The purpose of our study is to show that phonological knowledge is an important basis for making predictions during speech perception. Taking the phonological constraint in English that coda obstruent clusters agree in their value for voicing, we conducted two experiments using vowel–stop–fricative sequences, where the task was to identify the fricative. Stimuli included sequences that were either congruent or incongruent. Consistent with models of featural underspecification for voiceless obstruents, our results indicate that only voiced stops induced predictions for an upcoming voiced fricative, eliciting processing difficulty when such predictions were not met. In contrast, voiceless stops appear to induce no equivalent predictions. These results demonstrate the important role of abstract phonological knowledge in online processing, and the asymmetries in our findings also suggest that only specified features are the basis for generating perceptual predictions about the upcoming speech signal.

Recent research on the acquisition of natural vs. unnatural phonological processes provides some support for the idea that learning a natural process is easier than learning an unnatural one (Wilson 2003, 2006, Pycha et al.2003, Pater & Tessier 2005). This study extends those findings by comparing the acquisition of two stress patterns that are identical except in naturalness. Learners were native speakers of English, a language with variable stress, and French, a fixed stress language. Both English and French speakers learned the natural pattern significantly better than the unnatural. The artificial languages specifically neutralised the phonetic cues that might have given a perceptual advantage to the natural language. The findings suggest that a naturalness bias aids in the distinguishing and learning of a phonological pattern. To explain the results, I argue for an interaction between a general and a language-specific cognitive mechanism.

Kalam is a Trans New Guinea language of Papua New Guinea. Kalam has two distinct vowel types: full vowels /a e o/, which are of relatively long duration and stressed, and reduced central vowels, which are shorter and often unstressed, and occur predictably within word-internal consonant clusters and in monoconsonantal utterances. The predictable nature of the reduced vowels has led earlier researchers, e.g. Biggs (1963) and Pawley (1966), to suggest that they are a non-phonemic ‘consonant release’ feature, leading to lexical representations with long consonant strings and vowelless words. Here we compare Kalam to other languages with similar sound patterns and assess the implications for phonological theory in the context of Hall's (2006) typology of inserted vowels. We suggest that future work on predictable vowels should explore the extent to which clusters of properties are explained by evolutionary pathways.

This paper provides a detailed diachronic account of the adaptation of the English posterior coronal obstruents /ʃ ʧ ʤ/ in Contemporary Korean. These consonants are variably adapted with a glide (/j/ or /w/), and the distribution of the glides is conditioned by phonetic and phonological characteristics of the English input, as well as native phonotactic restrictions. The diachronic change in the occurrence of /w/ serves as an example of a variable phonetic detail in the input that is faithfully represented in loans in earlier stages, but which is subsequently eliminated in the emerging norm. Given this data, I propose how what, on the surface, may appear to be a ‘phonological’ adaptation can arise through regularisation of what is essentially a ‘phonetic’ adaptation. This study highlights the complexity of loanword adaptation and the importance of examining all of the different factors shaping this process.

Complex segments consisting of two phases are potentially ambivalent as to which phase determines their phonemic status – e.g. whether // is a stop or a nasal. This theoretical problem is addressed here with respect to a typologically unusual phoneme in Hiw, an endangered Oceanic language of Vanuatu. This complex segment, //, combines a velar voiced stop and a velar lateral approximant. Similar phonemes, in the few languages which have them, have been variously described as (laterally released) stops, affricates or (prestopped) laterals. The nature of Hiw // can be established from its patterning in tautosyllabic consonant clusters. The licensing of word-initial CC clusters in Hiw complies with the Sonority Sequencing Principle, albeit with some adjustments. Consequently, the well-formedness of words like /mejiŋə/ ‘berserk’ relies on // being analysed as a prestopped velar lateral approximant – the only liquid in the system.

The Northwest Caucasian language Kabardian displays a typologically unusual process of word formation, whereby two lexical roots fuse to form a single prosodic word whose phonological behaviour is parallel to prosodic words containing a single root. It is shown that this process of fusion, which is subject to a number of phonological and morphosyntactic restrictions, reflects a typologically unusual response to a cross-linguistically common minimal word requirement banning monomoraic prosodic words. Rather than employing segmental lengthening or insertion to ensure that minimality is satisfied, Kabardian chooses to violate the one-to-one mapping between grammatical and prosodic words. A further complication is the scalar nature of minimality in Kabardian: while the impossibility of fusion in certain prosodic and morphosyntactic contexts allows monomoraic prosodic words to surface, a more stringent minimality restriction ensures that all prosodic words have at least one mora.

In this paper, I present an analysis of the typology of laryngeal co-occurrence restrictions based on contrast markedness. The key ingredient of the analysis, for which I provide experimental support, is that laryngeal co-occurrence phenomena reflect a preference for maximising the perceptual distinctness of contrasts between words (Flemming 1995, 2004). An AX discrimination task finds that the contrast between an ejective and a plain stop is less accurately perceived in the context of another ejective in the word than in the context of another plain stop in the word. Pairs of words like [k'ap'i] and [k'api], which contrast 2 vs. 1 ejectives, are less reliably distinguished than pairs of words like [kap'i] and [kapi], which contrast 1 vs. 0 ejectives. The unifying factor of all laryngeal co-occurrence patterns is the neutralisation of the contrast between words with one and two laryngeally marked segments, exactly the contrast that is shown to be relatively perceptually weak.

