LINGUIST List 23.4022|
Fri Sep 28 2012
Review: Semantics; Syntax: Croft (2012)
Editor for this issue: Anja Wanner
From: Peter Arkadiev <alpgurevgmail.com>
E-mail this message to a friend
Discuss this message
Announced at http://linguistlist.org/issues/23/23-1788.html
AUTHOR: William Croft
SUBTITLE: Aspect and Causal Structure
PUBLISHER: Oxford University Press
Peter M. Arkadiev, Institute of Slavic Studies, Russian Academy of Sciences,
The new book by William Croft has been awaited by the linguistic community for
more than a decade, some preliminary chapters of it having appeared on the
author's website as early as in 2000. The book itself stems from and hinges upon
William Croft's work on argument structure and verbal semantics dating back to the
early 1980s. In this book Croft summarizes his ideas about the structure of events
and argument expression, which have been familiar to the linguistic audience at
least since Croft 1991 and have been further developed in his later publications such
as Croft 1998, and presents a fully-fledged general theory of event structure.
However, the book under review is by no means just an elaboration and summary of
older ideas, but contains a detailed and coherent presentation of a largely novel and
promising theoretical framework coupled with an insightful analysis of a rich body of
data (mainly from English), as well as an illuminating critical discussion of many of
the existing approaches to event structure and argument realization. Though the
conceptual basis of the book is shaped by the "functional" cognitive linguistic trend
of thought (in particular, Construction Grammar), Croft bases many of his insights
and proposals on the results achieved in the "formal" grammatical frameworks, and
especially acknowledges the impact of Levin & Rappaport Hovav (2005) on the
development of his ideas (p. xiii). The book is dedicated to the memory of Melissa
The main body of the book consists of nine chapters. In the first chapter
("Introduction", pp. 1-30), the main problems discussed in the book, viz. verbal
meaning and its relation to the realization of arguments and to constructions in
which verbs appear, are presented together with a brief discussion of the major
types of existing approaches to semantics, among which Croft chooses the
Cognitive Grammar approach with its geometrical diagrammatic representations.
(Here it is worth noting that graphic representations play a very important role in the
book, which necessarily means that some important aspects of Croft's theory
cannot be adequately reflected in the review.) Other issues tackled in the
introduction include the cognitive linguistic notions of frame and construal (the latter
understood as language-particular structuring of the extralinguistic experience in
semantic terms) and the problem of grammatical relations and constructions.
Following his own earlier work (e.g. Croft 2001), Croft proposes to abandon the
notion of "global" grammatical relations such as subject, object etc. in favour of
construction-specific grammatical relations, which may be different both across
languages and across different constructions within one language.
Chapters 2-4 mainly deal with aspectological issues. In chapter 2 ("The aspectual
structure of events", pp. 31-69) Croft discusses the existing approaches to the
classification of event types ("lexical aspect") and points out their empirical
problems and conceptual drawbacks. Croft revises the traditional Vendlerian
classification of event types and proposes the following more fine-grained system
(p. 45 and section 2.4.1):
a. Four types of states: inherent (permanent) states ("be Polish"), acquired
permanent states ("be cracked"), transitory states ("be ill"), and point states ("it is 5
b. Two types of activities: directed activities (for "degree achievements" such as "to
cool") vs. undirected activities ("to walk").
c. Two types of achievements: reversible achievements ("the door opened twice")
vs. irreversible achievements ("the mouse died (*twice)").
d. Two types of accomplishments: incremental accomplishments ("to write a letter")
vs. non-incremental accomplishments ("to repair a computer").
e. Cyclic achievements (semelfactives, "to cough").
As a formal framework for the analysis of aspectual types Croft proposes a two-
dimensional geometric representation involving the temporal and the qualitative axes
and modeling the presence and type of qualitative change as it occurs in time.
Alternative aspectual construals of predicates are represented as combinations of
the aspectual contour of the predicate with the aspectual profiling imposed by the
tense-aspect constructions (discussed in chapters 3 and 4). Thus the English
"inceptive state" verb "see" implies both a transitory state of seeing profiled by the
present tense and the punctual event of entering into this state profiled by the
simple past. The two-dimensional geometric representation allows Croft to motivate
his typology of aspectual types, in particular to link three types of states (acquired
permanent, transitory, and point states) to three types of achievements (denoting
entry into these states, viz. irreversible, reversible and cyclic achievements,
respectively) and two types of activities (directed and undirected) to two types of
accomplishments (incremental and non-incremental, respectively).
