Tai ngay!!! Ban co the xoa dong chu Meaning Change in Grammaticalization Für Pirmin der die Grenzen der Sprache erforscht Meaning Change in Grammaticalization An Enquiry into Semantic Reanalysis R E G I N E E C KA R D T 1 3 Great Clarendon Street, Oxford ox2 6dp Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York ß Regine Eckardt 2006 The moral rights of the author have been asserted Database right Oxford University Press (maker) First published 2006 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose the same condition on any acquirer British Library Cataloguing in Publication Data Data available Library of Congress Cataloguing in Publication Data Data available Typeset by SPI Publisher Services, Pondicherry, India Printed in Great Britain on acid-free paper by Biddles Ltd., King’s Lynn ISBN 019–926260–8 978–019–926260–1 1 3 5 7 9 10 8 6 4 2 Contents Acknowledgements vii 1 Introduction 1 2 Meaning Change under Reanalysis: Previous Views 22 3 Truth Conditional Semantics 59 4 What is Going to Happen 91 5 From Step to Negation: The Development of French Complex Negation Patterns 128 6 From Intensifier to Focus Particle 171 7 To Be or Not to Be a Determiner 202 8 Semantic Reanalysis: The Algebraic Backbone of Meaning Change 235 Appendix: Definitions from Logic and Model Theory 250 Source Texts by Chapter 255 References 264 Index 281 This page intentionally left blank Acknowledgements Many persons and events influenced the making of this book, some in quite unforeseen ways.
I want to thank all those friends and colleagues who were willing to discuss aspects of meaning change, and patient enough to follow the winding road of my expositions, pointing out short cuts or dead ends. Some deserve special mention. Miriam Butt who convinced me that grammaticalization was really a topic worth looking into. Urs Egli who followed my investigations with a patient fatherly eye (even though he may have had other ideas about what I should be doing at times).
Aditi Lahiri who taught me that the best was the least I could give. Irene Wolke without whom life would have been impossible. The lively crowd of Dutch semanticians who taught me that science and fun need not contradict each other, and who respected me even though I was ‘going historical’. Elizabeth Closs Traugott who respected me even though I was ‘insisting on being formal’, and who helped me with her sharp comments to remedy at least some of the inconsistencies in my thinking and writing.
Ekkehard König who showed me exciting new fields for exploration. Christoph Schwarze and Judith Meinschaefer who assisted me in the inter- pretation of Old French examples. Ede and Cécile who had their own special share in the project. Ewald Lang who not only chaperoned an earlier version of this book safely through the formalities of a Habilitationsverfahren but who even read and seemed to like it.
Renate and Claudia, invaluable companions on the thorny road to light in German academia. Manfred Krifka and the people at ZAS Berlin who offered me a new intellec- tual and institutional home, and who soon turned from colleagues into friends. The Deutsche Forschungsgemeinschaft which funded project A1, part of the SFB 471 Konstanz, and made this research possible. viii Acknowledgements John Davey of Oxford University Press who kept his constructive and helpful attitude in spite of unforeseen twists and turns in the project.
Manfred, Luzie, and Pirmin, who will always show me where real life is. Göttingen September 2005 1 Introduction 1.1 Meaning Change under Reanalysis The present monograph investigates meaning change under reanalysis in the paradigm of truth conditional semantics. I will defend the claim that this combination of framework and subject is a fruitful one and can help us to gain a better understanding of the meaning shifts and changes that occur in structural reanalysis. Importantly, detailed semantic representations can oVer an explanation for historical data that were viewed as unexplainable accidents by earlier approaches.
Let me start with an example to illustrate the issues that will be pursued, and consider the development of periphrastic perfect tenses in European languages. The stages are outlined in Meillet (1909) as follows. Classical Latin, as well as Celtic and Gothic, only used inXected past tenses. Contemporary European languages, in contrast, often possess a construction to report about the past on the basis of an auxiliary ‘have’ plus past participle of the verb.
Hence we will conjecture that the construction must have originated somewhere. How does such a construction emerge? Most authors who investigated this topic agree on the following basic steps, illustrated by (made-up) English examples.1 In an initial stage, the language can use participial forms of the verb only to ascribe a property, as in the example in (1.1) verb: to use participle: used car Such modiWed nouns could be used as objects of transitive verbs, in particular with the full verb have as in (1.2): 1 The invented example is backed up by the extensive survey of English data in Visser (1969, 1973) on the use of have + past participle. Focused discussions in Brinton (1988: 94 V.) and Carey (1994) oVer more detailed hypotheses about the actual progress of the development and Olbertz (1993, 1998) complements these in an interesting way with the development of complex perfects from Latin to Spanish. Composite perfect tenses in other Romance languages have likewise received thorough inves- tigation (e.
