Expressivity in European Languages 9781108834032, 9781108989084, 9781108984430

Citation preview

Expressivity in European Languages

There is an emerging perspective in the discipline of linguistics that takes expressivity as one of the key components of human communication and grammatical structure. Expressivity refers to the use of grammar in natural languages to convey sensory information in a creative way, for example through reduplication, iconicity, ideophones, and onomatopoeia. Expressives have been more commonly associated with non-European languages, so their presence in European languages has so far been under-documented. With contributions from a team of leading scholars, this pioneering book redresses that balance by providing copious, detailed information about the expressive systems of a set of European languages. It comprises a collection of original surveys of expressivity in languages as diverse as Hungarian, Finnish, Scots, German, Greek, Italian, Catalan, Breton, Basque, Georgian, and Russian, with the common goal of challenging structuralist assumptions about the role of syntax, and showing how expressivity is both typologically diverse and universal. jeffrey williams is Professor of Ethnology and Linguistics at Texas Tech University. He has previously edited Expressive Morphology in the Languages of South Asia (2021) and The Aesthetics of Grammar: Sound and meaning in the languages of Mainland Southeast Asia (2014).

Published online by Cambridge University Press

Published online by Cambridge University Press

Expressivity in European Languages Edited by

Jeffrey P. Williams Texas Tech University

Published online by Cambridge University Press

Shaftesbury Road, Cambridge CB2 8EA, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467 Cambridge University Press is part of Cambridge University Press & Assessment, a department of the University of Cambridge. We share the University’s mission to contribute to society through the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781108834032 DOI: 10.1017/9781108989084 © Cambridge University Press & Assessment 2023 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press & Assessment. First published 2023 A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data Names: Williams, Jeffrey P. (Jeffrey Payne), 1958- editor. Title: Expressivity in European languages / edited by Jeffrey P. Williams. Description: Cambridge, United Kingdom ; New York, NY : Cambridge University Press, 2023. | Includes bibliographical references and index. Identifiers: LCCN 2023005688 (print) | LCCN 2023005689 (ebook) | ISBN 9781108834032 (hardback) | ISBN 9781108984430 (paperback) | ISBN 9781108989084 (epub) Subjects: LCSH: Europe–Languages. | Expression. | LCGFT: Essays. Classification: LCC P380 .E97 2023 (print) | LCC P380 (ebook) | DDC 415–dc23/eng/20230512 LC record available at https://lccn.loc.gov/2023005688 LC ebook record available at https://lccn.loc.gov/2023005689 ISBN 978-1-108-83403-2 Hardback Cambridge University Press & Assessment has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Published online by Cambridge University Press

Contents

List of Figures List of Tables List of Contributors 1

Introduction jeffrey p. williams

page vii viii x 1

Uralic 2

3

Hypocoristic reduplications and embellished clippings in Hungarian and elsewhere mario brdar, rita brdar-szabo´ , nikolett f. gulya´ s and laura horva´ th Reduplication in Finno-Ugric languages iwona piechnik

13

53

Germanic 4

Expressivity in Scots: A study of echo words jeffrey p. williams

5

Reduplication as expressive morphology in German gerrit kentner

Hellenic 6

Expressivity in Modern Greek: Some morphological mechanisms for the expression of negative emotions haritini kallergi, georgia katsouda, magdalene konstantinidou and anastasios tsangalidis

91 103

121 123

v

Published online by Cambridge University Press

vi

Contents

Romance 7

Repetition and reduplication in Italian anna m. thornton

231

8

Analysing expressives in a spoken corpus of Majorcan Catalan nicolau dols

269

Celtic 9

A survey of expressive words in Breton me´ lanie jouitteau

295

Vasconian 10

Vindicating the role of ideophones as a typological feature of Basque iraide ibarretxe-antun˜ ano

313

Caucasian 11

Expressive constructions in Georgian and other Caucasian languages thomas r. wier

337

Comparative 12

Parameters of variation in the syntax of expressive suffixes: Case studies of Russian, German, Spanish, and Greek olga steriopolo, giorgos markopoulos, and vassilios spyropoulos

Index

Published online by Cambridge University Press

361

393

Figures

1.1 10.1 10.2 10.3 10.4 10.5

Languages in Europe Stress in polysyllabic ideophones: mara-mara and irrist The ideophone taka-taka The ideophone pla-pla The ideophone mara-mara and gesture Ideophones as a sign of Basqueness

page 3 319 322 323 324 331

vii

Published online by Cambridge University Press

Tables

2.1 2.2

2.3 2.4 7.1 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 8.10 9.1 10.1 10.2 10.3 10.4 10.5

Main characteristics of the linguistic periods of the Hungarian language page 26 Expressive morphology in Udmurt, Meadow Mari, Komi-Permyak, Northern Mansi, Northern and Eastern Khanty, and Finnish 37 The distribution of multiple diminutivization, embellished truncation, simple truncation and abbreviated reduplication 42 Expressive morphology in Udmurt, Meadow Mari, Komi-Permyak, Northern Mansi, Northern and Eastern Khanty, and Finnish 43 Semantic types of Italian reduplicated adjectives 245 Lowest places for speech rate 279 Pitch difference in fragments containing expressives 280 Maximum intensity in fragments containing expressives 280 Minimum intensity in fragments containing expressives 281 Speech rate in fragments containing expressives 281 Extreme cases of pitch difference, above and below average plus/minus absolute deviation 281 Extreme cases of maximum intensity, above and below average plus/minus average absolute deviation 282 Extreme cases of minimum intensity, above and below average plus/minus absolute deviation 282 Extreme cases of speech rate, above and below average plus/minus absolute deviation 282 Extreme cases of pitch difference in relation to total cases found for every type of expressive 283 Shocks and falls mimetics (adverbs and interjections) in Kerne 303 The ideophone mara-mara 315 Word-initial position phonemes in Basque ideophones 317 Consonant clusters in Basque ideophones 317 Semantic fields in Basque ideophones 325 The aesthetic power of Basque ideophones 328

viii

Published online by Cambridge University Press

List of Tables

11.1 11.2

Ideophones within light-verb constructions in Sanzhi Dargwa Floral and faunal terms derived from expressives in Caucasian languages 11.3 Georgian case assignment across TAM series and conjugational class 11.4 Formal criteria distinguishing different conjugation classes 11.5 Statistical distribution of features of morphosyntactic exponence across different expressive classes 11.6 Case assignment of different classes of expressives across tense-aspect series 11.7 Expressive verbs of emission of light or sound or of salient movement 11.8 Salience of voiced fricatives in expressives depicting emission of light 11.9 Ingressive expressive predicates 11.10 Case assignment of different classes of ingressive expressives across tense-aspect series 12.1 Semantic types of expressive suffixes in Russian 12.2 Syntactic types of expressive suffixes in Russian 12.3 Syntactic variation in attachment of the size suffixes in Russian and German 12.4 Syntactic variation in attachment of the size suffixes in Russian, German, and Spanish 12.5 Semantic and syntactic properties of the Greek expressive suffixes -ak1, -ak2, -ul1, -ul2 12.6 Syntactic variation in the attachment of size suffixes in Russian, German, Spanish, and Greek 12.7 Syntactic variation in the attachment of attitude suffixes in Russian and Greek 12.8 Basic properties of the Greek expressive suffixes -its1 and -its2 12.9 Basic properties of the Russian homophonous suffixes -ic1, -ic2, and -ic3

Published online by Cambridge University Press

ix

340 346 350 350 351 352 353 354 355 356 363 365 369 373 377 378 378 382 389

Contributors

Mario Brdar is Professor of English Linguistics in the Department of English Language and Literature at the University of Osijek. He was the president of the Croatian Applied Linguistics Society in 2008–2010, and its vicepresident in 2010–2012. He has been the editor of Jezikoslovlje and is a member of the editorial boards of Review of Cognitive Linguistics, ExELL, Proverbium, and Bosanski jezik. His research interests include cognitive linguistics, morphosyntax, and lexical semantics. Rita Brdar-Szabó is Professor of German Linguistics in the School of Germanic Studies and Head of the Intercultural Linguistics Doctoral Programme at Loránd Eötvös University, Budapest. Her main research interests include cognitive linguistics, morphology, lexical semantics, and contrastive linguistics. She has published on the cognitive theory of metaphor and metonymy, grammaticalization, blending, prototype theory, and usage-based models. Nicolau Dols is Professor of Catalan at the University of the Balearic Islands, Majorca, and president of the Philological Section of the Institut d’Estudis Catalans. His research interests lie mainly in the fields of phonology and, more generally, of grammar. He co-authored (with Max W. Wheeler and Alan Yates) Catalan: A comprehensive grammar (1999) and (with Richard M. Mansell) Catalan: An essential grammar (2017). His research has also addressed issues in sociolinguistics, language standardization, translation, and constructed languages. Nikolett F. Gulyás is an assistant professor at Eötvös Loránd University, Budapest. She is a co-author of the Typological Database of the Ugric Languages and the Typological Database of the Volga Area Finno-Ugric Languages. She has specialized in Permic and Ugric languages and her research interests include comparative morphosyntax, linguistic typology, and language contacts. Laura Horváth is a doctoral student at the Eötvös Loránd University researching aspect in Udmurt. She has worked on the Department of x

Published online by Cambridge University Press

List of Contributors

xi

Finno-Ugric Studies projects Multilingual Practices in Finno-Ugric Communities (2013–2017) and the Typological Database of the Volga Area Finno-Ugric Languages (2017–2022). Iraide Ibarretxe-Antuñano completed her PhD at the University of Edinburgh in 1999 and is Professor of General Linguistics at the University of Zaragoza, a researcher at the Institute for Heritage and Humanities, Zaragoza, and a fellow of the Academia Europaea (Linguistics section). Her research focuses on the relationship between language, cognition, and communication from a typological and psycholinguistic perspective. She has published on topics related to semantic typology, ideophones and sound symbolism, and the relationship between metaphor, embodiment, and culture. Mélanie Jouitteau is a researcher at the National Centre for Scientific Research, (CNRS), France. She is a formal linguist with a specialization in Breton syntax and open science, and has been developing a wikigrammar of the Breton language, ARBRES, since 2009. Haritini Kallergi holds a PhD in Linguistics from the Aristotle University of Thessaloniki and an MA degree in General Linguistics from the University of Amsterdam. She is author of Reduplication at the Word Level: The Greek facts in typological perspective (2015) and her main research interests are in the area of morphological description and expressivity. Georgia Katsouda is a research professor at the Research Centre for Modern Greek Dialects of the Academy of Athens, working on the Historical Dictionary of Modern Greek. She studied Greek Literature and Linguistics at the Faculty of Philology of the University of Athens and completed her PhD thesis entitled Morphological changes in the Verbal Stem (in Greek) in 2002. She has published on Modern Greek morphology, dialectology and dialectal lexicography. Gerrit Kentner is currently a lecturer in the Department of Linguistics at the Goethe University, Frankfurt. He has previously worked as a researcher at the Max Planck Institute for Empirical Aesthetics and at the University of Potsdam. His research is mainly concerned with the role of prosody in word formation, syntax, and language processing. Magdalene Konstantinidou is a research professor at the Research Centre for Modern Greek Dialects of the Academy of Athens, and a member of the editorial team of the Historical Dictionary of Modern Greek. She studied Classical and Modern Greek Literature and Linguistics at the University of Athens, and received her PhD in Linguistics from the University of

Published online by Cambridge University Press

xii

List of Contributors

Bielefeld. Her main topics of research are semantics, dialectology, and dialectal lexicography. She is the author of Sprache und Gefühl: Semiotische und andere Aspekte einer Relation (1997). Giorgos Markopoulos is Assistant Professor of Linguistics at the Department of Mediterranean Studies, University of the Aegean. His research interests lie primarily in the fields of theoretical linguistics and language typology and focus on morphosyntax and phonology, with particular emphasis on the cross-linguistic study of grammatical structures and phenomena. His recent publications include articles in the Catalan Journal of Linguistics, The Linguistic Review, and Transactions of the Philological Society. Iwona Piechnik is a professor at the Jagiellonian University, Cracow, and a researcher in Linguistics. She specializes in historical and typological linguistics, sociolinguistics, and historical-cultural studies, with particular interest in the evolution of communication through the media and the rise of electronic technologies and their use of language. Her main research areas focuses on the links between Indo-European and Uralic languages. Vassilios Spyropoulos is Associate Professor of Linguistics at the National and Kapodistrian University of Athens. His interests lie in the theoretical analysis of syntax, morphology, and their interfaces, as well as in the grammatical description of Greek and its dialects. He is a co-author of Greek: A comprehensive grammar (revised edition, 2012) and is author of many articles in journals such as The Linguistic Review, Transactions of the Philological Society, Linguistic Analysis, Linguistics, Studia Linguistica, Morphology, and Journal of Greek Linguistics. Olga Steriopolo is a senior researcher in Linguistics at the Leibniz-Zentrum Allgemeine Sprachwissenschaft (ZAS), Berlin, leading a research project on the interrelation between social and grammatical genders. Her research interests lie in the fields of expressive morphology, syntax, language typology, and gender studies. She has been published in The Annual Review of Linguistics, Folia Linguistica, The Journal of Slavic Linguistics, Languages, The Linguistic Review, Open Linguistics, Routledge Handbook of Pronouns, and Trends in Linguistics. Anna M. Thornton is Professor of Linguistics at the University of L’Aquila. Her main research interests are morphology (with a focus on non-canonical phenomena, such as overabundance and prosodically constrained reduction) and gender in language. She was president of the Italian Linguistic Society (SLI) in 2016–2019, and has been a corresponding member of Accademia della Crusca since 2022.

Published online by Cambridge University Press

List of Contributors

xiii

Anastasios Tsangalidis is a professor at the Department of Theoretical and Applied Linguistics, School of English Language and Literature, Aristotle University of Thessaloniki. His main research interests are in the area of syntactic and semantic description, especially in typological terms, as well as the relevance of grammar to language teaching. Further to the study of categories such as tense, aspect, mood, modality, evidentiality and mirativity, he has been particularly interested in social and expressive aspects of meaning. Thomas R. Wier is Assistant Professor of Linguistics at the Free University of Tbilisi. His work focuses on the theory and typology of polysynthetic languages, with special focus on morphological blocking, incorporation, (non)configurationality, and complex systems of case and agreement. He received his doctorate from the University of Chicago and has worked extensively on languages of the Caucasus, as well as some Native American language families, including his recent grammar and chrestomathy Tonkawa Texts. Jeffrey P. Williams is Professor of Ethnology and Linguistics at Texas Tech University in Lubbock, Texas. His research covers lesser-known varieties of English, expressivity, language endangerment, and language contact. His field research has been in the West Indies, Australia, Papua New Guinea, and Menorca. He has previously edited Expressive Morphology in the Languages of South Asia (2021), The Aesthetics of Grammar: Sound and meaning in the languages of Mainland Southeast Asia (2014), and (with Michael Aceto) Contact Englishes of the Eastern Caribbean (2003).

Published online by Cambridge University Press

Published online by Cambridge University Press

1

Introduction Jeffrey P. Williams1

1.1

Introduction to Expressivity

There is an emerging perspective in linguistics that takes expressivity as one of the key components of human communication and grammatical structure. The movement afoot challenges the discipline to reconsider its foundations, which have been based on structuralist assumptions of arbitrariness in grammar, as well as prevailing ideas about “types” of languages. The “types” of languages found in Europe have not figured as prominently as they deserve in this emerging discussion of expressivity in the world’s languages. Expressivity, as exemplified by the grammatical category called expressives, is a rapidly growing area of interest in both theoretical (Akita 2015, Dingemanse 2012, 2015, 2018, Gutzmann 2019, Iwasaki et al. 2017, McCready 2020, Potts 2005, 2007, Pullum & Rawlins 2007) and anthropological linguistics (Badenoch 2022, Badenoch et al. 2019, Webster 2008, Williams 2014). One feature that all approaches share is that expressives are somehow “special” in language. Expressivity is the grammaticalization of sensory qualities: it is used to characterize the way things appear to the senses and are articulated and conveyed in a linguistic message. Expressivity, through its iconicity, reduces the cumbersomeness of description in discourse: it allows the speaker to condense information into a single praxis. Expressivity is difficult to study on its own. The clearest path to gaining a better understanding of it lies in the study of expressives, which are also known as ideophones – the hallmarks of expressivity in language. Expressivity is no different from other properties of human language that find articulation in the grammar through principles, rules, and representations. The contributors to this volume contend that expressivity is governed by a principle of expressivity that is universal in human language, which states the following: a systematic feature of human language is the ability to articulate and communicate perception of natural and social worlds. I refer to this feature as the principle of expressivity. 1

Texas Tech University.

1

https://doi.org/10.1017/9781108989084.001 Published online by Cambridge University Press

2

Jeffrey P. Williams

The principle of expressivity is manifested through the grammatical resources of human languages. The grammar of each language has its own set of structures that can be employed by speakers to reflect on perceptions of actions, activities, states, and the social positions of individuals. This accounts for the variation we find across languages in terms of what structurally constitutes expressivity. The variation has been formally treated in a variety of ways: some are predominantly semantic, others syntactic, and a few attempt to unite both treatments through a singular formalization. Regardless of the treatment, all of these accounts focus on the category of expressives that are found in human language. The present volume stands as a significant and original contribution to the growing body of literature on expressivity in the world’s languages (e.g. Voeltz & Kilian-Hatz 2001, Dingemanse 2012, McCarthy et al. 2012, Williams 2014, Akita 2015, Dingemanse 2015, Iwasaki et al. 2017, Ibarretxe-Antuñano 2017, Dingemanse 2018, Haiman 2019, Akita & Pardeshi 2019, Williams 2021).2 It sets out to document the role of expressivity in a range of languages spoken in Europe of both Indo-European and nonIndo-European stock. The contributors to this volume demonstrate that European languages possess the intricate grammatical resources for creating elaborative, rhyming, and alliterative expressions that convey the emotions, states, conditions, and perceptions of speakers. Typically, expressives have been considered to belong to the domain of the languages of Africa, Asia, and other parts of the non-European world. As this volume’s contents demonstrate, this view is erroneous and grounded in a less than thorough description of the grammatical resources of the European languages discussed. Expressivity derives from the historical trajectory of the term “expressive” in the linguistics literature over the past century. Expressivity is the property of a language to deliver sensory information about an event, an entity, or other culturally determined category through a set of grammatical resources. This sensory information relates to culturally defined categories, categorized in the cognitive architecture of culture and grammar. Within this property of language, we find reduplication, expressives, ideophones, onomatopoeic forms, and other sound symbolic constructions). These forms and processes, unfortunately, have been relegated to the margins and footnotes of reference grammars and remain, as a result, relatively undocumented in many standard reference works. It is clear from the evidence presented by the authors contributing to this volume that the grammatically artistic usage of these forms enriches and enlivens discourse in the European sphere. 2

The following are recent exceptions to that observation: Freywald (2015), Jouitteau (2015), Kallergi (2015b), Keevalik (2010), Kentner (2017), Piechnik (2015), Steriopolo (2008).

https://doi.org/10.1017/9781108989084.001 Published online by Cambridge University Press

Introduction

3

Figure 1.1 Languages in Europe

The authors of each chapter explore expressivity in various ways, drawing on first-hand experiences either as native speakers, near-native speakers or as linguists with extensive field research experience. The contributors to this volume have been carefully chosen to provide data and analyses that cover the major language families of the region. It should be noted that at the time this proposal was conceived and developed, the COVID-19 pandemic was raging across the globe. This unfortunate situation hampered the ability of many scholars to access fundamental scholarly resources and work with native-speaker consultants. This volume stands as a tribute to those scholars who were able to maintain a high level of intellectual engagement in spite of global and local circumstances. The chapters are organized into parts according to genetic affiliation to the language families of the region, and Chapter 12 concludes the collection with comparative data drawn from a range of European languages. 1.2

Organization of the Volume

In Chapter 2, the authors examine both non-concatenative and concatenative means of signaling expressivity in Hungarian. These processes include

https://doi.org/10.1017/9781108989084.001 Published online by Cambridge University Press

4

Jeffrey P. Williams

reduplication, expressive suffixation (diminutive formation), plain truncation, and truncation followed by suffixation. The authors focus on two sets of items with hypocoristic3 function in Hungarian, both arising through the interaction of clipping or truncation with another process. There are, on the one hand, socalled embellished clippings, that is, truncations that are subsequently furnished with one or more of the diminutive or endearment suffixes of the grammar. As they point out, the array of variants found is not usual in IndoEuropean languages. The explanations they provide are that the functions seem to have either (i) exceptional simultaneous truncation of hypocoristics without embellishment but with reduplication in one set of items (as BronBron – LeBron; CoCo – Collette, from Inkelas (2012: 376), or (ii) truncation of hypocoristics with embellishment, but no subsequent reduplication (at least not exact reduplication), in the case of another set of items. The authors document the range of possibilities to account for this distribution, which is peculiar in terms of the morphological processes that interact with each other and the order in which they do so. The suggestion implicit in the distribution of the factors that may have led to or at least facilitated this state of affairs is also peculiar. The authors also examine the languages in the Volga-Kama linguistic area, such as Udmurt, Komi-Permyak and Meadow Mari, as well as Surgut Khanty and Northern Mansi, which are spoken in Siberia, and establish an inventory of reduplicative constructions which finds that very few types of these constructions seem to have been borrowed from neighboring Turkic languages. Reduplications are almost never of the exact type, but only partial and/or inexact, which are, of course, also abundant in Hungarian. In these languages, reduplication has a very important role as a tool for expressing intensification or other meanings. In Chapter 3, Iwona Piechnik provides a detailed examination of reduplication in the Finno-Ugric languages, drawing on her extensive knowledge of the languages as well as the written grammatical traditions of Russian and Finnish. As she points out, some of the languages of this large linguistic family use reduplication more productively than others. Finno-Ugric reduplication is a tool of word formation and does not serve grammatical ends, such as inflection. In other words, reduplication in Finno-Ugric belongs to lexical, not to inflectional morphology. It involves syllabic or morphemic doubling, and some reduplicated clusters are already so lexicalized that they are not perceived as expressive anymore. However, sometimes their lexicalization is weak because they have rather an affective nuance.

3

Hypocoristics are discussed in Chapter 2 §2.3.

https://doi.org/10.1017/9781108989084.001 Published online by Cambridge University Press

Introduction

5

As Piechnik makes clear, the morphological properties of Finno-Ugric languages favor reduplication: they are agglutinative, hence easily meld various stems, allowing them often to keep the stress on the initial syllable. In the Finno-Ugric languages, reduplication has a secondary function as expressivity. Sometimes, it is even its most salient feature. Reduplicated stems are never only denotative. They have their expressive part too, in other words, they are (more or less) expressives par excellence. Most of all, expressivity is in the very structure of echo-pairs (echo constructions). Their semantic value comes from their form (which is often of onomatopoeic or ideophonic origin, hence semantically iconic), or rather from the concentration of reduplicated units. Chapter 4 documents the fact that Scots possesses a wide range of forms of expressivity as part of its grammatical repertoire – including most notably echo word formations. The chapter focuses on the category of echo words in Scots expressivity. Scots, like too many other European languages, is viewed as possessing a dearth of expressivity in spite of evidence to the contrary. Nuckolls (2003) has claimed that ideophones as a class of expressions are underdeveloped as a category in Standard Average European languages. The perception is that ideophones are a common feature in many of the world’s languages but are underdeveloped in English and other Indo-European languages. As the chapters in this volume attest, that assertion is incorrect. In Chapter 5, Gerrit Kentner investigates reduplication in German. In opposition to what other scholars have noted about the systematicity of German reduplication, Kentner shows that German has various types of reduplication, chief among them rhyme reduplication (Schickimicki “posh person” < schick “posh”), ablaut reduplication (Mischmasch “jumble” < misch “to mix”), and full or total reduplication (Kaffeekaffee “coffee coffee” = real coffee/standard variety coffee). Kentner’s approach is to survey recent research into the various reduplicative constructions in German, with the aim to systematize them, describe their morphophonology and their conditions of use. He also discusses the expressive functions that the various kinds of reduplication may have. Kentner sets the stage by first revisiting the iconic meaning potential ascribed to reduplication in general and cross-linguistically. He goes on to discuss and describe the morpho-phonology and the use conditions of the various types of reduplication in German, suggesting that the specific conditions of use and the poetic form together engender the expressive function of reduplication in German. Against this background, the contradistinction between plain and expressive morphology (Zwicky & Pullum 1987) is taken up and critically discussed in the final section of his chapter. In Chapter 6, Haritini Kallergi, Georgia Katsouda, Magdaele Konstantinidou, and Anastasios Tsangalidis investigate aspects of expressivity in Standard Modern Greek, specifically cases in which expressivity exhibits

https://doi.org/10.1017/9781108989084.001 Published online by Cambridge University Press

6

Jeffrey P. Williams

the characteristic [+negative] or pejorative. Their research focuses on the level of morphology, particularly on productive word formation, both through compounding (first compound constituents with pejorative meaning, as in vromokánalo “filthy channel”) and derivation (derivational suffixes with pejorative functions, such as vuleftéi “bad ministers,” ipalilákos “insignificant clerk” and fititarjó “a student lot,” among others). It should be noted that although the particular SMG morphological phenomena/devices have been scatteringly studied in earlier Greek literature, negative expressive meaning, that is, pejoration, has not been systematically dealt with so far. The authors point out that further research could be conducted on the relation between expressivity and factors such as metaphor, irony, and type of register. One more specific (lexicographic) dimension of this research might concern the connection of register labelling, as used in dictionaries, and elements of pejorative meaning; admittedly, what might sometimes be presented as a “ludic use” (παικτική χρήση) of a dictionary entry in fact describes its status as a pejorative. In Chapter 7, Anna Thornton gives a thorough description of reduplicative elements and constructions in Italian. Her survey shows that Italian makes use of various sorts of reduplicative processes, targeting elements ranging from single segments to words belonging to various parts of speech to sentence fragments and even complete utterances. As Thornton points out, the pattern that emerges is quite familiar: many of the reduplicative constructions found in Italian have parallels in other languages, in particular languages of the Mediterranean area, such as Greek (Kallergi 2015a) and Maltese (Stolz et al. 2011). The conclusion that Thornton reaches is that certain types of reduplicative lexical items or constructions appear to have originated in Italo-Romance varieties spoken in regions bordering the Mediterranean, and spread beyond over time. Her conclusion seems to hold with other information presented in this volume. Nicolau Dols’s contribution (Chapter 8) provides a detailed study of expressives in Catalan relying on a body of twelve folk tales, which were voice-dramatized and broadcast on radio in Majorca in 1959. The total data set he draws upon amounts to a total of 63,000 words and just over five hours of recordings. Another source of the lexical material for his study has been elicited from the Diccionari Català-Valencià-Balear. Dols determines from among the features studied in his analysis the observation that pitch differences reveal a tendency for expressives to appear in fragments with uneven fundamental frequencies. The segmental features of expressives show matches of sound and meaning that have been observed in other languages. The corpus includes repetitions with morphological alternations and the adverbial use of the structure Verb + who + Verb expressing continuity or repetition.

https://doi.org/10.1017/9781108989084.001 Published online by Cambridge University Press

Introduction

7

In Chapter 9, Mélanie Jouitteau provides an organized inventory and a discussion of expressives in Breton, a modern Celtic language that exists in a bilingual sociolinguistic context with French in parts of western France. She defines expressives as forms whose morphophonology is not entirely arbitrary and is instead, partly iconic. She develops a discussion of the productivity of the operations of expressive morphology, its exclusive use for expressive means, and its degrees of iconicity. For each category she discusses what operations and structures might be exclusive to expressive words, focusing on the morphological hallmarks of Breton expressives: reduplication, apophonic alternations, and a trisyllabic recurring pattern. Breton expressive morphology uses reduplication, apophonic alternations, and/or a specific trisyllabic pattern, though none of those are exclusive to expressive morphology. Basque, a language spoken at the western edge of the Pyrenees in southwestern Europe, possesses a large, distinctive class of iconic elements known as ideophones. Iraide Ibarretxe-Antuñano succinctly describes the main typological characteristics of these words in Chapter 10, arguing that, due to their prominent status in Basque and in linguistic typology in general, they should be considered a main typological trait of Basque, similar to other specific linguistic features ascribed to the language such as ergativity, case alignment, and double marking. Basque’s large repertoire for ideophones grants this language its categorization as a highly-ideophonic language. This fact serves as a good case in point to refute the general claim that distribution of this type of words is limited to languages in Asia, Africa, and the Americas, as well as to justify ideophones as an important typological trait of Basque on the same level as its other morphosyntactic typological characteristics. In Chapter 11, Thomas Wier surveys the various kinds of expressive language present in the three autochthonous Caucasian families: AbkhazAdyghean, Kartvelian, and Nakh-Daghestanian. He also considers in depth the specific morphological and syntactic peculiarities of expressives in Georgian, which exhibit exuberant consonant clusters, processes of reduplication uncharacteristic of the language as a whole, and specific morphosyntactic alignment splits between different classes of expressive. As he states, expressives are not one thing, but many. Wier’s chapter also looks in considerable depth at the behavior of Georgian expressives, and finds that it is not uniform even within this one language. Different subclasses of expressives may manifest different kinds of exponence, either categorically or statistically, and other categories may crosscut these generalizations. Georgian thus shows that expressives are not a single thing, but instead, many different things, one thrown into sharp relief by the fascinating and typologically unusual system of the language’s morphosyntax.

https://doi.org/10.1017/9781108989084.001 Published online by Cambridge University Press

8

Jeffrey P. Williams

Chapter 12, by Olga Steriopolo, Giorgos Markopoulos, and Vassilis Spyropoulos, compares morphosyntactic properties of expressive suffixes in four European languages: Russian, German, Spanish, and Greek. It shows that although these suffixes share the same expressive meaning, they differ significantly in their syntactic structure, namely in the manner and place of attachment in the syntactic tree. Thus the Russian and Spanish expressive suffixes that refer to the size of a referent (or size suffixes) are syntactic modifiers, while the German size suffixes are syntactic heads. And in Greek, the two most productive expressive suffixes –ak and –ul have homophonous counterparts that possess contrasting syntactic properties: syntactic heads vs. syntactic modifiers. This shows that across languages as well as within single languages such as Greek, there is no one-to-one correspondence between the meaning and the structure of expressive forms. These findings are further supported by two novel case studies of the homophonous suffixes –its (in Greek) and –ic (in Russian). A final point that they emphasize is that syntactic variation emerges not only across languages with different syntactic structures, but also within the same language, occurring both across different vocabulary items. The authors reach the conclusion that expressive meaning is not associated with any specific syntactic structure, either across languages or within a single language. 1.3

Conclusion

The contributions to this volume take a variety of approaches to the study of expressivity – some more formal and others more sociolinguistic in nature. Regardless of the approach, the contributions here demonstrate the vitality of expressivity in the languages of the European linguistic sphere and the importance of the data provided to a general understanding of expressivity as a linguistic phenomenon. The information provided by the languages of Europe also demonstrates the complex nature of the expressive system in human languages. Further documentation is certainly needed but this volume stands as a first contribution to a body of work on comparative expressivity that draws on complex linguistic data to provide compelling conclusions. References Akita, Kimi. 2015. Sound symbolism. In Jan-Ola Östman & Jef Verschueren (eds.) Handbook of Pragmatics. Amsterdam: Benjamins, 1–24. Akita, Kimi & Prashant Pardeshi (eds.) 2019. Ideophones, Mimetics and Expressives. Amsterdam: Benjamins. Badenoch, Nathan. 2022. Silence, cessation, and stasis: The ethnopoetics of “absence” in Bit expressives. Journal of Linguistic Anthropology 32: 94–115.

https://doi.org/10.1017/9781108989084.001 Published online by Cambridge University Press

Introduction

9

Badenoch, Nathan, Madhu Purti & Nishaant Choksi 2019. Expressives as moral propositions in Mundari. Indian Linguistics 80: 1–17. Cruse, David A. 1986. Lexical Semantics. Cambridge: Cambridge University Press. Dingemanse, Mark. 2012. Advances in the cross-linguistic study of ideophones. Language and Linguistics Compass 6(10): 654–72. DOI: https://doi.org/10.1002/ lnc3.361 2015. Ideophones and reduplication: Depiction, description, and the interpretation of repeated talk in discourse. Studies in Language 39(4). 946–70. DOI: https://doi .org/10.1075/sl.39.4.05din 2018. Redrawing the margins of language: Lessons from research on ideophones. Glossa: A journal of general linguistics, 3(1): 4. DOI: https://doi.org/10.5334/gjgl .444 Freywald, Ulrike. 2015. Total reduplication as a productive process in German. Studies in Language 39: 905–45. Gutzmann, Daniel. 2019. The Grammar of Expressivity. Oxford: Oxford University Press. Haiman, John. 2019. Ideophones and the Evolution of Language. Cambridge: Cambridge University Press. Ibarretxe-Antuñano, Iraide. 2017. Basque ideophones from a typological perspective. Canadian Journal of Linguistics, 62(2): 196–220. DOI: https://doi.org/10.1017/ cnj.2017.8 Inkelas, Sharon. 2012. Reduplication. In J. Trommer (ed.) The Morphology and Phonology of Exponence. Oxford: Oxford University Press, 355–78. Iwasaki, Noriko, Peter Sells & Kimi Akita (eds.) 2017. The Grammar of Japanese Mimetics: Perspectives from structure, acquisition, and translation. Abingdon: Routledge. Jouitteau, Mélanie. 2015. Free-choice and reduplication: A study in Breton dependant indefinites. In Tomasz Czerniak, Maciej Czerniakowski & Krzysztof Jaskuła (eds.) Representations and Interpretations in Celtic Studies. Lublin: Wydawnictwo KUL, 978–83. Kallergi, Hartini. 2015a. Reduplication at the Word Level: The Greek facts in typological perspective. Berlin and New York: Mouton de Gruyter. DOI: https:// doi.org/10.1515/9783110365597 2015b. Total reduplication as a category of expressives. In Daniela Rossi (ed.) The Why and How of Total Reduplication. Amsterdam: Benjamins, 873–904. Kallergi, Hartini & Magdalene Konstantinidou. 2018. Reduplicative constructions involving distortion. In Aina Urdze (ed.) Non-prototypical Reduplication. Berlin: de Gruyter, 91–150. Keevalik, Leelo. 2010. Social action of syntactic reduplication. Journal of Pragmatics 42(3): 800–24. Kentner, Garritt. 2017. On the emergence of reduplication in German morphophonology. Zeitschrift für Sprachwissenschaft 36(2): 233–77. McCarthy, John J., Wendell Kimper, & Kevin Mullin. 2012. Reduplication in harmonic serialism. Morphology 22: 173–232. McCready, Elin. 2020. Expressives. In Daniel Gutzmann et al. (eds.) The Wiley Blackwell Companion to Semantics. Hoboken, NJ: Wiley-Blackwell, 1–29. Nuckolls, Janis. 1996. Sounds like Life: Sound-symbolic grammar, performance, and cognition in Pastaza Quechua. Oxford: Oxford University Press.

https://doi.org/10.1017/9781108989084.001 Published online by Cambridge University Press

10

Jeffrey P. Williams

2003. To be or not to be ideophonically impoverished. Texas Linguistics Forum 47: 131–42. Piechnik, Iwona. 2015. Reduplicative syllables in Romance languages. Romanica Cracoviensia 15: 30–55. Potts, Christopher. 2005. The Logic of Conventional Implicatures. Oxford: Oxford University Press. 2007. The expressive dimension. Theoretical Linguistics 33(2): 165–98. Pullum, Geoffrey K. & Kyle Rawlins. 2007. Argument or no argument? Linguistics and Philosophy 30: 277–87. Steriopolo, Olga. 2008. Form and function of expressive morphology. PhD dissertation, University of Vancouver. Stolz, Thomas, Cornelia Stroh, & Aina Urdze. 2011. Total Reduplication: The areal linguistics of a potential universal. Berlin: Akademie Verlag. Voeltz, F. K. Erhard & Christa Kilian-Hatz (eds.) 2001. Ideophones. Amsterdam: Benjamins. Webster, Anthony K. 2008. “To give an imagination to the listener”: The neglected poetics of Navajo ideophony. Semiotica 171: 343–65. Williams, Jeffrey P. (ed.) 2014. The Aesthetics of Grammar: Sound and meaning in the languages of Mainland Southeast Asia. Cambridge: Cambridge University Press. (ed.) 2021. Expressive Morphology in the Languages of South Asia. Abingdon: Routledge. Zwicky, Arnold M. & Geoffrey K. Pullum. 1987. Plain morphology and expressive morphology. In Proceedings of the Thirteenth Annual Meeting of the Berkeley Linguistics Society, 330–40. DOI: https://doi.org/10.3765/bls.v13i0.1817

https://doi.org/10.1017/9781108989084.001 Published online by Cambridge University Press

Uralic

https://doi.org/10.1017/9781108989084.002 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.002 Published online by Cambridge University Press

2

Hypocoristic reduplications and embellished clippings in Hungarian and elsewhere Mario Brdar*, Rita Brdar-Szabó{, Nikolett F. Gulyás{, and Laura Horváth{

2.1

Introduction

2.1.1

What is expressivity?

Like many other notions in linguistics and related disciplines, expressivity and expressives are shrouded in uncertainty and controversy, and are contested by linguistic camps of various interests and/or ideologies.1 The term ‘expressive language’ is used in the study of language acquisition and disorders, contrasted with the term ‘receptive language’ (Smolak 1982). The former, according to Frazier (2011: 620) ‘refers to the way a child expresses him/herself for everyday wants, needs, and feelings. Spoken, written, and body language, including facial expressions and sign language, are all abilities considered to be expressive language skills.’ In a way, this notion of expressive language is related to Searle’s Principle of Expressibility (PE), according to which ‘[f]or any meaning X and any speaker S whenever S means (intends to convey, wishes to communicate in an utterance, etc.) X then it is possible that there is some expression E such that E is an exact expression or formulation of X’ (Searle 1969: 20). In other words, whatever can be conceptualized, can also be said or expressed in one way or another. As pointed out by Haiman (2013: 61), ‘[i]t is a commonplace belief in the Western linguistic tradition that language exists for packaging information as clearly and as economically as possible.’ This idea that language has no other purpose than to transmit information (Haiman 2013: 61) is also implicitly present in Searle’s PE. However, Bisang (2009: 34) notes that ‘morphosyntactic structures and their properties can never fully express the meaning they have in a concrete speech situation’ (our emphasis), due to what Levinson calls ‘the { * University of Osijek Eötvös Loránd University, Budapest 1 The research was supported by the Hungarian Scientific Research Fund Grant No. K125282 for Nikolett F. Gulyás and Laura Horváth and No. FK143242 for Nikolett F. Gulyás. We are grateful to Tatiana Efremova, Yulia Speshilova, Larisa Ponomareva, Sampsa Holopainen, and Bogáta Timár for their useful comments. We also owe gratitude to the reviewers of the present chapter. All the remaining errors are the sole responsibility of the authors.

13

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

14

Brdar, Brdar-Szabó, F. Gulyás, Horváth

articulatory bottleneck’ (Levinson 2000: 29) arising because of the speed asymmetry between articulation (encoding) and interpretation (decoding), the latter being much quicker. Given this problem, linguists face a choice of two options. One possibility, embraced by formal linguists like Diffloth (1976, 1979) and Zwicky and Pullum (1987) is to separate plain or core morphology from a special area of expressive or extra-grammatical morphology (Dressler & Merlini Barbaresi 1994), the latter being irregular and allegedly playing no significant role in productive morphology (see also Inkelas & Zoll 2005, Potts 2007, Raimy 2009, McCarthy et al. 2012). The other possibility rests on the assumption that language has functions other than purely referential (Bühler 1934, Jakobson 1960). Talking about cognitive science undergoing an ‘emotional turn’ in recent years, Foolen (2015) points out that “besides the rational part of the brain, the ‘other half of the human mind’” should also be taken into consideration. But just as one half of the brain cannot function without the other, the emotive or expressive and the denotative function of language are likewise inseparable, that is, they are co-present in all utterances, though one may be more dominant than the other. In fact, a range of linguistic devices can be arranged on a continuum as being more or less expressive, and some of them may appear to do nothing other than convey an expressive sort of meaning, for example, interjections. It is in this sense that we use the term ‘expressive morphology’ to refer to various linguistic devices that arise as a result of some morphological operations which seem to be used primarily or to a considerable degree to indicate ‘attributional qualities, feelings and opinions, as well as a means of metacommentary’ (Williams 2021: 1), without tying expressivity exclusively to certain morphological phenomena. 2.1.2

How is expressivity expressed?

Now that we have outlined the notion of expressivity as a feature of language that is in theory present anywhere in language and not just confined to some areas of it, we should nevertheless try to determine its scope, that is, provide its most frequent forms of materialization. This can be done along either the intensional or extensional dimensions. As far as its intension is concerned, the term is delineated in approaches such as Cruse (1986) or Potts (2005) by setting up a diagnostics relying on the truth-conditional meaning. Propositional meaning is truth-conditional, while expressive meaning is truth-conditionally independent. Both Cruse (1986: 272) and Potts (2005) further characterize expressives as immediate and nondisplaceable. Potts (2005) also adds that expressives are descriptively ineffable and perspective-dependent, and that they can be repeated in an utterance. On the extensional side, it transpires that, as far as its formal side is concerned, the term has been used in two fairly narrow ways in the literature.

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

Hypocoristic reduplications & embellished clippings in Hungarian

15

Diffloth (1972) was the first to use the term for a separate class of complex words in the languages of Southeast Asia (Williams 2014: 6). Similarly, Burenhult (2002: 162) defines expressives as a category of words which form a distinct word-class in many Austroasiatic languages and ‘denote sensory perceptions of the speaker – visual, auditory, tactile, olfactory, gustatory, emotional or other – in relation to a particular phenomenon. . . . They often display peculiar phonological and morphological features, and they function syntactically like sentence adjuncts.’ On the other hand, expressives have often been equated with just one particular phenomenon, most notably reduplication. As observed by Sidwell (2013: 18), ‘a strong feature of these expressives is reduplication, either full or partial, often with segmental substitutions or alternations that behave like pseudomorphemes but are difficult or impossible to systematize.’ Expressives have therefore come to be equated with reduplication, as in Watson (1966), or Anderson (2008). Similarly, Kruspe (2004) lumps expressives together with ideophones. While practically any morphosyntactic and lexical phenomenon or their combination may exhibit expressive features, such as syntactic repetition or reduplication, onomatopoeia, phonaesthemes, swearwords or expletives, honorifics, particles, suffixation (especially diminutive or augmentative, or pejorative), clipping (or truncation), compounding, blending, or reduplication, we will focus in this chapter on just three of these that basically belong to word-formation: clipping, suffixation, and reduplication, as well as their interactions. 2.1.3

The goals and organization of this chapter

The present chapter focuses on two sets of items with hypocoristic function in Hungarian, both arising through the interaction of clipping or truncation with another process. There are, on the one hand, so-called embellished clippings: truncations that are subsequently furnished with one of the diminutive or endearment suffixes, or even more than one, as in Example (1): (1)

Feri/Fecó/Fercsi/Ferike/Ferkó (Ferenc) Zoli/Zotya/Zotyi/Zolesz/Zolcsó/Zoló/Zolika (Zoltán) Kata/Kati/Kató (Katalin) Karcsi/Karesz (Károly) doki (doktor ‘doctor’) gimi (gimnázium ‘grammar school’) kolesz (kollégium ‘dormitory’) ubi (uborka ‘cucumber’) pari (paradicsom ‘tomato’) ruci (ruha ‘dress’) naci (nadrág ‘trousers’)

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

16

Brdar, Brdar-Szabó, F. Gulyás, Horváth vinyó (vincseszter/Windows ‘hard disk’)

On the other hand, there are also reduplications, typically consisting of two identical CV syllables. These can be based on first names, surnames, or some common nouns referring to people, denoting kinship relationships, profession, or object, as in Example (2): (2)

Zozó (Zoltán) Fefe (Ferenc) Zsozsó (Zsófia, but perhaps also from Erzsó, a hypocoristic from Erzsébet) Kokó (Kovács) Sziszi (Szijjártó) Titi (Tibor) dodó (doktor) gigi (cigi, from cigaretta ‘cigarette’)

Except for reduplications that are based on surnames, which are not numerous, there is also at least one parallel hypocoristic form of the former type for all the others, that is, an embellished clipping, as in Example (3): (3)

Zoli/Zotya/Zotyi/Zolesz/Zolcsó/Zoló – Zozó (Zoltán) Zsófi/Zsóca – Zsozsó (Zsófia) Lajcsi – Lala (Lajos) Tibi – Titi (Tibor) doki – dodó (doktor)

This state of affairs, the array of variants we find, is not usual in Indo-European languages. They seem to have either (i) exceptional simultaneous truncation of hypocoristics without embellishment and reduplication in one set of items (as BronBron – LeBron; CoCo – Collette, from Inkelas (2012: 376)), or (ii) truncation of hypocoristics with embellishment, but no subsequent reduplication (or at least not exact reduplication), in the case of another set of items. In other words, the two sets do not overlap in these languages, but they do in Hungarian. Additionally, Hungarian can also have (hypocoristic) truncations without embellishment (prof – professzor). It is our goal in this chapter to document the range of possibilities that account for this distribution, which is peculiar in terms of the morphological processes that interact with each other and the order in which they do so. The suggestion implicit in the distribution of the factors that may have led to or at least facilitated this state of affairs is also peculiar. In §2.2, we take a closer look at some non-catenative, and catenative means of signaling expressivity in Hungarian. Specifically, these are reduplication (§2.2.1), plain truncation (§2.2.2) and truncation followed by suffixation (§2.2.3). In §2.3 we turn to Hungarian proper name hypocoristics of the types illustrated in Examples (1)–(3). We first examine hypocoristic truncations with embellishing suffixes, and then hypocoristic reduplications. The parallelism between the two is considered in the light of the Hamilton-

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

Hypocoristic reduplications & embellished clippings in Hungarian

17

Kager prediction and we suggest the most parsimonious explanation for this state of affairs. The genetic and/or areal factors leading to or facilitating this situation are considered in §2.4, where we look at hypocoristic reduplication and truncation in other languages. A set of brief conclusions rounds off the chapter in §2.5. 2.2

Some concatenative and non-concatenative means of signaling expressivity in Hungarian

2.2.1

Reduplication in Hungarian

Reduplication, that is, reduplicative construction (RC) is defined by Moravcsik (1978: 323) as ‘a pattern where the double or multiple occurrence of a sound string, syllable, morpheme, or word within a larger syntagmatic unit is in systematic contrast with its single occurrence, with the iterated elements filling functionally non-distinct positions.’ The outcome of reduplication as a morphological process is a single word, that is, a complex morphological construction consisting of at least two parts. In traditional terminology, one part is considered the base, that is, the part that is copied, and the other is considered to be the reduplicant, that is, a copy of the base. The base is typically an independently existing word, occurring alone, and possibly in combination with other free or bound lexical forms. What is reduplicated need not always be a free form, that is, a word. The base for reduplication can also be a bound form, that is, a bound morpheme, either inflectional or derivational. Kiefer (1995–1996) discusses the reduplication of verbal prefixes in Hungarian. The construction that is illustrated in Example (4) is also used to indicate iterativity: (4)

A városká-ban sétálva be-be-néz-t-ünk a templom-ok-ba. the little_town-ine walk-cvb red-pref-look-pst-1pl the church-pl-ill ‘While walking around in the little town, we looked into different churches.’

Reduplication has a whole range of functions, intensification being probably the most frequent and most important one in the case of adjectives and adverbs (Brdar & Brdar-Szabó 2012). Reduplication can, for example, be used to express plurality (5), collectivity (6), iteration (7), intensification (8), or diminution (9), as shown in the following set of examples from sundry languages: (5)

anak ‘child’ vs. anak-anak ‘children’ gunung ‘mountain’ vs. gunung-gunung ‘mountains’

(Malay; Moravcsik 1978) (6)

rumah ‘house’ vs. rumah-rumah ‘houses/houses collectively/various houses’

(Indonesian; Rafferty 2002)

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

18

Brdar, Brdar-Szabó, F. Gulyás, Horváth

(7)

guyon ‘jest’ vs. guguyon ‘to jest repeatedly’

(Sundanese; Regier 1998) (8)

sweet-sweet ‘very sweet’ bitter-bitter ‘very bitter’

(Singapore English; Wong 2004: 343) (9)

djidis ‘tooth’ vs. djidjidis ‘little tooth’

(Comox; Regier 1998)

According to whether the whole base is copied or not it is possible to distinguish between total, or full, and partial reduplication. Reduplication is total when the entire base is copied, as shown in Examples (5) and (6), from Malay and Indonesian, respectively. The total reduplication of adjectives and adverbs for intensification is attested in some cases in Hungarian: (10)

a. b. c. d.

sok-sok [many-many] ‘very many’ alig-alig [hardly-hardly] ‘with great difficulty’ néha-néha [sometimes-sometimes] ‘very seldom’ messze-messze [distant-distant] ‘very distant’

We have stated that reduplication is total when the entire base is copied and partial if it copies only a part of the base. However, taking a closer look at the issue, we realize that in theory there are at least four possibilities for reduplication considering the degree of identity between the base and the copy. The total reduplication can copy the entire base in an exact manner, that is, without any modification. This is what we have seen in the present section so far, the possibility that might be referred to as exact total reduplication. It is, however, also possible that it copies the whole base but that some parts get modified, that is, replaced by something, and/or that some elements are added to the base in the copy. This may be referred to as inexact total reduplication. Further, just some part of the base may be copied with or without any further modification and subsequent addition. The former may be called exact partial reduplication, the latter inexact partial reduplication. All four labels have been used in literature, but their delimitation may be quite problematic, particularly when a segmentally different copy that contains some additional material is involved. Therefore, all the reduplications in which the reduplicant is somewhat different from the base will be treated here as partial ones. Partial reduplication copies only a part of the base. The part that is reduplicated can be defined in terms of phonological or segmental units. For example in Agta, plurals of nouns can be formed by copying not the whole stem but just its first (consonant-)vowel-consonant sequence (Healey 1960: 7): (11)

takki ‘leg’ vs. tak-takki ‘legs’

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

Hypocoristic reduplications & embellished clippings in Hungarian

19

This state of affairs could alternatively be described in terms of truncation, too. We could say that the base is truncated and then copied in front of itself. This type of partial reduplication, in which there are clear signs of truncation in the copy, is found in a number of languages where reduplicated adjectives indicate extreme smallness. The copy (preceding the base) loses the initial plosive consonant, for example, in English, itsy-bitsy, itty-bitty, or in Dutch, ietsiebietsie ‘very little’. The same phenomenon is found in Hungarian: icipici, icilipicili, iciri-piciri, icur-picur ‘very small’, from pici ‘small’. The opposite process is also possible, that is, the extension of the base in the reduplicant, for example, in the English adjective easy-peasy or argy-bargy, where a plosive sound appears in the reduplicant. In what Moravcsik (1978: 327) considers to be an example of total reduplication in Hungarian, that is, teli-s-teli [full-der-full] ‘completely full’, we actually have the copy of the adjective teli ‘full’ extended by -s, elsewhere used as an adjectivizing suffix. Similarly, in rég-es-régi [old-der-old] ‘very old’, we assume that the base is preceded by an extended copy. In adverbs örökk-ön-örökk-é [eternal-supeternal-tra] ‘for all times, eternally’ and kör-ös-kör-ül [circle-der-circleder] ‘all around’, both constituents contain suffixes. They are similar to the English reduplication cloppety-clop, where the reduplicant is extended by an extra syllable. While the extensions in the above look like morphemes, the extension in Hungarian can be just a single phoneme, similar to the English easy-peasy, as in ingó-bingó ‘moving rhythmically,’ or in izé-mizé ‘thingy’. Finally, there are also cases in which there is an element that is modified within the copy, the two otherwise being of the same ‘size’, that is, not exhibiting truncation or extension. These come in two main types. The first type is characterized by onset alternation such that an initial segment of the base (typically a consonant) is replaced by a different segment in the copy (i.e. another consonant), which results in phonological overlap between the rest of the constituents, hence their being labelled as rhyming(-motivated) reduplications, as in English words walkie-talkie, hanky-panky, hurly-burly, or piggy wiggy. Compare also Hungarian tarka-barka ‘very colourful’, from tarka ‘colourful’, cserebere ‘swap’, csiribiri ‘insignificant’, csingilingi ‘ting-a-ling’, locspocs ‘slush/sludge’, or lárifári ‘nonsense/codswallop/bunkum’. In the other type, the ablaut reduplications, the vowel of the first constituent is almost always a high vowel while the second constituent contains a low vowel. There is also a tendency for the former to be a front and for the latter to be a back one. In the second type we observe vowel antiphony, that is, a vowel alternation pattern similar to ablaut, hence the label ablaut-motivated reduplications. Among the ablaut-motivated reduplicated adjectives and adverbs expressing intensity in English we find: (12)

a. teeny-tiny ‘very small’ b. tip-top ‘excellent’

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

20

Brdar, Brdar-Szabó, F. Gulyás, Horváth

Ablaut-motivated reduplicated adjectives expressing intensity can also be found in Hungarian. In some cases both constituents end in the suffix -(V)s, which is elsewhere used to derive adjectives from nouns. The pattern seems to be still productive, as is seen in the examples in (13): (13)

a. b. c. d. e. f.

rissz-rossz ‘very bad’, from rossz ‘bad’ fidres-fodros ‘very much curled, curly’, from fodor ‘wave, curl’ gidres-gödrös ‘full of pits and holes’, from gödör ‘pit’ girbegörbe, girbegurba ‘full of curves’ from görbe ‘crooked’ dimbes-dombos ‘hilly, have a wavelike appearance’, from domb ‘hill’ gizgazos or gizes-gazos ‘very weedy, overgrown with weeds’, from gaz ‘weed’

There are also nouns like mendemonda ‘hearsay, scuttlebutt’ and verbs like kipeg-kopog ‘produce rhythmic sound’. Both types of these reduplications can be easily found in the Late Old Hungarian period (Lőrinczi 1992: 854–57) but not in Early Old Hungarian: the reason for this could be the presumably later origin of these formations – or the type of texts attested for the Early Old Hungarian period, since these playful, expressive words hardly fit into the narrative of clerical texts (Zelliger 1991: 527). However, in addition to these two main types, we also find hybrid formations. In the case of irgum-burgum, an adjective expressing feigned anger, mainly found in fairy tales, or used in addressing children, it is difficult to decide whether we are dealing with an extended or truncated ablaut-motivated reduplication, as neither of its parts is used as an independent word. Idres-bodros, related to fidres-fodros above, could be a truncated ablaut-motivated reduplication.

2.2.2

Truncation/clipping in Hungarian: when less is more

Several methods of shortening words involve a reduction of the input word or words in one way or another, and can therefore be considered to be subtractive processes. However, in some cases subtraction may be followed or simultaneously accompanied (it is positively difficult to decide in some instances which is the case) by an additive word formation process, such as suffixation or compounding. The theoretical status of subtractive morphological processes seems to be unassailable in inflectional morphology, but scholars dealing with word formation have, on the whole, been less enthusiastic about admitting subtraction into the word formation part of morphology. In fact, a number of researchers have explicitly rejected some or most of such subtractive phenomena as forming part of word formation proper (see Haspelmath 2002: 25, Štekauer 1998: 23, Bauer & Huddleston 2002: 1632). Truncation, also called clipping, is a subtractive way of producing new words by cutting off part or parts of an already existing lexical item, resulting in a phonologically shorter form, as is seen in the instances in (14):

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

Hypocoristic reduplications & embellished clippings in Hungarian (14)

21

a. ad(vert(isement)) b. plane (from airplane) c. flu (from influenza)

Adopting the terminology used by Bauer & Huddleston (2002: 1634), we may call the source of the clipping the original, the phonological material that is cut away may be termed the surplus, while the remaining material forming the new word may be termed the residue. The residue is usually initial, as in (14a). This type of clipping is called back-clipping. Less commonly, the residue may be the final part of a word, as in (14b). This type of clipping is called fore-clipping. Occasionally, material can be clipped at both ends of a word, as in (14c). This type is referred to as mid-clipping, ambi-clipping, or syncope (Jamet 2009). An interesting property of clipping is that, unlike central word formation processes such as affixation or compounding which always bring about some relatively tangible semantic differences between the semantic ranges of their bases and outputs, Bauer (1983: 233) and Mattiello (2013: 15, 22) claim that it does not modify the denotative meaning of the source word: For instance, in the case of lab, reduced from laboratory, we can observe a reduced signans in the output, but not a reduced signatum. It is clear that the signantia are dissimilar, but not the signata, which are denotationally and referentially (although not contextually) identical. In this case, then, as we go from input to output there is a reduction in the degree of iconicity. (Mattiello 2013: 22)

However, it seems that in the case of clippings, again punning on Bisang (2009), less is actually more. It is of course often noted that there is a difference in the register between the source and the residue, the latter normally belonging to informal, colloquial speech, or to the special register of particular social groups. Clippings are thus often classified as belonging to expressive or extra-grammatical morphology. Many discussions of English word formation point out as a sort of ‘social meaning’ (Plag 2003: 18) the expressive value of clippings because they indicate familiarity with the denotatum of the clipping and simultaneously signal a certain degree of closeness between the speaker and the listener, thus changing its stylistic value: Clipping refers to the process whereby a lexeme (simplex or complex) is shortened, while still retaining the same meaning and still being a member of the same form class. Frequently clipping results in a change of stylistic level. (Bauer 1983: 233)

There is actually not much clipping proper in Hungarian. In addition to some borrowings that were clipped in the donor language, such as celeb, pronounced as /tseleb/, busz ‘bus’, labor, there are just a few genuine clippings in the present-day Hungarian. We have already mentioned above prof, to which we may add szitu from szituáció ‘situation’, tulaj from tulajdonos

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

22

Brdar, Brdar-Szabó, F. Gulyás, Horváth

‘owner’, mikró from mikrohullámú sütő ‘microwave oven’, bá from bácsi/ bátya ‘uncle’, kösz ‘thanks’ from köszönöm, mob or fon, the back- and foreclipping, respectively, from mobiltelefon ‘mobile phone’ (Veszelszki 2013: 190, 2017). Clippings are also found in hybrid compounds (sometimes considered to be blends, see Brdar-Szabó & Brdar 2008), like maszek ‘private’, from magán ‘own’ and szektor ‘sector/sphere’, in which both constituents were clipped, or compounds in which only one was clipped, for example, reptér ‘airport’, from repülőtér. As native speakers need not be familiar with the morphological structure of borrowed words, face and féjsz can also be considered as clippings of facebook and féjszbuk/fészbuk, respectively. 2.2.3

Truncation combined with suffixation in Hungarian: when more is not necessarily more

A peculiar feature of clipping is that in English (and many other languages) instances are often extended by hypocoristic suffixes (-y/-ie, -o, -s, -er or -ers/as). Consider the following examples, mainly from Australian English: (15)

telly (from television), rego (from car registration), Duncs (from Duncan), leccer (from lecture), Wimbers (from Wimbledon)

Hungarian embellished clippings of common nouns, found mostly in colloquial language and slang, almost invariably end in -(C)i, though there are some that end in ‑(C)o , -(C)a, in -(V)sz , or -(C)s, (Brdar-Szabó & Brdar 2008): (16)

a. gimi from gimnázium ‘grammar school’ töri from történelem ‘history’ koli from kollégium ‘dormitory’ cigi from cigaretta ‘cigarette’ aksi from akkumulátor ‘battery’ ubi from uborka ‘cucumber’ pari from paradicsom ‘tomato’ kapi from káposzta ‘cabbage’ nari from narancs ‘orange’ csoki from csokoládé ‘chocolate’ hambi from hamburger pali from palacsinta ‘pancake’ ajcsi from ajándék ‘present’ telcsi from televízió ‘TV set’ limcsi from limonádé ‘lemonade’ tünti from tüntetés ‘demonstration’ bünti from büntetés ‘fine, penalty’ b. teló from telefon ‘telephone’ szenyó from szendvics ‘sandwich’ brinyó from bringa ‘bike’ vinyó from vincseszter ‘hard disk drive’ (after IBM 3340 hard disk nicknamed Winchester after Winchester 30-30 rifle, as the disk was planned to have two 30MB spindles)

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

Hypocoristic reduplications & embellished clippings in Hungarian

23

c. szenya from szendvics ‘sandwich’ d. pálesz from pálinka ‘brandy’ pöresz from pörkölt ‘stew’ kolesz from kollégium ‘dormitory’ alkesz from alkoholista ‘alcoholic’ e. pelus from pelenka ‘nappy, diaper’

These clippings cannot be used without embellishing suffixes. They are found in compound words with embellishing suffixes, regardless of whether the truncation is the first or the second constituent of the compound: parilé, from pari ‘tomato’ + lé ‘juice’, or paradicsomsali, from paradicsom ‘tomato’ + sali (from saláta ‘salad’). Both constituents can be truncated at the same time, so that in addition to paradicsomsali, there is also parisali. Two embellished truncations are also found in adjacent positions in some conventionalized collocations or terms with the structure Adj + N, the first element can also be truncated in this fashion, for example kovászos uborka ‘pickled cucumber’ – kovi ubi. Some of these embellished clippings are not necessarily brand new: for example, the word cuki from cukor ‘sugar, candy’ in the meaning of ‘patisserie’ seems to have first been used in 1898, and in the meaning of ‘sweetie’ in 1933 (TESz: see Benkő 1967–1976), csoki from csokoládé ‘chocolate’ appears to be used first in 1932 (TESz.), but their number has been increased and they have become more frequently used. 2.3

Hypocoristics

Embellished clippings (the term was introduced by Bauer & Huddleston 2002: 1636) such as those discussed in §2.2.3 are often considered to be hypocoristics. According to Simpson (2008: 643), hypocoristics are ‘alternative forms of words or names which share part of the same form, have the same denotation, but have different connotations and different levels of formality.’ Strictly speaking, hypocoristics are terms used in linguistics for pet names (e.g. Harry for Harold), ‘usually for people but also occasionally for places’ (Bauer et al. 2013: 402). Embellished clippings and hypocoristics are not coextensive phenomena, as shown by the fact that hypocoristics are apparently more restricted in their ability to combine the suffix with certain bases: . . . suffixed hypocoristics do not allow non-homorganic second consonants in nasalinitial clusters, while clippings do. Derivatives exhibiting these two regular patterns are, for example, Winny ♦ *Winfy ♦ Winfred versus confy ♦ confidential. (Bauer et al. 2013: 402)

We could actually claim that the two intersect, as suffixed hypocoristics may be seen as a subset of clippings, while not all embellished clippings necessarily count as hypocoristics or terms of endearment. In this chapter we understand

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

24

Brdar, Brdar-Szabó, F. Gulyás, Horváth

the term hypocoristic as a term of endearment, close to Simpson’s definition above, but restrict it along the lines of Bauer et al. (2013) in applying it to variants of proper names. 2.3.1

Hypocoristic truncations combined with embellishing suffixation in Hungarian

Word formation in Hungarian is principally based on suffixation while truncation, which Ladányi (2007: 63) claims is a non-grammatical tool, is more marginal. However, truncated forms of nouns (and other nominal categories) are productively used with suffixation (Ladányi 2017: 515) especially with diminutive derivational suffixes to denote the speaker’s attitude towards the person or topic in question. As we have already shown in §2.2.3, there is a wide variety of diminutive suffixes that can be attached with truncated stems. Hypocoristic truncations with embellishing suffixation were productively used in earlier stages of the Hungarian language as well. The combined suffix -kó attached to the clipped stem is already attested in Early Old Hungarian, for example, Stephco (1251), Janko (1277) (Szegfű 1991: 213). The diminutive suffix -csa/-cse is presumably of Proto-Hungarian origin (Sárosi 2003: 143, Szegfű 1991: 215) and is attested in Old Hungarian, for instance tocza-k-ra (puddle-pl-subl) (1519) (TESz., s.v. tó; see Benkő 1967–1976), modern Hungarian tócsa ‘puddle’ < tó ‘lake’ + -csa, Zarchazek (1322), modern Hungarian Szárcsaszék (geographic name). In the Proto-Hungarian period the suffix is believed to have been attached to common nouns. For example, the Early Old Hungarian Scymche (1208–1235) is from the common noun scym (modern Hungarian szem ‘eye’). In the case of loanword names in Old Hungarian however, it was attached mainly to their truncated stems: Bencha (1245), Ancha (1275). There seems to be hardly any cases in the historical texts where it was attached to the whole stem in this period: one of these cases is from Iwanche from 1238 (Szegfű 1991: 243, 252). It thus behaved in a way similar to contemporary usage, namely it is attached to a clipped stem of a proper name often showing ablaut: Marcsa from Mária, or Julcsa from Júlia. However, ablaut is not obligatory, for example, Borcsa from Borbála. Not only proper names but common nouns and adjectives can also be the base for the same word formation, though these forms are unproductive in contemporary Hungarian but were previously used in some dialects: egércse ‘small mouse’ (< egér ‘mouse’), aprócsa ‘very small’ (< apró ‘small’) (Zsilinszky 2003: 183). A similar suffix -csi also co-occurs with truncated names and often results in ablaut: Karcsi from Károly, Julcsi from Júlia, but note also Ilcsi from Ilona. Hungarian diminutive suffixes used with truncated stems show variation according to their productivity. Following Ladányi (2007, 2017), we assume

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

Hypocoristic reduplications & embellished clippings in Hungarian

25

an item productive if it is transparent and frequently used. The diminutive suffix -cska/-cske is considered a basic schema since there are fewer restrictions on its usage than the similar diminutive ‑ka/‑ke has. While the latter ones cannot be attached to stems ending in -a or -e there are no such limitations in the case of -cska/-cske (Ladányi 2017: 550). Other diminutives suffixes cooccurring with truncation often show hypocoristic usage and they can usually be combined as in the case of -i, -csi, -ó, -u, -us and -ka/-ke, for example: Nórika from Nóra, Terike from Teréz, Karcsika from Károly, Melcsike from Melinda, Margóka from Margit, Zituka from Zita, Bencuska from Bence. Although these suffixes show different level of productivity in present-day Hungarian, their common feature is that they can all express the speaker’s positive stance towards the theme as is the case with ancika (< anya ‘mother’), kutyika (< kutya ‘dog’), édike () -i/-y denominal (diminutive) suffix, which is of ProtoFinno-Ugric (henceforth PFU) origin (Csúcs 2005: 291) and is considered to be rare and opaque suffix in the contemporary Udmurt (Alasheyeva 2011: 59), at least in cases other than hypocorisms. The same word formation process can also be observed in Komi-Permyak where proper names borrowed from Russian can get the suffix -i as in the case of the name Öndi which derived from Russian Andrey. Additionally, the suffix -u and -ka are widely used in the same manner, for example Vańu < Vańa, Śińka < Ökśiń < Russian Oksana (Batalova 2002: 48, Ponomareva 2010: 232). A rare example for truncation and suffixation is the word nyocka ‘little girl’ from nyv ‘girl’ (Batalova 2002: 48). Clipping is not characteristic for Meadow Mari either: hypocoristic suffixes ‑U, ‑Uš, ‑Uk can be added to truncated proper names: Liduš, Liduk < Lida; Vańu, Vańuk < Vańa; Oľu < Oľa. The suffix -Aš, -kdaj, -βaj, -kaj can be used with kinship terms: ergaš < ergе ‘son’, ergə̑ βaj < erge ‘son’ (Uchaev 1982: 92), but they do not seem to be used with other common nouns. In Finnish, clipping without any suffix attached to the truncated word is not as typical as the formation of clipping with a suffix attached to it, however, examples for back clipping (ale < alennus ‘discount’, ope < opettaja ‘teacher’) or syllable clipping (mopo ‘moped’ < moottori ‘motor’ + polkupyörä ‘bicycle’) can be quite easily found (Pitkänen-Heikkilä 2016: 3225). Clipping with suffixation is often used, and these type of formations are extremely productive, especially in slang: the source of the clipping with suffixes -ari/-äri and -is (: -ikse-) is often (but not always) a compound and the derivative has three (cases with -ari/-äri, as in Example 25a) syllables, while in

https://doi.org/10.1017/9781108989084.003 Published online by Cambridge University Press

34

Brdar, Brdar-Szabó, F. Gulyás, Horváth

the case of -is (: -ikse-) (< Swedish), two syllables (25b) (Hakulinen et al. 2008: §216–17, Pitkänen-Heikkilä 2016: 3225): (25)

a. väikkäri < väitöskirja ‘dissertation’, sydäri < sydänkohtaus ‘heart attack’, kommari < kommunisti ‘communist’, sytkäri < sytytin ‘lighter’, Hesari < Helsingin Sanomat (newspaper) (Hakulinen et al. 2008: §216) b. koris < koripallo ‘basketball’, korvis < korvakoru ‘earring’, mahis < mahdollisuus ‘chance’ (Hakulinen et al. 2008: §217)

Other hypocoristic suffixes are, for instance, the following: ‑(k)kA: enkka < ennätys ‘record’, -ppA: Timppa < Timo, remppa < remontti ‘renovation’, ‑(k) ki: veski < vessa ‘toilet’, ‑(t)si: partsi < parveke ‘balcony’, ‑(t)ski: jätski < jäätelö ‘ice cream’, -de: jäde < jäätelö ‘ice cream’, -e: Ile < Ilkka, -kkU-/-U: enkku < englanti ‘English as a subject in school’ (Hakulinen et al. 2008: §218–20). The expressive -(t)si(k)ka element can be interpreted as a combined suffix from the suffixes -(t)si and -(k)ka (see above) (Hakulinen et al. 2008: §220), for example, tietsikka from tietokone ‘computer’. 2.4.1.4 Suffixation In Udmurt literary language, the aforementioned suffix -i of PFU origin can be used in cases of the clipping of personal names (and very rarely, of kinship names) in hypocorisms, and usually, it cannot be used with common nouns or with non-truncated words. There is, however, another diminutive suffix (-kaj, borrowed from Tatar) used in the Southern and Peripheral-Southern Udmurt dialects that can be added to non-clipped words: for example, lulykaj (< lul ‘soul’+ 1sg) ‘my dear’ (Keľmakov 2006: 128). Again in dialects, complex comparative suffixes (formed from two different comparative suffixes: -gezgem < gez + gem, -gezʒ́yk < gez + ʒ́yk, -gemʒ́yk < gem + ʒ́yk) can express intensification of the comparative meaning of the comparative adjective: Central Udmurt vekcigesgem (< vekci [‘short’] -gesgem [comp]) ‘even shorter, significantly shorter’. There is also a dialectal suffix -ʒ́em that expresses the speaker’s negative stance so the derived word form has a pejorative meaning in the case of personal names: Koľa-ʒ́em ‘wicked Koľa’ (Keľmakov & Saarinen 1994: 109–10). The Komi-Permyak language has a large variety of diminutive suffixes. Some of them like -i and -u are used in the formation of proper names while others are used in different types of nominal derivations. Due to the long and extensive contacts with the Russian language, many of the Komi-Permyak diminutive suffixes seem to be borrowings, for example nylocka ‘little girl’ ( fke-fke / fkasne-fkasne (M) ‘each other’: the reduplicated numeral shares the same grammatical case or the same suffix; Erzya seems to have more examples of this type. ○ kavto (E) / kafta (M) ‘two’ kavtoń-kavtoń (E) / kaftoń-kaftoń (E) / kaftèńkaftèń (M) ‘doubly’ < (Paasonen 1990). ○ kemeń (E) / keməń (M) > kemoneń-kemoneń (M) ‘by tens’ (Tsygankin 1980: 133, Erelt & Punttila 1999: 10). A similar distributive value applies to nouns too: ijeń-ijeń (E) ‘annually, yearly’ (Feoktistov 1974: 132, Erelt & Punttila 1999: 10) < ije (E) / ij (M) ‘year’. Reduplication is also used to form distributive adverbs: pingoń-pingoń (M) ‘at times, occasionally’ < pinge (E, M) ‘time, period, century’, kosto-kosto (E) ‘sometimes, occasionally’ < kosto ‘from where, when’, škań-škań (E) ‘periodic, sporadic’, ‘sometimes, occasionally’ < ška ‘time’ (Paasonen 1990).  Superlative: repetition of the same adjectives or adverbs, for example, mazi-mazi (E, M) ‘the most beautiful’, paro-paro (E) ‘the best’ (Bartens 1999: 109). Also with additional elements: the first adjective in Ablative + enclitic -jak + adjective, which can be translated literally in English as: adjective + -er than adjective: parodojak paro (E) ‘the best’ (Bartens 1999: 109) < paro ‘good’, pokštojak pokš ‘the biggest’ < pokš (E, M) ‘big’ (Zaicz 1998: 196) , tjužadojak tjuža (E) ‘yellowest’ (Erelt & Punttila 1999: 4) < tjuža (E) ‘yellow’. Zaicz also mentions a kind of excessive superlative formed by syntactic threefold repetition linked by the conjunction di ‘and’: pokš, pokš di pokš ‘the very biggest’ (Zaicz 1998: 196).  Words of onomatopoeic origin, for example: kuku (E, M) ‘cuckoo’ (Paasonen 1990), gaga (M) ‘goose’ (Paasonen 1990). 3.3.2

Partial reduplication

Partial reduplication is rarer, and is mainly used for:  Distributive expressions: kaviń-kafta (E) / kauń-kafta (E) / kaftèń-kafta (M) ‘doubly’ < kavto (E) / kafta (M) ‘two’ (Paasonen 1990); stakanoś-stakan

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

66

Iwona Piechnik

(E) ‘glass by glass’ (Feoktistov 1974: 132, Erelt & Punttila 1999: 10) < stakan ‘glass’ (< Russ.).  Intensity of an action (e.g. speed), as Erelt and Punttila (1999) indicate it in repetition of verbal forms: lakaź lakaj (M) ‘boiling it boils’ (Erelt & Punttila 1999: 5) < lakams (E, M) ‘to boil, be boiling’, livtjaź livtjaś (E) lit. ‘leaving (s/he) left’ = (s/he) left very quickly’ < ľivtems (E) ‘to leave’, laśkoź laśkoś (M) lit. ‘running (s/he) ran’ = ‘(s’he) ran very fast’ (Feoktistov 1974: 130, Erelt & Punttila 1999: 7) < laśkəms (M) ‘to run’. Or in repetition of nouns: ideń id ‘grandson’ (lit. ‘child’s child’ < ejd (E) / id (M) ‘child’), pingeń pinks (E, M) ‘forever’ (Paasonen 1990) < pinge ‘time’ (in Genitive + Illative).  Ideophones or descriptive words: lij-laj (E) ‘fluttering’ (Paasonen 1990), lykom-lakom M ‘trembling and waving/swinging’ (Paasonen 1990). 3.4

Mari

The Mari language, formerly known also as Cheremiss, is spoken in the Mari Republic, and has two main varieties: Hill Mari (Western), and Meadow Mari (Eastern). Most examples in this section come from the excellent online Mari–English Dictionary (Bradley et al. 2014), created at the University of Vienna. This dictionary gives the Meadow Mari version. 3.4.1

Total reduplication

Complete repetition of a word is a very common occurrence in Mari, mainly to express intensity and distributivity.  Intensity by: ○ redoubling of adjectives, adverbs, or nouns,13 for example: kandekande ‘very blue’ < kande ‘blue’, ošyn-ošyn ‘very white’ < ošyn ‘white’ (adv.) < oš(o) ‘white’ (adj.), šeme-šeme ‘pitch-black’ (adj.) & šemynšemyn ‘pitch-black’ (adv.) < šemyn ‘black’ (adv.) < šeme ‘black’ (adj.), kužun-kužun ‘for a long time’ < kužun ‘long time, for a long time, for long; a long distance, far, long’, šuko-šuko ‘very much’ < šuko ‘a lot’, šöryn-šöryn ‘very crookedly, quite aslant’ < šöryn ‘slantwise, aslant, at an angle, obliquely; sideways, half-turned; at a slant, not straight, leaning, crooked, crookedly; askew, to one side; on one’s side; askance, sideways (glance); (fig.) in an unfriendly manner, suspiciously, with suspicion, with distrust; slanting’; šygyryn-šygyryn ‘overcrowded, very crowded,

13

Many words can function as various parts of speech.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

Reduplication in Finno-Ugric languages

67

jostling’ < šygyryn ‘crowded, cramped; (fig.) bad, difficult, uneasy, dreary’, toka-toka ‘barely, with difficulty’ < toka ‘recently; barely, just, just barely, with difficulty’, and so on. ○ onomatopoeic and descriptive words, which are very numerous in Mari, and they can serve as nouns, adjectives, adverbs, or simply onomatopoeias, e.g. kyšt-kyšt ‘quiet movement’, loč-loč ‘hit, noise’, ľop-ľop ‘sloshing of water, mud’, pydyr-pydyr ‘small and frequent’, pyz-pyz ‘rolling felt’, pyj-pyj ‘wood cracking’, žyž-žyž-ž ‘buzzing insects’, mömö ‘monster, fright, bogeyman’.  Distributivity by repetition of: ○ pronouns: kušto-kušto ‘where, which places, what places (no movement)’ < kušto ‘where’, mo-mo ‘what things, what’ < mo ‘what’, kö-kö ‘who exactly’ < kö ‘who’; ○ nouns or adverbs: ala-ala ‘hardly, doubtful, who knows’ < ala ‘maybe, probably, could be; hardly, doubtful, who knows; emphatic particle; sometimes . . . sometimes, either . . . or, now . . . now; or; about, roughly’, almaš-almaš ‘in turns, in shifts, by turns, alternately’ < almaš ‘replacements, successors; replacement, substitute; in turns, in shifts, by turns, alternately; shift change, change of shift; in turns, in shifts, by turns, alternately; duty, on duty’, ašyn-ašyn ‘at times, now and then, now and again, from time to time’ < ašyn ‘successfully’ < aš ‘power; success, benefit’, osyn-osyn ‘in fourths, in quarters’ < oso ‘fourth (part), quarter (of a pancake); piece’, poče-poče ‘one after another, in a line, in a file’ < poče ‘one after another, in a line, in a file; in succession, in a row’, šujen-šujen ‘in a few places, here and there; seldom, rarely, now and then, from time to time’ (Maitinskaja 1964: 125, Erelt & Punttila 1999: 6, Bradley et al. 2014) < šujen ‘sparsely, far apart; rarely, seldom, infrequently; slowly’, tapyryntapyryn ‘at times, from time to time, sometimes’ < tapyryn ‘rarely, seldom’ < tapyr ‘moment, instant, time’, veran-veran ‘in places, here and there’ < ver place’, and so on. ○ numerals: iktyn-iktyn ‘one by one, one at a time, separately, individually’ (Maitinskaja 1964: 130, Erelt & Punttila 1999: 10) < iktyn ‘alone’ < ikte ‘one’, koktyn-koktyn ‘in twos, by pairs, in pairs, two by two’ < koktyn ‘as a pair, as two, two of them’ < kokyt ‘two’. ○ verbs: muren-muren ‘singing’ < muraš ‘to sing’; an interesting word is a deverbal reduplicative form maneš-maneš ‘gossip; a type of folk-tales’ < maneš ‘it is said’ (lit. ‘s/he) says’ – 3rd pers.sing). However, when a word is polysemic, the meaning of reduplicated word is polysemic too, for example, distributivity and intensity at a clip: tugaj-tugaj ‘such-and such; so, so very’ < tugaj ‘such, that kind, like this; so, very’.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

68

Iwona Piechnik

3.4.2

Total reduplication with additional elements

Total repetition with additional elements occurs less frequently. This phenomenon may consist in slipping a preposition between two doubled words or adding an ending to one of them.  Superlative of adjectives in constructions with preposition deč ‘of, from’: saj deč saj ‘the best of the best’ (saj ‘good’), ošo deč ošo ‘the whitest of the white’ (ošo ‘white’), šeme deč šeme ‘the blackest of the black’ (šeme ‘black);  Repetition of the same stem, but with different endings: ○ adjectives or nouns: viklan-vik ‘directly, straight’ (Maitinskaja 1964: 132, Erelt & Punttila 1999: 10, Bradley et al. 2014) < vik ‘immediately, straight, directly’ + -lan ‘Dative case’, veryn-vere ‘in places, here and there’ < ver place’, lümeš-lüm ‘proper name’ < lümeš ‘in the name of’ + lüm ‘name’; ○ deverbal adverbs: temen-temyde ‘halfway up’ < temaš ‘to fill (up); to cover; to satiate; to reach; to fulfill’, tüknen-tüknyde ‘almost touching, barely touching, touching from time to time, grazing against’ < tüknen ‘stammering, stuttering, with a stammer; (fig.) blindly, at random, by guesswork, aimlessly’ + tüknyde ‘without stammering, without a stutter, smoothly (talking, reading)’ < tüknaš ‘to touch, to hit; to run into, (fig.) to stammer’, etc.  Intensity of an action by repetition of verbal forms, e.g. peltenak-pelta ‘burning burns, baking bakes (e.g. in the sun)’ (Maitinskaja 1964: 131, Erelt & Punttila 1999: 7) < pelaš ‘to burn, to be burnt’. 3.4.3

Partial reduplication

Descriptive words in partial reduplication are probably more numerous than those in total reduplication. The most numerous group is that of echo-pairs, which can express various meanings:  Intensification: šöryn-vuryn ‘from side to side’ < šöryn ‘slantwise, aslant, at an angle, obliquely; sideways, half-turned; at a slant, not straight, leaning, crooked, crookedly; askew, to one side; on one’s side; askance’, šöryn-türyn ‘swinging from one side to the other, reeling’ < šöryn + türyn ‘slantwise, askew, at an angle, sloping; drinking directly, from a bowl, from a vessel; sideways, facing no one, with one’s back turned’;  Indefiniteness: aki-tuki ‘somehow, any old way, carelessly’, aldi-buldi ‘unstable, unsteady, inconsistent, inconsequent; stupidly, inappropriately, unsuitably’, alik-pylik ‘disreputable, frivolous, unreliable’, kadyr-mugyr ‘curvy, winding, twisting, meandering, serpentine’ < kadyr ‘crooked,

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

Reduplication in Finno-Ugric languages

69

curved’ + mugyr ‘hump’, kyrča-marča ‘trifles, small things; petty, trifling, small’, lule-lole ‘wind-fallen trees’, pyldi-poldi ‘muddle-headed, lightheaded, thoughtless’, šyldyr-goldyr ‘knocking, rumbling’, and so on.  Distributivity: iktyn-koktyn ‘some alone and some in pairs’ < iktyn ‘alone’ < ikte ‘one’ + koktyn ‘as a pair, as two, two of them’ < kokyt ‘two’, tidattudat / tide-tudo ‘this and that, one thing and another, something’ < tide ‘this’ + tudo ‘that, it’, tembak-umbak ‘back and forth, hither and thither, to and fro’ < tembak ‘closer, toward the speaker (movement towards)’ + umbak ‘far, further, into the distance’  more rarely others that mainly express indefiniteness: ažme-tužmo ‘giddiness, dizziness (caused by alcohol), drunkenness, intoxication’, myge-mugo ‘vaguely, quietly, indistinctly, tyšte-tušto ‘here and there’. In these examples, the relation of the ending vowels is e:o. Sometimes the same meaning is expressed both by total and partial reduplication, e.g. jyvyž-jyvyž / juvyž-jovyž / jyvyž-juvyž ‘rapid movement’. A kind of partial reduplication is due to the Old Turkic influence, as Asztalos and other authors explain it: “adjectives beginning with CVC-structure can be reduplicated by doubling the CVC and by the substitution of the second C with p” (Asztalos et al. 2021: 267), for example, tap-taza ‘completely healthy, fit as a fiddle’ < taza ‘healthy’,14 sap-sare ‘completely yellow’ < sare ‘yellow’, šepšeme ‘pitch-black’ < šeme ‘black’, čup-čuvar ‘very mottled, very colourful’ < čuvar ‘variegated, multicoloured, colourful, speckled, piebald’, čap-čara ‘completely bare, completely naked, stark naked, buck naked; completely empty; indigent, poor’ < čara ‘bare, naked; bald; empty, destitute’. Such reduplicated adjectives can also form adverbs, for example, čap-čaran ‘completely bare, completely naked, stark naked, buck naked’. This reduplication undoubtedly serves to intensify the meaning of adjectives and adverbs. However in the case of the following examples, maybe we are dealing with additional phenomena: lep-leve ‘very warm’ < leve ‘warm’, with its corresponding adverb: lep-levyn ‘very warmly’ < levyn ‘warmly’ – but the reduplicant can also form a reduplication independently: lep-lep ‘warm’15; šyp-šyma ‘shy, bashful, modest’ < šyma ‘gentle, tender, affectionate, kind; soft, pleasant, light’ – but there is also an adjective šyp ‘quiet, calm’ (which has its reduplicative form too: šyp-šyp ‘very quiet’), so the lexeme šyp can be either a reduplicant or an adjective that adds its meaning to the whole word šyp-šyma. A similar case is in two other words: typ-tymyk ‘deep silence, complete silence,

14

15

It should also be noted that a similar adverb tüp ‘completely, entirely, totally’ exists too and it can be used to say tüp taza (written without a hyphen) ‘completely healthy, fit as a fiddle’, where it is not any reduplicant. However, the word lep alone means something else entirely: ‘spleen’ or ‘a horse disease’.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

70

Iwona Piechnik

perfect silence; silent, completely silent’ < tymyk ‘quiet, silence, stillness; peace, tranquillity, serenity; quiet, silent, noiseless’, typ-tynys ‘complete silence; completely silent’ < tynys ‘peace; quiet, tranquillity, calm; quiet, calm, serene, peaceful’. However typ alone also means ‘quiet, silent, calm; quietly, silently, in silence, calmly’, so it contributes to the enhancement of the meaning.16 Finally, we can say, that syntactic reduplication occurs in Mari too. It can be found mainly in verbs to express an iterative, durative or progressive aspect in many Mari tales. For example, in the description of protagonists’ long marching: kaja, kaja ‘(s)he/they walk(s), walk(s)’ (3rd pers. sing./plur.). 3.5

Permic

The Permic languages include Komi and Udmurt, which have regional varieties and are mainly spoken in the Komi Republic and Udmurtia in the Russian Federation. As far as reduplicative phenomena are concerned, the sound iconicity in Permic languages is very high, with many ideophones and descriptive words. 3.5.1

Komi

The Komi language includes two varieties: Komi-Zyrian and Komi-Permyak. Both are pluricentric, with many regional subvarieties, so they are difficult to describe in detail. For present purposes I have decided to limit my description to the Great Komi–Russian Dictionary (BKR 2020) as a source of examples from the Komi-Zyrian language, with some references to other sources.

3.5.1.1 Total reduplication A doubled word (noun, adverb, adjective) undergoes a process of lexicalization and takes on a new meaning, which is used to express:  Distributivity, for example: lun-lun ‘from day to day’ < lun ‘day’, korkökorkö ‘occasionally’ < korkö ‘sometime, once, someday’, paraön-paraön ‘in pairs’ < para ‘couple, pair’ (< Rus. para), radön-radön ‘interrow, by rows’ < rad ‘row’ (< Rus. rjad), vježonön-vježonön ‘weekly’ < vježon ‘week’, ötkön-ötkön ‘one by one, one at a time’ < ötkön ‘lonely, by oneself’ < öti ‘one’, kykön-kykön ‘in pairs’ < kykön ‘together, in two’ < kyk ‘two’;

16

It should be noted that šyp and typ are similar in their form and meaning, however šyp can intensify its meaning in reduplication šyp-šyp, while typ does not have that possibility (because typ-typ is a different descriptive word meaning ‘tapping, patting, rapping, footsteps’).

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

Reduplication in Finno-Ugric languages

71

 Indefiniteness, for example: könkö-könkö ‘somewhere, in places’ < könkö ‘somewhere, anywhere’;  Intensity, for example: ylö-ylö ‘far far away, very far’ < ylö ‘far’, kojtovkojtov ‘staggering from exhaustion’ < kojtavny ‘to stagger, can hardly stand on his feet’, and so on. This type usually forms adverbs, but can also form absolute superlative of adjectives, for example: ydžyd-ydžyd ‘very big, very large, huge’ < ydžyd ‘big, large’ (see Majtinskaja 1964: 125, Bartens 2000: 138). Ideophones and onomatopoeias can be doubled to express indefiniteness or to imitate sounds: kań-kań ‘inaudibly, silently, imperceptibly’, švac-švac ‘clap-clap’.

3.5.1.2 Total reduplication with additional elements Total reduplication with an additional element is rather rare in Komi. It is seen in some ideophones e.g.: cuk-cuki ‘clippety-clop’, veža-vež ‘rapidly alternating, intermittent; interspersed; disorderly’, or a numeral: kyka-kyk ‘both’ < kyk ‘two’. The ending -a is used to form adjectives. However, with the addition of the ending of the elative case -yś (‘from, out of’), the repetition is also used to intensify the meaning of adjectives and adverbs, even to create absolute superlative, for example: bur ‘good’ > buryśbur ‘very good’, jon ‘strong’ > jonyś-jon ‘very strong, mighty’, gyryś ‘large, big’ > gyryśyś-gyryś ‘very large, very big’, kuź ‘long; tall’ > kuźyś-kuź ‘very long; very tall’, kypyd ‘cheerful, joyful’ > kypydyś-kypyd ‘very cheerful, very joyful’, miśtöm ‘ugly’ > miśtömyś-miśtöm ‘hideous’, tölka ‘smart, clever’ > tölkayś-tölka ‘very smart; the cleverest’, śökyd ‘heavy’ > śökydyś-śökyd ‘very heavy’, adverb: śökydyś-śökyda ‘very hard, very heavily’, and so on. Furthermore, there is also the possibility of adding the possessive suffix of the 3rd pers. sing. to the elative case, the form of which becomes -śys: kokńi ‘light, easy, not difficult’ > kokńiyś-kokńi / kokńiśys-kokńi ‘very light, very simple’, ydžyd ‘big, large’ > ydžydyś-ydžyd / ydžydśys-ydžyd ‘huge, enormous’, and so on. Such a possessive form can be literally translated as ‘adjective of its adjective’.

3.5.1.3 Partial reduplication Partial reduplication is more common than total reduplication and includes onomatopoeias, ideophones, and other parts of speech. If reduplicative pairs differ in vowels, the first vowel is usually u, while the second is often a, o, or e.  Onomatopoeias: these can be nouns, adjectives, or adverbs (sometimes in the same form), for example: bur-bor ‘murmur, whisper’, kun-kon ‘resonant

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

72

Iwona Piechnik

like empty (adj.); resonantly, resoundingly’, kurc-karc / kuž-kaž / ruc-rac ‘crunch, crackle’, cyž-caž ‘with a bang, with a crack’, švica-švaca ‘whistling; with noise; clapping’, švuc-švac ‘clap-clap, slap-slap’, and so on.  Ideophones: these most often differ in vowels and express various meanings, for example: kuń-kań / kuńki-kańki ‘softly’, bryz-braz ‘to throw oneself (e.g. to water)’, vudžik-vedžik ‘in a waddling manner’, liza-löza ‘with noise’, ljuzi-ljozi / ljuź-ljaź ‘shaggy, dishevelled’, ljuvi-ljavi / ljuvja-ljavja ‘uneven, unevenly hanging’, and so on. But they can also sometimes differ in consonants: kundy-mundy ‘junk things, stuff’, kölyj-mölyj ‘things, stuff’ (BKR 2020), < kölyj ‘stuff’, kyjny-lyjny ‘to earn a livelihood (by) hunting’, kyjyś-lyjyś ‘hunter’, and so on. Or they can be echo-pairs, which differ in the first syllable: śuśsa-vośsa ‘quick-witted, sharp-witted, smart’.  Other parts of speech (often nouns and adverbs), which differ by one or two first syllables: cukyr-makyr ‘wrinkle’, cukyr-pakyr ‘wrinkled, crumpled’ (< cukyr ‘wrinkle, fold’). Echo-pairs may result from the composition of serial words, that is, two similar words with the same suffix: gymalöm-cardalöm ‘thunderstorm; disaster’ (BKR 2020) < gym ‘thunder, storm; bam!, clap!’) ~ gymalöm ‘thunder, thunderclap’ + card ‘lightning’, mamtöm-baťtöm / baťtöm-mamtöm ‘without mother and father’ < mam ‘mother’ + bať ‘father’ + suffix -töm, gyžja-vježja ‘cattle, hoofed animals’ < gyž ‘nail, claw, hoof’, ulyna-vylyna ‘at different heights, at different levels; uneven’ < ulö-vylö ‘up down’ < ulö ‘under; to’ + vylö ‘up; high’, śojöd-juöd / śojan-juan ‘food, meals’ < śojny-juny / śojyštnyjuyštny ‘to feed off, to have meals’ < śojny/śojyštny ‘to eat’ + juny/juyštny ‘to drink’, and so on. Two verbal forms can be repeated to express intensity: śojömön-śojö ‘eating eat’ = ‘eat up, annoy’ (see Majtinskaja 1964: 131) < śojny ‘to eat’. Many reduplicated pairs may serve as prefixes to create compound verbs. A lot of such preverbs are onomatopoeic or simply ideophonic. They have a kind of adverbial function of specifying or intensifying the meaning of the verbs such as: munny ‘to go’, kyvny ‘to hear; to feel’, kerny ‘to do; to process’, vöcny ‘to do, to execute, to make, to work out’ or vartny ‘to thresh; to hit’, which having largely lost their original meaning, usually serve to form compound verbs (see Fejes 2004). On rare occasions these verbs may have fully repeated prefixes, for example, gerc ‘quack’ > gerc-gerckerny / gercgercmunny ‘to quack one or twice’, gyž ‘nail, claw, hoof’ > gyž-gyžkjerny ‘to scratch’. Usually, Komi morphology prefers partial reduplication, which differs in vowels in the first syllable, for example: puvk-pevkvöcny ‘to ruffle’ (figurative), brutka-brotkakyvny ‘to grumble, to grouse’, and so on. We can find original elements for many reduplicated pairs, which also exist as independent onomatopoeias, ideophones or nouns: gudyr ‘sediment, sledge’

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

Reduplication in Finno-Ugric languages

73

> gudyr-gadyr ‘sediment, sledge, mud’ (BKR 2020) > gudyr-gadyrkerny / gudyr-gadyrvöcny ‘to blend hastily, to mingle hastily’, gudyr-gadyrmunny ‘to get mixed up’, raz ‘once’ > ruz-raz (also ruz-paz) ‘scattered; in disarray; in pieces’ > ruz-razkerny ‘to scatter; to shake’, ruz-razmunny ‘to collapse; to crackle’, vježyń ‘skew; curvature; ‘crooked, oblique, twisted’ > vužyń-vježyń ‘oblique, skewed, uneven; unevenly; awry; at random’ (BKR 2020). It can also form a compound verb: vužyń-vježyńmunny ‘to warp, to twist’. The source of verbs can also be independent reduplicated pairs: ľuv-ľav ‘woof, arf’ > ľuv-ľavmunny ‘to be heard from afar, to be heard in the distance (voices)’, ľuva-ľavakyvny ‘to be heard (noise, voices)’, puś-paś ‘to pieces, to smithereens’ > puś-paśmunny ‘to go to pieces, to get broken’, or puśpaśvartny / puś-paśkerny / puś-paśvöcny ‘to smash to smithereens, to crush’ (BKR 2020), ruč-rač ‘crack’ > ruč-račmunny ‘to crackle one time’,17 śuv-śav / śuvk-śavk ‘in a scattered manner, dispersedly’ > śuv-śavmunny / śuvkśavkmunny ‘to scatter, to shunt’, šaj-paj ‘in confusion’ > šaj-pajvajödny ‘to astonish, to astound’, šaj-pajmunny ‘to be confused, to be bewildered’, vuš-vaš ‘in a whisper’ > vuš-vaškerny ‘to whisper’; vuš-vašmunny ‘to whisper something to each other’; also in another variant: vuški-vaški ‘in a whisper’, but in preverbal position there is a vocalic change: vuška-vaškakyvny ‘to whisper’, vuz-vaz ‘crunching’ (BKR 2020) > vuz-vazkerny ‘to swallow biting with a crunch; to crush with a crunch’, vu-vazmunny ‘to crunch, to crackle’, ľug-ľeg / ľugi-ľegi / ľigi-ľegi ‘shaky, fragile’ > ľug-ľegvӧcny / ľug-ľegkerny ‘to wag; to shake’, śur-śar ‘bang-bang’ > śur-śarkerny / śur-śarvöcny ‘to make a crack’, śur-śarmunny ‘to go off (a crack)’, śura-śarakyvny ‘to make a knock’. Sometimes those preverbs have -a or -k(a) endings, but do not have a distinctive meaning, for example: pur-par / purk-park ‘quickly, hastily’ > purk-parkkerny / purk-parkvartny / purk-parkvӧcny ‘to get the job done quickly and efficiently’, pura-parakyvny / purk-parkakyvny ‘to flutter’, vurvar ‘quickly, in one fell swoop’ > vur-varkerny ‘to eat quickly, with a crunch; to crush something with a crunch’, vur-varmunny ‘to be crushed with a crack’, also in another variant: vurk-varkkerny ‘to crush with a crack’, vurk-varkmunny ‘to be crushed with a crack’, vura-varakyvny ‘to grumble’, vurkavarkakyvny ‘to crackle (when broken)’, buľ-boľ / buľs-boľs ‘glug-glug, with a gurgle sound’ > buľska-boľskakyvny ‘to gurgle’. It is not only verbs that can be formed with Komi reduplications, but also nouns, though less frequently, and they are derived from verbs. Usually these nouns contain the lexeme kylöm ‘feeling, perception’, for example: bulskabolskakylöm ‘squishing, slapping’ < bulska-bolskakyvny ‘to squish, to slap, to slurp’ < buls-bols ‘slap-slap, squelch-squelch’, gura-garakylöm ‘rumbling, 17

Let’s note that onomatopoeic single rač ‘crack’ exists too and it can form the verb račmunny with the same meaning, but less intensified.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

74

Iwona Piechnik

grumble’ < gura-garakyvny ‘to grumble, to growl (in the stomach)’ < gur-gar ‘with a roar, with noise’. We have only given some examples here of the many that can be found in Komi. As has been seen, this is a very rich language when it comes to various types of reduplication. The majority of reduplicated forms are ideophones, which are abundant in this language and can fulfil the functions of adverbs or verbs without being a clear morphological class. Rozhanskiy, who studied Komi ideophones, labels their behaviour as ‘morphological mimicry’ (Rozhanskiy 2018: 196). A strong tendency to reduplication can also be seen in Komi folklore, for example, in: 18  Folk names: Kirjan Varjan / Kir’jan-Var’jan ‘Komi-Zyrian epic hero’ 19 (KM) < kirjyny, girjyny ‘to untie loops; to tear a thread’ + varjyny, varködny, varödny20 ‘to crush, to squash, to press, to break through’, so this name can mean ‘putting (somebody) to rout’ (KM: 174–175). KokljaMoklja ‘a spirit living in the forest in a hunter’s hut’ can mean ‘a creature having large curved legs’: it is probably a Russian loanword, but in Komi cuklja-muklja / cukyľ-mukyľ ‘squiggle, zigzag; crooked-curved’ (KM: 178–179) < cuklja ‘crooked, curved, with a bend’ / cukyľ ‘bend, turn, bow’ (BKR 2020), tödyś-kužyś / kužyś-tödyś ‘sorcerer, healer’ (KM: 309–310), ‘master, craftsman’ (BKR 2020) < tödny ‘to know’ + kužny ‘to be able’.  Titles of folk songs: Tölöcu-völöcu; Ylyn-ylyn roc Kavkazyn ‘Far, far away in the Russian Caucasus’; Gögör-gögör me vidźödla ‘Around, around I look’; Votyśej daj votyśej ‘Berry picker, berry picker’; Olis-vylis vöryn ‘Once they lived in a forest’. Various parallelisms are also seen in such song titles: Bur baťköd-mamköd olöm ‘Good life with father and mother’; Saldatjasöj, mica vojtyrjasöj ‘My soldiers, my beautiful people’; Suskin sadjyn ‘In a cedar grove’; Śylim, jöktim, gažödcim ‘We sang, we danced, we had fun’, Šondiöj-mamöj ‘My mother, my sun’; Bereg doryn, ju doryn ‘On a shore, by the river’ (Siikala & Ulyashev 2016: 247–9).  Tales, which often start with the words olis-vylis ‘once upon a time there lived . . .’ (sing.) / olisny vylisny ‘once upon a time there lived . . .’ (plur.) (Wichmann 1916: 24, 38, 44, 48, 55, 57, 60, 64, 68, 81, 83, 85, 87, 89, 91, 96, 100, 104, 105, 111, 113, 114, 117, 120, 122, 310, 313), and they end with the words: olöny (da) vylöny ‘they are still living’ (Wichmann 1916: 13, 16, 29, 70, 85, 117) < ovny / övny ‘to live’ + vyvny / vövny ‘to be’; or in

18 19 20

Lower Vychegda dialect. BKR 2020 gives only the verb girjyny ‘to scrape (off ); to wipe; to drag; to undo loops’. BKR 2020 does not have this verb in any form.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

Reduplication in Finno-Ugric languages

75

syntactic repetitions for a long walk of tales’ protagonists: munas munas ‘is going, going’, munas da munas ‘is going and going’, munasny, munasny ‘they are going, going’ (Wichmann 1916: 20, 22, 23, 39, 54, 74, 84, 97) < munny ‘to go’. The abundance of reduplicative phenomena in Komi and their semantic capacity (even their ambiguity) make this language one of the most expressive in the Finno-Ugric family. 3.5.2

Udmurt

Like its Permic cognate, Komi, Udmurt has regional varieties. For present purposes I have decided to limit my description to the Great Udmurt–Russian Dictionary (BUR 2020) as a source of examples from the Udmurt language, with some references from EUD 2015.

3.5.2.1 Total reduplication Total reduplication is more frequent in Udmurt than in Komi. Doubles can be created:  on the basis of existing words: buj-buj ‘soft, fluffy; striped’ < buj ‘colour; stripe’, curin-curin ‘in rows, in parallel’, curjen-curjen ‘in a row, in rows; in stripes’ < cur ‘row’, ńimaz-ńimaz ‘separately’ < ńimaz ‘separate, special; separately, specially’, vajo-vajo ‘multi-branched’ < vajo ‘branched, forked, knotty, twiggy’, dyryn-dyryn < dyryn ‘sometimes; from time to time’ < dyr ‘time’, intyjen-intyjen ‘in places, in some areas’ < intyjen ‘in places’ (EUD 2015, BUR 2020). The latter example in a sentence shows the genuine popularity of this phenomenon in Udmurt: intyjen-intyjen turyn vožyk-vožyk adskyle ‘in places the grass is already turning green’ (vožyk-vožyk is a descriptive word).  ideophones expressing various meanings and having functions of various parts of speech, for example: byk-byk ‘expressing softness, fluffiness’, purpur ‘loose, fluffy; free, wide’, ciľtyr-ciľtyr ‘denoting the play of colours’, cupyr-cupyr ‘lively, briskly’, curt-curt ‘firmly, confidently’, duak-duak ‘impetuously’ (about a violent gusting of the wind), cutyr-cutyr ‘about something curled up, warped’, gyl-gyl ‘cleanly; smoothly’, völ-völ ‘free (ly), comfortable/y, spacious(ly)’. Sometimes doubling serves to intensify a single ideophone, for example, čaž ‘the noise of rain, the crackle of torn matter, the sound of boiling water’ > čažčaž ‘hum, hiss’. A kind of intensity is shown in an absolute superlative by repetition of adjectives or adverbs in the exactly same form: śöd-śöd ‘very black’ < śöd

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

76

Iwona Piechnik

‘black’, töď-töď / töďy-töďy ‘very white, snow-white’ < töď / töďy ‘white’, gord-gord ‘very red, bright red’ < gord ‘red’, čyž-čyž ‘very ruddy; bright red, intense red’ < čyž ‘ruddy, rosy’, ťjom-ťjom ‘gloom, darkness, unfolding darkness’ (EUD 2015), kudiz-kudiz ‘some (plural, e.g. some people)’ < kudiz ‘anybody, anyone’, myrdem-myrdem < myrdem ‘hardly, barely, with difficulty’, sapyr-sapyr / sopyr-sopyr < sapyr ‘abundantly’, bugyr-bugyr < bugyr ‘adverb intensifying the meaning of the verbs to swirl; to seethe’ (EUD 2015). Ideophones can also be repeated in phrases where they complete a verb: gyźgyź vešany ‘to caress wholeheartedly; to pat gently on the back’, gyź-gyź užany ‘to work with desire’, ciľ-ciľ śöd ‘pitch black’ (EUD 2015, BUR 2020) < ciľ-śöd ‘shiny black’, bugyľ-bugyľ karyny ‘to make plump; gurgling, bubbling’, vupuľyos bugyľ-bugyľ džutiśko ‘water is bubbling; gurgling, rising bubbles’. Of course, repetition can also apply to onomatopoeia: cingyľ-cingyľ ‘sound of the bell’, and so on.

