The Grammar of Copulas Across

Languages Maria J. Arche (Editor)

Visit to download the full and correct content document:
https://ebookmass.com/product/the-grammar-of-copulas-across-languages-maria-j-ar
che-editor/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Social Communication Across the Lifespan Heather J.

Ferguson

https://ebookmass.com/product/social-communication-across-the-
lifespan-heather-j-ferguson/

Communicating Across Cultures and Languages in the

Health Care Setting: Voices of Care 1st Edition Claire
Penn

https://ebookmass.com/product/communicating-across-cultures-and-
languages-in-the-health-care-setting-voices-of-care-1st-edition-
claire-penn/

Conectados (World Languages) 1st Edition Patti J.

Marinelli

https://ebookmass.com/product/conectados-world-languages-1st-
edition-patti-j-marinelli/

Trapped: Brides of the Kindred Book 29 Faith Anderson

https://ebookmass.com/product/trapped-brides-of-the-kindred-
book-29-faith-anderson/
Season of the Wolf Maria Vale

https://ebookmass.com/product/season-of-the-wolf-maria-vale-3/

Season of the Wolf Maria Vale

https://ebookmass.com/product/season-of-the-wolf-maria-vale-2/

The Grammar of Interactives Bernd Heine

https://ebookmass.com/product/the-grammar-of-interactives-bernd-
heine/

The Grammar of Expressivity Daniel Gutzmann

https://ebookmass.com/product/the-grammar-of-expressivity-daniel-
gutzmann/

Season of the Wolf Maria Vale

https://ebookmass.com/product/season-of-the-wolf-maria-vale/
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

The Grammar of Copulas Across Languages

OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

O X F O R D S T U D I E S I N T H E O R E T I C A L L I N GU I S T I C S

GENERAL EDITORS: David Adger and Hagit Borer, Queen Mary University of London
ADVISORY EDITORS: Stephen Anderson, Yale University; Daniel Büring, University of
Vienna; Nomi Erteschik-Shir, Ben-Gurion University; Donka Farkas, University of
California, Santa Cruz; Angelika Kratzer, University of Massachusetts, Amherst;
Andrew Nevins, University College London; Christopher Potts, Stanford University;
Barry Schein, University of Southern California; Peter Svenonius, University of
Tromsø; Moira Yip, University College London
RECENT TITLES
 The Morphosyntax of Imperatives
by Daniela Isac
 Sentence and Discourse
edited by Jacqueline Guéron
 Optimality-Theoretic Syntax, Semantics, and Pragmatics
From Uni- to Bidirectional Optimization
edited by Géraldine Legendre, Michael T. Putnam, Henriëtte de Swart, and Erin Zaroukian
 The Morphosyntax of Transitions
A Case Study in Latin and Other Languages
by Víctor Acedo-Matellán
 Modality Across Syntactic Categories
edited by Ana Arregui, María Luisa Rivero, and Andrés Salanova
 The Verbal Domain
edited by Roberta D’Alessandro, Irene Franco, and Ángel J. Gallego
 Concealed Questions
by Ilaria Frana
 Parts of a Whole
Distributivity as a Bridge between Aspect and Measurement
by Lucas Champollion
 Semantics and Morphosyntactic Variation
Qualities and the Grammar of Property Concepts
by Itamar Francez and Andrew Koontz-Garboden
 The Structure of Words at the Interfaces
edited by Heather Newell, Máire Noonan, Glyne Piggott, and Lisa deMena Travis
 Pragmatic Aspects of Scalar Modifiers
The Semantics-Pragmatics Interface
by Osamu Sawada
 Encoding Events
Functional Structure and Variation
by Xuhui Hu
 Gender and Noun Classification
edited by Éric Mathieu, Myriam Dali, and Gita Zareikar
 The Grammar of Expressivity
by Daniel Gutzmann
 The Grammar of Copulas Across Languages
edited by María J. Arche, Antonio Fábregas, and Rafael Marín

For a complete list of titles published and in preparation for the series, see pp. –.
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

The Grammar of
Copulas Across
Languages

Edited by
MA RÍ A J . A RC HE, A NTONIO F Á BREG A S,
AND RAFAEL MARÍN

1
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

3
Great Clarendon Street, Oxford, OX DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© editorial matter and organization María J. Arche, Antonio Fábregas, and Rafael Marín 
© the chapters their several authors 
The moral rights of the authors have been asserted
First Edition published in 
Impression: 
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
 Madison Avenue, New York, NY , United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 
ISBN –––– (hbk.)
ISBN –––– (pbk.)
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR YY
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Contents
General preface vii
List of abbreviations ix
The contributors xiii

. Main questions in the study of copulas: Categories, structures,

and operations 
María J. Arche, Antonio Fábregas, and Rafael Marín
. Copulas and light verbs as spellouts of argument structure:
Evidence from Dene languages 
Nicholas Welch
. The support copula in the left periphery 
Teresa O’Neill
. The copula as a nominative Case marker 
Kwang-sup Kim
. Number matching in binominal small clauses 
Susana Bejar, Jessica Denniss, Arsalan Kahnemuyipour, and
Tomohiro Yokoyama
. Agreement with the post-verbal DP in Polish dual copula clauses 
Anna Bondaruk
. On PERSON, animacy, and copular agreement in Czech 
Jitka Bartošová and Ivona Kučerová
. Aspects of the syntax of ce in French copular sentences 
Isabelle Roy and Ur Shlonsky
. The role of the copula in periphrastic passives in Russian 
Olga Borik
. The copula in certain Caribbean Spanish focus constructions 
Luis Sáez
. Variation in Bantu copula constructions 
Hannah Gibson, Rozenn Guérois, and Lutz Marten
. Predicational and specificational copular sentences in Logoori 
Nicoletta Loccioni

References 
Index of Terms and Languages 
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

General preface
The theoretical focus of this series is on the interfaces between subcomponents of the
human grammatical system and the closely related area of the interfaces between the
different subdisciplines of linguistics. The notion of ‘interface’ has become central in
grammatical theory (for instance, in Chomsky’s Minimalist Program) and in linguis-
tic practice: work on the interfaces between syntax and semantics, syntax and
morphology, phonology and phonetics, etc. has led to a deeper understanding of
particular linguistic phenomena and of the architecture of the linguistic component
of the mind/brain.
The series covers interfaces between core components of grammar, including
syntax/morphology, syntax/semantics, syntax/phonology, syntax/pragmatics,
morphology/phonology, phonology/phonetics, phonetics/speech processing,
semantics/pragmatics, and intonation/discourse structure, as well as issues in the
way that the systems of grammar involving these interface areas are acquired and
deployed in use (including language acquisition, language dysfunction, and language
processing). It demonstrates, we hope, that proper understandings of particular
linguistic phenomena, languages, language groups, or inter-language variations all
require reference to interfaces.
The series is open to work by linguists of all theoretical persuasions and schools of
thought. A main requirement is that authors should write so as to be understood by
colleagues in related subfields of linguistics and by scholars in cognate disciplines.
Copular constructions have been mysterious since Panini and Aristotle, and
the mysteries have only grown as our knowledge of the range of cross-linguistic
variation in copular clauses has developed. This volume both surveys where our
understanding of these constructions is, and further substantially extends the range
of cross-linguistic data. This brings new challenges, but also new possibilities of
deeper understanding. Overall, the chapters move towards an understanding
of copular elements as being manifestations of a range of functional categories in
the clausal domain, as opposed to being necessarily realizations of predication.
Beyond this, the chapters extend the range of relevant phenomena by looking
at how copulas enter into syntactic dependencies with other clausal elements,
including subjects and focalized constituents, highlighting the way that copular
elements are integrated into a wide range of clausal structures.
David Adger
Hagit Borer
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

List of abbreviations
// etc. st/nd/rd person or noun class number
ACC accusative
ADJ adjective
APPL applicative
ASC amalgam specificational copular
AspP aspectual phrase
AUG augment
AUX auxiliary
COMP complementizer
CONJ conjunction
CONN connective
COP copula
CopP copular phrase
CP complementizer phrase
CSC canonical specificational copular
DEF definite
DEM demonstrative
DET determiner
DM Distributed Morphology
DP determiner phrase
DUR durative
DV default vowel
ECM exceptional case marking
EPP extended projection principle
EVD evidentiality
EZ ezafe
FCCS focus construction in Caribbean Spanish
F/FEM feminine
FOC focus
FV final vowel
GEN genitive
HC host clause
IC interrupting clause
IL individual level
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

x List of abbreviations

IMPF/IPFV imperfect(ive)
INAN inanimate
INC inceptive
IND indicative
INDEF indefinite
INE inessive
INST instrumental
INT intensive
IP inflectional phrase
IV initial vowel
LOC locative
M/MASC masculine
NEG negative
NMR number matching requirement
N/NEU neuter
NOM nominative
NON-VIR non-virile
NP noun phrase
OBJ object
OM object marker
PC pronominal copula
PERS persistive
PFV perfective
PL plural
PLA plural addressee
PLUR pluractional
POSS possessive
PPP past passive participle
PredP predicate phrase
PREP preposition
PRES present
PRO pronoun
PROX proximal marker
PRS present
QN question marker
REF referential
REFL reflexive
REL relative
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

List of abbreviations xi

REP repetitive
SBJ subject
SBJV subjunctive
SC small clause
SG singular
SL stage level
SM subject marker
SOT sequence of tenses
TAM tense/aspect/mood
THM thematic (lexical) prefix
TOP topic
TopP topic phrase
TP tense phrase
TT topic time
VIR virile
vP light verb phrase
VP verb phrase
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

The contributors
MARÍA J. ARCHE is a Senior Lecturer at the University of Greenwich, UK. Her research focuses
on the syntax and semantics of tense and aspect and their acquisition. She is the author of a
monograph on the copulas of Spanish entitled Individuals in Time. Tense, Aspect and the
Individual/Stage Distinction published by John Benjamins in . She has edited special issues
on aspect and argument structure for Natural Language and Linguistic Theory and Lingua.
JITKA BARTOŠOVÁ was awarded her PhD by McMaster University, where she studied on the
Cognitive Science of Language program. Her dissertation explores theoretical phenomena of
the syntax–semantics interface, focusing on the empirical domain of specificational and
identificational copular clauses. Her main areas of interest are syntactic operations (movement,
inversion, agreement) and the way they affect semantic interpretation. She mainly works on
the Czech language.
SUSANA BEJAR is an Assistant Professor of Linguistics at the University of Toronto and co-
investigator of the SSHRC-funded research project ‘Copular Agreement Systems: Locality and
Domains’. Her areas of expertise are syntax and morphology, with a focus on complex
inflectional systems and theories of feature matching and valuation. She has published articles
in journals such as Linguistic Inquiry, Syntax, and Journal of Linguistics, and was co-editor of
the Oxford University Press volume Phi Theory: Phi-features across modules and interfaces.
ANNA BONDARUK is Professor at the John Paul II Catholic University of Lublin, Poland. She is
Head of the Department of Theoretical Linguistics. Her research interests cover the syntax of
Germanic, Celtic, and Slavic languages, and theoretical linguistics. She is an author of three
books, the first of which concerns comparative constructions in English and Polish, the second
focuses on PRO and control in English, Irish, and Polish, and the third centers around copular
clauses in English and Polish. She is currently working as a grant member on a comparative
analysis of psychological predicates in English, Polish, and Spanish.
OLGA BORIK is Associate Professor at the Universidad Nacional de Educación a Distancia
(UNED), Madrid. She obtained her PhD degree from Utrecht University in . During
her academic carrier she has held positions at the New University of Russia (Moscow),
Universidade Nova de Lisboa, and Universitat Autònoma de Barcelona. She has published a
monograph and several articles on aspect and tense, as well as articles on the morphology–
semantics interface and argument structure. Her most recent research interests include syntax
and semantics of participles, passives in Russian, semantics of bare nominals, and kind
reference.
JESSICA DENNISS is a PhD candidate at the University of Toronto and a member of the SSHRC-
funded research project ‘Copular Agreement Systems: Locality and Domains’. Her research
focuses on the syntax and morphology of Australian Aboriginal languages, in particular
Ngarinyman, a language on which she has conducted fieldwork. She is currently part of the
team that is working to produce a Ngarinyman dictionary. She has presented at the conferences of
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

xiv The contributors

the Societas Linguistica Europaea, Canadian Linguistic Association, and Australian Linguistic
Society, and has received a Vanier Canada Graduate scholarship and a John Monash scholarship.
ANTONIO FÁBREGAS got his PhD () from the Universidad Autónoma de Madrid, and has
been a Full Professor of Hispanic Linguistics at the University of Tromsø–The Arctic Univer-
sity of Norway since . His research concentrates on the syntax and semantics of word-
internal structures, with particular attention to grammatical categories, aspect and tense, and
the properties of affixes. He is the author of more than one hundred articles and book chapters
in journals like The Linguistic Review, Linguistic Analysis, and Journal of Linguistics for and
publishing houses such as Oxford University Press, De Gruyter, and John Benjamins. He has
also written several monographs, such as Morphology: From Data to Theories (, EUP),
with Sergio Scalise, and Las nominalizaciones (, Visor). He is currently associate editor of
the Oxford Research Encyclopedia of Morphology, and chief editor of Borealis: An International
Journal of Hispanic Linguistics.
HANNAH GIBSON is Lecturer in Linguistics at the University of Essex. Prior to this she was a
postdoctoral researcher in the Graduate School of Language and Culture at Osaka University
and a British Academy Postdoctoral Research Fellow in the department of linguistics at the
School of Oriental and African Studies (SOAS), University of London. Her doctoral research
examined aspects of the morphosyntax of the Tanzanian Bantu language Rangi. Her research
has included work within the Dynamic Syntax theoretical framework, as well as more broadly
in the fields of syntax, morphosyntax, and language contact, with a particular focus on the
Bantu languages.

ROZENN GUÉROIS is a postdoctoral researcher at Ghent University, working on the research

project ‘A typology of the passive voice in Bantu’ (BOF–Special research fund, April –
March ). Prior to this, she worked as a postdoctoral research assistant in the Leverhulme-
funded project ‘Morphosyntactic variation in Bantu: Typology, contact and change’ (March
–March ). As part of her PhD research, she produced A grammar of Cuwabo (,
Université Lyon ). Her research focuses on the morphosyntactic study of Bantu languages
from a descriptive and typological point of view.
ARSALAN KAHNEMUYIPOUR is an Associate Professor of Linguistics at the University of Toronto
and co-investigator of the SSHRC-funded research project ‘Copular Agreement Systems:
Locality and Domains’. His areas of expertise are syntax, morphology, and the syntax–prosody
interface. He has worked on a number of languages including his native Persian, as well as
English, Armenian, Turkish, and Niuean, among others. He is the author of a monograph with
Oxford University Press and articles in journals such as Natural Language and Linguistic
Theory, Linguistic Inquiry, Syntax, and Journal of Linguistics.

