We report on an articulatory study which uses an electromagnetic articulograph to investigate word-initial consonant clusters in Italian. In particular, we investigate clusters involving a sibilant, such as in spina ‘thorn’. The status of the sibilant in such clusters, referred to as ‘impure s’, is an unresolved problem for the syllable phonology of Italian. Coordination patterns of the gestural targets of consonantal and vocalic gestures reveal a structural difference between obstruent–liquid clusters, e.g. /pr/, and sibilant–obstruent clusters, e.g. /sp/. Whereas in /pr/, both /p/ and /r/ have distinct coordination patterns as compared to either /p/ or /r/ as a single consonant in the same (word-initial) position, this is not the case for /sp/. Here the /p/ patterns like a single consonant: /p/ in spina patterns with /p/ in Pina (proper name). Thus, although /s/ in spina constitutes a word onset, there is evidence against it being part of a syllable onset.

This paper examines the articulation of harmonically distinct classes of vowels in Standard Yoruba. There has been considerable disagreement as to whether the distinction between [e o] and [∊ ɔ] is one of vowel height or tongue-root advancement/retraction. This paper reports on an ultrasound investigation of Yoruba vowels. Results are consistent with harmonic classes distinguished by a tongue-root advancement/retraction feature, not by vowel height. We also investigate the relation between articulations of the tongue root and its neutral position between utterances, the inter-speech posture (ISP). We find more variability in ISP-to-articulation mapping than previous studies, but our results are still partially compatible with a postulated correlation between phonologically ‘active’ feature values and articulatory displacement from ISP. Overall, our results support an analysis of Yoruba vowels in terms of a tongue-root feature, and provide insight into the mapping between phonetics and phonology.

This article argues that, contrary to some recent proposals, a given phonological form may be organised into at most one array of metrical structure at a time. The bulk of the paper is dedicated to a case study of Huariapano, a language that has been claimed to motivate multiple, coexisting but autonomous, layers of metrical parsing. I show that this conclusion is premature: both stress and segmental patterning in Huariapano can be modelled within a single system of constituency, once context-dependent variation in foot form is taken into account. The reanalysis developed here also draws on the idea that foot-initial syllables may be targeted by augmentation or fortition processes even when unstressed. Independent evidence for foot-initial strengthening is furnished by segmental phonotactics in a range of other languages.

This paper explores the relative merits of constraint ranking vs. weighting in the context of a major outstanding learnability problem in phonology: learning in the face of hidden structure. Specifically, the paper examines a well-known approach to the structural ambiguity problem, Robust Interpretive Parsing (RIP; Tesar & Smolensky 1998), focusing on its stochastic extension first described by Boersma (2003). Two related problems with the stochastic formulation of RIP are revealed, rooted in a failure to take full advantage of probabilistic information available in the learner's grammar. To address these problems, two novel parsing strategies are introduced and applied to learning algorithms for both probabilistic ranking and weighting. The novel parsing strategies yield significant improvements in performance, asymmetrically improving performance of OT learners. Once RIP is replaced with the proposed modifications, the apparent advantage of HG over OT learners reported in previous work disappears (Boersma & Pater 2008).

In a detailed comparison of the intonational systems of two unrelated languages, Israeli Sign Language and American Sign Language, we show certain similarities as well as differences in the distribution of several articulations of different parts of the face and motions of the head. Differences between the two languages are explained on the basis of pragmatic notions related to information structure, such as accessibility and contingency, providing novel evidence that the system is inherently intonational, and only indirectly related to syntax. The study also identifies specific ways in which the physical modality in which language is expressed influences intonational structure.

This paper examines two related phenomena from Catalan that pose a challenge to a proposal in McCarthy (2012) that combines Harmonic Serialism (HS) with the theory of Optimal Interleaving (OI) of Wolf (2009). The first phenomenon, found with stems ending in a sibilant, is the selection of a marked masculine allomorph only before the plural morph, also a sibilant, which prevents an OCP violation. The second, phrasal, phenomenon is restricted to one dialect of Catalan: the plural morpheme is not realised in prenominal position only when it would surface between two consonants. It is shown that, with some crucial modifications on the way spell-out proceeds, HS/OI can account for the two phenomena when they occur separately, but when they co-occur within the same phrase, wrong predictions are made, due to the one-change-at-a-time basic property of HS. These problems do not arise in a parallel OT approach.

