In order to account for the accentual and rhythmical structure of English, a binary-branching prosodic constituent structur is assumed, in which minimally the syllable and the foot must be headed. Feet are potentially marked as accented. This representation makes it possible to describe the prominence patterns of word groups as resulting from three accent deletion rules, the Compound Rule, the Initial Accent Deletion Rule and the Rhythm Rule. It was shown that the structural change effected by Initial Accent Deletion cannot be expressed in theories which represent stress as a relative concept. Moreover, this rule, which like the Compound rule is a lexical rule, provided evidence for the existence of a stratum in the lexical phonology of English in which compounding and so-called ClassII derivation take place. The Rhythm Rule is a postlexical rule, which was shown to apply to the output of the other two rules. Without the aid of any conditions or constraints, it accounted effortlessly for the stress-shift data presented in the recent literature. It could moreover be shown that apparent cases of stress shift in unaccented speech (in which the Rhythm Rule does not apply) should not in fact be viewed as the output of any stress-shift rule at all, but should be explained as the effect of preboundary lengthening as applying to the different constituents in the prosodic hierarchy. It was argued that an analysis of sentence accentuation whereby focused constituents have to be assigned accents can run into problems that do not exist in a ‘deaccenting’ analysis, in which nonfocused constituents are deprived of their accents. Finally, it was argued that English, unlike Dutch, lacks phonological rules that refer to primary word stress, and that, at best, primary stress may reveal itself in low-level timing distinctions.

Three important strands of research characterise modern phonological theory; (i) the cyclic interaction of morphology and phonology; (ii) prosodic structure as it is built from segments into moras, syllables, feet and ultimately the phonological word; and (iii) the precise internal structure of segments, including feature geometry, how these features may reside on autosegmental tiers, and how they may be underspecified.

The theory of PROSODIC CIRCUMSCRIPTION (McCarthy & Prince 1990a) is a general approach to the problem of limiting the domain of rules to less than a morphological constituent. For example, in the Arabic singular/ plural pairs ndub/anaadib ‘locust’ and sulṭaan/salaaṭiin ‘sultan’, vowel length in the final syllable remains unaltered despite significant changes in the shape of the rest of the word. Prosodic circumscription theory partitions the singular base into affected (aun, sul) and unaffected (dib, ṭaan) portions, with only the affected portion mapped onto a light–heavy (or iambic) template.

Over the last few years, much work in phonology has been devoted to exploring the way features are specified for segments; in particular, to what extent feature specification may be underlyingly present and/or acquired by rule or default in the course of a derivation. While a number of proposals have been made attributing various degrees of underspecification to abstract levels of the phonology (Kiparsky 1985; Steriade 1987; Archangeli 1988), it has been generally assumed that where phonetic implementation comes into play, i.e. at the end of the derivation, segments are exhaustively specified.

An important contribution to our knowledge of tone sandhi among the Chinese dialects is Lü's (1980) article on the tones and tone sandhi behaviour of Danyang, a Wu dialect of Chinese. Lü's description of Danyang is, to date, our only source on the dialect. While it is a northern Wu dialect, the tone sandhi patterning in Danyang differs from Shanghai, Suzhou, Wuxi and other dialects in the vicinity. There are a number of interesting problems related to tone in the dialect. This paper restricts the topic to only one of these problems, namely the treatment of the six basic tone patterns in Danyang, focusing in particular on the pattern in which a contour tone is copied Onto adjacent syllables in the tone sandhi domain.

Phonological underspecification plays an important role in phonological theory. Some features are left blank in underlying representations. If they are relevant to the pronunciation of a segment, they are filled in at some point in the derivation; otherwise, they are left blank permanently. When underspecification was reintroduced to phonological theory in the 1980s, researchers originally assumed that all information that could be considered predictable had to be underspecified in underlying representations (Kiparsky 1982; Archangeli 1984). Information can be considered predictable in one of two ways. First, information is predictable if it is redundant or allophonic; the voicing and nasality of the vowel in the word grin [grin] in English are predictable (because vowels are always voiced, and are always nasalised before a nasal) and omitted from underlying representations. Second, given that a feature is binary, it is possible to leave one value of the feature blank in underlying representations; if the segment is not specified as e.g. [+ F], then it must be [− F], by default. The approach that defines predictability in this fashion is known as RADICAL UNDERSPECIFICATION

A basic assumption in current phonology is that features are not an unstructured set like (1a), but are arranged hierarchically into a constituent structure tree such as (1b), proposed in Sagey (1986: 61):

Other groupings of features have been proposed (Clements 1985; Steriade 1987; inter alii).

