The articles in this volume stem from papers presented at the conference Variation and Change in Argument Realisation (Naples–Capri, May–June 2010). In the introductory chapter, ‘Argument structure in flux’, Jóhanna Barðdal & Michela Cennamo provide a useful if brief overview of the principal desiderata in the study of argument structure and a chapter-by-chapter synopsis of the volume. The twenty chapters that follow test a number of established and novel theories against a wealth of data from a wide variety of languages. They include original analyses of much discussed questions, such as phonologically null arguments, as well as treatments of less debated issues, for example differential argument marking according to semantic role. The volume thus offers a very valuable contribution to the current understanding of argument structure and argument realisation in synchronic and diachronic perspective.

The chapters are organised in five thematic parts. Part I, ‘Argument structure and encoding strategies’, comprises four articles. In the first article, ‘Non-canonical subjects in clauses with noun predicates’, Marina Benedetti discusses a construction of Modern Standard Italian (gli (dative) prese paura ‘him (over)took fear’) where the dative NP lacks some behavioural properties of the Italian subject. Building on work in Relational Grammar, Benedetti proposes that this is an impersonal construction, where the dative NP is solely a subject in underlying syntax. The NP of psychological state, on the other hand, is claimed to be both a predicate and an object in underlying syntax. Although this proposal begs for further investigation of the very notions of argumenthood and predication, it does capture the data, paying some attention to diachronic and crosslinguistic evidence.

Diana Forker's contribution, ‘Differential agent marking in Hinuq’, provides a lucid and detailed analysis of argument realisation in Hinuq, Nakh-Daghestanian. Forker identifies four types of agents that lack agent entailments to different degrees, or have patient properties. These agents differ from canonical ones in case marking (they take AT-essive spatial case), reflexivization and cross-clausal control. The picture arising from this study is that of a role-dominated language (Van Valin Reference Van Valin1980), where argument realisation is predominantly motivated in semantic terms.

Leonid Kulikov's article, ‘Case variation and case alternation in Indo-European and beyond: A diachronic typological perspective’, explores the sources and outcomes of case competition, attempting to outline a hierarchy of parameters that can be held responsible for the resolution of this competition. While one could not dispute the author's view that it is impossible to settle for a single universal hierarchy of parameters, the ultimate concern of this kind of research ought to be the study of the typological variation in the relevant parameter hierarchies and the constraints on such variation. This chapter provides an important contribution to this pursuit.

The next chapter, ‘Constructional polysemy and argument realisation with the Irish GET verb’, by Brian Nolan, deals with two Irish constructions with the verb faigh ‘get’: NP₁+ recipient GET + NP₂ and NP₁ + passive GET + NP₂. The article has the merit of drawing attention to the role of argument semantics in event structure. Indeed, the realisation of the construction as one or the other type mentioned above is sensitive to the semantics of the postverbal NP. The analysis departs from other accounts of the passive, including that of the framework adopted by Nolan, Role and Reference Grammar (Van Valin & LaPolla Reference Van Valin and LaPolla1997: 294–302), in assuming that passivization can affect the semantics of the predicate. This departure could be avoided by analysing the second construction as a monoargumental intransitive structure, where faigh + NP₂ constitute a predicative nuclear juncture, in Role and Reference Grammar terms.

Part II, ‘Argument structure and verb classes’, includes five chapters. Barbara McGillivray, in ‘Latin preverbs and verb argument structure: New insight from new methods’, studies the realisation of the spatial argument of Latin preverbed verbs (e.g. abeo ‘go from, depart’, formed by the preverb ab ‘from’ and the verb eo ‘go’), convincingly arguing for the advantages of a corpus-based multivariate statistical analysis. The findings reveal a not entirely unidirectional development from prepositional marking to case marking and then again to prepositional marking. This development imperfectly matches the much debated evolution from Latin synthesis to Romance analysis, while also uncovering the orthogonal role of animacy in the realisation of the spatial argument.

