Time Depth in Historical Linguistics, edited by COLIN RENFREW, APRIL McMAHON & LARRY TRASK
The McDonald Institute for Archaeological Research. 2000. 681 pages, 2 volume set ISBN 1-902937-06-6. Vol 1 ISBN 1-902937-13-9. Vol 2 ISBN 1-902937-14-7 (£50.00/ $90.00)

Time depth represents one of the most challenging problems in historical linguistics. It is increasingly necessary to make comparisons with archaeological and genetic data to construct a broad historical interpretation of the past. However, unlike archaeology and genetics there is no commonly agreed method of dating either absolutely or relatively. To achieve any systematic comparison with these other types of data, historical linguistics has to achieve a consensus view of linguistic time depth. The McDonald Institute symposium and the subsequent two volume set of papers focuses on establishing some common ground between the competing techniques used to determine the sequence of a languages evolution.

Many of the papers in these volumes show a revival of interest in the much criticised techniques of glottochronology and lexicostatistics. Both sides of the debate were well represented. The proponents reformulated the more traditional approaches to produce a more sophisticated form of these methods. The greater quantification of linguistics may prove a fruitful example of the increase in sophistication being applied to time depth. The advocates of these methods, such as, Baxter, Pagel and Peiros, gave clear examples of why the methods were useful and how they could achieve the desired goal. The more sceptical researchers, for example, Blust, Dolgopolsky and Matisoff, subjected the techniques both new and old to critical examination.

There are also well reasoned critiques of all the methods used to determine time depth, for example, Campbell, Trask and Comrie. In the first of these papers, Campbell outlines the background of the main techniques and then gives a cogent analysis of the advantages and draw backs of each method. Comrie outlines the problems associated with rate at which languages may change. The proposal here is that language change may be accepted more readily in small communities than in large and therefore that social structure may be more important than previously credited. Trask examines the pitfalls associated with relative chronologies. He points out that linguistic events may not be a once and for all event but may form part of continuous processes that may mislead us. Linguistic reality may be far more complicated than anticipated with competing forms appearing in a language. This then means that loss and assimilation may occur simultaneously. Furthermore, sounds may revert through random forces to earlier forms again making relative time depth less easy to determine.

Despite the concentration on glottochronology and lexicostatistics in these volumes there were many other methods discussed including, phonetic comparisons, linguistic palaeontology, and morphosyntactic comparison. Good examples are drawn on to illustrate the various techniques such as Dolgopolsky’s in palaeontology, using numeric systems and combining this with cultural evidence determining the social conditions necessary for such developments, Gamkrelidze and Wiik use similar methods. Heggarty and Matras look at phonetics and grammar respectively to construct possible time depths for language evolution. Nichols uses a variety of linguistic features to argue for estimates of time depth using the linguistic diversity of the Americas. Her view is however, hotly contested by other contributors. The most reasoned arguments against her stance are proposed by Nettle, who points out that Nichols seems to have chosen the least plausible assumptions as the basis for her models.

Although the no consensus was reached on any of the issues discussed in these papers, a few general basic assumptions were agreed by most of the participants. These included the general premise that two populations speaking the same language would if separated for some reason, will from that point begin to diverge in linguistic features. These may include changes in grammar and phonation, and words may be lost or change meaning. All of these and many more features will mean that gradually the two groups will be come over time more linguistically different from each other. Researchers were also in agreement that a common tool-kit of methods was needed for linguistic comparison and classification. This was an encouraging view point of the volumes that rather than reject outright methods that were partially successful in favour of the elusive goal of a single all encompassing method researchers were prepared to refine existing methods and to import methods found to be successful in other related disciplines.

The volumes of papers are not an easy read. They present complex ideas and discussions to members of the same discipline and therefore do not always use language that would be immediately accessible to the non-specialist. However the volumes are of great benefit in bringing together the main views and theories current in historical linguistics. These volumes would be extremely useful for those trying to understand the arguments surrounding time depth in historical linguistics, particularly those such as archaeologists and geneticists seeking integration between their own discipline and the history of early language.

Margaret Clegg
University College London

Review Submitted: February 2004

The views expressed in this review are not necessarily those of the Society or the Reviews Editor.