I propose that the notions of segment and phoneme be enriched to allow some concurrent clustering, even in classical theories. My main application is the Khoisan language !Xóõ, where treating clicks as phonemes concurrent with phonemic accompaniments allows the inventory size to be radically reduced, so solving the problems of many unsupported contrasts. I show also how phonological processes of !Xóõ can be described more elegantly in this setting, with support from metalinguistic evidence and production-task experiments. I describe a new allophony in !Xóõ. I go on to discuss other applications, some rather speculative, of the concept of concurrent phoneme. The article also provides a comprehensive review of the segmental phonetics and phonology of !Xóõ, together with previous analyses.

In Korean compound nouns, the onset of the second element tensifies: /pjəl + picʰ/ → [pjəl.p*icʰ]. This process displays a substantial number of exceptions. We show that in the Yanbian dialect phonological factors influence compound tensification; their influence is confirmed by the results of a wug test. These results mostly mirror generalisations informed by lexical data, but a pattern not clearly observed in the existing lexicon also emerged (analytic bias or ‘overlearning’), while phonologically implausible interactions were not actively incorporated (‘underlearning’). Preliminary acoustic analysis motivates an OCP restriction against multiple laryngeally marked segments, due to the shared feature [long non-modal voicing]. But in simplex nouns the co-occurrence of two tense onsets is overrepresented. This apparent contradiction is explained by a directionality difference: compound tensification is progressive, as opposed to the regressive assimilation of creaky voice from the vowel of the following syllable in simplex words.

Nivkh, a linguistic isolate of Eastern Siberia, displays a typologically uncommon process which affects /e/ in the context of a following uvular consonant, and which reportedly results in palatalisation of the consonant preceding /e/, e.g. /pʰeq/→[pʰʲq]. Phonologically, there are indeed reasonable grounds for analysing the consonants as palatalised. Phonetically, matters are less clear. Acoustic evidence shows that the clearest correlate of palatalisation is a steep and consistent fall of F2 during the realisation of the vowel. The fact that F2 lowering is also typical before uvulars suggests that palatalisation is the result of a perceptually motivated reanalysis, whereby the palatality of /e/ has come to be associated with the preceding consonant, and the vowel has a lowered and retracted reflex.

In the paradigms of certain Tigre verbs, the final stem consonant and the vocalic suffix metathesise in some configurations. In this short paper, this is analysed as a hiatus-resolving strategy. The limits and consequences of this strategy are shown to follow from the representation of verbs, in an analysis conducted within CVCV phonology and Element Theory.

Many languages have restrictions on word-final segments, such as a requirement that any word-final obstruent be voiceless. There is a phonetic basis for such restrictions at the ends of utterances, but not the ends of words. Historical linguists have long noted this mismatch, and have attributed it to an analogical generalisation of such restrictions from utterance-final to word-final position. To test whether language learners actually generalise in this way, two artificial language learning experiments were conducted. Participants heard nonsense utterances in which there was a restriction on utterance-final obstruents, but in which no information was available about word-final utterance-medial obstruents. They were then tested on utterances that included obstruents in both positions. They learned the pattern and generalised it to word-final utterance-medial position, confirming that learners are biased toward word-based distributional patterns.

The Iambic-Trochaic Law (ITL) asserts that listeners associate greater acoustic intensity with group beginnings and greater duration with group endings. Some researchers have assumed a natural connection between these perceptual tendencies and universal principles underlying linguistic categories of rhythm. The experimental literature on ITL effects is limited in three ways. Few studies of listeners' perceptions of alternating sound sequences have used speech-like stimuli, cross-linguistic testing has been inadequate and existing studies have manipulated intensity and duration singly, whereas these features vary together in natural speech. This paper reports the results of three experiments conducted with native Zapotec speakers and one with native English speakers. We tested listeners' grouping biases using streams of alternating syllables in which intensity and duration were varied separately, and sequences in which they were covaried. The findings suggest that care should be taken in assuming a natural connection between the ITL and universal principles of prosodic organisation.

This paper presents a novel approach to capturing exceptional stress that relies on prespecification of foot edges in the input. Focusing on Turkish, this approach accounts for both regular and exceptional stress in a unified manner and within a single grammar, and unlike other approaches, does not overpredict. On this account, Turkish is a footless, but trochaic, language. Both regular and exceptional Turkish morphemes are subject to the same constraint ranking; exceptional morphemes are different only in that they have one or more syllables already footed in the input, although the type of foot (e.g. trochaicity, binarity) is determined by the constraints of the grammar. As regular morphemes vacuously satisfy these constraints (which act on the foot), trochees appear on the surface only if there is an input foot available (i.e. in words with exceptional morphemes), since the grammar itself has no apparatus to parse syllables into feet.

Within the autosegmental-metrical approach to French intonation, the existence of an intermediate phrase or ip is controversial. Our study provides strong evidence for its existence, by uncovering systematic pitch-scaling effects within this constituent. We first show that the presence of an ip break is responsible for blocking recursive downstep of subsequent AP-final LH* rises in declarative utterances, causing the return of the final H target to the pitch level set by the first accentual peak of the phrase (i.e. complete reset). Additional evidence for the internal structuring of the Intonation Phrase is also provided by partial reset on the postboundary H target as well as by preboundary lengthening on the last syllable of the ip. The pitch-scaling effects are claimed to result from control over the reference pitch level for the entire ip, which can alternatively be modelled through secondary association of the last pitch accent of the domain.