In chapter 3 ("Change, boundedness, and construal", pp. 70-126), various issues
pertaining to the aspectual behaviour of predicates are discussed from a cognitive
viewpoint. Croft starts with the discussion of the notion "directed change", linked to
such well-known theoretical concepts as "incremental theme" and "scale" and
forming an aspectual supercategory encompassing directed achievements, directed
activities and both incremental and non-incremental accomplishments. Further Croft
proposes to distinguish between two types of boundedness: qualitative boundedness
(q-boundedness) and temporal boundedness (t-boundedness). Q-boundedness is
inherent to the lexical semantics of the predicate and corresponds to the familiar
notion of telicity as involving a "natural endpoint" and the result state of an event. T-
boundedness implies the profiling of both the initial and the final endpoints of an
event in a particular tense-aspect construction, without indication as to whether the
result state has been achieved or even implied by the aspectual contour of the
predicate. A large part of the chapter (section 3.2) is devoted to the discussion of
the aspectual construals available to different English verbs, and notably of the
ways lexical and encyclopedic semantics of predicates affects and constrains their
aspectual potential. In this section the issue of the three existing approaches to
meaning variation (polysemy, derivation and vagueness) is raised for the first time in
the book (see below), and is resolved in favour of a cognitive usage-based approach
in which the aspectual potential of the verb depends on "asymmetries in the
frequency of use of one aspectual construal over another" (p. 91). Several
mechanisms of aspectual construal are identified, viz. aspectual selection or
metonymy "found with those predicates that allow either a directed achievement
construal or a transitory (resulting) state construal" (p. 93), "structural
schematization" found with cyclic achievements (e.g. "The light flashed") construed
as undirected activities (e.g. "The light flashed for five minutes"), and "scalar
adjustment" involving coarse-grained and fine-grained conceptualizations of the
same event found, e.g., with disposition predicates: the activity construal in "He is
being polite" is fine-grained whereas the inherent state construal in "He is polite" is
coarse-grained. This latter kind of alternative construal is applicable to various
aspectual types. Special subsections are devoted to an insightful analysis of
auxiliary and adverbial aspectual constructions in English and to a nice account of
aspectual types and aspectual construals of Russian verbs based on secondary
In chapter 4 ("The interaction of grammatical and lexical semantics: quantitative and
qualitative analyses", pp. 127-172), Croft approaches the mutual affinities and
tensions between lexical and grammatical aspect from both cross-linguistic and
language-internal perspective. A re-evaluation of the well-known typology of
temporal-aspectual categories of Dahl 1985 is based on the "multidimensional
scaling" approach to cross-linguistic data (Croft & Poole 2008) and yields some non-
trivial results, such as e.g. the existence of a typologically valid present
imperfective cluster and the lack of a sharp cross-linguistic separation between
perfective and perfect. With respect to the "alignment" of grammatical and lexical
aspect Croft's findings more or less confirm the expectation that qualitatively
unbounded situations (activities and state) would gravitate towards the imperfective
while qualitatively bounded situations more often occur with the perfective aspect.
The second part of the chapter is devoted to a detailed description of the basic
tense-aspect constructions of English (Present Tense, Simple Past Tense,
Progressive, Perfect) and "of the range of variation in aspectual potential of English
verbs across" these constructions (pp. 145-164) followed by a multidimensional
scaling analysis of the interaction of lexical and grammatical aspect in English and
Japanese (pp. 165-171). It must be said that the role the of Japanese data in this
section is not entirely clear, especially since no genuine Japanese examples are
provided. The major outcome of the analysis is the spatial model of English and
Japanese lexical aspect in fig. 4.4 on p. 166 showing a circular arrangement of
major aspectual types from transitory states to directed achievements to directed
activities to undirected activities to cyclic achievements to inactive actions and
back to transitory states. Croft concludes (p. 169-171) that "the
perfective/imperfective distinction in grammatical aspects corresponds to an
opposition" (p. 169) of aspectual types involving, respectively, transitory states and
directed achievements, on the one hand, and activities and cyclic achievements, on
the other. These apparent paradoxes are resolved under the assumption (cf.
chapters 2 and 3) that state and directed achievement result from different profiling
of the common directed aspectual contour, and that cyclic achievement and iterative
and undirected activity are different instantiations of the undirected aspectual
In chapters 5 and 6 Croft switches to the construction of a force-dynamic theory of
argument realization, expanding his own earlier proposals from Croft 1991 and 1998.