The following chapter, ‘Experiencer linking: Psych verbs at the interface’, is by Marijana Marelj. Adopting the Theta System, Marelj proposes that case assignment and the merging of arguments as internal or external are captured in terms of merging instructions that relate to clusters of two binary features, ±c(ause change) and ±m(ental state), which replace traditional theta roles. The strength of this approach is that feature assignment is properly tested, though not in the case of the dative argument of gefallen-type verbs in German, which is claimed to be underspecified for the feature ±m. I see no independent evidence in support of this claim. The analysis nonetheless resolves some case–theta linking problems arising from experiencer predicates, and provides support for the view that case is derivable from the lexical semantics of predicates.

The chapter authored by Anna Siewierska & Eva van Lier, ‘Introduce: Encoding a non-prototypical three-participant event across Europe’, deals with the realisation of non-agent human participants in constructions with verbs meaning ‘introduce’ in Slavic, Romance and Germanic. The treatment of human non-agents turns out to be sensitive to the relative referentiality and anaphoricity of these arguments. The findings also support the hypothesis that object-marking systems differentiating between thematic roles cope better than other systems with the constraints against sequences of clitic non-agents, since in such systems ambiguity does not arise between theme and recipient. Significantly, the analysis sheds light on the role of specific lexical choices, and associated event structures, in the realisation of theme and recipient.

The principal aim of the chapter by Gretel De Cuyper, ‘On the relationship between lexical aspect, verbal meaning, and (lexical) argument structure’, is to identify a genuinely lexical aspectual level that is not affected by the semantics of VP objects. A contribution of this article is the distinction of three types of lexical aspect: l-atelic, l-resultative, and l-telic. Unlike l-resultative lexical aspect, l-telic aspect does not entail the reaching of the final goal to be achieved. Lexical heads with l-telic aspect are thus found in combination with both in and for adverbials. It is unfortunate that De Cuyper does not factor Bertinetto & Squartini's (Reference Bertinetto, Squartini, Bertinetto, Bianchi, Higginbotham and Squartini1995) gradual completion verbs in her treatment of l-telic aspect.

The final chapter in Part II, ‘Four Romanian verbs of occurring: The effect of argument structure on verb meaning and use’, authored by Daniela Caluianu, is a helpful reminder that seemingly synonymous verbs may involve different argument structure and semantics. Thus, a avea loc ‘take place’ and a se petrece ‘occur’ are intransitive verbs of happening, which take event arguments contributing to sentence meaning, whereas a se întâmpla ‘happen’ and a (o) păţi ‘happen’ encode two-argument predicates, which lend themselves to an existential analysis whereby the main contribution to sentence meaning comes from the verb.

Part III, entitled ‘Unexpressed arguments’, consists of three chapters. In the first one, ‘The pro cycle’, Jan Terje Faarlund analyses pro as a lexical pronoun with no phonological substance. The article offers an original account of the loss and rise of pro, in Norwegian and français avancé, respectively, in the context of changes in nominal and verbal morphology. In assessing the correlation between null subjects and verbal inflection, the author claims that this correlation is pragmatic. This insight should be pursued with reference to zero anaphora in languages without verb agreement (e.g. Chinese). This type of unexpressed argument is instead simply said to be ‘of a different character’ (263), following a well-established view.

Sonia Cyrino's contribution, ‘Argument promotion and SE-constructions in Brazilian Portuguese’, deals with the rise of a Brazilian Portuguese structure with an internal argument in subject function (As revistas estão xerocando ‘the magazines are xeroxing’). Cyrino rejects a passive or middle analysis of this structure, relating its development to the loss of non-active morphology. Her analysis captures the animacy constraint on the argument of the construction under scrutiny, though not, as far as I can see, its association with quasi-instrumental verbs.

In the final chapter, ‘Unaccusativity and the diachrony of null and cognate objects in Greek’, Nikolaos Lavidas ascribes the lack of referential null objects with alternating unaccusatives (e.g. ‘dry’) in Greek to their causative semantics, claiming that a causative structure with a null object would lend itself to ambiguity between a causative and an anticausative interpretation.