It has been assumed that tonal alignment is not contrastive in contour tones (e.g. Hyman 1988, Odden 1995, Yip 2002). However, Remijsen (2013) and DiCanio et al. (2014) have recently reported evidence for this configuration. In relation to this controversy, we report on an acoustic analysis of the tone system of Shilluk. The dataset is built around the contrast of Low vs. Early-aligned High Fall vs. Late-aligned High Fall vs. High in closed monosyllabic stems with a short vowel. The results support the hypothesis that tonal alignment is contrastive in falling contour tones in Shilluk: the two falling contours differ consistently and significantly in terms of tonal alignment, relative both to one another and to phonetically similar level-tone configurations. The falling contours do not differ significantly in terms of a number of other phonetic parameters (F0 height, size of F0 change, duration).

Scottish Gaelic has been cited as providing an instance of vowel excrescence (Hall 2006). One of the defining properties of excrescent vowels is that they are phonologically inert and are not motivated by – nor do they contribute to – the syllable structure of a language. In this paper, we report on a series of experiments which tap into native speakers' intuitions of syllable structure in Scottish Gaelic. Insofar as intuitions about syllable count and syllabification reflect phonological structure, our results suggest that the relevant vowels of Scottish Gaelic are not phonologically inert, and contribute directly to native speaker intuitions involving the number of syllables and the affiliation of consonants to those syllables. However, our results also establish that the relevant vowels have an intermediate phonological status, which also distinguishes them from underlying vowels.

This paper documents and analyses stress and vowel length in Samoan words. The domain of footing, the Prosodic Word, appears to be a root and cohering suffixes; prefixes and most disyllabic suffixes form a separate domain. Vowel sequences that disrupt the normal stress pattern require constraints matching sonority prominence to metrical prominence, sensitive to degree of mismatch and to the number of vowels involved. Two suffixes unexpectedly have an idiosyncratic footing constraint, observable only in a limited set of words. We also discuss trochaic shortening and its asymmetrical productivity, and the marginal contrastiveness of some features in loans. While Samoan does not appear to be typologically unusual, it does offer arguments (i) in favour of alignment constraints on Prosodic Words rather than only on feet directly, and (ii) against simple cyclicity. Some of the strongest evidence comes from stress patterns of the rich inventory of phonotactically licit vowel sequences.

One of the major attributes of autosegmental phonology is the possibility of reducing procedural techniques of morphological exponence to a generalised concept of concatenation. This research programme, which equates the triggers of non-concatenative processes with affixes consisting of incomplete autosegmental or prosodic representations, is called Generalised Non-linear Affixation in Bermúdez-Otero (2012). In this paper, we argue that the Generalised Non-linear Affixation analysis of segmental lengthening by mora affixation extends naturally to subtractive morphology. Defective (phonetically uninterpretable) integration of an affix mora into the prosodic structure of its base triggers deletion and shortening. We show that this approach derives all major types of quantity-manipulating morphology (vowel shortening, segmental subtraction and vowel-length polarity), and thus demonstrate that Generalised Non-linear Affixation extends fully to subtractive morphology, which has been seen as the ultimate problem for a concatenative reanalysis (Anderson 1992).

This paper reassesses the hypothesis that early phonotactic learning of constraint-based grammars relies on the Identity Map – i.e. it uses observed surface forms as the inputs which cause errors and drive learning via constraint reranking. We argue that this approach's success is closely tied to Optimality Theory's fully parallel grammatical evaluation. In the constraint-based derivational framework of Harmonic Serialism (HS; McCarthy 2000, 2007b), reliance on observed surface forms as inputs can block the discovery of ‘hidden rankings’ between markedness constraints, preventing the learner from discovering a restrictive grammar. This paper illustrates the problem, using a pattern of positional vowel restrictions in Punu (Kwenzi Mikala 1980), and considers the role of various learning assumptions. We conclude that hidden rankings are a fundamental obstacle to restrictive error-driven learning in any HS-like framework, and that learning in such frameworks inevitably requires consideration of some unattested surface forms as inputs, even at the earliest learning stages.

Linguists have been attempting to define the range of locations in which infixes can occur since Ultan's pioneering work in 1975, but to date there has been no unambiguous evidence for infixation after the first syllable, despite previous (now controversial) claims of its existence by Ultan (1975) and Moravcsik (2000), as well as its predicted existence by Yu's Salient Pivot Hypothesis (‘phonological pivots must be salient at the psycholinguistic or phonetic level’) (2003, 2007). Previously examined potential examples are controversial due to restricted patterns and the acceptability of alternative analyses such as a first-vowel pivot or a foot-based pivot (Samuels 2010). In this article, I present strong evidence from fieldwork on Yeri, an endangered Torricelli language of Papua New Guinea, that imperfective and additive morphemes productively occur as infixes after the first syllable of the verb stem, and that a first-vowel or foot-based analysis cannot account for their position.

In his recent bid for the presidency of the Association for Mathematics of Language, Makoto Kanazawa writes:*