In any study of empirical phenomena, the unusual holds particular attraction. The Bird of Paradise with its amazing feathers, the monolithic baobabs and the joint nursing of emperor penguins are phenomena that catch the imagination of the specialist and the interested lay person alike. The scientific study of ‘outliers’ holds particular promise, often revealing a system's complexity that is not evident when studying simpler phenomena.

This paper argues that rather than just select the best candidate, EVAL imposes a harmonic rank-ordering on the full candidate set. Language users have access to this enriched information, and it shapes their performance. This paper applies this idea to variation. The claim is that language users can access the full candidate set via the rank-ordering imposed by EVAL. In variation, more than one candidate is well-formed enough to count as grammatical. Consequently, language users will access more than just the best candidate from the rank-ordering. However, the accessibility of a candidate depends on its position on the rank-ordering. The higher the position a candidate occupies, the more likely it is to be selected. In a variable process, variants that appear higher on the rank-ordering (i.e. are more well-formed) will therefore also be the more frequent variants. This model is applied to variation in the phonology of Faialense Portuguese and Ilokano.

This paper addresses theoretical issues confronting cross-height harmony systems through an experimental study of Kinande, a Bantu language of the Democratic Republic of Congo. Using a combination of acoustic analysis and lingual ultrasound imaging, we evaluate previous proposals concerning the phonetic correlates of the harmonic vowel feature and the transparency of low vowels. Results indicate that (i) although a multivalued scalar acoustic feature in F1/F2 space is not adequate to distinguish all vowel categories in Kinande, the cross-height feature does correlate acoustically with F1, (ii) the cross-height feature of Kinande involves systematic tongue-root articulations and (iii) low vowels in Kinande are not neutral to harmony in the way reported in earlier work, but exhibit significant and systematic tongue-root advancement and retraction according to the dictates of harmony.

The East Slavic languages have distinctive stress and no contrastive vowel length; the phonetic correlates of word stress are increased vowel duration, intensity and changes in vowel quality. But there are East Slavic dialects with an unusual type of word prosody: word stress is distinctive, yet the immediately pretonic vowel is as long as or longer than the vowel under stress, and there is a fixed tonal contour over the two syllables. I argue that the additional duration in the pretonic syllable is due to a type of stress-to-tone mapping in which a lexical high tone is assigned to the pretonic syllable. Variation across these and other closely related dialects reveals a typology of stress and tone mapping with increasing dependence on intrinsic vowel duration and sonority. The additional vowel length on neutralised/reduced vowels in the immediately pretonic position in some dialects raises questions about the relationship of vowel duration and vowel reduction in East Slavic. The prosody of these East Slavic dialects appears to be typologically unusual in having both lexical stress and non-contrastive tone, and in specifically not aligning high tone with stress.

Vowel sounds may be inserted into a word by two mechanisms: insertion of a vocalic articulatory gesture (epenthesis), or retiming of existing gestures to produce a vowel-like transition between consonants (intrusion). I argue that epenthetic vowels are phonological units but intrusive vowels are not. A representation using abstract gestures as well as segments can capture facts about the typology of vowel intrusion.

In the Yakima dialect of Sahaptin, the smallest forms of words are CCV and CVC, as established by a study of lexical patterns and the obligatory augmentation of subminimal roots. We argue that CCV words cannot be considered a type of disyllable, nor do the initial consonants in CCV words exhibit properties (apart from minimality) that would allow them to be classified as moraic. We therefore propose that Yakima Sahaptin represents a previously unrecognised type of minimality, biconsonantal minimality. This kind of minimality requires a theoretical framework in which consonants can be distinguished from vowels, as do other phonological phenomena.

In Hungarian, stems ending in a back vowel plus one or more neutral vowels show unusual behaviour: for such stems, the otherwise general process of vowel harmony is lexically idiosyncratic. Particular stems can take front suffixes, take back suffixes or vacillate. Yet at a statistical level, the patterning among these stems is lawful: in the aggregate, they obey principles that relate the propensity to take back or front harmony to the height of the rightmost vowel and to the number of neutral vowels. We argue that this patterned statistical variation in the Hungarian lexicon is internalised by native speakers. Our evidence is that they replicate the pattern when they are asked to apply harmony to novel stems in a ‘wug’ test (Berko 1958). Our test results match quantitative data about the Hungarian lexicon, gathered with an automated Web search. We model the speakers' knowledge and intuitions with a grammar based on the dual listing/generation model of Zuraw (2000), then show how the constraint rankings of this grammar can be learned by algorithm.

A dispreference for onsetless syllables is often manifested by processes of consonant insertion or vowel deletion which eliminate onsetless syllables, and it is sometimes manifested by disallowing prosodic distinctions which onsetful syllables make, e.g. onsetless syllables may resist bearing tone or stress. This paper investigates the interaction between minimality and the structure of onsetless syllables in the Bantu language Zinza, documenting a previously unobserved onsetless effect which is partially conditioned by minimality. In Zinza, word-initial vowels of disyllables are lengthened, so /o-ljá/ becomes [óó-lja] ‘you eat’ (compare [o-líma] ‘you cultivate’ and [bá-lja] ‘they eat’, with no lengthening). These facts test previous theories of onsetless effects (Downing 1993, 1998, Odden 1995, Orie 2000). I argue that a special representation of onsetless syllables is not warranted, and does not explain this pattern of lengthening. A constraint against word-initial short onsetless syllables interacting with word minimality suffices to explain this lengthening.

