Phonetic Data Analysis: An Introduction to Fieldwork and Instrumental Techniques (review)

Experimental phonetics has found increasing relevance in recent years, particularly via the importance of laboratory phonology. Phonetic fieldwork, however, is still very much a minority activity within linguistics, though growing awareness of the need to fully document languages, especially given the realization that a sizable majority of the world's languages are endangered, has begun to focus some attention on this linguistic activity. Peter Ladefoged is one phonetician who has engaged in fieldwork, and has produced detailed phonetic descriptions of a great many languages. And, as is well known, he is no stranger to experimental phonetics, and does a fine job of writing textbooks, as the success of his Course in phonetics and Elements of acoustic phonetics attests. Indeed, there is probably no single scholar today who is as well qualified to tackle the task of producing a book on the subject of "Phonetic data analysis: An introduction to fieldwork and instrumental techniques". There can be no question that a book of this nature is needed, as there is no comparable book available; and given Ladefoged's stature both as a phonetician and as an author, it comes as no surprise that it is a valuable contribution. Some readers will be familiar with some of the content, as it has predecessors in articles Ladefoged has published on phonetic fieldwork, though obviously the present volume is much more substantial, and no doubt it is in part a response to suggestions from interested scholars that he produce a textbook version.

The book comprises eight chapters, the first of which is "Recording the sounds of a language". This chapter covers such topics as determining which sounds in a language are, or should be, of interest to the phonetician; how to create sets of speech materials to use in an investigation; how to find appropriate speakers; how to go about actually recording people, particularly in a field setting; and what kind of recording equipment to use. Also discussed are three pieces of very pertinent advice: to check recordings as soon as possible once they are done (i.e., while speakers are still available); to make transcriptions of recording as soon as possible; and to keep detailed fieldnotes of all work done.

There are two aspects of this chapter I might take issue with. First, Ladefoged makes the point at the outset that "[w]ithout knowing the phonology of a language you cannot describe the phonetics. You need to know what it is you have to describe" (p. 1). While there can be no quarrel with such a statement in principle, Ladefoged goes on to say that usually a researcher will be able to find some previous work that will be of use, a dictionary, a descriptive account of the phonology, etc. Given that a substantial majority of the world's languages remain undescribed, this claim is somewhat optimistic—and certainly in my own experience not the case. Fieldworking phoneticians must also be prepared to work in the absence of such descriptive materials, that is, to do at least a basic analysis themselves.

A second aspect of this chapter I would modify is the discussion of recording equipment. Ladefoged favours recording on DAT (digital audio tape) or direct to computer, and of course I have no quarrel with this, my own preference being DAT. What is missing from the discussion, though, is consideration of other formats, such as minidisk. We frequently find questions in forums such as the Linguist List as to which type of recorder to use in a fieldwork situation. While minidisk recording may indeed be suitable for many purposes, it should be made clear in a book such as this that the compression it uses renders a minidisk recording unsuitable for acoustic analysis. In preparing to do acoustic analyses, we should obviously be aiming for the highest quality possible of the speech signal, but I would push this a step further and suggest that even fieldworkers not intending to do acoustic analyses of...