3.5.2.2 Total reduplication with additional elements Examples of total reduplication with an additional element include: ciľ ciľany ‘to gleam, to glisten’ < ciľany ‘to shine, to sparkle’ < ciľ is an ideophone reinforcing the meaning of twinkling, myd-mydlań ‘back and forth’ < myd ‘second, other’. However with the addition of the ending of ablative case -leś (‘from’),21 it is also used to intensify the meaning of adjectives and adverbs, even to create absolute superlative: śödleś no śöd ‘very black’ (Majtinskaja 1964: 131), töďyleś no töďy ‘very white, snow-white’ < töďy ‘white’ (Bartens 2000: 138).

3.5.2.3 Partial reduplication Echo-pairs are based mainly on ideophones having very various meanings, for example: cingyr-cangyr ‘thin’, cindyr-candyr / cindyr-vandyr / cińdyr-vańdyr ‘higgledy-piggledy; scrawny, thin’, ńikyr-ńakyr ‘weak, shaky’, ńiž-ńaž / ńižyrńažyr ‘slowly, without haste; lingering, quiet’, šuźi-maźi ‘silly, foolish; neither this nor that’ < šuźi ‘idiot, stupid, crazy’ (EUD 2015), žug-žag ‘weed, rubbish, waste’ < žug ‘leaves of root vegetables’ + žag ‘rubbish, weedy’ (EUD 2015), čyžy-vyžy ‘relative, kinsman’ < čyžy ‘genus; sediment’ + vyžy ‘root; family; offspring, generation’ (EUD 2015, KYV), mygyľ-magyľ ‘slurred, incomprehensible speech’, dubyr-dabyr ‘rumble, roar’, gyl-gal / gylym-galym ‘clumsy; clumsily’, gyľym-gaľym ‘lazily, slowly, carelessly’, cityr-cutyr / cityr-kotyr / cutyr-catyr ‘sinuous, with convolutions, with zigzags’, etc. Sometimes, such echo-pairs only exist in expressions, for example: śin aźyn ucyr-vočyr adske ‘(something) flashes before the eyes’.

21

It works in analogy to Komi elative case -yś.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

Reduplication in Finno-Ugric languages

77

Echo pairs are more rarely based on other words, for example: otyn-tatyn ‘here and there’ < otyn ‘there’, otcy-tatcy ‘back and forth’ < otcy ‘here’, otyśtatyś ‘from here and from there; from everywhere’ < otyś ‘thence, form there’, solan-talań / sopala-tapala ‘back and forth’ < solań ‘in that direction’. In the above-mentioned echo-pairs, the most productive relation of vowels in onsets is i : a, while o : a and u : a are rarer. However, the vowel a appears most often in the onset of the reduplicant. Often Udmurt reduplication forms serial verbs, that is, two verbs are compounded, forming an echo-pair by the same ending and creating a new meaning, for example, mynyny-vetlyny ‘to go back and forth, to pace back and forth’ > ‘to wander, to ramble’, kyľyny-myľyny ‘to stay; to throw away, to be thrown away’ < kyľyny ‘to fall behind, to lag behind; to stay’ + myľyny ‘to become superfluous’ (BUR 2020), and derivatives: kyľem-myľem ‘leftovers, waste, residue’, śiyny-juyny ‘to eat + to drink’ > ‘to have meals’, ulyny-vylyny ‘to live + to be’. Their continuations are also in expressions: ulon-vylon ‘living’ < ulon ‘life’), or ylem-vylem ‘lived; seasoned; life-being; being, existence’, which is seen at the beginning of many folk tales: ulem-vylem odig batyr ‘once upon a time there lived a hero’. Udmurt reduplication, and generally Permic reduplication is very rich and abundant. However, curiously, as far as Permic languages are concerned, Ivanov (2016) notes that the use of onomatopoeia without verbs occurs relatively rarely. See also the further studies on Udmurt reduplication by Shybanov (2010, 2017). An overview of the reduplication in Permic languages is also available, for example, in Shlyakhova (2013). 3.6

Ugric

The Ugric group includes Hungarian as well as Mansi and Khanty. On account of their geographical location between the Ural Mountains and the Ob River, Mansi and Khanty are also referred to as Ob-Ugric. 3.6.1

Hungarian

Hungarian lexical reduplication is exceptionally rich and has various types that have to be presented in a simplified manner.

3.6.1.1 Total reduplication Total reduplication is the easiest way to form double words, and one can find three main meanings in such constructions:  intensification: alig-alig ‘hardly’ + ‘hardly’ = ‘hardly, with great difficulty’, sok-sok (Kiefer & Németh 2019) ‘many’ + ‘many’ = ‘a lot, very many’

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

78

Iwona Piechnik

 distributivity: ki-ki ‘who’ + ‘who’ = ‘each’, olykor-olykor ‘sometimes’ + ‘sometimes’ = ‘rarely, seldom’, néha-néha ‘seldom’ + ‘seldom’ = ‘from time to time, very seldom’, egyszer-egyszer ‘once’ + ‘once’ = ‘once in a while, sometimes, every now and then, rarely’  indefiniteness: egy-egy fiú ‘some boys’, már-már ‘very nearly, almost’ < már ‘already, yet’.

3.6.1.2 Partial reduplication Partial reduplication is more frequently found in Hungarian than total reduplication.  Echo-pairs: replacement of an initial element (consonant or syllable) of the base by another one: ○ intensification: tarka-barka (Kiefer & Németh 2019) ‘very colourful, spotty’ < tarka ‘colourful, spotty’, csiga-biga (Kiefer & Németh 2019) / csigabiga (ES) ‘(tiny, sweet) snail’ < csiga ‘snail’; ○ diminutiveness: cica-mica (Kiefer & Németh 2019) ‘(tiny, sweet) kitten’ < cica ‘cat, kitten’  ideophones with vowel alternation: csip-csup ‘unimportant’, hercehurca ‘hassle’;  words with vowel alternation (similar to ablaut): dimbes-dombos (Kiefer & Németh 2019) ‘hilly, hummocky’ < domb ‘hill’ + adjectivizing suffix os, girbe-görbe / girbegurba ‘full of curves, sinuous, twisty’ < görbe ‘curved, wry’, rissz-rossz ‘sleazy’ < rossz ‘bad’. Other examples of this types: ringy-rongy ‘duds, gubbins’ < rongy ‘rag’, ripsz-ropsz ‘in a jiffy’ < rather onomatopoeic-based, alternatively from ripsz ‘rep’ (fabric), mendemonda (ES) ‘hearsay’  subducting of an initial consonant of the base: irul-pirul (Kiefer & Németh 2019) ‘blush, be blushful’ < pirul ‘blush’, ici-pici (Kiefer & Németh 2019) ‘very tiny’ < pici ‘tiny’, ihog-vihog ‘to chuckle’ < vihog ‘to chuckle’;  addition of an initial consonant to the copied base: for example, irkálfirkál ‘to scribble’ (< ír ‘to write’), and in many Hungarian diminutive names: Anna-Panna, Andi-Bandi, Ista-Pista, Erzse-Perzse, Öske-Böske.  only some of the base is copied: hébe-hóba (ES, Kiefer 1995–1996) ‘now and then’, csecsebecse (ES) ‘trinket, gewgaw’ < csecse ‘pretty, decorative’, cserebere ‘swapping’ < csere ‘exchange’,  rhyming compound verbs: ázik-fázik ‘to be rain-soaked’ < ázik ‘to soak’ + fázik ‘to feel cold’, csúszik-mászik ‘to crawl’ < csúszik ‘to slide’ + mászik ‘to climb’, tesz-vesz ‘to bustle’ < tesz ‘to do’ + vesz ‘to take’  rhyming adjectives and adverbs: össze-vissza ‘jumbled, confused’ < össze ‘together’ + vissza ‘backwards’, egyszer-ketszer ‘sometimes’ < egyszer ‘once’ + ketszer ‘twice’.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

Reduplication in Finno-Ugric languages

79

The Hungarian language has a huge number of these double words (ikerszók), and this phenomenon has been discussed many times in numerous publications (e.g. Szikszainé 1993, 2001 or Lipták & Saab 2019), which is why only a few examples are given here. As many reduplicated compounds in Hungarian indicate intensification, they ‘mainly serve as stylistic versions of their bases: they mostly express the endearing attitude of the speaker, thus they should be dealt with in a morphopragmatic framework as well’ (Kiefer & Németh 2019: 346). Another interesting phenomenon existing only in Hungarian reduplication applies to verbal prefixes (which could also be called preverbs). Kiefer (1995–1996: 175, 177, 178, 186, 190, 191), Lipták and Saab (2019: 531, 544, 552) and Keszler (2010: 300) give the following examples among others: meg-meg-áll ‘to stop from time to time’, vissza-vissza-néz ‘to look back from time to time / repeatedly’, vissza-vissza-ül ‘to sit down again’, át-át-jon ‘to come over from time to time’, át-át-megy ‘to go over from time to time’, bebe-rúg ‘to get drunk from time to time’, ki-ki-megy ‘to go out from time to time’, ki-ki-nyit ‘to open from time to time’, agyon-agyon-tapos ‘to trample to death from time to time’, agyon-agyon-üt ‘to strike dead from time to time’, agyon-agyon-lő ‘to shoot down from time to time’, agyon-agyon szúr ‘to stab to death from time to time’, agyon-agyon-ver ‘to beat from time to time’, belebele-néz ‘to look into something from time to time’, ki-ki-néz ‘to look out from time to time’, át-átnéz ‘to look across repeatedly’, be-benéz ‘to drop in repeatedly’, fel-felsóhajt ‘to sigh repeatedly’, hátra-hátranéz ‘to look behind one repeatedly’, ki-kimarad ‘to stay away repeatedly’, oda-odamond ‘to snap at somebody repeatedly’, and so on. Lipták and Saab (2019: 528) give such examples of verbs with complements in sentences: Peti rendszeresen be-be nézett az ablakon. ‘Peti looked in the window regularly.’, Fel-fel dobta az érmét a levegőbe. ‘He threw up the coin into the air from time to time.’ Of course, in all these examples, it is noteworthy that the meaning ‘from time to time’ is present in the meaning of reduplicated preverb. However there is a difference between normal preverbs and reduplicated preverbs: regular preverbs normally function as verbal prefixes, yet they can be separated from the verb and moved after it (like adverbs) as occurs in German; but, curiously, reduplicated preverbs cannot be detached from the verb (Kiefer 1995–1996: 188). Moreover, as Lipták and Saab (2019: 531–32) state, reduplicated particles cannot ‘occur in a sentence containing sentential negation’, ‘be focused or contrastively topicalized’ in a sentence, or ‘take place when the verb is elided’. Notwithstanding, there are constraints in the reduplication of Hungarian prefixes: first, they have to be monosyllabic or disyllabic (longer prefixes cannot be doubled); second, only preverbs expressing iterativity, telicity, and

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

80

Iwona Piechnik

resultativity can be reduplicated.22 Therefore preverbal reduplication is possible only with: át ‘through’, be ‘in’, ki ‘out’, fel ‘up’, vissza ‘back’, and meg ‘perfective’, as well as agyon ‘over, to death’ (not with all of the verbs).23 Curiously, these reduplicated verbal forms are never lexicalized and are not involved in processes of forming new words like derivation or compounding. Their meaning issues from the meaning of the prefixed verb, as well as the feature of iterativity (see Kiefer 1995–1996: 180, 194). And of course in Hungarian there is the possibility of creating syntactic reduplications (non-lexicalized), in emphatic goals, for example, csupa-csupa ‘complete’ + ‘complete’ > ‘full of something’ or like in the title of the novel by Frigyes Morvay: Sok-sok kérdés: Sok-sok válasz ‘Lots and lots of questions: Lots and lots of answers’ (2020). Moreover, according to Keszler, repeated numerals can express distributivity: két-két ember ment be ‘two persons went in at a time’ (< két ‘two’), három-három könyvet adtak a tanulóknak ‘the pupils were given three books each’ (< három ‘three’) (Keszler 2010: 300). In Hungarian, we can also find examples of the figura etymologica, for example, él(ni) az életet ‘to live the life’ or éneket énekel(ni) ‘to sing a song’, which are a kind of reduplication too. However in the following examples, Keszler rather sees reduplication expressing emphasis or augmentation: kérve kér (lit. ‘to ask asking’) ‘to beseech’ (< kér ‘ask for’), várva vár (lit. ‘to wait waiting’) ‘to look forward with impatience’ (< vár ‘wait for’), or concessivity: szépnek szép ‘pretty she may be [but . . .]’ (< szép ‘pretty’); tanulni tanultam ‘I did learn [but . . .]’ (< tanulni ‘to learn’) (Keszler 2010: 300). Nevertheless in my opinion they can be considered polyptoton, which is related to figura etymologica as well. Other examples of this type could be: szebbnél szebb ‘more beautiful than more beautiful’ (< szebb is comparative degree of szép ‘pretty, beautiful’) or óráról órára ‘hour by hour’ (< óra ‘hour’). A similar kind of stylistic reduplication can be found in poetry, like in the oldest Hungarian poem ‘Lamentations of Mary’ (Hun. Ómagyar Mária-siralom, dated to the thirteenth century): Világ világa, Virágnak virága ‘World’s light/world,24 Flower’s flower’.

22 23

24

Reduplication of prefixes is not therefore possible with stative verbs or with intransitive change of state verbs expressing irreversible changes. However, while preverbs express the iterativity of perfective events, reduplication also involves the prefix el ‘away’, which serves to express the lexical aspect (Aktionsart) of durativity rather than the aspect of perfectivity (Kiefer 1995–1996: 180, 181), e.g. el-el-ábrándozik ‘day-dream from time to time’, el-el-bámészkodik ‘stand gaping about from time to time’, el-el-dolgozgat ‘be working leisurely from time to time’, el-el-üldögél ‘sit about from time to time’, el-el-jár ‘visit from time to time’, el-el-olvas ‘read from time to time’. Hungarian világ can mean ‘world’ or ‘light’.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

Reduplication in Finno-Ugric languages

3.6.2

81

Ob-Ugric

We can now turn to discuss the two languages related to Hungarian that are spoken between the Ural Mountains and the Ob River. They are collectively known as Ob-Ugric: Mansi (formerly Vogul) and Khanty (formerly Ostyak) are spoken in the Khanty-Mansi Autonomous Okrug of the Russian Federation.

3.6.2.1 Mansi The Mansi language was previously known by the exonym ‘Vogul’. It has many regional varieties, which makes it difficult to provide an overview of the language or its general trends in reduplication. Most of the examples in this section come from Ahlqvist’s dictionary (1891). The other examples come from a new dictionary (Baxtijarova & Dinislamova 2016) containing materials from the area of upper Lozva River in Sverdlovsk Oblast, and a few examples are taken from Majtinskaja’s article (1964) and Forsberg’s book (2007) about eastern dialects.

3.6.2.1.1 Total reduplication The mere repetition of short words (one or two syllables) is very common and serves to express intensity and distributiveness, but also to create entirely new words that become lexicalized.  Intensity: āla-āla ‘a little bit; barely’ < āla ‘hardly, scarcely, near’ (Baxtijarova & Dinislamova 2016), janyj-janyj ‘big-big, very big’ (Baxtijarova & Dinislamova 2016) < janyj ‘big; main’, mań-mań ‘very small’ (Majtinskaja 1964: 125), lāščal-lāščal ‘silently, gently, on the sly’ < lāščal ‘slowly, quietly’ (Baxtijarova & Dinislamova 2016), tūp-tūp ‘slightly, barely’ (Baxtijarova & Dinislamova 2016) < tūp ‘a little’;  Distributivity: akval-akval ‘one by one, one at a time’ (Majtinskaja 1964: 130, Baxtijarova & Dinislamova 2016) < akv(a) ‘one’. Repeated numerals in distributive meaning: kit-kit kopeikäl ‘two kopecks each’, ät-ät punt ‘ten pounds each’, qurum-qurum quml ‘for each of the three men’, lou-lou līn ‘ten kopecks each’ (Ahlqvist 1894: 150);  Neutral repetition of lexemes: viť-viť / vit-vit ‘water bucket made of birch bark’ < viť / vit ‘water’, püv-püv ‘son’s son’,25 apyj-apyj ‘great-grandchild’ < apyj ‘grandchild’ (all Baxtijarova & Dinislamova 2016),

25

The same construction is in jege-püv-püv ‘brother’s son, nephew’ < jeg(e)-püv ‘brother’ < jeg ‘father’ + püv ‘son’.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

82

Iwona Piechnik

 Nouns of onomatopoeic origins: pišč-pišč ‘mouse’, rap-rap ‘swift (bird)’ (both Baxtijarova & Dinislamova 2016)  Other onomatopoeias and descriptive words: syr-syr ‘various, varied’ (Baxtijarova & Dinislamova 2016).26

3.6.2.1.2 Partial reduplication Partial repetition seems less frequent, but serves expressiveness and the formation of serial nouns.  Expressive echo-pairs: ščjohri-hōhri ‘dragonfly’ (Baxtijarova & Dinislamova 2016) < ščjohri ‘knife (with a thin, narrow blade)’;  Serial nouns, that is, compound words with the same morphological ending: āščij-ščānij ‘parents’ < āšč ‘father’ + ščāń ‘mother’, ètyj-hōtalyj ‘night-day’ < èt ‘night’ + hōtal ‘day’, vātāl-xaščtāl ‘completely unfamiliar, unknown’ < vātāl ‘unknown’ + xaščtāl ‘unknown’ (all Baxtijarova & Dinislamova 2016) – the latter example is tautological as both lexemes shares the same meaning;  Ideophones: ana-na ‘ow, ouch!, vos-vosńe ‘stint, sandpiper’ (both Baxtijarova & Dinislamova 2016). In Alqvist’s dictionary there are also words which seem to be reduplicated, but the supposed reduplicant is only similar to the other lexeme, for example, el-elam ‘to transfer, to translate’ < el- ‘far’ + elam ‘to tow’, xuljum-xul ‘ide, orfe’ < xuljum ‘the upper course of a small river’ + xul / qul ‘fish’, päľ-pāl ‘one of the ears’ < päľ ‘ear’ + pāl ‘side’. Regarding syntactic reduplication: Ahlqvist (1891) gives the following variants of Mansi greetings (salutations): pǟsi, pǟsi! / pasia, pasia! / pǟśe, pǟśe! (The present-day transliteration would be: pašča pašča.) Normally this reduplicated salutation is the answer of the addressee to the person who has greeted them with the words Pašča olen! In many folk tales, syntactic reduplication reveals a durative aspect of actions: mønøst, mønøst ‘they walked, they walked’, mønøs-mønøs ‘he walked, he walked’ (Forsberg 2007: 207–08, 216–17), kåjtøs, kåjtøs ‘he ran, he ran’ (Forsberg 2007: 218). Such repetitive use of verbs is quite common in Mansi stories, as repetition of words is generally common in Ob-Ugric poetry (see §3.6.2.2.2). However, Mansi seems to be a rather poor language when it comes to lexical reduplication.

26

This word only seems to be onomatopoeic because it cannot logically be a repetition of syr ‘perch’.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

Reduplication in Finno-Ugric languages

83

3.6.2.2 Khanty The Khanty language (formerly Ostyak) is the second Ob-Ugric language and has many dialects. The following examples are from the dialect of the area of the Kazym River (Solovar 2014).

3.6.2.2.1 Total reduplication As in Mansi, the mere repetition of words in Khanti serves to express the intensivity or distributivity, and some of these compounds become lexicalized:  Intensification Words: păsta-păsta ‘very quickly’ < păsta ‘quickly, rapidly’; Ideophones: syr-syr ‘diverse, various’, ťov-ťov ‘sandpiper species’, ščèvščèv ‘aquatic warbler’; Distributivity:  ajn-ajn ‘gradually’ < aj ‘small, little’, itn-itn ‘one by one’ < it ‘one’, kŭtakŭta ‘here and there’ < kŭt ‘distance; between’; Ideophones: vŷļ-vŷļ ‘from time to time’; Onomatopoeias: vov-vov-vov ‘dog’s bark’, ťoj-ťoj / răps-răps ‘sound of a cradle creak’.

3.6.2.2.2 Partial reduplication Partial reduplication in Khanty mainly consists of echo-pairs:  Ideophones and descriptive words: kăry-păry ‘diligent-industrious’, kătakŭta ‘negligently’, savêr-vavêr ‘agile, expeditious’, sérêmty-vorêmty ‘to mature’, uly-muly ‘stupid, foolish’, harêh-murêh ‘egoist’, ščŭŋki-ščaŋki ‘riddled with deep ruts, uneven road’; Serial words (compound words with the same endings): verbs: aļľêty-tŷty ‘to  wear-bear’, vŷļty-hoļty / vŷļêmty-hoļêmty ‘to live-sleep’ < vŷļty ‘to live’; nouns: aŋki-ašči / ašči-aŋki ‘parents’ < aŋki ‘mother’ + ašči ‘father’, imi-iki ‘man with woman’ < iki ‘man + imi ‘woman’, jajļam-johļam ‘human brothers’ < jaj ‘elder brother’ + joh ‘people’; adjectives: vantļy-vŷļy ‘(previously) unseenunknown (unfamiliar)’ < vantty ‘to see; to experience’ + vŷļy ‘to live’; adverbs: jăma-atma ‘badly-well’ < jăm ‘good’ + atêm ‘bad’, syta-pita ‘enough to satiety’ < syt ‘satiated’ (< Russ. sytyj) + pit (< Russ. piť ‘to drink’). As for syntactic reduplication, Shiyanova (2011), who studied the Shuryshkarsky District dialect (Yamalo-Nenets Autonomous Okrug), states that repeated words are frequent and occur in every variety of the Khanty language and have an intensifying or amplifying function:  Adjectives: kavrêm-kavrêm ‘hot-hot = very hot’, pity-pity ‘black-black = pitch-black’, navi-navi ‘white-white = snow-white’, vurty-vurty ‘readheaded-red-headed’ or ‘red-red’ (Shiyanova 2011: 39).

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

84

Iwona Piechnik

 Durative verbs: kănšty-kănšty ‘to search-search = looking for’, măntymănty ‘to go-go = riding, driving’; jity-jity ‘to go-go = to go easy’ (Shiyanova 2011: 39).  Pronouns: muj-muj ‘what-what = what so’, tumi-tumi ‘this-this = this/ that one’.  Adverbs: tup-tup ‘only-only = just (now)’, sora-sora ‘quickly-quickly = very quickly’, ščir-ščir ‘well-well = as it should be, appropriately’, hotyhoty ‘how-how = how so?’, šeŋk-šeŋk ‘very-very’ (Shiyanova 2011: 39). These few examples illustrate the popularity of this phenomenon, and in this context, Ob-Ugric poetry is worth mentioning too. In a study of Old Khanty songs, M. Sipos states: ‘Repetition is the most salient feature of the language of Ob-Ugric folklore’ (Sipos 2021: 132). On the basis of Old Khanty heroic epic songs collected by Hungarian scholar Antal Reguly,27 Sipos (2021) analysed examples of figura etymologica, polyptoton, and simple word repetitions used in northern Khanty folklore. Various repetition techniques were shown to play an important role in the oral epic tradition as they enabled both the singer to memorize and improvise, and the listeners to follow the songs. 3.7

Conclusions

In summarizing the observations of reduplication in Finno-Ugric languages, my main conclusions are:  Hungarian, Komi, Udmurt, and Mari have a more marked tendency towards reduplication than other Finno-Ugric languages. On this score, Saami languages are the poorest: they generally do not have lexical reduplication.  Lexical partial reduplication is more frequent than total reduplication. However, in syntactic reduplication, words are almost always repeated in whole.  Finno-Ugric languages have many onomatopoeias and descriptive words (ideophones) which can function as various parts of speech, mainly as adverbs and adjectives. They can sometimes also complete verbs that have a general meaning, like ‘to do’, ‘to make’, ‘to go’, ‘to take’.  Many of the Finno-Ugric languages share the same concepts, for example, reduplicative constructions with numerals to express distributivity. I have not devoted too much space to syntactic reduplication, which carries many meanings (e.g. emphasis, frequency, iterativity, durativity, continuation), as it would have been too broad a topic to discuss together with lexical 27

The collected songs were mainly published in the first half of the twentieth century.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

Reduplication in Finno-Ugric languages

85

reduplication in a single chapter. However, several researchers have already devoted in-depth works to this subject for some of these languages separately, for example, Keevallik (2010) for Estonian, Brdar and Brdar-Szabó (2014) for Hungarian. This chapter has presented an outline of the wide variety of phenomena associated with reduplication in the Finno-Ugric languages. More typological research is needed, since these languages are still little known to readers who cannot read Russian, Finnish, Estonian, or Hungarian. Fortunately, researchers are now working on this issue, though in a rather pointwise manner – focusing on only a single language, a few languages, or a subgroup – for example, the work of Rozhanskiy (2018), or the Typological Database of the Volga Area Finno-Ugric Languages by Asztalos, Gulyás, Horváth, and Timár (2021). Abbreviations E Eng. Est. Fin. Hun. M Russ.

Erzya (Mordvinic language) English Estonian Finnish Hungarian Moksha (Mordvinic language) Russian

References Ahlqvist, A. (1891). Wogulisches Wörterverzeichnis. Helsinki: SKS. (1894). Wogulische Sprachtexte nebst Entwurf einer wogulischen Grammatik, Yrjö Wichmann (ed.). Helsinki: SKS. Asztalos, E., Gulyás, N. F., Horváth, L., & Timár B. (2021). New aspects in the study of Mari, Udmurt, and Komi-Permyak: The typological database of the Volga area Finno-Ugric languages. In S. Szeverényi (ed.) Uralic Studies, Languages, and Researchers. Studia Uralo-altaica, Vol. 54. Szeged: University of Szeged, 255–74. DOI: https://doi.org/10.14232/sua.2021.54.255-274 Bartens, R. (1999). Mordvalaisten kielten rakenne ja kehitys. Helsinki: Suomalaisugrilainen seura. (2000). Permiläisten kielten rakenne ja kehitys. Helsinki: Suomalais-ugrilainen seura. Baxtijarova, T. P. & Dinislamova, S. S. (Бахтиярова, Т.П. & Динисламова, C.C.) (2016). Мансийско-Русский Словарь (Верхне-Лозьвинский Диалект) [Mansijsko-russkij slovar’ (verxne loz’vinskij dialekt)]. Tyumen: Format. BKR (2020). Большой Коми–Русский Словарь. [Bol’šoj komi-russkij slovar’ ] [Great Komi–Russian dictionary]. Institute of Language, Literature and History, Federal Research Centre Komi Scientific Center of the Ural Branch of the Russian Academy of Sciences. https://dict.fu-lab.ru/dict?id=5109

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

86

Iwona Piechnik

BMR (2020). Большой Марийско–Русский Словарь. [Bol’šoj marijsko-russkij slovar’] [Great Mari–Russian dictionary]. Mari Research Institute of Language, Literature and History. https://dict.fu-lab.ru/dict?id=37767 BUR (2020). Большой Удмуртско–Русский Словарь. [Bol’šoj udmurtsko-russkij slovar’] [Great Udmurt–Russian dictionary]. Udmurt Institute of History, Language and Literature of the Ural Branch of the Russian Academy of Sciences. https://dict.fu-lab.ru/dict?id=129449 Bradley, J., Riese, T., & Guseva, E. (2014). Mari–English Dictionary. Vienna: University of Vienna. https://mari-language.univie.ac.at Brdar, M. & Brdar-Szabó, R. (2014). Syntactic reduplicative constructions in Hungarian (and elsewhere): Categorization, topicalization and concessivity rolled into one. In G. Rundblad et al. (eds.) Selected Papers from the 4th UK Cognitive Linguistics Conference. London: UK Cognitive Linguistics Association, 36–51. EKSS (2009). Eesti keele seletav sõnaraamat [Explanatory dictionary of the Estonian language]. Tallinn: Eesti Keele Instituut. https://eki.ee/dict/ekss Erelt, M. (1997). Reduplication in Estonian. In M. Erelt (ed.) Estonian: Typological Studies II, Tartu: Department of Estonian of the University of Tartu, 9‒41. (2008). Intensifying reduplication in Estonian. Linguistica Uralica 44(4), 268‒77. Erelt, M. & Punttila, M. (1999). Suomalais-ugrilaisten kielten reduplikaatiosta. Lähivertailuja 10, 3‒12. EUD (2015). Eston–udmurt kylľukam [Estonian–Udmurt dictionary]. Tallinn: Eesti Keele Instituut. https://eki.ee/dict/eud Fejes L. (2004) ‘Compound verbs’ in Komi: Grammaticalisation without a grammatical morpheme? Acta Linguistica Hungarica 51 (1–2), 5–43. Feoktistov, A. P. (Феоктистов, А. П.) (1974). О конструкциях с повтором основы (figura etymologica) [O konstrukcijax s povtorom osnovy (figura etymologica)] [On constructions with stem repetition (figura etymologica)]. In L. Szűts (ed.) Jelentéstan és stilisztika. A Magyar Nyelvészek 2. Nemzetközi Kongresszusának előadásai. Budapest: Akadémiai Kiadó, 129–36. Forsberg, U.-M. (2007) Itämansin kielioppi ja tekstejä. Helsinki: Suomalais-Ugrilainen Seura. ISK = Hakulinen, A., Vilkuna, M., Korhonen, R., Koivisto, V., Heinonen, T. R., & Alho, I. (2004) Iso suomen kielioppi. Helsinki: Suomalaisen Kirjallisuuden Seura. Ivanov, V. A. (2016). Синтаксис звукоподражаний в финно-пермских языках [Sintaksis zvykopodražanij v finno-permskix jazykax] [Syntax of onomatopoeia in the Finno-Permian languages]. Acta Linguistica Petropolitana 12(1), 709–35. Keevallik, L. (2010). Social action of syntactic reduplication. Journal of Pragmatics 42, 800–24. Keszler, B. (2010). Reduplication and (especially general and indefinite) pronominal function. Studia slavica Academiae Scientiarum Hungaricae 55(2), 299–302. Kiefer, F. (1995–1996). Prefix reduplication in Hungarian. Acta Linguistica Hungarica 43(1‒2), 175‒94. Kiefer, F. & Németh B. (2019). Compounds and multi-word expressions in Hungarian. In B. Schlücker (ed.) Complex Lexical Units: Compounds and multi-word expressions. Berlin and Boston: De Gruyter, 337–58. KM = Konakov, N. D., Napol’skih, V. V., Siikala, A.-L., Hoppál, M., & Belyh, S. (2003) Komi Mythology, Budapest: Akadémiai Kiadó and Helsinki: Finnish Literature Society.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

Reduplication in Finno-Ugric languages

87

Lipták, A. & Saab A. (2019). Hungarian particle reduplication as local doubling. Acta Linguistica Academica 66(4), 527–74. Lönnrot, E. (1907). Kalevala: The Land of heroes. 2 vols. Translated by W. F. Kirby. London: Dent. Mäger, M. (1966). Intensiivistavad reduplikatiivsõnad eesti murretes. Emakeele Seltsi Aastaraamat 12, 91–107. Majtinskaja, K. E. (Майтинская, К. Е.) (1964). Структурные типы удвоений (повторений) в финно-угорских языках [Strukturnyje tipy udvojenij (povtorenij) v finno-ugorskix jazykax] [Structural types of doublings (repetitions) in the Finno-Ugric languages]. In B. А. Serebennikov (Б. А. Серебренников) (ed.) Вопросы финно-угорского языкознания: Грамматика и лексикология [Voprosy finno-ugorskogo jazykoznanija: Grammatika i leksikologija] [Issues of Finno-Ugric linguistics: grammar and lexicology]. Leningrad and Moscow: Nauka, 122‒35. Miestamo, M. (2011). Skolt Saami: A typological profile. Suomalais-Ugrilaisen Seuran Aikakauskirja 93: 111–45. Mikone, E. (2001). Ideophones in the Balto-Finnic languages. In F. K. Erhard Voeltz & C. Kilian-Hatz (eds.) Ideophones. Typological Studies in Language, Vol. 44. Amsterdam, Philadelphia: Benjamins, 223–33. Morvay F. (2020) Sok-sok kérdés: Sok-sok válasz. Budapest: Könyv Guru. Niemi, A. R. (ed.) (1919). Suomen kansan vanhat runot. [Old poems of the Finnish people]. Parts 3–4. Helsinki: SKS. Niilus, V. (1938). Reduplikatiivsõnadest. Eesti Keel 17, 224–32. Paasonen, H. (1990–1996). Mordwinisches Wörterbuch. The dialect dictionary of the Mordvin languages based on the materials of Heikki Paasonen. Helsinki: Suomalais-Ugrilainen Seura. Volumes 1–6 online. https://www.mv.helsinki.fi/ home/rueter/PaasonenMW.shtml Räsänen, M. (2010). Päivä päivältä enemmän ja enemmän. Suomen toistokonstruktioita. Helsinki: Yliopistopaino. Rießler, M. (2022). Kildin Saami. In M. Bakró-Nagy, J. Laakso & E. Skribnik (eds.) The Oxford Guide to the Uralic Languages, Oxford: Oxford University Press, 219–39. Rozhanskiy, F. (2018). Non-Canonical behavior of reduplicated ideophones in Komi. In A. Urdze (ed.) Non-Prototypical Reduplication. Berlin and Boston: De Gruyter Mouton. 177–99. Sammallahti, P. (1993). Sámi-suoma-sámi sátnegirji: Saamelais-suomalaissaamelainen sanakirja, Ohcejohka: Girjegiisá. (1998). Saamic. In D. Abondolo (ed.) The Uralic Languages, London and New York: Routledge, 43–95. Shiyanova, A. A. (Шиянова, А. А.) (2011). Парно-повторные слова в хантыйском языке (на материале шурышкарского диалекта) [Parno-povtornyje slova w xantyjskom jazyke (na materiale šuryškarskogo dialekta)] [Pair-repeated words in Khanty language (on material of shuryshkar dialect)]. Вестник Угроведения: Филология [Vestnik ugrovedenija: Filologija] 3(6), 38–40. Shlyakhova, S. S. (2013). Study of reduplication in Finno-Ugrian languages of Permic group. Middle-East Journal of Scientific Research, 16, 1329–333. Shybanov, A. A. (Шибанов, А. А.) (2010). Наречия и подражательные слова в удмуртском языке [Narečija i podražateľnyje slova v udmurtskom jazyke]

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

88

Iwona Piechnik

[Adverbs and imitative words in the Udmurt language]. Вестник Челябинского государственного университета [Vestnik Čeljabinskogo gosudarstvennogo universiteta] 34(215): Филология. Искусствоведение [Filologija. Iskusstvovedenie] 49: 133–36. (2017). Редуплицированные подражательные слова в удмуртском языке [Reduplicirovannyje podražateľnyje slova v udmurtskom jazyke]. ВосточноЕвропейский научный вестник [Vostočno-Evropejskij naučnyj vestnik] 12(4), 85–88. Siikala, A-L. & Ulyashev, O. (2016). Hidden Rituals and Public Performances. Helsinki: Finnish Literature Society. Sipos, M. (2021). Word and stem repetitions in the heroic epic songs collected by Antal Reguly. In S. Szeverényi (ed.) Uralic studies, languages, and researchers. Proceedings of the 5th Mikola Conference. 19–20, September 2019. Szeged: University of Szeged, 131–47. SMS (2022). Suomen murteiden sanakirja. [Dictionary of Finnish dialects]. Centre for Finnish Languages. https://kaino.kotus.fi/sms Solovar, V. N. (Соловар, В. Н.) (2014). Хантыйско-русский словарь (казымский диалект) [Xantyjsko–russkij slovar’ (kazymskij dialekt)] [Khanty–Russian dictionary]. Khanty-Mansiysk: Format. Szikszainé Nagy, I. (1993). Az ikerítés helye, szerepe, szabályszerűségei a magyar nyelvben, és szabályszerűségei a magyar nyelvben, Budapest: Magyar Nyelvtudományi Társaság. (2001). Az ikerszók a magyar szókincs rendszerében, Magyar Nyelvjárások 39: 77–87. Tsygankin, D. V. (1980). Грамматика Мордовских Языков. [Grammatika mordovskikh yazykov] [Grammar of the Mordovian languages]. Saransk: Mordovian State University. Tuomi, T. (1993). Täpötäysi ja patajuoppo. Kieliviesti 4/1993, 8–10. UKDW (2020). Uusi kielemme: Ruokaruoka Pikapikaa [Finnish for busy people: Double words – Reduplication]. https://bit.ly/3YpR215 Vartiainen, V. (2019). Kotikotiin syömään ruokaruokaa. Kielikello 3/2019. https://www .kielikello.fi/-/kotikotiin-syomaan-ruokaruokaa Wichmann, Y. (1916). Syrjänische Volksdichtung. Helsinki: Société Finno-Ougrienne. Zaicz, G. (1998). Mordva. In D. Abondolo (ed.) The Uralic Languages. London and New York: Routledge, 184–218.

https://doi.org/10.1017/9781108989084.004 Published online by Cambridge University Press

Germanic

https://doi.org/10.1017/9781108989084.005 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.005 Published online by Cambridge University Press

4

Expressivity in Scots: A study of echo words Jeffrey P. Williams*

4.1

Introduction

Scots has been spoken in Scotland for many centuries and is spoken today throughout the east and south of the country – the historic Lowlands – and also in Orkney and Shetland, which form the Northern Isles. The Scots language is part of the larger West Germanic branch of Indo-European and a rather close relative of Modern English. Both Scots and Modern English diverged from the Early Middle English that had begun in the eleventh century. Scots has maintained some of the features of this earlier variety of English in its more isolated, rural dialects. Scots originated with the language of the Angles who arrived in what is now south-east Scotland about 1,400 years ago. During the Middle Ages this language developed and grew apart from its sister tongue in England, until a distinct Scots language had evolved. At one time Scots was the dominant language of Scotland and was used to compose literary works as well as to inscribe official records. Scots, like too many other European languages, is viewed as possessing a dearth of expressivity in spite of evidence to the contrary. Nuckolls (2003) has claimed that ideophones as a class of expressions are underdeveloped as a category in Standard Average European languages. The perception is that ideophones are a common feature in many of the world’s languages but are underdeveloped in English and other Indo-European languages. Scots possess a wide range of forms of expressivity as part of its grammatical repertoire – including most notably echo word formations. This chapter will focus on the category of echo words in Scots expressivity. 4.2

Background to expressivity

Expressivity is the grammaticalization of sensory qualities: it is used to characterize the way things appear to the senses and are articulated and * Texas Tech University.

91

https://doi.org/10.1017/9781108989084.006 Published online by Cambridge University Press

92

Jeffrey P. Williams

conveyed in a linguistic message. Expressivity, through its iconicity, reduces the cumbersomeness of description in discourse: it allows the speaker to condense information into a single praxis. Expressivity is difficult to study on its own – the clearest path to gaining a better understanding of expressivity lies in the study of expressives – the hallmarks of expressivity in language, in which speakers transform grammatical resources in order to express and convey emotions, senses, conditions, and perceptions that enrich both vernacular and ritual discourse. Expressivity can be viewed as a mechanism to foreground characteristics of a semantic field defined by perceptual parameters. Expressivity in any language allows speakers to bring to the fore aspects of an entity or event through the structural properties of grammar. Components are given prominence through transformation, sometimes involving the rule of quantity whereby more elements demand more cognitive attention. Expressivity is the process in grammar that allows speakers to comment on their perceptions of the world – perceptions being defined as the state of being or process of becoming aware of something through the senses. Importantly, expressivity further provides the grammatical resources for commenting on the fleeting world of sensory perception. It is tied directly to our ability to draw upon our physiological apparatus to comment on the signals we receive from the outside world. Perception, or what we distinguish and how we recognize and mark those distinctions, is an issue that occupies the discourse of philosophers, cognitive scientists, psychologists, and others. These concerns bleed over into the discussion of grammar, especially when we delve into the realm of expressivity. Expressivity is the praxis of grammar. Expressivity provides a shorthand means of expressing a complex concept. It points out a perception, depicting it in the grammatical resources of the language. Expressivity is no different from other properties of human language that find articulation in the grammar through principles, rules, and representations. I contend that expressivity is governed by a principle of expressivity that is universal in human language, which states the following: a systematic feature of human language is the ability to articulate and communicate perception of natural and social worlds. I refer to this feature as the principle of expressivity. The principle of expressivity is manifested through the grammatical resources of human languages. The grammar of each language has its own set of structures that can be employed by speakers to reflect on perceptions of actions, activities, states, and the social positions of individuals. This accounts for the variation we find across languages in terms of what structurally constitutes expressivity. The variation has been formally treated in a variety of ways; some are predominantly semantic, others syntactic, and a few, attempt

https://doi.org/10.1017/9781108989084.006 Published online by Cambridge University Press

Expressivity in Scots

93

to unite both treatments through a singular formalization. Regardless of the treatment, all of these accounts focus on the category of expressives that are found in human language.1 Expressives, also referred to as ideophones, are the poster children for expressivity in grammar. The concept of expressives is a relatively young one in linguistics. Linguistic analyses of ideophones, reduplication, echo words, onomatopoeia, and other non-categorical forms have grown over the past two decades (see Voeltz & Kilian-Hatz 2001, McCarthy et al. 2012, Akita 2015, Iwasaki et al. 2017, Ibarretxe-Antuñano 2017, Dingemanse 2012, 2015, 2018, Haiman 2019, and, Akita & Pardeshi 2019, Williams 2004, 2021 to name only some). One of the earliest descriptions of an expressive can be found in the writings of the psychologist-phonetician Edward Wheeler Scripture (1864–1945). In his Elements of Experimental Phonetics, Scripture used the term ideogram for the holistic perception of a printed word. He states that ‘printed words are perceived in wholes as ideograms and not as combinations . . . . . . . words may be perceived under conditions that exclude any perception of the single elements’ (Scripture 1902: 128). He goes on further to create a parallel between the ‘image’ of the printed word and that of the auditory word: It may be suggested that auditory words and phrases form ‘ideophones’ just as printed ones form ‘ideograms’. The further distinctions may be made of ideograms and ideophones into sensory (visual words and auditory words) and motor ones (written words and spoken words). In all probability the most prominent features of a phonetic unit are first perceived, and the details are gradually filled in. (Scripture 1902: 132)

For Scripture then, an expressive is a unit of sound that is perceived as a whole rather than as a combination of some parts; and this whole represents one idea. This condition is what has driven linguists to consider expressives as somehow special in terms of their linguistic categorization. There are several types or forms of expressives found in the grammars of human languages. It is beyond the scope of the present chapter to go into these in any detail. Expressives are shape-shifting forms whose functions cross-cut grammatical categories and classes but, in general, serve to allow the speaker to grammaticalize perception and/or provide meta-commentary on an argument in the discourse. As the name conveys, expressives allow speakers to ‘express’ an

1

There is a chasm between two linguistic schools of expressivity – those who focus primarily on expressivity in the broadest terms and linguistic data from English, German, and other European languages (e.g. Gutzmann 2019) and those who focus on African and Asian languages from a more anthropological bent. Neither camp cites the other or seems to be concerned about how to unite the vast amount of information on expressivity that exists and has yet to be reconciled into a coherent theoretical account.

https://doi.org/10.1017/9781108989084.006 Published online by Cambridge University Press

94

Jeffrey P. Williams

opinion, an attitude, a perception, or other psychological state regarding a topic in a situated discourse (Williams 2021). The conclusion of all of these accounts is that expressives are somehow different and special and do not fit either within the traditional morphological component of grammar or the descriptive dimension of meaning; or that they are not part of productive morphology as Inkelas and Zoll (2005) contend.2 For many linguists, an expressive is a unit of sound that is perceived as a whole rather than as a combination of some parts; and this whole represents one idea. This condition is what has driven linguists to consider expressives as somehow special in terms of their linguistic categorization. Some scholars simply claim that expressivity is extra-grammatical in the sense that the principles of expressivity do not follow the regular rules of prosaic, or natural grammar (Mattielo 2013, Zwicky & Pullum 1987). 4.2.1

Echo words: Background

Echo word morphology has been a topic of particular interest in morphology and phonology for some years now. The phenomenon has been characterized by several different terminologies and from a theoretical perspective, with several attempts to generate the plethora of forms in relevant languages. In spite of the interest in this aspect of grammar, there is little consensus on what echo word morphology actually is, either descriptively or typologically. Echo words are a linguistic category that is widespread in the languages of the world, especially in Eurasia. We find them endemic throughout the IndoAryan, Dravidian, Tibeto-Burman and Munda languages of South Asia; pervasive in the languages of Mainland and Insular Southeast Asia; scattered throughout the languages of Central Asia, Tibet, and the Mongolian Plateau; and, finally, present in some of the languages of Europe. Echo words are not widely found in the indigenous languages of the Americas nor Australia nor are they present to any extent, in the languages of the Pacific. The earliest attribution of echo words in the Anglophone linguistic tradition can be found in Chatterji’s 1926 multi-volume work on the history and development of the Bengali language. While the concept of ‘echo words’ is most often attributed to Emeneau’s work (1938) on the ‘echo word motif’ in Dravidian folk tales, it is clear from Emeneau’s reference to Chatterji that the 1926 account is the earliest we have in English.3

2 3

Haiman (2019) has made the further claim that expressive morphology is purely decorative in language. It remains to be seen if there are accounts of echo words in other linguistic traditions, such as Sanskrit, Tibetan, or others.

https://doi.org/10.1017/9781108989084.006 Published online by Cambridge University Press

Expressivity in Scots

95

Chatterji describes the ‘presence of ‘echo words’ in Bengali and other South Asian languages. He defines these as: . . . a word [that] is repeated partially (partially in the sense that a new syllable, the nature of which is generally fixed, is substituted for the initial one of the word in question, and the new word so formed, unmeaning by itself, echoes the sense and sound of the original word), and in this way the idea of et cetera, and things similar to or associated with that, is expressed. Chatterji (1926: 176)

Chatterji provides the following early examples of echo words from Bengali and Gujarati, showing the cognate status of the forms. (1a)

ghōra – tōra

(1b)

ghōrō – bōrō Gujarati ‘horses, etc. – horses and the like’

Bengali

He provides the basic patterns of a single type of echo word in several languages but does not fully dissect the complete patterns in any one language. It is the first descriptive account of the process with examples that we find in the South Asian sphere. Chatterji goes on to note that these ‘echo words’ are different from compounds where one member is an obsolete word; instead, in ‘echo words’ there is an unmeaning echoant. This is an important distinction that must be held between compounds and echo words in South Asian languages. Compounds in many South Asian languages concatenate a word that is in currency with an obsolete, or archaic word that typically conveys the same, or a very similar meaning. We also find this sort of patterning in other languages with echo word morphology such as Jarai (Williams and Siu 2014) and Scots, which is the subject of the following sections.

4.2.2

Echo words: Terminology

Echo words are a word formation process involving the formation or generation of rhyming words through copying, change, and collocation. Echo word morphology goes by a wide range of names in the descriptive grammars where it is the focus in languages such as Vietnamese, Telugu, Mongolian, and many others. Rhyming is a tricky term to use in a universal fashion since the conceptualization of what it means to rhyme is governed by local grammatical conditions. Abbi has defined an echo word as ‘. . . a partially repeated form of the base word – partially in the sense that either the initial phoneme (which can be either a consonant or a vowel) or the syllable of the base is replaced by another phoneme or another syllable’ (Abbi 1992: 20). Nowadays, there a several descriptors used to cover the range of phenomena we find in the world’s languages, with some authors referring to these forms as

https://doi.org/10.1017/9781108989084.006 Published online by Cambridge University Press

96

Jeffrey P. Williams

rhyming reduplication, overwriting, fixed segment reduplication, and chameleon affixation. In this treatment I continue to use Chatterji’s original terms to describe this form of expressivity in Scots. Echo word formation is an example of reduplication with prosodic replacement in the copy. Replacement can involve any prosodic portion of the copy, including an onset, nucleus, coda, tonal pattern, and register. The copy with prosodic replacement is referred to as the “echoant. ” The expressivity of echo forms is evidenced by the lack of direct translatability for the echo “collocant.” The meaning of the collocant is part of the meaning of the entire collocation, which differs from the meaning of the root, or stem, alone. A diagnostic feature of echo formation is that the replacive morpheme directly impacts the meaning of the echo collocation. There can be, and is, a pattern of difference in echoants. For echo word formation where roots and other non-meaningful morphs have multiple collocative allomorphs, I would suggest that a common term “echoant” be used to categorize them on par with the use of the term “reduplicants,” which results in a difference in meaning. 4.3

Scots echo words

What has been stated for the class of echo words in general also holds for Scots forms. Scots exhibits a wide range of echo forms in its morphosyntax. A degree of sensory motivated allomorphy is also a property of expressivity in Scots. Just as in other languages, we must be able to distinguish compound forms from echo word forms: where fairly straightforward structural differences exist between the two, semantically, each conveys meaning and exists as a lexical item. The data for this study derive from primarily printed sources, including dictionaries, newspapers, and other linguistic materials. Expressives in Scots has not hitherto appeared as the topic or subject of any scholarly paper and the closest topical linguistic treatment is Williams (2004). (2)

catter-batter – A quarrel or a dispute. < batter ( V, to give repeated blows)

(3)

clitter-clatter – Particularly noisy and animated talk and chatter, a clattering noise, often confused and senseless conversation. < clatter (N, noisy talk)

(4)

currie-wurrie – A dispute which ends in violence. < wurrie (growling noise)

(5)

deedle-doodle – N, A meaningless song, 2. A badly played tune. from deedle – V, 1. To dandle an infant, 2. To train an infant, 3. To sing in a low tone, to hum an air without the words.

(6)

diddle-daddle – To waste or take your time. A lot of ineffectual activity with little to show for it.

(7)

eckle-feckle – ADJ, 1. Cheerful, merry, 2. Possessing a shrewd judgment.

https://doi.org/10.1017/9781108989084.006 Published online by Cambridge University Press

Expressivity in Scots

97

(8a)

eedle-deedle – ADJ, Easy going in business. N, an easy going person, a daydreamer.

(8b)

eedle-doddle – Same as eedle-deedle.

(9)

eeksy-peeksy – ADJ, Equal, exactly equal, alike.

(10a)

equal-aqual – ADJ, Equally balanced, 2. Exactly alike or equal. V, To make equal, 2. To balance.

(10b)

equals-aquals – ADV, On a strict equality.

(11)

feckle – N, Trouble, anxiety.

(12)

fiddle-diddle – N, Music of the fiddle, the movement of a fiddler’s arm when playing the fiddle.

(13)

fiddle-faddle – N, nonsense, whim. V, To dawdle.

(14a)

feery-fary – N, A great hubbub, an angry tumult. from feery – N, 1. Tumult, bustle, confusion. 2. Rage, passion.

(14b)

feery-farry – N, A fight. < feery (N, 1. tumult, bustle, confusion). 2. Rage, passion. N, A great hubbub, an angry tumult.

(15)

fick-facks – N, A fiddling, finicking or tedious piece of work, a fuss about nothing.

(16a)

figgle-faggle – N, Silly, ludicrous, or unbecoming conduct.

(16b)

figgle-faggler – N, One who destroys good morals. (In this form valence can be changed through the addition of the -er actor suffix to the collocation.)

(17)

glim-glam – N, Another name for the childhood game Blind Man’s Bluff.

(18)

haggerty-taggerty – ADJ, In a ragged state.

(19)

haggle-baggle – N, A dispute over prolonged bargaining by someone reluctant to accept the terms of a bargain.

(20)

hickertie-pickertie – N, Another phrase for higgledy-piggedly meaning confusion and disorder.

(21)

hiddie-giddie – ADJ, Topsy turvy.

(22)

hingum-tringum – V, To be in low spirits. N, Worthless, an unsavory character.

(23)

hink-stink – N, A small beer.

(24)

hinkie-pinkie – ADJ, Description of a weak beer.

(25)

hirrie-harrie – INJ, An outcry after catching a thief.

(26)

hish-hash – N, A mess or a muddle.

(27)

hockerty-cockerty – ADV, A phrase meaning to sit on another person’s shoulders.

(28)

hodge-podge – N, Very thick soup.

(29)

hudderie-dudderie – ADJ, To have a scruffy and unkempt appearance.

(30)

hurry-burry – V, To feel harassed, can also mean a dispute/rumpus.

(31)

mixter-maxter – ADJ, To be in a chaotic and jumbled up state.

https://doi.org/10.1017/9781108989084.006 Published online by Cambridge University Press

98

Jeffrey P. Williams

(32)

pitter-patter – V, To recite words quickly without giving them the careful thought and time they deserve.

(33)

reeble-rabble – N, A great confusion.

(34)

reel-rall – N, A state of confusion. V, To move or walk confusedly, to walk in an aimless manner. ADV, Topsy-turvy.

(35)

reemle-rammle – V, To make a great deal of noise – to behave noisily. N, A great noise. ADV, In a noisy, rude manner, with a jingling and confused sound.

(36)

reestlin-rustlin – ADJ, So dry as to rustle – (used of corn).

(37)

ribble-rabble – N, A great confusion. V, To crowd in great confusion. ADV, In great confusion.

(38)

snochter-dichter – N, Handkerchief.

There are two types of expressive compounds in Scots: echo words and pure expressives. I refer the reader back to §4.2.1 for a refresher on the general principles of echo words. In Scots, echo words tend to exhibit the following pattern: (39)

root ! root + echoant OR echoant + root

Scots has a proclivity to suffix the echoant to the root as shown in Example (40) below. (40a)

fiddle – N, A stringed musical instrument played with a bow.

(40b)

fiddle-diddle – N, Music of the fiddle; the movement of a fiddler’s arm when playing the fiddle.

However, there are some forms where the echoant is prefixed to the root. The echoant is a modification of the root in terms of phonological elements, which creates a rhyming compound. (41a)

rabble – V, 1. To speak indistinctly and quickly. 2. To work hastily and carelessly. 3. To mob riotously. N, 1. Confused, careless speech. 2. Incoherent reading or talking. 3. A gabble. 4. Hasty and careless work. 5. A ruinous mass, a building falling to decay. 6. A careless, hurried worker. 7. A foolish story.

(41b)

reeble-rabble – N, A great confusion.

(41c)

ribble-rabble – N, 1. A great confusion. V, 1. To crowd in great confusion. ADV, 1. In great confusion.

Example (41c) is a classic form of expressive whereby the meaning cross-cuts three derivational categories in Scots: noun, verb, and adverb. It is also important to note that both Examples (41b) and (41c) have the same meaning in the nominal scope. The expressivity of echo forms is evidenced by the lack of direct translatability for the echo collocant. The meaning of the collocant is part of the meaning of the entire collocation, which differs from the meaning of the root, or stem, alone.

https://doi.org/10.1017/9781108989084.006 Published online by Cambridge University Press

Expressivity in Scots

99

There are also forms that I am calling pure expressives for the fact that they are a rhyming pair in which neither has independent meaning. The meaning is conveyed by the collocation of the two forms into a single depiction. These forms often have multiple meanings and cross-cut derivational classes. 4.3.1

Analysis of compound forms

Scots, like other Germanic languages, makes use of extensive compounding as a feature of its morphology and word formation. A large number of roots can be combined to create new forms such as the examples shown in (42) through (50) below. (42)

gumple-face – N, A down cast face < gumple, V, 1. To become sulky. 2. To show bad humor. N, A surfeit.

(43)

tea-man – N, A tea drinker.

(44)

temper-thrawing – ADJ, Souring the temper > temper + thraw.

(45)

tear-blob – N, Tear drop.

(46)

reeving-wind – N, High wind < reeving (high, strong).

These forms, however, are not expressives in Scots. They are simply compounds. We know they are not expressives because the two roots have independent meanings that can be analyzed in isolation from each other. Expressive compounds, on the other hand, do not have two independent roots with isolatable meanings. (47)

dyke-slouch – N, A ditch or open drain at the bottom of a dyke.

(48)

ear-leather – N, The loin strap, passing through the crupper and over the kidneys of a horse.

(49)

ebb-bait – N, Shellfish used by fishermen for fishing bait.

(50)

even-hands – ADV, On equal terms. N, An equal bargain.

Some expressives are true expressives in that they do not possess any root form that can be isolated from the total collocation to provide a meaning. (51)

eastie-wastie – N, A vacillating person.

(52)

easy-osy ~ easy-oasy – ADJ, Easy-going. N, An easy-going person.

Scots expressivity also exhibits vowel disharmony in echo forms.4 As is typical of echo word formation in the world’s languages, a prosodic category of the echoant is altered to create the new collocation of the base plus the

4

Yip (1998) refers to this phenomena as “identity avoidance.”

https://doi.org/10.1017/9781108989084.006 Published online by Cambridge University Press

100

Jeffrey P. Williams

echoant. Scots expressivity also exhibits vowel harmony. For many forms, there are variants in which the vowels agree in terms of overall features of articulation The forms agree with other words in the utterance, creating vowel harmony across word boundaries, as shown in the examples that follow: (53a)

currie-wurrie – N, A violent dispute (probably currie (intensifier) plus wurr (wirr) growl).

(53b)

gurry-wurry – N, Dogfight, wrangle. ADJ, Snarling, growling.

(53c)

wirr – N, A dog’s growl. 2. An angry answer, a fit of bad temper, wrath, roughness. 3. A diminutive, peevish, person. V, To growl like a dog, to fret, to whine. INT, To call dogs to cause them to fight.

(54a)

holus-bolus – ADV, Completely, all at once.

(54b)

hollis-bollis – variant of holus-bolus.

Example (55b) contradicts some of the assumptions that have been made about expressives, most notably that they cannot undergo further morphological operations. In this case, a portion of the expressive can be completely reduplicated to alter the meaning of the complete expressive collocation further. (55a)

hoot-toot – INT, An exclamation of strong dissatisfaction or jocular contradiction.

(55b)

hoot-toot-toot – INT, An exclamation of annoyance.

Echo words can even be created out of onomatopoeic forms in Scots. Example (56) provides such a brief case study of the process. (56a)

mush – N, Negative – not a whisper, not a sound – onomatopoeic.

(56b)

cushle-mushle – N, A whispering, muttering (echo word formation of mushle).

4.4

Conclusion

This chapter has presented primary data on echo word forms in the Scots language that have not been introduced in the literature previously. The contribution is data-based to provide the wherewithal for others to provide more detailed and sophisticated analyses. The belief that the languages of Europe lack expressivity is contradicted by copious proof of the role of echo word morphology in Scots. As is demonstrated by the other contributors to this volume, the languages of Europe may be grouped with the languages of Asia into a larger category of Eurasian languages that make extensive use of echo word morphology to convey expressivity. While more data certainly needs to be collected and analyzed, we can for now correct the misconception about the existence of expressivity in European languages.

https://doi.org/10.1017/9781108989084.006 Published online by Cambridge University Press

Expressivity in Scots

101

References Akita, Kimi. (2015). Sound symbolism. In Östman, Jan-Ola & Verschueren, Jef (eds.) Handbook of Pragmatics. Amsterdam: John Benjamins, 1–24. Abbi, Anvita. (1992). Reduplication in South Asian Languages: An areal, typological, and historical study. New Delhi: South Asia Books. Akita, Kimi & Pardeshi, Prashant (eds.) (2019). Ideophones, Mimetics and Expressives. Amsterdam: John Benjamins. Chatterji, S. K. (1926). Origin and Development of the Bengali Language. Calcutta: Calcutta University Press. Dingemanse, Mark. (2012). Advances in the cross-linguistic study of ideophones. Language and linguistics compass 6(10), 654–72. DOI: https://doi.org/10.1002/ lnc3.361 (2015). Ideophones and reduplication: Depiction, description, and the interpretation of repeated talk in discourse. Studies in language 39(4), 946–70. DOI: https://doi .org/10.1075/sl.39.4.05din (2018). Redrawing the margins of language: Lessons from research on ideophones. Glossa: A journal of general linguistics 3(1), 4. DOI: https://doi.org/10.5334/gjgl .444 Emeneau, M. B. (1938). An echo-word motif in Dravidian folk-tales. Journal of the American Oriental Society 58(4): 553. Gutzmann, Daniel. (2019). The Grammar of Expressivity. Oxford: Oxford University Press. Haiman, John. (2019). Ideophones and the Evolution of Language. Cambridge: Cambridge University Press. Ibarretxe-Antuñano, Iraide. (2017). Basque ideophones from a typological perspective. Canadian Journal of Linguistics 62(2), 196–220. DOI: https://doi.org/10.1017/cnj .2017.8 Inkelas, Sharon & Zoll, Cheryl. (2005). Reduplication: Doubling in morphology. Cambridge: Cambridge University Press. Iwasaki, Noriko, Sells, Peter & Akita, Kimi (eds.) (2017). The Grammar of Japanese Mimetics: Perspectives from structure, acquisition, and translation. Abingdon: Routledge. McCarthy, John J., Kimper, Wendell & Mullin, Kevin. (2012). Reduplication in Harmonic Serialism. Morphology 22, 173–232. Mattiello, Elisa. (2013). Extra-Grammatical Morphology in English. Berlin: Walter de Gruyter. Nuckolls, Janis. (2003). To be or not to be ideophonically impoverished. Texas Linguistics Forum 47: 131–42. Robinson, Mairi (ed.) (1985). Concise Scots Dictionary. Aberdeen: Aberdeen University Press. Scripture, Edward Wheeler. (1902). Elements of Experimental Phonetics. New York: Scribners. Voeltz, F. K. Erhard & Kilian-Hatz, Christa (eds.) (2001). Ideophones. Amsterdam: John Benjamins. DOI: https://doi.org/10.1075/tsl.44 Williams, Jeffrey P. (2004). Ecky-becky: Evidence of Scots echo word morphology in Barbadian English. Journal of Pidgin and Creole Languages 8: 163–70.

https://doi.org/10.1017/9781108989084.006 Published online by Cambridge University Press

102

Jeffrey P. Williams

(ed.) (2021). Expressive Morphology in the Languages of South Asia. Abingdon: Routledge. Williams, Jeffrey P. & Siu, Lap. (2014). The aesthetics of Jarai echo morphology. In Williams, J. P. (ed.) The Aesthetics of Grammar: Sound and meaning in the languages of Mainland Southeast Asia. Cambridge: Cambridge University Press, 191–206. Yip, Moira (1998). Identity avoidance in phonology and morphology. In Lapointe, Steven G. & Brentari, Diane K. (eds.) Morphology and Its Relation to Phonology and Syntax. Stanford, CA: CSLI Publications, 216–46. Zwicky, Arnold & Pullum, Geoffrey (1987). Plain morphology and expressive morphology. In Proceedings of the thirteenth annual meeting of the Berkeley Linguistics Society, 330–40.

https://doi.org/10.1017/9781108989084.006 Published online by Cambridge University Press

5

Reduplication as expressive morphology in German Gerrit Kentner*

5.1

Introduction

The term ‘expressive’ refers to a varied class of semiotic phenomena with which interlocutors convey (in often depictive rather than descriptive terms) their sensory or aesthetic experience, and with it evaluative, emotive, and attitudinal content. The exponents of the expressive meaning might be entirely non-linguistic (e.g., facial expressions of joy, awe, or disgust), para-verbal (e.g., high-pitched tone of voice signaling endearment), or they might be genuinely linguistic items (e.g., an ideophonic interjection like German schwups, which roughly corresponds to Eng. ‘hey presto!’, or the diminutive/ evaluative morphology in German Hans.i.lein – Hans.dim.dim, Eng. ‘dear little Hans’). It is perhaps because expressive phenomena are so multifaceted and transgress the boundaries of language that even the expressive exemplars that make use of genuinely linguistic means are often deemed outside the realm of grammar proper, or at least distinct from ‘plainer’, more mundane linguistic constructions. This conception is most evident in the (by now traditional) contradistinction of plain versus expressive morphology as expounded by Zwicky and Pullum (1987). In this chapter, I address this contradistinction by taking recourse to investigations of reduplication in German. Reduplication is commonly understood as a morphological process in which a stem (or base) is extended by a copy of (a part of ) its segmental material; the copied portion, also known as the reduplicant (Rubino 2005), is often prosodically circumscribed, that is, it has a more or less pre-determined shape corresponding to, for example, a syllable, a phonological foot, or a phonological word. Reduplication thus differs from word formation processes in which segmentally specified morphs are concatenated. German has various types of reduplication, chief among them rhyme reduplication (Schickimicki ‘posh person’ < schick ‘posh’), ablaut reduplication (Mischmasch ‘jumble’ < misch ‘to mix’), and full or total * Goethe-University, Frankfurt.

103

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

104

Gerrit Kentner

reduplication (Kaffeekaffee ‘coffee coffee’ = real coffee/standard variety coffee). However, like the elusive role of expressives in the linguistic system in general, the status of reduplication as a regular (and productive) morphological process in German is contested. There are various reasons for this: first, in German, reduplication is mostly a phenomenon of colloquial and familiar language use and thus evades the norms that characterize the written standard. Hence, reduplication lies outside the focus of scholars who, for convenience or other reasons, use written sources as their object of study. Second, as noted above, there are various kinds of reduplicative constructions in the German language which cannot be captured with a single grammatical analysis. Third, there are patterns that may look like reduplication but, on closer inspection, turn out to be the product of quite different morphological or syntactic processes. The effect of this state of affairs seems to be utter confusion, culminating in the curt and dismissive statement that, as far as German is concerned, reduplication ‘can hardly be dealt with systematically’ (Barz 2015: 2407). In the following, I survey recent research into the various reduplicative constructions in German, and, with the aim of systematizing them, describe their morphophonology and their use conditions. In the course of this endeavour, I also discuss the expressive functions that (various kinds of ) reduplication may have. In order to set the stage, I first revisit the iconic meaning potential ascribed to reduplication in general and across languages (§5.2). Two expressive phenomena in which reduplication features prominently, namely diminutives and ideophones, will be discussed in this context as well. In §5.3, the morpho-phonology and the use conditions of the various types of reduplication in German are discussed. I suggest that the specific use conditions and the poetic form together give rise to the expressive function of reduplication in German (§5.4). Against this background, the contradistinction between plain and expressive morphology is taken up and critically discussed in §5.5. 5.2

Iconic meaning potential of reduplication

In cross-linguistic studies on reduplication, researchers have often remarked on the potentially iconic relationship between reduplicative form and meaning (‘more of the same form stands for more of the same meaning’, Kouwenberg & LaCharité 2005: 534). An iconic form–meaning relationship is illustrated by cases in which reduplication encodes concepts such as plurality (gula ‘sugar’ – gula-gula ‘sweets’, Indonesian), iteration (lat’ ‘to lie’ – la-lat’ ‘to lie a lot’, Trumai), intensification (mavi ‘blue’ – masmavi ‘azure’, Turkish), or augmentation (ngaru ‘wave’ – ngaru-ngaru, ‘large wave’, Maori). However, a simple iconic relationship is difficult to maintain, as reduplication appears to

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

Reduplication as expressive morphology in German

105

be used in a seemingly anti-iconic fashion in many languages, for example, for approximation and attenuation (yala ‘yellow’ – yala-yala ‘yellowish’, Jamaican Creole; maji ‘wet’ – majimaji ‘somewhat wet’, Swahili) or diminution (bəkw ‘man’ – ba:bagw ‘boy’, Kwak’wala). Still other morphological uses of reduplication lack any obvious relationship between form and meaning, for example, when reduplication encodes perfect tense (currere ‘run. inf’ – cucurri ‘run.1sg.perf’, Latin). Regier (1998) and Fischer (2011) suggest that both the apparently iconic reduplications as well as the putatively anti-iconic ones can be traced back to a common source in which the iconic maxim more of the same form stands for more of the same meaning holds. Specifically, Fischer works out a common iconic grounding of reduplication that she ascribes to i. the concept of quantity increase and ii. to the resemblance of reduplication with iterative babbling in baby talk. As for the concept of quantity increase, Fischer distinguishes an increase in the vertical dimension (augmentation, intensification) from an increase in the horizontal dimension (plurality, iteration, distribution). An increase in the horizontal dimension entails conceptualizations like spread, scatter, and dispersion. Observing a multitude of similar items spread out on a horizontal plane makes each individual item appear relatively small and blurry. Consequently, reduplicative constructions that make use of the ‘horizontal’ meaning increase may promote secondary meaning components that involve diminution, attenuation, lack of control and, via semantic shifts, pejoration and contempt. An increase in the vertical dimension, on the other hand, promotes concepts such as prototypicality and intensification: rather than blurring the view, the vertical quantity increase brought about by reduplication implies sharpening the focus on the individual referent, thereby making it appear more vivid. Apart from encoding, however abstractly, the concept of quantity increase, reduplication, or more generally, phonological iteration, is reminiscent of child language or stuttering. The iconic relationship to baby talk affords both positive and negative connotations: syllable iterations typical of baby talk evoke concepts like smallness and cuteness, which may promote caring affection; the similarity to playful reduplicative babbling fosters ludic semantic flavours (funniness) that, as we shall see, figure prominently in German reduplication. On the other hand, baby talk is also related to naiveté, and stuttering may provoke pejoration and contempt. In sum, Fischer’s proposal ascribes a common grounding to the pluripotential expressiveness of reduplication and the multitude of sometimes antithetic meanings associated with reduplication. Across languages, reduplicative words are represented excessively in the realm of diminutives (Jurafsky 1996) and expressive ideophones (the present volume vividly attests to this observation). Diminutive markers (be they reduplicative or not) denote relative smallness of the referent of the base word

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

106

Gerrit Kentner

(e.g., the German diminutive suffix -chen in Wäldchen forest. dim ‘grove’ < Wald ‘forest, woodland’, German). In addition, diminutives can be interpreted as evaluative expressives, that is, words that convey disdain (Eng. commie as pejorative term for communist) or affection (Lizzy as hypocoristic name for Elisabeth) towards the referent. Notably, kinship terms in many languages, which sometimes entail diminutive semantics, can be considered reduplicative in that they involve consonant repetitions (mommy, daddy, sister; Ital. nonno ‘uncle’; hung. neny ‘aunt’, Bulg. baba ‘grandmother’). Dingemanse (2015) remarks on the prevalence of reduplicative forms in the ideophone inventory across languages. Ideophones are expressive words that signify sensory imagery through phonological markedness. While ideophones may well have conventionalized meanings, the descriptive content of these vocabulary items takes a back seat to the benefit of the iconically foregrounded expressive content. Like speech-accompanying gestures, ideophones do not directly take part in the propositional content of an utterance but provide additional non-at-issue meaning (see, e.g., Barnes et al. 2022). They can be thought of as rhetorical devices that, when used appropriately, render the message more affectively engaging or stimulating by making the listener imagine the sensory experience or emotional involvement of the speaker. Reduplication, or repetition in general, is a conspicuous and therefore potent phonological marker for evaluatives and ideophones, and for expressives in general. Like all lexical items, expressives make use of the phoneme inventory of their language, but in contrast to ordinary words, they sometimes have variable shapes. This holds especially for some reduplicative expressives that may, in actual use, be extended by not only one but two or more copies, sometimes with no strict upper bound (Dingemanse 2015). Speakers and listeners tolerate or even purposefully use the variability to express gradient perceptions. However, not all reduplicative ideophones are variable in this sense. In fact, many are restricted to phonological doubling. The grammarian has to distinguish different kinds of iteration and carefully delineate morphological reduplication (commonly restricted to doubling) from lexical sequencing (potentially unrestricted). This holds also for reduplicative forms in German, which will be considered in detail in the following sections. 5.3

Reduplication in German

In German, the main habitat of reduplicative words like Mama ‘mum’, larifari ‘slipshod’, Mischmasch ‘mishmash’, or schickimicki ‘fancy-shmancy’ are colloquial registers of the language. These reduplicative words are used predominantly in oral and socially close communication (or in genres that pretend social closeness, e.g., advertisements), and express a variety of expressive, affective or evaluative meanings that correspond with their colloquial registers

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

Reduplication as expressive morphology in German

107

of use, namely jocular or affectionate diminution and/or depreciation or slight disdain. Quite often, reduplicative words are associated with jocularity, playfulness, and a lack of seriousness (think of brand names for sweets like Hubbabubba, or nicknames like Jojo, a hypocoristic of the given name Johannes). Full or total reduplication (a.k.a. contrastive focus reduplication or identical constituent compounding), while also used in nonstandard registers, encodes emphasis, prototypicality, or normality (Finkbeiner 2014, Frankowsky 2022, Freywald 2015). The variety of meanings ascribed to reduplicative words is mirrored in the promiscuity regarding the phonological or morphological targets of reduplication: Word-internal repetition may target syllables and phonological feet, word stems, and even whole words. Moreover, words may be repeated to form iterative lexical sequences that resemble reduplication. Because of the variable nature and their nonstandard registers of use, repetition and reduplication have been claimed to emanate from a ludic drive rather than from orderly linguistic competence. Correspondingly, grammarians have considered reduplication in German to be not only marginal, but also irregular and non-productive (e.g., Barz 2015, Schindler 1991, Wiese 1990), ‘pre-grammatical’ (Bzdȩga 1965: 22), or ‘extra-grammatical’ (Dressler 2000). When viewed as a family of related morphological types, reduplication clearly appears to be morphologically ‘extravagant’ (Eitelmann & Haumann 2022), a prime example of expressive morphology in the sense of Zwicky & Pullum (1987). However, disregarding the superficial resemblance among the various patterns and instead focusing on each individual morphological type, it is possible to give explicit formal accounts that capture their essential grammatical features (e.g., Finkbeiner 2014, Frankowsky 2022, Freywald 2015, Kentner 2017, 2022, Wiese 1990). Viewed from this perspective, the diverse reduplicative structures are perfectly regular and by no means extragrammatical. 5.3.1

A morphological taxonomy of reduplicative constructions in German

In the most comprehensive collection of German reduplicative words, Bzdȩga (1965) amassed approximately 1,880 lemmas gleaned from a broad range of dialectal and historical strata of German. This collection attests to the diversity of reduplicative structures in German. The two biggest classes of reduplicatives in Bzdȩga’s collection are rhyming reduplication (e.g., Ilsebilse < Ilse [proper name]) and ablaut reduplication (e.g., Wirrwarr ‘jumble’ < wirr ‘chaotic’). However, their morphological status is often ambiguous, as many apparently reduplicative forms either lack a synchronically transparent morphological base (e.g., Techtelmechtel ‘fling’, neither *techtel nor *mechtel are identifiable morphemes), or involve two stems (e.g., Schnippschnapp [name of

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

108

Gerrit Kentner

a card game] < schnippen ‘to snip’, schnappen ‘to snatch’) and may thus be more properly treated as a special kind of compound. Other tokens are morphologically simplex despite their reduplicative form (e.g., Mama ‘mum’, Kuckuck ‘cuckoo’). Given the great diversity of reduplicative structures and their marginal status in morphological descriptions of German, it is necessary to identify those patterns from the diverse set that are unambiguous instances of reduplication, and to assign the dubious cases a proper place in the lexicon and/or grammar of German. In order to systematize the various morpho-phonological types, I have proposed (Kentner 2017) a taxonomy of reduplicative constructions that is determined by the degree of lexicality of the reduplicated form and by the correspondence with an identifiable morphological base, if present. The proposed taxonomy distinguishes i. (phonotactically illegal) interjections and ii. iterative syntagmas from iii. reduplicative lexical items. Iterative interjections formally resemble reduplications but they may be phonotactically illegal; their status as normal lexical items is therefore doubtful. Items in this category are the iterative syllables characterizing laughter (hihi, haha, etc.) or onomatopoeic imitations, for example, of machine gun fire (ratatata), in which there is no strict upper bound to the iteration. Iterative syntagmas transgress the boundary of a single word. They are best analysed as word repetitions, as they do not abide by the requirement for lexical integrity, that is, they can be split up (dalli dalli / dalli, los, dalli – ‘quick!’). Also, as in the case of the interjections, there is often no strict upper bound regarding the number of repetitions (hopp (hopp hopp . . .) – ‘quick!’). Among the reduplicative lexical items are those that do not have a transparent relation to any current and synchronically available morphological base. Therefore, these items do not represent morphologically productive patterns of reduplication. This holds for word-like interjections like dingdong (onomatopoetic for the sound of a doorbell), ideophonic adverbs like ratzfatz ‘in a jiffy’, nouns like Techtelmechtel ‘fling’, and loans like Bonbon ‘candy’ (French, derived from bon ‘good’) or Dumdum ‘dumdum’. Similarly, a few predicative adjectives like plemplem ‘crazy’ show reduplicative structure. Cases like these are indeed difficult to classify as their provenance is varied and often unclear. Reduplicative paronomasias employ either blending or compounding of two near-homophonic stems (e.g., schlampampen ‘to be untidy/sloppy’ < schlampen ‘to skimp’ + pampe ‘mush’; Klimperwimper ‘person blinking one’s eyelashes’ < klimpern ‘to tinkle’, Wimper ‘eyelash’). Since the reduplicative surface is not due to segmental copying, these words have to be distinguished from proper reduplication (see §5.3.2). As suggested by Benczes (2019), it is most likely the conspicuous quasi-reduplicative surface structure which supports the lexicalisation of these special compounds and promotes their continuous use.

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

Reduplication as expressive morphology in German

109

The word-like reduplicative structures just mentioned either lack any relation to an identifiable morphological base, or they are related to more than a single base. In that, they differ from four kinds of word formation that have been analysed as proper reduplication, because those can be derived from a single morpho-phonological base via copying. These cases will be presented in the following sections.

5.3.2

Proper reduplication in German

5.3.2.1 Phonological doubling A familiar pattern for nickname formation (1) can be analysed as a two-step process, involving i. truncation of a full name to a light CV syllable and ii. subsequent doubling or copying of that syllable. Doubling serves to establish wordhood (Saba Kirchner 2010) because, in German, a light syllable cannot serve as a word on its own. The pattern is productive for name formation yet heavily constrained by segmental context: generally, names with complex (2a) laryngeal (2b) and (2c), or rhotic onsets (2d) do not undergo this truncation plus doubling process. In some cases, however, source forms with complex onsets can be used, but they then require reduction to singleton onsets (Britta > Bibi). (1)

a. Jojo < Johannes ! [jo] ! [joːjo] b. Lulu < Luise ! [lu] ! [luːlu, lʊlu] c. Vivi < Viola ! [vi] ! [viːvi, vɪvi]

(2) a. *Floflo < Florian ! [flo] ! *[floːflo] b. *Ii < Ina ! [ʔi] ! *[ʔiːʔi] c. *Haha < Hartmut ! [ha] ! *[haːha] d. ?Roro < Robert ! [ro] ! ?[roːro]

Furthermore, only names with cardinal vowels allow this truncation-plusdoubling process. Apparently, syllables with diphthongs (3a) front rounded vowels (3b) and (3c), or non-low lax vowels (3d) cannot be doubled in this way because the resulting structure would feature such a vowel in a final open unstressed syllable, which is ungrammatical in German. (3)

a. b. c. d.

*Meimei < Meike ! [maɪ̯ ] ! *[maɪ̯ maɪ̯ ] *Lyly < Lydia ! [ly] ! *[lyːly] *Hoehoe < Hoeness ! [hø] ! *[høːhø] *Käkä < Käthe ! [kɛ] ! *[kɛːkɛ]

Because of the loss of segmental content due to the truncation, different source forms can yield the same hypocoristic form, which, moreover, and in contrast to the full name, is underspecified with respect to gender (Jolanda (f.), Josefa (f.), Joachim (m.), Johannes (m.) > Jojo; Lorena (f.), Lorenz (m.) >

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

110

Gerrit Kentner

Lolo). Therefore, these names are felicitously used only in familiar contexts, among close friends or family, that is, when the gender of the name is known to the interlocutors and therefore redundant and irrelevant for the identification of the person referred to, that is, when the risk of confusion is minimal. Thus, the opaque form establishes strict conditions that restrict their use to those familiar contexts in which hypocoristic forms are appropriate. In addition, the reduplicative yet short and simple disyllabic form of the name potentially conveys its hypocoristic or affectionate meaning iconically via the association to reduplicative baby talk. 5.3.2.2 Rhyme and ablaut reduplication The two biggest classes of reduplicative words in German that were identified by Bzdȩga (1965) are rhyme reduplication (4a) and ablaut reduplication (4b). (4)

a. Hinkepinke (hinken ‘hobble’), Ilsebilse (Ilse [proper name]), Schickimicki (< schick ‘posh’) b. Quitschquatsch ( [a]. In this respect, ablaut reduplication differs from other means of word formation in which the morph order is regulated independently of the phonology (see Kentner 2017 for a grammatical analysis that captures the variable ordering of base and reduplicant). While monosyllabic and disyllabic bases are equally attested, ablaut reduplication requires strict segmental restrictions with respect to the base in order to apply. Ablaut reduplication is impossible if the stem vowel of the base cannot undergo ablaut, that is, ablaut reduplication is restricted to bases with [i, ɪ, o, ɔ] or [a]. In general, rhyme and ablaut reduplications are commonly found in nonstandard registers of oral language, for example, in playful conversation, not only with children. If these words are used in written language at all, they mark the text as informal. Correspondingly, they may be found either in chat conversations, in which an immediate interaction of the interlocutors is normal, or in poetic texts. Reduplication serves to foreground or amplify

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

112

Gerrit Kentner

expressive content. In the case of ideophonic or onomatopoetic stems (e.g., schwappen ‘to swash, to slosh’, platschen ‘to splash’), expressive meaning may already be part of the stem. Additionally, speakers use reduplication to depict their perspective on the referent, event, or action encoded by the stem. This perspective is bound to the utterance situation (“non-displaceable” in the words of Potts 2007), and oscillates between the poles of diminution, affection, lack of seriousness, and jocular pejoration. In that these words highlight a speaker-specific and utterance-specific perspective, they are akin to other useconditional expressions such as ideophones (Dingemanse 2018) or noninflectional constructions (Bücking & Rau 2013). 5.3.2.3 Total reduplication or identical constituent compounding As with English and several other languages, modern German features total reduplication in the form of identical constituent compounding (ICC, see Hohenhaus 2004, Finkbeiner 2014, Frankowsky 2022, Freywald 2015), a.k.a. contrastive focus reduplication (Ghomeshi et al. 2004). In this morphological pattern, word stems are doubled/reduplicated to form new words. (5)

a. Nimmst Du Basmatireis oder einfach Reis-Reis? ‘Do you take basmati rice or just rice-rice’ (i.e., prototypical rice, standard variety rice) b. Was meinst Du mit ‘jetzt’ – jetzt-jetzt oder in zwei Minuten? ‘What do you mean by ‘now’ – now-now or in two minutes?’ c. Der Typ ist echt schlau – nicht nur gewieft, sondern schlau-schlau. ‘This guy is really smart – not just slick, but smart-smart.’

As the term suggests, contrastive focus reduplications are used exclusively in contrastive contexts to denote the stem’s prototypical features vis-à-vis less prototypical but contextually available alternatives. Freywald (2015) adopts the term Real-X-reduplication by Stolz et al. (2011) to emphasize the prototypicality reading these items have in German. I follow Hohenhaus (2004) and argue that these words are best analysed as a special form of endocentric compound, hence the label identical constituent compounding (ICC): as in endocentric compounds, the first, accented, part restricts the meaning of the identical head – in this case by emphasizing the head’s prototypical or ideal properties. This compound analysis may seem at odds with the traditional concept of reduplication; reduplication usually involves copying of phonological material whereas compounding involves the concatenation of morphosyntactic units, namely word stems. However, adopting a broader concept of morphosyntactic reduplication (as espoused in Inkelas & Zoll’s 2005 Morphological Doubling Theory), one may stick to the term reduplication. Ghomeshi et al. (2004) discard the compound analysis as ICC may involve parts-of-speech not typically used in compounding. In fact, it may be that this type of word formation is more promiscuous than canonical compounds with

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

Reduplication as expressive morphology in German

113

respect to the stem that is used: ICC may target adverbs that are not typically used as stems in endocentric compounds. Also, in contrast to canonical compounds, linking elements are banned in ICC. However, the promiscuity regarding the stems involved and the lack of linking elements are by no means compelling arguments against the compound analysis. Note that German makes productive use of phrasal compounds (Meibauer 2007), which generally lack linking elements. Furthermore, (phrasal) compounds may involve parts of speech in head or modifier position that are not typically found in canonical compounds (e.g., pronouns: Über-Ich; Ich-AG, Wir-Gefühl, ‘superego’, ‘You Inc.’, ‘group identity’ or adverbs im Hier und Jetzt, ‘in the here and now’). Like most phrasal compounds, but in contrast to rhyme and ablaut reduplication, ICC do not become lexicalized – instead, they are created ad hoc as they are bound to a salient contrastive context in order to be used. The semantics and pragmatics of ICC has been discussed extensively by Finkbeiner (2014) and Horn (2018). Freywald (2015) specifically remarks on the ambiguity of words like Freundfreund (lit. friend-friend), which can be translated as ‘buddy, not romantic partner’ or ‘boyfriend, not just buddy’, depending on the context. These contradictory meanings prove that the prototypicality reading of these words is semantically underspecified and crucially depends on the context of use. It is specifically because of this semantic underspecification, which can only be resolved in the actual context, that these words cannot become lexicalised. Instead, these reduplicative compounds are nonce words created ad hoc. Importantly, in order to use this kind of total reduplication felicitously, the interlocutors need to (tacitly) agree on the relevant meaning dimension that the prototypicality reading targets. Therefore, and because of the ad hoc and spontaneous use, these words are confined to familiar registers of mostly oral language use, with sufficiently acquainted interlocutors that can trust each other to read the relevant context in the same way. 5.4

The expressive meaning of reduplication: Use conditions and poetic form

The four kinds of pattern identified as proper reduplication, namely phonological doubling, rhyme reduplication, ablaut reduplication, and total reduplication share important use conditions: all of them are found in familiar, colloquial, nonstandard, playful, spontaneous, and mostly oral language. They presuppose close acquaintance of the interlocutors and are, correspondingly, characteristic of a class of registers that Koch and Österreicher (1985) label as ‘Sprache der Nähe’ (language of the close environment), which affords a high degree of expressivity. Only in this type of register, reduplications can deploy

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

114

Gerrit Kentner

their full potential as words that convey, apart from the descriptive content, the (sensory or emotional) involvement of the speaker and his or her perspective on (aspects of ) the communicative situation. In more formal registers (as, e.g., in written communication), reduplication would be inappropriate and highly marked, as these registers do not afford the same degree of expressivity. Given that the four types of reduplication are amenable to grammatical formalization, I surmise that it is not so much the alleged morphological peculiarity that sets reduplication apart from plain or ordinary morphology, but the conditions that reduplication imposes on the users. That is, if reduplication is used in an otherwise formal or standard register, it may well ‘raise chuckles’ (one of the diagnostics that Zwicky & Pullum 1987 suggest to identify expressive as opposed to plain or ordinary morphology). But this does not imply that reduplication should be deemed ‘extragrammatical’ (as has been proposed by Dressler 2000). I think that, in the case of reduplication, the ‘chuckle’-diagnostic points to a register clash rather than to grammatically deviant behaviour of these forms. This is because, once the different morpho-phonological types of reduplication are properly sorted, reduplication is not a strange or particularly unruly phenomenon from a grammatical point of view. Apart from the use conditions, the highly marked phonological form of reduplication itself likely contributes to the expressive meaning. Since repetition in general is a hallmark of poetic language use (Fabb 2015, Görner 2015, Jakobson 1960, Menninghaus et al. 2017), reduplicative morphology may be considered a condensed form of poetic language confined to single words. It will therefore likely affect the evaluation of the respective words with regard to how euphonious or cacophonous they are perceived. In this regard, we tested affective meaning dimensions, cognitive effects and the aesthetic appeal associated with reduplicative morphology in a rating experiment (Kentner et al. 2022). Participants of the experiment were asked to rate various reduplicative and non-reduplicative non-words on six bi-polar rating scales. The scales covered the key emotional dimensions of valence (with the poles labelled appreciative – depreciative) and arousal (soothing – arousing), the perceived familiarity of the patterns (familiar – strange), euphony (euphonious – cacophonous), funniness (funny – serious), and perceived size (belittling – magnifying). Stimuli were non-words with different reduplicative patterns that either conformed to the reduplicative patterns that are conventional and productive in German morphology (rhyme reduplication, e.g., ‘jaffe-maffe’, ablaut reduplication ‘liff-laff’, total reduplication ‘miffe-miffe’), or differed from them (reverse ablaut with the vowel order [a]-[i]: ‘laffe-liffe’, post-vocalic consonant alternation ‘laff-lass’). In addition, a non-reduplicative baseline condition, in which none of the segments were repeated (‘liss-maff’), was presented. The results of this rating study suggest that, in the absence of descriptive content, reduplicative forms, are associated

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

Reduplication as expressive morphology in German

115

with several meaning characteristics that are generally considered positive: they are perceived to be more appreciative, less arousing, more familiar, distinctly more euphonious and funnier, and they are perceived as more affectionately belittling when compared to the non-reduplicative baseline. Among the various reduplicative patterns, total reduplication and ablaut reduplication boost these effects to a particularly pronounced degree. This boost especially affects the scales concerning euphony, funniness, familiarity, and positive belittling (cuteness). These findings are perfectly in line with the close and familiar social environment in which reduplicative words are mostly used. Apart from the euphony, these effects are compatible with the iconic grounding that Fischer (2011) suggests for reduplication. The jocularity and lack of seriousness as well as the diminutive effect are likely a reflection of the association with child language and baby talk. Taken together, it seems that the use conditions and the poetic form of the reduplicative words conspire and jointly give rise to the expressive meaning components that may be summarized as jocularity/lack of seriousness and/or affective diminution. Reduplication is therefore well suited to produce (mocking) nicknames, as it does in the case of phonological doubling (Jojo < Jolanda [proper name]), rhyme reduplication (Sillepille < Silke [proper name]) and ablaut reduplication (Frinzfranz < Franz [proper name]). The effect of diminution and jocularity is also noticeable in the case of reduplicative ideophones like plitschplatsch ‘splish-splash’, in which the event so depicted is felt to be less severe than the corresponding non-reduplicative platsch ‘splash’. In the case of total reduplication (identical constituent compounding) the core semantic effect of reduplication is the expression of prototypicality, but the reduplicative structure insinuates ludic semantic flavours as well, in line with the colloquial use conditions of these forms. 5.5

Discussion and conclusion

Zwicky and Pullum (1987) have proposed various criteria to define expressive morphology as a type of word formation that deviates from plain or normal morphology. The first criterion is the pragmatic effect. Expressive morphology, according to Zwicky and Pullum, has ‘an expressive, playful, poetic, or simply ostentatious effect of some kind’ (1987). As we have seen, reduplication in German, in all its forms, perfectly fits this description. However, as suggested above, it is not so much the peculiar morphological structure of these words, but, on the one hand, the use conditions, and on the other hand, the resulting phonological form which engender the poetic and expressive effects. Zwicky and Pullum (1987) also comment on the fact that expressive morphology is less strict than plain morphology regarding the environment in which

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

116

Gerrit Kentner

the rules apply that produce the derived words in question. Specifically, they note that expressive morphology is promiscuous regarding the morphosyntactic input category whereas plain morphology only applies to a specific, determinate input category. As demonstrated above (5), total reduplication may apply flexibly to nouns, adjectives, and adverbs; rhyme and ablaut reduplications (4a and 4b) can be derived from proper names, adjectives, or interjections alike. Correspondingly, this criterion is fulfilled as well. Furthermore, Zwicky and Pullum suggest that, in contrast to plain morphology, which all competent speakers master, expressive morphology resembles an art form: not all speakers produce words with expressive morphology, and the outputs sometimes vary between those who do. Again, this is true for the various forms of reduplication in German. There are speakers who actively produce newly coined reduplications but others who neither produce nor accept them as proper words. Rejection, however, might not be an effect of the morphological structure but a corollary of the nonstandard register in which these words are predominantly used. Not all speakers are fluent in all registers of a language, and not all speakers accept nonstandard language as ‘proper’ language. That is, the rejection can be considered a prescriptivist value judgement rather than a grammaticality judgement. It has to be noted, though, that such value judgements are typical for many if not most linguistic phenomena that are subject to considerable inter-speaker variation, that is, not only for instances of expressive morphology. Finally, Zwicky and Pullum (1987) suggest that phenomena of expressive morphology, while certainly within the ‘sphere of human linguistic abilities’, are not amenable to normal grammatical analysis but ‘lie in a domain orthogonal to grammar’. Regarding the cases of expressive reduplication discussed above, I disagree. Granted, when lumped together into a composite class of ‘reduplication’, it is impossible to come up with a single comprehensive grammatical analysis that can derive the various forms in equal measure. It is probably for this reason that reduplication has previously been considered to be irregular, unsystematic, and non-productive. However, once the subtypes (phonological doubling, rhyme/ablaut reduplication, total reduplication) are identified, it is very well possible to formulate grammatical accounts for each of them. In these accounts, morpho-phonological constraints (e.g., the strict prosodic circumscription in the case of phonological doubling and in rhyme/ ablaut reduplication) might well override otherwise important morphosyntactic regularities (cf. the promiscuity regarding the input categories). This however, does not suffice to justify the exclusion of these words from the realm of grammar and their expulsion to an ill-defined ‘extra-grammatical’ domain (as proposed by Dressler 2000). In sum, while the various subtypes of reduplication have to be kept apart as distinct morphological classes, each with its own grammatical characteristics,

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

Reduplication as expressive morphology in German

117

they may still have comparable expressive effects in that they all give rise to a sense of playfulness or jocularity, or diminution. This effect is due to the special use conditions to which these words are subject, and to their poetic phonological form which iconically foregrounds their affective, ludic and unserious semantic flavours. References Barnes, K. R., Ebert, C., Hörnig, R., & Stender, T. (2022). The at-issue status of ideophones in German: An experimental approach. Glossa: A journal of general linguistics, 7(1). DOI: https://doi.org/10.16995/glossa.5827 Barz, I. (2015). German. In Müller, P., Ohnheiser, I., Olsen, S. & Rainer, F. (eds.) Word-Formation: An international handbook of the languages of Europe, Vol. 4: Word formation in the individual European languages. Berlin and Boston: Walter de Gruyter, 2387–410. Benczes, R. (2019). Rhyme over Reason: Phonological motivation in English. Cambridge: Cambridge University Press. Bücking, S. & Rau, J. (2013). German non-inflectional constructions as separate performatives. In Gutzmann, D. & Gärtner, H.-M. (eds.) Beyond Expressives: Explorations in use-conditional meaning. Leiden: Brill, 59–94. Bzdȩga, Andrzej (1965). Reduplizierte Wortbildung im Deutschen. Poznań: Poznańskie Towarzystwo Przyjaciół Nauk. Dingemanse, M. (2015). Ideophones and reduplication: Depiction, description, and the interpretation of repeated talk in discourse. Studies in Language 39(4), 946–70. (2018). Redrawing the margins of language: Lessons from research on ideophones. Glossa: A journal of general linguistics 3(1), 4. DOI: https://doi.org/10.5334/gjgl .444 Dressler, W. U. (2000). Extragrammatical vs. marginal phonology. In Doleschal, U. & Thornton, A. (eds.) Extragrammatical and Marginal Phonology. Munich: Lincom, 2–10. Eitelmann, M. & Haumann, D. (2022) (eds.) Extravagant Morphology: Studies in rulebending, pattern-extending and theory-challenging morphology. Amsterdam: Benjamins. Fabb, N. (2015). What is Poetry? Language and memory in the poems of the world. Cambridge: Cambridge University Press. Féry, C. (1997). Uni und Studis: die besten Wörter des Deutschen. Linguistische Berichte (172), 461–89. Finkbeiner, R. (2014). Identical constituent compounds in German. Word Structure 7 (2), 182–213. Fischer, O. (2011). Cognitive iconic grounding of reduplication in language. In Michelucci, P., Fischer, O., & Ljungberg, C. (eds.) Semblance and Signification. Amsterdam: John Benjamins. Frankowsky, M. (2022). Extravagant expressions denoting quite normal entities. In Eitelmann, M. & Haumann, D. (eds.) Extravagant Morphology: Studies in rulebending, pattern-extending and theory-challenging morphology. Amsterdam: Benjamins, 155–79.

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

118

Gerrit Kentner

Freywald, U. (2015). Total reduplication as a productive process in German. Studies in Language. 39(4), 905–45. Ghomeshi, J., Jackendoff, R., Rosen, N., & Russell, K. (2004). Contrastive focus reduplication in English (the salad-salad paper). Natural language & linguistic theory, 22(2), 307–57. Görner, R. (2015). Ästhetik der Wiederholung. Göttingen: Wallstein Verlag. Grüter, T. (2003). Hypocoristics: The case of u-formation in Bernese Swiss German. Journal of Germanic Linguistics, 15(1), 27–63. Hohenhaus, P. (2004). Identical constituent compounding: A corpus-based study. Folia Linguistica 38(3–4), 297–331. Horn, L. R. (2018). The lexical clone: Pragmatics, prototypes, productivity. In Finkbeiner, R. & Freywald, U. (eds.) Exact Repetition in Grammar and Discourse. Berlin: De Gruyter, 233–64. Inkelas, S. & Zoll, C. (2005). Reduplication: Doubling in morphology. Cambridge: Cambridge University Press. Jakobson, R. (1960). Closing statement: Linguistics and poetics. In Sebeok, T. (ed.) Style in Language. Cambridge, MA: MIT Press, 350–77. Jurafsky, D. (1996). Universal tendencies in the semantics of the diminutive. Language 72(3), 533–578. Kentner, G. (2017). On the emergence of reduplication in German morphophonology. Zeitschrift für Sprachwissenschaft 36(2), 233–77. (2022). Do not repeat: Repetition and reduplication in German revisited. In Eitelmann, M. & Haumann, D. (eds.) Extravagant Morphology. Amsterdam: Benjamins. 181–205. Kentner, G., Franz, I., & Menninghaus, W. (2022). Poetics of reduplicative word formation: Evidence from a rating and recall experiment. Language and Cognition 14(3), 333–61. DOI: https://doi.org/10.1017/langcog.2021.27 Koch, P. & Oesterreicher, W. (1985). Sprache der Nähe – Sprache der Distanz. Mündlichkeit und Schriftlichkeit im Spannungsfeld von Sprachtheorie und Sprachgeschichte. Romanistisches Jahrbuch 36(1), 15–43. Kouwenberg, S. & LaCharité, D. (2005). Less is more: Evidence from diminutive reduplication in Caribbean Creole languages. In Hurch, B. & Mattes, V. (eds.) Studies on Reduplication. Berlin: De Gruyter, 11–29. Meibauer, J. (2007). How marginal are phrasal compounds? Generalized insertion, expressivity, and I/Q-interaction. Morphology, 17(2), 233–59. Menninghaus, W., Wagner, V., Wassiliwizky, E., Jacobsen, T., & Knoop, C. A. (2017). The emotional and aesthetic powers of parallelistic diction. Poetics 63, 47–59. Potts, C. (2007). The expressive dimension. Theoretical Linguistics 33(2), 165–98. Regier, T. (1998). Reduplication and the arbitrariness of the sign. In M. A. Gernsbacher & S. J. Derry (eds.) Proceedings of the 20th Annual Conference of the Cognitive Science Society. Mahwah, NJ: Lawrence Erlbaum Associates. Rubino, C. (2005). Reduplication: Form, function and distribution. In Hurch, B. & Mattes, V. (eds.) Studies on Reduplication. Berlin: De Gruyter, 11–29. Saba Kirchner, J. (2010). Minimal reduplication. Dissertation, University of California, Santa Cruz. Schindler, W. (1991). Reduplizierende Wortbildung im Deutschen. Zeitschrift für Phonetik, Sprachwissenschaft und Kommunikationsforschung 44(1–4), 595–611.

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

Reduplication as expressive morphology in German

119

Stolz, T., Stroh, C., & Urdze, A. (2011): Total Reduplication: The areal linguistics of a potential universal. Berlin: Akademie-Verlag. Wiese, R. (1990). Über die Interaktion von Morphologie und Phonologie– Reduplikation im Deutschen. Zeitschrift für Phonetik, Sprachwissenschaft und Kommunikationsforschung 43(1–4), 603–24. (2001). Regular morphology vs. prosodic morphology? The case of truncations in German. Journal of Germanic Linguistics 13(2), 131–77. Zwicky, A. M. & Pullum, G. K. (1987). Plain morphology and expressive morphology. In Aske, J., Beery, N., Michaelis, L. & Filip, H. (eds.) Proccedings of the Thirteenth Annual Meeting of the Berkeley Linguistics Society. Berkeley, CA: Berkeley Linguistics Society, 330–40.

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.007 Published online by Cambridge University Press

Hellenic

https://doi.org/10.1017/9781108989084.008 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.008 Published online by Cambridge University Press

6

Expressivity in Modern Greek: Some morphological mechanisms for the expression of negative emotions Haritini Kallergi, Georgia Katsouda, Magdalene Konstantinidou and Anastasios Tsangalidis

6.1

Introduction

The aim of this study is to investigate aspects of expressivity in Standard Modern Greek (hereafter SMG), specifically cases in which expressivity shares the characteristic [+negative] or pejorative, in the more technical sense adopted here. Our research is focused on the level of morphology, particularly on productive word formation, both through compounding (first compound constituents with pejorative meaning, as in vromokánalo ‘filthy TV channel’ and paʎoiós ‘old/damn virus’, among others) and derivation (derivational suffixes with pejorative functions, such as ipurʝéi ‘bad ministers’, ipalilákos ‘insignificant clerk’ and fititarjó ‘a student lot’, among others). It should be noted that although the particular SMG morphological phenomena/devices have been scatteringly studied in earlier Greek literature, negative expressive meaning, that is, pejoration, has not been systematically dealt with so far. The chapter is structured as follows: §6.2 explains our theoretical starting points regarding expressivity in general and negative expressivity (pejoration) in particular. Section 6.3 presents cases of pejorative compounding, whereas §6.4 is dedicated to pejorative suffixation. In §6.5, we conclude on the factors that motivate the rise of pejorative meanings in the devices explored and in §6.6, we make some suggestions for future study. 6.2

A theoretical overview of the notion of expressivity

6.2.1

Expressives vs. descriptives

With respect to the notion of expressives, we will offer a brief overview, focusing especially on those theoretical approaches which we will be using in this chapter. Expressives are linguistic items that encode expressivity, in the sense of ‘charged information about the speaker’s emotional state’ at the time of the 123

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

124

Kallergi, Katsouda, Konstantinidou, Tsangalidis

utterance.1 They constitute part of the linguistic code, and as such their expressive meaning is invariant, being the same for all the speakers of the language.2 They are understood as reliable signals of emotionality,3 arbitrarily and by linguistic convention connected with the content they express.4 When using an expressive term/device the speaker indicates his ‘heightened emotional state’5 or his (often negative) attitude towards a situation, but without literally saying anything about that state, or naming it.6 Expressives are opposed to descriptives, that is, to linguistic items that are not used for indicating the speaker’s emotional state or attitude, but for picking out items in the world. There is a fundamental distinction between expressives and descriptives7 that demands further specification. David Kaplan (1999) in his most influential lecture on ‘the meaning of ouch and oops’ has argued that expressives (like the interjection ouch) and descriptives (like the sentence I am in pain) both convey ‘semantic information’, that is, ‘conventionalized information that’s carried by the expressions of our language’, yet ‘they convey it through different modes of expression’.8 1

2

3

4 5 6 7

8

Potts (2019: 599 and 618). For a brief discussion of ‘expressivity’ in linguistics, see Amaral (2018) and Gutzmann (2015); for a more comprehensive presentation see Konstantinidou (1997: I–III). Referring to ‘suffixes of subjective evaluation’, Stankiewicz (1968: 97) has pointed out: ‘the term “subjective” should not mislead: expressive suffixes are part of the linguistic code, and their meaning is the same for all speakers; they may, furthermore, signal the emotive meaning independently of the actual emotional state of the speaker’; see also Kaplan (1999: 18, 17) who has acknowledged ‘the conventionalized significance of the expressive side of language’, pointing out that ‘the mode of expression, descriptive vs. expressive, does not correlate with the nature of the semantic information, objective vs. subjective’; the intrinsic/conventionalized character of linguistic expressivity is also noted by Hom (2008: 421) with a characteristic formulation: ‘The word [nigger] is derogatory regardless of one’s personal associations or feelings toward African Americans’. Yet, some scholars believe that expressivity is not constant and that it depends on the features of the context of use. In that sense, expressive content is ascribed to the pragmatic dimension, for instance, to presuppositions triggered by expressives (Schlenker 2007) or to felicity conditions on the use of the sentence (Predelli 2010). For arguments against this theory see, e.g., Frigerio and Tenchini (2014), Tenchini and Frigerio (2016), Amaral (2018). Potts (2012: 2532) puts it as follows: ‘As speakers, we have strong expectations that uses of [expressives such as] damn will correlate with the speaker’s being in a heightened emotional state (or wishing to create that impression). In turn, we use it only when we are in such a state (or wish to create that impression). The total effect of these assumptions is that [an expressive such as] damn is a reliable signal of emotionality’. See also Potts (2019: 618). See Kaplan (1999: 3, 17, 18), Gutzmann (2013: 2, 3), Gutzmann (2015: 6, 14). Potts (2007: 173) following a suggestion from Bill Ladusaw; see also Potts (2019: 618). See Kaplan (1999: 16). See Kaplan (1999), Potts (2007), and also Williamson (2009: 153), who points out: ‘What is most crucial is the separation of those aspects of meaning that contribute to truth-conditions from those that do not’. For a short overview and more literature on the subject see Croom (2011: 347). Cappelen and Dever (2019: 100) have this insight short and aptly formulated: ‘“Ouch” doesn’t describe your pain – it expresses it’. See also Tenchini and Frigerio (2016: 177) who have

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

125

Nevertheless, Kaplan (1999) does not provide any further clarification on the relevant distinction. Chris Potts (2007: 168) has explained this difference in semantic terms; he has identified two classes of semantic types, that is, descriptive types of descriptive content (like dog, woman) and expressive types of expressive content (like interjections, or lexemes of the type damn, bastard, fuck), and has integrated both kinds of meaning in the framework of a two-dimensional semantic theory (Potts 2005, 2007).9 A third semantic type of mixed, that is, descriptive and expressive, content, involving lexemes such as cur (vs. dog), Kraut (vs. German), and racial or ethnic slurs has been added and analysed by other scholars (e.g., Williamson 2009, McCready 2010, and Croom 2011). To illuminate the special nature of the meaning of expressives as opposed to the meaning of descriptives, Potts (2005, 2007) connected the notion of meaning with the notion of non-context-dependent entailments or commitments and analysed descriptive or at-issue meaning as at-issue entailments and expressive meaning as conventional implicatures, a concept that was initiated by Grice (1975 [1967]). A conventional implicature is conceived as part of the meaning of an expression that (unlike conversational implications) is not contextdependent and is not contributing to the truth conditions of a sentence (i.e., being ‘not at issue’); moreover, it is speaker-oriented, that is, it represents a conventionally encoded commitment made by the speaker of the utterance. In addition, Potts (2007: 166–83) offered a set of six characteristic properties of expressive content, which are also meant to serve as defining criteria of expressives, that is, as means for distinguishing expressive items from descriptive ones.10  Independence: Expressive content contributes a dimension of meaning that is separated from the regular descriptive content (see Examples 1 and 2). Thus, one can change or remove the expressive content of a phrase without affecting its descriptive content, i.e., without affecting its truth conditions. Note that this property has been acknowledged as the most important and identifying feature of expressives (Gutzmann 2013: 49, see also Hess 2019: 195).

