Sound guide to hearsay evidence

This hefty volume is the first in what promises to be a major series of linguistics handbooks from Blackwell. It is an authoritative reference work containing a collection of literature surveys as well as fresh and original articles on mainstream generative phonology in the early 1990s. The intended reader is a person with an advanced knowledge of phonological theory who wishes to consult a fairly concise, up-to-date account of the some sub-area of phonological theory or a sketch of the phonology of some well-studied language.

The contributions to the book run to 32 chapters and fall into two broad categories. The first 11 chapters all deal with central issues in recent generative phonological theory while the remaining 21 chapters examine a melange of topics. Some are studies in the phonology of languages or language groups that have figured prominently in the development of phonological theory over the past 20 years while others deal with topics as diverse as the acquisition of phonology, the phonological basis of sound change, rule ordering, vowel harmony, the syntax/phonology interface, sentence prosody and experimental phonology. Interspersed with these chapters are studies in the prosodic aspects of phonology. In addition, there are two chapters on what I will call "alternative" generative phonology models (Sanford A. Schane's theory of particle phonology and Colin J. Ewen's chapter on dependency phonology) .

The key issues that the volume is concerned with are: phonotactics (constraints on the configuration of phonological words and higher phonological domains); alternations (the differences in the phonetic manifestation of morphemes in different contexts); contrastiveness (the ways in which differences in sounds are used to mark lexical or grammatical distinctions) and the interaction between the different sub-components of the grammar.

Throughout the book one or other of these themes is never far from the centre of the exposition, be it in primarily theoretical chapters or in chapters which focus on the description and analysis of particular languages. This blend of the theoretical and the descriptive is one of the book's many strengths. I will illustrate it with the treatment of phonotactics. There are several contributions which are all concerned with different facets of the problem of defining phonotactic constraints.

They include a chapter on "Prosodic Morphology" which shows that in many languages the phonological word is minimally bi-moraic; a chapter on the role of skeletal positions and moras in determining the configuration of words; a study of cross-linguistic constraints on the distribution of segments in feet and syllables; an analysis of constraints on quantity in words, and a contribution on the distribution of features within and across syllable domains in Japanese. In addition, there are two chapters which investigate the phonological domains beyond the word. They show that a useful way of handling phonotactic constraints is to posit a prosodic hierarchy whose smallest unit is the individual autosegment or feature, going up the scale to the skeletal position, the syllable, the foot, the phonological word, the phonological phrase, the intonation phrase and the utterance.

The treatment of alternations in the realisation of sounds representing the same morpheme in different contexts is another dimension of phonology that receives extensive coverage. George N. Clements and Elizabeth V. Hume present an articulator-based version of feature geometry in which vocal tract constriction plays a key role. There are also articles on diphthongisation, on Semitic phonology, on skeletal positions and moras which also tackle alternations and the issues they raise especially in connection with the relation between segments and the skeletal tier. Likewise, distinctiveness receives methodical treatment, with language-specific studies complementing the more theoretical investigations.

Although overall this is a truly outstanding book, I have a few reservations. Space was found for interesting but marginal topics like the phonology of language games and the phonological basis of sign language at the expense of coverage of significant "alternative" generative approaches to phonology such as government phonology and declarative approaches to phonology. (The only "alternative" models presented are particle phonology and dependency phonology.) Furthermore the rise of nonlinear models of phonological representations has made batteries of language-specific derivational rules a thing of the past. In view of this, one may ask why room was found for a chapter on rule ordering.

Finally, after the volume went to press, there was a paradigm shift in mainstream phonology. The rule-free, constraints-based model of optimality theory, the seeds of which were already sown in the model of prosodic morphology proposed by McCarthy and Prince in this volume and elsewhere, quickly established itself as the new orthodoxy and supplanted the earlier rule-based, derivational model. The result is that some of the assumptions on which this volume is based are not considered valid anymore. In any event, there is much in this volume that will remain valuable.

Notwithstanding the above misgivings this book is a towering achievement which will no doubt become a classic. Every phonologist should have it as a constant companion. The snag is the £79 price tag. Most of us may have to resort to reading a library copy.

Francis Katamba is reader in linguistics, University of Lancaster.