This book "supplies a vocabulary of English words and idiomatic phrases 'arranged … according to the ideas which they express'. The thesaurus, continually expanded and updated, has always remained in print, but this reissued first edition shows the impressive breadth of Roget's own knowledge and interests."
Gussmann, Edmund (2002) Phonology: Analysis and Theory. Cambridge University Press, xiii+234pp, hardback ISBN 0-521-57409-9, $60.00. Announced at http://linguistlist.org/issues/13/13-159.html
Zoe Toft, SOAS, University of London
OVERVIEW Phonology: Analysis and Theory, by Edmund Gussmann is an introductory textbook which aims to provide an overview of the field of phonology, such that might be covered within one year of teaching, with minimal reference to theory-specific mechanisms. The book is divided into nine chapters, each with a summary and suggestions for further reading, covering both primary sources for data and further theoretical analyses in different frameworks.
Chapter 1, 'Sounds and segments', examines the assumption that the sound stream can be chopped up into independent segments, such as an orthographic or IPA transcription might lead us to believe. In fact, much of the chapter is devoted to undermining this notion of segmental independence, using data from Old and Modern English, Muskerry Irish and Icelandic to show how certain properties of sounds are often closely connected with neighbouring sounds or with the position within the word.
Chapter 2, 'The melody and the skeleton', focuses on the motivation for recognising a timing tier independent from segmental material. Adoption of the skeleton-melody distinction entails the theoretical possibility of (i) skeletal positions without segments attached and also (ii) melody unattached to skeletal positions, and examples of both these types of phonological objects are examined.
Chapter 3, 'Domains and phonological regularities', starts to explore the phonology-morphology interface. The author shows that morpheme boundaries may, but do not necessarily have to coincide with phonological domain boundaries.
Chapters 4, 'The syllable', and 5, 'More on codas', introduce a layer of organization above the skeleton, that of syllabic constituents. It is argued that constituent boundaries cannot be assumed to coincide with word boundaries. For example, each segment in a word initial consonant sequence does not necessarily belong to one and the same onset. The author also argues that word final consonants are attached to onsets, not codas, and that these are followed by empty nuclei.
In chapter 6, 'Some segmental regularities', the knowledge gained in the preceding 5 chapters is drawn together and used to investigate a variety of phonological processes including vowel harmony (Turkish), vowel reduction (English and Russian), spirantization (Icelandic) and final devoicing (German).
Chapter 7, 'Syllable structure and phonological effects: quantity in Icelandic', is devoted to an in-depth exploration of a variety of aspects of Icelandic phonology. New evidence is provided for the claim made in chapters 4 and 5 that word final consonants actually attach to onsets not codas, and the discussion concerning what may or may not form a branching onset or a coda-onset sequences is also further investigated.
Chapter 8, 'Segmental double agents', addresses the phenomenon where one phonetic object is best analysed as two phonological objects. We see, for example, how the Russian labio-dental consonant sometimes acts like an obstruent and sometimes like a sonorant. Rather than trying to struggle with the Jekyll and Hyde nature of a single segment, two phonological objects with one and the same phonetic realization are proposed.
In the final chapter of the book, Chapter 9, 'Words and feet: stress in Munster Irish', a level of phonological organization above the level of syllabic constituents is introduced. Stress placement in Munster Irish is variable and a detailed set of data is worked through step by step, gradually building up a detailed analysis of the phenomenon.
The book concludes with an appendix, the phonetic alphabet of the International Phonetic Association, a bibliography (pp. 227-232) and an index (pp. 233-234).
CRITICAL EVALUATION The book is very well structured, with the first 5 chapters providing an introduction to many key issues, with the exception of subsegmental units on the grounds that to do so would involve a considerable amount of model specific machinery. Chapter 2, on the skeleton, is particularly successful with a good mixture of familiar and new data (including Finnish nuclear simplification, Germanic and Turkish compensatory lengthening) providing strong motivation for this theoretical construct. Chapter 6 draws together all the different aspects of representation so far considered and provides some very interesting case studies, including spirantization of obstruent sequences in Icelandic and Russian vowel reduction. My only reservation about this chapter is the amount of ground it covers. Given that this book is designed to be 'covered within one academic year' (p. ix), I think it would have been possible to extend chapter 6 to include more, different phonological processes such as umlaut, coalescence, dissimilation, and metathesis.
In Chapter 7 students' skills are consolidated by means of an extended investigation into certain aspects of Icelandic. It becomes clear that the student must learn to juggle several phenomena at the same time - an invaluable skill for a student to develop! The analysis itself of Icelandic quantity and syllabification is extremely thought-provoking for those familiar with other analyses of the same set of problem (e.g. Ito 1986, Arnason 1980), although I believe it loses some of its elegance in the details. For example, it is simply stated that [lr], [mj] or [pn] sequences form rhyme-onset sequences: Given that these sequences exhibit rising sonority their proposed syllabification deserves some explanation or at least comment from a cross linguistic perspective.