9 10

pointed out that ‘expressing contempt by means of a slur [is] not equivalent to describing one’s attitude of contempt’; also Konstantinidou (1997: 81–84), who has argued that there are two fundamentally different modes or ways of communicating feelings linguistically: the mode of expression of feelings and the mode of description (presentation or thematization) of them; these modes become evident by comparing, e.g., a disappointment interjection to the corresponding notion ‘disappointment’ or the proposition I’m disappointed. For a modification of Potts’s theory see Gutzmann (2011). For a discussion and a critical evaluation of the system introduced by Potts (2005, 2007) see Amaral et al. (2007), Gutzmann (2013), Amaral (2018), and Hess (2019).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

126

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(1)

That bastard Kresge is famous (Potts 2007: 168)

(2)

a. Descriptive: Kresge is famous b. Expressive: Kresge is a {bastard/bad in the speaker’s opinion}

 Nondisplaceability: Expressives predicate something of the utterance situation, that is, their meaning ‘is valid only for the utterer, at the time and place of utterance’, just like a smile or a gesture of impatience (Cruse 1986: 272 quoted by Potts 2007: 169). So, ‘one cannot intend ouch or oops to be valid in the past’, to use an example offered by Amaral (2018: 328), inasmuch as expressive content cannot be displaced at a time different from the time of speech. In other words, as is also suggested by Amaral (2018: 328), expressives are to be conceived as indexical elements.  Perspective dependence: Expressive content is evaluated from a particular perspective. In general, the perspective is the speaker’s, but there can be deviations if conditions are right. A speaker’s expressives indicate that she is in a heightened emotional state, whereas the communicated emotion is directed either to a specific individual, or at some specific feature of the current state of affairs or it is just general, undirected emotion.  Descriptive ineffability: Speakers are never fully satisfied when they paraphrase expressive content using descriptive, that is, non-expressive, terms. When pressed for definitions, they resort to illustrating where the words would be appropriately used.  Immediacy: Like performatives, expressives achieve their intended act simply by being uttered; they do not offer content so much as inflict it. So, the act of uttering an expressive morpheme, such as that bastard Kresge in (1), is sufficient for conveying its content, that is, for expressing hostility towards Kresge.  Repeatability: If a speaker repeatedly uses an expressive item, the effect is generally one of strengthening the emotive content, rather than one of redundancy. Gutzmann (2013: 38–47) offered a refinement of the aforementioned criteria suggested by Potts. For example, to the property of independence, he added the following sub-properties (a)–(d), which function as diagnostics for the expressive (non-truth-conditional) content. Expressive meaning (‘use-conditional’ meaning in Gutzmann’s broader terminology),11 unlike descriptive meaning: (a) (b) (c) (d) 11

cannot be negated by ordinary negation cannot be denied directly in dialogue is not part of what is questioned by an interrogative does not affect the descriptive content, if not fulfilled.

See Gutzmann (2013) and Gutzmann and McCready (2016: 76–77).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

127

On Potts’s descriptive ineffability of expressive content, Gutzmann, assuming, for the sake of the argument, that (3b) is a perfect paraphrase of (3a), pointed out that even if they both (3a), (3b) mean the same, they express this differently. (3)

a. Oops! b. I just observed a minor mishap.

In this sense, he argued for a reformulation/modification of Potts’s descriptive ineffability, focusing not so much on the impossibility of paraphrasing use-conditional (‘expressive’ in our terminology) content in a satisfactory way, but on the fact that it is impossible to paraphrase use-conditional content using only truth-conditional expressions without changing the modus of expression. On the other hand, Meibauer (2013: 36) has added to the six properties of expressive meaning another one, namely, the degree of expressivity, pointing out that some expressive items are more powerful than others. But why are expressive meanings characterized by the properties identified by Potts and others (especially by the first three ones regarded as basic)? Hess (2019), looking for a ‘philosophically interesting’ answer to this crucial question, pointed out the special nature of expressive content. He argued that expressives are characterized by the three main properties (independence, non-displaceability and perspective dependence) because their meaning ‘is different in kind’ from descriptive meanings (Hess 2019: 208). Based on the observations that ‘expressive meanings seem to directly involve the speaker (her states, emotions, or attitudes) rather than just abstract (e.g., truthconditional) contents, and that the utterer of an expressive is responsible for the choice of loaded, often taboo vocabulary’ (2019: 193), Hess offered an alternative, more pragmatically oriented account of expressives. He argued that what characterizes expressives as a lexical class is that they always raise the issue of a speaker’s expressive commitment. Unlike assertoric commitments that are commitments to the truth of asserted contents, expressive commitments are commitments to the applicability or appropriateness of a certain term of expression. In this sense, by choosing an expressive the speaker undertakes an expressive commitment; he ‘signals that it adequately reflects his emotional state’ (Hess 2019: 210). As Hess puts it ‘expressive meanings are commitments to the appropriateness of strongly charged (often vulgar or taboo) vocabulary – which, in turn, can signal a speaker’s heightened emotional state, negative attitude etc.’ (Hess 2019: 193). A different approach to the problem, based on semiotic terms, is taken by Konstantinidou (1997). Starting from the semiotic distinction between symbolic and indexical signs, she identifies two fundamental different types of linguistic signs; symbol-signs that arbitrarily and by linguistic convention stand for

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

128

Kallergi, Katsouda, Konstantinidou, Tsangalidis

concepts,12 and index-signs (which are, in fact, of a symptomatic nature)13 that arbitrarily and by linguistic convention stand for emotions,14 emotional states, or emotional attitudes (of the source, i.e., of the person who is using them). A crucial point of this theory is the idea of the conventional semiotic connection of linguistic items with something being by its nature fundamentally different from concepts,15 that is, emotions or emotional attitudes.16 (Pure) descriptives, according to this theoretical approach, belong to the category of symbol-signs, having by convention concepts as their significata/significations, whereas (pure) expressives to the category of index-signs, having by convention emotions/ emotional attitudes as their significata/significations. Items belonging to a mixed semantic type and having a double semiotic face, symbolic and indexical at the same time, are also identified within the framework of this approach.17 12

13 14

15

16

17

Some additional clarifications: The thing signified by a linguistic symbol-sign is a conceptual magnitude, a certain concept. A word, like dog is a linguistic symbol-sign conventionally connected with the concept of dog, i.e., with that conceptual extra-linguistic element that helps humans (and animals) to identify dogs as a world object, to distinguish them from other world objects, e.g., cats, and other animals or objects. The concept which is semiotically connected with a linguistic sign represents its descriptive or truth-conditional meaning, the at-issue content of that sign in other terminology. By using linguistic signs of this kind, the speaker points out to world objects in an indirect way, i.e., through the mediation of concepts. A symbol-sign in virtue of its semiotic connection with a concrete concept, activates, when it is used, the relevant concept in the mind of the addressee, and the relevant world item is picked out. See for details Konstantinidou (1997). See for details Konstantinidou (1997: 90–91, 94–95, 128). It should be noted that the thing signified by a linguistic index-sign is not a conceptual but an emotional magnitude; it is a certain emotion, not the concept of an emotion. An interjection, like ouch, is a linguistic index-sign conventionally connected with an emotional state, with a feeling, in fact that feeling that one experiences when he is in pain. The emotion, which is arbitrarily, by linguistic convention and in semiotic terms connected with a linguistic sign represents its expressive (emotive) meaning, the non-truth-conditional meaning aspect of that sign in other terminology. By using a linguistic sign of that type, the speaker points out in a direct way (i.e., without the mediation of concept) to a certain emotional state that is being experienced by him hic et nunc. He indicates (and thereby he communicates in a conventionalized way) the presence of a certain emotion in him. We may find similar functions in signs of the type of a gesture, e.g., of a respect gesture that indicates feelings of respect on behalf of the sender towards the addressee. Compare other types of index-signs, like fever that indicates illness (but not by convention), or smoke, that indicates fire (yet, not by convention but in a natural way). See Konstantinidou (1997) for details. See in this respect Bühler’s intuition that ‘the axiom of language theory that all language signs must be symbols of the same kind is too narrow’ (1965 [1934]: 105), as well as Williamson’s statement that ‘the meaning of a word is [. . .] not exhausted by the concept that it expresses, and the study of meanings cannot be subsumed under the study of concepts. An adequate theory of concepts must resolve such questions’ (Williamson 2009: 155). The concept vs. emotion-attitude distinction that underlies this approach corresponds, mutatis mutandis, to the three-fold division of men’s souls into intellect (noûs), nobler affections (thumós), and the appetites or passions (epithumetikón) defined by Plato (Republic, Book IV), which constitutes a philosophical basis for the proposed semiotic differentiations. Linguistic signs of this mixed type, such as Boche (vs. German), are signs with double semiotic face, symbolic and indexical; they are simultaneously symbol- and index-signs, i.e., they are arbitrarily and by linguistic convention semiotically connected: (a) with a concrete concept

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

129

The special semiotic character of expressive items, that is, their correlation to emotion and not to cognition, may serve as an interpretative key for the majority of the distinctive properties of expressives, but also for the correlation/relationship/affinity between expressives and certain types of non-verbal communication, such as gestures, as has been observed by many scholars.18 It may also provide a framework for treating the neurobiological evidence based on research on aphasia that expressive language is located in a different area of the brain from other types of content, inasmuch as aphasia patients with damage to the brain’s left hemisphere are nonetheless able to curse, that is, to make use of non-propositional language (Jay 2000, Potts 2007: 177, Amaral 2018: 328). Be that as it may, it should be admitted that thanks to the seminal works of Kaplan (1999) and Potts (2005, 2007), followed by Gutzmann (2013, 2015) and other scholars, expressive linguistic elements ceased to be viewed as subjective (and therefore as not amenable to objective investigation) and entered the scope of systematic semantic analysis and description. So, in the last decades there have been a lot of works on linguistic expressivity focusing especially on expressive vocabulary, for example, pejoratives, taboo vocabulary (see only indicatively Cappelen & Dever 2019, Finkbeiner et al. 2016a, Croom 2013, Hom 2010, etc.), and expressive morphology (‘evaluative’ in another terminology),19 for example, affixation, compounding, reduplication (see only indicatively Urdze 2018, Grandi & Körtvélyessy 2015a, Körtvélyessy 2015, Meibauer 2013, Fortin 2011, Dressler & Merlini Barbaresi 1994, Zwicky & Pullum 1987), providing expressive data from many different language systems worldwide (see only indicatively Williams 2013, Williams 2021). 6.2.2

Negative expressives or pejoratives

As has been discussed in the previous section, we are interested in linguistic expressions that reflect the speaker’s emotions, dispositions, and attitudes, all of which are expressed, but not explicitly named. We are also specifically

18

19

(e.g., ‘German’) and (b) with a concrete feeling, emotion, or emotional attitude (e.g., negative feeling/attitude, pejoration) towards the notions signified by the relevant concept (e.g., ‘being a German’/’Germanhood’), so that the one who makes use of them is able: (a) to point to the relevant notions through activation of the relevant concept and (b) to express (indicate/communicate) directly his emotional attitude towards the relevant notions. Following Hornsby, Croom (2011: 344 fn 2) claims: ‘It may be that expressives serve as verbal ‘‘gestures’’ in that ‘‘the commitments incurred by someone who makes the gesture are commitments to targeted emotional attitudes, and not necessarily to thoughts’’ (Hornsby, 2001: 140–41)’; see also Gutzmann (2013: 51), Cruse (1986: 272 quoted by Potts 2007: 169, and Amaral 2018: 329) and Konstantinidou (1997: 91, 117–18). For the seemingly broader notion of evaluative morphology see the discussion in §6.2.2.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

130

Kallergi, Katsouda, Konstantinidou, Tsangalidis

dealing with derogatory attitudes, hence with expressions that, in the words of Hom (2010: 164), ‘conventionally convey negative, emotional content beyond the truth-conditional content that they are normally taken to encode (if any)’. In this sense, we are exploring negative expressivity or pejoration. Pejorative language can be characterized as expressive or evaluative, but we prefer the former term, because, following Kaplan (1999), we want to focus not on the kind of content being conveyed through pejoration, that is, negative evaluation (which incidentally can also be conveyed descriptively), but in the way in which that content is carried, which is the way of expression (not the way of description). On the other hand, evaluative morphology (Scalise 1984, Dressler & Merlini Barbaresi 1994, Grandi & Körtvélyessy 2015a, 2015b) seems to refer to a wider range of phenomena and constructions, including quantitative diminution, quantitative augmentation, age variation, approximation/reduction/attenuation, intensification, endearment, hypocorism, expression of social position, contempt, authenticity/prototypicality, the evaluativity of which is not defined explicitly or in a commonly accepted way, and the scholar’s intuition is ‘the only criterion governing discussion and interpretation of data’ (Grandi & Körtvélyessy 2015b: 8). Moreover, evaluativity as judgement may be narrower than expressivity; for example, Fortin (2011: 1, fn 1) argues that in recent studies evaluativity is discussed as a characteristic of ‘degree constructions that are evaluated against (and exceed) a contextuallyspecified standard’. Of course, evaluation is strongly linked to the expression of emotions. As Finkbeiner et al. (2016b: 8) suggest, ‘pejorative expressions categorize people [and] categorizing people often goes together with a derogatory attitude against them’. That is, it is plausible to think that negative emotions motivate negative attitudes, which are, in turn, the basis of negative evaluation and resulting verbal abuse. For this reason, the term evaluative is not avoided altogether, but is used in mention of the work of scholars who employ it in order to discuss phenomena very similar to pejorative expressivity, in particular pejorative word-formational processes.20 But what does pejorative mean, apart from ‘negative’? There seem to be cross-linguistic tendencies of expressive morphemes to convey certain meanings along several emotive axes. One such axis, relevant to pejoratives (as negative expressives, see §6.2.1) is (UN)PLEASANTNESS which, according to some early theorists (see Hübler 1998 and references therein), includes ANGER. Another emotive dimension is (DIS)APPROVAL or, to our interest, REJECTION, which, in a reductionist mode, includes CONTEMPT and DISGUST. However, according to Dressler and Merlini Barbaresi (1994: 30), emotions are far too subtle and complicated to be satisfactorily 20

Note, also, that there are some scholars who may use the two terms interchangeably (e.g., Hom 2010, Dammel & Quindt 2016).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

131

represented in verbal language. For this reason, we make use of the PEJ/[+/–PEJ] index, as a ‘lexical marker of pejoration’ (Finkbeiner et al. 2016b: 7),21 which reflects this expressive component typically additional to and distinct from the truth-conditional meaning of the expressions analysed. Apart from a ‘feeling’ or ‘tone’ of contempt, deprecation, rejection, dismissal, underestimation or other, [+PEJ] may include the ironic disposition of the speaker, which is ‘often used to convey pejorative speaker attitudes’ (Finkbeiner et al. 2016b: 11), as well as a mocking attitude, which may intend to ridicule the referent of the word formation. According to Finkbeiner et al. (2016a), pejoration can be traced and analysed at all levels of grammar, including phonology (pejorative prosody), morphology (e.g., derivation and compounding), and the lexicon, where we find word pairs like dog-cur as well as what Hom (2010) refers to as ‘pejorative words’, that is, slurs (words expressing discrimination against ethnic groups or social minorities), insults (e.g., bastard) and swear words (e.g., damn, fucking). Pejoration may also be expressed by (morpho)syntactic means, such as echo-constructions or schm-reduplication (see, e.g., Kallergi & Konstantinidou 2018 for Modern Greek) and by syntactic constructions, such as, for example, the Ich/Du-NP! construction in German and other languages (Finkbeiner et al. 2016a). Finally, pejoration is analysed at the level of semantics and pragmatics, exhibiting a tradition of competing theories regarding its locus and nature of meaning. The current descriptive study of pejorative word-formation mechanisms in SMG is restricted to compounding with first compound constituents appearing as swear words in SMG and to suffixation with a collection of nominalizing derivational suffixes, including diminutives, augmentatives, and collectives. As discussed by Hom (2010: 164–5), pejoratives relate to taboo language and the degree of formality of the context, but may have variable ‘expressive force’ themselves, that is, they may vary in terms of strength of their derogatory force, as well as the dimension of derogation; some of them function only as amplifiers of negative emotions or attitudes (like fucking in, e.g., John is a fucking [good/bad] lawyer, Hom 2010: 165). Indeed, cases examined here may be milder than slurs or swear words in different degrees (in Fortin’s words they exhibit less ‘semantic intensity’, Fortin 2011: 34), and some cases may only exhibit an amplifying/intensifying role. However, all in all, we generally agree with the criteria or the characteristics of pejoratives, as stated by Dammel and Quindt (2016: 43) for ‘evaluative derivation’, that is, that:

21

Regarding the PEJ/(PEJ) index, see Dressler and Merlini Barbaresi (1994), Fortin (2011), Hom and May (2013: 298) quoted by Finkbeiner et al. (2016b: 7), Gutzmann and McCready (2016: 88), Dammel and Quindt (2016: 41).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

132

Kallergi, Katsouda, Konstantinidou, Tsangalidis

i. the pejorative (sub)functions of the devices analysed here are productive (and possibly also type-frequent); ii. the pejorative functions may arise with expressively/semantically neutral bases; iii. the pejorative functions may occur independently of context (in most of the cases discussed); iv. the pejorative constructions stand in systematic opposition to nonevaluative (non-expressive in our terms) patterns. This practically means that, for the majority of our examples of pejorative functions, there are neutral counterparts (whether derivational or not), and the difference is evident at an expressive level (as far as the notorious descriptive ineffability of expressives allows for meaning equivalence, see §6.2.1). Finally, the majority of the morphological mechanisms dealt with here has been discussed in the Modern Greek literature, but relatively unsystematically, that is, either scatteringly in various case-studies (of individual constructions, mentioned in the following sections) or from different theoretical viewpoints and for different aims. Recent works that are based on extensive data collection and include data on pejoratives are, indicatively, Kamilaki et al. (2016) (on taboo language) and Christopoulou (2016) (on marginal vocabulary) and, with narrower scope, Efthymiou (1999, 2013, 2015, 2019) and (Melissaropoulou (2015) (both on evaluative affixation). Most of these works are written in Greek.22 The Modern Greek data used in this study come from dictionaries and grammars of the language as well as from personal records of the authors: these include data from everyday conversations in which we were present, data from computer-mediated communication collected through simple Google searches, and records of informal language in literature, theatre, and the media. 6.3

Expressive compounding in SMG – Pejoration by first constituents

Compounding is a particularly productive process in Modern Greek.23 According to Ralli (2013: chapter 2), typical Greek compounds are structures which are different from phrases on phonological, structural, and semantic grounds. They are one-word units which bear one stress (they are phonological words). Their two constituents are typically connected with the semantically empty linking vowel -o-. The first constituent is usually a stem – that is, the 22

23

The same is true of earlier works, such as Setatos (1971), Babiniotis (1969), AnastasiadisSymeonidis (1986), and Minas (1978). The lists of both recent and earlier works are far from exhaustive. For details on Greek compounding see Anastasiadis-Symeonidis (1983, 1986), Ralli (1992, 2013), Revithiadou (1997).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

133

part of a word which is stripped of its inflectional ending. As for the morphological category of the second constituent, this may vary from a stem (e.g., kalo-cér-i ‘summer’ < stems kal(ós) ‘good’ and cer(ós) ‘weather’) to a word (i.e., a full-word form, e.g., patat-o-ceftés ‘potato roll’ < stem patát(a) ‘potato’ and noun ceftés ‘roll’). The linking vowel -o- (glossed CM: Compound Marker) is systematically present, but when the second constituent begins with a vowel stronger than /o/, that is, an /a/ or /e/, it is deleted. Yet, compounds with a loose structure realize -o- against the norm, ‘for ensuring internal cohesion’ (Ralli 2013: 149).24 In this section we deal with a most common strategy in SMG expressive morphology, that is, the strategy of forming expressive one-word compounds displaying the structure in (4): (4)

[Constituent1]-[Compound Marker]-[Constituent2]

whereby [Constituent1] is a non-head element that encodes expressivity and is typically employed to form compounds with expressive meaning, used by Greek speakers in order to derogate, to express contempt towards the referent denoted by the second constituent (the right-hand member, usually the head)25 of the compound. It is not our aim to exhaust all relevant cases, but to focus on some representative ones – those that are used relatively often and are the most productive, and in that sense, are often to be found as distinct entries in SMG dictionaries. By convention and for reasons of economy, we will be using the term ‘first constituent’ in a rather loose way to include the linking vowel -o-, apart from the stem: for example, first constituent vlaxo- consists in the stem vlax- (from the word vláxos ‘Aromanian/Vlach’) and the linking vowel -o-. It is worth noting that in SMG there are also first compound constituents that encode positive expressivity, that is, meliorative meaning ([+MEL]), such as kalo- ‘good’ (e.g., kal-o-pandreménos good-CM-married ‘well married’; kal-o-faɣás good-CM-eater ‘gourmet, foodie’), but also expressive second compound constituents (usually heads) that may be associated with negative expressivity, that is, pejoration, such as -vlaxos, literally ‘Aromanian/Vlach’ (e.g., bastun-ó-vlaxos stick-CM-Aromanian.MET.PEJ ‘hillbilly’; burdz-ó-vlaxos fortress-CM-Aromanian.MET.PEJ ‘redneck’)26 and -voló27 < Ancient Greek -βολῶ [bolô], stem of the verb βάλλω [bál:ō] ‘throw’ (e.g., ʝen-o-voló give.birth-CM-throw.1SG.PR.PEJ ‘I give birth to

24 25 27

For details on the exceptional presence or absence of -o- see Ralli (2013: chapter 4). 26 Ralli (2013: 126). For these derogatory terms, see Efthymiou (2019: 30, 35). For a brief account of the negative expressive content of the second constituent -voló, see Efthymiou (2019: 164).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

134

Kallergi, Katsouda, Konstantinidou, Tsangalidis

one child after the other [+PEJ]’; ʝirn-o-voló ramble-CM-throw.1SG.PR.PEJ ‘I ramble here and there [+PEJ]’; ksern-o-voló throw.up-CMthrow.1SG.PR.PEJ ‘I vomit repeatedly [+PEJ]’).28 Expressive first compound constituents have not so far been studied systematically from the point of view of their expressive function, though there are occasionally brief reports on some of them in the framework of general studies on Modern Greek slang and taboo vocabulary or expressive morphology, especially expressive affixation (see, e.g., Christopoulou 2016, Christopoulou et al. 2017). In this section, we present and analyse five SMG morphological elements contributing [+negative]/(pejorative) expressivity, that is, the first constituents vlaxo- ‘Aromanian/Vlach.MET’, ʝifto- ‘Gypsy’, vromo- ‘stink/dirt’, skato- ‘shit’, and paʎo- ‘old’.29 From a semantic point of view all these items – except for paʎo-, which constitutes a special case that deserves a special interpretive treatment – are derived from lexemes that belong to the non-neutral vocabulary of SMG,30 namely they are [+negative] marked terms, especially when they are used metaphorically. Based on the analysis and the insights of Modern Greek negative terms provided by Efthymiou et al. (2013) we will use the term ‘negative’ in a rather broad sense, so as to include not only taboo and insulting terms, that is, terms that in most modern dictionaries are marked by usage labels, like ‘pejorative/derogatory’, ‘offensive’, ‘derisory’, ‘disparaging’, or even ‘colloquial’ or ‘informal’, but also terms whose negativity is expressed in the lemma’s definition (e.g., by including expressions, such as ‘offensive characterization’, ‘derogatory characterization’ or ‘negative characterization’). Note that according to Efthymiou et al. (2013), terms labelled as negative in Modern Greek dictionaries may be distributed on the basis of their semantics in categories such as nationality, mental abilities, appearance, behaviour, political or ideological beliefs, sexual orientation, derogatory words for women or men, sex, bodily effluvia, age, disease, religion, and so on. Regarding the morphological status of the first constituents under investigation it should be noted that in the recent Greek literature (see Christopoulou 28 29

30

Note that compounds with meliorative second constituents are to our knowledge not traced in SMG. Some other relatively frequent pejorative first constituents not dealt with in this chapter due to space limitations are kolo- ‘ass’ (e.g., kol-o-kanáʎa ass-CM-channels ‘fucking TV channels’), ðʝavolo- ‘devil’ (e.g., ðʝavol-o-koronoiós devil-CM-coronavirus ‘damn/devil coronavirus’), ɣamo- ‘fuck(ing)’ (e.g., ɣam-o-panepistímio fucking-CM-university ‘fucking university’), etc. As ‘non-neutral’ we understand here by and large those terms that are not orthophemisms in the terminology suggested by Cameron and Kulick (2003: 29). According to the authors, orthophemism is a term coined by them ‘to account for direct or neutral expressions that are neither sweet-sounding, evasive, or overly polite (euphemistic), nor harsh, blunt, or offensive (dysphemistic)’, i.e., an orthophemism is typically a more formal and more direct or literal term than the corresponding dysphemistic or euphemistic term (see, e.g., dysphemistic ‘shit’, euphemistic ‘poo’ vs. orthophemistic ‘faeces’). For an overview on this differentiation that focuses on Modern Greek data see Kamilaki et al. (2016: 14-22).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

135

2016: section 4.1.2, and references therein) some of them, such as skato-, are treated as so-called prefixoids (semi-prefixes, prefix-like items), that is, as morphemes of a mixed category between affixes and stems, mainly by virtue of the change that they have been undergone in their semantics (i.e., their nonliteral usage). We will not take this position in this study. Following Schmidt’s (1987) postulation for clear-cut categories that do not blur the necessary distinction between compounding and derivation, we will count SMG first constituents under investigation not as prefixoids, but as ordinary stems that may be used in the expressive compounding process in both literal and nonliteral senses. Note that the SMG items under investigation, when involved in the formation of expressive compounds as first constituents, are not semantically remote from the corresponding autonomous lexemes from which they are derived when used non-literally (except for the special case of paʎo-).31 6.3.1

Expressive compounding with first constituent vlaxo‘Aromanian/Vlach.MET’

In SMG the lexeme Βλάχος (also βλάχος) [vláxos] ‘Aromanian/Vlach’ refers literally to that person who has or whose ancestors had the neo-Latin Aromanian/Vlach language as their native language (Béis 2008); see Example (5). In dialectal Modern Greek it was also given the extended meaning of ‘a person who speaks some Modern Greek northern idiom’ (see ILNE, entry Βλάχος 3). Furthermore, in SMG the word has acquired the metaphorical meanings of ‘a villager versus a city dweller’ and of ‘an unrefined, uncouth person’; see (6). When metaphorically used, the word, besides its descriptive, truth-conditional meanings ‘villager’ or ‘unrefined person’, conveys systematically (i.e., conventionally) an additional expressive meaning aspect, that is, it functions as a pejorative itself (see ILNE, entry Βλάχος 4; LKN, entry βλάχος, where the negative content of the word is categorized as something conventional by the usage label ‘derogatory’). By using this term, the speaker expresses an attitude of contempt, derogation, and denigration towards the individual denoted by the base-word (head), for the properties ascribed to him, that is, for being a person who lives in a village, or a person deprived of couth, good manners, or refinement (see 6a, 7a, 6b, 7b). This expressed negative attitude represents the non-truth conditional meaning aspect or the lexeme vláxos in metaphorical use that vanishes when the pejorative term is being replaced by a neutral counterpart.32 31

32

See Meibauer’s (2013) relevant argumentation, according to which left-hand evaluative members of expressive compounds in German are still lexemes, but have undergone metaphorical extension (i.e., are used as metaphors). See in this respect Hornsby (2001: 128).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

136

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(5)

vláxos Aromanian ‘Aromanian/Vlach[–PEJ]’

(6)

vláxos Aromanian.MET.PEJ a. ‘a villager [+PEJ], a provincial’ b. ‘an unrefined, uncouth person [+PEJ], a bumpkin, a boor, a Neanderthal’

(7)

a. me léne vláxo, les ce θa me call.3SG.PR Aromanian.MET.PEJ say.2SG.PR and FUT prozvliθó an me léne vláxo33 be.offended.1SG if me they.call Aromanian.MET.PEJ ‘They call me a provincial, as if I could feel offended when they call me a provincial’ b. ótan ítane sto xorʝó, ítane evʝenicí, when be.3PL.PST in.the village be.3PL.PST polite. tóra pu píɣan ecí ʝínane Vláçi34 Now that go.3PL-PST there become.3PL.PST. Aromanians. MET.PEJ ‘When they were in the village, they were polite; now that they went there they became bumpkins!’

As a first compound constituent the word vláxos appears in the form of vlaxoformed by the stem vlax- and the linking or compound vowel -ο-. Note that occasionally the linking vowel -ο- appears in cases like (18) or (22) where its presence is non-phonologically conditioned (see the case of Example 15, where the absence of -o- is phonologically motivated). Semantically, the first constituent vlaxo- is found in several compounds, both endocentic (i.e., headed, such as 11) and rarely exocentric (where neither of the two constituents assumes the role of the head, e.g., Example 9),35 which refer to Aromanian literally, as in the following Examples (8)–(11). Note that these are all non-negative marked cases listed in SMG dictionaries. (8)

vlax-o-xór-i [LKN, entry βλαχο-] Aromanian-CM-village-SUFF ‘a village inhabited by Aromanians’

(9)

vlax-ó-fon-os Aromanian-CM-voice-SUFF ‘Aromanian speaking person’

[LKN, entry βλαχο-]

(10)

vlax-o-kalíva Aromanian-CM-hut ‘Aromanian hut’

[LNEG, entry βλαχοκαλύβα]

(11)

vlax-o-ðímarxos [ILNE, entry βλαχοδήμαρχος] Aromanian-CM-mayor ‘mayor of a village inhabited by Aromanians’

33 35

34 Personally attested by M. Konstantinidou. Personally attested by M. Konstantinidou. For this distinction in Greek, see Ralli (1992, 2013).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

137

Alternatively, vlaxo- can be used metaphorically for someone with uncouth, unrefined manners, or old-fashioned/anachronistic mentality, as in (12) and (13). It also forms endocentric compounds which refer to something rustic, naive, artless, inartistic, crude, or of bad quality (especially with regard to music or style genres), as in (14). When metaphorically used, first constituent vlaxo- has a mixed semantics; besides the descriptive (truth-conditional) meaning of ‘unrefined, inartistic, anachronistic’, ultimately, ‘bad in manners or in quality’, it also has the expressive (non-truth-conditional) meaning aspect of pejoration.36 By virtue of vlaxo- the speaker ascribes to what is denoted by the second compound constituent, the characteristics of ‘impoliteness’, ‘unreliability’, ‘anachronism’, or of ‘low quality’, and simultaneously he indicates his negative (abusive, derogatory) attitude towards the referent for the ascribed characteristics; he expresses this attitude, without literally saying, just by choosing the concrete linguistic item and not an expressively neutral counterpart of it. Thus, the metaphorically used element vlaxo- contributes both descriptive and expressive content;37 moreover, by virtue of being applicable to an expressively neutral base (e.g., ðímarxos ‘mayor’, ðiciɣóros ‘lawyer’ etc.) it contributes to the pejorative meaning of the whole compound,38 as is seen, for example, in (12)–(14). (12)

vlax-o-ðímarxos39 a. Aromanian.MET.PEJ-CM-mayor ‘mayor of a non-progressive village [+PEJ]’ b. Aromanian.MET.PEJ-CM-mayor ‘mayor with an authoritarian mentality and old-fashioned ideas [+PEJ]’ c. Aromanian.MET.PEJ-CM-mayor.MET ‘a newly wealthy man who has not managed to eliminate his unrefined manners [+PEJ]’

(13)

vlax-o-ðiciɣóros40 Aromanian.MET.PEJ-CM-lawyer ‘bad lawyer [+PEJ], pettifogger’

(14)

vlax-o-barók [LKN, entry βλαχο-] Aromanian.MET.PEJ-CM-baroque ‘a kind of style that imitates baroque in a way that is considered kitsch by the speaker [+PEJ]’

36 37

38 39 40

For the systematic connection of metaphorical meaning with expressive meaning see Meibauer (2013). Compare similar approaches by Gutzmann (2011) on nominal pejoratives, usually called slurs, Steriopolo (2008) on Russian expressive affixes, and Fortin (2011) on productive Spanish connotative affixes. For this quality of metaphorically used first constituents see Meibauer (2013: 23). ILNE, entry βλαχοδήμαρχος; LKN, entries βλαχο- and βλαχοδήμαρχος; LNEG, entry βλαχοδήμαρχος; XLNEG, entry βλαχοδήμαρχος. ILNE, entry βλαχοδικηγόρος; LKN, entry βλαχο-.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

138

Kallergi, Katsouda, Konstantinidou, Tsangalidis

When vlaxo- is an expressively neutral first constituent, that is, when it contributes the pure literal meaning relating to Aromanians, it seems to be nearly unproductive in SMG. Instead of the formation of a compound with non-head vlaxo- in its literal sense, other linguistic means are selected to convey the meaning ‘somebody/something related to Aromanians’. So, instead of a compound of the type vlaxokalíva ‘Aromanian hut’, constructions with a genitive NP modifier, for example, kalíva vláxon hut Aromanian.GEN.PL ‘hut of Aromanians’, or adjectival modification, for example, vláçici kalíva Aromanian.ADJ.FEM hut ‘Aromanian hut’, seem to be of common use in SMG. This leads us to assume that whenever vlaxo- is used in SMG, it is seemingly used metaphorically, that is, pejoratively. On the other hand, vlaxo- as an expressive first constituent, that is, when used metaphorically, contributing both descriptive and expressive meaning, is relatively productive in SMG, and can easily be found in informal, oral discourse. So, apart from the aforementioned examples, included in standard dictionaries of Modern Greek and being treated, occasionally, as autonomous/ independent lexical entries (e.g., case 12 of vlaxoðímarxos in both ILNE and LKN, entries βλαχοδήμαρχος and βλαχο-), we have attested a large number of compounds in computer-mediated communication, where vlaxo- functions as (mixed) expressive, that is, it indicates the negative attitude of the speaker towards the referent (expressive content), to which he ascribes qualities such as ‘impoliteness’, ‘unreliability’, ‘anachronism in ideas or in manners’, ‘low standards in terms of taste, aesthetics, level of education’ (descriptive content) considered typical of the Aromanians by that group of speakers who are using vlaxo- metaphorically, that is, to derogate/to express contempt; see cases in Examples (15)–(19). (15)

vlax-ánθropos41 Aromanian.MET.PEJ-man ‘uncouth person [+PEJ]’

(16)

vlax-o-trapezítis42 Aromanian.MET.PEJ-CM-banker ‘uncouth banker, i.e., behaving uncouthly while doing his job [+PEJ]’

(17)

vlax-o-taksidzís43 Aromanian.MET.PEJ-CM-taxi.driver ‘uncouth taxi driver, i.e., behaving uncouthly while doing his job [+PEJ]’

41 42 43

Ρε μπαοκια βλαχάνθρωποι κουράσατε ρε νούμερα πραγματικά κουράσατε με τα παπατζιλίκια σας τόσα χρόνια. At https://bit.ly/3FzXZ8p (accessed 29 January 2020). Από βλαχοδήμαρχοι έως βλαχοτραπεζίτες μοιράζουν χρήμα. At https://bit.ly/3TqNTwh (accessed 8 November 2021). Ἐχουν μαζευτεί η σάρα, η μαρα, οι βλαχοταξιτζήδες και τα σκυλιά του δρόμου!!! At https://3vita.gr › topiko-voulas-20141128-8558 (accessed 8 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

139

(18)

vlax-o-evropéos neoélinas44 Aromanian.MET.PEJ-CM-European.ADJ Modern.Greek ‘uncouth European Modern Greek, i.e., Greek imitating the European way of life badly and unsuccessfully [+PEJ]’

(19)

vlax-o-diméni kopéla45 Aromanian.MET.PEJ-CM-dressed.FEM girlfriend ‘old-fashioned dressed girlfriend [+PEJ]’

Vlaxo- in metaphorical usage is mainly combined with [+animate] nouns (see cases from 15 to 18). We have also found a case where the base is a [+animate] participle (e.g., 19). Nevertheless, the base-word (second compound constituent) can also be a [–animate] noun, to the referent of which the speaker ascribes the characteristics ‘unrefined’, ‘rough’, ‘provincial’ or ‘of bad taste or low quality’, for example, cases from (20) to (22). (20)

vlax-o-siberiforá 46 Aromanian.MET.PEJ-CM-behaviour ‘uncouth behaviour, manners considered typically as rude/impolite [+PEJ]’

(21)

vlax-o-víla47 Aromanian.MET.PEJ-CM-mansion ‘nouveau-riches or bad-taste mansion [+PEJ]’

(22)

vlax-o-ekpombí 48 Aromanian.MET.PEJ-CM-show.N ‘bad-taste or low quality (TV) show [+PEJ]’

When the second constituent is a taboo or highly colloquial lexeme, an abusive or insulting term (see, e.g., cases in 23 and 24), or a lexeme that is abusive only metaphorically (see, e.g., case in 25), the first constituent vlaxo- seems to play exclusively the role of an intensifier/amplifier of the expressive meaning of the second constituent.49 It in fact functions as a pure expressive item, that is, as an

44 45 46

47 48 49

Φρόντισαν να διαδίδουν τον παρωχημένο «διαφωτισμό» τους στον διψασμένο τότε «βλαχοευρωπαίο» Νεοέλληνα. At https://bit.ly/40bXHMS (accessed 1 Juli 2021). Bλαχοντυμένη νυν κοπέλα του πρώην σου. At https://bit.ly/3FBbULi (accessed 1 Juli 2021). Το να κάνεις ό,τι σου καπνίσει, χωρίς να σε νοιάζει αν ενοχλείς, δεν είναι μαγκιά αλλά ελληνική βλαχοσυμπεριφορά ‘Doing whatever you want, without caring about whether or not you are disturbing anyone, is not cunning, it is a typical Greek uncouth behavior’. At https://kzread .info/dron/golnqWOK8tm3Vl_9FYkgfQ.html (accessed 8 November 2021). Κατέστρεψαν το φυσικό περιβάλλον για να κτίσουν τις βλαχοβίλες τους. Personally attested by M. Konstantinidou. Δεν αποβλακώνεται με. . . βλαχοεκπομπές που ανακυκλώνουν την ί-δι-α σαθρότητα. At https:// diskoryxeion.blogspot.com/2012/12/76.html (accessed 1 Juli 2021). The notion of intensification as an increase in expressiveness is discussed by Fortin (2011: 80), with reference to Yakimova (2000: 153), Bergeton (2004), Navarro Ibarra (2009), and others. Compare also Potts’s (2007: 166) notion of ‘emphatic (expressive!) modifier’ with respect to the intensifying meaning aspect of the term fucking in really fucking brilliant, where the type of expressivity is of course not [+negative/+pejorative], but the opposite; see also Meibauer’s (2013: 31) notion of ‘pure intensifier’ with regard to the intensifying role of German Arsch- in

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

140

Kallergi, Katsouda, Konstantinidou, Tsangalidis

expletive pejorative (glossed EXPL.PEJ), that contributes no descriptive, but merely expressive meaning (i.e., pejoration),50 see cases from (23) to (25). Note that the negative expressive content encoded and expressed by vlaxo- whenever it is used as a pejorative intensifier/amplifier comes from the conventionalized metaphorical (pejorative) usages of vlaxo- described above51 (since vlaxo- has no expressivity, when literally used). This metaphor-based negative content of vlaxo- is added to the negative content encoded and expressed by the non-neutral (pejorative) second compound constituent, with the result being that the compound produced displays a higher degree of negative expressivity (pejoration/ deprecation)52 than its expressive basis (second constituent) by itself. Relying on Potts’s (2007) notion of repeatability as an identifying property of expressives, we could treat this emphatic pattern as a case of repetition of the pejorative aspect that leads to its strengthening. Similar emphatic functions of first constituents have been identified in expressive compounds in German (see Meibauer 2013: 31). Αs for the metaphoric descriptive meanings of vlaxo- related to ‘impoliteness’, ‘unreliability’, ‘anachronism in ideas or in manners’, ‘low standards in terms of taste, aesthetics, level of education’ etc., these seem to be eliminated or blocked as irrelevant whenever the second compound constituent is an expressive one,53 though this is a subject deserving further investigation. (23)

50

51

52 53

54

vlax-o-malákas54 Aromanian.EXPL.PEJ-CM-masturbator.PEJ ‘asshole/wanker [+INTENSE.PEJ]’

the adjectival compound arschbeschissen ‘arse rotten’. It should be noted that even if we are ready to acknowledge an intensification in ‘descriptiveness’, i.e., a superlative (descriptive) meaning aspect of the type ‘very’ attached to the first compound constituent when it is used as an intensifier, we will still have to explain what this meaning is based on, i.e., where it originates from. Following Cruse (2004) and other scholars (Gutzmann 2013: 26, 2015: 38), we call linguistic items that ‘possess only expressive and no descriptive meaning’ expletive (Cruse 2004: 57). For the terminological distinction between expletive pejoratives that share the characteristic [– MIXED] and mixed pejoratives, that share the characteristic [+MIXED] in their semantic profile, see Gutzmann and McCready (2016: 77). See also the respective discussions regarding vromo- and paʎo- (§6.3.3 and §6.3.5). Meibauer (2013: 31) argues that first constituents that work as pure intensifiers ‘are nevertheless metaphors’; that their intensifying meaning is metaphor-based. For the conventional character of metaphors, in the sense that they are not constructed ad hoc, see also Meibauer (2013: 31). Regarding quantification with pejoratives (i.e., weak and strong pejoratives) see Gutzmann and McCready (2016). For Greek, see also Xydopoulos and Christopoulou (2019). Compare the case of proper name interjections of the type of English Jesus Christ! used to express surprise or anger. See also Gutzmann (2015: 274–5) for a similar argumentation regarding the semantic changes of English boor which started as an ordinary descriptive predicate but has been developed into a functional expletive use-conditional (i.e., expressive in our terms) item through an intermediate stage, where it must have had a functional mixed content. Γιατί τόση κακία για την κοπέλα ρε βλαχομαλάκες νεοέλληνες. At https://kzfaq.info › get › kalomira-sings-greece-national. . . (accessed 8 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek (24)

vlax-o-vlax-ára55 Aromanian.EXPL.PEJ-CM-Aromanian-AUG.FEM.PEJ ‘extremely uncouth (female) person [+INTENSE.PEJ]’

(25)

vlax-o-teleftéi56 Aromanian.EXPL.PEJ-CM-last.PL.MASC.MET.PEJ ‘the absolute last ones [+INTENSE.PEJ], absolute trash’

141

Thus, when vlaxo- is associated with a pejorative base, it strengthens/amplifies the negativity of the second constituent, that is, it functions as a pejorative intensifier, as an exclusively non-descriptive item, contributing purely nondescriptive meaning (i.e., pejoration). 6.3.2

Expressive compounding with first constituent ʝifto- ‘Gypsy’

The SMG lexeme γύφτος [ʝíftos] ‘Gypsy’ (derived from the national name Αιγύπτιος [eʝíptios] ‘Egyptian’) is the non-neutral, pejorative/derogatory counterpart of the unmarked SMG term αθίγγανος [aθíŋganos] or τσιγγάνος [tsiŋgános], and their alteration ατσίγγανος [atsíŋganos] ‘Romani person, Romany’ (a Middle Greek word derived from Indic Athigan through the Romani language, following LKN, entries αθίγγανος, ατσίγγανος, τσιγγάνος). The lexeme ʝíftos belongs to the category of nominal pejoratives which are usually called (racial/ethnic) slurs57 and stand as one of the prime examples of expressives (Amaral 2018). Following the stereotype analysis, according to which the meaning of a slur is a mixture of descriptive and expressive (pejorative) content (Croom 2011, 2013, Gutzmann 2011, Williamson 2009), we may analyse the SMG lexeme ʝíftos as one having both the descriptive meaning of the corresponding neutral term aθíŋganos, tsiŋgános, that is, ‘member of a nomadic people of a certain ethnicity/nationality or race speaking Romani, a language related to Hindi’, and the expressive meaning aspect of pejoration, that is, the negative (derogatory) attitude of the speaker towards the people of that ethnicity/nationality or race, see (26). (26)

ʝíftos Gypsy ‘Gypsy/Romany [+PEJ]’

The expressive (pejorative) term ʝíftos has also acquired in SMG the metaphorical/extended derogatory meanings (27a)–(27d); see ILNE, LKN, LNEG, XLNEG, entry γύφτος. 55 56 57

H μια είναι Πρωτευουσιάνα και αυτή είναι βλαχοβλαχάρα. At https://bit.ly/3xM1PXB (accessed 8 November 2021). Τελευταίες! Βλαχοβλαχάρες! Βλαχοτελευταίοι. At https://twitter.com › Tommaso_Tsa › status (accessed 8 November 2021). For slurs and lexical pejoratives see Hom (2008, 2010), Croom (2011, 2013), Gutzmann (2015), Finkbeiner et al. (2016a).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

142

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(27)

ʝíftos Gypsy.MET.PEJ a. ‘a petty, mean or miser person [+PEJ]’ b. ‘an uncultured, uncouth/boorish person [+PEJ]’ c. ‘a person who lives in conditions of great disorder and filth [+PEJ]’ d. ‘a person with dark complexion or with a presentation stereotypically/by convention attributed to Gypsies [+PEJ]’

As a first compound constituent, the pejorative ʝíftos occurs in the form ʝiftomade up by the stem ʝift- of the lexeme ʝíftos and the linking vowel -o-. It (ʝifto-) appears in an unexpectedly large corpus of expressive compounds,58 predominantly endocentic (e.g., 28–37). It is combined most frequently with nominal bases (e.g., cases in 29–31 and 33–37) and rarely with adjectival ones (e.g., cases in 32 and 40). As can be seen in Examples (28) to (32), the second constituents combined with the first constituent ʝifto- may share the characteristic [+animate]/[+human], but they may also share the opposite characteristic [–animate] (see, e.g., cases in 35 and 37). Also, all bases in Examples from (28) to (37) are expressively neutral items themselves. (28)

ʝift-o-politikós59 Gypsy.MET.PEJ-CM-politician ‘unreliable/unfair politician [+PEJ]’

(29)

ʝift-o-ipurɣós60 Gypsy.MET.PEJ-CM-minister ‘unqualified/unreliable minister [+PEJ]’

(30)

ʝift-o-turístas61 Gypsy.MET.PEJ-CM-tourist ‘tourist who does not spend money [+PEJ]’

(31)

ʝift-o-evropéos62 Gypsy.MET.PEJ-CM-European ‘uncouth/uneducated European [+PEJ]’

(32)

ʝift-o-diménos63 Gypsy.MET.PEJ-CM-dressed ‘badly dressed [+PEJ]’

58 59

60 61

62 63

It is worth noting that as a first constituent, ʝifto- does not constitute a distinct entry in SMG dictionaries, except for the LKN dictionary. Η χώρα πράγματι βρίσκεται σε πρωτοφανή φτώχεια και την ευθύνη την έχουν οι γυφτοπολιτικοί που τόσα χρόνια και ακόμα και σήμερα κλέβουν δημόσιο χρήμα. Example personally attested by G. Katsouda. Τρόμος στην σκέψη ότι αυτός ο γυφτοϋπουργός είναι υπεύθυνος για την ασφάλειά μας. Example personally attested by G. Katsouda. Γυφτοτουρίστες από τα βαλκάνια με κρουασάν και αυγά μέχρι φασόλια και φακές από το σπίτι τους. . . νοικιάζουν δωμάτια να μείνουν τζαμπέ. At https://bit.ly/3XVtGPW (accessed 8 November 2021). At https://bit.ly/3Z8XhGg (accessed 8 November 2021). At https://bit.ly/3Tvr0I7 (accessed 8 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek (33)

ʝift-o-panepistímiο64 Gypsy.MET.PEJ-CM-university ‘university that does not have financial resources [+PEJ]’

(34)

ʝift-o-trápeza65 Gypsy.MET.PEJ-CM-bank ‘unreliable bank [+PEJ]’

(35)

ʝift-o-siberiforá66 Gypsy.MET.PEJ-CM-behaviour ‘unqualified/unprofessional behaviour [+PEJ]’

(36)

a. ʝift-elás67 Gypsy.MET.PEJ-Greece ‘uncivilized Greece [+PEJ]’ b. ʝift-o-eláða68 Gypsy.MET.PEJ-CM-Greece ‘uncivilized Greece [+PEJ]’

(37)

ʝift-o-páputs-a69 Gypsy.MET.PEJ-CM-shoe-SUFF.PL ‘miserable, old, dirty, badly made shoes [+PEJ]’

143

Firstly, we observe that the compound vowel -ο- appears between the first and the second constituent, even though the second constituent begins with a vowel stronger than /o/ (e.g., cases 31 ʝiftoevropéos and 36b ʝiftoeláða), that is, in cases where the presense of -o- should be phonologically prohibited (see §6.3). It is, nevertheless, worth dwelling on the cases of (36a) and (36b), where we have two compounds (ʝiftelás, ʝiftoeláða) with the same word-base in two different forms, both beginning with an /e/, namely the [+learned] form elás ‘Greece’ (Ancient Greek Ἑλλὰς) and the [–learned]/plain form eláða ‘Greece’ (derived from the accusative form Ἑλλάδα of the Ancient Greek noun Ἑλλάς); for the distinction [+/–learned] see for more details the introduction in §6.4). In the case of ʝiftelás (36a) the absense of -o- is phonologically conditioned, whereas its non-phonologically motivated presence in ʝiftoeláða marks the loose relation between the compound parts. Secondly, regarding semantics, it is observed that, in all the cases mentioned above, ʝifto- does not retain the literal meaning ‘Gypsy’, but is used metaphorically to ascribe to the referent denoted by the second constituent some properties considered typical of Gypsies by that group of speakers who use

64 65 66 67 68 69

At https://bit.ly/3JOVYHT (accessed 8 November 2021). Example personally attested by G. Katsouda. At https://bit.ly/3FEetMS (accessed 8 November 2021). At https://bit.ly/42u1sPQ (accessed 8 November 2021). At https://bit.ly/3JsyruV (accessed 8 November 2021). Φοράν τα ίδια γυφτοπάπουτσα από πενήντα χρόνια πριν, τα παιδιά τους. At https://bit.ly/ 3YZIMnN (accessed 8 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

144

Kallergi, Katsouda, Konstantinidou, Tsangalidis

the first constituent ʝifto- non-literally to derogate or hold in contempt their target. These stereotypical properties are: (a) (b) (c) (d) (e)

‘dirty, messy, very poor or miser’, e.g., (30), (33) ‘uncouth, uncultured/uneducated’, e.g., (31), (36) ‘unqualified/unprofessional, unreliable’, e.g., (28), (29), (34), (35) ‘of bad taste’, e.g., (32) ‘miserable, dirty, badly made’, e.g., (37)

Nevertheless, there are also rare cases in which ʝifto- is used with its literal meaning in order to ascribe the property of being a Gypsy or being related to Gypsies (see, e.g., case in 38). (38)

ʝift-ó-sporos70 Gypsy-CM-seed.MET ‘Gypsy’s offspring/child [+PEJ]’

Apart from neutral (non-expressive) word-bases, ʝifto- is also to be found in combination with non-neutral (offensive/insulting, taboo etc.) second constituents. As with vlaxo- in the previous §6.3.1, ʝifto- in these cases seems to lose its (metaphorical) descriptive meanings described above and to function as a pure (pejorative) intensifier/amplifier, namely as an expletive pejorative that increases the intensity/degree of pejoration communicated through the pejorative word-base (see cases from 39 to 41). (39)

ʝift-o-malákas71 Gypsy.EXPL.PEJ-CM-masturbator.PEJ ‘asshole/wanker [+INTENSE.PEJ]’

(40)

ʝift-o-póniros72 Gypsy.EXPL.PEJ-CM-cunning.PEJ ‘sly/cunning person [+INTENSE.PEJ]’

(41)

ʝift-o-kléftis73 Gypsy.EXPL.PEJ-CM-thief.PEJ ‘thief/cheater [+INTENSE.PEJ]’

6.3.3

Expressive compounding with first constituent vromo- ‘stink’

The lexeme βρόμα [vróma] is a semantically non-neutral (vernacular/slang) element in SMG, in the sense that it is seldom employed in polite colloquial speech. So, next to its descriptive (truth-conditionally relevant) meaning ‘unpleasant/bad smell’ and in an extended sense ‘impurity/excrement that 70 71 72 73

See LKN, entry γυφτο-. At https://www.pronews.gr/amyna-asfaleia/toyrkia/242363 (accessed 8 November 2021). Example personally attested by M. Konstantinidou. Example personally attested by M. Konstantinidou.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

145

emits a bad smell, disgusting dirt’ it also contributes a (slightly) derogatory/ abusive meaning aspect, like the English lexemes ‘stink, niff, pong’ and ‘filth’, ‘dirty’, respectively. Likewise, in a metaphorical sense, vróma encodes negative expressivity, in that it may be used as an insulting/abusive characterization of a person, usually for a woman considered obscene (viz. with the meaning ‘slut’), and this is so already since the late Middle Ages (see Sinaksarion Jinekon 850; Kriaras, entry βρόμα). As a first compound constituent, pejorative vróma appears in the form vromo- (and that since the fourteenth century ce, see cases 50–51) made up by the stem vrom- of the lexeme vróma and the compound vowel -o-. Note that, as with vlaxo-, there are exceptional cases like (44), (49), (54) where the presence of -o- is non-phonologically conditioned, that is, it appears even when the second constituent begins with a vowel stronger than /o/, that is, an /a/ or /e/. For instance, when the base begins with an /a/, the linking vowel -o- may or may not appear (vrom-ánθropos in Example 48, yet, also vrom-oánθropos74 stink.MET-CM-person ‘filthy/mean/hated person’). In virtue of the pejorative/abusive meaning aspect of the lexeme vróma, when combined with expressively neutral bases (usually heads), vromo- forms expressively non-neutral, that is (slightly) pejorative/insulting/offensive compounds,75 both endocentric (e.g., vrom-ó-ɣlosa stink.MET-CM-tongue.MET ‘abusively, indecent language’) and exocentric (e.g., vrom-ó-ɣlos-os stink. MET-CM-tongue-SUFF ‘foul-mouthed, foul-tongued’). These are items with a mixed semantics, that is, compounds which along with their descriptive meaning convey typically an additional negative meaning aspect, too. More specifically, by using vromo- in literal or metaphorical senses, the speaker ascribes to the referent denoted by the second compound constituent one of the following properties (a)–(f ). These constitute the descriptive content of the first constituent under investigation. When used literally vromo- assigns to the referred entity the properties (a) or (b): (a) ‘having an unpleasant smell’,76 e.g., vrom-ó-xorto stink-CM-herb ‘skinking/ stinky herb’ (LKN, entry βρομόχορτο; ILNE, entries βρωμόχορτο 1, βρωμο- 1); 74 75

76

At https://bit.ly/3JSFKgX (accessed 8 November 2021). Note that in so far as there are – to our knowledge – no empirical data concerning the degree (higher or lower) of offensiveness of first constituent vromo- in SMG, we will rely, regarding this, on our intuition as Modern Greek native speakers, taking also into account the usage labelling distinctions made by SMG dictionaries with respect to vromo- and other SMG expressive (pejorative) first constituents, such as skato- ‘shit’ (see §6.3.4), kolo- (‘ass’) etc. For the fluctuation of offensiveness in Modern Greek slang/taboo vocabulary and for scaling offensiveness as a continuum, that goes from [–offensiveness]/no offensiveness through [±offensiveness]/slight offensiveness to [+offensiveness]/increased offensiveness, see Christopoulou et al. (2017). In Modern Greek dialects there are numerous complex names with first constituent vromomeaning different bad smelling plants (ILNE, entry βρωμο- 1), such as vrom-ó-xorto stink-CMherb ‘in places, the plant Chenopodium murale or Datura stramonium or Heliotropium villosum

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

146

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(b) ‘being unclean, dirty, grimy’, e.g., vrom-ó-nera stink-CM-water.PL ‘dirty (stagnant) water’, vrom-ó-pan-o stink-CM-cloth-SUFF ‘dirty cloth’, vromó-çer-a stink-CM-hand-SUFF.PL ‘dirty hands’ (ILNE, entry βρωμο- 2), vrom-o-ɣúrun-o77 stink-CM-pig-SUFF ‘dirty/stinky pig’. Yet, more often vromo- is used non-literally to assign to the [+/–animate] referent, usually sharing the feature [+smell-free] or [+dirt-free], the properties from (c) to (f ): (c) ‘morally dirty, obscene, vulgar, bitchy’, e.g., vrom-o-θíliko stink.METCM-female.N ‘insultingly, (little) woman of bad reputation’, vrom-ó-ɣlosa stink.MET-CM-tongue.MET ‘abusively, indecent language or metonymically a person who uses bawdy talk’, vrom-ó-loɣ-o stink.MET-CM-wordSUFF ‘obscenity’, vrom-o-ðuʎá stink.MET-CM-job ‘indecent job/action’ (ILNE, entry βρωμο- 4a); (d) ‘unpleasant, annoying, pesky’ (usually of state of affairs), e.g., vrom-ókrio stink.MET-CM-cold.N ‘dirty coldness, uncomfortably cold’, vrom-óceros stink.MET-CM-weather ‘dirty/awful weather’, vrom-o-aéras stink. MET-CM-wind ‘awful wind’, vrom-o-táksið-o stink.MET-CM-journeySUFF ‘a journey in which one experiences hardships’ (ILNE, entries βρωμο- 5, βρωμοτάξιδο); (e) ‘of very poor quality, of very low standards, worthless, useless, bad’ (usually of [–animate], [–human] referents), e.g., vrom-o-áloɣo stink. MET-CM-horse ‘worthless, bad horse’, vrom-ó-cer-o stink.MET-CMcandle-SUFF ‘bad, worthless candle’, vrom-ó-ksið-o stink.MET-CM-vinegar-SUFF ‘vinegar of bad/poor quality’, vrom-o-fústan-o stink.METCM-dress-SUFF ‘dress of bad/poor quality’, vrom-ó-psom-o stink.METCM-bread-SUFF ‘bread of bad/poor quality’ (ILNE, entry βρωμο- 3), vrom-o-stiló stink.MET-CM-pen ‘pen of very poor quality’, vrom-omáθima stink.MET-CM-course ‘course of very low standards’ (LNEG, entry βρομο-). It is under this special sense (e) that first constituent vromoshows a very high frequency of occurrence in SMG, certified by the large number of findings in simple Google searches. See Examples from (42) to (45) from both everyday and computer-mediated conversations:

77

etc.’, vrom-o-kláði stink-CM-branch ‘the plant Anagyris foetida, commonly Mediterranean stinkbush’, vrom-apíɣanos ‘the plant Ruta graveolens’; yet these plant names presumably have no expressive meaning aspect, due to lack of respective evidence as well as of evidence for derogatory nuances of the word vróma in dialectal Modern Greek. To be noticed that all mentioned dialectal examples are provided by ILNE in the respective entries (first constituent written βρωμο-). Ποῦ πᾶς μὲ αὐτὸ τὸ βρωμογούρουνο, χοιροβοσκὲ χαμένε; (Kazantzakis & Kakridis Odysseia: 232.219).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek (42)

vrom-o-trápeza78 stink.MET-CM-bank ‘filthy bank [+PEJ]’

(43)

vrom-o-vivlío79 stink.MET-CM-book ‘filthy book [+PEJ]’

(44)

vrom-o-efimeríða80 stink.MET-CM-newspaper ‘broom newspaper/brochure [+PEJ]’

(45)

vrom-o-kánal-o81 stink.MET-CM-channel-SUFF ‘bad/unreliable/spreading fake news TV channel [+PEJ]’

147

(f ) ‘obnoxious, abhorrent, hated, hateful, detestable’ (usually of [+animate], [+human]); see cases from (46) to (49), also ILNE, entry βρωμο- 4b where many other relevant examples are to be found. Notice that under this special sense the first constituent vromo- is also very productive in SMG nowadays. (46)

vrom-ó-turkos82 stink.MET-CM-Turk ‘dirty/filthy/hated Turk [+PEJ]’

(47)

vrom-o-ðanistís83 stink.MET-CM-lender/creditor ‘dirty/filthy creditor [+PEJ]’

(48)

vrom-ánθropos84 stink.MET-CM-man ‘filthy person [+PEJ]’

(49)

vrom-o-evropéos85 stink.MET-CM-European ‘filthy European [+PEJ]’

In all six meaning cases (a)–(f ), apart from the aforementioned properties assigned to the referent by means of first constituent vromo- (whether in literal or non-literal use) which constitute the descriptive meaning aspects of vromo-, an attitude of disgust, annoyance, or contempt on the part of the speaker is also 78 79 80 81 82 83 84 85

Έκλεισε αυτή η βρωμοτράπεζα. Personally attested by M. Konstantinidou. Παράτα τα βρωμοβιβλία στο σχολείο και ξεκίνα το γλέντι. ‘Leave the filthy books at school and get started party’. At http://polyk.blogspot.com/2017/ (accessed 8 November 2021). At https://bit.ly/3FCIk8b (accessed 8 November 2021). At https://bit.ly/3JQg0lq (accessed 8 January 2022). At https://bit.ly/3yRmwSB (downloads PDF; accessed 8 November 2021). See also ILNE, entry βρωμο- 4b. Το ΣτΕ έκρινε τις μειώσεις συντάξεων αντισυνταγματικές, οπότε τι συζητάτε με τους βρωμοδανειστές; Example personally attested by M. Konstantinidou. Πρέπει να πέσουν πρόστιμα να συνετιστούν μερικοί βρωμάνθρωποι! At https://attikosparatiritis .wordpress.com (accessed 8 November 2021). At https://bit.ly/3JnHdKA (accessed 8 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

148

Kallergi, Katsouda, Konstantinidou, Tsangalidis

expressed by means of the compound. This attitude is directed towards the referent (because the referent has to do with vróma, in literal or non-literal sense) or is directed towards some properties of the referent, such as those exemplified above, following Croom’s (2011: 353) respective argumentation. This expressed attitude on the part of the speaker constitutes the expressive meaning aspect of the compound, and by being not-truth-conditionally relevant, it cannot be cancelled or denied, even though the properties ascribed to the referent of the compound, being themselves descriptive (truthconditionally relevant), are amenable to denial. It should also be noted that expressive compounds with the first constituent vromo- in literal and metaphorical use and with a pejorative/insulting meaning aspect make their appearance timidly since the late Middle Ages; see cases in (50), that is, βρωμόστομος [vromóstomos] ‘abusively, one who has a stinky mouth’ and βρωμόχνωτος [vromóxnotos] ‘abusively, one whose breath stinks’, (51), that is, βρωμόπαπον [vromópapon] ‘filthy duck’, and (52), that is, βρωμοαρμένιοι [vromoarménii] ‘filthy Armenians’. (50)

Medieval Greek [Poulologos 1.43–44; 14th century] Βρωμόστομε, βρωμόχνωτε, τί ἔν’ τὰ λαλεῖς, πρὸς τίναν, ἐσὺ ὁποὺ τρῶς ἐκ τὸ πουρνὸν ὁλῶμα τὰ ὀψάρια vromóstome, vromóxnote, ti en ta lalís, pros tínan, esí opú tros ek to purnón olóma ta opsária ‘You, stinky mouth! You, stinky breath! What are you talking about; and to whom do you tell them? You, who have been eating uncooked fish since morning’

(51)

Medieval Greek [Poulologos 2.35–36; 14th century] Γυρίζει ὁ σμυρίλιος καὶ λέγει πρὸς τὴν πάπιαν: Τίνος τὰ λές, βρωμόπαπον, τίνος τὰ τσαμπουνίζεις; ʝirízi o zmirílios ce léʝi pros tin pápçan tínos ta les, vromópapon, tínos ta tsambunízis ‘The thrush turns around and says to the duck: To whom are you saying these things, you filthy duck, to whom are you telling this nonsense?’

(52)

Early Modern Greek [Procopius of Nazianzus Narrative 1.1084–5; beginnings of nineteenth century] Οἱ γὰρ βρωμοαρμένιοι πάλιν δὲν ἡσυχάζουν, πουγγεῖα ἀναρίθμητα τοῖς ἐντοπίοις τάζουν i γar vromoarménii pálin ðen isixázun, Puŋɟía anaríθmita tis endopíis tázun ‘For, the filthy Armenians do not stay quiet again, they promise countless money bags to the locals’

On the other hand, when SMG vromo- combines with an expressively nonneutral base, for example, with a pejorative or an otherwise negatively marked

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

149

second constituent, it seems to contribute none of the descriptive meanings (a)– (f ) mentioned above (literal or metaphor-based), but appears to merely strengthen/intensify the expressive content, that is, the negative emotion or attitude conveyed by the second constituent. Note that the described increase in expressiveness results from the combination of the expressivity of the two constituents, inasmuch as both of them are expressively marked (i.e., they are intrinsically pejoratives).86 The allegedly descriptive character of the intensification observed, which has been suggested by dictionary approaches through the usage of the quantifier marker ‘very’, ‘extremely’ and the like within the definitions of vromo- compounds,87 is questionable, in so far as the emergence of the notion ‘very’/’extremely’ from the notion of SMG ‘vróma’ (i.e., ‘stink/dirt’) remains unexplained in descriptive terms, and the quantification susceptibility of the notions represented by the second constituents remains to be seen. In this sense, vromo- appears to function as a pure intensifier/amplifier, as an expletive pejorative term.88 See Examples from (53) to (57), where vromo- modifies a taboo word/an obscenity (53), a racial slur (54), insults (55)–(56), or a term in nonneutral metaphorical/extended use (57). (53)

vrom-o-putána89 stink.EXPL.PEJ-CM-prostitute.PEJ ‘filthy whore [+INTENSE.PEJ]’90

(54)

vrom-o-arápis91 stink.EXPL.PEJ-CM-nigger.PEJ ‘filthy nigger [+INTENSE.PEJ]’

(55)

vrom-o-malákas92 stink.EXPL.PEJ-CM-masturbator.PEJ ‘filthy asshole/wanker [+INTENSE.PEJ]’

(56)

vrom-o-ɣúrun-o93 stink.EXPL.PEJ-CM-pig.MET.PEJ-SUFF ‘filthy pig [+INTENSE.PEJ]’ (insultingly, applied to a person, ‘boorish guy’)

86 87

88 89 90 91 92 93

See the related discussion with regards to vlaxo- (§6.3.1). According to ILNE (entry βρωμο- 6), for example, first constituent vromo- acts in standard and in dialectal Modern Greek as an intensifier of the meaning of the second constituent, e.g., SMG vrom-ó-ksilo (literally stink-CM-wood) ‘strong/powerful beating, violent thrashing’; dialectal, Cyprus vrom-ó-kutos (literally stink-CM-obtuse) ‘very obtuse’; dialectal, Cyprus vrom-ó-kutsos (literally stink-CM-lame) ‘very lame’ etc. See in this regard §6.3.1 and §6.3.5. Υποτίθεται ότι είναι (γυναίκες) ανώτερες και πλουσιότερες, αλλά είναι βρωμοπουτάνες (Papataxiarchis Tautotites: 274). Compare synonym putan-arjó explained by LKN (entry πουταναριό 2) as ‘whore (with intensity)’. At https://bit.ly/3ySqzhy (accessed 15 December 2021). At www.press-gr.com/2007/12/blog-post_9047.html (accessed 8 November 2021). Τα βρωμογούρουνα δίνουν το 2ο μάτς της σεζόν. At https://bit.ly/3EuWHuD (accessed 8 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

150

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(57)

vrom-ó-ksilo94 stink.EXPL.PEJ-CM-wood.MET ‘violent beating, thrashing [+INTENSE.PEJ]’

6.3.4

Expressive compounding with first constituent skato- ‘shit’

The lexeme σκατό [skató] belongs to the non-neutral vocabulary of SMG, since it refers to a taboo element (‘excrement, faeces’).95 It derives from the lexical stem σκατ- [skat] of the Ancient Greek substantive σκῶρ [skôr] (genitive σκατὸς [skatós]) ‘excrement, ordure, of men and cattle’. During the Middle Ages, the plural form σκατὰ skatá acquired the metaphorical pejorative meaning ‘bullshit, nonsense’ (see Kriaras, entry σκατόν 2). Today, the lexeme skató, further to its literal meaning ‘shit’ (see Example 58) has a range of new metaphorical meanings; it may be used (generally in the plural form skatá) to present something as miserable, insignificant, or worthless (see Example 59); or (in the singular form) to show that someone is too young and therefore inexperienced, but nevertheless cheeky, presumptuous, and arrogant (see Example 60). Note that in both its literal and metaphorical uses, the word typically conveys strong negative feelings on the part of the speaker, such as strong dislike or aversion towards the referent. (58)

ta pezoðrómia ítan ʝemáta me skatá scílon96 the pavements were full.PL with shit of.dogs ‘The pavements were full of dog shit’

(59)

SMG [LNEG, entry σκατό] pái ce ksoðévi ta leftá tis s aftá ta skatá Goes and spends the money her in these the shit.PL.MET ‘She spends her money on this shit’

(60)

SMG [LNEG, entry σκατό] éna tóso ða skató ce vɣázi ce ɣlósa a such.a PRTCL shit.SG.MET and sticks.out and tongue ‘Such a shit, and he answers back!’

Compounds with first constituent skato- (compiled from the stem of the lexeme skat-ó and the compound vowel -o-) make their first appearance in ancient times contributing the literal meaning of σκῶρ, genitive σκατὸς ‘excre-

94 95

96

Θέλεις όμως ένα βρομόξυλο, είπε γελώντας, πολύ ξύλο θέλεις για να συνέλθεις (Douka Ploti poli: 263). See, e.g., LKN where the term is marked as vulgar or derogatory in both its literal and metaphorical senses; also LNEG where skató is marked with an (!) used to mark vulgar or taboo terms Personally attested by M. Konstantinidou.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

151

ment’, for example, Ancient/Koine Greek σκατοφάγος [skatofáɣos] ‘eating dung or dirt’, σκατοφόροι [Skatofóroi] ‘dung-carriers’ (see LSJ, s.v.). In Medieval and Early Modern Greek, we have many attestations of compounds with first constituent skato- that convey a derogatory meaning towards the intended referent (see Examples 61–67, which are all taken from Kriaras). The attested compounds are both endocentric (e.g., 61 to 65) and exocentric97 (e.g., 66 and 67). Moreover, skato- occurs in literal (e.g., 67) as well as in non-literal uses (e.g., 61, 62). (61)

σκατόγερος [skatóʝeros] (Spanos 145; 16th century edition of a 14th–15th century text) ‘abusive characterization of an aged man’

(62)

σκατόγραια [skatóɣrea] (Poulologos 39 app.; 14th century) ‘abusive characterization of an aged woman’

(63)

σκατοποντικός [skatopondikós] (Diigisis Pediofrastos 144; 14th century) ‘abusive characterization, filthy, disgusting mouse’

(64)

σκατόχοιρος [skatóçiros] (Diigisis Pediofrastos 436; 14th century) ‘abusive characterization, pig that lives in the dirt and feeds on them, dirty pig’

(65)

σκατογάδαρος [skatoɣáðaros] (Diigisis Pediofrastos 747; 14th century) ‘abusive characterization of a donkey’

(66)

σκατοπρόσωπος [skatoprósopos] (Spanos 556; 16th century edition of a 14th–15th century text) ‘abusive characterization, filthy, ugly-faced’

(67)

σκατογένης [skatoʝénis] (Spanos 152; 16th century edition of a 14th–15th century text) ‘abusive characterization, one who has impurities in his beard’

In SMG, first constituent skato- is literally but, mostly, non-literally employed to form a multitude of derogatory compounds of insult, more often endocentric (e.g., 68–78) and less often exocentric (e.g., 79–80) (see LKN and LNEG, entry σκατο-; Kamilaki et al. 2016: 156).98 Regarding the semantics of SMG skato- the following distinctions can be made. When combined with a neutral base (i.e., a base that conveys no expressivity), skato- as an item derived from a non-neutral (taboo) lexeme constantly carries a mixed semantics, that is, it contributes both descriptive and expressive content, and this is so regardless of whether skato- is used literally or non-literally. By the literal use of skato- the speaker ascribes to the base of the compound the property (a):

97 98

For the terminological distinction see §6.3.1. See also Christopoulou (2016: sections 4.2.4 and 6.6.1.3) for an elaborate discussion of the functions of skato- in the ‘slang vocabulary’ of Modern Greek, i.e., in the vocabulary used by marginalized social groups.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

152

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(a) ‘related with excrements’,99 e.g., (68); see also case (75); (68)

SMG [LNEG, entry σκατόμυγα 1] skat-ó-miɣa shit-CM-fly.N ‘a large fly that feeds on faeces and transmits infectious diseases [+PEJ]’

whereas by using skato- non-literally the speaker ascribes to the (expressively neutral) base the properties (b) or (c): (b) for [–animate] referents: ‘bad, adverse, unpleasant, troublesome’, e.g., cases from (69) to (74); (69)

skat-o-mixánima100 shit.MET-CM-machinery ‘shitty/fucking machinery (i.e., computer)’

(70)

skat-o-cinitó101 shit.MET-CM-mobile.N ‘shitty/fucking mobile phone’

(71)

skat-ó-mɲalo102 shit.MET-CM-mind ‘shitty/fucking/dirty mind’

(72)

skat-o-eforía103 shit.MET-CM-tax.office ‘shitty/fucking tax office’

(73)

skat-o-ðuʎá104 shit.MET-CM-job ‘shitty/fucking job’

(74)

skat-o-kánal-o105 shit.MET-CM-channel-SUFF ‘shitty/fucking TV channel’

99

100 101 102 103 104 105

Note that the seemingly unmarked term skatoloʝía ‘the frequent use of the word shit or other related words in speech’ (see LKN, LNEG, entry σκατολογία) is a neoclassical compound (for this term see, e.g., Ralli 2013: 151–2) formed on the basis of stems of Ancient Greek origin (σκατο- + -λογία), either directly or indirectly, through a west European language (e.g., French scatologie, English scatology), and in that sense it does not fall within this classification pattern. The same applies to the medical term skatofaʝía ‘the pathological condition of eating faeces’ (see LNEG, entry σκατοφαγία) < Ancient Greek σκατο- + -φαγία; see also English skatophagy. Μας έχουνε βάλει σκατομηχανήματα που δεν δουλεύουνε. Example personally attested by M. Konstantinidou. Σβήσε τον αριθμό μου από το σκατοκινητό σου! ‘Delete my phone number from your shitty mobile phone!’ At https://bit.ly/3yTtE0G (accessed 30 December 2022). Λέει ό,τι του κατέβει στο σκατόμυαλο του! At https://katohika.gr/gn-blog/milisoski-skpia-giftovergina (accessed 8 November 2021). Εγώ δεν είμαι εργαζόμενος που δίνω στη σκατοεφορία το 50% και στους υπαλλήλους μου και στο κράτος. At www.press-gr.com/2009/10/blog-post_8450.html (accessed 8 November 2021). Πρέπει να χαμογελάνε και να είναι ευχάριστοι, μολονότι η σκατοδουλειά που κάνουν είναι κουραστική και κακοπληρωμένη. At https://bit.ly/3ZeYoV9 (accessed 15 November 2021). Ύπουλο σκατοκάναλο ‘sneaky shitty TV channel’. At http://www.press-gr.com/2009/10/ george-galloway.html (accessed 15 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

153

(c) for [+animate]/[+human] referents: ‘bad (from a moral point of view), despicable, cruel, odious, loathsome, obnoxious’, or in general ‘disgusting’, for example, cases from (75) to (80). (75)

SMG [LNEG, entry σκατόμυγα 1] skat-ó-miɣa shit.MET-CM-fly.Ν ‘fly [+PEJ], shitty/fucking fly’

(76)

skat-o-ʝatrós106 shit.MET-CM-physician ‘shitty/fucking doctor/physician’

(77)

skat-ánθropos107 shit.MET-man ‘shitty/fucking guy’

(78)

skat-o-afendikó108 shit.MET-CM-boss ‘shitty/fucking boss’

(79)

skat-ó-mɲal-os109 shit.MET-CM-brain/mind-SUFF ‘shitty-brained (person)’

(80)

skat-ó-psix-os110 shit.MET-CM-soul-SUFF ‘dirty-souled (person)’

While properties (a), (b), and (c) constitute the descriptive meaning of the first constituent skato-, the feeling of strong annoyance, irritation, the attitude of deep deprecation, or contempt on the part of the speaker towards the intended referent constitutes its expressive meaning aspect. On the other hand, as observed in the cases of vlaxo-, ʝifto-, vromo- above, as well as paʎo- below, when combined with a non-neutral base, mostly an insulting or otherwise expressively charged term, skato- seems to contribute exclusively expressive content, namely only a strong negative feeling, an attitude of contempt, deprecation, annoyance, or disgust directed against the referent of the compound on the part of the speaker. This negative feeling or attitude being added to the inherent negativity of the base enhances the pejorative meaning of the base and of the compound in general (see cases 81 to 86).

106 107 108 109

Μας πήρε πολλά αυτός ο σκατογιατρός! Personally attested by M. Konstantinidou. At https://bit.ly/3KvYiUT (accessed 8 November 2021). Σκατοαφεντικά! Σκατοτράπεζες! Δολοφόνοιιιι. At https://bit.ly/3Isck7B (accessed 8 November 2021). 110 At https://bit.ly/3SytDs7 7 December 2018 (accessed 8 November 2021). See fn 109.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

154

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(81)

skat-o-ceratás111 shit.EXPL.PEJ-CM-cuckold ‘shitty cuckold [+INTENSE.PEJ]’

(82)

skat-o-fasístas112 shit.EXPL.PEJ-CM-fascist ‘shitty fascist [+INTENSE.PEJ]’

(83)

skat-ó-vlaxos113 shit.EXPL.PEJ-CM-Aromanian.MET.PEJ ‘shitty bumpkin’

(84)

skat-o-ɣúrun-o114 shit.EXPL.PEJ-CM-pig.MET.PEJ-SUFF ‘(figuratively, applied to a person) shitty pig [+INTENSE.PEJ]’

(85)

skat-o-ʝelíos115 shit.EXPL.PEJ-CM-ridiculous/ludicrous ‘shitty ludicrous’

(86)

skat-o-ilíθios116 shit.EXPL.PEJ-CM-idiot ‘shitty idiot’

Just like paʎo- (see §6.3.5), first constituent skato- in the functions mentioned above, appears to be very productive in SMG since the relevant compounds are abundantly attested in computer-mediated everyday conversations. Moreover, skato- is treated as an autonomous dictionary entry by major lexicographical works of SMG, such as LKN, LNEG, and XLNEG. Finally, it should be noticed that the phonologically conditioned linking vowel -o- is realized even though the second constituent begins with a vowel weaker than /o/, for example, an /e/ or /a/ (see cases 72 and 78), such as in the cases of loose compounding. However, there are also cases like Example (77), where -o-, following the norm (see §6.3), is eliminated when the second constituent begins with an /a/. 6.3.5

Expressive compounding with first constituent paʎo- ‘old’

The SMG first constituent παλιο- [paʎo] originates from the older unreduced form παλαιο- [palaio] from Ancient/Koine Greek adjective παλαιὸς [palaiós] ‘old in years (of persons or things)’, ‘of old date, ancient (of persons or things)’, also in a negative sense ‘a dotard’ or ‘antiquated, obsolete, worn by use, worn-out, useless, worthless’ (see LSJ, LSK, Thayer 1995, Strong 1890: 111 112 113 114 115 116

Με γέλασε όμως ο σκατοκερατάς. (Fleischer & Stergellis Maidonis: 71). At https://bit.ly/3Ez0vv5 (accessed 8 November 2021). At https://bit.ly/3LFIH5N (accessed 8 November 2021). Ναι, ορέ σκατογούρουνο. Έτσι δεν ντύνονται κι οι βρομιάρηδες οι πρίντζιπές σας; (Detsikas Karaiskakis: 349). Mία σκατογελοία κυβέρνηση ‘a shitty ludicrous government’. At https://sarantakos.wordpress .com/2020/12/05/meze-456/ (accessed 8 November 2021). At https://osr55.wordpress.com/2012/09/12/ (accessed 8 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

155

entry παλαιός). It is attested (also in the form paleo-) from the fourteenth century ce in a series of compounds as a means of contributing negative expressivity (pejoration); see cases from (88) to (94) marked lexicographically (see Kriaras, s.v.) as derogatory or abusive, either by usage labels or by description of their negativity within the definition.117 Middle as well as late Middle Greek derogatory/abusive compounds with first constituent paleo-, paʎo- display a nominal (e.g., 88–90 and 93–94) or an adjectival base (e.g., 91, 92) and are related either to a [+animate] referent (e.g., 88, 89, 91, 92) or to a [–animate] one (e.g., 93, 94). Notice that in Middle Greek, apart from the reduced expressive form παλιο- paʎo-, the older unreduced form Παλαιοpaleo- may also contribute a derogatory meaning aspect (see cases in 90, 91, 93), and that unlike SMG [+learned] unreduced form παλαιο- paleo-, for the neutral semantics of which, see footnote 131. It is also to be noticed that in the late Middle Ages the adjective παλαιὸς [paleós], παλιὸς [paʎós] is also encountered as an abusive term, ascribing to the referent of the noun modified by it (mostly a person) the properties of being ‘contemptible, vile, worthless’ and indicating the disrespectful or abusive attitude of the speaker for referent (see the case in 87). (87)

Late Middle Greek [Katzourbos Ε΄ 40, 17th century; see Kriaras, entry παλαιός 2e] τῆς Πουλισένας, τῆς παλιᾶς πουτάνας tis pulisénas, tis paʎás putánas ‘of Pulisena, the vile whore’

The non-neutral meaning aspects of the lexeme palaiós, paleós, paʎós traced in Ancient/Koine and (late) Medieval Greek appear to be the basis for the observed negative expressivity of the first constituent paleo-/paʎo- both in Middle and late Middle Greek, as exemplified in cases (88)–(94), all taken from Kriaras. (88)

παλιοτσουτσουβάχα [paʎotsutsuváxa] (Sachlikis 77; 14th century) ‘abusive characterization for a loose woman’ (second constituent τσουτσουβάχα < Venetian chiuchiavache ‘a nocturnal bird’, in metaphorical use)

(89)

παλιοδρεπανομύτης [paʎoðrepanomítis] (Poulologos 24 app.; 14th century) ‘abusive characterization for one who has a sickle-shaped nose’ (second constituent δρεπανομύτης ‘one who has a sickle-shaped nose’)

(90)

παλαιοξέρασμα [paleoksérazma] (Poulologos 155; 14th century) ‘metaphorically, as abusive characterization, the one who is disgusting’ (second constituent ξέρασμα ‘vomit, nausea’)

(91)

παλαιοξερασμένος [paleokserazménos] (Poulologos 99 app.; 14th century) ‘abusive characterization, disgusted’ (second constituent ξερασμένος ‘abusive characterization, nauseating, disgusted, disgusting’; see Kriaras, entry ξερνώ)

117

For a brief account of derogatory labelling see §6.3, and for details see Efthymiou et al. (2013).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

156

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(92)

παλιογιβεντισμένη [paʎoʝivendizméni] (Sachlikis 403; 14th century) ‘of a prostitute, humiliated, humbled’ (second constituent γιβεντισμένη ‘ridiculed’)

(93)

παλαιοκόσκινο [paleokóscino] (Krasopateras I 12; 16th–18th century edition of a 15th century text) ‘abusively, the sieve’ (second constituent κόσκινο ‘sieve’)

(94)

παλιομπάλωμα [paʎobáloma] (Katzοurbos Ε΄ 408; 17th century) ‘derogatory/abusively, old and useless piece of fabric for patchwork’ (see Kriaras, s.v.; second constituent μπάλωμα ‘piece of fabric for patchwork’)

Nowadays first constituent paʎo- still functions as a ‘pejorativizer’118 of the meaning of the elements that constitute the second constituents of SMG compounds, despite the fact that the attested negative meaning aspects of the Ancient/Koine/Medieval Greek lexeme palaiós, paleós, paʎós are no longer retained in SMG lexeme paʎós (see in this regard below). Specifically, it is found combined mostly with nominal (e.g., 95–116) and rarely with adjectival bases (e.g., 117–119) and forming numerous compounds, mainly endocentric (e.g., 95–103, 105–110) but also exocentric (e.g., 104). In combination with an expressively neutral base, apart from its descriptive (cancellable) content, which we will specify below, paʎo- adds a derogatory, non-cancellable (i.e., surviving under negation),119 meaning aspect to the head of the compound. In other words, it indicates that an attitude of contempt, a feeling of annoyance or displeasure is present on the part of the speaker towards the referent, or, more concretely, following the specification made by Croom (2011: 353), towards some set of the referent’s properties. It should be noted that there are many expressive compounds with first constituent paʎo- attested in SMG dictionaries (see LKN, LNEG, XLNEG), treated as autonomous lexical entries, such as the cases from (95) to (100). Those having [+animate] second constituents are used as insulting terms (e.g., 95–97), as long as paʎo- ascribes to the referents denoted by the second constituents the notion of ‘(morally) bad, worthless’, whereas the others having [–animate] second constituents are used as non-neutral (pejorative) terms (cases 98–100), as long as paʎoascribes to the referents the notion ‘bad’, ‘troublesome’, or ‘undesirable’. (95)

paʎ-ánθropos old.MET.PEJ-person ‘malicious man, wretched, bad guy’

(96)

paʎ-ó-peð-o old.MET.PEJ-CM-child-SUFF ‘naughty boy’

(97)

paʎ-ó-scil-o old.MET.PEJ-CM-dog-SUFF ‘cur, damn dog’

(98)

paʎ-ó-ðromos old.MET.PEJ-CM-road ‘abusively, road in a bad condition’.

(99)

paʎ-ó-ceros old.MET.PEJ-CM-weather ‘bad weather’

(100)

paʎ-o-ðuʎá old.MET.PEJ-CM-job ‘difficult job, reprehensible activity’

118 119

The term is borrowed from Gutzmann and McCready (2016: 88 fn 9). See Fortin (2011), Calvo (2020: 101).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

157

Apart from these fixed lexicalized cases many other relevant examples, such as those presented in cases from (101) to (123), are identified very easily in everyday conversations and other informal sources – an indicator for the great productivity of this type of compounding in SMG. From the semantic analysis of these cases, it becomes evident that, by using the first constituent paʎo-, the speaker ascribes to the referent denoted by the second constituent the properties/the characteristic features (a) and (b): (a) for [+animate]/[+human] referent: ‘being worthless’, see Examples (101)– (104); (101)

paʎ-ó-turkos120 old.MET.PEJ-CM-Turk ‘old Turk [+PEJ], damn Turk’

(102)

paʎ-o-evropéos121 old.MET.PEJ-CM-European ‘old European [+PEJ]’

(103)

paʎ-o-civérnisi122 old.MET.PEJ-CM-government ‘old government [+PEJ]’

(104)

paʎ-ó-stom-os123 old.MET.PEJ-CM-mouth-SUFF ‘bad mouthy/Rabelaisian [+PEJ]’

and (b) for [–animate] referent: ‘being bad, troublesome’, ‘causing difficulty or annoyance’ to the speaker, see Examples (105)–(110). (105)

paʎ-o-iós124 old.MET.PEJ-CM-virus ‘old virus (in this instance, the coronavirus CoVid-19) [+PEJ]’

(106)

a. paʎ-aróstça125 old.MET.PEJ-malady ‘old malady (in this instance, the coronavirus CoVid-19) [+PEJ]’ b. paʎ-o-asθénia126 old.MET.PEJ-CM-disease ‘old disease (in this instance, cancer) [+PEJ]’

120 121 122 123 124 125 126

Τι λέτε, μωρ’ παλιότουρκοι και σεις παλιοζαγάρια; (folk song Tis Lenos Botsari). At www.sarajevomag.net/entipa/teuhos_103/i103_p09_geo.html (accessed 8 November 2021). At www.sarajevomag.net/entipa/teuhos_103/i103_p09_geo.html (accessed 8 November 2021). Μίλα μόνο με θετικά λόγια (θα ζοριστείς παλιόστομε το ξέρω). At https://bit.ly/3KxyQhW (accessed 8 November 2021). Το φάρμακο γι’ αυτόν τον παλιοϊό είμαστε εμείς! Αν Καθίσουμε στα σπίτια μας. SKAI RadioJournal, 25 March 2021. Δεν μπορείς να παίξεις με την παλιαρρώστια! ERT1 Radio-Journal, 19 August 2021. Είναι πράγματι τρομερό αυτό που συμβαίνει με την παλιοασθένεια. . . Ο καρκίνος φαίνεται πως «προτιμά» κάποιες περιοχές της Ελλάδας. At www.makeleio.gr 29 July 2015 (accessed 8 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

158

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(107)

paʎ-o-eforía127 old.MET.PEJ-CM-tax.office ‘old tax office [+PEJ]’

(108)

paʎ-o-leoforío128 old.MET.PEJ-CM-bus ‘old bus [+PEJ]’

(109)

paʎ-ó-kras-o129 old.MET.PEJ-CM-wine-SUFF ‘plonk, cheap wine [+PEJ]’

(110)

paʎ-o-vivlío130 old.MET.PEJ-CM-book ‘old book [+PEJ]’

These notions constitute the descriptive (truth-conditional) meaning aspect of the first constituent paʎo-, conveyed along with the aforementioned expressive (derogatory) meaning aspect, when paʎo- is combined with a neutral base-word. It is important to point out that in all cases mentioned above, whether lexicalized or not (95–110), paʎo- is not used literally to convey the notion of ‘oldness’, of ‘being old’, ‘bygone’, or ‘old-time’, typically connected with the lexeme paʎós in SMG.131 On the contrary, it is used to convey other non-literal/extended meanings, such as ‘bad’, ‘worthless’, ‘unpleasant’, ‘troublesome’, annoying’, which, it is worth noting, are not typically connected with the SMG paʎós. A similar observation is made by Stephany and Thomadaki (2017: 120), who also point out that in the family of SMG paʎo- compounds the adjective paʎós has developed pejorative meanings (‘unworthy’, ‘bad’) as part of the construction, that is, which have not originated from the meaning of the lexeme.132 This finding leads us to assume that 127

128 129 130 131

132

Η παλιοεφορία (πολύ επιεικής ο χαρακτηρισμός!!!), το μόνο που έχει κατά νου είναι πως θα μας πάρει με κάθε τρόπο και το τελευταίο Ευρώ. At https://bit.ly/3Zjnmml (accessed 8 November 2021). Για να μιλάς για ταλαιπωρία πρέπει να έχεις στριμωχτεί εν καιρώ κορωνοϊού σε ένα παλιολεωφορείο. At https://bit.ly/3FBQsWt (accessed 8 November 2021). Mας υποσχέθηκαν ότι θα πιούμε ένα παλιό καλό κρασί, αλλά μας σέρβιραν ένα παλιόκρασο. At https://bit.ly/3YW5aPb (accessed 8 November 2021). Με κάτι τέτοια παλιοβιβλία διδάσκουνε τα Θρησκευτικά στα παιδιά μας. At https://bit.ly/ 3Iu4zho (accessed 8 November 2021). The notion of ‘oldness’ is notably conveyed in SMG by the [+learned] unreduced first constituent form paleo-, which is usually expressively neutral, contributing only descriptive (truth-conditional) content (see, e.g., neoclassical compounds pale-o-vivliopolío old-CMbookshop ‘second-hand bookshop’, pale-o-komatizmós old-CM-partisanship ‘old fashioned partisanship’, pale-o-ɣrafía old-CM-description.of ‘palaeography’). For the usually neutral status of [+learned] forms see Kamilaki et al. (2016). The [+learned] variation paleo- is directly derived from the Ancient Greek first constituent παλαιο- (< Ancient Greek adj. παλαιὸς [palaiós]) or from the Ancient Greek originated international term palaeo-, paleo(see, e.g., French paléographie < paléo- + -graphie). For the fundamental insight that ‘a construction may have holistic properties that are not derivable from the properties of its constituents and/or its structure’ see, e.g., Booij (2013: 260) quoted by Stephany & Thomadaki (2017: 120); for Greek, see Ralli (2013: 258).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

159

the derogatory meaning aspect of paʎo- in SMG may be directly inherited from the attested non-literal expressive (derogatory) uses of Middle and late Middle Greek first constituent paleo-, paʎo- (see cases 88–94). This is however a point deserving further investigation. Consider in advance that, whereas the SMG nominal phrase παλιό κρασί [paʎó krasí] ‘old wine’ is a semantically neutral phrase with a rather meliorative meaning aspect (‘aged’ as ‘good’ or ‘rare’), the compound παλιόκρασο [paʎókraso] (109), compounded from the same lexemes as the aforementioned phrase, that is, paʎós ‘old’ and krasí ‘wine’, is typically an expressive (derogatory) compound conveying the meaning ‘of bad quality, cheap wine’ towards which the speaker expresses a negative feeling (pejoration) similar to that conveyed by the English phrase cheap old plonk. Αt present, we take for granted that SMG compounds with first constituent paʎo- combined with expressively neutral second constituents constantly carry a mixed semantics; along with the descriptive meaning ‘bad, worthless, troublesome’, an expressive (derogatory) meaning aspect is typically contributed as well. This expressive aspect vanishes when, instead of the compound pattern with paʎo-, an expressively neutral transcription/counterpart of it is used, for example, instead of the pejorative paʎókraso ‘plonk, cheap wine [+PEJ]’ (109) a neutral syntactic phrase, such as kacís piótitas krasí ‘of bad quality wine’ or fθinó krasí ‘low-priced wine’, is used. On the other hand, when combined with non-neutral, namely negative second constituents (see cases from 111 to 123) paʎo-, just like vlaxo-, ʝifto- vromo-, skato-, appears in the role of a pure pejorative intensifier/amplifier, that is, as an element that strengthens the inherent/conventionalized expressive (pejorative) meaning aspect of the second constituent, without conveying any descriptivity at all.133 The negative expressivity of paʎo- repeats, in a way, the negative expressivity conveyed by the second constituent and makes it thereby stronger. Thus, an intensely negative attitude, a strong feeling of contempt or deprecation towards the referent is expressed on the part of the speaker through this morphological device (paʎo-); this strong negative feeling or attitude represents the expressive meaning aspect of the respective paʎo- compounds. On the other hand, the descriptive non-literal/extended meanings of paʎo- (‘bad’, ‘worthless’, ‘unpleasant’, ‘troublesome’, ‘annoying’) attested above, that is, when paʎocombines with expressively neutral second constituents, seem to be blocked as irrelevant when it combines with expressively non-neutral bases.134 In this sense, paʎo- is to be treated as an expletive item, that is, as an expletive pejorative that contributes solely expressive [+negative] content. The expressive [+PEJ] meaning aspect of the second constituents in cases (111–113, 123) is based on the conventionalized metaphorical usages of the 133 134

See the related discussions with regards to vlaxo- and vromo- (§6.3.1 and §6.3.3). See also §6.3.1 on this subject.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

160

Kallergi, Katsouda, Konstantinidou, Tsangalidis

second constituents skrófa (literally ‘female.pig’ > figuratively ‘bitch’, ‘prostitute’; LNEG, XLNEG), ɣurúni (literally ‘pig’ > figuratively ‘uncouth or fat person’; LKN), and vláxos (literally ‘Aromanian/Vlach’ > figuratively ‘provincial’, ‘bumpkin’; LNEG, XLNEG), respectively. In Example (114), it is the pejorative suffixation that charges the base-word with negative meaning aspects, whereas in cases from (115) to (122) the expressivity of the second constituents seems to relate with the insulting or slurring status of the respective lexemes in SMG.135 That is, the second constituents may be used per se as autonomous or independent insulting terms (e.g., Skrófa! Γurúni! Vláxo! Kumúni! Fasísta! Maláka! Sixaméne! Vlaméne! etc.) since to the referents of such terms are ascribed stereotypical negative properties by those speakers who use them to insult.136 It is also worth noting that paʎo- compounds of this type are often used in combination with the insulting or slurring terms, forming a reduplication pattern of the type [X paʎo-X], as in Example (123). (111)

paʎ-o-skrófa137 old.EXPL.PEJ-CM-fimale.pig.MET.PEJ ‘old bitch, slut [+INTENSE.PEJ]’

(112)

paʎ-o-ɣúrun-o138 old.EXPL.PEJ-CM-pig.MET.PEJ-SUFF ‘old pig [+INTENSE.PEJ]’

(113)

paʎ-ó-vlaxos139 old.EXPL.PEJ-CM-Aromanian.MET.PEJ ‘a fucking bumpkin [+INTENSE.PEJ]’

(114)

paʎ-o-kumún-i140 old.EXPL.PEJ-CM-communist-DIM.PEJ ‘fucking/old commie [+INTENSE.PEJ]’

(115)

paʎ-o-fasístas141 old.EXPL.PEJ-CM-fascist.PEJ ‘fucking/old fascist [+INTENSE.PEJ]’

(116)

paʎ-o-malákas142 old.EXPL.PEJ-CM-masturbator.PEJ ‘a fucking asshole/wanker [+INTENSE.PEJ]’

135 136 137 138 139 140 141 142

Regarding the stark negative expressivity of the base-word fasístas in case (115), see the questionnaire-based survey of Xydopoulos and Christopoulou (2019). This position conforms with the relevant argumentation by Meibauer (2013: 31). Στον αγύριστο, παλιοσκρόφα! (Fakinou Astradeni: 109). See also LKN, entry παλιοσκρόφα. Ρε παλιογουρουνο, τι λες τώρα; At www.jimnyclub.com/forum/index.php?topic=717.0 20 July 2006 (accessed 8 November 2021). Άιντε να χαθείτε Από’δώ μέσα, παλιόβλαχοι! (Greek comedy movie O Fandazmenos, 1973; with Lambros Konstantaras). At www.rizospastis.gr/story.do?id=5725865 (accessed 8 November 2021). At https://bit.ly/3ITCykG (accessed 8 November 2021). Εξελίσσεται σε ‘παλιομαλάκας’ όταν βρίσκεται προς το τέλος της στράτευσης. At https://bit.ly/ 3lVSDgC (accessed 8 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek (117)

paʎ-o-sixaménos143 old.EXPL.PEJ-CM-disgusting ‘damn disgusting [+INTENSE.PEJ]’

(118)

paʎ-o-vlaménos144 old.EXPL.PEJ-CM-stupid ‘bloody idiot [+INTENSE.PEJ]’

(119)

paʎ-ó-xodros145 old.EXPL.PEJ-CM-fat ‘damn fat guy [+INTENSE.PEJ]’

(120)

paʎ-o-rezílis146 old.EXPL.PEJ-CM-laughingstock.PEJ ‘bloody laughingstock [+INTENSE.PEJ]’

(121)

paʎ-o-kléftis147 old.EXPL.PEJ-CM-thief ‘old thief [+INTENSE.PEJ]’

(122)

paʎ-o-kleft-ar-éi148 old.EXPL.PEJ-CM-thief-AUG-SUFF.PEJ ‘old fucking thieves [+INTENSE.PEJ]’

(123)

ái skrófa, paʎ-o-skrófa149 get.lost fimale.pig.MET.PEJ old.EXPL.PEJ-CM-fimale.pig.MET.PEJ ‘Get lost, you bitch, old bitch! [+INTENSE.PEJ]’

161

The intensifying meaning aspect of paʎo- when combined with non-neutral bases (abusive/insulting terms, etc.) is traced back to Middle Greek, as can be seen in cases such as (90), (91), and (92). This pattern of SMG compounding could be treated as similar to expressives such as fucking bastard in English, where fucking intensifies the expletive pejorative bastard.150 Finally, one last remark regarding first constituent paʎo- should be made from a morphological point of view. The linking vowel -o- is exceptionally preserved between the two members of the compound, even if the second constituent begins with a vowel stronger than /o/, that is, with an /a/ (see, e.g., 143

144 145 146 147 148

149 150

Βρε παλιοσιχαμένε. ‘You damn disgusting one!’ (Karistiani Suel: 133). Δε διάβαζε αυτός ένα μήνα για να πας στο τέλος να του αντιγράψεις εσύ παλιοσιχαμένε, ε παλιοσιχαμένε! At http:// ariscap.blogspot.com/2014/05/blog-post.html (accessed 8 November 2021). Παλιοβλαμμένε, λες βλακείες! ‘You bloody idiot, you are talking nonsense!’ At https://bit.ly/ 3FEK9BM (accessed 8 November 2021). Εμείς δεν τρώμε μπριτζόλες, γρούνια, αρνιά και λουκάνικα, για να γίνουμε σαν τον παλιόχοντρο από την Καρδίτσα. . . At https://bit.ly/3IPZCke (accessed 8 November 2021). Example personally attested by M. Konstantinidou. Μὴν τὸν ψηφίσετε, γιατ’ εἶναι παλιοκλέφτης! (Loukatos Laografika: 68). Ντροπή σας ληστές, απατεώνες, παλιοκλεφταραίοι, αποβράσματα. At www.oparlapipas.gr/ 2013/11/blog-post_1193.html (accessed 8 November 2021). Παλιοκλέφτες! Παλιοκλεφταραίοι! (Greek film Fonazi o Kleftis, 1965; with Dinos Eliopoulos). Ἄι, σκρόφα, παλιοσκρόφα! (Papachristodoulou Saranta Ekklisies: 220). See Gutzmann and McCready (2016: 84), also Geurts (2007) quoted in Gutzmann and McCready (2016).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

162

Kallergi, Katsouda, Konstantinidou, Tsangalidis

106b or with an /e/ (see, e.g., 102, 107), just like in the cases of compounds with a loose structure; yet, when the second constituent begins with an /a/, there are also typical cases attested where -o- is deleted on phonological grounds (see, e.g., 95, 106a). 6.3.6

Summary of expressive compounding

In §6.3, we have presented and analysed five representative cases of first compound constituents that contribute pejoration, that is, vlaxo- ‘Aromanian/ Vlach.MET’, ʝifto- ‘Gypsy’, vromo- ‘stink/dirt’, skato- ‘shit’, and paʎo- ‘old’. The expressive first compound constituents studied are derived from nouns (vláxos, ʝíftos, vróma, skató) and from adjectives (paʎós) and are combined most frequently with nouns and less frequently with adjectives (or perfect passive participles behaving as adjectives), producing either nominal or adjectival compounds. We have also noted that, with respect to the presence or absence of the linking vowel -o-, the expressive compounds in question exhibit both possibilities in a similar way to other types of SMG compounds, but the sometimes non-phonologically motivated presence of -o- points to characteristics of a loose compound structure (see Ralli 2013). It has been observed that the first constituents studied, in addition to their descriptive meanings, systematically encode expressivity, partly by non-literal (metaphorical/extended) usage (i.e., [Constituent1.MET.PEJ]) and partly by either literal (i.e., [Constituent1.PEJ]) or non-literal usage. Their expressivity is based on the expressivity attached to the lexemes from which they derive, mostly when these are used non-literally (e.g., vlaxo- being expressive only by non-literal use),151 but also by literal usage (e.g., skato- or vromo-), except for the case of paʎo-. The expressivity of vlaxo- is based on the negative expressivity of the lexeme vláxos ‘Aromanian/Vlach’ when used metaphorically, whereas the expressivity of ʝifto-, vromo- and skato- is based on the negative expressivity of the respective lexemes ʝíftos ‘Romany[+PEJ]/Gyspy’, vróma ‘stink/dirt’ and skató (plural skatá) ‘shit’ whether literally or figuratively used. Regarding the pejorative aspect conveyed by paʎo- ‘old’, this is presumably inherited from Medieval Greek since the SMG lexeme paʎós does not contribute any negative expressive content. Regarding the status of the second compound constituent with respect to expressivity, we distinguished the following combination patterns: Pattern 1: A negative expressive form [Constituent1] combined with an expressively neutral form [Constituent2] produces a negative expressive compound (124):

151

This finding conforms with respective insights by Meibauer (2013).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek (124)

163

[Constituent1.PEJ]-[CM]-[Constituent2.NEUTRAL] > [Compound.PEJ] e.g. [stem skat(ó) ‘shit’]-[o]-[noun eforía ‘tax office’] > [skatoeforía ‘shitty/fucking tax office’]

Pattern 2: A negative expressive form [Constituent1] combined with an expressively non-neutral, namely, a negative expressive form [Constituent2] produces a highly/ intensely negative expressive form (125): (125)

[Constituent1.PEJ]-[CM]-[Constituent2.PEJ] > [Compound.INTENSE.PEJ] e.g. [stem skat(ó) ‘shit’]-[o]-[noun fasístas ‘fascist’] > [skatofasístas ‘shitty fascist’]

Overall, the use of negative expressive first constituents, such as those described in §6.3, is a very common way to ‘pejorativize’, that is, to contribute negative (pejorative) meaning to a neutral base or to amplify or strengthen the negative expressivity of a non-neutral base. 6.4

Expressive derivation in SMG – Pejoration by suffixation

There is a very extensive number of suffixes in SMG, since it is a highly inflected language, where the majority of words are formed on the basis of a stem+affixes combination. Those derivational suffixes that express pejoration constitute a rather heterogeneous category, in at least two senses. First, a portion of them inherently carry some negative expressivity. As Dressler and Merlini Barbaresi (1994: 92) observe, ‘pejoratives [. . .] have a standard negative connotation’ and this seems to be part of their meaning. Arguably, this group of suffixes also have an exclusively pejorative expressive function, in that the negative expressivity is always present in the meaning of their derivatives, and any positive connotations or neutral effects are excluded (see, e.g., -ácas below). On the other hand, many of the suffixes discussed here produce words with positive or neutral meanings as well. This difference seems to partly correlate with the semantic type of the suffix in question. Generally, the suffixes discussed can be categorized into three groups according to their basic semantics (the descriptive meaning they contribute to the derivation). One large group includes nominalizers which add the meaning ‘human characteristic/activity, personal trait of a more or less general property’ (as in spir-ʝáris pimple-SUFF ‘a pimply person [+PEJ]’ or ipurʝ-ilíci minister-SUFF ‘the occupation of a minister [+PEJ]’). This group largely exhibits pejorative functions and almost half of the cases included here are inherently negatively expressive. In the second group, one may also find several suffixes that refer to characteristics or properties and express negative feelings and attitudes, but these suffixes have another primary function, such as augmentation, intensification or diminution (e.g., -arú as in but-arú thigh-SUFF ‘woman

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

164

Kallergi, Katsouda, Konstantinidou, Tsangalidis

with big thighs’). As has been observed (e.g., Dressler and Merlini Barbaresi 1994, Efthymiou 2003), augmentative and intensive affixes may generally have both positive and negative connotations, such as approval or admiration (cf. amaks-ára car-AUG ‘big and/or nice car’, oló-aspros AUG-white ‘totally white’) and displeasure respectively (e.g., θeó-stenos AUG-narrow ‘too narrow’). In fact, augmentatives, as well as diminutives, have been argued to present contradictory meanings across languages (e.g., Jurafsky 1996). For example, diminutives are known for their involvement in expressive meanings, such as endearment (see, e.g., Dressler & Merlini Barbaresi 1994, Jurafsky 1996), but their size semantics (the meaning SMALL) may also extend to attenuation or belittlement. Thus, for example, the diminutive suffix combination -uðáci may have pejorative or meliorative effect, as in ʝatruðáci (‘unimportant doctor’) and popuðáci (‘small/cute baby bum’), respectively. As also stated by Fortin (2011: 17), ‘the [diminutive and] augmentative tend to conflate with the affectionate and pejorative forms, crosslinguistically, depending on context’. Thus, it is reasonable to expect that derogatory connotations are not exclusive to the derivations involving these suffixes. Finally, a third group consists of suffixes that denote a collection or group (mostly, of people). Most cases examined in this category seem to inherently and exclusively involve a pejorative meaning component and an explanation for this is offered in the analysis below. A second parameter along which the various suffixes are different is that some of them do not appear in standard dictionaries of SMG, whereas most others exhibit lexicalized products in dictionaries. For instance, the augmentative suffix -úmba (as in laðúmba ‘undesired/excessive large amount of oil’) can only be encountered in Google search results, which implies that: i. it generally appears in informal, oral discourse; ii. it mostly creates nonce forms or hapax legomena; and iii. it acts in the periphery of SMG lexicon (in, perhaps, slang). By contrast, a suffix like -íla with the meaning ‘unpleasant smell or taste of something’ has plenty of representatives in dictionaries. The latter suffix, in fact, seems to be more productive than -úmba, and due to its polysemy (with meanings extending to unpleasant situations in general) is far more central in SMG vocabulary (see Dangas & Manasis 2018 for a similar distinction between πρότυπα ‘standard’ and marginal suffixes). In sum, although all of the suffixes discussed are productive, they may differ in degrees of productivity and in register (neutral vs. colloquial or slang, standard vs. marginal).152 152

On the basis of the criterion of productivity, as well as due to space considerations, we will not deal with suffixes that have very few representatives, such as -éŋgo (as in psiléŋgo ‘tall-FEM [+PEJ]’, treléŋgo ‘crazy-FEM[+PEJ]’, murléŋgo ‘crazy-FEM[+PEJ]’) and -iakas (as in

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

165

It is also important to point out the multifunctionality of many of the suffixes examined. Some of them can be said to be ‘polyanaphoric’, that is, the same derivative may refer to different referential classes. For instance, as noted by Dangas and Manasis (2018: 328 following Anastasiadis-Symeonidis 1997), kartácas card-ácas ‘a person with cards [+PEJ]’ can refer to a football referee who notoriously gives cards or to a player who notoriously gets cards. Alternatively, the suffix itself can be said to be polysemous among several meanings. Without taking a position on the theoretical issue of polyanaphora and polysemy, we observe that most suffixes under discussion have a wide range of meanings or interpretations (see, e.g., -íla or -iáris in LKN) and in some cases (especially in non-lexicalized instances of the relevant derivations), the precise meaning of the derivative is indeterminate out of context. This characteristic has also been noted in the relevant literature (see, e.g., Grandi & Körtvélyessy 2015a). In what follows, a list-like description is offered along these lines: first, the input for the suffix is presented; then, in terms of the output, the syntactic category and the semantics of the derivative are discussed. Note that only nominal suffixes (forming nouns and, less centrally, adjectives) are dealt with, since the nominal category is prominent in pejorative derivation, both as input and output (see also Melissaropoulou 2015: 271).153 Last, we explore the connection of the function of the suffix to pejoration and of its pejorative function to other functions that the same suffix or other, similar, suffixes may have. Note that the examples, drawn from the LKN and Google, are not exhaustive. Finally, with respect to notation, we will be using the feature [+PEJ] for pejorative, without going into an in-depth analysis of its status as a semantic or pragmatic feature. As mentioned above, the pejorative function may be inherent or not inherent in the semantics of a suffix. Following Dressler and Merlini Barbaresi (1994), we also mention pragmatic features in brackets, such as [familiar], [ludic], or [informal]. A characteristic of affixes and words which are particularly prominent in SMG is the feature [+/–learned] (see, e.g., Anastasiadis-Symeonidis and Fliatouras 2004, 2018, 2019, Kamilaki 2009), on which a few preliminary notes are in order. The term learned, being born from the long history of Greek and its earlier issues with diglossia,154 refers to at least two characteristics:

153 154

xazúʎakas ‘stupid-MASC[+PEJ]’, stravúʎakas ‘squint eyes-MASC[+PEJ]’, gavúʎakas ‘blindMASC[+PEJ]’). The latter suffix also fails to meet the criterion of neutral bases, viz. all of its attested base-forms have negative meaning or belong to marginal vocabulary. SMG also exhibits verb-forming suffixes with a moderately productive pejorative function (e.g., -iázo, see Efthymiou 2013). See, e.g., Ferguson (1959) and Fishman (1967), and for diglossia in Greek, Mackridge (1990, 2004).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

166

Kallergi, Katsouda, Konstantinidou, Tsangalidis

i. An archaic identity or origin, that is, etymologically coming from earlier stages of Greek and, most prominently from Hellenistic Greek (see Manolessou et al. 2019: 208). In this sense, the feature [+learned] relates to diachrony and refers to formal, fossilized and now marginal characteristics of language (we are focusing on morpholexical characteristics here, but, according to various scholars, [+learned] characteristics include, further to lexical features, all levels of grammar, ranging from phonology to syntax (see, e.g., AnastasiadisSymeonidis & Fliatouras 2019: 18–20). ii. It refers to the use of an item at a high register or style, typically, but not necessarily, characterized by formality (see, e.g., Kamilaki et al. 2016). This second interpretation of the [learned] feature refers to the synchronic state and to the socio-pragmatic or discourse level of use, often contrasting [learned] to [colloquial/vernacular] or even [folk]. In sum, the term [learned] is Janus-faced, in including the archaic and the formal (i.e., formalistic, standard, or academic), so that Anastasiadis-Symeonidis and Fliatouras (2019: 12) name it ‘the archaic formal’ (‘το αρχαιοπρεπές επίσημο’). Not all Greek scholars agree upon a specific use of the SMG term λόγιο [lóʝio]: Anastasiadis-Symeonidis and Fliatouras (2004, 2019) contend that the internationally recognized feature [learned] represents both diachronic and synchronic features of style and that it can best be described as a continuum from [–learned] to [+learned], having the linguistic norm (the neutral terms) in between. By contrast, Manolessou et al. (2019: 207 ff ) argue that the synchronic and diachronic levels of analysis are far too complicated to be combined under the same term [learned], which they consider anachronistic (as opposed to the also anachronistic [folk]), and propose that [+learned] should refer only diachronically to [archaic], as opposed to [vernacular]. For the level-of-use/register/style dimension of the feature, Manolessou et al. (2019) suggest the term [high/superior level of use]. In this chapter, we support a clearer use of the term, reserving the terms [archaic] for inherited and indeed marginal types, and the term [high register] for the relevant socio-pragmatic/discourse uses of an item. However, it may be the case that items can be characterized as both, which is why the term [learned] is not abandoned altogether. Regarding pejoratives, the majority of which are generally placed in the [–learned] area in the sense of Anastasiadis-Symeonidis and Fliatouras (see, e.g., Anastasiadis-Symeonidis & Fliatouras 2019: 32–3 for the relation of [–learned] with neologism and expressivity), we will use the term [+colloquial] (instead of [–high register]).155 155

This means that [–colloquial] would refer to [+high register], and possibly [+archaic], or to neutral/default terms.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

6.4.1

167

Suffixes denoting a characteristic or property

In this group, we have included derivational suffixes which mostly form nouns and adjectives and which contribute various descriptive meanings (e.g., a recurrent one is that of a ‘profession or occupation’ and another is ‘a person who tends/likes/prefers to do X’). More specifically, we discuss -íla, -iá, -iáris/-iára/-iáriko, -dzís/-dzú, -ácas, -ístikos, -éos/-éa, -ás/-ú, and -ilíci. As is evident, in some of them, the inflectional suffix (coming at the periphery of the derivative) carries a fixed gender value (typically masculine or feminine), whereas others may receive gender markers of different values. Also, most of them attach to nouns, but a small number has bases other than nouns, that is, adjectives, and less frequently, verbs. According to theorists of expressive language (e.g., Scalise 1984 on evaluative morphology), such suffixes are located at the periphery of evaluative morphology for two reasons, one formal and one functional. First, some of them change the category of the base they attach to, whereas typical evaluative affixes do not (see, e.g., Scalise 1984, Steriopolo 2008, Grandi & Körtvélyessy 2015a; cf. Bauer 1997, Dammel & Quindt 2016). Second, the suffixes examined below combine more than one function, for example, in a single suffix, such as -ácas, three functions are conflated: indicating a human agent (changing the [–animate] value of the base into [+animate]), conveying a property (such as extreme liking of the referent or inclination towards something) and expressing the speaker’s negative attitude towards the agent or the particular property of the agent. Melissaropoulou (2015: 271) argues that such affixes are marginally evaluative or non-prototypical, as typical evaluative affixes should bear the function of evaluating only. The exact status of the suffixes discussed here will not be of great concern, since, taking a semanticfunctional viewpoint leads us to a vast number of items widely used to serve online pejoration. In other words, the conflation of expressive meanings with grammatical functions within a suffix will not be an impeding factor to our descriptive purposes. Pejorative meanings arise through the use of the suffixes explored here, and this is so, not only due to special contextual conditions, but also very often due to the semantic make-up of the suffixes themselves, as well as other motivating factors and parameters to be reviewed in the conclusion to §6.4. Besides, augmentative and diminutive suffixes (typically discussed under expressive morphology) also often involve their descriptive meaning (roughly BIG/SMALL respectively), a situation which is not very different from the case of this first group of suffixes.

