Rolandas Mikulskas. “Copular constructions in Lithuanian. Valency, argument realization and grammatical relations in Baltic”

Straipsniai / Articles 257 GÖTZ KEYDANA Georg-August-Universität Göttingen Fields of research: phonological reconstruction: reconstruction of phonological grammars (accentology, syllabification), phonetic grounding of diachronic phonological processes; historical syntax, argument structure and case, subordination (mainly Vedic Sanskrit). ROLANDAS MIKULSKAS COPULAR CONSTRUCTIONS IN LITHUANIAN. VALENCY, ARGUMENT REALIZATION AND GRAMMATICAL RELATIONS IN BALTIC Amsterdam: Benjamins, 2017, xv, 280 pp. ISBN 9789027259127 The book under review is the fourth publication in a series on argument structure in Baltic. The author presents a thorough investigation of copular constructions in Lithuanian, i.e. constructions with būti and, to a lesser degree, other copular-like verbs like virti and tapti. The analyses given are couched in the framework of Cognitive Linguistics. Copular constructions have been studied extensively in the last decades, both from a typological and from a theoretical point of view. This interest is partly due to the vastly different semantics covered by CCs, ranging from predicative to equational, and from existential to locative and possessive. CCs are also challenging from a syntactic point of view. How can we account for their structure? Are small clauses involved? If so, what is their exact syntactic status? How are cases like the dative in the mihi est-construction or the genitive in constructions like Onutės žalios akys (p.24) licensed? The author gives essentially synchronic answers to at least some of the questions mentioned (the caveat “essentially” being due to the fact that it sometimes GÖTZ KEYDANA 258 Acta Linguistica Lithuanica LXXVII remained unclear to me if the claimed dependencies between different constructions are to be understood as diachronic or synchronic. Genuine attempts at diachronic explanations remain rather meek. E.g., the scenario for the development of pseudoclefts on p.129 is largely irrelevant since the author does not search for any evidence for the alleged precursors of pseudoclefts in Old Lithuanian). Contrary to a trend in functional linguistics in recent years, he draws his data almost exclusively from intuition. Corpora, which might have shed light on the distribution of the various constructions discussed, were not consulted. This may be considered a drawback, especially when pseudoclefts are concerned. As a non-native of Lithuanian, I may well be misled, but I am under the impression that this type is marginal at best. In the first part of his book, Mikulskas develops a detailed account of the syntax and the semantics of copular constructions of Lithuanian. The first chapter discusses problems of role assignment and linking. The gist of this chapter is convincing: Attempts to derive the attested constructions from verbal subcategorization do face serious problems in the case of verbs like būti, whose “semantic resources […] are not sufficient” (p.26). In the second chapter, the author looks into the semantics of būti in greater detail. He illustrates the inadequateness of what he calls the “verbocentric model” with the spray/load-alternation. Mikulskas follows Goldberg (1995) in tying this alternation to different constructions and thus ultimately to different semantic frames. Turning to būti, he discusses the semantic contribution of the verb in various copular constructions. He concludes that the copula essentially predicates identity; the semantics of the various constructions, then, is due to its predicative complements. Mikulskas’ arguments for the identity-semantics of the verb, however, are rather weak. They seem to be based essentially on his observation that “we use this traditional term [i.e. copula, G.K.] having borrowed it from the analysis of the proposition of identity as a logic structure” (p.42). Apart from this petitio principii we find claims that “its [i.e. the copula’s, G.K.] structural function in constructions of identity is also obvious, and it allows us to speak of a relative independence of such verbs with respect to the second component of the complex predicate and assign the meaning of ‘identity’ to them” (p.43). To me, this is far from obvious, and I would have liked to see some evidence supporting this assumption. What then, about well-formed sentences like ponas ne brolis, his ex. 2, p.2, or the epitome of identity statements, Maironis − tai Jonas Mačiulis, his ex.49, p.19? Chapter 3 is an extensive study of the various attested copula constructions of Lithuanian. The chapter culminates in a new classification distinguishing ascriptive constructions (with adjective complements) from equative constructions (with nouns) which “may be further classified into descriptional-identifying Rolandas Mikulskas, Copular constructions in Lithuanian. Valency, argument