacob Reiske, Moses Mendelssohn und Gotthold Ephraim Lessing. 2 Theile. Berlin: Voss und Sohn. Littmann, Enno. (1875–1958). 1942. Der deutsche Beitrag zur Wissenschaft vom vorderen Orient. Stuttgart & Berlin: W. Kohlhammer. Meusel, Johann Georg. (1743–1820). 1811. Lexikon der vom Jahre 1750 bis 1800 verstorbenen Teutschen Schriftsteller. Vol. XI, pp. 192-208. Leipzig: J. G. Neubert. [Contains most complete bibliography.] Mommsen, Theodor. (1817–1903). 1871. “Bruchstücke des Johannes von Antiochia und des Johannes Malalas”. Hermes 6:323-383. Morus, Samuel Friedrich Nathanael. (1736–1792). 1777. De Vita J. J. Reiskii Medicinae Doctoris. Lipsiae: Georg Niebuhr, Carsten. (1733–1815). 1772. Beschreibung von Arabien: aus eigenen Beobachtungen und im Lande selbst gesammelten Nachrichten. Kopenhagen: N. Möller; Leipzig: B. C. Breitkopf und Sohn, Pfeiffer, Rudolf. ((1869 –1979). 1976. History of Classical Scholarship from 1300 to 1850. Oxford: Clarendon Press. Sacy, Antoine Isaac Silvestre de. (1758–1838). 1854-[65]. “Biography de Johann Jacob Reiske”. Joseph Fr. Michaud (1767–1839) and Louis Gabriel (1773–1858) Michaud. Biographie universelle ancienne et moderne. 35.381-387. Deuxième édition par Eugène Ernest Desplaces (b. 1828). Paris: Madame C. Desplaces. Sandys, John Edwin, Sir. (1844–1922). 1998[1903–1908]. A History of Classical Scholarship. Vol. III: The Eighteenth Century in Germany, and the Nineteenth Century in Europe and the United States of America. Bristol [England]: Thoemmes Press. Wilamowitz-Moellendorff, Ulrich von. (1848–1931). 1982. History of Classical Scholarship. Transl. from the German by Alan Harris; edited with introduction and notes by Hugh Lloyd-Jones. Baltimore, Md.: The Johns Hopkins University Press.

196

KURT R. JANKOWSKY

197

THE CONTEXT AND SENSE OF HUMBOLDT’S STATEMENT THAT LANGUAGE ‘IST KEIN WERK (ERGON), SONDERN EINE TÄTIGKEIT (ENERGEIA)’ HANS AARSLEFF Princeton University During the last years of his life Wilhelm von Humboldt produced the three-volume work entitled Ueber die Kawi-Sprache auf der Insel Java (1836-39). The bestknown part of this work is the introduction, often called the Kawi-Introduction (or for short KI), which, he said, would be ‘devoted to more general considerations’ with a view to reading the rest of the large work (OL 1988:20; GS 7:13).1 The KI presents the summary of some thirty years’ work on language, with much repetition and steady commitment to basic views that were formed early. His basic conception was that ‘we must look upon language, not as a dead product, but far more as a producing,’ a thought to which he gave this formulation on the next page: ‘In itself [language] is no product (Ergon), but an activity (Energeia)’ (‘Sie [die Sprache] ist kein Werk (Ergon), sondern eine Tätigkeit (Energeia)’) (OL 1988:48-9; GS 7:45-6). Four aspects of this statement merit attention: (a) The use and contrasting of the Greek terms, of which Energeia occurs nowhere else in Humboldt’s writings. (b) The correlation of Ergon with the German ‘Werk,’ which means ‘finished work,’ ‘work already done,’ and Energeia with ‘Tätigkeit,’ which signals ‘action,’ ‘activity,’ with the implication of creativity and creation. (c) The package-like quality of the formulation, like a quotation that suddenly flashed into the context. (d) The primal emphasis placed on energy. After the word Energeia the text continues about language, ‘its true definition can therefore only be a genetic one.’ This means that languages must be understood in terms of their formation thus becoming more or less successful depending on the formative capabilities of their speakers. Language is not a rational construct, but ‘an act, a truly creative performance of the mind’ (OL 1988:183; GS 7:211). Speech and language must be understood in aesthetic terms, a conception that is splendidly argued in Diderot’s Lettre sur les sourds-et-muets (1751), which, basing on Condillac’s anti-rationalist Essai sur l’origine des connaissances humaines (1746), is the pivotal text in Diderot’s 1. For abbreviations, see the list at head of References.

198

HANS AARSLEFF

œuvre.2 The Essai has related chapters on ‘Music’ and on ‘Inversions,’ that is, the transposition of words away from any sentence order that might be considered natural or normal. The subject of the Lettre is inversion and its source in the energetic rhetorical expressivism that produces it. Inversion undermines the rationalist doctrine that the prevailing word order of, say, modern French is the natural order.3 The sense of the ergon/energeia statement has been the subject of endless and aimless speculation without ever leading to a viable answer. In accordance with the canons of intellectual history, the remedy for this failure lies in attention to the context in which Humboldt was living and working, and this takes us back to the time when his exposure to Basque first made language the heart of his anthropology. As we shall see, it was also in this context that the ergon/energeia conception first entered his mind.4 When he arrived in Paris with his family in November of 1797, Humboldt had gone through a long period of depression that found disturbing expression in his letters. Full of self-reproach, he felt unproductive and unable to write because he could not collect his thoughts. In an especially dark letter to Friedrich Schiller, he wrote in the late summer of 1797 that he lacked the fundamental notions for what he was trying to do – ‘dass mir noch Grundgedanken fehlen’ (Freese 1986:228). At the time he was trying to write two things, one on the eighteenth century and another entitled ‘Plan einer vergleichenden Anthropologie.’ Both remained unfinished. He worked on the latter during 1797, but in spite of efforts to define the subject as well as aims and method, language is wholly absent except once when it is listed among ‘exterior’ features that reveal ‘Verschiedenheit unter Menschen,’ such as ‘Körperbau und Betragen, Gestalt, Farbe des Gesichts und des Haars, Physiognomie, Sprache, Gang und Gebehrden überhaupt’ (GS 1:399).5 But within a short time the depression lifted while his thinking also took a radical turn. 2. On the centrality of the Lettre in Diderot’s aesthetics, see the magisterial Chouillet 1973, esp. pp. 153-257. Also Chouillet (1984). Baudiffier (1982) offers a succinct demonstration of Diderot’s closeness to Condillac. Becq (1994) treats Condillac’s aesthetics, pp. 444-64. 3. On rhetorical expressivism, see Aarsleff (2001: xviii-xxiii), and also Aarsleff, forthcoming. 4. On contextualism, see Skinner (2002) and Tully (1988). 5. To take an example of the absence of language, in ‘Vergl. Anthrop.’ we read concerning ‘Kenntniss der Charaktereigentümlichkeit’ that, ‘der Charakter ensteht nicht anders, als durch das beständige Einwirken der Tätigkeit der Gedanken und Empfindungen’ (GS 1:386) – the confident ‘not otherwise than’ leaves no role for language whatsoever. ‘Vergl. Anthrop.’ was no not begun until early 1797, but this causes conflict with the editor Leitzmann’s dating of the set of theses on language which he gave the title ‘Ueber Denken und Sprechen’ (GS 7:581-3) and dated 1795/96, a conflict that did not occur for him because he mistakenly dated ‘Vergl. Anthrop.’ 1795, which is one of several of his datings that have proved untenable. Lammers (1935:30) also questions the dating of ‘Ueber Denken und Sprechen.’ Menze (1965:211-13) accepts the dating but then says that, ‘erst der Aufenthalt in Paris und die Reisen nach Spanien bringen neue Beschäftigung, Vertiefung, und Auseinandersetzung mit der Sprache,’ which then again raises the question why, written in the meantime, ‘Vergl. Anthrop.’ has no role for language. The commentary at FG 5:674 (to a letter of 1803) remarks that, ‘Die Anfänge von Humboldts Sprachphilosophie sind offenbar schon in die Pariser Jahre zurückzuführen.’ Howald (1944:97-106) is good on Humboldt’s depression and on his recovery once in Paris. bringen neue Beschäftigung, Vertiefung, und Auseinandersetzung mit der Sprache,’ which then again raises the question why, written in the meantime, ‘Vergl. Anthrop.’ has no role for language.

THE CONTEXT AND SENSE OF HUMBOLDT’S STATEMENT

199

Only a few weeks after arriving in Paris Humboldt felt exuberant improvement and firm confidence about getting on with the writing. In a searching self-assessment written at the end of 1797 he contrasted past failure with present assurance that he now had ‘das unverkennbare Bewusstsein einer lebendigen und emporstrebenden Tätigkeit’ (GS 14:391). About a month earlier he had written with enthusiasm about Paris to a friend; he was full of cheer, ‘und vielleicht nie noch war ich so sehr zur Beobachtung und Tätigkeit bestimmt,’ more than ever eager to get on with writing on the eighteenth century and on comparative anthropology (Freese 1986:234-5). From then on his letters from France and Spain are optimistic, with regular mention of his new-found linguistic anthropology. The question is: what had caused this sudden and dramatic change? In print Humboldt several times declared that it all happened because he ‘by chance’ came upon Basque and its speakers on his way toward Madrid late in 1799 (GS 3:292; 6:137). But the story is very different. The special status of Basque was already being discussed in the Parisian circles in which Humboldt moved before he left for Spain. Chief among his informants was the political figure and man of letters DominiqueJoseph Garat, who, born at Bayonne, was a native speaker of Basque and a dedicated advocate of his home region’s language and culture. Garat is mentioned in a letter to Schiller already in late January 1798, with the flattering comment that he is equipped ‘mit einer ausserordentlich guten Kopf’ (Seidel 1962:147), and by the summer of that year Humboldt noted that Garat had written for Mercure de France, and that he had been one of the initiators of the first collected edition of Condillac published in 1798, the same year as Garat’s ‘Discours préliminaire’ appeared in the fifth edition of the dictionary of the French Academy. Both in practical and linguistic matters Garat became Humboldt’s cicerone in the new territory he was now entering. He paid close attention to two of Garat’s essays that had appeared in Mercure de France in 1783 and 1785, even to the point of copying out parts of those essays. The first, ‘Sur Bayonne & sur les Basques,’ contains a eulogy of the Basque language, with emphasis on the intimate relationship between a nation’s entire mode of thought, feeling, and culture on the one hand and its language on the other: (1) Basque has all the admired qualities of Greek and Latin and is probably older than either, by which he meant original and free of any admixture from other languages. (2) Basque shows perspicuous analogy in the formation of words. (3) And most of all, the syntax of Basque shows ‘une hardiesse incroyable pour entrer dans la pensée de tous les côtés, & cependant dans les mots des marques sûres & infaillibles pour fixer rapidement leurs rapports au milieu même des mouvements les plus passionés & les plus convulsifs de l’inversion.’ To these qualities, Garat declared, the Basque nation chiefly owed ‘tout ce qu’il y a de raison & d’étendu dans les idées..., d’énergique, de fin & de délicat dans ses sentiments’ (Garat 1783:66). Thus Basque displayed the most fruitful fit between speaking and thinking, between thought and expression. Garat also noted that the great rhetorician Quintilian was a native Basque, an observation that has deeper implications

200

HANS AARSLEFF

than I can go into here. Garat defined the question that informs the entire KI, namely how ‘language is connected with the nation’s mental power’ (OL:21; GS 7:14).6 Garat’s essay of 1785 is a review of Antoine Rivarol’s rationalist Discours sur l’universalité de la langue française (1784). Rivarol and Garat shared the principle that ‘la parole est aussi nécessaire à la pensée que la pensée à la parole.’ This insight, wrote Garat, was ‘une de plus belles découvertes du siècle,’ while at the same time blaming Rivarol for not crediting the discovery to Condillac, whose glory it was to have perceived it in his first work, the Essay, and to have expounded it more fully in his later writings (1785b:21). It was with these things in mind that Rivarol made the much-cited claim that ‘ce qui n’est pas clair n’est pas français’ (Rivarol: 49). French did not suffer the disorder of inversions which, he said, flourishes when people become more powerfully governed by their passions than by reason. Celebrating the rationality of his language, Rivarol admitted without regret that French had no affinity for music and the arts (Rivarol: 50). Naturally, as the dedicated standard-bearer for Condillac, Garat did not agree. Garat answered that, on the contrary, inversion favored clarity more than the direct rationalist order: M. de Rivarol prétend que dans le Latin, le sens est suspendu jusqu’à la fin de la phrase. Mais j’ignore quelle est la langue dans laquelle le sens peut être achevé avant la phrase; & c’est -là encore une des choses que tout le monde répète sans que personne s’avise une fois au moins d’examiner ce qu’il dit. Dans toutes les langues du monde, ce n’est qu’avec la phrase que le sens peut être terminé, & jusqu’au dernier mot le sens est suspendu. (1785b:33; GS 15:45-6; OL 1988: xliii-xlvi)

This entire passage was copied out in French by Humboldt in a note entry from 1799 that was first published in 1918, but not identified until I did so (OL 1988:xliv). At this point Humboldt became so excited that he also copied out in German a sentence on the facing page: ‘Die Sprachen mit Inversionen sind der Klarheit vorteilhaft, weil man mehr Mittel sich auszudrücken hat’ (GS 15:46). In Garat’s French this sentence was immediately preceded by these words: ‘Dans les langues à inversions, dans la langue Latine, par example, il y a vingt manières de construire la même phrase; & on n’est jamais obscur si on donne aux mots les désinences qui en marquent les rapports.’

And it was followed by this sentence on the deficiency of the rationalist order: ‘& les langues à ordre direct lui sont contraires, puisqu’il n’y a souvent qu’une seule manière d’être clair, & qu’il y en a vingt d’être obscur’ (1785b:32). In his critique Garat also 6. Garat (1783) is identified in Humboldt’s Nachlass in Mueller-Vollmer (1993:260). Aarsleff (1988) treats Garat’s importance for Humboldt in some detail, though without use of Garat (1783), which was not known to me. Aarsleff (1988) also demonstrates Diderot’s impact on Humboldt’s thought; for his ecstatic response to the reading of Diderot, see p. li.


201

recalled that he was himself, both as speaker and listener, familiar with a language – his own Basque of course – in which inversions were much bolder than in the classical languages. Indeed he believed it was the language that is spoken and understood most quickly of all, thus exhibiting the ideal quality that Humboldt would later find preeminent in Greek and Sanscrit. Humboldt may also have known Garat’s entry ‘Cantabrie’ (1785a) in the second of the three volumes of the Encyclopédie méthodique that were devoted to Grammaire et littérature (EMGL). All entries in this work were signed, and there can be no doubt that the EMGL was busily pillaged without credit. In this entry, which was not in the original Encyclopédie, Garat was very explicit. Basque, he wrote, has more declensions and conjugations than the classical languages, often using declensions even where Greek and Latin use prepositions. He immediately continued to state the consequence of this richness. In Basque, ‘[les] inversions sont infiniment plus hardies que celles du latin & du grec; & cependant avec quelque rapidité qu’on parle cette Langue devant moi, je ne suis jamais ni arrêté ni suspendu dans l’intelligence d’une phrase!’ (EMGL 2:446b). Garat was again saying that it is the entire utterance, the sentence that is the unit of meaning, not the individual words as the rationalists maintained. This principle of the unit-sentence is prominent also in Condillac and Diderot, and it has important implications for logic, a point well understood by Jeremy Bentham. In this entry Garat also celebrated the analogy that rules in Basque because it is an unmixed, ‘original’ language. This is yet another key notion that is firmly stated by Condillac: mixed languages have weak analogy (Essay 2001:187; Essai 1973:261). As in Condillac and Garat, this notion also becomes prominent in Humboldt’s work. In the KI he wrote, echoing Garat: ‘In Basque there flourishes a wealth of verb-inflection forms which it is difficult to survey. But the analogy of their formation is adhered to with admirable care, so that the understanding runs through them by way of an easily recognizable guide-line’(OL 1988:134; GS 7:150-1). And in the context of celebrating inflection as the condition for the ideal he repeatedly called ‘wahre innere Festigkeit,’ Humboldt observed that, ‘in its word-construction [i.e., analogy] and speechordering [i.e., syntax], Basque has a special power arising from its brevity and boldness of expression [i.e., as in inversion]’ (OL 1988:146; GS 7:164). It is obvious that Humboldt responded with unmistakable evidence of interest, engagement, and discipleship on the conceptual coherence of the aesthetic view of language, on energetic expression, on inversions, inflectionality, true inner fixity, the weakening of analogy in mixed languages, and most of all on the combined effect of these qualities in producing the intimate union of language and culture. Garat was not an original mind, yet he applied all these qualities in his eulogy of the Basque language and nation. What is the source of all this? Where lies the center? As Garat made clear it lies in Condillac’s Essay, and especially in the chapter on inversions, with weighty support from the chapter on ‘Music.’ The freedom of word order that is made possible by high inflectionality allows the imagination to create a compactness of expression that we, with Diderot, recognize in a hieroglyph or a painting. In French Condillac and

202

HANS AARSLEFF

Diderot used the word ‘tableau’ for this effect; in German Humboldt used ‘Gemälde’ or more often ‘Bild,’ and when he wrote in French ‘tableau.’ The entire conception is well stated in one of Humboldt’s French writings: L’avantage à cet égard est entièrement du côté des langues qui regardent l’expression comme un tableau de la pensée dans lequel tout est continu et fermement lié ensemble, et ou cette continuité est imprimée aux mots mêmes...et qui permettent à celui qui écoute, de suivre, toujours à l’aide des sons prononcés, l’enchainement des pensées, sans l’obliger à interrompre ce travail pour remplir les lacunes qui laissent les paroles. (GW 7:345-6; cf. OL 1988:88-9; GS 7:94-6, and often elsewhere)

One of Humboldt’s charges against Chinese was precisely that it imposed on the listener the burden of filling in the lacunae or, in the terminology of rationalist grammar, of ‘subauditing’ what was not expressed. All this was something that Garat’s claims for Basque had made clear long before Humboldt decades later began to apply these conceptions to Sanskrit and Greek. We are now at the threshold of ergon/energeia. In the chapter on inversions Condillac cited some lines from a Horatian ode (Odes Bk. I, no. 28) by way of illustration: ...nec quicquam tibi prodest aerias tentasse domos, animoque rotundum percurrisse polum, morituro

The expressiveness of these lines, he argued, is caused by the need to join ‘tibi’ in the first line with the word on dying in the last. If these words were put into the direct order of prose, there would be no role for the imagination to create the synthesis that Condillac and Diderot, and with them Humboldt, so greatly prized. Inversion, said Condillac, has much the same quality as the language of action, which is the protolanguage that exhibits the poetic origins of speech in which ‘a single sign is often equivalent to an entire sentence.’ From this potency it followed that inversions ‘create a picture [tableau], that is, that they in a single word unite the circumstances of an action, much as the painter unites them on the canvas’ (Essay 2001:175-6; Essai 1973:249). Condillac called this effect ‘vivacité.’ The quick effect of Condillac’s exposition was Diderot’s celebration of inversion and energy in the Lettre sur les sourds-et-muets (Baudiffier 1982). In 1782 the EMGL (1:713a) had a new entry ‘Energie,’ written by Nicolas Beauzée. It explained that, Energie est cette qualité qui, dans un seul mot ou dans un petit nombre des mots, fait apperçevoir ou sentir un grand nombre d’idées; ou qui, au moyen du petit nombre d’idées exprimées par les mots, excite dans l’âme des sentiments d’admiration, de respect, d’horreur, d’amour, de haîne, &c. que les mots seuls ne désignent point.