6.4.1.1 The suffix -íla This suffix is inherently marked for feminine gender (involving the -a.FEM.SG inflectional suffix) and it combines with nominal bases (nouns

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

168

Kallergi, Katsouda, Konstantinidou, Tsangalidis

and adjectives) to create nouns. It carries the pragmatic feature [+colloquial] and, according to the LKN, it mainly contributes the meaning of ‘an unpleasant property’ (typically, smell, taste, or colour) of what is denoted by the base, as in the following examples: (126)

ðén andéxo tin kapn-íla not stand the.FEM smoke-íla.FEM ‘I can’t stand the smell of smoke[+PEJ]’

(127)

éxo mía sokolatíla sto stóma have.1SG a.FEM chocolate-íla.FEM in.the mouth ‘I’ve got a(n intense/unpleasant) taste of chocolate in the mouth’

Instances such as kapníla and citriníla (‘yellow stain’) are found in the LKN as lexicalized examples of the meaning of the suffix and the component UNPLEASANT seems to be a steadily present feature of the meaning of the suffix, that is, it is not the result of contextual or pragmatic factors. In fact, -íla is highly productive and according to the LKN, it may refer to unpleasant situations in general, as in sapíla ‘lit. rottenness, a situation where social decadence is evident’ (probably by extension of the more concrete meanings pertaining to physically unpleasant situations). Thus, it can also be encountered in combination with abstract entities in order to express the speaker’s disapproval or contempt to what the base describes; a brief Google research reveals examples such as aristeríla left-íla ‘left-ness[+PEJ]’, ðeksiíla right-íla ‘right-ness[+PEJ]’, and cinezíla Chinese-íla ‘Chinese-ness[+PEJ]’. Regarding the first two of these examples, we observe a tendency of -íla to attach to bases denoting political affiliations or general ideologies. The last example refers to products made in China or the property of something being made in China. However, it should be noted that, more often than not, the derivatives still retain an assumed relationship to the sense of smell or taste, which is the most basic meaning of the suffix, even in cases such as the following: (128)

bóxa, vróma, pasoc-íla156 stench, stink, PASOK-íla ‘Stench, stink, the bad smell of PASOK’157

(129)

vrom-ís-ame mitsotac-íla158 stink-PFV-3.PL Mitsotakis-íla ‘We got [=have become] stinking of Mitsotakis’

156 157

158

At www.stoxos.gr/2015/09/12_30.html (accessed 12 November 2021). ΠΑΣΟΚ (PASOK) is an acronym for Panhellenic Socialist Party and refers to a socialdemocratic political party, which, under that name, was one of the two major electoral forces in Greece from 1974 to 2014. At https://bit.ly/41tZRZx (accessed 12 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

169

However, the derivatives may also refer to concrete entities, as in citriníla ‘yellow stain’ and cinezíla above (‘a product made in China’, with the negative implications of low cost and inferior quality). In yet other cases, the meaning of the suffix is even more vague, referring to a general, unspecified, but definitely negative property (an ‘essence’) of the base, as in (130): (130)

aftó ci an íne mitsotacíla ci arniticí enérʝia159 this CONJ if is Mitsotakis-íla and negative energy ‘this is a typical case of mitsotacila (a negative essence of Mitsotakis) and negative energy’

6.4.1.2 The suffix -iá The suffix -iá,160 a highly productive and polysemous suffix, is studied in detail by Efthymiou (1999, 2013), who mainly focuses on its semantics and expressive dimension. It forms only feminine nouns with nominal, adjectival, and verbal bases and, according to Efhtymiou (2013), has a wide range of meanings, including ‘a single instantiation of an action’ (e.g., pineʎá painting.brush-iá ‘an individual act of brushing’), a ‘negative quality’, similarly to -íla above (laðʝá oiliá ‘an oil stain’), ‘quantity contained in what the base denotes’ (e.g., tiγaɲá frying.pan-iá ‘food at the amount of a frying pan’), ‘blow with the instrument denoted by the base’ (sfirʝá hammer-iá ‘blow with a hammer’), ‘characteristic property or behaviour’ (e.g., ponirʝá cunning-iá ‘cunningness’), and ‘collectivity’ (e.g., turcá Turk-iá ‘the Turkish nation [+PEJ] /a group of Turks [+PEJ]’). All of the above are lexicalized instances of word-formation with -iá. Regarding expressivity, Efthymiou (1999) extensively discusses the positively expressive function of -iá as an index of familiarity and perhaps ludic or ironic disposition (of the type kafeðʝá coffeeFAM161 ‘a nice coffee’, according to Efthymiou’s notation). With respect to negative expressivity or pejoration, we will isolate two uses of the suffix (in fact, the latter two mentioned above) in the body of the non-lexicalized instances we encountered, as more interesting. As mentioned above, the suffix expresses (human) characteristics and properties. More often than not, these are presented as individual acts, as in lexicalized γaiðurʝá donkey-iá ‘lit. donkey-ness’,162 but typically ‘an 159 160

161

162

At https://temaxidon.blogspot.com/2011/06/blog-post.html (accessed 12 November 2021). The suffix has a range of different phonetic realizations, according to the consonant that precedes it. This is why the orthographic -i- of -iá appears with different glosses (mostly the palatal counterparts of the preceding consonant). According to Efthymiou (2013: 155), FAM seems to refer to the familiarity between speaker and addressee, but according to her analysis in Efthymiou (1999: 148–9), this intimate relationship between interlocutors is the framework for an expressive/emotional use of the word in question on the part of the speaker. Unfortunately, the noun γaiðúri ‘donkey’ has a range of metaphorical meanings, all of which are more or less negative. Thus, when used to refer to a person, it may become a swear-word,

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

170

Kallergi, Katsouda, Konstantinidou, Tsangalidis

unthoughtful/insensitive/selfish act’. As noted by Efthymiou (2013: 155), deadjectival nouns with -iá may ‘denote negative or undesirable human properties. Interestingly, these properties are usually conceived actualized through concrete events’. Thus, γaiðurʝá most typically refers to something that someone did. As Efthymiou puts it, these nouns ‘do not exclusively express properties, but can also refer to occurrences of events. In other words, deadjectival nouns in -iá give rise to eventive readings (see Efthymiou 1999)’ (Efthymiou 2013: 155, citation original). By extension of the transfer from ‘general property’ to ‘individual event’, and in line with a general tendency of connotative affixes163 to require a [+bounded] interpretation of entities, the non-lexicalized derivatives of the suffix may refer to an actual object (131) or a particular institution or habit – in fact, any kind of social attitude or manifestation thereof (132–137) – and even individual persons (138). (131)

cinezʝá Chinese-iá ‘Chinese product[+PEJ]’, as in ti se káni na apokalís éna smartphone cinezʝá ce éna álo aplá ‘smartphone’164 ‘What makes you call a smartphone a Chinese-product [+PEJ] and another (smartphone) just ‘smartphone’?’

(132)

amerikaɲá ‘any product, institution or habit of U.S. origin [+PEJ]’165

(133)

amerikaɲá ‘an act/behaviour with American characteristics’ as in i amerikaɲá tu proθipurɣú ‘the Americanness[+PEJ] of the prime minister’166

163

164

165

166

connoting that the person is insensitive, coarse, or of bad manners. See also ILNE, entries γάιδαρος, γαϊδούρι, γαϊδουριά. According to Fortin (2011), connotative affixes (CAs) are expressive affixes (EAs) that have both descriptive and expressive meaning (e.g., a diminutive which means small and cute [+affection] at the same time). Fortin’s claim is that ‘the descriptive meaning of a CA is only available for a unitized substance or aggregate’ (2011: 128). Otherwise, the meaning of an EA is only expressive (cf. SMG ner-áci water-DIM ‘small [bottle/individual container of] water’, but if water is viewed as mass, ner-áci can only be interpreted as ‘water [+affection]’). At www.digitallife.gr/kinezika-oxi-kinezika-kai-kinezies-71514 (accessed 15 November 2021). Note that an internet search may easily reveal a range of meanings for cinezʝá, e.g ‘(metaphorically) a smell of cheap, low-quality product’ (typically, of technological equipment), or the product itself, which may refer to a cheap mobile phone or an expensive superproduction coming from the Chinese film industry, imitating Hollywood. A possible generalization among all these meanings (excluding that of smell) is that cinezʝá refers to any non-original, replica products made in China. This could be a film of a particular type (either too fancy or epic or melodramatic), a USAbased company in Greece, such as Starbucks, an institution such as Black Friday and a custom like Halloween or Sweet Sixteen celebrations (sources: https://bit.ly/3xWgI9Y and www.slang .gr/lemma/2171-amerikania (accessed 15 November 2021)). The act refers to the policy of economic motivation given to young people in order to get vaccinated against CoVid-19, which according to the speaker was an act imitating a strategy initiated by a USA governor. At https://bit.ly/3IBQWN2 (accessed 15 Juli 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

171

(134)

kulturʝá culture-iá ‘a pretentiously sophisticated/intellectual attitude, philosophy, social situation [+PEJ]’167

(135)

turist-çá tourist-iá ‘typically touristic [+PEJ]’168

(136)

xaritomeɲés cute-iá.PL ‘cute actions or words, spoken with affectation/frill’ (ironic)

(137)

lafazaɲés Lafazanis-iá.PL ‘actions or words typical of Lafazanis [+PEJ]’169

(138)

kulturʝá ‘a pretentiously sophisticated/intellectual person’170 as in ton θeorúsa pço kulturʝá ap’ tin ipólipi paréa ‘I considered him more intellectual[+PEJ] than the rest of the company’

The second pejorative use of -iá is its function of denoting ethnic groups, as in turcá (Turk-iá ‘the Turks [+PEJ]’), ʝiftçá (Gypsy-iá ‘the Gypsies [+PEJ]’), and vlaçá (Aromanian-iá ‘the Aromanians/Vlachs [+PEJ]’). According to Nick Nicholas,171 in the early nineteenth century, the suffix used to be the vernacular (neutral) term for a nation or group of nations (whether official or unofficial). Thus, for example, venetçá ‘Venice’, fraŋɟá ‘Western countries (collectively)’, and vlaçá (‘Wallachia’) were used to refer to areas. With the introduction of Puristic Greek, -iá for toponyms was abandoned in favour of the more archaic form -ía, which exhibits hiatus (cf. venetía ‘Venice’, ʝermanía ‘Germany’).172 Nicholas argues that -iá with the function of referring to actions (such as amerikaɲá) was homophonous to the old one denoting nations, and it may have gradually taken over the latter. In line with this analysis, we may conclude that -iá, when applied to bases denoting nations, is ambiguous between the action meaning (more specially, a negative behaviour, typical of a nation) (139) and a (derogatory) meaning of ‘a collection of

167

168

169

170

171 172

For example, in https://bit.ly/3Y4Cz9G (accessed 16 November 2021), kulturʝá is the fashionable ‘free relationship’ attitude of youngsters, a ‘we are not dating just hanging out’ philosophy in relationships. The word seems to refer to a restaurant, since it was attested as a title for a restaurant review, discussing the mediocre or low quality of the food. At https://bit.ly/3EHxeOK (accessed 14 November 2021). The name refers to a contemporary Greek politician, in an apparent pun with the allegedly original meaning of the name, that of ‘liar’. At https://bit.ly/41ooQO9 (accessed 4 March 2022). At https://bit.ly/3SvITWG (accessed 16 November 2021). Reference to persons seems to be a rare function among lexicalized instances. However, there are a few relatively opaque cases, such as afendʝá boss-iá ‘oneself, one’s majesty (ironic)’ and levendʝá brave.man-iá in the more or less fixed phrase íse levendʝá ‘you are very brave/you are a good fellow!’ At https://bit.ly/3F2KDRZ (accessed 20 November 2021). For a discussion on the connection of hiatus with high register, on the one hand, and glide formation/palatalization with low register, informality and the [–learned] feature, see Efthymiou (2013) and the rich references therein. Efthymiou (2013) argues that -iá as well as the suffix -iáris discussed immediately below, and the verbalizing suffix -iázo, exhibit phonetic iconicity, viz. a correlation between the phonetic features of the suffixes and their role in negative evaluation.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

172

Kallergi, Katsouda, Konstantinidou, Tsangalidis

people with the same national identity’ (140), with the context resolving the issue. (139)

aftó pu ékane ítan ʝiftçá this that did was Gypsy-iá ‘What s/he did was Gypsy-like [+PEJ], indecent’

(140)

mazéftice óli i ʝiftçá gather.3SG.PST all the Gypsy-iá ‘All the Gypsies [+PEJ] gathered’

6.4.1.3 The suffix -iáris/-iára/-iáriko Ample information on this suffix is given in Anastasiadis-Symeonidis (2000) and Efthymiou (2013). According to LKN, -iáris (marked for masculine, and -iára, marked for feminine) creates adjectives having (mainly) nouns as bases. The suffix may also derive adjectives from other adjectives. Note that the neuter forms in -iáriko arise only in combination with the adjectival suffix -iko (i.e., *ftiɲár-o cheapy-NEUT but ftiɲ-iáriko cheapy-NEUT). According to Anastasiadis-Symeonidis (2000: 65), -iáris attaches to bases that denote external or internal (personality) properties, which are (conceived as) negative (viz. illnesses or imperfections of the body or character). The semantic contribution of the suffix is the establishment of a permanent relationship between the property denoted by the base and the referent, in a way perceptible with the senses (sight, smell, etc.) and divergent from the norm (Anastasiadis-Symeonidis 2000: 67, 69, 70). Also, Efthymiou (2013: 154) suggests that derivatives with -iaris have aspectual meanings, such as iterativity. For example, a griɲáris (145) is a person who tends to whine all the time and this occurs more than what is usually expected or desired, inviting a derogatory attitude on the part of the speaker. (141)

spirʝáris pimple-iáris ‘pimply[+PEJ]’

(142)

vromɲáris dirt-iáris ‘dirty[+PEJ]’

(143)

kseðοndʝáris kse-tooth-iáris ‘toothless[+PEJ]’

(144)

xticáris tuberculosis.PEJ-iáris ‘a sickly person [+PEJ]’

(145)

griɲáris whining-iáris ‘whiner’

(146)

martirʝáris testimony.PEJ-iáris ‘informer[+PEJ], a fink’

Note that, unlike cases from (141) to (146), the bases of -iáris do not necessarily bear a negative meaning. Derivatives from (147) to (151) are examples of the fact that it may attach to neutral bases as well.173 These bases can be adjectival or nominal: 173

In fact, there are few instances where the bases of -iáris are positive, and, for this reason, the overall meaning seems to be positive (Anastasiadis-Symeonidis 2000: 71): xaðʝáris caressiáris ‘cuddly’, pexniðʝáris playing-iáris ‘playful’.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek (147)

citriɲáris (< cítrinos ‘yellow’) ‘yellow[+PEJ]’174

(148)

aspruʎáris (< asprulós ‘whitish’) ‘pale white [+PEJ]’

(149)

ftiɲáris (< ftinós ‘cheap’) ‘a cheapy person [+PEJ]’

(150)

skupiðʝáris (< skupíði ‘garbage’) ‘garbage man [+PEJ]’

(151)

traγuðʝára (< traγúði ‘song’) ‘a (mostly folk) singer-FEM [+PEJ]’

173

In the cases of the denominal -iáris derivatives that refer to a job or occupation (150, 151),175 the pejorative function consists in the implication that these occupations have a low socio-economic status (AnastasiadisSymeonidis 2000: 69–70). However, at least traγuðʝára has a lexical equivalent with the same base, but without pejorative implications (traγuð-ístria singer-FEM ‘woman singer’). Thus, pejoration (the implication of low status, plus a subtle expression of contempt) seems to come from the suffix. In fact, Anastasiadis-Symeonidis argues that -iáris is inherently pejorative, in the sense of having a standard role across various types of word-formation and in carrying, in her notation, the semantic features [+subjective] and [+negatively evaluative] (Anastasiadis-Symeonidis 2000: 71, following KerbratOrecchioni 1980). On the other hand, -iáris bears two characteristics that, according to the relevant theory (e.g., Scalise 1984, Grandi 2005), seems to draw it far from the typical case of expressive morphology (evaluative in Scalise’s and Grandi’s terminology). One is the fact that the relation between the base-form and the derivative with -iáris may not reflect a relation between a ‘standard meaning’ and a ‘standard meaning plus an evaluative value’ (see, Grandi 2005). As observed by Efthymiou (2013: 158), for instance, the base-form spirí ‘pimple’ does not have the standard meaning involved in the derivative spirʝáris ‘pimply[+PEJ]’. This has to do with the fact that ‘pimple’ refers to an object, whereas ‘pimply’ refers to a person. This is the second point in which -iáris is divergent from the norm of expressive affixes. As mentioned in the introduction to this section, it is untypical of expressive suffixes to change the syntactic category of the base they attach to. However, as Bauer (1997 quoted in Dammel & Quindt 2016: 43) observes, there seem to be ‘no hard and fast formal criteria for evaluative affixes’ across languages. Thus, instead of 174 175

The word citriɲáris, although based on an adjective, is mostly used as a noun, pejoratively referring to an East Asian person. Anastasiadis-Symeonidis argues that such derivatives are nouns only at a surface level, but they are in fact adjectives, similarly to all the derivatives with -iáris (2000: 67). Note, also, that she makes reference to jobs that are now obsolete (e.g., arkuðʝáris bear-iáris ‘kind of street entertainer, who carried a bear, which seemingly danced’, maimuðʝáris monkey-iáris ‘kind of street entertainer, who used monkeys’). Since many of these jobs were based on deception or exercised by minorities (e.g., Gypsies), they have been marked negatively and are nowadays used mostly as slurs/insults.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

174

Kallergi, Katsouda, Konstantinidou, Tsangalidis

placing -iáris in the periphery of expressive suffixes, we may consider it as a morphological device with a mixed semantics: a descriptive meaning (very roughly, ‘someone who has been/is inclined towards something’) and an expressive meaning or function, viz. pejoration (see mixed use-conditional items, Gutzmann 2011). With respect to productivity, derivatives with -iáris are typically not occasional, novel formations, but lexicalized ones. More specifically, it is difficult to attest derivatives in -iáris that have not been heard before in colloquial SMG. Still, it seems to be particularly productive in slang, taboo, or obscene vocabulary, with nominal and even verbal bases: burðeʎáris brothel-iáris ‘man who likes going to brothels’, partuzʝáris orgy-iáris ‘man who likes orgies’, psoʎára dick-iára ‘promiscuous, obscene woman’, tsiliburðʝáris ‘fuck. around-iáris’ ‘womanizer’, γamɲáris ‘fuck-iaris’ ‘womanizer’. These bases do not necessarily have negative semantics, but are definitely marked for register. We also observe that with bases that are already negative, the expressive role of the neuter type -iáriko may be attenuating or pragmatically positive, especially if the neuter derivative is used for non-neuter human referents. For instance: crazy.iár.NEUT ‘crazy in a fun way, fun.ADJ[+ludic]’

(152)

treʎáriko

(153)

palavʝáriko crazy.iár.NEUT ‘crazy[+ludic]’176

Here, -iáriko seems to add the pragmatic feature [+ludic] (Dressler & Merlini Barbaresi 1994) which might be responsible for the resulting decrease in the pragmatic force of the utterance. The neuter gender might also play a role in the relatively low offensiveness (or absence thereof ) of these derivatives (see Christopoulou et al. 2017). However, the issue cannot be further investigated here. In both this and other cases, where -iáris creates deadjectival adjectives (152–153), its function is only to add some expressive load (positive or negative). In other words, -iáris may in certain cases behave as a typical or pure expressive affix, in that, without inducing category change, it is used as a tool for the manipulation of meaning and the pragmatic or discursive effects of utterances.

176

A nice example of this use is the following utterance, attested on an instagram profile page: The word palavʝáriko is immediately preceded by the word trel-útsik-o crazy-DIM-NEUT ‘crazy[+affection]’, with both adjectives modifying the noun ‘creature’, which refers to a girl, implicating sympathy (for the role of the attenuating suffix -útsiko, see Efthymiou 2019): Λοιπόν πριν από 19 χρόνια γεννήθηκε ένα υπέροχο πλασματάκι λίγο τρελούτσικο παλαβιάρικο αλλά με υπέροχη ψυχή. ‘Well, 19 years ago there was born a wonderful little creature, somewhat crazy, but with a wonderful soul’.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

175

6.4.1.4 The suffix -(a/i)dzís/-(a/i)dzú The two forms presented in the title correspond to the masculine and feminine form of a nominalizing suffix, which, according to the LKN, originates in the Turkish suffix -ci, -çi177 that primarily denotes profession (e.g., Turk. taksi-ci ‘taxi-driver’) (but also see Kiranoudis 2001 for a multi-level analysis).178 Therefore, many of its representatives in the lexicon: i. involve bases of Turkish origin; ii. concern a professional occupation. However, the SMG suffix -dzís, attaching to Greek bases as well, is [–learned], [–formal] and [+colloquial], hence the occupation typically relates to low economic classes and is usually perceived as of low status in society, as with -iáris: fortiɣadzís ‘lorry driver’, laikadzís ‘open-market seller’, suvladzís ‘suvlaki-maker’, kulurdzís ‘kuluri vendor’. The professions mentioned often have a periphrastic equivalent of neutral status (e.g., politís laikón aγorón seller folk. GEN.PL market.GEN.PL ‘open-market seller’), but in pairs with lexical equivalents (e.g., violistís ‘violinist’ – violidzís ‘violin player’), it becomes evident that -dzís profiles the job as one that is done only for a living. It has been noted in the SMG literature (as early as Kazazis 1972 and Tzitzilis 1997) that morphological elements borrowed from Turkish are stylistically marked and this is why there are lexical pairs in which the SMG part is neutral, whereas the equivalent part based on Turkish loans is stylistically marked and expressively richer, if not semantically negative: for example, ðiasceðázo ‘to be entertained’ – ɣlendó ‘to party [+colloquial]’, orofí ‘ceiling’ – taváni ‘ceiling [+colloquial]’ (Tzitzilis 1997: 105), but also ðeksamení ‘tank’ – xavúza ‘tank [+colloquial] (usually for dirty water)’, eristikós ‘aggressive’ – kavγadzís ‘brawler’, anóitos ‘stupid’ – budalás ‘stupid [+colloquial], [+ludic]’. As Kazazis (1972: 113) puts it, ‘the stylistically marked Turkisms have enriched the expressive and stylistic potential of every Balkan language’. Also, Anastasiadis-Symeonidis (2010) and Anastasiadis-Symeonidis and Fliatouras (2019: 38) observe that, in general, SMG affixes borrowed from foreign languages are used with colloquial bases and turn out to be vastly [+colloquial], even if they were stylistically unmarked in the language of origin. These factors justify to some degree the development of pejorative meanings in suffixes such as -dzís and other Turkish loans mentioned in this section.

177 178

This is a very tentative phonetic transcription for the Turkish suffix, which has a number of allomorphs (-ci, -cι, -cü, -çi, -çι, -çü) and allophones. For the problem of the exact form of the suffix and its allomorphs in Greek (including dialects), the reader is referred to Kiranoudis (2001).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

176

Kallergi, Katsouda, Konstantinidou, Tsangalidis

Apart from professional nouns, the suffix -dzís, following its Turkish predecessor -ci, -çi (see Kiranoudis 2001: 368), is used to form denominal nouns that denote a stable but dynamic human property: it expresses the agent or actor in a relational role with the base, which can be paralleled to French -eur or English -er (Kiranoudis 2001: 443, 449, where he suggests that -dzís denotes a ‘typical activity’ ‘χαρακτηριστική δραστηριότητα’). More specifically, the property or activity expressed is often an inclination towards or a liking or preference of a person for what the base denotes, as in the following examples: (154)

ɣlikadzís ‘a person with a sweet tooth’

(155)

kavɣatzís ‘a person who likes fights and quarrelling’

(156)

fasardzís ‘a person who tends to make a fuss or razzmatazz’

(157)

fiɣuradzís ‘a person who likes making an impression’

(158)

paokdzís/paoktsís ‘a fan of PAOK’179

(159)

pasokadzís/pasoktsís ‘a PASOK voter’

As Examples (155) and (157) imply, the suffix is quite productive with [+colloquial] bases already denoting a negative quality or coming from marginal/slang vocabulary, as in (160)–(164), and, in these cases, it is [+pejorative] and, perhaps, [+ludic]: (160)

teknadzú (< teknó ‘very young man’) ‘woman who prefers too young men [+PEJ]’

(161)

burðeladzís (< burðélo ‘brothel’) ‘man who likes going to brothels [+PEJ]’

(162)

banistirdzís (< banistíri ‘voyeurism’) ‘a voyeur [+PEJ]’

(163)

dzabadzís (< dzába ‘for free’) ‘person who avoids paying; cheapskate [+PEJ]’

(164)

trakadzís (< tráka ‘using or borrowing ‘a person who tends to use or borrow without paying or giving back’) things without paying/giving back [+PEJ]’

However, the suffix is attested in pejorative derivations with relatively neutral bases, and the same is observed in nonce formation (although nonce forms with -dzís may not be very frequent), which involves an abstract noun as a base (169): (165)

paramiθadzís (< paramíθi ‘fairy tale/lie’) ‘a liar [+PEJ]’180

(166)

trakadzís (< tráka ‘car crash’181) ‘a driver who tends to crash [+PEJ]’182

179 180 181 182

PAOK is an acronym name for one of the most famous football teams of Thessaloniki, Greece. Of course, in connection with the original meaning of the suffix (to denote professions), paramiθadzís may refer to the job of telling fairy tales (the now obsolete storyteller). This meaning is presented in the LKN as the primary meaning that gave rise to that of Example (164). Referring to an F1 driver. At https://bit.ly/3Z5SuWA (accessed 20 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

177

(167)

etimadzís (< étimo ‘ready, instant’) ‘a person who likes easy, ready-made, instant things [+PEJ]’183

(168)

paniʝirdzís (< paniʝíri ‘open fair’) ‘ridiculous person [+PEJ]’

(169)

ipervoladzú (< ipervolí ‘exaggeration’) ‘a girl who is inclined to exaggerate [+PEJ]’184

However, positive meanings cannot be excluded, especially if the base has a positive meaning itself, as in: (170)

plakadzís (< pláka ‘fun’) ‘a funny person/one who makes jokes [–PEJ]’

(171)

kalaburdzís (< kalabúri ‘joke’) ‘a person who likes/tells jokes [–PEJ]’

(172)

kataferdzís (< kataférno ‘succeed’) ‘a person who tends to get his way [–PEJ]’

The highly colloquial style of the suffix -dzís/-dzú seems to involve the [+ludic] feature here, and this seems to play a role in the rather meliorative function of these cases. Moreover, the semantics of the base play a crucial role in the meaning of a derivative, as is possible with other cases of expressive affixes.

6.4.1.5 The suffix -ácas The suffix -ácas has been studied in detail by Dangas and Manasis (2018). According to their analysis, it is marked for masculine for both men and women and attaches to semantically positive, negative or neutral bases, which are either: nouns (as in Example 173, mamá ‘mother’), nominalized adjectives (as in 175, éksipnos ‘clever’), verbs (as in 178, periméno ‘wait’), adverbs (as in 179, puθená ‘nowhere’), lexical phrases or even elliptical sentences (as in 180, s’tá’leɣa ‘I told you’), acronyms (PAOK, Example 176), and loan words (gádzet ‘gadget’, Example 177) and produces the respective derivatives (examples mostly taken from Dangas and Manasis 2018): (173)

mamácas ‘a person attached to his mother [+PEJ]’

(174)

ʝalácas ‘a person who wears glasses [+PEJ]’

(175)

eksipnácas ‘wiseguy [+PEJ]’

(176)

paokácas ‘a PAOK maniac [+PEJ]’

(177)

gadzetácas ‘a person who likes gadgets too much [+PEJ]’

(178)

perimenácas ‘someone who waits instead of another person in queues for a small amount of money [+PEJ]’185

183 184 185

The more frequent derivative adjective etimadzíðiko is a pejorative term for anything ready made, looking down or mass production. Personally attested by H. Kallergi. At www.slang.gr/lemma/22355-perimenakias (accessed 5 December 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

178

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(179)

puθenácas ‘a person who is not there, when s/he should [+PEJ]’

(180)

staleɣácas ‘a person who always says “I told you so” [+PEJ]’186

(181)

kaloperasácas ‘a person who always wants to have fun [+PEJ]’

It is therefore a highly productive suffix, with a descriptive meaning. Its semantic core seems to be ‘a person who is prone to/inclined to/tends to do something or has a very strong inclination/liking for something, which exceeds the norm’.187 In this sense, it is similar to -(a)dzís/-(a)dzú, as well as -iáris, discussed above. However, as Dangas and Manasis (2018) argue, -ácas is special in that: i. although it is generally less restricted in terms of selection of syntactic categories, it forms lexemes that are only [+human]; ii. its derivatives always have a pejorative meaning, unlike those of -adzís/-adzú and -iáris (see Examples 170–172 and footnote 173); iii. it is even more informal and colloquial than the others, which makes it particularly preferable in SMG argot or slang, and iv. it is newer than the other suffixes, which, according to Dangas and Manasis (2018), means that while older suffixes, such as -iáris, lose their ‘expressive power’, -ácas satisfies the need of speakers for expression and nonce formation. This is why -ácas is widely attested in relatively restricted registers and in certain contexts, for example, familiar conversations between young people on sports, shopping (esp. of electronic equipment) or criticisms of human behaviour. Its strong participation in nonce formation also seems to play a role in another special feature of -ácas; in comparison to suffixes like -adzís/-adzú and -iáris, it creates lexemes with a degree of semantic opacity. Attaching to bases which are themselves interpreted according to context, -ácas forms lexemes whose meanings are not fully predictable, or are ambiguous between different referents or interpretations. For instance, a brief internet search on kutácas, deriving from kutí (‘box’), reveals that it may mean either a person preferring cars (‘box-shaped’ vehicles) as a means of transport, or a journalist that often appears on television (often metonymically referred to as ‘the box’) or one that presents people on panels. In fact, such a derivative could refer to anything, according to the range of possible interpretations of its base, and the different interpretations typically have a semantic element in common (e.g., rectangular shape) (see also Stavrianaki 2009 for multi-reference in SMG diminutives).

186 187

According to www.slang.gr/lemma/1430-stalegakias (accessed 5 December 2021). Interestingly, Daltas (1985) mentions -ácas among diminutive suffixes of SMG.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

179

6.4.1.6 The suffix -ás/-ú Another suffix that denotes property, as well as extreme liking or inclination is –ás (marked for masculine) and its equivalent feminine -ú, which create denominal nouns.188 According to Hatzidakis (1905: 420–1), -ás developed from the suffix -éas, which formed common nouns but already from the Hellenistic period it started attaching to various nominal bases with a variety of meanings.189 These suffixes are encountered with the relatively neutral function of creating words for jobs, such as psarás/psarú (‘fisherman or man who sells fish/ woman who sells fish’), peripterás/peripterú ‘man/woman behind the counter of a kiosk’, Pacetás or deliverás (‘delivery-boy’) and nixú (‘manicurist, nail technician’). According to the LKN, its primary function is to describe a person with a property in a high degree, as in çilás lip-ás ‘man with big lips’ and ðondás teeth-ás ‘man with big teeth’. In this sense, it is very close to augmentatives (see §6.4.2.1). However, its primary function is not to express augmentation; instead, the meaning elements BIG or IN A HIGH DEGREE are combined with other functions, such as that of the agent (Melissaropoulou 2015: 271). Based on this fact, Melissaropoulou considers -ás a marginally expressive suffix (evaluative in her terminology). Also, Anastasiadis-Symeonidis (1996) terms the suffix ‘a relational suffix’ (‘συσχετιστικό επίθημα’), that is, one that conveys a relation of the agent with what the base denotes and in AnastasiadisSymeonidis (1997) (referred to in Dangas & Manasis 2018: 329) she suggests that -ás is only potentially pejorative. However, even in the apparently neutral function of forming professional nouns, the bases indicate jobs associated with low socioeconomic status, with the feminine formations sounding particularly derogatory compared to other alternatives that refer to the same job.190 Thus, -ás and -ú may be considered as mildly pejorative. As already mentioned, -ás may also describe extreme liking or inclination as a property characteristic of a person, and, with this function, both -ás and its feminine equivalent -ú are involved in a number of pejorative lexicalizations: (182)

leftás money-ás ‘a loaded guy’

(183)

makaronás pasta-ás’ ‘a person who likes pasta very much’/ ‘an Italian [+PEJ]’

188 189

190

The suffix can be also marked for neuter, having the form -úðiko/-áðiko (see the LKN). However, the neuter type does not seem to participate in pejoration in any significant way. It is perhaps worth noting that in the Modern Greek dialect of Kythera, -éas participates in the formation of pejorative nouns, e.g., tifléas ‘blind[+PEJ]’, zavéas ‘stupid’, çezéas ‘coward’ etc. (Katsouda 2020: 179–80). See, e.g., a very elaborate account of how the term nixú may be insulting at https://bit.ly/ 3mbbXqB (accessed 28 January 2022).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

180

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(184)

poliloɣás/poliloɣú lots.of.words-ás/-ú ‘a garrulous man/woman’

(185)

ʝinekás woman-ás ‘womanizer’

(186)

ɣlosú tongue-ú ‘a girl that speaks too much or brazenly’

(187)

sokakú street[+colloquial]-ú ‘a streetwalker’

Regarding nonce formation, it is equally pejorative and in fact more pejorative than neutral (if at all), as the following attested instances indicate: (188)

sirizás SYRIZA-ás ‘a SYRIZA191 voter/fan [+PEJ]’192

(189)

pasokás PASOK-ás ‘a PASOK voter/fan [+PEJ]’193

(190)

emvoli-á-ðes vaccine-á-PL ‘people who vaccinate or get vaccinated [+PEJ]’194

In productive word-formation the suffix also attaches to phrases or elliptical sentences: (191)

aspromaʎás (< áspra maʎá ‘white hair’) ‘a man with white hair [+PEJ]’

(192)

sapçociʎás (< sápça ciʎá ‘rotten belly’) ‘a man with a fat belly [+PEJ]’

(193)

vapsomaʎás (< vápso maʎá ‘dye hair’) ‘a man who dyes his hair [+PEJ]’

(194)

nemenalás (< ne men alá ‘yes, but’) ‘a man who says ‘yes, but’ [+PEJ]’.

6.4.1.7 The suffix -éos/-éa The suffix -éos/-éa (masculine and feminine respectively, Greek -αίος/-αία) produces nouns and adjectives from nominal bases and, according to the LKN, it retains the Ancient Greek form -αῖος[aîos]. It bears a strong connection to the concept of place, since one of its major functions is to create nouns that denote a person’s descendance (the place where he/she comes from): (195)

cercir-éos Corfu-éos ‘someone from Corfu’

(196)

evrop-éos Europe-éos ‘a European’

(197)

aθin-éos Athens-éos ‘an Athenian’

The LKN also notes that -éos forms denominal adjectives that also denote the place where something is located or comes from (e.g., akréos edge-éos ‘extreme.ADJ’, ɣoniéos corner-éos ‘one at the corner.ADJ’), but manner as well (e.g., piʝéos spring-éos ‘spontaneous’, astrapiéos lightning-éos ‘fast as lightning’).

191 192 193 194

This is an acronym for Συνασπισμός (sinaspizmós) Ριζοσπαστικής (rizospasticís) Αριστεράς (aristerás), meaning ‘Coalition of the Radical Left’. At https://gianniotis.blogspot.com/2014/04/blog-post_28.html (accessed 27 January 2022). At https://petefris.blogspot.com/2006/10/blog-post_11.html (accessed 28 January 2022). With the meaning ‘vaccinators/vaccine companies/pro-vaccine officials’.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

181

In addition, the plural masculine form of the suffix (-αίοι -éi) appears as a separate lemma with special functions. It is essentially used as a secondary, more expressive, type of plural in order to denote (extended) families, or a collective of relatives (similarly to the plural of family names in English, such as, for example, the Johnsons): (198)

i Katsimiç-éi the.PL Katsimixas-éos.MASC.PL ‘the Katsimixas brothers’195

(199)

i Papaðopul-éi the.PL Papadopoulos-éos. MASC.PL

‘the (whole or part of ) Papadopoulos family’

In this latter sense, the suffix expresses identity (as the property of belonging to a particular community or family) with a slightly coloured tone ([+familiar], and, perhaps, [+pejorative] with non-celebrity names, at least in comparison to the more neutral inflectional plural forms, such as i Papaðópuli the.PL Papadopoulos.PL ‘the Papadopoulos family’). Its collective meaning here may be responsible for its pejorative connotations (as suggested in §6.4.3 and in Mutz 2015, who, however, also posits a diachronic connection between pejoratives and locatives, viz. expressions that denote place). The LKN also observes that the plural -éi attaches to nouns denoting professions (as in kapetanéi ‘captains’), which, as the dictionary suggests, may sometimes be derogatory (skupiðʝaréi ‘garbage men [+PEJ]’). Indeed, pluralization with -éi is frequently encountered in Greek dialects, with common nouns and without being stylistically or expressively marked (e.g., East Thrace: arɣatéi ‘workers’; Macedonia: ðaskaléi ‘teachers’; Peloponnese: ʝiftéi ‘Gypsies’).196 In SMG, however, -éi and -éos in productive word-formation seems to be mainly pejorative. By extension of the notions of a family, community, group located somewhere or a fára ‘social class [+PEJ]’, the suffix can be productively used to show the identity of a person designated by their preference to what the base denotes, plus a negative attitude on the part of the speaker towards this preference. To put it differently, whatever identity is expressed by the derivative, it is definitely not shared by the speaker. Consider the examples: (200)

sirizéos SYRIZA-éos ‘a SYRIZA fan/voter [+PEJ]’

(201)

mitsotacéos Mitsotakis-éos ‘a supporter of Mitsotakis [+PEJ]’197

(202)

aristeréi leftwing-éos.MASC.PL ‘leftist-PL [+PEJ]’198

195 196 197 198

A famous Greek musical duo, consisting of two brothers. See ILNE, entries ἀργάτης, δάσκαλος, γύφτος, respectively. Ο Κουλικος μητσοτακαιος Τσιτσιπάς παίζει; At https://twitter.com/kokokokokokokob (accessed 22 November 2021). Η διαρκής αντιδραστικότητα των αριστεραίων. At www.phorum.com.gr/viewtopic.php?t= 28254&p=1729200 (accessed 22 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

182

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(203)

papaðaréi priest-éos.MASC.PL ‘priests [+PEJ]’199

(204)

ðimarçéi mayor-éos.MASC.PL ‘mayors [+PEJ]’200

(205)

aftí these

(i emvoliazméni enoó ipurʝéi proθipurʝéi proeðréi vuleftéi (the vaccinated mean.1SG minister-éi prime.minister-éi president-éi MP-éi pos ce bénun se aeroplána201 now and enter-3PL in aeroplanes

kulupú) etcetera [+colloquial] ‘These (I mean, this vaccinated bunch of ministers, prime ministers, presidents, MPs, etc.) how come they get on planes?’

In most of the examples above, the suffix does not provide the literal meaning (that of ‘of X origin’ or ‘belonging to family X’). Instead, the suffixed nouns carrying -éi simply substitute the base of the derivation, in order to express pejoration. For instance, aristeréi (202) would be mutually interchangeable with the neutral form aristerí as nominalized adjectives, ðimarçéi (204) would be interchangeable with ðímarçi as nouns, and so on. Thus, in its pejorative use, -éos does not change the category of the base (as is the case with the lexicalized adjectives mentioned above). Also, in this use, -éos and -éi act on perfectly neutral bases and function similarly to collective suffixes, discussed in §6.4.3. It may be assumed that its presence in non-standard (e.g., dialectal) varieties make it [+colloquial], and, as such, it is characterized by low status and easily acquires pejorative meanings (see, e.g., Turkish loans in §6.4.1.4). However, the plural marking of -éi also seems to play a role in its pejorative function (for the involvement of the plural in expressive constructions see, for example, Stankiewicz 1964, for Greek Tsiakmakis et al. 2021, see also Konstantinidou 2004, 2005, Kallergi & Konstantinidou 2018).

6.4.1.8 The suffix -(i)líci Another Turkish loan, originating from the Turkish nominalizer -lik, the suffix -(i)líci/-íci202 is always marked for neuter (the -i.N.NEUT.SG ending is the result of the adaptation of the loan into Greek) and creates nouns on nominal bases (see Dangas (2012) and Kiranoudis (2001) for an extensive analysis). Similarly to -(a)dzís/-(a)dzú (also of Turkish origin, see §6.4.1.4), suffixation with -ilíci typically points to a profession or occupation, or, more generally, to a property or an activity with which a person occupies their time 199 200

201

202

Example personally attested by G. Katsouda. In Δημαρχαίοι, δεδδηαίοι και λοιποί νταραβεριτζήδες μπράβο! Κάνατε τους πολίτες χιονοπρόσφυγες και πετάτε το μπαλάκι ο ένας στον άλλο. At https://bit.ly/3Z6aMHj (accessed 18 February 2021). Αυτοί (οι εμβολιασμένοι εννοώ υπουργαίοι πρωθυπουργαίοι προεδραίοι βουλευταίοι δημοσιογραφαίοι κουλουπου) πως και μπαίνουν σε αεροπλάνα; At https://diodotos-k-t .blogspot.com/2021/06/blog-post_79.html (accessed 24 November 2021). See, e.g., Kiranoudis (2001), Dangas (2012) and the references there for the different forms (-líci and -íci) of the suffix -ilíci.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

183

(see Dangas 2012: 1025 for a wide array of semantic categories of the bases). The base-forms are semantically neutral and the derivatives can be pejorative (206–208) or, rarely, expressively neutral (209), but, in any case, highly [+colloquial] and [–formal], [–archaic], hence [–learned].203 (206)

ðiciɣorilíci (< ðiciɣóros ‘lawyer’) ‘the lawyer profession [–formal], [–learned]’

(207)

ipurʝilíci (< ipurɣós ‘minister’) ‘ministership [–formal], [–learned]’

(208)

proeðrilíci (< próeðros ‘president’) ‘presidentship [–formal], [–learned]’

(209)

mastorilíci (< mástoras ‘craftsman’) ‘craftsmanship [–formal], [–learned]’

Apart from the afore-mentioned function, which focuses on the description of abstract states, -ilíci also seems to denote individual events, actions or instances of behaviour (see also Dangas (2012: 1026) for examples that denote ‘object’, ‘action’ or ‘result of an action’, such as mezeklíci ‘small savoury dish’, facirílici ‘trick of a fakir’ and papadzilíci ‘gimmick’). According to Dangas (2012: 1026), this individuating function (the acquisition of the feature [+concrete] by the derivatives) is secondary to the meaning of human property, initially given by the suffix. Especially with respect to human action or behaviour, -ilíci often connotes persistency in the action on the part of the doer (the negative sense of typical) and expresses the speaker’s contempt towards the act itself or towards the person who acts. Depending on the base, it may also imply that the action is unfavourable, of low quality or indecent (even humiliating), and these implications point to a pejorative function. A large number of cases (some of which are present in dictionaries) involve bases with negative meanings (the base may actually be an insult) or with already pejorative connotations. (210)

karaɟoz(i)líci (< karaɟózis ‘a ridiculous person’) ‘ridiculous behaviour [+PEJ]’

(211)

maskarilíci (< maskarás ‘a ridiculous person’) ‘ridiculous behaviour [+PEJ]’

(212)

ʝiftilíci (< ʝíftos ‘Gypsy [+PEJ]’) ‘Gypsy-like behaviour [+PEJ]’

(213)

rezilíci (< rezíli ‘humiliation [+colloquial]’ ‘humiliating behaviour [+PEJ]’

In some cases, with either positive or negative nominal bases, the suffix, according to Kiranoudis (2001: 538), has a merely expressive function; for instance, ɣoitilíci in (215) is considered by Kiranoudis a case where the suffix does not contribute anything to the semantics of the derivation, but adds a

203

Kazazis (1972) and Tzitzilis (1997), referred to in Kiranoudis (2001: 542–3), argue that in the case of -ilíci derivations with a base-word that denotes profession, the professions are of a high socio-economic status, and this status is ironically undermined by the speaker through the use of the suffix (i.e., low quality is implied, despite the high status of the profession).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

184

Kallergi, Katsouda, Konstantinidou, Tsangalidis

component of emotionality or ironic disposition on the part of the speaker (examples from Kiranoudis 2001: 538): (214)

koroiðilíci (< koroiðía ‘mockery’) ‘mockery (emphatic)’

(215)

ɣoitilíci (< ɣoitía ‘charm’) ‘charm (ironic)’

In nonce formation, -ilíci is also widely used in its plural form -ilíca and with expressively neutral bases as well, in which case it has, almost pervasively, a pejorative function.204 In terms of register and style, -ilíca is even more informal and colloquial: it is observed in special argots (e.g., among fans of cars and racing) or marginal vocabulary, and presupposes a great amount of familiarity between speakers. Some instances of the ample availability of the suffix in nonce formation are the following attested examples: (216)

xoreftilíca ‘dancing[+PEJ]’

(217)

zorilíca ‘rough stuff, playing rough [+PEJ]’

(218)

ksilíca ‘beatings, aggressive acts [+PEJ]’

(219)

ɟeilíca ‘homosexual-like behaviour [+PEJ]’

(220)

gomenilíca ‘flirts, flirting[+PEJ]’

(221)

dzabadzilíca ‘acts or behaviour of cheapskates [+PEJ]’

(222)

ðeks(i)ilíca ‘typical behaviour of a right-winger [+PEJ]’

(223)

aristerilíca ‘typical behaviour of a left-winger [+PEJ]’

(224)

kanguro-alvanilíca ‘kitsch acts typical of Albanians [+PEJ]’205

(225)

cinezilíca ‘acts typical of the Chinese [+PEJ]’206

(226)

evropailíca ‘acts imitating North/West Europeans [+PEJ]’207

The strong pejorative effect of Examples (216)–(226) is the result of several factors: the meaning of the base, the effect of the plural, but most importantly the role of the suffix, which is inherently pejorative (Dangas 2012: 1028) and which is more prominent when the base is neutral (see Example 226). It is noteworthy that, as with -ácas above and nonce formation in general, the derivatives with -ilíci or -ilíca have a number of possible interpretations, 204

205 206 207

An exception to the pervasiveness of the pejorative function would be the noun badilíca sideilíca ‘car high-speed drifting’ (when the car goes so fast, that its sides drift). Its meaning can be considered neutral, unless we suppose a nuance of expressivity (in the sense of emotional involvement) as in ɣoitilíci in Example (215). At https://bit.ly/3ZmCPSv (accessed 18 December 2021). In Κλασικά κινεζιλίκια που πουλάν σε ασχετους συσκευες υποτιθετε κλεμενες iphone5–5s σε πολύ καλή τιμη. At https://agriogourouna.espivblogs.net . In Mετατρέψαμε την ταυτότητά μας σε μια ανάκατη σούπα απο αρχαιοελληνιλίκια, ευρωπαϊλίκια, αριστεριλίκια και άλλα παρόμοια. At https://amynagr.blogspot.com/2020/01/ blog-post_65.html (accessed 16 December 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

185

only resolved by context. Also, such derivatives are vague enough to refer to a large number of features relevant to the meaning of the base-form, as long as the features in question are perceived as contemptuous by the speaker. For example, cinezilíca in (225) may simultaneously refer to different (kinds of ) actions or behaviours, but all of them are characterized by a negative, stereotypical perception of the Chinese culture or of particular activities coming from China.

6.4.1.9 The suffix -(í)stikos A final case of suffixes that contribute the meaning of property or identity is -(í)stikos and its feminine and neuter allomorphs -(í)stici (FEM) and -(í)stiko (NEUT). The reason why three genders are available for the suffix is that it forms adjectives, which, in SMG, are generally marked for these three genders. Although the focus of the discussion is largely set on nominal suffixes, we mention -(í)stikos because its function is very similar to some of the nominalizers encountered earlier. As with other suffixes so far, there are lexicalized examples of derivation with -ístikos on nominal bases and these derivatives can be expressively neutral or even positive. This is observed when the base is positively expressive or stylistically unmarked itself, as in: (227)

kuklístikos (< ‘doll’) ‘dolly-like’

(228)

aɣorístikos (< ‘boy’) ‘boy-like, for boys’

(229)

koritsístikos (< ‘girl’) ‘girl-like, for girls’

(230)

pareístikos (< ‘company of friends’) ‘relating to companies of friends’

By contrast, when the base is stylistically or expressively marked (typically already suffixed by diminutives), negative connotations arise: (231)

θeatrinístikos (< actor.DIM ‘actor[+PEJ]’) ‘actorish, pretentious [+PEJ]’

(232)

ʝinekulístikos (< woman.DIM ‘woman[+PEJ]’) ‘typical of weak women [+PEJ]’

(233)

aðerfístikos (< sister ‘sister/faggot’) ‘typical of gay men [+PEJ]’

(234)

psonístikos (< crank ‘snob, crank’) ‘snobbish, eccentric [+PEJ]’

Examples from (231) to (234) have neutral equivalents, and for some of them the equivalent involves a totally different word, e.g.: θeatrikós ‘theatrical’, ʝinecíos ‘feminine, for women’, aðerfikós fraternal, brotherly’. Thus, with respect to the pejorative meanings of these instances, both the base and the suffix seem to play a role. However, in nonce formation, -ístikos may be used to mark property or identity by being attached to semantically and stylistically neutral bases as well, as Example (235) indicates. The suffix alone may evoke or intensify

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

186

Kallergi, Katsouda, Konstantinidou, Tsangalidis

negative connotations. Note that the meaning of the suffix, as already implied by Examples (231) θeatrinístikos and (232) ʝinekulístikos, involves an implication of resemblance with a prototype or stereotype (the -ish or -like effect in English). This, in turn, may sometimes imply that the referent is stereotypical and at other times that he/she is not original enough or not qualified enough to be named with the proper, neutral term. (235)

o pço sibaθís apó tus tris, an ce arcetá laicístikos, the most agreeable from the three if and quite populist-ístikos, aristerístikos, sirizístikos208 left-ístikos, SYRIZA-ístikos ‘the most agreeable of the three, although rather populistic, left-like, typical of SYRIZA’

(236)

o the

‘eklektós’ tu kózmu o proθipurɣós se stiláci chosen of. world the prime. in style. the minister DIM kutsavacístiko ce polí proθipurʝístiko209 pseudo-macho-ístiko and very prime.minister-ístiko ‘the one as the chosen of the world, the prime-minister with manners resembling pseudo-macho-men and very prime-minister-like’ énas one

o the

In any case, the function of -ístikos is not merely to create an adjective, but to express the speaker’s attitude towards the quality described or, more specifically, the degree to which the referent corresponds to a stereotype.

6.4.1.10 Summary of suffixes denoting property The first group of suffixes under discussion has exhibited a number of semantic, formal, and pragmatic regularities. Firstly, many suffixes denote an occupation or job, and others describe a human property or propensity, a tendency based on intense liking. A typical behaviour based on preference, intense liking, or a tendency implies repetitiveness, namely, repeated occurrences of specific actions and choices. Repetition, in turn, relates with emotions such as annoyance (see, e.g., Kakridi-Ferrari 1998 for pragmatic repetition, Dammel and Quindt 2016 for disturbance connected to frequentative meanings), thus the pejorative function of these suffixes seems to have a semantic basis. Secondly, most of the suffixes examined are nominalizers, but only a small portion of them changes the category of the base they attach to, which, for some theorists, makes them less prototypical in the sphere of expressive or evaluative morphology (see Bauer 1997). The category shifters we have discussed are -iá, -iáris/-iára, -ácas, and -ístikos. On the other hand, many suffixes (including some of the category-shifters, such as 208 209

At https://athens.indymedia.org/post/1108835/ (accessed 29 January 2022). At https://halyvourgos.wordpress.com/2013/11/page/2/ (accessed 29 January 2022).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

187

-ácas) are inherently pejorative and very productive in creating words with generally negative meaning, as well as very frequent in nonce formation for the expression of contempt and derogation in ludic, familiar, and highly informal contexts. In terms of style/register, all suffixes are unanimously [+colloquial], with some of them also being diachronically [–learned], viz. [–archaic], and they usually appear in oral language, often alongside slang or taboo vocabulary. However, as we have seen, a small part of this first group of suffixes exhibits formations with positive/meliorative meanings, which seems to be a lexical fact, that is, it depends either on the base of the derivation or the lexicalization process. In the case of -iáriko, which seems to mitigate the force of a negative base (with a meliorative effect), the attribution of neuter gender to non-neuter entities may also play a role (e.g., Melissaropoulou 2015 argues that the neuter is the unmarked gender type for diminutives and diminutives are often involved in the mitigation of the utterance force). Besides, in these formations the role of suffixes is primarily amplifying (since -iáriko forms adjectives and these negative bases are already adjectival), thus the amplified or emphasized meanings may be finally manipulated by context towards the desired direction. The role of context in the interpretation of intensification belongs to the discussion immediately following. 6.4.2

Augmentative and diminutive suffixes

As mentioned in the introduction to §6.4, augmentation, intensification, and diminution often associate with expressivity and especially the bipolarity of positive and derogatory meanings. In addition, the distinction between augmentation and intensification is not absolutely clear or agreed upon between theorists. Indeed, both intensification and augmentation relate to an ‘increase’ in quantity or quality (see, e.g., Dressler & Merlini Barbaresi 1994: 416, 437), and some scholars contend that augmentation (increase in size/‘enlargement’) is part of intensification (as semantic upgrading, see, e.g., Dressler & Merlini Barbaresi 1994, Gavriilidou 2010, 2013 for SMG), whereas others suggest the opposite (e.g., Minas 1978 for SMG). For our purposes, we will stick to the notion of BIGNESS or ‘increase in amount’ for augmentation. Here, we focus on augmentation, instead of intensification, as the former concept seems to be more clearly descriptive. As suggested in footnote 49, intensification has been dealt with as an increase in expressiveness, as a kind of amplification or emphasis of the expressive meaning intended. When an entity is qualitatively evaluated, ‘BIG X’ can be interpreted as ‘important X’ and express admiration (see also Merlini Barbaresi & Dressler 2020: 418). On the other hand, if augmentatives are exploited to mean excess

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

188

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(‘too big/too much of an X’), contempt and rejection is easily connoted.210 The pejorative uses of the diminutive may also be said to arise through its basic semantics and, in particular, the denotative feature SMALL (see Dressler & Merlini Barbaresi 1994, Regier 1998). These effects seem to result from the pragmatic exploitation of the semantics of the word-formation devices presented here. Thus, only the pejorative uses of such suffixes will be discussed below, and these uses cannot be claimed to form inherent parts of the suffix semantics.

6.4.2.1 Augmentative suffixes 6.4.2.1.1 The suffixes -a, -ára and -arás/-arú In the LKN, the augmentative suffix -a is presented as a separate lemma from the feminine inflectional suffix -a. However, it must be noted that generally the feminine gender in SMG is marked with respect to the masculine and adds or increases offensiveness, both in standard and marginal vocabulary (see Christopoulou et al. 2017). To give a characteristic example, scilí ‘dog. NEUT’ and scílos ‘dog.MASC’ are expressively neutral, whereas scíla ‘dog. FEM’ means both ‘female dog’ and ‘bitch’ in the insulting sense. Considering a crosslinguistic tendency of augmentatives to express negative expressivity (Klimaszewska 1983), the involvement of the feminine in lexical offensiveness or, even, the correlation of the meaning ‘female’ to pejoration (e.g., Mutz 2015), it is perhaps not surprising that the unmarked grammatical gender value for augmentatives in SMG has been claimed to be the feminine (Daltas 1985, Melissaropoulou 2015: 271). On the other hand, the masculine is equally strong or even prevailing, and according to Melissaropoulou (2007, 2009) it correlates with a stronger augmentative effect. The augmentative -a typically turns neutral nouns to feminine, adding the descriptive meaning ‘big’, as shown in the following examples.211 However, in some cases, actual big size is not the case. The derivatives with -a imply aggressiveness, grossness, nastiness, or unpleasantness within a familiar context, hence their meaning can be solely pejorative, although mildly so. 210

211

Of course, there are various views on the connection between augmentation/intensification and negative qualitative evaluation. For example, Efthymiou (2015: 59), drawing on Körtvélyessy’s proposal, mentions a possible metaphor at play, such as BIG IS NASTY (cf. also Körtvélyessy 2015). However, unlike other languages, e.g., Italian, where augmentative suffixes tend to have negative connotations (Dressler & Merlini Barbaresi 1994: 101), SMG augmentation seems to point to positive expressivity, at least with general vocabulary (see e.g., Daltas 1985). With respect to offensive vocabulary, augmentatives contribute to greater offensiveness (Christopoulou et al. 2017). The derivative cefála may be further suffixed with the masculine suffix -(a)s and the masculine cefálas ‘big-head’ is often used as an insult meaning ‘stubborn’ or ‘stupid’. Interestingly, -as appears as a separate suffix in the LKN that creates masculine nouns from feminine nouns and describes males related to the property described by the base, e.g., stafíðas (‘raisin-man’), xondropatátas (‘fat potato-man’). Since it adds no particular meaning to the formation, apart from assigning masculine gender and the feature [+animate], it can be considered a kind of classifier.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

189

(237)

maçéri ‘knife.NEUT’ maçéra knife.AUG.FEM ‘big knife [+PEJ]’

(238)

çéri ‘hand.NEUT’ çéra ‘big hand.FEM [+PEJ]’

(239)

cefáli ‘head.NEUT’ cefála ‘big head.FEM [+PEJ]’

(240)

ðáxtil-a ‘finger-NEUT.PL’ ðaxtíl-es finger-a.FEM.PL ‘big fingers [+PEJ]’

Example (240) is characteristically used in combination with the pejorative first constituent vromo- (see §6.3.3), as in (241), which is not accidental considering an observed tendency of expressive devices to combine with each other in order to maximize expressive effects (see, e.g., Example 24 and the discussion following Examples 346 and 348 in §6.4.3.1). (241)

vápste ta níça apó tis vromoðaxtíles paint.IMP the nails of the vromo-finger.AUG.FEM.PL ‘Colour the nails on your dirty damn fingers!’

sas212 yours

With bases that already carry a negative meaning, augmentatives seem to add intensity and to amplify pejoration (examples and translations are from Daltas 1985: 70): (242)

bekrís ‘drunkard’ békra ‘hopeless drunkard [+PEJ]’

(243)

mazoçistís ‘masochist’ mazóxa ‘hopeless masochist [+PEJ]’

Similarly, one of the most frequent SMG augmentative suffixes, -ára,213 functions as an amplifier with negative vocabulary, but typically expresses admiration with neutral vocabulary, as in (244) and (245) respectively: (244)

trelí ‘crazy.FEM’ trelára ‘totally crazy [+PEJ]’

(245)

tenía ‘movie’ teniára ‘great movie [+MEL]’

However, there are instances where -ára is pejorative with general vocabulary or neutral bases:214 (246)

éçis pjási ólo to kreváti me tis poð-áres su215 have.2SG catch.2.SG.PFV all the bed with the leg-AUG.PL yours ‘You have taken up the whole bed with your damn legs!’

(247)

esí ʝirnás tis selíðes pu íçe karaɣlípsi me tis ðaxtil-áres tu o proiɣúmenos216 you turn the pages that had kara-licked with the finger-AUG.PL his the precedent ‘You turn the same pages that someone before had licked[+PEJ] with his damn fingers’

212 213

214 215 216

At https://twitter.com/moglis__/status/1050098767037427712 (accessed 3 March 2023). Etymologically, -ára seems to relate to the augmentative -a, in that, according to the LKN, the former includes the latter in its structural make-up (viz. the hypocoristic -ári plus the augmentative -a). A lexicalized such case is elináras, which refers to a range of negative attributes of the contemporary Greek, including inconsiderateness, ethnocentricity, corruption, and others. At https://bit.ly/3Tu6kjA (accessed 5 March 2023). At https://twitter.com/kostasvaxevanis/status/1461276242586607624 (accessed 29 December 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

190

Kallergi, Katsouda, Konstantinidou, Tsangalidis

The sentential context in these examples is also negatively expressive, if one considers the imperative and the ‘right now’ expression in (246), as well as the use of another expressive device (the intensive prefix kara-) in (247). Moreover, there is nothing in the context that warrants the quantitative interpretation ‘big’ (especially in Example 247, where the owner of the fingers is most likely unknown). A final case that belongs to this family of suffixes is -arás. According to the LKN, like -ára, -arás also originates from the older hypocoristic -ári (see footnote 213), plus the suffix -ás/-ú discussed in §6.4.1.6. Thus, unlike the other two suffixes in this group, -arás has a feminine (-arú) and a neuter (-aráðiko/-arúðiko) form, that is, it does not carry a fixed gender value. Also, its semantics is not limited to augmentation, but includes the meaning ‘a person who has big X/does X a lot’ (as in ipnarás ‘one who sleeps a lot [+PEJ]’), that is, it adds the characteristic [+animate] and the semantic element of physical property or propensity (similarly to suffixes of the first category in this section). Along with these descriptive meanings, non-descriptive meaning aspects are also contributed by the suffix, ranging from melioration (with neutral/non-negative bases, as in 248– 249) to pejoration (with negative bases as in 250–251, or with bases referring mainly to body parts in highly informal or familiar situations, as in 252–256). (248)

pextarás (< péxtis ‘player’) ‘a great player [+MEL]’

(249)

ðuleftarás (< ðuleftís ‘worker’) ‘a hard worker [+MEL]’

(250)

pseftarás (< pséftis ‘liar’) ‘big liar [+PEJ]’

(251)

alitarás (< alítis ‘tramp’) ‘scum’

(252)

cilarás (< ciʎá ‘belly’) ‘a man with a big belly [+PEJ]’

(253)

kolarás (< kólos ‘ass’) ‘a man with a big ass [+PEJ]’

(254)

butarú (< búti ‘thigh’) ‘a woman with big thighs [+PEJ]’

(255)

mitarú (< míti ‘nose’) ‘a woman with a big nose [+PEJ]’

(256)

vizarú (< vizí ‘breast, tit’) ‘a woman with big tits [+PEJ]’

Pejorative uses of this suffix can also be found in the fossilized derogatory instances γermanarás German-arás ‘Kraut’ and elinarás Greek-arás ‘ultra Greek’ (ironic). In nonce formation, it seems typical of -arás to select nouns that denote occupation or profession, with which it functions pejoratively, mainly through irony. That is, while in its typical use it implies importance or superiority (the ‘great X’ meaning), in ironic contexts it implies arrogance or bad quality and expresses the speaker’s contempt, instead of admiration: (257)

217

ɣravat-oménos vuleftarás, poté ðuleftarás217 tie-PRTCPL.MASC MP.arás never worker.arás ‘A MP with a tie is never a great worker’

At https://bit.ly/3ECebFC (accessed 30 December 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek (258)

191

fitítria sti saloníci, ti strímokse énas kaθiʝitarás sto asansér218 student.FEM in.the Thessaloniki her cornered a professor.arás in.the elevator ‘When she was a student in Thessaloniki, a professor[+PEJ] came on to her in the elevator’

In sum, the meaning of derivatives with the augmentative -a, -ára, and -arás can be quantitative (‘big X’, ‘X a lot’) or qualitative/expressive (having a meliorative or ‘great X’ and a pejorative or ‘bad X’ meaning), depending on two factors: i. the semantic category of the base; ii. the sentential and pragmatic context.

6.4.2.1.2 The suffixes -úra and -urʝá These two suffixes are relatives in that, according to the LKN, the latter (-urʝá) consists of the former (-úra) plus the suffix -iá (discussed earlier in §6.4.1.2). In addition, they are both presented in dictionaries in similar ways: for -úra, the LKN suggests that it ‘intensifies pejoratively’ and, for -urʝá, that it ‘augments pejoratively’. Both also form feminine nouns. However, they exhibit some differences as well. The suffix -úra possibly originates in a nominalizing and collectivizing suffix (Italian and Latin -ura, as in Lat. strictura and Ital. armatura). In SMG it may form nouns from verbs, adjectives or nouns (see, e.g., 259, 260). (259)

xáno/xáso219 ‘lose’ xasúra ‘losing’

(260)

θolós ‘blurred’ θolúra ‘blur.N’

Although it is not mentioned in the LKN, it must be noted that even in this function, the suffix tends to add a negative or derogatory meaning to the derivative220 (but see Diamantidou 2017). In support of this claim, one may consider that the above words have other, neutral, nominal derivatives, viz. xásimo ‘loss’ and θolótita ‘blur.N’. In part, the stylistic profile of the suffixes involved may be responsible for the difference. For instance, -ótita is of archaic origin and neutral with respect to the [formal/informal] parameter (a ‘default’ choice in the sense of Manolessou et al. 2019 or close to the ‘norm’ in

218 219 220

At https://bit.ly/3Y5NQXt (accessed 30 December 2021). Here, the perfective stem xas- of the verb xáno is used. The participation of the perfective stem is the case across most instances of derivation with -úra. On the other hand, there seem to be a few lexicalized instances with -úra which are completely neutral, such as perpatúra (< perpató ‘walk’) ‘walker’ (device) and simaðúra (< simáði ‘mark’) ‘floating marker, buoy’. Also, the suffix can productively render a positive meaning, as in pextúra (< péxtis ‘player’) ‘great player’ and freskaðúra (< freskáða ‘freshness’) ‘extremely fresh’. Nevertheless, a positive meaning would very reasonably arise if we accepted an augmentative role for -úra in these cases. It must also be noted that all instances with positive meaning do not derive from the attachment of the suffix to verbal bases, but to nominal ones.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

192

Kallergi, Katsouda, Konstantinidou, Tsangalidis

the [+/–learned] continuum, Anastasiadis-Symeonidis & Fliatouras 2019). On the other hand, -úra is definitely [–learned]/[–archaic], [–formal], and [+colloquial]. Still, the effect of the suffix is pejorative in that xasúra and θolúra seem to express some kind of emotion, specifically displeasure with what the base denotes. Interestingly, xasúra ‘specializes’ in referring to loss of money in gaming or gambling, whereas θolúra is also metaphorically used to refer to mental haze or confusion.221 In this sense, -úra is a negative expressive marker (as a marker of emotions, attitudes, and personality elements of speakers, see Ochs & Schieffelin 1989). Diamantidou (2017) discusses -úra within a specific theoretical model,222 argues for its status as an intensifier and suggests that it carries the meaning of ‘an unpleasant and/or annoying sensation of the speaker’ (Diamantidou 2017: 966). As an augmentative/intensive pejorative marker, -úra also appears in nouns that refer to abstract entities, such as klapsúra ‘whining [+PEJ]’ and psematúra ‘lie [+PEJ]’, but also to actual objects, such as psomúra ‘a lot of/big bread [+PEJ]’, malúra (‘a lot of/big/unorderly hair [+PEJ]’), laspúra (‘a lot of mud [+PEJ]’), the latter of which are mentioned in the LKN in a separate lemma for the augmentative pejorative use of the suffix. In the same entry, the dictionary also mentions abstract denominal or deadjectival nouns, such as elinikúra (< eliniká ‘Greek’) and ɣalikúra (< ɣaliká ‘French’), which refer to ‘a pretentious Greek/French expression’ or more generally to ‘an annoyingly Greek/French characteristic’, laikúra ‘(too) folk [+PEJ]’, miðenikúra ‘an absolute zero [+PEJ]’, and others. Here, as Diamantidou (2017: 960) suggests, it is counterintuitive to suppose that augmentation is at play, since the nominal bases are not gradable (viz. magnifiable). In addition, the meaning of these derivatives does not involve any quantifiable increase, but an aspect of intensity, plus a negative stance. Augmentation may perhaps be assumed only as a metaphorical element (the excess or exaggeration implied by these derivatives) of the descriptive meaning of the derivation. With this expressive role, the suffix is also productive in nonce formation, with attested instances such as plastikúra ‘plastic object [+PEJ]’, cinezúra ‘chinese product [+PEJ]’ and politikúra ‘annoying political discussion’ or ‘pretentious political verbalism’. Notably, as with hapax legomena involving

221

222

Similarly, the noun zalúra ‘dizziness’ (< záli ‘dizziness’ ‘get dizzy’) has abstract interpretations as well (referring to mentally unpleasant situations, apart from physical dizziness), whereas the neutral terms záði, zaláða ‘dizziness’ does not. The same is observed for anakatosúra (< anakatévo/anakatos- (perfective stem) ‘stir’) where the abstract pejorative meaning ‘mess, commotion’ is lexicalized and does not appear with the neutral nominal counterpart anakátema ‘stirring’. For a similar analysis and relevant examples, see Diamantidou (2017). The particular model was postulated by D. Corbin as presented in Anastasiadis-Symeonidis (1992).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

193

other suffixes, the meaning of nonce formations with -úra may be unclear/ ineffable and open to different interpretations, which are resolved through context (see, e.g., -ácas in §6.4.1.5). Unlike -úra, the suffix -urʝá selects only nouns as bases and involves an extra interpretation, which -úra lacks. Apart from augmentation or intensification, it may also have an expressive collective meaning. In both meanings, derogation arises though the suffix, since the bases are relatively neutral:223 (261)

laspurʝá (< láspi ‘mud’) ‘a lot of mud [+PEJ]’

(262)

plastikurʝá

(< plastikó ‘plastic.N’) ‘plastic object [+PEJ]’

(263)

vlaxurʝá (< vláxos ‘Aromanian/Vlach’) ‘the Aromanians/Vlachs [+PEJ]’

(264)

ʝifturʝá

(265)

elinurʝá (< élinas ‘Greek’) ‘the Greeks [+PEJ]’224

(266)

alvanurʝá (< alvanós ‘Albanian’) ‘the Albanians [+PEJ]’

(267)

papaðurʝá (< papás ‘priest’) ‘the priests [+PEJ]’

(< ʝíftos ‘Gypsy’) ‘the Gypsies [+PEJ]’

In this latter use, it is reminiscent of its origin (Lat./Ital. -úra) as well as its other consisting element, -iá which also has a collective interpretation. As in the case of -iá, of course, the collective meaning of -urʝá is limited to highly informal, familiar and colloquial contexts.

6.4.2.1.3 The suffixes -ákla, -úmba, and -óŋga There are other augmentative suffixes involved in negative expressivity which exhibit moderate or limited productivity. These include -ákla, -úmba, and -óŋga, all of which are typically marked for feminine gender and can be found only in highly informal and colloquial speech (including slang). For -ákla, the same observations as with typical augmentatives can be made; the suffix may create words that express admiration, as in: (< arçiɣós ‘chief’) ‘a big/amazing chief [+MEL]’

(268)

arçiɣákla

(269)

korákla (< kóri ‘daughter’) ‘a big/beautiful daughter [+MEL]’

However, derivatives with -ákla may have connotations of undesired size or undesired presence. For instance, mitákla in (270) is a big nose for which the 223

224

Note, however, that certain ethnic names have negative connotations, drawn from the stereotypes related to the specific ethnic groups. For instance, as discussed in §6.3.2 and §6.3.1, ʝíftos ‘Gypsy’ is already negatively loaded, whereas vláxos ‘Aromanian/Vlach’ is often used as an insult itself (meaning ‘peasant, redneck, hillbilly’). In Έχετε το δίκιο σας βέβαια, με την αλητεία που δέρνει την Ελληνουριά ‘But you have the right to complain, of course, with all the vagrancy characterizing the Greeks’. At https://bit.ly/ 3xYIMtk (accessed 31 December 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

194

Kallergi, Katsouda, Konstantinidou, Tsangalidis

speaker expresses negative feelings. However, big size may not be the case at all and the suffix may be only expressive of contempt, as in (273). (270)

mitákla (< míti ‘nose’) ‘big nose [+PEJ]’

(271)

spirákla

(< spirí ‘pimple’) ‘big pimple [+PEJ]’

(272)

çerákla

(< çéri ‘hand’) ‘big hand [+PEJ]’

(273)

aplóni ti çerákla tu ce aómatos eleíste ton aómato!225 stretch.3SG.PR the hand.AUG his and blind mercy the blind ‘He stretches his goddamn hand and (shouts) . . . ‘Blind . . . help the blind!’’

The other two suffixes -óŋga and -úmba seem to retain augmentation or intensification as a standard meaning, along with pejorative connotations. The -óŋga suffix seems relatively restricted with neutral bases, as only the few instances following have been found.226 However, it is interesting that -óŋga may attach to a base-form already involving augmentative suffixation. That is, the derivatives mitaróŋga and vizaróŋga in (277) and (278) exhibit double augmentative marking, which maximizes the intensifying effect. (274)

psilóŋga ‘very tall [+PEJ]’227

(275)

mitóŋga ‘big nose [+PEJ]’

(276)

vizóŋga ‘big tit [+PEJ]’

(277)

mit-ar-óŋga nose-AUG-AUG ‘big nose [+INTENSE.PEJ]’

(278)

viz-ar-óŋga tit-AUG-AUG ‘big tit [+INTENSE.PEJ]’

Finally, the suffix -úmba may have positive effects according to the meaning of the base, as in salatúmba (< saláta ‘salad’) ‘big, nice salad’228 and fredúmba (< frédo ‘(a type of ) cold coffee’) ‘a nice freddo coffee’.229 It also seems to be used quite productively in military argot (see Asimopoulos 2016), in which it is usually found in combination with bases describing positive 225

226

227 228 229

In (Μιλαω παντα για τους ψευτοαναπηρους). Απλωνει την χερακλα του και ... Αοματοςςςς . . . Ελεηστε τον αοματοοοοο . . . ‘(I’m talking about the false handicaps). He stretches his hand [+PEJ] and . . . Bliiiind . . . Help the bliiiiind . . .’. At https://bit.ly/3xXecAh (accessed 3 January 2022). However, it seems that -óŋga often appears with negative or insulting vocabulary, as in trelóŋga ‘crazy[+PEJ]’, murlóŋga ‘crazy[+PEJ]’, paparóŋga ‘something stupid [+PEJ]’, vromóŋga ‘dirt[+PEJ]’. Also, in LNEG (entry τρελαίγκω), trelóŋga is paralleled with treléŋgo, thus, -óŋga and -éŋgo (see fn 152) may possibly have arisen from similar developmental paths and a common root. At https://bit.ly/3mfrNAb (13 March 2022). In Και μια σαλατουμπα και μια χαρα!!!! ‘And [we will make/eat] a big nice salad, and it will be fine!’ At https://bit.ly/3masawb (accessed 3 January 2022). In Να έχουμε να θαυμάζουμε απίθανα κόλπα από τα πιτσιρίκια, καθώς θα πίνουμε τη φρεντούμπα μας το καλοκαίρι. ‘In order for us to admire the gimmicks of the youngsters, while we will be drinking our nice freddo coffee in the summer’. At https://peravre.blogspot.com/2010/02/blogpost_18.html (accessed 13 March 2022).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

195

concepts for the army (exoðúmba < ‘exit [+MEL]’, aðiúmba < ‘day-off, leave [+MEL]’). However, it can also have negative connotations and may express contempt or annoyance. These feelings are presumably directed towards the large amount or big size of the referent, denoted through suffixation with -úmba, or, rather more specifically, at the excessive amount connoted by its use in certain cases. For example, with the base spirí ‘pimple’, the already negative connotations seem to be reinforced. In the rest of the following instances, excess is implied as the target of pejoration: (279)

laðúmba (< láði ‘oil’) ‘(too) large amount of oil’230

(280)

xartúmba (< xartí ‘paper’) ‘(too) large number of documents’231

(281)

spirúmba (< spirí ‘pimple’) ‘a big/annoying pimple’

(282)

nixúmba (< níçi ‘nail’) ‘long nail [+PEJ]’232

(283)

vizúmba (< vizí ‘tit’) ‘big tit [+PEJ]’233

(284)

malúmba (< malí ‘hair’) ‘big/bulky/lots of hair [+PEJ]’234

Interestingly, -úmba is not mentioned in the LKN, nor is there, to our knowledge, etymological research available for it. Nevertheless, it seems quite productive in nonce formation with bases belonging to general vocabulary. Kamilaki et al. (2016) mention it among suffixes present in taboo derivation and suggest that it appears more frequently in the plural (2016: 33). It also seems to have considerable ludic effects (together with the rest of the suffixes in this category and perhaps more than the most typical or frequent augmentative in the previous categories), but this intuition should be empirically tested.

6.4.2.2 Diminutive suffixes It is a common ground of studies on diminutives that they play a significant role in verbal expressivity and that physical smallness is only a minor part of

230

231 232

233

234

In Αυτές με το λεπτό ζυμαράκι χωρίς λαδούμπα, με τις ελαφριές γεμίσεις που δεν έχουν επάνω την Άρτα και τα Γιάννενα. ‘Those [pizzas] with the thin little dough, without too much oil, with the light stuffing which does not include all sorts of (redundant) materials’. At https://bit.ly/ 41JcjVJ (accessed 19 March 2022). Example personally attested by H. Kallergi. With reference to long toenails in men, in Μείνε καλύτερα με τη νυχούμπα, να σε βοηθάει να κρέμεσαι και από τα κλαδιά των δέντρων. ‘Better stay with the long nails [+PEJ], in order to be able to hang from the trees more easily’. At https://bit.ly/3Z7vxSS (accessed 19 March 2022). In Κάνε κάμποσα λάικ στο χαζομισοπούτ@νο που σου πετάει φωτό με τη βυζούμπα έξω. ‘Give several likes to the stupid little slut who throws a picture to you with her big tit visible’. At https://bit.ly/3mcYn5Z (accessed 19 March 2022). In [. . .] τις κουκλίτσες ολόγυμνες με τα τατού και την μαλούμπα-απλύτου για κάνα μήνα. ‘[. . .] the dollies all-naked with their tattoos and the long hair unwashed for about a month’. At https://bit.ly/3SyoHDs (accessed 20 March 2022).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

196

Kallergi, Katsouda, Konstantinidou, Tsangalidis

their semantics cross-linguistically (e.g., Dressler & Merlini Barbaresi 1994, Jurafsky 1996 and references there). Efthymiou (2015: 61), reporting on Körtvélyessy’s (2012)235 model of evaluative semantics, mentions diminutive meanings that range from quantitative evaluation (e.g., attenuation), representing a descriptive or truth-conditionally relevant meaning aspect, to qualitative evaluation (e.g., affection, or a positive or negative attitude) and discursive meanings (e.g., mitigation of utterance force, politeness, intimacy, or irony) that represent non-descriptive meaning aspects. There is a large number of diminutive suffixes in SMG, all of which may potentially have all of the aforementioned meanings (see, e.g., Babiniotis 1969, Efthymiou 2015, Melissaropoulou 2015, Katsaros 2018).236 In this section, however, we will be dealing with only a small number of diminutives, indicative of the different shades of pejoration that SMG diminution is available for. More specifically, we will discuss the pejorative functions of certain diminutive suffixes, which will be grouped into two categories, on the basis of a formal or functional property. The first category includes the suffixes -áci, -ákos, -uðákos/-uðáci and -ácis, which all involve the basic diminutive morpheme -ak-. The second group consists of the suffixes -ário and -ískos, which are both [+learned], in the sense of being [+archaic] and belonging to a high register of use (see the introduction to §6.4), and which, at least in productive word-formation, seem to be inherently pejorative. Across both groups, what is meant of pejoration is primarily the meaning of mitigation or underestimation (viz. the meaning of ‘unimportant’ or ‘worthless’, instead of ‘bad’ with augmentatives), although, on closer inspection, there may be extra or unique functions for some of these suffixes.

6.4.2.2.1 The suffixes -áci, -ákos, -uðáci/-uðákos, and -ácis One of the most basic and frequent diminutive suffixes in SMG is -áci (Stavrianaki 2002, Melissaropoulou 2015). The rest of the suffixes in this

235 236

Körtvélyessy (2012) is an unpublished paper. Elsewhere, we refer to the same work, in its published version of 2015. For instance, Katsaros (2018) offers examples in which various diminutive suffixes, such as -úkos (‘DIM.MASC’), -úlis (‘DIM.MASC’) and -úla (‘DIM.FEM’) are pragmatically exploited to express various discourse conditions, such as humour, friendliness, irony, or ‘friendly irony’ (Katsaros 2018: 319). Some diminutives, also, have special ironic effects due to their gender marking. On the one hand, linguistic gender (whether lexical or grammatical) is often exploited for social purposes and especially derogatory speech, hate speech or insult (see §6.4.2.1.1 for cases of marked lexicalization of feminine forms); on the other hand, irony is generally created by the co-presence of incompatible elements. Thus, feminine diminutives, such as -úla and -ína, are often involved in formations where the referent is masculine (e.g., stratiot-ína soldier-DIM.FEM ‘woman soldier’ referring to a man), which has strong ironic and derogatory effects.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

197

paragraph formally derive from -áci: specifically, -ákos and -ácis involve -akplus the masculine inflectional suffixes -os and -is respectively. The suffixes -uðákos and -uðáci, on the other hand, are compound suffixes consisting of a weakened diminutive (-úð-i) and -ákos and -áci respectively. Being marked for neuter gender, -áci can turn nouns marked for masculine or feminine into neuter. According to scholars (e.g., Corbett 1991: 227–8), the neuter correlates with the diminutive, since it often reflects the less than typical gender distinctiveness (or ‘sex indefiniteness’) characterizing young animate entities (including human babies, who similarly to young animals, seem to lack distinctive features of their sex and, so, all look alike). Also, as Melissaropoulou (2015: 274) claims, the neuter is the unmarked gender value for diminutives in SMG. For this reason, and for our descriptive purposes, we shall consider -áci central to the specific group of suffixes under discussion. Note that other suffixes in this ‘family’ seem to carry fixed gender values, since they may select bases with different genders than their own (e.g., katérina ‘Catherine’ > katerin-ákos ‘Catherine-DIM.MASC’). Stavrianaki (2002) focuses on -áci, studying the relevant lexicographic resources as well as its different functional manifestations as a diminutive (e.g., a conveyor of the meaning SMALL, as in trapez-áci ‘table-DIM’) and a hypocoristic (a marker of endearment, affection and intimacy, as in nicknames such as mar-áci ‘Maria-DIM.NEUT’) and a large number of in-between cases, including lexicalized forms (e.g., sakáci ‘blazer’) and lexicalized pejoratives (e.g., xorʝat-áci villager-DIM ‘peasant[+PEJ]’). She analyses her data within Corbin’s (1987, 1991) theoretical framework, exploring the ways in which all the different meanings can be unified around a semantic centre. A clear-cut distinction between descriptive and connotative meanings of -áci is not her objective, since, as she concludes, it may be the case that both meanings are at play at the same time.237 Instead, based on several semantic regularities of the various instantiations, she suggests (Stavrianaki 2002: 612–3) that the semantic instruction given by the suffix in derivation is the following: ‘mark the referent as a person, object or situation with which the speaker feels familiar (ized) enough to express an evaluative judgement on the basis of a quantitative or qualitative norm (size, emotional value, competence and so on)’. Depending on the meaning of the base and the context of utterance, it is decided whether actual small size applies, and along which axis the speaker appropriates his knowledge of the referent and expresses a judgement. In this way, Stavrianaki

237

Stavrianaki (2002: 608) correctly argues that in many cases it is not easy to exclude the possibility of the literal diminutive reading in expressive, connotative uses of the suffix (as in ʝatruð-áci doctor-DIM, which may refer to a young doctor), nor can one exclude the connotative use of a diminutive form for a referent that is actually big (e.g., spit-áci house-DIM for a big house, which is nevertheless endeared by its owners).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

198

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(2002: 613) argues, there is no need to define whether the meaning of an -áci derivative is either denotative or connotative, a choice that might miss the point of its special function (according to which, it may both denote and connote). In the case of the pejorative use of the suffix, the speaker, according to Stavrianaki’s analysis, expresses a stance characterized by features such as knowledge of the field of experience upon which the evaluation is made and an assumption that the situation referred to by the base is easily handled or carried out. This is particularly evident in instances where the base refers to intellectual or artistic products, such as ptiçi-áci (a university degree-DIM), traɣuð-áci (song-DIM), arθr-áci (article-DIM), which may all express underestimation or contempt. Another case where -áci has pejorative function is where the base denotes place of origin or the derivative describes a property belonging to a group. For example, apart from xorʝat-áci (villager-DIM ‘peasant[+PEJ]), lexicalized cases include amerikan-áci (American-DIM) and eparçiot-áci (provincialDIM). These may refer to people of any age (preferably young) who strongly exhibit properties of the culture in which they were raised. These properties are evaluated on the basis of negative stereotypes (e.g., eparçiotáci as someone raised in the provinces is a naïve person or one lacking manners or high education). Similarly, the nonce form pasok-áci PASOK-DIM has been attested with the meaning ‘a person belonging to the political party of PASOK, to which they show great loyalty’ and seems to be uttered ironically or contemptuously by the speaker.238 One more area in which -áci functions pejoratively is nouns denoting professions and occupations, most of which are in fact considered of high status. Among the most-frequently used items involve -ákos and -uðákos/-uðáci: (285)

kaθiʝitáci (< kaθiʝitís ‘professor’) ‘professor-DIM[+PEJ]’

(286)

fititáci (< fititís ‘university student’) ‘student-DIM[+PEJ]’

(287)

ðiciɣoráci (< ðiciɣóros ‘lawyer’) ‘lawyer-DIM[+PEJ]’

(288)

ðaskalákos (< ðáskalos ‘teacher’) ‘teacher-DIM[+PEJ]’

(289)

ðimosioɣrafákos (< ðimosioɣráfos ‘journalist’) ‘journalist-DIM[+PEJ]’

(290)

ipalilákos (< ipálilos ‘employee, clerk’) ‘clerk-DIM[+PEJ]’

(300)

meternixákos239 (< méternix ‘Metternich’)240 ‘Metternich-DIM[+PEJ]’

238 239 240

At www.xanthipress.gr/anipomono-pasokaki-berdepse-tis-thesis (accessed 8 January 2022). Καὶ Μετερνιχάκος ἀνέγνων χθὲς ἔν τινι ἐφημερίδι (Hatzidakis 1916: 34). Klemens von Metternich (1773–1859), Austrian politician and diplomat.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek (301)

ʝatruðáci (< ʝatrós ‘doctor’) ‘doctor-DIM[+PEJ]’

(302)

ʝatruðákos (< ʝatrós ‘doctor’) ‘doctor-DIM[+PEJ]’

199

As typical ‘reductors’ of size, -ákos/-áci suffixes reduce the prototypical characteristics attributed to a profession, so that the derivatives above present people who are ‘less than’ good versions/instantiations of profession X. Thus, such derivatives generally express the speaker’s idea that the referents lack experience, status, or professionalism. More specially, they are used to mitigate the referent’s importance. Finally, reference should be made to the particular role of -ácis. The LKN mentions its context-dependent pejorative function, giving the lexicalized example kozmácis (< kózmos ‘people’) ‘poor people’. In fact, kozmácis, typically referring to the stereotypical proletariat or ordinary people who lack opportunities to enjoy goods, expresses empathy or condescension, but not contempt as Examples (285)–(302) do. Moreover, -ácis seems to have acquired a derivative function, especially rich in nonce formation, which focuses on framing a situation and expressing irony over it. The former is particularly evident in the nominalization of phrases, as in (304)–(306), in which irony is successfully rendered, in our view, by means of the specific semantic instruction of -ácis. (303)

baxalácis (< báxalo ‘fuss, ‘(young) people usually involved in riots and vandalisms razzmatazz’) [+PEJ]’

(304)

isapostácis (< ísi apóstasi ‘people who keep the same distance from every side of an ‘equal distance’) argument, moderate[+PEJ]’

(305)

kaloçimonácis (< kaló çimóna ‘people who tend to wish “happy winter” all the time ‘good winter’) [+PEJ]’

(306)

citapsiláciðes (< cíta psilá ‘look up’) ‘those who look up all the time [+PEJ]’241

(307)

savatáciðes (< sávato ‘Saturday’) ‘people who prefer going out on Saturdays [+PEJ]’

In this use, -ácis is very similar to -ácas (discussed in §6.4.1.5). First, its interpretations relate with human propensity, preference, and repetitive behaviour, as those of -ácas. Second, it is used as a classifier (ταξικός σηματοδότης), that is, it signals the category type of nonce forms, especially when these come from the univerbation of syntactic phrases. More specifically, -ácis is used instead of a simple inflectional suffix in this case (*kalo-çimóna-s ‘goodwinter-MASC’). In this process of nominalization, the output is initially masculine, as with -ácas, however, feminine forms exist for the -ácis derivatives (e.g., baxalácisa ‘a woman who often takes part in riots’). Last, but not least, the two suffixes share the same plural form -áciðes. Thus, it might be assumed that the plural forms of -ácas have been reanalysed into forms in -ácis. Besides, -ácas

241

Referring to the film ‘Don’t look up’ (2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

200

Kallergi, Katsouda, Konstantinidou, Tsangalidis

originates in -ácis (see Dangas & Manasis 2018: 324 and references therein). Thus, the two suffixes may be considered synonyms with respect to pejoration. On the other hand, we agree (together with Dangas & Manasis 2018) that -ácas is not diminutive any more (cf. Daltas 1985). By contrast, -ácis retains its diminutive semantics, having the effect of ironic endearment, paradoxical belittlement (in the case of serious issues being referred to) and a concomitant high degree of ludicrousness. Note that equivalent forms may co-exist; for instance, both baxalácis and baxalácas are synchronically attested. However, according to our native intuitions, baxalácis sounds more ironic (and perhaps, humorous), whereas baxalácas focuses on the meaning ‘a person with an extreme liking towards or obsession with doing X’, and, on this basis, it is perhaps more pejorative.

