LINGUIST List 24.833

Fri Feb 15 2013

Review: Semantics; Syntax; Typology: Suihkonen et al. (2012)

Editor for this issue: Monica Macaulay <>

Date: 13-Jan-2013
From: Roxana Popescu <>
Subject: Argument Structure and Grammatical Relations
E-mail this message to a friend

Discuss this message

Book announced at

EDITOR: Pirkko SuihkonenEDITOR: Bernard ComrieEDITOR: Valery SolovyevTITLE: Argument Structure and Grammatical RelationsSUBTITLE: A crosslinguistic typologySERIES TITLE: Studies in Language Companion Series 126PUBLISHER: John BenjaminsYEAR: 2012

REVIEWER: Roxana Luliana Popescu, University of Bacău


“Argument Structure and Grammatical Relations: A cross linguistic typology” isa volume of papers from a symposium on this topic. It comprises 14 articles,discussing different aspects of argument structure and grammatical relationsapplied to languages from all over the world (Europe, North and Central Asia,and western North America). After a concise introduction by one of thevolume’s editors, Bernard Comrie, there is a series of papers on the notion ofalignment typology, case marking (nominative-accusative), ergative verbs,agent-patient noun phrase arguments, typology of grammatical relations andmany other notions directly or indirectly related to the topic of the volume.

In his paper “A deceptive case of split-intransitivity in Basque”, AsierAlcázar focuses on the asymmetry of Basque unergatives and transitives basedon the absolute construction and reduced relative clauses, questioning theuniversality of unergatives as syntactically intransitive. Generally,unergative verbs are considered intransitive in Basque and they are eitherverbs (objectless) or light verbs (their object is separable from ‘egin’‘do’). However, due to the fact that Basque uses a range of morphemes forcase marking, person indication and number agreement with some intransitives(although generally associated with transitivity), the author demonstratesthat split-intransitives tend to become ergative.

Bernard Comrie brings into discussion “Some argument structure properties ofthe verb ‘to give’ in the languages of Europe and Northern and Central Asia”.Case marking and other morphosyntactic properties are analyzed inmonotransitive and ditransitive constructions, focusing on the non-Agentarguments, such as Recipient, Theme, and Patient. Presenting examples frommore than 33 different languages, the author concludes that the indirectobject (the Recipient) is more frequent in ditransitive constructions with theverb ‘give’. He brings as an argument the ‘syntactic primacy’ in aditransitive construction of non-Agent arguments, considering that it maysometimes match case marking alignment. However, the author demonstrates thatthe Recipient is mostly preferred over the Theme, on the one hand, due to theproximity to the verb, on the other hand semantically. The author shows thatthe translation of the word ‘to give’ can be influenced by the features of therelation of the Recipient argument with the speaker in terms of person/numberand power/solidarity. Therefore, syntactic primacy is also given a semanticchoice, the studied languages offering plenty of examples to show thepopularity of the Recipients over the other objects.

Greville G. Corbett, in “Grammatical relations in a typology of agreementsystems”, argues that grammatical relations are not sufficient to determine atypology of agreement relations. Thematic roles and verbal semantics,communicative functions and surface case also help identify the type ofagreement. After presenting a theoretical framework on grammatical relationsbased on the Relational Hierarchy (Johnson 1977) and the AccessibilityHierarchy (Keenan & Comrie 1977), the author mentions some problems withgrammatical relations which restrict agreement, such as dummy subjects(English ‘there’), backward agreement/attraction (copula and the noun phrasein the predicate), and possessive constructions. The paper proceeds bydiscussing to what extent all the other features combined (thematic roles,communicative functions and case) help in establishing agreement. Comrie’s(1975) and Robblee’s (1993) Predicate Hierarchies are further applied to someRussian prose works to analyze agreement according to thematic roles whichinfluence it. In addition, the author suggests that languages havecommunicative functions that influence agreement and that the subject does notnecessarily control agreement, especially when the subject is focused and notthe topic. There can also be object agreement, when the topic is emphasized.Further, the author brings into discussion cases, such as nominative, genitiveand accusative that influence number and gender agreement by giving theresults of a study on 415 examples from literary Russian texts. To sum up, thepaper concludes that grammatical relations do not account for agreemententirely, but they also depend on the meaning of the clause and theircommunicative functions.