203

To illustrate this passage Beauzée immediately cited the same Horatian lines that Condillac had quoted to show the vivacity of inversion, which, Condillac had affirmed, is made possible only by the high inflectionality of Latin. Beauzée’s explication of energy inescapably calls to mind Humboldt’s statement on ergon and energeia, and this also was in Beauzée’s entry, which began thus: ‘Ce mot est grec, ejnevrgeia, actio, efficacia: il a, dans ce sens, pour racines, ejn, (in, dans, en), et – e[rgon (opus, ouvrage, œuvre). ‘The distinction here made between the Greek words along with their Latin and French equivalents recurs in Humboldt’s distinction between ‘Tätigkeit’ for energeia and ‘Werk’ for ergon. So here we are back with the famous Humboldt passage that was quoted in the first paragraph of this essay, and what I called its package-like quality should now be obvious. Energeia is the pivot of Humboldt’s notion of language, but the conception of that pivot was not his own. He owed it to Garat, to Condillac, to Diderot, and to Beauzée’s formulation. When speech and language popped into the center of Humboldt’s anthropology, the depression lifted, and he was successfully launched on his life’s work. The intimate relation between thinking and speaking is a principle that Garat proclaimed in all his writings on language. In the ‘Discours préliminaire’ to the Academy’s dictionary he wrote that once languages had come to be considered in a philosophical light, ‘on a vu...que les mots ne nous servaient pas seulement, comme on le croyait, à nous communiquer nos pensées, mais qu’ils nous étaient nécessaire pour penser’ (p. iv). Words and language are not merely, as the rationalists held, the instrument of communication, but a fortiori the creative and necessary instrument of thinking, or, to use the Greek word for instrument, the organ of thinking. There is space for two concluding observations. In a book published as late as 1996, one reads that, in Humboldt’s view, in order to provide the study of language with ‘eine neue, angemessenere Grundlage, darf man die Sprache nicht als Instrument, sondern sie als Organ betrachten. Diese Gegenüberstellung von Metaphern bezeichnet die Kluft zwischen Humboldts Sprachansicht und jener traditionellen Auffassung, die im Rationalismus der Aufklärungszeit ihre letzte Ausformulierung gefunden hatte’ (Di Cesare 1996:278).

This seems to be the conventional wisdom in Humboldt scholarship, and it is false. As I said at the outset, for a strict Cartesian, speaking is merely the instrument for communication of the pre-fabricated discourse of the mind, but this conception is overturned by the anti-rationalist argument of Condillac’s Essai in favor of the aesthetic and creative conception of the nature of language. The making of speech in interpersonal dialogue is the primal human creation (Essay 2001:6; Essai 1973:101-2). Furthermore, Humboldt knew that the rationalist doctrine had been overcome, and he said so: “Der Irrthum ist längst verschwunden, dass sie [die Sprache] ein Inbegriff von Zeichen von, ausser ihr, für sich bestehenden Dingen, oder auch nur Begriffen sei’ (GS 7:621, quoted

204

HANS AARSLEFF

with comment OL 1988:xlii). This is a succinct statement of the position that Condillac attacked when he showed that there can be no thinking and knowledge without discursivity, no discourse without speech, no speech without sociability and dialogue, and no dialogue without creativity. I have pointed out that Humboldt’s fresh thinking did not suddenly arise from his chance encounter with Basque, even though he more than once chose to give that impression. Still, against the evidence, his words are still trusted. It has been claimed that Humboldt’s new orientation, ‘ergibt sich ganz aus den in Anthropologie und Aesthetik angelegten Gedanken und ohne erkennbaren neuen Anstoss von Aussen’ (Borsche 1981:202). To the contrary, as we have seen, there was plenty of outward impulse, and Humboldt knew well where it came from. This goes, we have seen, for the anthropology, but it is also true of the related aesthetic conception of language – the energeia conception – that is so magnificently presented in the Lettre sur les sourds-et-muets and in many of Diderot’s other writings. The case of Garat shows what attention to the richly documented context will do for the intellectual history of Humboldt’s sources and developments. It also bares the impotence of much that passes for secure conventional wisdom in Humboldt studies.

REFERENCES Abbreviations

GW GS EMGL Essai 1973 Essay 2001 FG OL 1988

Humboldt 1841-52 Humboldt 1903-37 Beauzée & Marmontel Condillac 1973 Condillac 2001 Humboldt 1960-81 Humboldt 1988

Aarsleff, Hans. 1988. Introduction. OL 1988: vii-lxv. . 2001. Introduction. Essay 2001: ix-xlv. . Forthcoming. Philosophy of language’ in Cambridge History of Philosophy in the Eighteenth Century, ed. Knud Haakonssen. Baudiffier, Serge. 1982. ‘Diderot et Condillac.’ Jean Sgard, ed., Condillac et les problèmes du langage, 115-36. Genève/Paris: Slatkine. Beauzée, Nicolas. 1782. ‘Energie’. EMGL 1:713a. . & Jean-François Marmontel, eds. 1782-86. Encyclopédie méthodique. Grammaire & Littérature. Paris: Panckoucke. Becq, Annie. 1994 [1984]. Genèse de l’esthétique française moderne 1680-1814. Paris: Albin Michel. Borsche, Tilman. 1981. Sprachansichten. Der Begriff der menschlichen Rede in der Sprachphilosophie Wilhelm von Humboldts. Stuttgart: Klett-Cotta. . ed. 1996. Klassiker der Sprachphilosophie, von Platon bis Chomsky. München: C. H. Beck.


205

Chouillet, Jacques. 1973. La Formation des idées esthétiques de Diderot. Paris: Armand Colin. . 1984. Diderot: Poète de l’énergie. Paris: PUF. Condillac, Etienne Bonnot de. 1973. Essai sur l’origine des connaissances humaines. ed. Charles Porset. Paris: Editions Galilée. . 2001. Essay on the Origin of human Knowledge, ed. & tr. by Hans Aarsleff. Cambridge: CUP (Cambridge Texts in the History of Philosophy). Di Cesare, Donatella. 1996. ‘Wilhelm von Humboldt (1767-1835).’ Borsche 1996: 275-89. Freese, Rudolf, ed. 1986. Wilhelm von Humboldt. Sein Leben und Wirken, dargestellt in Briefen, Tagebüchern und Dokumenten seiner Zeit. Darmstadt: Wissenschaftliche Buchgesellschaft. Garat, Dominique-Joseph. 1783. ‘Lettre sur Bayonne et les basques.’ Mercure de France, 8. Feb. pp. 55-77. . 1785a. ‘Cantabrie.’ EMGL 2:446-47. . 1785b. Review of Rivarol 1784. Mercure de France. 6 Aug. pp. 1035; 13 Aug. 63-73. . 1798. ‘Discourse préliminaire.’ Dictionnaire de l’Academie française, 2 vols. Fifth ed., 1:i-x. Howald, Ernst. 1944. Wilhelm von Humboldt. Erlenbach-Zürich: Rentsch. Humboldt, Wilhelm von. 1841-52. Wilhelm von Humboldt’s Gesammelte Werke. 7 vols. ed. Carl Brandes. Berlin: G. Reimer . 1903-37. Wilhelm von Humboldts Gesammelte Schriften. 17 vols. ed. Albert Leitzmann et al. Berlin: Behrs Verlag. . 1960-81. Wilhelm von Humboldt. Werke in fünf Bänden, eds. Andreas Flitner & Klaus Giel. Darmstadt: Wissenschaftliche Buchgesellschaft. . 1988. On Language. The Diversity of human Language-Structure and its Influence on the mental Development of Mankind, tr. by Peter Heath, with Introduction by Hans Aarsleff. Lammers, Wilhelm. 1935. Wilhelm von Humboldts Weg zur Sprachforschung 17851801. Berlin: Juncker und Dünnhaupt. Menze, Clemens. 1965. Wilhelm von Humboldts Lehre und Bild vom Menschen. Rattingen: A. Henn-Verlag. Mueller-Vollmer, Kurt. 1993. Wilhelm von Humboldts Sprachwissenschaft: Ein kommentiertes Verzeichnis des sprachwissenschaftlichen Nachlasses. Paderborn: Schöningh. Rivarol, Antoine. [1784]. Discours sur l’universalité de la langue française. Paris: Classiques Larousse, n.d. Seidel, Siegfried, ed. 1962. Der Briefwechself zwischen Friedrich Schiller und Wilhelm von Humboldt. 2 vols. Berlin: Aufbau-Verlag. Skinner, Quentin. 2002. Visions of Politics. 3 vols. Vol. 1 Regarding Method. Cambridge: Cambridge Univertisity Press. Tully, James, ed. 1988. Meaning and Context. Quentin Skinner and his Critics. Princeton: Princeton Univertisity Press.

206

207

ON THE NOTION OF STRUCTURE AND STRUCTURALISM IN BRAZIL ENI P. ORLANDI Universidade Estadual de Campinas Considering structuralism as one of the most expressive theories in the study of language, and the notion of structure as a notion that marks the history of science, affecting the relation among disciplines, I will situate structuralism in Brazil through texts and authors that, in our history, characterize our tradition of language studies. I will try not to repeat here what I call integrated bias of file reading in science: I will attempt to avoid the interpretations that attribute to a founder ideas that are only possible now, with a nonlinear development, both in time and space. I prefer to make my position in history of science explicit and try to interpret what is characteristic of the different authorships. My objective is, therefore, to elaborate some meanings of Brazilian structuralism. Since my perspective is that the history of science occurs simultaneously in various places, with its different production conditions and its different effects on society in politics and history, I will attempt to show how this movement of ideas took shape and produced its effect in Brazil without digressing from occurrences in other places. As Puech (1995) said , and that is no different in Brazil, structuralism is exactly this “precipitation” that, at the end of the ‘50s, allowed the hope for a unity in human sciences to sediment, even beyond the scientific culture/ literate/ natural cuts, recomposing relations among disciplines. As the same author says (1995), we are facing the challenge of reasoning and the permanent need to renew the forms and the representations of rationality. However, if Puech (1995) is going to distinguish linguistic structuralism, structuralism and philosophy, structuralism and materialism to comprehend them, our route will be a different one, also passing similar places, but taking us in another direction. And this is perhaps due to what can be denominate more appropriately as the history of science, which is not indifferent to the conditions in which its practice is constituted, is not homogeneous and is sensitive to the scientific processes derived from particular, regional, or better said, territorialized histories. Aside from this, and this is also a reality of all countries in their history of knowledge, science carries along with it aspects wich we can call cognitive, in their generality, others that are social-cultural and disciplinary. The design of the disciplines together

208

ENI P. ORLANDI

with the conjuncture in which science is produced constitutes a tradition, that is, in my terms, a network of affiliation of senses in a memory that cannot be anything but singular in its particular conditions. It is this conjuncture, this tradition, this disciplinary design that I attempted to outline here in a few words, without losing view of the nucleus of my reflection: structuralism. We will see how references to foreign authors, made by our authors, are a manner in which to argument in relation to our own history. They are neither simple influences nor mere reception. They are elaborated in function of a tradition of our own in the study of language. They are forms in which to argue in function of ideas that give the specificity of a Brazilian linguistics tradition in relation to science in general. This leads me to introduce one of our most renowned authors of linguistic history in Brazil, Joaquim Mattoso Câmara Jr. And I introduce him citing one of his texts that is a reflection on João Ribeiro, one of our expressive XIX – century grammarians: (...)I believe it to be of the utmost importance in our present conditions a systematic exam, of a critical and interpretive nature, on the ideas of our most relevant grammarians of the past (João Ribeiro). Linguistic studies in Brazil are at a crossroad, ready to take a new route, and various studies that are appearing give us much to hope for in the future. (...) At the same time, the method of interpretation and grammatical exposition is being perfected, freer of logistic preoccupation, as well as the psychologism that followed it. Therefore, at a moment such as this, it is particularly important to take a stance in relation to our ancestors. (...) Only in this way will we be able to harmonize present thinking with that of the past, a condition essential to the development of a well-rooted linguistic science among us and without a problem of continuity in time and that is, at the same time, imperative to avoid imitation of foreign doctrinary thought, elaborated on the basis of other languages that (such as the behaviorism of Bloomfield) does not coadunate with our experience on this particular point (Mattoso Câmara: 1972).

We see here auto-reference to a work model in which criticism, singularity and affiliation are presented. As we will see in the continuation, Mattoso Câmara Jr will be a strong critic of psychologism and, above all, of mechanicism and Bloomfieldian behavior although, in the same text, he will say that it is a quality of João Ribeiro to establish, “according to the present descriptive orientation, the separation between the morphic plan and the phonetic plan, fleeing from the “confusion of levels” of which Bloomfield and his disciples were so cautious”. This means that it is not a question of dogmatically adhering or refuting but of reflecting through the authors in function of a history of our own, on knowledge in which, not infrequently in face of our own tradition, inaugurated with our grammaticization (S. Auroux 1992) in the XIX century, Mattoso will relate grammar and linguistics: Descriptive grammar (...) is part of pure linguistics. Therefore, as every pure and disinterested science, linguistics has on its side a normative discipline that is part of what we can call applied linguistics for a purpose of social behavior. (...) the fact is that normative grammar

STRUCTURE AND STRUCTURALISM IN BRAZIL

209

depends on synchronic linguistics or, in short, descriptive grammar, not to be capricious or counterproductive (Mattoso Câmara:1970).

And he goes further, criticizing the fact that many grammarians do not distinguish between grammar and style and mix “the expressive and personal traits of a given author (...that) are presented as imperative formulas of well speaking”. As we can see, our linguist is zealous of our tradition, but knows the foreign authors and also works in the interior of dichotomies that are the nodal point of our exposition on his authorship relative to structuralism: grammar/style, morphology/phonetics, description/ normativity, but also synchrony/diachrony, mentalism/mechanicism, as we will see further on. It is there that he clearly takes a stance: situating himself in relation to American structuralism, he will favor Sapir and his mentalism (whose work he translated into Portuguese), moving away from Bloomfield and his mechanicism. We know the extension of what is called structuralism; extension from the point of view of its importance, of its expansion in space and time and its amplitude in the field of sciences. Mattoso Câmara Jr (1967) gives a definition by Hrabák: Structuralism is not a theory or a method; it is an epistemological point of view. Part of the observation that every concept in a given system is determined by all of the other concepts of the same system and alone signifies nothing. (...) The scientific work of structuralism is the synthesis of the romantic vision – whose cognitive basis is the deduction based on a philosophical system that classifies and evaluates the facts a posteriori, and the position contrary to positivism that, on the contrary, constructs its philosophy based on the facts that it proved by experience. (...) In the same manner that no concept is unequivocal before being integrated into its particular structure, the facts are not unequivocal in themselves.

We can say that this structuralism, thus defined, covers everything (this is what I call generalized structuralism). And this occurs in all places where science is done. In this sense we pass by the common place (and here I use this expression in its effects of senses): the Saussurian heritage, the structuralism of Prague, of the Americans, of the Glossematics of Hjelmslev etc. And names such as Jakobson, Lévy Strauss, Troubetzkoy, Sapir, Bloomfield, and Martinet will certainly be present in any history on the subject. But they are not present in the same manner. And it is this difference of manners which interests me to understand in Brazil. I also like to keep Saussure, or better, the text of the Course as a memorable place, as we need these representations in science. And I keep Saussure as the “place of origin and beginning of the closure of regression in time and dispersion in space” (Puech,1995). This is what I have called, à quelques différences près, the founding discourse: that which opens discursiveness and creates, at the same time, a tradition, where the senses make other senses, in this case, that of General Linguistics in relation to Compared Grammar. Saussure is the author of this discourse. We know, by discourse analysis, that Saussure, as founder, is also the work of other authors (external evaluation, prestige, legitimization, affiliation, tradition, heritage, and projection on the horizon of science). But we cannot deny that in him, epistemologically, an object, a theory is constructed.

210

ENI P. ORLANDI

And this without forgetting that Saussurian linguistics is made present by structuralism. And, in our structuralism, we build our relations with Saussure on this side of the Atlantic. 1. How is this history made in Brazil? In Brazil, although from early on these authors were in general cited here, the structuralism to which our authors were initially related is the American structuralism: Bloomfield, Sapir – more Sapir than Bloomfield. And various citations by Jakobson that ally in their structuralism two striking elements: communication and functionalism, plus a theory for sounds. Next, to the European side, to Martinet (and his functionalism). Why? To answer this question, I am presenting my hypothesis in this text. I believe that in Brazil I can mention two crucial moments: one is the formation of linguistics by structuralism, connected to authors who preserve the synchrony/diachrony relation, internal form/external form, language/culture, sound/senses and another later one, in which the investment is strongly on synchrony, form and language. At this moment, Saussure and Hjelmslev appear, as well as Jakobson, Martinet, Greimasian semiotics, Pottier and Coseriu accompanied by a strong presence of the formation in historic linguistics in the style of Benveniste (and his work on the Indo-European). I will discuss this history based on texts. I selected texts by Mattoso Câmara Jr (1972) (in Dispersos) and one by T.H. Maurer (1956). In addition, a special issue of the journal Tempo Brasileiro on structuralism called my attention, the main article of which, by Mattoso Câmara Jr (1967), concerns Linguistics, which is in a manner an inaugural synthesis of the presence of structuralism in Brazil at its beginning. 2. A scientific territory of our linguistic tradition: Rio de Janeiro Based on the reading of Mattoso Câmara (1967), we can conclude that the arguments were organized to guide thinking of an authorship in which the question of the senses, the culture and the history could be elaborated, based on the principles of structuralism. In other words, in this perspective, synchrony and diachrony are distinct but articulate themselves in function of the notion of structure. Around this, varied considerations are woven regarding the psychological, the philosophical, etc. Respecting the structuralist principles that separate the organism and the culture, that see the language as fact/structure/institution in synchrony, there is the relation sound/ senses, the relation form/function, the principle of communication, where phonology is the model, having the rest beside it. (A variation, coming from the other side, thinks the form/material relation). Bloomfield, cited many times, is however criticized for his mechanicism.