6.4.2.2.2 The suffixes -ário and -ískos These two suffixes are moderately productive, perhaps due to their strong [+learned] character. Both of them are [+archaic], hence marginal, and characterized as [+high register] (e.g., used in formal speech), whereas, as we have seen, pejoration mostly arises in informal, familiar, and colloquial speech (often combining marginal, taboo, or insulting vocabulary), which is not suitable ground for the [+learned] feature. However, precisely because of the incompatibility between learned style and informal contexts, -ário and -iskos may have powerful pejorative effects.242 The suffix -ário is marked for neuter gender and forms part of several lexicalized instances, whereby the base is [–animate] and the diminutive meaning (‘small’) is relatively opaque: vivliário ‘health/account booklet’, oário ‘ovum’, orário ‘working time/opening hours’, solinário ‘tube’ (as of toothpaste). However, in its use with [+animate], [+human] bases, -ário is typically expressive of contempt and derogation: (308)

fititário (< fititís ‘university student’) ‘student[+PEJ][+learned]’

(309)

kaθiʝitário243 (< kaθiʝitís ‘professor’) ‘professor[+PEJ][+learned]’

(310)

iθopiário244 (< iθopiós ‘actor’) ‘actor[+PEJ][+learned]’

(311)

vuleftário245 (< vuleftís ‘MP’) ‘MP[+PEJ]’

(312)

ðespinário (< ðéspina or ðespinís ‘(young) lady’) ‘young lady [+PEJ]’

242 243 244 245

See also Kamilaki et al. (2016) for the copresence of [+learned] items with taboo vocabulary in SMG for a variety of functions including humour and pejoration. Πριν αναδειχτεί σε διάφορες θέσεις, διετέλεσε ‘καθηγητάριο’. At https://bit.ly/42p4rZJ (accessed 8 November 2021). Το φιλόδοξο, διαπλεκόμενο συγγραφάριο/σκηνοθετάριο/ηθοποιάριο. At www.kulturosupa.gr/ theatromania/2000-48489/ (accessed 20 January 2023). In Δουλοπρεπές βουλευτάριο. At https://sarantakos.wordpress.com/2012/10/25/glezos/ (accessed 8 November 2021).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

201

The pejorative function of -ário with [+animate]/[+human] bases seems to be inherited from earlier stages of the language, since analogous instances can be found in Ancient Greek: ἀνθρωπάριον [anthrōpárion] ‘pitiful fellow’, γυναικάριον [gunaikárion] ‘woman[+PEJ]’, ἀνδράριον [andrárion] ‘man[+PEJ]’, δουλάριον [dulárion] ‘female slave [+PEJ]’. In fact, the SMG data seem to revive an earlier tradition, that of a connection between the Ancient Greek diminutive árion and pejoration. In this connection, it may be assumed that the suffix -ion may have played a role, because, as noted by Stylianou (2018: 346–7, fn 1; Papyrus Larousse 29.475), -ion used to have a similative pejorative meaning (e.g., Ancient Greek ἀνδρίον [andríon] ‘similar to a man, but so small, cowardly or worthless that he should not be called a man’). As Chantraine (1933) suggested (noted in Grandi 2011: 17), the notion of similarity may relate either to inferiority or small size.246 Thus, for example, SMG vuleftário (as in 311) retains a nuance of meaning, according to which the referent is similar to an MP, but not a proper MP, according to speaker standards. Despite the archaic and learned character of -ário, SMG instances as in (308–312) are not used only by people who exhibit a learned profile or speak within a framework that cherishes [+learned] language (such as, e.g., the speech of the clergy). They can be used by ordinary people in everyday conversations and the [+learned] -ário appears beside [–learned] vocabulary, as in (313): (313)

i mariána, éna elefθeriázon ce plúsio ðespinário, daraverízete erotiká me ton mário247 ‘Marianna, a liberal and rich young lady[+PEJ][+learned], has an affair[–learned] with Marios’

The paradoxical appearance of a [+learned] element in a [–learned] context leads to the reinforcement of pejoration; the speaker not only expresses contempt towards the referent by the use of the -ário derivative, but also places himself at a greater distance from the referent, retaining the [+high register] for himself. In this way, the speaker seems to express the least identification with or intimacy towards the referent. Similar observations can be made about -ískos, which has been studied by Stylianou (2018). She notes that the archaic form -ískos is found today in words of two types: i. lexicalizations where -ískos has a neutral and more or less transparent diminutive function, such as lofískos (< lófos ‘hill’) ‘small hill’,

246 247

In the words of Chantraine (1933: 64): ‘ce qui ressemble à une chose peut lui être inférieur, ou être plus petit.’ [‘that which resembles something may be inferior or smaller’, translation ours]. At https://bit.ly/3IBdoGj (accessed 8 January 2022).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

202

Kallergi, Katsouda, Konstantinidou, Tsangalidis

kolpískos (< kólpos ‘bay’) ‘small bay’ and ðromískos (< ðrómos ‘road’) ‘small road’; and ii. instances whereby the base denotes a [+animate], [+human] entity, and, in this case, the meaning is pejorative (examples mostly taken from Stylianou 2018): (314)

ðiciɣorískos (< ðiciɣóros ‘lawyer’) ‘lawyer[+PEJ][+learned]’

(315)

ðimosioipalilískos (< ðimósios ipálilos ‘civil servant’) ‘civil servant[+PEJ][+learned]’

(316)

vuleftískos (< vuleftís ‘MP’) ‘MP[+PEJ][+learned]’

(317)

arçiʝískos (< arçiɣós ‘leader’) ‘leader[+PEJ][+learned]’

(318)

iʝemonískos (< iʝemónas ‘hegemon’) ‘hegemon[+PEJ][+learned]’

(319)

apateonískos (< apateónas ‘fraud’) ‘fraud[+PEJ][+learned]’

(320)

filoloʝískos (< filóloɣos ‘philologist’) ‘philologist[+PEJ][+learned]’

According to the LKN, -ískos appears in Ancient Greek, where it was already used to denote similarity, hypocorism or diminution and pejoration (e.g., Ancient Greek ὀβελ-ίσκος [obelískos] (< ὀβελ-ὸς ‘skewer’) ‘small skewer/obelisk’ (a built artefact that resembles a skewer), νεαν-ίσκος [neanískos] (< νεαν-ίας ‘young’) ‘a youngster’, ἀνθρωπ-ίσκος [anthrōpískos] (< ἄνθρωπ-ος ‘(hu)man’) ‘a small/worthless person’). Its etymology goes back to Indo-European -isko (‘similar to’) (Stylianou 2018: 346), hence, etymologically related are also the English suffix -ish, the German -isch, the Lithuanian -iškas, and suffixes in many other European languages. Thus, SMG has inherited these functions, keeping the pejorative one as the most prominent. In addition, the doubtless [+archaic] status of the suffix makes it [+learned] in a stylistic way, that is, as a [+high register] item. Again, as with -ário, the referent of the base-word is presented as similar to, but less than prototypical X, as non-original, as a fake. According to Stylianou, in this pejorative function, where the status and importance of the referent is being questioned, suffixation with -ískos is productive in nonce formation, whereas, with -ískos as a diminutive suffix, it is much more difficult to get new lexicalizations (Stylianou 2018: 352). However, according to our native speaker intuition, -ískos as a [+learned] pejorative marker seems to be used mainly by older people, whereas its productivity in the speech of the young and other marginal vocabularies seems to be low. The same assumption would be worth testing for -ário.

6.4.2.3 Summary of augmentatives and diminutives In §6.4.2.1 and §6.4.2.2, we examined the pejorative functions of indicative cases of augmentative and diminutive suffixation. Regarding augmentative suffixes, we have discussed two groups consisting of etymologically related suffixes, plus a heterogeneous group consisting

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

203

of rather marginal suffixes. The majority of the first group (-a, -ára, and -arás/-arú) carry a fixed gender value, viz. feminine. Meaning-wise, -arú and its equivalent masculine -arás can be considered a mixed case, since they not only encode augmentation, but their augmentative meaning is also conflated with the meaning ‘a person who has (large) X’, that is, with the meaning of human property or propensity typically characterizing the first group of suffixes (§6.4.1). In this function, which may be pejorative, these suffixes mainly attach to bases denoting body parts. In other augmentatives (including the -úra/-urʝá group but also -ákla, -óŋga, and -úmba, all standardly marked for feminine), pejoration arises mainly through the pragmatic exploitation of augmentation, irony, and the semantics of the base, but we have also observed pejorative meanings productively arising outside augmentative interpretations and with neutral bases. Notably, -úra appears to bear an inherently expressive component in standardly connoting displeasure with the abstract entity denoted by the derivative. Finally, -urʝá additionally involves a collective meaning, a non-existent option among other augmentatives. With respect to diminutive suffixes, we have again divided our cases into groups. We may call the first group the -ak- family (-áci, -ákos, -uðáci/-uðákos and -ácis), since all of these suffixes morphologically connect to the basic diminutive -áci. Gender values on these suffixes are in some cases fixed (-áci and -ákos). Their bases with respect to pejorative functions are very often nouns denoting professions. The same is observed with the other diminutive group, -ário and -ískos. Pejorative meanings arise, in the case of the first group, through the pragmatic exploitation of diminutive semantics or arguably through the semantic instruction of -áci, which, according to certain analyses, points to subjective quantitative or qualitative evaluation. More specifically to pejoration, -áci and its related suffixes express contempt and underestimation, often accompanied by an ironic attitude on the part of the speaker. Irony is strongly evident in the second group of diminutives as well (-ário and -ískos). Note that their pejorative functions seem to be inherited from earlier stages of Greek. however, their [+learned]/ [+archaic] status marks a sharp contrast with their use in colloquial, informal contexts, forming a good ground for irony and the expression of negative attitudes. 6.4.3

Collective suffixes

Under this rubric, we will discuss several suffixes that generally convey the meaning ‘a group of/a collection of X’ (περιληπτικά ‘inclusive’ in SMG). Again, these are presented in two subcategories; in the first, the suffixes are etymologically related (-aría and -arʝó), whereas, in the second (-lói, -máni), they bear strong similarities with respect to function and productivity. The collective meaning seems to relate to pejoration cross-linguistically. As various scholars observe, the lack of specific reference to the individual entities

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

204

Kallergi, Katsouda, Konstantinidou, Tsangalidis

of the group or collection and the implicated lack of distinctive identity within the group easily motivate the assumption of ‘non-noteworthiness’ of the individual entities of the group (Koo & Rhee 2016: 317). Hence, many constructions or morphological devices across languages, such as suffixes or constructions that mean ‘X and the like’ or other ‘reductive’ constructions (where many referents are ‘squeezed’, as it were, into one word or expression) acquire dismissive meanings (‘X and other similar objects/people, not worth mentioning’) and/or pejorative functions, such as the expression of contempt (whereby unimportance of the referent is implied).248 This does not, of course, imply that all formations with reference to a group of similar entities have pejorative connotations (see, e.g., Efthymiou 1999 for other SMG cases, such as eleónas ‘olive grove’, xristçanosíni ‘christianity’, peθeriká ‘in-laws’). However, as Efthymiou (1999: 337) suggests for such cases, the collective meaning probably does not come from the process of suffixation but seems to arise at a later point or for different reasons. The collective suffixes described below have at least two characteristics in common: i. they apply to bases which do not have a collective meaning themselves, nor are their derivatives marked for plural to denote plurality of entities; ii. the suffixes in question seem to have the collective meaning (‘a group of/a collection of X’) as their primary meaning, so that the collective meaning of the derivative arises from the application of the particular suffixes. These facts may be assumed as relevant factors for the acquisition of pejorative meanings in the collective suffixes. It should be noted, also, that pejoration is inherently involved in the semantics of the derivatives, at least as far as productive or nonce formation is concerned.

6.4.3.1 The suffixes -aría, -arʝó Petrounias (1991) has conducted a detailed analysis of the suffix -aría, making reference to its origin, semantics, and distribution in SMG in comparison to its Italian cognate, -eria. The etymology of both SMG -aría and -arʝó directly or indirectly relates to the Latin suffix -arius. The same etymon seems to appear in similar suffixes across various European languages, for example, German -(er)ei (see Dammel & Quindt 2016), French -erie, and English -ery. 248

See, e.g., Aliquot-Suengas (1996) for French -aille; Suzuki (1998) for expressions meaning ‘the likes of’ in Japanese; Efthymiou (1999) for the collective meaning in SMG; Koo and Rhee (2016) for pejorative suffixes in Korean; Finkbeiner (2016) for the pejorative force of ‘blah blah blah’ expression in German; Dammel and Quindt (2016) for German suffixes that mean ‘total of instances of V-ing’ whereby repetitiveness implies speakers’ dismissive attitude; Wiese and Polat (2016) for m-reduplication; in urban German and Kallergi and Konstantinidou (2018) for m-reduplication and other echo-constructions in SMG.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

205

In Italian, the primary role of the suffix -eria is to denote: i. a workshop or place where things are being made or sold (as in, e.g., gelateria ‘ice-cream factory/shop’); Petrounias (1991: 58) also mentions that it denotes ii. an action (a rare use) iii. a group of people (pejoratively) or iv. a human property (again, pejoratively).249 Note, also, that due to the productive function under i., in Italian the suffix may appear in oral speech as well as in formal and written language. In SMG, instances of the primary meaning (‘shop/place’) are only lexicalized: for example, pitsaría (‘pizzeria’), biraría (‘beer bar’) can be found in dictionaries.250 Also, there are a few lexicalized cases denoting a group of objects (e.g., trapezaría ‘dining table set’, dzamaría ‘set of glass windowpanes’), where the overall meaning is neutral. Therefore, such cases may appear in formal registers and may be encountered in written language. In addition, -aría appears in lexicalizations, mainly in informal speech, which directly characterize individual persons (kocetaría ‘a posh person’, letsaría ‘a shabby/dirty person’) or individual actions/states as in ksaplotaría ‘lying lazily’, ksipolitaría ‘being barefooted’ (Petrounias 1991: 59), arlumbaría ‘(talking) nonsense’ (ILNE, entry ἀρλουμπαρία). On the other hand, in productive word-formation (viz. typically, in oral, informal speech, which is often absent from dictionaries) the suffix has a collective meaning with pejorative connotations. This occurs mostly when the suffix attaches to colloquial bases marked for singular, which often have negative or derogatory meanings themselves. All of the following examples (taken from Petrounias 1991 and ILNE, entry -αρία) are negative characterizations for a group of people: (321)

kurel-aría (< kurelís ‘rag /shabby, disreputable person’)

(322)

tsoɣlan-aría (< tsoɣláni ‘bastard’)

249

250

Petrounias (1991) does not give Italian examples for these uses, but some of them can be exemplified by instances like fannulloneria ‘being idle’, massoneria ‘masonry, group of Masons’, zozzeria ‘dirty thing/dirty person’ (we thank Michele Magliocchetti for these examples). With respect to names of shops, the borrowed suffix -erí (from French -erie) is sometimes used to create terms such as uzerí ‘tavern offering ouzo and related food’ and suvlacerí ‘souvlaki shop’: especially in the latter example, the formality of the French suffix, which would be expected for high class entities, is combined with the folk or colloquial character of the word suvláci, creating ludic effects. See also Christopoulou (2016: 558) for the participation of this and other borrowed suffixes from French in the formation of marginal vocabulary items.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

206

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(323)

alit-aría (< alítis ‘tramp’)

(324)

bats-aría (< bátsos ‘policeman[+PEJ]’)

(325)

cirats-aría (< cirátsa ‘lady[+PEJ], rumour-monger’)

(326)

basklas-aría (< basklás ‘low-class[+PEJ]’)

(327)

snob-aría (< znob ‘snob’)

However, there are instances where the suffix selects semantically neutral bases, but still has a pejorative collective function. As in Examples (321) to (327), the speaker refers to a category as a whole or a sub-group of the category referred to by the base-word, expressing his negative feelings or attitude towards it. A relatively good English equivalent of the descriptive meaning that the suffix offers here would be ‘a bunch of Xs, an X lot’: (328)

fitit-aría (< fititís ‘university student’) ‘a bunch of students [+PEJ]’

(329)

maθit-aría (< maθitís ‘pupil’) ‘a bunch of pupils [+PEJ]’

(330)

vuleft-aría (< vuleftís ‘MP’) ‘a bunch of MPs [+PEJ]’

(331)

kaθiʝit-aría (< kaθiʝitís ‘professor’) ‘a bunch of professors [+PEJ]’

Some attested instances consist of bases with already collective meanings. The suffix in (332) seems to negatively intensify the collective meaning of the word: (332)

ðen ánikse poté to stóma tis óli aftí i epitrop-aría251 not open.PFV.PST.3SG never the mouth its all this the committee-aría ‘All this (damn) committee has never opened its mouth’

In (333), however, the pejoratively amplified collective meaning of the acronym PASOK (referring to a political party) is extended to refer to a specific behaviour. The connection is perhaps motivated if behaviour is conceived as a collection of individual actions. Alternatively, the individuating function may arise through a metonymical use of the whole (‘a group of people who support PASOK’) for the part (‘the specific mentality or behaviour of the people who support PASOK’): (333)

i pasokaría íne i nootropía pu epikratí stus enapomínandes tu civernitikú sçímatos252 ‘The PASOK-aría is the mentality prevailing to the last ones left in the governmental scheme’

Moving on to the suffix -arʝó (written -αριό or less often -αρειό),253 it should be noted that it already appears in its unreduced form -arío(n) (written -αρεῖο(ν) 251 252 253

At SKAI radio station, Aris Portosalte’s talking show, 19/1/2021. At https://bit.ly/3EHP842 (accessed 17 January 2022). In both the ILNE and Petrounias (1991), the written form of -arʝó is presented orthographically as -αρειό,, instead of the now common -αριό form, similarly to its originating Ancient Greek -

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

207

or -αρίο(ν)) in Medieval Greek. According to the ILNE (entry -αρειό) and Petrounias (1991: 66), it originates in an earlier Greek suffix denoting professions (-áris/-άρης as in maceláris ‘butcher’) in combination with an also early Greek locative suffix (-ío(n)/-εῖο(ν)). Thus, in at least Medieval Greek derivatives with -arío(n) > -arʝó(n) used to refer to a place (e.g., kabanarío(n), kabanarʝó(n) ‘bell tower’) or a professional space, viz. ‘a place where many people/objects of the same profession/kind are gathered’, as in ascitarío ‘a place where monks are gathered’. Today, -arʝó retains this local meaning in rare, lexicalized instances such as plistarʝó ‘washing room’ and ceramiðarʝó ‘ceramics workshop’.254 According to ILNE, in many Modern Greek dialects -arʝó has also a function of creating abstract nouns from nouns or adjectives, such as ksaplotarʝó ‘lying down, being laid back’ (< ksaplotós ‘streched out, relaxed’), polimilitarʝó ‘prattle’ (< polimilitís ‘prater’). Its contemporary functions, however, are primarily pejorative, applying to two kinds of bases. One is bases that denote ethnic groups. The most frequently attested cases point to ethnic groups of people (or specific properties thereof ) that the speaker clearly does not identify with. In this case, reference to an actually large group of people is not necessarily the case, as the suffix projects a negative feeling of the speaker that the referents are many (or perhaps, too many). (334)

ʝiftarʝó/tsiganarʝó (< ʝíftos/tsigános ‘Gypsy/Romany’) ‘Gypsies/a lot of Gypsies together [+PEJ]’

(335)

alvanarʝó (< alvanós ‘Albanian’) ‘Albanians/a lot of Albanians [+PEJ]’

(336)

ʝermanarʝó (< ʝermanós ‘German’) ‘Germans/a lot of Germans [+PEJ]’

(337)

elinarʝó (< élinas ‘Greek’) ‘Greeks/a lot of Greeks [+PEJ]’

Apart from this collective meaning, derivatives involving -arʝó may also have individual reference, similarly to -iá (§6.4.1.2). That is, they often refer to actual persons, emphasizing their ethnic identity in a negative way. Again, the context determines whether the interpretation is collective (typically, when the derivative is modified by the quantifier ólo ‘all’, Example 338) or individuating, in which case there is no descriptive meaning added, only a pejorative meaning component contributed by -arʝó, as in (339). (338)

254 255

évale ólo to alvanarʝó sti xóra255 put.PFV.PST.3SG all the Albanian.COLL[+PEJ] in.the country ‘He put all the Albanians[+PEJ] into the country’

εῖον. For an etymological analysis that explains how -arʝó roots back in Koine Greek see the ILNE and for its connection to Lat. -arius, see Petrounias (1991: 66). See Efthymiou (1999: 338, fn 4). At www.phorum.com.gr/viewtopic.php?t=691&start=45 (accessed 17 January 2022).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

208

Kallergi, Katsouda, Konstantinidou, Tsangalidis

(339)

tsíbisan stin pátra to alvanarʝó256 bust.PFV.PST.3PL in.the Patras the Albanian-arʝó ‘They busted the Albanian[+PEJ] in Patras’

The second type of bases selected by -arʝó in productive derivation is lexemes of general vocabulary (frequently those denoting profession or occupation), but a tendency towards lexemes with negative meanings is encountered here again (Examples 344–348). (340)

fititarʝó (< fititís ‘university student’) ‘a student lot’

(341)

ðaskalarʝó (< ðáskalos ‘teacher’) ‘a teacher lot’

(342)

kaθiʝitarʝó (< kaθiʝitís ‘professor’) ‘a professor lot’

(343)

papaðarʝó (< papáðes ‘priest.PL.’) ‘a priest lot’

(344)

aplitarʝó (< áplita ‘dirty laundry’) ‘an unwashed lot’

(345)

fasistarʝó (< fasístas ‘fascist’) ‘a fascist lot’

(346)

citsarʝó (< cits ‘kitsch’) ‘a kitsch lot’

(347)

katinarʝó (< Katína ‘rumour-monger’) ‘a rumour-monger lot’

(348)

putanarʝó (< putána ‘whore’) ‘a whore lot’

In the latter group of examples, where the bases already carry a negative meaning, the function of the collectivizing or individuating -arʝó is amplifying, which is supported by the fact that such derivatives are often encountered in prefixation with the intensifier kara- (e.g., karacitsarʝó ‘bunch of extremely kitsch people/objects’, karaputanarʝó ‘bunch of whores/extremely slutty person or behaviour’). Prefixation with kara- is observed by Petrounias (1991: 65) for derivatives with -aría as well (e.g., karacitsaría ‘bunch of extremely kitsch people/objects’), which indicates that, at least regarding this particular function, -aría and -arʝó are synonymous.

6.4.3.2 The suffixes -lói and -máni These suffixes, both marked for neuter, singular, effect a kind of ‘lexical plural’ on nominal bases; that is, although their derivatives can be marked for plural (inflectional -a for neuter), their (pejorative) collective meaning already arises from suffixation:257 256

257

At www.makeleio.gr, 21 May 2016 (accessed 17 January 2022). Note that in this individuating function, the derivative may appear in the plural, e.g., ta alvanarʝá (the.PL.NEUT Albanianarʝá.PL.NEUT) ‘the Albanians[+PEJ]’), ta alitarʝá (the.PL.NEUT tramp-arʝá.PL.NEUT) ‘the tramps[+PEJ]’. The term ‘suffix’ is used conventionally for these two bound morphemes, since both -lói and -máni attach to stems, with which they are linked via an -o- that resembles the compound marker (possibly a trace of compounding at earlier stages in the life of -lói and -máni as free morphemes/lexemes, such as the SMG bound second constituent -lóʝio, as in pelatolóʝio

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek (349)

scilolói (< scílos ‘dog’) ‘a pack of dogs/all sorts of dogs’

(350)

siɟenolói (< siɟenís ‘relative, kin’) ‘a pack of relatives’

(351)

papaðolói (< papáðes ‘priest.PL’) ‘a pack of priests’

(352)

ʝiftolói (< ʝíftos ‘Gypsy’) ‘a pack of Gypsies’

(353)

xartománi (< xartí ‘paper’) ‘a pack of papers’

(354)

ruxománi (< rúxo ‘item of clothing.SG’) ‘a pack of clothes’

(355)

ʝinekománi (< ʝinéka ‘woman’) ‘a pack of women’

(356)

ɣatománi (< ɣáta ‘cat’) ‘a pack of cats’

(357)

peðománi (< peðí ‘child’) ‘a pack of children’

(358)

turkománi (< túrkos ‘Turk’) ‘a pack of Turkish people’

(359)

siɟenolóʝa ‘pack-of-relatives.PL’ ‘packs of relatives’

(360)

xartomáɲa ‘pack-of-papers.PL’ ‘packs of papers’

(361)

scilolóʝa ‘pack-of-dogs.PL’ ‘group of loud and rowdy people’

209

In the latter examples, where the collective suffixes are pluralized, the plural meaning is either pleonastic (i.e., it does not correspond to an actually larger number of groups) but is more emphatic or it acquires additional meanings, as Example (361), which may refer to people, instead of dogs. The additional meaning is perhaps reached by means of the negative connotations of the word scilolói ‘pack of dogs’, such as strayness and lack of control (of the group or within the group). Similarly to other suffixes with collective meaning (see -iá, -urʝá, and -aría, -arʝó above), -lói and -máni frequently appear in a phrase with the quantifier ólo ‘all.NEUT.SG’ (specifically in ólo to ‘all the X-COLL’) and present the entities of the group as indistinctive, hence not special or important (an implication that, as we have seen above, motivates the pejorative). Nevertheless, -lói and -máni seem to be specialized in the meaning ‘too many Xs aggregated in the same place’ (see Efthymiou 1999). The pejorative meaning arises from a standard connotation about the speaker’s view on the number of entities being referred to: they are annoyingly a lot and they are present at the same time, in the same place. This seems to be the feature that ‘collection of customers/clientele’). The LKN (entry -λόι) connects -lói to Middle Greek -lóʝin (< Koine Greek -lógion; e.g., Middle Greek arxondo-lóʝin ‘collect. persons of noble birth or of distinguished social rank, nobility, gentry’) and -máni to Lat. manus, which apart from ‘hand’, also meant ‘crowd’. Since the meaning of these early lexemes is now completely opaque, but their morphological status resembles that of second compound constituents, -lói and -máni can be considered suffixoid items (see Efthymiou 1999). However, the exact status of the -o- and the limits of the bound morphemes would require diachronic research and cannot be further investigated here (see also Kiranoudis (2001) for the phonological adaptation of borrowed suffixes in the process of grammaticalization).

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

210

Kallergi, Katsouda, Konstantinidou, Tsangalidis

distinguishes collectives from inflectional plural. In addition, it may be a slightly distinctive feature of -lói and -máni with respect to their peers. For instance, fititaría (328) and papaðarʝó (343) have been glossed ‘a bunch of Xs’, where X is presented as unimportant or bad, viz. contempt is expressed. The meaning of corresponding terms such as fititolói and papaðománi is best described as ‘too many Xs together’, and in this case, annoyance seems to be the most central pejorative effect. Moreover, -lói and -máni exhibit an individuating meaning (i.e., reference to a person or property) to a much lesser degree, if at all, than other collectives. Finally, with respect to their special pejorative collective function, -lói and -máni seem to be highly interchangeable, as is evident from the distribution in nonce formation in the attested examples below.258 (362)

sirizolói / sirizománi ‘a bunch/pack of SYRIZA voters or members’

(363)

pasokolói / pasokománi ‘a bunch/pack of PASOK voters or members’

(364)

komunistolói / komunistománi ‘a bunch/pack of communists’

(365)

elinolói / elinománi ‘a bunch/pack of Greeks’

(366)

cinezolói / cinezománi ‘a bunch/pack of Chinese people’

6.4.3.3 Summary of collectives As with other groups of suffixes, collective suffixes have been presented in categories that either involve etymologically related items (-aría and -arʝó) or functionally related ones (-lói and -máni). All collectives, however, create denominal nouns acting on singular bases as a kind of lexical plural (specifically contributing the general meaning ‘a collection/group of Xs’). Apart from this denotation or descriptive meaning, we have seen that collectives are inherently pejorative, with the first group mainly expressing contempt (in the sense expressed by English expressions such as ‘a bunch/pack of Xs, an X lot’), whereas with the second group annoyance seems more relevant (out of the meaning ‘too many Xs together’). The role of collectives may rarely be redundant in semantic terms, in the sense that only one instance (Example 332, epitrop-aría committee-aría.PEJ) exhibits a base with already collective meaning. Here, although epitropí ‘committee’ is not a pejorative term itself, the function of the collective suffix -aría is amplifying of the collective meaning, hence more emphatic and pejorative. In other cases, due to the pejorative meanings developed in combination with assumingly negative bases, collectives may have individuating meaning, that is, their

258

However, it is possible that -máni is slightly more productive, since for some instances, such as ðimosioγrafománi ‘pack of journalists’ and anarxománi ‘pack of anarchists’, a brief Google search did not return equivalents with -lói.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

211

derivatives may refer to an individual (putan-arʝó whore-arʝó ‘whore[+PEJ]’, alvan-arʝó Albanian-arʝó ‘Albanian[+PEJ]’). Again, the role of the collective suffix is amplifying or emphatic of the negative meaning or connotation of the base and, in this sense, purely expressive. We have also observed that, although -arʝó seems to have inherited its pejorative role from earlier Greek (among other meanings), productive word formation with it and other collectives is restricted to the pejorative meaning. 6.5

General conclusions

We conclude our study with the main findings from §6.3 and §6.4, discussing several formal and functional parameters of negative expressivity in SMG compounding and suffixation. In §6.3, we have presented a very frequent and productive morphological mechanism at the disposal of SMG speakers for expressing negative feelings or attitudes towards a specific target without literally naming those feelings or attitudes. It is the mechanism of deriving negative expressive elements (pejoratives), through one-word compounding. As shown by the preceding analysis, the speaker creates a compound structure of the following type: [Constituent1]–[Compound Marker]–[Constituent2]

For example: [stem skat(ó) ‘shit’]–[o]–[noun ʝatrós ‘doctor/physician’] >skatoʝatrós ‘shitty/fucking doctor/physician’

This places at the second member position the linguistic element denoting the intended target against whom he experiences at the time of the utterance a negative feeling or attitude (e.g., deprecation, contempt, annoyance, anger, or disgust)259 and at the first member position an expressively charged element that contributes the pejorative aspect of the compound. Due to the expressivity of the first constituent, the derived compound obtains expressivity and is thereby capable of indicating the speaker’s negative feeling or attitude towards the entity denoted by the base, as well as contributing descriptive content. In semiotic terms, the formed expressive compound functions as an indexical sign that points towards the speaker’s pejorative attitude. The semantic groups from which the derogatory first constituents analysed are drawn are those denoting ethnic, racial, or cultural group, smelliness or dirtiness, or bodily effluvia (see Meibauer 2013: 24–25, 30 for similar findings about German pejorative non-heads; see also, Norri’s 2000 insight that

259

‘Or at least intends to create such an impression’ in Potts’s (2019: 618) formulation.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

212

Kallergi, Katsouda, Konstantinidou, Tsangalidis

negativity consistently accompanies terms for nationality in at least sixty per cent of the entries). As is evident from dictionary labelling systems, negatively marked terms are systematically derived from these and some other semantic domains, such as those denoting animals, tabooed bodily organs, acts of sex, diseases, etc. (see for Greek Efthymiou et al. 2013. Kehayia 1997, see also Allan & Burridge 2006). As already observed by scholars, such as Meibauer (2013), there are morphological expressive items (first constituents) that seemingly convey a higher degree of expressivity than others. Our intuitions as SMG native speakers, are in conformity with these insights. In §6.4, we have described the repertoire of pejorative functions of many SMG suffixes, dividing them into large groups according to their basic, descriptive meaning. Thus, we have discussed cases of suffixes that very roughly denote a property of the referent relating to what the base encodes, suffixes that belong to the functional categories of augmentation and diminution, and finally, collective suffixes. A generalization that can be made with respect to their formal properties is that their striking majority forms nominal classes (nouns and, to a lesser degree, adjectives), with a strong tendency towards human referents, but also with reference to states and objects (e.g., -íla, -ilíci, -úra). This seems to support an observation made by Dammel and Quindt (2016: 49, fn 3 citing Bauer 1997), that, across languages, the morphological evaluation of ‘referents’ is more frequent than that of ‘actions’ (specifically, expressive mechanisms on verbs). Also, the target of evaluation is strongly nominal, in that most of the suffixes in total select nominal bases (of course, verbs as well as phrases have been also observed as bases for pejorative derivation). As summaries of individual sections have been provided at the end of each section, we will devote some more space here to make a few generalizations with respect to (pejorative) meaning and suffixation. Regardless of whether the pejorative function of the suffixes is inherited from earlier stages of Greek, we observe that the following general factors correlate with or motivate the development of pejorative meanings in pejorative suffixation. Pejoration is either an (intrinsic) meaning aspect of the base or an (intrinsic) meaning aspect of the suffix or both. Here we attempted to isolate the role of the suffix itself, keeping the variable of the base-word semantics stable, that is, by focusing on cases where the bases are semantically and/or expressively neutral. However, two observations made throughout the section must be emphasized; first, negative or already pejorative semantics in the base-word seem to affect the interpretation of the whole derivation; second, pejorative suffixes (inherently or not) tend to attach to bases of colloquial, [–learned], taboo, slang, or otherwise marginal vocabulary, which are often used for expressive and pejorative purposes anyway (see Christopoulou 2016). In that

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

213

case, pejorative suffixes assume the role of an amplifier, that is, they are used as indices of a higher degree for the overall expressive effect or indicate emphasis. Regarding the semantics of the suffixes, several other factors seem to be responsible for the development of pejorative functions. An important mechanism involves metaphoric extensions. In various cases of suffixation, concrete, negative meanings tend to be generalized or transferred to abstract domains (see, e.g., -íla and -arʝó). In the domains of diminution or augmentation, metaphor seems to play a role in the extensions from SMALL to UNIMPORTANT, leading to the expression of contempt or from BIG to NASTY, in turn leading to the expression of annoyance and other negative attitudes (see, e.g., Jurafsky 1996, Körtvélyessy’s 2015, Prieto 2015, Mutz 2015). Also, in terms of semantics, we have seen that many of the nominalizing suffixes we have discussed fall into broad semantic categories that refer to the human body and its functions, as well as the human senses (typically, smell, as in the case of -íla). Further, a large number of pejorative suffixes mean ‘a person who has X or who tends to do X/extremely likes X/-ing’ (-ás/-ú, -ácas, -adzís, -iáris), where the characteristics attributed often point to human bodily functions or imperfections. The ‘strong tendency/propensity/extreme liking’ meaning as well as the meaning of professional (frequent) activity encoded by many pejorative suffixes and the descriptive meaning of augmentatives and diminutives have implications such as: i. excessive or inadequate amount/size (which already involves a negative connotation); and/or ii. a repetitive action/behaviour that easily motivates the expression of feelings such as annoyance and induces negative, dismissive attitudes (see, e.g., Dammel & Quindt 2016). Similar dismissive dispositions that undergo conventionalization seem to arise from collective meanings. We have discussed that plurality and lack of specificity of individual identities, which are combined in the case of collectives, seem to motivate the rise of pejoratives on the basis of a negative assumption about the group (viz. lack of specific spatial limits) and about the individual entities included in the group, viz. lack of identity, hence unimportance and unworthiness. The plural, on the other hand, whether inflectionally marked or lexically involved in the suffixes and the morphological processes in question plays an important role for the development or reinforcement of negative expressivity. Of course, diachronic change and grammaticalization has not been the objective of this descriptive study. Nevertheless, we cannot ignore the role of pragmatics, in most of the processes examined, which may be at work

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

214

Kallergi, Katsouda, Konstantinidou, Tsangalidis

synchronically as well. As Dammel and Quindt (2016: 69) argue, ‘expressive affixes hardly grow completely independent of context and pragmatic inferencing’, since their descriptive meaning is still accessible for pragmatic exploitation towards pejorative effects of various kinds. Most scholars on expressive or evaluative morphology mention the role of conventionalization of implicatures in the establishment of negative meanings, at the utterance level and at the word level later. A rather simple example would be to suppose the work of inferencing from the notion BIG/MANY to TOO BIG/TOO MANY,260 or from REPETITION to ANNOYANCE, or, even, from SIMILAR to NOT ORIGINAL and INFERIOR (encountered in cases such as -ískos, -ário and -ístikos). Another possibly motivating pragmatic force, from a synchronic point of view, is irony. Although irony has not been dealt with systematically in this paper, it has been encountered at least as the incongruence between contradictory registers and styles, specifically the SMG-idiosyncratic [–learned] – [+learned] combination. For example, in the case of -ário and -ískos, the use of [+learned] or [+high register] elements in informal, familiar, colloquial contexts seems to enhance the pejorative effects of inherited pejorative items. Finally, an SMG-specific factor, relating to the lexical nature of pejorative suffixes, is their status as loans. It has been noted that foreign loans in general are inserted into the language as [–learned], [+colloquial] elements, even though they may have been items of a default register type or style in the originating language, which may be French, English, Italian, Turkish, a Balkan language or others (see, e.g, Anastasiadis-Symeonidis 2010). And we have seen that [–learned]/[+colloquial] style and informal/familiar speech is the general framework for the expression of pejoration and other kinds of expressivity or evaluation. The most characteristic instantiation of this phenomenon (the acquisition of a marked profile for loans) is Turkish loans (or so called turkisms, see Kazazis 1972, Tzitzilis 1997): at least two of the pejorative suffixes we have examined can be neutral in Turkish (-ci, -çi, and -(i)lik), whereas in SMG they are specialized in denoting profession or occupation and human activity or property in negative terms (viz. inclinations/states that invite negative attitudes on the part of the speaker). This factor is in fact independent of the (original) semantics of the suffix but relates to the usage of an item in everyday speech and the wider socio-pragmatic conditions of this usage.261 Thus, it seems that idiosyncratic, lexical features of the suffixes (such as their origin) make an independent contribution to their status as pejoratives.

260 261

Or to EXAGGERATION, as suggested in Mutz (2015). Kiranoudis (2001: 548), for instance, mentions a reaction against turkisms during the establishment of Katharevousa, the high variety of Greek in the period of diglossia.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

6.6

215

Suggestions for further study

As we mentioned earlier, this study is far from exhaustive. Negative expressivity could be studied with respect to other morphological mechanisms, but also at other levels of grammar, such as syntax or phonology. Also, there are several other interesting aspects of pejoration and expressivity in general that would certainly be worth investigating. First, with respect to the connection between the two different mechanisms examined here (compounding and suffixation), it would be interesting to examine the possibility of combining the two mechanisms in order to maximize the pejorative effects intended. We have already seen that such a possibility exists (as in, e.g., vlaxo-vlax-ára, kara-cits-arʝó; see §6.3.1 and §6.4.3.1, respectively), but we have not studied the exact effects, possible grammatical restrictions, and other details. Combinability may also take place within a single of the mechanisms in question, for example, between different suffixes, and this issue would require further investigation as well. Second, more detailed study could be devoted to the grammatical aspects of the mechanisms in question. For instance, it would be useful to know the full set of restrictions on the use of one suffix over the other, which may clarify when some suffixes are interchangeable (if at all), and when they are not. With respect to grammatical categories, also, it would be interesting to explore verbal pejorative suffixes and verbal expressive compounds. Grammatical gender, finally, is a parameter that deserves independent research (more specifically, the relation of each of the SMG gender types with expressivity). Third, further research could be conducted on the relation between expressivity and factors such as metaphor, irony and type of register. The former two phenomena, being very broad themselves, would require a separate discussion; moreover, the different register labels (e.g., [informal], [colloquial], [slang]) as well as the parameter of humour (through, e.g., the pragmatic feature [ludic]) could be further analysed so that their role in the effects described can be elucidated. One more specific (lexicographic) dimension of this research might concern the connection of register labelling, as used in dictionaries, and elements of pejorative meaning; admittedly, what might sometimes be presented as a ‘ludic use’ (παικτική χρήση) of a dictionary entry in fact describes its status as a pejorative. In addition, we have seen that expressive items are [+informal], and often [+colloquial] items. But it is to be further examined whether all [+colloquial] items are expressive in nature. Last, systematic diachronic research would be necessary for a fully fledged study of the phenomenon. Apart from the rise of each pejorative item in the course of time (which might include the strengthening of already pejorative words or a move from meliorative to pejorative and vice versa), it would be interesting to study the semantic change of pejorative items to other types of

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

216

Kallergi, Katsouda, Konstantinidou, Tsangalidis

lexical items, once the former are weakened. Besides, the nature or extension of non-expressive meaning as well as of melioration is as interesting as that of negative expressivity, that is, pejoration. Abbreviations ADJ AUG CM COLL DIM EXPL FEM GEN IPFV MASC MEL MET NEUT N NP PRTCL PRTCPL PST PEJ PFV PL PR SG SMG SUFF

Adjective Augmentative Compound Marker Collective Diminutive Expletive Feminine Genitive Imperfective Masculine Meliorative Metaphorical Neuter Noun Noun Phrase Particle Participle Past Pejorative Perfective Plural Present Singular Standard Modern Greek Suffix

Primary Sources Detsikas, K. (2005). Karaiskakis = Δέτσικας, Κ. Καραϊσκάκης. Ο στρατάρχης, τόμος Β΄ [Karaiskakis. The Field Marshal, Vol. 2]. Athens: Iliotropio. Diigisis Pediofrastos (1972). (14th century) = V. Tsiouni (ed.) Παιδιόφραστος διήγησις τῶν ζώων τῶν τετραπόδων [Playful story about quadrupeds]. Critical edition. Μiscellanea Byzantina Monacensia 15. Munich: Institut für Byzantinistik und Neugriechische Philologie der Universität München. Douka, M. (2007). Ploti poli = Δούκα, Μ. Η πλωτή πόλη [The floating city]. Athens: Patakis. Fakinou, E. (1982). Astradeni = Φακίνου, Ε. Αστραδενή [Astradeni]. Athens: Kedros. Fleischer & Stergellis (1984) Maidonis = H. Fleischer & A. Στεργέλλης (eds.) Ημερολόγιο Φαίδωνα Μαηδώνη (24.6–10.9.1944) [Phaedon Maidonis diary]. Μνήμων, 9, 33–156. DOI: https://doi.org/10.12681/mnimon.249 Karistiani, I. (2006). Suel = Καρυστιάνη, I. Σουέλ [Suel]. Athens: Kastaniotis.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

217

Katzourbos (1964). (17th century) = Λ. Πολίτης (ed.) Γεωργίου Χορτάτση, Κατζοῦρμπος, Κωμωδία [Georgios Chortatzis’ Katzourbos, a comedy]. Critical edition. Heraklion: Ἑταιρία Κρητικῶν Ἱστορικῶν Μελετῶν [Society of Cretan Historical Studies]. Kazantzakis, N. & Kakridis, I. Th., (eds.) (1965). Odysseia = Ν. Καζαντζάκης & Ι. Θ. Κακριδῆς (eds.) Ὁμήρου Ὀδύσσεια, μετάφραση [Homer’s Odyssey, translation]. Athens: Ι. Θ. ΚΑΚΡΙΔΗ – ΕΛΕΝΗΣ Ι. ΚΑΚΡΙΔΗ [I. Th. Kakridis and Helen I. Kakridis Publications]. Krasopateras (1988). (16th–18th century edition of a 15th century text) = H. Eideneier Krasopateras, kritische Ausgabe der Versionen des 16.–18. Jahrhunderts. [Krasopateras: Critical edition of the versions of the 16th–18th centuries]. Cologne: Romiosini. Loukatos, D. S. (1963). Laografika = Δ. Σ. Λουκᾶτος, Σύγχρονα λαογραφικὰ [Contemporary folklore issues]. Athens: Private edition. Papachristodoulou, P. (1936–1937). Saranta Ekklisies = Π. Παπαχριστοδούλου, Γλωσσάριο Σαράντα Ἐκκλησιῶν [Saranta Ekklisies glossary]. Ἀρχεῖον Θράκης 3, 214–25. Papataxiarchis, E. (1998). Tautotites = Ε. Παπαταξιάρχης, Tαυτότητες και φύλο στη σύγχρονη Ελλάδα: ανθρωπολογικές προσεγγίσεις [Identities and sex in modern Greece: Anthropological approaches]. Athens: Alexandreia. Poulologos (1987). (14th century) = Ι. Tsavari, Ὁ Πουλολόγος, κριτικὴ ἔκδοση μὲ εἰσαγωγή, σχόλια καὶ λεξιλόγιο [Poulologos, lit. ‘Bird Book’, critical edition with introduction, comments and glossary]. Athens: National Bank Educational Institution. Procopius of Nazianzus Narrative (2012). (early 19th century) = M. G. Varvounis & P. Tzoumerkas, Η ‘διήγησις’ Προκοπίου του Ναζιανζηνού για την πυρκαγιά και την ανοικοδόμηση του ναού της Αναστάσεως στα Ιεροσόλυμα (1808–1810): η μαρτυρία του χφ. 107 της Ιεράς Μητροπόλεως Σάμου [The narrative of Procopius Nazianzus about the burning and rebuilding of the Church of the Holy Sepulchre in Jerusalem (1808–1810): The testimony of manuscript 107 of the Holy Metropolis of Samos]. Thessaloniki: Athanasios Altinzis Publications. Sachlikis (1960). (14th century) = M. Vitti, Il poema parenetico di Sachlikis nella tradizione inedita del cod. Napoletano. Kρητικὰ Χρονικά, 14, 173–200. Sinaksarion Jinekon (1905). (16th century) = Συναξάριον τῶν εὐγενικῶν γυναικῶν καὶ τιμιωτάτων ἀρχοντισσῶν [Synaxary of noble women and honorable ladies]. K. Krumbacher, Ein vulgärgriechischer Weiberspiegel [A vulgar Greek women’s mirror]. Munich: Verlag der Akademie der Wissenschaften. Spanos (1977). (16th century edition of a 14th–15th century text) = H. Eideneier (ed.) Spanos: Eine byzantinische Satire in der Form einer Parodie. Einleitung, Kritischer Text, Kommentar und Glossar [Spanos: A Byzantine satire in the form of a parody. Introduction, critical text, commentary and glossary]. Berlin and New York: De Gruyter. .

References Aliquot-Suengas, S. (1996). Référence collective / sens collectif: la notion de collectif à travers les noms suffixés du lexique français. Doctoral thesis. University of Lille III.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

218

Kallergi, Katsouda, Konstantinidou, Tsangalidis

Allan, K. & Burridge, K. (2006). Forbidden Words: Taboo and the censoring of language. Cambridge: Cambridge University Press. Amaral, P. (2018). Expressive meaning. In F. Liedtke & A. Tuche (eds.) Handbuch Pragmatik [Handbook of Pragmatics]. Stuttgart: Springer/Metzler, 325–33. DOI: 10.1007/978-3-476-04624-6 Amaral, P., Roberts, Cr. & Allyn Smith, E. (2007). Review of The Logic of Conventional Implicatures by Chris Potts. Linguistics and Philosophy, 30(6), 707–49. DOI: 10.1007/s10988-008-9025-2 Anastasiadis-Symeonidis, A. (1983). La composition en grec moderne d’un point de vue diachronique. Lalies, 2, 77–90. (1986). Η νεολογία στην Κοινή Νέα Ελληνική [Neology in Standard Modern Greek], Journal of the Faculty of Philosophy of the Aristotle University of Thessaloniki, Supplement 65. (1992). Η νεοελληνική παραγωγή κατά το μοντέλο της D. Corbin [Derivation in Modern Greek according to D. Corbin’s model]. Μελέτες για την Ελληνική Γλώσσα [Studies in Greek Linguistics], 13, 505–26. (1996). Η νεοελληνική σύνθεση [Modern Greek compounding]. In G. Catsimali & F. Kavoukopoulos, (eds.) Ζητήματα νεοελληνικής γλώσσας. Διδακτική προσέγγιση [Aspects of Modern Greek language: A teaching approach]. Rethymno: University of Crete, 97–120. (1997). Το επίθημα -άδικο στη νεοελληνική [The suffix -áðiko in Modern Greek]. Μελέτες για την Ελληνική Γλώσσα [Studies in Greek Linguistics], 17, 157–71. (2000). Το επίθημα -(ι)άρ(ης) στη νεοελληνική [The suffix -(i)áris in Modern Greek]. Μελέτες για την Ελληνική Γλώσσα [Studies in Greek Linguistics], 20, 65–74. (2010). Pourquoi une langue emprunte-t-elle des suffixes? L’exemple du grec et du latin. META 55(1). Mélanges en hommage à André Clas, 147–57. Anastasiadis-Symeonidis, A. & Fliatouras, A. (2004). Η διάκριση λόγιο και λαϊκό στην ελληνική γλώσσα: Ορισμός και ταξινόμηση [The learned vs. colloquial distinction in Modern Greek: Definition and classification]. In Proccedings of the 6th International Conference on Greek Linguistics (ICGL6). Rethymno: Department of Philology of the University of Crete. (2018). Από το λόγιο επίπεδο της Νέας Ελληνικής στην Αρχαία Ελληνική [From the learned register of Modern Greek to Ancient Greek]. Μελέτες για την Ελληνική Γλώσσα [Studies in Greek Linguistics], 38, 37–50. (2019). Το λόγιο επίπεδο της σύγχρονης Νέας Ελληνικής: Συγχρονικές και διαχρονικές τάσεις [The learned register in contemporary Modern Greek: Synchronic and diachronic tendencies]. In Α. Fliatiuras & Α. Anastasiadis-Symeonidis (eds.) Το λόγιο επίπεδο στη σύγχρονη Νέα Ελληνική: Θεωρία, ιστορία, εφαρμογή [The learned register in contemporary Modern Greek: Theory, history, applications]. Athens: Patakis, 15–55. Asimopoulos, P. (2016). Μορφολογικές και σημασιολογικές παράμετροι του φρασεολογικού πλούτου της στρατιωτικής κοινωνιολέκτου [Morphological and semantic parameters of the phraseological richness of the military sociolect]. Μελέτες για την Ελληνική Γλώσσα [Studies in Greek Linguistics], 36, 467–74. Babiniotis, G. (1969). Ὁ διὰ συνθέσεως ὑποκορισμὸς εἰς τὴν ἑλληνικὴν [Hypocorism through composition in Greek]. Athens: National and Capodistrian University of Athens.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

219

Bauer, L. (1997). Evaluative morphology: In search of universals. Studies in Language, 21(3), 533–75. Béis, S. (2008). Aroumain. In E. Adamou (ed.) Les noms des langues ΙΙ: Le patrimoine plurilingue de la Grèce. Louvain-la-Neuve: Peeters, 31–46. Bergeton, U. (2004). The independence of binding and intensification. PhD thesis, University of Southern California. Booij, G. (2013). Morphology in construction grammar. In T. Hoffmann & G. Trousdale (eds.) The Oxford Handbook of Construction Grammar. Oxford: Oxford University Press, 255–73. Bühler, K. (1965 [1934]). Sprachtheorie: Die Darstellungsfunktion der Sprache. Stuttgart: Fischer. Calvo, M. (2020). Evaluative derivatives and evaluative forms: Two distinct groups of affixed lexical units in evaluative morphology. In G. Dreijers, J. Sīlis, S. Sviķe & J. Veckrācis (eds.) Βridging Languages and Cultures II: Linguistics, translation studies and intercultural communication. Berlin: Frank and Timme, 83–112. Cameron, D. & Kulick, D. (2003). Language and Sexuality. Cambridge: Cambridge University Press. Cappelen, H. & Dever, J. (2019). Bad Language. Oxford: Oxford University Press. Chantraine, P. (1933). La formation des noms en grec ancien. Paris: Champion. Christopoulou, K. (2016). Μία λεξιλογική προσέγγιση στο περιθωριακό λεξιλόγιο της Νέας Ελληνικής [A lexicological approach of the marginal vocabulary of Modern Greek]. Doctoral thesis, University of Patras. Christopoulou, K., Xydopoulos, G. J. & Tsangalidis, A. (2017). Grammatical gender and offensiveness in Modern Greek slang vocabulary. In T. Georgakopoulos, T.-S. Pavlidou, M. Pechlivanos, A. Alexiadou, J. Androutsopoulos, A. Kalokairinos, S. Skopeteas & K. Stathi (eds.) Proceedings of the 12th International Conference on Greek Linguistics (ICGL12), Vol. 1. Berlin: Edition Romiosini, 291–305. Corbett, G. (1991). Gender. Cambridge: Cambridge University Press. DOI: 10.1017/ CBO9781139166119 Corbin, D. (1987). Morphologie dérivationnelle et structuration du lexique. Tübingen: Niemeyer. (1991), La formation des mots: Structure et interprétation. Lexique, 10, 7–30. Croom, A. M. (2011). Slurs. Language Sciences, 33(3), 343–58. DOI: 10.1016/j. langsci.2010.11.005 (2013). How to do things with slurs: Studies in the way of derogatory words. Language & Communication, 33(3), 177–204. https://doi.org/10.1016/j.langcom .2013.03.008 Cruse, D. A. (1986). Lexical Semantics. Cambridge: Cambridge University Press. (2004). Meaning in Language: An introduction to semantics and pragmatics. 2nd ed. Oxford: Oxford University Press. Daltas, P. (1985), Some patterns of variability in the use of diminutive and augmentative suffixes in spoken Modern Greek Koine (MGK). Glossologia, 4, 63–88. Dammel, A. & Quindt, O. (2016). How do evaluative derivational meanings arise? A bit of Geforsche and Forscherei. In R. Finkbeiner, J. Meibauer & H. Wiese (eds.) Pejoration. Amsterdam and Philadelphia: Benjamins, 41–74. DOI: 10.1075/ la.228.03dam

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

220

Kallergi, Katsouda, Konstantinidou, Tsangalidis

Dangas, N. (2012). Το επίθημα -(ι)λίκι στην κοινή νεοελληνική [The suffix -(i)líki in Standard Modern Greek]. In Z. Gavriilidou, A. Efthymiou, E. Thomadaki & P. Kambakis-Vougiouklis (eds.) Selected Papers of the 10th International Conference of Greek Linguistics (ICGL10). Komotini: Democritus University of Thrace, 1021–29. Dangas, N. & Manasis, V. (2018). Το επίθημα -άκια(ς) στη νέα ελληνική [The suffix ákias in Modern Greek]. Μελέτες για την Ελληνική Γλώσσα [Studies in Greek Linguistics], 38, 323–34. Diamantidou, Ε. (2017). Μια απóπειρα προσέγγισης του επιθήματος -ούρ(α) της νέας ελληνικής [An attempt to approach the suffix -oúra in Modern Greek]. Μελέτες για την Ελληνική Γλώσσα [Studies in Greek Linguistics], 37, 959–72. Dressler, W. U. & Merlini Barbaresi, L. (1994). Morphopragmatics: Diminutives and intensifiers in Italian, German, and other languages. Berlin: De Gruyter. Efthymiou, A. (1999). Περιληπτική σημασία/αναφορά στα επιθηματοποιημένα ουσιαστικά της Νέας Ελληνικής [Collective meaning/reference in the suffixed nouns of Modern Greek]. In Α. Mozer (ed.) Πρακτικά του 3ου Διεθνούς Συνεδρίου Ελληνικής Γλωσσολογίας [Proceedings of the 3rd International Conference on Greek Linguistics]. Athens: Greek Letters, 333–40. (2003). Προθήματα ή πρώτα συνθετικά που δηλώνουν επίταση στη ΝΕ [Prefixes and first constituents denoting intensity in Modern Greek]. Μελέτες για την Ελληνική Γλώσσα [Studies in Greek Language], 23, 519–28. (2013). On the interaction between semantics and phonetic iconicity in evaluative morphology. SKASE Journal of Theoretical Linguistics, 10, 152–66. (2015). Modern Greek diminutive and augmentative adjectives (in a cross-linguistic perspective). SKASE Journal of Theoretical Linguistics, 21(1), 57–71. (2019). Η μορφολογική ποικιλία στα αξιολογικά επίθετα και ρήματα της Νέας Ελληνικής [Morphological variation in evaluative adjectives and verbs in Modern Greek]. In Α. Archakis, Ν. Koutsoukos, G. Xydopoulos & D. Papazahariou (eds.) Γλωσσική Ποικιλία. Μελέτες Αφιερωμένες στην Αγγελική Ράλλη [Language variation. Studies dedicated to Angeliki Ralli]. Athens: Kappa Ekdotiki, 155–76. Efthymiou, A., Gavriilidou, Z., & Papadopoulou, E. (2013). Labeling of derogatory terms in Modern Greek dictionaries. In N. Lavidas, T. Alexiou & A.-M. Sougari (eds.) Major Trends in Theoretical and Applied Linguistics 2: Selected papers from the 20th ISTAL. London: Versita, 27–40. https://doi.org/10.2478/ 9788376560885.p12 Ferguson, C. A. (1959). Diglossia. Word, 15, 325–40. Finkbeiner, R. (2016). Bla, bla, bla in German: A pejorative construction. In R. Finkbeiner, J. Meibauer, & H. Wiese (eds.) Pejoration. Amsterdam and Philadelphia: Benjamins, 269–99. Finkbeiner, R., Meibauer, J. & Wiese, H. (eds.) (2016a), Pejoration. Amsterdam and Philadelphia: Benjamins. DOI: 10.1075/la.228.08ten Finkbeiner, R., Meibauer, J. & Wiese, H. (2016b). What is pejoration, and how can it be expressed in language? In R. Finkbeiner, J. Meibauer & H. Wiese (eds.) Pejoration. Amsterdam and Philadelphia: Benjamins, 1–20. DOI:10.1075/ la.228.08ten Fishman, J. (1967). Bilingualism with and without Diglossia; Diglossia with and without Bilingualism. Journal of Social Issues, 23, 29–38.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

221

Fortin, A. (2011). The Morphology and Semantics of Expressive Affixes. PhD thesis, University of Oxford. Frigerio, A. & Tenchini, M.-P. (2014). On the semantic status of connotation: The case of slurs. In P. Stalmaszczyk (ed.) Issues in Philosophy of Language and Linguistics. Lodz: University of Lodz Press, 57–75. Gavriilidou, Z. (2010). Aπό την στέρηση στην επίταση: η περίπτωση του α-/αν- [From deprivation to intensification: the case of a-/an-]. In Proceedings of the 8th International Conference on Greek Linguistics, Ioannina 30 August–2 September 2007. Ioannina: University of Ioannina, 674–83. (2013). Όψεις επίτασης στη Nέα Eλληνική [Aspects of intensification in Modern Greek]. Thessaloniki: Kyriakides. Geurts, B. (2007). Really fucking brilliant. Theoretical Linguistics, 33(2), 209–14. DOI:10.1515/TL.2007.013. Grandi, N. (2005). Sardinian evaluative morphology in typological perspective. In I. Putzu (ed.) Sardinian in Typological Perspective. Berlin: Akademie Verlag, 188–209. Grandi, N. (2011). Renewal and innovation in the emergence of Indo-European evaluative morphology. Lexis. E-Journal in English Lexicology, 6, 5–25. Grandi, N. & Körtvélyessy, L. (eds.) (2015a). Edinburgh Handbook of Evaluative Morphology. Edinburgh: Edinburgh University Press. Grandi, N. & Körtvélyessy, L. (2015b). Introduction: Why evaluative morphology? In N. Grandi & L. Körtvélyessy (eds.) Edinburgh Handbook of Evaluative Morphology. Edinburgh: Edinburgh University Press, 3–20. Grice, H. P. (1975 [1967]). Logic and conversation. In P. Cole & J. Morgan (eds.) Syntax and Semantics, Vol. 3. New York: Academic Press, 41–58. Gutzmann, D. (2011). Expressive modifiers and mixed expressives. In O. Bonami & P. C. Hofherr (eds.) Empirical Issues in Syntax and Semantics 8: Selected papers from the Colloque de Syntaxe et Sémantique à Paris CSSP. Bern: Peter Lang, 123–41. (2013). Expressive and beyond: Introduction to the varieties of use-conditional meaning. In D. l. Gutzmann & H.-M. Gärtner (eds.) Beyond Expressive: Explorations in use-conditional meaning. Leiden: Brill, 1–58. (2015). Use-Conditional Meaning: Studies in multidimensional semantics. Oxford: Oxford University Press. Gutzmann, D. & McCready, E. (2016). Quantification with pejoratives. In R. Finkbeiner, J. Meibauer, & H. Wiese (eds.) Pejoration. Amsterdam and Philadelphia: Benjamins, 75–102. DOI:10.1075/la.228.08ten Hatzidakis, G. N. (1905). Μεσαιωνικὰ καὶ Νέα Ἑλληνικὰ [Medieval and Modern Greek], vol. 1. Athens: Sakellariou. (1916). Ἀπάντησι [A response]. Ἐπιστημονικὴ Ἐπετηρὶς Ἐθνικοῦ καὶ Καποδιστριακοῦ Πανεπιστημίου [Scientific Yearbook of the National and Kapodistrian University],12, 28–47. Hess, L. (2019). Expressive meanings and expressive commitments. A case of meaning as use. In P. Stalmaszczyk (ed.) Philosophical Insights into Pragmatics. Berlin and Boston: De Gruyter, 193–224. Hom, C. (2008). The semantics of racial epithets. Journal of Philosophy, 105(8), 416–40. (2010). Pejoratives. Philosophy Compass, 5(2), 164–85.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

222

Kallergi, Katsouda, Konstantinidou, Tsangalidis

Hom, C. & May, R. (2013). Moral and semantic innocence. Analytic Philosophy, 54, 293–313. DOI:10.1111/phib.12020 Hornsby, J. (2001). Meaning and uselessness: How to think about derogatory words. Midwest Studies in Philosophy, 25(1), 128–41. Hübler, A. (1998). The Expressivity of Grammar: Grammatical devices expressing emotion across time. Berlin: De Gruyter. ILNE = Academy of Athens (1933–). Ἱστορικὸν λεξικὸν τῆς Νέας Ἑλληνικῆς τῆς τε κοινῶς ὁμιλουμένης καὶ τῶν διαλέκτων [Historical dictionary of Modern Greek, both of the standard language and the dialects]. 7 vols (Α–ðóʝis). Athens: Academy of Athens. Jay, T. (2000). Why we Curse. Philadelphia: Benjamins. Jurafsky, D. (1996). Universal tendencies in the semantics of the diminutive. Language, 72(3), 533–78. Kakridi-Ferrari, M. (1998). Επανάληψη: η λειτουργία της ως βασικού μηχανισμού στη γλώσσα. [Repetition: Its function as a basic mechanism in language]. PhD dissertation, National and Capodistrian University of Athens. Kallergi, H. & Konstantinidou, M. (2018). Reduplicative constructions involving distortion: An overview of Greek data with emphasis on the echo-word construction. In A. Urdze (ed.) Non-prototypical Reduplication. Berlin and Boston: De Gruyter, 91–150. Kamilaki, Μ. 2009. Τα λόγια στοιχεία στη νεανική επικοινωνία. Κοινωνιοπραγματολογική διερεύνηση της ποικιλίας [±ΛΟΓΙΟ] στον γραπτό λόγο νεαρών ομιλητών της Νέας Ελληνικής [Learned elements in youth communication. Socio-pragmatic investigation of the [±learned] variety in the written speech of young speakers of Modern Greek]. PhD dissertation, National and Capodistrian University of Athens. Kamilaki, M., Katsouda, G., & Vrachionidou, M. (2016). Πιπέρι στο στόμα: Όψεις των λέξεων-ταμπού στη Νέα Ελληνική [Pepper in the mouth: Aspects of taboo words in Modern Greek], 2nd edn. Athens: Kalligrafos. Kaplan, David (1999): The meaning of ouch and oops: Explorations in the theory of meaning as use. Howison Lecture in Philosophy delivered at UC Berkeley. Transcribed by Elizabeth Coppock. Online. https://eecoppock.info/ PragmaticsSoSe2012/kaplan.pdf www.youtube.com/watch?v=iaGRLlgPl6w Katsaros, P. (2018). Υποκορισμός στην Ελληνική και Ιταλική. Θεωρία της μορφοπραγματολογίας [Hypocorism in Greek and Italian. Theory of morphopragmatics]. PhD dissertation, Aristotle University of Thessaloniki. Katsouda, G. (2020). The Dialect of Kythera: Description and analysis. Athens: Society of Kytherian Studies. Kazazis, K. (1972). The status of Turkisms in the present-day Balkan languages. In H. Birnbaum & S. Vryonis (eds.) Aspects of the Balkans: Continuity and Change. Paris and The Hague: Mouton, 87–116. Kehayia, E. (1997). Insulting words in Modern Greek. In Proceedings of the 2nd International Conference on Greek Linguistics. Graz: Neugebauer Verlag, 297–304. Kerbrat-Orecchioni, C. (1980). L’énonciation de la subjectivité dans le langage. Paris: Armand Colin. Kiranoudis, P. (2001). Μορφολογία των τουρκικών δανείων της ελληνικής [Morphology of the Turkish loanwords in Greek]. PhD dissertation, Faculty of Philology,

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

223

Aristotle University of Thessaloniki. www.didaktorika.gr/eadd/handle/10442/ 13615 Klimaszewska, Z. (1983). Diminutive und augmentative Ausdrucksmöglichkeiten des Niederländischen, Deutschen und Polnischen: Eine konfrontative Darstellung. Wrocław: Wydawnictwo Polskiej Akademii Nauk. Konstantinidou, M. (1997). Sprache und Gefühl: Semiotische und andere Aspekte einer Relation. Hamburg: Helmut Buske Verlag. (2004). Παραμορφωτική αναδίπλωση λέξεων. Ένα διαλεκτικό σχήμα λόγου και ο λειτουργικός του ρόλος [Distortive lexical reduplication: A dialectical rhetorical figure and its functional role]. Μελέτες για την Ελληνική Γλώσσα [Studies in Greek Linguistics], 24, 345–56. (2005). Degenerative lexikalische Reduplikation durch [m] im Anlaut: Eine eigenartige rhetorische Figur im Mundartgriechischen und Türkischen. In Ε. Butulussi, E. Karagiannidou & K. Zachu (eds.) Sprache und Multikulturalität. Festschrift für Professor Dorfmüller-Karpusa. Thessaloniki: University Studio Press, 253–74. Koo, Hyun Jung & Rhee, S. (2016). Pejoratives in Korean. In R. Finkbeiner, J. Meibauer, & H. Wiese (eds.) Pejoration. Amsterdam and Philadelphia: Benjamins, 301–23. Körtvélyessy, L. (2012). Evaluative morphology from a cross-linguistic perspective. Habilitation thesis, Eötvös Loránd University Budapest. Körtvélyessy, L. (2015). Evaluative Morphology from a Cross-Linguistic Perspective. Newcastle upon Tyne: Cambridge Scholars Publishing. Kriaras = Kriaras, E. (1968–). Λεξικό της μεσαιωνικής ελληνικής δημώδους γραμματείας, 1100–1669 [Dictionary of Medieval Greek vernacular literature 1100–1669]. Thessaloniki: Centre for the Greek Language. LKN = Λεξικό της Κοινής Νεοελληνικής [Dictionary of Standard Modern Greek]. (1998). Thessaloniki: Institute for Modern Greek Studies. LNEG = Babiniotis, G. (2019). Λεξικό της νέας ελληνικής γλώσσας [Dictionary of Modern Greek Language]. 5th ed. Athens: Centre of Lexicology. LSJ = Liddell, H. G. & Scott, R. (1996). A Greek–English Lexicon. Revised and augmented by Sir H. S. Jones. Oxford: Clarendon Press. LSK = Α. Konstantinidis (ed.) (1909). Μέγα λεξικὸν τῆς ἑλληνικῆς γλώσσης [Great Dictionary of the Greek Language] Athens: I. Sideris. Mackridge, P. (1990). Katharevousa (c.1800–1974): An obituary for an official language. In M. Sarafis & E. Martin (eds.) Background to Contemporary Greece. London: Merlin Press, 25–51. (2004). Diglossia and the separation of discourses in Greek culture. In Теоретические проблемы языкознания: сборник статей к 140-летию кафедры общего языкознания Филологического факультета СанктПетербургского государственного университета [Teoreticheskie problemy yazykoznaniya. Sbornik Statey k 140-letiyu Kafedry obshchego yazykoznainya Filologicheskogo fakulteta Sankt-Peterburgskogo gosudarstvennogo universiteta]. [Theoretical problems of linguistics: A collection of articles dedicated to the 140th anniversary of the Department of General Linguistics of the Philological Faculty of St Petersburg State University]. St Petersburg: Faculty of Philology, St Petersburg State University, 112–30.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

224

Kallergi, Katsouda, Konstantinidou, Tsangalidis

Manolessou, I., Katsouda, G., & Bassea-Bezantakou, C. (2019). Ο χαρακτηρισμός ‘λόγιο’ στη λεξικογραφία: η περίπτωση του Ιστορικού Λεξικού της Ακαδημίας Αθηνών [The label ‘learned’ in lexicography: The case of the Historical Dictionary of Modern Greek of the Academy of Athens]. In Α. Fliatouras & Α. AnastasiadisSymeonidis (eds.) Το λόγιο επίπεδο στη σύγχρονη Νέα Ελληνική: Θεωρία, ιστορία, εφαρμογή [The learned register in contemporary Modern Greek: Theory, history, applications]. Athens: Patakis, 200–17. McCready, E. (2010). Varieties of conventional implicature. Semantics and Pragmatics, 3(8), 1–57. Meibauer, J. (2013). Expressive compounds in German. Word Structure, 6(1), 21–42. DOI:10.3366/word.2013.0034 Melissaropoulou, D. (2007). Μορφολογική περιγραφή και ανάλυση του μικρασιατικού ιδιώματος Κυδωνιών και Μοσχονησίων [Morphological description and analysis of the Asia Minor dialect of Kydonies and Moschonisia (Aivaliot)]. PhD dissertation, University of Patras. (2009). Augmentation vs. diminution in Greek dialectal variation: An optimal system. In G. Boyé, F. Montermini, & J. Tseng (eds.) Selected Proceedings of the 6th Décembrettes: Morphology in Bordeaux. Somerville, MA: Cascadilla, 125–37. (2015). Modern Greek. In N. Grandi & L. Körtvélyessy (eds.) Edinburgh Handbook of Evaluative Morphology. Edinburgh: Edinburgh University Press, 269–77. https://www.jstor.org/stable/10.3366/j.ctt1g09zkx.29 Merlini Barbaresi, L. & Dressler, W. U. (2020). Pragmatic Explanations in Morphology. In V. Pirrelli, I. Plag & W. U. Dressler (eds.) Word Knowledge and Word Usage: A cross-disciplinary guide to the mental lexicon. Berlin: de Gruyter, 405–51. Minas, K. (1978). Ἡ μορφολογία τῆς μεγεθύνσεως στὴν ἑλληνικὴ γλώσσα. [The morphology of augmentation in the Greek language]. Ioannina: University of Ioannina. Mutz, K. (2015). Evaluative morphology in a diachronic perspective. In N. Grandi & L. Körtvélyessy (eds.) Edinburgh Handbook of Evaluative Morphology. Edinburgh: Edinburgh University Press, 142–55. Navarro Ibarra, Ía. (2009). Predicados complejos con le en español mexicano. PhD thesis, Universitat Autònoma de Barcelona. Norri, J. (2000). Labelling of derogatory words in some British and American dictionaries. International Journal of Lexicography, 13(2), 71–106. Ochs, E. & Schieffelin, B. (1989). Language has a heart. Text, 9, 7–25. https://dx.doi .org/10.1515/text.1.1989.9.1.7 Papyrus Larousse (1981–1994) = Εγκυκλοπαίδεια Πάπυρος Λαρούς Μπριτάννικα: Ερμηνευτικό και ετυμολογικό λεξικό της ελληνικής γλώσσας [Encyclopaedia Papyrus Larousse Britannica: Interpretive and Etymological Dictionary of the Greek Language] 61 vols., Athens: Papyrus. Petrounias, E. (1991). Νεοελληνικό επίθημα -αρία από τα βενετσιάνικα και συγγενικά. Πολλαπλές πηγές δανεισμού και αναδανεισμός [The Modern Greek suffix -aría from Venetian and relatives. Multiple loan sources and reborrowing]. Μελέτες για την Ελληνική Γλώσσα [Studies in Greek Linguistics], 11, 53–69. Potts, C. (2005). The Logic of Conventional Implicatures. Oxford: Oxford University Press.

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

225

(2007). The expressive dimension. Theoretical Linguistics, 33, 165–97. (2012). Conventional implicature and expressive content. In K. von Heusinger, C. Maienborn, & P. Portner (eds.) Semantics: An international handbook of natural language meaning, Vol. 3. Berlin and Boston: De Gruyter, 2516–36. (2019). Conventional implicature and expressive meaning. In C. Maienborn, K. von Heusinger, & P. Portner (eds.) Semantics – Interfaces. Berlin and Boston: De Gruyter, 598–622. Predelli, St. (2010). From the expressive to the derogatory: On the semantic role for non-truth-conditional meaning. In S. Sawyer (ed.) New Waves in Philosophy of Language. Basingstoke: Palgrave Macmillan, 164–85. Prieto, V. M. (2015). The semantics of evaluative morphology. In N. Grandi & L. Körtvélyessy (eds.) Edinburgh Handbook of Evaluative Morphology. Edinburgh: Edinburgh University Press, 21–31. Ralli, A. (1992). Compounding in Modern Greek. Rivista di Linguistica, 4(1). 143–73. (2013). Compounding in Modern Greek. Dordrecht: Springer. Regier, T. (1998). Reduplication and the arbitrariness of the sign. In M.-A. Gernsbacher & S. J. Derry (eds.) Proceedings of the Twentieth Annual Conference of the Cognitive Science Society. Mahwah, NJ: Lawrence Erlbaum Associates, 887–92. Revithiadou, A. (1997). Prosodic domains in Greek compounding. In G. Drachman, A. Malikouti-Drachman, C. Klidi, & J. Fykias (eds.) Greek Linguistics ’95: Proceedings of the 2nd International Conference on Greek Linguistics. Graz: Neugebauer Verlag, 107–16. Scalise, S. (1984). Generative Morphology. Dordrecht: Foris. Schlenker, P. (2007). Expressive presuppositions. Theoretical Linguistics, 33(2), 237–45. DOI:10.1515/TL.2007.017 Schmidt, G.-D. (1987). Das Affixoid. Zur Notwendigkeit und Brauchbarkeit eines beliebten Zwischenbegriffs in der deutschen Wortbildung. In G. Hoppe, A. Kirkness, E. Link, I. Nortmeyer, W. Rettig & G.-D. Schmidt (eds.) Deutsche Lehnwortbildung. Beiträge zur Erforschung der Wortbildung mit entlehnten WBEinheiten im Deutschen. Tübingen: Narr, 53–101. Setatos, M. (1971). Transformations of emotive meaning in Modern Greek Koine. Orbis, 20, 458–69. Stankiewicz, E. (1964). Problems of emotive language. In T. Sebeok, A. Hayes & M. C. Bateson (eds.) Approaches to Semiotics: Cultural anthropology, education, linguistics, psychiatry, psychology; transactions of the Indiana University Conference on Paralinguistics and Kinesics. The Hague: Mouton, 239–64. (1968). Declension and Gradation of Russian Substantives. The Hague: Mouton. Stavrianaki, K. (2002). Κατασκευασμένες λέξεις σε -άκι: είναι δυνατή μια ενιαία ανάλυση [Constructed words in -áki: is a single analysis possible?]. Μελέτες για την Ελληνική Γλώσσα [Studies in Greek Linguistics], 22, 605–15. (2009). -άκι και -ette: Τα υποκοριστικά επιθήµατα της Ελληνικής και της Γαλλικής. Συγκριτική µελέτη και προεκτάσεις για τη γλωσσική εκµάθηση και διδασκαλία [-áki and -ette: The diminutive suffixes in Greek and French. Comparative study and implications for language learning and teaching]. Πρακτικά διεθνούς συνεδρίου 2008, ευρωπαϊκό έτος διαπολιτισμικού διαλόγου: Συνομιλώντας με τις γλώσσεςπολιτισμούς. [Proceedings of the International Conference 2008, European Year for Intercultural Dialogue. Conversing with languages-cultures]. Thessaloniki:

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

226

Kallergi, Katsouda, Konstantinidou, Tsangalidis

School of French Language and Literature, Aristotle University of Thessaloniki, 759–69. Stephany, U. & Thomadaki, E. (2017). Compounding in early Greek language acquisition. In W. U. Dressler, F. N. Ketrez & M. Kilani-Schoch (eds.) Nominal Compound Acquisition. Amsterdam: Benjamins, 119–43. Steriopolo, O. (2008). Form and function of expressive morphology: A case study of Russian. PhD thesis, University of British Columbia. Strong, J. (1890). The Exhaustive Concordance of the Bible: Showing every word. New York: Hunt & Eaton, Cranston & Curts. Stylianou, E. (2018). Το επίθημα -ίσκ(ος) στη νέα ελληνική [The suffix -ísk(os) in Modern Greek]. Μελέτες για την Ελληνική Γλώσσα [Studies in Greek Linguistics], 38, 345–54. Suzuki, S. (1998). Pejorative connotation: A case of Japanese. In A. H. Jucker & Y. Ziv (eds.) Discourse Markers: Descriptions and theory. Amsterdam: Benjamins, 261–76. DOI:10.1075/pbns.57.13suz Tenchini, M. P. & Frigerio, A. (2016). A multi-act perspective on slurs. In R. Finkbeiner, J. Meibauer & H. Wiese (eds.) Pejoration. Amsterdam and Philadelphia: Benjamins, 167–86. DOI:10.1075/la.228.08ten Thayer, J. H. (1995). Thayer’s Greek–English Lexicon of the New Testament: Coded with Strong’s Concordance Numbers. Peabody, MA: Hendrickson Academic. Tsiakmakis, E., Borràs-Comes, J. & Espinal, M.-T. (2021). The interpretation of plural mass nouns in Greek. Journal of Pragmatics, 181, 209–26. Tzitzilis, C. (1997). Die türkischen Elemente im Neugriechischen verglichen mit den türkischen Elementen in anderen Balkansprachen. Zeitschrift für Balkanologie, 33 (1), 101–12. Urdze, A. (ed.) (2018). Non-prototypical Reduplication. Berlin and Boston: De Gruyter. Wiese, H. & Polat, N.-T. (2016). Pejoration in contact: m-reduplication and other examples from urban German. In R. Finkbeiner, J. Meibauer & H. Wiese (eds.) Pejoration. Amsterdam and Philadelphia: Benjamins, 243–67. Williams, J. P. (ed.) (2013). The Aesthetics of Grammar: Sound and meaning in the languages of Mainland Southeast Asia. Cambridge: Cambridge University Press. (ed.) (2021). Expressive Morphology in the Languages of South Asia. Abingdon: Routledge. Williamson, T. (2009). Reference, inference, and the semantics of pejoratives. In J. Almog & P. Leonardi (eds.) The Philosophy of David Kaplan. Oxford: Oxford University Press, 137–58. DOI:10.1093/acprof:oso/9780195367881.003.0009 XLNEG = Χρηστικό λεξικό της νεοελληνικής γλώσσας [Utilitarian Modern Greek dictionary]. Athens: Academy of Athens, 2014. Xydopoulos, G. & Christopoulou, K. (2019). Μέτρηση της προσβλητικότητας σε λέξεις του περιθωριακού λεξιλογίου της νέας ελληνικής: εκτιμώντας τα πρώτα αποτελέσματα [Measuring offensiveness in marginal vocabulary items in Modern Greek: assessing the first results]. In T. Markopulos, C. Vlachos, A. Archakis, D. Papazachariou, G. I. Xydopoulos & A. Roussou (eds.) Proceedings of the 14th Conference for the Greek Language, Patras, 5–8 September 2019. Patras: University of Patras, 936–46. https://pasithee.library.upatras.gr/icgl/article/ download/3728/3771 DOI: 10.26220/icgl.v1i1.3728

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Expressivity in Modern Greek

227

Yakimova, N. (2000). Markedness and intensity. In Z. Catalan, E. Pancheva, A. Shurbanov, K. Stamenov & T. Ward (eds.) Seventy Years of English and American Studies in Bulgaria. Sofia: St. Kliment Ohridski University Press. Zwicky, A. M. & Pullum, G. K. (1987). Plain morphology and expressive morphology. In J. Aske, N. Beery, L. Michaelis & H. Filip (eds.) Proccedings of the Thirteenth Annual Meeting of the Berkeley Linguistics Society. General Session and Parasession on Grammar and Cognition. Berkeley, CA: Berkeley Linguistics Society, 330–40. https://doi.org/10.3765/bls.v13i0.1817

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.009 Published online by Cambridge University Press

Romance

https://doi.org/10.1017/9781108989084.010 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.010 Published online by Cambridge University Press

7

Repetition and reduplication in Italian Anna M. Thornton*

The linguist’s task is [. . .] to study the whole range of repetition in discourse, and in doing so to seek out those regularities which promise interest as incipient sub-systems. (Hopper 1987: 142)

Many of the examples [. . .] lie at the uncomfortable boundary between parole and langue, where it is not always clear when grammar has emerged from discourse. (Evans 2007: 369)

7.1

Introduction

A distinction between repetition and reduplication is considered of primary importance by many scholars, and there have been numerous attempts at drawing the boundary between two clearly related phenomena (see e.g., Gil 2005, Stolz & Levkovych 2018).1 For some authors this boundary coincides with the boundary between (traditionally conceived) morphology and syntax: reduplication would operate within word boundaries, hence belong to ‘morphology’, repetition above the word level, hence belong to syntax. Such a view is exposed, for example, by Rubino (2005: 11, my emphasis), who defines reduplication as the ‘systematic repetition of phonological material within a word for semantic or grammatical purposes’. Under such a definition, little could be said about reduplication in Italian: it is by now a commonplace observation that contemporary Western European languages do not exploit (partial) reduplication in their inflectional morphology (Rubino 2005: 23),2

* University of L’Aquila 1 I acknowledge with thanks all the friends and colleagues who have provided references, discussed specific points with me, and/or read drafts of the text. In alphabetical order, they are: Federica Casadei, Paolo D’Achille, Sara Di Giovannantonio, Maria Grossmann, Francesca Masini, Pietro Maturi, Andrea Sansò, and Miriam Voghera. None of them are responsible for the views expressed in this chapter. 2 Contrary to older stages of these languages, where partial reduplication in inflection is attested, e.g., to form the perfect in Ancient Greek (léluka ‘I loosened’, léloipa ‘I left’, etc.) and residually in some Latin verbs (didici ‘I learned’, tetendi ‘I tended’, etc.). According to Rubino’s (2013) WALS survey, Italian does not display reduplication at all, not even total reduplication;

231

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

232

Anna M. Thornton

and Italian is no exception. However, drawing the boundary between repetition and reduplication in this way faces immediate problems when lexeme formation (commonly taken to be part of morphology) is considered (as pointed out by Stolz and Levkovych 2018: 30, among others); it is notoriously difficult to decide whether certain sequences constitute a compound word or a phrase, and cases of repetition or total reduplication in this area abound, even in Italian. In (1) I list a few Italian examples in rough order of decreasing wordhood (detailed discussion of some of these cases will be offered below; the list in (1) is not meant to be exhaustive, but only representative of the Italian phenomena that must be considered in this contribution). (1)

a. lemme lemme ‘(walking) very slowly’, *lemme not attested independently leccalecca ‘lollipop’, lit. lick lick b. fuggifuggi ‘stampede’ lit. runaway runaway c. passo passo ‘step by step’ lit. step step d. piccolo piccolo ‘very small’ lit. small small e. cammina cammina ‘as time goes by’ lit. walk walk

A different approach to drawing the boundary between repetition and reduplication is proposed by Stolz and Levkovych (2018): they assume that ‘repetition fulfills exclusively pragmatic tasks whereas reduplication is responsible for the expression of grammatical categories’. This definition shifts the boundary from that between morphology and syntax to that between ‘grammar’ and pragmatics. However, this second boundary is no more clear-cut than the first one: as forcefully argued by Mauri and Masini (2022), at least some ‘grammar’ emerges from spontaneous speech, where speakers exploit all the resources of their language to convey meaning, and even express emotions, and hearers interpret what speakers say, often by means of ‘pragmatic inferences [. . .] thus adding new layers of meaning, which may eventually conventionalize’ (Mauri & Masini 2022: 104; for background literature on this approach, see at least Hopper 1987, Tannen 1987). Therefore, I agree with Hopper (1987) and Mauri and Masini that a more productive approach to understanding the phenomena that are variously labelled as repetition or reduplication (and also iteration, doubling, replication and the like)3 is to assume a continuity between more pragmatically motivated discourse

3

however, this view is plainly wrong, as repeatedly observed in work by Thomas Stolz and colleagues (Stolz et al. 2011, 2015), and as well known by Italianists since at least the 1950s. Practically every contribution on reduplicative constructions contains observations on the often interchangeable usage of several different terms to label phenomena which might be considered the same; see, e.g., Stolz (2018: 201, fn 1). Many authors propose terminological distinctions to refer to what they take to be different phenomena (see in particular Gil 2005, Stolz & Levkovych 2018), but not all these distinctions converge on separating the same phenomena under the same label. Kim (2012: 441) aptly summarizes the issue: for ‘reduplication’ there are ‘as many different definitions as sources consulted’.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

233

repetition and more grammaticalized examples of repeated occurrence of identical elements, which may be called reduplication for convenience.4 A description of the kind of phenomena encountered at a given stage of a single language, such as the one attempted for Italian in the present contribution, will hopefully be able to assess at what point on a discourse to grammar continuum each phenomenon is collocated. An almost inescapable ingredient of such an approach is the adoption of a view which does not assume clear-cut boundaries between the lexical, morphological, syntactic and pragmatic levels, such as that offered by Construction Grammar in its various declinations (Hoffmann & Trousdale 2013). So the real boundary to search for would be that between already grammaticalized constructions, regardless of whether they form words, phrases or even sentences, and single instances of repetition which are motivated in the specific discourse context in which they occur, but are not (yet?) instances of a construction that is represented also at a more abstract level. In the present contribution, I will adopt this approach in describing the repetitive or reduplicative constructions of Italian. Another distinction which must be discussed before proceeding to the description of the Italian facts is that between total and partial reduplication. Since partial reduplication, that is, reduplication of elements smaller than a full word form, is commonly employed to express inflectional category values (‘plural’, ‘past’, etc.), which, as already observed, are not usually expressed by reduplication in Western European languages, one might hypothesize that Italian will not offer examples of partial reduplication. However, a number of cases of repetition of elements smaller than a full word are attested in Italian, and some of them have clearly expressive functions, and hence are of interest in the context of the present volume; they will be briefly illustrated in §7.3. Finally, I must state which potentially relevant phenomena I will address only cursorily in the present contribution. A useful classification of types of structures involving repetition is offered by Fischer (2011), who distinguishes between the following categories:5 (i) (ii) (iii) (iv) 4

5

repetition at clause or phrase level repetition at word or morpheme level repetition at the sub-morphemic level ‘Ablaut reduplication’

As observed by Moravcsik (1978: 300, fn 3) the term reduplication is ‘infelicitous’ in that it implies there being ‘only two copies of the same thing’, while in principle there might be constructions requiring three (or more) occurrences of the same element (such cases are further discussed by Moravcsik (1978: 312–13)); however, Moravcsik, like most other authors, retains reduplication because of its widespread usage. Ghomeshi et al. (2004: 309) also offer a list of reduplicative phenomena found in English; their list partially overlaps with Fischer’s one, but since it is focused on and limited to English, I will not use it as a guide here.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

234

Anna M. Thornton

(v) ‘onomatopoeic reduplication’ (vi) reduplication in babytalk/motherese (vii) hypocoristic use In §7.2 I will discuss only very briefly types (iv)–(vii), not only because of space limitations, but also because I think they are less interesting in the context of Italian, for a number of reasons. The bulk of the present article will be devoted to types (i)–(iii), described in §7.3–§7.5. Section 7.6 discusses the findings and concludes. 7.2

Repetition and reduplication in ideophones, onomatopoeic formations, child language, and baby talk

‘Ablaut reduplications’ (type iv), that is, words like tic tac ‘tick-tock’, zig-zag ‘id.’ (often ideophones), are attested in Italian, but they do not appear to present peculiarities with respect to other languages; many of these elements are attested in a phonologically identical or very similar form also in other Western European languages, and are often loanwords from one language to the other (Mioni 1990: 257 mentions ‘i tipi europei zigzag, dindon, tictac’ [the European types zigzag, dindon, tictac], implicitly suggesting that these types have little or no language-specific properties). A similar situation holds with what Fischer calls ‘onomatopoeic reduplication’ (type v). This category, like the previously mentioned one, includes ideophones, that is, words which represent sounds emitted by animals, natural sources or artifacts6 (some of which are constructed by means of ‘Ablaut reduplication’ rather than full reduplication of identical elements, and would therefore fall within type (iv) in Fischer’s classification) and properly onomatopoeic lexemes, that are based on ideophones but are phonologically and grammatically more integrated lexemes. A few Italian examples of reduplicated ideophones and onomatopoeic lexemes are listed in (2);7 note that the verbs in (2b) contain a full (gloglottare) or partial reduplication followed by /tː/ or /ʎː/ (orthographically and ) before the inflectional endings: (2)

6

7

a. ideophones pio pio ‘sound emitted by chicks’ cip cip ‘sound emitted by little birds’

In principle, ideophones may represent various sorts of sensory properties, such as colour, taste, smell, etc. (Dingemanse 2015). However, Mioni (1990), in a corpus-informed study of ideophones in Italian, found that only ideophones depicting sounds and noises were attested in his corpus (comics, both translated from English and originally written in Italian). It is not clear whether bla bla (also bla bla bla), which occurs in Italian in functions similar to those that have been described for parallel expressions in other languages (see in particular Finkbeiner 2018), has an onomatopoeic origin. The etymology of bla bla deserves a separate full treatment and must be left for future research.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

235

chicchirichì ‘sound emitted by roosters’ tutù ‘sound emitted by trains’ tuttù ‘sound emitted by a busy telephone landline’ toc toc ‘sound of knocking on door’ drin drin ‘sound of telephone ringing’ b. onomatopoeic lexemes pissi pissi ‘the act of saying psst’ gloglottare ‘gobble, gurgle’ borbottare ‘mutter’ bisbigliare ‘whisper’ tartagliare ‘stutter’

It is possible that the idiom fare giacomo giacomo ‘wobble; lit. to do James James’, predicated of legs, also originates like type (2a), from an ideophone giac /dʒak/ representing the sound of shuffling feet, reduplicated and crossed with the anthroponym Giacomo ‘James’; this etymology is debated; Castellani Pollidori (2002) offers a review of etymological proposals and a different interpretation. Fischer’s types (vi) and (vii) are also closely related, as observed by Thornton (1996: 104–09). Italian hypocoristics usually have the shape of a minimal prosodic word, that is, a disyllabic trochee ending in a vowel. This shape can be arrived at either by mapping the base (the full proper name) to the minimal word template from left to right (3a), or by circumscribing the base’s final foot (3b) (see Thornton 1996 for many examples and details). In productively formed hypocoristics, reduplication of phonemes or syllables is not a requirement, and occurs only if the base already accidentally contains sequences which present some element repeatedly (for example, Ceci < Cecilia). Besides these types, however, Italian also has many hypocoristics which contain reduplication of syllables or at least of phonemes (3c): these have been analysed by Thornton (1996) as formed by very young children (or possibly by caregivers speaking in baby talk, a variety which has many phonological characteristics in common with children’s speech) and then frozen as lexical representations of proper nouns. This last type of hypocoristics displays many characters typical of child language, among which are reduplication, and consonant and vowel harmony, which also result in reduplication of some of the phonological material in the output (see Thornton 1996: 104–09 for a list of phonological properties of child language). (3)

a. productive hypocoristics formed by mapping the base to the minimal word template Ale < Alessandra / Alessandro Eli < Elisabetta Emi < Emilia Edo < Edoardo Nico < Nicoletta / Nicola Simo < Simona / Simone

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

236

Anna M. Thornton b. productive hypocoristics formed by prosodic circumscription of the base’s final foot Sandro < Alessandro, Sandra < Alessandra Betta < Elisabetta Nando < Ferdinando Renzo < Lorenzo Tore < Salvatore Vico < Lodovico c. hypocoristics which display features of child language Lallo < Alessandro, Claudio Lilli < Elisabetta Titti < Tiziana Checco /ˈkekko/ < Francesco Dado < Edoardo Mimma < Emilia Teta < Teresa

One and the same name can give rise to hypocoristics of all three types (e.g., Elisabetta > Eli, Betta, Lilli; Alessandro > Ale, Sandro, Lallo), but only type (3c) always contains identical consonants in the onset of the two syllables, yielding apparent partial reduplication, and sometimes even identical vowels (as in Lilli, Titti); most often, however, the final vowels -a and -o, typically occurring in names for females and males respectively, are retained, so that reduplication in the output is not full.8 Italian lexemes typically used in baby talk frequently display syllable reduplication (sometimes with a word-internal geminate, like the hypocoristics in (3c)); a representative list (from Bernini 2010) is given in (4), where (4b) contains examples of ideophones metonymically used as nouns, linking type (vii) to type (v) above. (4)

Italian baby talk lexemes (Bernini 2010) a. pappa ‘food, meal’ pipì ‘pee’ popò / pupù ‘pooh’ nanna ‘sleep’ b. ciufciuf ‘train’ brumbrum ‘automobile’ baubau ‘dog’

I will not elaborate further on these types, since they all display properties which have been recognized to be cross-linguistically common, and even universal, in their respective domains, so that Italian does not offer any new insights on their formation. 8

Most of the hypocoristics in (3c) have a long consonant word-medially; normally such segments are considered to constitute the coda of the first syllable and the onset of the second one, so that the words do not display exact full syllable reduplication, since the first syllable has a coda and the second one does not.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

237

In §7.3–§7.5 I will describe and discuss what Italian has to offer in terms of Fischer’s types (iii), (ii), and (i) respectively. 7.3

Repetition and reduplication of units below the word

7.3.1

Consonant reduplication

There is some evidence that doubling of letters in the spelling of nouns denoting products is exploited for expressiveness. I will only discuss the two Examples in (5) and (6), both from advertising: (5)

[Young man holding a bottle of Biorepair mouthwash (the advertised product) speaking]: Io suggerisco a Biorepair di scrivere colluttorio con due ‘t’ e non con una, perché il suo è più [t’tɔrjo], più denso, perché è l’unico che ha il microrepair per riparare lo smalto e in più ha un’azione antibatterica senza la clorixidina. ‘I suggest that Biorepair should write colluttorio [‘mouthwash’] with two t’s instead of one, because their mouthwash is more [t’tɔrjo], denser, because it is the only one that has micro-repair to restore the enamel, and also has an antibacterial effect without chlorhexidine.’ [voice-over]: Da oggi Biorepair anche in colluttorio. ‘From today on Biorepair also comes as a mouthwash [kollut’tɔrjo]’.9

Colluttorio is a common variant, frowned-upon by grammarians and often also by medical professionals, of the word collutorio ‘mouthwash’, from Latin colluĕre ‘wash out, rinse’, whose third stem is collut-; Latin deverbal derivatives in -ōrius are based on the verb’s third stem (as defined by Aronoff 1994), and Italian learned derivatives in -orio are also based on the Latin third stem (see Thornton 2015: 794–6 for details); therefore, collutorio is the regular formation, and it is unclear how the variant with /tː/ developed; an influence from the noun colluttazione ‘scuffle’ is suggested,10 although there does not seem to be any semantic connection between the two words. Be that as it may, the existence of the variant colluttorio has been exploited by the commercial described and transcribed in (5). The next example, instead, does not appear to exploit an existing variant of a word,11 but creates a new word, a trademark name: (6)

9 11

a. la RiCcotta, quella con due “c”, gustosa come sempre, con solo 186 kcal e una ricetta innovativa che la mantiene fresca per 30 giorni, e la RiCcotta Spalmabile,

10 www.youtube.com/watch?v=fPcpkPWlumA For example, by https://bit.ly/3y4L1v1 Even if riccotta is attested as an antiquated variant of ricotta; however, contemporary speakers are not aware of this variant, whose only attestation in GDLI dates back to the first half of the seventeenth century at the latest; on the contrary, most speakers actually prefer the variant colluttorio and often ignore the fact that collutorio is the ‘correct’ form.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

238

Anna M. Thornton nuovo prodotto fatto senza conservanti e con soli 3 ingredienti naturali: siero di latte, crema di latte e sale.12 ‘the RiCcotta, the one with two “c’s”, tasty as always, with only 186 kcal and an innovative recipe that keeps it fresh for 30 days, and the RiCcotta spread, a new product made without preservatives and with only 3 natural ingredients: whey, cream of milk and salt.’ (boldface in the original text) b. con RiCcotta il gusto raddoppia13 ‘with RiCcotta the taste doubles’

By reading the presentation in (6a) one may wonder why the noun ricotta ‘id.’ was manipulated to create the trademark name RiCcotta, by blending the cheese’s name with the adjective ricco ‘rich’,14 since richness makes one think of increased content, added ingredients, while the advertising presents RiCcotta as having ‘only’ three natural ingredients and ‘only’ 186 kcal [per 100 grams, presumably], so as poor rather than rich in content. The ‘poorness’ in ingredients and calories is at odds with the ‘richness’ implied by the name. The mystery is solved when one looks at the television advertisement, which ends with the slogan reported in (6b): the spelling with a double C is meant to convey the idea that taste ‘doubles’ – so what is richer is taste rather than the list of ingredients or the protein and fat content. It is important to note that these exploitations of reduplication as an iconic device to express the presence of ‘more of something’ that is considered desirable in a product (be it effectiveness of a mouthwash, or taste of a food) only work at the orthographic level, where Italian spelling represents a long consonant by doubling the letter that represents it; phonologically, nothing ‘doubles’ in colluttorio and RiCcotta, where orthographic and represent phonological /tː/ and /kː/ respectively.15

12 14

15

13 https://bit.ly/3mh8c2y www.youtube.com/watch?v=tmiKqmVYLBA The form riccotta could also represent the feminine singular form of the adjective riccotto ‘rather rich’, which could be formed by adding the diminutive suffix -otto to ricco ‘rich’; however, this adjective does not seem to be attested. Another possible analysis would consider riccotta as just a blend of ricco and ricotta, and no orthographic doubling would need to be called into question; however, this blend would be unusual in having the adjective in prenominal position, while normally in Italian adjectives are postnominal; besides, most blends involve shortening of at least one of the two source words, while here there is no shortening (at most, ricco appears as a full stem without an agreeing inflectional ending); finally, the operation of orthographic doubling is emphasized by the official spelling, in which a capital C is inserted inside ri__cotta. In any case, riCcotta is less compelling than colluttorio as a case of purely orthographic doubling. Stolz et al. (2011: 59) also discuss consonant gemination as a candidate reduplicative phenomenon, and observe that in deciding the issue ‘too much depends on the stance the analyst takes in terms of phonological theories’. Italian is uncommon among languages of Europe in having distinctive consonantal length, which is represented by double letters orthographically (except in the case of inherently long segments, in which a variety of orthographic solutions is employed).

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

7.3.2

239

Vowel iteration

A similarly purely orthographic repetition or iteration occurs with vowels. Iteration (usually three, sometimes even more times) of a stressed vowel is a spelling device used to represent prolonged realization of a vowel in speech for stylistic purposes of emphasis and expressiveness.16 It is relatively common with adjectives expressing dimension or value, for example, in customers’ reviews of places and products, where it is easy to find entries containing (often as titles) words like ‘big, lit. biiig’ (used to convey positive appreciation, almost a synonym of ‘good’) or ‘small, little, lit. liiittle’, as in the review of a car in (7): (7)

Il display centrale è piuttosto ricco di informazioni, anche se scritte in carattere piiiiiccolo che rende difficile la lettura a colpo d’occhio17 ‘The central display is quite rich in information, although written in a smaaaaall font that makes it difficult to read at a glance’

Graphic vowel iteration is also commonly used to ‘depict’ (Dingemanse 2015) in writing the phonetic realization of interjections, where vowel duration can vary continuously. Examples are legion, even in canonical Italian literature; the example in (8) shows that the number of vowel iterations can vary for stylistic effects:18 La Marchesa sbadigliando parlò con ammirazione della badessa: ‘Come s’è portata!’ diss’ella ‘non mi aspettava tanto; ah! che contegno! aah! che dignità! aaah! che disinvoltura!’ ‘[The Marquise yawned and spoke admiringly of the abbess: ‘How she behaved!’ she said, ‘I didn’t expect so much; ah! what demeanour! aah! what dignity! aaah! what ease!’]’

(8)

(Alessandro Manzoni, Fermo e Lucia, 1823)

7.3.3

Syllable repetition in songs

The last kind of repetition of units below the word level worth mentioning here is the repetition of syllables in songs. The repeated units are mostly meaningless, and used for purely rhythmic purpose.19 An example is the text of the 16 17 18

19

Prolonged realization in speech of a stressed vowel is an instance of ‘analog acoustic expression’, used ‘to emphasize analogically the meaning of a lexical item’ (Shintel et al. 2006: 171). https://bit.ly/3IO9rhj A search in the Biblioteca Italiana Zanichelli, which contains over 1,000 texts of Italian literature from its origins to the early twentieth century, yields 10 tokens of aah, 8 tokens of aaah, and 3 tokens of aaaah; all these types have their first attestation in sixteenth-century authors, although frequency increases in the nineteenth and twentieth century. They appear to be an instance of ‘the reduplication of a phonetic string regardless of its meaning’, a type unattested according to Moravcsik (1978: 305).

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

240

Anna M. Thornton

song Ba-ba-baciami piccina, which is available online;20 the web resource also provides an English translation of the text,21 where word-initial syllable reduplication is rendered as consonant reduplication: for example, ba-babaciami ‘red-red-kiss=me’ is ‘translated’ as k-k-kiss me). In the whole song, there is repetition of word-initial CVs or full syllables (ba-ba-baciami, bo-bobocca ‘red-red-mouth’, tan-tan-tanti ‘red-red-many’), and once even of a syllable consisting of a consonant realizing a proclitic pronoun and the following word’s initial syllable (tin-tin-t’interessa red-red-you.dat=interests ‘it interests you’); besides, semantically completely empty syllables are reiterated as a refrain (tarataratarataratatà, tereteretereteretete, bi-a-be, bi-obu, bi-a-ba); only one semantically loaded (monosyllabic) word is repeated: il cuor cuor cuor ‘the heart heart heart’. This song was first recorded in 1940; however, the stylistic devices it uses are still common in Italian popular music; the song Mille (‘a thousand’) sung by Fedez, Orietta Berti, and Achille Lauro, which was a great hit in 2021, contains the lines in (9): a. sa red b. suoni you honk

(9)

sa red il the

sabato saturday cla red

sera night cla ca cabrio nera red red cabriolet black

In (9b) the sequence cla cla is presumably the iteration of the onset and nucleus of the first syllable of the word clacson ‘horn’, which is not even realized in the following part of the text, which proceeds with the reduplication of the initial syllable of the next word, cabrio. Apparently, the reiterated syllables are either open or end in a sonorant, while syllables ending in an obstruent are truncated and realized without their coda (clacson /ˈklakson/ yields cla cla, not *clac clac). 7.4

Repetition and reduplication at the word level and reduplication as a lexeme-formation device

This section describes Italian lexemes that display a reduplicative structure. Some are isolated lexemes, each one with its own history; some are the product of lexeme formation rules, or instantiate constructions that are productive, or at least available for coining new items, and ‘vital’ in the sense of Stolz et al. (2011: 108), that is, a reduplicated and a non-reduplicated item coexist and can be associated by speakers. My main aim is that of giving information on what kinds of structures are attested. I have drawn data from the literature and from my native speaker’s competence, but also from the itTenTen16

20

https://bit.ly/41ufnot

21

https://bit.ly/3KIvcC1

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

241

corpus,22 from which I have extracted the 500 most frequent sequences of two identical words: the types of constructions described in this article and the examples used to illustrate them are primarily the ones that are attested among these 500 sequences. The contexts in which the sequences occurred have been inspected to check for errors (such as dittographies) and peculiarities. Other corpora and sources of data have been used occasionally, and are cited when necessary. 7.4.1

Nouns

7.4.1.1 Loanwords and nouns formed with non-native material Italian has a number of nouns made up of two identical parts, usually monosyllabic but sometimes disyllabic. These nouns are all loanwords, usually from French or English (in which they are often loanwords in their turn, from languages of Asia or Africa). The reduplicated part is not attested in isolation, and the process of forming such nouns is not productive. Examples of common nouns of this kind are given in (10):23 (10)

agar agar, can can, cha cha / cha cha cha, cin cin ‘chin-chin, used as a toast before drinking; toast’, couscous, dumdum ‘dum-dum bullet’, pon pon, tam tam, ylang ylang, yo yo

Two words circulating more specifically in Italian are bunga bunga ‘Silvio Berlusconi’s sex parties’24 and aut aut ‘an either-or choice/situation’. It is notable that both are formed by reduplicating non-native material (foreign or learned); aut aut is a delocutive formation based on the Latin conjunction aut ‘either/or’. 7.4.1.2 Proper names A reduplicated structure is common in proper names of musical bands, fictional characters, companies, and associations. Of course, many of these names of foreign origin circulate in Italian (among the most frequent ones in the itTenTen16 corpus are Duran Duran, Candy Candy, Mau Mau, Tintin, Tom Tom), and this structure is also exploited for the formation of names of Italian popular music bands, associations, and companies. Frequently occurring examples are Dik Dik (name of an Italian band, from the common

22 23

24

The itTenTen16 (Italian Web 2016) corpus is part of the TenTen corpus family; it contains texts collected from the Internet for about 4.9 billion words. I add glosses only when necessary; most of these nouns exist also in English with the same meaning. I list just one spelling variant for each type, but the corpus and Italian dictionaries often attest several spelling variants: e.g., pon pon, pom pom and pompon ‘pompom’. The meaning of this expression given in the text originated in Italian, but the word is a loanword, attested in English first, in several different meanings. Ample information on the history of this expression can be found in the Wikipedia entry about it (the Italian version is at https://it.wikipedia.org/wiki/Bunga_bunga).

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

242

Anna M. Thornton

noun dikdik ‘small African antelope’); Liber liber, a non-profit-making organization whose purpose is making available free access to non-copyrighted books – the name is a word play exploiting two senses of Latin liber, which means ‘book’ as a noun and ‘free’ as an adjective; Miu Miu, a fashion brand founded by Miuccia Prada, whose hypocoristic nickname is Miu. All these examples seem to indicate that reduplication of an item whose phonological shape is not typically Italian25 is a strategy exploited to form brand names. 7.4.1.3 Deverbal nouns Italian has a number of nouns formed by reduplication of a verb stem. Representative examples are given in (11): (11)

a. leccalecca ‘lollipop’ b. fuggifuggi ‘stampede’, pigiapigia ‘lit. push push’, spingispingi ‘lit. push push’ c. arraffa arraffa ‘lit. snatch snatch’, magna magna ‘lit. eat eat’

While the noun in (11a) is probably the only case of a reduplicated VV noun designating a concrete object, and is possibly a calque from Spanish chupachupa (Thornton 2008: 3), the action nouns in (11b)–(11c) are representative of a moderately productive construction, whose origins and development have been thoroughly investigated by Thornton (2008). These action nouns are formed through the reduplication of a verbal base (which is homophonous to the morphomic verb stem used in compounding, and to the singular imperative form of the verb). They originate as lexicalization of reported speech acts, consisting of repeated imperatives. The first attestations of such speech acts, dating back to the fifteenth century, are found in contexts describing battles or stampedes, in which many people participate; the speech acts refer to orders uttered several times and/or by several speakers in such contexts.26 By the nineteenth century, the formation of VV reduplicative action nouns had become slightly productive, and nouns for which no previous attestation in

25

26

Italian native words do not end in stops or in unstressed /u/, and only function words or words that have undergone troncamento, a post-lexical process of final vowel deletion that applies only after nasals and liquids, can end in a liquid: so dik, miu, and liber are not possible native content words in Italian. According to Spitzer (1951–52: 16–17), ‘Les impératifs sont ceux que s’adressèrent (ou auraient pu s’adresser à eux-mêmes) les agents de l’action décrite’ [the imperatives are those that the agents of the described action addressed to each other (or could have addressed to themselves)]. An example of such a direct speech antecedent of fuggifuggi is in the Historia dell’Europa by Pierfrancesco Giambullari (1495–1555), in a passage which depicts a rather twisted linguistic situation during a battle: ‘alcuni de’ soldati di Ottone, che sapevano parlare franzese, cominciarono a gridare in quella lingua: “fuggi, fuggi”. Il che sentendo i Lotteringi, e pensando per la finta favella che ciò dicessero i compagni loro, subitamente volsero le spalle’ [some of Otto’s soldiers, who spoke French, started shouting in this language: ‘Run away, run away’. The Lotharingians, hearing this, and thinking, given the language, that the command had been uttered by their fellows, immediately turned away].

