Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives

SRB Highlights

This review appeared in Volume 9 (3) of The Semiotic Review of Books.

The Evolution of Symbols

by Michael Ruse

The Symbolic Species: the Co-evolution of Language and the Brain, by Terrence W. Deacon, New York: W.W. Norton & Company, 1997, 527 pp., ISBN 0-393-03838-6, $29.95.

Terrence Deacon is interested in the question of language. In particular, Deacon -- a well-known researcher in neuroscience and anthropology associated with Boston University and with the Harvard Medical School -- is concerned to offer us a theory about how language has evolved, and the way in which it functions today. Evolutionists take seriously the notion that language is a natural phenomenon which has appeared by normal forces of nature, and Deacon is a Darwinian inasmuch as he thinks that the chief mechanism of organic change is natural selection. But Deacon believes strongly that discussions hitherto have not simply provided wrong answers, but in fundamental ways have not even asked the right questions. Deacon believes that there is something special and unique about language in humankind and this has not been recognized adequately by previous researchers. He thinks however, that given now our knowledge of brain structure together with new understandings of the nature of language and of the forces of evolution, the time has come when one can put together a full and satisfying synthesis.

Deacon's book, The Symbolic Species is divided into three parts. The first deals with the whole question of the nature of language. Then the second, which is the longest in the book, talks of the nature of the brain, and why human brains are different from animal brains. The third and final part of the book deals with what Deacon refers to as "co-evolution," where he takes up such issues as the origination of language and its causes. In a way, although the second part is by far the most detailed and draws on Deacon's great knowledge and expertise, it is the beginning and end of the book which are really important for putting things in context, and hence it is on these that I shall concentrate here. At the end of my review, I shall make some comments about the second part of the book, although more about style than about content.

It is now pretty much a commonplace that there is something biological about language. Virtually everybody today, even the most non-biological of social scientists, recognizes that human language is not a particularly rational thing. Languages are not put together in the best possible way for understanding ideas, nor are languages learnt in the most sensible way with the simplest and most basic parts coming first and the more complex and difficult parts coming second. In addition, there are all sorts of puzzles, for instance, about why it is that really intelligent people who are good at mastering complicated things later in life find it absolutely impossible to learn perfectly a new language. Yet, looking at things from the other side, children immersed in that language, even if they are right down at the lower level of the IQ scale, pick up the language easily, quickly, and perfectly. The answer to these and like puzzles surely lies in evolutionary biology.

Thanks, as everybody knows, to the work of Noam Chomsky, it is recognized more and more that it is not simply a question of language being biological, but that in some deep sense it is a question of language being evolutionarily biological. Chomsky and his school in the last forty years have done heroic work in demonstrating the ways and extent to which language is deeply rooted in human biological nature. However, as is just as well known, there is a paradox in dealing with Chomsky: although an evolutionist, for whatever reasons, he draws back from Darwinian evolution, seeming rather to find the development of language to be an almost one-step jump or "saltation". For Chomsky, biology is all important, but Darwinian biology is irrelevant. One cannot therefore discuss things like language structure in terms of natural selection, for the mechanism simply does not apply.

There are today many who now want to pick up the Darwinian branch, even if Chomsky himself refuses to do so. Evolutionary biologists and students of language from an evolutionary perspective increasingly want to take Chomsky's insights about the biological nature of language and to relate them to the mechanism of natural selection: a connection which, in their eyes, is no less tight than that of the anatomist wants to connect things like hands and teeth and eyes to natural selection, and the sociobiologist wants to relate morality and social behaviour generally to natural selection. Most notably recently the MIT-based evolutionary psychologist and popular writer Steven Pinker has argued that a thoroughly Darwinian approach to language is not only possible but demanded. His best selling work The Language Instinct is designed precisely to show that natural selection rules okay among the verbs and nouns just as much as it does among the sperm and the ova.

Deacon simply feels that this kind of approach is, not so much inappropriate, as inadequate. He thinks that there are problems and objections with trying to root language directly in biology in the way that an ultra-Darwinian like Pinker suggests.

