Jump to Content Jump to Main Navigation

Linguistics Semantic Roles
by
Beth Levin

Introduction

The meaning of a predicate, especially a verb, may be characterized via the relations that its arguments bear to it. Semantic roles—also known as thematic relations, theta roles, participant roles, and deep cases—are labels for certain recurring predicate-argument relations. They have proved attractive because they provide a way of representing commonalities across different uses of the same predicate or across uses of distinct but semantically related predicates that may be obscured because arguments with certain semantic roles may have various syntactic realizations. Thus they provide a level of abstraction for the statement of generalizations concerning a variety of linguistic phenomena. In particular, argument realization generalizations are often stated over a thematic hierarchy, a ranking of semantic roles. However, semantic roles have not lived up to their initial promise. It has proved impossible to find a small set of roles that can be applied across all verbs in a language, let alone across languages. Yet this desideratum must be met if semantic roles are to figure effectively in accounts of linguistic phenomena. Further, some generalizations involving semantic roles seem to require reference to coarse-grained roles, whereas others require reference to fine-grained roles. Moreover, reliable diagnostics are difficult to identify even for the roles cited most often. Although these difficulties have led some researchers to reject semantic roles, others have taken alternative approaches, including the use of generalized semantic roles, which are inspired by the notion of prototype, with no single property being necessary or sufficient for an argument to bear such a role. Despite these drawbacks, semantic roles continue to be useful in stating linguistic generalizations, and so descriptive, if not theoretical, uses of semantic role labels persist across subfields, including language acquisition, psycholinguistics, and neurolinguistics. Furthermore, semantic roles are useful in natural language processing. Since semantic roles have been implicated in phenomena involving argument structure, the separate Oxford Bibliographies article Argument Structure should be consulted for additional relevant resources. Acknowledgments: For discussion of the material in this article, the author is grateful to Scott Grimm, Chris Manning, Malka Rappaport Hovav, and two reviewers, as well as the students in her autumn 2012 lexical semantics class.

General Works

There are a considerable number of handbook and encyclopedia articles devoted to semantic roles, and they also receive treatments in many introductory semantics textbooks, especially those aimed at undergraduates. Such textbook treatments typically present a particular inventory of semantic roles and show their applicability to the semantic representation of particular sentences; further, they usually include some discussion of the problems that face semantic roles as a semantic representation. In contrast, the handbook and encyclopedia treatments provide more sustained discussion of the notion of semantic role. They often trace the development of the notion, as well as its place within current linguistic theory. In addition, they may introduce several approaches to semantic role inventories, highlighting the similarities and differences among them, and they may discuss the limitations both of particular approaches and of semantic role approaches in general. Due to this added depth, representative handbook chapters and encyclopedia articles are the focus of this section. Bruce and Moser 1992 provides the most basic treatment of semantic roles, while Van Valin 1994 provides a more extensive introduction encompassing both traditional and generalized semantic roles. Wechsler 2006 complements these by also providing an overview of the development of semantic role approaches. Davis 2011 is the most extensive of the handbook chapters, introducing semantic roles from a formal semantic perspective. Butt 2006 introduces semantic roles in the context of a discussion of morphological case, thus drawing attention to the relation between the two notions. Levin and Rappaport Hovav 2005, a survey of argument realization, provides a detailed introduction to both traditional and generalized semantic roles, as well as a thorough discussion of thematic hierarchies. More specialized encyclopedia articles and handbook chapters are mentioned elsewhere in this article, including Fillmore’s own retrospective view on case grammar, Fillmore 2003 (cited under Case Grammar: Development). Campe 1994 includes an extensive list of references, written in several languages, on various topics that fall under the notion of “semantic role.”

  • Bruce, Bertram, and Margaret G. Moser. 1992. Grammar, case. In Encyclopedia of artificial intelligence. 2d ed. Edited by Stuart C. Shapiro, 563–570. New York: Wiley.

    Save Citation »Export Citation »E-mail Citation »

    This brief introduction, written for researchers in artificial intelligence, provides a good starting point for anyone with little previous background. It explains the motivation for semantic roles and reviews several approaches current in the late 1960s and early 1970s, including case grammar.

    Find this resource:

  • Butt, Miriam. 2006. Theories of case. Cambridge, UK: Cambridge Univ. Press.

    DOI: 10.1017/CBO9781139164696Save Citation »Export Citation »E-mail Citation »

    A textbook providing a comprehensive introduction to morphological case, which also includes a discussion of semantic roles, introduced in the work of Fillmore as deep cases (see various works of Fillmore cited under Case Grammar: Development). Reviews the place of semantic roles in several linguistic theories, especially as they figure in argument realization.

    Find this resource:

  • Campe, Petra. 1994. Case, semantic roles, and grammatical relations: A comprehensive bibliography. Amsterdam: John Benjamins.

    Save Citation »Export Citation »E-mail Citation »

    Despite its publication date, this bibliography still offers entry points to work on various approaches to semantic roles, work on specific semantic roles, and work on linguistic phenomena whose description apparently involves semantic roles.

    Find this resource:

  • Davis, Anthony R. 2011. Thematic roles. In Semantics: An international handbook of natural language meaning. Vol. 1. Edited by Claudia Maienborn, Klaus von Heusinger, and Paul H. Portner, 399–420. Berlin: Mouton de Gruyter.

    Save Citation »Export Citation »E-mail Citation »

    An introduction to semantic roles and related notions from a formal semantic perspective. Probably best appreciated by those with some background in semantics and syntax. Reviews basic issues confronting any attempt to define a set of semantic roles that figures in the explanation of linguistic phenomena, particularly argument realization.

    Find this resource:

  • Levin, Beth, and Malka Rappaport Hovav. 2005. Argument realization. Cambridge, UK: Cambridge Univ. Press.

    DOI: 10.1017/CBO9780511610479Save Citation »Export Citation »E-mail Citation »

    Chapters 2, 3, 5, and 6 provide detailed discussions of semantic roles, generalized semantic roles, the place of semantic roles in approaches to argument realization, and the thematic hierarchy.

    Find this resource:

  • Van Valin, Robert D., Jr. 1994. Functional relations. In The encyclopedia of language and linguistics. Edited by R. E. Asher, 1327–1338. Oxford: Pergamon.

    Save Citation »Export Citation »E-mail Citation »

    The first part of this introduction to the syntactic, pragmatic, and semantic relations that arguments can bear to predicates is devoted to semantic roles. It motivates and describes both traditional and generalized semantic roles and reviews the treatment of semantic roles in several linguistic theories. Reprinted in second edition of The Encyclopedia of Language and Linguistics, edited by Keith Brown (Oxford: Elsevier, 2006), 683–696.

    Find this resource:

  • Wechsler, Stephen. 2006. Thematic structure. In Encyclopedia of language and linguistics. 2d ed. Edited by Keith Brown, 645–653. Amsterdam: Elsevier.

    Save Citation »Export Citation »E-mail Citation »

    An introduction to semantic roles, as well as other forms of lexical semantic representation of verb meaning as they bear on argument realization. Provides some historical perspective on the development of the notion.

    Find this resource:

The Theoretical Notion

Given the prevalence of semantic roles in linguistic representations and descriptions, some researchers have stepped back to consider both the notion of semantic role as a theoretical construct and its place in linguistic theory. Perhaps the best-known paper along these lines is Dowty 1989, which considers the formal semantic underpinnings of semantic roles (some of the ideas in this paper are foreshadowed in Ladusaw and Dowty 1988, cited under Other Phenomena). Dowty 1989 sets the stage for the introduction of proto-roles in Dowty 1991 (cited under Proto-roles), which develops important conceptual points in Dowty 1989 further. Although Dowty 1989 takes semantic roles to arise from the lexical entailments that verbs impose on their arguments, Carlson 1998, which also takes a formal semantic perspective, proposes that semantic roles contribute to event individuation. Parsons 1990 discusses semantic roles and their significance in the context of a neo-Davidsonian event semantics. Bierwisch 2006 provides a comparative assessment of two leading perspectives on the notion of semantic role after laying out desiderata for any account of this notion. Jackendoff 1987 presents a different position, suggesting that semantic roles emerge from the nature of a conceptual representation. Further discussion of the notion of semantic role can be found in Sources of Semantic Roles.

  • Bierwisch, Manfred. 2006. Thematic roles—Universal, particular, and idiosyncratic aspects. In Semantic role universals and argument linking: Theoretical, typological, and psycholinguistic perspectives. Edited by Ina Bornkessel, Matthias Schlesewsky, Bernard Comrie, and Angela D. Friederici, 89–126. Berlin: Mouton de Gruyter.

    DOI: 10.1515/9783110219272Save Citation »Export Citation »E-mail Citation »

    Lays out criteria that any account of the theoretical construct “semantic roles” must satisfy. Evaluates and compares two perspectives on semantic roles with respect to these criteria: the extrinsic perspective, which takes them to be independent notions, and the intrinsic perspective, which takes them to be defined over event structures.

    Find this resource:

  • Carlson, Greg. 1998. Thematic roles and the individuation of events. In Events and grammar. Edited by Susan Rothstein, 35–51. Dordrecht, The Netherlands: Kluwer.

    DOI: 10.1007/978-94-011-3969-4Save Citation »Export Citation »E-mail Citation »

    Taking a formal semantics perspective, this paper suggests that semantic roles are important within linguistic theory because they provide a means of individuating events. Its starting point is an examination of the common assumption that every argument of a verb is assigned a distinct semantic role.

    Find this resource:

  • Dowty, David R. 1989. On the semantic content of the notion “thematic role.” In Properties, types and meaning. Vol. 2. Edited by Gennaro Chierchia, Barbara Partee, and Raymond Turner, 69–129. Dordrecht, The Netherlands: Kluwer.

    Save Citation »Export Citation »E-mail Citation »

    Considers the formal semantic foundations of semantic roles. Introduces the notion of L-thematic role, based on the intuition that certain sets of lexical entailments recur across verbs. Discusses ordered argument and neo-Davidsonian approaches to predicate-argument associations.

    Find this resource:

  • Jackendoff, Ray S. 1987. The status of thematic relations in linguistic theory. Linguistic Inquiry 18:369–411.

    Save Citation »Export Citation »E-mail Citation »

    Considers the place of semantic roles, and of semantics more generally, in linguistic theory, taking as a point of comparison the treatment of semantic roles within the Government-Binding framework. Argues that semantic roles are defined over positions in a conceptual representation taking the form of a predicate decomposition.

    Find this resource:

  • Parsons, Terence. 1990. Thematic roles. In Events in the semantics of English. By Terence Parsons, 68–104. Cambridge, MA: MIT Press.

    Save Citation »Export Citation »E-mail Citation »

    Takes semantic roles to be relations between events and their participants within a neo-Davidsonian theory of events. Discusses recurring issues and problems relating to the notion of semantic roles in this context.

    Find this resource:

Approaches Antedating Fillmore’s “The Case for Case”

Fillmore 1968, “The Case for Case” (cited under Case Grammar: Development), is taken to be the foundational work that has inspired research on semantic roles However, semantic roles or similar notions have long figured in work in linguistics. The kārakas introduced in the work of the Sanskrit grammarian Pāṇini are often cited as the earliest antecedents of today’s semantic roles. Kiparsky and Staal 1969 provides an overview of the place of kārakas in Pāṇini’s grammar, in a larger discussion of the rules that determine the surface morphological case of each of a verb’s arguments. A briefer introduction to Pāṇini’s kārakas may be found in section 2.3 (pp. 15–18) of Butt 2006 (cited under General Works). Moving considerably forward in time, Blake 1930 recognized a notion of case relationship, which like a semantic role is distinct from its morphosyntactic instantiations. At about the same time, Hjelmslev 1935 proposed a unified, localist approach to morphological cases. Although not explicitly using semantic roles, the valency approach pioneered in Tesnière 1959 is also relevant, as it shares with work on semantic roles the goal of characterizing a verb’s argument-taking potential. Anderson 2006 reviews several centuries of work on the semantic functions of morphological case in an effort to understand current approaches to semantic roles. For further discussion see Butt 2006, cited under General Works.

  • Anderson, John M. 2006. Modern grammars of case: A retrospective. Oxford: Oxford University Press.

    DOI: 10.1093/acprof:oso/9780199297078.001.0001Save Citation »Export Citation »E-mail Citation »

    Considers the development of semantic roles in linguistic theory since the 1960s from the perspective of earlier conceptions of the function of morphological case, with an emphasis on treatments in the nineteenth century and the first half of the twentieth century.

    Find this resource:

  • Blake, Frank R. 1930. A semantic analysis of case. In Curme Volume of Linguistic Studies. Edited by James T. Hatfield, et al., 34–49. Language Monograph 7. Baltimore: Waverly.

    Save Citation »Export Citation »E-mail Citation »

    Argues for a notion of case relationship—roughly, a semantic role—that is distinct from a case form—the morphosyntactic instantiation of a case relationship, whether as a morphological case or as some other grammatical category.

    Find this resource:

  • Hjelmslev, Louis. 1935. La catégorie des cas: Étude de grammaire générale. Part 1. Acta Jutlandica 7. Aarhus, Denmark: Universitetsforlaget.

    Save Citation »Export Citation »E-mail Citation »

    Provides a unified semantics of morphological cases from a localist perspective, and thus can be viewed as setting the stage for the introduction of semantic roles in their localist instantiations.

    Find this resource:

  • Kiparsky, Paul, and J. F. Staal. 1969. Syntactic and semantic relations in Pāṇini. Foundations of Language 5:83–117.

