Customer Reviews

The Rise and Fall of Languages

5 star:		(1)
4 star:		(3)
3 star:		(1)
2 star:		(1)
1 star:		(0)

 

 

R M Dixon is a well known linguist who specializes in the aboriginal languages of Australia. In this captivating book, Dixon presents his theory of punctuated equilibrium (adopted from the idea of the same name by evolutionary theorists Stephen Jay Gould and Niles Eldredge) to describe how languages change. Dixon challenges linguists to dedicate more time to the study and description of the thousands of languages on the verge of extinction, rather than devote their energies to arcane formalisms. The author is also highly critical of those historical linguists who claim to have found evidence for the "mother of all languages", accusing them of poor methodology. Historical linguistics involves slow and painstaking analysis of language forms, and Dixon is not the first to chastise newcomers for shoddy work. Dixon's book is not overly technical, and is thus suited for both a professional and a lay audience. Anyone interested in learning more about the evolution of language should read Dixon's latest work.

The book is a somewhat short essay by Dixon arguing that a theory of "Punctuated Equilibrium" best explains linguistic relationships. That is, he believes that for most of the time language has existed, languages have been in states of equilibrium with their neighbors, during which time they would gradually converge on an areal prototype (through exchange of lexical items, grammatical concepts and categories--though not many actual grammatical forms--and features of pronunciation). These states of equilibrium would be occasionally punctuated by some "catastrophic" event (e.g., a volcanic eruption, an invasion by some other group into the area, the invention of some new technology, or the development of agriculture). During these short periods of punctuation, languages would rapidly change and split into daughters.

According to Dixon, it is these situations which would create the type of "family tree" relationship among languages which the Comparative Method is designed to reconstruct, and which so beautifully fits with the IE family. However, this was not the normal situation, and after a few millennia, these daughter languages would gradually start to enter into states of equilibrium with their new neighbors, gradually becoming more like the areal prototype of their new location, and eventually obscuring their genetic relationships.

Dixon's decades of experience with Australian languages clearly influenced this idea greatly, and it certainly seems like it could apply well to the situation there. Intuitively, furthermore, the idea seems to me like a plausible one, and I had already assumed something vaguely similar to it before I read the book. However, the book turned out to be quite dissapointing. I felt like Dixon did a poor job of backing up his hypothesis. The book is only abour 150 pages, but even at that length it quickly began to feel very repetitive, with Dixon mostly repeating the theory over and over with new wording. Hardly any examples from actual languages appear, and when they do, it's usually just in the form of a statement by Dixon with no actual forms from the language itself provided. I doubt there are more than four or five occurences of actual native language forms in the entire work. It is a true shame that an idea that seems to me so promising was justified so weakly (and particularly since the author was a man of such intellect and eminence as Bob Dixon).

Not that PE does not have merits. It has been clear for a long time that the Family Tree model of linguistic relationships is weak in several areas. The PE hypothesis could provide for a fuller picture of proto-languages, language splits, and linguistic relationships.

And when Dixon is not arguing for his PE hypothesis, he is at his best. His refutations of the Nostraticists and Proto-Worlders are succinct and direct (I just wish he would back them up with concrete evidence, rather than vague statements along the lines of "they allow a great deal of phonetic leeway in comparisons." Show me proof that they do this! Don't just say it!).

His final chapter (ch. 9), basically a discussion of how the discipline of linguistics has lost its way and become lost from its original purpose (i.e., describing languages) and why field documentation of dying languages is so important, is by far the most persuasive and elegant of the work. I agree with all of the points Dixon makes here (the ridiculousness of the dozens of new fad theories appearing for a few years and then silently fading away, the huge importance of documenting dying and endangered languages before we lose them forever, the obligation of all linguists to undertake this task, etc.), and I find many of his statements (such as "If this work [documenting undocumented languages] is not done soon, it can never be done. Future generations will then look back at the people who call themselves 'linguists' at the close of the twentieth and beginning of the twenty-first century, with bewilderment and disdain", p. 138) to be extremely moving and haunting. How I wish the rest of the book had been written with the same passion and persuasiveness of the final chapter!