Steven Pinker, a proponent of the Universal Grammar view of language abilities and an articulate champion of many of Chomsky's original insights about the uniqueness of language, argues in a recent book (The Language Instinct) that innate grammatical knowledge is not at all incompatible with an adaptationist interpretation of its origins. He argues that a language instinct could have gradually evolved through the action of natural selection. On the one hand, this is a far more biologically plausible alternative to miraculous accidents and it challenges us to face some of the difficult problems ignored by theories relying on miraculous accidents to fill in the gaps. On the other hand, an adequate formal account of language competence does not provide an adequate account of how it arose through natural selection, and the search for some new structures in the human brain to fulfil this theoretical vacuum, like the search for phlogiston, has no obvious end point. Failure to locate it in such a complex hierarchy of mechanisms can always be dismissed with the injunction: look harder. (38)

So what is Deacon providing as an alternative? Basically his is a co-evolutionary scenario. He believes that, in some sense, we should stop looking in the brain for the operation of natural selection and paradoxically start looking outside the brain for the Darwinian evolutionary component. In particular, we should look at language itself. (There are analogies here with Richard Dawkins's notion of an "extended phenotype".)

In his crucial discussion, "Outside the brain," in the fourth chapter of Symbolic Species, Deacon suggests that what happens is that in some sense languages themselves evolve. They throw out new mutations and so on and so forth, and then in some way these are selected by language users. In other words, it is not so much a question of the brain changing directly to produce a language, but rather languages are changed directly to accommodate themselves to brains. Deacon writes:

The world's languages evolved spontaneously. They were not designed. If we conceive of them as though they were invented systems of rules and symbols, intentionally assembled to form logical systems, then we are apt either to assign utility and purpose where there is none, or else to interpret as idiosyncratic or inelegant that for which we cannot recognize a design principle. But languages are far more like living organisms than like mathematical proofs. The most basic principle guiding their design is not communicative utility but reproduction -- theirs and ours. So, the proper tool for analyzing language structure may not be to discover how best to model them as axiomatic rule systems but rather to study them the way we study organism structure: in evolutionary terms. Languages are social and cultural entities that have evolved with respect to the forces of selection imposed by human users. (110)
Deacon therefore suggests that, in respects, a proper way to think of languages is as parasitic. A language in some sense is something which settles itself into the human framework, but has to adapt and evolve in order to succeed. Those languages which do this do well, and those which fail to do this go extinct.

How then does one explain the kind of similarities or "deep structure" that people like Chomsky have identified in languages and which a Darwinian evolutionist would take as the best of all possible evidence that languages reflect some common structure of humankind? Simply put, Deacon argues that similarities between languages are convergent, that is to say they are rather like the fin of the dolphin and the fin of the shark, which -- although without common evolutionary ancestry (except a long way back) -- have come to be very similar because they are adapting to similar problems, namely propulsion through water. So likewise, two languages which (say) have roughly the same colour terms should be understood, not as reflecting some common ancestry, but rather as reflecting human needs to separate out different colours, and as having come independently to roughly the same solutions. The fact that we think in terms of, and use words for, black, white, red, green and so forth in different languages is not a function of a common origin, or at least not necessarily a function of a common origin. It is much more a function of the fact that humans, for biological purposes, have found it worthwhile to distinguish black, white, red, green. Hence, languages have adapted themselves in order to accommodate our needs.

Deacon backs this theory with a great deal of detailed exposition and (in his second section) discussion of the nature of the brain. But, this is not so much the crucial part of his argument. Rather it is support for it. Really what we need to do next is to jump straight to the third and final section to pick up the thread of argumentation, particularly in Chapter 12: "Symbolic Origins." This is where Deacon discusses why there is the need for language and what selective forces might have pushed us forward on this. Here we find an elaborate sociobiological argument suggesting that language evolved as a tool: a tool which enables human beings to live together socially and yet to pursue their separate reproductive goals.

The key biological problem in need of solution (in Deacon's opinion) was that of the need of males to ensure paternity. If, as in the human species, it is necessary for both males and females to be involved in child rearing -- given such obvious things as the long period of time that children need to be trained and nurtured before they can handle things on their own -- then there is the question of how males can know that they are indeed the biological fathers of their offspring. (This is the kind of question which obviously presupposes the so-called "selfish gene theory.") Deacon points out that one thing which is surely going to be biologically advantageous for males is going to be the urge and ability to provide provisions for their mates, but that the best of all possible provisions for mates (and for offspring also) will undoubtedly involve much effort and labour by these males. This applies particularly to the highest protein form of food available, namely meat.