Michael A. Daniel, Timur A. Maisak and Solmaz R. Merdanova discuss the issueof “Causatives in Agul”. After introducing Agul with its sociolinguisticsituation, its genetic affiliation and its grammar, the authors divide theirpaper into 3 sections: section 2 describes formal properties of causativeverbs and productive constructions (e.g. infinitives with the verb ‘aq’as’‘do’ and periphrastic ‘do’ causatives), and non-productive constructions (e.g.‘do’ causatives, labile verbs and lexical causatives); section 3 describessemantic features of Agul causatives and contrasts between the attestedpatterns; and section 4 comprises some typological background and highlightsthe common and uncommon features of Agul causatives. The authors give atypology of the dichotomy between the non-productive and periphrasticcausatives. The discrimination of semantic roles of the Causer and the Causeeare debated as part of the causative constructions (direct or indirect) andcase markers. Finally, there is an overview and an East Caucasian perspective,periphrastic causatives being considered intermediate between bi- andmonoclausal constructions.

Andrei Filchenko, in “Continuity of information structuring strategies inEastern Khanty: Definiteness/topicality”, discusses how discourse is renderedcohesive based on grammatical relations. He analyzes information structuringboth pragmatically and semantically by means of elision, word order,agreement, case, voice and possessive markers, applying all these to corpusdata from Khanty, a Finno-Ugric language. His paper shows that there is astrong correlation between reduced morphological complexity and discoursereferents, which are pragmatically dominant and focuses on the possessivemarkers, as pragmatic markers of the referents. The author notes the frequencyof objective/definite vs. subjective/indefinite conjugations in Khanty,understood as involving the formal properties of the Object (O) Argument, suchas: possessive constructions, pronouns, demonstratives, embedded clauses,elided/zero objects. The position of the argument in O (transitivenon-Subject) may vary, due to its pragmatic properties: a fixed SOV positionfor an unidentifiable Object referent or variations (either OSV or SVO) for anidentifiable and pragmatically active Object referent. Embedded non-finiteclauses are also identified in Khanty as subject-controlled, but this is notmandatory (for example some adverbial, complement or relative clauses do nothave a coreferential S of the non-finite clause with the S of the matrixclause). This choice depends on the pragmatic properties of the referent(identifiability and activation). Thus, the author demonstrates that the mainstrategies for information structuring in Eastern Khanty are based onpragmatic features: identifiability, activation, definiteness and topicality.

John A. Hawkins, in “Patterns of asymmetry in argument structure acrosslanguages: Some principles and puzzles”, discusses the co-occurrence of ruleapplicability asymmetries, formal marking asymmetries and linear orderingasymmetries, described in terms of hierarchies which can be explained on theone hand according to processing and performance, and on the other hand bydeclining levels of frequency and accessibility. The author discusses thequestions raised by hierarchies and their correlating properties: favoringsyntactic and morphologic rules for higher positions (e.g. verb agreement withNominative before Accusative); zero marking preferred for higher positions;linear ordering related to hierarchy positions or case hierarchy. The authorshows that the hierarchies among the arguments of a predicate depend on theprinciples of complexity, citing different views on this topic in theliterature, such as Hawkins (2004), Blake (1990), Primus (1993, 1995, 1999).

Taeho Jang and Thomas E. Payne, in “Topic marking and the construction ofnarrative in Xibe”, present a corpus-based study on the functions of the topicmarker ‘da’ in spoken Xibe (a language spoken in Northwestern China),identifying the instances where it occurs syntactically as ‘candidatepositions’. They illustrate these syntactic environments, such as: following aclause-initial Noun Phrase; following a clause-initial adverb; following thesentence-initial conjunctives (‘tumake’ ‘and then’ and ‘dam’ ‘however’); andbetween clauses, following one of three converb endings or a perfective aspectending. Structurally, these are the contexts in which ‘da’ may occur, but itis never obligatory. Discourse factors are also discussed in terms of thechoice of the topic marker ‘da’: expressing temporal sequence or addition. Theauthors argue for the importance of this marker in building narrativediscourse or characterizing a specific genre, register and style. They arguethat the notion of limitation connects all its usages and identify thetendency among proficient speakers to use it in order to link clauses intemporal sequence forming ‘clause chains’.

