Customer Reviews

The Origin of Language: Tracing the Evolution of the Mother Tongue

5 star:		(10)
4 star:		(13)
3 star:		(7)
2 star:		(5)
1 star:		(3)

 

 

When the Tower of Babel was being constructed, so the story goes, God was so incensed at the presumption of humans that he condemned them to speak a multitude of tongues. Ever since then, we've often needed translators to speak to each other, and imperfectly at that. Many are the battles fought because of misunderstandings caused by language differences.

Merritt Ruhlen has a different take on the language schism. In his book, called *The Origin of Language*, appropriately enough, he explains the theory that all of today's languages had a common origin, many thousands of years ago, and that linguistic drift accounts for all the differences we see today.

The way that he arrives at this point is fascinating. He allows the reader to play along in the linguistics game, providing sample words that work nicely to group languages together in ever larger categories, until they all tie together in one world glotknot. It's all so obvious that you can't believe that anyone could think differently.

Of course people think differently. In fact, a lot of linguists (Eric Hamp at the University of Chicago, for one) think differently. Many of them think that Ruhlen and his sometime mentor, Joseph Greenberg, are kind of nuts. For one thing, picking ten words at a time to group languages together is a risky endeavor. Even if Ruhlen believes he picked the ten words at random, you can't get around the fact that Ruhlen *knows* what conclusion he wants to reach, and that could taint the whole process. Anecdotal evidence is a notoriously bad way to come up with general theories.

Furthermore, Ruhlen doesn't really go into the quantitative business of assessing how great an effect phonetic drift has in muddying up the genetic relationships between languages, and when he does do it, the mathematics are misleading or simply wrong. Richard Feynman made a big point of telling his students not to use an observation that suggested a theory as confirming evidence of that same theory, a lesson Ruhlen seems to have missed.

All of which Ruhlen probably doesn't worry too much about. He and Greenberg are more concerned about getting the big picture together first, and addressing the details later. Nothing wrong with that, but in his effort to gain converts to the "lumpers" faction (as opposed to the "splitters"), he has an alarming tendency to denigrate the work of others. Like the guy who runs his coworkers down behind their backs, he tends paradoxically to lose a lot of support.

It's too bad, because linguists often do seem polarized around this question of origin, and the field really could use a solid, balanced book that looks at what new work needs to be done without ignoring or downplaying work that's already been done. This book isn't it, though. The casual reader will learn a lot about the way that lumpers work, because it's more exciting; more inquisitive readers will hanker for something with more study and less politics.

This is not a book about comparative linguistics. Instead, it is a book devoted to Ruhlen's personal fantasies.

Comparative linguistics, like all linguistics, and indeed like all serious scholarly work, is done by applying rigorous and scrupulous methods to carefully obtained data. The right way of doing comparative linguistics was worked out only at the end of the 18th century, and it has been developed and refined ever since.

Before that time, people had no idea how to compare languages, and they worked wholly in the dark. Their favored "method" was nothing more than the assembly of miscellaneous resemblances among miscellaneous languages, in the hope that this might shed light on language origins. But it didn't, and it doesn't: miscellaneous resemblances are meaningless and worthless, as has been amply demonstrated countless times. See any decent textbook of historical linguistics.

But this Dark Age procedure is exactly what Ruhlen wants his readers to accept, believe in, and follow. Ruhlen shows no understanding of the numerous and serious obstacles to the comparison of languages, and no understanding of the formidable pitfalls that must be avoided if useful work is to be done.

In place of rigor, Ruhlen offers us only lists of miscellaneous resemblances, which, like the forlorn scholars of the past, he wants us to take seriously, and to use as the sole basis for spectacular conclusions.

Worse, Ruhlen wants his eager readers to believe that they too can do serious work in linguistics: "Don't believe the blinkered professionals when they tell you that good work requires years of training and experience, or that it requires a comprehensive knowledge of the languages you want to compare. Just trust me when I tell you that any idiot with a bilingual dictionary can do real linguistics, better even than the professionals."