211

In general, we can affirm that, in Brazil, there is a certain reluctance to leaving mentalism. The sound/senses relation is thought in relation to the phonic material and the semantic structure (meaning: ethnolinguistics, cultural anthropology), a step from diachrony. When this separation, as in Bloomfield – who is principally criticized for behaviorism, the preponderance of the external over the internal of the subject – drains the semantic content, it is avoided, as is Hjelmslev for his excessive abstraction (linguistic algebra), that leaves the substance aside (the contents of thought). Sapir, however, accepts these articulations, as does Jakobson, although diversely, who speaks of a diachrony made of the description of states of structure. A strong example, given by Mattoso Câmara Jr, is Meillet, a disciple of Saussure, who maintains a relation with history. In my interpretation, these authors maintain the notion of history in the XIX century as a reference, although the practice of structuralism, in a certain manner, makes them surpass it. Theoretically, this passage is made explicit (moments that permit the reinvention of the past: history and epistemology and science and art) by the Hjelmslev group, which Saussure himself read in a more radical manner, I might say, as Harris did, who was a disciple of Bloomfield. Mattoso Câmara Jr did not neglect the history/culture relation or the poetic relation (and this is in relation to the affiliation with Sapir and Jakobson). He criticizes the manner in which the notion of system dominates that of structure. According to him, Structuralism, submitted to the notion of system, closes, abstracts and leaves no place to give greater importance to the relation of synchrony with diachrony. According to the interpretation of Mattoso Câmara Jr, the notion of structure, although it has appeared in an implicit interpretation by Schleicher, based on the comparatist typology of Bopp, is lost due to his reification of language considered “literally a living organism”. For this reason, according to Mattoso Câmara Jr, the most positive origins of the structural conception of language are found in the theory of form that Humboldt developed. In the interpretation of Mattoso, Humboldt took the form as an ideal configuration and with this created a mentally existing object, not having the concrete existence that Vossler refuted in the implications of the organicism of Schleicher and in the positivistic comprehension of a language as the sum of concretely considered linguistic facts. But there is the advantage in relation to Vossler of not being anti-structuralist; contrary to this form, Humboldt’s is a structural concept. For Mattoso Câmara Jr, the internal form/external form distinction is the first nitid and coherent affirmation of linguistic structuralism. And he adds more and this, in the direction in which I want to develop my reflection, is very important. He states: “we may say in passing that in this initial manifestation the semantic structuration was admitted as a prior denial to the arbitrary identification that is made at times between structuralism and anti-mentalism”. In this statement we see the knot of the question formulated: the relation of linguistic structuralism, of semantics and of mentalism, which maintains the relation with history (poetic and cultural).

212

ENI P. ORLANDI

Mattoso Câmara Jr then affirms that “the first rigorous and conscientiously structuralist position is that of Saussure”. To refer to this, he chooses to mention Saussure’s affirmation on “the states of the language”. In this, Saussure moves away from the neogrammarians who refused this concept because, to them, any pause to take permanent facts into consideration was artificial and anti-scientific. Continuing, Mattoso Câmara Jr explores the distinction made by Saussure between diachrony and synchrony. It is here that the notion of system enters – through the relation with the notion of Whitney’s social institution, present in the Course. The notion of system is contained more in the notion of institution than in the notion of organism and it is playing with the idea of system and structure, insisting on the arbitrariness of the linguistic phenomenon, by a reflection that comes from a sociological comprehension – presence of the French sociology in which the Durkheim school prevails – and in which, according to Mattoso Câmara Jr, Saussure eviscerates the concept of system for the language on structuralist lines for the first time in linguistics. The existence of language as a group of relations supports a position on the equivocality of facts (Hrabák, above). These are the relations that Saussure considers in the institution that is language and not the empirical facts or, in other words, as Mattoso Câmara Jr states – “it consubstantiates the facts arising out of existing relations. However, the relational concept in Saussure does not appear in terms of structure but as a closed system: “la langue est un système où tout se tient” ”. According to Mattoso Câmara Jr, this interpretation of language as a system is what takes Saussure to different doctrinary positions, depending on his situation on the plane of synchrony or diachrony. In historical linguistics, as Jakobson also states, he remains fundamentally neogrammatic and changes position when he speaks of synchrony. Mattoso Câmara Jr states that if we consider the concept of system as a complete organization with sufficient elements, we soon realize that, strictly speaking, we cannot obtain an explanation for a change from it. Change, viewed from the perspective of neogrammatics, will always be an isolated fact, and with it, we pass from the system to chaos. And, taking sides, Mattoso Câmara Jr states: “From this specific point of view, the Saussurian thought has caused some uneasiness”. According to Mattoso Câmara Jr, to overcome this uneasiness it is necessary to introduce a notion of structure lato sensu without the implications that the notion of system (as in Saussure) necessarily requires. Saussure overcame naturalistic phonetics by introducing a grammatical order in what seemed to be a capricious and inconsistent variety. He did not consider the isolated facts of minimum vocal emissions but only the resulting interrelations and structuration. In the same manner, for Mattoso Câmara Jr, the Saussurian structuralism unfolds on the planes of associative relations and syntagmatics, which neglect to emphasize the elements themselves to consider the points of relation, such as paradigms and, in the same manner, privilege the concept of syntagm which considers the linear structuration. Mattoso Câmara Jr continues stating that Bally, disciple


213

of Saussure, makes the structural nature of the syntagm concept emerge, at the same time in which he explores the field of affection, left aside by his master, creating his stylistics. The importance given to the syntagmatic structure leads to the interpretation of the phrase as the genuine linguistic reality. Therefore, as Mattoso Câmara Jr says, the phrasal structure is, in last analysis, a relationship of functions, arriving at a modality of linguistic structuralism that is functional, and its varieties, quite familiar to Brazilian authors. Advancing in his interpretation of structuralism, Mattoso Câmara Jr speaks of the school of Prague and refers to B. de Courtenay in his effort to classify the phoneme as an abstract linguistic reality but criticizes him for his psychologism. Always emphasizing the necessity of appreciating the importance of the interrelations of language, he will show the relevance of authors such as Jakobson (who was his professor in the USA), Karcevsky, Troubetzkoy and Mathesius. He emphasizes the fact that structuralist Czechs practice a conception of structuralism that is ample, as stated by P. Garvin, and is not limited to linguistics. However, “by linguistics, the Circle gains European dimension”. This interests us since in Brazil there will be a research to deeply explore these relations of linguistics with poetry and anthropology, working closely with the question of communication and the functions of language at USP in the decade of 1960, when Jakobson was received in Brasil. According to Mattoso Câmara Jr, Jakobson simplified the classification of the phonological oppositions with their reduction to binary conjugates and with the concept of “mark”. This new structural comprehension opens the way to a universal typology, creating particular, contingent structures for different languages. We cannot forget that Mattoso Câmara Jr is the linguist who practices systematically and with excellence the description of the Brazilian languages. This brings us then to what causes a strong impression on Mattoso Câmara Jr: The Sapirian structuralism. As the author says, this structuralism, strange to the European Saussurian structuralism, is created in the USA; it is the linguistic structuralism of Sapir. Instead of connecting him directly to Boas (who is very empiric), with whom Sapir was intimate, Mattoso Câmara Jr, in a gesture of interpretation that is charged with his affiliation to a Brazilian tradition, ties him preferentially to Humboldt, who, according to him, had inspired Sapir in a way of seeing language in the face of culture and the organization of thought. Humboldt’s concept of form is, according to Mattoso Câmara Jr, the leitmotif of Sapir in the interpretation of language. What is interesting then are the patterns that make the element a linguistic form. The comprehension of the form, that is, a pattern in linguistic concepts, opens the way for a structural semantics, although it has not been carried out by Saussure’s followers, nor by Sapir, but permeates his structuralism. What separates Mattoso Câmara Jr from Bloomfield is the structuralism confined to Humboldt’s external form (in Bloomfield, supported by behaviorism). In general, he says that the linguist reserved himself for the study of phonic material and its combinations as being the language in the strict sense, and banished semantics to

214

ENI P. ORLANDI

cultural anthropology. Cultural anthropology took a structuralist orientation. And the structuralization of culture will come together in the structuration of languages through the interdisciplinary study of ethnolinguistics, which is, according to Mattoso Câmara Jr, the common denominator between anthropologists and linguists in North America. However, European structuralism approaches semantics within linguistics (notion of “semantic fields”). This American linguistics, be it Sapir’s or Bloomsfield’s, has in common the weight of the interest of Boas in American indigenous languages and sees in linguistics a wide anthropological study that, however, sidetracked in the sense of a structural line, says Mattoso Câmara Jr. To my thinking, this is a sign that in the scientific territory in which the linguistics of Rio de Janeiro is constituted, the background is anthropologic, when not pragmatic. In relation to the tradition that is nourished in the scientific territory of São Paulo, in at least one of the branches, an idea of order, subject to flaws, is developed. When thinking the language that leads to the comprehension of an object as discourse is not an anthropological object such as speech (Saussurian or the notion of speech in the American tradition) and does not inscribe itself, in the notion of history in the XIX century. It is clearly post-structuralist, as we will see further on. Returning to Mattoso Câmara Jr, we can perceive that, going in the direction of abstraction, he accentuates in Bloomfield – who, more than by his highly praised behaviorism, is known for considering the relational nature of the language, with disciples such as Harris – the exclusion of the semantic done by the generalization of his distributional method. Mattoso Câmara Jr then recals that, in the abstractionist line, the recourse to notations of symbolic logic becomes usual in the North American descriptive linguistics. But what interests Mattoso Câmara Jr is the fact that a language is seen as a hierarchy of structures or levels (phonology, syntax, semantics) that should not interfere with one another. Declaring himself against the relation of a language to the concept of system only, and assuming the perspective of Humboldt, Sapir and even Saussure, he affirms the search for objective and real structures. And this will be his work in the description of Portuguese. He refers favorably to Guillaume, saying that it was he who tried to “clearly and coherently create a structural linguistic psychology”. And he shows how, leaving psychologism, he proposes a structuralism of the psychological. Guillaume, says Mattoso Câmara Jr, distanced from psychologism and proposed a structuralism of the psychological, putting ahead of the linguistic structure a psychic structure which is his object: “The external forces, of historical-social order, can break the linguistic system but are only arbitrary in reference to the system at the first moment because they are soon integrated into the system”. He tries to go beyond two difficulties of the linguistic structuralism in the face of history, according to Mattoso Câmara Jr: by the conciliation between the idea of the system and the idea of change and between the external change and the internal change


215

which, since it comes from the social, begins to participate in the system. Sapir’s mentalism, according to Mattoso, delved into the subconscious and the intuitions of the subconscious to find the roots of the language. He longed for a deep psychology that could serve as a base for linguistics since, according to Sapir, Current psychology does not seem exactly adequate to explain how these formal submerged systems that reveal themselves in the languages of the world are constituted and transmitted. (...) Everything indicates that the languages must be cultural deposits of a vast and complete network of psychic processes that have not yet been clearly defined to us.

As a matter of fact - here one more of the unfoldings that linguistics in Brazil will suffer is outlined: on the one side, the relation language/culture (unconscious, internal/ external form) and on the other, the relation with a memory that is not just cultural but of another nature - the relation language/symbolic/history which introduces the notion of material form. Hjelmslev belongs to the group that separates the linguistic from psychology and sees a system that is self-sufficient: a three-part concept is established in which the system transcends the particular consciences of the speakers of the language and that only the linguist can comprehend through objective hypotheses, exempt of preconceived ideas. As can be imagined, Mattoso Câmara Jr criticizes him for his exaggerated abstraction. Mattoso Câmara Jr is decidedly Sapirian. This is a double authorship – Bloomfield and Sapir – who dominate the American scene and with whom our first structuralists, via Mattoso Câmara Jr, relate. But not exclusively since even Mattoso Câmara Jr has an interest in Martinet and his functionalism. The concept of linguistic structure is essentially connected to the comprehension of the functioning of language, he says. Structural relations do not exist in the air. They occur according to the role their oppositie terms play in the process of communication. Therefore, he cites Martinet congenially, for whom “all structuralists take into consideration the function of the linguistic units” (1964). And he cites Martinet not just congenially. In his Estrutura da Língua Portuguesa (Structure of the Portuguese Language) (Mattoso Câmara 1970), he divides his work into chapters whose titles are: Part one “The Second Articulation or Phonology” and Part two “The First Articulation or Morphology”, citing Martinet explicitly on page 23, where he speaks of the technique of description. Mattoso Câmara Jr refers next to Chomsky but, at this point, we arrive at the limit of the form of structuralism in which we are thinking and stop here. But not before referring to what he is going to call diachronic structuralism. If Saussure, the first ostensive structuralist, was only a structuralist as a synchronist (as were Bally, Hjelmslev, Firth and a large part of the Bloomfieldians), it should be noted that someone such as Meillet, disciple of Saussure, took advantage of the Saussurian structuralism for a new vision of diachrony. He interprets the linguistic evolution as a passage from one state of language to another instead of the

216

ENI P. ORLANDI

continuous flow in which there would be no opportunity for structuration (as in the case of the neogrammarians). In the same manner, says Mattoso Câmara Jr, Sapir involved the history of language in a wide structural vision. For him, language obeys a “deviation” and is a structure in permanent elaboration. From early on, Jakobson also claims the need to study phonetic evolution in structural lines. Once again, Mattoso Câmara Jr advocates against the idea of system and prefers that of structure since “in the face of the concept of strictly considered linguistic system, it would be difficult to explain diffusion” (here he refers to Jakobson’s concept of diffusion). Once again, he refers to Martinet for the importance he gives to historic phonology in structural lines, with its three forces: the structural force, strictly speaking, the economy (of psychological order) and the condition of biological order (asymmetry of the phonetic organs). To these three linguistic forces he joins the external ones and admits the importance, as does Jakobson, of borrowing. Finalizing, he cites Braque: “I do not believe in things; I only believe in the relations between them”. This heritage had echoes, as we will see, even in those who, in Brazil, such as I, did not read Mattoso Câmara Jr in their linguistic formation, but read Saussure, Hjelmslev and Martinet. We can say that what we have shown up until now was the Brazilian structuralism produced in Rio de Janeiro (without forgetting that, in the number in which Mattoso’s article on structuralism appears, there is already an article by Miriam Lemle (1967) on “The new structuralism in Linguistics: Chomsky”). 3. The scientific territory of São Paulo The situation that will be experienced in São Paulo is totally different, more specifically at USP. In this work I will give only a outline of a much deeper history. At USP, the Indo-European linguist Thedoro Henrique Maurer Jr. also has a relation with American structuralism: he was a student of Bloomfield at Yale University in 1945 and 1946. Maurer refers to Bloomfield as one of the deepest and most competent linguists of the century, greatly knowledgeable in Sanskrit and the work of Pânini. Maurer (1956) makes this affirmation in a reflection on grammar, description and historical linguistics in an article published in Bulletin 157 of USP’s Faculty of Philosophy, Science and Letters, as an introduction to a work on the description of the phonology of the uaiuai language by W.N. Hawkins. We would like to highlight a few passages from this work of Maurer. His definition of grammar – “description of the categories of a language and study of their functions and use” – and his criticism of the deviation from objectivity of the grammatical studies anterior to structuralism since they dealt with the description through logical and philosophic preoccupations which, according to him, made a perfectly objective analysis of linguistic phenomena difficult.


217

Language analyses that look for categories similar to those that exist in Latin and want to distinguish forms by the simple reason that we are accustomed to them as heirs of a system that came to us from Indo-European, he calls them artificial. We may not have a category in a language but, through it, still express the corresponding notion whenever necessary. He says that the XIX century created historical linguistics with the works of Bopp, Rask, Grimm, Diez and others but, in general, “neglected descriptive linguistics, whose object would be the rigorous analysis and the exact description of the essential elements of phonology, morphology and syntax of a language spoken by a determined group”. This is a definition in the American spirit of language definitions. It does not lack reference to indigenous languages that - as is always said in Brazil - must be described before they disappear and for this require specialists trained in this technique by Americans (presence of the Summer Institute of Linguistics - SIL), represented by Pike and Nida. Maurer observes that “isolated initiatives occurred in the XIX century in this effort of observation based on the admirable method encountered in the practice of Hindus, elaborated more than 2000 years ago”. In 1851, Böhtlinck (European editor of Pânini grammar) applies the descriptive method to a Russian language, iacute. As Leskien proceeds in the field of Baltic and Slavic languages and, at the beginning of XX century, Finck in the same manner describes eight different languages, insisting on the importance of the descriptive study as a base for historical investigation and of philosophical generalizations. We can say, as I. Blikstein (2000), that Maurer had a great presence in the entrance of Linguistics. (...) The chair of Classic Glottology later became Indo-European Linguistics and the classes that Maurer gave, and I say this in an article that I wrote, were, in truth, classes that breathed Structuralism. He was already doing Comparative Linguistics from a structural perspective. (...) His vision was that of American Linguistics.