Juha A. Janhunen, in “On the hierarchy of structural convergence in the AmdoSprachbund”, offers a diachronic perspective on alignment systems. The authorpresents arguments in terms of the structural features of the languages of theAmdo Sprachbund belonging to four linguistic stocks: Turkic, Mongolic, Siniticand Bodic, all being influenced by interactive adaptations and sharing manyproperties at all levels (phonological, morphological, syntactic andpragmatic). However, there is a distribution of these features according tosome typological spheres, such as: the Altaic (Turko-Mongolic), the Sinitic(Chinese) and the Bodic (Tibetan). The author contrasts them to find out whichmakes up the general substratum and which is more restrictive in use, tryingto establish the language boundaries. The paper also deals with a basicdistinction that “has not crossed the language boundaries in Amdo Sprachbund”(p. 184), namely the difference between the accusative and ergative strategiesof argument structure. The author notices that the structural markers in theselanguages have the tendency to be identical with the genitive markers in otherlanguages. This genitive-ergative feature of Amdo Tibetan is also considered a‘connective’ occurring both adnominally and adverbially. The author concludesthat there are structural features which are more easily borrowed or lost, andother structures remaining as such (ergativity and nominal phrase word order).

Lars Johanson, in “Pyramids of spatial relators in Northeastern Turkic and itsneighbors”, deals with basic topological relations at different stages(older/recent) of the languages spoken in Siberia and Mongolia. The authorpresents an overview of the classes which combine with the nominals and areadded to predicates as spatial markers. He distinguishes between non-dynamicand dynamic situations for spatial relators expressing on the one hand,location (referring to a place: ‘in’, ‘at’, ‘on’) and translocation (‘along’,‘through’ a situation), and on the other hand, adlocation (‘motion towards’)and ‘delocation’ (‘motion away from’). The stages of historical developmentsof these basic spatial relators are further presented as a pyramid-likefigure, each changing or altering the preceding stage, with a slope from thegeneral to the particular. At each level he includes the following items withexemplifications from the studied languages: (A) markerless constructions(nominals with intransitive verbs); (B) simple case suffixes (simple localcase markers added as suffixes to nominals); (C) composite case suffixes; (D)simple postpositions (mostly converb forms added); (E) complex postpositions(nominal phrases with spatial meaning: ‘interior’, ‘back’, ‘side’, ‘bottom’,etc.). However, the author demonstrates that the diachronic relations betweenthe levels do not offer a chronological development: items at a lower levelmay be older than others. The paper concludes that the older markers developedinto more refined markers in terms of semantic content, spatial relatorsacquiring dynamic and non-dynamic interpretations according to the movementcharacter of the predicate verb which can be inherited by a language.

Andrej A. Kibrik, in “What’s in the head of head-marking languages?”, focuseson phenomena that identify the semantic roles in a clause, claiming that thehead-marking technique is functionally equivalent to nominal cases rather thanto grammatical relations (subject, direct object, etc.). The author arguesthat head-marking languages combine the typology of locus of marking and thetypology of argument type that considers personal affixes on the verb aspossible pronominal arguments. He also identifies the semantic roles in thelanguages studied as marked by linear positions in the morphological structureof the verb in which pronominal elements are inserted. These positionsrepresent functional correspondents of case affixes in dependent-markinglanguages, known and labeled as cases. He concludes that in head-markinglanguages grammatical relations and role marking tend to overlap due to therelative linear order of morphemes.

