In this paper, I analyze two clause combining strategies in Ossetic that exhibit mixed properties between coordination and subordination. I argue that the ‘mismatch approach’ proposed by Culicover & Jackendoff (1997) and Yuasa & Sadock (2002) is best suited to account for their properties. However, in order to adequately describe the behavior of these constructions in terms of the mismatch approach, appealing to three levels of grammar is required instead of two levels (syntax and semantics) discussed in previous works. This provides a clear argument in favor of models of grammar such as Lexical Functional Grammar (LFG), where the syntactic level is split between constituent structure (c-structure) and functional structure (f-structure). The properties of semantic coordination and subordination that have been proposed in earlier work mostly belong to the level of f-structure, and not semantics proper. I argue that the only substantial semantic difference between coordination and adverbial subordination is that the former introduces discourse relations between speech acts, while the latter introduces asserted predicates that link two propositions within the same speech act. I provide definitions of coordination and subordination at all the three levels of grammar formalized in terms of the LFG framework, and discuss the tests that can be used for each of these levels.

This paper examines a non-canonical morphophonological vowel alternation in the roots of Russian verbs that is conditioned by aspectual information (derived imperfectivization). This aspectual morpheme is usually expressed as a suffix, but in the forms of interest appears as a vocalic nucleus in the root (whereas there is no vocalic nucleus in the perfective form). In a manner broadly compatible with Distributed Morphology (DM), I argue that this alternation is part of a more general phonological process – yer realization – special only in that it is triggered by morphosyntactic, rather than phonological, information. I propose an analysis of this pattern in which autosegmental representations – in this case, a mora – can be the exponents of morphosyntactic features. This approach obviates the need for DM readjustment rules, which have been criticized on empirical and theoretical grounds (Siddiqi 2006, 2009; Bye & Svenonius 2012; Haugen & Siddiqi 2013). I demonstrate that the required allomorphic interaction between the root and the derived imperfective morpheme is local, despite surface appearances: the intervening vowel is a theme vowel, inserted post-syntactically. This approach makes sense of broader patterns involving this theme vowel, and vindicates theories of allomorphic interaction that impose strict locality conditions (e.g., structural and/or linear adjacency).

Of late, a controversy has arisen over the internal structure of Korean syllables. While there is general agreement that non-phonotactic criteria argue for left-branching, Lee & Goldrick's (2008) left-branching phonotactic analysis is contradicted by Berg & Koops's (2010) claim as to a phonotactically symmetrical syllable structure. A comparison of the methodologies of the two studies, a revisit of the previous data and a new analysis cement the conclusion that there is neither a left-branching nor a right-branching phonotactic effect in Korean syllables. An investigation of the phonotactic structure of Finnish CVC syllables, which exhibit a psycholinguistic left-branching bias much like Korean, reveals that word-initial syllables possess a largely symmetrical organization whereas word-final syllables tend to show a right-branching slant. This curious set of results is consistent with the following three hypotheses: (i) The phonotactic criterion has an inherent VC bias. (ii) Symmetrical syllable structures represent a compromise between left- and right-branching effects. (iii) The strength of phonotactic constraints increases from earlier to later portions of words. The bottom line of this analysis is that, contra all previous claims, phonotactic constraints cannot be used as an argument for sub-syllabic constituency. We discuss the proposal that the basis of the left-branching bias in Korean syllables is instead to be found in the high degree of coarticulation between the onset consonant and the following vowel.

In French and Spanish, both parts of the causative–anticausative alternation can be formally encoded in two ways: Depending on the form of the verb, marked and unmarked causatives and marked and unmarked anticausatives can be distinguished. The goal of this paper is to verify whether causalness is a factor in the encoding and whether the two languages differ in this respect (verbs used more often as causatives than as anticausatives have a high degree of causalness, while verbs used more often as anticausatives than as causatives have a low degree of causalness). On the basis of a corpus study of 20 French and 20 Spanish verbs, it will be shown that in both languages a strong correlation between causalness and encoding exists. A high degree of causalness increases the likelihood that a verb’s anticausative is marked and the causative is unmarked, and a low degree of causalness increases the likelihood that a verb’s anticausative is unmarked and the causative is marked.

Modern Persian conjugation makes use of five periphrastic constructions with typologically divergent properties. This makes the Persian conjugation system an ideal testing ground for theories of inflectional periphrasis, since different types of periphrasis can be compared within the frame of a single grammatical system. We present contrasting analyses of the five constructions within the general framework of a lexicalist constraint-based grammatical architecture (Pollard & Sag 1994) embedding an inferential and realizational view of inflectional morphology (Stump 2001). We argue that the perfect periphrase can only be accounted for by assuming that the periphrase literally fills a cell in the inflectional paradigm, and provide a formal account drawing on using valence for exponence. On the other hand, other periphrastic constructions are best handled by using standard tools of either morphology or syntax. The overall conclusion is that not all constructions that qualify as periphrastic inflection from the point of view of typology should receive the same type of analysis in an explicit formal grammar.