There is more, much more. The very first duty of a scholar is to get the data right, but Ruhlen can't even do that. For example, of the 13 Basque items presented on page 65 (as "language B"), four are wrong, and two more are not even native Basque words, but are words borrowed from Latin or Spanish. And there are also some profound problems concerning the origins and earlier forms of several of the others, problems which Ruhlen ignores because he doesn't even know about them: he's just extracted his forms incomprehendingly from a bilingual dictionary. But Ruhlen doesn't care about such humdrum tasks as getting the facts right: he merely wants to persuade readers of the spectacular success of his primitive and wrongheaded approach. So what if the data are wrong: it's the Big Picture that's important, right? Professional linguists, who are all too aware of the enormous difficulty of establishing links between any languages at all, have no time for this sort of nonsense.

This shabby book is made even shabbier by Ruhlen's practice of making nasty remarks about those linguists who have quite properly criticized his work -- which means just about every linguist who has ever commented on it at all. He even goes so far as to say nasty things about long-dead linguists of the past, like Meillet (on page 79), apparently on the ground that they too would have condemned his work if they had lived to see it.

You will learn nothing about doing comparative linguistics by reading this dreadful book. You will learn only how to join the massed ranks of the linguistic cranks. And we already have more than enough of those. There are thousands and thousands of cranks churning out useless and pathetic "comparisons" like Ruhlen's every year. Ruhlen is more prominent than most, but he is no better.

R. L. Trask
Professor of Linguistics

(mail addresses withheld at Amazon's request)

Ruhlen's approach to historical linguistics is an extreme manifestation of the use of mass comparison across language families to arrive at family trees which could not be demonstrated using the more clearly reliable traditional comparative methods - because the posited time-depths are too great. Using those methods, we can reconstruct forms and families only a little earlier than the earliest written records. Greenberg and the Nostraticists, reviving some of the 'glottochronological' notions of the 1950s, represent a more moderate version of Ruhlen's view. Some of their proposals have been generally accepted (eg, Greenberg's on Africa). But Ringe and others have argued persuasively (with statistics) that in general their methods are unreliable, because - given enough time - chance similarities are very likely and are indistinguishable from genuine cognates when traditional methods are unavailable. This would apply even more strongly to Ruhlen. He acknowledges that his position is very controversial; but readers should be aware that his book has not succeeded in persuading more than a few linguists that he is right in thinking that we can date or identify 'Proto-World' (the universal ancestor language), still less reconstruct any of it. It is not even regarded as certain (though it is perhaps probable) that there WAS just one Proto-World. Ruhlen's position is not wholly untenable, but beware of regarding it as the best currently available; the consensus is that it is dubious.

Even if Ruhlen should have a case, this would NOT support those who posit links between apparently unrelated languages on the basis of a few unsystematic instances of similar words for similar concepts, eg, very roughly similar words for 'god', 'father' etc around the world, wrongly seen as showing that all languages derive from Sanskrit, Latvian or Hungarian, or that two isolated languages such as Zuni and Japanese are in fact linked (all these examples are from actual proposals).

BTW: most linguistics programs will happily accept a student who knows as many as three languages. Even monoglots can study the subject with profit, learning about more languages as they go.

I am by no means an expert on languages or the history of languages. I picked up this book on a whim awhile back. After reading the reviews on here, I can understand why the author warned the reader that his view is controversial.

Basically Ruhlen gives you a list of about 10-15 words from 10 or so different languages. Basic words like "hand" or "head" or "water," words that would've been around for awhile. He lets the reader group the words accordingly while he would give a few pointers on how words change over time and what one sound would tend to change to. He starts with the Indo-European languages and proceeds to do the same with Native American, African, and Asian languages as well. Eventually we find out that all the languages have come from a single mother tongue. We also find out that his theory coinsides considerably with current genetic theories on the spread of humans. It was very interesting and fun to do, and Ruhlen never talks down to the reader.

That said, there were some problems. He kept referring to "Indo-Europeanists" who, according to him, dismiss his theories in-hand without even looking at them. I find this hard to believe. Sometimes it seemed to me that he all but called them racists. Then again, after reading some of the quotes from those who disagree with his theory, it all seems a bit petty, like someone who disagreed with a person's theory personally attacked that person.

All in all, it is a good book to read to get a different take on the story of language. If you're really interested after this, you probably need to get a book that has the "traditional" viewpoint of linguists.