    Save Citation »Export Citation »E-mail Citation »

    Introduces Pāṇini’s grammar, presenting the levels of representation recognized, analogizing them to levels of representation in Chomsky’s Standard Theory. Focuses on the level of the kārakas, discussing each kāraka, the rules assigning it to arguments of verbs of particular semantic types, and the rules governing its morphosyntactic realization.

    Find this resource:

  • Tesnière, Lucien. 1959. Éléments de syntaxe structurale. Paris: Klincksieck.

    Save Citation »Export Citation »E-mail Citation »

    Introduces a comprehensive theory of verb valency—basically, argument-taking potential. After distinguishing between arguments and adjuncts, classifies verbs by number of selected arguments.

    Find this resource:

Case Grammar

Fillmore’s case grammar remains a particularly influential approach to semantic roles, although some other approaches cited in this bibliography were developed at around the same time. Case grammar opened the way for exploring the use of semantic roles in the statement of linguistic generalizations, particularly in argument realization generalizations. Development reviews representative work on case grammar, while Critiques presents some well-known critiques of this approach.

Development

Fillmore 1969, first circulated in 1966, presents the motivation for positing the (deep) cases or semantic roles at the heart of case grammar. Fillmore 1968 is the seminal work setting out case grammar. Despite its use of semantic roles to associate arguments with their verb, this paper is inspired by Chomsky’s Standard Theory of syntax, adopting transformations for argument realization. Fillmore 1977a discusses reactions to case grammar and sets the stage for the frame semantics introduced in Fillmore’s later work. Fillmore 1977b proposes an alternative, nontransformational approach to subject and object selection, contrasting with Fillmore 1968. Two handbook articles, Anderson 2006 and Fillmore 2003, provide good introductions to case grammar. A thorough overview can also be found in Cook 1989, cited under Other Approaches.

  • Anderson, John M. 2006. Case grammar. In Encyclopedia of language and linguistics. 2d ed. Edited by Keith Brown, 220–233. Amsterdam: Elsevier.

    Save Citation »Export Citation »E-mail Citation »

    A detailed overview of the fundamentals of case grammar, its subsequent development, and related approaches.

    Find this resource:

  • Fillmore, Charles J. 1968. The case for case. In Universals in linguistic theory. Edited by Emmon Bach and Robert T. Harms, 1–88. New York: Holt, Rinehart, and Winston.

    Save Citation »Export Citation »E-mail Citation »

    Motivates semantic roles by showing that the semantic relations of arguments to their verb are obscured on the surface. Introduces a semantic role inventory and transformations converting semantic role representations of sentences into their surface syntax. Reprinted with a foreword by the author in Fillmore’s Form and Meaning in Language. Vol. 1, Papers on Semantic Roles (Stanford, CA: CSLI Publications, 2003), 23–122.

    Find this resource:

  • Fillmore, Charles J. 1969. Toward a modern theory of case. In Modern studies in English. Edited by David A. Reibel and Sanford A. Schane, 361–375. Englewood Cliffs, NJ: Prentice-Hall.

    Save Citation »Export Citation »E-mail Citation »

    Initially circulated in 1966, sets the stage for Fillmore 1968. Motivates the notion of case, although the term is not used, to capture the common semantic relation an argument may bear to its verb across alternate syntactic realizations. Reprinted with a foreword by the author in his Form and Meaning in Language. Vol. 1, Papers on Semantic Roles (Stanford, CA: CSLI Publications, 2003), 1–21.

    Find this resource:

  • Fillmore, Charles J. 1977a. The case for case reopened. In Grammatical relations. Edited by Peter Cole and Jerrold M. Sadock, 59–81. Syntax and Semantics 8. New York: Academic Press.

    Save Citation »Export Citation »E-mail Citation »

    Revisits the author’s earlier work on case grammar, clarifying his intentions. Confronts challenges raised by other researchers, suggesting that they might be resolved by recognizing that verb meanings are “relativized to scenes” (p. 59)—that is, verb meanings are construals of events. Reprinted with a foreword by the author in his Form and Meaning in Language. Vol. 1, Papers on Semantic Roles (Stanford, CA: CSLI Publications, 2003), 175–199.

    Find this resource:

  • Fillmore, Charles J. 1977b. Topics in lexical semantics. In Current issues in linguistic theory. Edited by Roger W. Cole, 76–138. Bloomington: Indiana Univ. Press.

    Save Citation »Export Citation »E-mail Citation »

    The second of these lectures elaborates on Fillmore 1977a, arguing subject and object are chosen via a set of ranked statements, each defining the relative salience of two event participants with respect to properties of the event or its participants. Reprinted with a foreword by the author in his Form and Meaning in Language. Vol. 1, Papers on Semantic Roles (Stanford, CA: CSLI Publications), 199–260.

    Find this resource:

  • Fillmore, Charles J. 2003. Valency and semantic roles: The concept of deep structure case. In Dependenz und Valenz: Ein internationales Handbuch der zeitgenössischen Forschung/Dependency and valency: An international handbook of contemporary research. Vol. 1. Edited by Vilmos Ágel, Ludwig M. Eichinger, Hans-Werner Eroms, Peter Hellwig, Hans Jürgen Heringer, and Henning Lobin, 457–475. Berlin: Walter de Gruyter.

    DOI: 10.1515/9783110141900.1Save Citation »Export Citation »E-mail Citation »

    Fillmore’s own retrospective overview of case grammar. Covers the historical development of the approach, its motivation, its basic claims, and its challenges. Discusses such issues as defining semantic roles, determining the best grain-size for semantic roles, and choosing a semantic role inventory. Contains a fairly lengthy bibliography.

    Find this resource:

Critiques

Since its inception, case grammar has faced numerous critiques, which draw attention both to the limitations of semantic role approaches in general and to the drawbacks of case grammar in particular, including its embedding within a transformational grammar. This subsection lists several often-cited early critiques, Dougherty 1970, Huddleston 1970, and Mellema 1974, which are directed towards “The Case for Case” (Fillmore 1968, cited under Case Grammar: Development) in particular. Chomsky 1972 argues for a deep structure syntactic representation over a case grammar representation. Anderson 1971 also presents evidence for deep structure as the locus of certain interpretive semantic phenomena, concomitantly highlighting a problem for case grammar. In fact, Fillmore 1977a (cited under Case Grammar: Development specifically mentions (p. 93) that Anderson 1971 affected Fillmore’s thinking on the nature of semantic roles. There are also numerous critiques of semantic roles that are not directed specifically at case grammar. Dowty 1991 contains an influential and often-cited critique of semantic roles, which particularly effectively presents the key difficulties facing attempts to identify viable semantic roles, while DeLancey 1991 points to a methodological misunderstanding underlying many problems of semantic role assignment. For further critiques see the references cited under General Works; see also Thematic Relations for critiques of thematic relations and Thematic Hierarchies for critiques of thematic hierarchies.

  • Anderson, Stephen R. 1971. On the role of deep structure in semantic interpretation. Foundations of Language 7:387–396.

    Save Citation »Export Citation »E-mail Citation »

    Shows that the holistic/partitive effect found with locative alternation verbs and related effects are best stated in terms of deep grammatical relations and cannot be stated in terms of case grammar.

    Find this resource:

  • Chomsky, Noam. 1972. Some empirical issues in the theory of transformational grammar. In Goals of linguistic theory. Edited by Stanley Peters, 63–130. Englewood Cliffs, NJ: Prentice-Hall.

    Save Citation »Export Citation »E-mail Citation »

    Section 6.8 (pp. 98–105) provides a critique of case grammar, arguing instead for the use of deep structure representations to handle some phenomena that case grammar proposed handling with semantic roles. Reprinted in Chomsky’s Studies on Semantics in Generative Grammar (The Hague: Mouton, 1972), 120–202.

    Find this resource:

  • DeLancey, Scott. 1991. Event construal and case role assignment. In Proceedings of the Seventeenth Annual Meeting of the Berkeley Linguistics Society, February 15–18, 1991. Edited by Laurel A. Sutton, Christopher Johnson, and Ruth Shields, 338–353. Berkeley: Berkeley Linguistics Society, Univ. of California.

    Save Citation »Export Citation »E-mail Citation »

    Argues that many problems that have plagued efforts to assign semantic roles arise due to a failure to recognize that semantic role assignments should encode construals of events rather than facts about actual events in the world. Revisits the notion of agent and some well-known examples in this context.

    Find this resource:

  • Dougherty, Ray C. 1970. Recent studies on language universals. Foundations of Language 6:505–561.

    Save Citation »Export Citation »E-mail Citation »

    Part I (pp. 506–531) is a much-cited, detailed critique of Fillmore 1968 (cited under Case Grammar: Development), which raises some challenges that have been repeatedly leveled against semantic roles. Includes considerable discussion of issues arising in Fillmore’s proposed derivations of surface structures from underlying deep structures, which embody case grammar analyses.

    Find this resource:

  • Dowty, David R. 1991. Thematic proto-roles and argument selection. Language 67:547–619.

    Save Citation »Export Citation »E-mail Citation »

    Section 3 (pp. 553–559) contains a detailed critique of semantic roles, which particularly effectively lays out often-cited problems of role fragmentation and unclear boundaries together with illustrative examples.

    Find this resource:

  • Huddleston, Rodney. 1970. Some remarks on case grammar. Linguistic Inquiry 1:501–511.

    Save Citation »Export Citation »E-mail Citation »

    An early critique of case grammar that raises issues that reappear in subsequent critiques.

    Find this resource:

  • Mellema, Paul. 1974. A brief against case grammar. Foundations of Language 11:39–76.

    Save Citation »Export Citation »E-mail Citation »

    A lengthy, often-cited critique of case grammar, which also argues that the phenomena Fillmore discusses could be given an equally perspicuous analysis within Chomsky’s Standard Theory.

    Find this resource:

Thematic Relations

The set of thematic relations introduced in Gruber 1965 offers an alternative to Fillmore’s deep cases. Thematic relations did not receive widespread recognition, particularly within work in generative grammar, until their applicability to the explanation of a range of linguistic phenomena was demonstrated in Jackendoff 1972, although Hust and Brame 1976 provides a critique of these applications. Thematic relations are motivated by taking a localist perspective on the semantic structure of a clause, which takes motion and location events as primary and analogizes all other events to such events. General arguments for localism are found in Gruber 1965, while a case study of a localist analysis of perception verbs is presented in Gruber 1967. Gruber 2000 extends the author’s localist analysis to a wider range of data and deals with argument realization. Jackendoff 1983 takes thematic relations in a somewhat different direction, setting them in the context of a theory of conceptual structure, and emphasizing their grounding in positions in the predicate decompositions that represent such structures. Jackendoff 1987 proposes that conceptual structures should include an “action tier” to cope with the notions of agent and patient. A later monograph, Jackendoff 1990, presents conceptual structure analyses of a considerable range of data, as well as delineating how argument realization can be formulated over conceptual structures.

  • Gruber, Jeffrey S. 1965. Studies in lexical relations. PhD diss., Massachusetts Institute of Technology.

    Save Citation »Export Citation »E-mail Citation »

    Introduces thematic relations, a set of localist semantic roles motivated by the structure of motion and location events. Grounds these notions in terms of decomposed representations of verb meaning, which supplement the localist notions with notions of agency and causation. Reprinted in Jeffrey S. Gruber’s Lexical Structures in Syntax and Semantics (Amsterdam: North-Holland, 1976), 1–210.

    Find this resource:

  • Gruber, Jeffrey S. 1967. Look and see. Language 43:937–947.

    DOI: 10.2307/411974Save Citation »Export Citation »E-mail Citation »

    Further illustrates the viability of the localist approach by motivating a localist analysis of the two perception verbs of the title—two verbs whose meanings might not seem readily amenable to an analysis in terms of motion or location.

    Find this resource:

  • Gruber, Jeffrey S. 2000. Thematic relations in syntax. In The handbook of contemporary syntactic theory. Edited by Mark Baltin and Chris Collins, 257–298. Oxford: Blackwell.

    Save Citation »Export Citation »E-mail Citation »

    Provides a comprehensive introduction to thematic relations, covering a wide range of verbs and constructions. Shows how particular sets of thematic relations are projected asymmetrically onto syntactic configurations encoding grammatical relations, and argues that these asymmetries rather than thematic hierarchies explain various linguistic phenomena.

    Find this resource:

  • Hust, Joel R., and Michael K. Brame. 1976. Jackendoff on interpretive semantics: A review of Semantic interpretation in generative grammar by R. Jackendoff. Linguistic Analysis 2:243–277.

    Save Citation »Export Citation »E-mail Citation »

    Section 1 (pp. 244–256) of this review of Jackendoff 1972 discusses thematic relations and the thematic hierarchy, raising questions about their effectiveness in the explanation of certain linguistic phenomena.

    Find this resource:

  • Jackendoff, Ray S. 1972. Semantic interpretation in generative grammar. Cambridge, MA: MIT Press.

    Save Citation »Export Citation »E-mail Citation »

    Chapter 2 (pp. 25–46) presents a much-cited summary of Gruber 1965 and discusses the advantages of thematic relations over Fillmore’s case grammar. This chapter and subsequent sections scattered through the book show how thematic relations can be used in the explanation of linguistic phenomena, including passivization, control, and reflexivization.

    Find this resource:

  • Jackendoff, Ray S. 1983. Semantics and cognition. Cambridge, MA: MIT Press.

    Save Citation »Export Citation »E-mail Citation »

    Develops the conceptual notions that undergird a theory of thematic relations as part of a larger effort to argue that a semantic representation is a conceptual structure; that is, that the study of semantics is the study of cognition.

    Find this resource:

  • Jackendoff, Ray S. 1987. The status of thematic relations in linguistic theory. Linguistic Inquiry 18:369–411.

