There's a reference in "Vanishing Voices" to a magazine ad that promises instruction in "most of the world's languages" -- a total of 76. That might seem pretty impressive, except that even the lowest estimates put the number of languages in the world at roughly 5,000. That doesn't include dialects or regional variations; it represents the number of bona fide languages spoken in the world, each as complex and distinct as English, Mandarin Chinese and Hindi.
If that seems hard to imagine, it's because the great majority are local tongues such as Rotokas, Sim'algax and Kurux, used by only a handful of people. (There are, for example, fewer than 500 native speakers of Kurux.) These languages are, "Vanishing Voices" explains, disappearing from the world at an astounding rate -- as many as half might become extinct in the next century.
A worldwide trend toward language extinction, according to the authors, has been going on ever since Europeans conquered the Americas and began to spread out across Africa, the Pacific and Australia. The trend has been accelerating in recent decades thanks to the global economic juggernaut -- through the leveling of regional distinctions and the ongoing displacement of indigenous populations, and (not least) through the rise of English as the lingua franca of business and commerce. The field of linguistics believes that the shrinking number of languages is a bad thing, but aside from the fact that it means fewer languages to study, and thus less for linguists to do, there is no consensus as to why it's bad -- or what should be done about it.
But Daniel Nettle and Suzanne Romaine are not linguists; they're from somewhat more sanguinary, more activist fields (anthropology and English, respectively). And they've taken up language extinction as their cause, claiming that indigenous languages have vital cultural knowledge encoded within them that's being lost to the world forever.
"Vanishing Voices" is well-written and engaging, and it makes you feel that something unique and irreplaceable is being lost whenever a language dwindles into oblivion. The book has photos of the last speakers of several languages, including Briton Ned Maddrell, who took the ancient language Manx to his grave in 1974, and Red Thundercloud, the last speaker of Catawba Sioux, who died in 1996.
As a book on the science of language, however, "Vanishing Voices" is a bit wacky. The point the authors try the hardest to make is that indigenous languages are somehow associated with biodiversity, and that their extinction is a symptom of the global ecological crisis. Which might be true if you squint at it the right way: Languages, along with flora, fauna and indigenous peoples, are dying out -- and it's no great secret that geocapitalism is the culprit. But is there a genuine, necessary connection between biodiversity and linguistic diversity? The authors never establish one, but they repeat the idea a number of times, coyly and at odd moments, as though they weren't entirely convinced of it themselves.
Even if you believe, as Noam Chomsky does, that the ability to learn and use language is innate, it's quite a different thing to say that languages themselves can be "ecological." It's like saying that since sex is innate and natural, so are strip joints and S/M clubs: The basic impulses behind them might be present in all of us, but the forms in which they're expressed depend on all sorts of complex cultural forces. Yet much of the book is tied together by this slender premise. Eventually, Nettle and Romaine's discussion of language tapers off altogether, into a narrative of indigenous peoples' struggle against the forces of globalization, which is important and makes for absorbing reading but isn't what the book purports to be about.
But Nettle and Romaine make a pretty good argument that it's easier in some languages than others to conceive of certain useful relationships among things, thanks to classifier systems that organize words into categories (like gender in French and German or the Japanese system of using different words to count differently shaped objects). In the dying Australian language Dyirbal, for instance, there are four categories for nouns, which reveal subtle shared similarities among the words, as well as cultural judgments about the objects to which they refer. "If some members of a set differ in some important way from the others," the authors note, "they are put into another group. Thus, while fish belong to Class I bayi words, the stone fish and gar fish, which are harmful and therefore potentially dangerous, are in the balan class." There is thus no mistaking, for a Dyirbal speaker, that the stonefish is dangerous. "The rationale for the categorization," the authors continue, "tells us something about how Dyirbal people conceive of their social world and interact with it."
Other examples follow, including that of a complex calendar system used by Balinese farmers to synchronize irrigation. But here's the hook: The claim that language operates in this fashion goes against the grain of mainstream linguistics, which holds as an article of faith that all languages are basically equivalent in terms of conveying meaning -- that none is better or more efficient than any other. Linguists today have to reiterate this point a lot: Early language researchers once tramped across strange terrain and called the local tongues barbarous and inferior.
Nettle and Romaine, instead, make a good showing at demonstrating that there are questions of better and worse regarding language: A language, or a language group, can and often will be superior to all others in its own natural and social environment. But there's still a long way to go to prove that the noun classes in Dyirbal, for example, make any difference in its speakers' consciousness -- in the way they think about fish or anything else. And if linguistics is right that all languages are equivalent, then even after reading "Vanishing Voices" you're still left with a difficult, even untenable question: If most of the world's languages are dying, so what?