KWANG-SUP KIM received his PhD from the University of Maryland, College Park in  and is
currently Professor of General and English Linguistics at Hankuk University of Foreign
Studies, Korea. His main interests are in syntactic theory, comparative syntax, the syntax–
semantics interface, and the syntax–phonology interface. He has written a book on last resort
strategies in minimalism, Minimalism and Last Resort (), and has published numerous
articles in journals such as Linguistic Inquiry, Lingua, and Studies in Generative Grammar.
He was president of the Korean Generative Grammar Circle in –.
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

The contributors xv

IVONA KUČEROVÁ is Professor of Linguistics at McMaster University. She specializes in theor-

etical syntax and semantics, and their interface. Her work explores the nature of phi-features at
the syntax–semantics interface, information structure and its morphosyntactic correlates,
definiteness systems and their relation to aspect, the morphosyntax and morpho-semantics
of case, agreement, and case splits, the syntax of null languages, and the syntax of copular
clauses. She works mainly on Slavic, Germanic, and Romance languages.

NICOLETTA LOCCIONI is a graduate student in the linguistics department of the University of

California, Los Angeles. Her theoretical interests include theoretical syntax and the syntax–
semantics interface. In the past few years, she has worked on the interaction between so-called
“individual-level predicates” and past tense in Italian, pseudo-relative constructions in
Romance, and copular constructions cross-linguistically. For her dissertation she is currently
working on the syntax and semantics of superlatives in Romance.
RAFAEL MARÍN is a researcher in linguistics at the laboratory STL (UMR ), CNRS/
Université de Lille . His work focuses on lexical aspect and related phenomena. He has
mainly worked on nonverbal predication (adjectives and participles, copular constructions),
psychological predicates, and the morphology–semantics interface. Since , he has been the
Director of the Catalan Foundation for Research and Innovation.
LUTZ MARTEN is Professor of General and African Linguistics at SOAS, University of London.
His research focuses on the description and analysis of structural, social, and functional aspects
of language, with a specific focus on African languages, and he is currently directing a
Leverhulme-funded research project on morphosyntactic variation in Bantu. His publications
include At the Syntax-Pragmatics Interface (OUP, ), A Grammatical Sketch of Herero
(with Wilhelm Möhlig and Jekura Kavari, Köppe, ), The Dynamics of Language (with
Ronnie Cann and Ruth Kempson, Elsevier, ), and Colloquial Swahili (with Donovan
McGrath, Routledge, /).
TERESA O’NEILL (PhD, The Graduate Center, CUNY) has served as Adjunct Assistant Professor
of Linguistics at the City University of New York and Columbia University. She is the former
Assistant Director of the Center for Integrated Language Communities, a National Language
Resource Center. Her work has focused on the syntax of tense, agreement, and case, particu-
larly in understudied languages and varieties of English. She is also active in endangered
language documentation.
ISABELLE ROY is an Associate Professor of Linguistics at the University of Paris VIII, and has
previously held a position at CASTL–University of Tromsø. Her main research interests are in
linguistic theory, with a focus on the syntax–semantics interface, in the areas of predication,
copular constructions, adjectives, categories and categorization, and linguistic ontology. She is
the author of Nonverbal Predication: Copular Sentences at the Syntax–Semantics Interface
(OUP, ).
LUIS SÁEZ is Professor at the Departamento de Lengua Española at the Universidad Complu-
tense de Madrid. He received his PhD in  from the Universidad Autónoma de
Madrid with a dissertation on comparative constructions supervised by Carlos Piera, and he
is co-editor (with Cristina Sánchez López) of the book Las construcciones comparativas ().
His work also includes ellipsis-related articles like ‘Sluicing with Copula’ and ‘Peninsular
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

xvi The contributors

Spanish pre-nominal possessives in ellipsis contexts: A Phase-based account’ (LSRL  and 

respectively,) and articles focusing on Spanish clitics (‘Applicative phrases hosting accusative
clitics’, ‘Restrictions on enclitics and the imperative in Iberian Spanish’).
UR SHLONSKY is Professor of Linguistics at the University of Geneva. His main interests are in
linguistic theory and comparative syntax, with special focus on Semitic and Romance.
NICHOLAS WELCH is an Assistant Professor of Linguistics at Memorial University of Newfound-
land. His research investigates the relationship between structure and interpretation, particu-
larly with respect to copular clauses, temporal grammar, and the fine structure of the clausal
periphery. Other key interests include language documentation and revitalization, particularly
in the context of the indigenous languages of Canada, in which he currently holds a Tier 
Canada Research Chair.

TOMOHIRO YOKOYAMA is a PhD candidate in linguistics at the University of Toronto and a

member of the SSHRC-funded research project ‘Copular Agreement Systems: Locality and
Domains’. For his dissertation, he works on Person Case Constraint effects in various lan-
guages. His approach to combinatorial restrictions on weak elements involves feature valuation
in conjunction with articulated person features, which diverges from the traditional Agree
analyses. He has presented at the conferences of the North East Linguistic Society, the
Linguistic Society of America, and the Canadian Linguistic Association in the areas of
morphosyntax, semantics, and pragmatics.
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study

of copulas
Categories, structures, and operations

M A RÍ A J. A R C H E , A N TO N I O F Á B R E G A S ,
AND R A F AE L M ARÍN

. Introduction: why copulas?

This volume is dedicated to copulas, and more specifically, to how their syntactic and
semantic properties can inform fundamental issues in linguistics. As we will show in
this chapter, copulas and copular clauses are one of the areas of grammar with the
greatest degree of variation attested. They vary both in forms, as they surface under
different categories (verbs, prepositions, pronouns) and head different functional
elements (T, Pred, C); and in behavior, since they participate in a diversity of
agreement patterns (e.g., dual patterns such as in Polish) and non-canonical con-
structions (e.g., amalgams). For this reason, copulas and copular clauses are a
privileged ground to explore essential theoretical issues concerning categorization,
formal mechanisms of the grammar of agreement and late insertion, as well as clause
structure. In sum, they are an unparalleled window into the study of the innermost
mechanisms and properties of human language.
The chapters presented here are reviewed versions of a selection of talks offered at
a workshop held at the University of Greenwich in June . The chapters all
demonstrate that the analysis of copulas is far from clear within individual languages
and even less so when a given analysis is applied to more than one language. Since the
cross-linguistic diversity in copulas is vast, the theoretical accounts need to embrace
acute subtlety to capture all the nuances. Our main goal in this first chapter is to
contextualize the contributions gathered in this volume by identifying both the
main empirical facts that a theory on copulas should account for and the theoretical
issues that such analyses have immediate consequences for. We will make reference
to the empirical issues, the accounts existing to date, and the views that the authors

The Grammar of Copulas Across Languages. First edition. María J. Arche, Antonio Fábregas,
and Rafael Marín (eds).
This chapter © María J. Arche, Antonio Fábregas, and Rafael Marín . First published in  by
Oxford University Press.
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

in the volume propose. The general issues at the core of the analyses of copulas are
the following:
a) The nature of grammatical categories; specifically, what kinds of heads are
made compulsory by universal grammar and what the need for support
elements is. As we will show, copulas have been understood as semantically
empty inflectional supports, light verbs or raising verbs. In any of these
approaches, the existence of copulas highlights the question of what connection
there should be between meaning and lexical categorization. Copulas are used
in contexts where their function appears to go beyond simple support for a
subject-predicate structure. Their use in defining information structure and
passive voice across a variety of languages is relevant in order to understand
their nature. How can copulas be defined so all these other uses are accounted
for, while not predicting that they should be used anywhere where verbal
inflection could in principle be useful? Are there truly semantically empty
verbs? What is the nature of support elements in general?
b) The working of agreement. Copular sentences can involve two nominative NPs
sharing one single verb, which is a unique situation leading to unexpected
agreement patterns. This makes copular constructions an unparalleled ground
to study the functioning of agreement, agreement probes in contexts where
there is more than one candidate goal, and to explore whether there is a
matching requirement between two NPs that are related through predication.
c) The contribution of light elements to defining the type of clause. Some languages
seem to have only one copula, while others have more than one element that can
be used in nonverbal predicate contexts. Why are there languages that have more
than one copula? How does this interact with the different types of copular
sentences described in the literature? How many different types of copular sen-
tences are there, and how are the empirical distinctions codified?
This chapter is organized as follows. In §., we present a detailed survey of the main
facts that a global theory of copulas should account for. This section discusses four
aspects of the grammar of copulas: the behavior of (prototypical) copulas and the
difficulties in delimiting the concept itself (§..), the taxonomy of copular sen-
tences (§..), the existence of two or more copulas in a given language (§..), and
other roles that copulas are associated with across languages. Later, in §., we focus
on two fundamental theoretical problems at the core of these facts: the morphosyn-
tactic role of copulas (§..) and how the classification of copular sentences is to be
analyzed (§..). Finally, in §. we discuss the current points of agreement and
disagreement in the study of copulas, as represented in the chapters of this volume.

. Main facts about copulas

The purpose of this section is to describe the empirical facts that theories of copulas
should account for. Given the significant disagreements found in the literature about
the proper characterization of the empirical aspects of copulas, we will also refer to
the different perspectives on the issues discussed.
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

.. The behavior of copulas

As is the case with many concepts borrowed from traditional linguistics,¹ copulas
turn out to be an extremely challenging notion to define, and most works that involve
an analysis of be and its cross-linguistic equivalents simply take the notion for
granted. () gives a prototypical example of a bona fide copula:

() John is sick.

This English example displays the properties that are prototypically associated with
copulas: (i) copulas carry verbal inflection, (ii) copulas appear in contexts where the
predicate is nonverbal, (iii) copulas are elements used to link the predicate and the
subject—as the term itself suggests—from Latin copula ‘link’, and (iv) copulas are
semantically light, possibly empty.
Thus, by virtue of (i), copulas should be inflected in whatever morphological prop-
erties verbs display in a language (e.g., tense, aspect, number and person, gender . . . ).
By virtue of (ii), () would correspond to predicating the adjective sick of the referring
expression John: λx[sick’(x)](j). In relation to (iii), copulas cannot define a predicate on
their own (*John is). Finally, in relation to (iv), copulas are necessary to allow the
adjective to define the predicate (*John sick). However, as we will show in this chapter,
all these prototypical properties are debatable, and are in fact the subject of enormous
cross-linguistic variation and disagreement in how they are analyzed.
In typological studies (such as Stassen  and Pustet ), the question of what
is the set of properties that characterize copulas becomes central. Different proposals
have been made and most of them agree that the definitions traditionally given on the
basis of Romance languages and English are both too restrictive and too broad.
Consider the definition below, from Pustet (: ):
() A copula is a linguistic element which co-occurs with certain lexemes in certain
languages when they function as predicate nucleus. A copula does not add any
semantic content to the predicate phrase it is contained in.
First of all, observe that the definition does not specify that the copula is a verb, or
that it combines with nonverbal predicates. With respect to the first property, in fact,
it has been noted that in many languages copular elements are historically related to
pronouns (Hengeveld : ; Stassen : ; Heine and Kuteva : )² or

¹ Latin and Medieval grammars since Priscian’s Institutiones used the term verbum substantivum
‘substantive verb’ to refer to Latin esse ‘be’. The term ‘copula’ was coined later, by Abelard, and was used
in the Grammaire de Port-Royal. It became widespread after Meillet (–), who emphasized that a
copular verb did not have most of the properties of verbs in a given language.
² We leave aside the nature of so-called pronominal copulas, illustrated in (i) for Maltese (Central
Semitic Creole), rd person pronouns that in some languages are compulsory to build some types of
copular sentences. See Doron (), Borg (), Pereltsvaig (), Dalmi (). Bondaruk, this
volume, briefly touches on the issue.
(i) Malta hi gzira.
Malta PC island
‘Malta is the island.’
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

deictic elements in general. In !Xuun (Lionnet ), a K’xa language spoken in

Namibia and Angola, the deictic proximal marker e (a) has developed a use as a
non-locative copula (b):
() a. me n|ee ti e.
SG head IMPF PROX
‘This is my head.’
b. m ba !uu ti e Jor-El.
SG father name impf COP Jor-El
‘My father’s name is Jor-El.’
Stassen (: –) also notes that pronouns and discourse markers are frequently
reanalyzed as “abstract linking morphemes in predicate nominal sentences” in African
languages like Shona, Temne or Zulu, among others. Similarly, the Lakota copular
verb hécha derives from the pronouns hé ‘this’ and cha ‘such’ (Pustet : ).
Note, next, that in Pustet’s definition there is no claim that copulas do not
combine with verbs. This contrasts with other available definitions, where the
combination with nonverbal predicates is taken to be central, as in (), from
Hengeveld (: ):

() A copula enables a nonverbal predicate to act as a main predicate in those

languages and under those circumstances in which this nonverbal predicate
could not fulfil this function on its own.

Based on her sample of  languages, Pustet defends the implicational hierarchy in
(): if a language uses copulas for verbal predicates (e.g., participles), it will also use
copulas for adjectives and nouns, but not viceversa.