This is not the best first book to read in linguistics, but if you already know some general linguistics and historical linguistics, it's good. What it significantly does is offer a critique of the Nostratic School of linguistics (which has been done elsewhere, most notably on a PBS "Nova" show) and offer a new critique, a punctuated equilibrium theory borrowed from the natural sciences. The p-e theory is quite cogent, but Dixon doesn't develop it in too much detail (which is actually his point, that most of the history of human language is irrecoverable). What Dixon does do is offer a research program for practicing linguistics as well as a huge putdown of the theoretical obsessions of most linguists practicing today. This part of the book is actually the most entertaining and courageous; I agree with the author in thinking that too much work is being devoted to Chomskyan (and other purely theoretical) linguistics and not enough to descriptive studies. Whether the money is actually available to record the 1500 or so as-yet unrecorded languages is highly debatable. Dixon himself puts forward a figure of about $300,000 per language. I have a feeling that if any government agency decides to follow through with Dixon's proposals, they'll have to pick and choose which languages to save. Although Dixon puts himself forward as a tough-minded reactionary railing against the vaporisms in contemporary linguistics, he winds up being in the same boat as impractical environementalists to whom any suggestion of, say, which chunk of rainforest to save is complete sacrilege.

In 1972, evolutionary biologists Niles Eldredge and Steven Jay Gould proposed the theory of punctuated equilibrium. In contrast to the more generally accepted theory of gradualism, they proposed that most of the change over the evolutionary history of a species occurs in relatively short bursts, called punctuations, followed by long periods of relatively little change, known as equilibria. Comparative linguistics employs many of the same models as evolutionary biology. Languages, like species, change in their characteristics over time; both languages and species diverge, and both eventually go extinct. As linguists trace the histories of languages and attempt to reconstruct their ancestors, most take a gradualist approach to language change. In "The Rise and Fall of Languages," however, linguist R. M. W. Dixon argues that punctuated equilibrium is a better model for considering language change.

The delineation of the Indo-European family tree and the reconstruction of Proto-Indo-European are arguably the two greatest accomplishments of nineteenth century linguistics. They also provided the model for doing comparative linguistics in the rest of the world. Using the same methods (and underlying assumptions), many other languages have also been grouped into families and their protolanguages reconstructed. One important assumption in this endeavor has been that languages change, from generation to generation, at a constant rate; this enables linguists to create a timeline of change for building family trees. Dixon agrees that languages are always in flux, but he argues that the rate of change is not steady.

On Dixon's view, periods of rapid linguistic change are rare; they are the punctuations of the punctuated-equilibrium model. Punctuations typically occur when speech communities split and expand into new territories. Because of geographic isolation, the speech of each group diverges from the others over generations. Some five to seven thousand years ago, the proto-Indo-Europeans spread out across most of Europe and much of western Asia, probably due to the invention of agriculture and the domestication of the horse. This expansion and the ensuing isolation of each group then led to the development of the various languages of the Indo-European family. Thus, punctuations due to expansion typically lead to a divergence of languages.

This model has worked well for reconstructing a number of other language families. For example, the reconstructed Austronesian family tree fits neatly with the archaeological data showing how the Austronesian people expanded across the islands of the Pacific from an original homeland on Taiwan. The Afro-Asiatic family tree has been delineated using the tools of comparative linguistics as well.

However, there are also many cases in which linguists cannot agree on familial relationships among languages, and Dixon argues that this is because these languages have remained in a state of equilibrium for some time. In situations where a geographical area is inhabited by a number of different speech communities, none of which having any sort of political or technological advantage over the others, a sort of geo-political stasis will occur, and changes within each language will slow down. In addition, each language in the area will influence the others, as borrowings of vocabulary and even syntactic structures take place. Thus, during times of equilibrium, there tends to be a convergence of languages.

In the case of the indigenous languages of Australia, one of Dixon's areas of expertise, equilibrium has been the norm for millennia. Internal changes in each language have been slow, and areal diffusion makes it difficult to distinguish cognates from borrowings in the languages. Thus, the comparative method fails, and it may not be possible to construct family trees or to reconstruct protolanguages.

Dixon criticizes blind faith in the comparative method, and especially attempts to use the comparative method to search for putative relationships farther and farther back into the past. He deems both Greenberg's mass comparison method and Swadesh's lexicostatistics and glottochronology as inappropriate methods for establishing genetic relationships. These methods do have a number of weaknesses, but most importantly for the current discussion is the assumption of a constant rate of linguistic change, which Dixon disputes.

Dixon is especially vitriolic in his attack on the proponents of Nostratic, a hypothesized super-family incorporating many of the established language families. He argues that the criteria used are so lax that any two languages compared by this method would appear related. Furthermore, he chastises "a credulous media ... [that] have embraced the most crackpot ideas about language relationships" (p. 43). Alas, this observation is sadly true of all scientific disciplines, not just linguistics. (And some biologists, such as Richard Dawkins, have made the same comment about the media and punctuated-equilibrium.)