I agree with Blikstein that works such as the Personal Infinitive of the Problem of Vulgar Latin or the Grammar of Vulgar Latin are historical studies in which description occupies a large space. Prof. T.H. Maurer Jr plays an ample role in the institutionalization of Linguistics, in Brazil, at USP, not only through his relation (which he will relax with time) with the North-American structural linguistics but, above all, due to the fact that he created the institutional conditions for Linguistics (historical, descriptive) to be implemented in Brazil, in his specific case, at USP. This he did by creating courses, contracting personnel, supporting training periods of students abroad, participating in examining boards, and writing introductions to works in this field. He will always have history as his main objective, Indo-European linguistics. Even when speaking of the importance of Bloomfield, he says: “His classic work, Language, constitutes a magnificent exposition,

218

ENI P. ORLANDI

although a bit heavy, of descriptive linguistics, side-by-side with historical linguistics” (Maurer 1956). What he appreciates in this movement of the history of the reflection on language is the method of description that can lead to a scientific knowledge of the language, as he says, “without the interference of the personal interpretation of your own language. In this way, a person can obtain, with the maximum of spontaneity, the image of the language spoken by him”. An expression that he repeats is that, with this method, one can obtain the description, elaborate the phonology, the morphology and the syntax “according to formal and structural elements significant to the language”, without philosophical or even linguistic presuppositions that “may disfigure the real image of the language being studied”. And praising the science of structuralism, he says: “Only in this way can we observe the linguistic facts with the objectivity with which a physicist describes the facts he studies in the laboratory”. According to him, “the descriptive analysis of languages has an extraordinary scientific importance, even for historic-comparative linguistics”. This in an indelible mark in Uspian linguistics: its relation with Indo-European Linguistics, its relation with Philology, but that does not impede the different forms to practice structuralism there. In the ‘60s, in fact, structuralism flourishes at full force at USP. We have then various contributions. And this occurs at the same time in which the institutionalization of linguistic studies is consolidated. If Maurer did not teach Bloomfield, he did however open space for the institutionalization of linguistics – through the Roman Philology chair – by the work of various of his disciples. In this way, we are going to have the practice of Linguistics by professors who, such as Izidoro Blikstein and Cidmar Paes, have classic formation and carry out work dedicated to structuralism in linguistics, and I myself, initially a Saussurian structuralist and, later, decidedly taken by the relation to Hjelmslev’s Glossematics. My works are dedicated to non-contentism, the non-separation between morphology and syntax, the notion of figure, the non-functionalism, the relation between the abstract form and the material form, the text. Izidoro Blikstein will be responsible for the approximation, structuralist and functionalist, of the themes studied in Linguistics, Communications, Literature, Anthropology (with authors such as Saussure, Jakobson, Coseriu, Barthes and Greimas), having done a representative work in the direction of a structuralism that placed linguistics in the relation with other areas of language study, a relation that gave wide space to semiological research. He was at the origin of the preparations for Jakobson’s coming to Brazil, and whose work has resonance even in secondary studies (which shows that linguistic knowledge has repercussions outside of the strictly academic ambit). Cidmar Paes, working with structuralism directed to Phonology and Lexicology, as well as Semiotics, worked with authors such as Saussure, Martinet, Greimas, and Pottier, introducing statistic treatment in linguistics at USP. In this direction, he


219

will develop a whole line of works supported by the method of statistics carried to effect by Z. Zapparoli. Professors such as Diana L.P. de Barros and, later on, J.L. Fiorin will also reinforce Semiotics. Phonetic and phonological studies have their impulse with Francis H. Aubert and Norma Hochgreb. As professors, we all did general, basic linguistics, with authors such as Saussure, Martinet, Jakobson, Pottier and even Benveniste and later returned to our specializations. In this sense, we can say that the work done in structuralism at USP is amply related to European structuralism. I believe we can distinguish two lines in the interior of this structuralism: one who that maintained a preoccupation with the content (function), with a certain form of working with the relation of the language with history (carried out by Indo-European linguistics), of structuralism with functionalism and another that did not. I include myself in this second case. Attracted by the work with material form, I did not remain very long in the dichotomic opposition between form and content. And I think that it was this insistence of mine in thinking a semantic that was neither just abstract nor of content that led me to the discourse analysis of French affiliation in which I situate M. Pêcheux. And if I refer to this, it is to show that authorship in science always has a long route that is not made of “influences” but of reflections within a “tradition” under certain conditions. In this sense, it is necessary to say that the question of materiality was also present in the studies of semiotics (in which the Greimas’ affiliation is situated) at USP (Diana L.P. de Barros and J.L. Fiorin). Therefore, we can say that there is an initial structuralism (of the ‘40s and ‘50s) that I exemplified here with Mattoso Câmara Jr in Rio de Janeiro and T.H. Maurer Jr in São Paulo) and a structuralism of the ‘60s and ‘70s at USP that are differentiated by their relation with theoretical affiliations internal to structuralism, quite distinct in their manners of inscribing linguistics in the Institution and that will determine, given the nature of structuralism itself (generalized or specific), different disciplinary profiles in the interior of human sciences. What I have said with these points and brief references on structuralism at USP, due to a practical question of time and space for my exposition, remains as a suggestion for a deeper investigation, which I give to my colleagues and that I intend to continue on another occasion. 4. The Inflections of the Science and the Political In my opinion, something that refers to politics and more directly to the history of science should be said here to conclude this exposition - something on the history of science and its interpretations. Uspian structuralism brings with it a fine inflection of thought on politics that can be comprehended if we think that, in the elaboration of a fundamental moment in the science of language in Brazil and of its institutionalization, it indicates

220

ENI P. ORLANDI

modernity, a futurity that resists, including the dictatorship that we had at the time (dictatorship realized by democratic capitalism), surprising it and surpassing it. A difference in relation to the European position that, from Dosse’s point of view, considers the ambiguity of the student movement of ’68 to be like a continuation of structuralism by other means or, to the contrary, like the first symptom of the reaction of “leftwing” anti-structuralists. I cannot help but remember the interview that Sarte gave to the l’Arche magazine, where he affirms that structuralism was the last barrier that the bourgeoisie raised against communism, with which I strongly disagree, due to my practice at the time in Brazil, more specifically at USP. Ideologically, at USP in the decade of 1960, we were structuralists because we were leftwing, although Philosophy at USP always remained aloof to structuralism, and is until today, with rare exceptions. However, what was visible to the linguists was that structuralism was beginning another explicit intellectual history for the study of language (in the widest and most general sense) and for the human sciences, such as the courses of Letters that were being implemented at the time, both in Rio de Janeiro and São Paulo. Going back to the question of the conceptions of science and its representations and interpretations, I would like to affirm that it is not sufficient to introduce the time (and the space) and, treating it as something that has it own objectivity, fill it with representations that are no more than representations of epistemology a posteriori (S. Auroux 1992). Time is history. And, in this history, in which science is the fact – the objects of science, the scientific disciplines are the objects of observation – as P. Henry (1984) says: “the facts demand meaning”. And it is in this meaning-making that the history of science is produced. Talking then of “totalitarianism” in regard to the structure is not something that can be done automatically and outside of a history that connects subject, language and ideology. In Brazil, being a structuralist in the ‘60s and ‘70s meant resisting the irrationality of the dictatorship. Therefore, instead of accepting a (history of) science with its presupposed contents as a consumed fact, we criticize the content and prefer to think this history discursively and then it is not its content but its functioning that interests us. In this way, we do not presuppose its previous existence. There is an effect of prior objectivity but we consider it in its contradictory material objectivity, discursive memory, in short, we observe it in its interdiscursivity. We can, then, appreciate the movement of scientific production, not as something monotonous but, on the contrary, full of presuppositions and implications and very dynamic consequences, frequently controversial. From this perspective, structuralism as practiced at USP (that is not the same as in the ‘40s and ‘50s), awakened persons in an acute manner to criticism of the system, of the social-political structure established and led them to work the discursive side of the possibility of, under censure, signifying in silence. Sign of resistance, of a work of history on history, of an effort to sustain an effective,


221

energetic intellectual life, in an institution dominated at the time by the strong hands of the dictatorship. Which leads me to conclude: contents matter little; it is the very possibility of thought (of thinking) that is at play in the games of scientific history. REFERENCES

Auroux, S. 1992. A Revolução Tecnológica da Gramatização, Campinas: Ed.Unicamp. Blikstein, I. 2000. Uma Visita do estruturalismo: Jakobson na USP, interview given to Eni P. Orlandi, Diana L. P. de Barros, Eduardo Guimarães and J.L. Fiorin, Relatos, 6. Campinas. Henry, P. 1984. “L’Histoire n’existe pas?”. Actes du Cheiron, European Society for the History of the Behavioral and Social Sciences. Lemle, M.1967 “The new structuralism in Linguistics: Chomsky”. Estruturalismo, 15/16. Rio de Janeiro: Tempo Brasileiro. Mattoso Câmara Jr, J. 1967. “O Estruturalismo Lingüístico”. Estruturalismo, 15/16. Rio de Janeiro: Tempo Brasileiro. . 1970. Estrutura da Língua Portuguesa. Petrópolis:Vozes. . 1972. Dispersos. Selection and Introduction of Carlos Eduardo Falcão Uchôa. Rio de Janeiro: FGV. Martinet, A. 1964. Éléments de linguistique générale. Paris: Armand Colin. Maurer, T.H. 1956. “Introdução” to the work of W. N. Hawkins, A Fonologia da Língua Uáiuái, Bulletin 157 of the faculty of Philosophy, Sciences and Letters, Ethnography of Tupi-Guarani 25. Puech, C. 1995. “La Linguistique Structurale, Du Discours de Fondation à l´Emergence Disciplinaire”. Langages. Paris: Larousse.

222

223

LES AVANCÉES DE NOTRE DISCIPLINE (SEANCE PLÉNIÈRE DE CLOTURE D’ICHOLS 2002)

SYLVAIN AUROUX Ecole Normale Supérieure Lettres et Sciences Humaines (Lyon) Nous sommes peu nombreux ici à avoir été présents au premier ICHoLs organisé à Ottawa par Konrad Koerner. Et pourquoi sommes-nous si peu nombreux à avoir été présents ? Parce que c’était en 1978. Et je vois beaucoup de jeunes gens dans cette assemblée. Cela fait 25 ans que nous étions à Ottawa; je crois qu’il est temps, au moment où nous allons décider la localisation du10e ICHoLs, et passer ainsi à une institution trentenaire, de faire quelque peu le bilan de ces 25 ans de travail. S’il n’y avait pas de bilan à présenter cela signifierait que nous avons travaillé pour rien pendant ces années. Je n’y peux croire. Reportons-nous à la situation autour de 1978. Je vais la rappeler brièvement pour ceux qui n’étaient pas là à cette époque. Nous avions eu la création en 1975 par l’initiateur du colloque d’Ottawa de la revue Historiografia Linguistica; en 1978, de la Société d’histoire d’épistémologie des sciences du langage à la Sorbonne. En 1979, nous avons eu la création de la revue Histoire d’épistémologie du langage, publiée par cette société. Ces institutions existent encore aujourd’hui. A partir du colloque d’Ottawa nous avons assisté à une croissance qu’on peut qualifier d’exponentielle. Viendra une troisième revue Beiträge zur Geschichte der Sprachwissenschaft, chez Nodus. On compte aujourd’hui 7 sociétés locales dédiées à notre discipline, la Géorgie, l’une des dernières nées, la Grande-Bretagne, l’Espagne, les Pays-Bas, les Etats-Unis, l’Allemagne et deux sociétés au moins ont déployé des activités consacrées au domaine sans que cela soit leurs préoccupations exclusives, je pense à la société italienne de philosophie du langage, mais aussi à nos amis brésiliens qui nous accueillent aujourd’hui. Il y a des éditeurs qui vivent de nos travaux, je pense à John Benjamins d’Amsterdam, le plus ancien de tous, celui qui publie les actes des colloques ICHoLS, à Gunter Narr, à Nodus. De nombreux autres éditeurs ont tenté leur chance dans le domaine; toutes les collections n’ont pas réussi. Il y a des laboratoires spécialisés. Celui du CNRS en France, dont la création date de 1984. Il y a eu beaucoup de travail en Allemagne autour des chaires de romanistique. Je pense à ce qu’il y avait à Tübingen du temps de Coseriu (1967), mais surtout à ce qu’a développé notre regrettée amie Brigitte Schlieben-Lange, en particulier autour

224

SYLVAIN AUROUX

des idéologues. Je pense au groupe qui travaillait avec Herbert Brekle (1985) à Ratisbonne et aux recherches pour le dictionnaire des linguistes allemands du 18e siècle. Je reviendrai sur cet aspect quantitatif. Je pense à Trèves et à ce qui c’est fait autour de H.-J. Niederehe sur la linguistique espagnole. Je pense aussi, et il faut rendre hommage à cette université, à ce laboratoire, au laboratoire de Campinas, et au projet qui a associé ces dernières années les chercheurs de Campinas (déjà très engagés dans le domaine) et l’université de São Paulo. Je n’oublierai pas Munster qui n’a pas su garder P. Schmitter ou Louvain, un peu plus tardif, qui reprit très vite les problématiques internationales. Il ne s’agit pas de faire un palmarès. Un aperçu quantitatif sera plus clair. En se basant sur les 9 colloques ICHoLS, qui attirent à chaque fois 100 à 150 participants, sur les sociétés savantes, sur les abonnements aux revues, on peut risquer les estimations suivantes: - l’histoire des sciences du langage regroupe à peu près de par le monde 1 000 participants intéressés, dont environ 300 producteurs actifs (je veux dire ceux qui publient des articles, sans appréciation qualitative sur ces articles). - Il y a tous les ans, au moins 100 articles spécialisés, ce qui veut dire en 25 ans 2500 articles sur l’histoire de la linguistique. - Il y a environ une cinquantaine de thèses, ce qui fait à peu près 250 à 300 thèses en 25 ans. - Il y a une dizaine de rencontres annuelles entre les sociétés locales, les manifestations comme celles de nos jours, ce qui donnent en 25 ans 250 rencontres.

On est donc face à un bon quantitatif. Il est désormais impossible de se tenir au courant de l’actualité sans recourir aux services des sociétés savantes, des banques de données et des sites spécialisés. Dire qu’il y a eu une croissance exponentielle, c’est, comme toujours, témoigner d’une institutionnalisation. Comme partout dans ces cas-là, on a vu se dérouler prises de pouvoir, luttes, carrières. C’est la vie. Leur description relèvera (pour peu que cela intéresse quelqu’un) de la sociologie de la science, quand les gens de ma génération livreront leurs archives privées. Mais derrière toute cette écume du développement scientifique, il faut poser la question de savoir quelles sont les connaissances qui ont été produites, poser la question de savoir si cet investissement de nombreuses universités de la planète et de tous ces chercheurs qui ont consacré leur temps et leurs talents à notre discipline, su porter ses fruits. La question que je vais donc essayer de traiter est celle-ci: qu’est-ce qui a changé en 25 ans au niveau du contenu des connaissances ? Je tacherai de rester très général, afin d’éviter de citer des noms. L’aboutissement, c’est aujourd’hui vous par exemple, mais ce sont aussi les grandes synthèses que nous avons vues ou que nous voyons paraître ces dernières années: l’histoire de la linguistique en italien, l’histoire des idées linguistiques en français (une centaine de collaborateurs). Des synthèses multilingues sont en cours, celle de Peter

LES AVANCÉES DE NOTRE DISCIPLINE

225

Schmitter (1982, 1987), qui est extrêmement intéressante, parce que régionalisée : elle ne choisit pas pour totaliser la somme de l’histoire le composant temporel linéarisé, mais aussi le composant spatial et le composant thématique. Une grande synthèse plus classique, je dirai moins bouleversante théoriquement, mais sans doute la plus importante au niveau de la quantité des informations, est en cours de parution chez De Gruyter. Un jeune chercheur qui se spécialise dans l’histoire des sciences du langage doit commencer par lire une somme importante de littérature secondaire. Quand tout a commencé, il y avait bien des choses que l’on ne retrouve plus dans les bibliographies de nos élèves. Reprenez les bibliographies, dans les années 60/70, il y avait un certain nombre de manuels. Le meilleur d’entre eux, Robert Robins (1967) est resté longtemps l’instrument le plus technique, quoique succinct. Robins était encore avec nous lors de notre dernière conférence (ICHoLs 8) en 1999 à Fontenay. Il y avait, le plus drôle parce que le moins intéressant scientifiquement, le Mounin (1967), l’Histoire de la linguistique des origines au 20e siècle. Ce livre entendait couvrir toute l’histoire de la linguistique en prenant pour base de référence interprétative la théorie de Martinet. A telle date (les connaissances empiriques étaient très élémentaires) il nous disait ce qu’il en était des connaissances concernant la première articulation, puis la deuxième articulation, etc. Les jugements de valeur étaient à l’emporte pièce. On peut dire que ce type d’ouvrage ne comprenait rien à l’histoire. Mounin n’était pas un historien, c’était un honnête homme qui racontait des choses, non sans intelligence, par rapport à un certain préjugé. Il y avait déjà plus professionnel, par exemple E. Coseriu. Il y avait eu aussi dans ces années-là, l’ouvrage collectif de D. Hymes qui contient les premiers grands textes méthodologiques de notre discipline, la première synthèse bibliographique de celui que l’on doit considérer comme le documentaliste essentiel de notre discipline, K. Koerner (1978). Ce dernier ouvrage parvient à lister toutes les histoires de la linguistique, mais avec une hypothèse importante élément qui va me permettre d’aborder notre travail. La période retenue tient en deux dates : 1822-1967; la seconde correspond parfaitement au renouveau que l’on vient d’aborder; la première nous dit “l’état de l’art” : à l’époque, on avait tendance à penser qu’en matière de sciences du langage les choses sérieuses ne commençaient qu’avec le comparatisme. Autrement dit dans les années 60-70, quelles étaient les caractéristiques de notre discipline? - premièrement, une vision globale à orientation exclusivement occidentale; on dispose de la même configuration dans l’histoire des sciences ou dans l’histoire de la philosophie qui ont hérité de la tradition germanique l’idée du “miracle grec” et ses conséquences drastiques (il n’y a de science qu’occidentale). - deuxièmement, je dirais une vision construite, en ce qui concerne les résultats et la périodisation, par les linguistes du 19e siècle, qui, en se faisant les historiens des sciences

226

SYLVAIN AUROUX

du langage, ont considéré que l’histoire tout entière avait pour finalité leurs propres travaux. L’histoire a pour cœur et pour point de vue sur l’ensemble des siècles une histoire héroïque du comparatisme. Il en reste encore des traces aujourd’hui, autant chez ceux qui n’ont pas refait leur documentation que chez ceux qui ne recherchent pas les origines du comparatisme lui-même. - troisièmement, des thèses dogmatiques qui engendrent un modèle canonique de développement des sciences du langage : avant le 19° siècle nous sommes dans une période pré-scientifique, la linguistique est une science née au 19e siècle avec la découverte du sanskrit et la naissance de la grammaire comparée et historique, science réorientée en synchronie au début du 20e siècle par Saussure.