() NOUNS > ADJECTIVES > VERBS

The scale is underpinned by the notions of valence, transience, and dynamicity found
in Givón () and Croft (): “within minimal pairs [in a given language], the
lexical item that is compatible with the copula is always less transitive, less [tempor-
ally] transient and less dynamic than its counterpart that does not admit copula use”
(Pustet : ). However, copulas do co-occur with verbs. The example in ()
shows a case from Bambara (Pustet : ) where the copula combines with a
verbal predicate:
() ne bε taa
SG COP leave
‘I am leaving.’
In light of these cases, it might be questioned whether English or Spanish passive
constructions are instances of the same pattern (copula + verb), rather than one
where the copula is treated as an auxiliary verb.
() Rorschach fue atacado por un perro.
Rorschach was attacked by a dog
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

Pustet’s definition in () makes the claim that copular verbs do not contribute any
semantic information to the predicate, in contrast to auxiliaries, which could contribute
modal or aspectual information. However, this claim is also problematic. Cross-linguistic
surveys have proposed a class of semi-copulas (also called pseudo-copulas), namely those
verbal forms which, like copulas, cannot form a predicate independently, but add an
identifiable meaning to it. An often-cited example of semi-copula is the English verb
become, which contributes a change of state meaning to the predicate.
() Tony Stark became *(a millionaire).
Even among prototypical copulas, it is not always clear that there is no meaning
contribution. In Spanish, as it is well known, two verbs have been considered copular:
ser and estar, the first associated to individual level (IL) properties and the second
associated to stage level (SL) properties (see Milsark , Carlson  for the
distinction). IL-adjectives must, then, combine with ser, while SL-adjectives combine
with estar ().
() a. Anacleto es español.
Anacleto isser Spanish
b. Anacleto está desnudo.
Anacleto isestar naked

The adjectives that allow for both copulas show a systematic meaning difference:
with estar, they pattern with SL predicates referring to stages of the individual, and
with ser, they pattern with IL predicates predicating the property of the individual as
such (see, among many others, Leborans , ; Arche ; Camacho ;
Gallego and Uriagereka ).
() a. Roberto Alcázar es guapo.
Roberto Alcázar isser handsome
b. Roberto Alcázar está guapo.
Roberto Alcázar isestar handsome
Unless we are willing to duplicate the entries for the adjective guapo ‘handsome’ and
all the others that combine with both copulas, cases such as () strongly suggest that
the verb estar (or the structure associated to it) introduces aspectual information that
defines the predicate as SL (for instance, as Arche , Brucart , and Camacho
 argue).
Spanish estar also constitutes a potential counterexample to another prototypical
copula property: the inability to define a predicate independently. The example in
() shows that in a locative meaning, estar can be used without any other (overt)
constituent. Unless we do not consider this verb a copula in locative uses, this
property is at odds with the traditional definition.
() Estoy.
I.amestar
‘I am here.’
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

The use and function of estar is in too many ways parallel to that of Spanish ser—
including the ability to be used in passive sentences—which suggests that estar
should belong to the same class as ser.
The facts just surveyed suggest that a much less restrictive definition of copula
would be more appropriate. The proposal in () illustrates the spirit of what is
needed:
() A copular element is an element needed to define a predication structure.
Such an element is typically a verb, but not always; it typically combines with
nonverbal categories, and it typically carries minimal meaning, which is connected
with its inability to define a predicate alone. However, none of these properties are
necessary to define a copula, as we have seen.
The definition in () is admittedly descriptive: it defines an object through its
surface role and says nothing about its theoretical status, why it is needed or even
about the grammatical category that it instantiates. We believe that this is a positive
result that is sustained by the chapters in this volume. They lead to the conclusion
that copula is not a distinct grammatical category, but rather the label that has been
given to a number of distinct objects in different languages. In the remainder of this
chapter, we examine the main current theoretical proposals about the nature of
copulas and the structures they participate in. We will show that none of them are
free of problems, but, more crucially, that all of them have clear facts supporting their
claims empirically. One important point to bear in mind when approaching this
tension is, precisely, that what we call ‘copula’ in one language is quite likely different
from what we call ‘copula’ in another; copulas seem to be involved in different
syntactic constructions cross-linguistically.

.. Types of copular sentences

It is far from clear whether there is only one kind of copular construction or whether
copulas can participate in different kinds of structures where a subject is related to a
nonverbal predicate. Different answers have been provided, partially depending on
whether the distinction between different kinds of copular sentences is argued to be
purely semantic or to have an impact on syntax.
The classical division of copular sentences comes from Higgins (), who proposes
a four-way split, depending on whether—in different combinations—the nonverbal
categories combined by the copula are referential or not: predicational (a), specifica-
tional (b), equative (‘identity statement’, c), and identificational (d).
() a. The winner is a man with a red beard.
b. The winner is Charlie Brown.
c. Britt Reid is the Green Hornet.
d. That woman is Susan.
In (a), the subject NP is referential and the post-copular NP is predicative,
ascribing some properties to the subject. In contrast, in (b) the subject is not
referential in the sense that it is not used to identify a referent in the context, and
the post-copular NP identifies such referent. In (c), an identity statement, both
NPs are equally referential. Finally, according to Higgins, in (d) the subject is
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

referential, but does not provide the identity of the referent; the post-copular NP
provides the identity.
A lot of descriptive and theoretical work has been conducted on this issue; see,
among many others, Halliday (), Akmajian (), Keizer (), den Dikken
(b), Lahousse (), Heycock (), for different interpretations and discussion.
Higgins’s () taxonomy has been questioned from two sides. On the one hand,
some authors have argued that the division is insufficient. It either needs other
(semantic) classes of copular sentences to be added (e.g., a definitional class A zombie
is a fictional undead being, Declerck ) or it is irrelevant in accounting for
syntactic phenomena (see Bejar et al. this volume for a critique along these lines).
On the other hand, other authors have argued that Higgins’s () classification is
over-specific and should be simplified. Mikkelsen () proposed that identifica-
tional sentences like (d) should be reduced to either identity statements or
specificational clauses. Mikkelsen argues that the typology of copular sentences
reduces to whether the two NPs denote an individual (type <e>) or a predicate
(type <e,t>).
() a. <e> is <e,t> (predicational)
b. <e,t> is <e> (specificational)
c. <e> is <e> (equational)
Identificationals like (d) are instances of (c), that is, equational/identity state-
ments; in contrast, those whose subject is simply a demonstrative () reduce to the
specificational type.
() That is Susan.

In cases such as ()—Mikkelsen claims—the first nominal has a predicational

semantics. She provides the following reasoning: that, as a demonstrative, cannot
refer to humans, so it does not make sense to claim that in () that refers to an
entity, since that entity would presumably be Susan, a human individual. Thus, ()
is an instance of a specificational sentence.
Other approaches have reduced Higgins’s typology even more, positing only two
classes: predicational and specificational (or inverse), depending on whether the
more referential NP is the first or the second in the clause. Predicational and
specificational clauses, as we will see in the following pages, are taken by many
scholars to be the two basic categories of copular sentences to which all the other
noted subtypes should be reduced to. Loccioni, in this volume, presents a study of the
two copulas in Logoori, a Bantu language, and argues that their distribution captures
the basic distinction between predicational and specificational, giving further support
to the claim that these are the two types that must be distinguished in the grammar of
natural languages. An exhaustive divide between predicational and specificational
clauses immediately accounts for sentences like John is my friend and My friend is
John, but the equative/identity statement type still needs to be accounted for. Moro
() in fact argued that copular sentences are never truly equative. His reasoning is
the following, starting from a bona fide equative sentence like ():
() The morning star is the evening star.
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

If we try to make a possessive pronoun in the second NP refer to the first NP, we
obtain ungrammaticality:

() *[The morning star]i is [itsi source of light]

This is surprising if neither of the two NPs is a predicate and both are referential
arguments. In a true identity statement where both NPs are referential arguments—
Moro argues—this coreference is possible:
() [The morning star]i is equal to [itsi source of light]

So, why is () uninterpretable? Binding theory shows that a pronoun contained in a
predicative nominal cannot be bound by the clausal subject: *Johni is hisi cook. This
is the same ungrammaticality that we find in (); hence, () is, in actuality, a
predicational sentence, and by parity also (). This position is contended by
Heycock and Kroch (), who, on the basis of semantic facts, argue that equative
sentences do in fact exist, because neither of the two NPs can be taken as really
predicative (e.g., My opinion about Alan Moore is your opinion about Frank Miller).
In contrast, Adger and Ramchand () argue that there is always an asymmetry in
referentiality between the two NPs involved in the construction. See Williams ()
and Pereltsvaig () for similar observations about the asymmetry. Their position
is opposed to Carnie (: –), who argues that the asymmetries identified
in equative sentences follow from a distinction in theta-marking, something that
necessarily implies that the equative copula cannot be interpreted as the logical
identity operation.
Other approaches also arguing for just two types of clauses are present in Blom
and Daalder (), Heggie (), Verheugd (), Moro (, ), and den
Dikken (, a). Interestingly, these theories tend to concentrate on the
syntactic properties of the structure, rather than on the semantics of the NPs
involved. As we will see in §.., in fact, it has been argued that there is only one
type of copular sentence, namely, the predicational one, whereas the specificational
one is syntactically derived from it. Before we examine this issue, let us consider the
question of whether the typology of copular sentences leads us to conclude that there
is more than one verb be.

.. Languages with multiple copulas

If we take the definition proposed in () of a copula being an element that relates a
subject to a nonverbal predicate (with the complications noted before), it can be
concluded that some languages have more than one copular element. Gibson et al.
(this volume) offer quite a comprehensive overview of the different morphological
makeup of copulas in a sample of Bantu languages. They note that even typologically
very close languages differ in the number of copulas they have and the regulations
underpinning their distribution. The variation of the copulas in correlation with the
constructions they appear in is crucial in advancing our knowledge about the nature
and behavior of copulas. Typological studies can inform syntactic and semantic
theoretical proposals in, at least, the following respects:
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

a) The taxonomy of copular constructions. Some copulas have been argued to

appear only in particular types of copular sentences. This can be taken as
evidence that the simple distinction between predicational and inverse copular
sentences needs to be enriched on principled grounds.
b) The nature of the copulas. The factors governing the distribution of multiple
copulas can be informative with respect to the kind of head(s) that the copulas
spell out in each language and the distinct kinds of elements that can be
subsumed under the traditional label of ‘copula’.
Let us briefly address the first aspect: what multiple copulas can tell us about the
typology of copular sentences. If there are distinct, lexically differentiated copulas,
whose distributions pattern with distinct types of copular sentences, the idea that at
least those types of copular sentences must be grammatically distinguished becomes
plausible. In this sense, consider the discussion below.
Some authors have proposed additional types of copular sentences based on the
existence of additional copulas. For example, Bolinger () proposed a locative
type in addition to the predicational and equative copular sentences. This has been
sustained by evidence from, for example, Kinyarwanda (Jerro ), where of the two
copular elements, ni and –ri, one (ni) has a variety of uses and –ri is restricted to
locational predications:
() a. Karemera a-ri m’u Rwanda.
Karemera SG-COP in Rwanda
‘Karemera is in Rwanda.’
b. *Mukamana a-ri umwarimu.
Mukamana SG-COP teacher
Intended: ‘Mukamana is a teacher.’
c. *Mukamana a-ri munini.
Mukamana SG-COP big
Intended: ‘Mukamana is big.’
Also, according to Wauters (), in Sereer (Niger-Congo) there are four copular
elements. The copula -oo is used in equative sentences (a); jeg is used in existential
constructions (b); ref marks NP copulas and is specialized in individual-level
predicates (c); xe can combine with stage-level predicates, among them locatives
(d)—which are also compatible with ref.
() a. Mark, Musaa Juf=oo.
Mark Musaa Juf-COP
‘Mark is Musaa Juf.’
b. a=jeg-a wiin faafaf.
=COP-DV people doctor
‘There are male doctors.’
c. osiriñ um a= ref-a osiriñ maak.
imam SG.POSS =COP-DV imam big
‘His imam was an important imam.’
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

d. obox ole a=xe tafil andok um.

dog DEF =COP outside room SG.POSS
‘The dog is outside his room.’
Even though the existence of this variety of copulas and their ties to distinct copular
sentences favors the idea that grammar differentiates more than a two-way distinc-
tion in the sentence taxonomy, the data do not necessarily enforce that conclusion.
Alternative explanations can be offered. For instance, the copula for locative predi-
cates might involve some kind of P incorporation to a copular element, yielding
a different spellout, as Benveniste () argued for French avoir and Gallego
and Uriagereka () have argued for Spanish estar. Copulas might in some cases
be misanalyzed functional projections dominating adjectives, nouns, or prepositions,
which necessarily project when they are used as predicates. In other words, () and
() are compatible with a proposal whereby locative copular sentences have a structure
distinct from ascriptional NP/AP sentences, but do not force that conclusion.
In sum, the evidence coming from the association between different copulas and
different types of clauses deserves further investigation, so that the empirical evi-
dence supporting different theoretical views can be clearly discerned.
Consider now the second issue: the distribution of multiple copulas. Here, we
largely follow the typological description in Pustet (). We understand that the
different kinds of factors governing the distribution of the copulas adds plausibility
to the claim that the copula in different languages might spell out different heads.
One of the factors involved in the multiplicity of copulas has just been discussed:
the type of copular sentence at hand. In line with the above, consider another
language, Lakota, which distinguishes identificational sentences (This is Mary)
from predicational ones (He is a soldier) by two distinct lexical copulas. We under-
stand that this might be due to two factors: either there are distinct types of small
clause codifying this distinction (which would be amenable with a syntactic proposal
whereby the copula is low in the structure) or the distinct properties of each type with
regard to information structure triggers a different copula choice. Since the identi-
ficational type generally involves focalization of the second NP, the copula could be
conceived as an element sensitive to the global information structure of the clause,
occupying a high position in the syntactic structure.
A second common factor in the choice of copulas is the grammatical category of the
post-copular constituent. As Pustet () notes, Bambara (Niger-Congo) distin-
guishes copular sentences with nominal predicates from those containing adjectival
predicates:
() a. nìn ye námása ye.
this COP banana COP
‘This is a banana.’
b. So ka sùrun.
house COP big
‘The house is big.’
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

In such cases, it would seem that the copular elements must be introduced low in the
structure, in a position where they can be sensitive to the lexical category of the
predicate of the small clause. In some cases, the sensitivity is to the concept expressed
by the predicate: Pustet reports that in Imonda (spoken in Papua New Guinea) the
choice of the copula is sensitive to whether the referent of the NP is intrinsically tall
or erect, which is reminiscent of a division of nouns based on shape that is typical of
some classifier systems (such as the one in Diné Bizaad/Navajo); moreover, in
Imonda one of the copulas is preferred when combined with predicates that refer
to females. Both properties suggest that in this language copulas could be related to
nominal classifiers.
Spanish is, without doubt, the best researched case of multiple copulas
(Luján , Porroche Ballesteros , Arche , among many others). Part of
the distribution of the two copulas is sensitive to the grammatical category of the
predicate: if it is a nominal element, ser is compulsory; if it is a gerund, estar is
compulsory. However, with adjectives, prepositional phrases, and (to some extent)
locations, both copulas are attested. In the case of adjectives, as in (), the standard
account is that ser combines with individual level predicates (Carlson ), which
generally (but not necessarily) are associated with an implication of temporal per-
sistence (a); estar associates with stage-level predicates (b). However, estar also
involves other effects, such as an evidential use according to Roby (), illustrated
in (c), whereby the property is stated to be characteristic, but relative to the direct
perception of the speaker. For reasons of space, we will not review here the different
accounts of how the sensitivity to this distinction is implemented; see Arche ()
and Roy () for overviews.

() a. Juan es gordo.