But if males are off hunting and otherwise gathering food for their mates, what guarantee can they have that (in their absence) other males will not be servicing their mates in their stead? Deacon suggests that what has evolved is some kind of marriage contract, where mates pledge fidelity to each other and at the same time mark themselves off from the rest of the community as sexuality exclusive. In order to have this, that is in order to have a marriage contract rather than simply a male/female pair bonding, it is necessary to have some symbolic tools at hand: language! In some way, therefore, language has evolved to protect male biological investment in that which they themselves provide physical investment.

About the need, within groups of hominids, to mark out who may or who may not have access to whom, Deacon writes as follows:

The first requirement, then, is that there must be a means for marking exclusive sexual relationships in a way that all members of the group recognize. Sexual access and a corresponding obligation to provide resources are not just habits of behavior; they cannot be more or less predictable patterns, or just predictions of probable future behaviors. Sexual access is a prescription for future behaviors. No index or memory of past behaviors can represent this. Nor can any index of present social status or reproductive state mark it. Even the refusal or avoidance of sexual activity only indicates a current state and is not necessarily predictive. Sexual or mating displays are incapable of referring to what might be, or should be. This information can only be given expression symbolically. The pair-bonding relationship in the human lineage is essentially a promise, or rather a set of promises that must be made public. These not only determine what behaviors are probable in the future, but more important, they implicitly determine which future behaviors are allowed and not allowed; that is, which are defined as cheating and may result in retaliation. (399)
From this apparently stems the beginnings of a full-blown language: something which when developed is used not simply in a sexual context, but more generally. In a way therefore, although a Darwinian, Deacon sees language as being somewhat along the lines of what Stephen Jay Gould has termed a "exaptation," that is to say, a by-product of something which had an initial adaptive function but not necessarily fully adaptive at first in the way that it might function now.

What now can we say of Deacon's co-evolutionary theory for the origins of language and the brain? I will say little or nothing about Deacon's views on brain evolution; although (as noted) I shall be returning to their style at the end of this review. I am prepared to regard them as well taken. (For some such comments see Ralph Holloway's review of Deacon in the American Scientist, 86 (1998), 184-186.) As far as the question of language is concerned, I confess I find myself considerably more enthused about his beliefs on the co-evolutionary nature of language itself -- the extent to which language must evolve to fit human needs, as well as the brain evolving to fit language -- than I do about Deacon's scenario for the evolution of language itself. Quite frankly, with respect to the evolution of language and its nature, I would like to see more attention paid to possible criticisms of more orthodox Darwinians like Pinker. I am not at all convinced that Deacon has shown the inadequacies of the straightforward Darwinian approach to be as crippling as he rather assumes. Most particularly, I am always a little uncomfortable when biological similarities are put down to convergence, rather than to common ancestry. Of course we know such convergence does occur -- I have instanced the fins of the porpoise and of the shark -- but one would like more evidence that this is what is going on in the case of language. Obviously, convergence is not everything: the fact that many words in English and French are the same is not a question of convergence, but of common ancestry, or descent of one from the other.

But worries like these notwithstanding, overall it does seem to me at least a plausible line of attack to suggest that language evolves in a biological way, as much as the brain evolved in a biological way. I am not sure that I would want to say that this is quite as revolutionary as Deacon suggests, or even that it is quite the paradigm switch that is rather suggested. Rather it seems to me that this might be something an orthodox Darwinian could embrace and make much of. That is to say, it is an insight which might be absorbed within existing theory, rather than demanding a completely new alternative theory.

Less convincing is Deacon's rather touching domestic story about the evolution of language (and its connection to the need of a marriage contract!) I would have felt more comfortable had Deacon at least tried to relate his work to the thinking of some of today's leading human sociobiologists or evolutionary psychologists like Leda Cosmides and John Tooby (neither of whom are referenced in the bibliography). For instance, I would like to know just how strong the marriage bond is in various societies. It is well known that in some societies it is so weak that, in fact, when it comes to which male will take responsibility for baby, humans go for a mother's brother's care rather than for the social father's care. Maternal uncles are responsible for the well being of children rather than the social fathers. As people like Richard Alexander have pointed out, there are good biological reasons for this: namely, one is exchanging a dubious connection at the parental level for a certain (if somewhat reduced) connection at the avuncular level.