C’est cela que nous avons appris, c’est de ce modèle que nous sommes partis. Il n’y avait pas à l’époque d’articulation entre la grammaire que l’on considérait comme étant une discipline purement normative développant (ce qui est incompatible avec la science) et les événements ultérieurs des connaissances scientifiques. On ne se posait même pas la question de savoir ce que signifiait le mot “linguistique”. Il n’y qu’une quinzaine d’années, que l’on dispose enfin d’une histoire un peu fine du terme, et que l’on sait que c’est un mot qui a été inventé (ou remotivé par J. S. Vater) pour désigner l’histoire des langues et leur classification, ce qui ne saurait recouvrir l’ensemble des sciences du langage. On a eu aussi des fantasmes. Ceux d’entre nous qui lisaient l’espagnol, l’espagnol de Montevideo, se souviennent d’un célèbre article sur le logicisme et l’antilogicisme dans la linguistique. Un thème qui était récurrent dans l’histoire de la linguistique consistait à soutenir que tout ce qui s’était fait en contact avec la logique était mauvais. Il a fallu se débarrasser de tout cela pour comprendre que la grammaire n’a introduit la notion de sujet qu’au XI° siècle (évidemment elle l’a emprunté à la logique !) et pour entrevoir que sans la logique de Port-Royal et plus tard l’apport de Frege et Russell, la grammaire n’aurait pas avancé sur la notion de détermination. Je n’ai cité que quelques exemples, mais l’histoire dans son ensemble était constituée de cette façon. A une exception près. Il y avait un ouvrage, qui n’était pas l’œuvre d’un linguiste traditionnel, qui n’était pas l’œuvre d’un historien, et, je dirai, au niveau du contenu, qui est probablement totalement erroné. Il s’agit, vous avez deviné de qui je voulais parler, de Noam Chomsky dont la Cartesian linguistics paraît en 1966. Les historiens érudits des sciences du langage et les philosophes ont montré de façon certainement définitive que les thèses historiques sont fausses. Mais ce n’est pas l’essentiel. Chomsky n’a jamais vraiment lu Descartes, et s’il l’a lu, il ne l’a pas compris : il faut des années pour avancer sur le sujet. La polémique qui entoura le livre, montre la véritable question épistémologique et cette question épistémologique est quasiment comme le point aveugle de l’histoire des théories linguistiques. Quel rapport pouvons-nous avoir entre notre présent scientifique et son passé ? Quelle profondeur pouvons-nous donner à ce passé ? L’intérêt du livre provient de ce que le présent de Chomsky n’était plus le comparatisme, ni le présent de ses héritiers structuralistes. Il n’a pas un projet d’historien : il se cherche des


227

justificatifs pour sa propre polémique parmi ses contemporains. Ce faisant il libère l’histoire de la barrière qu’instaurait le modèle traditionnel entre le “pré-scientifique” et la “scientifique”. Avant lui, on ne se posait même pas la question de savoir comment s’était constitué le concept de “proposition” ou celui de “verbe transitif”. Après lui, on pouvait plus légitimement former le projet d’appliquer les méthodes générales de l’histoire des sciences à toutes les connaissances qui d’une façon ou d’une autre, en quelque lieu que ce soit, à quelque époque que ce soit ont abordé la question du langage. J’exagère sans doute le rôle de Chomsky. On peut considérer que la chose est plus complexe : il a rendu légitimes les études sur le 17° et le 18° siècle (la grammaire générale, notamment), pour le reste le phénomène est plus diffus. Il tient sans doute d’abord à la multiplication des études et à l’accumulation des résultats et, par conséquent, au fait même de l’existence de chercheurs en histoire des sciences du langage. En tout état de cause, je dirai que le développement de l’histoire des théories linguistiques, dans son aspect quantitatif, a amené une croissance de la population qui s’y consacrait, et donc une croissance des connaissances, laquelle entraîne quasi mécaniquement des changements au niveau des contenus de connaissances. Nos recherches ont eu des résultats. Qu’est-ce qu’un résultat en matière de recherche dans l’histoire des théories linguistiques; autrement dit, comment pouvonsnous évaluer les travaux ? On peut présenter une grille d’analyse. Elle n’est sans doute pas parfaite, c’est celle que je propose à mes étudiants lorsque j’accepte de diriger leur thèse. Je leur demande d’apporter quelque chose de nouveau dans un des cas suivants: - éditions critiques; traductions des textes anciens. - connaissance de nouveaux textes. - extraction des faits nouveaux au moment de l’apparition de leur connaissance. - évaluation quantitative et qualitative des productions. - description conceptuelles; évaluation de la structure et de la solidité des représentations. - modélisation des changements théoriques à différentes échelles. - explication des changements. - description du contexte social et du soubasement institutionnel du développement scientifique. - éventuellement, modélisation de phénomènes généraux de l’évolution scientifique.

1.

Quelles connaissances nouvelles avez-vous apportées ?

1 – D’abord l’ouverture sur d’autres traditions. Il est difficile de dire aujourd’hui, que les tablettes de bilingues sumérien/akkadien contenant des paradigmes verbaux ou pronominaux (tournant des troisième et second millénaires avant notre ère) ou les paradigmes égyptiens (1600 av. J.C.), étaient inconnus dans les années 70. Les spécialistes de l’Egypte ou de Babylone les connaissaient parfaitement. Pratiquement aucun historien des sciences du langage n’en faisait état. Alors que dans n’importe quel ouvrage général d’histoire des mathématiques, vous avez toujours quelque chose sur le calcul des aires ou

228

SYLVAIN AUROUX

des volumes en Mésopotamie. Chacun devrait connaître les tablettes grammaticales. Cela nous fait aussi réfléchir à la nature de notre discipline. Il est probable que dans les années soixante, on aurait dit que mettre un paradigme grammatical à côté d’un texte n’a rien à voir avec de la science, alors que, comme nous le montrent l’histoire des mathématiques, c’est ce genre de chose qui est fondamental pour l’histoire de l’humanité. La tradition chinoise n’était connue que par des ouvrages généraux, et je me souviens que dans ces années-là, c’était par les ouvrages de philosophie que des gens comme moi avait accès à la théorie mohiste, voire à l’ensemble des réflexions sur la logique bouddhiste. Mais c’est l’époque, où Joseph Needham lance la publication des études de son équipe sur la science et la civilisation chinoise (1965). La tradition arabe était un petit peu plus connue. Dès Ottawa nous avions des arabisants (Versteegh, par exemple, qui venait de publier son ouvrage très controversé sur l’influence de la tradition grecque chez les arabes). En 1976, Georges Troupeau venait de publier son lexiqueindex du Kitab de Sibawaihi. Le second colloque de la Société d’histoire et d’épistémologie des sciences et du langage, en 80, était consacré à la tradition linguistique arabe. Le sanscrit était sans doute mieux connu à cause des travaux du 19e siècle et d’une continuité de publications importantes tout au long du 20° siècle. Mais les grandes éditions qui rendent Panini et sa suite accessibles aux non-spécialistes datent de la fin des années 80. Et ce n’est que dans les vingt dernières années du vingtième siècle que la tradition tamoul vint s’agréger à nos connaissances communes. 2 – Ensuite une extension considérable du champ de recherche. Nous avons eu de nouveaux objets. Pensez à la linguistique des langues amérindiennes, en 1978, nous n’avions que l’ouvrage de Victor Hanzeli (1969), pour nous faire comprendre que leur connaissance ne remontait pas au tournant des 19° et 20° siècle et aux travaux de Sapir et Boas. C’est dans les années 80, qu’on s’est mis à travailler sur cette question, en recherchant systématiquement des sources. Et régulièrement, maintenant dans les colloques d’ICHoLs, il y a des communications sur la grammaire des langues amérindiennes. Il faut bien se représenter l’énormité du phénomène. Si on s’en tient au seul patrimoine espagnol (excusez-moi, devant cette assemblée de laisser de côté le patrimoine “portugais” et votre célèbre Anchietta), à la fin du 17e siècle on disposait d’une trentaine de grammaires, donc, c’étaient 30 langues de “grammatisées” (voir plus loin sur cette notion); à la fin de 17e c’étaient 80; et à la fin du 18e c’étaient 150. En trois siècles, l’humanité a conquis 150 langues qu’elle a agrégé à un patrimoine commun. En matière de progrès scientifique, peu de disciplines ont connu de tels résultats. Nous commençons seulement depuis une dizaine d’années aussi à entrevoir des éléments extrêmement importants comme le développement de la connaissance des langues africaines. Il y a trente ans quelqu’un comme moi aurait été prêt à soutenir (sans véritable connaissance empirique) que tout avait commencé à la fin du 19e siècle. Mais, vous avez, par exemple, eu dans ce colloque une communication sur un texte du 17° siècle; plusieurs articles ont montré ces dernières années l’existence d’une tradition continue.


229

3 – Enfin, une accessibilité nouvelle des sources, autant que la production de nouvelles sources. L’accroissement du nombre des chercheurs qui se sont consacrés à l’histoire des sciences du langage (je ne suis pas sûr qu’il perdurera dans le nouveau contexte universitaire) nous a conduit à un bon quantitatif. Pensez, par exemple, à tout ce qui a été fait sur l’antiquité, nous possédons des éditions critiques des grands textes (dans quantité de langues : il existe une traduction annotée de Denys le Thrace en serbo-croate), pensez à la quantité de textes du moyen-âge qui ont été sortis des archives, édités, traduits, expliqués, etc. Pensez aux archives linguistiques de la renaissance, aux chantiers sur le 19e siècle, pensez à tout ce qu’a apporté la connaissance de la querelle Meyer/Ascoli à propos des frontières dialectales et, donc, sur la nature de ce que l’on entend par “langue”. Autrement dit, c’est vrai, je crois que quelqu’un qui n’aurait pas vécu ces 25 années, et se retrouverait aujourd’hui avec la connaissance de l’époque, ne s’y retrouverait pas. En particulier, parce qu’il y a des choses nouvelles que l’on peut voir, des questions que l’on peut désormais poser et que l’on n’envisageait pas. - premièrement, un nouveau cadre de référence pour l’histoire globale. A partir de 1982, quand nous avons lancé le projet d’histoire des idées linguistiques, nous avions proposé après bien des discussions, un plan qui a beaucoup circulé et qui est devenu un plan canonique. C’est un plan en trois tomes: dans le premier, on aborde les origines, on met à égalité, les grecs, les arabes, les chinois, les sancrits, etc.; dans le second on traite de la grammaire latine et de son transfert aux vernaculaires, jusqu’aux grandes compilations de la fin du 18° siècle et du début du 19°; enfin, dans le troisième tome on aborde la grammaire comparée et tout le développement de la linguistique moderne. Ce plan connaît des variantes (dans le manuel italien ou dans le manuel de De Gruyter) qui peuvent être significatives, notamment quant à l’évaluation de la place historique de la grammaire comparée. Quoiqu’il en soit de ces questions qui méritent des discussions qui n’ont pas toujours eu lieu, on peut dire que c’est mis en place un nouveau cadre de référence concernant la périodisation des sciences du langage. - deuxièmement, “un acquit pratique”, l’idée qu’il y a de la science du langage dans le long terme. Je dis “acquit” parce qu’on ne conteste plus aujourd’hui qu’on fait de l’histoire des sciences du langage lorsque l’on étudie les dictionnaires de rimes chinois, les Canones de Théodose, ou une tablette “grammaticale” babylonienne. Je dis “pratique” car cela n’empêche pas qu’à l’intérieur de certains des grands manuels contemporains, certains auteurs expliquent encore qu’il n’y a de science du langage qu’à partir du 19e siècle. On dispose d’un cadre de travail, on n’en tire pas nécessairement toutes les conséquences théoriques. Il est clair que nous manquons encore des discussions fines sur la naissance des sciences du langage, comme en disposent les mathématiques ou l’astronomie. Nous n’avons pas non plus tiré vraiment les conséquences de l’histoire comparée des grandes traditions. Ces questions sont pourtant essentielles pour les sciences du langage. Je pense, en particulier, à la question philosophique essentielle de cette discipline qui est celle de l’universalité. Y-a-t-il un rapport fort (comme je le crois) entre la structure des langues concernées et la forme des outils linguistiques construits dans telle ou telle civilisation? Y-a-t-il des liens de type cognitif entre le fait que certaines traditions (en fait, une seule, la

230

SYLVAIN AUROUX

nôtre) a pu aborder toutes les langues de la planète? La tradition chinoise (qui n’a pas connu de grammaire indigène) aurait-elle pu le faire en conservant ses spécificités? - troisièmement, nous avons rendu visible des phénomènes que nous ne connaissions pas. La recherche soutenue qui accumule des résultats rend visible des phénomènes qu’un regard accommodant de trop près ne peut voir. C’est ainsi qu’avant que l’on ait réuni les grands corpus d’inscriptions sur les monuments italiens, on n’imaginait guère ce qu’étaient les langues italiques. Avant l’existence d’une communauté de chercheurs en histoire des théories linguistiques, il était anecdotique de considérer que Donat avait été traduit en syriaque, voire en géorgien, en breton ou en provençal. Nous n’avions pas d’espace de représentation où réunir ces informations. Les historiens des sciences du langage se sont mis à s’intéresser aux premières grammaires ou aux premiers dictionnaires de telle ou telle langue. Nous pouvons maintenant faire des séries. On remarque aussitôt plusieurs choses. D’abord, la multiplication de cet “outillage” des langues (ce que nous appelons “grammatisation”) n’est le fait que de la tradition occidentale; deuxièmement, elle ne devient significative (croissance exponentielle) qu’à partir de la Renaissance; troisièmement, il y a simultanéité entre la grammatisation des vernaculaires européens et celle d’autres langues (par exemple, les langues amérindiennes, voir supra). Autrement dit, nous avons mis au jour un phénomène inconnu, parce qu’invisible.

De cet ensemble de travaux résultent évidemment de nouvelles questions ou si l’on veut de nouveaux problèmes à résoudre. Je n’en citerai que quelques uns, dans le seul but de montrer comment notre discipline évolue dans ses formulations théoriques quand elle progresse dans la connaissance des faits : - nous considérons que la grammaire est un moyen d’apprendre les langues (bien que la thèse soit controversée chez les théoriciens de l’apprentissage, il s’agit de la conception vulgaire du rôle de la grammaire); ce statut ne remonte pas vraiment en deçà de la Renaissance. Auparavant les grammaires (grecques et latines, mais aussi arabe, sanskrite) ne s’adressaient qu’à des locuteurs natifs. Comment en est-on arrivé là ? Pourquoi principalement en Occident ? On a bien entendu des esquisses de schéma explicatif : existence de grammaire latine; rôle du latin dans la constitution du savoir; disparition du latin comme vernaculaire; nécessité d’apprendre le latin comme langue seconde; utilisation des grammaires comme instruments; traduction en vernaculaire des grammaires latines, etc. - nous envisageons que la grammaire est une représentation théorique de ce qu’est une langue (la “grammaire générative” repose sur ce postulat). Or, les études de suivi nous montrent tout autre chose : les grammaires (comme les dictionnaires) sont transférées (traduites) d’une langue dans une autre, notamment avec leur corpus d’exemples; le renouvellement se fait par parties, très lentement. Leur mode d’existence est celui d’objets techniques (les historiens des techniques ont parfaitement décrit ce mode). La grammatisation est essentiellement un transfert technologique. Dans ces conditions, il faut réévaluer bien des jugements négatifs sur les premières grammaires qui adaptent les schéma gréco-latin aux vernaculaires du monde entier (certains d’entre nous parlent de “grammaire latine étendue”). L’histoire de l’évolution et de l’adaptation de ce cadre théorique (y compris aux vernaculaires européens) est à refaire.


231

- nous disposons d’une histoire convenue du comparatisme : enfin W. Jones vint, découvrit le sanskrit et fonda le comparatisme. Malheureusement, on sait que les ressemblances entre le grec et le sanskrit sont évoquées depuis le 16° siècle; on remarque que dès que des connaissances linguistiques de langues apparentées sont disponibles pour une même communauté savante on assiste quasi-mécaniquement à des travaux de comparaison (pour les langues sémitiques dès le 17° siècle, pour les langues caribes ou finno-ougriennes dès le 18°, etc.); si on envisage les travaux contemporains de Port-Royal et de G. Hickes (Institutiones grammaticae anglo-saxonicae et moesogothicae, 1689; remarquez la reprise du titre de Priscien !), on est obligé de penser que la grammaire générale et la grammaire historique et comparée répondent probablement au même problème théorique de ramener le divers à l’unité. Qu’est-ce qui a changé avec Grimm ? Simplement pas la simple découverte de nouveaux faits, mais la réinterprétation théorique de faits dont certains bien connus, avec des compétitions de modèles théoriques (diffusionnisme vs généalogisme) et des reconstructions d’éléments conceptuels (loi phonétique).