Part IV, ‘Split intransitivity’, comprises a cohesive set of four chapters, which I will comment on after brief synopses of their contents. In the first article, ‘Split intransitivity in Irish and the syntax–semantics interface’, Aidan Doyle tests Sorace's (Reference Sorace2000) Auxiliary Selection Hierarchy against Irish evidence. He argues for a minor revision of this hierarchy, in agreement with a previous proposal advanced by Cennamo (Reference Cennamo1999).

The second article, ‘Semantic constraints on the Latin impersonal passive: On telicity and agentivity’, is authored by Maria Napoli. Working with the Auxiliary Selection Hierarchy, Napoli claims that agentivity is the principal constraint on the licensing of a prepositional agentive phrase in the Latin impersonal passive.

In the following article, ‘Auxiliary selection in German: Constructional gradience with perfect formation’, Elke Diedrichsen takes issue with the universality claims of previous accounts of perfect auxiliary selection based on the Auxiliary Selection Hierarchy, proposing a refinement of this hierarchy, which captures synchronic and diachronic German evidence.

In the final chapter, ‘Tornar and volver: The interplay of frequency and semantics in compound tense auxiliary selection in Medieval and Classical Spanish’, Malte Rosemeyer successfully pursues the hypothesis that the directionality of change in perfect auxiliary selection is influenced by the combined effect of verb semantics and frequency of usage, reporting findings from a quantitative corpus-based diachronic study of BE/HAVE selection with two Spanish verbs meaning ‘return’.

For brevity, I will only mention two issues arising from Part IV. First, the crosslinguistic variation in the rankings on the Auxiliary Selection Hierarchy strongly suggests that the rationale of this variation should be sought in the conflict between the semantic parameters (telicity, agentivity, etc.) that characterise the different verb classes on the hierarchy. This conflict is resolved differently in different languages. While one cannot aim to determine a single, universally valid, hierarchy of verb classes, one can seek universality in the constraints on the variation in the relevant parameter rankings.

The issue of the crosslinguistic comparability of verb meanings also arises, as pointed out by Doyle in his chapter. Some of the findings presented in this very volume strongly suggest that seemingly synonymous verbs may have different event and argument structure. The lesson to be learnt from this non-trivial result is that crosslinguistic comparison should rely on prior testing of the semantics of the relevant verbs in each of the languages under investigation.

Part V, ‘Existential and presentational constructions’, includes four chapters. In the first one, ‘Control and the evolution of possessive and existential constructions’, Denis Creissels explores argument realisation in possessive constructions in a typological perspective, highlighting the role of the semantic parameter of the control exerted by an individual in the development of possessive constructions from existentials, comitatives, or transitives.

In ‘Existential constructions: A semasiological perspective’, Livio Gaeta argues for a semasiological analysis of existential constructions, which makes explicit connections between the wealth of synchronic variation in existential constructions and the different sources and paths of development underlying this variation.

The findings reported by Creissels and Gaeta suggest that change in the domains of possession and existentiality does not lead unidirectionally from one of these domains to the other: while the implicit argument (Francez Reference Francez2010) of an existential structure may derive from a possessor, the reverse process also obtains. This result further highlights an issue mentioned above, namely the role of argument semantics in event structure, in diachronic terms.

The third chapter, Mair Parry's ‘Variation and change in the presentational constructions of north-western Italo-Romance varieties’, offers a sophisticated analysis of the development of presentational constructions in dialects of the North-West of Italy, uncovering important albeit subtle differences in the morphosyntax of existential, unaccusative and unergative structures.

In the final chapter, ‘Argument realisation and existential pro-forms in early Italo-Romance’, Francesco Maria Ciconte convincingly argues that the pro-form of Italo-Romance existentials emerges as the spell-out of a location, in non-existential constructions with a definite postcopular NP, but loses its locative function as it spreads to existential structures. Like Gaeta's article, Ciconte's contribution seriously challenges a synchronic analysis of existentials as locatives.

The Naples–Capri papers are dedicated to the memory of one of the conference participants and chapter authors, Anna Siewierska. The attention to empirical detail and the lack of theoretical bias in the selection of the papers honour Siewierska's intellectual legacy, while the impressive range of languages and grammatical domains explored in the volume makes it an important reference for future research on argument structure.