Hee-Soo Kim, in “Transitives, causatives and passives in Korean and Japanese”,analyzes the structures enumerated in the title in terms of ‘event control’(EC) and ‘root event’ (RE), as well as the status of noun phrases in rootevents. After mentioning the interest in the similarities and dissimilaritiesbetween these concepts by different linguists (Dowty 1979, Jackendoff 1972,Grimshaw 1990, etc.), the author draws attention to the parallelism betweentransitive, causative and passive structures, based on Korean and Japanese.The author deals with semantic roles in terms of event/root control which comeeither from the subject, from both the subject and object/non-subject or justfrom the object/non-subject. Different types of causatives and thesimilarities or differences between causative and transitive/passive areprovided, however noting some ambiguity for Korean and Japanese. Furthermore,the semantic roles of the noun phrases involved in a sentence are identifiedas agent/patient (in transitive structures), causer/causee (in causativestructures) and patient/non-argument and agent (in passive structures) andcompared with some English structures. The author demonstrates that allsemantic roles depend more on event control than on root control.

Marianne Mithun, in “Core argument patterns and deep genetic relations:Hierarchical systems in Northern California”, argues that core argumentmarking is directly connected to deep genetic relationships and stronglyresistant to areal influence. The author offers a comparison of thehierarchical systems in four California languages as evidence for borrowing.However, the author suggests that these languages differ in the strategiesused to avoid low ranking arguments and lower potential ambiguity. The authoridentifies differences in the bases of the systems (agent/patient versussubject/object), in the manners for maintaining them (passivization, omissionof arguments, alternative pronominal shapes) and in their degree ofpenetration into grammars. Thus, she suggests the structural features of theselanguages are not descendants of a fully formed hierarchy system, but theyhave been stimulated by language contact.

Anna Siewierska and Dik Bakker, in “Three takes on grammatical relations: Aview from the languages of Europe and North and Central Asia”, discussdifferent approaches to grammatical relations (GRs), found in FunctionalGrammar, Relational Typology and Role and Reference Grammar, all being labeledas functional-typological. The authors suggest that GRs are not universal,based on the common considerations of these approaches. The paper alsopresents different typologies within these approaches: directly related to GRsand related to various languages in terms of the presence, nature and strengthof GRs. All these typologies are applied to languages in Europe, Northern andCentral Asia. However, all of the languages considered (in Europe and Northernand Central Asia) share nominative/accusative subjects, in spite of the factthat subjects are variable (in Europe) or not (in Northern and Central Asia).The result is that these typologies distinguish languages in Europe fromNorthern and Central Asia, and the presence of these GRs is considered a normin Europe but an exception in North and Central Asia.

Pirkko Suihkonen, in “On aspect, aspectual domain and quantification inFinnish and Udmurt”, offers a case study on quantification and aspectualdomain. Providing examples from Finnish and Udmurt, the author claims that thechoice of aspect depends on argument structure, which is, in its turn, relatedto event structures. On the one hand, argument structure adds information tothe structure-building elements; on the other hand, event structure involvesstructure-building operations such as passivization or nominalization.According to the author, lexical predicates, different derivational suffixesand quantifying adverbs are mutually influential in terms of the aspectualdomain of sentences, the derivation of verbs being used to bind a sentence ina context. The verbal semantic changes influence quantification and aspect.The author also embraces Dowty’s approach (1979, 1989) to semantic structuresof predicates and sentence types, which suggests a method of formalizing thesemantics based on Montague Grammar. Finally, the author concludes that thereare no clear boundaries between lexical and grammatical means used to markaspect and quantification.


The papers in this volume interrelate in terms of scientific content on thetopic of argument structure and grammatical relations. All the papers areconsistent in volume and supported by a wide range of bibliographicalreferences, using genuine exemplifications for the theoretical issuesdiscussed. The theoretical framework of each composition offers clearexplanations for understanding their applied studies on language.

This book addresses linguists with experience in the areas of argumentstructure and grammatical relations, and introduces details on languages moreor less familiar to its readers. The research on argument structurecomplements that on grammatical relations. This volume is to be appreciatedfor providing application of the main theoretical issues, presented throughoutthe entire volume, which makes its chapters cohere. This shows that it ispossible to have a common theoretical pattern for argument structure appliedto all languages, but still with some geographically-related linguisticdissimilarities.

The thematic hierarchy is also addressed because it allows for reference toarguments in terms of their relative ranking, e.g., the argument bearing the‘highest’, ‘second highest’, or ‘lowest’ ranked semantic role, obviating theneed to refer directly to arguments by semantic role.