This book, not be confused with the same author's "On the origin of languages", which is a far more technical work on a similar theme directed towards professionals, is intended for the general reader, and is easy to read. Far too easy, some may think, as it is a book that enrages many (nearly all, in fact) professional linguists, who consider his methods to be worthless and his conclusions nonsense.

Merritt Ruhlen not only believes that all human languages have a common origin, a reasonable enough hypothesis as long as one takes it no further, but he also goes much further, arguing that the common origin can be demonstrated and that it is obvious enough to be recognized in similarities that exist in modern languages. He presents various examples, but one is enough to illustrate the basic point: he reports that the words for water, in five of twelve language families that are normally considered to have no discernable relationship with one another, are (omitting diacritics and some other complications) "ka", "akwa", "okhwa", "akwa" and, again, "akwa". By restricting himself to words like "water", "two", "finger" etc., that have been part of human experience since the earliest times, and are not very likely to be borrowed from other languages, Ruhlen tries to avoid the danger that the similarities result from borrowing. It would hardly do to use words like "telephone" that are similar in many languages simply because the thing itself is something that has been transferred from one culture to another in recent times.

Ruhlen's argument is that the similarity of these words across language families is not a coincidence; his opponents' argument is that he has selected his data to the point that he has just gone through long lists of possibilities in dictionaries until he finds examples that suit his purpose. In the Indo-European family, for example, English "water" doesn't look very promising for including in the above list, but Italian "acqua" looks just fine. The argument then resolves into the question of whether there is a big enough range of possibilities in language families that are unfamiliar to the ordinary reader to allow this sort of selection to seem successful even though there is no genuine relationship to be found. This is a question that the ordinary reader cannot answer, and so what I should like to see would be a book by one of Ruhlen's opponents who sets out the opposite case in an accessible manner. I know of no such book: the late and much regretted Larry Trask (whose own review appears elsewhere in this collection) could have written one, but that is no longer possible.

Comparative linguists can be grouped into two mutually antagonistic camps: the "splitters" who maintain distinction between language families unless rigorous criterion are met and "lumpers" who readily accept the grouping of disparate languages into a single language family on the basis of having somewhat minimal commonality. Ruhlen's controversial work is an engaging account of how "lumpers" have developed a model which distills all extant languages into twelve language families that are direct descendants of a postulated mother tongue.

Roughly two-thirds of the book involves the classification of languages into families and into families of families. The author does this by presenting tables containing selected words in various languages and asking the reader to classify the languages based on similarity. Ruhlen continues on this sort of path until arriving at the final direct descendants of the mother tongue. Throughout this journey Ruhlen guides the reader to the inner workings of language classification. The final third of the book is an attempt at justification of his classification method citing both genetics and archeology as sources of corroborative evidence while also citing sources of disagreement with other linguists.

Ruhlen's work is an insightful introduction to historical linguistics for the lay reader. Unfortunately, the book at times seemed to have less to do with historical linguistics and more with serving as a platform by which to blast opposing schools of thought who believe Ruhlen's judgement as to what constitutes a cognate is more subjective than is desirable. These frequent attacks were a significant detraction from what was otherwise an enjoyable read. Accepting the logic behind Ruhlen's language taxonomy at times required a significant leap of faith and he was unconvincing in dealing with the time depth issues pertaining to linguistic reconstruction. However ethnographic surveys of blood proteins and other genetic data indicate that the movement of peoples from an ancestral homeland in Africa can be reconstructed and this reconstruction corresponds to Ruhlen's linguistic family tree. This remarkable degree of correspondence is compelling evidence that Ruhlen is on the right track.

As an anthropologist teaching introductory courses in a community college, I give several general lectures on linguistics each semester. My own background is in the social aspects of language use, although I've always been fascinated by historical linguistics. I've been long aware of the ongoing debates about the origin of human language and usually mention the controversy about Nostratic ("our language"--the hypothetical original human language) in my classes. I picked this book up to give me a little more background. Unfortunately, I found it a very disappointing read.