Ces exemples rapides n’ont qu’un seul but : montrer qu’en histoire des théories linguistiques, non seulement le paysage a changé, mais surtout que nous avons affaire à des problèmes théoriques spécifiques passionnants. Il y a un travail considérable à faire. Je n’ai que deux regrets. Qu’il me soit permis, pour clore ce survol, de les exprimer. D’abord, le peu d’intensité des débats méthodologiques. Il faut remonter à la fin des années 70 (Persival 1976) ou au début des années 80 (Grotsch 1982, Schmitter 1982) pour trouver des discussions consistantes sur la notion de paradigme (on n’est pas allé très loin) ou le rôle de la narrativité (mais l’introduction de l’histoire sérielle, par B. Schlieben-Lange notamment) n’a guère soulevé les passions. On a avancé sur la notion d’influence, tout en restant souvent bloqué sur des questions de sources textuelles (qui a écrit le premier la formule X) et sans mettre en avant le fait que nous avions affaire à des connaissances. Peut-être manque-t-il dans nos milieux une information suffisante sur l’histoire des autres disciplines (mathématiques, physique, chimie) et de ses méthodes. Mon second regret est peut-être essentiel à l’avenir de notre discipline. Il concerne le peu d’impact de nos travaux sur le mouvement de progression de la science contemporaine. Lorsque nous constatons qu’il a fallu 25 siècles pour grammatiser environ 900 langues, est-ce que ce résultat massif n’a pas d’intérêt pour évaluer la faisabilité de certains programmes de recherche, pour soutenir, par exemple, les grammaires d’unification (aptes à “récupérer” les connaissances traditionnelles), plutôt que certains modèles (la plupart de ceux de Chomsky) qui proposent de tout reconstruire à nouveaux frais ? Le rôle des découvertes empiriques dues à la connaissance de nouvelles langues (je pense à des choses comme l’apparition de concepts, qui, dès le début du 17° siècle tournent autour de l’ergatif, avec la découverte de l’absence d’accusatif en basque et la présence d’un “casus agentivus”), ne doit-il pas être versé au dossier de la question de l’innéisme ou du caractère empirique des sciences du langage ? Probablement eût-il fallu que nous discutions davantage les thèses de Itkonen (1990). Bref, il importe de travailler avec en perspective le fonctionnement de la science contemporaine et ce que nous pouvons en dire. Il faut probablement parvenir autant à des synthèses théoriques lisibles par des non-spécialistes

232

SYLVAIN AUROUX

qu’à être capable de lister et dater les inventions qui ont été faites tout au long de l’histoire en matière de connaissances linguistiques (la découverte des cordes vocales, l’opposition sourde/sonore, les différents cas, l’ergatif, les temps verbaux, la notion d’aspect, etc. ne sont-elles pas des “inventions” au même titre que la chute des graves ou la circulation du sang?). Il en va du statut contemporain des sciences du langage. La science fonctionne avec un discours dénué de référence temporelle à ses actes propres. L’histoire n’y est pas une modalité d’argumentation. On en conclue trop vite qu’elle n’y a aucune place. Je me souviens de discussions avec des collègues linguistes praticiens, lors d’un colloque sur le lexique. Ils m’expliquaient doctement que la démarche historique n’avait pas d’intérêt pour eux car chaque théorie possédait une définition intrinsèque de ses catégories et que lorsqu’ils reprenaient une nomenclature existante (par exemple, celle des parties du discours), il y avait simple homonymie. L’atemporalisme pourtant à un prix exorbitant : l’incapacité de dépasser l’instant. De mes études de philosophe, je me souviens d’une définition que Leibniz donnait de la pierre. Il disait qu’elle était mens momentanea, “esprit instantané”. C’est dire que finalement si on n’a pas d’histoire, si on n’est pas dans la temporalité, on est comme un caillou. Il est de l’essence des activités de l’esprit d’avoir toujours lieu dans le temps. Et quand bien même ce mode d’existence ne se confond pas avec le rapport des propositions à la vérité (mais les sciences ne sont pas non plus simplement des systèmes de propositions), notre tâche intellectuelle, en quelque sorte notre finalité philosophique, est d’éclaircir ce rapport de l’esprit au temps. Je vous remercie. Question d’un auditeur : Puisqu’on est au moment du rendez-vous avec les idées forces, j’en profite pour signaler, pour rappeler, pour dire comme on le sait tous, qu’il n’y a pas d’histoire autonome des cités intellectuelles, celle-ci se rattache à des faits plus matériels. Je me disais, effectivement, qu’il était peut-être temps, qu’on commence à se poser la question de la connexion entre, les mouvements intellectuels dans l’histoire de la linguistique, et la production et la reproduction de la vie matérielle, pour employer des termes qui percutent toujours. Alors je me demande si la question de l’historicisation, c’est-à-dire de l’émergence des disciplines des sciences du langage, des disciplines du langage, de différents contextes de production, est bien avancée. Est-ce que on a de l’espoir ou de nouveaux espoirs, ou un espoir de renouveau dans la direction de cette analyse? S.A.: Oui, bien sûr. L’un de mes grands regrets dans l’histoire des théories linguistiques, c’est de ne pas avoir pu faire l’histoire économique précise du développement des sciences du langage. Mais il est tout à fait évident que le développement des technologies linguistiques est dans l’histoire de l’humanité quelque chose d’aussi important que le développement des technologies, du feu, de l’acier ou des choses comme ça. Et que, notamment, tout ce qui se développe à partir du 15e siècle, l’organisation du pillage du monde par l’occident chrétien triomphant, est connecté au développement de la linguistique européenne (“la révolution technologique de la grammatisation”). Et derrière tout ça, il y a aussi des enjeux idéologiques. Je prends toujours l’exemple qui concerne la différence entre la France et l’Allemagne au


233

tournant du 18e siècle. Vous avez en France un problème de norme, le problème du pouvoir central qui est à l’origine de l’Académie française et de son dictionnaire (1694), totalement synchronique. Vous trouvez une définition de l’usage comme étant quelque chose qui est rapportée à une autorité, qui n’a de valeur qu’instantanée, toujours référée à un centre de décision. En Prusse, comme dans les autres pays de langue allemande, le rapport à la langue “nationale” ne peut fonctionner de cette façon. Là où on cherchera l’unité, c’est dans l’histoire. La grammaire de Grimm est une grammaire des dialectes germaniques, et si vous lisez l’ouvrage de 1948 sur l’histoire de la langue allemande, lisez la préface, vous y verrez comment se définit l’empire (Reich) de la langue allemande. Les enjeux économiques qui se cachent derrière la grammatisation des vernaculaires du monde entier à partir de la tradition gréco-latine ne concerne pas seulement les relations de pouvoir et le pillage du monde. Pensez au développement de l’industrie du livre, au développement de tout ce qu’il y a derrière (la science moderne, notamment). L’équipement grammatical et lexical d’une langue (une langue n’est pas spontanément apte à être le véhicule de la physique nucléaire et de la biologie moléculaire) est un investissement considérable. Il peut rapporter directement. Pensez, par exemple, à ce que coûte au monde entier actuellement la domination de l’anglais. La croissance de l’histoire de la linguistique doit elle-même être reliée aux enjeux économiques de la discipline (pensez au bouleversement technologique que représente l’automatisation du traitement du langage naturel) et à son fort développement dans les années soixante dix. C’est un petit peu comme le fait que dans les années 50 s’est développée à Pittsburgh, par exemple, toute une école d’analyse des sciences de la nature et des mathématiques. C’est le moment où vous avez des programmes à choisir et où vous avez des besoins d’avoir une idée, de savoir où l’on va dans la linguistique. Evidemment, il y a aussi une utilisation rhétorique et opportuniste de l’histoire. On a vu des linguistes qui faisaient de l’histoire pour justifier la domination de leurs hypothèses dans la discipline. Aujourd’hui avec l’histoire, on peut expliquer quand, où, pourquoi, comment, il y a domination et aussi nous pouvons expliquer que toutes les dominations n’ont qu’un temps.

RÉFÉRENCES

Auroux, Sylvain. 1987. “The first Uses of the French Word ‘Linguistique’ (1812-1980)”. Aarsleff, Hans. Kelly Louis G. Niederehe, Hans-Josef. eds. Papers in History of Linguistics: 447-459. Amsterdam: John Benjamins. Auroux, Sylvain. 1994. La révolution technologique de la grammatisation. Liège : Mardaga. . ed. 1989-2000. Histoire des idées linguistiques. 3 vols. Liège : Mardaga. . Koerner, E.F.K. Niederehe, Hans-Josef. Versteegh, Kees, eds. 20002004. History of the Language Sciences. An International Handbook on the Evolution of the Study of Language from the Beginnings to the Present. 3 vols. Berlin/New York: Walter de Gruyter.

234

SYLVAIN AUROUX

Brekle, Herbert. 1985. Einführung in die Geschichte der Sprachwissenschaft. Darmstadt. Coseriu, Eugenio. 1967. “Logicismo y antilogicismo en la gramatica”. Teoria del lenguaje y Linguistica General. 235-260. Madrid: Gredos. . 1969-1972. Die Geschichte der Sprachphilosophie von der Antike bis zur Gegewart. 2 vols. Tübingen: Gunter Narr Verlag. Grotsch, Klaus. 1982. Sprachwissensschaftgeschichtsschreibung. Ein Beitrag zur Kritik und zur historischen und methologischen Sebstvergewisserung der Disziplin. Göppingen: Kümmerle Verlag. Hanzeli, Victor. E. 1969. Missionary Linguistics in New France: A Study of Seventeenth- and Eighteenth Century Descriptions of American Language. The Hague: Mouton. Itkonen, Esa. 1990. Universal History of Linguistics. Amsterdam & Philadelphia: John Benjamins. Koerner, Konrad. 1978. Western Histories of Linguistic Thought. An annotated chronological bibliography. Amsterdam: John Benjamins. Mounin, Georges. 1967. Histoire de la linguistique des origines au XX° siècle. Paris: Presses Universitaires de France. Persival, Keith. 1976. “The Applibility of Kuhn’s Paradigms to the History of Linguistics”. Language. 52: 285-294. Robins, Robert. 1967. A short History of Linguistics. London: Longman. Schmitter, Peter. 1982. Untersuchungen zur Historiographie der Linguistik. StrukturMethodik-theoretische Fundierung. Tübingen: Gunter Narr Verlag. . ed. 1987. Geschichte der Sprachwissenschaft. Tübingen: Gunter Narr Verlag. Troupeau, Georges. 1976. Lexique-index du Kitab de Sibawahi. Paris.

235

INDEX OF NAMES Aarsleff, Hans 198,200,204,205,233 Abaev 93,99 Aimard, P. 159, 164, 165,166 Alkmim, Tania Maria 88, 92 Alpatov, V. M. 94, 99 Althusser, L. 109, 120 Andreev A. P. 95, 96 ,99 Arnauld, A. 41, 43 ,44 ,61,62 Auroux, Sylvain 3, 15, 39, 40, 75, 85, 86, 131, 208, 220, 221, 232, 233 Authier-Revuz, J. 166, 167 Bachmann, Peter 194 Baddeley, Susan 7, 9, 15 Bakhtine, M.M. 98, 99 Bar-Hillel, Yehoshua 129, 130 Barros, Diana L. P. de 85, 219, 221 Baudiffier, Serge 198, 202, 204 Beauzée, Nicolas 23, 24, 202, 203, 204 Becq, Annie 198, 204 Bennholdt-Thomsen, Anke 194 Bentley, Richard 183, 184, 194 Benveniste, E. 210, 219 Blikstein, I. 217, 218, 221 Bloomfield, L. 110, 111, 112, 113, 114, 115, 116, 119, 120, 145, 208, 209, 210, 211, 213, 214, 215, 216, 217, 218 Bonnet, C. 66, 158, 159, 165, 166 Borsche, Tilman 204 Bosquet, Jean 3, 7, 10, 15, 21, 24 Bouquet, Simon 133, 139 Bowerman, M. 163, 166 Brekle, Herbert E. 101, 104, 105, 184, 194, 223, 234 Brücke, Ernst. 104, 105 Brunot, F. 104, 105 Buarque de Holanda, A. 166 Cagnin, Antonio Luiz 87, 92 Campos, Ernesto de Souza 142, 148 Canguilhem, G. 98, 99 Cauchie, Antoine Caucius 3, 8, 10, 11, 13, 15, 20, 24

Chapelain, Chiflet 61 Chevalier, Jean-Claude 71, 85 Chiflet, Chapelain 45, 46, 51, 52, 53, 61 Chomsky, N. 110, 111, 114, 115, 116, 117, 118, 119, 120, 129, 204, 215, 226, 227, 231 Chouillet, Jacques 198, 205 Christy, Craig 133, 135, 136, 139, 163 Clark, E. 163, 164, 165, 167 Coignard, J.-B. 61 Colombat, Bernard 3, 15, 24 Condillac, Etienne Bonnot de 64, 65, 66, 67, 68, 69, 71, 197, 198, 199, 200, 201, 202, 203, 204, 205 Coseriu, Eugenio 210, 218, 223, 225, 234 Cram, David 124, 125, 129, 131, 139 Cunha, Luiz Antônio 148 De Lemos, C. 157, 166, 167 Delesalle, S. 41 Della Volpe, G. 109, 119, 120, 121 Demaizière, Colette 16 Désirat, Claude 71 Desmet, P. 131, 139 Di Cesare, Donatella 203, 205 Diderot 24, 197, 198, 200, 201, 202, 203, 204, 205 Dodart, Denis 104, 105 Duarte, Paulo 143, 148 Dubois, Jacques 15 Dudley, Homer 101, 102, 103, 104, 105 Dyce, Alexander 194 Eckstein, Friedrich August 194 Engels, F. 98, 100 Eschyle, Sophocle 175, 182 Estienne, Henri 24 Estienne, Robert 3, 7, 9, 11, 12, 13, 15, 21, 24 Fagyal, Zsuzsanna 101, 104,105 Figueira, R. A. 147, 157, 160, 162, 163, 167 Fiorin, J.L. 141, 218, 219, 221 Firth, John Ruppert 127, 131, 215 Flanagan, James L. 103, 104, 105 Fleischer, Heinrich Leberecht 186, 195

236

INDEX

Flournoy, Théodore 135, 139 Fodor Jerry, A. 126, 130, 131 Förster, Richard 194, 195 Forthcoming 198, 204 Foucault, Michel 33, 40, 61 Fournier, J.-M. 14, 15, 17, 18, 20, 21 23, 25 Fournier, Nathalie 14, 16, 25 França, Jean Marcel Carvalho 30, 35, 39 Freese, Rudolf 198, 199, 205 Fück, Johann 184, 185, 186, 195 Galet, Y. 19, 25 Ganault, Joël 71 Garat, Dominique-Joseph 199, 200, 201, 202, 203, 204, 205 Garnier, Jean 3, 7, 9, 10, 13, 15 Gattel, Claude Marie 64, 65, 71 Girard 61 Gleitman, L. R. 166 Goold, George P. 194 Grotsch, Klaus 231, 234 Guimarães, Eduardo 28, 40, 86, 149, 150, 151, 156, 221 Guzzoni, Alfredo 194 Habermas, J. 107, 108, 120 Haller, Albrecht von 104, 105 Hans-Josef 131, 233, 234 Hanzeli, Victor E. 228, 234 Harper 120 Harris, Roy 139 Hausmann, Franz Josef 16 Henry, P. 221 Henry, Victor 139 Herder, Johann Gottfried von 185, 195 Hesiode 182 Hjelmslev, L 209, 215 Homere 182 Hubert, Carrier 62 Humboldt, Wilhelm von 63, 94, 95, 98, 111, 197, 198, 199, 200, 201, 202, 203, 204, 205, 211, 213, 214 Ikobava 94, 99 Irson, C. 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 54, 56, 57, 58, 60, 61, 62 Itkonen, Esa 231, 234 Jakobson 95, 210, 211, 212, 213, 215, 216, 218, 219, 221 Jankowski, Kurt R. 183

Jespersen, O. 157, 161, 163, 167 Johannes, Malalas 95 Jooken, L. 131, 139 Karmiloff-Smith, A. 167 Kasbarian, Jean-Michel 85 Katz Jerrold, J. 130, 131 Kempelen, Wolfgang von 101, 102, 103, 104, 105, 106 Kibbee, Douglas A. 3, 16 Koerner, Konrad 131, 223, 225, 233, 234 Lammers, Wilhelm 198, 205 Lancelot, Claude 41, 42, 43, 44, 47, 47, 61, 62 Léon, Jacqueline 123, 129, 131 Lessing, Karl Gotthelf 186, 195 Lévy-Bruhl 99, 100 Lévy-Strauss, C 209 Lima, Herman 87, 92 Littmann, Enno 195 Maat, Jaap 129, 131 Macherey, Pierre 63,71 Maciel, Maximino 75, 77, 80, 81, 82, 83, 84, 85, 86 Maire, Catherine 62 Manner, E. 64, 76, 78, 118, 135, 161, 166, 208, 209, 210, 211, 212, 215, 217, 220 Marcuse, H. 108, 121 Mariani, Bethania S.C. 27, 40 Marmontel, Jean-François 204 Marr, N.Ya. 93, 94, 95, 96, 97, 100 Martinet, A 209, 215, 221 Masset 21, 22, 23, 24, 41, 62 Masterman, Margaret 123, 126, 127, 128, 131 Mattoso Câmara Jr. J. 152, 153, 154, 155, 156, 208, 209, 210, 211, 212, 213, 215, 219, 221 Maupas, Charles J. 8, 10, 11, 15, 17, 21, 22, 23, 24, 42, 45, 46 ,62 Maurer, T.H. 145, 210, 216, 217, 218, 219, 221 Mazière, F. 39,40, 41, 42, 47, 85 Meigret, Louis 3, 7, 9, 11, 15, 19, 20, 22, 24 Mendes, Raimundo Teixeira 142, 148 Menze, Clemens 198, 205 Mešaninov 93, 100 Mesquita Filho, Júlio de 142, 148 Meusel, Johann Georg 195 Michel, M. J. 85 Mommsen, Theodor 186, 195 Mongin, François Bernard 66,71 Morus, Samuel Friedrich Nathanael 195

INDEX Mounin, Georges 225, 234 Mueller-Vollmer, Kurt 200, 205 Neves, M.H.M.171 Nicot, J. 21, 42, 62 Niebuhr, Carsten 186, 195 Niederehe, Hans-Josef 131, 233, 234 Niederehe, Kelly Louis G. 131, 233, 234 Nieuhoff, John 28, 37, 38,39 Normand, Cl. 59 Ondrejovi, Slavomír 101, 102, 105 Orlandi, Eni P. 28, 33, 39, 40, 85, 86, 149, 151, 156, 207, 221 Oudin, C. 42, 45, 46, 62 Palsgrave, John 3, 7, 9, 15 Pêcheux, Michel 28, 34, 40, 219 Pellat, Jean-Christophe 4, 5, 6, 16 Pellisson; Olivet 41, 46, 61, 62 Persival, Keith 231, 234 Pfeiffer, Rudolf 184, 195 Phillips, D.L. 109, 121 Pillot, Jean 3,6, 7, 9, 10, 12, 13, 15, 18, 20, 21, 23,24 Pindare 174, 175, 182 Platon 177, 178, 180, 181, 182, 204 Pompino-Marschall, Bernd 101, 103, 105 Port-Royal 24, 41, 43, 45, 47, 61, 226, 231 Puech, C. 162, 207, 209, 221 Rafail 93, 100 Ramus, Petrus 3, 7, 8, 9, 11, 15 Ribeiro, Julio 86 Richens R.H. 123, 124, 125, 126, 128, 129, 131 Riegel, Martin 4, 5, 6, 16 Rioul, René 4, 5, 6, 16 Rivarol, Antoine 200, 205 Robins, Robert 234 Roget, Peter Mark 123, 127, 131 Rosier, I. 86 Sacy, Antoine Isaac Silvestre de 195 Salmon, Vivian 123, 124, 126, 131 Sandys, John Edwin, Sir 195 Sapir, E 209, 211 Saussure, Ferdinand de 63, 109, 110, 121, 133, 134, 135, 136, 137, 138, 139, 164, 167, 179, 180, 209, 210, 211, 212, 213, 214, 215, 216, 218, 221, 226 Schmitter, P. 131, 139, 224, 231, 234 Schuchardt, H. 98, 100 Schwartzman, Simon 142, 148

237

Seidel, Siegfried 199, 205 Serbat, Guy 5, 16 Serreius, Johannes, 10, 11, 15 Skinner, Quentin 198, 205 Sodré, Nelson Werneck 87, 92 Städtler, Thomas 16 Stéfanini, Jean 7, 8, 16 Steinthal 139 Sully, J. 157, 159, 160, 161, 162, 163, 167 Swiggers, Pierre 16, 131, 139 Sylvius, Iacobus 3, 6, 9,11, 12,13 ,15 Tamine-Gardes, J. 158, 159, 165, 166, 167 Tarnoczy, T. H. 101, 102, 103, 105 Tillmann, H. G. 102, 105 Todorov, Tzvetan 135, 137, 139 Traunmüller, Hartmut 102, 103, 106 Tristan Howald, Ernst 71 Troupeau, Georges 228, 234 Trubeckoj, N.S. 95, 100 Trubetskoy 95, 100, 209 Tully, James 198, 205 Ungeheuer, Gerold 105, 106 Varvolsen, S. 86 Vaugelas (C. Favre de) 48, 51, 52, 53, 54, 62 Versteegh, Kees 131, 228, 233 Veselovskij 98, 100 Weaver, Warren 124, 131 Weinrich, H. 19, 25 Wilamowitz-Moellendorff 196 Wildgen, Wolfagang 101, 104, 105 Wilkins, John 68, 123, 125, 126, 127, 131 Wilks, Yorick 123, 129, 130, 131 Wittgenstein, L. 109, 121, 123, 126 Zvegincev 93, 100, 238