Chapter 8, on segmental double agents, provides an extremely enjoyable and stimulating excursus into the relationship between phonology and phonetics. It picks up a thread that is woven throughout the book - that as researchers we should always watch our assumptions. First students learn not to assume that phonological and morphological domains coincide. They then learn that constituent boundaries do not necessarily coincide with word boundaries. Finally, one more assumption is tested in this chapter - that just because something sounds the same, a uniform phonological analysis cannot be assumed. Rather, as ever, evidence must be sought and provided in favour of any analysis adopted.
With Chapter 8 I believe the author could have brought the book to a close. Chapter 9, on foot structure, feels as though it is an add-on, almost an afterthought: Perhaps the author felt that an introductory textbook would be incomplete without reference to metrical structure. Unfortunately I think the cursory introduction to foot structure is not successful, and unlike many of the other chapters in this book I believe it could only work really well with considerable additional support, either from lectures or other reading material.
Each chapter comes with a good selection of suggested further reading. I was particularly happy to see references to data sources as well as to analyses in a variety of frameworks. Given the aim of the book that the student should 'try and see what qualifies as a phonological issue and how it may be interpreted'(p.ix.), I think the book could have benefited from the inclusion of problem sets at the end of each chapter. Whilst it is true that many introductory textbooks do not choose to provide problem data sets, I believe this is a lost opportunity, both from the point of view of the student and the teacher.
I think the book would also have benefited from inclusion of more of the standardly used terms in phonology; 'phonemes', 'allophones', 'sonority', 'OCP' (Obligatory Contour Principle), 'MOP' (Maximalization of Onset Principle) are just some of the terms which are never mentioned in this book. Whilst I do believe it is important that students (and established researchers) be able to critically think about competing theoretical approaches, I do not believe that the best way to achieve this is to avoid arming students with any terminology, however historically weighed down with baggage such terminology might be. Rather, students should be equipped with all they need to read outside of the particular approach adopted by the teacher.
To be fair, Edmund Gussmann clearly states his intention 'to avoid model specific machinery and theory internal issues' (p.x) on the grounds that whilst 'models come and go, problems remain'(p.xi). I wholeheartedly applaud this reverence for data above theory, and yet it is my personal feeling that a theory neutral approach is neither possible nor desirable. Whilst Gussmann does indeed succeed in avoiding much model specific theory, the cost, to my mind, is high. Perhaps most seriously he is forced to avoid any reference to subsegmental material; we find no discussion of features, elements or any sort of unit smaller than the 'sound', although in description of Turkish vowel harmony (ch. 6) some vague subsegmental units are introduced. Without a notion of subsegmental material, the concept of 'natural classes' never arises, even though for many phonologists this would be seen as a bedrock for understanding phonological processes.
Attempts at theory neutrality also result in a high number of stipulations when trying to 'explain' phonological phenomena. For example, in chapter 4 on the syllable, an onset cluster (in English) is stipulated as containing an 'obstruent followed by a non-homorganic non- nasal sonorant' (p.74). Whilst this is descriptively adequate, a phonologist normally tries not only to describe but also to explain. Unfortunately, in trying to mention as few theory-dependent constructs as possible (e.g. the sonority sequencing generalization or constituent government), description and stipulation is sometimes all we are left with.
In the preface to this book the author warns that '...no specific theoretical doctrine / approach / theory is explicitly adopted or adhered to, and hence practitioners of any particular model may be disappointed or may want to take issue with the particulars of what follows'. Perhaps I am simply one of these such practitioners. Perhaps, too, I was naive in my hope that this book would offer more of a case for a theoretical position to challenge the current near hegemony that exists in phonological theory.
Despite my misgivings about the (a-)theoretical premise of this book, I found it very clearly written and refreshing to read and shall be using it in future teaching. Many of the data sets discussed are new and interesting, and any approach that encourages students to think twice about their assumptions rather than to uncritically adopt the theoretical fashion of the day has to be a good thing.
REFERENCES Ito, J (1986) Syllable theory in prosodic phonology. Doctoral dissertation. University of Massachusetts. Arnason, K (1980) Quantity in historical phonology: Icelandic and related cases. Cambridge: Cambridge University Press.
ABOUT THE REVIEWER:
Zoe Toft is a PhD student at the School of Oriental and African Studies, University of London, where she is researching syllables without vowels, such as those containing 'syllabic' consonants and also 'empty headed' syllables, from both a phonological and phonetic perspective. For the last three years she has jointly taught introductory phonology courses to both BA and MA students.