Harmonic Grammar is a model of linguistic constraint interaction in which well-formedness is calculated in terms of the sum of weighted constraint violations. We show how linear programming algorithms can be used to determine whether there is a weighting for a set of constraints that fits a set of linguistic data. The associated software package OT-Help provides a practical tool for studying large and complex linguistic systems in the Harmonic Grammar framework and comparing the results with those of OT. We first describe the translation from harmonic grammars to systems solvable by linear programming algorithms. We then develop a Harmonic Grammar analysis of ATR harmony in Lango that is, we argue, superior to the existing OT and rule-based treatments. We further highlight the usefulness of OT-Help, and the analytic power of Harmonic Grammar, with a set of studies of the predictions Harmonic Grammar makes for phonological typology.

Opaque interactions have long been recognised as a challenge for Optimality Theory. We show that although there has been considerable effort to bring opacity into the scope of Optimality Theory, some types of process interactions are still problematic for the theory. Based on data from Tundra Nenets, we present and analyse a case of fed counterfeeding in which a process A feeds a process B, and B counterfeeds A. We argue that such interactions present a challenge to Optimality Theory with Candidate Chains (OT-CC; McCarthy 2007) since the two interactions impose contradictory ranking requirements. We propose an extension of the theory that does not abandon its main assumptions and that makes fed counterfeeding analysable in OT-CC. This extension is based on the assumption that constraints can make reference to the position specified in a previous step in the derivation.

Abstract phonological patterns and detailed phonetic patterns can combine to produce unusual acoustic results, but criteria for what aspects of a pattern are phonetic and what aspects are phonological are often disputed. Early literature on Romanian makes mention of nasal devoicing in word-final clusters (e.g. in /basm/ ‘fairy-tale’). Using acoustic, aerodynamic and ultrasound data, the current work investigates how syllable structure, prosodic boundaries, phonetic paradigm uniformity and assimilation influence Romanian nasal devoicing. It provides instrumental phonetic documentation of devoiced nasals, a phenomenon that has not been widely studied experimentally, in a phonetically underdocumented language. We argue that sound patterns should not be separated into phonetics and phonology as two distinct systems, but neither should they all be grouped together as a single, undifferentiated system. Instead, we argue for viewing the distinction between phonetics and phonology as a largely continuous multidimensional space, within which sound patterns, including Romanian nasal devoicing, fall.

This paper proposes a model of stress assignment in which metrical structure is built serially, one foot at a time, in a series of Optimality Theory (OT)-style evaluations. Iterative foot optimisation is made possible in the framework of Harmonic Serialism, which defines the path from an input to an output with a series of gradual changes in which each form improves harmony relative to a constraint ranking. Iterative foot optimisation makes the strong prediction that decisions about metrical structure are made locally, matching attested typology, while the standard theory of stress in parallel OT predicts in addition to local systems unattested stress systems with non-local interactions. The predictions of iterative foot optimisation and parallel OT are compared, focusing on the interactions of metrical parsing with syllable weight, vowel shortening and constraints on the edges of prosodic domains.

Despite differences in parsimony and philosophical orientation, physical and abstract theories of phonology often make similar empirical predictions. This study examines a case where they do not: gemination restrictions in Hungarian. While both types of theory correctly prohibit the lengthening of a consonant when flanked by another consonant, they make different predictions regarding both the relative duration changes within a target consonant and the applicability of restrictions to lengthening processes besides gemination. In two speech-production experiments, these predictions are evaluated by measuring stop and frication durations within affricates. Results show that relative duration changes occur, and that the restriction holds only for gemination, supporting an abstract theory. Yet results also indicate that gemination exhibits sensitivity to inherent durational differences between affricates, providing some support for a physical theory. Thus I argue that an adequate theory of phonology must include abstract constituents, alongside a limited, principled set of physical landmarks.

Phonological patterns often have phonetic bases. But whether phonetic substance should be encoded in synchronic phonological grammar is controversial. We aim to test the synchronic relevance of phonetics by investigating native Mandarin speakers' applications of two exceptionless tone sandhi processes to novel words: the contour reduction 213→21/—T (T≠213), which has a clear phonetic motivation, and the perceptually neutralising 213→35/—213, whose phonetic motivation is less clear. In two experiments, Mandarin subjects were asked to produce two individual monosyllables together as two different types of novel disyllabic words. Results show that speakers apply the 213→21 sandhi with greater accuracy than the 213→35 sandhi in novel words, indicating a synchronic bias against the phonetically less motivated pattern. We also show that lexical frequency is relevant to the application of the sandhis to novel words, but cannot account alone for the low sandhi accuracy of 213→35.