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

243

reported speech exists have come into usage. The twenty-some such nouns investigated by Thornton (2008) all require that the Agent of the action refer to a plurality of individuals; most of these nouns belong to two semantic frames, called by Thornton (2008) the Stampede frame (11b) and the Illegal appropriation frame ((11c); note that in magna magna the action of ‘eating’ is used metaphorically to refer to illicit appropriation of public funds). Besides, there is a phonological constraint on the verbal bases, that must be disyllabic.27 The action nouns in (11b)–(11c) are a clear example of how a grammatical construction that makes use of reduplication (in this case, a lexeme-formation rule, with specific phonological conditions on the input and a semantically coherent output) can develop from repetition in speech. 7.4.2

Adjectives

7.4.2.1 Intensification through reduplication Reduplication has long been recognized as one of the means of intensification of adjectives available in Italian; in Italian grammars, the phenomenon is usually discussed in sections devoted to the ‘Superlative’ (e.g., Serianni 1988: § V.74, Lepschy & Lepschy 1981: 103), although not all authors concede that reduplication is one of the ways to form a ‘superlative’.28 The question is complicated by the uncertain status of the notion ‘superlative’, and even of the very category ‘Degree’ as a grammatical category. As for Degree, Stump (2008) discusses the issue of whether it should be considered an inflectional or a derivational category, and concludes that there is no universally valid answer, but the issue must be established language by language. For Italian, although descriptive grammars usually (but not always: see Maiden & Robustelli 2000 for an exception) inertially discuss Degree (with the values ‘comparative’ and ‘superlative’) as a grammatical category of adjectives, specialized literature dispels with this myth. Colombo (2019) considers the existence of a category ‘Degree of Adjectives’ a ‘grammatical superstition’: in his view, the so-called comparative in Italian is a syntactic construction (formed by più ‘more’ + adjective), and the so-called absolute superlative formed with the suffix -issimo (or in some cases -errimo, on which see Thornton 2019) is derivational. Authors agree that the possibility of intensifying the degree in which a quality is possessed is virtually unlimited 27 28

Arraffa is the only apparent exception, which can be explained by Plenat’s principle of ‘extrametricality of initial vowels’ (see Thornton 2008: 214–15 for details). Sorrento (1951: 332) explicitly considers reduplicated adjectives, adverbs, and adverbial PPs as expressing an ‘absolute superlative’, but his notion of ‘superlative’ seems rather ample and more metaphorical than core-grammatical. Sorrento argues that all reduplicative processes in Italo-Romance express a kind of ‘superlative’ meaning, no matter what part of speech or kind of phrase the reduplicated entities belong to.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

244

Anna M. Thornton

(Serianni 1988: § V.56), and Italian has many ways of achieving this goal (see at least Rainer 1983 and Grandi 2017 for ample surveys): reduplication is only one among others, not a grammaticalized way of forming ‘the (absolute) superlative’ as a specific value of the grammatical category of Degree. Rainer (1983, following Sorrento 1951) distinguishes two types of reduplicated adjectives, which he calls ‘emphatic reduplication’ and ‘real reduplication’ (echte Reduplikation); the two would be distinguished by the fact that in the first type the two adjectives are separated by a pause, or a comma in writing, and the intonation is ‘emphatic’, while in the second type there is no pause or comma between the two adjectives and the intonation is neutral. Rainer himself observes, however, that diachronically the ‘real’ reduplication derives from the constructionalization of the emphatic one. Various authors consider even the ‘real’ reduplication of adjectives a syntactic process, as opposed to a morphological one (Grossmann & D’Achille 2016: 41, Grandi 2017: 60); however, if one does not assume a clear-cut boundary between morphology and syntax, the question loses importance. ‘Real’ adjective reduplication is certainly a productive construction to express a high degree of possession of a quality. Compared to other means of intensification, reduplication has some specific features. Rainer, based on a corpus of intensified adjectives in Italian literature and press (see Rainer 1983: I for details), considers it ‘ein affektisch“volkstümlichen” Intensivierungsverfahren’ [an affective ‘popular’ means of intensification]. Its domain is not limited to certain semantic classes of adjectives: Table 7.1 shows that the most frequent reduplicated adjectives in itTenTen16 are distributed over practically all of Dixon’s (2004) semantic types. It must be observed that some adjectives, mostly in the Human Propensity class, occur preferably in reduplicated form: this is certainly the case for quatto quatto ‘silent red’ and mogio mogio ‘sad red’, whose preference for reduplication is even recorded by dictionaries (e.g., NDM, s. vv.). Another relatively frequent reduplicated sequence is tomo tomo, often occurring in the sequence tomo tomo, cacchio cacchio, an expression made famous by the Neapolitan comic actor Totò (1898–1967); tomo tomo is lemmatized as such and said to occur only in reduplicated form by NDM;29 cacchio does not seem to exist in Italian as an adjective at all,30 so the relatively frequent tokens of the sequence tomo tomo, cacchio cacchio (also in reverse order) must be 29

30

Although this is not the case in Neapolitan, the Italo-Romance variety from which the adjective was borrowed into Italian; Neapolitan dictionaries list tomo, glossed as ‘serio, grave’ [serious, grave] (Andreoli 1966, s.v.), ‘strano, flemmatico, serio, compassato’ [strange, phlegmatic, serious, composed] (D’Ascoli 1993). The word has other meanings as a noun: it is a technical term in agriculture, ‘vine bud’, and is used euphemistically in place of cazzo, a vulgar noun designating the male organ. On the etymology of cacchio and the semantic relation between its various senses see Loporcaro (2017).

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

245

Table 7.1 Semantic types of Italian reduplicated adjectives Semantic type

Most frequent examples

DIMENSION

grande grande ‘big’, piccolo piccolo ‘small, little’, lungo lungo ‘long’, grosso grosso ‘big’, sottile sottile ‘thin’, piccino piccino ‘tiny’ nuovo nuovo ‘new’, fresco fresco ‘fresh’ bello bello ‘beautiful’, bravo bravo ‘good, capable’ , buono buono ‘good, well behaved’, brutto brutto ‘ugly’ nero nero ‘black’ stretto stretto ‘tight’, paro paro ‘precise, exact’, tondo tondo ‘round’, fitto fitto ‘thick’, fine fine ‘thin’, caldo caldo ‘warm, hot’, dolce dolce ‘sweet’, leggero leggero ‘light’ zitto zitto ‘silent’, mogio mogio ‘sad’, quatto quatto ‘silent’ veloce veloce ‘fast’, lento lento ‘slow’ facile facile ‘easy’, semplice semplice ‘simple’ uguale uguale ‘equal’ falso falso ‘fake’ tanti tanti ‘many’ lontano lontano ‘far’, vicini vicini ‘near’ –

AGE VALUE COLOUR PHYSICAL PROPERTY

HUMAN PROPENSITY SPEED DIFFICULTY SIMILARITY QUALIFICATION QUANTIFICATION POSITION CARDINAL NUMBERS

understood as more or less conscious quotations from Totò. This sequence also co-occurs with other reduplicated adjectives in the semantic field of ‘silent, calm’, and often in contexts in which one is said to do something silently with an intent of pursuing their own hidden agenda, or some ambitious goal: (12)

Come direbbe Totò: zitto zitto, tomo tomo, cacchio cacchio, Zapatero è riuscito ‘As Totò would say: zitto zitto [silent red], tomo tomo , cacchio cacchio, Zapatero succeeded’

So it seems that Italian has a wealth of means to express the senses ‘silent, calm’ through reduplicated adjectives, some of which do not even occur in non-reduplicated form. In some cases, an adjective is not exactly reduplicated but replicated in two slightly different forms: the base adjective is followed by another one derived from it. Frequent cases are the ones in (13): (13)

solo soletto solo soletto alone alone dim ‘all alone’ piccino picciò piccino picci- ò little little- ? ‘tiny little’

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

246

Anna M. Thornton

While soletto occurs even by itself,31 picciò occurs only after piccino, sometimes as the final member of a number of reiterations, as in the traditional fairy tale La donnina piccina piccina picciò ‘the woman.dim little (piccina) little (piccina) picciò’; the element -ò does not correspond to any evaluative suffix. The meaning of these sequences is intensifying as in other reduplications of adjectives, with maybe an even more affective nuance (De Santis 2011 maintains that piccino picciò is typical of baby talk). Colour terms can be reduplicated, as in occhi azzurri azzurri ‘lit. eyes blue red, that is, very blue eyes’; in this case, the reduplicative construction is used to indicate ‘the presence of a colour at the highest degree of brightness and saturation’ (Grossmann & D’Achille 2019: 69), so the function of intensification is still present, but the affective tinge is less pronounced. Even colour terms can occur in a sequence in which the unmodified adjective is followed by an elative derivative in ‑issimo, as in rosso ross-issimo ‘lit. red red-elative, i.e., very red’. Finally, I mention syndetic constructions in which an adjective is followed by the conjunction e ‘and’ and another adjective derived from the first one, as in (14) (De Santis 2011); all these examples have an intensifying value. (14)

vecchio e stravecchio old and extra old ‘really old, very old’ fritto e rifritto fried and refried ‘very old, stale’ (metaphorically speaking)

7.4.2.2 Meaning-changing reduplication of adjectives In some cases, a reduplicated adjective functions as an adverbial expression, and its meaning is not (any more) transparently related to the base adjective. A case in point is papale papale ‘said explicitly’ from papale ‘papal, concerning the Pope’. Papale is a relational adjective, and as such non-gradable, so papale papale is not a case of intensifying reduplication like the ones discussed in §7.4.2.1; its meaning is certainly not ‘very papal’. Such examples are best understood as cases of lexicalization from emphatic reduplications occurring in discourse; each example would deserve a diachronic investigation. 7.4.2.3 Adjective-forming reduplication There is an isolated case of NN reduplication that is used in adjectival function: terra terra ‘low quality, not complex; lit. earth earth’. This adjectival use probably developed from an

31

As in the first line of the poem In cortile ‘In the courtyard’ by Umberto Saba: In cortile quei due stavan soletti ‘in the courtyard those two were alone’. On the usage of solo soletto as a means of intensification, see Jaberg (1947).

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

247

adverbial use (for which NN reduplications are more commonly used, cf. §7.4.3.2), in contexts in which both adjectives and adverbs could occur, like (15a). As Sorrento (1951: 337) observes, this specific expression is found all over Italy, not only in the standard language but also in other Italo-Romance varieties: its meaning is roughly ‘skimming the earth surface’, also metaphorically. The metaphoric sense is the only one in standard Italian. Terra terra distributes like adjectives: it occurs both in attributive (15b) and predicative (15c) position, and can even be modified by an adverb like quasi ‘almost’ (15d): (15)

a. Tutti i segreti dell’orto spiegati ‘terra terra’ ‘All the secrets of the vegetable garden explained ‘terra terra’ [lit. earth red, i.e., in a simple way, with a pun because of the connection between ‘earth’ and growing vegetables] b. Prima argomentazione terra terra ‘First simple argument’ c. le loro omelie sono terra terra, non elevano più lo spirito. ‘their [priests’] homilies are low level, they don’t elevate your spirit anymore’ d. Detta in modo quasi terra terra, le lauree utili sono quelle che alla fine degli studi vi fanno avere uno stipendio ‘speaking in an almost too simple way, the useful degrees are the ones that will get you a job after you graduate’

7.4.3

Adverbs

Discussion of reduplicative constructions involving or forming adjectives has paved the way for the presentation of reduplicative constructions involving or forming adverbs. It is notoriously difficult to define and delimit the class of adverbs, and adjectives and adverbs have many features in common, even in languages, like Italian, where in most syntactic contexts it is easy to tell them apart (see Ramat & Ricca 1994). Indeed, some reduplicative constructions that apply to adjectives have a parallel with adverbs; others, however, are specific to adverbs. Besides, some reduplicated sequences can function both as adjectives and adverbs (e.g., così così ‘so-so’: compare quest’anno è andata così così ‘this year it went so-so’ and abbiamo passato momenti felici e momenti un po’ così così ‘we spent happy moments and moments that were a little so-so’). 7.4.3.1 Intensification through reduplication of adverbs Intensification by means of reduplication, already discussed with respect to adjectives, applies also to adverbs. Actually, Rainer (1983: 65) found that in his corpus of written Italian intensification through reduplication is more frequent with adverbs than with adjectives; besides, he observes that for some adverbs, like the ones in (16), the only means of intensification is reduplication: (16)

allora allora, almeno almeno, appena appena, forse forse, or ora, quasi quasi, subito subito

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

248

Anna M. Thornton lit. then RED ‘right then’, lit. at least RED ‘at the very least’, lit. a little RED ‘just a little’, lit. maybe RED ‘maybe maybe’, lit. now RED ‘right now’, lit. almost RED ‘almost’, lit. right now RED ‘right now’ (Rainer 1983: 64)

Adverbs can be divided in several functional and semantic categories: manner adverbs, degree adverbs, adverbs expressing spatial and temporal location, sentence adverbs, and focalizers or focus modifiers (Traugott 2006). All these categories are represented among the most frequent reduplicated adverbs in the itTenTen16 corpus: (17)

a. manner: piano piano ‘slowly red’, forte forte ‘loud red’, bene bene ‘well red’ b. degree: molto molto ‘very red’, tanto tanto ‘very red’, poco poco ‘a little red’, troppo troppo ‘too much red’, quasi quasi ‘almost red’, appena appena ‘just red’ c. spatial: sotto sotto ‘under, below red’, su su ‘up red’, giù giù ‘down red’, intorno intorno ‘around red’ d. temporal: mai mai ‘never red’, sempre sempre always red’, subito subito ‘rightnow red’ e. sentence adverbs: forse forse ‘maybe red’ f. focalizers and focus modifiers: proprio proprio ‘exactly red’, davvero davvero ‘really red’

An isolated case of reduplicated manner adverb is lemme lemme ‘very slowly’, where lemme is unattested outside of this collocation; the etymology is not clear; according to several dictionaries, lemme possibly proceeds from Latin sollemnem ‘solemn’, while Nobile and Lombardi Vallauri (2016: back cover) consider it an ideophone. 7.4.3.2 Adverb-forming reduplication In some cases adverbs are formed by reduplication of lexemes belonging to other parts of speech. Stolz et al. (2011: 120, fn. 117) observe that ‘adverbialization is indeed a very important function of T[otal]R[eduplication]’. The most significant sub-group is that of adverbs formed by reduplicating nouns. This type has been widely discussed in the literature with respect to Southern-ItaloRomance varieties, particularly Sicilian,32 where reduplication of nouns denoting geographical entities and features of a landscape or environment is used to refer to movement within or along the entity denoted by the reduplicated noun: (18)

32

a. firriari strati strati wander street street ‘wander around the streets’

See at least Sorrento (1951: 332–8), Amenta (2010).

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

249

b. Si nni iàva trenu trenu he/she went train train ‘wandered around inside the train’ c. facìanu bbaldòria dda caserma caserma they did party there barracks barracks ‘they partied all around the barracks’ d. u dutturi va paisi paisi the doctor goes village village ‘the doctor goes from village to village’

(data from Amenta 2010)

Standard Italian offers only a few examples, studied by Migliorini (1968). Migliorini pronounces himself in favour of a Greek origin of the construction, which from Southern Italo-Romance varieties in close contact with Greek spread to the North. Other scholars have considered the hypothesis of an Arabic rather than Greek origin of the type. According to Migliorini, three distinct kinds of meaning are attested in Italian for this construction, which he calls ‘“moto per luogo” o più esattamente [. . .] “moto rasente luogo”’ [‘movement through place’ or, more exactly, ‘movement along a place’],33 ‘modo di procedere’ [manner of moving ahead] and ‘distanza minima’ [minimal distance] (Migliorini 1968: 186). He also observes that this construction is related to another adverb-forming construction, where PPs with the preposition a ‘at, to’ governing a noun are reduplicated; often, the same noun is found in a purely reduplicative (asyndetic) construction and in the prepositional one, sometimes with dropping of the first preposition, as illustrated in (19) for (a) man(o) (a) mano: (19)

a. a mano a mano, lit. at hand at hand b. mano a mano, lit. hand at hand c. mano mano, man mano34 lit. hand hand all ‘gradually, step by step, as time goes by’

Migliorini (1968: 189) rightly observes that it is often difficult to establish which type appeared first – this was certainly true at the time in which

33

34

Total reduplications expressing movement along a line (a coastline, a wall, etc.) is attested in several languages of the Mediterranean area, including Greek (Kallergi 2015a) and Maltese (Stolz et al. 2011: 313). The semantics of this construction is often dubbed prolative, although, as shown by Haspelmath (2009: 515), this is only one of no less than eight different labels used in the literature to refer to (case values expressing) movement along a path. Man mano from mano mano is formed by applying the optional postlexical rule of truncation (Ital. troncamento), which allows to delete a word-final vowel when it is preceded by a nasal or liquid sonorant. Man mano is the most frequent variant, followed by the two variants including the preposition, and finally by untruncated mano mano (frequency counts from the la Repubblica 1985–2000 corpus (roughly 330 M tokens) are: man mano 2779, a mano a mano 553, mano a mano 501, mano mano 66).

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

250

Anna M. Thornton

Migliorini wrote, over half a century ago; availability of diachronic corpora would make the task easier nowadays. The adverbial reduplicated PPs (thoroughly investigated in several contributions by Piunno; see at least Piunno 2016) are more frequent than the adverbial NN reduplications; simple NN adverbial reduplications are attested mainly with nouns of high frequency. In (20) I list the most common NN adverbial reduplications, divided by rough ‘semantic type’: (20)

a. Spatial progression giro giro lit. circle circle ‘all around’, punto punto lit. point point ‘point by point’ b. Temporal progression mano mano lit. hand hand ‘gradually’, volta volta lit. time time ‘time after time’ c. Temporal progression expressed metaphorically as spatial progression passo passo lit. step step ‘step by step’, tratto tratto lit. segment segment ‘occasionally’, via via lit. way way ‘gradually’ d. Minimal distance pelo pelo lit. hair hair ‘very close, by a hair’s breadth’

Although the ‘spatial progression’ type is most likely the original meaning of the construction, the metaphoric extension to temporal progression seems to be more widespread. According to Benigni and Lo Baido (2020) the examples that indicate ‘minimal distance’ can have a ‘counter-avertive’ meaning, and are used to stress the fact that success was attained notwithstanding the fact that some obstacle was in the way. Indeed, pelo pelo often modifies verbs such as riuscire ‘succeed’, arrivare ‘arrive’ (just in time before something starts, or a train leaves, etc.), starci ‘fit’ and the like. Some examples of these adverbial NN reduplications are given in (21): (21)

a. Prendete poi il pompelmo rosa e tagliatelo a rondelle, buccia e tutto, e disponete le fette giro giro al vassoio ‘Then take the pink grapefruit and cut it into slices, peel and all, and arrange the slices all around the tray’ b. Per contestarla dettagliatamente, punto punto, [. . .] occorrerebbero anni. ‘To contest it in detail, point by point, would take years’. c. a chi osservi l’opera [. . .] sarà consentito seguire il maestro passo passo e di scoprire, volta volta, le meraviglie dell’immaginazione e dell’altrove ‘those who observe the work will be allowed to follow the master step by step and to discover, time after time, the wonders of the imagination and of elsewhere’ d. Mano mano ti verranno chiesti i dati che lo comporranno ‘Gradually you will be asked for the data that will make it up’ e. È pertanto opportuno porsi dei sotto-obiettivi e cercare via via di raggiungerli ‘It is therefore advisable to set sub-goals and gradually try to achieve them’ f. Arriverò pelo pelo ‘I’ll get there just in time’

A complex example related to the expression of a minimal measure is minimo minimo ‘at least’, where the adjective minimo ‘minimal’ has first been converted to a noun ‘minimum’ (probably by ellipsis of a nominal head numero ‘number’), which then developed the adverbial function when

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

251

reduplicated, probably from sequences such as sono come numero minimo tre > sono come minimo tre ‘they are at least three, lit. they are three as a minimum’ > sono minimo tre ‘id.’, where the preposition governing the noun has been dropped, > sono minimo minimo tre ‘id.’, with N reduplication for emphasis. This hypothesis is supported by the following considerations: to express the sense ‘at least’, the sequence come minimo ‘as a minimum’ is overwhelmingly more frequent than minimo minimo (with a ratio of 37.5 to 1 in the itTenTen16 corpus); minimo minimo occurs most frequently when it quantifies over something, for example, money, time, units of measure, as respectively in the examples in (22a)–(22c), but can then extend to more abstract situations, where no exact quantification is involved, and it is not even clear what the domain of quantification would be, as in the examples in (23). (22)

a. un botto di minimo minimo 20 milioni di dollari ‘a lot of minimo minimo 20 million dollars’ b. vuole impegnarsi minimo minimo 20 ore a settimana nel progetto ‘she wants to devote minimo minimo 20 hours a week to the project’ c. La Fiorentina vera pesa minimo minimo 1 kg. ‘the real Florentine steak weights minimo minimo 1 kg.’

(23)

a. domanda che prevede minimo minimo la sfera di cristallo ‘a question that requires minimo minimo a crystal ball’ b. Cioè minimo minimo le tiri una badilata in testa, no? ‘I mean, minimo minimo, you hit her on the head with a shovel, wouldn’t you?

Finally, it must be observed that *come minimo minimo is virtually unattested (three tokens in the ItTenTen 16 corpus, vs. hundreds of tokens of minimo minimo and thousands of come minimo). Examples of reduplicated adjectives forming an adverb are pari pari lit. equal red, and giusto giusto lit. right red, both meaning ‘exactly’, in contexts where reference is made to copying something exactly, or fitting exactly in a space, and the like.

7.4.4

Interjections

Reduplication or even triple iteration is also found in interjections. In particular, there are minimal pairs of single and reiterated Italian interjections whose segmental make-up is the same, but whose meaning is completely different. A good example is in (24): (24)

a. [ʔɛ] b. [ʔɛʔɛʔɛ]

While (24a) indicates assent, (24b) ‘vale da ammonimento scherzoso o da rimprovero bonario’ [is used to convey a joking admonition or a good-natured

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

252

Anna M. Thornton

reproach] (Villani 1986: 41). The two interjections, however, also differ in their prosody. 7.4.5

Contrastive Focus Reduplication

Contrastive Focus Reduplication (CFR) is the label adopted by Ghomeshi et al. (2004) for a phenomenon of total word reduplication that can be based on words belonging to several parts of speech, and, in English at least, can also apply to some kinds of phrases. Other labels adopted to denote this phenomenon by other scholars are répétition distinctive (Gougenheim 1935), Real-X reduplication (Stolz et al. 2011: 199, Freywald 2015). CFR ‘restricts the interpretation of the copied element to a ‘real’ or prototypical reading’ (Ghomeshi et al. 2004: 307), ‘narrowing down the meaning of a word to its core meaning’ (Freywald 2015: 916), or at least signalling that ‘one meaning of the word is being contrasted with other possible meanings’ (Ghomeshi et al. 2004: 317). Ghomeshi et al. (2004) mention the fact that CFR has been observed in other languages, including Italian, for which their source is Wierzbicka (2003 [1991]).35 Indeed, the phenomenon is attested in Italian, and has been described by Italian scholars writing in Italian long before it made its appearance in the international literature: it is discussed by Sorrento (1951), Medici (1959, 1961),36 Fabi (1964) and particularly Poggi Salani (1971); it is also considered by Rainer (1983).37 In Italian linguistics, this phenomenon is called il tipo caffè-caffè, lit. ‘the type coffee-coffee’; caffè caffè started to be used during the Second World War to refer to real coffee as opposed to surrogates, at a time when real coffee was scarce and rationed and surrogates were used. Various scholars have been at pains to backdate the attestation of this construction in Italian. Poggi Salani (1971) considers the phrase in (25a) as a diachronic and logical precursor of the construction, and the phrase in (25b), nowadays a frozen expression, as an early instance of it:

35 36

37

The chapter on Italian in Wierzbicka 2003 is a revised version of Wierzbicka 1986. Medici (1959) describes the phenomenon and its function in the following way: ‘raddoppiamento del sostantivo per affermare, per contrapposizione, la pienezza delle qualità inerenti al sostantivo stesso’ [doubling of the noun to affirm, by contrast, the fullness of the qualities inherent in the noun itself]. The same phenomenon had been detected in French by Gougenheim (1935: 346), who proposed to call it ‘répétition distinctive’ and characterized it as follows: ‘en répétant un substantif on insiste sur le fait qu’il s’agit bien de ce qui est véri[t]ablement désigné par ce mot pour le distinguer [. . .] d’autre chose qui par un abus du langage pourrait se parer de la même dénomination. La répétition équivaut alors aux épithètes «véritable» ou «authentique»’ [by repeating a noun we insist on the fact that we are indeed talking about what is truly designated by this word, to distinguish it [. . .] from something else which by an abuse of language could take the same denomination. The repetition is then equivalent to the epithets ‘genuine’ or ‘authentic’].

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian (25)

253

a. ai barri veri barri ‘to the buccheros real buccheros’38 b. alla fin fine ‘ultimately, in the end, after all’, lit. at the end end

I list in (26a)–(26c) a few examples: these are all nouns, denoting persons (26a), artistic genres or abstract concepts (26b), concrete objects, places or substances (26c); I have observed the possibility of using the constructions even with adjectives (26d). (26)

38

CFR in Italian a. Don Marino è un prete prete ‘Father Marino is a real priest’, lit. priest priest (Sorrento 1951: 332) donna donna lit. woman woman (Fabi 1964, citing a work by Nievo completed in 1858; Medici 1959, citing a 1952 newspaper article by Virginio Lilli) ingegnere ingegnere (vs. geometra) ‘a real engineer (who has a college degree) vs. a simple surveyor’ (Fabi 1964) b. vita vita lit. life life (Medici 1959, citing a 1951 newspaper article by Eugenio Montale) pittura pittura lit. painting painting (Medici 1959, citing a 1953 newspaper article by Leonardo Borgese) romanzo romanzo lit. novel novel (Medici 1959; example found in Oreste Del Buono, ‘L’inesauribile Simenon fabbrica un romanzo al mese’, Oggi 50, 15 December 1955; it is unclear whether the original expression was uttered by the interviewee, Georges Simenon (1903–1989), in French or in Italian) prosa prosa, teatro teatro lit. prose prose, theater theater (Fabi 1964) c. caffè caffè ‘authentic coffee (vs. a wartime surrogate)’ lit. coffee coffee (Medici 1959, backdating the expression to 1941–42) Non il violino violino, ma il violino zigano ‘not the real violin (lit. violin violin), but the gypsy violin’ Ma vino vino, non del vino sofisticato come questo ‘but peasant’s wine (lit. wine wine), not a fancy wine like this’ Un Toscano purissimo di Firenze Firenze ‘a super-pure Tuscan man from Florence city (lit. Florence Florence)’ (Rainer 1983) Tutte le Russie sono tre: la piccola Russia, cioè l’Ucraina, la Russia bianca, cioè la Bielorussia, e la Russia Russia, cioè la Russia ‘All the Russias are three: little Russia, that is Ukraine, white Russia, that is Belarus, and Russia Russia, that is, Russia’ (Paolo Nori, Sanguina ancora, Milan, Mondadori, 2021: 77)

Bucchero is a kind of Tuscan pottery.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

254

Anna M. Thornton d. ce li avete blu blu? ‘have you got them [socks] in dark blue (lit. blue blue)?’ [not in pale blue] (uttered by me in September 2021 talking to a street vendor carrying socks)

Clearly CFR exists in Italian; I suspect that it is not limited to nouns;39 a corpus-based study would be essential to determine the range of bases to which it can apply; it would also be extremely time consuming because of the need to examine all the contexts of reduplicated sequences manually to assess their semantic value. Glossing and translating the examples in (26) has proved not an easy task; this might be interpreted as evidence that this construction conveys a range of meanings showing the property of ‘descriptive ineffability’, considered by Potts (2007) as one of the hallmarks of expressive constructions.40 There has been much discussion in the literature on whether CFR should be unified with intensifying reduplication of the kind most often encountered with adjectives (§7.4.2.1). Wierzbicka (2003 [1991]) proposes a unifying ‘semantic formula’ for the two kinds of reduplication, which in her view share the speaker’s intention of using the reduplicated word in ‘strict correspondence’ with what is meant, ‘no exaggeration’; therefore there would be ‘no need to postulate a separate meaning’ for the reduplication of gradable adjectives: ‘the connotation of “high degree” is due to an implicature, calculable from the combined effect of the reduplication and the invariant meaning of the base’ (Wierzbicka 2003 [1991]: 264–5).

7.5

Repetition and reduplication in discourse

In this section I will focus on repetitions and reduplications that have clausal rather than word status. Clearly this is an area in which the distinction between grammatical and stylistic devices is most difficult. A commonly found proposal is to distinguish repetition, which in writing would be signalled by a comma and in speech by the (possibility of ) a pause between the two items, from reduplication, where no comma and no pause appear (see e.g., Wierzbicka 2003 [1991]: 255 for such a somewhat oversimplifying distinction). Stolz and Levkovych (2018) propose to distinguish ‘pragmatically motivated repetition from grammatically motivated reduplication’, thus

39

40

However, restrictions on which parts of speech can participate in CFR probably exist; Ghomeshi et al. (2004: 313) observe that functional items like AUX and DET cannot be subject to CFR because they ‘lack the appropriate sort of semantic variation’. Potts (2007: 166) defines descriptive ineffability as follows: ‘Speakers are never fully satisfied when they paraphrase expressive content using descriptive, i.e., nonexpressive, terms’.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

255

incurring in the problem of drawing the boundary between grammar and pragmatics, as already observed.41 Many studies observe that repetition is more typical of oral communication (Tannen 1987: 226) and is universally found both in natural speech and in literary texts that mimic speech. One variant of repetition happens when the same element is uttered repeatedly by different speakers, or by the same speaker but addressed to different addressees (Poggi Salani 1971: 69–70 discusses at length these cases with examples from the diachrony of Italian). One would be tempted to disregard these cases, were it not for the fact that even such cases have been shown to be able to give rise to grammaticalized constructions, such as the action nouns described in §7.4.1.3. A full treatment of stylistic devices exploiting repetition in Italian discourse exceeds by far the scope of the present contribution and the space available, so I will hereafter only discuss the phenomena which appear to me as most noteworthy. 7.5.1

Repetition and reduplication in discourse markers

Repetition is often used in speech acts which are variously denominated by different scholars, and that can perhaps be unified under the label of ‘discourse markers’; another useful characterization of the kind of expressions that often exploit reduplication is ‘Formulae of social exchange’ (Kaltenböck et al. 2011); Langacker (2008) calls a number of such formulae, including the ones used to exchange greetings, answering questions, and general markers of politeness, ‘expressives’. Bazzanella (1995: 231) observes that ‘immediate repetition’ of discourse markers is common particularly with agreement markers and focalizers. Among the most frequent ‘immediate repetitions’ of such expressions in the ItTenTen2016 corpus we find the ones in (27): (27)

a. agreement/disagreement markers: sì sì ‘yes’, no no ‘id’ b. greetings: ciao ciao ‘hello/bye’, baci baci ‘kisses’42 c. markers of politeness: grazie grazie ‘thank you’

7.5.2

Reduplicated imperatives as clauses

Spitzer (1918, 1951–52) has drawn attention to ‘metaphorical usages’ of reduplicated imperative verb forms in Romance languages, and particularly in Italian. He proposes to distinguish two sub-types, which he calls ‘historical’ or ‘descriptive’ and ‘gerundial’: the historical or descriptive type has temporal 41 42

Stolz & Levkovych (2018: 51, fn 20) admit that the boundaries between grammar and pragmatics are ‘fuzzy’. Baci (and baci baci) is a common formula of farewell in informal writing, such as e-mailmessages.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

256

Anna M. Thornton

meaning, and is used in narrative texts to report events that have occurred, while the gerundial one is used like an adverbial modifier, with hypothetical, concessive, causal or modal meaning. Thornton (2009) has studied these constructions in contemporary Italian, and has called into question the possibility of neatly distinguishing the two types as Spitzer proposed, since it can be demonstrated that some ‘gerundial’ usages historically derive from the grammaticalization of a ‘narrative’ (a term used by Pisani 1933: 247) usage. An early example of this type is the one in (28): (28)

Gira, rigira, non trovo né il padrone né la padrona ‘Go around, go around again, I cannot find my master nor my mistress’ (Carlo Goldoni, Gli amanti timidi, 1765)

The speaker who utters (28) is a servant, who is going around the house to look for his employers. Here the two imperatives are certainly ‘descriptive’ or ‘narrative’, since they are used to describe the servant’s going around the house. However, in this context, they can also be interpreted as having a concessive meaning: notwithstanding the fact that he goes around looking for them, the servant cannot find his employers. The double repetition of the imperative of girare ‘go around’ is attested in three variants: gira gira, gira e gira, gira e rigira. The syndetic variant with the second imperative from the prefixed verb rigirare ‘go around again’ is the most common (Thornton 2009) and it has become a fixed expression used in concessive constructions; in present-day Italian, the context in which these expressions are used need not involve a physical movement of going around, showing that the expression has been lexicalized to express concessivity: (29)

Gli argomenti, gira gira, sono pressappoco sempre gli stessi ‘the topics, gira gira, are more or less always the same’ (la Repubblica 1985–2000 corpus)

Already Spitzer (1951–52: 456) had observed that this expression (in all its variants) has specialized as antecedent to clauses that declare that the result of a process is not different from the initial state. Many more instances of reduplicated imperatives, instead, are used as antecedents of clauses that report actual results, conceived as a consequence of the action expressed by the verb. Thornton (2009) has identified a ‘hypothetical’ subgroup (30a) and a subgroup of types that function as ‘antecedents of a consecutive clause’ (30b): (30)

a. gratta gratta ‘lit. scratch scratch’ stringi stringi ‘lit. squeeze squeeze’ scava scava ‘lit. dig dig’ b. dai e dai ‘lit. give and give’ batti e ribatti ‘lit. hit and hit again’ cammina cammina ‘lit. walk walk’

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

257

These types differ from the concessive one because they are used as antecedents of clauses that report a result, not the lack thereof; they differ from each other in that the result is presented as hypothetical, conditioned by the action expressed by the reduplicated imperatives, after the types in (30a), and as actual (a consequence of the action expressed by the imperatives) after the types in (30b). In present-day Italian, most of these types are used with a purely metaphorical conventionalized meaning, completely severed from the verb’s basic meaning, which however in many cases can be shown to be at the origin of the lexicalization of a construction. For example, scava scava is first attested as a ‘narrative’ repetition (31a), but is nowadays used also in contexts in which the original meaning ‘dig’ is at best only metaphorically present (31b): (31)

a. Pinocchio restò a bocca aperta, e non volendo credere alle parole del Pappagallo, cominciò colle mani e colle unghie a scavare il terreno che aveva annaffiato. E scava, scava, scava, fece una buca così profonda, che ci sarebbe entrato per ritto un pagliaio: ma le monete non ci erano più. ‘Pinocchio remained speechless, and not wanting to believe the Parrot’s words, he began with his hands and nails to dig up the ground he had watered. And he digs, he digs, he digs, he made a hole so deep that a haystack would have gone straight in: but the coins were gone.’ (Carlo Collodi, Le avventure di Pinocchio, 1883; translation by translate.google.it)43 b. scava scava, sono proprio i meccanismi sociali a determinare il reale ‘digs digs, it is precisely the social mechanisms that determine reality’ (la Repubblica 1985–2000 corpus, translation by translate.google.it)

The English translations offered by translate.google.it do not recognize the imperative forms in (31), and take scava as a 3sg.prs.ind form. This is clearly wrong: in (31a) the whole narration is in the Past Absolute, not in the Present; in (31b) there is no 3sg subject. Scava scava (scava) is a sequence of forms homophonous with 2sg.imp, which is used to report a narrated fact in (31a) and to represent a condition in (31b): a more satisfactory translation for (31b) would be ‘if you look carefully enough, you’ll find that it is social mechanisms that determine reality’. This difficulty in translating the reduplicated imperative construction in another language suggests that the construction possesses the property of ineffability considered by Potts (2007) as typical of expressives. Thornton (2009) analyses these reduplicative constructions as converbs, that is, ‘nonfinite verbal form[s] whose main function is to mark adverbial subordination’ (Haspelmath 1995: 3), and proposes that the reduplicated element is not any more a 2sg.imp form, but a verb stem homophonous with it. Under this

43

De Santis (2014) has studied this construction with particular attention to how it is used in Collodi’s Pinocchio.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

258

Anna M. Thornton

analysis, this would be a further case of grammaticalization of a reduplicative construction originating from repetition in discourse. 7.5.3

Repetition as expression of frequency of occurrence of entities and events

Sornicola (1981: 225–7) has drawn attention to the frequent occurrence in speech of multiple repetitions (at least three tokens, possibly more) of nouns, expressing a speaker’s feeling of boredom or annoyance at the repeated occurrence in their experience of objects, events or discourse topics. Her examples are given in (32):44 (32)

a. Non fanno altro che parlare di / almeno per quello che io seguo / canzoni / canzoni / canzoni / e poi si parla /. . . / poi si parla sempre Napoli / Napoli / Napoli ‘They keep speaking of / at least for what I follow / songs / songs / songs / and then they talk / . . . / then they always talk Naples / Naples / Naples’ b. Sempre repliche repliche repliche repliche repliche repliche ‘Always re-runs re-runs re-runs re-runs re-runs re-runs’

Similar examples are easy to find also in more recent corpora of spoken Italian: (33)

poi ancora pioggia pioggia pioggia ‘then again rain rain rain’ (corpus PAISÀ)

Poggi Salani (1971: 70) reports such constructions also in written texts, both ancient and modern, both literary (34a) and journalistic (34b): (34)

a. e non si vede se non acqua, acqua, acqua ‘and one can see only water, water, water’ (Niccolò Machiavelli, Mandragola, 1524) b. Le città ammalate: Catanzaro. Una volontà univoca: uffici uffici uffici ‘the sick cities: Catanzaro. A univocal will: offices, offices, offices’ (Title in the daily newspaper Il giorno, 1 May 1971)

All the examples in (32)–(34) do indeed convey a feeling of boredom or annoyance by the speaker or writer. Boredom, annoyance, and contempt have been reported as speakers’ attitudes expressed through reduplicative constructions or repetition in discourse also in other languages (e.g., Modern Greek, Kallergi & Konstantinidou 2018, English, Fischer 2011). However, it is possible that these attitudes are not a necessary part of the meaning of these constructions in Italian; they might just be an implicature of the most frequent

44

Sornicola uses a semi-phonetic transcription, and uses / to indicate intonational breaks; I have transcribed the examples in conventional Italian orthography for ease of reading.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

259

contexts in which the device is used, which is cancellable in the appropriate context, as shown by the examples in (35), from customers’ reviews of seaside resorts (contained in the itTenTen16 corpus), in which the speakers’ attitude is happiness and satisfaction rather than boredom and annoyance: (35)

a. Ottimo albergo e ottimo posto dove andare per spendere (come abbiamo fatto noi) 2 settimane di relax, ottima cuci[n]a e mare mare mare ‘Very good hotel and great place to go to spend (as we did) 2 weeks of relax, great food and sea sea sea’’ b. il Coral bech [sic] è in una posizione magnifica [. . .] dal balcone del ristorante si gode di una vista mozzafiato, da qualsiasi parte ci si giri mare mare mare mare . . . ‘the Coral beach resort is in a magnificent position [. . .] from the balcony of the restaurant you can enjoy a breath-taking view, wherever you turn sea sea sea sea’

7.5.4

‘Framing’ reduplication

Various scholars (Sornicola 1981: 226, Sabatini 2011 [1985]: 125–6, D’Achille 2005: 241–4) have drawn attention to a construction in which a verb and its clitics (if there are any) which occur clause initially are repeated clause-finally. This construction is variously called: struttura a cornice ‘frame structure’, frase foderata ‘sentence in a dust jacket’, epanalepsis (though Mortara Garavelli 1994 argues that it should rather be called epanadiplosis). It has been observed that this construction is particularly common in romanesco (the vernacular spoken in Rome), but it is attested also in colloquial and substandard varieties of Italian, and in comedies. In (36) I list some examples cited in the literature: (36)

a. Mi voglio adirar, mi voglio ‘I want to get angry, I want’ (Pietro Aretino, La cortigiana, early sixteenth century, cited by D’Achille 2005: 242) b. Non me fa’ venì il nervoso. . . no’ me fa ‘Don’t make me nervous, don’t make me’ (Diego Fabbri, La bugiarda, 1956, cited by D’Achille 2005: 243) c. Saranno pure stati dei falsi saranno ‘they could well have been fakes, they could’ (spoken corpus, Sornicola 1981: 226) d. So’ stata all’opera so’ stata ‘I’ve been to the opera, I’ve been’ (spoken corpus, Sornicola 1981: 226)

A comprehensive study of the usage of this construction outside of romanesco is a desideratum. It is unclear what syntactic constraints hold on the reduplication of material: Sabatini characterizes it as repetition of a whole predicate, or just a modal verb and its clitics; D’Achille concurs, and adds that in periphrastic verb forms only the auxiliary is repeated; however, even the few examples in (36) show that a negation can be included in the repetition (36b),

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

260

Anna M. Thornton

and that a whole periphrastic form can be repeated (36d; however, while repetition of the auxiliary only in (36d) would indeed be fully grammatical, omitting the negation in (36b) is inconceivable; Ugolini (1983: 74) lists numerous examples in seventeenth-century romanesco in which an initial negation is included in the repeated sequence). Even the semantics of the construction is debated: while D’Achille (2005) maintains that, besides being more common in certain local varieties and in low registers, it can be considered an ‘idiosyncratic’ characteristic of certain speakers, devoid of any specific expressive value, Sabatini (2011 [1985]: 126) proposes that this kind of repetition ‘ribadisce la verità di quanto asserito dal verbo, e in questo modo esprime un atteggiamento di sicurezza di giudizio del parlante’ [reaffirms the truth of what is asserted by the verb, thereby expressing the speaker’s attitude of confidence in their judgement]. Di Giovannantonio (2023), studying the occurrence of the construction in seventeenth-century romanesco, surmises that one of the functions of the construction is to express a speaker’s attitude of anger or frustration, which also seems an appropriate characterization for the examples in (36a)–(36b). 7.6

Discussion and conclusion

In the previous sections I have attempted to give a description (as complete as possible within the space limits of a book chapter) of reduplicative elements and constructions in Italian. The survey has shown that Italian makes use of various sorts of reduplicative processes, targeting elements which range from single segments (§7.3.1) to words belonging to various parts of speech (§7.4, §7.5.2, §7.5.3) to sentence fragments (§7.5.4) and even complete utterances (§7.5.1). The picture that emerges is quite familiar: many of the reduplicative constructions found in Italian have parallels in other languages, in particular languages of the Mediterranean area, such as Greek (Kallergi 2015a) or Maltese (Stolz et al. 2011). In this regard, it must be stressed how certain types of reduplicative lexical items or constructions appear to have originated in Italo-Romance varieties spoken in regions bordering the Mediterranean (e.g., the lexicalized expression tomo tomo cacchio cacchio in Neapolitan, the adverbial reduplication discussed in §7.4.3.2 in Sicilian).45 The abundance and variety of phenomena surveyed is already a result: as often observed also for other languages, reduplicative expressions are rarely or not at all thematized in descriptive grammars of Italian, and a book-length 45

This is consonant with the findings by Stolz et al. (2011) that Neapolitan is the Romance language displaying the highest token frequency of total reduplications in their parallel corpus of translations of Le petit prince.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

261

study of reduplication in this language (at the level of that conducted by Kallergi 2015a for Greek) is a desideratum. Our survey confirms once again, after Stolz et al. (2011), that Italian does make use of total reduplication, contrary to what was claimed by Rubino (2013) on the basis of a sketchy four-page long description of the language. The meanings expressed by Italian reduplications are also similar to those found in other languages. Meanings connected to ‘iconicity of quantity’ (Stolz 2018) prevail: while system–adequacy pressure prevents the exploitation of reduplication to express nominal plurality, intensification of adjectives and adverbs by reduplication is common to the point of being recognized even by descriptive grammars as a grammaticalized way of expressing the elative (or ‘superlative’) degree, and verb reduplications expressing repeated or continuous action, although not very abundant in terms of type-frequency, are well entrenched in the language and have given rise to a variety of grammaticalized or lexicalized constructions, such as the action nominals described in §7.4.1.3 and the various verb reduplications originating as repeated imperatives which have become converbs specialized to express hypothetical, concessive or causal meanings (§7.5.2). A number of reduplicative constructions with adverbial meanings, expressing manner of motion, temporal progression, or minimal distance (§7.4.3.2) are less immediately recognizable as iconic, but the iconic foundation of these types of meanings has been argued for by several scholars. Contrastive Focus Reduplication is also well attested – indeed, as observed in §7.4.5, this phenomenon was identified in Italian (and French) long before attention to its existence in English was drawn by Ghomeshi et al. (2004). Repetition in discourse to convey frequent, numerous or repeated occurrence of items and events is also common (§7.5.3), and can convey speakers’ attitudes that range from boredom and annoyance to bliss, depending on the speakers’ assessment of the desirability or lack thereof of the items or events repeatedly mentioned. Conspicuously absent from Italian, instead, is echoword reduplication of the table-shmable kind. Many of the lexicalized or grammaticalized reduplicative constructions of Italian are canonical according to the criteria of canonicity proposed by Stolz (2018) and Stolz and Levkovych (2018: 32), since they consist of two immediately adjacent words, identical in form and meaning, but forming a construction whose meaning is different from that of each single constituent. Often the reduplication produces a change of part of speech (verb ! noun: fuggifuggi, noun ! adjective or adverb: terra terra, man mano, etc.). Since the present contribution appears in a volume devoted to expressivity, it is now necessary to evaluate which of the Italian reduplications I have described have expressive value. For some authors, the very fact of using a repetition or a reduplicative construction is expressive by definition (De Santis 2014, Grandi 2017: 60), although a precise definition of ‘expressivity’ is not

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

262

Anna M. Thornton

always provided by those who use this label. As aptly pointed out by Kallergi (2015b: 875–7) among others, there is no unique universally agreed-upon definition of expressivity, and the notion is used to include elements which range from the expression of sensory reality (ideophones, onomatopoeic lexemes) to the expression of non-propositional meaning and of speakers’ feelings; this last sense is captured by Jakobson’s (1960) ‘emotive’ function, centred on the Speaker; the same function was called Ausdruck (literally ‘expression’) by Bühler (1982). Italian certainly has ideophones and onomatopoeic lexemes that express sensory reality and have a reduplicative structure (§7.2), but these are not usually considered reduplicative constructions, since the unreduplicated elements do not usually exist independently (a condition for the recognition of reduplication posited by many scholars, amongst whom is Moravcsik 1978). Most of the grammaticalized constructions and lexicalized items formed by reduplication of existing lexemes do not appear to be particularly expressive, unless one is prepared to grant expressive value to any construction which involves iconicity of quantity at its origins. It has been shown in the course of the chapter that even completely grammaticalized reduplicative constructions of Italian, such as the action nouns of the fuggifuggi type or the hypothetical, causal, and concessive converbs based on reduplicated verb stems originated from the repetition of ‘descriptive imperatives’ in discourse. It is this repetition in discourse which often took place in contexts in which speakers were in an emotional state, like during fights and battles (fn 26), or writers wanted to convey some emotional state of their characters (like Pinocchio’s disappointment at the discovery that his money was gone, described in (31)). But once these discourse repetitions have grammaticalized into a lexeme-formation rule or a way of expressing various converbal meanings, the expressive charge of the contexts in which the reduplicated items are used is lowered or even absent (see e.g., example (31b)). So not all reduplicative constructions in Italian can be considered to express a high level of emotional involvement by the speaker. A higher level of speaker’s emotion, instead, is found in repetitions such as the ones described in §7.5.3, which are used to express speakers’ extreme psychological states, both negative (boredom: Examples (32)–(34)) and positive (satisfaction, delight: Examples (35a)–(35b)), and possibly also in the ones described in §7.5.4. So it seems that when structures that originate as expressive repetitions in discourse grammaticalize or lexicalize, they lose at least part of their expressivity. At present, the structures discussed in §7.5.3 and §7.5.4, which are non-canonical reduplications in many respects according to Stolz’s criteria (there are more than two repetitions in §7.5.3, the repeated elements can be more than a single word and are by definition not adjacent in §7.5.4), appear to have a higher expressive charge than most of the grammaticalized or lexicalized more canonical reduplications.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

263

A final point that deserves discussion is whether it is possible to unify the analysis of at least some of the constructions which have been described separately so far. Wierzbicka (2003 [1991]) strongly argues in favour of a unifying analysis of Contrastive Focus Reduplication of nouns (§7.4.5) and intensifying reduplication of adjectives and adverbs (§7.4.2.1 and §7.4.3.1). She proposes that in both cases the reduplication has the same function (or ‘illocutionary force’): ‘By repeating a word (‘XX’) the speaker draws attention to that word, and insists on its strict correspondence with what is meant (‘I mean X, not something a little different from X’)’ (Wierzbicka 2003: 262). The interpretation of reduplicated adjectives as expressing ‘high degree’ would then be due to an implicature, resulting from the combined effect of the reduplication and the meaning of a gradable base; both constructions could be described by the formula ‘no exaggeration’ (Wierzbicka 2003: 265). I remain agnostic with respect to this analysis, which in any case cannot be extended to all kinds of grammaticalized reduplications in Italian; since Contrastive Focus Reduplication of nouns also occurs in languages which do not have intensifying reduplication of adjectives (such as English), there does not seem to be a necessary link between the two constructions, apart from the fact that they both make use of reduplication to express a specific kind of meaning. Besides, according to Wierzbicka’s analysis, AA reduplication has an ‘emotional component’ which Contrastive Focus Reduplication lacks, so the two structures are not completely unifiable even in her analysis. In conclusion, Italian uses repetitions in discourse, and has used various sorts of repetitions all along its history: some of these repetitive constructions have grammaticalized, giving rise to a lexeme formation rule forming deverbal action nouns, a construction expressing high degree for adjectives and adverbs, and constructions expressing various adverbial and converbal meanings; these constructions are closer to morphology than to syntax, creating new lexemes or periphrastic structures considered by some as periphrastic inflectional forms of lexemes (the ‘superlative’ or ‘elative’, converbs). Other repetitive structures (those described in §7.5.3 and §7.5.4) are rather in the domain of syntax, but appear to have been already constructionalized, and convey an additional speaker’s oriented meaning. As Hopper (1987: 147–8; 150) observed: ‘no principled line can be drawn between the emergent regularities designated to be “grammatical” and other regularities deemed to be “rhetorical”, “formulaic”, etc. [. . .] the more useful a construction is, the more it will tend to become structuralized’. References Amenta, Luisa. (2010). La reduplicazione sintattica in siciliano. Bollettino – Centro di studi filologici e linguistici siciliani, 22, 345–58. Andreoli, Raffaele. (1966). Vocabolario napoletano – italiano. Naples: Arturo Berisio Editore.

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

264

Anna M. Thornton

Aronoff, Mark. (1994). Morphology by Itself: Stems and inflectional classes. Cambridge, MA: MIT Press. Bazzanella, Carla. (1995). I segnali discorsivi. In Lorenzo Renzi, Giampaolo Salvi, & Anna Cardinaletti (eds.) Grande grammatica italiana di consultazione. III. Tipi di frasi, deissi, formazione delle parole. Bologna: il Mulino, 225–57. Benigni, Valentina & Lo Baido, Maria Cristina. (2020). La reduplicazione nella codifica della maniera. Testi e linguaggi 14, 151–79. Bernini, Giuliano. (2010). baby talk. In Raffele Simone (ed.) Enciclopedia dell’italiano. Vol. I. Rome: Istituto della Enciclopedia Italiana, 139–40. https://bit.ly/3n3dpLU Bühler, Karl. (1982). Sprachtheorie. Die Darstellungsfunktion der Sprache. Stuttgart: Gustav Fischer Verlag. Castellani Pollidori, Ornella. (2002). Per la storia del detto Le gambe fanno giacomo giacomo. In L’Accademia della Crusca per Giovanni Nencioni. Florence: Le Lettere, 333–56. Colombo, Adriano. (2019). Superstizioni grammaticali. Italiano a Scuola, 1(1), 91–104. DOI: 10.6092/issn.2704-8128/9998 D’Achille, Paolo. (2005). Sintassi e fraseologia dell’italiano contemporaneo tra diacronia e diatopia. In Klaus Hölker & Christiane Maaß (eds.) Aspetti dell’italiano parlato. Münster: Lit, 235–49. D’Ascoli, Francesco. (1993). Nuovo vocabolario dialettale napoletano. Naples: Adriano Gallina Editore. De Santis, Cristiana. (2011). reduplicazione espressiva. In Raffele Simone (ed.) Enciclopedia dell’italiano, Vol. II. Rome: Istituto della Enciclopedia Italiana, 1224–5. https://bit.ly/3ELIklN (2014). Cresci, cresci, cresci . . . La reduplicazione espressiva come strumento di espressione di relazioni transfrastiche. In Cristiana De Santis, Angela Ferrari, Gianluca Frenguelli, Francesca Gatta, Letizia Lala, Marco Mazzoleni, & Michele Prandi (eds.) Le relazioni logico-sintattiche: teoria, sincronia, diacronia. Ariccia: Aracne, 191–219. Di Giovannantonio, Sara. (2023). «La vuoi finì la vuoi?»: la ripetizione nella sintassi del romanesco pre-belliano. In Davide Mastrantonio, Valentina Bianchi, Marianna Marrucci, Orlando Paris, Ibraam Abdelsayed, & Martina Bellinzona (eds.) Repetita iuvant, perseverare diabolicum: un approccio multidisciplinare alla ripetizione. Siena: Università per Stranieri, 177–85. Dingemanse, Mark. (2015). Ideophones and reduplication. Depiction, description, and the interpretation of repeated talk in discourse. Studies in Language, 39(4), 946–70. DOI: 10.1075/sl.39.4.05din Dixon, R. M. W. (2004). Adjective classes in typological perspective. In R. M. W. Dixon, & Alexandra Y. Aikhenvald (eds.) Adjective Classes: A cross-linguistic typology. Oxford: Oxford University Press, 1–49. Evans, Nicholas. (2007). Insubordination and its uses. In Irina Nikolaeva (ed.) Finiteness: Theoretical and empirical foundations. New York: Oxford University Press, 366–431. Fabi, Angelo. (1964). Ancora del tipo «caffè caffè». Lingua nostra 25(1), 18. Finkbeiner, Rita. (2018). Bla(h), bla(h), bla(h): Usage and meaning of a repetitive allrounder. In Aina Urdze (ed.) Non-prototypical reduplication. Berlin & Boston: De Gruyter Mouton, 71–89. DOI: 10.1515/9783110599329-003

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

265

Fischer, Olga. (2011). Cognitive iconic grounding of reduplication in language. In Pascal Michelucci, Olga Fischer, & Christina Ljungberg (eds.) Semblance and Signification. Amsterdam and Philadelphia: John Benjamins, 55–82. DOI: 10.1075/ill.10.04fis Freywald, Ulrike. (2015). Total Reduplication as a productive process in German. Studies in Language 39(4), 905–45. DOI: 10.1075/sl.39.4.06fre GDLI = Battaglia, Salvatore (ed.) (1961–2002) Grande Dizionario della Lingua Italiana, Turin: UTET. Gil, David. (2005). From repetition to reduplication in Riau Indonesian. In Bernard Hurch (ed.) Studies on Reduplication. Berlin and New York: Mouton de Gruyter, 31–64. Ghomeshi, Jila, Jackendoff, Ray, Rosen, Nicole, & Russell, Kevin. (2004). Contrastive Focus Reduplication in English (The Salad-salad paper). Natural Language & Linguistic Theory 22: 307–57. DOI: 10.1023/B:NALA.0000015789.98638.f9 Gougenheim, Georges. (1935). La répétition distinctive. Le Français moderne 3(4), 345–6. Grandi, Nicola. (2017). Intensification processes in Italian: A Survey. In Maria Napoli & Miriam Ravetto (eds.) Exploring Intensification: Synchronic, diachronic and cross-linguistic perspectives. Amsterdam and Philadelphia: John Benjamins, 55–77. DOI: 10.1075/slcs.189.04gra Grossmann, Maria & D’Achille, Paolo. (2016). Italian colour terms in the BLUE area: Synchrony and diachrony. In João Paulo Silvestre, Esperança Cardeira, & Alina Villalva (eds.) Colour and Colour Naming: Crosslinguistic approaches. Lisbon: Centro de Linguística da Universidade de Lisboa/Universidade de Aveiro, 21–50. (2019). Compound color terms in Italian. In Ida Raffaelli, Daniela Katunar & Barbara Kerovec (eds.) Lexicalization Patterns in Color Naming: A crosslinguistic perspective. Amsterdam and Philadelphia: John Benjamins, 61–79. DOI: 10.1075/sfsl.78.04gro Haspelmath, Martin. (1995). The converb as a cross-linguistically valid category. In Martin Haspelmath & Ekkehard König (eds.) Converbs in Cross-Linguistic Perspective. Berlin and New York: Mouton de Gruyter, 1–55. DOI: 10.1515/ 9783110884463-003 (2009). Terminology of case. In Andrej Malchukov & Andrew Spencer (eds.) The Oxford Handbook of Case. Oxford: Oxford University Press, 505–17. DOI: 10.1093/oxfordhb/9780199206476.013.0034 Hoffmann, Thomas & Trousdale, Graeme (eds.) (2013). The Oxford Handbook of Construction Grammar. Oxford: Oxford University Press. DOI: 10.1093/ oxfordhb/9780195396683.001.0001 Hopper, Paul. (1987). Emergent grammar. Proceedings of the Thirteenth Annual Meeting of the Berkeley Linguistics Society. 139–57. DOI: 10.3765/bls.v13i0.1834 Jaberg, Karl. (1947). Elation und Komparation. In Festschrift für Edouard Tièche zum 70. Geburtstage am 21 März 1947. Berne: H. Lang, 41–60. Jakobson, Roman. (1960). Closing statement: Linguistics and poetics. In Thomas Sebeok (ed.) Style in Language. Cambridge, MA: MIT Press, 350–77. Kallergi, Haritini. (2015a). Reduplication at the word level: The Greek facts in typological perspective. Berlin and New York: Mouton de Gruyter. DOI: 10.1515/ 9783110365597

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

266

Anna M. Thornton

(2015b). Total Reduplication as a category of expressives.(Counter)evidence from Modern Greek. Studies in Language 39(4), 873–904. DOI: 10.1075/sl.39.4.04kal Kallergi, Haritini & Konstantinidou, Magdalene. (2018). Reduplicative constructions involving distortion. In Aina Urdze (ed.) Non-Prototypical Reduplication. Berlin and Boston: De Gruyter Mouton, 91–150. DOI: 10.1515/9783110599329-004 Kaltenböck, Gunther, Heine, Bernd, & Kuteva, Tania. (2011). On thetical grammar. Studies in Language 35(4), 852–897. DOI: 10.1075/sl.35.4.03kal Kim, Yuni. (2012). Review of Stolz et al. (2011). Studies in Language 36(2), 440–8. DOI: 10.1075/sl.36.2.08kim Langacker, Ronald W. (2008). Cognitive Grammar: A basic introduction. Oxford: Oxford University Press. Lepschy, Giulio & Lepschy, Anna Laura. (1981). La lingua italiana: storia, varietà dell’uso, grammatica. Milan: Bompiani. Loporcaro, Michele. (2017). Cacchio! Una nuova etimologia. In Annette Gerstenberg, Judith Kittler, Luca Lorenzetti, & Giancarlo Schirru (eds.) Romanice loqui. Festschrift für Gerald Bernhard zu seinem 60. Geburtstag. Tübingen: Stauffenburg, 321–31. Maiden, Martin & Robustelli, Cecilia. (2000). A Reference Grammar of Modern Italian. London: Arnold. Mauri, Caterina & Masini, Francesca. (2022). Diversity, discourse, diachrony: A converging evidence methodology for grammar emergence. In Miriam Voghera (ed.) From Speaking to Grammar. Berne: Peter Lang, 101–50. Medici, Mario. (1959). Il tipo «caffè caffè». Lingua nostra, 20(3), 84. (1961). Ancora di «caffè caffè». Lingua nostra, 22(1), 28. Migliorini, Bruno. (1968). Il tipo sintattico «camminare riva riva». In Cesare Segre (ed.) Linguistica e filologia. Omaggio a Benvenuto Terracini. Milan: Il Saggiatore di Alberto Mondadori Editore, 183–90. Mioni, Alberto M. (1990). Fece splash e, glu glu, affondò. L’ideofono come parte del discorso. In Monica Berretta, Piera Molinelli, & Ada Valentini (eds.) Parallela 4. Morfologia/Morphologie. Tübingen: Narr, 255–67. Moravcsik, Edith A. (1978). Reduplicative constructions. In Joseph H. Greenberg (ed.) Universals of Human Language. Vol. 3: Word structure. Stanford, CA: Stanford University Press, 297–334. Mortara Garavelli, Bice. (1994). Epanadiplosi. In Gian Luigi Beccaria (ed.) Dizionario di linguistica e di filologia, metrica, retorica. Turin: Einaudi. NDM = Il Nuovo De Mauro. https://dizionario.internazionale.it Nobile, Luca & Lombardi Vallauri, Edoardo. (2016). Onomatopea e fonosimbolismo. Rome: Carocci. Pisani, Vittore. (1933). Pānini, Māgha e l’imperativo descrittivo. Rendiconti della _ dei Lincei. Classe di scienze morali, storiche e R. Accademia nazionale filologiche, VI(9), 246–67. Piunno, Valentina. (2016). Multiword modifiers in some Romance languages: Semantic formats and syntactic templates. Yearbook of Phraseology, 7, 3–34. DOI: 10.1515/ phras-2016-0002 Poggi Salani, Teresa. (1971). Il tipo caffè caffè. Lingua nostra, 32(3), 67–74. Potts, Christopher. (2007). The expressive dimension. Theoretical Linguistics, 33(2), 165–98. DOI 10.1515/TL.2007.011

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

Repetition and reduplication in Italian

267

Rainer, Franz. (1983). Intensivierung im Italienischen. Salzburg: Institut für Romanistik der Universität Salzburg. Ramat, Paolo & Ricca, Davide. (1994). Prototypical adverbs: On the scalarity/radiality of the notion of ADVERB. Rivista di Linguistica, 6(2), 289–326. Rubino, Carl. (2005). Reduplication: Form, function and distribution. In Bernard Hurch (ed.) Studies on Reduplication. Berlin and New York: Mouton de Gruyter, 11–29. (2013). Reduplication. In Matthew S. Dryer & Martin Haspelmath (eds.) The World Atlas of Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology. https://wals.info/chapter/27 Sabatini, Francesco. (2011) [1985]. «I popolari discorsi svolti nella mia poesia». Sintassi del parlato nei Sonetti di Belli. In Francesco Sabatini, L’italiano nel mondo moderno, Vol. I. Naples: Liguori, 123–48. Serianni, Luca. (1988). Grammatica italiana. Italiano comune e lingua letteraria. Suoni forme costrutti. Turin: Utet. Shintel, Hadas, Nusbaum, Howard C., & Okrent, Arika. (2006). Analog acoustic expression in speech communication. Journal of Memory and Language, 55(2), 167–77. DOI: 10.1016/j.jml.2006.03.002 Sornicola, Rosanna. (1981). Sul parlato. Bologna: il Mulino. Sorrento, Luigi. (1951). Sintassi romanza. Ricerche e prospettive. II edizione. Varese and Milan: Istituto Editoriale Cisalpino. Spitzer, Leo. (1918). Über den Imperativ im Romanischen. In Leo Spitzer, Aufsätze zur romanischen Syntax und Stilistik. Halle: Niemeyer, 181–231. (1951–52). Sur quelques emplois métaphoriques de l’impératif. Un chapitre de syntaxe comparative. Romania, 72, 433–78 and 73, 16–63. Stolz, Thomas. (2018). (Non-)Canonical Reduplication. In Aina Urdze (ed.) NonPrototypical Reduplication. Berlin and Boston: De Gruyter Mouton, 201–77. DOI: 10.1515/9783110599329-007 Stolz, Thomas & Levkovych, Nataliya. (2018). Function vs. form – On ways of telling repetition and reduplication apart. In Rita Finkbeiner & Ulrike Freywald (eds.) Exact Repetition in Grammar and Discourse. Berlin and Boston: De Gruyter Mouton, 29–66. DOI: 10.1515/9783110592498-002 Stolz, Thomas, Stroh, Cornelia, & Urdze, Ania. (2011). Total Reduplication: The areal linguistics of a potential universal. Berlin: Akademie Verlag. Stolz, Thomas, Urdze, Aina, Nintemann, Julia, & Tsareva, Marina. (2015). When some dots turn a different color . . . Thoughts on how (not) to determine whether or not reduplication is universal. Studies in Language, 39(4), 795–834. DOI: 10.1075/ sl.39.4.07sto Stump, Gregory T. (2008). Comparison. In A. D. Cruse, F. Hundsnurscher, M. Job, & P. Lutzeier (eds.) Lexikologie: Ein internationales Handbuch zur Natur und Struktur von Wörtern und Wortschätzen, 2. Halbband. Berlin and Boston: De Gruyter Mouton, 1655–7. DOI: 10.1515/9783110171471.2.36.1655 Tannen, Deborah. (1987). Repetition in conversation as spontaneous formulaicity. Text – Interdisciplinary Journal for the Study of Discourse, 7(3), 215–43. DOI: 10 .1515/text.1.1987.7.3.215 Thornton Anna M. (1996). On some phenomena of prosodic morphology in Italian: accorciamenti, hypocoristics and prosodic delimitation. Probus 8(1), 81–112. DOI: 10.1515/prbs.1996.8.1.81

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

268

Anna M. Thornton

Thornton, Anna M. (2008). Italian Verb-Verb reduplicative Action Nouns. Lingue e linguaggio 7(2), 209–32. DOI: 10.1418/28096 (2009). Imperativi raddoppiati nell’italiano contemporaneo: un tipo di converbi. In Angela Ferrari (ed.) Sintassi storica e sincronica dell’italiano. Subordinazione, coordinazione, giustapposizione, Vol. II. Florence: Cesati, 1189–206. (2015). Paradigmatically determined allomorphy: The ‘participial stem’ from Latin to Italian. In Peter O. Müller, Ingeborg Ohnheiser, Susan Olsen, & Franz Rainer (eds.) Word-Formation: An international handbook of the languages of Europe, Vol. 1. Berlin and Boston: De Gruyter Mouton, 780–802. (2019). Sulle forme in -errimo nell’italiano contemporaneo. Studi di grammatica italiana 38, 301–32. Traugott, Elisabeth Closs. (2006). The semantic development of scalar focus modifiers. In Ans van Kemenade & Bettelou Los (eds.) The Handbook of the History of English. Malden, MA: Blackwell, 335–59. DOI: 10.1002/9780470757048.ch14 Ugolini, Francesco. (1983). Per la storia del dialetto di Roma nel Cinquecento. I Romani alla Minerva, un’improbabile ‘madonna Iacovella’ e un pronostico di un conclavista. Contributi di dialettologia umbra, 3(1), 3–99. Villani, Paola. (1986). Note teoriche per lo studio dei fonosimboli. Linguaggi, 3 (1–2), 32–44. Wierzbicka, Anna. (1986). Italian reduplication: Cross-cultural pragmatics and illocutionary semantics. Linguistics, 24(2), 287–315. DOI: 10.1515/ling.1986.24.2 .287 (2003 [1991]). Cross-Cultural Pragmatics: The semantics of human interaction. Berlin and New York: Mouton de Gruyter.

Corpora itTenTen16 (Italian Web 2016) www.sketchengine.eu/ittenten-italian-corpus la Repubblica 1985–2000 https://bit.ly/3JP1HNQ PAISÀ www.corpusitaliano.it

https://doi.org/10.1017/9781108989084.011 Published online by Cambridge University Press

8

Analysing expressives in a spoken corpus of Majorcan Catalan Nicolau Dols*

8.1

Materials and methods

The materials under analysis here are a collection of twelve folk tales, randomly selected for inclusion from among thirty-five in the Corpus Oral de la Llengua Catalana, an oral corpus by Institut d’Estudis Catalans, the official academy of the Catalan language.1 Initially collected and composed in written form by Antoni M. Alcover (1862–1932), they have been repeatedly reprinted since initial publication in 1880, when they rapidly became a classic that replaced local oral story-telling (Guiscafrè 2007). The recordings by Francesc de B. Moll in the late 1950s follow the texts collected and arranged by Alcover. The titles of the folk tales included in the sub-corpus are as they appear in the Grimalt and Guiscafrè (Alcover 1996–2022) edition of Alcover’s Rondaies mallorquines (Majorcan folk tales), and are given with their ATU type (AarneThompson-Uther classification of folktales, Uther 2004): (1) ‘En Joanet de sa gerra’ (ATU 555. The Fisherman and his Wife2) (2) ‘En Salom i es batle’ (ATU 1525A. Stealing the Count’s Horse, Sheet, and Parson) (3) ‘Es metge Guinyot’ (ATU 1641B. Physician in Spite of Himself ) (4) ‘Es pou de sa lluna’ (ATU 301A. Quest for a Vanished Princess) (5) ‘Es reim del rei moro amb set pams de morro’ (ATU 425E. Enchanted Husband Sings Lullaby) (6) ‘Es tres mantells d’or’ (ATU 400. The Quest for a Lost Bride) (7) ‘La reina banyuda’ (ATU 566. Fortunatus) (8) ‘Sa coeta de na Marieta’ (ATU 310. Rapunzel) (9) ‘Sa llampria meravellosa’ (ATU 561. Aladdin)

* University of the Balearic Islands and the Institute for Catalan Studies, Barcelona 1 This research has been funded by the Institut d’Estudis Catalans, project PRO2023-S04-DOLS. 2 This and the following titles in English are not glosses of the original Catalan titles, but those recorded in the ATU index.

269

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

270

Nicolau Dols

(10) ‘S’aucellet’ (ATU 235. The Jay Borrows the Cuckoo’s Skin) (11) ‘Una madona que enganà el dimoni’ (ATU 1175. Straightening a Curly Hair) (12) ‘Un pare i quatre fills’ (ATU 653. The Skillful Brothers) Five hours and thirteen minutes of voice recordings with neither music nor sound effects have been acoustically analysed using Praat, a computer program for acoustic analysis (Boersma & Weenink 2022) and a series of Praat scripts to extract values for pitch, intensity, and speech rate over the whole of the subcorpus, previously transcribed on TextGrid tiers. All recordings have undergone a process of segmentation automatically conducted by a boundary-insertion script modelled on Lennes (2002). The script discards empty segments (minimum duration = 0.25 s; maximum intensity = 46 dB) and places boundaries leaving a lapse of 0.1s between each boundary and the commencing/ending sound in order to ensure a homogeneous counting of time (which is especially important for speech-rate measurements). The whole subcorpus has yielded 4,918 fragments, which, after segmentation, have all been analysed for intensity, pitch (maximum and minimum), and speech rate (number of words per second). 8.2

Iconic elements under analysis

Alongside the iconicity of phonetic features of speech such as pitch contrasts, intensity, or speech rate, iconicity can be detected in the phrasing of the message. It has already been suggested that iconicity prefers narratives (Dingemanse 2012: 664). The oral corpus analysed here is no exception to this tendency. I concentrate on several resources, mainly based on iteration (either repetition or reduplication). Non-lexicalised imitations of nature by voice and exclamations will also be examined, with examples of each type following. Not only lexical ideophones, but also morphological and syntactic constructions will appear, in which iteration adds a specific meaning to the sentence. R1: Reduplicative ideophones Nyam-nyam repetition (mastication) Zas-zas repetition (climbing movements) Mèu-mèu repetition (miaowing) Toc-toc repetition (knocking) Tutup intensity (strike) Rac-a-rac repetition (biting [a fruit]) Xiu-xiu intensity (swift flying) Bitlo-bitlo continuity (speed)

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

Expressives in a spoken corpus of Majorcan Catalan

271

R2: Partial reduplication (reduplicative patterns with alternating vowels) Tris-tras continuity (walking) (eight times, always duplicated: ‘tristras tris-tras’) Ti-tu-ti repetition (flute playing) Xelim-i-xelam repetition (to struggle) The formula ‘X va, X ve’ (‘X goes, Y comes’), as in ‘fitorada va, fitorada ve’ (‘stab’), ‘bocinada va, bocinada ve’ (‘bite’), ‘pedra va, pedra ve’ (‘stone’), ‘destralada va, destralada ve’ (‘ax blow’). R3: Repetition of the same word or phrase (syntactic repetition) Demana qui demana (‘asks who asks’) repetition (to ask) Per amunt, per amunt! (‘up and up’) intensity i tots estirades d’aquí i estirades d’aquí deçà (‘and they all pull (noun) from here and pull (noun) from there’) repetition (pull) i ja va esser partit a espipellar cireres i cireres (‘and it started biting cherries and cherries’) quantity vos he mester forçat forçat (‘I need you so much’) intensity I heu de pensar i creure i creure i pensar (‘You must think and believe, and believe and think’) intensity (think and believe). This is a traditional formula to link passages of a tale. It is highly conventionalised, that is, not in use out of this genre. i ja el me tornaren tenir més adormit que un socot de figuera de moro i allà dorm qui dorm (‘and there they had him again, sleeping beside the stump of a big prickly pear, and sleep who sleeps there’) continuity En Joanet s’acabussa a sa pomera que tenia més a prop i daça qui daça pomes i més pomes (‘Joanet rushes to the nearest apple tree and there takes who takes apples and more apples’) repetition R4: Repetition with alternating morphemes (morphological base repetition) Camina caminaràs continuity (walking) ara és hora de metges ni metjons! (‘now is the time for doctors! (ironic)’) quantity/diversity (no care at all) Que consulta ni consultenons! (‘No consult at all’) quantity/diversity (no care at all) senyorum, senyoram i senyorim (‘gentlemen’) quantity/diversity criadum, criadam i criadim (‘servants’) quantity/diversity que gegants ni gegantes! quantity/diversity (no care at all) Ara és hora de cosetes ni cosetons! (‘now is the time for details!’ (ironic)) quantity/diversity (no care at all) tot es viudim, viudam i viudum (‘all widows’) quantity/diversity

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

272

Nicolau Dols

tot aquell peixam, peixim i peixum (‘all those fish’) quantity/ diversity cap casal ni casull (‘no house at all’) quantity/diversity s’asseu ben assegut (‘he sits well sat’) quality (confirmation, emphasis) se donava per pagat i redepagat (‘he considered himself paid enough’) quantity (confirmation, emphasis) xiu-xi-bi-biu! re-xiu-xiu! intensity (swift flying) Most of the cases listed above deal with notions similar to verbal aspect (as with repetition or continuity) or to general quantity, diversity or intensity; in any case, let us say ‘degree’. I1: Interjections, exclamation, emphasis Ai! Hahahaha! Ah! Zas! (sudden appearance) Tot-toc Plaf! Tutup (dull knock) Oh, això sí que no! (‘Oh, not at all!) No res, idò! (‘Ok, then!’) Hala! (‘Come on!’) Sí, que la veig! (‘Yes, I can see it!’) Això és un metge! (‘What a doctor he is!’) Ja ho crec que sí! (‘Of course, I will!’) Sí fa! (‘Of course’) I2: Direct sound imitation mmmm . . . hesitation xiu-xiu frying oil ti-tu-ti flute [whistling] whistling [screams] fighting

8.3

Research questions

Two questions now arise: i. What is the role of acoustic traces such as pitch, intensity, and speed (speech rate) in terms of expressivity? ii. Can any correlation be found between acoustic cues and linguistically based expressives (lexical items, iteration)?

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

Expressives in a spoken corpus of Majorcan Catalan

8.3.1

273

Acoustic traces as expressives

8.3.1.1 Pitch After examination of the most outstanding fragments in terms of pitch (the 10 per cent of fragments with the highest pitch, the 10 per cent with the lowest pitch, and the 10 per cent of the greatest differences between high and low pitches within the same fragments) mean-pitch deviation appears mainly in direct imitation of nature and in the emotional featuring of characters (the way in which actors depict the emotions of the characters they play by using voice inflections). Several of the most salient differences in pitch appear in tales involving animals, (9) and (10), whereas in one of them (10) those differences are due to whistling to call animals, which cannot therefore be analysed as speech sounds. In the other case, the sound imitation of a little bird’s escape makes use of speech sounds: flight, swiftness, and height are mirrored by the vowel with the highest Formant 2 (F2 )3 ([i]) and with sibilant [ʃ] (see §8.4.1 on sibilants): El se posa an es cap, pega volada per amunt com un estel, i per amunt! per amunt!, i xiu-xiu-xi-xiu! xi-xi-bi-biu! ‘It puts it on its head and starts flying like a kite, up and up! and [ʃiwʃiwʃiˈʃiw] [ʃiʃibiˈbiw]’, the last two words being only imitative sounds. Interestingly enough, the rise in pitch also affects the immediately previous ‘i per amunt! per amunt!’, which is an instance of iteration – the most widely used expressive mechanism in the whole corpus. In the lowest ranks of pitch measurements, bass voice appears also in imitations. Such is the case of a scene where a group of devils appear making noise. It must be said that this fragment of Tale (11) is included among the ten fragments with the highest differences in pitch, in this case between 561.78 Hz and 47.87 Hz, in this way transmitting a sense of trouble or disorder. In a more level bass voice, that is to say bass with smaller pitch differences, we find the voice of the Moor King in Tale (5), a telling of Beauty and Beast, where the Beast is a shadow that turns into human form only once a month. For this character a voice is provided as in the samples found at the bottom of the list for maximum pitch (seven samples out of the ten ending the list of 4,918): com aquestes que hi ha aquí dins que te serveixen (‘as the ones in here who serve you’, highest pitch = 100.57 Hz, pitch difference = 41.73Hz) si t’acostes a mi i me toques

3

Formant 2 (F2) is the second concentration of acoustic energy in a spectrogram, counting vertically from the bottom to the top, and is responsible for the position of the tightest constriction in the mouth during the production of a sound: the higher F2, the more advanced the tongue position.

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

274

Nicolau Dols

(‘if you approach me and touch me’, highest pitch = 100.29 Hz, pitch difference = 35.80 Hz) Jo som un encanteri (‘I am a spell’, highest pitch = 96.35 Hz, pitch difference = 9.61 Hz) Escolta bé (‘Listen carefully’, highest pitch = 95.93 Hz, pitch difference = 22.34 Hz) No som més que una ombra (‘I am nothing more than a shadow’, highest pitch = 95.49 Hz, pitch difference = 15.05 Hz) no esperis més que aquesta hora cada trenta dies (‘do not expect more than this hour every thirty days’, highest pitch = 90.19 Hz, pitch difference = 19.43 Hz) Catalineta, lo que ara te diré (‘[. . .] Catalineta, what I am going to tell you’ [continuing after ‘Listen carefully’ cited above], highest pitch = 86.44 Hz, pitch difference = 35.14 Hz) The character is depicted as sombre by means of low and level pitch. Especially interesting among the samples cited are cases such as ‘si t’acostes a mi i em toques’, and the self-defining ‘Jo som un encanteri’ and ‘No som més que un ombra’, where nasals add their dullness to the overall descriptive effect (see Aryani et al. 2015, Albers 2008, Auracher et al. 2010, Wiseman & van Peer 2003 on perception of nasals). In general, differences in pitch do not seem to point unambiguously to any given expressive in particular: no blind search for expressives can be directly guided by pitch values, whereas some type of relation can be observed, as will be shown below, between the most extreme cases of pitch (highest and lowest) and pitch difference values, and expressives. As in the examples above, other instances of pitch deviation from the average can be interpreted in terms of emotional featuring. High pitch can relate to joy, but also to anger, and low pitch to gravity. These are examples extracted from the 10 per cent of samples with the highest maximum pitch and the lowest minimum pitch (the average maximum pitch of the corpus = 361.40 Hz; average minimum pitch in the corpus = 100.58). I have purposely chosen examples with a lexical content easily interpretable as portraying the same intention attributable to the acoustic factor under analysis: Joy – Per la senyora reina quan tendrà fruit de bendició! Sí, sííí. (‘For the queen, when she will be blessed with a fruit of her womb! Yes, yeeees.’) (Maximum pitch = 633.91 Hz).

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

Expressives in a spoken corpus of Majorcan Catalan

275

Sa pobra viuda agafa es set infantons i s’agenolla amb ells davant en Joanet per besar-li es peus, donant-li milions de gràcies amb un plors desfet. [Plors Plors] (‘The poor widow takes her seven children and kneels down in front of Joanet to kiss his feet while thanking him while crying loudly. [Cries Cries]’) (Max. pitch = 630.51 Hz). Podeu fer comptes quina alegria no degué tenir na Catalineta i si el se degué cuidar a menjar de besades, an aquell filló des seu cor. No sabia què li passava. (‘You can imagine Catalineta’s joy and the way she covered with kisses that beloved son of hers. She could not understand what was happening to her’) (Max. pitch = 625.35 Hz). Anger se posa a cridar, fet un dimoni, dient: – Ah gran polissona que me fuigs amb aqueix polissó! (‘He starts shouting like a devil, saying : – Ah you scoundrel fleeing with that other scoundrel’) (Max. pitch = 630.15 Hz). i ja s’hi aborda com una lleona, plorant i cridant: – Polissó! Gran polissó! Ara pagaràs totes ses que has fetes! Venen a menar-te’n de part del rei . . .! (‘And she runs to him like a lioness, crying and shouting: – Scoundrel! You big scoundrel! You’ll pay now for all you’ve done! Now they come to arrest you by order of the king . . .!’) (Max. pitch = 625.35 Hz). Agafau-lo’m entre tots i fregiu-lo’m dins sa pella, i el me menjaré! (‘You catch him and fry him in a pan, and I will eat him up!’) (Max. pitch = 602.01 Hz). Gravity que no vull fer hereu es major (‘I don’t want to name my eldest son the heir’) (Minimum pitch = 44.59 Hz). Arriba a les portes del cel i toc-toc (‘He arrives at Heaven’s doors and knock knock’) (Min. pitch = 48.04 Hz). El rei se’n tem i ho prengué molt tort (‘The king realises and gets deeply disappointed’) (Min. pitch = 49.87 Hz). Although the values listed above can be easily associated with emotions, the relation between pitch and emotions is not constant. However, even if a direct

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

276

Nicolau Dols

correlation between pitch deviation and iconic marking cannot be fully proven by our data (See §8.3.2), some tendencies of pitch average deviation can be initially related to specific expressives. 8.3.1.2 Intensity Leaving aside the outsets and endings of stories, where differences in intensity can be caused by the beginning and ending of recording, the most striking differences in this dimension appear as expected when indirect style turns into direct, but also when the narrating voice addresses the listener: Ara figurau-vos si li degué tornar venir de nou, a na Catalineta, tot allò ‘Now imagine how Catalineta was amazed by all those things’ (Intensity difference: 69.80 dB) (Tale 5, fragment 801) Idò heu de creure i pensar i pensar i creure ‘And now you must think and believe, and believe and think’ [a traditional linking formula in folk tales] (Intensity difference: 69.51 dB) (Tale 3, fragment 8) Heu de creure i pensar que com na Catalineta s’hagué acabat aquell remarro ‘You must think and believe that when Catalineta had eaten up that huge grape’ (Intensity difference: 69.24 dB) (Tale 5, fragment 347)

If we turn now to the highest intensity peaks throughout the corpus, it can be seen that intensity is used for quite different purposes, as in the ten fragments including the highest peaks: i hala un raig d’aigua des set pous pes cap de sa serp! I un altre raig ben afavorit pes cap des lleó! Llavò etziba pes cap des dimoni cucarell tota sa que li quedava dins sa redoma! ‘And [interjection] another jet of water from the seven wells onto the snake’s head! And yet another thick jet on the lion’s head! Then he throws on to the head of the devil the rest of the water in the flask’ (90.00 dB) (Tale 4, fragment 622) dispara i va ferir cada ou just en es cap fent-hi un foradí ‘he shot and hit each egg right on the top, making a hole in it.’ (89.79 dB) (Tale 12, fragment 526) Un dematí hi van i no troben cap flor a s’arbre ‘One morning they go there and find no flowers on the tree’ (89.65 dB) (Tale 7, fragment 24) O Barrufet! ‘Oh, Barrufet [a comic name for a devil]’ (89.27 dB) (Tale 2, fragment 520)

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

Expressives in a spoken corpus of Majorcan Catalan

277

Fan un gran dinar i un ball ben vitenc per celebrar s’alegria grossa que hi havia en es castell. I, des cap de tres dies de festa, la princesa amb en Bernadet i aquell minyó, se despedeixen de tota la família, se posen damunt aquell mantell d’or de la princesa, que diu: – Val Déu i can Bernadet! ‘They give a great and lively ball to celebrate the great joy in the castle. And, after three days of celebration, the princess with Bernadet and that child, say goodbye to the whole family, step on her golden mantle, and she says: “Good God to Bernadet’s!” [a magic formula to direct the magic mantle, in this case to Bernadet’s home]’ (89.25 dB) (Tale 6, fragment 440) que dugué son pare a fer tot lo que ella volia ‘what led her father to do everything she wanted’ (89.19 dB) (Tale 6, fragment 392) Prenen sa fua a l’aire i amb un instant foren a can Bernadet ‘They sped into the air and suddenly arrived at Bernadet’s’ (89.10 dB) (Tale 6, fragment 442) Ell tot d’una s’aborda an es tres mantells; però just arribà a un, perquè dues fadrinetes d’aquelles n’agafen un perhom i ja no les veren pus ‘He suddenly rushed to the three mantles; but he just got to one, because two of those little girls grabbed one of them before vanishing’ (89.04 dB) (Tale 6, fragment 112) I es bous?! ‘And the bulls?!’ (88.99 dB) (Tale 2, fragment 76) – Barrufet, dic! – Què dimonis tens, tant de cridar? – Puja de pressa! – Deixa’m pegar un parell de tionades abans! ‘Barrufet [the Devil’s name]! – What the hell makes you shout that way? – Come up quickly! – Let me throw some logs before!’ (88.99 dB) (Tale 2, fragment 522)

As we can see, the highest levels of intensity are found in a quite diverse range of fragments: narration of swift actions (but of plain events too), surprise, vocative, and exclamations. 8.3.1.3 Speech rate Speech rate is calculated by dividing the number of words in a fragment by its duration in seconds. A high speech rate represents fast speech, that is, many words in a given time. There does not seem to be a clear link between high speech rates and meaning, apart from a sense of excitement during arguments. Clitics attached to a main word and graphically separated by a dash or an apostrophe are counted as different items throughout the corpus. Among the ten samples with the highest speech rate, six reflect direct speech, and all ten are sentence-final or whole sentences:

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

278

Nicolau Dols

I qui l’ha d’haver fet metge, an es meu homo? ‘And who must have made my husband into a doctor?’ (6.88 w s) (Tale 3, fragment 429) – I què m’és a mi si se’n tem? ‘– And what if he realises?’ (6.54 w s) (Tale 8, fragment 976) a veure si n’heu fet cap! ‘if you have done any!’ (6.38 w s) (Tale 5, fragment 114) I no he fet més que lo que vostè m’ha dit. ‘And I have not done anything but follow your orders.’ (6.52 w s) (Tale 2, fragment 152) tot d’una t’hi du . . .? ‘does it take you there at once . . .?’ (6.25 w s) (Tale 7, fragment 436) i ja li va haver estret cap a ca son pare. ‘And he fled towards his father’s’ (6.05 w s) (Tale 12, fragment 308) i sa corrent el se’n va dur riu avall. ‘and the current took him downstream’ (5.95 w s) (Tale 8, fragment 1304) A la fi en troba un que li entrà per s’ull dret. ‘Finally he finds a suitable one’ (5.93 w s) (Tale 12, fragment, 274) I ningú el veia a ell. ‘And no one could see him’ (5.90 w s) (Tale 9, fragment 1182) i sé ben cert que no ho he fet a posta, d’arrabasar-lo-t’ho! ‘And I know for sure he has pulled it off from you on purpose’ (5.89 w s) (Tale 8, fragment 1176)

The lowest places in the list are for vocatives, where a name is called to be heard from far: three instances of ‘Marieta!’ at 0.52 w s, 0.584 w s and 0.586 w s from Tale 8 close the list. The tendency is clear in this part of the list: one- or twoword fragments (almost all of them vocatives) are pronounced at a strikingly low speed, resulting from vowel lengthening, as when shouting a name. The average for the last twenty places is of 1.14 words and 0.85 w s. It should be recalled that pauses defining fragments are automatically located by the script applied. Duration is therefore calculated on a regular basis, as shown in Table 8.1.

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

Expressives in a spoken corpus of Majorcan Catalan

279

Table 8.1 Lowest places for speech rate Words

Seconds

Speech rate (w s)

– Marieta! – Marieta! – Marieta! [whistling] – Ah! Sobretot, ‘Above all’ M’heu redoblat s’encantament! ‘You’ve doubled the spell on me! [thunder]’ Joanet, – Sobretot, ‘Above all’ – Mmmm . . . Amen. Mentrestant ‘Meanwhile’ Selleta, ‘little saddle,’ Barrufet! Marieta! Ma nina! ‘Marieta! My girl!’

1.896 1.712 1.704 1.416 1.368 1.248 5.792

0.52742616 0.58411215 0.58685446 0.706214689 0.730994152 0.801282051 0.863259669

1.152 1.152 1.128 1.121 1.104 1.104 1.088 3.264

0.868055556 0.868055556 0.886524823 0.891437944 0.905797101 0.905797101 0.919117647 0.919117647

The seventh fragment in Table 8.1 (‘M’heu redoblat s’encantament!’) can be discarded: even if words are pronounced at quite a high speed, they are followed by a sound effect (one of the very few in the corpus), adding to the total duration of the fragment. The rest are vocatives, interjections, and linking adverbs. The speech rate alone, even if the emic content of sentences were not understood, might help the listener to understand if an argument is happening or a name is being called. We cannot claim a specific meaning is attached to this feature, but only a general intention. 8.3.2

At the intersection between acoustic cues and expressives

After checking the interpreting possibilities of three phonetic features (pitch, intensity, and speed), we now turn to the second question asked at the beginning of the analysis: is there any sort of correlation between specific expressive resources as those pointed out in §8.2? They are listed here for the sake of clarity: R1 Reduplicative ideophones R2 Partial reduplication (reduplicative patterns with alternating vowels)

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

280

Nicolau Dols

Table 8.2 Pitch difference in fragments containing expressives Pitch difference Average = 260.90294 Hz

Above average

Below average

Total

R1 R2 R3 R4 I1 I2

33 9 75 29 69 19

31 3 45 11 67 3

64 12 120 40 136 22

Table 8.3 Maximum intensity in fragments containing expressives Maximum intensity in every fragment Average = 83.793419 dB

Above average

Below average

Total

R1 R2 R3 R4 I1 I2

50 11 94 34 84 20

14 1 26 6 52 2

64 12 120 40 136 22

R3 Repetition of the same word or phrase (syntactic repetition) R4 Repetition with alternating morphemes (morphological base repetition) I1 Interjections, exclamation, emphasis I2 Direct sound imitation After counting each of the items in every class and calculating pitch difference (change of pitch within a fragment) (Table 8.2), maximum and minimum intensity (Tables 8.3 and 8.4), and the speech rate of all 4,918 fragments of the corpus (Table 8.5), the average for each feature has been calculated, as well as the number of fragments showing a given expressive resource above and below the average quantity. In order to detect the mostP N extreme cases, we calculate the average absolute deviation (AAD = Dm ¼ N1 jx1  xj) for each feature and then we add it to i¼1the number of items above this point. We then the average value, and count subtract the AAD from the average value and count the number of items below this point (see Tables 8.6–8.9). Inspection of Tables 8.6–8.9, reveals that of all factors analysed, the difference in pitch within a fragment seems to be the most relevant for

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

Expressives in a spoken corpus of Majorcan Catalan

281

Table 8.4 Minimum intensity in fragments containing expressives Minimum intensity in every fragment Average = 27.681966 dB

Above average

Below average

Total

R1 R2 R3 R4 I1 I2

36 8 62 20 74 14

28 4 58 20 62 8

64 12 120 40 136 22

Table 8.5 Speech rate in fragments containing expressives Speech rate Average = 3.3339458 w s

Above average

Below average

Total

R1 R2 R3 R4 I1 I2

26 5 65 10 35 6

38 7 55 30 101 16

64 12 120 40 136 22

Table 8.6 Extreme cases of pitch difference, above and below average plus/ minus absolute deviation Pitch difference (AAD = 136.10447)

Above average +AAD (>397.004581 Hz)

Below average – AAD (8.997927 dB)

Below average – AAD (31.372012 dB) 15 6 23 10 25 2

Below average – AAD (3.9662795 w s)

Below average – AAD ( *trah- > *trā- > Latin trans-, Sanskrit tirati, tarati, ‘he crosses over’. Basque trinkulin-trinkulin ‘staggering, tottering, reeling’, draka-draka-draka ‘horse galloping’, traka-traka, ‘trot’, ‘trotting’ (Ibarretxe-Antuñano 2017: 204). 8.4.2

Grammatical aspects

It is not uncommon to state that a typical feature of ideophones is their marginality in the sentence. Dingemanse and Akita (2017: 506) propose the concept of grammatical integration, based on position, optionality, and embedding in morphosyntactic structure. Other proposals seem to adhere to the idea that ideophones cannot play the role of a syntactic constituent if they cannot take affixes ensuring morphosyntactic processes such as, for instance, agreement: ... in his recent work on ideophones in Chichewa, Kumeleka argues that the ideophone in languages like Chichewa are not verbs because, whereas verbs require concordial agreement with the nominals with which they pattern, ideophones do not (1992). (Moshi 1993: 187)

A similar position can be read in Lee (2017), although on different empirical grounds, after collecting cases of inflected ideophones: ‘[. . .] some ideophones are verbal. That means that they can be affixed with an actor voice marker.’ (Lee 2017: 188). There is no doubt that inflection matters as evidence of grammatical integration in the sense of Dingemanse and Akita (2017) just mentioned. However, inflection cannot be the only proof of verbal (or nominal) behaviour of an ideophone. In this respect, Ibarretxe-Antuñano (2017) takes advantage of Nuckolls’s (2006: 40) pragmatic interpretation of verbal function and analyses the following example in the same way: Upel-a oso-oso-rik atzaparr-etan artu barrel- full-fullpaw-LOC take. ABS PART PRF

ta zanga, zanga! ustu and zanga zanga empty. PRF

arte Until

He grabbed the barrel with his paws and zanga, zanga! until it was empty’ (OEH) [. . .]. The ideophone not only describes the drinking event in a very expressive and vivid way, but also takes the place of a possible verb ‘to drink’. (Ibarretxe-Antuñano 2017: 206)

Similar cases, that is, of an ideophone pragmatically functioning as a verb, can be found in our corpus:  i, zas!, una bona fregada a sa llàntia amb s’anellet des dit des mig de sa mà esquerra, (Tale 9) ‘and, zas!, a good rub on the lamp with the ring of the lefthand middle finger’.  I es homos, zas!, una bona destralada (Tale 2) ‘and the men, zas! a good axe blow’.

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

288

Nicolau Dols

 i es moixos pes camí, mèu-mèu!, fins que no me’n deixaren bocí (Tale 4) ‘and the cats on my way, mèu-meu! [=they ate it up] until they left me without a single bit’.  S’hi aborden amb ses dents i, rac-a-rac, bones bocinades! (Tale 7) ‘they attack them [= the apples] with their teeth and, rac-a-rac, good bites!’  Arriba a ses portes del cel i toc-toc! (Tale 1) ‘He arrives at heaven’s doors and toc-toc!’ Other interesting aspects of ideophones in the corpus under study is the repetition of syntactic structures and the affix alternation, a sort of incomplete reduplication based on affixes instead of single sounds (see a list of them under R4 in §8.2). Their function is simply quantitative and no morphological recategorisation of the original root is involved. In this sense, grammatical integration is not complete: affixes are added, but only for an expressive, not a grammatical function. A deeper level of grammatical integration can be seen in child expressive speech, as in the traditional riddle describing a cow: ‘Dos lluentos, dos punxentos, quatre tripo-trapos i un espolsador de mosques’ (‘Two shine+nom+pl meaning ‘eyes’, two prick+nom+pl meaning ‘horns’, four ‘tripotrapos’ meaning ‘legs’, and a fly swatter’ meaning ‘tail’)

‘Tripo-trapos’ is here no more than an expressive used for walking (the same as the ‘tris-tras’ seen in §8.4.1 with a noun categoriser and a plural morpheme). Full word repetition, sometimes with a conjunction or a relative pronoun in the middle stands for continuity as in ‘demana qui demana’ (‘asks who asks’) (R3 in §8.2). What is interesting here is the adverbial function of the whole sequence, finally translatable as ‘asking once and again’, and its lack of agreement with the subject. A sentence such as the following is correct: ‘Demana qui demana vaig arribar al castell’ (‘Asks who asks [3rd person singular, no explicit pronoun], I reached the castle’) ‘Canta qui canta farem la feina’ (‘Sings who sings [3rd person singular, no explicit pronoun] we shall do the job’).

8.5

Concluding remarks and further research

There is, in general, no way of directly relating acoustic features with ideophones in Catalan or, at least, not according to the analysis of the corpus used here. However, from among the features studied, the observation of pitch difference reveals a tendency for expressives to appear in fragments with an uneven fundamental frequency. The segmental features of expressives show

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

Expressives in a spoken corpus of Majorcan Catalan

289

matches of sound and meaning already observed in other languages. The corpus includes repetitions with morphological alternations and the adverbial use of the structure verb + who + verb expressing continuity or repetition. Further attention should be paid to the phonetic foregrounding of expressives. Working with voice corpora seems the only feasible way to reach this point, as in Dingemanse’s and Akira’s assertion: ‘Corpus evidence is crucial for getting at expressive features of the speech signal, which would be hard to elicit and are best identified on the basis of recorded data available for repeated inspection’ (Dingemanse & Akita 2017: 508). If ‘[i]deophones are marked words that depict sensory imagery’ (Dingemanse 2011: 41), and the concept ‘sound symbolism’ meets ‘the proposal that linguistic sounds such as phonemes, features, syllables, or tones can be meaningful’ (Nuckolls 1999: 225), then studies on voice quality and altogether on acoustic and articulatory aspects of speech have a contribution to make. As observed by High in Amazonia, voice and body have a close relationship and emotions become physical by means of the voice: ‘Language, as a form of inter-subjectivity that corresponds specifically to human bodies, is both a product and catalyst of this process of affective relations’ (High 2018: 67). Recent works, such as Ponsonnet (2018) on the expressive capability of prosodic contour, Sondhi et al. (2015) on stress detection by pitch analysis, and Aryani et al. (2015) on the basic affective tone of poems measured by phonological saliency and iconicity have proposed that voice effects can add to the meaning of an utterance the way expressives do – an argument also supported by the findings of this chapter.

References Albers, S. (2008). Lautsymbolik in Ägyptischen Texten. Mainz: Von Zabern. Alcover, A. M. (1996–2022). Aplec de Rondaies Mallorquines d’En Jordi d’es Racó, edited by J. A. Grimalt & J. Guiscafrè. 9 vols. Palma: Editorial Moll. Auracher, J., Albers, S., Zhai, Y., Gareeva, G., & Stavniychuk, T. (2010). P is for happiness, N is for sadness: Universals in sound iconicity to detect emotions in poetry. Discourse Processes, 48, 1–25. Aryani, A., Ullrich, S., Kraxenberger, M., & Jacobs. A. M. (2015). Measuring the basic affective tone of poems via phonological saliency and iconicity. Psychology of Aesthetics, Creativity, and the Arts, 10(2), 191–204. Boersma, P. & Weenink, D. (1992–2022). Praat: Doing Phonetics by Computer [Computer program]. Version 6.2.06, retrieved 23 January 2022 from www.fon .hum.uva.nl/praat/ Choksi, N. (2020). Expressives and the multimodal depiction of social types in Mundari. Language in Society, 49(3), 379–98. Diffloth, G. (1979). Expressive phonology and prosaic phonology in Mon-Khmer. In T. L. Thongkum et al. (eds.) Studies in Tai and Mon-Khmer Phonetics and

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

290

Nicolau Dols

Phonology in Honour of Eugénie J. A. Henderson. Bangkok: Chulalongkorn University Press, 49–59. Dingemanse, M. (2011). Ezra Pound among the Mawu: Ideophones and iconicity in Siwu. In P. Michelucci, O. Fischer, & C. Ljungberg (eds.) Semblance and Signification. Amsterdam: Benjamins, 39–54. (2012). Advances in the cross-linguistic study of ideophones. Language and Linguistics Compass, 6(10), 654–72. Dingemanse, M. & Akita, K. (2017). An inverse relation between expressiveness and grammatical integration: On the morphosyntactic typology of ideophones, with special reference to Japanese. Journal of Linguistics, 53, 501–32. Dingemanse, M., Blasi, D. E., Lupyan, G., Christiansen, M. H., & Monaghan, P. (2015). Arbitrariness, iconicity, and systematicity in language. Trends in Cognitive Sciences, 19(10), 603–15. Dols, N. (2020). Phonology. Phonetics. Intonation. In J. A. Argenter & J. Lüdtke, (eds.) Manual of Catalan Linguistics. Berlin and Boston: De Gruyter, 101–28. Guiscafrè, Jaume. (2007). Antoni M. Alcover, father of the Mallorcan folktale canon. Ooohéee: Estudis sobre la creació i edició infantil i juvenil, 3, 58–80. Heinz, J. M. & Stevens, K. N. (1961). On the properties of voiceless fricatives. Journal of the Acoustica Society of America, 33(5), 589–96. High, C. (2018). Bodies that speak: Languages of differentiation and becoming in Amazonia. Language & Communication, 63, 65–75. Hiraga, M. K. (1994). Diagrams and metaphors: Iconic aspects in language. Journal of Pragmatics, 22, 5–21. Ibarretxe-Antuñano, I. (2017). Basque ideophones from a typological perspective. Canadian Journal of Linguistics/Revue canadienne de linguistique, 62(2), 196–220. Jongman, A., Wayland, R., & Wong, S. (2000). Acoustic characteristics of English fricatives. Journal of the Acoustical Society of America, 108(3), 1252–63. Kwon, N. & Yu, S. (2018). Experimental evidence for the productivity of total reduplication in Japanese ideophones and ordinary vocabulary. Language Sciences, 66, 166–82. Lee, A. P. (2017). Ideophones, interjections, and sound symbolism in Seediq. Oceanic Linguistics, 56(1), 181–209. Lennes, M. (2002). calculate_segment_durations.praat [Script for Praat]. Zenodo. https://bit.ly/3kR54tY Masuda, K. (2007). The physical basis for phonological iconicity. In E. Tabakowska, C. Ljungberg, & O. Fischer (eds.) Insistent Images. Philadelphia: Benjamins, 57–72. Matsuo, K. & Palmer, J. B. (2009). Coordination of mastication, swallowing and breathing. Japanese Dental Science Review, 45, 31–40. Moshi, L. (1993). Ideophones in KiVunjo-Chaga. Journal of Linguistic Anthropology , 3(2), 185–216. Nuckolls, Janis B. (1999). The case for sound symbolism. Annual Review of Anthropology, 28(1), 225–52. (2006). The neglected poetics of ideophony. In Catherine O’Neil, Mary Scoggin, & Kevin Tuite (eds.) Language, Culture and the Individual. Munich: LINCOM Europa, 39–50.