Juan isser fat ‘Juan is (characteristically) fat, is a fat person’
b. Juan está gordo.
Juan isestar fat ‘Juan is (at this point) fat’
c. Esta paella está deliciosa.
this paella isestar delicious

Other languages reported by Pustet () to have similar distinctions are Barasaano,
Ndyuka, Limbu, and Maltese; a comparable sensitivity to the aspectual information
of the AP is known to apply also to languages like Portuguese and Catalan, although
the distribution is not identical to Spanish.
Finally, copulas can also be sensitive to the nature of the subject: Dumi (Sino-
Tibetan language from Nepal; van Driem ) differentiates between sentences with
animate and inanimate subjects with different copulas. This kind of sensitivity is
compatible with an analysis where the copula is introduced high, at the level of TP;
however, it does not force this conclusion, as the sensitivity could also be obtained if
the copula is introduced immediately above the predicate, on the assumption that the
subject in copular sentences is also base-generated below TP.
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

.. Copulas beyond copular sentences

Finally, copulas are known to be used in other constructions that do not obviously
involve a subject and a nonverbal predicate. The existence of such cases can be dealt
with in two ways: either the other structures where copulas appear are analyzed
as (less obvious) instances of a subject/nonverbal predicate structure, or the use
of copulas in such cases is taken at face value, leading to a redefinition of what a
copula is. Both venues have been explored. Here we will concentrate on just two
cases: the use of copulas to mark information structure, specifically associated to
focalization structures, and the use of copulas to express passive voice.
... Copulas and information structure: clefting Copulas are generally involved
in the articulation of a particular kind of information structure of the clause, namely,
the one that assigns exhaustive focus to one constituent in the clause: the so-called
clefted and pseudo-clefted sentences. The examples in () illustrate the type:
() a. It was John who brought beer.
b. John was who brought beer.
c. Who brought beer was John.
According to some authors, the existence of such a structure suggests that
copulas can be analyzed as discourse markers. The structure has some standard
copular properties: there is a nominal expression (the free relative) which acts
as a predicational constituent describing the properties of a referential nominal
expression; the inverse order is allowed, as in copular clauses, and the element
relating the two is at least completely syncretic with a copular verb. However,
some properties of (pseudo-)clefted sentences are not found in run-of-the-mill
copular sentences. First, while inverted copular sentences tend to assign focus
to the post-copular expression (Blom and Daalder , Roy and Shlonsky this
volume), pseudo-clefted sentences assign focus to the expression that assigns
value to the free relative, independently of its position: both (b) and (c)
above treat John as focus. The contrast below shows that while this is not the
case for predicational copular sentences such as that in (a), it is the case in
the clefted (b).
() a. Jean est mon ami.
Jean is my friend
b. Mon ami, c’est Jean.
my friend it-is Jean
Second, in contrast with inverted copular sentences, focus must be exhaustive. While in
() it is understood that the only person who brought beer was John, in an inverted
clause such as (), it does not follow that Jean is the only friend of the speaker.
() Jean, c’est mon ami.
Jean it-is my friend
Some authors (Lambrecht ; Lehmann ) argue that the exhaustive focus
exhibited by (pseudo-)clefted sentences is an indecomposable part of their meaning,
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

without attributing any of it to the nature of the copula. In contrast, other authors
have tried to divide such structures into independent components. In this regard,
three main proposals can be found. In the first proposal, the pseudo-clefted structure
is an instance of a copular sentence where one of the members has undergone partial
ellipsis (Peters and Bach ). The second proposal involves movement (Chomsky
, Moreau ), and the third analysis proposes base-generation of both con-
stituents (Higgins , Akmajian ). We refer the reader to den Dikken (b)
for a detailed discussion of these competing analyses.³ What is crucial for our
purposes is that the second and the third perspectives are essentially forced to say
that, in a pseudo-cleft, the copula is not just supporting a subject-predicate structure,
but also assigning exhaustive focus to one of the members. In the first analysis it is at
least in principle conceivable that becoming the exhaustive focus is a condition for a
referential DP to escape ellipsis (Merchant ), but nothing in principle forces this
in the other two analyses.
In fact, there are attested structures where the copula is the only surface marker
of exhaustive focus. This has been suggested by Zellou () for Tigrinya, for
example, where copulas are shown to be information structure markers. In this
volume, O’Neill and Sáez study the issue in English and Spanish respectively.
As Saéz (this volume) discusses, in Caribbean Spanish we find structures such as
() below.
() Juan compró fue un libro.
Juan bought was a book
‘A book is what Juan bought.’
Whether these structures are subjacently pseudo-clefts where the wh-element is not
expressed (Toribio , Sedano ) (as represented in (a)) or monoclausal
structures where the copula is a focus marker (Bosque b) (b) has been a long-
debated question within Spanish linguistics.
() a. [what Juan bought] was a book.
b. Juan bought [FocP was [a book]]
The analysis in (a) has several problems noted in Bosque (b) and summarized
by Sáez in this volume. The one in (b) has its own set of problems, such as the fact
that the complement of FocP cannot be extracted (*What bought Juan was?). Sáez
argues for a proposal where () is an instance of an amalgam structure, specifically a
subclass of Horn amalgams, whereby the copula heads a topic phrase (TopP).

³ Note that in the ellipsis account the structure would have to be taken to be distinct from a run-of-the-
mill copular, since, without ellipsis, cases such as (i) would never be a well-formed copular sentence.
(i) [Who brought the beer] was [John brought the beer]
Cases where the ellipsis requires ungrammatical source sentences have also been noted in the literature
(Blom and Daalder , Akmajian ). The movement approach has been considered problematic for
similar reasons (but has been resurrected as movement at LF by Bošković ).
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

O’Neill, this volume, studies similar structures, such as those in () for English.
Since the analysis of pseudo-clefts without a wh-word is obviously not available, she
also argues for an analysis involving an amalgam (following den Dikken et al. ),
where the copula is introduced as a support element in Fin(iteness)P, in the boundary
between the TP and the CP.
() He needs a break is what he needs.
The use of copulas as heads of information structure raises a number of questions for
the nature of copulas in grammars that allow structures comparable to () and ().
To begin with, it would provide arguments in favor of a view of copulas as pure
support elements that are either merged high or are insensitive to the position where
they are introduced, provided that it is a position where phi features and verbal
morphology might be required. Second, it brings up the issue of whether copulas are
designated to support predication or, more generally, are used to divide the clausal
structure in two branches, perhaps along the lines of a figure-ground configuration
that can be used both for predication (subject-predicate) and information structure
(focus-presupposition).

... Passives The use of copulas as elements involved in the expression of

passive voice is also widespread across languages. Abraham () shows that next
to reflexive morphemes, structures involving a copular verb and a non-finite form
of the verb (typically a participle) are used very frequently cross-linguistically to
express passive voice: German sein ‘be’, used for stative passives, Dutch zijn or
Spanish ser ‘be’ are just a few examples of this situation. Let us illustrate it here
with Spanish, as this will allow us to hint at some of the theoretical questions
related to the fact:

() Juan fue atacado por un perro.

Juan wasser attacked by a dog
‘Juan was attacked by a dog.’

In a sense, it is unsurprising that copulas are used in passives. As in a copular

sentence, it can be argued that the existence of a participle denoting a predicate
requires a support element to carry subject agreement and temporal-aspectual
marking. Participles are, after all, non-finite forms of the verb, and, in Spanish,
inflect as adjectives, that is, for gender and number but not for person. However,
there is also a surprising side to this situation, highlighted in Crespí (), among
other authors. Consider ().

() a. El libro fue escrito.

The book wasser written.
‘The book was written.’ (event)
b. El libro está escrito.
The book wasestar written
‘The book is (already) written.’ (result state)
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

The two copulas in Spanish can be used in the passive construction.⁴ In (a), the
interpretation is eventive (there was an event of ‘writing’ that affects the book), while
(b) is stative (after the writing event, the book is now in a ‘written’ state). The
copula in (a) is taken to be an auxiliary verb in traditional grammar, but why
should it be identical to the copula ser in Spanish?
The problem is that if (a) is used to express a passive event whereby there is a
change in the subject, we should have expected to find another kind of verb, with an
eventive semantics. German does precisely this, using werden ‘become’ to express the
eventive passive and leaving sein ‘be’ for the stative one.
() a. Goethes Name wird mit ‘oe’ geschrieben.
Goethe’s name becomes with ‘oe’ written
‘Goethe’s name is written with ‘oe’.’ (eventive)
b. Goethes Name war mit ‘ae’ geschrieben.
Goethe’s name was with ‘ae’ written
‘Goethe’s name was found written with ‘ae’.’ (result)
Spanish also has equivalents of werden ‘become’, which like werden can be taken to
be dynamic auxiliaries. For instance, ponerse, literally ‘put-SE’, is one such example.
Why are such verbs not used in passives instead of the copula ser, which does not add
the dynamicity that is required in an event passive?
() a. Juan se puso gordo.
Juan se became fat
b. *El libro se puso escrito.
the book SE became written
Intended: ‘The book was written.’ (eventive)
There are at least two ways to address this problem. The first one is to reduce the
eventive passive to a copular sentence where the predicate happens to be a non-finite
verbal category which contains more aspectual information than a usual adjective,
triggering the eventive reading (as argued in Crespí  with facts relating to
Aktionsart, temporal anchoring, and grammatical aspectual information).
The second one is to take (a) at face value and accept that, in addition to the
predication support function of the copula, there is a distinct use as a passive
auxiliary. Given that this use is typologically frequent, treating auxiliary be and
copular be as a case of homophony would not be a plausible analysis. Instead, the
existence of such cases would support a view where estar is associated to specific
aspectual information, while ser is in Spanish an underspecified support element for
verbal morphology that can be introduced at different positions in the structure,
including the area where diathesis is defined, independently of the aspectual infor-
mation related to the predicate.

⁴ We gloss over the controversy of whether (b) is a case of adjectival passive, or even whether
adjectival passives would be significantly different from verbal passives; see Emonds (), Bruening
(), Gehrke (), and Arche et al. () for discussion.
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

Support for this view that the plurifunctionality of the copula is linked to its nature as
a mere support element (at least in some languages) is brought to this volume by Borik.
This author says that Russian copulas must be overt to support overt tense morphology
and their presence cannot be due to a semantic contribution involving eventivity, as
Paslawska and von Stechow () had previously proposed. Borik shows that there is
no empirical correlation between eventivity and presence of the copula in Russian.

. Main theoretical debates about copulas

In this section we will concentrate on the main theoretical debates that the empirical
facts just revised have addressed. We will focus on two fundamental issues: the nature
of the copula as a linguistic object, and how the different types of copular structures
are differentiated, namely, either lexically—that is, through the use of distinct
copulas—syntactically through different base structures or different derivational
sequences, or semantically, deriving the different interpretations from the contribu-
tions of the subjects and predicates. The first question concerning their nature is
informed not only by the facts reported in §.. about the different categorization of
copulas but also by the uses that copulas have (§..) in different languages,
including their role as auxiliaries or focus-markers. Similarly, the properties that
regulate the distribution of copulas when more than one such element is available in
a language (§..) is also obviously relevant, including the type of copular sentence
(§..). The second question mainly refers to the facts in §.. and §.., where
different copulas are used in different types of sentences.

.. The nature of the copula

The nature of copulas seems deeply connected to their role within clause structure.
While all analyses share the intuition that the copula is a support element, there is
disagreement with respect to what exactly it is a support of. There are three main
proposals about the category that a copula instantiates as enumerated below; we will
study them in turn in what follows.
(i) Copulas are the spellout of T/Infl
(ii) Copulas are the spellout of a type of v/V
(iii) Copulas are the spellout of predication heads
... Copulas above vP Several authors claim that a copular verb like be is
introduced in T⁰ (or Infl) directly (for instance, Dik , ; Hengeveld ;
Ouhalla ; Carnie ; see also Kiss  for Hungarian; Doron  and
Greenberg  for Hebrew hu; Doherty  for Irish is; Baker  for English;
and Roy  for Spanish). Thus, it is not a verb: rather, it is a head that realizes the
verbal inflectional categories (tense, subject agreement . . . ) when the predicate can-
not combine with them, either because it is a nonverbal category or because it is a
non-finite verb (participle, gerund or infinitive, for instance). O’Neill (this volume)
provides evidence that in English all forms of the copula be are inflectional support
items. Other researchers, like Becker (), have argued that only some copular
forms are inflectional support, while others are verbs.
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

The view that be is plain inflectional support regards copulas as dummy elements
(to pass the Stray Affix Filter), supporting inflectional information that cannot be
morphologically carried by the nonverbal predicate category. This accounts for several
facts. First, it provides an explanation as to why copulas are necessary in inflectional
languages to form independent sentences with categories that reject tense and subject
agreement morphemes. Second, it also explains why copulas (generally) do not make a
semantic contribution. And third, it plausibly explains why in some languages copular
elements are only required when there is a polarity or temporally marked form, on the
assumption that the unmarked interpretation (positive and present, respectively) is
obtained without overt affixes. Arabic, for example, does not use a copula in the
affirmative present tense as shown in (a), but it requires a form to carry inflection
in the past; see (b), and the negative present tense, as illustrated in (c).
() a. L-madi:natu ?a:miratun.
the town busy
‘The town is busy.’
b. Ka:nat l-madi:natu ?a:miratun.
was the-town busy
‘The town was busy.’
c. Laysat l-madinatu ?a:miratun.
is.not the-town busy
‘The town is not busy.’
The idea is also largely compatible with the use of copulas in focus and passive
constructions, to the extent that it involves viewing it as a support element which,
then, is expected to be underspecified.
However, the claim that copulas are universally inflectional support elements cannot
account for the fact that copulas exist in languages that lack morphological expression of
tense, aspect, and subject agreement. Welch (this volume) shows, for instance, that
Chinese has a copular element shi, which is indisputably analytic. Another problem
faced by the idea that copulas are introduced directly in T (or Infl) is the one presented
by cases such as (), where the copula follows a root or dynamic modal. As is known,
root and dynamic modals are introduced below TP in the structure, perhaps immedi-
ately above vP or AspP (see, e.g., Picallo ). If all copular forms are merged in T,
cases such as (), instances of root modality semantics, are unexpected.
() a. Puedes ser agresivo.
you.can be aggressive
‘You are allowed to be aggressive.’
b. Es capaz de ser cruel.
she.is able to be cruel
‘She is able to be cruel.’
Also, those cases where the copula distinctions are shown to be sensitive to lexical
properties of the predicate, such as its category or its conceptual information (see
§. above) show that the approach is not without complications.
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

... Copulas as (light) verbs Facts discrediting copulas as mere materializations

of Inflection have led other researchers to propose that copular verbs are introduced
lower in the structure and can be considered light verbs. Copulas are analyzed as
verbal supports in Halle and Marantz (), Moro (), Rothstein (), Schütze
(), Mikkelsen (), and Cowper (b). Note also that, in a sense, any system
that uses vbe as a flavor of v, such as Harley (), can be said to adhere to this view
somehow.
In addition to their ordering with respect to modal verbs, the proposal that copulas
are introduced as verbs has other arguments in its favor. For English, Schütze ()
makes the point that be can be combined with the aspectual affixes -en and -ing, as
shown in (), which importantly in English combine with full verbs, but not modal
auxiliaries. In fact, Schütze argues that the English copula is introduced in these
contexts as a support for such affixes.
() Diana Prince had be-en be-ing complimented.
Theories that claim that copulas are full verbs have to face the problem that it
is not clear that copulas contribute any meaning. Note, for instance, that in
Harley’s () system, the behavior associated to vbe is defined by the absence of
the properties that other flavors of v have: unlike vdo, it does not introduce an agent;
unlike vbecome, it does not define a change, and unlike vcause, it does not initiate a
process.
The problem is solved by some authors by proposing that copulas lack ‘concep-
tual’ content, but contribute structural semantics. Welch, this volume, argues that
copulas can be viewed as devices used to introduce argument structure in contexts
where the predicate is nonverbal. Another solution is to associate the copula with a
distinct meaning, something that some authors have also proposed.
For some, at least identificational copulas (Peter Parker is Spiderman) have a theta-
grid (see Pereltsvaig , who adheres to this theory, for an overview). Others have
argued that all uses of the copula involve some meaning. In this vein, Rothstein
() has argued for English that the copula be contributes a characterizing mean-
ing, as seen in (a) as opposed to (b).
() a. Bruce Wayne seems to be clever.
b. Bruce Wayne seems clever.
Along similar lines, Becker () has argued for a semantically rich copula, even
though she restricts the claim to non-finite forms (e.g., English be). In () below,
(b) must mean that Lois forced Clark to act politely, while (a) means that she
tutored Clark to make him into a polite person.
() a. Lois made Clark polite.
b. Lois made Clark be polite.
However, a generalization based on this view is difficult to maintain, as there is no
apparent meaning difference in (). To explain the lack of difference in () one
could assume for () that English has a null copula form also present in (a), but
then the question would be why the same null copula is not available in (b).
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

() a. This compound seems coordinative.

b. This compound seems to be coordinative.