My first problem was with Ruhlen's method, or at least with the examples of it that he gives. Many of the chapters present the reader with a series of words in different languages and encourage the reader to work out the historical relationships between them. This "do it yourself" approach is supposed to demonstrate the legitimacy of Ruhlen's reconstruction of linguistic history. An intelligent reader, according to Ruhlen, will come up with the same set of common sense relationships as a linguist. Well, after the first couple of chapters, I found his purported relationships dicier and dicier. I'll leave it to the professional linguists who have reviewed the book to comment on the accuracy or Ruhlen's reconstructions, but frankly, this PhD couldn't follow them.

My major problem with the book, however, has to do with the entire idea of Nostratic as being the sole mother tongue of our species. It's hard to talk about the origin of human language without talking about human origins in general, and I don't think that Ruhlen adequately does this. By using Cavalli-Sforza's genetic data to support his arguments, he allies himself implictly with the Repalcement theory of modern human origins. This theory, sometimes called the "Out of Africa" or "Eve" theory) states that modern homo sapiens originated in Africa around 100,00 years ago, migrated out of Africa and into the rest of the Old World, where they replaced existing populations of Neanderthals and Homo erectus (who may have lasted far longer in Asia than in Europe). The opposing school of thought, the Multiregional Theory, suggests that existing regional groups evolved into modern homo sapiens in parallel, interbreeding with each other and with any newcomers from Africa. The idea of Nostratic fits far better with the Replacement theory than with the Multiregional. The Replacement theory is currently the favored one, but there are still enough questions and new discoveries to keep the debate going. For instance, the question of how much, if any, interbreeding (and communicating) took place between existing populations and the newcomers is the subject of considerable debate.

But Ruhlen leaves this question and other important issues untouched. When precisely did our ancestors begin to communicate via linguistic means? Certainly full-blown language capabilities couldn't have developed until relatively late (some argue that even Neanderthals did not have fully human speech). But, given the success of experiments teaching sign languages to Koko, Kanzi, and other great apes, some minimal linguistic ability was already present before our line split from theirs approx. 5 million years ago. Some kind of vocal linguistic behavior would have developed before full-fledged languages. Most likely this would have appeared independently among different hominid populations as an outgrowth of existing gesture and call systems. The fossil record does hint at an ongoing expansion of the speech area of the brain over several million years. Homo sapiens and its predecessor populations were widespread geographically. To expect them all to develop a single original tongue from which all other languages evolved just makes no sense to me. Even assuming an extreme Replacement model of human origins, it seems unlikely scenario. The idea of multiple independent origins of human languages, some of which survived and ultimately grew into the language families we know today (and possibly some that we don't), makes a lot more sense to me than Ruhlen's One Original Language argument.

Leaving aside the merits (no pun intended) of the author's argument for the moment, my final criticism has to do with his style. His constant complaints about how mainstream historical linguists have "done him (and his mentor Joseph Greenberg) wrong" grow stale very quickly. Ruhlen's comments take on a whiny air and get in the way of his arguments. It's true that some of Greenberg's controversial theories about American Indian languages were eventually proved to be correct, but that doesn't mean that his theories on Nostratic will prove to be equally correct, or that the methods Greeberg and Ruhlen use actually do reveal deep historical connections. Ruhlen has topnotch credentials, and rather than just condemning those scholars who disagree with him, I would have preferred a cooler headed approach.

I wish I could have given this book a better review. As I tell my students, linguistics is a fascinating field. It takes a special kind of detail-oriented person to do linguistic research. But in the end, the conclusions linguists reach often have breathtaking implications. Unfortunately, I just can't buy the arguments in Ruhlen's book.

Ruhlen is a Greenbergite and a Proto-Worldist. This fact alone is a guarantee of controversy in linguistic circles. If you are really interested in the origin of language, by all means read this interesting book, but try to balance it with a more traditional view.

That said, this is the most comprehensive statement of the Proto-Worldist/Nostratic viewpoint that I've seen. One can criticize it on various grounds (note R. Trask's objections elsewhere on this page). While the occasional incorrect citation can be forgiven--after all, Ruhlen is trying to establish broad affinities here--I find more to criticize in the fact that Ruhlen's super-families are reconstructions at second or third hand. For instance, to get to his precious reconstructed Proto-World roots, he has to go through reconstructed Indo-European and reconstructed Eurasiatic. To take that many steps without hard facts to back them all the way is asking for trouble.