realization and grammatical relations in Baltic Recenzijos / Reviews 259 and properly identificational ones” (81). This classification is based on insightful interpretations, not however on distribution or grammaticality judgements. See below on this point. Chapter 4 serves to complete the picture. Mikulskas discusses more marginal types of copula constructions before adding “Final remarks” to the first part of his book in chapter 5. Here, the author again turns to the existential construction which, in his convincing analysis, is the basis for locative and possessive constructions. Thus, at the core, būti has two functions. On the one hand, it denotes existence (allowing for locative and possessive constructions); on the other, it serves as a copula denoting identity (allowing for ascriptive constructions). The second part of the book is devoted to specificational constructions. An example of such a construction is Mikulskas’ ex. 211, p.98: Mūsų anglų kalbos mokytojas yra Jonas Petraitis. Constructions like these stand out because of their peculiar information structure: “[T]he nominal with role definiteness [i.e. a stage-level predicate, G.K.] is always the topic whereas that with individual definiteness [i.e. an individual-level predicate, G.K.] is its comment” (p.97). The order in specificational constructions is fixed. Rather unsurprisingly, the author claims that these constructions are based on “the meaning of identity” (p.110) and that they differ from identity or equation due to “the semantic and referential properties of their nominals” (ibid.). These properties, however, are not spelled out explicitly. Chapter 7 is devoted to pseudoclefts like Jeigu ko nemėgstu, tai cepelinų, Mikulskas’ ex.255, p.120, or Ką Marija padarė, tai bėgo maratoną. The latter type is remarkable when compared to Engl. What Maria did was run a marathon (p.121) because of the finite verb in the second part. The author does not account for this difference. Pace Mikulskas it is hardly surprising that “the postverbal (subject) nominal is always the nominative” (p.121). This is simply due to the fact that the instrumental in predicative constructions is tied to a stage-level reading – certainly not to the position relative to the copula. Also, “case attraction” in reduced pseudoclefts, which the author discusses on p.123ff., seems to be default in languages with rich case systems. His analysis of the allegedly attracted nominals as attributive is convincing. In chapter 8 Mikulskas raises the question of subjecthood in specificational constructions. Despite clear agreement patterns he concludes that the subject is the preverbal nominal. The point will be taken up below. The third and final part of the book is devoted to aspect in copular constructions. Apart from lexicalized forms like būna or inflectional forms like būdavo, aspect is generally encoded by preverbs or adpositions. GÖTZ KEYDANA 260 Acta Linguistica Lithuanica LXXVII In chapter 11 he applies Stassen’s stratificational model to Lithuanian copular constructions. The results are hardly surprising. Cf. the following statement: “[T]he nominal encoding strategy, particularly its variety with a non-verbal copula in identificational sentences, offers the least possibilities for the expression of aspect” (p.212). That copular-like verbs like e.g. tapti offer greater aspectual potential than būti almost amounts to a truism. In chapter 12, various aspectual and temporal possibilities with different types of copular constructions are illustrated. The survey is impressive. However, it hardly sheds new light on the TA-system of Lithuanian: The observed possibilities are, as one would have suspected, essentially tied to the semantics of the complements. The book ends with “general conclusions” and indices. Mikulskas’ book is a major contribution to the study of Lithuanian syntax. It has, however, severe shortcomings which lessen its potential impact. A minor one concerns Mikulskas’ attitude towards work from authors of what he calls “the formal persuasion” (58,262). The author discusses formal accounts as a background to his own approach. However, given the plethora of papers he could have chosen from, his selection seems to have been guided by chance. Worse still, Mikulskas does not do the work he reviews justice. Both his account of formal semantic and minimalist approaches (others are ignored) are grossly inadequate. He obviously has no grasp of type theory, nor does he understand the intricacies of minimalist analyses. But his treatment of formal analyses also reveals a more fundamental problem: One of the hallmarks of work in the generative tradition is the way data are used in a systematic fashion to elucidate patterns. Take, for instance, the following examples adapted from data Mikulskas quotes in his review of analyses by “representatives of the formal trend” (p.62): (1) They considered alcohol (to