238

239

INDEX RERUM A. Ambiguity, ambiguïté, ambigüidade 36, 124, 130, 136, 219 Absence, absence, ausência 36, 92, 96, 98, 109, 133, 134, 135, 137, 138, 139, 189, 198, 231 Acoustics, acoustique, acústica 105 Acquisition, acquisition, aquisição 83, 157, 158, 159, 160, 161, 162, 163, 165, 166, 167 Action, action, ação 7, 8, 19, 22, 79, 97, 102, 113, 119, 162, 173, 189, 191, 197, 202 Anagram, anagramme, anagrama 45, 137, 138 Analogy, analogie, analogia 136, 162, 179, 180, 181, 199, 201 Animism, animisme, animismo 97 Anthropology, anthropologie, antropologia 198, 199, 203, 204, 211, 213, 218 Arabic, arabe, árabe 184, 185, 186, 187, 188, 189, 190, 191, 192, 228, 230 Autonimy, autonimie, autonímia 159 B. Basque, basque, basco 94, 198, 199, 201, 202, 204, 231 Behaviorism, béhaviorisme, comportamentalismo 112, 208, 211, 213, 214 Biblical studies, études bibliques, estudos bíblicos 183, 187 Black language, Langage des Noirs, linguagem dos negros 87, 88, 92

D. Data, données, dados 58, 59, 91, 224 Descriptive linguistics, linguistique descriptive, lingüística descritiva 114, 118, 153, 214, 217 Designation, désignation, designação 24, 35, 38, 39, 159 Diachrony, diachronie, diacronia 51, 209, 210, 211, 212, 215 Dialect, dialecte, dialeto 38, 82, 83, 145 Dictionary, dictionnaire, dicionário 21, 41, 47, 48, 56, 58, 60, 61, 62, 65, 157, 158, 160, 166, 167, 199, 203, 223, 224 Difference, différence, diferença 18, 19, 21, 23, 24, 42, 45, 48, 53, 78, 83, 95, 113, 117, 126, 134, 138, 150, 162, 164, 166, 219, 233 Discourse, discours, discurso 27, 28, 29, 31, 33, 34, 35, 36, 37, 39, 40, 43, 47, 49, 51, 52, 60, 64, 65, 69, 70, 71, 85, 87, 89, 107, 110, 111, 133, 135, 167, 173, 176, 177, 178, 199, 200, 203, 204, 205, 209, 210, 214, 219, 221, 232

E. Empiricism, empirisme, empirismo 75 English, anglais, inglês 3, 5, 15, 28, 59, 76, 125, 127, 128, 130, 131, 155, 157, 159, 160, 161, 163, 167, 183 Enlightenment, Lumières, Iluminismo 69, 70 Epistemology, épistémologie, epistemologia 65, 107, 108, 109, 111, 118, 211, 220 Ergon, Energeia, energy, énergie 187, 197, 198, C. 202, 203, 204 Chronicle, chronique, crônica 27 Etymology, étymologie, etimologia 70, 137 Creation, création, criação 8, 19, 39, 119, 141, Evolution, évolution, evolução 48, 53, 56, 61, 81, 142, 145, 147, 148, 165, 172, 180, 197, 203,223 93, 94, 95, 96, 97, 98, 99, 108, 131, 143, 152, Culture, culture, cultura 39, 70, 86, 144, 171, 172, 215, 234 185, 192, 199, 201, 207, 210, 211, 213, 215 Extended Latin grammar, grammaire latine étendue, gramática latina estendida 3

240

INDEX

F. Form, forme, forma 4, 5, 6, 7, 8, 10, 11, 13, 14, 16, 17, 18, 19, 20, 21, 23, 24, 25, 27, 28, 29, 32, 34, 35, 41, 42, 43, 44, 45, 46, 47, 48, 52, 58, 63, 69, 75, 77, 80, 82, 83, 84, 91, 94, 97, 98, 107, 108, 112, 113, 115, 118, 119, 126, 134, 138, 151, 155, 176, 178, 210, 211, 213, 215, 218, 219, 229 French, français, francês 3, 4, 5, 6, 7, 8, 9, 10, 11, 14, 15, 16, 17, 18, 24, 25, 27, 37, 38, 39, 41, 42, 43, 44, 45, 46, 47, 48, 49, 51, 54, 59, 60, 61, 62, 63, 64, 65, 66, 79, 85, 86, 96, 97, 99, 104, 133, 135, 136, 138, 144, 146, 158, 159, 161, 163, 164, 165, 166, 198, 199, 200, 201, 202, 203, 212, 219, 224, 233 Function, fonction, função 24, 27, 38, 47, 50, 58, 79, 80, 97, 107, 130, 134, 142, 144, 154, 174, 208, 210, 215, 219

Hyperlanguage, hyperlangue, hiperlíngua 39, 40, 75, 76, 77, 78, 79, 81, 83, 85 I. Identity, identité, identidade 133, 134, 137, 138, 151, 160 Ideologists, idéologues, ideólogos 63, 64, 65, 66, 67, 68, 70, 71, 223 Indigenous language, langue indigène, língua indígena 30, 39, 146 Institutionalization, institutionnalisation, institucionalização 27, 149, 151, 152, 156, 217, 218, 219, 224 Intermediary language, langue intermédiaire, língua intermediária 123, 124, 126, 127 Italian, italien, italiano 4, 87, 161, 224, 229

G. German, allemand, alemão 3, 94, 95, 96, 98, 136, 183, 184, 187, 189, 195, 197, 200, 201 General grammar, grammaire générale, gramática geral 21, 24, 41, 42, 43, 45, 47, 49, 60, 61, 63, 64 Grammar, grammaire, gramática 3, 4, 5, 6, 14, 32, 41, 63, 64, 65, 68, 69, 71, 72, 75, 76, 77, 78, 79, 80, 81, 82, 85, 86, 91, 116, 118, 125, 131, 143, 145, 149, 150, 151, 156, 179, 180, 181, 183, 201, 202, 204, 208, 209, 216, 217, 226, 227, 228, 229, 233 Grammaticization, grammatisation, gramatização 17, 75, 77, 149, 208 Greek, grec, grego 5, 7, 8, 20, 25, 41, 42, 44, 47, 60, 145, 171, 172, 183, 184, 185, 187, 189, 192, 197, 199, 201, 202, 203, 225, 234

L. Latin, latin, latim 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 19, 20, 23, 27, 41, 42, 44, 46, 47, 48, 51, 55, 56, 59, 60, 62, 133, 136, 145, 147, 155, 183, 184, 185, 189, 190, 191, 192, 199, 200, 201, 203, 216, 217, 230 Lexicon, lexique, léxico 47, 48, 53, 55, 60, 61, 81, 82, 86, 166, 167, 232, 234 Linguistic colonization, colonisation linguistique, colonização lingüística 32, 38 Linguistic families, familles linguistiques, famílias lingüísticas 3, 146, 230 Literary language, langage littéraire, linguagem literária 78 Logic, logique, lógica 43, 44, 61, 65, 99, 108, 109, 110, 111, 114, 116, 119, 120, 121, 126, 162, 177, 178, 180, 181, 201, 214, 226, 228 Logicism, logicisme, logicismo 108, 226, 234

H. Hebrew, hébreu, hebraico 183, 185, 187, 192 Historical linguistics, linguistique historique, Lingüística histórica 144, 212, 216, 217 Historiography, historiographie, historiografia 223, 234 History, histoire, história 17, 24, 27, 28, 39, 46, 62, 75, 86, 87, 92, 93, 95, 99, 105, 117, 118, 120, 131, 137, 139, 142, 143, 147, 149, 150, 151, 154, 155, 156, 157, 158, 161, 166, 179,187, 190, 192, 195, 196, 198, 204, 205, 207, 208,209, 210, 211, 214, 215, 216, 217, 219, 220, 221, 223, 224, 225, 226, 227, 230, 231, 233, 234

M. Machine, machine, máquina 101, 102, 103, 104, 105, 123, 129, 131, 160 Manual, manuel, manual 43, 46, 102, 229 Mathematization, mathématisation, matematização 108 Mechanism, mécanisme, mecanismo 102, 103, 105, 112, 133, 136, 138, 155, 180 Mentalism, mentalisme, mentalismo 112, 114, 209, 211, 214

INDEX Metalanguage, métalangage, metalinguagem 150, 156, 176 Metaphor, métaphore, metáfora 59 Metaphysics, métaphisique, metafísica 108, 113, 116 Method, méthode, método 41, 42, 43, 44, 45, 46, 47, 48, 51, 53, 54, 60, 61, 62, 64, 65, 66, 67, 68, 71, 72, 98, 108, 109, 112, 119, 120, 123, 129, 134, 135, 167, 198, 205, 208, 209, 214, 217, 217 Morphology, morphologie, morfologia 47, 49, 53, 56, 58, 78, 124, 144, 209, 215, 217,218

241

Presence, présence, presença 27, 102, 133, 134, 135, 137, 138, 160, 165, 210, 212, 217, 231 Primitive, primitive, primitivo 10, 97, 108, 114, 116, 127, 172 Prose, prose, prosa 46, 136, 161, 202 Psychologism, psychologisme, psicologismo 208, 213, 214 R. Reason, raison, razão 66, 67, 70, 80, 85, 116, 118, 121, 137, 138, 159, 164, 188, 200, 211, 216 Record, register, registro 129, 161, 165 Russian, russe, russo, 93, 95, 96, 97, 98, 99

N. Narrative, narrative, narrativa 28, 158, 173, 173 Nature, nature, natureza 10, 19, 21, 27, 31, 32, 39, S . 65, 67, 96, 97, 101, 108, 109, 111, 118, 120, 155, Sanskrit, Sanskrit, sânscrito, 135, 202, 216, 226, 157, 160, 162, 163, 167, 176, 177, 178, 179, 231 180, 181, 203, 208, 212, 214, 215, 219, 227, Science, science, ciência 47, 58, 60, 69, 71, 100, 229, 233 104, 107, 109, 110, 111, 112, 113, 114, 115, 116, Neologism, néologisme, neologismo 164 117, 118, 119, 120, 121, 122, 135, 150, 151, 152, Norm, norme, norma 43, 47, 50, 54, 85, 86, 90, 91, 166, 177, 179, 181, 184, 192, 207, 208, 209, 211, 97 , 218 216, 218, 219, 220, 224, 225, 226, 228, 229, 231, 235 P. Semantics, sémantique, semântica 112, 113, 114, Paradigm, paradigme, paradigma 11, 63, 64, 67, 115, 129, 130, 133, 139, 144, 211, 213, 214 68, 71, 167, 227, 231 Spanish, espagnol, espanhol 27, 65, 228 Philology, philology, filologia 78, 150, 184, 187, Speech, parole, fala 4, 6, 23, 24, 32, 43, 44, 66, 69, 218 70, 101, 102, 103, 104, 105, 110, 111, 113, 115, Philosophy, philosophie, filosofia 46, 64, 68, 72, 134, 135, 137, 138, 159, 160, 163, 165, 166, 167, 107, 108, 111, 114, 120, 141, 142, 143, 147, 171, 172, 173, 174, 175, 180, 197, 200, 202, 203, 171, 176, 180, 181, 184, 204, 205, 207, 209, 204, 214, 221 216, 220, 221, 223 Structuralism, structuralisme, estruturalismo 109, Phonetics, phonétique, fonética 78, 81, 101, 105, 110, 111, 152, 207, 209, 210, 211, 212, 213, 214, 144, 145, 152, 231, 212 215, 216, 217, 218, 219, 220, 221 Phonology, phonologie, fonologia 81, 113, Structure, structure, estrutura 11, 35, 47, 59, 83, 155, 210, 214, 215, 216, 217, 218, 221 94, 95, 97, 115, 116, 117, 118, 120, 124, 126, Phrase, syntagme, sintagma 14, 24, 76, 77, 80, 131, 135, 146, 155, 160, 166, 178, 205, 207, 209, 210, 211, 212, 213, 214, 215, 216, 217, 84, 92, 118, 136, 200, 201, 213, 241 218, 219, 220, 221, 223, 227, 229 Poetry, poésie, poesia 171, 172, 173, 174, 175, Synchrony, synchronie, sincronia 57, 75, 180, 176, 187, 213 209, 210, 211, 212, 226 Portuguese, portugais, português 27, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 75, 76, 77, 78, 79, 80, Syntax, syntaxe, sintaxe 8, 9, 15, 21, 42, 47, 49, 50, 60, 62, 76, 77, 78, 79, 82, 83, 84, 91, 115, 81, 82, 83, 84, 85, 88, 89, 90, 91, 92, 142, 143, 124, 125, 128, 130, 144, 199, 201, 214, 217, 218 144, 145, 146, 147, 149, 150, 151, 152, 153, System, système, sistema 19, 20, 22, 66, 67, 68, 70, 81, 154, 155, 156, 160, 162, 163, 167 96,97,110,111,113,118,129,134,135,136,138,139, Positivism, positivisme, positivismo 107, 108, 150, 153, 154, 155, 156, 190, 209, 211, 212, 214, 215, 109, 111, 112, 113, 114, 116, 119, 142, 209 216,220

242

INDEX

T. Technology, technologie, tecnologia 14, 148 Tense, temps, tempo 4, 6, 8, 11, 17, 18, 19, 20, 21, 22, 23, 24, 31, 42, 43, 50, 51, 59, 61, 62, 95, 142, 144, 145, 172, 173, 179, 210, 221, 223, 224, 232, 233 Theology, théologie, teologia 41, 57, 184, 185, 188 Tupi, tupi, tupi 143, 145, 146, 147, 221 U. Universal language, langue universelle, língua universal 123, 124, 126, 129, 131 Use/usage, usage, uso 21, 48, 49, 50, 51, 53, 54, 56, 57, 58, 60, 61, 62, 71, 79, 80, 82, 86, 126, 137, 154, 180 V. Value, valeur, valor 14, 18, 20, 22, 24, 47, 54, 87, 93, 133, 137, 138, 154, 164, 179, 189, 192, 225, 233 Vocabulary, vocabulaire, vocabulário 35, 46, 56, 60, 159, 163

STUDIES IN THE HISTORY OF THE LANGUAGE SCIENCES E. F. K. Koerner, Editor

Zentrum für Allgemeine Sprachwissenschaft, Typologie und Universalienforschung, Berlin [email protected] The series Studies in the History of the Language Sciences (SiHoLS) has been established as a companion to the journal Historiographia Linguistica. The series intends to meet the revival of interest in the history of linguistic thought and to provide an organized reservoir of information concerning our heritage of linguistic ideas of more than two millennia. SiHoLS will publish book-length scholarly studies on (the evolution of) human reflection about the nature of language and the many ways in which it can be analyzed and used. These studies may concern particular aspects of language study, entire traditions, or special periods of their development. In addition, the series will include re-editions or entirely new translations into English of ‘classic’ works in the field that have been out of print for many years. These new editions will be introduced by a present-day specialist who places the book in its intellectual and socio-historical context, and highlights its significance in the evolution of our thinking about language. A complete list of titles in this series can be found on the publishers website, www.benjamins.com 111 Zwartjes, Otto, Gregory James and Emilio Ridruejo (eds.): Missionary Linguistics III / Lingüística misionera III. Morphology and Syntax. Selected papers from the Third and Fourth International Conferences on Missionary Linguistics, Hong Kong/Macau, 12–15 March 2005, Valladolid, 8-11 March 2006. viii, 357 pp. Expected August 2007 110 Guimarães, Eduardo and Diana Luz Pessoa De Barros (eds.): History of Linguistics 2002. Selected papers from the Ninth International Conference on the History of the Language Sciences, 27-30 August 2002, São Paulo - Campinas. 2007. x, 242 pp. 109 Zwartjes, Otto and Cristina Altman (eds.): Missionary Linguistics II / Lingüística misionera II. Orthography and Phonology. Selected papers from the Second International Conference on Missionary Linguistics, São Paulo, 10–13 March 2004. 2005. vi, 292 pp. 108 Niederehe, Hans-Josef: Bibliografía cronológica de la lingüística, la gramática y la lexicografía del español (BICRES III). Desde el año 1701 hasta el año 1800. 2005. vi, 474 pp. 107 Luhtala, Anneli: Grammar and Philosophy in Late Antiquity. A study of Priscian's sources. 2005. x, 171 pp. 106 Zwartjes, Otto and Even Hovdhaugen (eds.): Missionary Linguistics/Lingüística misionera. Selected papers from the First International Conference on Missionary Linguistics, Oslo, 13–16 March 2003. 2004. vi, 288 pp. 105 Formigari, Lia: A History of Language Philosophies. Translated by Gabriel Poole. 2004. x, 252 pp. 104 Koerner, E.F.K.: Essays in the History of Linguistics. 2004. x, 271 pp. 103 Joseph, John E.: From Whitney to Chomsky. Essays in the history of American linguistics. 2002. viii, 240 pp. 102 Koerner, E.F.K. and Aleksander Szwedek (eds.): Towards a History of Linguistics in Poland. From the early beginnings to the end of the 20th century. 2001. xxii, 335 pp. 101 Kelly, L.G.: The Mirror of Grammar. Theology, philosophy and the Modistae. 2002. x, 243 pp. 100 Koerner, E.F.K. and Hans-Josef Niederehe (eds.): History of Linguistics in Spain/Historia de la Lingüística en España. Volume II. 2001. xxii, 463 pp. 99 Auroux, Sylvain (ed.): History of Linguistics 1999. Selected papers from the Eighth International Conference on the History of the Language Sciences, 14–19 September 1999, Fontenay-St.Cloud. With the assistance of Joscelyne Arpin, Elisabeth Lazcano, Jacqueline Léon. 2003. xii, 403 pp. 98 Graffi, Giorgio: 200 Years of Syntax. A critical survey. 2001. xiv, 551 pp. 97 Hutchins, W. John (ed.): Early Years in Machine Translation. Memoirs and biographies of pioneers. 2000. xii, 400 pp. 96 Joseph, John E.: Limiting the Arbitrary. Linguistic naturalism and its opposites in Plato's Cratylus and modern theories of language. 2000. x, 224 pp. 95 Cram, David, Andrew R. Linn and Elke Nowak (eds.): History of Linguistics 1996. Volume 2: From Classical to Contemporary Linguistics. 1999. xx, 390 pp. 94 Cram, David, Andrew R. Linn and Elke Nowak (eds.): History of Linguistics 1996. Volume 1: Traditions in Linguistics Worldwide. 1999. xx, 341 pp.