Ruhlen is on much safer ground when he supports the theory that the Indo-Europeans originated in Anatolia (Chapter 8), and he is up to date regarding the findings of genetics and archaeology. This is definitely the work of a "lumper," full of interesting ideas, some supported weakly, others more strongly. Worth reading, even if it's not mainstream scholarship.

The jacket of Mettitt Ruhlen?s book THE ORIGIN OF LANGUAGE suggests that he is one of the world?s foremost linguists, however, the contents of his book suggest he is a leader of one faction of linguists, and by his own words, not in the majority.

Apparently, linguists form two camps (terminology from Colin Renfrew) ?lumpers? (those who subscribe to the Nostratic (Russian)-Petersen-Greenberg-Ruhlen perspective that all languages probably evolved from a common source which they can identify, hence the search for the ?origin?of language), and the ?splitters? (those who believe one simply cannot tell if there was a mother tongue because if it existed, the original language is so far back in time as to be unknowable.

Ruhlen makes a fairly compelling case for his argument. He takes the reader through a series of exercises designed to illustrate how with 10-12 words per language one can identify and classify (taxonomy) commonalities that reflect family groupings, and that using the root words for each classification one can then compare and aggregate families into supra-families or groupings. Most interesting to me are the links between languages of the Eskimo-Aleuts and other Indo-European groups.

I was with Ruhlen until he began to discuss Sforza-Cavalli?s work with genetics, which he suggests supports his theory (it does partially). However, in his most recent book (2001) Sforza-Cavalli suggests the Indo-European homeland was probably in the steppes of Russia (Ukraine), not Anatolia, thus agreeing with the traditionalists (Child-Gimbutas), not Renfrew with whom Ruhlen associates himself in this book. Sforza-Cavalli?s synthesis suggests timing is the core issue regarding the movement of the nomadic Kurgan culture into Europe and the expansion of the peoples of agricultural Anatolia into Europe.

This book, written for lay readers, ventures two arguments. One, which seems plausible enough, is that existing linguistic families correspond to genetic markers in the peoples who originally spoke them, and are related in a similar tree that relates human populations. The other, far more controversial, is that just as *Homo sapiens sapiens* is the offspring of a single stock and a common group of ancestors, so also are all human languages related; and in fact some root words of Proto-Human can be reconstructed by the comparative method.

Amateur philologists like myself will naturally jump at such a tantalizing suggestion; and perhaps one test of how good an amateur you are is to see how many problems you can find with his development of the themes.

His basic method, like Joseph Greenberg's, is bulk comparison of vocabulary. Few attempts to reconstruct underlying forms are made here, with some conspicuous exceptions. His method is to present words in lists, and invite readers to perceive similarities themselves; an interesting rhetorical ploy that makes the reader the accomplice in setting up the main thesis. In these lists, reconstructed forms from protolanguages appear unmarked next to vocabulary items taken from wide ranges of existing lnaguages; there's no reassurance that any of the words in one of his rows are from the same language, or are actually attested forms. The same is true of the long lists of words cited in text: protolanguages appear besides dozens of obscure languages. Nouns, verbs, and grammatical particles or inflections appear alongside one another in the same list.

The rest of the book is taken towards presenting an argument that explains why reconstructions or analyses of the history of his roots will not be forthcoming. Ruhlen argues that no such analysis was necessary to identify Indo-European, which was discovered on the strength of word lists alone. One flaw in this argument is that scientific hypotheses are not thumbs-up or thumbs-down propositions; rather, they gain or lose confidence depending on the depth, power, and detail of their explanatory power. It isn't that these word lists aren't enough to get to "maybe," it's that they aren't enough to go further than that.

In any case, phonological and historical rules *do* appear in the text. Ruhlen notes correctly that often a /k/ sound is voiced to /g/, or that /t/ sounds can be palatized to /c/ or /s/ sounds. These phonological rules let Ruhlen cast a wider net, allowing more and more words from different languages to be mustered as evidence. They *never* appear as historical grounds to reject a purported cognate, on the ground that the inherited form in this particular branch must have changed in an intermediate stage prior to the observed form in one language.

The bottom line is that Ruhlen's hypothesis stalls at the "maybe" stage. Someone is going to have to do the harder work of actual historical linguistics here if this hypothesis is going to be able to move past that stage.