The standard method of describing phonation for tone, vocal register, stress and other linguistic categories relies on the ‘continuum hypothesis’ that linguistic sounds are produced by means of glottal states determined by the aperture between the arytenoid cartilages, the endpoints of the voiceless–voiced continuum being ‘open glottis’ and ‘closed glottis’. This paper takes a different view, pointing out that many languages make use of a number of valves, and that these valves are not articulations on a glottal continuum but represent a synergistic and hierarchical system of laryngeal articulations. These valves constitute a principal source of phonological contrast, with an influence on how oral articulatory events are characterised.

Nubi is a language with culminative word prominence, spoken in Uganda and Kenya. The prominence, or accent, falls within a right-edge three-syllable window, and at the phrasal level deaccenting and rhythm-related accent shift occur in specific phonological and morphological contexts. Since accent is used to express morphological categories, and deaccenting and accent shift are morphologically conditioned, the functional load of accent is comparable to that of morphologically conditioned stress in other languages. Accented syllables are associated with H tones, with default L tones appearing between them. Typologically, Nubi is not a stress-accent language, because unlike for example English, it has no stress distinctions independently of the distinction between accented and unaccented syllables. It is also non-tonal, because accent marking is not lexically idiosyncratic, and because it has no option other than to supply a H tone is accented syllables, it is unlike a prototypical intonation language that has pitch accents with discoursal meanings.

Numerous proposals have been advanced as to how prosodic systems should be typologised. In this paper I distinguish two prototype systems, tone and stress accent, which, unlike systems analysed as ‘pitch accent’, have two inviolable, definitional properties: (i) obligatoriness (every word has at least one stress accent); (ii) syllable-dependency (the stress-bearing unit is necessarily the syllable). In contrast, the oft-cited criterion of culminativity (every word has at most one tone/accent) not only includes tone/accent systems that are neither obligatory nor syllable-dependent, but also culminative non-prosodic features that are clearly not accentual. I argue that there is no one pitch-accent prototype. Instead, since tone and stress accent may co-occur, and since languages may ‘pick and choose’ between the non-definitional properties that tend to cluster within the tone vs. stress-accent prototypes, there is a range of intermediate (and possibly indeterminate) word-prosodic systems which may or may not be best seen as ‘types’.

This paper examines the cues for and interaction between the metrical and tonal systems of Kera. Kera has no word-level stress, but the heads of its quantity-sensitive iambic feet are cued by duration, intensity and vowel allophony; in addition, foot boundaries are identified by vowel-harmony and tone-spreading domains. The tonal system has three underlying tones, which are enhanced by differences in voice onset time (VOT). Kera demonstrates an interaction between the iambic foot and tone. Words have one or two tones. For words of three or more syllables the tone-bearing unit is the foot, but for shorter words it is the syllable. This dichotomy is accounted for by a faithfulness constraint requiring all tones to surface, which overrides a constraint limiting each foot to one tone. Kera shows that a tonal system can be sensitive to metrical structure, while maintaining a certain independence between the two systems.

Acoustic evidence suggests that contemporary Seoul Korean may be developing a tonal system, which is arising in the context of a nearly completed change in how speakers use voice onset time (VOT) to mark the language's distinction among tense, lax and aspirated stops. Data from 36 native speakers of varying ages indicate that while VOT for tense stops has not changed since the 1960s, VOT differences between lax and aspirated stops have decreased, in some cases to the point of complete overlap. Concurrently, the mean F0 for words beginning with lax stops is significantly lower than the mean F0 for comparable words beginning with tense or aspirated stops. Hence the underlying contrast between lax and aspirated stops is maintained by younger speakers, but is phonetically manifested in terms of differentiated tonal melodies: laryngeally unmarked (lax) stops trigger the introduction of a default L tone, while laryngeally marked stops (aspirated and tense) introduce H, triggered by a feature specification for [stiff].

The Northern and Western dialects of the Mon-Khmer language Kammu use fundamental frequency to distinguish words, while the Eastern dialect relies on the contrastive voicing of initial consonants to distinguish words which display an F0 contrast in the other dialects. The production and perception of tones in the three dialects is investigated, using recordings and perception tests. As expected, Northern and Western Kammu speakers do use F0 to distinguish words, as seen both in their production and in perception tests, which show that they could distinguish small F0 differences. In contrast, Eastern Kammu speakers neither produce nor perceive F0 differences at the lexical level. Furthermore, the voiceless and voiced stops in Eastern Kammu cause smaller F0 perturbations in the following vowel than might be expected for a language undergoing tonogenesis. The significance of these results for tonogenesis is discussed, as well as their relevance for the notion of ‘tone language’.