be) excitatory. (2) They considered alcohol (to be) water. (3) They considered Cicero *(to be) Tully. Data like these show unambiguously and independently of any interpretation or framework that at least in English there is a syntactic difference between equatives (ex.3) and predicatives (exs.1,2). Under embedding, the copula is mandatory in the latter, while it may well be omitted in the former. Observations like these are invaluable for two reasons: First, they are independent of premises. Second, to be descriptively adequate, any model of English copula constructions should be able to account for them. Mikulskas, however, hardly ever contributes observable and testable data like these. Concerning the example quoted, he even misses the point, since his own analysis predicts Rolandas Mikulskas, Copular constructions in Lithuanian. Valency, argument realization and grammatical relations in Baltic Recenzijos / Reviews 261 that copular constructions with postverbal noun behave exactly alike – and he seems to suggest that this is not supposed to be understood as a peculiarity of Lithuanian. Instead of sifting data, Mikulskas tends to resort to interpretations guided by his cognitive framework or even to argumenta ab auctoritate (e.g. p.59). In the few instances where he adapts a strictly data-driven methodology, he misses the impact of his observations: Inspired by English data he quoted in a review of formal analyses, he introduces Lithuanian binding data on p.130. Unnoticed by him, they challenge his analysis as they show that Lithuanian clefts form a binding domain. Thus, treating their parts as syntactically independent (p.121) raises serious questions. In one of the rare instances when the author himself actually argues with distributional patterns, he fails in a rather spectacular fashion: Commenting on ex.149 on p.67, he claims the phrase yra dėžėje to be “discontinuous” because “the morpheme of the locative case representing the LOC” is not directly adjacent to the verb. Since there is no way case marking could possibly show up on the copula, this observation is astounding. Even more surprising is the fact that Mikulskas continues this line of thought by remarking on it being “noteworthy” that “the locative prepositional phrase [prie ežero in ex.153] remains indivisible” (p.67). Preposition stranding is unknown in Lithuanian. How, then, could this pattern be an issue? An example of the impressionistic methodology Mikulskas applies elsewhere is his discussion of subjects in specificational constructions. Agreement in data like his ex.289, p.136, Šios mokyklos direktorius esanti tokia Janina Petraitienė, points to subjecthood of the postverbal constituent. Now, the author is certainly right in claiming that agreement is no sufficient criterion (though it is certainly a strong one). Since the seminal work of Andrews (1976) and Zaenen, Maling & Thráinsson (1985) various subject tests have been developed which go far beyond simple agreement. Mikulskas, however, uses none of these. Rather, he resorts to the simple intuition that an agreement-based analysis clashes with information structure. Dwelling on this example, it struck me as noteworthy that, whatever the cognitive humdrum, the author’s analysis tends to regress to extremely conservative premises based essentially on traditional grammar: The subject in specificational constructions is the preverbal constituent mainly because of the author’s general uneasiness with the notion of postverbal subjects in Lithuanian. Equative and predicative constructions with nouns are opposed to predicative constructions with adjectives (ascriptive constructions according to Mikulskas) simply because of the parts of speech involved. A final point I would like to mention is the lack of formal explicitness in Mikulskas’ work. Though the author claims to give a construction grammar GÖTZ KEYDANA 262 Acta Linguistica Lithuanica LXXVII account of the data, we neither find an explicit lexical entry for būti, nor any account of the interaction of this entry with constructions. While the level of grammatical analysis may be a bit disappointing to some (including myself), the book nevertheless offers the most detailed account of copular constructions in Lithuanian today. This makes it an indispensable tool for anyone interested in the syntax of Lithuanian. REFERENCES Andrews Avery 1976: The VP complement analysis in Modern Icelandic. – North Eastern Linguistic Society 6, 1–21. Goldberg Adele E. 1995: Constructions. A Construction Grammar Approach to Argument Structure. Chicago: University of Chicago Press. Zaenen Annie, Joan Maling, Höskuldur Thráinsson 1985: Case and Grammatical Functions: The Icelandic Passive. – Natural Language and Linguistic Theory 3: 441–483. Įteikta 2017 m. lapkričio 10 d. GÖTZ KEYDANA Georg-August-Universität Göttingen Sprachwissenschaftliches Seminar Käte-Hamburger-Weg 3, 37073 Göttingen, Germany [email protected]