An insightful route to deep thought 1

Why does language change? That question always interests students. But how does language change interact with the grammar? And how should we integrate what we know about linguistic change with the development of linguistic theory? If you ask students these questions, you're likely to face blank stares.

This book will be a boon to educators who teach courses that go beyond a basic introduction to historical linguistics and who seek to link issues in linguistic change to questions in theoretical phonology or syntax. It seeks to answer these latter questions, and it is engaging and insightful as it does so.

As the author acknowledges, the book would not work for an introductory historical linguistics course. There are no datasets to practise on, and some linguistic sophistication is assumed. Rather, it is a conceptual work, focusing on what historical linguists say they do, what they actually do, and what Mark Hale thinks they should do. It aims to make readers think hard about the foundations of historical linguistics, so considerable space is devoted to defining what language is and how we should relate texts from previous periods to the linguistic systems that we reconstruct from them.

The volume has five parts. Parts two and three are the core of the book, dealing with how (Hale argues) phonological and syntactic change should be understood. They discuss previous approaches and then exemplify Hale's approach by investigating relevant topics. Part one is foundational, discussing the nature of language, historical texts and linguistic descent. Part four deals with issues in linguistic reconstruction; part five summarises and concludes.

Hale defends the standard generative understanding of "language" as the internalised grammar in speakers' minds. This, coupled with the common assumption that language acquisition ceases relatively early, means that language change does not literally exist. "Change" is just the acquisition by one generation of a slightly different language from that of their parents. Hale shows that, while many historical linguists might overtly agree with these positions, their practice can be inconsistent with them. He argues, for example, that a phonological change is not the same as a synchronic phonological process. This position can be rejected only if Hale's fundamental assumptions can be shown to be moot (and some can be, I believe). The book challenges readers to accept Hale's arguments, or to be as coherent and as clear in arguing against them as Hale is in arguing in their favour.

Who is it for? Anyone with an interest in linguistic change.

Presentation: Clear and neat.

Would you recommend it? This book is firmly on my list of the few crucial texts for historical linguists, alongside Roger Lass's Historical Linguistics and Language Change and William Labov's Principles of Linguistic Change (neither of which would work as a straightforward course textbook, either). Students should be encouraged to read this volume critically to broaden their horizons and deepen their thoughts.

Patrick Honeybone, lecturer in linguistics and English language at Edinburgh University.