    Save Citation »Export Citation »E-mail Citation »

    Proposes that to better accommodate notions such as agent and patient into a thematic relations account, the conceptual structure of a sentence includes an action tier representation in addition to a representation in terms of thematic relations.

    Find this resource:

  • Jackendoff, Ray S. 1990. Semantic structures. Cambridge, MA: MIT Press.

    Save Citation »Export Citation »E-mail Citation »

    Extensively elaborates and develops Jackendoff’s earlier work on conceptual semantics. Includes a fuller exposition of the action tier, provides treatments of a range of verb classes and adjunct types, and lays out a theory of the mapping from conceptual structure to the syntax, which references a thematic hierarchy.

    Find this resource:

Feature-Based Approaches

There have been various efforts to define semantic roles in terms of more primitive features, in the hope that such definitions would provide a basis for dealing with generalizations that involve finer or coarser levels of detail than typical semantic roles provide. Two representative efforts are Reinhart 2002 and Rozwadowska 1988. Specifically, Reinhart 2002 provides the most detailed exposition available currently of theta theory, a theory of argument structure and argument realization introduced by Reinhart and colleagues, which posits two equipollent semantic features. Ostler 1979 takes a localist approach that builds on the works cited under Thematic Relations to identify a set of features that can be used to define semantic roles. Anderson 2006, cited under Other Approaches, posits a distinct set of localist-inspired features.

  • Ostler, Nicholas M. 1979. Case-linking: A theory of case and verb diathesis applied to Classical Sanskrit. PhD diss., Massachusetts Institute of Technology.

    Save Citation »Export Citation »E-mail Citation »

    Inspired by Gruber 1965 and Jackendoff 1972 (both cited under Thematic Relations), Ostler defines semantic roles via two localist features, source and goal, which may be instantiated across different semantic fields and predicate types. He embeds these semantic role representations into a larger theory of argument realization, supported by data from Japanese and Sanskrit.

    Find this resource:

  • Reinhart, Tanya. 2002. The Theta System—An overview. Theoretical Linguistics 28:229–290.

    Save Citation »Export Citation »E-mail Citation »

    Introduces a theory of argument structure that distinguishes arguments in terms of two equipollent features: cause change (c) and mental state (m). The various combinations of these features stand in for typically recognized semantic roles. The features also figure in operations on argument structures.

    Find this resource:

  • Rozwadowska, Bożena. 1988. Thematic restrictions on derived nominals. In Thematic relations. Edited by Wendy Wilkins, 147–165. Syntax and Semantics 21. New York: Academic Press.

    Save Citation »Export Citation »E-mail Citation »

    Argues that semantic roles should not be taken to be unanalyzable, but rather should be defined in terms of the equipollent features Cause, Change, and Sentient. Supports this proposal with analyses of English and Polish nominalizations.

    Find this resource:

Other Approaches

Semantic roles play a part in a number of linguistic frameworks. Cook 1989 provides an excellent introduction to various approaches to semantic roles, including several approaches covered in this article. Lexicase is another tradition that owes an intellectual debt to case grammar; the most thorough exposition is Starosta 1988, but see Starosta 2003 for an overview. Integral to functional grammar is a set of participant—or semantic—roles; a representative introduction is Dik 1997. Cognitive grammar, represented by Langacker 2002, focuses on event conceptualizations, introducing a notion of role archetype to identify certain privileged event participants. Theta theory, presented in Reinhart 2002 (cited under Feature-Based Approaches) is a theory of argument realization in which semantic roles, defined in terms of two binary-valued features, play a central part. While theta theory does not use localist-inspired features, Anderson introduces two binary-valued features that are localist in their inspiration, resulting in a localist approach that is somewhat different in orientation than the approach cited under Thematic Relations. His ideas are developed in a series of books and articles, represented here by Anderson 2006. Published at about the same time as “The Case for Case” (Fillmore 1968, cited under Case Grammar: Development), Halliday 1967 provides a thorough and influential characterization of the argument-taking properties of verbs within systemic grammar, including the semantic roles posited. Further approaches are discussed under Generalized Semantic Roles.

  • Anderson, John M. 2006. Modern grammars of case: A retrospective. Cambridge, UK: Cambridge Univ. Press.

    DOI: 10.1093/acprof:oso/9780199297078.001.0001Save Citation »Export Citation »E-mail Citation »

    Taking a dependency-grammar perspective to sentence structure, Anderson motivates a spatial grounding for more traditional semantic roles via the features source and locative and combinations thereof (e.g., absolutive, the role of a located entity, is defined as the absence of these two features). Discusses various phenomena, including raising and control.

    Find this resource:

  • Cook, Walter Anthony. 1989. Case grammar theory. Washington, DC: Georgetown Univ. Press.

    Save Citation »Export Citation »E-mail Citation »

    A textbook that provides a comprehensive overview of Fillmore’s case grammar and also reviews various other approaches to semantic roles, including Anderson’s localist approach and Gruber and Jackendoff’s thematic relations, as well as other approaches directly inspired by Fillmore’s work.

    Find this resource:

  • Dik, Simon C. 1997. The theory of functional grammar. Part 1, The structure of the clause. 2d ed. Edited by Kees Hengeveld. Berlin: Mouton de Gruyter.

    Save Citation »Export Citation »E-mail Citation »

    Chapter 5 (pp. 105–126) introduces a set of semantic functions, as he calls semantic roles, relevant to the set of predication or “state-of-affairs” types recognized in functional grammar. Later chapters consider how semantic functions figure in subject and object assignment and other linguistic processes.

    Find this resource:

  • Halliday, M. A. K. 1967. Notes on transitivity and theme in English, Part I. Journal of Linguistics 3:37–81.

    DOI: 10.1017/S0022226700012949Save Citation »Export Citation »E-mail Citation »

    This influential paper lays out the various semantic clause types recognized in systemic grammar, as well as the semantic roles associated with each. Unlike case grammar, noun phrases may bear more than one semantic role in their clause.

    Find this resource:

  • Langacker, Ronald W. 2002. Transitivity, case, and grammatical relations. In Concept, image, and symbol: The cognitive basis of grammar. 2d ed. By Ronald W. Langacker, 209–260. Berlin: Mouton de Gruyter.

    Save Citation »Export Citation »E-mail Citation »

    Introduces cognitive grammar’s notion of role archetype: a participant type that recurs across event conceptualizations, which serves as the prototype for a morphological case or grammatical relation. Suggests events are conceptualized as action chains, with role archetypes organized according to their place in such chains, leading to a thematic hierarchy.

    Find this resource:

  • Starosta, Stanley. 1988. The case for Lexicase. London: Pinter.

    Save Citation »Export Citation »E-mail Citation »

    Chapter 4 of this introduction to Lexicase, a dependency grammar framework, lays out an inventory of five semantic roles that gives prominence to the notion of patient; each role has “inner” and “outer” versions, corresponding roughly to argument and adjunct instantiations. Subsequent chapters analyze argument realization phenomena in various languages.

    Find this resource:

  • Starosta, Stanley. 2003. Lexicase grammar. In Dependenz und Valenz: Ein internationales Handbuch der zeitgenössischen Forschung/Dependency and valency: An international handbook of contemporary research. Vol. 1. Edited by Vilmos Ágel, Ludwig M. Eichinger, Hans-Werner Eroms, Peter Hellwig, Hans Jürgen Heringer, and Henning Lobin, 526–545. Berlin: Walter de Gruyter.

    DOI: 10.1515/9783110141900.1Save Citation »Export Citation »E-mail Citation »

    A brief introduction to the highlights of Starosta’s Lexicase, including its treatment of semantic roles.

    Find this resource:

Generalized Semantic Roles

The term “generalized semantic role” is introduced in Van Valin 1999 (cited under Macroroles) to refer to several conceptions of semantic roles that are unified in rejecting the traditional assumption that a semantic role is characterized by a set of jointly necessary and sufficient conditions. Such characterizations are introduced to deal with the multiplicity of semantic roles that are realized as subject, and to a lesser extent as object; for this reason, most such approaches usually recognize two such roles. Proto-roles and Macroroles each focus on an influential conception of generalized semantic role. Other Conceptions covers further perspectives on generalized semantic roles. An overview of generalized semantic roles can be found in section 3.1 (pp. 51–68) of Levin and Rappaport Hovav 2005, cited under General Works.

Proto-Roles

The best known generalized semantic roles are the two proto-roles introduced in Dowty 1991 as part of a theory of subject and object selection. Each proto-role brings together a cluster of properties that verbs lexically entail of their arguments which are relevant to argument realization: the agent proto-role properties have been associated with the semantic role agent and are relevant to subjecthood, while the patient proto-role properties are associated with the semantic role patient and are relevant to objecthood. The entailments identified in Dowty 1991 overlap in significant ways with the properties that figure in subject selection in Fillmore 1977b, cited under Case Grammar: Development. The proto-role approach has been applied to various linguistic phenomena beyond those discussed in Dowty 1991, with these applications sometimes requiring extensions or refinements of the approach. Grimm 2011 recasts the proto-role entailments as privative features, which can be used in the explanation of the distribution of morphological cases. Primus 1999 proposes a third, recipient proto-role, to deal with phenomena involving dative case. Blume 1998 builds on Primus’s work to handle verbs with a second oblique argument. Ackerman and Moore 2001 extends the argument realization capabilities of proto-roles so they can better handle multiple realizations of the arguments of a single verb; in so doing, the authors also propose an additional patient proto-role entailment. Aissen 1999 uses proto-roles in an account of morphosyntactic marking of arguments across languages. Davis and Koenig 2000 refines the proto-role entailments and uses them within a hierarchically organized lexicon that is inspired by head-driven phrase structure grammar (HPSG). Barker and Dowty 1993 proposes a comparable set of lexical entailments for nouns. Critiques of proto-roles can be found in Croft 1998 (cited under Thematic Hierarchies) and Newmeyer 2001 (cited under Argument Realization).

  • Ackerman, Farrell, and John Moore. 2001. Proto-properties and grammatical encoding: A correspondence theory of argument selection. Stanford, CA: CSLI Publications.

    Save Citation »Export Citation »E-mail Citation »

    Refines the proto-role approach to better handle the range of argument realization options for a particular verb, especially those involving an argument that can be realized as an object or oblique. Argues such alternations require adding “bounding entity” to the set of patient proto-role entailments.

    Find this resource:

  • Aissen, Judith. 1999. Markedness and subject choice in Optimality Theory. Natural Language and Linguistic Theory 17:673–711.

    DOI: 10.1023/A:1006335629372Save Citation »Export Citation »E-mail Citation »

    Uses reified proto-roles within an Optimality Theoretic account of morphological markedeness in subject and object choice across languages.

    Find this resource:

  • Barker, Chris, and David Dowty. 1993. Non-verbal thematic proto-roles. In NELS 23: Proceedings of the 23rd Annual Meeting of the North East Linguistic Society. Vol. 1. Edited by Amy J. Schafer, 49–62. Amherst: Graduate Linguistics Student Association, Univ. of Massachusetts.

    Save Citation »Export Citation »E-mail Citation »

    Applies Dowty’s 1991 proto-role approach to relational noun phrases, positing proto-part and proto-whole lexical entailments, which enter into the choice of head versus prenominal possessor/of complement in such noun phrases.

    Find this resource:

  • Blume, Kerstin. 1998. A contrastive analysis of interaction verbs with dative complements. Linguistics 36:253–280.

    Save Citation »Export Citation »E-mail Citation »

    Through an examination of several languages, shows there is a significant class of two-argument verbs with an animate second argument expressed in an oblique case, typically the dative. Argues such arguments do not show patient proto-role entailments and explains their oblique case using a precursor of Primus 1999.

    Find this resource:

  • Davis, Anthony R., and Jean-Pierre Koenig. 2000. Linking as constraints on word classes in a hierarchical lexicon. Language 76:56–91.

    Save Citation »Export Citation »E-mail Citation »

    Integrates the proto-role entailments into a theory of argument realization inspired by head-driven phrase structure grammar that recognizes hierarchically related verb classes. Includes a lengthy discussion of the problems with using thematic hierarchies in argument realization—a construct that the proposed analysis does not require.

    Find this resource:

  • Dowty, David R. 1991. Thematic proto-roles and argument selection. Language 67:547–619.

    Save Citation »Export Citation »E-mail Citation »

    Proposes that certain entailments that a verb may lexically impose on its arguments play a role in subject and object selection with transitive verbs. These privileged entailments fall into two sets—the agent and patient proto-role entailments—according to whether they are typically associated with agent or patient.

    Find this resource:

  • Grimm, Scott. 2011. Semantics of case. Morphology 21:515–544.

    DOI: 10.1007/s11525-010-9176-zSave Citation »Export Citation »E-mail Citation »

    Reenvisions Dowty’s agent proto-role entailments as privative properties. Uses the possible combinations of these properties to define a lattice. Proposes that morphological cases are associated with continuous regions in this lattice, so that the lattice essentially defines a typology of case systems.

    Find this resource:

  • Primus, Beatrice. 1999. Cases and thematic roles: Ergative, accusative and active. Tübingen, Germany: Niemeyer.

    DOI: 10.1515/9783110912463Save Citation »Export Citation »E-mail Citation »

    Notes that certain agent and patient proto-role entailments are paired with each other in an asymmetric control relation. Using data from several languages, proposes proto-roles figure in morphological case assignment and the control relation figures in word order determination. Argues for a recipient proto-role. Chapter 3 (pp. 32–60) is particularly relevant.