A fundamental assumption of theories like (1b) is that rules operate on constituents, an assumption which is formalised by limiting rules to spreading or delinking a single node.

The role which the pharynx plays in the production of speech sounds has long been the subject of debate, in particular as regards the independence of the pharynx from other speech articulators and in particular the larynx. Given that the linguistic and anatomical information about the larynx and the pharynx is quite complex, the objective of this paper is to review the literature relevant to the relationship between these two articulators and to recast some of this information in terms of current articulator-based theories of phonological features such as the ones reviewed in McCarthy (1988). 1 presents two pharyngeal features [ATR/ RTR] ‘advanced tongue root’ and ‘LL/RL’ ‘lowered larynx’separately in two subsections, with definitions, examples and arguments in favour of having two pharyngeal features rather than one. § 2 argues that pharyngeal features are independent of the primary point of articulation features.

The functional equivalence of CVV and CVC syllables, as opposed to CV syllables, is a time-honoured observation holding true for numerous languages over a variety of phonological and morphological phenomena, including stress assignment (cf. Newman 1972 for a review). Traditionally, the opposition between the two types of syllables has been informally described by reference to syllable weight: CVV and CVC syllables are heavy, CV syllables are light (e.g. La Grasserie 1909: 31–32). It has also been observed, however, that in languages sensitive to the CV/CVV distinction, CVC syllables do not necessarily pattern with CVV syllables, but may instead pattern with CV syllables, thus counting as light rather than heavy (Hyman 1985: 5–6; McCarthy & Prince 1986: 32–34; Hayes 1989: 255–256).

Hayes (1990) proposes the adoption of a new notation for representing hierarchical phonological structure, in response to a notational problem which he calls the DIPHTHONGISATION PARADOX. More generally, he seeks to find a replacement for the current method of depicting feature trees, which is not ideally suited to the clear expression of rules and derivations. While I am in complete agreement with this standpoint and fully support Hayes' assertion that certain ambiguities of interpretation have crept into the graphical conventions currently in widespread use, I nevertheless believe his proposal suffers from a number of problems. Once identified, these problems can be solved with further notational and formal refinements.

Although primary word stress regularly falls on the penult in Polish and on the antepenult in Macedonian, there are a number of lexical exceptions in both languages. In the first generative treatment of such exceptions, Comrie (1976) suggested two unrelated diacritic features, [± stressable] for Polish and [ ± never posttonic] for Macedonian, in order to accommodate the accentual paradigms exhibited by exceptional words within the framework of Chomsky & Halle (1968). More recently, metrical accounts of exceptional stress have been proposed in Franks (1985), Halle & Vergnaud (1987) and Rubach & Booij (1985) for Polish and in Franks (1987, forthcoming) and Halle & Vergnaud (1987) for Macedonian. These analyse deviations from the regular patterns in the two languages in completely unrelated ways – in Polish exceptional stress is a consequence of idiosyncratic extrametricality, whereas in Macedonian it results from the idiosyncratic presence of an inherent accent. Responding to this type of analysis, Hammond (1989) argues that an alternative treatment in which exceptional stress in both languages is treated similarly is conceptually more elegant and descriptively superior. He accomplishes this by employing roughly the same set of stress rules for Polish and Macedonian, with the exception that lexical accent is interpreted differently in the two languages.

It has been demonstrated that metaphony is triggered by underlying final /-i/ and /-u/. The important feature that unifies the two posttonic vowels, which are responsible for the tonic vowel alternations, is [+ high[. The spreading of this feature to the tonic vowel accounts for metaphony. Vocalisms defined without the feature [high], such as those of particle phonology and Goldsmith's model, fail to mirror the metaphony process.

Hall (1989) introduces a rule of Fricative Assimilation (FA) in German, which, he claims, poses a challenge to the principle of Structure Preservation in Lexical Phonology, as presented in Kiparsky (1985). This claim is based on the observation that FA is demonstrably lexical because it respects morpheme boundaries, but nonetheless introduces a nondistinctive feature, thus violating a marking condition. However, Hall has not appreciated the force of the analysis of Catalan in Kiparsky (1985), which suggests that assimilated sequences may show special behaviour with respect to marking conditions. In this paper we show first, based on arguments in Kiparsky (1985), Hayes (1986) and Itô (1988), that a general constraint on the interpretation of autosegmental formalism specifically rules out the application of the marking condition to the output of FA.