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

Expressives in a spoken corpus of Majorcan Catalan

291

O’Meara, C., Kung, S. S., & Majid, A. (2019). The challenge of olfactory ideophones reconsidering ineffability from the Totonac-Tepehua perspective. International Journal of American Linguistics, 85(2), 173–212. Ponsonnet, M. (2018). Expressivity and performance: Expressing compassion and grief with a prosodic contour in Gunwinyguan languages (northern Australia). Journal of Pragmatics, 136, 79–96. Pradilla Cardona, M. À. (2002). Ensordiment, espirantització i fenòmens que afecten les sibilants. In J. Solà, M. R. Lloret, J. Mascaró, & M. Pérez Saldanya (eds.) Gramàtica del Català Contemporani, Vol. 1. Barcelona: Empúries, 287‒318. Recasens, D. (2013). On the articulatory classification of (alveolo)palatal consonants. Journal of the International Phonetic Association, 43(1), 1–22. Reidy, P. F. (2016). Spectral dynamics of sibilant fricatives are contrastive and language specific. Journal of the Acoustical Society of America, 140(4), 2518–29. Sapir, E. (1921). Language: An Introduction to the study of speech. New York: Harcourt, Brace and Company. Sondhi, S., Khan, M., Vijay, R., Salshan A. K., & Chouhan, S. (2015). Acoustic analysis of speech under stress. International Journal of Bioinformatics Research and Applications, 11(5), 417–32. Uther, H.-J. (2004). The Types of International Folktales: A classification and bibliography, based on the system of Antti Aarne and Stith Thompson. 3 vols. Helsinki: Suomalainen Tiedeakatemia-Academia Scientiarum Fennica. Wheeler, M., Yates, A., & Dols, N. (1999). Catalan: A comprehensive grammar. London: Routledge. Wiseman, M. & van Peer, W. (2003). Roman Jakobsons Konzept der Selbstreferenz aus der Perspektive der heutigen Kognitionswissenschaft. In H. Birus, S. Donat, & B. Meyer-Sickendiek (eds.) Roman Jakobsons Gedichtanalysen. Göttingen: Wallstein, 277–306.

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.012 Published online by Cambridge University Press

Celtic

https://doi.org/10.1017/9781108989084.013 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.013 Published online by Cambridge University Press

9

A survey of expressive words in Breton Mélanie Jouitteau*

9.1

Methodology and road map

Breton is an Indo-European language, with a grammatical tradition in French.1 Mimetics and ideophones were singled out in diachronic studies of Breton of the nineteenth century because their diachronic derivation showed irregularities. As a result, dictionaries tend to mention expressives, often with an accompanying French translation. I began this study by organizing the results of exhaustive searches in the available lexicology material adapted for automatized searches with the keywords interjection, interj. or onomatopée, onom. (Matasovi
c 2009, Deshayes 2003, Le Gonidec 1821, Henry 1900, Ernault 1927, Cornillet 2020). I created a dedicated page in my wikigrammar of the Breton language (Jouitteau 2009–2022) for each variety of expressives mentioned in the formal literature: interjections, mimetics (= onomatopée in French tradition), taboo word camouflage, and phonoesthemes (= ideophones in French tradition), and sorted the collected data accordingly. This systematic ranking process ensured that an expressive type was not accidentally neglected. It confirmed that these varieties cannot be understood as categorial, mutually exclusive classes because a given expression may belong to several of them, but that they do show consistent morphological, syntactic and semantic properties. I obtained a premodern Breton inventory containing mostly mimetics, ideophones, and some minimal interjections like aiou! to express pain, some of which are now outdated. The twentieth century saw a new academic interest in the study of orality and familiar registers. Pedagogical descriptions of the Breton language gradually took stock of the decline of Breton oral practices, and were subsequently

* National Centre for Scientific Research, France (CNRS) 1 In this chapter, R in the glosses stands for the preverbal particle ‘rannig’. The Breton dialects (Kerne, Leon, Treger, Gwenedeg, or Standard) are mentioned in brackets. For reasons of space, the sources of every occurrence cannot be cited here, but they are all carefully referenced on the wikigrammar of Jouitteau (2009–2023). I have favoured for discussion in this chapter the forms that I could cross-reference in several corpuses. Examples are presented in their original orthography.

295

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

296

Mélanie Jouitteau

designed to assist Breton learners to attain near-native speaker proficiency. This triggered greater attention to a wide variety of expressives, especially taboo expressions, interjections, and all sorts of focalizing strategies seldom reported before. I integrated with my survey the notes, examples, and remarks of Gros (1974), a stylistic treatise of great descriptive value. I next enriched my data collection by a manual study of sixteen Standard Breton comic books (listed in the Appendix). Most of them are translated from an available French comic, which allowed me to observe the translation strategies of different authors. Finally, I conducted three elicitations with two native speakers of the Breton language.2 At all steps of the data organization process, I have enriched the description by targeted searches of Menard and Bihan (2016–), Favereau (2016–), Jouitteau (2009–2023), and with internet search engines. I present in §9.2 the morphological hallmarks of Breton expressives: reduplication, apophonic alternations, and a trisyllabic recurring pattern. In §9.3, I turn to each syntactico-semantic variety in turn. I discuss the productivity of the expressive operations: do they operate across all categories? Are they lexically restricted?; their exclusivity: does a given expressive operation always result in expressives or not?; and their iconicity: is it iconic, and if so, in what sense? For reasons of space, I have mostly set aside discussion of phonoesthemes, which would require a dedicated study.3 9.2

Morphological hallmarks

9.2.1

Reduplication

Reduplications are to be found in mimetics (1), phonoesthemes (2), and taboo word camouflage (an dipadapa ‘diarrhoea’). (1)

ur c’harr éh ober kwik-kwik-kwik (East Kerne) a car at to.do /kwik-kwik-kwik/ ‘a car doing kwik-kwik-kwik’

(2)

Te’ vad a zo gwigour ez potou! (Treger) you ! R is creak in.your shoes ‘But you, you have bloody creaky shoes!’

2

3

The raw elicitation data is available online at the elicitation centre of the wikigrammar (Jouitteau 2009–2022), and is also redistributed across the wikigrammar. My deepest thanks and gratitude go to the speakers A.-M. Louboutin (Kerne) and Janig Bodiou-Stephens (Treger). Thanks also to Marijo Louboutin for her kind help in setting up the elicitations during the COVID-19 pandemic, and to the IKER (CNRS) laboratory for support. Phonoesthemes result mostly from articulatory coincidences, with some candidates for perceptive phonoesthemes. They are documented, following the French terminology, under ‘idéophones’ in the wikigrammar Jouitteau (2009–2023).

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

A survey of expressive words in Breton

297

The reduplication structure in (3) bears alone the expressive dimension, whereas (4) adds apophonic alternations. Neither /bardi/ nor /barda/ are lexical entries, and the repetition of these nonsensical words iconically denotes the action of uttering unimportant language. (3)

hag evel-henn hag evel-hont (Standard) and like-this and like-that ‘and yadeyackyack . . .’

(4)

a. ha bardi, ha barda . . . ha bardi ha and /bardi/ and /barda/ and /bardi/ and b. ha flip ha flap, ha jip ha and /flip/ and /flap/ and /ʒip/ and ‘and yadeyackyack . . . and yadeyackyack . . .’

barda . . . (Standard) /barda/ jap . . . (Kerne) /ʒap/

I found only weak evidence for the reduplication of interjections. The attention attractor C’hep! ‘Hey!’ shows an extra initial /p/ in reduplicated Pep pep! ‘Hey! Hey! I’m talking to you!’, expressing impatience. Other minimal interjections can appears twice side by side (6), (7), but without morphological fusion or rearrangements they could simply be repeated (5). The change in meaning is consistent with the pragmatic effect of repetition: the speaker behaves as if the interlocutor has not heard or paid attention to the previous occurrence. The verum focus interjection A! repeated in (6) and (7) expresses intensification while implying the interlocutor does not fully realize the extent of an intensity. (5)

ur pladad eus ar c’hentañ! . . . Menam! Menam! (Standard) a plate of the fist yum yum ‘An abundant dish of first quality! Yum yum!’

(6)

Deuet tomm din ken a oa, HaHa! (Treger) came hot to.me as.much R was Ah! Ah! ‘I had an intense heat stroke (You can’t imagine)!’

(7)

Aaaa! Me meus bet tomm ayayaylh! (Kerne) Ah! Ah! Ah! I have had hot Ouch! Ouch! Ouch! ‘I had an intense heat stroke (You can’t imagine)!’

Iconicity of reduplication is clear in intensifiers: more of the linguistic material obtains a greater degree of its meaning (berr ‘short’, berr-berr ‘very short’). Reduplication is also iconic when it targets dynamic movement or change of state verbs (8) and prepositions (9) to obtain iterative meaning: the greater the repetition, the greater the intensification of the meaning. The separative di- prefix at the heart of the reduplicated verbal structure in (8) is fully productive with all verbs. (8)

Goude e vezont bloñset ha dibloñset tout evel-just. (Leon) after R are hit and prefix.hit all of-course ‘Afterwards, they are bruised of course.’

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

298

Mélanie Jouitteau

(9)

Hezh skôe ket war ar youd, oa ‘biou-’biou bep taol. (Kerne) this.one hit not on the porridge was beside-beside each hit ‘He didn’t hit the porridge, but next to it every time.’

Reduplication does not always produce expressive words. The adjective berr ‘short’ has no adverbial counterpart, but its reduplication around coordination does (berr-ha-berr /short-and-short/ ‘briefly’). Reduplication is here exocentric, with an expressive impact due to rhythm, but without an iconic dimension. Reduplication of a verb with a diminutive (10) obtains ‘less of the same meaning’, the opposite of (8). In (10), the two occurrences do not construct a greater degree. They have a consecutive reading (living well and then less than well). Reduplication obtains an iterative reading on this alternation. (10)

Bevañ-bevaik a rae, kalonek atav . . . (Standard) to.live-to.live.small R did brave always ‘He was barely getting by, always brave . . .’

No expressive semantics is present in reduplication of the head noun of an analytic demonstrative that creates a free choice item: ar plac’h-mañ-plac’h, / the girl-here-girl/ ‘any girl, whatever girl’. The left of the reduplicated structure ar plac’h-mañ would be a demonstrative, ‘this girl’, if in isolation. Jouitteau (2015) has shown this would not be the right analysis for the reduplicated structure. The reduplicated structure exists independently of the determiner (11), but the analytic demonstrative does not (absence of mutation on kêr shows that the article is syntactically absent, cf. ar gêr ‘the house’). The adverbial clitic -mañ ‘here’ is a deictic. The reduplication of the head noun on its right obtains a less than clearly identified referent ( house here or any house, really). (11)

Pa veze dornadeg, e kêr-mañ-kêr . . . (Kerne) when was threshing.collective in house-here-house ‘when the wheat was threshed in such and such a house . . .’

9.2.2

Apophonic alternations

Expressive apophonic alternations are present in mimetics, phonoesthemes, interjections, and taboo word camouflage. Mimetics include the common tiktak ‘sound of a clock’ or balingbalom ‘sound of bell’ (12). The noun chuchumuchu ‘whisper, murmur’ relies on the reduplication of a voiceless fricative in articulatory coincidence with the act of murmuring, which reveals its phonoesthetic dimension. Phonoesthemes typically use apophonic alternations in monosyllabic minimizers like in (13) or tremen ku-ha-ka, to.pass /ky/-and-/ ka/, ‘to narrowly pass’. Interjections sporadically use apophonic alternations like Menam-menam! ‘Yum yum!’ (5). They extensively resort to it for taboo

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

A survey of expressive words in Breton

299

word camouflage. In (14), Fitamdaoula! camouflages /faith in my God/ ‘Goodness me!’. The equivalent interjection Satordallik!, Satordistac’h! camouflages the adjective sakre ‘sacred’ and Doue ‘God’; Tankerru! camouflages Kurun! ‘thunder’; Nondidiko!, Nondididisteg! camouflages the French borrowing Nom de Dieu!, literally ‘name of God’, and so on. (12)

Baling Balom, Marrig zo klaoñv . . . (Kerne song) Baling Balom Mary.little is sick ‘Ding-dong, Little Mary is sick . . .’

(13)

N’ o deus ket bet tro da lavaret na bu na ba! (Standard) neg 3PL has not had time to say neither /by/ nor /ba/ ‘They didn’t have time to say “phew!”’

(14)

Hañ, fitamdaoula! Setu tapet Fulup avat! (Kerne) right! taboo.word here caught Philippe exclamative ‘Right! Goodness me! Philippe is caught out!’

The vowels of apophonic alternations in expressive morphology consist mainly of the maximally distinctive vowels, /i, a, u/, as well as /e/ and the central vowel /y/. None of them are nasals. This set of vowels contrasts sharply with that of fillers and hesitation marks Añ . . ., Beñ . . ., Bo . . ., Eee . . ., Eump . . ., Hañ . . ., Hmmm . . ., Ma . . ., Oc’h . . ., Oñm . . . The later represent more typically the Breton non-expressive vowel system with nasals /ã, ɛ̃, ø̃ , õ/ or the vowels /a, o, ɔ, e/. This contrast in vowel systems between expressive apophonic alternations on the one hand, and fillers and non-expressive morphology on the other, shows that expressive morphology can resort to a distinctive, dedicated vowel system. Breton expressive morphology does not have monopoly on apophonic alternations. Several non-expressive paradigms make use of apophonic alternations, like the nominal -ed suffixation of adjectives (klañv /klaõw/ ‘sick’ > kleñved /klɛ̃vɛt/ ‘sickness’ and yac’h /jaX/ ‘healthy’ > yec’hed /jehɛt/ ‘health’). The set of vowels can clearly exceed the expressive set, as illustrated by internal plurals (troad /troat/ ‘foot’ > treid /trɛjt/ ‘feet’, or roc’h /rɔX/ ‘rock’ > reier /rejɛX/ ‘rocks’, or askorn /askɔrn/ ‘bone’ > eskern /eskɛrn/ ‘bones’), or infinitival heads and their participial (sevel /sevɛl/ ‘to rise’, savet / savɛt/ ‘risen’, sentiñ /sɛnti/ to obey > santet /sãntɛt/ ‘obeyed’, or lemel /lemɛl/ ‘to remove’ > lamet /lãmɛt/ ‘removed’). Guerssel and Lowenstamm (1994, 1996) have studied the relationships between the different verbal patterns of classical Arabic and have proposed an apophonic path ordering the melodic primitives. This path is implicational and derivational: ø => I => A => U => U. Ségéral (1995), and Ségéral and Scheer (1998) have extended these results to strong German verbs, and proposed that this path and its implicative meaning are universals of human language. Since then, as noted in Scheer (2000: 7), ‘other works have revealed the existence of apophonic systems which conform to the predictions. Such is

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

300

Mélanie Jouitteau

notably the case of the Ge’ez (classical Ethiopian, Ségéral 1995, 1996), Acadian (Ségéral 1995, 2000), Berber (Bendjaballah 1998, 1999), Bédja (Cushite, Bendjaballah 1999), Italian, French, and Spanish (Boyé 2000), Somali (Cushitic, Ségéral & Scheer 1997 [. . .]) and English (Ségéral & Scheer 1996), the system of weak verbs in Classical Arabic (Chekayri & Scheer 1996, 1998, 2004)’. No expressive Breton word collected for this paper contradicts this proposition of universal; we have flik-flak ‘flic-floc’, but not */ # flak-flik, Menam-Menam ‘yum yum’, but no */# manem-manem. Only two cases seemed to go upstream the path, inside non-reduplicated words: cholori ‘racket’ and mont e belbi /mõn e bɛlbi/, /go in futilities/ ‘to lose one’s mind’, but their expressive dimension is up for debate. 9.2.3

A trisyllabic pattern

A trisyllabic pattern with apophonic alternation emerges as distinctive across all expressive categories. It is particularly salient in ideophones that evoke a disorderly fall in several consecutive movements because their wild morphological variation preserves the trisyllabic pattern: Badadav! Badadaou!, Badadouilh!, Boudoudoum! Boudadoof! Boudoudouf!, and maybe also Fataklev!. Like their French equivalent Patatras! and Badaboum!, these interjections are mostly constructed with plosive consonants and vowels patterning in x-x-y (Paradaouf! allows for a liquid). Their speech act is paraphrasable (she fell!), and their temporal anchoring is consecutive or coinciding with the time of the paraphrase. They are not mimetics if the fall is silent. The aspectual structure of Paradaouf! in (15) is loosely ideophonic over the three consonants p-r-d (> not exactly three consecutive movements). (15)

Lod-all ‘meus bet gwelet koz-lammat ag . . . paradaouf! var an douar. (Leon) some-other have been seen bad-to.jump and badaboum on the earth ‘I’ve seen others jump badly and badaboum! down.’

The same pattern is observed in (16), a song sung with a child on one’s knees, making it jump to the rhythm of a horse that walks, then trots, then gallops. The ternary consonant rhythm reproduces that of the gallop – letting the vowels rhyme. Finally, the pattern is found in taboo word camouflages (ar vadadailh, an deur-deur-deurt ‘the diarrhoea’), and in the noun talabao ‘tumult’ (17). (16)

Didedoup! Didedoup! Da Vontroulez da ‘vit stoup! (Kerne) /didØdup didØdup / to Montroulez to get tow ‘Let’s go to Montroulez to look for tow.’

(17)

youc’hadennoù an dud ha talabao al loened (Standard) cries the people and tumult the animals ‘the cries of humans and the tumult of the beasts’

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

A survey of expressive words in Breton

301

Such a trisyllabic pattern is not exclusively expressive (cf. bodadeg ‘meeting’, talaspik ‘stool’, talatenn ‘headband’, etc.). 9.3

Mimetics, phonomimes

Mimetics (onomatopées in the French descriptive tradition) mimics extralinguistic sounds and strictly denotes sounds.4 They are derivationally productive and furnish raw material feeding lexical categories (nouns, verbs, adverbs) with various loose associative meanings. Menard and Le Bihan (2016–) offer a representative sample of examples using the mimetic /flip/, which denotes the ‘noise of a whip’ and can be the bare object of the verb ober ‘to do’ (ober flip ‘to do an action producing the noise /flip/), alone or reduplicated (18). (18)

. . . ken ra flip-flip-flap lost he liviten paour. (Standard) so.much does /flip, flip, flap/ tail his jacket poor ‘. . . so much that his poor jacket goes /flip, flip, flap/.’

The noun flip also denotes the object producing /flip/, ur flip ‘a whip’, which in turn derives into several expressions with associative meaning to: i. the gesture of throwing a flip, in strinkañ e flip /to throw in /flip//, ‘to throw on the fly’; ii. the aspectual structure of the action of throwing a whip (mont e flip, /to go in /flip//, ‘to leave quickly’ or diwar ar flip, /from the /flip//, ‘hastily’ or ‘on the flight’); iii. the reference of something that has the effect of a whip (ur flip, a drink with hot cider, sugar, and brandy); iv. several denotations, pictorial representations of dangerous tongue movements (gossip, fire movements). Each of the derived results can productively enter further regular morphological derivation. The nominal suffix -ad yields flipad ‘whiplash, gossip’, and even ‘long path’. Ur flipad means ‘a lot, a lot of time, a long way’, probably from the elongated structure of a physical whip together with the intensifying effect of the hit ‘whiplash’. The aspectual construction achap en ur flipad / escape in a flip.N/ means the contrary: ‘in not a lot of time’. The verb flipañ has meanings as different as ‘to slip away’, ‘to drink’, ‘to spend’, ‘to slander’, and so on. These meanings do not seem to compete with each other in the lexicon, nor does it with the non-expressive flip ‘ear-lobe’, as if the image of the whip was still convoked each time, instead of conventionalized in the lexicon. In (19), the noun flip denotes a fast, possibly silent, out-and-back trajectory. It 4

Mimetics include some conventionalized speech to animals, which I set aside here. The wikigrammar presents them at ‘huchements’.

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

302

Mélanie Jouitteau

commutes with a mimetic (bare object) or an infinitival verb heading a small clause like lammat ‘to jump’, but not a deverbal noun lamm ‘jump’ (Ne ri nemet lamm*(-at) hag er-meaz). In (20), flip appears in a narrative infinitive, again commutable with lammat ‘to jump’ (. . . ha lammat d’e wele). (19)

Ne ri nemed flip hag er meaz. neg will.do only /flip/ and in.the out ‘You will just pop in.’

(20)

. . . ha flip d’ e wele. (Standard) and /flip/ to his bed ‘. . . and he jumped in bed.’

(Leon)

Mimetics of shocks and impacts productively obtain aspectual adverbs such as plouf, splash, flav, krak, pfiouff, and so on. In the following examples, the aspectual adverbs are fully integrated into the syntactic structure, between the subject and the predicate of the sentence, where aspectual adverbs otherwise appear (21). The derived mimetic is not paraphrased by the sentence, it modifies it. The adverb appears alone in a tensed sentence (21), but it is introduced by a coordination marker in narrative infinitives, like a realized subject does in (22) and (23), or a participial clause in (24). (21)

Kaor Morwena dioustu nun taol piouff a zo kollet. (Kerne) goat Morwena now in.one hit /pjuf/ R is lost ‘Morwenna’s goat, suddenly, pfiouff, disappeared.’

(22)

. . . ha me ha badadav da vont d’ar bord all. (Treger) and I and /badadaw/ to to.go to.the edge other ‘. . . and boom! I fainted!’

(23)

. . . ha me ha splash da gouezhañ ‘ba ‘n dour. (Treger) and me and /splaʃ/ to fall into the water ‘. . . and splash! I fell in the water!’

(24)

. . . ha me ha splash kouet ‘ba ‘n dour. (Treger) and me and /splaʃ/ fallen in the water ‘. . . and splash! I fell in the water!’

Expressive aspectual adverbs built on shocks and impact mimetics differ from non-expressive aspectual adverbs because they contain information about the physical properties of the materials that impact each other (plouf, splash liquid, flav sticky, flexible solid, krak rigid solid, pfiouff gas, etc.). Other aspectual adverbs do not (cf. a-greiz-tout ‘suddenly’, ingal ‘permanently’, dalc’hmat ‘constantly’, adarre ‘again’, a-bep-eil ‘alternatively’, etc.), even the expressive ones if they do not mimic shocks and impact (lip-ha-lip, treha-tre, penn-da-benn ‘completely’). Expressive aspectual adverbs of shocks and impact show a gradability in iconicity, from mimetics to ideophones as illustrated in Table 9.1, which organizes data from two elicitations with the same speaker in Locronan (Kerne). Flav

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

A survey of expressive words in Breton

303

Table 9.1 Shocks and falls mimetics (adverbs and interjections) in Kerne

liquid

egg yolk, glue

baby (even sticky)

plate, cupboard, bench

matches

Flav!

*/OK

OK

*

*

*

Klak!

*

*

*

OK

Badadav!

*

#

OK

OK

OK with gesture OK

A fall on to hard ground of:

chocolate powder - not tested * OK

appears only for liquids ‘if they are sticky enough’, like egg yolk or glue. The trisyllabic Badadav! /badadao/ is more generalist than the other two and covers all solids. It tolerates an egg yolk impact (in contrast to klak!, clearly rejected), and the fall of matches. The silent fall of chocolate powder further signals that the trisyllabic Badadav is not a mimetic. The lack of mass articulation in chocolate powder even thins the ideophonic value of the trisyllabic structure. (25)

Ale (Klak!/ Krak! / Badadav!) An alumetez ‘zo kouet war an douar! (Kerne) come.on /klak/ /krak/ /badadaw/ the matches is fallen on the ground ‘Patatras, the matches have fallen on the floor!’

These complex derived mimetics can also semantically encode the result of an action like the separation of subparts. Klak! is accepted both for the fall of hard furniture and for the fall of matches and their scattering on the ground (25). This speaker uses a reduplicated glililing for the fall of marbles, but /klak/ in (25) is not reduplicated. Instead, the speaker accompanies klak! with an ostensible distributive gesture of the hands (a palm spread). In (25), krak did not require a distributive gesture to express the separation of its subparts. The klak / krak contrast is confirmed below with a bench that falls and breaks (krak), or that falls but stays intact (klak). In (26), the syntactic structure is a small clause without a realized subject. The consecutive temporal reading is brought by the coordination marker. The preposition war is static, so all the information (he) fell and broke is brought by krak. (26)

Kouet eo ‘ bank ha { krak / klak } war an douar! (Kerne) fallen is the bench and /krak/ / /klak/ on the earth ‘The bench fell on the ground, (and broke / and didn’t break)’

Now if one compares Klak! with Dao! ‘Paf!’ associated with the gesture of knocking, the former has a patient argument, whereas the later semantically links two arguments (hitting agent, hit patient). They are still not syntactic verbs: Dao! could not take an object (* Dav an nor! ‘Knock at the door!’) or be passivized (*/# Dao! gant Frank. ‘He was hit by Frank’).

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

304

Mélanie Jouitteau

Meinard (2015) distinguishes interjections from mimetics (in which she includes nouns directly derived from mimetics like ur flip ‘a whip’). The present data confirms her generalisation that mimetics and their derived nouns are a productive source of lexical creation, and that they can refer, as opposed to interjections that are predicative in nature and never refer. However, mimetics of shocks and impacts constitute an inter-class. These derived mimetics contain aspectual information, thematic relationships (patient, agent, etc.), as well as fine-grained information on the parameters of the impact (materials involved, end result), like verbal predicates would (to unstick). These ‘semantically rich’ aspectual adverbs can always have the distribution of paraphrased interjections like (25). Adjunct adverbs resemble interjections because they can have lexical content related to emotions, and have neither argument nor inflected form. The Breton narrative infinitive structures in (22, 23, 24, 26) allow for non-tensed matrix sentences, and seem to provide a bridge for mimetics of shocks and impacts to move from fully integrated aspectual adverbs to interjections.

9.4

Interjections

Interjections are not derivationally productive and can replace a sentence, which differentiates them from mimetics (Meinard 2015). The interjections are generally invariable, with minor variations related to the address (gender, formal mode of address) that are not instantiated by expressive interjections. They vary in semantic size. The lightest seem the interpellation interjections Eh!, C’hep!, Hep!, Yao!, Yo!, You!, Alo!, Ola!, Oc’hola!, Orê!, and others, which are all interchangeable. We saw C’hep! ‘Hey!’ can be repeated in Pep pep! ‘Hey! Hey! I’m talking to you!’. Next come a set of interjections with an opaque and minimal morphology but some sort of translatable proto-semantic content (27), in which we also find the mimetics of shocks and impacts like (25). They are syntactically optional, and only allowed in the left and right peripheries of the sentence, not in the middle field of tensed sentences (27). Most of those minimal interjections are of arbitrary morphology (cf. A! ‘Oh! So . . .’ or ‘That’s for sure!’, or Ac’ha!, Ac’hañ! Oc’ho!, Ac’haaaa! ‘I was right!’ or ‘I got you!’, etc.), to the exception of an enclitic -X ideophonic on the act of spitting that expresses disgust: Ec’h!, Oc’h! Fac’h! Fec’h! or Ac’h in (27), or to the ideophonic flavour of Fou! Hou! Fow! ‘Phew!’ expressing relief. (27)

(Ac’h!) Henn neus lakaet din (*Ac’h!) e zaorn war ma foñs (Ac’h!) (Treger) Eww! he has put to.me Eww! his hand on my foundation Eww! ‘He put his hand on my ass, eww!’

A third set of interjections takes from grammaticalisation of lexical material. They mostly show a non-iconic morphology, with the exception of the

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

A survey of expressive words in Breton

305

ideophonic monosyllabic minimizers Mik!, Grik!, Chik! ‘Hush!’, which derive respectively from the adjective mik ‘inert’, or from a noun denoting a ‘word’ (unless Chik! is from chik ‘chin’, or chik ‘quid of tobacco’). The three variants however show a convergence of forms, with the same single syllable in -ik which is also homophonous to the diminutive suffix. Interjections seldom have derived forms, which is not surprising if they are sentences (Meinard 2015), but again the agentive interjection of shock and impact Dao! in (28) shows a suspiciously verbal behaviour because the prefix ad- is normally reserved for verbal or nominal roots (adober ‘to do again’, adkoan ‘second supper’). However, both interviewed speakers had the Badadav! crash interjection, but neither accepted it with the -et participle suffix (*Badavet, obtained from the made-up French *Badaboumé). The verb-flavoured Yao! ‘Gee up!’ is to be treated separately, as a stylistic use of the language normally dedicated to giving orders to horses (29). (28)

Dao! Taol kaer! ad-Dao! Kaerat! (Standard)5 Paf! blows beautiful again-Paf! handsome.optative ‘Wham! . . . Well-done! And again! Bravo!’

(29)

Ha yao da vro Vec’hiko! (Standard) and gee.up to land Mexico ‘. . . and gee up! Back to Mexico!’

9.5

Taboo word camouflages

Taboo Breton words mainly concern the sexual, scatological, or religious domain, and occasionally some references to poverty or dirt (kutez ‘slum, hovel’, lastez ‘garbage’), some infirmities and diseases (moñs ‘stump’), as well as some violent natural phenomena (kurun ‘thunder’, foeltr ‘lightning’). These words, on which language documentation is usually silent, have been compiled in specialized lexicons, such as the Cryptological Glossary of Breton by Ernault (1884–1902, 1999) or the dictionary of taboo words by Menard (1995). The taboo dimension of a word seems to be enough to give it an interjection value (Gast!, /prostitute/, ‘Fuck!’) and/or an intensification value (ur c’hastad hini, /a prostitute.content one/, ‘an enormous one’ or alkool ar c’hast, /alcohol the prostitute/, ‘fucking alcohol’ or Petra ar c’hast eo? /what the prostitute is / ‘What the fuck is it?’), but only camouflaged taboo words show expressive morphology like Fidamdoustik!, Fidambie! or Fitamdaoula! seen in (14) camouflaging Feiz d’am Doue! ‘Faith in my God!’. Camouflaged taboo words are motivated by a necessity to exhibit avoidance, rather than 5

The example is only found once on p. 7 of the Tintin Album ‘Flight 714’ translated into Standard Breton in the An Here edition. The English translation here is from the English album. French was PAF! Bien ça! Re-PAF! Bravo!.

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

306

Mélanie Jouitteau

genuinely avoid offence. An offending word is commonly replaced by another offending word. Fidamdoull! avoids pronouncing the name of /God/ only to replace it by toull ‘hole’, giving ‘#Faith in my hole’. Both a taboo word and its ostensible avoidance are compatible (30). They flourish in colloquial language, but can still appear in polite usages. In (30), the speaker is a business manager in a youth album (Tintin). (30)

Atoe! Ma Doue! Pegen plijet on ouzh ho kavout . . . (Standard) to.#God My God! how happy am at you find ‘Oh Gosh! Oh my god! What joy for me to find you . . .’

Interjections only marginally give derived adjectives (She is a wow! Bottineau 2013). Breton taboo camouflage interjections have an evaluative semantic dimension which could allow for an adjectival derivation. The camouflage of the interjection Feiz d’am Doue! appears after the determiner like a noun in ar fidamdie a blantenn-mañ /the /fidamdije/ of plant-here/ ‘this fucking plant’. The semantic structure is predicative (this plant is /fidamdije/), but fidamdie is illicit as a predicative noun (* Fidamdie eo ar blantenn-mañ, //fidamdije/ is this plant /). Any expressive morphological camouflage allows for the structure, also written in comics as ar ☠ ☠ ☠ ☠ ☠ a blantenn-mañ. Non-expressive interjections are here ungrammatical (*an ac’hanta a blantenn-mañ, with ac’hanta! ‘Well!’, or ar *ar memestra a blantenn-mañ, with memestra! ‘all the same!’), as are mimetics (*an dao a blantenn-mañ, with Dao! ‘knock, hit’). The equivalent construction /det N1 of NP2/ is known in French as the qualitative construction with a ‘pure degree’ interpretation (cette sapristi de bonne femme). It involves a nominal group without predicate inversion, inside an evaluative adverbial projection (Doetjes & Rooryck 2003). In contrast to French, the Breton construction also allows adjectives, including nonevaluatives (*une longue de plante, but un hir a blantenn ‘a long plant’, an hir a blantenn-mañ /the long of plant-here/ ‘this long plant’), leaving open the categorial nature of Fidamdoue.

9.6

Conclusion

Breton expressive morphology uses reduplication, apophonic alternations, and/or a specific trisyllabic pattern, none of them being exclusive to expressive morphology. Apophonic alternations use a dedicated vowel system, which however still obeys the universal apophonic path ordering the melodic primitives of Ségéral and Scheer (1998). Mimetics derive referring nouns, as opposed to interjections that are predicative in nature (Meinard 2015), but the Breton matrix infinitives (narrative infinitives) provide a bridge for mimetics of shocks and impacts like klak, dao or badadav to move from fully integrated aspectual adverbs to interjections. Mimetics of shocks and impacts

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

A survey of expressive words in Breton

307

constitute an inter-class with more verbal-like properties, including a morphological derivation normally reserved for verbs or nouns, but they still cannot passivize or be derived as participles (compare with Don’t yuck somebody else’s yum), and a gradability in iconicity. Appendix An Here (ed.). 2003. Nij 714 da Sydney, translation of Hergé (1963). Vol 714 pour Sydney, Casterman (ed.). Ar Menn, Brieg. 2015. Ar pevar gringo Dalton, Bzh5 (ed.), translation of Morris and Goscinny (1967) Tortillas pour les Dalton, Dupuis (ed.). Bannoù-Heol (ed.). 2000. Sell ’ta!, Boulig ha Billig, translation of Roba (1988) 22! V’là Boule et Bill!, Roba SPRL, Dargaud (ed.). Biguet, Olier. 2017. Tintin en Amerika, translation of Hergé (1973) Tintin en Amérique, Casterman (ed.). Bzh5 (ed.). 2007. Ar pevar Sant Dalton, translation of Goscinny and Morris (1971) Les Dalton se rachètent, Dargaud (ed.). Kervella, Divi. 2001. Troioù-kaer Tintin: Bravigoù ar Gastafiorenn, An Here (ed)., translation of Hergé (1963) Les bijoux de la castafiore, Casterman (ed.). 2002. Troioù-kaer Tintin: Al Lotuz Glas, An Here (éd.), translation of Hergé. 1946. Le Lotus Bleu, Casterman (ed.). 2002b. Troioù-kaer Tintin: An Enez du, An Here (éd.), translation of Hergé. 1963. L’île noire, Casterman (ed.). 2006. Ar c’hazh e Breizh, translation of Geluck, Philippe (2000) Le chat est content, Casterman (ed.). Keit Vimp Bev (ed.). 1984. Yakari hag an estranjour, translation of Derib & Job. 1982. Yakari et l’étranger, Casterman (ed.). Le Saëc, Erwan. 1990. Ar skarzherien, Keit Vimp Bev (ed.). Monfort, Alan. 2006. Gaston 14, Yoran Embanner (ed.), translation of a selection from four albums Gaston Lagaffe by Franquin, Dupuis (ed.). 2007. Gaston 10, Yoran Embanner (ed.), translation of Gaston 10, a selection from three albums Gaston Lagaffe, copyright Marsu 2007 by Franquin, Dupuis. Moulleg, Loeiz. 1978. An Ankou, troioù-kaer Spirou ha Fantasio, Dupuis (ed.), translation of Fournier (1976) L’Ankou, Dupuis (ed.). Preder & Armor. 1977. Emgann ar Pennoù, Preder (ed.), Armor diffusion, translation of Goscinny and Uderzo (1966) Le combat des chefs, Dargaud (ed.). Skol an Emsav 1977. Pare Paotred Dalton, translation of Morris and Goscinny (1975) La guérison des Dalton, Dargaud (ed.).

References Boyé, Gilles. 2000. Problèmes de morpho-phonologie verbale en français, en espagnol et en italien. Thesis, University of Paris 7. Bendjaballah, Sabrina. 1998. Aspects apophoniques de la vocalisation du verbe berbere (kabyle). In Patrick Sauzet (ed.) Langues et Grammaire II-III, Phonologie. Paris: Université Paris 8, 5–24.

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

308

Mélanie Jouitteau

1999. Trois figures de la structure interne des gabarits. Thesis, University of Paris 7. Bottineau, Didier. 2013. OUPS! Les émotimots, les petits mots des émotions: des acteurs majeurs de la cognition verbale interactive. Langue française 180, 99–112. Chekayri, Abdellah & Tobias Scheer. 1996. The apophonic origin of glides in the verbal system of Classical Arabic. In J. Lecarme, J. Lowenstamm, U. Shlonsky (eds.) Studies in Afroasiatic Grammar, The Hague: Holland Academic Graphics, 62–76. 1998. La provenance apophonique des semi-voyelles dans les formes verbales en Arabe Classique. Langues et Linguistique 2, 15–54. 2004. The appearance of glides in Classical Arabic defective verbs. Folia Orientalia 40, 7–33. Cornillet, Gérard. 2020. Geriadur Brezhoneg-Galleg. Manuscript. Deshayes, Albert. 2003. Dictionnaire étymologique du breton, Douarnenez: Le Chasse Marée. Doetjes, Jenny & Johan Rooryck. 2003. Generalizing over quantitative and qualitative constructions. In Martine Coene & Yves D’hulst (eds.) From NP to DP, Vol. I: The syntax and semantics of noun phrases. Amsterdam: Benjamins, 277–95. Ernault, Émile. 1884–1902. Glossaire cryptologique du breton, avec un additif et corrections et 2 suppléments. In G. Le Menn (ed.) Kryptidia, vols II, III, VI, VIII. 1927. Geriadurig brezhoneg-galleg. Saint-Brieuc: Prudhomme. 1999. Glossaire cryptologique breton (expressions érotiques, scatologiques, etc.), avec index breton-français et français-breton et notes par Gwennole Le Menn. Bibliothèque Bretonne 7. Saint-Brieuc: Skol. Favereau, Francis. 2016–2022. Grand dictionnaire bilingue breton-français, françaisbreton. https://geriadurbrasfavereau.monsite-orange.fr Le Gonidec, J.-F., 1821. Dictionnaire celto-breton ou breton-français, Angoulême: Trémenau. Gros, Jules. 1974. Le trésor du breton parlé III. Le style populaire (Éléments de stylistique trégorroise). Barr-Heol, Lannion: Giraudon. Guerssel, Mohand & Jean Lowenstamm. 1994. Ablaut in Classical Arabic measure I active verbal forms. Communication à la Second Conference on Afro-Asiatic Languages, Nice. 1996. Ablaut in Classical Arabic measure I active verbal forms. In J. Lecarme, J. Lowenstamm, U. Shlonsky (eds.) Studies in Afroasiatic Grammar, The Hague: Holland Academic Graphics, 123–34. Henry, Victor. 1900. Lexique étymologique des termes les plus usuels du breton moderne. Rennes: Plihon et Hervé. Jouitteau, Mélanie. 2009–2023. ARBRES, site de recherche sur la syntaxe formelle de la microvariation syntaxique de la langue bretonne. CNRS. https://arbres.iker .cnrs.fr Jouitteau, M. 2015. Free choice and reduplication, a study of Breton dependent indefinites. In Tomasz Czerniak, Maciej Czerniakowski & Krzysztof Jaskuła (eds.) Representations and Interpretations in Celtic Studies, Lublin: Wydawnictwo KUL, 201–30. Matasovi
c, Ranko. 2009. Etymological Dictionary of Proto-Celtic. Leiden: Brill. Meinard, Maruszka Eve Marie. 2015. Distinguishing onomatopoeias from interjections. Journal of Pragmatics 76, 150–68.

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

A survey of expressive words in Breton

309

Menard, Martial. 1995. Alc’hwez Bras ar Baradoz Vihan, Geriahudur ar brezhoneg. An Here. Menard, Martial & Hervé Le Bihan. 2016–. Devri: Le dictionnaire diachronique du breton. Université Rennes II and Kuzul ar Brezhoneg. http://devri.bzh Scheer, Tobias. 2000. De la Localité, de la Morphologie et de la Phonologie en Phonologie. Thesis, University of Nice Sophia Antipolis. Ségéral, Philippe. 1995. Une théorie généralisée de l’apophonie. Thesis, University of Paris 7. 1996. L’apophonie en ge’ez. In J. Lecarme, J. Lowenstamm, U. Shlonsky (eds.) Studies in Afroasiatic Grammar, The Hague: Holland Academic Graphics, 360–91. 2000. Théorie de l’apophonie et organisation des schèmes en sémitique. In Jacqueline Lecarme, Jean Lowenstamm & Ur Shlonsky (eds.) Research in Afroasiatic Grammar. Amsterdam and Philadelphia: Benjamins, 263–99. Ségéral, Philippe & Tobias Scheer. 1996. Modern German and Old English strong verbs: two ways of running apophony. Communication au colloque Generative Grammatik des Südens, Berlin, 17–19 May. 1997. Apophonic theory and Cushitic languages. Communication au colloque GLOW, Rabat, 19–21 March. 1998. A generalized theory of Ablaut: The case of Modern German strong verbs. In Albert Ortmann, Ray Fabri & Teresa Parodi (eds.) Models of Inflection, Tübingen: Niemeyer, 28–59.

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.014 Published online by Cambridge University Press

Vasconian

https://doi.org/10.1017/9781108989084.015 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.015 Published online by Cambridge University Press

10

Vindicating the role of ideophones as a typological feature of Basque Iraide Ibarretxe-Antuñano*

10.1

Basque and its ideophone repertoire: A brief history

Basque is an isolate language in southwestern Europe.1 It is currently spoken by around 900,000 speakers distributed across both sides of the western edge of the Pyrenees. Apart from a very rich dialectal variation (see Zuazo 2013, Camino 2019) and an unknown phylogenetic origin (see Trask 1997, Martínez Areta 2013, Gorrochategui et al. 2018), Basque is an often-cited and wellknown language among language scientists due to some of its distinctive linguistic characteristics; namely, its ergativity, case alignment, and double marking (see, Igartua 2020 for a review). Another distinctive feature of the Basque language is its rich ideophone repertoire. Current dictionaries of Basque ideophones (Ibarretxe-Antuñano 2006, Santisteban 2007) have compiled more than 4,000 ideophonic entries, which is just an estimate of the real size of such a repertoire, since Basque speakers, as happens with other ideophonic speakers, create ideophones on the spot depending on their needs (e.g., tita-tita for typing, tak tak tak for computer typing). The large repertoire of ideophones bestows on Basque its categorisation as a highly-ideophonic language. This fact serves as a good case in point to refute the general claim for the limited distribution of this type of words to languages in Asia, Africa, and the Americas (see Dingemanse 2022) as well as to vindicate ideophones as an important typological trait of Basque, on the same level as its other morphosyntactic typological characteristics. The lack of visibility of ideophones in these two contexts may be partly explained if one considers the little interest that these linguistic items have attracted in typological studies in general, and in Basque linguistics in particular. This neglect is nevertheless natural since, until very recently, ideophones were hardly explored in these cross-linguistic general linguistics contexts. This situation has been gradually reversed since the late 1990s due, among other * University of Zaragoza 1 This research has been funded by the Spanish Government (AEI/FEDER FundsFFI2017–82460-P; PID2021–123302NB-I00), the Government of Aragon (Psylex H11–17R; MultiMetAR LMP143_21), and the Iberus Campus (ICON action group).

313

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

314

Iraide Ibarretxe-Antuñano

factors, to the publication of several key books that include descriptive research on ideophones from all over the world (see, for example, Hinton et al. 1994, Voeltz & Kilian-Hatz 2001, Akita & Pardeschi 2019). In Basque linguistics, the study of ideophones shares a similar path. After a shiny period at the beginning of the twentieth century, when the key role of ideophones (usually called onomatopoeia) appeared in traditional descriptive Basque studies, these linguistic elements fell into oblivion in dominant linguistic approaches to Basque linguistics until the end of the century. At present, ideophones are enjoying a timid renaissance in Basque linguistics (see Ibarretxe-Antuñano 2017: 198–9 for an overview). This chapter aims at contributing to this ‘ideophonic renaissance’ by bringing to the fore the singularity of this special word class in Basque. Section 10.2 briefly explains the concept of ideophone from a typological perspective and illustrates what these words feel like in Basque by means of the ideophone mara-mara. In the rest of the chapter, §10.3 is devoted to the exploration of the multimodal structure, §10.4 the meaning, and §10.5 the use of Basque ideophones. 10.2

Ideophones in general and ideophones in Basque

Ideophones are marked linguistic elements due to their foregrounded structural characteristics and their compact and multifaceted depictive meaning (Dingemanse 2012, Ibarretxe-Antuñano 2017, 2019, 2022a, Akita & Dingemanse 2019). In other words, ideophones display salient linguistic characteristics that distinguish them from other word categories in their respective languages. What counts as ‘linguistically salient’ depends on each language, but usually among these prominent features one may find unusual phonotactics; specific phonemes, morphemes, and word orders; reduplication; syntactic looseness; multi-categorial properties; gestural patterns; inter alia. In addition, at the same time, ideophones provide a vividly expressive and precise depiction of the entity or event they refer to. That is, they descriptively inform about the entity or event, as well as depictively enact that information. This is why ideophones fulfil representative in addition to expressive and aesthetic functions. Basque ideophones fit into this general typological characterisation. Table 10.1 shows a speaker describing a typical Basque ideophone: maramara. Formally, this ideophone exhibits a total reduplication of a two-syllable morph with a nasal bilabial consonant in initial position, a second-syllable stress pattern together with an iconic gesture. As far as its meaning is concerned, mara-mara refers to a downward, soft, continuous movement of small light objects. Mara-mara, when used in a literal sense, usually refers to snow as in Table 10.1; when used metaphorically, mara-mara highlights the ‘gentleness’, ‘softness’, and ‘abundance’ character of the event described.

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

Ideophones as a typological feature of Basque

315

Table 10.1 The ideophone mara-mara mara-mara https://osf.io/gk9ev

Oral text Mara-mara eztau inoiz eitten eurik, mara-mara eiten deu beti elurrek eta da ba, luman modure, xuabe xuabe berantza datorrelako. Mara-mara eiteko, elurre ixan behar dau, ez dau balio txotorrak edo ez dau balio harrik edo ez dau . . . beste elurre mota batzuk ez dute mara-mara eitten; mara-mara dator elur, elur-lumaka beratuten direnen, es mara-mara English (semi-literal) translation ‘Rain never makes mara-mara, it is snow that always makes mara-mara and it is, well, similar to feathers, because it comes softly softly downwards. In order to make maramara it has to be snow, hail is not possible or hailstone is not possible or no . . ., any other type of snowfall does not make mara-mara; mara-mara comes, when snow, em, snowflakes descend, it is mara-mara.’

The ideophone mara-mara in Table 10.1 is a good illustrative example of how ideophones work in Basque. The following sections provide a succinct typological characterisation of these words in Basque from three viewpoints: their multimodal structure, their meaning, and their use. 10.3

The multimodal structure of Basque ideophones

The structure of Basque ideophones shares some of the key typological features that are generally used in characterising these words from a crosslinguistic perspective. This section offers a description of the phonological, morphosyntactic, and gestural structure of Basque ideophones. 10.3.1

Mimetic vowel harmony

This process refers to the repetition of the same vowel sound (Akita et al. 2013) and it is frequently found in ideophones as illustrated in (1a). Disharmony or vowel alternation is also typical. The i-a vowel alternation is the most frequent pattern, followed by the i-o pattern as exemplified in (1b): (1)

a. Plasta-plasta ‘crashing down’, tetele-metele ‘without thinking’, pil-pil ‘simmer; snowflake; palpitation’, bro-bro ‘boil hard’, zurrumurru ‘gossip; whisper’ b. Plisti-plasta ‘wade’, zirt-zart ‘slashing; crackling; shine’, binbili-bonbolo ‘gently; rocking’

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

316

Iraide Ibarretxe-Antuñano

10.3.2

Palatalisation

This phonetic process is a general feature of the Basque language (Michelena 1985: ch. 10, Hualde & Ortiz de Urbina 2003, 37–40, de Rijk 2008: 13–15). Conditioned or automatic palatalisation occurs in contexts where the vowel /i/ precedes /n/, /l/, and /t/ as in baina /´baɲa/. This palatalisation shows a great deal of dialectal variation in both usage and distribution. Western dialects are known for their palatalisation preference. Affective and expressive palatalisation, on the other hand, is a general widespread productive mechanism to create diminutives, augmentatives, and affective words with a positive or negative valence (e.g. gozo ‘sweet’ and goxo ‘very sweet, pleasant, lovely’). In the case of ideophones, palatalisation is also used with expressive purposes. That is to say, when the meaning of the ideophone needs to be intensified. For instance, tipi-tapa means ‘to walk in small and quick steps’. Its palatalised form, ttipi-ttipa, keeps the same meaning but adds an extra degree of smallness and quickness to these steps. All ideophones whose phonetic structure admits palatalisation may undergo this process for expressive purposes. This palatalisation is reflected in the orthography in writing.2 Example (2) presents some illustrative cases. (2)

a. Voiceless apico-dental stop /t/ t ! voiceless predorsal palatal stop /c/ tt Ter-ter ‘little by little; disorderly way’ ! tter-tter ‘small steps’ b. Voiced apico-dental stop /d/ d ! voiced predorsal palatal stop /ɟ/ dd Dalan-dalan ‘completely full’ ! ddalanddal ‘even more completely full’ c. Voiceless lamino-alveolar fricative /s̻/ z ! voiceless palato-alveolar fricative /ʃ/ x Zafla ‘slice’ ! xafla ‘poultice, plaster; slice’ d. Voiceless apico-alveolar fricative /s̺/ s ! voiceless palato-alveolar fricative /ʃ/ x Brist ‘sudden and quick movement’ ! brixt ‘sudden disappearance’ e. Apico-alveolar lateral /l/ l ! palato-alveolar /ʎ/ ll Lapa-lapa ‘lick’ ! llapa-llapa ‘slurp down’ f. Apico-alveolar nasal /n/ n ! palato-alveolar nasal /ɲ/ ñ Nirro ‘half-closed eye person’ ! ñirro (ñirro-ñarro) ‘short-sighted’

Palatalisation also occurs in other sets of ideophones whose meaning is not necessarily loaded with an extra charge of expressivity. This is the case with ttakun-ttakun ‘the sound of the Basque instrument txalaparta’ or ñir-ñir ‘gleam, twinkling’. 10.3.3

Unusual phonology

This refers to cases where some sounds, sound clusters or sound phonotactic environments are exclusive (or frequent) to ideophones. In Basque, a 2

For the sake of exposition, whenever needed, the phonemic symbol will be followed by its grapheme. Expressive palatalisation is often orthographically represented with a repetition of the consonantal grapheme as in Examples (2a) and (2b).

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

Ideophones as a typological feature of Basque

317

distinctive sound that usually only occurs in ideophonic words is the voiced lamino-alveolar affricate /dz/ dz. Another example is the diphthong /iṷ/, usually absent in prosaic words but present in some ideophones such as briu-brau ‘energetically and quickly’. The appearance of some sounds and sound clusters in specific phonotactic positions is also recurrent in ideophones. This is the case of affricates (/ts̻-/ tz-, /tʃ-/ tx-) and palatals (/c-/ tt-, /ɲ-/ ñ-, /ʃ-/ x-) in word-initial positions (Hualde 1991: 12). Similarly, the word-initial position of sounds such as plosives, the voiceless labio-dental fricative, the voiceless laminal alveolar fricative and nasals is also typical of these words. These sounds are frequent in other positions in Basque, but their word-initial position is a clear sign of either an ideophone or non-native words and recent loanwords. In addition, consonant clusters formed by plosives or /f/ f plus liquids in wordinitial and word-final positions are also a give-away for ideophones (Trask 1997: 258). Tables 10.2 and 10.3 summarise these phonological characteristics together with illustrative examples. Table 10.2 Word-initial position phonemes in Basque ideophones Phonemes and clusters

Example

Affricates

dzist-dzast ‘poke repeatedly; work carelessly’ txir-txir ‘fry; creak; chatter’ tzillo-tzallo ‘shuffle’ pilpil-pulpul ‘palpitation’ giri-giri ‘dive, swim’ briu-brau ‘energetically and quickly’ furrust-farrast ‘roaring’ xixta-mixta ‘lighting’ zipirti-zaparta ‘left, right, and centre’ zurrust ‘swallow, gulp’ mela-mela ‘completely soaked’ nistiki-nastaka ‘hodgepodge, jumble’

Voiced lamino-alveolar /dz-/ dzVoiceless palato-alveolar /tʃ-/ txVoiceless lamino-alveolar /ts̻-/ tz-

Stops (/p-/ p-, /t-/ t-, /k-/ k-, /b-/ b-, /d-/ d-, /ɡ-/ g-) Diphthongs Fricatives Voiceless labiodental /f-/ fVoiceless prepalatal sibilant /ʃ-/ xVoiceless lamino-alveolar sibilant /s̻/ zNasals (/m-/ m-, /n-/ n-, /ɲ-/ ñ-)

Table 10.3 Consonant clusters in Basque ideophones Position

Consonant cluster

Example

Onset

Stop + liquid

Coda

/f-/ f- + liquid Liquid/sibilant/nasal + stop

blai-blai ‘soaked’ gran-gran ‘train clattering’ fliu-flau ‘quickly and energetically’ zart ‘snap; crack’ tenk ‘steady; tense, taut; come to a stop’ laprast ‘slip, slide’

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

318

Iraide Ibarretxe-Antuñano

10.3.4

Syllabic structure

The number of syllables in Basque ideophones is varied. The great majority of ideophones are one-, two-, and three-syllable, but four- and even five-syllable items are also possible as illustrated in (3). Stress varies across dialectal varieties. It is fixed on the penultimate syllable in Souletin. For most of the other dialects, it is first syllable in monosyllabic stems (zárt ‘slap’ ! zárta ‘a slap’) and in vowel-final disyllabic (záfla ‘slice’). In other cases, it follows a second-syllable stress pattern (irríst ‘slide, slid’) and, when reduplicated, it falls on the first morph as shown in Figure 10.1. Di-da ‘beat; proceed’, zart ‘snap, break; slap, hit; suddenly’ Mara-mara ‘softly, gently’, irrist ‘slide, slid’, zart ‘slap’ Zingulu-zangulu ‘shuffle’ Hirrinbili-harranbala ‘move awkwardly; clumsy’ Tibiribiri-tibiribiro ‘chattering away’

(3)

a. b. c. d. e.

10.3.5

Reduplication

This morphological process is one of the ubiquitous characteristics of ideophones across languages. Reduplication refers to the repetition of phonological material within a word for semantic or grammatical purposes (Rubino 2011). When the repetition of a single morph takes place in its total form, it is called total reduplication; partial reduplication, on the other hand, refers to the partial repetition of a single form and it comes in a wide array of forms (vowel lengthening, consonant gemination, consonant insertion, and so on). Reduplication, both total and partial, is a pervasive process in Basque. It is commonly used with a distributive meaning (egunean-egunean ‘each day’), quantifier meaning (handi-handia ‘very big’) as well as an expressive meaning (saski-naski ‘mess, jumble, mixup’) (Hualde 2003). Both total and partial reduplication are common in Basque ideophones as illustrated in (4a) and (4b), respectively. (4)

a. Total reduplication CCVC1CV2 – CCVC1CV2: plasta-plasta ‘crashing down’ CV1CV2 – CV1CV2: kili-kili ‘tickle’ CVC1CV2 – CVC1CV2: gurka-gurka ‘in gulps’ b. Partial reduplication CiC1Ci2Ci3 – CaC1Ca2Ca3: pinpili-panpala ‘favourite’ Ci1CiC2 – Co1CaC2: bilin-bolan ‘tumbling, toppling, turning over’ CVC1CV2CVC3 – mVC1CV2CVC3: zirkurin-mirkurin ‘complaint, moan’ V1CV2 / mV1CV2: irrimirri ‘weak’ Ci1Ci2Ci3 – pa1Ca2Ca3: txitxili-patxala ‘complaint, moan’

As shown in (4b), in the case of partial reduplication, the second morph undergoes different types of processes: vowel alternation (see §10.3.1),

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

Figure 10.1 Stress in polysyllabic ideophones: mara-mara and irrist

320

Iraide Ibarretxe-Antuñano

consonant alternation, consonant insertion, and all of these processes at the same time. In consonant insertion cases, the most frequent pattern is the inclusion of a bilabial nasal /m-/ m- (see Igartua 2013). Repetition of morphs in Basque ideophones is not restricted to duplicated forms. Basque ideophones allow the multiplication of forms for iconic expressive functions: the higher number of repetitions, the more intensive the meaning. For instance, pla means ‘one slap’, pla-pla ‘two slaps’, but pla-pla-pla ‘continuous slapping’. There are, however, other ideophones where the number of repetitions is lexicalised as in ter-ter-ter ‘in line; one after the other’, dra-dra-dra ‘monotonous speech’, and fil-fil-fil ‘fall down in circles and slowly’. 10.3.6

Open lexical word class

One of those much-debated questions in ideophone literature has been the word class status of ideophones. Some authors have argued for a distinctive and separate word class (e.g., Doke 1935, Awoyale 1989, Kabuta 2001), whereas others have distributed ideophones across traditional categories (noun, verb, adjective, adverb, etc.) since they fit their core definitional functional properties (Amha 2001, de Jong 2001). In Basque, as the ideophone marmar illustrates in (5),3 ideophones may be said to form a distinctive lexical word class of their own. They are used as bare elements, that is, they do not take any explicit morphological marking, and their syntactic function is only revealed once they are put into context (see Exx. (5a) and (5b)). That is to say, some ideophones are multi-categorial. However, on the other hand, some of these ideophones may also be found in compound formations by means of support verbs such as egin ‘do, make’ (Ex. (5c)) or derived by means of different morphemes such as the deadverbial suffix -ka ‘iterative action’ (Ex. (5d)) or spatial cases such as the locative -n (Ex. (5e)). (5)

Marmar a. Marmar as noun ‘rumour; whisper; grumble’ Beraz, herritarren errua ere izango da therefore citizen.gen fault.abs too be.fut aux.3sg marmar honekin jarraitzen badugu ide this.com continue.hab if.aux.1pl ‘Therefore, it will be people’s fault, if we keep on [spreading] this rumour’ b. Marmar as adverb ‘whispering; grumbling’ Bere buruarekin mintzo zen marmar 3sg.gen head.com hitz aux.3sg ide Lit. ‘S/he talked marmar (grumbling) with his/her head’ ‘S/he was grumbling to him/herself’

3

All examples are taken from the Corpus of Contemporary Basque (ETC) at www.ehu.eus/etc

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

Ideophones as a typological feature of Basque c. Marmar + egin ‘to groan; to whisper; to grumble’ Nire ondoan doan gizonak marmar I.gen side.loc goes.rel man.erg ide du: ‘Ez dago eskubiderik’ aux.3sg neg is right.part ‘The man next to me grumbled: “It’s not fair”’ d. Marmar + -ka ‘whispering; grumbling’ Mutilek hartaz marmarka jardun boys.erg that.inst ide.ite be.busy.doing ‘The boys were busy gossiping about that’ e. Kexarik badute, marmarrean hasten complaint.part if.have.3pl ide.loc start ‘If they have complaints, they start grunting’

321

egin make

zuten aux.3pl

dira aux.3pl

As recently argued by Dingemanse (2022), the increasing body of crosslinguistic and typological research on ideophones has made clear the need for considering ideophones as an open lexical word class on their own. Ideophones may share with traditional word classes some of their grammatical and functional properties. However, despite these similarities, their structurally marked characteristics, versatile grammatical nature, productivity ratios, and their depictive sensory meanings provide ample evidence to justify considering them as a distinctive lexical word class. 10.3.7

Morphosyntactic integration

Another controversial issue in the ideophone literature is the degree of ideophone integration in the morphosyntactic structure of the utterance. Some authors have argued that ideophones are independent, non-integrated items, which stand aloof in the sentence (Kunene 1965). Other authors have proposed that it might be a question of degree and the type of ideophone construction: the more expressive ideophones are, the less grammatical integration they will show (Dingemanse & Akita 2017). Basque ideophones cover a wide range of integration possibilities. As shown in Example (5) in §10.3.5, marmar illustrates cases where the ideophone is fully integrated in other constructions as a nominal complement in (5a) or as a predicative do-verb in (5c) (marmar egin). Thanks to different derivational and compounding processes, which are highly productive in Basque, ideophones become more syntactically integrated forms as in tinkitankari ‘what produces a continuous beating’ from tinki-tanka ‘beating; hammering’ + -(l)ari ‘agent’ or zarraparratsu ‘noisy’ from zarraparra ‘turmoil’ + -tsu ‘characteristic’. As also illustrated in §10.3.5, the same ideophone marmar may also occur in less syntactically integrated cases as in Example (5b): marmar acts as a non-

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

322

Iraide Ibarretxe-Antuñano

Figure 10.2 The ideophone taka-taka

compulsory adverbial element. This is also the case of Example (6) represented in Figure 10.2. The ideophone taka-taka ‘walk in regular and small steps’ is performed at the end of the utterance together with a co-speech gesture4 that iconically represents each step. This results in a more vivid (expressive) depiction of the scene. (6)

Eta ixil-ixilik ixil-ixilik han joan dira, and quiet-quietly quiet-quietly there.loc go aux.3sg [taka taka taka taka] ide ‘And there they went very quietly taka taka taka taka [walking in regular and short steps]’

Ideophones in Basque can also stand independently in Example (7). The speaker uses the ideophone pla-pla together with the co-speech gesture (both 4

The speech and gesture co-articulation moment is marked with [ ].

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

Ideophones as a typological feature of Basque

323

Figure 10.3 The ideophone pla-pla

hands turning up and down quickly) to describe a traditional way of preparing anchovies (open and battered). The ideophone depicts the moment when the cook takes the open anchovy and swiftly batters it in flour before deepfrying it. (7)

Pla-pla ra Ondarruko berbi. Pla-pla eindde, ide aux.3sg ondarroa.adn word ide make.aux.3sg [pla-pla pla-pla] ide ide ‘Pla-pla is a word from Ondarroa (Western dialect village). Pla-pla is done. Pla-pla pla-pla’

All of these examples seem to indicate that ideophones might be situated alongside a cline of syntactic integration. A cline that, as suggested by Dingemanse and Akita (2017), might go hand in hand with the degree of expressiveness. Ideophones such as pla-pla are highly expressive, whereas ideophones in derived forms such as marmarka might be perceived as less expressive by speakers. In fact, some of ideophonic derived forms (e.g., ide+-tu/-du verbs such as labandu (laban ‘slip’-du) ‘to slip’, bildu ‘cover, wrap’ (bil(-bil) ‘circle/wrapping’-du), have undergone such an de-ideophonisation process, that they are no longer considered ideophones but prosaic words by speakers. 10.3.8

Gesture

When orally produced, as already mentioned in §10.3.6, ideophones are usually co-articulated with gestures (Kita 1993, Dingemanse 2013, IbarretxeAntuñano 2004, 2017). In Figure 10.4, the speaker, while explaining the meaning of mara-mara (see Table 10.1), moves both his hands in a zig-zag soft and downward movement that clearly depicts the essence of this ideophone.

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

324

Iraide Ibarretxe-Antuñano

Figure 10.4 The ideophone mara-mara and gesture Source: https://osf.io/gk9ev

10.4

The meaning of Basque ideophones

The non-arbitrary gradual relation between meaning and form in ideophones allows the distinction between ‘conventional’ and ‘sensory’ ideophones. The former are closer to onomatopoeia, that is, there is a direct imitation of a sound. The latter do not necessarily reproduce the sound of the event, entity, or characteristic they refer to. Basque is rich in both types. Conventional ideophones cover a wide range of semantic fields such as weather phenomenona (burrunba ‘thunder’), animal sounds (kurrin ‘pig grunt’), instruments (txintxirri ‘rattle’), and physical events such as hitting (zapla ‘slap’), ingestion (klaska-klaska ‘wolf down’), and destruction (zirrist-zarrast ‘saw’) (see Ibarretxe-Antuñano 2023). Sensory ideophones are also central to the ideophone repertoire for light (dir-dir ‘shine; dazzle’), events (bristi-brasta ‘work

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

Ideophones as a typological feature of Basque

325

Table 10.4 Semantic fields in Basque ideophones Category

Subclass

Example

Actions and events

Motion Communication Light Sound

tinkun-tankun ‘walk clumsily’, antxintxi egin ‘run’ xuxu-muxu ‘whispering’, txatxala-patxala ‘ramble’ nir-nir egin ‘shine’, zirrinta ‘ray, beam’ brinbraun ‘clang’, gilin-gilin ‘sound of small cowbell’, zirris-zarras ‘sound of sawing’ zausta-zausta ‘wolf down’, hurrup ‘slurp’ birrin-birrin ‘devastate; tear’, sisti-sasta ‘sting’ blisti-blasta ‘slapping’, furrust-farrast ‘roar’, panpapanpa ‘hit continuously’ gal-gal ‘boil’, txil-txil ‘soft boiling’ txir-txir ‘fry’, pla-pla ‘battered’ irri egin ‘laugh’, intziri-mintziri ‘sob’ tarrat ‘fart’, pilpil-pulpul ‘palpitate’ firristi-farrasta ‘work carelessly’ mela-mela ‘soaked’, pla ‘withered’ bil-bil egin ‘wrap tightly’ burrun burrun ‘bumblebee’ karramarro ‘crab’ txantxangorri ‘robin’ klunklun ‘toad’ perpelete ‘gilthead’ igiri-migiri ‘otter’ txantxar ‘henbane’, ziza ‘mushroom’ xirimiri, zirzira ‘drizzle’, xixta-mixta ‘lighting’ dunbala ‘drum’, txintxirri ‘rattle’ farras ‘slovenly’, zirriborro ‘smudge’ kokolo-mokolo ‘idiot’, sinkulin-minkulin ‘wimpy’ garranga ‘hook, bait’, firinda ‘pulley’ kiribil ‘spiral’ tunt ‘not a thing’, tzirtzil ‘unimportant thing’ txitxi ‘meat’, mau-mau ‘eat’, ttotto ‘dog’ barrasta-barrasta ‘profusely’, dalan-dalan ‘full’ brenk ‘precipitous mountain’ txitxil ‘penis’ kinkirrinkon ‘champaine’

Ingestion Destruction Hitting

Animals

Boiling Cooking Emotions Body functions Badly-performed States Miscellany Insects Crustaceans Birds Anphibians Fish Others

Plants Weather Musical instruments Characteristics Physical Psychological Gadgets Things General Low value Child language Quantity Nature Sexual terms Miscellaneous

Adapted and updated from Ibarretxe-Antuñano, 2017: 205–6.

carelessly’), and descriptive characteristics (sinkulin-minkulin ‘wimpy’, hilihili ‘clumsily; with no energy’, barra-barra ‘profusely’), among others. Table 10.4 summarises some of the main semantic areas covered by ideophones in Basque. From a semasiological viewpoint, ideophones are quite polysemous. They may encode a wide array of meanings and contexts. For instance, bar-bar

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

326

Iraide Ibarretxe-Antuñano

means ‘drink in gulps; sound of bubbling water; cracking; sound of an insect’s walk; jabbering, talking; rhythmic falling of a light body; in droves; and noise, din’. Ideophones often encode metaphorical meanings as in bolo-bolo ‘spreading (news, gossip. . .)’. Sometimes, as happens for prosaic words, these metaphorical semantic extensions are conceptually motivated and select some of the defining properties of the non-metaphorical meaning. For example, parraparra means ‘gush out’; when used in relation to spending, for instance, it means ‘profusely’ (that is, it selects the ‘quantity’ part of the meaning), but when apply to speaking, it could be talkative but also ‘frankly (speaking)’ (that is, it selects the ‘suddenness’ part, instead). From an onomasiological viewpoint, ideophones present a great deal of (partial) synomymy. It is usual to find groups of ideophones that describe the same general event, and therefore, they can be taken as synonyms in certain contexts. However, on many occasions, their interchangeability is quite partial since their semantics usually includes certain specific nuances, which are sometimes overlooked in dictionaries but evident to speakers. An illustrative example in Basque is the group of ideophones that refer to ‘boiling’. All ideophones in (8) refer to this event, but as shown in their definitions, some refer to specific types of boiling. Definitions, adapted and expanded, are drawn from Azkue (1905) and Mitxelena’s (1987–2008) dictionaries. (8)

Boiling a. pil-pil (pill-pill; pir-pir) ‘movement of a liquid when boiling softly, small bubbles; superficial boiling; slow boiling typical of sauces when cooking’ b. txil-txil ‘soft boiling’ c. pul-pul ‘sound of external boiling; different from bol-bol’ d. bol-bol ‘sound of internal boiling; different from fil-fil ‘simmer’ or txir-txir ‘soft boiling, usually in oil (fry)’ e. pol-pol (poll-poll) ‘noisy boiling’ f. bor-bor (gor-gor; bro-bro) ‘deep boiling’ g. bal-bal ‘strong boiling’ h. gar-gar(-gar) ‘boiling on strong fire’ i. gal-gal(-gal) (kal-kal; khal-khal) ‘boiling with big bubbles’

10.5

The use of Basque ideophones

Ideophones in Basque are ubiquitous: they turned up in all registers (formal-informal), in all contexts (oral-written), at all ages, and in all dialects. In this respect, the dialectal distribution of ideophones may offer different possibilities. Some ideophones may be typical in one variety and less known to others (e.g. pla-pla ‘anchovy cooking’ (see Figure 10.4) in Ondarroa (Bizkaia, Western Basque) or dzingua ‘sudden seawater increase’ in Lekeitio (Bizkaia, Western Basque). Some other ideophones are known cross-dialectally but

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

Ideophones as a typological feature of Basque

327

spelled out differently; for instance, the ideophone dizdiz (dirdir, dizdir, birbir, sirt-sart . . .) for ‘shine, dazzle’. Finally, several ideophones may share similar meanings as ‘slide’ and ‘slip’, but not the dialectal distribution: laban is the preferred choice in western Basque and irrist in central Basque. As far as their linguistic functions are concerned, ideophones fulfil a representational function as has been shown in the examples used in the preceding sections. Some objects (taratulu ‘drill’), some animals (karramarro ‘crab’), some phenomena (xirimiri ‘drizzle’), and some actions (irrist ‘slide’) are only encoded by means of ideophones. There is no other way of encoding them in the language, or at least, the ideophone is the prototypical way of naming them. What is more, ideophones are indispensable vehicles that represent complex, detailed, and specific meanings as shown in the case of boiling in Example (8). Ideophones, nevertheless, also fulfil an important aesthetic function due to their compact meaning and expressivity (Ibarretxe-Antuñano 2022a). This is why they frequently occur in all sorts of creative manifestations of culture, both traditional and modern (songs, poems, riddles, etc.; see IbarretxeAntuñano 2020, Lecuona 1964) as well as in all the forms of the linguistic landscape (public and private signs, advertising, etc.; see, Ibarretxe-Antuñano 2022b). Table 10.5 illustrates some real examples of the use of ideophones in Basque. Finally, yet importantly, ideophones can be also considered aesthetic catalysers that produce an intimate interlocutive involvement among the ideophonic language speakers (Ibarretxe-Antuñano 2020, 2022b), an involvement that facilitates intertextuality and a group bonding effect within their linguistic community. As Childs (1998) has suggested, ideophones are often recognised by their speakers as part of their identity and, in Basque, ideophones are felt as a manifestation of ‘Basqueness’. Consequently, they are not only frequently used in the local linguistic landscape as shown in Table 10.5, but also in ‘outside’ contexts where what needs to be foregrounded is the Basque ‘essence’ or ‘character’. Figure 10.5 shows an example of this function of ideophones as an illustration of ‘Basqueness’. It reproduces an advertisement for the low-cost airline company Volotea. It belongs to a marketing campaign the airline launched when it started to operate from Bilbao International Airport in 2018. One of the new destinations was Málaga, Spain. The main text reads: ¿Txombo en la Malagueta y volvemos? ‘Txombo on the Malagueta [Malaga famous beach] and we come back?’ The advertisement includes the ideophone txonbo (spelled with an m according to Spanish orthographic rules). It means ‘dive headfirst into the water’ and it is part of the Bilbao lexicon.

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

328

Iraide Ibarretxe-Antuñano

Table 10.5 The aesthetic power of Basque ideophones Songs, poems

Public signs

Popular song Tiriki tauki tauki / Arrankin trankin trankun with different variations5 and adaptations (Txirri, Mirri eta Txiribiton clowns version for kids and the poet Bitoriano Gadiaga’s poem)

Tiriki-tauki ‘hit on an anvil’ and tauki-tauki ‘hammering’ Road and traffic signs

Lit. ‘Careful with the one in front, risk of making talka [collide, crash]’ ‘Minimum space between cars’

5

The Xenpelar Dokumentazio Zentroa, the information centre for bertsolaritza [improvised Basque verse singing] compiles up to five different versions of this popular folk song. All information about this song (sources and downloadable music files) available from https://bdb .bertsozale.eus/en

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

Ideophones as a typological feature of Basque

329

Table 10.5 (cont.) Public campaigns

European-founded anti-rumours network for diversity (C4iCommunication for Integration). Bilbao campaign.

Lit. ‘Zurrumurruak [rumours] (Sp. Don’t let them seep through you! Rumours?) Blai [soaked] in information!

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

330

Iraide Ibarretxe-Antuñano

Table 10.5 (cont.) Advertising

Foreign language school Bai&By

Branding and naming

Lit. ‘The Marquis of Bai&By delivers certificates barra-barra [profusely]! Cambridge First and Advanced and all HABE levels [Basque certification]’ Shops and business

Mara-mara [softly, abundance] bookstore; ZurruMurru [rumour] bookstore; Dir-dir [shine] jewel designer’s; Tiriki-tauki [strike on an anvil] jewellery; Talka [collide] shoe repairs; Dzingua [sudden seawater increase] surf shop; Tipi-tapa [walk in small and quick steps] cafe; Txurrut [swallow, gulp] cafe; Simiriri [drizzle] hotel.

10.6

Conclusions

This chapter has offered an overview of ideophones in Basque and has shown that Basque ideophones form an open lexical word class in this language. They are ubiquitous: they turn up in all contexts, registers, and dialects. They share a set of structural properties such as vowel harmony (plasta-plasta), palatalisation (ttipi-ttapa), specific phonotactics (dzist-dzast), or reduplication (fil-fil-fil). As lexical items, they may undergo morphological processes such as derivation (zarraparratsu) and compounding (irrist egin). However, they can also turn up in bare conditions, that is, lacking any morphological overt

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

Ideophones as a typological feature of Basque

331

Figure 10.5 Ideophones as a sign of Basqueness

marking (furrust-farrast). This is why ideophones, and sometimes the same ideophone, may fulfil different functions associated with other word classes (nouns, verbs, adjectives, adverbs, interjections) and, consequently, show different degrees of syntactic integration in the structural contexts they appear

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

332

Iraide Ibarretxe-Antuñano

in. Ideophones are typical in semantic domains that describe actions and events (motion, communication, destruction, ingestion, light, etc.), objects (instruments, gadgets, etc.), nature (animals, plants, weather, etc.), and physical or psychological characteristics, among others. Their meaning moves along a scale of iconicity: it ranges from the direct onomatopoeic imitation of physical sounds in conventional ideophones (zapla) to the sensory-based concepts in sensory ideophones (hili-hili). Similar to any other open lexical word class, ideophones exhibit semasiological and onomasiological relations. Some ideophones are highly polysemic and figuratively extend their meanings into more abstract domains (e.g. mara-mara). At the same time, groups of partial synonymic ideophones are frequent (e.g. boiling ideophones). As lexical items, ideophones fulfil a representational function. However, quite distinctively from other word categories, their meanings do not just represent concepts (e.g. boiling); meanings are depicted in a compact and vivid manner (pil-pil vs. gar-gar). This is why they also cover expressive functions sometimes relegated to other parts of the linguistic structure of a language (e.g. affective morphology, syntactic topicalisation, or prosody, inter alia). It is precisely their multifaceted nature that turns them into aesthetic catalysers; that is, interlocutive involvement and ‘Basqueness’ group bonding facilitators within and outside the Basque-speaking community. They are useful instruments that have been widely exploited in the creation of culture artifacts, linguistic landscape, and marketing. As proclaimed in the title, the main purpose of this chapter has been to vindicate the role of ideophones as a distinctive typological feature of this language. The comprehensive overview provided in this chapter has demonstrated that Basque ideophones make up a prominent and singular lexical word class in this language. It is therefore about time they regain the place they deserve at the heart of the Basque language. References Akita, K. & Dingemanse, M. (2019). Ideophones (mimetics, expressives). In M. Aronoff (ed.), Oxford Research Encyclopedia of Linguistics. Oxford: Oxford University Press. DOI: 10.1093/acrefore/9780199384655.013.477 Akita, K. & Pardeshi, P. (eds.) (2019). Ideophones, Mimetics, and Expressives. Amsterdam: Benjamins. Akita, K., Imai, M., Saji, N., Kantartzis, K. & Kita, S. (2013). Mimetic vowel harmony. Japanese/Korean Linguistics, 20, 115–29. Amha, A. (2001). Ideophones and compound verbs in Wolaitta. In F. K. E. Voeltz & C. Kilian-Hatz (eds.), Ideophones. Amsterdam: Benjamins, 49–62. Awoyale, Y. (1989). Reduplication and the status of ideophones in Yoruba. Journal of West African Languages, XIX(1), 15–34. Azkue, R. M. (1905–1906). Diccionario Vasco–Español–Francés I–II. Bilbao: La Gran Enciclopedia Vasca.

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

Ideophones as a typological feature of Basque

333

Camino, I. (2019). An overview of Basque dialects. Linguistic Minorities in Europe Online (LME). https://doi.org/10.1515/LME Childs, G. T. (1998). Ideophone variation is tied to local identity. In M. K. Verma (ed.), The Sociolinguistics of Language and Society: Selected papers from SS IX. London: Sage, 36–46. De Jong, N. (2001). The ideophone in Didinga. In F. K. E. Voeltz and C. Kilian-Hatz (eds.), Ideophones. Amsterdam: Benjamins, 120–38. De Rijk, R. P. G. (2008). Standard Basque: A Progressive Grammar. Vol. 1: The Grammar. Cambridge, MA: MIT Press. Dingemanse, M. (2012). Advances in the cross-linguistic study of ideophones. Language and Linguistics Compass, 6(10), 654–72. (2013). Ideophones and gesture in everyday speech. Gesture, 13(2), 143–65. (2022). Ideophones. In E. van Lier (ed.), The Oxford Handbook of Word Classes. Oxford: Oxford University Press. Dingemanse, M. & Akita, K. (2017). An inverse relation between expressiveness and grammatical integration: on the morphosyntactic typology of ideophones, with special reference to Japanese. Journal of Linguistics, 53(3), 501–32. Doke, C. M. (1935). Bantu Linguistic Terminology. London: Longman, Green and Co. Euskaltzaindia. (2005–2008). Euskararen Herri Hizkeren Atlasa (EHHA). Bilbao: Euskaltzaindia. www.euskaltzaindia.eus/component/ehha Gorrochategui, J., Igartua, I. & Lakarra, J. (eds.) (2018). Euskararen Historia/Historia de la Lengua Vasca. Vitoria-Gasteiz: Eusko Jaurlaritza/Gobierno Vasco. Hinton, L., Nichols, J. & Ohala, J. (eds.) (1994). Sound Symbolism. Cambridge: Cambridge University Press. Hualde, J. I. (1991). Basque Phonology. London and New York: Routledge. (2003). Compounds. In J. L. Hualde & J. Ortiz de Urbina (eds.), A Grammar of Basque. Berlin: Mouton de Gruyter, 351–62. Hualde, J. I. & Ortiz de Urbina, J. (eds.) (2003). A Grammar of Basque. Amsterdam: Benjamins. Ibarretxe-Antuñano, I. (2004). Language typologies in our language use: The case of Basque motion events in adult oral narratives. Cognitive Linguistics, 15(3), 317–49. (2006). Hizkuntzaren Bihotzean: Euskal Onomatopeien Hiztegia. EuskaraIngelesera-Gaztelania. Donostia: Gaiak. (2017). Basque ideophones from a typological perspective. Canadian Journal of Linguistics/Revue Canadienne de Linguistique, 62(2), 196–220. (2019). Towards a semantic typological classification of motion ideophones: The motion semantic grid. In K. Akita & P. Pardeshi (eds.), Ideophones, Mimetics, and Expressives. Amsterdam: Benjamins, 137–66. (2020). Ideófonos y poesía. Tropelías: Revista de Teoría de la Literatura y Literatura Comparada, Extra 7, 411–25. (2022a). The aesthetics of Basque ideophones and beyond. Workshop on Typology of Ideophones (WTI-2022). York University, Toronto, and Nagoya University, Japan. 24–25 June. Online. www.youtube.com/watch?v=U274L3Ro5t8 (2022b). El papel de los ideófonos en el paisaje lingüístico bilingüe. I Congreso Internacional de Paisaje Lingüístico. Universidad Pablo de Olavide, Seville. 9–11 November. (2023). Onomatopoeias in Basque. In L. Körtvélyessy & P. Štekauer (eds.), Onomatopoeia in the World’s Languages. Berlin: Mouton de Gruyter.

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

334

Iraide Ibarretxe-Antuñano

Igartua, I. (2013). La reduplicación compleja en euskera: notas acerca de su formación y sus paralelos en otras lenguas. Fontes Linguae Vasconum, 116, 5–29. (2020). Basque among the world’s languages: A typological approach. In E. Santazilia, D. Krajewska, E. Zuloaga & B. Ariztimuño (eds.), Fontes Linguae Vasconum 50 urte. Ekarpen Berriak Euskararen Ikerketari. Nuevas Aportaciones al Estudio de la Lengua Vasca. Pamplona: Gobierno de Navarra/Nafarroako Gobernua, 329–49. Kabuta, N. S. (2001). Ideophones in Cilubà. In F. K. E. Voeltz & C. Kilian-Hatz (eds.), Ideophones. Amsterdam: Benjamins, 139–54. Kita, S. (1993). Language and thought interface: A study of spontaneous gestures and Japanese mimetics. PhD thesis, University of Chicago. Kunene, D. P. (1965). The ideophone in Southern Sotho. Journal of African Languages, 4, 19–39. Lecuona, M. de. (1964). Literatura Oral Vasca. Donostia: Auñamendi. Martínez Areta, M. (2013). Basque and Proto-Basque: Language-internal and Typological Approaches to Linguistic Reconstruction. Bern: Peter Lang. Michelena, L. (1985). Fonética Histórica Vasca. Donostia: Gipuzkoako Foru Aldundia. Mitxelena, K. (1987–2005). Orotariko Euskal Hiztegia. Bilbao: Euskaltzaindia. www .euskaltzaindia.eus/oeh Rubino, C. (2011). Reduplication. In M. S. Dryer and M. Haspelmath (eds.), The World Atlas of Language Structures Online. Munich: Max Planck Digital Library. https:// wals.info/chapter/27 Santisteban, K. (2007). Onomatopeia eta Adierazpen Hotsen Hiztegia. Bilbao: Gero. Trask, R. L. (1997). The History of Basque. London: Routledge. Voeltz, F. K. E. & Kilian-Hatz, C. (eds.) (2001). Ideophones. Amsterdam: Benjamins. Zuazo, K. (2013). The Dialects of Basque. Reno, NV: The Center for Basque Studies University of Nevada, Reno.

https://doi.org/10.1017/9781108989084.016 Published online by Cambridge University Press

Caucasian

https://doi.org/10.1017/9781108989084.017 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.017 Published online by Cambridge University Press

11

Expressive constructions in Georgian and other Caucasian languages Thomas R. Wier*

The study of expressive language has long labored, both literally and figuratively, under an image problem. Ideophones depicting or evoking particular kinds of sensory or perceptual experiences of languages’ speakers have often been shunted to the metagrammatical margins, in part because Standard Average European languages do not grammaticalize them in quite the same way or to the same extent that some languages of Africa, Asia and the Americas often do, but also because early scholars found it difficult to characterize precisely how sensory semantics became iconically encoded as a morphosyntactic class distinct from other parts of the lexicon. In many ways, the languages of the Caucasus at Europe’s outer edge epitomize this tendency. Despite constituting some one-quarter of the linguistic diversity of the continent, the typological properties of these languages such as baroque polysynthesis, ergativity and other non-nominative alignments, extremes in the number of grammatical categories and phonological inventories coded in verbal and nominal morphosyntax have often been treated as anthropologically ‘other’, and not taken into consideration in discussions of Europe’s linguistic heritage. Surveys such as this chapter on expressives in Abkhaz-Adyghean, Kartvelian and Nakh-Daghestanian languages can help place Europe’s diversity in a new light. 11.1

Variation across Caucasian languages

Expressive language in Caucasian languages manifests itself in many different ways. Dingemanse (2012, 2015, 2017, 2018) and Dingemanse and Akita (2017) define ideophones as ‘marked words that depict sensory language’ in the sense that:  they are conventionalized expressions with consistent forms and meanings and not nonce formations or interjections;

* Free University of Tbilisi

337

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

338

Thomas R. Wier

 they typically possess structural properties (of phonology or morphosyntax) that distinguish them from other classes of words;  they depict rather than merely describe an event or state in an affective, performative or mimetic way;  their semantic content encodes sensory information of sight, sound, or a speaker’s internal psychological state. In all these respects, ideophones and related expressive constructions behave in some ways as hybrid classes of words, often partaking in some but not all of a language’s usual repertoire of grammatical properties. In some cases, expressives do not constitute a formally distinct class but nonetheless adopt a kind of statistical profile or suite of traits that sets them apart from other classes of words in a language. Consider two different sets of Georgian lexical items: (1)

Georgian expressive clusters a. bdğvr-ial-eb-s glitter-suff-th-subj.3.sg ‘it glitters/twinkles’ b. bžğlet-a sob-mas.nom ‘sobbing’ c. brč’val-i stabbing.pains-nom ‘stabbing pains, colic’

(2)

Georgian expressive reduplication and consonant gradation a. t’q’ap’a-t’q’up’-i rattle-red-nom b. tkapa-tkup-i clop-red-nom c. dgapa-dgup-i / dgaba-dgub-i tromp/stomp-red-nom

In the first set (1), each word expresses some salient sensory experience, whether by sight (1a), sound (1b), or internal sensation (1c). Each word also manifests another feature for which Georgian is quite famous: extremely complex syllable onsets (Butskhrikidze 2002). Georgian allows up to eight consonants in a row, for example, gv-prckvn-i-s [obj.1.pl-peel-th-subj.3.sg] ‘he is peeling us’. In the second set of words in (2), we see three different kinds of expressive processes: reduplication and consonant and vowel gradation. In all the forms in both (1) and (2), expressive imagery is encoded by a specific kind of phonological process to the extent that it is a salient hallmark of expressives and can be identified as such by native speakers. At the same time, in none of these cases do expressives actually constitute a separate syntactic or morphological category distinct from nouns, verbs, and so on. This is a trait that we will find in many other languages of the Caucasus examined here.

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

Georgian and other Caucasian languages

11.1.1

339

Variation of exponence

Researchers have found other features common among expressive constructions (Dingemanse 2012: 655). Expressives are usually intonationally and phonationally foregrounded, in that they are characterized by higher or lower pitch ranges and/or by phonation types that are not typical or are not grammaticalized as part of a given language’s phonological system, such as breathy voice, creaky voice, or whispered voice. Expressives often also manifest specific kinds of morphological contrasts not found (or not found as productively) in nonexpressive language. One very common feature of such morphological expressiveness is total or partial reduplication, as in (2), but many languages also have dedicated expressive affixes. So for example in Megrelian (Zan; Kartvelian), in (3), an -in formant is often found with expressive verbs denoting emission of sound, often in combination with partial reduplication (Kadshaia & Fähnrich 2001): (3)

Megrelian a. zirzin- buzz b. č’urč’in-whine, cry c. ğumin- whine, howl d. švit’in- whistle e. zuzin- whistle

f. g. h. i.

c’k’irinpičinjujink’ižin-

clank, jingle stop breathing while crying mumble cry, shout

In Bagvalal (Avar-Andic; Nakh-Daghestanian), speakers may indicate that an object somehow fails to live up to expected behavior for that object by the encliticization of expressive particles =dan or =q’ā to the nominal constituents under discussion (Kibrik et al. 2001: 155): (4)

a. muk’u=b=dan hal-i-la hinča=b=dan small=n=expr stalk-obl-sup big=n=expr ‘Such a small stalk for such a big gourd!’ b. o-̄šu-r š’an=r=q’ā ʕarb-da-r this-obl.m-erg small-npl-expr letter-pl-erg ‘He writes in such small letters’

dažik ek’wa gourd be ek’wa qwa-rā-X write-ms.ipf-conv be

In some Caucasian languages, expressiveness has even become grammaticalized with a dedicated slot on the verb. In his discussion of the verbal templatic morphology of Budukh (Lezgic; Nakh-Daghestanian), Alexeev (1994) notes that an expressive slot follows spatial preverbs but comes before negation and gender agreement prefixal slots: It is also possible to identify in Budukh a specific series of expressive prefixes -t’, -č’, ǯ, -c to express either the intensity with which an action takes place or the unpleasant emotions connected with the action, e.g. čet’erˁi, to throw around; čot’onšu, to kick down; yet’erˁi, to strew the lot; q:at’orğu, to rip right apart; q:et’erˁi, to winnow; q: ot’olq’ol, sot’olq’ol, to be torn apart; q:ot’onšu, to trample; sart’arħar, to run; set’erˁi, to spill out; sot’onšu, to make felt; ʕat’orğu, to hem (stitch); ʕut’orğu, to close up hole with a lace; vot’onšu, to crumple; vot’onku, to knead (dough) – all with prefix -t’-, and

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

340

Thomas R. Wier

q:uč’onx̌u, to graft; voč’onx̌u, to shrug; ʕoč’onx̌u, to chase; oč’onx̌u, to clean – all with prefix -č’-, and so on. (Alekseev 1994: 272)

Thus in Budukh, morphosyntactic expressiveness may be seen as a category related to, but distinct from, other systems of spatial and deictic reference. On the other hand, in Sanzhi Dargwa (Dargic; Nakh-Daghestanian), which has an exceptionally rich inventory of expressive verbs (for some examples, see Table 11.1; Forker 2019: 235), ideophones bear little to any discrete reduplication or other morphological cue, but instead are primarily flagged by their presence in a light-verb construction, as in (5): (5)

a. amma ʁaˁʁ r-ik’-ul ca-r but scream f-say.ipfv-icvb cop-f ‘But she is screaming.’ b. liil xurt’ aʁ-ib ca-b all swallow do.pfv-pret cop-hpl ‘[The wolf] swallowed all her sisters.’

ik’ dem.up hel-i-la ruc-be that-obl-gen sister-pl

Indeed, in this variety of Dargwa, in many cases, the only formal cue to the ideophonic character of these constructions is that the ideophone often bears some marked laryngeal feature, such as pharyngealized vowels, uvular consonants or glottalization, since some ideophones may be complements to multiple different light verbs, and vice versa. Table 11.1 Ideophones within light-verb constructions in Sanzhi Dargwa Ideophone

Light verb

Translation

č’aˁm čaˁχ

b-arq’- (n-do.pfv-) b-ik’ʷ- (n-say.ipfv-)

‘chew’ ‘pour’

c’ip

či-r-aʁ- (spr-abl-do.pfv-) b-arq’- (n-do.pfv-)

‘chop off, cut off’

laˁħ, lap’ paˁqaˁr, p’aq’

(ha-)b-arq’- (up-n-do.pfv-) b-uq- (hpl-go.pfv-)

‘flap, wave’ ‘shake off’

Pas Pirχ

b-ik’ʷ- (n-say.ipfv-), b-arq’- (n-do.pfv-) b-arq’- (n-do.pfv-)

‘scatter’ ‘light up’

qːeh q’ac’

b-ik’ʷ- (n-say.ipfv-) b-ikː- (n-bite.pfv-)

‘cough’ ‘gnaw, bite’

b-ik’ʷ- (n-say.ipfv-) b-ax- (n-go-) sːurk’ sːurk’

b-arq’- (n-do.pfv-) b-ik’ʷ- (n-say.ipfv-)

‘press’ ‘rub, polish’

t’aˁq’ χuˁrχ

b-ertː- (n-burst.pfv-) b-ik’ʷ- (n-say.ipfv-)

‘crack, split’ ‘snore’

χʷaˁrt

b-uq- (hpl-go.pfv-)

‘flinch, cringe, wince’

Source: Forker 2019: 236.

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

Georgian and other Caucasian languages

11.1.2

341

Typology as a constraint on exponence of expressives

To a great extent, the differences in how ideophones are expressed in the three autochthonous families are understandably constrained by the overall typological profiles of the languages in question. Abkhaz-Adyghean languages manifest extremes in terms of the phonological inventories that they possess and in the extent to which they morphologize grammatical categories on syntactic heads. Thus while Abkhaz (Abkhaz-Abaza; Abkhaz-Adyghean) has between fifty-eight and sixty-four consonant phonemes (depending on whether we are examining the literary or nonstandard dialects respectively), it has only two underlying phonemic vowels, /a/ and /ə/, among the fewest of any language. As a result of this, expressive constructions tend to be marked as gemination of verb stem consonants, not vowels: -ss- in á-pssa-ra ‘sweep’ (Abzhywa dial., cf Bzyp a-psa-ra; Chirikba 1996: 23), pssħa-rá (Tapant dial.; Chirikba 1996: 25). Thus the kinds of vowel gradations employed to create ideophonic lexical items in Georgian in (2) play little to no role in Abkhaz expressive language. The languages of this family also display extremely high degrees of verbal head-marking polysynthesis – with 11 categories inflected on the verb, they rank in the second highest category of the World Atlas of Language Structures (Haspelmath et al. 2005: Ch. 22), but nouns are marked with few case or other categories. In closely related Kabardian (Circassian; Abkhaz-Adyghean), expressives are usually found as particles (Colarusso 1992: 158–9): (6)

Kabardian expressive particles a. q’əʔa ‘please!’ b. ʔaw (astonishment) c. aλawaha (astonishment) _ d. dədəd-dədəd (pain, sorrow) (expressive of a blow on the cheek) e. śaaky f. t’aay (expressive of a fall in a struggle)

Because of this overall morphosyntactic orientation, dependents in this family simply lack the categories with which expressives in other languages might be manipulated. Nakh-Daghestanian languages in contrast often have a dozen or more vowels and their morphosyntactic profile is much more highly dependentmarking. Consequently, a Nakh language like Chechen, with its twenty-three simple vowel distinctions (not including diphthongs) and its eight case forms, has a phonological profile that allows ideophones to manifest themselves in vowel gradations, umlaut, and ablaut. Frequently, as noted above in the case of Sanzhi Dargwa, this consists of constructions of light verbs with complements bearing ideophonic content. For example, in (7), the light verb is ælla ‘say’ and its object is a reduplicated noun varq-varq ‘croaking’ (Muslim Khulando, p.c.):

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

342

Thomas R. Wier

(7)

Chechen Varq-varq ælla pħid-aš y-ara y-exaš ʔ oma čoħ ˤ croak-croak say.cv frog-pl ii-prog.past ii-cry.cv.sim pond in ‘Frogs were croaking in the pond.’

In Chechen, this ideophone for ‘croak’ can be modified to express the pitch of the sound: varq-varq, vorq-vorq, vurq-vurq. In some Caucasian languages, this use of light verbs is effectively the only strategy possible for integrating ideophonic expressions into the language. In Udi, a Lezgic language of Azerbaijan and Georgia, independent verbs have become a closed class such that almost all verb constructions actually involve a light verb stem as in (8) plus an incorporated element of some kind, either a noun, adverbial or some dependent verb stem. Thus in (9), a light verb p- ‘say’ has incorporated a stem äit-‘word’ to create a new predicate ‘speak’(Harris 2008): (8)

Udi light verbs b- ‘do’, bak- ‘become’, d- ‘make, causative’, p- ‘say’, ak’- ‘see’, aq’- ‘take, get, receive’, ef- ‘hold’, p’ur- ‘die’, sak- ‘push’, (u)k’- ‘eat’, uˤğ- ‘drink’, etc.

(9)

Luiza-n Udi-n muz-in äit-ne-p-e Louisa-erg Udi-gen language-inst word-3sg-say-aor.II ‘Louisa spoke in Udi.’ (Harris 2008: 214)

In Udi, this morpholexical constraint has the consequence that there are no (new) simplex expressive verbs: all expressive verbs consist of an expressive element (ideophone, noun, etc.) incorporated into a light-verb stem: giˁzgiˁz-pesun [laugh-say-mas] ‘laugh, chuckle’, q’aśq’aś-p-esun [crunch-say-mas] ‘eat, crunch on’, q’aˁq’aˁ-p-esun [ideo-say-mas] ‘drown, strangle’, t’ut’u-p-esun [ideo-say-mas] ‘tremble’, ğaˁğaˁ-p-esun [ideo-say-mas] ‘snarl, snap’ (Schulze 2005: 165). In most languages though, expressive elements vary in precisely how integrated they are in clauses. Thus in Archi (Lezgic; Nakh-Daghestanian), incorporated expressives in some cases behave like simple stem formatives and thus are invisible for purposes of case assignment of other arguments in the clause (Testelec 1987): (10)

Archi a. t’ant’ ziz-war bee.abs buzz-say.prs ‘The bee buzzes.’ b. za-ri diq’ u-bu χ I-erg soup.abs drink-say.pst ‘I eat the soup’ c. jaIt’i-li zatːi-k ƛiɬ-bo snake-erg I-loc hiss-say.pst ‘The snake hissed at me.’

Like most languages of the family, Archi is an ergative language, and so the subject t’ant’ ‘bee’ of the intransitive predicate in (10a) takes

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

Georgian and other Caucasian languages

343

absolutive case (here, null marking); in a corresponding transitive like (10b), the incorporated form χu- ‘drink’ does not prevent the verb from assigning absolutive case to the other argument diq’’ ‘soup’. In (10c) however, the expressive element ƛiɬ-‘hiss’ appears to absorb the assignment of absolutive case normally assigned to direct objects of transitive verbs, making it unavailable for the notional patient, which receives locative case. In a different ergative Lezgic language of Daghestan and Azerbaijan, Lezgian, expressives have completely generalized the absorption of absolutive case so that originally onomatopoetic expressives like those in (10) uniformly take ergative subjects with a single light verb awun ‘do’ (Haspelmath 1993: 181), as in (11)–(12): (11)

C1V(r)C2 + awun a. murr-murr awun b. č’arx-č’arx awun c. žirt’- žirt’ awun d. ziw-ziw awun

(12)

C1V1C2rV1C2 + awun a. t’aq’raq’ awun ‘crunch’ b. bağrağ awun ‘rumble’ c. lešreš awun ‘splash’ d. č’wäqräq’ awun ‘snap, crack’

‘purr’ ‘crunch’ ‘squelch’ ‘clink’

Like the Georgian examples in (2), Lezgian expressives show varying kinds of vowel or consonant gradation, reminiscent of Semitic templatic morphology (see also §11.1.4 and §11.2). In some languages, a distinction between different levels of integration of incorporated expressives has become grammaticalized as a marker of animacy. In Bezht’a (Tsezic; Nakh-Daghestanian) and Lak (Lakic; NakhDaghestanian), incorporating constructions mark subject arguments with ergative case if they are animate (13a, 14a), but absolutive case if they are inanimate (13b, 14b; Testelec, 2021): (13)

Bezht’a a. is-t’i hic-ƛo-jo (Animate subject) brother-obl.erg sneeze-say-pst ‘Brother sneezed.’ b. okko c’im-ƛo-jo (Inanimate subject) coin.abs jingle-say-pst ‘The coin jingled.’

(14)

Lak a. čːit-ul maIw-ukunni (Animate subject) cat-erg meow-say.pst ‘The cat meowed.’ b. granata p’aIq’-ukunni (Inanimate subject) grenade.abs explode-say.pst ‘The grenade exploded.’

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

344

Thomas R. Wier

Thus in these languages the encoding of expressive forms represents a kind of covert category of animacy. In yet other languages, expressive-like forms are fully integrated into the language’s formal gender-marking system. Like most northeastern Caucasian languages, nouns in Hinuq (Tsezic; NakhDaghestanian) fall into a number of different gender classes depending on how they agree with certain verbs, adjectives, adverbs and some other wordclasses. In Hinuq, expressive verbs depicting the emission of sound may be nominalized by the addition of a suffix -ni, as in (15), and when doing so they are consistently classed with Hinuq’s fifth gender otherwise associated with tools, abstract nouns, and meteorological phenomena (Forker 2013: 120): (15)

Hinuq a. ħapya:- ‘bark’ b. babaya:- ‘bleat’ c. dodoya:- ‘shake’ d. bubuya:- ‘bellow’ (bull) e. p’ap’aya:- ‘chatter’

ħap-ni ‘barking’ (Forker 2013: 109) baba-ni ‘bleating’ dodo-ni ‘shaking’ bubu-ni ‘bellowing’ p’ap’a-ni ‘chattering’

Thus in Hinuq one can largely predict which formal gender class an expressive nominalization will be. 11.1.3

Category innovation exclusive to expressives

Such cases reflect the recruitment of a pre-existing grammatical category for expressive purposes, but in many languages, expressive language manifests entirely new grammatical or phonological categories that are either not otherwise present in the language, or are very marginal. Thus many languages in the Caucasus lack a distinct voiceless labiovelar fricative /f/, but according to Sumbatova and Mutalov (2003: 3), in Icari Dargwa (Dargic; Nakh-Daghestanian) this phone exists marginally only within the class of ideophones like uf- in uf.bik’waraj ‘blow (about someone), utter a fie’. In the core lexicon of Khwarshi (Tsezic; NakhDaghestanian) vowel-length (Khalilova 2009: 20) and of Budukh both vowellength and nasalization (Azmaiparashvili et al. 2016: 148) are unknown, and yet both processes occur occasionally in those languages in expressive constructions. In Udi, neither consonants nor vowels exhibit length contrasts, but in some forms glottalized velar or uvular stops can become lengthened: ekː’a ‘all what’, eqː’ara ‘how much’, t’eqː’ara ‘so much’, features that Schulze (2005: 71) attributes to potential expressive usage. Likewise, in Udi reduplication plays virtually no role in the core grammar of the language, but does appear as a productive process in expressive forms (according to Schulze 2005: 137–8): (16)

a. b. c. d.

č’uč’up’ k’ak’ala gugum k’ok’oc

‘curl’ ‘excrement’ ‘horsefly’ ‘hen’

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

Georgian and other Caucasian languages

11.1.4

345

Expressives as a source of lexical innovation: flora and fauna

Another almost ubiquitous way in which Caucasian languages use expressive language is as a source for names of salient animal and plant species, or words closely associated with them. This form of expressive language is indeed so widespread within the Caucasus that it may be rightly distinguished as a separate kind of lexical process. Though there is variation both within and between languages and language families, a common trend is for species names to have a reduplicative template of . In many cases, the species names can be reconstructed to their respective protolanguage, for example, Georgian k’ak’abi ‘partridge’ and Megrelian k’ok’obe ‘partridge’, since they underwent separately identifiable sound-changes (in this case, the raising/backing of Proto-Kartvelian *a to /o/ in Megrelian). In other cases, it seems likely that a single lexical item was loaned between languages (including closely related ones), or was innovated separately according to a similar expressive template. This is true of the words for ‘nightingale’ across the Caucasus, which have often borrowed the root bulbul- from Persian. 11.1.5

Expressives as a locus of diachronic change

Finally, expressives may also be a source for or a locus of diachronic change. For one, ideophones may become grammaticalized into other constructions and subsequently lose any semantically decompositional meaning that is accessible to speakers. Thus in the Sanzhi Dargwa example above, it is difficult to isolate the difference in meaning between sːurk’ b-arq’- [ideo-do. pfv-] ‘press’ and sːurk’ b-ik’ʷ-[ideo-say.ipfv-] ‘rub, polish’, despite the fact that each element of these constructions is formally quite distinct. Over time, such ideophones may be fully grammaticalized as part of some other element. The reverse is also true: in some cases, non-ideophonic content may begin to behave like ideophones, at least in a formal sense. In recent work, Gilles Authier has documented so-called pseudo-ideophones in Archi, in which a lexical contrast arises by reanalysis of some part of a simplex verb as complex (Authier 2022). He gives examples like ƛum-mus ‘pull’ and ł:um-mus ‘flee’ in which the stem’s initial consonant has innovated what appears to be an expressive contrast: (17)

lo-bur χ:ams:-i-ƛ’iš b-ułne child-pl.nom bear-obl-subel hpl-flee.pf ‘The children fled from the bear’

Because the form with initial ł- appears cognate to a lexeme -ĝn in the related Kryts language (Lezgic; Nakh-Daghestanian) without expressive

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

346

Thomas R. Wier

Table 11.2 Floral and faunal terms derived from expressives in Caucasian languages Family

Language

Species name

Abkhaz-Adyghean

Abkhaz

a-ħʷǝħʷ ‘pigeon’, patpatǝ- ‘to flitter, quaiver (of a bird)’, a-c’ǝpc’ǝp ‘sp. of bird’, a-č’wǝnč’wǝħa ‘bird’s stomach’, ğyarğyar- ‘to frighten birds in fields or kitchen gardens’, a-k’ark’alamǝ́šw‘sp. of small lizard’, a-c’ǝrc’ǝrǝ́ ‘sound made by dragon-fly or snake’

Kartvelian

Abaza Georgian

ħʷǝħʷ ‘pigeon’ k’irk’it’a ‘kestrel’, guguli ‘cuckoo’, bulbuli ‘nightingale’, but’but’a ‘warbler’,k’ok’orina ‘sandpiper’,c’ivc’iva ‘chickadee’, k’ak’abi ‘partridge’, k’ač’k’ač’i ‘magpie’, tutiq’uši ‘parrot’, q’arq’at’i ‘stork, spoonbill’,xoxobi ‘pheasant’, baq’aq’i ‘frog’

Megrelian

ğerğet’i ‘goose’, ğorğonji ‘goose’, k’ok’obe ‘partridge’, dadulia ‘little chick’, t’urt’urue ‘turtle dove’, č’k’ič’k’it’ia ‘ant’, jgijgit’ia ‘ant’ parpānd ‘blackbird’, č’inč’wer ‘bird’s nest’, metetx ‘bat’

Svan Nakh-Daghestanian

Aghul Akhvakh

q:Iaraq:Ial ‘magpie’, c’urc’ul ‘marten’, mizmiz ‘mosquito’ koq̇:o ‘frog’, žožo ‘fly’

Archi

žužuḳ ‘quail’, ƛ̣:iƛ̣:í ‘kind of song bird’, ḳʷatḳʷat _ _ ‘woodpecker’ zimzi ‘mosquito’

Bezht’a Budukh

q̇öq̇ötö ‘woodpecker’, šašu ‘swallow’ _ kakɨl ‘partridge’, čič ‘lizard’

Chechen Dargwa

šoršal ‘thrush, blackbird’, qoqa ‘pigeon’ š:aḳaḳi ‘quail’, wadwadi ‘hoopoe’ (Akusha), č̣ič̣ala ‘snake’, imiʔala ‘stinging insect, ant’ (Akusha), zimizal ‘stinging insect, ant’ (Chiragh), q:ulq:a ‘crow’

Karata Khinalug

č’ʷanč’ʷara ‘butterfly, quail’ k’ak’id ‘partridge’, čänč ‘pigeon’

Hunzib Lak

q’alq’ala ‘magpie’ q:aqnu ‘partridge’, ʁanʁarat’i ‘beetle’

Lezgian Rutul

kutkut’ ‘hoopoe’, čurčul ‘snake, worm’ c̣uc̣ul sɨrɨc̣al ‘marten’

Tabasaran

q:aq:uba ‘partridge’, mizmiz ‘mosquito’, q:Iarq:Iar ‘magpie’ kɨc̣ac̣aj ‘sparrow’,t’aq̇t’aq’aj ‘woodpecker’ bɨzbɨza ‘mosquito’

Avar

Tsakhur

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

Georgian and other Caucasian languages

347

function (cf. ji-ĝn-iǯ ‘pull’), the Archi form with initial ƛ-may represent an incipient expressive morphological contrast. In languages with long written histories, we can also trace the evolution of the formal exponence of such expressive constructions by looking at how they change in frequency and function in historical texts. The Caucasian language with by far the longest written history in the region, Georgian, possesses a rich wealth of historical texts dating from the fifth century ce onward. These can allow us to see how reduplicative patterns for expressive verbs like ğağadeba ‘call out to, wail to’ (first att. fifth century) or k’amk’ameba ‘sparkling’ (first att. eleventh century) expand and change over many centuries: (18)

The Martyrdom of Queen Shushanik (c. 476–482 ce) sen-i . . . mo-i-c’i-a mis zeda,romel-sa=c=igi illness-nom pvb-prv-contract-3sg3sg.gen on which-dat=rel=det c’inaysc’ar u-ğağa-d-eb-d-i before.hand prv-wail.out-put-th-impf-1/2 ‘She contracted an illness about which you were crying out before . . .’