    Find this resource:

Macroroles

Prominent in Role and Reference Grammar are two generalized semantic roles, the macroroles actor and undergoer, which are assigned algorithmically to the arguments of a verb based on their relative positions in an aspectually inspired predicate decomposition of a verb’s meaning. Conventional semantic roles, too, can be defined over these predicate decompositions. Van Valin 2004 provides an introduction to the notion of macrorole, while Van Valin 2006 lays out the predicate decompositions that are the basis for macrorole definitions. Van Valin 1999 provides an illuminating comparison of macroroles and proto-roles.

  • Van Valin, Robert D., Jr. 1999. Generalized semantic roles and the syntax-semantics interface. In Empirical issues in formal syntax and semantics. Vol. 2. Edited by Francis Corblin, Carmen Dobrovie-Sorin, and Jean-Marie Marandin, 373–389. The Hague: Thesus.

    Save Citation »Export Citation »E-mail Citation »

    Reviews the motivation for positing macroroles, and then compares them to other conceptions of generalized semantic role, particularly the proto-roles of Dowty 1991 (cited under Proto-roles).

    Find this resource:

  • Van Valin, Robert D., Jr. 2004. Semantic macroroles in Role and Reference Grammar. In Semantische Rollen. Edited by Rolf Kailuweit and Martin Hummel, 62–82. Tübingen, Germany: Narr.

    Save Citation »Export Citation »E-mail Citation »

    Explicates the notion of marcrorole in Role and Reference Grammar and addresses two fundamental questions: whether macroroles are syntactic or semantic notions, and why there are only two of them. Reviews dissociations between the number of surface arguments found with a verb and the number of macroroles it selects.

    Find this resource:

  • Van Valin, Robert D., Jr. 2006. Semantics in Role and Reference Grammar. In Encyclopedia of language and linguistics. 2d ed. Edited by Keith Brown, 158–167. Amsterdam: Elsevier.

    Save Citation »Export Citation »E-mail Citation »

    This overview of semantics in Role and Reference Grammar presents the two notions of semantic role that it assumes: a number of thematic relations analogous to typical semantic roles, defined over positions in predicate decompositions, and two macroroles, generalized semantic roles that are algorithmically defined over the thematic relations.

    Find this resource:

Other Conceptions

Although proto-roles and macroroles are probably the best-known generalized semantic roles, notions that fall under this rubric in that they bring together clusters of more conventionally sized semantic roles are posited in a range of other work. A representative sample is included here. The notions of A and O, most explicitly introduced in Dixon 1979, might be considered the antecedents of generalized semantic roles, although they are not usually presented in this way; they are now commonly used in the functional-typological literature to refer to the arguments of a transitive verb. The A-case in Schlesinger 1995 bears considerable similarity to the agent proto-role in being a cluster concept; Schlesinger also introduces a C-case, which does not have an analogue in other generalized semantic role approaches, to handle comitative-like and instrument-like notions. Kibrik 1997 introduces four hyperroles to deal with accusative versus ergative languages. Several previously mentioned works also include generalized semantic roles. Lexicase, represented in Starosta 1988 and Starosta 2003 (cited under Other Approaches), includes macroroles inspired by Role and Reference Grammar. Dik 1997 (also cited under Other Approaches) introduces a notion of first argument, which encompasses several semantic roles, including agent and force, to deal with their uniform realization as subject. The action tier, introduced in Jackendoff 1987 and elaborated in Jackendoff 1990 (both cited under Thematic Relations), serves a purpose very similar to generalized semantic roles.

  • Dixon, R. M. W. 1979. Ergativity. Language 55:59–138.

    DOI: 10.2307/412519Save Citation »Export Citation »E-mail Citation »

    As part of a larger study of ergativity, section 5.2 (pp. 102–109) introduces the “syntactic-semantic” labels A and O to refer to the semantic classes of arguments of transitive verbs realized as subject and direct object, respectively, in a language like English with a nominative-accusative system of morphological case.

    Find this resource:

  • Kibrik, Aleksandr E. 1997. Beyond subjects and objects: Towards a comprehensive relational typology. Linguistic Typology 1:279–346.

    Save Citation »Export Citation »E-mail Citation »

    Introduces a notion of hyperrole in delineating a typology of the organization of clause structure. Besides recognizing agent and patientive hyperroles comparable to Role and Reference Grammar’s actor and undergoer, respectively, introduces principal and absolutive hyperroles, which are key in accusative and ergative languages, respectively.

    Find this resource:

  • Schlesinger, Izchak M. 1995. Cognitive space and linguistic case. Cambridge, UK: Cambridge Univ. Press.

    DOI: 10.1017/CBO9780511551321Save Citation »Export Citation »E-mail Citation »

    Argues for agent-based and comitative/instrument/manner-based generalized semantic roles, the A-case and C-case, respectively, while rejecting a comparable patient-based role. Supports these proposals with analyses of subject and object selection and preposition choice. Draws support from acceptability rating studies, primarily of English.

    Find this resource:

Sources of Semantic Roles

Many researchers see semantic roles as notions derived from more elaborated semantic representations. Jackendorf was the first to argue (in Jackendoff 1972) that semantic roles are simply labels for positions in a predicate decomposition, a proposal that continues to receive attention. Dowty 1991, cited under Proto-Roles, sees them as names for certain recurring clusters of entailments imposed by verbs on their arguments. Croft 1991, section 4.3.4 (pp. 155–159), suggests instead that semantic roles are defined via positions in a “causal chain” representation of an event; see also Langacker 2002, cited under Other Approaches. Along similar lines, Haiden 2012 proposes that semantic roles reflect cognitively and perceptually salient properties of event participants. Various works cited under Theoretical Notion also consider the nature of semantic roles.

  • Croft, William A. 1991. Syntactic categories and grammatical relations. Chicago: Univ. of Chicago Press.

    Save Citation »Export Citation »E-mail Citation »

    Chapter 4 argues that the representation of verb meaning takes the form of a causal chain, with semantic roles defined in terms of distinguished positions in this causal chain. Croft discusses the assumptions behind semantic role approaches and uses these as a basis for a critique of such approaches.

    Find this resource:

  • Haiden, Martin. 2012. The content of semantic roles: Predicate-argument structure in language and cognition. In The Theta System: Argument structure at the interface. Edited by Martin Everaert, Marijana Marelj, and Tal Siloni, 52–77. Oxford: Oxford Univ. Press.

    DOI: 10.1093/acprof:oso/9780199602513.001.0001Save Citation »Export Citation »E-mail Citation »

    Drawing on the cognitive and developmental psychology literature, argues semantic roles are grounded in properties that figure in the perception of causality and intention. Haiden uses these properties to give substance to the two features that Reinhart 2002 (cited under Feature-Based Approaches) employs to define the set of semantic roles.

    Find this resource:

  • Jackendoff, Ray S. 1972. Semantic interpretation in generative grammar. Cambridge, MA: MIT Press.

    Save Citation »Export Citation »E-mail Citation »

    Section 2.4 (pp. 36–43) argues that semantic roles are labels for certain recurring positions in the predicate decomposition representations of events, an idea that has since been more widely adopted.

    Find this resource:

Specific Semantic Roles

Although semantic role lists have been largely discredited as a semantic representation, specific semantic roles are often referred to in discussions of linguistic phenomena. This section provides references to work that deals with some of the most often posited semantic roles, especially work that deals with their definition, their identification, and their place in the description and explanation of linguistic phenomena. These roles include agent and patient/theme, perhaps the two most referenced roles, which are identified with the subject and object of transitive verbs, respectively. Croft 1991 makes a noteworthy proposal to impose some organization on semantic roles other than agent and patient/theme: these semantic roles fall into two linguistically significant groups, the antecedent and subsequent roles. Luraghi 2001 shows the relevance of these two groups to the understanding of case syncretisms, while also recognizing a third group of roles.

  • Croft, William A. 1991. Syntactic categories and grammatical relations. Chicago: Univ. of Chicago Press.

    Save Citation »Export Citation »E-mail Citation »

    Section 5.2 (pp. 184–192) introduces the “causal order hypothesis,” which identifies two subsets of semantic roles, the antecedent and subsequent roles, according to whether a participant bearing a given role precedes or follows the patient in an event’s causal chain. Croft argues that these classes of roles figure in linguistic phenomena such as case syncretisms.

    Find this resource:

  • Luraghi, Silvia. 2001. Syncretism and the classification of semantic roles. Sprachtypologie und Universalienforschung 54:35–51.

    Save Citation »Export Citation »E-mail Citation »

    Uses case syncretisms as a tool to look at relations among semantic roles, and argues for a refinement on Croft’s antecedent and subsequent roles that also recognizes a group of concomitant roles (see Croft 1991).

    Find this resource:

Agent

The agent role, typically defined as the role assigned to the animate volitional instigator of an event, is one of the most cited semantic roles, figuring in a wide range of linguistic phenomena, but this role defies easy analysis. Within linguistics, there is ongoing debate about the best characterization of the agent role in particular and the concept of agency in general, which is reflected in considerable work on agentivity. The notion of agent has also received attention in the literature on the philosophy and psychology of action and events, which is outside the scope of this article. Cruse 1973 considers the nature of the agent role by showing that a much used diagnostic for agentivity delineates a set of arguments that go beyond the narrow understanding of agent, and then identifying a subset of them as true agents. Folli and Harley 2008 argues that agents are distinguished by more than animacy; they are “teleologically capable” entities, which may or may not be animate. Clark and Carpenter 1989 provides a localist perspective on the notion of agent, treating it as a type of source. Van Valin and Wilkins 1996 reviews various definitions of agent and proposes that agentivity arises from an implicature, adopting a proposal made in Holisky 1987 on the basis of different data; it argues that most instances of the agent role are actually instances of a broader effector role. Further work concerning the scope of the agent role is presented in Broad versus Narrow Conceptions, while Non-prototypical Instances considers agents that depart from the role’s prototype. Not only is the characterization of the agent role a matter of controversy, but so is the place of this role in the explanation of linguistic phenomena. This controversy stems from the typical realization of the agent as the subject of an active sentence. As discussed in Keenan 1976, there is considerable overlap in the noun phrases picked out by the semantic role agent, the grammatical relation subject, the morphological case nominative, and the discourse notion of topic, and the notion relevant to a given phenomenon can be hard to identify, so that sometimes phenomena said to involve one of these notions actually involves another. These complications arise for semantic roles in general, and further references may be found in Place in Linguistic Generalizations.

  • Clark, Eve V., and Kathie L. Carpenter. 1989. The notion of source in language acquisition. Language 65:1–30.

    DOI: 10.2307/414840Save Citation »Export Citation »E-mail Citation »

    Taking a localist approach, argues that agents and causes are conceptualized as sources based on the use of the English preposition from to express nonsubject agents as well as causes in child speech.

    Find this resource:

  • Cruse, D. A. 1973. Some thoughts on agentivity. Journal of Linguistics 9:11–23.

    DOI: 10.1017/S0022226700003509Save Citation »Export Citation »E-mail Citation »

    In an attempt to come to grips with the notion of agentivity, discusses the problems plaguing definitions of agent and the what X did was diagnostic for agents. Proposes that four semantic notions—volitive, effective, initiative, and agentive—figure in any characterization of the NPs picked out by this diagnostic.

    Find this resource:

  • Folli, Raffaella, and Heidi Harley. 2008. Teleology and animacy in external arguments. Lingua 118:190–202.

    DOI: 10.1016/j.lingua.2007.02.004Save Citation »Export Citation »E-mail Citation »

    Proposes that teleological capability rather than animacy is criterial to the agent role. The result is a broader than usual notion of agent, encompassing, for instance, sound emitters. Also shows agents are set apart from causes, citing differences in the linguistic behavior of the two types of entities.

    Find this resource:

  • Holisky, Dee Ann. 1987. The case of the intransitive subject in Tsova-Tush (Bats). Lingua 71:103–132.

    DOI: 10.1016/0024-3841(87)90069-6Save Citation »Export Citation »E-mail Citation »

    Concludes that agentivity arises via an inference through an analysis of the complex and sometimes variable distribution of the ergative marker on the single argument of intransitive verbs in the Caucasian language Tsova-Tush. Takes volition and conscious control to be properties of agents.

    Find this resource:

  • Keenan, Edward L. 1976. Towards a universal definition of “subject.” In Subject and topic. Edited by Charles N. Li, 303–333. New York: Academic Press.

    Save Citation »Export Citation »E-mail Citation »

    Reviews a wide range of properties attributed to subjects, and argues that they are of three types: coding (i.e., morphological case, agreement), behavior (i.e., syntactic), and semantic properties. The semantic properties include those properties ascribed to subjects by virtue of their agentivity.

    Find this resource:

  • Van Valin, Robert D., Jr., and David P. Wilkins. 1996. The case for “effector”: Case roles, agents, and agency revisited. In Grammatical Constructions. Edited by Masayoshi Shibatani and Sandra A. Thompson, 289–322. Oxford: Clarendon.

    Save Citation »Export Citation »E-mail Citation »

    Argues that few verbs entail agentivity of an argument; rather, in most instances, agentivity arises via an inference. Considers the factors that favor such an inference and revisits the notions of instrument and cause in this context. Includes critical reviews of major previous approaches to agentivity.