Post-tonic synope in English (Received Pronunciation) optionally deletes a schwa between a stressed and an unstressed vowel (gén(e)ral), but it cannot apply if the vowel following the schwa is stressed (gén*(e)ràte), or if no vowel follows (hápp*(e)n). Syncope is thus triggered by a metrical lapse of unstressed vowels. In addition, short stressed vowels cannot occur in an open syllable in English (Stress-to-Weight), except when preceding a single consonant and a vowel. Hammond (1997a) analyses such seemingly open stressed syllables in words like gén(e)ral as closed by a virtual geminate. I argue that post-tonic syncope can be understood as another means of satisfying the Stress-to-Weight requirement, closing the stressed syllable in a different way, at the same time avoiding a metrical lapse. In addition, surprisingly, English post-tonic syncope is sensitive to the quality of the flanking consonants: the consonant following the alternating vowel must be a sonorant which is more sonorous than the consonant preceding it (dél*(i)cate, cól*(o)ny). These are the same conditions as those applying to syllabic consonant formation, which can be regarded as a stage preceding syncope, explaining the melodic restrictions. I analyse the interplay of different forces in Stratal Optimality Theory, employing Government Phonological representations.

This paper argues that Danish verb-second clauses have two structural instantiations and that each structure is associated with distinct information-structural properties. Information-structurally undifferentiated V2 clauses are realized as TPs, whereas information-structurally differentiated V2 clauses are CPs. The evidence for this correlation comes from the behavior of the overt VP anaphor det, which exhibits a complex, but principled, positioning pattern in V2 clauses. I develop a feature-driven analysis of V2 clauses that accounts for previously unnoticed restrictions on the initial position in declarative V2 clauses.

English binominal NPs (BNPs) (e.g., a giant of a man, a skullcracker of a headache) are of empirical and theoretical interest due to their complex syntactic and semantic properties. In this paper, we review some basic properties of the BNP construction, focusing on its headedness, semantic relations, and the role of the preposition of. We argue that these properties suggest an account in the spirit of construction grammar. In particular, we argue that the English BNP is a nominal juxtaposition construction with particular special syntactic constraints, linked to a semantic interpretation reminiscent of the subject–predicate relation.

This paper discusses how information structure can be seen as a subjective and intersubjective concept in Verhagen's (2005) and Breban's (2010) definitions, though less so in Traugott's (2010) use of the terms. More difficult is the question of whether markers of information structure can be characterised as (inter)subjective; this is more easily determined for morphological markers than for prosody or word order. For unambiguous markers of information structure, I suggest that their emergence (e.g. copula > focus marker) is typically accompanied by (inter)subjectification, whereas their further development (e.g. topic marker > subject marker) displays objectification. The paper not only shows that grammatical items can undergo an increase as well as a decrease in (inter)subjectivity – thus denying strict unidirectionality – but also confirms that these processes are independent of grammaticalisation.

The first part of this review article summarises and evaluates the contents of the book, attempting to do justice to the wealth of perspectives it offers. The book considers phonological, morphological, syntactic and semantic features from the vantage points of approaches as diverse as typology, computational linguistics and formal theories like Head-driven Phrase Structure Grammar (HPSG) and Minimalism. The second part of the article discusses several of the unifying threads that run through the volume, including the internal and cross-linguistic validity and correspondence of features, as well as the boundaries between morphology, syntax and semantics. I argue for a syntactic treatment of what has been referred to as periphrastic tense constructions in Bulgarian (including the future and the perfect), which I believe ensures greater language-internal and cross-linguistic consistency in proposing features and assigning their values. After briefly examining animacy in Bulgarian, I conclude that the operation and classification of features can make drawing boundaries between semantics and morphosyntax especially difficult. Finally, a case is made for treating tense as at least partly morphosyntactic in English, in contrast to prevalent current assumptions about the strictly morphosemantic nature of tense cross-linguistically.

Formal models of syntax typically accord the structural position external to the verb's domain a privileged status in the overall syntactic makeup of a language, either by assuming that external arguments are always S or A, or by linking external argument position to syntactic pivothood. This paper demonstrates that the Oceanic language Äiwoo has an ergative verb phrase – i.e. A as the VP-internal argument and S/O as external arguments – but no corresponding S/O pivot. That is, the ergative structure of the verb phrase in Äiwoo does not entail any syntactically privileged status of the VP-external arguments; rather, it is simply a by-product of various diachronic developments. This situation shows that what has traditionally been perceived as fundamental differences in grammatical organisation – the difference between an accusative and an ergative pattern of VP structure – need not in fact be associated with any broader differences in syntactic or pragmatic structure. More importantly, it goes against the assumption that it is possible to assign universal functions to syntactic configurations. Instead, it can be seen as providing support for the view argued for by Evans & Levinson (2009: 444) that ‘most linguistic diversity is the product of historical cultural evolution operating on relatively independent traits’.