Finally, there is an intermediate analysis that combines intuitions from the two
accounts just described: that copulas are introduced higher than verbs and that
copulas carry meaning. Examples of this view are the works by Brucart (,
) and Camacho () about Spanish. They have claimed that the SL-related
copula estar ‘be’ in Spanish is a manifestation of AspP, and is actually introduced to
assign an inchoative aspectual value to the adjectival predicate.

() Corto está triste.

Corto isestar sad
AspP

Asp AP
estar
[uAsp [inch]] A ...
triste
[iAsp [ ]]

Although this analysis escapes some of the counterarguments to the previous pro-
posals, it presents other issues. For example, there are prototypical cases of estar that
are not inchoative in any sense, such as locative ones like Gotham está en América
‘Gotham isestar in America’. Also, it is difficult to see how the idea can apply to the
other copula, ser, which does not seem to carry any aspectual value, and is, in fact,
compatible with stative-like and activity-like predicates. See examples in () in this
regard and Arche () and references therein for cases of the sort of (b), with
evaluative adjectives. In fact, Camacho () is forced to label ser’s projection as
Cop(ular)P, unable to identify it with an independent category.

() a. Jessica Jones {es/*está siendo} atlética.

Jessica Jones isser/is beingser athletic
b. Jessica Jones {es/está siendo} comprensiva.
Jessica Jones isser is beingser understanding

... Copulas lower than vPs The third existing proposal about what kind of head
copulas instantiate is that they are support elements that establish a predicative
relation between an adjective or noun and its (thematic) subject. This approach
presupposes a view of small clauses as asymmetric structures where subject and
predicate are mediated by a designated head as shown in (–); see Kayne (),
Hornstein and Lightfoot (), Bowers (), Svenonius (), Pereltsvaig
(). The copula can be analyzed as that intermediate head, and therefore is treated
as a predicational support. The approach contrasts with both symmetric accounts of
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

small clauses (famously, Moro , Pereltsvaig ; a) and traditional approaches
where the subject of a nonverbal predicate was introduced as a specifier or adjunct
inside the predicate’s lexical projection (Stowell a, Manzini ; b).
() [PredP [DP Matt Murdock] Pred⁰ [AP blind]]]
() a. [SC[DP Matt Murdock] [AP blind]]
b. [AP[DP Matt Murdock] A⁰ blind]
Baker (: ), who explicitly denies that English be is a manifestation of Pred⁰,
argues that copular particles in a language like Edo (Niger-Congo) are manifestations
of this head. The diagnostic relies on his own proposal, where PredP is only necessary
to define nonverbal categories as predicates, since verbs are able to introduce their
own subjects. For example, copulas in Edo must be used when the predicate is a noun
or an adjective.
() a. Èmèrí *(yé) mòsèmòsè.
Mary COP beautiful
b. Úyì *(re) òkhaèmwèn.
Uyi COP chief
Baker notes that, crucially, Edo copulas never double as auxiliaries, and they are also
required when the nonverbal small clause is embedded under a main verb in ECM
constructions. These are precisely the properties that we would expect if Edo copulas
are not verbs or manifestations of higher functional projections or inflectional features.
Baker, finally, suggests that nonverbal copulas in languages like Somali, Berber, Chinese,
Niuean, and Paumari are also instantiations of PredP. Other studies that have argued
that copulas are projections of PredP (or an equivalent head used to support predica-
tion) include Bowers () for Irish copular particles, Nishiyama () for the
predicative copula in Japanese, Adger and Ramchand () for pronominal augments
in Scottish Gaelic, and van Gelderen (b). Den Dikken (a) also proposes that
the English copula be can be an instantiation of Rel(ator), which heads a predication
structure, although we must hasten to add that den Dikken (a: –) is very clear
in his book that Rel is not the label of a specific category comparable to Pred, but rather
a function that can be satisfied by different kinds of elements; den Dikken’s claim, thus,
is about the function of the copula, not its category.
Thus, we have seen that there are arguments to support the claim that languages
have copular elements that instantiate T, v, Asp, and Pred, at least, but none of these
heads can account for all of the different kinds of copulas attested cross-linguistically.
It has also been claimed that within one single language there is more than one kind
of copula (e.g., ser/estar in Spanish). Even if we give up the idea of identifying one
single category to cover all copular elements, the category of the copula in a language
might still be problematic, as it is the case of English be, where, as we have seen, some
authors have claimed that not all of its forms instantiate the same category, or that
distinct uses of the verb must be differentiated. In contrast, the function of copulas is
much clearer: they are support elements that, depending on other factors, are needed
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

to license agreement, other inflectional elements, aspectual information or predica-

tion in contexts where a fully fledged verbal predicate is missing.

.. The nature of copular structures

In §.. we saw that most approaches that reduce copular sentences to two kinds are
of a syntactic nature. From the semantic side (and sometimes the purely philosoph-
ical side), however, there are several approaches where there is at least a division
between equative copular sentences and other sentences. Interestingly, this has had
consequences for the nature of copulas, specifically with respect to whether several be
elements must be distinguished within the same language.
John Stuart Mill () and Bertrand Russell (: ) famously contended
that English be stands for two distinct elements: the real copular use, which is
possibly semantically empty, and a transitive verb (in Montagovian terms, of type
<e,<e,t>>) that takes two arguments and establishes an identity relation between
them; thus, there is a copular be and an equative be that happen to be spelled out in
the same way. The same position is adopted in Seuren () and Higginbotham
(). Bolinger () goes further and proposes a third be, the locational one. As
we saw in §.., this division is motivated for at least some languages through the
existence of distinct copular elements that are specialized for different functions,
but the claim is made also for languages that do not morphologically differentiate
between these three uses.
Theories that reduce the typology of copular sentences to two classes, one basic
(the predicational type) and one derived from the other (the specificational or inverse
one) forcefully claim that there is only one verb be, of essentially empty semantics.
The view that there is only one be is argued for in Montague (), Partee (),
Moro (, , ), and den Dikken (, a). In these accounts, whatever
is empirically distinct across the copular classes has to be a function of (i) the
semantic definition of the two elements related by the copula and (ii) the syntactic
operations that derive one type of copular sentence from the other. One prototypical
example of this line of research is Adger and Ramchand (), who study Scottish
Gaelic facts and derive the empirical differences between (apparent) equatives and
predicational copular sentences from the syntactic-semantic structure of the NPs
involved.
Note, however, that it is in principle possible to claim that equative sentences and
predicational sentences (inverse or not) are syntactically and semantically different
and still defend an analysis where there is only one copula be. One prototypical
example of this is the proposal by Heycock and Kroch (: –), who contend
that there is a real distinction between predication and equation. In particular, they
claimed that there are two types of small clauses: predicational and equative, inde-
pendent of the copula. The example below illustrates an equative small clause with
the verb make, with no copula involved.

() If what you say is true, that would make [the real murderer John].
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

Obviously, this position would presuppose that the copula is not the spellout of a
predicational head.
Independently of the question of whether an equative copular type must be
differentiated, one additional area where there has been debate is whether the
division between predicational and specificational copular sentences is syntactically
reflected. Most theories that accept the divide in some form acknowledge that they
are differentiated by the referentiality or predicationality of the NPs involved,
roughly as in ():

() a. Predicational: [More referential NP] is [Less referential NP]

b. Specificational: [Less referential NP] is [More referential NP]
However, a number of proposals have argued that both types of copular sentence are
derived from the same underlying structure, roughly the one in (), where there is a
small clause:
() be [SC [More referential NP] [Less referential NP]]

This underlying structure is closer to the predicational scheme in (a), assuming

that the more referential NP (subject of predication) is closer to higher projections,
like InflP/TP, than the predicate NP. Hence, (b) involves some kind of syntactic
inversion, a claim made by, among others, Williams (), Heggie (),
Moro (, , ), Heycock (, ), and den Dikken (, a).
As hinted at above, this idea constrasts with theories where identity and predication
involve two different base generated structures, as was the case of Heycock
and Kroch () mentioned above. The derivational analysis predicts syntactic
differences in the predicational and specificational order. Here are some of those
that have been noted:
a) When the small clause is selected by a main verb, only the non-inverted
order is allowed. This follows if under the main verb there is no structural
space to move the second NP above the first one. Moro () noted that
once the copula is present, the inversion is possible. This can be analyzed in
a number of ways, but the common intuition is that the presence of the
copula introduces additional syntactic space that creates a position for the
second NP to be moved across the first one. This is shown in the examples
() and ():

() a. I consider Superman the best superhero. (predicational)

b. *I consider the best superhero Superman. (specificational)
() I consider the best superhero *(to be) Superman.
b) Extraction out of the second NP in a predicational copular sentence is possible,
while it is banned in the specificational sentence (Grosu , Moro ).
This follows if the second NP in the specificational is a subject; only then
would cases such as () be a case of subject-island.
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

() a. I think that a picture of Batman was the cause of the riot (predicational)
b. What do you think that a picture of Batman was the cause of ?
(’) a. I think that the cause of the riot was a picture of Batman. (Inverse)
b. *Who do you think that the cause of the riot was a picture of ?
We refer to den Dikken (a, b) for an exhaustive list of further syntactic and
semantic differences. What is relevant for us here is that an inversion analysis of
specificational copular sentences looks plausible in light of these facts.
() [NP . . . [[NP ] [NP]]]
This analysis, however, raises some questions. The first one has to do with the
potential intervention effects triggered by the first NP. Assuming that the subject
NP c-commands the second, as any asymmetric analysis of small clauses states, how
can the lower NP cross over it? Some of the articles in this volume discuss this
problem. Loccioni, who, as we mentioned, researches the distribution of two copular
elements in Logooni (Bantu), argues that kuva is a real copula associated to predica-
tional clauses; in contrast ne is compulsory in specificational clauses. She claims that
ne is the spellout of a functional head that makes the dislocation of NP possible. In
her view, the intervention effect that () should be subject to is avoided in the
following way: ne spells out an F⁰ head that selects TP; FP allows either of the two
NPs, depending on their information value, to move to Spec,FP. Hence, the inter-
vention is solved by the proposal that the movement that displaces NP above NP is
motivated by features that, in the derivation where this happens, NP is not associ-
ated to. In the absence of FP/ne, NP cannot cross above NP. Roy and Shlonsky,
also in this volume, focus on the intervention problem, and study it within a number
of contexts where French copular sentences require the presence of the expletive ce
(Jean, c’est mon meilleur ami, lit. ‘Jean, it’s my best friend’). They claim that this
expletive is introduced when, for whatever reason, TP cannot establish a direct
agreement relation with an NP in the small clause below it. One such case is directly
relevant for the problem of the intervention effect: the inverse order is obtained when
NP moves to a low focus position, and then the small clause (containing the overt
NP, and an unpronounced copy of NP) moves to a position above focus. In
general, whenever NP is in a focus position, agreement of TP with it will be
impossible, triggering insertion of ce to satisfy a compulsory subject position. This
triggers in some cases singular agreement even when there is an overt plural NP:
() Les enfants c’est bruyant.
the children it is noisy
‘Children are noisy.’
Agreement facts are introduced as part of the argumentation for one or another
structure. In inverse copular sentences, it has been argued that the fact that the NP
triggers agreement () is an argument in favor of it moving above NP to an
argument position (Heycock , Moro ), against analyses where it moves to
an A’ position (Blom and Daalder , Heggie ).
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

() a. The mafia bosses are our biggest problem.

b. Our biggest problem is the mafia bosses.
However, as noted by some authors (cf. for example Declerck : –), these
facts are not systematic. Languages such as Spanish show a strong tendency to still
agree with the NP in inverse copular sentences. Consider (), the direct translation
of () to Spanish. While (a) poses no problem, the contrast in (b) shows that
agreement with the NP is strongly preferred. It has the flavor of a propositional
interpretation of mafia bosses, along the lines of ‘the fact that there are mafia bosses’.
() a. Los capos mafiosos son nuestro mayor problema.
the bosses mafia-ADJ are our biggest problem
b. Nuestro mayor problema {son / ?es} los capos mafiosos.
our biggest problem are / is the bosses mafia-ADJ
The same happens with st or nd person agreement as shown below.
() a. El Estado {soy / *es} yo.
the State am is I
b. El que ganó {eres / *es} tú.
he who won are is you
This might suggest that different languages use different strategies to derive the
inverse ordering, with English possibly moving NP to an A-position and Spanish
displacing it to an A’-position, although there are other conceivable solutions in
terms of feature prominence.

. The proposals in this volume

In this final section, we will provide a closer overview of the points made in the rest of
the chapters of this volume, highlighting the points of agreement and disagreement. The
chapters can be grouped in three parts, according to what their main claims are about:
• Chapters dealing with the nature of copulas as linguistic objects: the condi-
tions for identifying them, their role as support elements, and their place
within the structure of the clause. In addition to the current chapter, the
contributions by Welch, Kim, and O’Neill, chapters one to four, compose
this part.
• Chapters about the formal operations of the grammar that copular
structures can crucially inform: agreement and the interaction between
phi-feature deficiency and agreement. The contributions by Bejar et al.,
Bartošová and Kučerová, Bondaruk, and Roy and Shlonsky, chapters five
to eight, are included here.
• Finally, the set of chapters that refer to the different structures where
copulas appear, including passive sentences and focus configurations, and
how the nature of these structures might determine the choice of copula.
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

The chapters included here are nine to twelve, authored by Borik, Sáez,
Gibson et al., and Loccioni.