Chapter 5 ("Toward a force-dynamic theory of argument realization", pp. 173-219)
starts with a critical evaluation of some of the existing approaches to argument
realization, showing empirical and conceptual problems of theories operating with
semantic (thematic) roles and their hierarchies, as well as limitations of such event-
based accounts as Dowty's (1991) proto-role approach. Croft's own proposal hinges
upon Talmy's (1988) notion of force-dynamic relations and already mentioned Croft's
earlier work. Instead of (generalized) semantic roles and their hierarchies, realization
of arguments such as subject, object and different kinds of obliques is determined
by the causal force-dynamic structure of the event and its profiling by the verbal
predicate. The major innovation to the earlier theory proposed in the book is the
integration of causal and aspectual representations of event structure in a
tridimensional space where each participant in the event is associated with its own
subevent characterized by an aspectual contour, and force-dynamic relations link
these subevents rather than participants themselves. The unity of event is secured
by the fact that its subevents, having distinct qualitative dimensions, share a
common temporal axis.
Chapter 6 ("Causal structure in verbal semantics and argument realization", pp. 220-
282) elaborates on the theoretical postulates of the previous chapter and explores
their consequences from a cross-linguistic perspective. A large part of the chapter is
devoted to the discussion of various kinds of construal of predicate relations which
either are noncausal (spatial and possessive) or show noncanonical (cyclic or
branching) causal chains (mental events, reflexive, reciprocal and comitative
situations). Other issues approaches in this chapter include voice, alignment
(accusative, ergative and active), causative and applicative constructions, and a
diachronic typology of case syncretisms elaborating on Croft's earlier distinction
between two types of oblique semantic relations, which he calls antecedent (those
which precede the object in the causal chain, e.g. instrument) and subsequent
(those which follow the object in the causal chain, e.g. beneficiary or goal), and the
generalization that case markers in languages will not conflate relations from
different domains. A tentative conceptual space for participant roles uniting both
causal and noncausal (spatial and intentional) relations is proposed in fig. 6.2 on p.
In chapter 7 ("The interaction of aspect and causal structure in verb meaning", pp.
283-319) the force-dynamic theory of argument realization is integrated with the
theory of event structure and aspect developed in chapters 2-4. The overall
aspectual type of a complex event consisting of several subevents each with its
own aspectual contour is determined by the following principle (p. 286): "the
aspectual type of the overall event is the type of the subevent that ranks highest in
the (…)" Verbal Aspectual Hierarchy : "directed change > undirected change >
state." Intralinguistic and cross-linguistic differences in the lexicalization of complex
events are discussed, such as the well-known distinctions between result verbs and
manner verbs and between verb-framing and satellite-framing type of lexicalization,
which in Croft's view are largely dependent on the presence of the directed change
component in the predicate's semantics. The chapter contains many interesting
observations about the behaviour of different verbs and constructions in English.
The last two chapters of the book are devoted to the interaction of verbal semantics
with different constructions. In chapter 8 ("Complex predicate constructions and the
semantics of simple verbs", pp. 320-357), various kinds of constructions are
discussed which express complex events whose structure exceeds the limits
imposed by the "constraints on the semantic structure of simple verbs --
nonbranching causal chain, temporal unity, a single directed change subevent" (p.
320). Constructions discussed include the English and Japanese resultative
constructions, which receive a very detailed and insightful treatment, depictive,
serial verb and converb constructions. Croft concludes the chapter by observation
that simple verbs prototypically encode maximally individuated events, a notion
linked to the well-established transitive action prototype (Hopper & Thompson 1980).