    Find this resource:

Broad versus Narrow Conceptions

As arguments bearing a range of semantic characterizations can be realized as subjects just as agents are, the question arises of whether a range of semantic roles can be realized as subject, whether the definition of agent should be broadened, or whether the agent role should be dispensed with in favor of either a broader semantic role or a generalized semantic role. All these positions have been taken. Folli and Harley 2008 (cited under Agent) argues for a broader notion of agent that encompasses, for instance, inanimate sound emitters. Van Valin and Wilkins 1996 (also cited under Agent) argues that most instances of the agent role are actually instances of Role and Reference Grammar’s broader effector role. Dowty 1991 (cited under Proto-Roles) takes a generalized semantic role approach to the agent role, as does Schlesinger 1995, which argues for a cluster concept notion of agent, the A-case, that has significant similarities with Dowty’s agent proto-role. The difficulties in identifying a necessary and sufficient definition of the agent role that encompasses all purported instances of this role are not the only motivation for a cluster concept or prototype approach to this role; attempts to provide a unified characterization of subject selection suggest a similar approach. The scope of the notion of agent has also been debated as a consequence of efforts to understand the relation of the agent role to the semantic roles cause, instrument, and natural force, which appear to belong to the same larger semantic domain, especially since they all can also be realized as subjects in English as well as some other languages. Since the subject of the ideal transitive event is considered to instantiate a prototypical agent, DeLancey 1984 considers the range of semantic notions that can and cannot be expressed as the subjects of transitive verbs in various languages to probe the nature of agentivity, distinguishing agents, which the author defines as ultimate volitional causes, from instruments, and other types of causes in terms of their argument realization possibilities. Croft 1991 (cited under Sources of Semantic Roles) takes a similar perspective, defining the notion of agent in terms of its position in a causal chain. Alexiadou and Schäfer 2006 also argues for distinct agent and cause roles. This work, then, takes a different position from Levin and Rappaport Hovav 1995, which argues for a notion of external cause that subsumes agents and other types of causers, or Schlesinger 1995, which encompasses some instruments and even patients under the cluster concept, the A-case.

  • Alexiadou, Artemis, and Florian Schäfer. 2006. Instrument subjects are agents or causers. In Proceedings of the 25th West Coast Conference on Formal Linguistics. Edited by Donald Baumer, David Montero, and Michael Scanlon, 40–48. Somerville, MA: Cascadilla Proceedings Project.

    Save Citation »Export Citation »E-mail Citation »

    Argues, using data from English, German, and Greek, that distinct agent and causer roles should be recognized due to their distinct morphosyntactic properties, and that the Role and Reference Grammar notion of effector overly underspecifies the role associated with subjects. Considers instrument subjects in this context.

    Find this resource:

  • DeLancey, Scott. 1984. Notes on agentivity and causation. Studies in Language 8:181–213.

    DOI: 10.1075/sl.8.2.05delSave Citation »Export Citation »E-mail Citation »

    Critically examines the notion of agent and the related notions of cause and instrument. Argues that the prototypical agent is an ultimate volition cause, through an examination of departures from transitivity in Hare (Athabaskan), Newari (Tibeto-Burman), and English.

    Find this resource:

  • Levin, Beth, and Malka Rappaport Hovav. 1995. Unaccusativity: At the syntax-lexical semantics interface. Cambridge, MA: MIT Press.

    Save Citation »Export Citation »E-mail Citation »

    Drawing on data from the causative alternation, argues for a notion of external cause that subsumes agents, natural forces, instruments, and emitters.

    Find this resource:

  • Schlesinger, Izchak M. 1995. Cognitive space and linguistic case. Cambridge, UK: Cambridge Univ. Press.

    DOI: 10.1017/CBO9780511551321Save Citation »Export Citation »E-mail Citation »

    Chapter 2 proposes a cluster concept notion of agent, the A-case, involving the features cause, control, and change. Chapter 9 considers how this characterization of agent figures in verb classification and experimentally verifies this classification by testing paraphrasability by a do verb form.

    Find this resource:

Non-prototypical Instances

Although volitionality is taken by some as criterial for agentivity, there has been increasing recognition that volitionality is not essential to agentivity. Nevertheless, the prototypical agent is volitional, as Dowty 1991 (cited under Proto-Roles) acknowledges in including volitionality as one of the entailments of the agent proto-role. Kittilä 2005 argues for a distinction between volitional versus nonvolitional agents, citing a range of crosslinguistic data that illustrates that nonvolitional agents receive a different expression than volitional agents. Fauconnier 2011 makes a similar point based on a systematic survey of over 150 languages. Budwig 1989 shows the relevance of the notion of control over an event using data from child language. Saksena 1980 considers a different type of non-prototypical agent, so-called affected agents, which share the property of affectedness with patients and show behavior that is distinct from that of prototypical agents.

  • Budwig, Nancy. 1989. The linguistic marking of agentivity and control in child language. Journal of Child Language 16:263–284.

    DOI: 10.1017/S0305000900010412Save Citation »Export Citation »E-mail Citation »

    Using data from six children two to three years of age, Budwig shows they prefer the pronoun my to I in sentences with increased agentivity. She further argues that in action sentences, some of these children use my where their control over the environment is at issue (e.g., in commands, requests), and I in other instances.

    Find this resource:

  • Fauconnier, Stefanie. 2011. Differential agent marking and animacy. Lingua 121:533–547.

    DOI: 10.1016/j.lingua.2010.10.014Save Citation »Export Citation »E-mail Citation »

    Demonstrates the importance of the notion of involuntary animate agent through a study of its encoding in 150 languages. Highlights the use of anticausative verb forms to indicate the presence of involuntary agents, arguing that such forms represent the uncontrolled nature of events with such agents.

    Find this resource:

  • Kittilä, Seppo. 2005. Remarks on involuntary agent constructions. Word 56:381–419.

    Save Citation »Export Citation »E-mail Citation »

    Illuminates agentivity by showing that, in some languages, descriptions of events with an involuntary animate agent are not given the morphosyntactically transitive expression typical of events with prototypical agents. Shows involuntary agents are distinguished from natural forces, and discusses the interaction of the animacy hierarchy with construal as an involuntary agent.

    Find this resource:

  • Saksena, Anuradha. 1980. The affected agent. Language 56:812–826.

    DOI: 10.2307/413490Save Citation »Export Citation »E-mail Citation »

    Proposes that the agent of some two-arguments verbs, particularly verbs of physical and mental ingesting, should be distinguished from prototypical agents in being affected. Supports this proposal by pointing to morphosyntactic parallels between such agents and patients, the prototypical affected arguments.

    Find this resource:

Patient, Theme, and Affectedness

The semantic role associated with the non-agent argument of a transitive verb is a second semantic role that has received considerable attention. This role, called the object in case grammar, is more commonly referred to as patient or theme. The labels “patient” and “theme” are sometimes used interchangeably, but they originally had different extensions. The label “theme” was introduced for the semantic role of a moved or located entity in the inventory of thematic relations introduced in the work of Gruber and Jackendoff (cited under Thematic Relations), but the use of the label was broadened until it came to refer to the semantic role of almost any argument realized as direct object within the Government-Binding framework. Thus, the scope of the semantic role “theme” has been the subject of controversy, and Potashnik 2012 is representative of efforts to delimit the use of this label. The semantic role “patient,” in contrast, was introduced within work that hewed more closely to case grammar, whose focus was verbs of change of state. The critical property of the patient role is affectedness, as described in Anderson 1977 (cited under Other Phenomena). Anderson 2004, a handbook article, examines linguistic phenomena said to involve this notion, although Tenny 1992 (cited under Other Phenomena) proposes that the relevant phenomena actually have aspectual explanations. Further experimental support for this notion is provided in Gropen, et al. 1991. Affectedness also figures among the patient proto-role entailments in Dowty 1991 (cited under Proto-roles). Its importance is also recognized in the action tier introduced to complement the thematic relations analysis of a sentence in Jackendoff 1987 (cited under Thematic Relations), and in subsequent work on conceptual structure, such as Jackendoff 1990 (also cited under Thematic Relations). Beavers 2011 proposes that the notion of affectedness should be given an aspectual grounding. Schlesinger 1995 (cited under Other Conceptions), in contrast, argues against identifying a patient role, pointing to the difficulties in finding a unified semantic characterization of the arguments that are realized as direct objects, as well as the realization of certain prototypical patients as subjects of intransitive verbs.

  • Anderson, Mona. 2004. Affectedness. In The Blackwell companion to syntax. Vol. 1. Edited by Martin Everaert and Henk van Riemsdijk, 121–141. Oxford: Blackwell.

    Save Citation »Export Citation »E-mail Citation »

    A handbook article that examines the notion of affectedness and the part it plays in the explanation of various linguistic phenomena.

    Find this resource:

  • Beavers, John. 2011. On affectedness. Natural Language and Linguistic Theory 29:335–370.

    DOI: 10.1007/s11049-011-9124-6Save Citation »Export Citation »E-mail Citation »

    Proposes that the notion of affectedness should be aspectually grounded, where the relevant aspectual characterization is formulated in terms of the notion of scalar change. Recognizes degrees of affectedness, defined in terms of monotonically weaker truth conditions that characterize the result state of the patient/theme.

    Find this resource:

  • Gropen, Jess, Steven Pinker, Michelle Hollander, and Richard Goldberg. 1991. Affectedness and direct objects: The role of lexical semantics in the acquisition of verb argument structure. Cognition 41:153–195.

    DOI: 10.1016/0010-0277(91)90035-3Save Citation »Export Citation »E-mail Citation »

    Using experimental evidence from children and adults, argues that affectedness, defined as either change of location or change of state, is a key notion in determining the realization of an argument of a verb as its object.

    Find this resource:

  • Potashnik, Joseph. 2012. Emission verbs. In The Theta System: Argument structure at the interface. Edited by Martin Everaert, Marijana Marelj, and Tal Siloni, 251–278. Oxford: Oxford Univ. Press.

    DOI: 10.1093/acprof:oso/9780199602513.001.0001Save Citation »Export Citation »E-mail Citation »

    Written in the context of Reinhart’s theta theory (see Reinhart 2002, cited under Feature-Based Approaches), reexamines the use of the label “theme” for the emitter argument of emission verbs, and argues that it has been wrongly attributed to them: they are better characterized as a type of cause.

    Find this resource:

Instrument

The instrument role has received considerable attention, and the references listed here are only a sample of those available, chosen to highlight the facets of this role that have provoked theoretical interest. Koenig, et al. 2008 presents an extensive survey of English verbs that take instrument arguments, grouping them into subclasses according to the way the instrument participates in the event denoted; thus, it provides a starting point for future analyses. Nilsen 1973 is a monograph devoted to the instrument role from a case grammar perspective, but much of the discussion remains relevant. Wojcik 1976 proposes that the notion of instrument is not primitive, but rather should be viewed as a label for certain positions in predicate decompositions. Lakoff 1968 lays out one of the most used diagnostics for instruments, whose effectiveness is further probed in Schlesinger 1979, a paper that tries to better understand the notion of instrument by considering where to place a dividing line between the notions of instrument and comitative. Studies of the comitative role, including some cited under Comitative, often investigate whether there should be a comitative role distinct from the instrument role. This question has been raised on both conceptual grounds and morphosyntactic grounds: both instruments and comitatives appear to depend on the presence of an agent, and both have the same oblique morphosyntactic realization in various languages. Arguments that apparently bear the instrument role may also be realized as subjects in English and some other languages—that is, they have the same realization as agents and natural causes. This shared expression raises questions about the boundaries among these roles. Schlesinger 1989 investigates the semantic properties that make this realization of instruments possible. Alexiadou and Schäfer 2006 (cited under Broad versus Narrow Conceptions) suggests that instrumental subjects are not all alike: some are more like agents, and others are more like causes. The interrelations among the notions of agent, cause, and instrument are probed further in Van Valin and Wilkins 1996 (cited under Agent), and in DeLancey 1984 and Schlesinger 1995 (cited under Broad versus Narrow Conceptions). Stolz 2001 considers the affinities of instruments with agents on the one hand, and comitatives on the other, through a crosslinguistic investigation of case syncretisms.

  • Koenig, Jean-Pierre, Gail Mauner, Breton Bienvenue, and Kathy Conklin. 2008. What with?: The anatomy of a (proto)-role. Journal of Semantics 25:175–220.

    DOI: 10.1093/jos/ffm013Save Citation »Export Citation »E-mail Citation »

    Systematically surveys English verbs that allow or require an instrument, proposing semantic representations for verbs of each subtype identified. Argues that the majority of verb types identified involve three subsituations precisely because of the nature and involvement of the instrument.

    Find this resource:

  • Lakoff, George. 1968. Instrumental adverbs and the concept of deep structure. Foundations of Language 4:4–29.

    Save Citation »Export Citation »E-mail Citation »

    Introduces and argues for a much adopted and discussed diagnostic for the instrument role: the availability of a use NP to paraphrase.

    Find this resource:

  • Nilsen, Don L. F. 1973. The instrumental case in English: Syntactic and semantic considerations. The Hague: Mouton.

    Save Citation »Export Citation »E-mail Citation »

    A detailed and wide-ranging investigation into the instrument role, which reviews a considerable number of syntactic and semantic diagnostics, arguing in favor of the latter.

    Find this resource:

  • Schlesinger, Izchak M. 1979. Cognitive structures and semantic deep structures: The case of the instrumental. Journal of Linguistics 15:307–324.

    DOI: 10.1017/S0022226700016418Save Citation »Export Citation »E-mail Citation »

    Shows English speakers judge sentences with a with phrase to form a continuum according to whether this phrase is understood as instrument or comitative. A survey of speakers of other languages reveals that the domains of their instrument markers also respect this continuum.

    Find this resource:

  • Schlesinger, Izchak M. 1989. Instruments as agents: On the nature of semantic relations. Journal of Linguistics 25:189–210.

    DOI: 10.1017/S0022226700012147Save Citation »Export Citation »E-mail Citation »

    Uses an investigation into the conditions that license apparent instrument subjects to argue for a prototype notion of agent, by virtue of which instrument subjects can be viewed as a type of agent. This paper is a precursor to Schlesinger 1995, cited under Broad versus Narrow Conceptions.