The results of two artificial grammar experiments show that individuals learn a distinction between identical and non-identical consonant pairs better than an arbitrary distinction, and that they generalise the distinction to novel segmental pairs. These results have implications for inductive models of learning, because they necessitate an explicit representation of identity. While identity has previously been represented as root-node sharing in autosegmental representations (Goldsmith 1976, McCarthy 1986), or implicitly assumed to be a property that constraints can reference (MacEachern 1999, Coetzee & Pater 2008), the model of inductive learning proposed by Hayes & Wilson (2008) assumes strictly feature-based representations, and is unable to reference identity directly. This paper explores the predictions of the Hayes & Wilson model and compares it to a modification of the model where identity is represented (Colavin et al.2010). The results of both experiments support a model incorporating direct reference to identity.

Obstruents in Russian have been claimed to assimilate in voicing in clusters when a sonorant consonant intervenes, e.g. ot mgly [dmg] ‘from the haze’. This phenomenon (‘sonorant transparency to voice assimilation’) is controversial: it is claimed to be a phonological rule of fast speech by some linguists, while its existence is denied by others. Previous studies have shown that voicing in presonorant obstruents (C1) in Russian is consistent with that of prevocalic obstruents in slow speech; however, no research has examined whether voicing in presonorant obstruents changes either as a function of the voicing of the rightmost (C2) obstruent in a cluster or in faster speech. This paper presents experimental results supporting the claim that the voicing of C2 obstruents does not affect voicing in presonorant C1 obstruents in slow or fast speech. The results suggest that obstruents do not assimilate through a sonorant in obstruent–sonorant–obstruent clusters in Russian.

This article discusses the syntax–prosody interface, drawing on evidence from Stockholm Swedish. It is shown that a Swedish main clause containing an embedded clause has three prosodic correlates, two of which are non-isomorphic to the syntactic bracketing. However, two coordinated clauses have only one – isomorphic – prosodic correlate. Optimality-theoretic constraints (Prince & Smolensky 1993) are used to derive this variation. A new markedness constraint, EqualSisters, is argued to be responsible for a preference for flat prosodic structures. This constraint requires that sister nodes in prosodic structure belong to the same prosodic category, and therefore sometimes conflicts with match constraints, which call for syntax–prosody correspondence (Selkirk 2009, 2011). When high-ranked, EqualSisters forces syntax–prosody non-isomorphism if the input syntactic structure contains embedding, whereas full isomorphism is predicted in coordinated structures. The previously suggested markedness constraints Non-recursivity and Exhaustivity (Selkirk 1996) cannot replace EqualSisters, and in the present account are rendered redundant.

Through analyses of laryngeal co-occurrence restrictions in two varieties of Aymara, this article shows that contrastively specified representations are crucial in shaping phonological patterning. The article argues for a model of contrastive specifications in which features are hierarchically ordered (Dresher 2009). This results in asymmetries between features such that, for a given inventory, some features are contrastively specified in a greater number of segments than others. This asymmetry between features plays a central role in accounting for the interaction of place of articulation features and laryngeal features in Bolivian Aymara. The article also demonstrates that contrastive representations can be achieved as output forms in Optimality Theory and that the constraints which determine contrastive representations can be integrated with constraints which motivate restrictions on the co-occurrence, ordering and location of laryngeal features in Peruvian and Bolivian Aymara.

Vowel deletion in Yine crucially refers to both morphological and phonological information. It has been argued that the process is only analysable in a theory where the phonology has access to morphology, either on the assumption of different morphological domains of constraint evaluation (Lin 1987, 1997a, b) or on the assumption of morphologically indexed optimality-theoretic constraints (Pater 2009). In contrast, I propose a phonological analysis of vowel deletion in Yine in a parallel Optimality Theory model. The phonology, I assume, has only limited access to morphological information, and can only distinguish between affix and stem material. I argue that the morphemes that trigger deletion of a preceding vowel have a defective underlying representation: they lack a mora, and ‘usurp’ the mora of a preceding vowel.

Phonological rules can be variable in two ways: they can apply to a subset of the lexicon (lexical variation), or apply optionally, with a probability that depends on the phonological environment (stochastic variation). These two types of variation are occasionally seen as mutually exclusive. We show that the vowel–zero alternation in Russian prepositions ([s trudom] ‘with difficulty’ vs. [sə stinoj] ‘with the wall’) exhibits both types of variation. In two corpus studies and a nonce-word experiment, we document novel stochastic factors that influence the alternation: similarity avoidance, stress position and sonority profile. These constraints interact additively, lending support to a weighted-constraints analysis. In addition to phonologically determined stochastic variation, we find significant lexical variation: phonologically similar nouns differ in the rate at which they condition the alternation in the prepositions. We analyse this pattern by augmenting the weighted-constraints approach with lexical scaling factors.