Different cake, similar taste

This book aims to account for the genesis of the "new" English colonial varieties of the Southern hemisphere: New Zealand English, Australian English, South African English and the Englishes of the Falkland Islands and of Tristan da Cunha. Phonology rather than grammar (and certainly not lexis) is the prime focus of the study, and New Zealand English the main object of attention.

Peter Trudgill's main source of data is provided by a selection of recordings made in 1946-48 for the National Broadcasting Corporation of reminiscences by New Zealanders who had been the children of the first settlers and who were born between 1850 and 1900. The recordings as a whole, and their analysis, are known as the Onze project ("Origins of New Zealand English"), led by Elizabeth Gordon in Christchurch. Since she and Trudgill have also this year published a book together on this project, the work discussed here could in some sense be seen as a "companion" volume.

Both books are likely to be of great interest to linguists and sociolinguists engaged in the study of Englishes around the world, while Trudgill's book in particular will be of interest also to historians of the English language of the 19th and early 20th centuries.

The key to both the hypotheses and the major imagery with which they are clothed lies in the subtitle of the book: the "inevitability of colonial Englishes". Trudgill's theme and motifs are likely to raise a few academic eyebrows. His argument is deterministic; he is confident that the dialects (accents) are "predictable", although this must surely be with hindsight.

The major motif is of "mixing": he talks about children in New Zealand around 1855 "selecting at will from a kind of supermarket" of sounds, and, more significantly refers to mixing the ingredients of a cake. Like cakes, with "the same ingredients in roughly the same proportions in roughly similar conditions for roughly the same length of time", you will get similar Englishes. In this kind of cake, however, it is the ingredients themselves presumably doing the cooking, since they, as speakers, are also the agents of change. By this essentially mechanistic argument, however, there is no need for the sociolinguist Trudgill to bring in any of the traditional arguments of social prestige, stigma or identity, and so on, or even salience. Judgements of what exactly are the right "proportions", however, for any one sound at any one time in any one place remain subjective; and in the end we may conclude, to paraphrase another linguist cited, that Trudgill's theory "works unless it doesn't".

Katie Wales is professor of modern English language, Leeds University.