    Find this resource:

  • Stolz, Thomas. 2001. Comitatives vs. instrumentals vs. agents. In Aspects of typology and universals. Edited by Walter Bisang, 153–174. Berlin: Akademie Verlag.

    Save Citation »Export Citation »E-mail Citation »

    Examines morphological case syncretisms involving the instrument, agent, and comitative roles, showing that languages either have instrument/agent (i.e., ergative case) syncretisms or instrument/comitative syncretisms, but not comitative/agent syncretisms, unless the instrument is involved too. Ends with a speculative discussion of the semantic implications of this pattern.

    Find this resource:

  • Wojcik, Richard. 1976. Where do instrumental NPs come from? In The grammar of causative constructions. Edited by Masayoshi Shibatani, 165–180. Syntax and Semantics 6. New York: Academic Press.

    Save Citation »Export Citation »E-mail Citation »

    Written from a generative semantics perspective, argues for defining the notion of instrument with respect to certain positions in predicate decompositions. Motivated by some difficult data involving verbs apparently taking two distinct instruments.

    Find this resource:

Comitative

The efforts to understand the semantic roles agent and instrument have drawn attention to the semantic role “comitative.” This role, as the role associated with animate entities that help the agent in performing an event, has affinities to both the agent and instrument roles. Most discussions of this role have centered around whether there are distinct comitative and instrument roles, especially because in some languages a single preposition (e.g., English with) or morphological case can be used with noun phrases bearing either of these roles. Two early papers on this topic, which are also seen as contributions to case grammar, are Walmsley 1971, which argues for treating the comitative and instrument roles as a single role, and Buckingham 1973, which argues for a distinct comitative role. These works, as well as Seiler 1974 and Schlesinger 1995, also introduce a variety of diagnostics for comitatives. Seiler 1974 subsumes both comitative and instrument under a general notion of concomitance to help explain why both can have the same morphological marking. Luraghi 2001 (cited under Specific Semantic Roles) also identifies a set of comitative-like roles through an examination of case syncretisms. Schlesinger 1995 deals with the shared morphosyntactic realization, as well as the semantic affinities, by positing a generalized semantic role, the C-role. Stolz 1996 brings crosslinguistic data to bear on the question of whether there is a distinct comitative role, arguing that such data do not support collapsing the comitative and instrument under a single role; Stolz, et al. 2006 is a more comprehensive conceptual and typological study that builds on this earlier work. References to the comitative role may also be found in some works listed under Instrument.

  • Buckingham, Hugh W. 1973. The comitative and case grammar. Foundations of Language 10:111–121.

    Save Citation »Export Citation »E-mail Citation »

    Written as a reply to Walmsley 1971, argues for an independent comitative role, pointing out that the comitative has a dependence on the agent that Walmsley’s analysis does not capture.

    Find this resource:

  • Schlesinger, Izchak M. 1995. The comitative. In Cognitive space and linguistic case. By Izchak M. Schlesinger, 60–91. Cambridge, UK: Cambridge Univ. Press.

    DOI: 10.1017/CBO9780511551321.005Save Citation »Export Citation »E-mail Citation »

    Introduces a generalized semantic role, the C-case, which subsumes the features accompaniment, manner, and instrument to explain the distribution of English with. Proposes that the presence of one of these features is sufficient for the use of with. Draws support from paraphrase and acceptability studies of English and Arabic.

    Find this resource:

  • Seiler, Hansjakob. 1974. The principle of concomitance: Instrumental, comitative, and collective. Foundations of Language 12:215–247.

    Save Citation »Export Citation »E-mail Citation »

    Using German data, proposes diagnostics for identifying comitatives and instruments, though both are marked by the preposition mit (with). Subsumes both under a general notion of concomitance.

    Find this resource:

  • Stolz, Thomas. 1996. Some instruments are really good companions—some not: On syncretism and the typology of instrumentals and comitatives. Theoretical Linguistics 23:113–200.

    Save Citation »Export Citation »E-mail Citation »

    Provides evidence for not reducing the comitative and instrument to a single role, based on their morphosyntactic realization in a large, areally-diverse sample of languages. Argues that instrument-comitative case syncretisms are an areal feature of European languages.

    Find this resource:

  • Stolz, Thomas, Cornelia Stroh, and Aina Urdze. 2006. On comitatives and related categories: A typological study with special focus on the languages of Europe. Berlin: Mouton de Gruyter.

    Save Citation »Export Citation »E-mail Citation »

    A comprehensive typological study of the notion of concomitance in terms of its meaning and its grammatical realization via cases and adpositions. Includes a corpus study of concomitance in Le Petit Prince (The Little Prince) and its translations into various languages of Europe.

    Find this resource:

  • Walmsley, John B. 1971. The English comitative case and the concept of deep structure. Foundations of Language 7:493–507.

    Save Citation »Export Citation »E-mail Citation »

    Taking a case grammar perspective, argues that there is no independent notion of comitative, but rather that comitative and instrument are different facets of the same semantic role. Makes this argument by reconsidering previously proposed diagnostics for these two roles.

    Find this resource:

Experiencer

The “experiencer” is the semantic role assigned to the sentient argument of predicates of perception, sensation, or, most often, psychological state. Bossong 1998 and Haspelmath 2001 shed light on the notion of experiencer using data from European languages. Few studies, however, focus on experiencers in their own right; rather, they typically discuss them in the context of the appropriate analysis of verbs of psychological state, so-called psych-verbs. Recognizing an experiencer role provides a way of capturing what experiencer-subject psych-verbs like fear and experiencer-object psych-verbs like frighten have in common despite their “flipped” argument realization, but a question then arises: why do arguments bearing the experiencer role have distinct morphosyntactic realizations? Croft 1993 addresses this question by suggesting there are distinct types of psych-verbs, whose experiencers occupy different positions in the causal chain representation of the event these verbs denote. Pesetsky 1995 presents results that dovetail with Croft’s proposals, arguing that frighten-type psych-verbs are covert causatives. Schlesinger 1992 argues that the experiencer of a fear-type psych-verb is no different from an agent. Landau 2010, instead, proposes a localist analysis of psych-verbs, where experiencers are analyzed as abstract locations. Experiencers also receive considerable discussion in Reinhart 2002, cited under Feature-Based Approaches.

  • Bossong, Georg. 1998. Le marquage de l’expérient dans les langues d’Europe. In Actance et valence dans les langues de l’Europe. Edited by Jack Feuillet, 259–294. Berlin: Mouton de Gruyter.

    DOI: 10.1515/9783110804485Save Citation »Export Citation »E-mail Citation »

    Notes that experiencers, like agents, are animates and, like patients, are affected. Proposes this duality is reflected in their argument realization options, and shows how these options are distributed across the languages of Europe, with areal preferences emerging.

    Find this resource:

  • Croft, William. 1993. Case marking and the semantics of mental verbs. In Semantics and the lexicon. Edited by James Pustejovsky, 55–72. Dordrecht, The Netherlands: Kluwer.

    DOI: 10.1007/978-94-011-1972-6Save Citation »Export Citation »E-mail Citation »

    Introduces the semantic subtypes of psych-verbs from the perspective of a causal model of events, showing how experiencers have a different place in the causal chain depending on the type of mental event involved. Points are supported using data from a variety of languages.

    Find this resource:

  • Haspelmath, Martin. 2001. Non-canonical marking of core arguments in European languages. In Non-canonical marking of subjects and objects. Edited by Alexandra Y. Aikhenvald, R. M. W. Dixon, and Masayuki Onishi, 53–83. Amsterdam: John Benjamins.

    Save Citation »Export Citation »E-mail Citation »

    Distinguishes experiencers of cognition, sensation, and emotion predicates based on differential preferences for their surface expression across European languages: experiencers of cognition predicates are most likely to be expressed like agents, experiencers of emotion predicates are most likely to be expressed like patients, and experiencers of sensation predicates are intermediate.

    Find this resource:

  • Landau, Idan. 2010. The locative syntax of experiencers. Cambridge, MA: MIT Press.

    Save Citation »Export Citation »E-mail Citation »

    Argues that experiencers should be taken to be locations and are syntactically instantiated in prepositional phrases, perhaps headed by null prepositions. Proposes this analysis will explain many of the puzzling properties of psych-verbs.

    Find this resource:

  • Pesetsky, David. 1995. Zero Syntax: Experiencers and cascades. Cambridge, MA: MIT Press.

    Save Citation »Export Citation »E-mail Citation »

    A study of the problematic behavior of psych-verbs, which comprehensively reviews previous work, and then proposes that causative experiencer-object psych-verbs such as frighten involve a zero causative morpheme, an analysis that has repercussions for their linguistic behavior.

    Find this resource:

  • Schlesinger, Izchak M. 1992. The experiencer as an agent. Journal of Memory and Language 31:315–332.

    DOI: 10.1016/0749-596X(92)90016-QSave Citation »Export Citation »E-mail Citation »

    Argues that the subject of experiencer-subject psych-verbs such as fear is better analyzed as an agent, since participants in several studies find it to have more control and intention than the object of experiencer-object psych-verbs such as frighten.

    Find this resource:

Benefactive

The benefactive, or beneficiary, role is assigned to the argument that benefits in some way from the action denoted by the verb. Drawing on corpus data, Fellbaum 2005 considers what should fall under the label “benefactive,” while Kittilä and Zúñiga 2010 brings crosslinguistic data to bear on this question. Zúñiga 2011 most directly confronts the notion of benefactive, proposing an event structure analysis that illuminates the paradoxical set of properties that characterize the benefactive role and result in similarities between this role and several other roles. Pylkkänen 2008 distinguishes between those benefactive arguments that are related to the patient/theme of an event and those that are related to the event as a whole. The benefactive role shows close ties to the goal and recipient roles discussed in Location, Source, Goal, Path, and Recipient, while other work subsumes the benefactive role under the goal or recipient role and other work subsuming recipients—animate goals––under the benefactive role, including Halliday 1967 (cited under Other Approaches). Folli and Harley 2006 argues for a benefactive role distinct from recipient and goal, as does Kittilä 2006 (cited under Location, Source, Goal, Path, and Recipient). Luraghi 2005 points out that benefactives also have affinities with noun phrases expressing the reason or purpose of events.

  • Fellbaum, Christiane. 2005. Examining the constraints on the benefactive alternation by using the World Wide Web as a corpus. In Linguistic evidence: Empirical, theoretical, and computational perspectives. Edited by Marga Reis and Stephan Kepser, 207–236. Berlin: Mouton de Gruyter.

    Save Citation »Export Citation »E-mail Citation »

    This corpus study of the English benefactive alternation (bake a cake for Pat/bake Pat a cake) includes considerable discussion of what a benefactive is and what verbs are found with benefactive arguments.

    Find this resource:

  • Folli, Raffaella, and Heidi Harley. 2006. Benefactives aren’t goals in Italian. In Romance languages and linguistic theory 2004. Edited by Jenny Doetjes and Paz González, 121–142. Amsterdam: John Benjamins.

    Save Citation »Export Citation »E-mail Citation »

    As its title implies, argues that benefactives should be distinguished from recipients and locations in Italian, although all three can be marked by the same preposition. Proposes that the three semantic roles are associated with three distinct syntactic configurational positions instantiated in a manner inspired by Pylkkänen 2008.

    Find this resource:

  • Kittilä, Seppo, and Fernando Zúñiga. 2010. Introduction: Benefaction and malefaction from a cross-linguistic perspective. In Benefactives and malefactives: Typological perspectives and case studies. Edited by Fernando Zúñiga and Seppo Kittilä, 1–28. Amsterdam: John Benjamins.

    Save Citation »Export Citation »E-mail Citation »

    An introduction to an edited collection of studies of the expression of benefactives and malefactives across languages. Considers the nature of the linguistic notion of benefactive by bringing data from its crosslinguistic expression to bear.

    Find this resource:

  • Luraghi, Silvia. 2005. Paths of semantic extension: From cause to beneficiary and purpose. In Historical linguistics 2003: Selected papers from the 16th International Conference on Historical Linguistics. Edited by Michael Fortescue, Eva Skafte Jense, Jens Erik Mogensen, and Lene Schøsler, 142–157. Amsterdam: John Benjamins.

    Save Citation »Export Citation »E-mail Citation »

    Argues that benefactives can not only be construed as recipients, but also as reasons (or purposes) of controlled events, giving rise to two types of case syncretisms involving benefactives, one involving the antecedent roles and the other the subsequent roles detailed in Croft 1991 (cited under Specific Semantic Roles).

    Find this resource:

  • Pylkkänen, Liina. 2008. Introducing arguments. Cambridge, MA: MIT Press.

    Save Citation »Export Citation »E-mail Citation »

    Distinguishes benefactives that relate an individual to an event from those that relate an individual to the verb’s object through a study of the properties of applicative constructions. Proposes a syntactic account in which the first type of benefactive is adjoined in clause structure higher than the second type.

    Find this resource:

  • Zúñiga, Fernando. 2011 Why should beneficiaries be subjects (or objects)?: Affaction and grammatical relations. In Case, animacy and semantic roles. Edited by Seppo Kittilä, Katja Västi, and Jussi Ylikoski, 329–348. Amsterdam: John Benjamins.

    Save Citation »Export Citation »E-mail Citation »

    Proposes a bi-eventive event structure analysis, which accounts for the characteristic properties of benefactives: their animacy, affectedness, and peripheral status in an event. Uses this analysis to explain why benefactives are sometimes realized as arguments and sometimes as adjuncts.