.. Copulas as formal support elements

As should hopefully have become clear from this chapter, copulas cannot be treated
as normal verbs without losing relevant generalizations. From the articles contained
in this volume, a clear consensus emerges that copulas are devices that are required in
order to support other elements in situations where a full verbal predication is not
defined.
However, the disagreement emerges when one tries to identify what exactly
copulas are a support for. Welch, this volume, provides a detailed analysis of
semantically light verbs in the Dene languages. In his proposal, copulas and light
verbs should be considered members of the same general class, roughly syntactic
devices that license argument structure. Copulas, then, are used in nonverbal predi-
cation because the predicates involved in such situations are unable to syntactically
define the structures that allow natural languages to introduce arguments. In his
view, then, copulas are introduced as syntactic support devices. The different copulas
studied in his chapter lack lexical semantics, and are differentiated solely on the base
of the context where they are introduced—for instance, by the eventivity of their
complement.
The view of copulas as syntactic support elements sharply contrasts with the view
of O’Neill in several respects. O’Neill argues that the English copula is a
morphological—not syntactic—support element which, moreover, is not associated
to the V category, or always introduced in T. Through an analysis of amalgam
specificational copular sentences (see Sáez, this volume, for another analysis of
amalgam constructions), she argues that copulas are introduced in any kind of
context where inflection is left stranded without a morphological host. This predicts
that copulas will not only be found in V or T nodes, but crucially also in nodes
belonging to the C domain, which is what she argues is the case in amalgam
specificational clauses, as they do not show any evidence of carrying a T node.
Welch and O’Neill’s positions can be contrasted in other ways. If copulas are
syntactic devices used to license argument structures, we expect that any language
with light verbs will have copulas, while if it is a morphological support device we
expect more variation, to the extent that morphology varies in more radical (and less
systematic ways) than syntax. Kim’s contribution in this volume argues, in fact, that
Korean lacks a copula in the sense that we understand this term for English or
Spanish. The goal of her article is to show that the Korean morpheme -I, traditionally
described as a copula, is in fact the manifestation of nominative case marking. In
addition to illustrating the difficulties that one encounters when trying to classify
semantically empty elements across languages, Kim shows that nominative case
marking and the alleged copula cannot co-occur and have a similar distribution;
the reason for treating it as a copula in the past is due to -I being introduced at the TP
level, and the compatibility with PPs, which normally are analyzed as not receiving
case. Kim, however, convincingly argues that in Korean PPs must get case in some
contexts.
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

Even though the rest of the articles in this volume do not set the identification of
the status of copulas as their main goal, it is interesting that they also show some
disagreement with respect to the treatment that the copula has to receive in the
language they study. Bejar et al., Bondaruk, and Loccioni assume Stowell’s (a)
treatment of copulas as raising verbs that combine with a small clause, suggesting
that they implicitly view copulas as filling a compulsory VP layer within the clause.
Bartošová and Kučerová, Roy and Shlonsky, Borik, and Sáez, in contrast, assume that
the copula is introduced at the TP level, which suggests—sometimes explicitly, as in
Borik, this volume—that they view them as support devices for inflection. The fact
that the contributions in this volume are divided between these two views makes it
clear that the syntactic position of the copulas is still an issue where there is
disagreement.

.. Copulas and agreement

The fact that copulas are used in contexts where both predicate and subject are
nominal elements makes copular sentences a unique source of information for the
study of agreement relations and phi-feature operations, as several of the chapters in
this volume show. From reading the contributions by Bejar et al., Bartošová and
Kučerová, Bondaruk, and Roy and Shlonsky, a clear consensus emerges that agree-
ment cannot be viewed as a monolithic process.
Bejar and colleagues focus on situations where there has to be number matching
between the two NPs inside a copular sentence ().
() Mary and Jane are {violinists / *a violinist} in two orchestras.
They argue that this feature-matching between the two DPs cannot be treated by
agreement understood in the regular syntactic sense. Apparent number takes place
within a small clause, before structurally introducing the copula, as a result of
feature-matching under merge when combining the two DPs. This happens only
when one of the two DPs lacks a value for number. There are other cases where
number matching does not take place ().
() [When making a snowman with a child]
The banana is the {eyebrow / eyebrows}.
Absence of number-matching emerges whenever neither of the DPs has an unvalued
number feature, which can be due to a number of different reasons: if the predicate
DP is referentially complex, it is introduced with a valued number feature, which
prevents matching; in other situations, the nominal expression is functionally
reduced and lacks the number layer, and finally in other cases the noun gets a
propositional interpretation, which involves it being embedded under a CP that
makes number inaccessible.
In her contribution, Bondaruk discusses Polish agreement patterns that are
partially reminiscent of the Spanish facts reported in (). In so-called dual copula
clauses, where the pronominal copula and the verbal copula co-occur, the verbal
copula agrees with the nominal that follows it if both are third person (). If one of
 OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

Main questions in the study of copulas 

the two nominals is st or nd person, the verbal copula agrees with it independently
of whether it is placed before or after the verb ().
() Ci zawodnicy to {była /*byli} drużyna.
these players-M.PL COP F.SG M.PL team-F.SG
‘These players were a team.’

() My to jesteśmy złodzieje.

we cop PL thieves-PL
‘We are thieves.’

In her analysis, tense is a multiple probe for agreement, which enters simultaneously
in a syntactic relation with both DPs, assigning nominative to both. One DP checks
T’s EPP feature, while another satisfies phi-feature valuation. The difference between
inverted and non-inverted copular clauses is rooted in which DP satisfies each
property when both are third person. In contrast, if one of the DPs is st or nd
person, given the feature hierarchy of person, this feature will block a third person
nominal from valuing the phi features.
In Bondaruk’s contribution, EPP satisfaction and phi-feature valuation have to
be dissociated. Bartošová and Kučerová propose an analysis that ultimately involves
dissociating nominative case assignment from phi-feature agreement. In Czech, the
neuter singular pronoun to, which they take to be phi defective, can be used to refer
to entities of any gender, not just neuter ones. In copular sentences where the verbal
copula agrees in gender with an NP, however, the antecedent of the pronoun must
match that gender specification. The explanation for this pattern is related to the
deficiency of the pronoun, which makes it unable to satisfy the phi features of the
copula. This has two results: to establishes an agreement link with the copula, but
does not value its phi features; as the features are still active, the copula establishes a
second agree link with the second NP and values it. This creates a multiple agree
chain between the pronoun, the copula and the NP, which will then share the same
value for phi features. On the assumption that gender features act as presupposi-
tions that restrict the reference of pronouns, the Czech pattern follows. The account
makes non-trivial predictions for the treatment of expletive pronouns, the relation
between person and animacy, and the larger distribution of the phi-deficient
pronoun to.
Roy and Shlonsky’s chapter is also concerned with situations where pronouns and
copulas interact in situations where agreement is not possible, although they make a
proposal that, in a sense, is the opposite of Bartošová and Kučerová’s, where the
introduction of a neuter pronoun in fact satisfies agreement when a DP cannot do so.
Their empirical data set involves the contrast between sentences where the expletive
ce pronoun can be introduced, and those where it must be introduced.
() a. Jean (c’)est mon meilleur ami.
Jean CE is my best friend
‘Jean is my best friend.’
OUP CORRECTED PROOF – FINAL, 21/12/2018, SPi

 Arche, Fábregas, and Marín

b. Mon meilleur ami *(c’)est Jean.

my best friend CE is Jean
‘My best friend is Jean.’

They show that ce is required in inverse copular sentences, and optional in the canonical
ones, assuming Moro’s () view, but whenever it is present, the pre-verbal DP is
interpreted as a topic. Their proposal is that in cases where ce is used the structure
involves configurations where, for different reasons, no DP in the sentence can move to
the agreement checking position, which they identify as a low subject position, and the
subject moves to a higher position where it is interpreted as presuppositional or generic.
Ce is introduced in such cases in the low subject position in order to check agreement of
the copula. As can be seen, their approach implies deconstructing the notion of subject,
distributing the properties associated to this function across distinct heads.

.. Copulas as part of different constructions

The final four chapters in this volume study the nature of copulas in different
structures, including a detailed examination of languages with more than one copula
where the distribution at least partially reflects differences in copular sentence types.
The goal of Borik’s contribution is to address the role of Russian copulas in
periphrastic passive constructions. Adhering to a view of copulas as morphological
support elements—specifically, needed whenever tense is spelled out with a marked
value—Borik argues against Paslawska and von Stechow (), who suggest that the
null copula of Russian forces a stative interpretation of the passive. This would imply
that the copula carries with it a semantic contribution when overt. Borik shows that
passives without an overt copula can have an eventive interpretation, and that in
terms of aspect both passives with an overt copula in the past and passives with an
empty copula in the present can receive the same interpretation. Providing semantic
formulas that account for the interpretive differences, when they exist, she shows that
the copula does not make any semantic contribution, and thus concludes that the
generally held view in Russian linguistics—that the copula is present when there is a
marked temporal value—can be maintained. Importantly, this implies that the
copula does not have an additional value as an auxiliary, as the reason for inserting
it in such sentences is identical to what we find in nominal predicate clauses.
Sáez analyzes the presence of copulas in focus constructions in some Spanish
varieties (), where an inflected form of the copula used for individual level
predicates introduces an element that is forcefully interpreted as focus.

() Juan compró fue UN LIBRO.

Juan bought was A BOOK
‘It was a book that Juan bought.’

After critically reviewing some previous approaches, he concludes that the reason for
finding the copula here is that the structure involves a bi-clausal cleft structure with
the properties of Horn-amalgams (Kluck ). In his analysis, as was the case in
Borik’s analysis, therefore, the copula is used in such contexts not because it develops
Another random document with
no related content on Scribd:
Western physics exhibits the “how” and the “how long” as distinct in
essence. As soon as the question is pressed home, causality
restricts its answer rigidly to the statement that something happens
—and not when it happens. The “effect” must of necessity be put
with the “cause.” The distance between them belongs to a different
order, it lies within the act of understanding itself (which is an
element of life) and not within the thing or things understood. It is of
the essence of the extended that it overcomes directedness, and of
Space that it contradicts Time, and yet the latter, as the more
fundamental, precedes and underlies the former. Destiny claims the
same precedence; we begin with the idea of Destiny, and only later,
when our waking-consciousness looks fearfully for a spell that will
bind in the sense-world and overcome the death that cannot be
evaded, do we conceive causality as an anti-Fate, and make it
create another world to protect us from and console us for this. And
as the web of cause and effect gradually spreads over the visible
surfaces there is formed a convincing picture of timeless duration—
essentially, Being, but Being endowed with attributes by the sheer
force of pure thought. This tendency underlies the feeling, well
known in all mature Cultures, that “Knowledge is Power,” the power
that is meant being power over Destiny. The abstract savant, the
natural-science researcher, the thinker in systems, whose whole
intellectual existence bases itself on the causality principle, are “late”
manifestations of an unconscious hatred of the powers of
incomprehensible Destiny. “Pure Reason” denies all possibilities that
are outside itself. Here strict thought and great art are eternally in
conflict. The one keeps its feet, and the other lets itself go. A man
like Kant must always feel himself as superior to a Beethoven as the
adult is to the child, but this will not prevent a Beethoven from
regarding the “Critique of Pure Reason” as a pitiable sort of
philosophy. Teleology, that nonsense of all nonsenses within
science, is a misdirected attempt to deal mechanically with the living
content of scientific knowledge (for knowledge implies someone to
know, and though the substance of thought may be “Nature” the act
of thought is history), and so with life itself as an inverted causality.
Teleology is a caricature of the Destiny-idea which transforms the
vocation of Dante into the aim of the savant. It is the deepest and
most characteristic tendency both of Darwinism—the megalopolitan-
intellectual product of the most abstract of all Civilizations—and of
the materialist conception of history which springs from the same
root as Darwinism and, like it, kills all that is organic and fateful. Thus
the morphological element of the Causal is a Principle, and the
morphological element of Destiny is an Idea, an idea that is
incapable of being “cognized,” described or defined, and can only be
felt and inwardly lived. This idea is something of which one is either
entirely ignorant or else—like the man of the spring and every truly
significant man of the late seasons, believer, lover, artist, poet—
entirely certain.
Thus Destiny is seen to be the true existence-mode of the prime
phenomenon, that in which the living idea of becoming unfolds itself
immediately to the intuitive vision. And therefore the Destiny-idea
dominates the whole world-picture of history, while causality, which is
the existence-mode of objects and stamps out of the world of
sensations a set of well-distinguished and well-defined things,
properties and relations, dominates and penetrates, as the form of
the understanding, the Nature-world that is the understanding’s “alter
ego.”
But inquiry into the degree of validity of causal connexions within a
presentation of nature, or (what is henceforth the same thing for us)
into the destinies involved in that presentation, becomes far more
difficult still when we come to realize that for primitive man or for the
child no comprehensive causally-ordered world exists at all as yet
and that we ourselves, though “late” men with a consciousness
disciplined by powerful speech-sharpened thought, can do no more,
even in moments of the most strained attention (the only ones, really,
in which we are exactly in the physical focus), than assert that the
causal order which we see in such a moment is continuously present
in the actuality around us. Even waking, we take in the actual, “the
living garment of the Deity,” physiognomically, and we do so
involuntarily and by virtue of a power of experience that is rooted in
the deep sources of life.
A systematic delineation, on the contrary, is the expression of an
understanding emancipated from perception, and by means of it we
bring the mental picture of all times and all men into conformity with
the moment’s picture of Nature as ordered by ourselves. But the
mode of this ordering, which has a history that we cannot interfere
with in the smallest degree, is not the working of a cause, but a
destiny.