Next, Dixon turns his attention to the issue of language death. In the past, natural disasters, disease, genocide and conquest have all been factors in the demise of language. In recent times, however, a new factor has arisen: the development of mass communication. As the languages of political power groups gain prestige, speakers of non-prestige languages will tend to learn them. Broadcast media expedite this process on the regional level, keeping mega-languages like English, Chinese, Arabic and Spanish, each of which are spoken by hundreds of millions of people across wide areas, from splitting into daughter languages. Instant global communication pushes this process even further, leading to the rise of English as a global language, a truly unprecedented event.

Dixon views the period of linguistic history since the beginning of European colonial expansion as one of punctuation. It is true that linguistic changes have been rapid during this time period. Unlike other periods of punctuation, however, in the current era we are witnessing a convergence of languages, not a divergence. First, the number of different languages is rapidly dwindling. And second, those languages that are thriving are nevertheless converging on English through the borrowing of vocabulary and syntactic structures. Conditions are so unlike anything in the past that historical models of linguistic change may no longer be relevant.

In "The Origin of Species," Charles Darwin emphasized a constant-rate gradual change of species because he wanted to place his theory of evolution by natural selection in direct opposition to the received view of that time that species arose suddenly as special creations. However, as Richard Dawkins points out in "The Blind Watchmaker," gradualism does not necessarily imply change at a constant rate. In "The Rise and Fall of Languages," Dixon makes an important contribution to comparative linguistics by making exactly the same point with regards to linguistic change. It is in the nature of languages to change from generation to generation, but there are a number of factors that can influence just how quickly they change.

I'm interested in linguistics, and I took a few classes in it in college twenty years ago. This was a very interesting book for someone like me, since it presents new ideas in diachronic linguistics, without assuming more background than I had. It was great to be taken out to the edge of the field, where science is being done, without having to battle through a mass of technicalities. Plus, it's very well-written and fun to read.

Dixon has a clear writing style and is easy to read. The content of the book is definitely targeted at those with at least a passing knowledge of linguistic theory. There are lots of interesting little tidbits - references to fieldwork the author has undertaken himself, as well as references to work that's going on around the world. However, the work tries to cover its title "The Rise and Fall of Languages" from a few different directions, with quite a lot of repetition. The author even admits at one stage that this work is really only a prelude to a forthcoming work of his own.

The main topic of is how languages evolve, and how this limits how far back in the past linguistics will ever be able to peer. This is the most interesting topic that the work covers, and should have really been its only focus. To be fair, it is by far the major one, but I felt like I needed to see more examples, or a model of language evolution, or something to back it up. To me, his theory of "punctuated equilibrium of language evolution" - which simply put means that sometimes languages split and develop separately, and sometimes they come together and borrow from each other, the former being the major model used in linguistics and to which some linguists try to fit every language in the world. Basically, the author makes the point that some information is lost/blurred over significant periods of time and it just can't be reconstructed. Although I feel I understand what the author is trying to get at, the argumentation seems to largely consist of somewhat common sense assertions repeated several times. However, there is enough information in there for this to be what the author intended - a launching point for the field to expand into and study in more detail.

The author's discussion of the origin of language itself left me somewhat cold - this section just didn't really "do it" for me. There is no real information here other than "perhaps a few words were added at a time, with a few grammatical features". Actually, if I recall correctly, he may have contradicted himself on this point by saying the exact opposite a few pages earlier by saying that it would all have needed to have happened very quickly. In a way I think his assumption is that language belongs to thoroughly modern humans and no modern humans have a truly primitive form of language. If you don't start with this assumption though, the picture could look quite different. It is only a minor topic in the book. For a thorough and, I feel, more illuminating look at this topic, I recommend Deacon's "The Symbolic Species: The co-evolution of Language and the Brain".

I do hope to see more on this theory from Dixon himself one day.

Dixon also passionately makes the case for recording the languages we have left on earth. I totally agree with him that if the world doesn't thoroughly document the linguistic and cultural situation of all that we have now, future generations will curse us. How many linguists, archaeologists or indeed many other kinds of scientists, wish now that early explorers had done much more recording and documenting, and much less shooting? We can't have that time back, but we can at least bequeath a different legacy to future generations.

Dixon is quite grim on the prospects of completely stopping language death. I'm more hopeful - I believe that government supported full bilingualism can save languages as well as integrating the youth from disadvantaged communities into the broader economic community, but Dixon has a lot more first-hand experience so he is probably right.

This book is only a light read to a thought provoking topic. It serves the linguistics community as a guide for new players to steer clear of those who think that statistics can recover lost information in any accurate fashion, and that they can reconstruct any stage of past language from current ones. One day we might hear more on this subject from Dixon and others.