A recent development in Dutch concerns the deictic interpretation of the second-person singular pronoun je, which may refer to the speaker only. In such examples the subject refers to the speaker – not the hearer – but at the same time, these examples come along with an implicature stating that the hearer would have done the same thing if s/he were in the speaker's situation. Why is it the case that a second-person singular pronoun may refer to the speaker only? And why is it that when speaker-referring je is used, it always comes along with an implicature of the kind described above? In this article I argue that this behavior of Dutch je is a consequence of its semantically unmarked status with respect to the first-person singular pronoun ik. Along the lines of Sauerland (2008), I propose that Dutch je only carries one feature, [PARTICIPANT], whereas ik carries two features: [SPEAKER] and [PARTICIPANT]. Consequently, je may in principle refer to all participants in the conversation, enabling je to refer to the speaker as well. The fact that je does not normally refer to the speaker but to the hearer only then follows as some kind of blocking effect resulting from application of the principle of Maximize Presupposition. The paper concludes by spelling out the predictions that this analysis makes for the cross-linguistic variation with respect to the readings that participant and other pronouns may yield.

Russian has a family of reflexive constructions that have non-canonical syntax and express a variety of meanings that range from disposition (‘I feel like working’) to ability (‘I cannot work here’) and generic assessment of quality (‘I work well here’). Previous analyses assume that these constructions are derived by a regular syntactic rule and postulate a null modal in the structure to account for their semantics (Benedicto 1995, Franks 1995, Rivero & Arregui 2012). Focusing on the ‘feel like’ construction, I show that derivational analyses have difficulty explaining its idiosyncratic properties, including non-canonical agreement (independent of the structural subject), as well as the interpretation of aspect. Moreover, derivational analyses overgeneralize, since only a subset of predicates occur in the ‘feel like’ construction in Russian, as the data from the Russian National Corpus indicate. In order to account for their idiosyncratic properties and semi-productivity, I propose that the ‘feel like’ construction and its kin are stored in the lexicon as constructions (Goldberg 1995; Jackendoff 1997, 2008). The proposed analysis clarifies the status of reflexive constructions in Russian and establishes the scope of cross-linguistic semantic variation by comparing reflexives in Russian to that in other Slavic languages.

One issue in the analysis of degree achievements is whether or not what are called degree achievements are in fact achievements (Hay, Kennedy & Levin 1999, Kearns 2007, Rothstein 2008a). In this paper, we offer evidence from Malay that they are. Our evidence involves findings about the aspectual effect of the verbal prefix meN- in degree achievement sentences, which may receive a natural account under an approach where degree achievements are lexically specified as achievements, but are difficult to explain if they are not. We propose that meN- merges with a verbal projection (VP) that describes eventualities with stages, in the sense of Landman (1992, 2008). This requirement explains meN-'s apparent effect on telicity in degree achievement sentences and the absence of such an effect in non-degree achievement sentences. It also accounts for the restricted distribution of meN- in stative sentences (Soh & Nomoto 2009) and regular achievement sentences. While certain aspectual parallels exist between the verbal prefix meN- and the English progressive, we argue that meN- is not a progressive marker, and that the parallels with the English progressive are due to the subcategorization requirement of meN-, which makes event stages more prominent in sentences with meN- compared to ones without. Our analysis supports the treatment of meN- as a light verb (v) (Aldridge 2008; Nomoto 2008, 2011; Sato 2012), rather than a marker of voice (Voice) (Cole, Hermon & Yanti 2008).

The see-in construction, exemplified in sentences like John sees/finds in Mary his best friend involves a relation of predication between the DPs Mary and his best friend. It provides evidence against the hypothesis that predication relations are established uniformly within the syntactic context of clausal structure. The empirical facts show that [in Mary] and [his best friend] in this example are both complements of see/find. It follows that Mary and his best friend could not possibly be syntactically related as the subject and predicate of a clause, despite the fact that they are engaged in a semantic relation of predication.

This paper considers data from Niuean, a Polynesian language with VSO word order and an extensive range of grammatical particles. We focus on three question particles, nakai, ka and kia, examining their possible historical origins. In related languages the preferred means of forming a yes–no question is by intonation alone, while in the Polynesian languages that have yes–no question particles, the forms are lexically distinct from those found in Niuean. We argue that the Niuean unmarked question particle nakai is derived from the negative, the pragmatically marked kia construction from the polite form of the imperative, and the ka construction from a lexical item which signals confirmation. In all three cases, the question particles do not replace their original grammatical or lexical source words but rather co-exist with them in new contexts. The three question particles have all undergone a process of semantic bleaching, increased syntactic bonding, and in some cases, phonetic reduction. While two of the processes have occurred early in Niuean, one is a very recent development.