II

The way to the problem of Time, then, begins in the primitive

wistfulness and passes through its clearer issue the Destiny-idea.
We have now to try to outline, briefly, the content of that problem, so
far as it affects the subject of this book.
The word Time is a sort of charm to summon up that intensely
personal something designated earlier as the “proper,” which with an
inner certainty we oppose to the “alien” something that is borne in
upon each of us amongst and within the crowding impressions of the
sense-life. “The Proper,” “Destiny” and “Time” are interchangeable
words.
The problem of Time, like that of Destiny, has been completely
misunderstood by all thinkers who have confined themselves to the
systematic of the Become. In Kant’s celebrated theory there is not
one word about its character of directedness. Not only so, but the
omission has never even been noticed. But what is time as a length,
time without direction? Everything living, we can only repeat, has
“life,” direction, impulse, will, a movement-quality (Bewegtheit) that is
most intimately allied to yearning and has not the smallest element in
common with the “motion” (Bewegung) of the physicists. The living is
indivisible and irreversible, once and uniquely occurring, and its
course is entirely indeterminable by mechanics. For all such qualities
belong to the essence of Destiny, and “Time”—that which we actually
feel at the sound of the word, which is clearer in music than in
language, and in poetry than in prose—has this organic essence,
while Space has not. Hence, Kant and the rest notwithstanding, it is
impossible to bring Time with Space under one general Critique.
Space is a conception, but time is a word to indicate something
inconceivable, a sound-symbol, and to use it as a notion,
scientifically, is utterly to misconceive its nature. Even the word
direction—which unfortunately cannot be replaced by another—is
liable to mislead owing to its visual content. The vector-notion in
physics is a case in point.
For primitive man the word “time” can have no meaning. He simply
lives, without any necessity of specifying an opposition to something
else. He has time, but he knows nothing of it. All of us are conscious,
as being aware, of space only, and not of time. Space “is,” (i.e.
exists, in and with our sense-world)—as a self-extension while we
are living the ordinary life of dream, impulse, intuition and conduct,
and as space in the strict sense in the moments of strained attention.
“Time,” on the contrary, is a discovery, which is only made by
thinking. We create it as an idea or notion and do not begin till much
later to suspect that we ourselves are Time, inasmuch as we live.[102]
And only the higher Cultures, whose world-conceptions have
reached the mechanical-Nature stage, are capable of deriving from
their consciousness of a well-ordered measurable and
comprehensible Spatial, the projected image of time, the phantom
time,[103] which satisfies their need of comprehending, measuring and
causally ordering all. And this impulse—a sign of the sophistication
of existence that makes its appearance quite early in every Culture—
fashions, outside and beyond the real life-feeling, that which is called
time in all higher languages and has become for the town-intellect a
completely inorganic magnitude, as deceptive as it is current. But, if
the characteristics, or rather the characteristic, of extension—limit
and causality—is really wizard’s gear wherewith our proper soul
attempts to conjure and bind alien powers—Goethe speaks
somewhere of the “principle of reasonable order that we bear within
ourselves and could impress as the seal of our power upon
everything that we touch”—if all law is a fetter which our world-dread
hurries to fix upon the incrowding sensuous, a deep necessity of
self-preservation, so also the invention of a time that is knowable
and spatially representable within causality is a later act of this same
self-preservation, an attempt to bind by the force of notion the
tormenting inward riddle that is doubly tormenting to the intellect that
has attained power only to find itself defied. Always a subtle hatred
underlies the intellectual process by which anything is forced into the
domain and form-world of measure and law. The living is killed by
being introduced into space, for space is dead and makes dead.
With birth is given death, with the fulfilment the end. Something dies
within the woman when she conceives—hence comes that eternal
hatred of the sexes, child of world-fear. The man destroys, in a very
deep sense, when he begets—by bodily act in the sensuous world,
by “knowing” in the intellectual. Even in Luther[104] the word “know”
has the secondary genital sense. And with the “knowledge” of life—
which remains alien to the lower animals—the knowledge of death
has gained that power which dominates man’s whole waking
consciousness. By a picture of time the actual is changed into the
transitory.[105]
The mere creation of the name Time was an unparalleled
deliverance. To name anything by a name is to win power over it.
This is the essence of primitive man’s art of magic—the evil powers
are constrained by naming them, and the enemy is weakened or
killed by coupling certain magic procedures with his name.[106]
And there is something of this primitive expression of world-fear in
the way in which all systematic philosophies use mere names as a
last resort for getting rid of the Incomprehensible, the Almighty that is
all too mighty for the intellect. We name something or other the
“Absolute,” and we feel ourselves at once its superior. Philosophy,
the love of Wisdom, is at the very bottom defence against the
incomprehensible. What is named, comprehended, measured is ipso
facto overpowered, made inert and taboo.[107] Once more,
“knowledge is power.” Herein lies one root of the difference between
the idealist’s and the realist’s attitude towards the Unapproachable; it
is expressed by the two meanings of the German word Scheu—
respect and abhorrence.[108] The idealist contemplates, the realist
would subject, mechanize, render innocuous. Plato and Goethe
accept the secret in humility, Aristotle and Kant would open it up and
destroy it. The most deeply significant example of this realism is in
its treatment of the Time problem. The dread mystery of Time, life
itself, must be spellbound and, by the magic of comprehensibility,
neutralized.
All that has been said about time in “scientific” philosophy,
psychology and physics—the supposed answer to a question that
had better never have been asked, namely what is time?—touches,
not at any point the secret itself, but only a spatially-formed
representative phantom. The livingness and directedness and fated
course of real Time is replaced by a figure which, be it never so
intimately absorbed, is only a line, measurable, divisible, reversible,
and not a portrait of that which is incapable of being portrayed; by a
“time” that can be mathematically expressed in such forms as √t, t², -
t, from which the assumption of a time of zero magnitude or of
negative times is, to say the least, not excluded.[109] Obviously this is
something quite outside the domain of Life, Destiny, and living
historical Time; it is a purely conceptual time-system that is remote
even from the sensuous life. One has only to substitute, in any
philosophical or physical treatise that one pleases, this word
“Destiny” for the word “time” and one will instantly see how
understanding loses its way when language has emancipated it from
sensation, and how impossible the group “time and space” is. What
is not experienced and felt, what is merely thought, necessarily takes
a spatial form, and this explains why no systematic philosopher has
been able to make anything out of the mystery-clouded, far-echoing
sound symbols “Past” and “Future.” In Kant’s utterances concerning
time they do not even occur, and in fact one cannot see any relation
which could connect them with what is said there. But only this
spatial form enables time and space to be brought into functional
interdependence as magnitudes of the same order, as four-
dimensional vector analysis[110] conspicuously shows. As early as
1813 Lagrange frankly described mechanics as a four-dimensional
geometry, and even Newton’s cautious conception of “tempus
absolutum sive duratio” is not exempt from this intellectually
inevitable transformation of the living into mere extension. In the
older philosophy I have found one, and only one, profound and
reverent presentation of Time; it is in Augustine—“If no one
questions me, I know: if I would explain to a questioner, I know
not.”[111]
When philosophers of the present-day West “hedge”—as they all
do—by saying that things are in time as in space and that “outside”
them nothing is “conceivable,” they are merely putting another kind
of space (Räumlichkeit) beside the ordinary one, just as one might, if
one chose, call hope and electricity the two forces of the universe. It
ought not, surely, to have escaped Kant when he spoke of the “two
forms” of perception, that whereas it is easy enough to come to a
scientific understanding about space (though not to “explain” it, in the
ordinary sense of the word, for that is beyond human powers),
treatment of time on the same lines breaks down utterly. The reader
of the “Critique of Pure Reason” and the “Prolegomena” will observe
that Kant gives a well-considered proof for the connexion of space
and geometry but carefully avoids doing the same for time and
arithmetic. There he did not go beyond enunciation, and constant
reassertion of analogy between the two conceptions lured him over a
gap that would have been fatal to his system. Vis-à-vis the Where
and the How, the When forms a world of its own as distinct as is
metaphysics from physics. Space, object, number, notion, causality
are so intimately akin that it is impossible—as countless mistaken
systems prove—to treat the one independently of the other.
Mechanics is a copy of the logic of its day and vice versa. The
picture of thought as psychology builds it up and the picture of the
space-world as contemporary physics describes it are reflections of
one another. Conceptions and things, reasons and causes,
conclusions and processes coincide so nicely, as received by the
consciousness, that the abstract thinker himself has again and again
succumbed to the temptation of setting forth the thought-“process”
graphically and schematically—witness Aristotle’s and Kant’s
tabulated categories. “Where there is no scheme, there is no
philosophy” is the objection of principle—unacknowledged though it
may be—that all professional philosophers have against the
“intuitives,” to whom inwardly they feel themselves far superior. That
is why Kant crossly describes the Platonic style of thinking “as the art
of spending good words in babble” (die Kunst, wortreich zu
schwatzen), and why even to-day the lecture-room philosopher has
not a word to say about Goethe’s philosophy. Every logical operation
is capable of being drawn, every system a geometrical method of
handling thoughts. And therefore Time either finds no place in the
system at all, or is made its victim.
This is the refutation of that widely-spread misunderstanding which
connects time with arithmetic and space with geometry by superficial
analogies, an error to which Kant ought never to have succumbed—
though it is hardly surprising that Schopenhauer, with his incapacity
for understanding mathematics, did so. Because the living act of
numbering is somehow or other related to time, number and time are
constantly confused. But numbering is not number, any more than
drawing is a drawing. Numbering and drawing are a becoming,
numbers and figures are things become. Kant and the rest have in
mind now the living act (numbering) and now the result thereof (the
relations of the finished figure); but the one belongs to the domain of
Life and Time, the other to that of Extension and Causality. That I
calculate is the business of organic, what I calculate the business of
inorganic, logic. Mathematics as a whole—in common language,
arithmetic and geometry—answers the How? and the What?—that
is, the problem of the Natural order of things. In opposition to this
problem stands that of the When? of things, the specifically historical
problem of destiny, future and past; and all these things are
comprised in the word Chronology, which simple mankind
understands fully and unequivocally.
Between arithmetic and geometry there is no opposition.[112] Every
kind of number, as has been sufficiently shown in an earlier chapter,
belongs entirely to the realm of the extended and the become,
whether as a Euclidean magnitude or as an analytical function; and
to which heading should we have to assign the cyclometric[113]
functions, the Binomial Theorem, the Riemann surfaces, the Theory
of Groups? Kant’s scheme was refuted by Euler and d’Alembert
before he even set it up, and only the unfamiliarity of his successors
with the mathematics of their time—what a contrast to Descartes,
Pascal and Leibniz, who evolved the mathematics of their time from
the depths of their own philosophy!—made it possible for
mathematical notions of a relation between time and arithmetic to be
passed on like an heirloom, almost uncriticized.
But between Becoming and any part whatsoever of mathematics
there is not the slightest contact. Newton indeed was profoundly
convinced (and he was no mean philosopher) that in the principles of
his Calculus of Fluxions[114] he had grasped the problem of
Becoming, and therefore of Time—in a far subtler form, by the way,
than Kant’s. But even Newton’s view could not be upheld, even
though it may find advocates to this day. Since Weierstrass proved
that continuous functions exist which either cannot be differentiated
at all or are capable only of partial differentiation, this most deep-
searching of all efforts to close with the Time-problem
mathematically has been abandoned.
 III
Time is a counter-conception (Gegenbegriff) to Space, arising out
of Space, just as the notion (as distinct from the fact) of Life arises
only in opposition to thought, and the notion (as distinct from the
fact) of birth and generation only in opposition to death.[115] This is
implicit in the very essence of all awareness. Just as any sense-
impression is only remarked when it detaches itself from another, so
any kind of understanding that is genuine critical activity[116] is only
made possible through the setting-up of a new concept as anti-pole
to one already present, or through the divorce (if we may call it so) of
a pair of inwardly-polar concepts which as long as they are mere
constituents, possess no actuality.[117] It has long been presumed—
and rightly, beyond a doubt—that all root-words, whether they
express things or properties, have come into being by pairs; but
even later, even to-day, the connotation that every new word
receives is a reflection of some other. And so, guided by language,
the understanding, incapable of fitting a sure inward subjective
certainty of Destiny into its form-world, created “time” out of space as
its opposite. But for this we should possess neither the word nor its
connotation. And so far is this process of word-formation carried that
the particular style of extension possessed by the Classical world led
to a specifically Classical notion of time, differing from the time-
notions of India, China and the West exactly as Classical space
differs from the space of these Cultures.[118]
For this reason, the notion of an art-form—which again is a
“counter-concept”—has only arisen when men became aware that
their art-creations had a connotation (Gehalt) at all, that is, when the
expression-language of the art, along with its effects, had ceased to
be something perfectly natural and taken-for-granted, as it still was in
the time of the Pyramid-Builders, in that of the Mycenæan
strongholds and in that of the early Gothic cathedrals. Men become
suddenly aware of the existence of “works,” and then for the first
time the understanding eye is able to distinguish a causal side and a
destiny side in every living art.
In every work that displays the whole man and the whole meaning
of the existence, fear and longing lie close together, but they are and
they remain different. To the fear, to the Causal, belongs the whole
“taboo” side of art—its stock of motives, developed in strict schools
and long craft-training, carefully protected and piously transmitted; all
of it that is comprehensible, learnable, numerical; all the logic of
colour, line, structure, order, which constitutes the mother-tongue of
every worthy artist and every great epoch. But the other side,
opposed to the “taboo” as the directed is to the extended and as the
development-destiny within a form-language to its syllogisms, comes
out in genius (namely, in that which is wholly personal to the
individual artists, their imaginative powers, creative passion, depth
and richness, as against all mere mastery of form) and, beyond even
genius, in that superabundance of creativeness in the race which
conditions the rise and fall of whole arts. This is the “totem” side, and
owing to it—notwithstanding all the æsthetics ever penned—there is
no timeless and solely-true way of art, but only a history of art,
marked like everything that lives with the sign of irreversibility.[119]
And this is why architecture of the grand style—which is the only
one of the arts that handles the alien and fear-instilling itself, the
immediate Extended, the stone—is naturally the early art in all
Cultures, and only step by step yields its primacy to the special arts
of the city with their more mundane forms—the statue, the picture,
the musical composition. Of all the great artists of the West, it was
probably Michelangelo who suffered most acutely under the constant
nightmare of world-fear, and it was he also who, alone among the
Renaissance masters, never freed himself from the architectural. He
even painted as though his surfaces were stone, become, stiff,
hateful. His work was a bitter wrestle with the powers of the cosmos
which faced him and challenged him in the form of material, whereas
in the yearning Leonardo’s colour we see, as it were, a glad
materialization of the spiritual. But in every large architectural
problem an implacable causal logic, not to say mathematic, comes to
expression—in the Classical orders of columns a Euclidean relation