In linguistic theory, thematic roles have traditionally been regarded asdeterminant in expressing generalizations about the syntactic realization of apredicate’s arguments. Most characterizations of thematic roles have beencarried out in terms of primitive semantic properties of predicates. Forexample, Grimshaw (1990: 2-6) argues for “a theory where it is not the rolesthemselves that are important, but their place in a thematic hierarchy. Alexical entry consists of a list of arguments, with no thematic information;the thematic roles can be deduced from the lexical meaning of the head (viaits lexical semantic structure). The labelling of roles does not, in fact,play a role at any stage; rather it is their place in the thematic rolehierarchy which is important.” The hierarchical ordering of thematic roles isone part of what she calls argument structure, the other part being theaspectual structure which divides an event into parts, therefore showing aclose connection between argument structure and grammatical relations.

The perspective this volume’s topic opens is towards a logical interpretationof argument structures and grammatical relations (see Ioan 1995, Montague’scollected papers in Thomason 1974). Case grammar and actantial grammar takelogic closer to the structure and functionality of natural language. Theformal analyses of the predicational and relational logic have their semanticand pragmatic origin in the distinctions of grammar that place syntax higherthan morphology and guide interest in deep structures.


Blake, B.J. 1990. Relational Grammar [Croom Helm Linguistic Theory Guides].London: Croom Helm.

Comrie, B. 1975. Polite plurals and predicate agreement. Language 51: 406-418.

Dowty, D. 1979. Word Meaning and Montague Grammar. The Semantics of Verbs andTimes in Generative Semantics and in Montague’s PTQ [Synthese Language Library7]. Dordrecht: Reidel.

Dowty, D. 1989. On the semantic content of the notion of ‘Thematic role’. InProperties, Types and Meaning II, Semantic Issues [Studies in Linguistics andPhilosophy 39], G. Chierchia, B.H. Partee & R. Turner (eds.), 69-129.Dordrecht: Kluwer.

Grimshaw, Jane. 1990. Argument Structure. Cambridge MA: MIT Press.

Hawkins, J.A. 2004. Efficiency and Complexity in Grammars. Oxford: OUP.

Ioan, Petru. 1995. Logical Horizons. Perspectives and meaning readjustments inthe present universe of formalisms. (Romanian version). Bucharest: Didacticand Pedagogic Publishing House.

Jackendoff, Ray S. 1972. Semantic Interpretation in Generative Grammar.Cambridge MA: MIT Press.

Johnson, D.E. 1977. On relational constraints on grammars. In Syntax andSemantics 8: Grammatical relations, P. Cole & J.M. Sadock (eds.), 157-178. NewYork: Academic Press.

Keenan, E.L. & Comrie, B. 1977. Noun phrase accessibility and universalgrammar. Linguistic Inquiry 8: 63-99.

Primus, B. 1993. Syntactic relations. In Syntax, Vol.1. J. Jacobs, A. vonStechow, W. Sternefeld & T. Vennemann (eds.), 686-705. Berlin: Walter deGruyter.

Primus B. 1995. Relational Typology. In Syntax, 2, J. Jacobs, A. von Stechow,W. Sternefeld & T. Vennemann (eds.), 1076-1109. Berlin: Walter de Gruyter.

Primus B. 1999. Cases and Thematic Roles: Ergative, accusative and active.Tübingen: Niemeyer.

Robblee, K.E. 1993. Individuation and Russian agreement. Slavic and EastEuropean Journal 37: 423-441.

Thomason, Richmond E. (ed.). 1974. Formal Philosophy. Selected Papers ofRichard Montague. New Haven and London: Yale University Press.


Roxana-Iuliana Popescu works as an Assistant Lecturer for the Department ofForeign Languages and Literatures, at “Vasile Alecsandri” University of Bacău,Romania. She is also a Ph.D. Student in Logic at “Al. I. Cuza” University ofIasi, Romania. Her main research interests include Linguistics, Logic andEnglish for Specific Purposes. She teaches practical courses in English(translations, grammar exercises, academic writing and communicationactivities), seminars in Contemporary English language and English forSpecific Purposes (engineering, mathematics, medicine, biology, occupationaltherapy, kinesiotherapy, physical education and sport).

Page Updated: 15-Feb-2013