(19)

The Sixth Meeting of the Ecclesiastical Council of Carthage (c. 1010–1030 ce) xolo ğmert-t-mtavrob-it-ta k’amk’am-eb-a-ta=gan but God-gen-government-inst-gen.pl twinkle-th-mas-gen.pl=from es-oden-i natel-i a-ku-s ğirs-ta this-quant-nom light-nom prv-have.inan-3sg worthy-dat.pl ‘But from the twinklings of the Divine Government those who are worthy have so much light . . .’

Such a long textual heritage allows scholars to know that apparent expressive constructions with particular features like reduplication exclusive to expressives are not new innovations in recent centuries, but long-enduring, stable features of the language. 11.2

Expressive language in Georgian

The most extensive previous treatment of expressive language in Georgian is that of Holisky (1988), who examines a number of familiar features from the previous literature. For Holisky, expressives are a syntactically distinct form-class that denote ‘varied kinds of sensation, the impingement of the material world . . . upon the senses’ (citing Emeneau 1969) that also ‘convey with vivid clarity and eloquence the perceptual qualities of objects and events’ (citing Johnson 1976). In addition to the criteria noted by Dingemanse (2012), Holisky found a variety of other traits either universally or statistically likely to appear, including: (20)

a. they do not always observe the same phonotactic patterns established for the rest of the lexicon; b. some phonological rules may not hold for this class; or vice versa, special rules may apply only to this class; c. they are restricted to a few syntactic positions, often in isolation;

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

348

Thomas R. Wier d. they tend not to be negatable or quantifiable; e. they describe situations in a holistic way, closer to being semantic propositions than predicates or arguments; f. they often exhibit stylistic restrictions, e.g. to children’s speech, or oral and not written language

Holisky’s investigation (undertaken with Nana Kakhadze) focused exclusively on one subclass of expressives: manner of speaking verbs. After examining more than one hundred such verbs divided into loud sounds, highpitched sounds, indistinct sounds, and so on, she found that many of them lack lexical discreteness, in the sense that they form quasi-paradigmatic structures amongst themselves. On the one hand, there are verbs like those in (21) through (23), in which a fixed consonantal templatic matrix alternates with different vowel qualities, and high vowels reflect higher, louder pitch and lower vowels reflect lower pitch (Holisky 1988: 57): Verbs of shrill, squeaky complaint (21)

č’uč’q’un-eb-s squeal-th3sg ‘S/he is squealing’

č’ič’q’in-eb-s squeak-th3sg ‘S/he is squeaking’

č’ač’q’an-eb-s squawk-th3sg ‘S/he is squawking’

Verbs of low, loud complaining (22)

jujğun-eb-s ‘speaks peevishly, nasally’

jijğin-eb-s ‘mutters’

jajğan-eb-s ‘grumbles loudly’

Verbs of soft, child-like talking (23)

t’ut’un-eb-s ‘mumbles softly’

t’it’in-eb-s ‘burbles, drivels’

[cf. t’a’tana ‘scolder, remonstrator’]

On the other hand, there are also expressive verbs in which the vowel stays fixed, while we see various kinds of consonant gradation, often reflecting an onomatopoetic hierarchy of [voiced > aspirated > glottalized]: Verbs of children’s crying (25)

jğav-i-s ‘weeps loudly, howls’

čxav-i-s ‘screeches, yowls’

č’q’av-i-s ‘(owl, jackal) yowls, howls’

Verbs of sobbing (26)

zlukun-eb-s ‘bawls, sobs loudly’ slukun-eb-s ‘sobs loudly’

Verbs of soft, child-like talking (27)

bluq’un-eb-s ‘stutters, stammers’ bluk’un-eb-s stutters, stammers’

Holisky also found that glottalization is associated with high pitch, with seventeen of the eighteen words associated with high-pitch having glottalized sounds, while voiced consonants are associated with lower pitch and negative attitudinal perspectives of the speakers (1988: 60):

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

Georgian and other Caucasian languages Glottalized

349

Voiced

c’rip’in-eb-s c’iv-i-s

squeaks, peeps screeches, shrieks

burt’q’un-eb-s burdğun-eb-s

mumbles, growls mutters, grumbles

č’uč’q’un-eb-s č’q’iv-i-s

squeals screeches, shrieks

dudğun-eb-s bluq’un-eb-s

talks through one’s nose stutters, stammers

k’iv-i-s

shrieks

bubun-eb-s

bellows, lows

However, perhaps the most significant generalization in Holisky’s work is that these manner of speaking verbs also overwhelmingly bear the same morphosyntactic profile: they belong to a class of ‘medial’ verbs, or atelic, agentive intransitive predicates that pattern differently from transitive predicates and stative intransitive predicates in terms of case, agreement and tense/ aspect/mood morphology (see also Holisky 1983). To understand this, it is important to understand the complicated system of Georgian case and agreement. Georgian is a Split-S language in terms of its morphosyntactic alignment (Harris 1981, Harris 1985, Wier 2005, Wier 2011, Wier 2014, Baker 2017): it has two classes of intransitive predicates, one of which patterns like the subject of transitive verbs, and another which patterns separately (though not quite like the objects of transitives). But in addition to this, Georgian features crosscutting tense-aspect-mood (TAM) splits, and different conjugations pattern differently in different TAM-combinations. In other words, one cannot say that a particular case desinence directly reflects a specific grammatical function, because there is a many-to-many-to-many relationship of case, agreement and grammatical function. So for example in (28) we see case-shifts from one tense to the next: (28)

a. bavšv-i male i-č’uč’q’un-eb-s child-nom soon prv-squeal-th-3sg ‘The child will squeal soon.’ b. bavšv-ma i-č’uč’q’un-a child-narr prv-squeal-aor.3sg ‘The child squealed.’ c. bavšv-s u-č’uč’q’un-i-a child-dat prv-squeal-pf-3sg ‘The child has apparently squealed.’

In (28a), the subject in the present/future series form is bavšv-i ‘child’ with nominative case -i, while in the aorist series form in (28b) it is bavšv-ma with narrative (ergative) -ma, and in the perfect-evidential series it is bavšv-s with dative -s. To get a holistic concrete sense of this, Tables 11.3 and 11.4 display the general system of case assignment and tense-aspect formation for Georgian

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

350

Thomas R. Wier

Table 11.3 Georgian case assignment across TAM series and conjugational class

Transitive PresentFuture Aorist

Subj: nom IObj: dat DObj: dat Subj: narr IObj: dat DObj: nom

PerfectEvidential

Subj: dat IObj: pp DObj: nom

‘Medial’ Intransitive

Stative Intransitive

DativeAffective

Subj: nom

Subj: nom

Subj: narr

Subj: nom

Subj: dat DObj: nom Subj: dat DObj: nom

Subj: dat

Subj: nom

Subj: dat DObj: nom

Sources: Harris 1981, Holisky 1983, Wier 2011

Table 11.4 Formal criteria distinguishing different conjugation classes

Transitive

‘Medial’

Stative Intransitive

DativeAffective

Formation of the Future subseries 3Sg suffix in Present

Preverb -s

i- … -eb -s

Preverb -a

e- . . . -eb -s

3Sg suffix in Future

-s

-s

-a

-a

Sources: Harris 1981, Holisky 1983, Wier 2011

verbs across different series and conjugation classes. In Table 11.3, the present-future series of verb paradigms, the first conjugation of transitives assign nominative case (nom) to their grammatical subject/agent (subj), dative case (dat) to their indirect object/goal (IObj), and also dative to their direct object/theme (DObj), if they are ditransitive. However, when we shift to the aorist series, verbs in this series of paradigms now assign narrative (aka ‘ergative’) to their subject, dative to their indirect object, and nominative to their direct object. When we shift yet again to perfect-evidential series of verb forms, the case array changes yet again: the notional subject is marked by dative case, the notional direct object by nominative case, and the notional indirect object not by a case at all but by a postpositional phrase headed by =tvis ‘for’. Table 11.4 likewise identifies the formal criteria by which these classes are distinguished. Very many formal explanations have been proffered for the technical origin and synchronic functioning of this complex system (e.g. see Harris 1981 arguing for a change in grammatical function; cf. Wier 2011 for an account involving two distinct levels of argument structure; see also Nash 2017 for an account based on theta-licensing). For the purpose of expressives however the

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

Georgian and other Caucasian languages

351

significance of all of this is that the two classes of intransitive verbs differ in each of these different tense-aspect-mood series, a class of ‘medial’/agentive intransitives which patterns like the subject of transitives, and a class of stative intransitives that patterns separately. The expressive verbs that Holisky examined pattern like the first of these two intransitive classes: as ‘medial’, atelic, agentive intransitive verbs. This generalization by Holisky is significant, because it shows that expressive manner-of-speaking verbs belong to a distinct formal class not just in their morphology, but also in their patterns of nominal case assignment and verbal tense-aspect formation and agreement. 11.3

Extensions and implications

This leaves open the question: do all Georgian expressives manifest these features, or just the manner-of-speaking verbs? To investigate this, I collected more than 150 expressives outside this domain of manner-of-speaking verbs to examine the extent to which they matched Holisky’s internally consistent generalizations (see Table 11.5). The database focuses on expressives of the emission of light, emission of sound, and salient movement, and includes not only verbs, but a number of nominals or adverbials which also depict sensory data and share some of the same features as verbs. For example in (29), among verbs of emission of sound, we find nominal triplets with many of the expressive traits discussed above: (29)

a. xrial-griali ‘banging, clatter’ b. xrial-ğriali ‘very loud and constant noise’ c. xrial-xriali ‘prolonged rattle, roar, thundering’

Table 11.5 Statistical distribution of features of morphosyntactic exponence across different expressive classes

Emission of light

Emission of sound

Salient movement

Control sample

TOTAL • verbs

52 40

51 19

50 41

175 53

Reduplication • full

18 (35%) 16 (31%)

23 (45%) 7 (14%)

30 (60%) 10 (20%)

2 (1%) 0 (0%)

• partial Root cluster >3

2 (4%) 10 (19%)

16 (31%) 5 (10%)

20 (40%) 5 (10%)

2 (1%) 13 (7%)

Medial verb Voiced fricative

11 (28% of verbs) 19 (37%)

8 (42% of verbs) 4 (8%)

19 (46% of verbs) 9 (18%)

1 (2% of verbs) 41 (23%)

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

352

Thomas R. Wier

Table 11.6 Case assignment of different classes of expressives across tenseaspect series

Pres

Aor

Perf-Evid

Emission of light

Emission of sound

Salient movement

cecxl-i bdğvrial-eb-s fire-nomglow-th3sg ‘The fire is glowing’ cecxl-ma i-bdğvrial-a fire-narrprv-glowaor.3sg ‘The fire glowed’

panjara raxrax-eb-s window.nom rattle-th3sg ‘The window is rattling’ panjara-mi-raxrax-a window.narrprvrattle-aor.3sg ‘The window rattled’

bavšv-i braxun-eb-s child-nomstomp-th3sg ‘The child is stomping’ bavšv-ma i-braxun-a child-narrprv-stompaor.3sg ‘The child stomped’

cecxl-s u-bdğvrial-i-a fire-datprv-glowpf-3sg ‘The fire has apparently glowed’

panjara-s u-raxrax-i-a window-datprv-rattlepf-3sg ‘The window has apparently rattled’

bavšv-s u-braxun-i-a child-datprv-stomppf-3sg ‘The child has apparently stomped’

As a control, the database also included 175 basic lexical items of all different lexical classes against which the features of the expressive sets could be compared. These basic lexical items were created from the list of basic semantic primitives used by the Loan Word Typology database (Haspelmath & Tadmor 2009), which was expressly designed to make the creation of semantically commensurable basic word lists possible. When we compare these additional classes of expressives to that earlier work, we find that they do largely cohere with Holisky’s findings, but with interesting twists. We see similar kinds of morphological exponence in the form of reduplication, expressive ablaut and consonant gradation, as well as expressive affixal formatives such as -in, -un or -ial. We also see that, like manner of speaking verbs, many verbs that denote emission of light or inanimate sources of sound also fall into Holisky’s medial class of verbs in terms of their case-assignment and tense/aspect/mood desinences. However, it is also clear that there arise other salient features that are either absent, or rarer, in the class of manner-of-speaking verbs. For example, while the data published by Holisky examined included forms with complex onsets with two segments like /č’q’/, /dğ/, /jğ/, /čx/, /t’q’/, and so on, in lexical items depicting emission of light, we often find extremely complex onsets of three, four, and even five segments in length, at more than double the frequency (19 percent) of the control sample (7 percent), and elevated rates in the other two classes:

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

Georgian and other Caucasian languages

353

Table 11.7 Expressive verbs of emission of light or sound or of salient movement Expressive with complex onset

English gloss

bdğvrial-eb-s

glitters, glares

k’rtol-av-s mbdvinvar-eb-s

glitter, quiver, shiver flicker, glimmer

mbzin-av-i mbžut’-av-i

brilliant flickering, glimmering

ckrial-eb-s bžğvrial-eb-s

moves gracefully, quickly, shimmers sparkles, glitters

brč’q’vial-eb-s brč’q’vin-av-s

sparkles, glitters, is radiant glitters, shines

bdğvin-av-s mžğer-i

walk about roaring (of lions) voiced (e.g. consonant)

mbdğvin-av-i ʣgrial-eb-s

roaring crashes, bangs

ʣgnar-av-s žğl-i-s

gnaws on s.th. cracks s.th. (e.g. glass)

brtq’el-brtq’el-i k’rt’n-i-s

high-flown, bombastic speech (bird) preens for lice

prckvn-i-s čkro

peels s.th. noise, sound of s.o.’s voice

t’q’lap’i t’q’laša-t’q’luši

fruit-leather (made by crushing fruit) continuous cracking, thwacking sound

This stands in contrast to the verbs of emission of sound or salient movement, fewer of which manifest extremes of consonant clusters in onset position. While such long consonant clusters can be found occasionally in nonexpressive parts of the lexicon (e.g. bžğali ‘light grey’, prckili ‘reddish stone’, dngra ‘fit, strapping, imposing’, p’rc’k’ali ‘splinter’, etc.), they are considerably more frequent among expressives. It seems likely that this reflects an exaptation from Georgian’s typologically unusual syllable structure toward making expressives more salient in discourses. But it also difficult to describe in purely morphological terms, since if there is any generalization to be had here, it is on a highly abstract prosodic level – not unlike the prosodic quasitemplates seen in Holisky’s earlier study described in §11.2. Another at least ostensible feature found extensively in verbs of emission of light, in addition to the manner of speaking verbs which Holisky investigated, is the presence of one or more voiced fricatives, often /ğ/, /ž/, /v/ or /z/:

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

354

Thomas R. Wier

Table 11.8 Salience of voiced fricatives in expressives depicting emission of light Expressive with voiced fricative

English gloss

živživ-eb-s

twinkles, glimmers; tweets

žik’žik’-eb-s žol-av-s

twinkles, glimmers flickers, glimmers

žužğ-av-s bzin-av-s

flickers, glimmers sparkles, glitters

ğudğud-eb-s ğadğad-eb-s

sparkles, dazzles with brightness sparkles, glows

vasvas-eb-s ğvelp-av-s

flashes, sparkles, shimmers burns down to embers

varvar-a na-ğverd-al-i

incandescent glow glowing embers/coals

ğviv-i-s ğuzğuz-eb-s

glows, flashes red, simmers (fire) crackles

gizgiz-eb-s elvar-eb-s

blazes up, roars dazzles, shines brightly

Do these voiced fricatives have the same causal role in expressive formation that reduplication, medial verbs or root clusters do? Here we can see the usefulness of the control sample, where voiced fricatives occur at approximately the same rate (21 percent of all expressives in Table 11.5) as the control sample (23 percent). Thus we can exclude this feature as an artefact of the sample creation, a merely apparent, and not actual, causal factor in expressive formation. Lastly, there is one category of expressive that did not cohere with Holisky (1988)’s results, but in an indirect way they throw her generalizations into sharper relief. These are the set of expressive verbs with ingressive aspect, which in Georgian are treated morphosyntactically as a kind of synthetic passive verb with an ingressive -d suffix:

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

Georgian and other Caucasian languages

355

Table 11.9 Ingressive expressive predicates Ingressive expressive

English gloss

ga-bdğvin-d-eb-a

will (begin to) sparkle, dazzle

ga-ğadğad-d-eb-a ča-nak’vercxl-d-eb-a

(fire) will (begin to) sparkle/glow (wood) will turn into glowing embers

ga-picx-d-eb-a t’q’d-eb-a

will get hot, heated break, shatter

še-mo-braxun-d-eb-a čačxap’un-d-eb-a

will come stomping in will rush down noisily

č’ianur-d-eb-a a-baban-d-eb-a

drags itself out endlessly; is badly delayed will stagger, shiver from fever

a-ğant’al-d-eb-a dzundzul-d-eb-a

will stagger (from one’s bed) come out at a trot

c’a-ğerğet’-d-eb-a a-q’vir-d-eb-a

waddle about like a goose will begin to cry out, clamor

a-raxrax-d-eb-a a-prut’un-d-eb-a

will begin to rumble will snort

Although in some cases these verbs are lexically underived (e.g. t’q’d-eb-a ‘it will break, shatter’) many of these verbs are effectively ingressivized forms of noningressive alternants seen above and in the rest of the database. Such forms often have reduplication, large consonant clusters, and depictive semantics like noningressive expressives. They differ however in one very important point: they are never medial verbs. Instead, they have the morphological and syntactic properties of the second class of intransitive verb in Georgian: the stative intransitives (see Table 11.3). Like statives, they form their future desinences with a lexically specific preverb, inflect their present tense forms with stative -a (unlike the transitive and medial -s; see Table 11.4), and they assign exclusively nominative case to their single argument, as in the examples in Table 11.10. Thus the exceptions to Holisky’s generalization about expressives themselves constitute a formal class. Thus unlike the forms in Table 11.6 which change from one series to the next, the expressives in Table 11.10 take -i nominative case subjects across tense/aspect series and consistently take preverbs characteristic of ingressives and stative intransitive verbs. The implication of this second class of expressive predicate is that the morphosyntactic features of expressives – the way they assign case to their arguments, the way aspectual desinences are assigned in a given paradigm – cannot be a direct expression of any one of their thematic roles, or of grammatical function, or of argument structure, or of their lexical semantics. Instead, the two different classes of expressives illustrate how the many-tomany relationship of case, agreement and tense-aspect marking is an

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

356

Thomas R. Wier

Table 11.10 Case assignment of different classes of ingressive expressives across tense-aspect series

Pres/ Fut

Aor

Perf-Evid

Emission of light

Emission of sound

Salient movement

cecxl-i ga-bdğvrin-d-eba fire-nom pvbsparkle-ingr-th-3sg ‘The fire will start to sparkle’ cecxl-i ga-bdğvrin-d-a fire-nom prv-sparkleaor.3sg ‘The fire started to sparkle’

panjara a-raxrax-d-eb-a window.nom pvbrumble-ingr-th-3sg ‘The window will start to rumble’ panjara a-raxrax-d-a window-nom pvbrumble-ingr-aor.3sg ‘The window rumbled’

bavšv-i še-mo-braxun-deb-a child-nom pvbventstomp-th-3sg ‘The child will come stomping in’ bavšv-i še-mo-braxun-da child-nom pvbvent-stomp-ingraor.3sg ‘The child came in stomping’

cecxl-i ga-bdğvrin-ebul-a fire-nom prvsparkle-th-pf-3sg ‘The fire has apparently started to sparkled’

panjara a-raxrax-eb-ul-a window. nom pvbrumble-th-pf-3sg ‘The window has apparently started to rumble’

bavšv-i še-mo-braxuneb-ul-a child-nom pvb-vent-stomp-thpf-3sg ‘The child has apparently come in stomping’

underlying one, in which different parts of the grammatical architecture can trigger assignment to one or more of these paradigmatic subsystems. Thus expressives may be assigned to the medial class of atelic, agentive intransitive verbs, but if they take on ingressive semantics, this triggers wholesale reassignment to the class of ingressive predicates, with a consequent total shift of case-assignment, agreement and tense-aspect formation. 11.4

Concluding remarks

This chapter has surveyed the different ways in which expressive predicates behave across the understudied languages of the Caucasus. We have seen how expressive constructions exhibit marked variation in exponence, ranging from verbal stem suffixation, to nominal cliticization, to dedicated templatic slots on the verb, to complicated systems ideophones being incorporated in different ways into light-verbs. We have also seen how the extreme typological diversity of the Caucasus has acted as a constraint on how expressive language may manifest itself in different languages. Because the three autochthonous families differ so radically from each other in their phonological, morphological and syntactic profiles, it is difficult to find traits common to expressive constructions across the region which are yet not also common cross-linguistic properties found in expressives around the world. What works for a dependentmarking language with many phonemic vowels in Nakh-Daghestanian simply will not work for more polysynthetic languages with few vowels found in

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

Georgian and other Caucasian languages

357

Abkhaz-Adyghean, nor yet again for Kartvelian which is both polysynthetic and yet has a famously complicated system of dependent case-assignment. This paper also looked in considerable depth at the behavior of one Caucasian language, Georgian, and found that the behavior of expressives is not uniform even within this single language. Different subclasses of expressives may manifest different kinds of exponence, either categorically or statistically, and other categories may crosscut these generalizations. Thus while verbs for manner of speaking, emission of light, emission of sound, and salient kinds of movement in Georgian all have a tendency to behave like ‘medial’ verbs which are atelic and agentive, verbs of emission of light are far more likely to have extreme consonant clusters in verb stems. On the other hand, all ingressive expressives, irrespective of their lexical semantics, belong to another morphosyntactic class altogether. Because the different classes of expressive behave in different ways in different contexts, Georgian thus shows that expressives are not one thing, but many different things, one thrown into sharp relief by the fascinating but typologically unusual system of Georgian morphosyntax more generally. References Alekseev, M. E. 1994. Budukh. In Rieks Smeets (ed.) The Indigenous Languages of the Caucasus, Vol. 4. Delmar, NY: Caravan Books, 367–406. Authier, Gilles. 2022. Les verbes composés à idéophones en archi. [Compound verbs with Ideophones in Archi.] Public lecture at the École Pratique des Hautes Études, 3 May. Azmaiparashvili, Levan, et al. 2016. ნახურდადაღესტნურენათაშედარებითი გრამატიკა, I ნაწილი [Comparative grammar of Nakh-Daghestanian languages, Vol. I]. Tbilisi: Chikobava Institute. Baker, James. 2017. How Georgian is (not) like Basque: A comparative case study of split-S languages. SLE Book of Abstracts 2017. 50th Annual meeting of the Societas Linguistica Europaea, University of Zurich, 10–13 September. Butskhrikidze, M. 2002. The Consonant Phonotactics of Georgian. Amsterdam: Netherlands Graduate School of Linguistics. Chirikba, V. A. 1996. A Dictionary of Common Abkhaz. Leiden: Brill. Colarusso, John. 1992. A Grammar of the Kabardian language. Calgary: University of Calgary Press. Dingemanse, Mark. 2012. Advances in the cross-linguistic study of ideophones. Language and Linguistics compass, 6(10), 654–72. 2015. Ideophones and reduplication: Depiction, description, and the interpretation of repeated talk in discourse. Studies in Language. International Journal sponsored by Foundations of Language, 39(4), 946–70. 2017. Expressiveness and system integration: On the typology of ideophones, with special reference to Siwu. STUF-Language Typology and Universals, 70(2), 363–85. 2018. Redrawing the margins of language: Lessons from research on ideophones. Glossa: A Journal of General Linguistics, 3(1). Dingemanse, M. & Akita, K. 2017. An inverse relation between expressiveness and grammatical integration: on the morphosyntactic typology of ideophones, with special reference to Japanese Journal of Linguistics, 53(3), 501–32.

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

358

Thomas R. Wier

Emeneau, M. B. 1969. Onomatopoetics in the Indian linguistic area. Language, 45(2), 274–99. Forker, Diana. 2013. A Grammar of Hinuq. Berlin: De Gruyter Mouton. 2019. A Grammar of Sanzhi Dargwa. Berlin: Language Science Press. Harris, Alice C. 1981. Georgian Syntax: A study in relational grammar. Cambridge: Cambridge University Press. Harris, A. C. 1985. Diachronic Syntax: The Kartvelian case. Orlando: Academic Press. 2008. Light verbs as classifiers in Udi. Diachronica, 25(2), 213–41. Haspelmath, M., 1993. A Grammar of Lezgian. Berlin and Boston: de Gruyter. Haspelmath, M., Dryer, M. S., Gil, D. & Comrie, B. 2005. The World Atlas of Language Structures. Oxford: Oxford University Press. Haspelmath, M. & Tadmor, U. 2009. The loanword typology project and the world loanword database. In Haspelmath, M. & Tadmor, U. (eds.) Loanwords in the World’s Languages: A comparative handbook. Berlin: De Gruyter Mouton, 1–34. Holisky, Dee Ann. 1983. Aspect and Georgian medial verbs. Bulletin of the School of Oriental and African Studies, 46(3), 564–5. 1988. On the study of expressives in Kartvelian languages. In Proceedings of the First International Symposium in Kartvelian Studies. Tbilisi: Tbilisi State University Press, 52–64. Johnson, M. R. 1976. Toward a definition of the ideophone in Bantu. OSU Working Papers in Linguistics 21, 240–53. Kadshaia, Otar & Heinz Fähnrich. 2001. Mingrelisch–Deutsches Wörterbuch. Wiesbaden: Reichert Verlag. Khalilova, Zaira. 2009. A grammar of Khwarshi. Doctoral dissertation, Leiden University. Kibrik, A. E., Kazenin, K. I., Lyutikova, E. A. & Tatevosov, S. G. 2001. Багвалинский язык: грамматика, тексты, словари [Bagvalinskij jazyk] [The Bagvalal language]. Moscow: Nasledie. Nash, L. 2017. The structural source of split ergativity and ergative case in Georgian. In Jessica Coon, Diane Massam, & Lisa deMena Travis (eds.) The Oxford Handbook of Ergativity. Oxford: Oxford University Press. Schulze, Wolfgang. 2005. A Functional Grammar of Udi. Unpublished manuscript. Sumbatova, N. R. & Mutalov, R. O. 2003. Grammar of Icari Dargwa. Munich: LINCOM. Testelec, Yakov. 1987. Эргативнообразные построения в Нахско-дагестанских языках. [Ergative constructions in Nakh-Daghestanian languages.] In Questions in Linguistics. Moscow, 109–21. 2021. Ergativity in East Caucasian. Lecture presented in the 2021 Online Course on East Caucasian Languages, Linguistic Convergence Laboratory. www.youtube .com/watch?v=w2_i-MIu2EE Wier, Thomas. 2005. Pivots and subjects in Georgian. In M. Butt & T. Holloway King (eds.) Proceedings of the Lexical Functional Grammar ’05 Conference. Stanford, CA: CSLI Publications, 487. 2011. Georgian morphosyntax and feature hierarchies in natural language. PhD dissertation, University of Chicago. 2014. Nonconfigurationality and argumenthood in Georgian. Lingua, 145, 36–64.

https://doi.org/10.1017/9781108989084.018 Published online by Cambridge University Press

Comparative

https://doi.org/10.1017/9781108989084.019 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.019 Published online by Cambridge University Press

12

Parameters of variation in the syntax of expressive suffixes: Case studies of Russian, German, Spanish and Greek Olga Steriopolo,* Giorgos Markopoulos,† Vassilios Spyropoulos{

12.1

Introduction

Expressive suffixes cross-linguistically express the speaker’s attitude (positive or negative) towards a referent; a subset of them, here termed ‘size suffixes’, can both express an attitude and indicate the size of a referent (small or big).1 In this chapter, we compare expressive suffixes in four European languages: Russian, German, Spanish, and Greek. All four languages are genetically related (despite belonging to different genera, they are all members of the Indo-European family), fall into the same morphological type (they are all fusional), and exhibit similar affixation mechanisms for expressive formation (predominantly suffixing). Nevertheless, as will be shown below, all four languages differ significantly with respect to the syntactic properties of expressive suffixes. In particular, the parameters of syntactic variation concern the manner and place of attachment in the syntactic tree. For example, Russian size suffixes are syntactic modifiers (manner of attachment) that can only attach to a noun category (place of attachment). The German size suffixes -chen and -lein are syntactic heads that, similarly to Russian, can attach to nouns. In Spanish, the productive size suffix -(c)it is a modifier that can attach to various syntactic categories. And, finally, in Greek, the three main expressive suffixes -ak, -ul and -its have homophonous counterparts that differ in terms of both the manner and the place of syntactic attachment. This shows that, cross-linguistically, although expressive suffixes can share the same expressive meaning (or function), they differ significantly in their syntactic structure (or form). This in turn supports the claim that there is no

† * Leibniz-Zentrum Allgemeine Sprachwissenschaft, Berlin University of the Aegean { National and Kapodistrian University of Athens 1 This work was supported by a DFG research grant to Olga Steriopolo (STE 2361/4-3).

361

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

362

Steriopolo, Markopoulos, Spyropoulos

universal one-to-one correspondence between the form and the function of linguistic objects across languages (Wiltschko 2008, 2014). The research has been conducted within the framework of Distributed Morphology (Halle & Marantz 1993, 1994; see also the overviews in Harley & Noyer 1999, Embick & Noyer 2007, Siddiqi 2010), which distinguishes between word formation from roots and word formation from syntactic categories (Marantz 1997, Harley & Noyer 1998, Embick 1998; among many others), a division which provides us with the appropriate formal tools for handling the variation in the morphosyntactic structure of expressive forms. The current work is a continuation and development of Steriopolo et al. (2021) and Steriopolo (2015). In §12.2, we discuss previous work on expressive suffixation in Russian, German, Spanish, and Greek. In §12.3, we propose two novel case studies of the homophonous suffixes -its (-ιτσ) in Greek and -ic (-иц) in Russian.2 Finally, in §12.4, we provide a conclusion and discussion of the findings. 12.2

Previous research

12.2.1

Russian

Steriopolo (2008, 2009) argues that expressive (expr) suffixes in Russian differ in both their meaning and form. In terms of the meaning, there are two major semantic types of expressive suffixes: attitude and size suffixes. 12.2.1.1 Semantic types of expressive suffixes in Russian Attitude suffixes (expr attitude) convey a speaker’s attitude towards a referent, while size suffixes (expr size) can both convey an attitude and indicate the size of a referent. Consider first the attitude suffixes. According to Steriopolo’s terminology (2008, 2009), there are two subtypes, ‘affectionate’ (affect) and ‘vulgar’ (vulg). For example, in (1b), the expressive attitude suffixes -ul’ and -us’ convey an affectionate (positive) speaker attitude towards the referent ‘grandfather’. In (2b), the expressive vulgar suffix -an conveys a vulgar (negative) attitude towards the referent ‘old man’. (1)

a. d’ed3 b. d’ed-ul’/us’-ja grandfather.n.sg grandfather-expr-n.sg ‘grandfather’ ‘grandfather (affect)’ (Steriopolo 2008: 2)

2 3

Throughout the chapter, Russian data are given in Latin transliteration using scientific transliteration of Cyrillic and Greek data in a broad phonemic transcription. Palatalization of consonants is indicated by an apostrophe in Latin transliteration of the Russian data.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek

363

Table 12.1 Semantic types of expressive suffixes in Russian Attitude suffixes (exprattitude)

Size suffixes (exprsize)

affectionate

-an’, -aš, -on, -ul’, -un’, -ur, -us’, -uš

vulgar

-ag, -ak, -al, -an, -ar, -ax, -il, -in, -ob, -ot, -ox, -ug, -uk, -ux -k (allomorphs: -ok/-ek/-ik), -c (allomorphs: -ec/-ic) -išč’

diminutive augmentative

Source: Steriopolo (2008: 117).

(2)

a. star’-ik b. star’-ik-an old-nom.sg old-nom-expr.sg ‘old man’ ‘old man (vulg)’ (Steriopolo 2008: 79)

In contrast to the attitude suffixes, size suffixes in Russian can both express an attitude and indicate the size of a referent. Their subtypes are ‘diminutive’ (dim) and ‘augmentative’ (aug). For example, in (3b), the diminutive suffix -ik conveys a positive attitude towards the referent ‘house’, while at the same time referring to its small size (e.g., a nice small house). In (4b), the augmentative suffix -išč’ can convey a negative attitude towards the referent ‘wolf’, but also indicates its large size (e.g., a big scary wolf ). (3)

a. dom b. dom’-ik house.n.sg house-expr.sg ‘house’ ‘house (dim)’ (Steriopolo 2008: 18)

(4)

a. volk wolf.n.sg ‘wolf’

4

b. volč’-išč’-e wolf-expr-n.sg ‘wolf (aug)’ (Steriopolo 2008: 110)

Table 12.1 illustrates the categorization of expressive suffixes in Russian proposed by Steriopolo (2008, 2009) based on the two main semantic types (attitude and size), with the two semantic subtypes for each expressive type (affectionate/vulgar and diminutive/augmentative, respectively). 12.2.1.2 Syntactic types of expressive suffixes in Russian Steriopolo (2008, 2009) shows that the two main semantic types of Russian suffixes (expr attitude and expr size) map onto two different syntactic types: syntactic heads and

4

There is a k ~ č’ alternation in this word that is typical for Russian.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

364

Steriopolo, Markopoulos, Spyropoulos

syntactic modifiers. Thus, expressive suffixes vary syntactically along two dimensions: (i) the manner of syntactic attachment (how the suffix merges in a syntactic tree: as a head or as a modifier) and (ii) the place of syntactic attachment (where it merges in a syntactic tree: below or above a syntactic category). Regarding the manner of attachment (head vs. modifier), Steriopolo (2008, 2009) shows that attitude suffixes merge as syntactic heads, as in (5a),5 whereas size suffixes merge as syntactic modifiers, as in (5b). In order to distinguish between syntactic heads and modifiers, Steriopolo (2008, 2009) assumes, following Schütze (1995), Bierwisch (2003), and Bachrach & Wagner (2007), that syntactic heads project and can thus determine the syntactic category and/or grammatical features (e.g., grammatical gender) of the output. By contrast, modifiers do not project, and thus cannot determine the categorial properties of the output. (5)

b.

a.

(Steriopolo 2008: 63)

As for the place of attachment, attitude suffixes can merge either with a root (√),6 as in (6a), or with various syntactic categories (e.g., nouns, adjectives, or verbs: n/a/v), as in (6b). In doing so, they always form nouns (n), acting as nominalizers. (6)

a.

b.

(Steriopolo 2008: 118)

5 6

A list of abbreviations for the symbols used in the syntactic diagrams throughout the article is given at the end of this chapter. The notation √ is from Pesetsky (1995).

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek

365

Table 12.2 Syntactic types of expressive suffixes in Russian Merging with roots exprattitude (heads) exprsize (modifiers)

Merging with syntactic categories

-an’, -aš, -on, -ul’, -un’, -ur, -us’, -uš, -ag, -ak, -al, -an, -ar, -ax, il, -in, -ob, -ot, -ox, -ug, -uk, -ux -k/-ek/-ok/-ik, -c/-ec/-ic, -išč’

Adapted from Steriopolo (2008: 150).

By contrast, size suffixes can only merge with a noun category, acting as noun modifiers, as in (7). (7)

The two-dimensional syntactic variation between attitude and size suffixes in Russian is presented in Table 12.2. The table shows that attitude suffixes are syntactic heads that can merge either with roots or with syntactic categories, while size suffixes are syntactic modifiers that can only merge with nouns. Notice that, in Russian, there are no syntactic modifiers that can merge directly with roots (as indicated by the empty cell in the table). Bearing in mind the syntactic differentiation of Russian expressive suffixes presented above, the following two questions arise. First, do the two types of expressive suffixes cross-linguistically share the same manner of attachment – that is, is it the case that attitude suffixes are universally syntactic heads, while size suffixes are syntactic modifiers? Second, do the two types of expressive suffixes crosslinguistically share the same place of attachment – that is, is it the case that attitude suffixes can universally attach to either roots or categories, while size suffixes can only attach to nouns? In what follows we will show that the answers to these two questions are negative and that expressive suffixes do not universally share the same manner or place of syntactic attachment, but present a great variety of different syntactic structures across languages as well as within a single language, as in Greek.7

7

See also Steriopolo (2017) on the variation of attachment places for size suffixes in three genetically unrelated languages: Russian, Kolyma Yukaghir, and Itelmen.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

366

Steriopolo, Markopoulos, Spyropoulos

12.2.2

German

Wiltschko and Steriopolo (2007) and Steriopolo (2015, 2016) show that, although the German diminutive suffixes -chen and -lein share the same place of syntactic attachment with the Russian size suffixes, the suffixes in these two languages differ in the manner of syntactic attachment. Whereas the Russian size suffixes are syntactic modifiers, as has been discussed above, the German ones are syntactic heads (see also Wiltschko 2006). This demonstrates that size suffixes across languages do not universally share the same manner of syntactic attachment. 12.2.2.1 Manner of syntactic attachment in German Wiltschko (2006) proposes that, in terms of the manner of attachment, the diminutive size suffixes chen and -lein in Standard German (as well as the suffixes -erl and -l in Austrian German) act as syntactic heads, as structured in (8). (8)

The evidence for the head status of the German diminutive suffixes comes from the fact that they can change the grammatical properties of the base to which they attach. First, they can change the grammatical gender of the nominal base; and second, they can turn mass nouns into count nouns. Consider first a change in grammatical gender. The diminutive suffixes -chen and -lein always form neuter nouns, regardless of the grammatical gender of a noun to which they attach. For example, in (9b), the suffixes merge with the masculine noun Tisch ʻtable’. The masculine gender of the noun is evidenced by masculine agreement with the determiner der and the adjective klein-er ʻsmall’ in (9a). The resulting diminutive nouns Tisch-chen ʻlittle table’ and Tisch-lein ʻlittle table’ in (9b) are neuter, as shown by neuter agreement with the determiner das and the adjective klein-es ʻsmall’. (9)

a. der/klein-er Tisch the.masc/little-masc table ‘the/little table’

b. das/klein-es Tisch-chen/-lein the.neut/little-neut table-dim/-dim ‘the/little table (dim)’ (Steriopolo 2015: 7)

In (10b), the same diminutive suffixes attach to the feminine noun Flasche ʻbottle’. The gender of the noun is evidenced by feminine agreement

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek

367

with the determiner die and the adjective klein-e ʻsmall’ in (10a). The resulting diminutive forms in (10b), Fläsch-chen ʻlittle bottle’ and Fläsch-lein ʻlittle bottle’, are neuter, triggering neuter grammatical agreement, as shown in (9b). (10)

a. die/klein-e Flasche b. das/klein-es Fläsch-chen/-lein the.fem/little-fem bottle the.neut/little-neut bottle-dim/-dim ‘the/little bottle’ ‘the/little bottle (dim)’ (adapted from Wiltschko & Steriopolo 2007: 2)

Syntactic structures for the diminutive nouns in (9b) and (10b) are proposed in (11a) and (11b), respectively. As shown in the structures, the diminutive suffixes -chen and -lein are syntactic heads that determine the neuter gender of the resulting diminutive forms. (11)

a.

b.

(Steriopolo 2015: 7)

In addition, as Wiltschko (2006: 669) shows, the diminutive suffixes -chen and -lein (as well as some other diminutive suffixes in German) can turn mass nouns into count nouns, indicating that they have an individuating function and behave as numeral classifiers. Wiltschko (2006: 676) views this property of the diminutive suffixes as evidence for their status as syntactic heads. For example, in (12b), the suffixes attach to the mass noun Wein ‘wine’ and turn it into count nouns: Wein-chen and Wein-lein ‘portion of wine’.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

368

Steriopolo, Markopoulos, Spyropoulos

(12)

a. viel Wein b. viel-e Wein-chen/-lein much wine many-pl wine-dim/-dim ‘much wine (mass noun)’ ‘many portions of wine (count noun)’ (adapted from Wiltschko 2006: 669)

12.2.2.2 Place of syntactic attachment in German With respect to their place of attachment in the syntactic tree, Wiltschko & Steriopolo (2007) argue that the diminutive suffixes -chen and -lein can only merge with a categorized noun, as structured in (13). (13)

The evidence comes from nominal morphology preceding the diminutive suffixes. For example, in (14a), the noun Ge-heim-nis ‘secret’ contains the nominal suffix -nis. In (14b) and (14c), the diminutive suffixes -chen and -lein attach outside the nominal suffix, forming the diminutive nouns Ge-heim-nischen ‘little secret’ and Ge-heim-nis-lein ‘little secret’, respectively. A structure for these diminutive forms is proposed in (14d), in which the noun Ge-heimnis ‘secret’ (n1) is formed first, and the diminutive suffixes -chen and -lein merge above it, forming the diminutive nouns (n2): Ge-heim-nis-chen and Geheim-nis-lein, respectively. As Wiltschko & Steriopolo (2007: 6) show, the diminutive suffixes -chen and -lein cannot merge with other syntactic categories besides nouns.8 (14)

8

a. Ge-heim-nis pref-secret-nom ‘secret’ b. Ge-heim-nis-chen pref-secret-nom-dim ‘little secret’ c. Ge-heim-nis-lein pref-secret-nom-dim ‘little secret’

See also Steriopolo 2008: 153.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek

369

Table 12.3 Syntactic variation in attachment of the size suffixes in Russian and German

Size suffixes Russian -k/-ek/-ok/-ik, -c/-ec/-ic, -išč’ German -chen, -lein

Manner of attachment: Syntactic modifiers

Place of attachment: Attaching only to nouns

✓

✓

*

✓

d.

(adapted from Wiltschko & Steriopolo 2007: 5, 6)

The syntactic properties of the German diminutive size suffixes -chen and -lein are presented in Table 12.3 in comparison with those of the Russian size suffixes, discussed earlier, in §12.2.1.2. The Russian size suffixes are syntactic modifiers, whereas the German size suffixes are syntactic heads. Thus, the suffixes in these two languages differ in their manner of attachment in the syntactic tree. Regarding the place of attachment, the suffixes in both languages share the same place of syntactic attachment: they can only merge with nouns.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

370

Steriopolo, Markopoulos, Spyropoulos

12.2.3

Spanish

Steriopolo (2015) provides a detailed analysis of the Spanish productive diminutive suffix -(c)it9 and argues that it shares the manner of syntactic attachment with the Russian size suffixes as they are all syntactic modifiers and thus differ from the German size suffixes, which are syntactic heads. Moreover, the Spanish suffix -(c)it is syntactically different from both the Russian and German size suffixes in the place of syntactic attachment. While the Spanish suffix can attach to various syntactic categories, such as nouns, adjectives, and verbs, the Russian and German ones can only attach to nouns. This demonstrates that size suffixes across languages do not universally share the same manner or place of syntactic attachment. 12.2.3.1 Manner of syntactic attachment in Spanish In its manner of attachment, the diminutive size suffix -(c)it behaves like a syntactic modifier with a structure such as the one proposed in (15). (15)

(Steriopolo 2015: 9)

The evidence that this suffix is a syntactic modifier comes from the fact that it never produces a grammatical change, either in the syntactic category, or in the grammatical features of the base to which it attaches, including grammatical gender (although it does provide a change in meaning, adding an expressive component). For example, in (16b), the suffix attaches to the masculine noun perr-o ʻdogmasc’, and the resulting diminutive form perr-it-o ʻlittle doggy’ is a masculine noun. In (17b), the same suffix attaches to the feminine form perr-a ʻdog-fem’, and the resulting diminutive noun perr-it-a ʻlittle doggy’ is a feminine noun. (16)

9

a. El perr-o de Juan the.masc dog-masc of Juan ‘Juan’s (male) dog is ill.’ b. El perr-it-o de Juan the.masc dog-dim-masc of Juan ‘Juan’s little (male) doggy is ill.’ (Fortin 2011: 31)

está enferm-o. is ill-masc está enferm-o. is ill-masc

The allomorph -it of the suffix is used with the base forms ending in -o or -a, while the allomorph -cit is used with the base forms ending in -e or a consonant.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek (17)

a. La perr-a de the.fem dog-fem of ‘Juan’s (female) dog is ill.’ b. La perr-it-a de the.fem dog-dim-fem of ‘Juan’s little (female) doggy is (Steriopolo 2015: 10)

371

Juan está enferm-a. Juan is ill-fem Juan está enferm-a. Juan is ill-fem ill.’

The suffix -(c)it can also attach to adjectives, adverbs, and verbs without any categorial change. For example, in (18b), it attaches to the adjective chic-o ʻsmall’ and the resulting diminutive form chiqu-it-o ʻvery small’ remains an adjective. (18)

a. El zapat-o era tan chic-o the.masc shoe-masc was so small-masc ‘The shoe was so small that she/he could not put it b. El zapat-o era tan chiqu-it-o the.masc shoe-masc was so small-dim-masc ‘The shoe was so small that she/he could not put it

que that on.’ que that on.’

no lo pudo calzar. not one could put.on no lo pudo calzar. not one could put.on (Steriopolo 2015: 10)

In addition, the suffix productively uses so-called repeated application (in the sense of Scalise 1984) to intensify the expressive meaning, which is taken as further evidence for the modifier status of the suffix, as shown in (19a)– (19c).10 (19)

a. Es-a cas-it-a era la más chiqu-it-a del barrio. that-fem house-dim-fem was the most small-dim-fem of.the neighbourhood ‘That little house was the smallest in the neighbourhood.’ b. Es-a cas-it-a era la más chiqu-it-it-a del barrio. that-fem house-dim-fem was the most small-dim-dim-fem of.the neighbourhood ‘That little house was the very smallest in the neighbourhood.’ c. Es-a cas-it-a era la más chiqu-it-it-it-a del barrio. that-fem house-dim-fem was the most small-dim-dim-dim-fem of.the neighbourhood ‘That little house was the very tiniest in the neighbourhood.’ (Steriopolo 2015: 11)

12.2.3.2 Place of syntactic attachment in Spanish In terms of the place of syntactic attachment, Steriopolo (2015) shows that the Spanish diminutive suffix -(c)it can merge with various syntactic categories, including nouns, adjectives, adverbs, and verbs. The evidence comes from the fact that the suffix always merges outside the category-forming morphology. Consider first a nominal category: in (20a), the suffix -(c)it attaches outside the nominal suffix -or, as in pint-or-cit-o ʻpainter (derogatory)’. The structure for this noun is proposed in (20b). In the structure, the noun pint-or ʻpainter’ is formed first, and then the diminutive suffix -(c)it attaches above it. 10

See also Steriopolo (2008), which shows that some Russian size suffixes that act as syntactic modifiers can also productively use repeated application to intensify their expressive meanings.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

372

Steriopolo, Markopoulos, Spyropoulos

(20)

a. Picasso piensa/dice que los cuadros del pint-or-cit-o Dalí son malos. Picasso thinks/says that the paintings of.the painter-nom-dim-masc Dalí are bad ‘Picasso thinks/says that the paintings of the painter (derogatory) Dalí are bad.’ (Fortin 2011: 30) b.

Steriopolo (2015) also observes that, if there are multiple nominal suffixes in a single word, the diminutive suffix -(c)it systematically merges above the last nominal suffix, as exemplified in (21). (21)

El dormitorio principal tiene un vest-id-or-cit-o. the.masc bedroom main has a.masc dressing.room-nom-nom-dim-masc ‘The main bedroom has a little dressing room.’ (Steriopolo 2015: 13)

Now consider an adjectival category. When merging with an adjective, the suffix -(c)it attaches outside the categorial morphology. For example, in (22a), the suffix is outside the adjectival suffix -ent, as in the adjective cal-ent-it-a ʻwarm (expressive)’. In the structure (22b), the adjective cal-ent-a ʻwarm’ is formed first, and then the suffix -(c)it merges above it. (22)

a. La mant-a es the.fem blanket-fem is ‘The blanket is very warm.’

muy very

cal-ent-it-a warm-adj-dim-fem

b.

(Steriopolo 2015: 13)

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek

373

Table 12.4 Syntactic variation in attachment of the size suffixes in Russian, German, and Spanish

Size suffixes Russian -k/-ek/-ok/-ik, -c/-ec/-ic, -išč’ German -chen, -lein Spanish -(c)it

Manner of attachment: Syntactic modifiers

Place of attachment: Attaching only to nouns

✓

✓

*

✓

✓

*

Table 12.4 summarizes the syntactic properties of the Spanish size suffix -(c)it, compared with those of Russian and German size suffixes. In its manner of attachment, the suffix -(c)it acts as a syntactic modifier, and in this regard is similar to the Russian size suffixes but different from the German ones, which are syntactic heads. The suffix -(c)it can attach to various syntactic categories, differing from both the Russian and German size suffixes which can only attach to a noun category. 12.2.4

Greek

In Greek, we find many different expressive suffixes with various levels of productivity (see Melissaropoulou 2015 for an overview). Focusing on the two most productive ones, namely -ak and -ul, Steriopolo et al. (2021) have proposed a classification of Greek expressive suffixes according to their semantic type and syntactic behaviour. In the following sections, we briefly review this classification and compare it with the properties we have already discussed with regard to the expressive suffixes in Russian, German, and Spanish. 12.2.4.1 Semantic types of expressive suffixes in Greek The suffix -ak is usually employed to form diminutives, that is, nouns that may denote both the small size of the referent and the (affectionate or derogatory) attitude of the speaker, as exemplified in (23)–(24).11 (23)

11

a. aftokinit-o b. aftokinit-ak-i car-neut.sg car-expr-neut.sg ‘car’ ‘toy car’

All of the examples in §12.2.4 are from Steriopolo et al. (2021).

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

374

Steriopolo, Markopoulos, Spyropoulos

(24)

a. mor-o baby-neut.sg ‘baby’

b. mor-ak-i baby-expr-neut.sg ‘little baby (affect)’

However, there is another -ak suffix that attaches only to masculine or feminine [+human] nominal bases and yields derived nouns that convey only the positive or negative attitude of the speaker towards the referent and, crucially, do not bear information about the size. (25)

a. mam-a mum-fem.sg ‘mum’

b. mam-ak-a mum-expr-fem.sg ‘mummy’

(26)

a. ipalil-os clerk-masc.sg ‘clerk’

b. ipalil-ak-os clerk-expr-masc.sg ‘unimportant clerk (derogatory)’

This distinction in terms of semantic content and distribution suggests the existence of two homophonous suffixes: -ak1 is a size suffix that, as will become evident in §12.2.4.2, may merge with various types of bases, and -ak2 is an attitude suffix that combines only with [+human] nominal bases. An analysis along the same lines is proposed also for the suffix -ul, which is the surface realization of two distinct underlying morphemes, -ul1 and -ul2. The former is an attitude suffix that reveals only the speaker’s emotions or opinion about the referent and has a wide distribution (27)–(28), whereas the latter is a size suffix that attaches only to [–human] nominal bases (29)–(30). (27)

a. andr-as man-masc.sg ‘man/husband’

b. andr-ul1-is man-expr-masc.sg ‘dear husband’

(28)

a. zil-jar-is b. zil-jar-ul1-is jealousy-adj-masc.sg jealousy-adj-expr-masc.sg ‘jealous’ ‘jealous (affect)’

(29)

a. limn-i lake-fem.sg ‘lake’

b. limn-ul2-a lake-expr-fem.sg ‘little lake’

(30)

a. ɣefir-a bridge-fem.sg ‘bridge’

b. ɣefir-ul2-a bridge-expr-fem.sg ‘little bridge’

12.2.4.2 Syntactic types of expressive suffixes in Greek According to Steriopolo et al. (2021), the four suffixes identified above differ from each other not only in their semantic content and distributional properties but also, in their syntactic structure. The structural differences concern both the type of syntactic node these morphemes may merge with (place of attachment) and how the merged constituents interact with each other (manner of attachment).

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek

375

To take the manner of attachment first, there is a basic contrast between size and attitude suffixes. The size suffixes -ak1 and -ul2 behave as heads, as they are able to change the grammatical gender of the base, as shown in examples (31)–(32). (31)

a. karekl-a chair-fem.sg ‘chair’

b. karekl-ak1-i chair-expr-neut.sg ‘little chair’

(32)

a. skat-o shit-neut.sg ‘shit’

b. skat-ul2-a shit-expr-fem.sg ‘little shit’

On the other hand, the attitude suffixes -ak2 and -ul1 are inserted in the syntactic structure as modifiers, which means that, although they can modify the semantic content of the base they attach to (by adding an expressive meaning), they cannot contribute to the grammatical make-up of the derived word; as a result, the base retains its features, for example, its grammatical gender. (33)

a. anθrop-os human-masc.sg ‘human’

b. anθrop-ak2-os human-expr-masc.sg ‘insignificant person (pejor)’

(34)

a. aðerf-i sister-fem.sg ‘sister’

b. aðerf-ul1-a sister-expr-fem.sg ‘little sister (affect)’

As far as the place of attachment is concerned, all four suffixes may merge with a nominalized base, as indicated by the affix ordering in the presence of an overt nominalizer: nom-expr / *expr-nom. (35)

a. anap-tir-as light-nom-masc.sg ‘lighter’

b. anap-tir-ak1-i / *anap-ak1-tir-i light-nom-expr-neut.sg ‘little lighter’

(36)

a. traɣuð-ist-is sing-nom-masc.sg ‘singer’

b. traɣuð-ist-ak2-os / *traɣuð-ak2-ist-os sing-nom-expr-masc.sg ‘unimportant singer (pejor)’

(37)

a. erɣat-ri-a b. erɣat-ri-ul1-a / *erɣat-ul1-ri-a worker-nom.fem-fem.sg worker-nom.fem-expr-fem.sg ‘female worker’ ‘unimportant female worker (pejor)’

(38)

a. sigendro-s-i gather-nom-fem.sg ‘gathering’

b. sigendro-s-ul2-a / *sigendro-ul2-s-a gather-nom-expr-fem.sg ‘small gathering’

However, two of these suffixes, -ak1 and -ul1, can merge not only with nouns, but also with other types of syntactic nodes. The first piece of evidence that points to this conclusion comes again from affix ordering and, in particular, from derived verbs and adjectives where the overt categorizer is placed after

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

376

Steriopolo, Markopoulos, Spyropoulos

and not before the suffix -ul1 (39)–(40). The affix ordering in these examples suggests that -ul1 enters the syntactic structure at the root level. (39)

a. psax-n-o search-ipfv-1sg ‘I search’

b. psax-ul1-ev-o / *psax-ev-ul1-o search-expr-verb-1sg ‘I search with my hands in a specific area’

(40)

a. omorf-os beautiful-masc.sg ‘beautiful’

b. omorf-ul1-ik-os / *omorf-ik-ul1-os beautiful-expr-adj-masc.sg ‘beautiful (affect)’

Furthermore, -ul1 can also attach to adjectival bases, as was shown in (28) and repeated here as (41). (41)

a. zil-jar-is b. zil-jar-ul1-is / *zil-ul1-jar-is jealousy-adj-masc.sg jealousy-adj-expr-masc.sg ‘jealous’ ‘jealous (affect)’

With respect to the suffix -ak1, we find pairs of derived nouns such as the one given in (42), which show that the suffixation of -ak1 may yield derived forms with idiosyncratic meanings, as opposed to the attachment of other suffixes like -ul2, which cannot have any significant bearing on the semantic content of the output. (42)

a. bot-ak1-i boot-expr-neut.sg ‘ankle boot’

b. bot-ul2-a boot-expr-fem.sg ‘small boot (e.g., for kids)’

The semantic difference between the two derived nouns bot-ak1-i and bot-ul2-a suggests that there is also a structural one: -ul2 attaches to an already categorized base that has acquired a concrete meaning, while -ak1 merges directly with the root, thus allowing the output bot-ak1-i to depart from the original semantic content. The structures of the two nouns are presented in (43). (43)

a.

b.

Table 12.5 summarizes the semantic and syntactic properties of the Greek expressive suffixes -ak1, -ak2, -ul1 and -ul2 (see also Steriopolo et al. 2021: 679). What we observe is that Greek expressive morphology exhibits the opposite pattern to what we find in Russian. In Greek, size suffixes are heads, whereas attitude suffixes are modifiers.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek

377

Table 12.5 Semantic and syntactic properties of the Greek expressive suffixes -ak1, -ak2, -ul1, -ul2

12.2.5

Suffix

Semantic Type

Merges as

Merges with

-ak1 -ak2 -ul1 -ul2

size attitude attitude size

head modifier modifier head

n, √ n[+human] n, a, √ n[–human]

Summary

Summarizing what we have seen in the previous sections, our case studies of Russian, German, Spanish, and Greek have revealed that expressive suffixation exhibits extensive syntactic variation both within and across languages. Table 12.6 (for the size suffixes) and Table 12.7 (for the attitude suffixes) provide a comparison of the syntactic properties of expressive suffixes in these language. 12.3

Homophonous expressive suffixes: Case studies of -its in Greek and -ic in Russian

In this section, we focus on two expressive suffixes: -its in Greek and -ic in Russian which seem to have a similar phonological form.12 We show that these suffixes are structurally different both across languages and within the same language. 12.3.1

The Greek expressive suffix -its

The analysis of the suffixes -ak1,2 and -ul1,2 in Steriopolo et al. (2021) finds further support in another productive expressive suffix, namely -its. In what follows, we first discuss the distribution and variation in the semantic content of -its, which suggests the existence of three separate homophonous suffixes. We then focus on the syntactic properties of these suffixes and, in particular, 12

Despite their shared phonology, the two suffixes are not, for the most part, etymologically related. However, according to the online Dictionary of Modern Greek (Institute of Modern Greek Studies, https://bit.ly/3YQLITZ), some instances of the Greek suffix -its3 (see §12.3.1) have a Slavic origin. In this section, we indicate stress in both Greek (§12.3.1) and Russian (§12.3.2) data, because these are new data that could be useful for further research on phonological properties of homophonous suffixes, which we do not investigate in this chapter.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

378

Steriopolo, Markopoulos, Spyropoulos

Table 12.6 Syntactic variation in the attachment of size suffixes in Russian, German, Spanish, and Greek

Size Suffixes Russian -k/-ek/-ok/-ik, -c/-ec/-ic, -išč’ German -chen, -lein Spanish -(c)it Greek -ak1, -ul2

Manner of attachment: Syntactic modifiers

Place of attachment: Attaching only to nouns

✓

✓

*

✓

✓

*

*

*

Table 12.7 Syntactic variation in the attachment of attitude suffixes in Russian and Greek

Attitude suffixes Russian -an’, -aš, -on, -ul’, -un’, -ur, -us’, -uš, -ag, -ak, -al, -an, -ar, -ax, -il, -in, -ob, -ot, -ox, -ug, -uk, -ux Greek -ak2, -ul1

Manner of attachment: Syntactic modifiers

Place of attachment: Attaching only to nouns

*

*

✓

*

on the manner and place of their attachment to the syntactic structure. The upshot of the discussion is that the behaviour of the suffix -its buttresses our earlier claim (ibid.) that size suffixes in Greek are syntactic heads, while attitude suffixes are modifiers. 12.3.1.1 Meaning The suffix -its is almost exclusively found in feminine nouns ending in -a.13 These nouns may:

13

The only exceptions are the neuter noun korítsi ‘girl’ (< kóri ‘daughter’) and some toponyms (e.g., Kalamítsi, Vasilítsi).

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek

379

i. convey information about the small size of an inanimate object or a nonhuman entity, as in (44), and additionally the speaker’s attitude towards it, as in (45)–(46); ii. express only an affectionate or derogatory meaning that the speaker associates with a human referent, as in (47)–(48); iii. be used in the names of towns and other places, as in (49). (44)

a. víð-a screw-fem.sg ‘screw’

b. við-íts-a screw-expr-fem.sg ‘small screw’

(45)

a. pleksúð-a braid-fem.sg ‘braid’

b. pleksuð-íts-a braid-expr-fem.sg ‘small and cute braid’

(46)

a. arkúð-a bear-fem.sg ‘bear’

b. arkuð-íts-a bear-expr-fem.sg ‘little bear (affect)’

(47)

a. θí-a aunt-fem.sg ‘aunt’

b. θ-íts-a aunt-expr-fem.sg ‘auntie (affect)’

(48)

a. panaɣí-a b. panaɣ-íts-a Virgin Mary-fem.sg Virgin Mary-expr-fem.sg ‘Virgin Mary’ ‘sweet Virgin Mary’

(49)

a. paleokastr-íts-a ‘Paleokastritsa (village in Corfu, Greece)’ b. mil-íts-a ‘name of various villages in Greece’

We can thus conclude that -its is a surface form shared by three distinct underlying morphemes with different distributional and semantic properties:14 -its1 attaches to [–human] bases and derives diminutives that denote size and attitude (hence, according to the classification we discussed in §12.1 and §12.2, it is a size suffix); -its2 attaches to [+human] bases and yields nouns with affectionate or derogatory meanings (attitude suffix); and -its3 is found only in fossilized forms used as toponyms. Given that -its3 is not productive and its distribution is limited to proper nouns, the focus in the remainder of this section will be on the other two suffixes, namely -its1 and its2. 12.3.1.2 Structure The suffixes -its1 and -its2 differ not only in their meanings and distribution but also in their syntactic structure. To turn first to the manner

14

In the DM framework that is followed in the present chapter, the term ‘morpheme’ does not refer to specific sublexical units but to abstract morphosyntactic features (or feature bundles) that are realized by phonological exponents (see Embick & Noyer 2007).

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

380

Steriopolo, Markopoulos, Spyropoulos

of syntactic attachment, let us compare on the one hand the examples in (50)– (51), and on the other, (52)–(53). (50)

a. karf-í nail-neut.sg ‘nail’

b. karf-íts1-a nail-expr-fem.sg ‘pin’

(51)

a. papá-s priest-masc.sg ‘priest’

b. papað-íts1-a15 priest-expr-fem.sg ‘great tit (Parus major)’

(52)

a. kopél-a girl-fem.sg ‘girl’

b. kopel-íts2-a girl-expr-fem.sg ‘girl (affect)’

(53)

a. kúkl-a doll-fem.sg ‘doll, beautiful girl’

b. kukl-íts2-a doll-expr-fem.sg ‘little doll, beautiful girl (affect)’

We observe that -its1 derives feminine nouns even when the base noun is neuter or masculine, whereas -its2 cannot trigger any change in the grammatical gender since it attaches only to feminine bases. This difference suggests that -its1 functions as a categorizing head that projects its features to the derived structure, whereas -its2 behaves as a modifier that attaches to the structure as an adjunct, as illustrated in (54a) and (54b). (54)

a.

b.

So far, we have seen that both -its1 and -its2 combine with nominal bases and form feminine nouns. But if we take a closer look at the specifics of the derivations each morpheme is involved in, we can detect differences with respect to the place these suffixes occupy in the syntactic structure. Consider, for instance, the data in (50) (repeated here for convenience in (55)) and (56), and then compare them with the data in (57)–(58). (55)

15

a. karf-í nail-neut.sg ‘nail’

b. karf-íts1-a nail-expr-fem.sg ‘pin’

The stem allomorph papað- also emerges in the plural forms of the base noun, e.g., papáð-es ‘priests’.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek (56)

a. karf-í nail-neut.sg ‘nail’

b. karf-ák1-i nail-expr-neut.sg ‘small nail’

(57)

a. gofrét-a wafer-fem.sg ‘wafer’

b. gofret-íts1-a wafer-expr-fem.sg ‘small wafer’

(58)

a. gofrét-a wafer-fem.sg ‘wafer’

b. gofret-ák1-i wafer-expr-neut.sg ‘wafer bite’

381

We can see a clear distinction between the two sets of data. In fact, one seems to be the mirror image of the other. In (55), the attachment of the suffix -its1 to the base karf- ‘nail’ yields a noun with a significantly different meaning: karfítsa does not denote a ‘small nail’ but a different object, that is, a ‘pin’. By contrast, the suffixation of -ak1 in (56) results in a derived noun with transparent semantics. In DM terms, the difference between the two derived forms can be captured as a difference in the place the expressive suffix occupies: -its1 merges directly with the root realized as karf and forms a noun with non-transparent semantics, as in (59a), whereas -ak1 attaches to a base that has been already categorized and has thus acquired a fixed meaning as “nail” in (59b). (59)

a.

b.

The situation is reversed with the pair gofretítsa – gofretáki. Here, it is the suffix -ak1 that attaches to the root and yields a non-compositional meaning (gofretáki is a differently shaped wafer and not just a small wafer), whereas -its1 combines with a nominal base in the output that has predictable semantics. The two sets of examples suggest that -its1 has two potential places of attachment – it can either merge directly with roots (60a) or it can attach to nominalized bases (60b). (60)

a.

b.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

382

Steriopolo, Markopoulos, Spyropoulos

The variable place of attachment of the suffix -its1 stands in contrast to the syntactic behaviour of -its2, which can only modify a nominal base and cannot interact directly with a root. As shown in the examples in (61)–(62), the suffixation of -its2 can only add an affectionate or derogatory meaning and does not alter the semantic content of the base noun. (61)

a. filenáð-a b. filenað-íts2-a girlfriend-fem.sg girlfriend-expr-fem.sg ‘girlfriend’ ‘girlfriend (affect)’

(62)

a. ðaskál-a b. ðaskal-íts2-a teacher-fem.sg teacher-expr-fem.sg ‘female teacher’ ‘female teacher (pejor)’

12.3.1.3 Summary In §12.3.1, we have discussed various manifestations of the expressive suffix -its. Based on its distribution and the attested semantic differences, we have concluded that -its is a surface form shared by three different morphemes: -its1, -its2, and -its3. Given that the last one is no longer productive and is found only in fossilized toponyms, we have focused on the syntactic properties of the first two, summarized in Table 12.8. We have shown that the behaviour of -its1 and -its2 suggests that, unlike what happens in Russian, Greek size suffixes are syntactic heads and Greek attitude suffixes are syntactic modifiers, corroborating the conclusions in our earlier work (Steriopolo et al. 2021). 12.3.2

The Russian suffix -ic

The Russian suffix -ic, like the Greek suffix -its, has at least three homophonous counterparts. However, unlike the Greek suffix, only one instance of -ic has an expressive content. The other two instances have non-expressive meanings. We will first discuss the different meanings and distributions of the suffix -ic, and then propose syntactic analyses of all three homophonous suffixes. We will show that they differ in both the manner and the place of syntactic attachment, arguing that the expressive homophone -ic is a size suffix whose syntactic behaviour supports the earlier claim (§12.2.1.2) that Russian size suffixes systematically act as syntactic modifiers.

Table 12.8 Basic properties of the Greek expressive suffixes -its1 and -its2 Suffix

Semantic type

Merges as

Merges with

-its1 -its2

size attitude

head modifier

n, √ n[+human]

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek

383

12.3.2.1 Meaning and distribution The suffix -ic is mostly found in feminine nouns ending in -a (as with the Greek suffix -its, discussed above). The exception is the expressive instance of the suffix -ic, which can attach to both feminine and neuter nouns, unlike the Greek suffix. In what follows, we will describe the meanings and the distribution of all three homophonous counterparts of the suffix -ic, which for convenience we will call here -ic1, -ic2 and -ic3. The suffix -ic1 forms (mostly) abstract nouns from other syntactic categories, such as adjectives, prefix + noun collocations, and other nouns. The resulting noun with the suffix -ic1 is always feminine. For example, in (63b) and (64b), the suffix -ic1 attaches to adjectives, as indicated by the adjectival suffix -n that precedes it. This results in feminine nouns with an abstract meaning. In (65b), the suffix attaches to a prefix + noun collocation (‘without work’), yielding the abstract noun ‘unemployment’. And in (66b), it attaches to the male first name Cyril, forming a feminine noun that denotes the Slavic writing system (‘Cyrillic script’). (63)

a. ráz-n-yj different-adj-masc.n.sg ‘different’

b. ráz-n’-ic-a different-adj-ic1-n.sg (fem) ‘difference’

(64)

a. jed’í-n-yj single-adj-masc.n.sg ‘single’

b. jed’i-n’-íc-a single-adj-ic1-n.sg (fem) ‘unit/number one’

(65)

a. bez rabót-y b. bez-rabót’-ic-a without work-acc.sg (fem) without-work-ic1-n.sg (fem) ‘without work’ ‘unemployment’

(66)

a. Kiríll Kirill.n.sg (masc) ‘Cyril16 (male name)’

b. Kiríll’-ic-a Kirill-ic1-n.sg (fem) ‘Cyrillic script’

The second instance of the suffix -ic, namely -ic2, forms nouns that denote females. Such nouns can refer to female occupations or personal qualities (e.g., beauty, laziness, and stubbornness), or denote the female sex of a biological species. The resulting nouns with the suffix -ic2 are always feminine. For example, in (67c) and (68b), the suffix -ic2 attaches to adjectives, as evidenced by the adjectival suffixes -n and -av, respectively. The resulting forms are feminine nouns that denote a female occupation (‘female mushroomer’) or quality (‘beautiful woman’). Notice that the suffix -ic2 in such nouns is a feminine equivalent of the male forms that end in -ik, as in (67b), and -ec, as in (68a). For instance, in (67b), the noun gr’ib-n’-ík denotes a male mushroomer

16

Saint Cyril and his brother, Saint Methodius, are considered to be the creators of the Slavic alphabet based on Greek letters (later called ‘the Cyrillic alphabet’).

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

384

Steriopolo, Markopoulos, Spyropoulos

(or it could also refer to a mixed group of people, as evidenced by the plural form; see §12.3.2.2 for an explanation), while in (67c), the noun gr’ib-n’-íc-a denotes a female mushroomer.17 In (69b) and (70b), the suffix -ic2 attaches to the nouns denoting a biological species, such as ‘bear’ and ‘eagle’. The resulting forms are feminine nouns for the female of the species, such as ‘female bear’ and ‘female eagle’, respectively. The exceptions are a few nouns with the suffix -ic that denote a species in general (some animals and birds), such as kun’-íc-a ‘marten (male or female)’, s’in’-íc-a ‘tomtit (male or female)’, and pt’-íc-a ‘bird (male or female)’. (67)

a. gr’ib-n-ój mushroom-adj-masc.n.sg ‘mushroom (adj)’ b. gr’ib-n’-ík c. gr’ib-n’-íc-a mushroom-adj-n.sg (masc) mushroom-adj-ic2-n.sg (fem) ‘mushroomer’ ‘a female mushroomer’

(68)

a. kras-áv’-ес b. kras-áv’-iс-a red/beauty-adj-n.sg (masc) red/beauty-adj-ic2-n.sg (fem) ‘handsome person’ ‘beautiful woman’

(69)

a. m’edv’éd’ bear.n.sg (masc) ‘bear’

b m’edv’éd’-ic-a bear-ic2-n.sg (fem) ‘female bear’

(70)

a. or’ël eagle.n.sg (masc) ‘eagle’

b. orl’-íc-a eagle-ic2-n.sg (fem) ‘female eagle’

The third instance of the suffix, the homophone -ic3, forms expressive diminutive nouns. It can attach to feminine or neuter nouns and create expressive forms with the same grammatical gender as the base forms. The resulting nouns can denote a small size or have an endearing function. Following the semantic classification of Steriopolo (2008, 2009), the suffix -ic3 is classified as an expressive size suffix (expr size), because it can convey both an attitude (endearment) and the small size (diminutive) of a referent. For example, in (71b) and (72b), the suffix -ic3 attaches to the feminine nouns róšč’-a ‘grove’ and vod-á ‘water’, forming the expressive forms: róšč’-ic-a ‘small grove’ and vod’-íc-a ‘water (endearing)’. The resulting expressive nouns are feminine (ending in -a) and preserve the grammatical gender of the base (feminine). In (73b) and (74b), the same suffix attaches to the neuter nouns plát’-e ‘dress’ and másl-o ‘butter’. The resulting expressive nouns plát’-ic-e ‘little dress’ and másl’-ic-e ‘butter (endearing)’ are also neuter (ending in -e), thus preserving the grammatical gender of the base to which they attach.

17

This meaning is from Efremova (2006). Another meaning of this Russian noun is ‘mycelium’.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek (71)

a. róšč’-a grove-n.sg (fem) ‘grove’

b. róšč’-ic-a grove-ic3-n.sg (fem) ‘small grove’

(72)

a. vod-á water-n.sg (fem) ‘water’

b. vod’-íc-a water-ic3-n.sg (fem) ‘water (endearing)’

(73)

a. plát’-e dress-n.sg (neut) ‘dress’

b. plát’-ic-e dress-ic3-n.sg (neut) ‘little dress’

(74)

a. másl-o butter-n.sg (neut) ‘butter’

b. másl’-ic-e butter-ic3-n.sg (neut) ‘butter (endearing)’

385

To summarize, there are at least three homophonous suffixes -ic in Russian: (i) -ic1 that forms abstract nouns (always feminine); (ii) -ic2 that forms nouns with a female denotation (always feminine); and (iii) -ic3 that forms expressive diminutive nouns (feminine or neuter, dependent on the gender of a base form). Thus, like the Greek suffix -its discussed above, the surface form of the Russian suffix -ic is shared by at least three homophonous morphemes with different meanings and distribution. However, unlike the Greek suffix -its, which has two expressive homophones, the Russian suffix -ic has just one, namely -ic3, with the other two being non-expressive. In addition, the Greek suffix -its is almost exclusively found in feminine nouns ending in -a, while in Russian, the expressive homophone -ic3 is found in both feminine nouns ending in -a and neuter nouns ending in -e. 12.3.2.2 Structure We will argue that the non-expressive suffixes -ic1 and -ic2 are syntactic heads, as illustrated in (75a), while the expressive size suffix -ic3 is a syntactic modifier, as in (75b). (75)

a.

b.

This proposal stems from the fact that the non-expressive suffixes -ic1 and -ic2 can change the syntactic category or grammatical gender of the base to which they attach, while -ic3 does not produce any grammatical change. Consider first the suffix -ic1 that forms abstract nouns. It can attach to different syntactic categories (adjectives or nouns), always forming a noun of feminine gender. For example, in (76b), repeated from (63b), the suffix attaches to an adjective, as evidenced by the adjectival suffix -n which is present in the resulting

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

386

Steriopolo, Markopoulos, Spyropoulos

form ráz-n’-ic-a ‘difference’ and precedes the suffix -ic1. A structure for the resulting feminine noun is given in (76c). Here, an adjective is formed first with the adjectival suffix -n, and then the nominal suffix -ic1 merges above it, forming a noun with feminine grammatical gender, nfem. (76)

a. ráz-n-yj different-adj-masc.n.sg ‘different’

b.

ráz-n’-ic-a different-adj-ic1-nsg (fem) ‘difference’

c.

In (77b), repeated here from (66), the same suffix merges with a masculine noun denoting the male first name Kiríll ‘Cyril’. The resulting feminine noun is Kiríll’-ic-a ‘Cyrillic script’. We propose that in the structure in (77c), the nominal suffix -ic1 merges above the male first name, changing the grammatical gender of the base noun from masculine to feminine: Kiríll (masc) – Kiríll’-ic-a (fem). (77)

a. Kiríll Kirill.n.sg (masc) ‘Cyril (male name)’

b. Kiríll’-ic-a Kirill-ic1-n.sg (fem) ‘Cyrillic script’

c.

Consider now the suffix -ic2 that denotes females. Similarly to the suffix -ic1, discussed above, this suffix can attach to various syntactic categories (adjectives or nouns), always forming a feminine noun. We propose that the difference between the suffixes -ic1 and -ic2 is that -ic2 is specified for the feature [+female],

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek

387

while -ic1 is not. The reason for assuming this is that the suffix -ic2 can only refer to females and never to males or mixed groups of people. For example, in (78a), repeated from (68a), the masculine noun kras-áv’-ес ‘handsome person’ can refer to a male person or any person in general (without specifying the referent’s gender), as evidenced by the plural form kras-áv-с-y ‘handsome people’, which can refer to a mixed group of people consisting of male and female individuals. However, the feminine noun kras-áv’-iс-a ‘beautiful woman’ can only refer to a female (the plural form kras-áv’-iс-y ‘beautiful women’ can only refer to refer to females, never to males or mixed groups of people). In the Russian language, there are many examples of the suffix -ic2 referring solely to female individuals, especially female occupations and personal qualities, such as korm’il’-ec ‘feeder’ – korm’il’-ic-a ‘female feeder’, polkovn’-ik ‘colonel’ – polkovn’-ic-a ‘colonel’s wife’,18 len’iv’-ec ‘lazy person’ – len’iv’-ic-a ‘lazy woman’, upr’jam’-ec ‘stubborn person’ – upr’jam’-ic-a ‘stubborn woman’, among many others (these data are from the Russian explanatory dictionary by Efremova (2006)). The structure for (78b) is given in (78c). In this structure, the suffix -ic2 merges on top of an adjective formed by the adjectival suffix -av, and it is specified for the feature [+female], which makes the whole derivation krasáv’-iс-a ‘beautiful woman’ not only grammatically feminine (similarly to -ic1), but also semantically female, referring exclusively to women (unlike -ic1, which can form abstract feminine nouns). (78)

a. kras-áv’-ес red/beauty-adj-n.sg (masc) ‘handsome person’

b. kras-áv’-iс-a red/beauty-adj-ic2-n.sg (fem) ‘beautiful woman’

c.

So far, we have shown that both homophonous suffixes -ic1 and -ic2 act as syntactic heads, because they consistently form nouns from other syntactic categories (e.g., adjectives). Thus, they produce a change in the syntactic 18

Historically, there were no female colonels. The noun denotes a colonel’s wife.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

388

Steriopolo, Markopoulos, Spyropoulos

category. In addition, when attaching to other nouns, they can change the grammatical gender of the base, always resulting in feminine nouns. We conclude that these suffixes are noun heads, specified for feminine grammatical gender, nfem. We have also shown that one of them, namely -ic2, is also specified for the feature [+female]. The third homophonous suffix, -ic3, is different from the other two in its syntactic properties. Crucially, unlike -ic1 and -ic2, it acts as a syntactic modifier. The reason for this assumption is the absence of any grammatical change in either the syntactic category or the grammatical gender of the base noun to which it attaches. This suffix can only attach to feminine or neuter nouns and form expressive nouns of the same grammatical gender. It never attaches to other syntactic categories and never causes a change in the grammatical gender of the input. Based on these facts, we conclude that -ic3 is a modifier of feminine or neuter grammatical gender nouns, that is, nfem or nneut respectively, as illustrated in the structure in (79). (79)

Consider, for example, the data in (80b), repeated from (71b). In the structure in (80c), the feminine noun nfem is formed first, then the expressive size suffix -ic3 merges as a noun modifier, yielding no syntactic change in the syntactic category (noun) or the gender feature (feminine), but adding a diminutive meaning to the whole formation: róšč’-ic-(a) ‘small grove’ (the suffix merges with neuter nouns in the same syntactic manner – see Examples (73) and (74)). (80)

a. róšč’-a grove-n.sg (fem) ‘grove’

b. róšč’-ic-a grove-ic3-n.sg (fem) ‘small grove’

c.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek

389

Table 12.9 Basic properties of the Russian homophonous suffixes -ic1, -ic2, and -ic3 Suffix

Semantic type

Merges as

Merges with

-ic1

non-expressive (abstract nouns)

a, n

-ic2

non-expressive (feminine nouns) expressive size (diminutive)

head (nfem) head (nfem [+female]) modifier

-ic3

a, n nfem, nneut

12.3.2.3 Summary We have shown that the Russian suffix -ic has at least three homophones: (i) -ic1 that forms abstract nouns (always feminine); (ii) -ic2 that denotes females (always feminine); and (iii) -ic3 that forms expressive diminutive nouns (feminine or neuter, dependent on the grammatical gender of the base noun). Only one of them, namely -ic3, is an expressive size suffix (expr size). We have argued that these homophonous suffixes differ in both the manner and place of syntactic attachment. Concerning the manner of attachment, the suffixes -ic1 and -ic2 are syntactic noun heads, nfem, specified for feminine grammatical gender. In addition, the suffix -ic2 is specified for the feature [+female]. By contrast, the expressive suffix -ic3 is a syntactic modifier. Concerning the place of attachment, the suffixes -ic1 and -ic2 can merge with various syntactic categories, while -ic3 can only merge with feminine and neuter nouns. These properties are summarized in Table 12.9. These findings support our earlier claim (§12.2.1) that the Russian size suffixes consistently act as noun modifiers. They also offer additional evidence for Steriopolo’s (2014) earlier claim that homophonous objects systematically differ not only in their meanings, but also – crucially – in their syntactic structures. 12.4

Conclusions

We have presented a comparison of four case studies of expressive suffixes in Russian, German, Spanish, and Greek. Although expressive suffixes across these languages have been shown to have a similar function, as they all express the speaker’s attitude (positive or negative) and can also indicate the referent’s size (big or small), they differ significantly in the syntactic properties, as they exhibit divergent patterns regarding the manner and place of syntactic attachment. For example, in Russian, attitude suffixes are consistently syntactic heads, while size suffixes are syntactic modifiers. However, the Greek

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

390

Steriopolo, Markopoulos, Spyropoulos

expressive suffixes -ak and -ul exhibit exactly the opposite pattern: attitude suffixes are consistently syntactic modifiers, while size suffixes are syntactic heads. This shows that expressive suffixes, regardless of whether they fall into the same semantic category, do not universally share the same manner of syntactic attachment. We also observe a similar picture when we look at the place of attachment. Again, there is no uniformity with regard to the place an expressive suffix may occupy in the syntactic structure. For instance, the size suffixes in Russian and German merge only with categorized nouns, but in Spanish, the productive suffix -(c)it merges with various syntactic categories; and in Greek, the suffix -ak attaches both to categorized bases and acategorial roots. For further work, it would be worth expanding this research to other branches of the Indo-European family and other language families with different morphological types in order to explore whether there is further syntactic variation and whether there are systematic patterns in how languages differ with respect to the manner and place of syntactic attachment of expressive suffixes. As a final point, we would like to emphasize that according to the findings of this study, syntactic variation emerges not only across languages with different syntactic structures, but also within the same language, occurring both across different vocabulary items (cf., for example, the Greek suffixes -ak1 and -ul2 denoting size, and -ak2 and -ul1 denoting attitude) and across homophonous items (cf., for example, the homophonous suffixes -ak1/-ak2, -ul1/-ul2, -its1/its2/-its3 in Greek and -ic1/-ic2/-ic3 in Russian). With this, we conclude that expressive meaning is not associated with a specific syntactic structure, either across languages or within a single language. Abbreviations 1sg acc adj (also a) affect aug dim (also dim) expr fem ipfv masc n n.sg neut nom pl

1st person singular Accusative adjective affectionate augmentative diminutive expressive feminine imperfective masculine noun Nominative singular neuter nominal plural

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Case studies: Russian, German, Spanish, Greek ptcp pref v vulg √

391

participle prefix verb vulgar root

References Bachrach, Asaf & Wagner, Michael. 2007. Syntactically driven cyclicity versus output– output correspondence: The case of adjunction in diminutive morphology. Presented at the 30th Annual Penn Linguistics Colloquium, Philadelphia, 24–26 February. Bierwisch, Manfred. 2003. Heads, complements, adjuncts: Projections and saturation. In E. Lang, C. Maienborn & C. Fabricius-Hansen (eds.) Modifying Adjuncts. Berlin and New York: de Gruyter, 113–59. Efremova, T. F. (Ефремова, Т. Ф.) (ed.) 2006. Современный толковый словарь русского языка [Sovremennyj tolkovyj slovar russkogo jazyka] [The modern explanatory dictionary of the Russian language]. 3 volumes. Moscow: Ast, Astrel’. Embick, David. 1998. Voice systems and the syntax/morphology interface. In Heidi Harley (ed.) MIT Working Papers in Linguistics 32: Papers from the UPenn/MIT roundtable on argument structure and aspect, Cambridge, MA: MITWPL, 41–72. 2010. Localism versus Globalism in Morphology and Phonology. Cambridge, MA: MIT Press. Embick, David & Rolf Noyer. 2007. Distributed morphology and the syntaxmorphology interface. In Gillian Ramchand & Charles Reiss (eds.) The Oxford Handbook of Linguistic Interfaces. Oxford: Oxford University Press, 289–324. Fortin, Antonio. 2011. The morphology and semantics of expressive affixes. Doctoral dissertation, University of Oxford. Halle, Morris. 1997. Distributed morphology: Impoverishment and fission. In Benjamin Bruening, Yoonjung Kang & Martha McGinnis (eds.) MIT Working Papers in Linguistics 30: Papers at the interface, Cambridge, MA: MITWPL, 425–49. Halle, Morris & Alec Marantz. 1993. Distributed morphology and the pieces of inflection. In Kenneth Hale & Samuel J. Keyser (eds.) The View from Building 20: Essays in linguistics in honor of Sylvain Bromberger. Cambridge, MA: MIT Press, 111–76. 1994. Some key features of distributed morphology. In MIT Working Papers in Linguistics 21, Cambridge, MA: MITWPL, 275–88. Harley, Heidi & Rolf Noyer. 1998. Licensing in the non-lexicalist lexicon: Nominalizations, vocabulary items and the encyclopedia. In Heidi Harley (ed.) MIT Working Papers in Linguistics 32: Papers from the UPenn/MIT roundtable on argument structure and aspect. Cambridge, MA: MITWPL, 119–37. 1999. Distributed morphology. Glot International 4(4), 3–9. Marantz, Alec. 1997. No escape from syntax: Don’t try morphological analysis in the privacy of your own lexicon. University of Pennsylvania Working Papers in Linguistics 4(2), 201–25. 2001. Words and things. Manuscript. New York University.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

392

Steriopolo, Markopoulos, Spyropoulos

2007. Phases and words. In S. H. Choe (ed.) Phases in the Theory of Grammar. Seoul: Dong-In Publishing, 191–222. Melissaropoulou, Dimitra. 2015. Modern Greek. In Nicola Grandi & Lívia Körtvélyessy (eds.) Edinburgh Handbook of Evaluative Morphology. Edinburgh: Edinburgh University Press, 269–77. Pesetsky, David. 1995. Zero Syntax: Experiencers and cascades. Cambridge, MA: MIT Press. Scalise, Sergio. 1984. Generative Morphology. Dordrecht: Foris Publications. Schütze, Carson T. 1995. PP attachment and argumenthood. In MIT Working Papers in Linguistics 26. Cambridge, MA: MITWPL, 95–151. Siddiqi, Daniel. 2010. Distributed morphology. Language and Linguistics Compass 4(7), 524–42. Steriopolo, Olga. 2008. Form and function of expressive morphology: A case study of Russian. Doctoral dissertation, University of British Columbia. 2009. Form and function of expressive morphology: A case study of Russian. Russian Language Journal 59, 149–94. 2014. Parameters of variation in the syntax of homophones. Poljarnyj vestnik: Norwegian Journal of Slavic Studies 17, 46–73. 2015. Syntactic variation in expressive size suffixes: A comparison of Russian, German, and Spanish. SKASE Journal of Theoretical Linguistics 12(1), 2–21. 2016. Expressives across languages: Form/function correlation. Annual Review of Linguistics 2, 293–324. 2017. Syntactic variation in diminutive suffixes: Russian, Kolyma Yukaghir, and Itelmen. Languages 2(4), 1–23. Steriopolo, Olga, Markopoulos, Giorgos & Spyropoulos, Vassilios. 2021. A morphosyntactic analysis of nominal expressive suffixes in Russian and Greek. The Linguistic Review 38(4), 645–86. Wiltschko, Martina. 2006. Why should diminutives count? In H. Broekhuis, N. Corver, R. Huijbregts, U. Kleinhenz, & J. Koster (eds.) Organizing Grammar: Linguistic studies in honor of Henk van Riemsdijk. Berlin: Walter de Gruyter, 669–79. 2008. The syntax of non-inflectional plural marking. Natural Language & Linguistic Theory 26(3), 639–94. 2014. The Universal Structure of Categories: Towards a formal typology. Cambridge: Cambridge University Press. Wiltschko, Martina & Olga Steriopolo. 2007. Parameters of variation in the syntax of diminutives. In Milica Radisic (ed.) Proceedings of the 2007 Canadian Linguistics Association Annual Conference, 1–12.

https://doi.org/10.1017/9781108989084.020 Published online by Cambridge University Press

Index

Abkhaz, 7, 337, 341, 346, 357 Abkhaz-Adyghean, 7, 337, 341, 346, 357 aesthetic power, 328 affixes, 30, 135, 137, 163, 164, 165, 167, 170, 173, 175, 177, 214, 271, 287, 288, 339, 391 amplification, 55, 62, 63, 187 Ancient Greek, 143, 152, 158, 180, 201, 202, 218 arbitrariness, 1, 50, 118, 225 Archi, 342, 345, 346, 357 Aromanian augmentation, 80, 104, 110, 130, 163, 179, 187, 188, 190, 192, 193, 194, 203, 212, 213, 224 baby talk, 105, 110, 115, 234, 235, 236, 246, 264 Basque, 7, 101, 287, 290, 313, 314, 315, 316, 317, 318, 320, 321, 322, 324, 325, 326, 327, 328, 330, 332, 333, 334, 357 Bezht’a, 343, 346 borrowings, 21, 34, 38 Breton, 7, 295, 296, 299, 300, 304, 305, 306, 308 Budukh, 339, 344, 346, 357 Catalan, 6, 227, 269, 284, 286, 288, 290, 291 Cheremiss. See Mari clipping, 4, 15, 16, 20, 21, 25, 27, 28, 29, 32, 33, 34, 39, 48 embellished, 4, 13, 15, 22, 23, 37, 41, 45 coda deletion, 29 cognitive architecture, 2 collective suffixes, 182, 204, 209, 210, 212 compounding, 6, 15, 20, 21, 80, 99, 107, 108, 112, 115, 118, 123, 129, 131, 132, 135, 141, 144, 150, 154, 157, 161, 208, 211, 215, 218, 222, 226, 242, 321, 330 concatenative, 3, 16, 17 COVID-19, 3 descriptive ineffability, 127, 132, 254 descriptives, 123, 124, 125, 128 diminutive, 4, 15, 16, 24, 33, 34, 35, 38, 39, 40, 41, 45, 46, 51, 78, 100, 103, 106, 115, 118, 164, 167, 170, 178, 187, 188, 196,

197, 200, 201, 202, 203, 219, 222, 225, 238, 298, 305, 363, 366, 367, 368, 369, 370, 371, 372, 384, 385, 388, 389, 390, 391, 392 diminutive formation, 4, 16 Distributed Morphology, 362, 391, 392 distributive numerals, 31, 63 distributivity, 54, 55, 59, 62, 63, 67, 70, 78, 80, 84 doubling, 4, 53, 69, 87, 101, 106, 109, 113, 115, 116, 232, 237, 238, 252 echo word formations. See echo words echo words, 5, 91, 93, 94, 95, 96, 98 echoant defined, 96 echo-pairs. See echo words elative, 71, 76, 246, 261, 263 English, xiii, 5, 18, 19, 21, 22, 27, 28, 34, 36, 39, 40, 42, 44, 46, 48, 49, 50, 51, 54, 55, 56, 61, 65, 66, 85, 91, 93, 94, 101, 111, 112, 117, 118, 140, 145, 152, 159, 161, 176, 181, 186, 202, 204, 206, 210, 214, 216, 221, 222, 223, 226, 227, 233, 234, 240, 241, 252, 257, 258, 261, 263, 265, 268, 269, 286, 290, 300, 305, 309, 315, 353, 354, 355 Estonian, 54, 55, 61, 62, 63, 85, 86 evaluativity, 130 expressive function, 5, 39, 104, 134, 163, 169, 183, 345 expressive meaning, 6, 8, 14, 47, 103, 112, 113, 114, 115, 123, 124, 125, 127, 133, 135, 137, 138, 139, 141, 146, 148, 153, 159, 170, 174, 187, 216, 225, 318, 361, 371, 375, 390 expressive morphology, 5, 7, 14, 36, 52, 94, 102, 103, 104, 107, 115, 116, 119, 129, 133, 134, 167, 173, 226, 227, 299, 305, 306, 376, 392 expressives, 1, 2, 5, 6, 7, 13, 14, 15, 47, 50, 92, 93, 96, 98, 99, 100, 101, 104, 106, 117, 123, 124, 125, 126, 127, 128, 129, 130,

393

https://doi.org/10.1017/9781108989084.021 Published online by Cambridge University Press

394

Index

132, 140, 141, 161, 221, 255, 257, 266, 269, 272, 274, 276, 279, 280, 283, 286, 288, 289, 295, 296, 332, 333, 337, 338, 341, 342, 343, 344, 345, 346, 347, 348, 350, 351, 352, 353, 354, 355, 356, 357, 358, 392 expressivity principle of, 1, 2, 3, 5, 6, 8, 13, 14, 16, 17, 53, 91, 92, 94, 96, 98, 99, 100, 113, 118, 123, 124, 127, 129, 130, 133, 139, 140, 145, 149, 151, 155, 159, 160, 162, 163, 166, 169, 184, 187, 188, 193, 195, 211, 212, 213, 214, 215, 216, 261, 262, 272, 316, 327 extragrammatical, 114 extra-grammatical morphology, 14 Finnish, 4, 30, 32, 33, 35, 40, 43, 44, 49, 53, 54, 55, 56, 57, 59, 60, 61, 62, 63, 64, 85, 88 Finno-Ugric, 4, 29, 31, 36, 38, 41, 44, 47, 51, 53, 54, 55, 61, 75, 77, 84, 85, 88 folk song, 157, 328 French, 7, 27, 36, 39, 56, 108, 152, 158, 176, 192, 204, 214, 225, 241, 242, 252, 253, 261, 287, 295, 296, 299, 300, 301, 305, 306 gender, 109, 167, 174, 187, 188, 190, 193, 196, 197, 200, 203, 215, 219, 304, 339, 344, 364, 366, 367, 370, 375, 380, 384, 385, 386, 388, 390 Georgian, 7, 337, 338, 341, 343, 345, 346, 347, 349, 350, 351, 353, 354, 357, 358 German, 5, 8, 42, 47, 79, 93, 103, 104, 105, 106, 107, 109, 112, 114, 115, 116, 117, 118, 119, 125, 128, 131, 135, 140, 190, 202, 204, 207, 211, 220, 224, 226, 265, 286, 299, 309, 361, 366, 367, 369, 373, 377, 378, 389, 392 Greek, 6, 8, 38, 123, 131, 132, 133, 134, 135, 136, 138, 139, 140, 141, 143, 145, 148, 149, 150, 151, 154, 155, 156, 158, 159, 160, 161, 162, 165, 171, 175, 179, 180, 181, 182, 189, 190, 192, 193, 201, 202, 203, 207, 209, 211, 212, 214, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 231, 249, 258, 260, 265, 361, 362, 365, 373, 374, 376, 377, 378, 382, 383, 385, 389, 390, 392 Hamilton-Kager Conundrum, 26 Hinuq, 344, 358 honorifics, 15 Hungarian, 3, 4, 13, 15, 16, 17, 18, 19, 20, 21, 22, 24, 26, 28, 29, 31, 32, 36, 38, 39, 40, 42, 44, 45, 47, 48, 54, 77, 78, 79, 80, 84, 85, 86

hypocoristics, 4, 16, 23, 25, 27, 28, 36, 39, 41, 44, 235, 236, 267 Icari Dargwa, 344 iconicity, 1, 7, 21, 70, 92, 171, 220, 261, 262, 270, 283, 289, 290, 291, 302, 307, 332 ideophones, 1, 2, 5, 7, 15, 55, 61, 70, 71, 72, 74, 75, 76, 78, 84, 87, 91, 93, 101, 104, 105, 106, 112, 117, 234, 236, 262, 270, 279, 283, 284, 285, 286, 287, 288, 290, 295, 300, 302, 313, 314, 315, 316, 317, 319, 320, 321, 323, 325, 326, 327, 328, 330, 332, 333, 337, 340, 341, 344, 345, 356, 357 ideophonic language, 7, 313, 327 indefiniteness, 54, 55, 63, 69, 71, 78, 197 intensification, 4, 17, 18, 30, 31, 34, 37, 44, 53, 77, 78, 79, 104, 130, 139, 149, 163, 187, 188, 193, 194, 221, 243, 244, 246, 247, 261, 297, 305 interjections, 14, 108, 110, 115, 116, 125, 140, 239, 251, 279, 290, 295, 296, 297, 298, 300, 303, 304, 306, 308, 331, 337 Italian, 6, 27, 36, 38, 40, 46, 47, 51, 179, 188, 191, 204, 214, 220, 222, 231, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 247, 249, 251, 252, 253, 255, 256, 257, 258, 259, 260, 261, 263, 265, 266, 267, 268, 300 Kartvelian, 7, 337, 339, 345, 346, 357, 358 Khanty, 4, 30, 32, 35, 44, 47, 50, 54, 77, 81, 83 Khwarshi, 344, 358 kinship terms, 33, 35, 106 Komi, 4, 30, 31, 33, 34, 37, 43, 44, 46, 49, 50, 54, 70, 71, 72, 73, 74, 75, 76, 84, 85, 87 Lak, 343, 346 Latvia, 62 Lezgian, 343, 346 Lithuanian, 40, 202 Majorca, x, 6 Maltese, 6, 249, 260 Mansi, 4, 29, 30, 31, 35, 44, 46, 49, 50, 54, 77, 81, 82 Mari, 4, 30, 31, 32, 33, 35, 37, 43, 44, 46, 50, 51, 61, 66, 67, 70, 84, 85 Medieval Greek, 207 metacommentary, 14 metaphor, 6, 140, 149, 188, 213, 215 Middle Greek, 155, 209 mimetics, 295, 298, 301, 302, 306, 332, 333 Mordvinic languages, 64

https://doi.org/10.1017/9781108989084.021 Published online by Cambridge University Press

Index Nakh-Daghestanian, 7, 337, 339, 341, 342, 343, 344, 345, 346, 356, 357, 358 nominative case, 349, 355 non-concatenative, 1, 17 Ob-Ugric, 29, 30, 35, 57, 84 Old Hungarian, 20, 24, 25, 45 onomatopoeia, 15, 77, 93, 284, 314, 324 palatalisation, 316, 330 pejorative, 6, 15, 34, 35, 106, 123, 130, 131, 132, 134, 135, 137, 140, 141, 144, 145, 148, 150, 153, 156, 158, 159, 160, 161, 162, 163, 164, 165, 171, 173, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 188, 189, 191, 192, 194, 196, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 220 performatives, 117, 126 Permic, 30, 33, 47, 50, 54, 70, 75, 77, 84, 87 pitch, 6, 66, 69, 76, 83, 270, 272, 273, 274, 275, 279, 280, 281, 288, 289, 339, 342, 348 prefixoids, 57, 135 principle of expressivity defined, 92 reduplication, 2, 4, 5, 7, 15, 16, 17, 18, 20, 26, 27, 29, 31, 32, 36, 38, 39, 40, 41, 42, 44, 45, 47, 48, 50, 51, 53, 54, 55, 56, 57, 58, 60, 61, 62, 63, 64, 65, 66, 68, 69, 70, 71, 72, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 93, 96, 101, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 129, 131, 160, 204, 223, 226, 231, 232, 233, 234, 235, 236, 237, 238, 240, 242, 243, 244, 246, 247, 248, 251, 252, 254, 255, 259, 260, 261, 263, 264, 267, 268, 270, 271, 286, 288, 290, 297, 298, 302, 306, 308, 314, 318, 330, 338, 339, 340, 344, 347, 352, 354, 355, 357 ablaut, 5, 19, 103, 107, 110, 111, 113, 114, 115, 116 echo. See echo words partial, 18, 31, 39, 54, 63, 233, 234, 286, 318, 339 rhyme, 5, 103, 113, 114 total, 5, 18, 38, 61, 68, 83, 84, 104, 113, 114, 115, 116, 232, 261 registers, 106, 107, 111, 113, 116, 178, 205, 214, 260, 295, 326, 330 repetition, 6, 15, 32, 54, 55, 58, 61, 63, 65, 66, 67, 68, 71, 75, 81, 82, 84, 106, 107, 114,

395 118, 140, 186, 231, 232, 233, 239, 243, 252, 254, 255, 256, 257, 258, 259, 261, 265, 267, 270, 271, 280, 288, 289, 297, 315, 316, 318 rhyming, 2, 19, 78, 95, 96, 98, 99, 107 Russian, 4, 8, 33, 34, 39, 40, 46, 53, 54, 70, 74, 75, 85, 88, 137, 225, 226, 361, 362, 363, 365, 369, 370, 371, 373, 376, 378, 382, 384, 385, 387, 389, 390, 391, 392 Saamic, 54, 63, 64, 87 Sanzhi Dargwa, 340 Scots, 5, 91, 94, 95, 96, 98, 99, 100, 101 Searle’s Principle of Expressibility, 13 semantic opacity, 178 slang, 22, 27, 33, 44, 134, 144, 145, 151, 164, 170, 174, 176, 177, 178, 187, 193, 212, 215, 219 Spanish, 8, 27, 28, 36, 56, 137, 242, 300, 313, 327, 361, 369, 371, 373, 377, 378, 389, 392 Speech rate, 277, 279, 281, 282 Standard Modern Greek, 5, 123, 216, 218, 220, 223 suffixation, 4, 15, 16, 20, 22, 24, 26, 29, 32, 33, 40, 41, 44, 123, 131, 160, 163, 182, 194, 195, 202, 204, 208, 211, 212, 213, 215, 299, 356, 362, 376, 377, 381, 382 syntactic doublings, 30 syntactic heads, 8, 341, 361, 364, 365, 366, 367, 369, 370, 373, 378, 382, 385, 387, 389 syntactic modifiers, 8, 361, 364, 365, 369, 370, 371, 382, 389 syntactic variation, 8, 361, 365, 377, 390 taboo language, 131, 132 toponyms, 171, 378, 379, 382 truncation, 4, 15, 16, 19, 23, 24, 25, 26, 27, 28, 29, 33, 39, 40, 41, 42, 44, 45, 46, 109, 111, 249 truth conditions, 125 Turkic, 4, 30, 31, 44, 45, 47, 69 Turkish, 39, 104, 169, 175, 176, 182, 209, 214, 222 Udi, 342, 344, 358 Udmurt, 4, 30, 31, 32, 34, 37, 43, 44, 46, 54, 70, 75, 77, 84, 85 verbal prefixes, 17, 32, 79 Vlach. See Aromanian vocative, 33, 34, 35, 277 vowel harmony, 100, 235, 315, 330, 332

https://doi.org/10.1017/9781108989084.021 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989084.021 Published online by Cambridge University Press