In chapter 9 ("Verb meaning and argument structure constructions", pp. 358-393)
Croft returns to the issue of the semantic interaction of verbs and argument
structure constructions and the problem of polysemy, derivation and vagueness
approaches to this interaction. English ditransitive and locative constructions and
verbs appearing in them are analysed in great detail. Croft argues that it is not
possible to completely disentangle the semantic contributions of verbs and
constructions and arrive at their "basic" or "unitary" meanings and that the contrast
between a lexical rule analysis and a constructional analysis is a false dichotomy
(cf. Croft 2003). Instead, Croft proposes to analyze verb-specific and narrow verb-
class specific constructions fully specified for particular semantic features and draw
generalizations from them. A usage-based exemplar model of verb + construction
meaning is developed on pp. 383-392, which hinges upon token frequency of co-
occurrences of particular verbs and constructions.
In the short "Envoy" (pp. 394-395), Croft briefly summarizes the main results of his
study and urges the reader that since his argumentation was mostly based on the
data from English, the generalizations proposed must be evaluated against broad
The "Glossary of terms" (pp. 397-407) is a welcome and useful addendum.
Croft's "Verbs" is undoubtedly a very important book for all linguists interested in
aspect, event structure, argument realization and verb semantics. The book
develops a whole new theory of event structure, comprising many of the core issues
of the semantics-syntax interface, such as constraints on the lexical semantic
structure of simple verbs, linguistic situation types and predicate classes,
interaction between lexical and grammatical aspect and between verbal and
constructional semantics, thematic roles and argument realization, voice and
various complex predicates, polysemy of case markers etc. Though the
argumentation is largely based on data from English, the discussion is typologically
well-informed and in many aspects draws upon cross-linguistic generalizations
(including those made by the author himself).
The theory presented in the book is a first coherent and all-encompassing
conception of event structure and argument realization in cognitive linguistics,
couched in a usage-based constructional approach to semantics and syntax and
sophisticated enough to challenge the other existing theories of these phenomena,
especially those developed in the "formalist" tradition. Croft can only be praised for
doing justice to competing approaches and for incorporating many of their insights
into his own conception instead of simply rejecting them as "aprioristic" or
"reductionist." The book, in addition to presenting the author's own ideas, contains
detailed and useful summaries and discussions of many of the existing approaches
to the phenomena in question, where both strong and weak points of different
proposals as well as their similarities and differences with Croft's own theory are
highlighted. Finally, many pages of the book are devoted to a really illuminating
analysis of a various English data, ranging from verbal aspect to intricacies of
resultative constructions. All this makes the book a fascinating reading rewarding
both theory-oriented and empirically-oriented audiences.
However, every important contribution to science has some weak points, and Croft's
book is no exception. My major criticism concerns Croft's failure to take into
account some of the important recent (and even not so recent) proposals in the
domain of aspect and event structure, which are in some respects parallel to Croft's
In his discussion of aspectual types in chapter 2, Croft ignores the approach to
event types and aspectual classes proposed by Sergei Tatevosov (2002), as well as
Tatevosov's critical survey of various proposals in this domain. This is indeed
unfortunate, since Tatevosov has developed a cross-linguistically applicable non-
aprioristic theory of aspectual types allowing to analyze data from any human
language and to arrive at directly comparable results. In addition, several of Croft's
"new" aspectual types, such as e.g. "inceptive states", have been already
recognized by Tatevosov as "cross-linguistic actional classes" supported by data
from many languages. This is also important because Croft's own approach to
aspectual types does not seem to be conceptually or methodologically superior to
Tatevosov's; Croft does not actually explain how his methodology of arriving at
event types and aspectual classes is constrained, whether the event types and
aspectual classes he postulated for English can be extended to other languages and
how such an extension can be achieved in a non-circular and non-aprioristic way.
This leads him to saying that "there is in fact an indefinitely large number of
predicate classes each having its own unique aspectual potential or range of
possible aspectual construals" (p. 57), which is not a very desirable result, since
linguistic theory must instead constrain possible predicate classes, and this is
precisely what Tatevosov's approach does (see especially Tatevosov 2010, where
cross-linguistic hypotheses about possible aspectual classes and language-specific
systems of aspectual classes are proposed).