    Find this resource:

Location, Source, Goal, Path, and Recipient

Location, source, goal, and path roles have been posited for motion, location, putting, and removing events. These notions are at the heart of localist approaches to semantic roles, such as those discussed in Thematic Relations, and are extensively discussed in the references there. A point of contention is whether the recipient, that is, an animate goal, in a giving event should be subsumed under the notion of goal or taken to represent a distinct semantic role. The localist work cited under Thematic Relations takes recipients to be a type of goal, since giving events are analyzed as events of motion in a possessional semantic field. Rappaport Hovav and Levin 2008 argues instead for a notion of recipient that is distinct from goal. Blansitt 1988 also argues indirectly for this conclusion in surveying attested case syncretisms that involve locations, goals, and recipients. Kittilä 2006 argues for an independent notion of recipient based on data involving the expression of the notion of benefactive. See Benefactive for further consideration of the recipient and goal roles. Location and goal also figure prominently in some work on the thematic hierarchy; see Baker 1996 and Bresnan and Kanerva 1992 (cited under Thematic Hierarchies).

  • Blansitt, Edward L. 1988. Datives and allatives. In Studies in syntactic typology. Edited by Michael Hammond, Edith Moravcsik, and Jessica Wirth, 173–191. Amsterdam: John Benjamins.

    Save Citation »Export Citation »E-mail Citation »

    Presents a typological study of morphological case or adpositional syncretisms involving the expression of locations, goals, and recipients, showing that possible syncretisms are constrained by a hierarchy of semantic roles. Consequently, when a syncretism involves the location and recipient, it also includes the goal.

    Find this resource:

  • Kittilä, Seppo. 2006. On distinguishing between “recipient” and “beneficiary” in Finnish. In Grammar from the human perspective: Case, space and person in Finnish. Edited by Marja-Liisa Helasvuo and Lyle Campbell, 129–152. Amsterdam: John Benjamins.

    Save Citation »Export Citation »E-mail Citation »

    Illuminates the nature of the recipient role by contrasting it with the beneficiary role, using data from the expression of these two notions in Finnish. Contains considerable discussion of the semantic definitions of the two roles.

    Find this resource:

  • Rappaport Hovav, Malka, and Beth Levin. 2008. The English dative alternation: The case for verb sensitivity. Journal of Linguistics 44:129–167.

    Save Citation »Export Citation »E-mail Citation »

    Argues that there is a notion of recipient that is distinct from spatial goal, on the basis of data involving the English dative alternation (give Pat a book/give a book to Pat).

    Find this resource:

Other Roles

Over the years, numerous other semantic roles have been proposed, though few of them have had staying power. This section mentions a handful that are chosen because they have been widely recognized, or because they pick up on important semantic notions that do not fit comfortably under other semantic roles. Blansitt 1978 and Talmy 1985 propose a stimulus role for the non-experiencer argument of predicates involving psychological states, although later studies of these predicates raises questions about whether all such arguments bear the same semantic role; see Experiencer as well as the Oxford Bibliographies article Argument Structure. Perhaps the best known nontraditional semantic role is incremental theme, introduced in Dowty 1991. Incremental theme is an argument of a verb which plays a key part in the determination of telicity. Turning to less widely adopted roles, Wechsler 1995 proposes a notion role that figures in certain events of cognition and perception, while Hook 1983 introduces an abstrument role that figures in removing events.

  • Blansitt, Edward L., Jr. 1978. Stimulus as a semantic role. In Valence, semantic case, and grammatical relations. Edited by Werner Abraham, 311–326. Amsterdam: John Benjamins.

    Save Citation »Export Citation »E-mail Citation »

    After considering several semantic role assignment options, including patient, instrument, and source, argues for a semantic role of stimulus for the non-experiencer argument of both experiencer-subject (i.e., fear-type) and experiencer-object (i.e., frighten-type, psych-verbs); see Experiencer.

    Find this resource:

  • Dowty, David R. 1991. Thematic proto-roles and argument selection. Language 67:547–619.

    Save Citation »Export Citation »E-mail Citation »

    Argues for an incremental theme semantic role in section 6 (pp. 567–571), on the basis of its contribution to object selection. This role is assigned to the argument of a verb that defines a homomorphism from its spatial extent to the time course of the event.

    Find this resource:

  • Hook, Peter Edwin. 1983. The English abstrument and rocking case relations. In CLS 19: Papers from the 19th regional meeting of the Chicago Linguistic Society. Edited by Amy Chukerman, Mitchell Marks, and John F. Richardson, 183–194. Chicago: Chicago Linguistic Society.

    Save Citation »Export Citation »E-mail Citation »

    Argues for an abstrument semantic role for the thing removed from a surface—a role that could be considered to be the converse of the notion of source. Hook supports this position with data from English and other languages that involves the removal form of the locative alternation.

    Find this resource:

  • Talmy, Leonard. 1985. Lexicalization patterns: Semantic structure in lexical forms. In Language typology and syntactic description. Vol. 3, Grammatical categories and the lexicon. Edited by Timothy Shopen, 57–149. Cambridge, UK: Cambridge Univ. Press.

    Save Citation »Export Citation »E-mail Citation »

    Proposes the semantic role of stimulus for the non-experiencer argument of both experiencer-subject (i.e., fear-type) and experiencer-object (i.e. frighten-type, psych-verbs); see Experiencer.

    Find this resource:

  • Wechsler, Stephen. 1995. Semantic constraints on argument structure. In The semantic basis of argument structure. By Stephen Wechsler, 31–62. Stanford, CA: CSLI Publications.

    Save Citation »Export Citation »E-mail Citation »

    Argues for a notion semantic role to deal with beliefs of cognitive agents with verbs of mental state and perception, as well as certain verbs of necessarily volitional action.

    Find this resource:

Place in Linguistic Generalizations

Since the work of Pāṇini, semantic roles have been motivated by their ability to facilitate the statement of argument realization generalizations, with Fillmore 1968 (cited under Case Grammar: Development) reinforcing this point. Argument Realization considers work discussing the place of semantic roles in argument realization; further references may also be found in Thematic Hierarchies. Semantic roles have also figured in the statement of other linguistic generalizations, as reviewed in Other Phenomena. However, there has been significant disagreement about whether semantic roles do figure in the characterization of particular linguistic phenomena. Such disagreements can be traced to the difficulties in disentangling whether a phenomenon should be attributed to a noun phrase by virtue of its semantic role, its grammatical relation, its surface morphosyntactic manifestation in terms of case, agreement, or word order, or its contribution to the information structure of a sentence. Keenan 1976 (cited under Agent) first raised these issues with respect to agentivity and subjecthood. Comrie 1989 discusses them more generally in a textbook chapter. Dixon 1979 cited under Other Conceptions makes comparable points, as these same issues must be dealt with in a particularly complex instantiation in understanding phenomena that fall under the rubric of ergativity. Maling 2001 addresses these issues as they relate to the dative case. Evans 1997 raises a similar issue: the need to distinguish properties of a semantic role from properties of the filler of this role, with a focus on animacy. It is this confusion that is behind the discussion of agency and animacy in Folli and Harley 2006 (cited under Benefactive).

  • Comrie, Bernard. 1989. Theoretical prerequisites. In Language universals and linguistic typology. 2d ed. By Bernard Comrie, 57–85. Chicago: Univ. of Chicago Press.

    Save Citation »Export Citation »E-mail Citation »

    This chapter of a language typology textbook presents ways in which the relation of an argument to its verb can be characterized: in terms of (1) semantic role, (2) grammatical relation, (3) pragmatic or discourse function, and (4) morphological case. Comrie illustrates the interactions among these variations with studies of English and Russian.

    Find this resource:

  • Evans, Nick. 1997. Role or cast. In Complex predicates. Edited by Alex Alsina, Joan Bresnan, and Peter Sells, 397–430. Stanford, CA: CSLI Publications.

    Save Citation »Export Citation »E-mail Citation »

    Shows via an analysis of noun incorporation in Mayali that the relevant generalizations cannot be stated solely in terms of semantic roles, but must also make reference to whether the noun phrases bearing the roles meet the prototypical expectations about animacy; hence the notion of cast in the title.

    Find this resource:

  • Maling, Joan. 2001. Dative: The heterogeneity of the mapping among morphological case, grammatical functions, and thematic roles. Lingua 111:419–464.

    DOI: 10.1016/S0024-3841(00)00039-5Save Citation »Export Citation »E-mail Citation »

    Using data from three Germanic languages, Maling shows that the mappings among semantic roles, grammatical relations, and morphological case are more complex than often assumed. She argues that depictive predication, adnominal genitives, and middle formation are all sensitive to semantic roles.

    Find this resource:

Argument Realization

Despite their shortcomings, the staying power of semantic roles can most likely be attributed to their usefulness in the statement of argument realization generalizations. Even researchers who do not use semantic roles as theoretical constructs often use semantic role labels pretheoretically to refer to arguments of verbs. The use of semantic roles in argument realization can be traced back to the work of Pāṇini (see Approaches Antedating Fillmore’s “The Case for Case”), though in more current work it dates to Fillmore 1968. This paper establishes the use of semantic roles as a starting point for argument realization generalizations, introducing transformational rules for this purpose. As part of a larger effort to develop a transformational grammar of English, Stockwell, et al. 1973 offers explicit transformations similar to those in Fillmore 1968, but providing coverage of a larger set of data. Dowty 1991 posits subject and object selection rules formulated over lexical entailments associated with the notions of agent and patient. Croft 1991 proposes that subject and object selection identifies privileged participants in the causal chain representation of an event. Semantic roles figure in argument realization generalizations in two principal ways, as reviewed in Chapter 5 (pp. 131–153) of Levin and Rappaport Hovav 2005. This work distinguishes equivalence class–preserving approaches from prominence-preserving approaches, according to whether all instances of a semantic role are given the same argument realization options or whether the relative semantic prominence of a pair of semantic roles is reflected in the relative prominence of their morphosyntactic realizations. Prominence-preserving approaches are discussed in Thematic Hierarchies, as they typically use thematic hierarchies; some other relevant approaches are included here. A strong form of equivalence class preservation is embodied in the Uniformity of Theta Assignment Hypothesis, which is the topic of Baker 1997. This paper presents an instantiation of this hypothesis that is also consistent with prominence preservation. Newmeyer 2001 argues against the use of the Uniformity of Theta Assignment Hypothesis. Further references to the place of semantic roles in argument realization may be found in the Oxford Bibliographies article Argument Structure.

  • Baker, Mark C. 1997. Thematic roles and syntactic structure. In Elements of grammar: Handbook in generative syntax. Edited by Liliane Haegeman, 73–137. Dordrecht, The Netherlands: Kluwer.

    DOI: 10.1007/978-94-011-5420-8Save Citation »Export Citation »E-mail Citation »

    Investigates the consequences of the Uniformity of Theta Assignment Hypothesis, which requires arguments with the same semantic (proto-)role to map onto the same syntactic position. Examines unaccusatives, psych-verbs, ditransitives, thematic hierarchies, and grammatical relations. Argues agents are projected into the highest syntactic position, followed by themes, and then goals.

    Find this resource:

  • Croft, William A. 1991. Syntactic categories and grammatical relations. Chicago: Univ. of Chicago Press.

    Save Citation »Export Citation »E-mail Citation »

    Represents the event denoted by a verb as a causal chain relating the participants in the event. Proposes that the participants that delimit the causal chain, typically the agent and patient, are selected as subject and object.

    Find this resource:

  • Dowty, David R. 1991. Thematic proto-roles and argument selection. Language 67:547–619.

    Save Citation »Export Citation »E-mail Citation »

    Introduces subject and object selection rules that determine which argument of a transitive verb will be its subject and which one its object on the basis of the number of agent and patient proto-role entailments associated with each argument.

    Find this resource:

  • Fillmore, Charles J. 1968. The case for case. In Universals in linguistic theory. Edited by Emmon Bach and Robert T. Harms, 1–88. New York: Holt, Rinehart, and Winston.

    Save Citation »Export Citation »E-mail Citation »

    Introduces transformational rules that make reference to semantic roles that determine a verb’s subject and object from a configurational syntactic structure in which the verb takes each of its arguments identified by semantic role as sisters. Reprinted with a foreword by the author in his Form and meaning in language. Vol. 1, Papers on semantic roles (Stanford, CA: CSLI Publications), 23–122.

    Find this resource:

  • Levin, Beth, and Malka Rappaport Hovav. 2005. The mapping from lexical semantics to syntax. In Argument realization. By Beth Levin and Malka Rappaport Hovav, 131–153. Cambridge, UK: Cambridge Univ. Press.

    DOI: 10.1017/CBO9780511610479.006Save Citation »Export Citation »E-mail Citation »

    Surveys approaches to the semantics/syntax mapping. Divides them into those whose mappings preserve semantic equivalence classes (e.g., where the semantic role of an argument determines its realization) and those whose mappings preserve semantic prominence relations among arguments (e.g., where the mapping respects a thematic hierarchy).

    Find this resource:

  • Newmeyer, Frederick J. 2001. Grammatical functions, thematic roles, and phrase structure: Their underlying disunity. In Objects and other subjects: Grammatical functions, functional categories and configurationality. Edited by William D. Davies and Stanley Dubinsky, 53–75. Dordrecht, The Netherlands: Kluwer.

    DOI: 10.1007/978-94-010-0991-1Save Citation »Export Citation »E-mail Citation »

    Newmeyer reviews the motivations for positing that each semantic role is associated with a specific underlying syntactic position, a proposal that follows from the Uniformity of Theta Assignment Hypothesis. He then argues against this proposal.

