...and the Earth had one tongue

David Crystal assesses three views of a global linguistic crisis.

Three books with but a single thought: the past, present and future of languages. But three very different treatments, suiting three different reader temperaments.

For a quick, accessible, level-headed read, it would be difficult to better Tore Janson's Speak, an adaptation of a book written in Swedish. It is indeed a "short history". Do not be misled by "301 pages": these are small-format pages. But quantity is not everything, and Janson manages to compress an extraordinarily wide perspective into its scope. Indeed, there could be nothing broader: he begins with Adam and ends 2 million years from now.

His basic idea is very simple: to survey "the role of languages in history" - how languages come and go, and why their paths differ so much. He sees the process as an interaction: "History is affected by languages, and languages are a part of history." Surprising as it may seem, this is a rather innovative idea. Histories of languages were popular a century ago, but have rather fallen out of favour since.

The organisation of the book is logical and familiar: reflections on the origin and evolution of language, the emergence of the Indo-European family and of writing, the roles of Greece and Rome, the rise of Germanic and English, national languages in Europe, and colonial expansion. An account of how new languages (for example pidgins) appear is followed by one on how they "disappear"; Janson does not like the term language death because, he says, it suggests that languages have lives - though this does not stop him talking about language "birth" elsewhere. The book then picks up speed, as he describes trends towards multilingualism and world English; and he ends in an explosion of speculation about the linguistic future of the race, in which all conceivable scenarios are explored, from political change (with English becoming less dominant) to human catastrophe.

This is an honest, sensitive and sensible account: honest, because Janson tells us plainly when we do not or cannot know something; sensitive, for he is acutely aware of the value of language diversity and the crisis of extinction; sensible, because he emphasises a functional perspective for language use ("Languages are no museum pieces. They are tools"). A recurrent theme is that the naming of a language is a crucial step in its emerging identity. "A language without a name does not exist"; "what is crucial for the existence of a language is whether people think it exists."

We might expect such a perspective from a writer in Scandinavia, where mutually intelligible languages are distinguished primarily by their names. One speaks Swedish because one is Swedish.

On the back jacket of Janson's book is the famous quote from Samuel Johnson, "Languages are the pedigree of nations." This is one of the driving forces behind Andrew Dalby's apocalyptic vision, Language in Danger - a book that focuses on just one of the topics in Janson's survey, but a topic that is arguably the most important of all, as it deals with the "disappearance" of the world's languages. Dalby does not like the term language death either, because the people involved do not normally die but switch to another language; he prefers "loss". However, like Janson, he cannot resist the anthropomorphic metaphor, slipping back into it and adding to it with "the after-life of languages".

Language endangerment has attracted three book-length treatments in the past two years, but another one is to be welcomed, for no ecological message can be repeated too often. "A language dies every two weeks" seems now to be standard rhetoric, in focusing the mind, and it is reiterated here on the back cover in prominent red, followed by the question "What are we going to do about it?" This promises a lively exposition, and the preface does indeed introduce a suitably dramatic tone. "Languages are being lost, month by month, year by year. Does it matter? Will it matter if we eventually become a one-language species?"

And then the steam seems to go out of the narrative. We are given a potted history of language, and an account of why languages spread and decline, including an "extended case study" of the Roman Empire and its languages. Dalby does this because it is a period of language where the whole story is known - but it is a lengthy and detailed story, told with copious illustrations from Plutarch, Cicero and others, and many detailed language examples, so that by its close the excitement induced by the preface is a distant memory. There are then further excursuses about the issues involved: the nature of bilingualism, multilingual policy, the link between language and nationhood, the nature of minority languages. All very relevant, and well explained, but the drama has now evaporated.

We eventually get to language loss on page 206. The book will be over in 80 pages. As a result, the treatment of what is lost when we lose a language is disappointingly brief, and even though the book picks up speed again in the last chapter, on the loss of diversity, I feel an opportunity has been missed. For Dalby is indeed making dramatic claims. His master claim is that, because of the link between language and nationhood, the human race is on course to have its languages reduced to about 200 (because there are some 200 nations, and likely to remain so) and that this will happen in the next 200 years. With eight pages to go, he says: "No more bilingualism then".

This is breathtaking stuff, and I just hope readers will plough through the earlier chapters to get to the crunch, but I have my doubts. Dalby does not make it easy for them: he does not go in much for summarising his argument, and it becomes difficult to see the wood for the trees. There are some pretty substantial trees, too, with long quotations from earlier writers and detailed lists of language examples. The book tries to do too much, and as a dramatic appeal for our attention it does not quite work.

But Language in Danger does not live up to the promise of its cover in a second way. Recall that the blurb highlighted language loss and asks: "What are we going to do about it?" Dalby does not answer. The final sentence of the book says, regarding the spread of major national languages, "our great-grandchildren need us to stop doing it and find another way". But we are left in the dark over what such another way might be. Yet there is so much to be said here, not least about the amazing revitalisation projects that are taking place today. Whether these are going to be successful is of course an issue, but that is beside the point. We need to know about them, for it shows what people are doing and how we might help, with time, money or resources. Dalby tells us nothing about all this. But without a forward-looking perspective, any book on language endangerment loses its energy and ultimately its persuasiveness.

The Babel story has surely now become a cliche for writers on language, but all three authors use it. Most writers still get it wrong. Dalby says: "At Babel we became multilingual," but we did not, for there were languages before Babel (as we learn from Genesis 10). I thought it a pity that John McWhorter should have used the cliche for the title of his engaging study The Power of Babel , but do not be put off by that. His approach is no cliché. Ignore also the trivial account of the book that the publishers have put on the jacket, where the author's insightful observations are reduced to "fascinating oddities". Fascinating, yes: but the whole point of his book is that the linguistic observations he makes are not oddities.

The scope of McWhorter's book resembles Janson. It is more natural history than political history, but it has the same kind of logical progression. There are similar observations about language numbers, variety, change, creation and decline. McWhorter opts for the more usual figure of about 6,000 languages in the world (Dalby opts for 5,000), and several times in the early pages we are told that an original single language developed into these 6,000 languages. This is true in a sense, but do not be misled by his phrasing into thinking there have been only 6,000 languages in the history of the race. There have of course been thousands of others.

The Power of Babel is written with stylistic panache, tabloid rather than Times . McWhorter's epilogue on linguistic reconstruction is titled "Extra, extra! The language of Adam and Eve", and the book begins: "I fell in love for the first time when I was four years old." Excuse me? This is a linguistics book? It is indeed, and the author's infectious enthusiasm, freshness of approach and eye for fascinating detail makes language study really come alive.

It will not be to everyone's taste, though. In particular, I can imagine many British readers reacting against its American parochialism. Quite often I had a problem wondering what McWhorter meant, simply because I am not familiar with the US media world that he assumes we all know: for example, he lost me in his discussion of language change when he started talking about Barney Google and others. Who he? A joke about a Richard Rhodes left me cold. Like many Americans, McWhorter takes it for granted that the rest of the world understands what they mean. We do not. A UK publisher's reader should have sorted this out.

In the end, these three voices - Swedish, British and American - are preaching the same gospel: language history is both fascinating and relevant. They are all part of a new mindset in language study, stemming from a developing sense of a global linguistic crisis, in which people look to the past to understand the present and to predict the future. Something important is happening to language, and these books in their different ways help us grasp what it is. Read one or all; but read one at least.

David Crystal is honorary professor of linguistics, University of Wales, Bangor, and the author of Language Death (2000).