The second recent proposal in the domain of event structure which Croft fails to take
into account is the work by Gillian Ramchand (2008), which, though couched in the
generative syntactic framework, is in many ways parallel to Croft's functional
cognitive theory. Like Croft, Ramchand proposes an integrated theory of aspect,
event structure and argument realization, where each event participant is associated
with its own subevent represented as a node in the syntactic tree linked to a specific
semantic interpretation; the relations between different participants and subevents of
the same event in Ramchand's system are causal in nature, thus resembling Croft's
force-dynamic relations. Similarly to Croft, Ramchand is concerned with the issue of
what constrains the semantic potential of simple verbs and with such constructions
as causative, resultative etc. A critical evaluation of Ramchand's theory and its
comparison with his own would have constituted a highly relevant part of Croft's
The third major piece of pertinent work ignored by Croft (and, unfortunately, by many
of his colleagues) belongs to the Russian linguist Elena Paducheva, who has
developed a sophisticated derivational theory of event structure and aspect (basing
mainly on the Russian data), in many respects similar to that of Croft's (including
parallels in graphic representations of event types). The relevant publications in
English include Paducheva 1995, 1997, 1998, 2003 (see
http://lexicograph.ruslang.ru/03MembersPadu.htm); among the important insights
made by Paducheva is the recognition of the principled correlation between
predicate classes and semantic types of verbal arguments, cf. Croft's cursory
remarks on p. 378 of his book.
Further, in many parts of the book Croft fails to take into account and refer to the
recent important work on the problems he is dealing with. Work which should have
been considered include Carlota Smith's papers on English tense and aspect, e.g.
Smith 1978 and Smith 1986, in section 4.3 on the basic tense-aspect constructions
in English; the major typological work (e.g., Nedjalkov ed. 2007) on reciprocals and
reflexives, their semantics, polysemy, diachrony and expression across languages
in section 188.8.131.52; recent insights in the semantics of comitative constructions and
comitative relations by Alexandre Arkhipov (2009); recent developments in the
typology of the so-called "active/stative" languages (Donohue & Wichmann eds.
2008) in section 6.3.1; recent proposals concerning transitivity such as Næss 2007
and Malchukov 2006. All these lacunae are rather unfortunate; not invalidating
Croft's own proposals, they nevertheless make them weaker and less supported by
the existing body of data and literature than desired.
There are not many errors and typos in the book, and I will point out only those
which pertain to the data. On p. 121 the Russian verb 'be interested in' is
'interesovat'sja,' not 'interesovat,' and the alleged Russian verb grančit' 'cut, facet'
does not exist. On p. 193 the notation from Wunderlich 1997 is mixed up:
accusative must be specified as [+hr], not [+lr], while the ergative is assigned [+lr],
not [+hr]. In the Turkish ex. (90b) on p. 267 a wrong accusative case marker
appears. In the Finnish ex. 85 on p. 317 GEN instead of PRTT must appear in the
glosses; in addition, the abbreviations list on pp. xvi-xvii does not contain PART and
To conclude, the new book by William Croft, despite certain drawbacks and
weaknesses, is a major contribution to linguistic theory, which should be read by all
linguists interested in aspect, event structure, and argument realization, regardless
of particular theoretical frameworks they adhere to. The book is rich in ideas and
empirical data and is written in a persuasive and appealing fashion, making it a
fascinating and smooth reading.
Arkhipov, Alexandre. 2009. Comitative as a cross-linguistically valid category. In:
New Challenges in Typology: Transcending the Borders and Refining the
Distinctions, Alexandre Arkhipov and Patience Epps (eds.), 223-246. Berlin, New
York: Mouton de Gruyter.
Croft, William. 1991. Syntactic Categories and Grammatical Relations. The
Cognitive Organization of Information. Chicago & London: The University of Chicago
Croft, William. 1998. Event structure in argument linking. In: The Projection of
Arguments: Lexical and Compositional Factors, Miriam Butt and Wilhelm Geuder
(eds.), 1-43. Stanford: CSLI.
Croft, William. 2001. Radical Construction Grammar. Syntactic Theory in
Typological Perspective. Oxford: Oxford University Press.
