Versed in oafish subtlety

If language's narrative is syntax, phonology is its verse. Syntax has traditionally explored the combinability or otherwise of larger chunks of meaning as they are expressed and perceived through time, as in the grammaticality of "Peter sought the root cause of the problem" versus "Sought of root problem the cause the Peter". Phonology, in contrast, has primarily concerned itself, since Roman Jakobson's Copenhagen lectures in 1939, with studying a language's sounds in their own right, probing, analysing, even attempting to explain the organisation and relations of a language's sounds to one another: their significant form. In recent decades, general phonology - the study of phonology not with regard to a particular language but sub specie aeternitatis - has concerned itself chiefly with theory-building, sometimes on the basis of data from, and analyses of, particular languages, but unfortunately at least as often on the discarded remnants of earlier theories.

This volume is one of a series from Oxford University Press. Though its focus is the phonology of Hungarian, it is also deeply committed to formulating its message in terms of the code of theoretical phonology.

Much about the synchronic phonology of Hungarian has been known for about a century. This book re-states much of this knowledge, occasionally with misattributions to the literature (Robert Austerlitz's 1950 publication on Hungarian phonemics, cited on page four, was an MA dissertation, not a PhD thesis). The re-statements are made in metaphors from a dialect of the metalanguage of generative grammar and its sequelae; metaphors that have been most helpful in pinpointing complex subtleties of English syntax, but which have thrown little or no new light on subtleties such as the occurrence of, say, f or v in "oaf-oafs" but "loaf-loaves".

Hungarian has a great deal of phonology, and much of the complexity of its phonology is due to the fact that it is tied up, as in the English examples above, with its morphology - a morphology that is not as classically agglutinative as the handbooks would have it, by the way. The authors kick analogous problems in Hungarian (for example, nominative/ accusative dal / dalt "song", but fal / falat "wall") upstairs into the lumber-room of the lexicon (their term is "lowering stems"), as if they cannot see a connection with what they call "final vowel shortening stems" - or because they see the connection in reverse. In similar fashion, the connection with "quaternary harmony" is merely noted in passing, although the unroundedness of the second vowel of hölgyek "ladies" is intimately connected with the quality of the second vowel of inak "sinews" (anti-harmonic stems). They marginalise even further stems that behave with more complexity or unpredictability, such as tó "lake" (accusative tavat ) and choosing to "ignore" them in a footnote.

The authors assume that it is the underlying identicalness among the world's languages that is important and that the differences are peripheral. But the opposite viewpoint is equally valid.

Nevertheless, this is an important book. It makes its statements clearly, albeit in a metalanguage that restricts its accessibility. In the twilight of the generative paradigm, it might have communicated more competently if the authors had framed what they have to say in more classical terms. Nevertheless, specialists in theoretical phonology will find much grist for their mills. And the "Preliminaries" chapter is the best concise general introduction to Hungarian in any language.

Daniel Abondolo is senior lecturer in Hungarian studies, School of Slavonic and East European Studies, University of London.