of beam and load, in the “analytically” disposed thrust-system of
Gothic vaulting the dynamic relation of force and mass. Cottage-
building traditions—which are to be traced in the one and in the
other, which are the necessary background even of Egyptian
architecture, which in fact develop in every early period and are
regularly lost in every later—contain the whole sum of this logic of
the extended. But the symbolism of direction and destiny is beyond
all the “technique” of the great arts and hardly approachable by way
of æsthetics. It lies—to take some instances—in the contrast that is
always felt (but never, either by Lessing or by Hebbel, elucidated)
between Classical and Western tragedy; in the succession of scenes
of old Egyptian relief and generally in the serial arrangement of
Egyptian statues, sphinxes, temple-halls; in the choice, as distinct
from the treatment, of materials (hardest diorite to affirm, and softest
wood to deny, the future); in the occurrence, and not in the grammar,
of the individual arts, e.g., the victory of arabesque over the Early
Christian picture, the retreat of oil-painting before chamber music in
the Baroque; in the utter diversity of intention in Egyptian, Chinese
and Classical statuary. All these are not matters of “can” but of
“must,” and therefore it is not mathematics and abstract thought, but
the great arts in their kinship with the contemporary religions, that
give the key to the problem of Time, a problem that can hardly be
solved within the domain of history[120] alone.
 IV
It follows from the meaning that we have attached to the Culture
as a prime phenomenon and to destiny as the organic logic of
existence, that each Culture must necessarily possess its own
destiny-idea. Indeed, this conclusion is implicit from the first in the
feeling that every great Culture is nothing but the actualizing and
form of a single, singularly-constituted (einzigartig) soul. And what
cannot be felt by one sort of men exactly as it is felt by another
(since the life of each is the expression of the idea proper to himself)
and still less transcribed, what is named by us “conjuncture,”
“accident,” “Providence” or “Fate,” by Classical man “Nemesis,”
“Ananke,” “Tyche” or “Fatum,” by the Arab “Kismet,” by everyone in
some way of his own, is just that of which each unique and
unreproduceable soul-constitution, quite clear to those who share in
it, is a rendering.
The Classical form of the Destiny-idea I shall venture to call
Euclidean. Thus it is the sense-actual person of Œdipus, his
“empirical ego,” nay, his σῶμα that is hunted and thrown by Destiny.
Œdipus complains that Creon has misused his “body”[121] and that
the oracle applied to his “body.”[122] Æschylus, again, speaks of
Agamemnon as the “royal body, leader of fleets.”[123] It is this same
word σῶμα that the mathematicians employ more than once for the
“bodies” with which they deal. But the destiny of King Lear is of the
“analytical” type—to use here also the term suggested by the
corresponding number-world—and consists in dark inner
relationships. The idea of fatherhood emerges; spiritual threads
weave themselves into the action, incorporeal and transcendental,
and are weirdly illuminated by the counterpoint of the secondary
tragedy of Gloster’s house. Lear is at the last a mere name, the axis
of something unbounded. This conception of destiny is the
“infinitesimal” conception. It stretches out into infinite time and infinite
space. It touches the bodily, Euclidean existence not at all, but
affects only the Soul. Consider the mad King between the fool and
the outcast in the storm on the heath, and then look at the Laocoön
group; the first is the Faustian, the other the Apollinian way of
suffering. Sophocles, too, wrote a Laocoön drama; and we may be
certain that there was nothing of pure soul-agony in it. Antigone goes
below ground in the body, because she has buried her brother’s
body. Think of Ajax and Philoctetes, and then of the Prince of
Homburg and Goethe’s Tasso—is not the difference between
magnitude and relation traceable right into the depths of artistic
creation?
This brings us to another connexion of high symbolic significance.
The drama of the West is ordinarily designated Character-Drama.
That of the Greeks, on the other hand, is best described as
Situation-Drama, and in the antithesis we can perceive what it is that
Western, and what it is that Classical, man respectively feel as the
basic life-form that is imperilled by the onsets of tragedy and fate. If
in lieu of “direction” we say “irreversibility,” if we let ourselves sink
into the terrible meaning of those words “too late” wherewith we
resign a fleeting bit of the present to the eternal past, we find the
deep foundation of every tragic crisis. It is Time that is the tragic, and
it is by the meaning that it intuitively attaches to Time that one
Culture is differentiated from another; and consequently “tragedy” of
the grand order has only developed in the Culture which has most
passionately affirmed, and in that which has most passionately
denied, Time. The sentiment of the ahistoric soul gives us a
Classical tragedy of the moment, and that of the ultrahistorical soul
puts before us Western tragedy that deals with the development of a
whole life. Our tragedy arises from the feeling of an inexorable Logic
of becoming, while the Greek feels the illogical, blind Casual of the
moment—the life of Lear matures inwardly towards a catastrophe,
and that of Œdipus stumbles without warning upon a situation. And
now one may perceive how it is that synchronously with Western
drama there rose and fell a mighty portrait-art (culminating in
Rembrandt), a kind of historical and biographical art which (because
it was so) was sternly discountenanced in Classical Greece at the
apogee of Attic drama. Consider the veto on likeness-statuary in
votive offerings[124] and note how—from Demetrius of Alopeke (about
400)[125]—a timid art of “ideal” portraiture began to venture forth
when, and only when, grand tragedy had been thrown into the
background by the light society-pieces of the “Middle Comedy.”[126]
Fundamentally all Greek statues were standard masks, like the
actors in the theatre of Dionysus; all bring to expression, in
significantly strict form, somatic attitudes and positions.
Physiognomically they are dumb, corporeal and of necessity nude—
character-heads of definite individuals came only with the Hellenistic
age. Once more we are reminded of the contrast between the Greek
number-world, with its computations of tangible results, and the
other, our own, in which the relations between groups of functions or
equations or, generally, formula-elements of the same order are
investigated morphologically, and the character of these relations
fixed as such in express laws.
 V
In the capacity of experientially living history and the way in which
history, particularly the history of personal becoming, is lived, one
man differs very greatly from another.
Every Culture possesses a wholly individual way of looking at and
comprehending the world-as-Nature; or (what comes to the same
thing) it has its own peculiar “Nature” which no other sort of man can
possess in exactly the same form. But in a far greater degree still,
every Culture—including the individuals comprising it (who are
separated only by minor distinctions)—possesses a specific and
peculiar sort of history—and it is in the picture of this and the style of
this that the general and the personal, the inner and the outer, the
world-historical and the biographical becoming, are immediately
perceived, felt and lived. Thus the autobiographical tendency of
Western man—revealed even in Gothic times in the symbol of
auricular confession[127]—is utterly alien to Classical man; while his
intense historical awareness is in complete contrast to the almost
dreamy unconsciousness of the Indian. And when Magian man—
primitive Christian or ripe scholar of Islam—uses the words “world-
history,” what is it that he sees before him?
But it is difficult enough to form an exact idea even of the “Nature”
proper to another kind of man, although in this domain things
specifically cognizable are causally ordered and unified in a
communicable system. And it is quite impossible for us to penetrate
completely a historical world-aspect of “becoming” formed by a soul
that is quite differently constituted from our own. Here there must
always be an intractable residue, greater or smaller in proportion to
our historical instinct, physiognomic tact and knowledge of men. All
the same, the solution of this very problem is the condition-precedent
of all really deep understanding of the world. The historical
environment of another is a part of his essence, and no such other
can be understood without the knowledge of his time-sense, his
destiny-idea and the style and degree of acuity of his inner life. In so
far therefore as these things are not directly confessed, we have to
extract them from the symbolism of the alien Culture. And as it is
thus and only thus that we can approach the incomprehensible, the
style of an alien Culture, and the great time-symbols belonging
thereto acquire an immeasurable importance.
As an example of these hitherto almost uncomprehended signs we
may take the clock, a creation of highly developed Cultures that
becomes more and more mysterious as one examines it. Classical
man managed to do without the clock, and his abstention was more
or less deliberate. To the Augustan period, and far beyond it, the
time of day was estimated by the length of one’s shadow,[128]
although sun-dials and water-clocks, designed in conformity with a
strict time-reckoning and imposed by a deep sense of past and
future, had been in regular use in both the older Cultures of Egypt
and Babylonia.[129] Classical man’s existence—Euclidean,
relationless, point-formed—was wholly contained in the instant.
Nothing must remind him of past or future. For the true Classical,
archæology did not exist, nor did its spiritual inversion, astrology.
The Oracle and the Sibyl, like the Etruscan-Roman “haruspices” and
“augurs,” did not foretell any distant future but merely gave
indications on particular questions of immediate bearing. No time-
reckoning entered intimately into everyday life (for the Olympiad
sequence was a mere literary expedient) and what really matters is
not the goodness or badness of a calendar but the questions: “who
uses it?” and “does the life of the nation run by it?” In Classical cities
nothing suggested duration, or old times or times to come—there
was no pious preservation of ruins, no work conceived for the benefit
of future generations; in them we do not find that durable[130] material
was deliberately chosen. The Dorian Greek ignored the Mycenæan
stone-technique and built in wood or clay, though Mycenæan and
Egyptian work was before him and the country produced first-class
building-stone. The Doric style is a timber style—even in
Pausanias’s day some wooden columns still lingered in the Heræum
of Olympia. The real organ of history is “memory” in the sense which
is always postulated in this book, viz., that which preserves as a
constant present the image of one’s personal past and of a national
and a world-historical past[131] as well, and is conscious of the course
both of personal and of super-personal becoming. That organ was
not present in the make-up of a Classical soul. There was no “Time”
in it. Immediately behind his proper present, the Classical historian
sees a background that is already destitute of temporal and therefore
of inward order. For Thucydides the Persian Wars, for Tacitus the
agitation of the Gracchi, were already in this vague background;[132]
and the great families of Rome had traditions that were pure
romance—witness Cæsar’s slayer, Brutus, with his firm belief in his
reputed tyrannicide ancestor. Cæsar’s reform of the calendar may
almost be regarded as a deed of emancipation from the Classical
life-feeling. But it must not be forgotten that Cæsar also imagined a
renunciation of Rome and a transformation of the City-State into an
empire which was to be dynastic—marked with the badge of duration
—and to have its centre of gravity in Alexandria, which in fact is the
birthplace of his calendar. His assassination seems to us a last
outburst of the antiduration feeling that was incarnate in the Polis
and the Urbs Roma.
Even then Classical mankind was still living every hour and every
day for itself; and this is equally true whether we take the individual
Greek or Roman, or the city, or the nation, or the whole Culture. The
hot-blooded pageantry, palace-orgies, circus-battles of Nero or
Caligula—Tacitus is a true Roman in describing only these and
ignoring the smooth progress of life in the distant provinces—are
final and flamboyant expressions of the Euclidean world-feeling that
deified the body and the present.
The Indians also have no sort of time-reckoning (the absence of it
in their case expressing their Nirvana) and no clocks, and therefore
no history, no life memories, no care. What the conspicuously
historical West calls “Indian history” achieved itself without the
smallest consciousness of what it was doing.[133] The millennium of
the Indian Culture between the Vedas and Buddha seems like the
stirrings of a sleeper; here life was actually a dream. From all this our
Western Culture is unimaginably remote. And, indeed, man has
never—not even in the “contemporary” China of the Chóu period
with its highly-developed sense of eras and epochs[134]—been so
awake and aware, so deeply sensible of time and conscious of
direction and fate and movement as he has been in the West.
Western history was willed and Indian history happened. In Classical
existence years, in Indian centuries scarcely counted, but here the
hour, the minute, yea the second, is of importance. Of the tragic
tension of a historical crisis like that of August, 1914, when even
moments seem overpowering, neither a Greek nor an Indian could
have had any idea.[135] Such crises, too, a deep-feeling man of the
West can experience within himself, as a true Greek could never do.
Over our country-side, day and night from thousands of belfries, ring
the bells[136] that join future to past and fuse the point-moments of the
Classical present into a grand relation. The epoch which marks the
birth of our Culture—the time of the Saxon Emperors—marks also
the discovery of the wheel-clock.[137] Without exact time-
measurement, without a chronology of becoming to correspond with
his imperative need of archæology (the preservation, excavation and
collection of things-become), Western man is unthinkable. The
Baroque age intensified the Gothic symbol of the belfry to the point
of grotesqueness, and produced the pocket watch that constantly
accompanies the individual.[138]
Another symbol, as deeply significant and as little understood as
the symbol of the clock, is that of the funeral customs which all great
Cultures have consecrated by ritual and by art. The grand style in
India begins with tomb-temples, in the Classical world with funerary
urns, in Egypt with pyramids, in early Christianity with catacombs
and sarcophagi. In the dawn, innumerable equally-possible forms
still cross one another chaotically and obscurely, dependent on clan-
custom and external necessities and conveniences. But every
Culture promptly elevates one or another of them to the highest
degree of symbolism. Classical man, obedient to his deep
unconscious life-feeling, picked upon burning, an act of annihilation
in which the Euclidean, the here-and-now, type of existence was
powerfully expressed. He willed to have no history, no duration,
neither past nor future, neither preservation nor dissolution, and
therefore he destroyed that which no longer possessed a present,
the body of a Pericles, a Cæsar, a Sophocles, a Phidias. And the
soul passed to join the vague crowd to which the living members of
the clan paid (but soon ceased to pay) the homage of ancestor-
worship and soul-feast, and which in its formlessness presents an
utter contrast to the ancestor-series, the genealogical tree, that is
eternalized with all the marks of historical order in the family-vault of
the West. In this (with one striking exception, the Vedic dawn in
India) no other Culture parallels the Classical.[139] And be it noted that
the Doric-Homeric spring, and above all the “Iliad,” invested this act
of burning with all the vivid feeling of a new-born symbol; for those
very warriors whose deeds probably formed the nucleus of the epic
were in fact buried almost in the Egyptian manner in the graves of
Mycenæ, Tiryns, Orchomenos and other places. And when in
Imperial times the sarcophagus or “flesh-consumer”[140] began to
supersede the vase of ashes, it was again, as in the time when the
Homeric urn superseded the shaft-grave of Mycenæ, a changed
sense of Time that underlay the change of rite.
The Egyptians, who preserved their past in memorials of stone
and hieroglyph so purposefully that we, four thousand years after
them, can determine the order of their kings’ reigns, so thoroughly
eternalized their bodies that today the great Pharaohs lie in our
museums, recognizable in every lineament, a symbol of grim triumph
—while of Dorian kings not even the names have survived. For our
own part, we know the exact birthdays and deathdays of almost
every great man since Dante, and, moreover, we see nothing
strange in the fact. Yet in the time of Aristotle, the very zenith of
Classical education, it was no longer known with certainty if
Leucippus, the founder of Atomism and a contemporary of Pericles
—i.e., hardly a century before—had ever existed at all; much as
though for us the existence of Giordano Bruno was a matter of
doubt[141] and the Renaissance had become pure saga.
And these museums themselves, in which we assemble
everything that is left of the corporeally-sensible past! Are not they a
symbol of the highest rank? Are they not intended to conserve in
mummy the entire “body” of cultural development?
As we collect countless data in milliards of printed books, do we
not also collect all the works of all the dead Cultures in these myriad
halls of West-European cities, in the mass of the collection depriving
each individual piece of that instant of actualized purpose that is its
own—the one property that the Classical soul would have respected
—and ipso facto dissolving it into our unending and unresting Time?