Croft, William. 2003. Lexical rules vs. constructions: a false dichotomy. In:
Motivation in Language: Studies in Honour of Günter Radden, Hubert Cuyckens,
Thomas Berg, René Dirven and Klaus-Uwe Panther (eds.), 49-68. Amsterdam,
Philadelphia: John Benjamins.
Croft, William & Keith T. Poole. 2008. Inferring universals from grammatical
variation: multidimensional scaling for typological analysis. Theoretical Linguistics
Dahl, Östen. 1985. Tense and Aspect Systems. Oxford: Blackwell.
Donohue, Mark & Søren Wichmann (eds.). 2008. The Typology of Semantic
Alignment. Oxford: Oxford University Press.
Dowty, David R. 1991. Thematic proto-roles and argument selection. Language
Hopper, Paul J. & Sandra A. Thompson. 1980. Transitivity in grammar and
discourse. Language 56(2). 251-299.
Levin, Beth & Malka Rappaport Hovav. 2005. Argument Realization. Cambridge:
Cambridge University Press.
Malchukov, Andrei. 2006. Transitivity parameters and transitivity alternations:
Considering co-variation. In Case, Valency and Transitivity, Leonid I. Kulikov, Andrei
L. Malchukov and Helen de Hoop (eds.), 329-358. Amsterdam, Philadelphia: John
Næss, Åshild. 2007. Prototypical Transitivity. Amsterdam, Philadelphia: John
Nedjalkov, Vladimir P. (ed.) 2007. Reciprocal Constructions. Vols. 1-4. Amsterdam,
Philadelphia: John Benjamins.
Paducheva, Elena. 1995. Taxonomic categories and semantics of aspectual
opposition. In: Temporal Reference, Aspect and Actionality, Pier-Marco Bertinetto,
Valentina Bianchi, Östen Dahl, James Higginbotham and Mario Squartini (eds.), Vol.
I, 71-90. Torino: Rosenberg & Sellier.
Paducheva, Elena. 1997. Verb categorization and the format of a lexicographic
definition. In: Recent Trends in Meaning-Text Theory, Leo Wanner (ed.), 61-74.
Amsterdam, Philadelphia: John Benjamins.
Paducheva, Elena. 1998. Thematic roles and the quest for semantic invariants of
lexical derivation. Folia Linguistica 31(3-4). 349-363.
Paducheva, Elena. 2003. Lexical meaning and semantic derivation: the case of
image creation verbs. In: Second International Workshop on Generative Approaches
to the Lexicon. May 15-17, 2003, Geneva, 230-237.
Ramchand, Gillian C. 2008. Verb Meaning and the Lexicon. A First-Phase Syntax.
Cambridge: Cambridge University Press.
Smith, Carlota. 1978. The syntax and interpretation of temporal expressions in
English. Linguistics and Philosophy 2(1). 43-99.
Smith, Carlota. 1986. A speaker-based approach to aspect. Linguistics and
Philosophy 9(1). 97-115.
Talmy, Leonard. 1988. Force dynamics in language and cognition. Cognitive
Science 12. 49-100.
Tatevosov, Sergei. 2002. The parameter of actionality. Linguistic Typology 6(3). 317
Tatevosov, Sergei. 2010. Akcional'nost' v leksike i grammatike [Actionality in
Lexicon and Grammar.] Unpublished Habilitation Thesis, Moscow State University.
Wunderlich, Dieter. 1997. Cause and the structure of verbs. Linguistic Inquiry 28(1).
ABOUT THE REVIEWER
Peter M. Arkadiev, PhD in linguistics (2006), is a senior research fellow
in the Department of Typology and Comparative Linguistics of the
Institute of Slavic studies of the Russian Academy of Sciences and an
assistant professor at the Centre for Linguistic Typology of the Institute
of Linguistics of the Russian State University for the Humanities,
Moscow. His main interests are linguistic typology with a focus on case
marking and argument structure and its formal realization, and tense-
aspect-modality. He works mainly on Lithuanian and Adyghe.
Read more issues|LINGUIST home page|Top of issue
Page Updated: 28-Sep-2012
While the LINGUIST List makes every effort to ensure the linguistic relevance of sites listed
on its pages, it cannot vouch for their contents.