    Find this resource:

  • Stockwell, Robert P., Paul Schachter, and Barbara H. Partee. 1973. The syntactic structures of English. New York: Holt, Rinehart and Winston.

    Save Citation »Export Citation »E-mail Citation »

    Inspired by Fillmore 1968, Chapter 2 (pp. 32–64) presents transformations that convert an underlying syntactic representation in which a verb takes its arguments, identified by semantic role, as sisters into a surface syntactic representation. Chapter 12 (pp. 717–810) provides a sample lexicon with entries containing the semantic roles associated with various verbs.

    Find this resource:

Other Phenomena

Semantic roles have been implicated in generalizations involving a range of linguistic phenomena that go beyond argument realization. Thus, Ladusaw and Dowty 1988 suggests that they figure in control, and Everaert and Anagnostopoulou 1997 proposes that via a thematic hierarchy they play a part in reflexivization, a point also made in Jackendoff 1972 (cited under Thematic Relations). Maling 2001 (cited under Place in Linguistic Generalizations) argues that semantic roles are implicated in middle formation, nominalizations, and depictive secondary predication. Nevertheless, in almost every instance where semantic roles have been implicated, another generalization that does not involve a reference to semantic roles has been proposed. In some instances these alternative generalizations appeal to alternative semantic notions. Thus, Tenny 1992 argues that a range of phenomena that have implicated affectedness––the semantic property underlying the patient role—actually involve telicity; see also Beavers 2011 (cited under Patient, Theme, and Affectedness). In other instances, generalizations said to involve semantic roles are shown to have a nonsemantic basis, as foreshadowed in Place in Linguistic Generalizations. For instance, Anderson 1977 suggests that the effects of certain lexical operations, including derivational morphology, on argument structure can be characterized with reference to semantic roles. Williams 1981 pursues this line of argument more generally, while Levin and Rappaport 1986 uses adjectival passive formation to argue that the effects of operations on argument structure are best stated without reference to semantic roles. For more on the nature of operations on argument structure, see the Oxford Bibliographies article Argument Structure.

  • Anderson, Stephen R. 1977. Comments on the paper by Wasow. In Formal syntax. Edited by Peter Culicover, Thomas Wasow, and Adrian Akmajian, 361–377. New York: Academic Press.

    Save Citation »Export Citation »E-mail Citation »

    Argues that certain lexical operations affecting the realization of the arguments of the verb, such as the causative alternation and object-oblique alternations, as well as certain derivational operations such as adjectival passive formation, require reference to semantic roles, with the theme semantic role figuring particularly prominently.

    Find this resource:

  • Everaert, Martin, and Elena Anagnostopoulou. 1997. Thematic hierarchies and binding theory: Evidence from Greek. In Empirical issues in formal syntax and semantics. Edited by Francis Corblin, Carmen Dobrovie-Sorin, and Jean-Marie Marandin, 43–59. Bern: Peter Lang.

    Save Citation »Export Citation »E-mail Citation »

    Argues that the thematic hierarchy plays a part in local anaphor binding in Greek.

    Find this resource:

  • Ladusaw, William A., and David R. Dowty. 1988. Toward a nongrammatical account of thematic roles. In Thematic relations. Edited by Wendy Wilkins, 62–73. Syntax and Semantics 21. New York: Academic Press.

    Save Citation »Export Citation »E-mail Citation »

    Argues the controller can be identified based on lexical entailments of verbs in conjunction with nonlinguistic principles of human action. Semantic roles may appear to figure in a theory of control, but only because they may label certain sets of entailments that are critical to a theory of control.

    Find this resource:

  • Levin, Beth, and Malka Rappaport. 1986. The formation of adjectival passives. Linguistic Inquiry 17:623–661.

    Save Citation »Export Citation »E-mail Citation »

    Shows that a closer examination of the data reveals that adjectival passive formation does not make reference to semantic roles, but rather to notions defined over argument structure such as external argument.

    Find this resource:

  • Tenny, Carol L. 1992. The Aspectual Interface Hypothesis. In Lexical matters. Edited by Ivan A. Sag and Anna Szabolcsi, 1–27. Stanford, CA: CSLI Publications.

    Save Citation »Export Citation »E-mail Citation »

    Argues that the semantic notion that matters to argument realization broadly construed is the aspectual notion of measuring out, a cousin of the incremental theme of Dowty 1991 (cited under Other Roles). Substantiates this proposal by showing that phenomena attributed to affectedness are better characterized in terms of measuring out.

    Find this resource:

  • Williams, Edwin. 1981. Argument structure and morphology. Linguistic Review 1:81–114.

    Save Citation »Export Citation »E-mail Citation »

    Considers how a verb’s argument structure carries over to morphologically related words. Takes argument structure to be a list of arguments, labeled by semantic roles, with a distinguished external argument. The effects of derivational morphology on a verb’s argument structure are stated in terms of semantic roles.

    Find this resource:

Thematic Hierarchies

Semantic roles often figure in the characterization of linguistic phenomena, particularly argument realization, through the mediation of a theoretical construct known as the “thematic hierarchy,” a ranking of semantic roles. Subject and object selection rules, for instance, are stated in terms of positions in the thematic hierarchy rather than in terms of particular semantic roles, allowing for the formulation of more general rules with wider applicability. Levin and Rappaport Hovav 2005 thoroughly discusses the thematic hierarchy as a theoretical construct, as well as its place in theories of argument realization, and also provides considerable references to other work. The thematic hierarchy is introduced in Jackendoff 1972, although the subject selection rule in Fillmore 1968 (cited under Case Grammar: Development can be seen as implicitly making reference to such a hierarchy. Dik 1997 shows how the thematic hierarchy can be used to state typological generalizations. There has been considerable controversy over the appropriate statement of the thematic hierarchy. One instantiation of this controversy involves the relative ranking of the theme, goal, and location roles, with Baker 1996 and Bresnan and Kanerva 1992 representing different positions on the appropriate ranking. Some work proposes that the thematic hierarchy is an artifact; thus, Dowty 1991 (cited under Proto-Roles) proposes that it reflects properties of certain recurring clusters of agent proto-role entailments, whereas Croft 1998 proposes that it is better replaced by force dynamic relations among arguments. Croft 1998 and Gee 1974, as well as Hust and Brame 1976 (cited under Thematic Relations) and Davis and Koenig 2000 (cited under Proto-Roles), provide critiques of thematic hierarchies.

  • Baker, Mark C. 1996. On the structural positions of themes and goals. In Phrase structure and the lexicon. Edited by Johan Rooryck and Laurie Zaring, 7–34. Dordrecht, The Netherlands: Kluwer.

    DOI: 10.1007/978-94-015-8617-7Save Citation »Export Citation »E-mail Citation »

    Argues that themes should be ranked above locations, goals, and benefactives in a thematic hierarchy in order to account for the unavailability of an unaccusative analogue of the dative alternation.

    Find this resource:

  • Bresnan, Joan, and Jonni Kanerva. 1992. The thematic hierarchy and locative inversion in UG: A reply to Paul Schachter’s comments. In Syntax and the lexicon. Edited by Timothy Stowell and Eric Wehrli, 111–125. Syntax and Semantics 26. New York: Academic Press.

    Save Citation »Export Citation »E-mail Citation »

    This reply to a reply to an earlier paper by the authors argues for a universal thematic hierarchy in which recipients and locations are distinguished, with recipient ranked above and location below patient/theme.

    Find this resource:

  • Croft, William A. 1998. Event structure in argument linking. In The projection of arguments: Lexical and compositional factors. Edited by Miriam Butt and Wilhelm Geuder, 21–63. Stanford, CA: CSLI Publications.

    Save Citation »Export Citation »E-mail Citation »

    Proposes a theory of argument realization based on causal chains that updates Croft 1991 (cited under Sources of Semantic Roles); in so doing, Croft argues that it is better to make reference to the force-dynamic relationships among arguments in argument realization than to their relationships on a thematic hierarchy.

    Find this resource:

  • Dik, Simon C. 1997. The theory of functional grammar. Part 1, The structure of the clause. 2d ed. Edited by Kees Hengeveld. Functional Grammar Series 20. Berlin: Mouton de Gruyter.

    Save Citation »Export Citation »E-mail Citation »

    Introduces functional grammar’s Semantic Function Hierarchy, basically a thematic hierarchy. Uses it to capture generalizations about the range of possible semantic roles that can be associated with subject and object across languages. Includes a useful general discussion of hierarchies and markedness. See especially chapters 2, 10, 11, and 16.

    Find this resource:

  • Gee, John Paul. 1974. Jackendoff’s Thematic Hierarchy Condition and the passive construction. Linguistic Inquiry 5:304–308.

    Save Citation »Export Citation »E-mail Citation »

    This squib presents passivization and reflexivization data that does not conform to the predictions of the account of these phenomena that appeals to a thematic hierarchy presented in Jackendoff 1972.

    Find this resource:

  • Jackendoff, Ray S. 1972. Semantic interpretation in generative grammar. Cambridge, MA: MIT Press.

    Save Citation »Export Citation »E-mail Citation »

    The first explicit introduction of the notion of a thematic hierarchy, which is motivated through its contribution to the explanation of passive and reflexivization facts.

    Find this resource:

  • Levin, Beth, and Malka Rappaport Hovav. Thematic hierarchies in argument realization 2005. In Argument realization. By Beth Levin and Malka Rappaport Hovav, 154–185. Cambridge, UK: Cambridge Univ. Press.

    DOI: 10.1017/CBO9780511610479.007Save Citation »Export Citation »E-mail Citation »

    Provides a critical overview of the considerable work that makes reference to thematic hierarchies, particularly as they figure in subject and object selection, and attributes the controversy over the correct statement of the hierarchy to the different goals and theoretical orientations of previous work.

    Find this resource:

Use in Computational Linguistics

Computational linguists have been interested in semantic roles as a way of encoding the relations between predicates, particularly verbs, and their arguments, using them in various natural language processing applications ranging from machine translation to information extraction to sentiment analysis. Significant efforts have been devoted in recent years to semantic role labeling—the automatic identification of the semantic roles of the arguments of a verb or other predicate using sophisticated statistical machine learning techniques. These efforts require a training data set that is already labeled for semantic roles, and the Berkeley FrameNet project described in Baker, et al. 1998 offers such a training corpus, which is exploited in Gildea and Jurafsky 2002, the first largely successful effort at semantic role labeling. FrameNet itself is inspired by Fillmore’s frame semantics, a descendant of his case grammar, envisioned in Fillmore 1977b (cited under Case Grammar: Development). FrameNet is designed so that its semantic roles are relativized to particular semantic frames, which could each apply to several semantically related verbs. In contrast, another much used training data set, PropBank, presented in Palmer, et al. 2005, was developed to provide verb-specific semantic role labeling. A special issue of Computational Linguistics (Màrquez, et al. 2008) brings together articles describing influential semantic role labeling efforts. Palmer, et al. 2010 provides an extensive review of all facets of the semantic role labeling task.

  • Baker, Collin F., Charles J. Fillmore, and John B. Lowe. 1998. The Berkeley FrameNet project. In COLING-ACL ’98: 36th Annual Meeting of the Association for Computational Linguistics and 17th International Conference on Computational Linguistics, Aug. 10–14, 1998, Montreal, Quebec, Canada; Proceedings of the conference. Edited by Christian Boitet and Pete Whitlock, 86–90. Stroudsburg, PA: Association for Computational Linguistics.

    Save Citation »Export Citation »E-mail Citation »

    Introduces a large-scale lexicography effort that provides abstract semantic frames, each describing an event type and the semantic roles of its participants. The applicability of the frames is demonstrated via the annotation of numerous sentences in the British National Corpus with frame elements.

    Find this resource:

  • Gildea, Daniel, and Daniel Jurafsky. 2002. Automatic labeling of semantic roles. Computational Linguistics 28:245–288.

    DOI: 10.1162/089120102760275983Save Citation »Export Citation »E-mail Citation »

    Presents the first automatic semantic role labeling system. Shows how the system can assign the domain-specific semantic roles found in FrameNet frames, and then shows how it can also assign coarser-grained semantic roles that more closely approximate typical semantic roles.

    Find this resource:

  • Màrquez, Lluís, Xavier Carreras, Kenneth C. Litkowski, and Suzanne Stevenson, eds. 2008. Special issue: Semantic role labeling. Computational Linguistics 34.2.

    Save Citation »Export Citation »E-mail Citation »

    Collects papers describing influential and representative semantic role labeling efforts.

    Find this resource:

  • Palmer, Martha, Daniel Gildea, and Nianwen Xue. 2010. Semantic role labeling. Synthesis Lectures on Human Language Technologies 6. San Rafael, CA: Morgan and Claypool.

    Save Citation »Export Citation »E-mail Citation »

    Provides an overview of major facets of the semantic role labeling task, including the linguistic background, the training corpora available, the principles behind automatic systems, and recent advances in such systems. While the initial chapters focus on English, the final chapter considers other languages, using Chinese as an illustration.

    Find this resource:

  • Palmer, Martha, Paul Kingsbury, and Daniel Gildea. 2005. The Proposition Bank: An annotated corpus of semantic roles. Computational Linguistics 31:71–106.

    DOI: 10.1162/0891201053630264Save Citation »Export Citation »E-mail Citation »

    Introduces PropBank, a corpus that adds semantic role labels to the arguments of all verbs in the much used Penn TreeBank. Discusses the semantic roles chosen, and describes a semantic role labeling system trained on the corpus and the effects of several types of information on its performance.

    Find this resource:

LAST MODIFIED: 01/13/2014

DOI: 10.1093/OBO/9780199772810-0141

back to top

Article

Up

Down