93 Coblin, W. South and Joseph A. Levi: Francisco Varo's Grammar of the Mandarin Language (1703). An English translation of ‘Arte de la lengua Mandarina’. With an Introduction by Sandra Breitenbach. 2000. liv, 282 pp. 92 Koerner, E.F.K.: Linguistic Historiography. Projects & prospects. 1999. x, 236 pp. 91 Niederehe, Hans-Josef: Bibliografía cronológica de la lingüística, la gramática y la lexicografía del español (BICRES II). Desde el año 1601 hasta el año 1700. 1999. vi, 472 pp. 90 Esparza Torres, Miguel Ángel and Hans-Josef Niederehe: Bibliografía Nebrisense. Las obras completas del humanista Antonio de Nebrija desde 1481 hasta nuestros días. 1999. vi, 374 pp. 89 Jones, William J.: Images of Language. Six essays on German attitudes to European languages from 1500 to 1800. 1999. x, 299 pp. 88 Koerner, E.F.K. (ed.): First Person Singular III. Autobiographies by North American scholars in the language sciences. 1998. x, 267 pp. 87 Stein, Dieter and Rosanna Sornicola (eds.): The Virtues of Language. History in language, linguistics and texts. Papers in memory of Thomas Frank. 1998. viii, 232 pp. 86 Darnell, Regna: And Along Came Boas. Continuity and revolution in Americanist anthropology. 1998. xviii, 333 pp. 85 Taylor, Daniel J.: De Lingua Latina X. A new critical text and English translation with prolegomena and commentary. 1996. x, 205 pp. 84 Verburg, Pieter A.: Language and its Functions. A historico-critical study of views concerning the functions of language from the pre-humanistic philology of Orleans to the rationalistic philology of Bopp. Translated by Paul Salmon in consultation with Anthony J. Klijnsmit. 1998. xxxiv, 577 pp. 83 Wollock, Jeffrey: The Noblest Animate Motion. Speech, physiology and medicine in pre-Cartesian linguistic thought. 1997. l, 461 pp. 82 Bekkum, Wout Jac. van, Jan Houben, Ineke Sluiter and Kees Versteegh: The Emergence of Semantics in Four Linguistic Traditions. Hebrew, Sanskrit, Greek, Arabic. 1997. ix, 322 pp. 81 Lee, Penny: The Whorf Theory Complex. A critical reconstruction. 1996. x, 324 pp. 80 Nerlich, Brigitte and David D. Clarke: Language, Action and Context. The early history of pragmatics in Europe and America 1780–1930. 1996. xiv, 497 pp. 79 Koerner, E.F.K.: Professing Linguistic Historiography. 1995. viii, 274 pp. 78 Jankowsky, Kurt R. (ed.): History of Linguistics 1993. Papers from the Sixth International Conference on the History of the Language Sciences (ICHoLS VI), Washington DC, 9–14 August 1993. 1995. xx, 380 pp. 77 Salmon, Vivian: Language and Society in Early Modern England. Selected essays 1982–1994. 1996. viii, 276 pp. 76 Niederehe, Hans-Josef: Bibliografía cronológica de la lingüística, la gramática y la lexicografía del español (BICRES). Desde los principios hasta el año 1600. 1994. vi, 457 pp. 75 Versteegh, Kees: The Explanation of Linguistic Causes. Az-Zağğāğī's Theory of Grammar. Introduction, translation, commentary. 1995. xvi, 310 pp. 74 Formigari, Lia and Daniele Gambarara (eds.): Historical Roots of Linguistic Theories. 1995. viii, 309 pp. 73 Goldziher, Ignaz: On the History of Grammar among the Arabs. Translated and edited by Kinga Dévényi and Tamás Iványi. 1994. xx, 153 pp. 72 Radwańska-Williams, Joanna: A Paradigm Lost. The linguistic thought of Mikołaj Kruszewski. 1994. xii, 200 pp. 71 Law, Vivien A. (ed.): History of Linguistic Thought in the Early Middle Ages. 1993. viii, 255 pp. 70 Formigari, Lia: Signs, Science and Politics. Philosophies of language in Europe 1700–1830. 1993. x, 218 pp. 69 Murray, Stephen O.: Theory Groups and the Study of Language in North America. A social history. 1994. xx, 598 pp. 68 Ahlqvist, Anders (ed.): Diversions of Galway. Papers on the history of linguistics from ICHoLS V. 1992. xxviii, 384 pp. 67 Subbiondo, Joseph L. (ed.): John Wilkins and 17th-Century British Linguistics. 1992. xvi, 376 pp. 66 Naumann, Bernd, Franz Plank and Gottfried Hofbauer (eds.): Language and Earth. Elective affinities between the emerging sciences of linguistics and geology. 1992. xvi, 445 pp. 65 Itkonen, Esa: Universal History of Linguistics. India, China, Arabia, Europe. 1991. x, 368 pp. 64 Noordegraaf, Jan, Kees Versteegh and E.F.K. Koerner (eds.): The History of Linguistics in the Low Countries. 1992. vi, 400 pp. + ills.

63 Sundby, Bertil, Anne Kari Bjørge and Kari E. Haugland: A Dictionary of English Normative Grammar 1700–1800 (DENG). 1991. x, 486 pp. 62 Leitner, Gerhard (ed.): English Traditional Grammars: an International Perspective. 1991. x, 392 pp. 61 Koerner, E.F.K. (ed.): First Person Singular II. Autobiographies by North American Scholars in the Language Sciences. 1991. x, 303 pp. 60 Kibbee, Douglas A.: For to Speke Frenche Trewely. The French language in England, 1000–1600: its status, description and instruction. 1991. viii, 268 pp. 59 Nerlich, Brigitte: Semantic Theories in Europe, 1830–1930. From etymology to contextuality. 1992. xi, 359 pp. 58 Dinneen, Francis P. S. J. and E.F.K. Koerner (eds.): North American Contributions to the History of Linguistics. 1990. xii, 238 pp. 57 Starnes, De Witt T. and Gertrude E. Noyes: The English Dictionary from Cawdrey to Johnson 1604–1755. With an introductory article and a bibliography byGabriele Stein. 1991. cxii, xxii, 299 pp. 56 Versteegh, Kees and Michael G. Carter (eds.): Studies in the History of Arabic Grammar II. Proceedings of the second symposium on the history of Arabic grammar, Nijmegen, 27 April–1 May, 1987. 1990. x, 320 pp. 55 Hall, Jr., Robert A.: A Life for Language. A biographical memoir of Leonard Bloomfield. 1990. x, 129 pp. 54 Antonsen, Elmer H., James W. Marchand and Ladislav Zgusta (eds.): The Grimm Brothers and the Germanic Past. 1990. x, 162 pp. 53 Owens, Jonathan: Early Arabic Grammatical Theory. Heterogeneity and standardization. 1990. xvi, 294 pp. 52 Juul, Arne and Hans Frede Nielsen (eds.): Otto Jespersen: Facets of his Life and Work. 1989. xviii, 154 pp. 51:2 Niederehe, Hans-Josef and E.F.K. Koerner (eds.): History and Historiography of Linguistics. Proceedings of the Fourth International Conference on the History of the Language Sciences (ICHoLS IV), Trier, 24–28 August 1987. Volume 2: 18th–20th Century. 1990. x, 397-873 pp. 51:1 Niederehe, Hans-Josef and E.F.K. Koerner (eds.): History and Historiography of Linguistics. Proceedings of the Fourth International Conference on the History of the Language Sciences (ICHoLS IV), Trier, 24–28 August 1987. Volume 1: Antiquitity–17th Century. 1990. xxv, 1-396 pp. 50 Koerner, E.F.K.: Practicing Linguistic Historiography. 1989. xiv, 455 pp. 49 Mauro, Tullio De and Lia Formigari (eds.): Leibniz, Humboldt, and the Origins of Comparativism. Proceedings of the international conference, Rome, 25–28 September 1986. 1990. vii, 329 pp. 48 Formigari, Lia: Language and Experience in 17th-century British Philosophy. 1988. viii, 169 pp. 47 Hall, Jr., Robert A. (ed.): Leonard Bloomfield: Essays on his Life & Work. 1987. x, 237 pp. 46 Taylor, Daniel J. (ed.): The History of Linguistics in the Classical Period. 1987. xii, 294 pp. 45 Owens, Jonathan: The Foundations of Grammar. An introduction to medieval Arabic grammatical theory. 1988. xii, 371 pp. 44 Amsler, Mark E.: Etymology and Grammatical Discourse in Late Antiquity and the Early Middle Ages. 1989. xi, 280 pp. 43 Bursill-Hall, Geoffrey L., Sten Ebbesen and E.F.K. Koerner (eds.): De Ortu Grammaticae. Studies in medieval grammar and linguistic theory in memory of Jan Pinborg. 1990. x, 372 pp. 42 Buzzetti, Dino and Maurizio Ferriani (eds.): Speculative Grammar, Universal Grammar, Philosophical Analysis. Papers in the Philosophy of Language. 1987. x, 269 pp. 41 Cowan, William, Michael Foster and E.F.K. Koerner (eds.): New Perspectives in Language, Culture, and Personality. Proceedings of the Edward Sapir Centenary Conference (Ottawa, 1–3 October 1984). 1986. xiv, 627 pp. 40 Hartmann, Reinhard (ed.): The History of Lexicography. 1986. viii, 265 pp. 39 Dinneen, Francis P. S. J.: Peter of Spain: Language in Dispute. An English translation of Peter of Spain's 'Tractatus' called afterwards SUMMULAE LOGICALES, based on the critical edition by L.M. de Rijk. 1990. xl, 271 pp. 38 Aarsleff, Hans, L.G. Kelly and Hans-Josef Niederehe (eds.): Papers in the History of Linguistics. Proceedings of the Third International Conference on the History of the Language Sciences (ICHoLS III), Princeton, 19–23 August 1984. 1987. xxvi, 680 pp. 37 Ó Mathúna, Seán P.: William Bathe, S.J., 1564–1614: a pioneer in linguistics. (English translation from the Irish edition, Dublin, 1981). 1986. iv, 211 pp. + 16 ill. 36 Sapir, Edward: Appraisals of his life and work. Edited by E.F.K. Koerner. 1984. xxviii, 224 pp.

35 Salmon, Vivian and Edwina Burness: A Reader in the Language of Shakespearean Drama. 1987. xx, 523 pp. 34 Quilis Morales, Antonio and Hans-Josef Niederehe (eds.): The History of Linguistics in Spain. 1986. viii, 360 pp. 33 Ramat, Paolo, Hans-Josef Niederehe and E.F.K. Koerner (eds.): The History of Linguistics in Italy. 1986. x, 364 pp. 32 Manchester, Martin L.: The Philosophical Foundations of Humboldt's Linguistic Doctrines. 1985. xii, 216 pp. 31 Christy, T. Craig: Uniformitarianism in Linguistics. 1983. xiv, 139 pp. 30 Gordon, W. Terrence: A History of Semantics. viii, 284 pp. Expected Out of print 29 Arens, Hans: Aristotle's Theory of Language and its Tradition. Texts from 500 to 1750, sel., transl. and commentary by Hans Arens. 1984. v, 525 pp. 28 Versteegh, Kees, E.F.K. Koerner and Hans-Josef Niederehe (eds.): The History of Linguistics in the Near East. 1982. xii, 265 pp. 27 Breva-Claramonte, Manuel: Sanctius' Theory of Language. A contribution to the history of Renaissance linguistics. 1982. viii, 294 pp. 26 Koerner, E.F.K., Hans-Josef Niederehe and Robert H. Robins (eds.): Studies in Medieval Linguistic Thought. Dedicated to Geofrey L. Bursill-Hall on the occassion of his 60th birthday on 15 May 1980. 1980. vi, 321 pp. 25 Hymes, Dell H.: Essays in the History of Linguistic Anthropology. 1983. xxiii, 406 pp. 24 Carter, Michael G. (ed.): Arab Linguistics. An introductory classical text with translation and notes. 1981. x, 485 pp. 23 Householder, Fred W.: Apollonius Dyscolus. The Syntax of Apollonius Dyscolus. 1981. vi, 281 pp. 22 McDermott, A. Charlene Senape (ed.): Godfrey of Fontaine's Abridgement of Boethius of Dacia's 'Modi Significandi sive Quaestiones super Priscianum Maiorem'. An text edition with English translation and introduction. 1980. ix, 237 pp. 21 Davis, Boyd H. and Raymond K. O’Cain (eds.): First Person Singular. Papers from the Conference on an Oral Archive for the History of American Linguistics. (Charlotte, N.C., March 1979). 1980. xiv, 239 pp. 20 Koerner, E.F.K. (ed.): Progress in Linguistic Historiography. Papers from the International Conference on the History of the Language Sciences, Ottawa, 28–31 August 1978. 1980. xiv, 421 pp. 19 Koerner, E.F.K.: Toward a Historiography of Linguistics. Selected Essays. With a foreword by Robert H. Robins. 1978. xx, 222 pp. 18 Hayashi, Tetsuro: The Theory of English Lexicography 1530–1791. 1978. xii, 168 pp. 17 Salmon, Vivian: The Study of Language in 17th-Century England. Second Edition. 1988. x, 218 pp. 16 Panconcelli-Calzia, Giulio (1878–1966): 'Geschichtszahlen der Phonetik' (1941), together with 'Quellenatlas der Phonetik' (1940). New edition. With an introduction by E.F.K. Koerner. 1994. xxxviii, 88 pp. 15 Pseudo-Albertus Magnus, and L.G. Kelly: 'Quaestiones Alberti de Modis significandi.' A critical edition. 1977. xxxvii, 191 pp. 14 Sigerus De Cortraco, and Jan Pinborg: Summa modorum significandi; Sophismata. New edition, on the basis of. G. Wallerand's editio prima, with additions, critical notes, an index of terms, and an introduction by Jan Pinborg. 1977. xli, 108 pp. Small-4to. 13 Drake, Glendon F.: The Role of Prescriptivism in American Linguistics 1820–1970. 1977. x, 130 pp. 12 Paulinus A.S. Bartholomaeo,: Dissertation on the Sanskrit Language. Translated, edited and introduced by Ludo Rocher. 1977. xxviii, 224 pp. 11 Koerner, E.F.K.: Western Histories of Linguistic Thought. An annotated chronological bibliography, 1822–1976. 1978. x, 113 pp. 10 Kilbury, James: The Development of Morphophonemic Theory. 1976. viii, 155 pp. 9 Niederehe, Hans-Josef and Harald Haarmann (eds.): In Memoriam Friedrich Diez: Akten des Kolloquiums zum Wissenschaftsgeschichte der Romanistik/Actes du Colloque sur l'Histoire des Etudes Romanes/ Proc. 1976. viii, 508 pp. 8 Stengel, Edmund (1845–1935) (ed.): Chronologisches Verzeichnis französischer Grammatiken vom Ende des 14. bis zum Ausgange des 18. Jahrhunderts, nebst Angabe der bisher ermittelten Fundorte derselben. 1976. x, 240 pp. 7 Pedersen, Holger (1867–1953): A Glance at the History of Linguistics. with particular regard to the historical study of phonology. 1983. xxxii, 100 pp.

History of Linguistics 2002: Selected Papers from the Ninth International Conference on the History of the Language Sciences, 27-30 August 2002, Sao Paulo ... in the History of the Language Sciences)

History of Linguistics: Selected Papers from the Eighth International Conference on the History of the Language Sciences, 14-19 September 1999, Fontenay-St.Cloud (Studies in the History of the Language Sciences 99

History of Linguistics 2005: Selected Papers from the Tenth International Conference on the History of the Language Sciences (ICHOLS X), 1-5 September ... in the History of the Language Sciences)

A History of Language Philosophies (Studies in the History of the Language Sciences)

Missionary Linguistics linguistica Misionera: Selected Papers From The First International Conference On Missionary Linguistics, Oslo, 13-16 March 2003 ... in the History of the Language Sciences)

Missionary Linguistics IV Linguistica Misionera IV: Selected papers from the Fifth International Conference on Missionary Linguistics, Mérida, Yucatán, ... in the History of the Language Sciences)

History of the French Language

The Fontana History of the Environmental Sciences

The Blackwell History of the Latin Language

The Blackwell History of the Latin Language

The Mirror of Grammar (Studies in the History of the Language Sciences)

The Mirror of Grammar (Studies in the History of the Language Sciences)

The History of History

The History of History

The History of History

The History of History

The History of History

The History of History

The History of History

The History of History

The History of History

The Lost History of the Ninth Amendment

A History of the English Language

A History of the Japanese Language

A Brief History of the Spanish Language

A History of the Korean Language

Towards a History of the Basque Language

A History of the Greek Language

A History of the English Language (2006)

A History of the French Language

A History of the French Language

History of Linguistics 2002: Selected Papers from the Ninth International Conference on the History of the Language Sciences, 27-30 August 2002, Sao Paulo ... in the History of the Language Sciences)

History of Linguistics: Selected Papers from the Eighth International Conference on the History of the Language Sciences, 14-19 September 1999, Fontenay-St.Cloud (Studies in the History of the Language Sciences 99

History of Linguistics 2005: Selected Papers from the Tenth International Conference on the History of the Language Sciences (ICHOLS X), 1-5 September ... in the History of the Language Sciences)

A History of Language Philosophies (Studies in the History of the Language Sciences)

Missionary Linguistics linguistica Misionera: Selected Papers From The First International Conference On Missionary Linguistics, Oslo, 13-16 March 2003 ... in the History of the Language Sciences)

Missionary Linguistics IV Linguistica Misionera IV: Selected papers from the Fifth International Conference on Missionary Linguistics, Mérida, Yucatán, ... in the History of the Language Sciences)

History of the French Language

The Fontana History of the Environmental Sciences

The Blackwell History of the Latin Language

The Blackwell History of the Latin Language

The Mirror of Grammar (Studies in the History of the Language Sciences)

The Mirror of Grammar (Studies in the History of the Language Sciences)

The History of History

The History of History

The History of History

The History of History

The History of History

The History of History

The History of History

The History of History

The History of History

The Lost History of the Ninth Amendment

A History of the English Language

A History of the Japanese Language

A Brief History of the Spanish Language

A History of the Korean Language

Towards a History of the Basque Language

A History of the Greek Language

A History of the English Language (2006)

A History of the French Language

A History of the French Language

Recommend Documents