Meillet 1909, Benveniste 1960, 1968, Dietrich 1987, Harris and Ramat 1987, Salvi 1987).2) Tom has used cars. At that stage, example (1.2) unambiguously had the structure in (1. The verb have denoted possession and related the subject—the possessor— to the possessed object. In this case, the possessed object was described by a noun and further qualiWed by an adjectival form of the verb use.
[Tom]NP [hasmain verb [used cars]NP ]VP Under this analysis, the sentence leaves the agent of the verb use open. Tom may have used the cars, but he may also be in the car trade and own cars that were used by others.2 At some point, however, speakers saw a new syntactic analysis of (2), namely the one in (1.2b) where the two verbs were tied closer together, their subjects were uniWed, and have, rather than denoting possession, made a more abstract temporal contribution.2) received a new syntactic analysis and a new tense construction emerged: have + participle. The verb have changed its status from full verb to auxiliary, and the status of the participle in the overall grammatical system changed likewise: while only transitive verbs allow for a predicative participle, the perfect can be formed for intransitive verbs as well (as shown by the contrast between *the laughed man and the man has laughed). Soon, the construction in (1.2b) would spread and a wealth of new perfect tense constructions would Xourish.
So far the broad outline of the development. The reorganization of syntactic structure between (1.2b) is an instance of reanalysis. The following classical characterization by Langacker (1977) is still valid: ‘change in the structure of an expression or class of expressions that does not involve any immediate or intrinsic modiWcation of its surface manifestation’ (Langacker 1977: 58). Importantly, the structural reanalysis from (1.2b) must be accompanied by appropriate meaning changes in the partaking words and constructions.
The semantic contributions of have, participle, and their combination with the subject noun phrase needed to change under reanalysis. Apparently, speakers were able to abstract suitable new meanings from the Wrst instances of reanalysis they encountered, and apply them in new contexts. 2 Obviously the have + participle construction emerged long before any of the communities knew cars. I used this ahistorical example, however, because ‘used cars’ is a salient category today, hence facilitating access to the older reading of example (1.
Introduction 3 In fact, many instances of structural reanalysis require concomitant appropriate semantic changes in order for the new construction to be mean- ingful. While the causal dependencies between changes in meaning and in form are still subject to debate, it is apparent and never has been doubted that speakers are somehow able to make the required semantic shifts. The fact that all competent speakers seem to be able to perform the necessary semantic operations does, however, not imply that we need not investigate what they do, and how they do it. In fact, the great ease and reliability with which semantic changes under reanalysis occur should have far-reaching repercussions on our conception of semantic processing.
The semantic side of reanalysis witnesses the great analytical potential and logical systematicity of human speakers, as well as the ingenious way in which natural languages combine reliable stability and Xexible elements, allowing adaptation to new communicative needs in reliable ways without disturbing the function of language as a tool for communication. The present monograph aims at contributing to our understanding of meaning change under reanalysis. The next section will brieXy relate the topic of the book to major trends in diachronic linguistics.3 oVers a closer view on the well-deWned steps taken by speakers in reanalysis, speakers who are conWdent that their indi- vidual actions cohere with the linguistic practices of their language commu- nity and will not lead them into linguistic obscurity.4 I will deWne the speciWc new perspective of the present monograph. Semantic change under reanalysis will be approached with the tools of formal, truth value based semantics, a means of investigation that has so far not been brought to bear on diachronic linguistics.
The Wnal section oVers an outline of the further course of the book.2 Some Background The investigation of meaning change dates back to at least the work of Hermann Paul, who could himself draw on the wealth of material that was collected to Wll the Wrst volumes of the largest etymological dictionary in German, the Deutsche Wörterbuch of Wilhelm and Jacob Grimm. Paul proposed the four-way distinction of meaning change through generalization, speciWcation, metaphor, and metonymy as the basic categories of meaning change (Paul 1880). More elaborate theories followed, e. Apart from these comprehen- sive studies in meaning change, some visibly productive modes of meaning change have also inspired scholarly investigation, notably meaning change by 4 Introduction metaphor (Indurkhya 1992, LakoV and Johnson 1980), metonymy (e.
Pan- ther 1999), and via prototype shifts (Dik 1977, Geeraerts 1997). In spite of all attested regularities, it is undeniable that meaning change is driven by language external factors perhaps more than any other mode of language change and therefore has a notoriously contingent quality. This feeling is nicely expressed in the resigned remark of Ferdinand de Saussure that si le français poutre ‘jument’ a pris le sens de ‘pièce de bois, solive’, cela est dû à des causes particulières. ce n’est qu’un accident parmi tous ceux qu’enregistre l’histoire d’une langue (Saussure 1962: 131 f.
(‘That French poutre ‘mare’ has acquired the meaning ‘piece of wood, rafter’ is due to particular causes .