As things stand, Deacon's whole treatment of this topic very much at the level of the "Just So Stories" that have drawn the derision of the critics of human sociobiology: critics like Stephen Jay Gould and Richard Lewontin (eg 19??). Not that I am against human sociobiology as such, but I do feel that we really have got beyond the stage where one can just spin pretend scenarios and expect to be taken seriously. We must now start to root our work in the achievements which have already been made by such people as Barkow, Tooby and Cosmides (1992) or by the Canadian researchers Martin Daly and Margo Wilson (1984). (There is a reference in Deacon's book to an early work by Daly and Wilson, but not to their more recent work (1992)).

Deacon is making some most interesting points about the evolution of language. There are ideas here which should interest anyone who cares about such subjects: not simply scientists, but also philosophers and students of literary theory. I think however, that there is much which should be approached with some care. At times, Deacon just simply lets his imagination run altogether too wild. However, at this point I must open things up and admit that, since much of the book is on the evolution of the brain, I have to confess that I found that much of it was quite beyond my critical powers. This is of course in part a reflection on me, but I suspect also is indicative of a deeper problem.

The Symbolic Species is a large work, published by a commercial publisher, and obviously designed for the general market. One has only to look at the titles of the chapters to realize this. For instance the first four chapters are entitled: "The Human Paradox," "A Loss for Words," " Symbols Aren't Simple," "Outside the Brain." It is clear that Deacon and his publishers, Norton, are trying to cash in on the success on earlier works like this book is intended to be: most obviously Pinker's The Language Instinct. No doubt there is a market for these works. But despite its glamorous outside appearance, there is a huge amount of really very technical material which, perhaps, ought to be presented exclusively to the scholars in the field.

Certainly, the huge amount of material on the brain could be condensed down to a chapter or two at this level. But more than this: even though I would have preferred a popular book which was only half the size of the one I have before me, at the same time some expansion of some of the discussions would have been very welcome. Let me give just one example: in the discussion of the evolution of language, Deacon makes reference to the distinction between pidgin languages and creole languages. What he tells us is that a pidgin language is one which is a simple combination of other languages, whereas a creole language is one which develops from this merger. On the basis of this, Deacon wants to make some important points about the convergence of creole languages. Now, although these were terms I had heard of before, I certainly did not know much about what they were. Moreover, not only were they interesting ideas in their own right, but here particularly would have been a place where (for the general reader) a much more detailed and extensive discussion would have been very welcome. I think that Deacon could have given us some actual examples of pidgin languages and of creole languages and shown us how they came into being and how they evolved and how convergence came about and so on and so forth. In other words, what I am suggesting is that here is a case where Deacon could have given us quite a bit more discussion than he does, in a very interesting way, which would have been of great appeal to the general reader -- and underlined his points, to boots!

But there is much which is of interest in this book. I hope that, at the least, I have suggested to readers of this Review, who might be expected to have a more professional interest than most in the nature of language, that it is a book which they will ignore at their own loss.


Alexander, R. D. (1987) The Biology of Moral Systems. Hawthorne, N.Y.: Aldine de Gruyter.

Barkow, J.H., Cosmides, L. & Tooby, J. (1992) The Adaptive Mind. New York: Oxford University Press.

Cosmides, L. & Tooby, J. (1989) "Evolutionary Psychology and the Generation of Culture." Ethology and Sociobiology. 10, (51-97)

Daly, M. & M. Wilson (1984) Sex, Evolution and Behavior. Boston: Wadsworth

Daly, J & Wilson, M. (1992) "The Man who Mistook his Wife for a Chattel." In The Adaptative Mind, J. Barkow et al. (eds.) New York: Oxford University Press

Gould, S.J. & Lewontin, R.C. (1979) The Spandrels of San Marco and the Panglossian Program: A Critique of the Adaptationist Program. Proceedings of the Royal Society of London, 205, (281-288)

Pinker, S. (1994) TheLanguage Instinct. New York: William Morrow.

Michael Ruse is Professor of Philosophy and Zoology at the University of Guelph. He is the author of numerous articles and books on evolutionism, among which The Darwinian Paradigm: Essay on its History, Philosophy and Religious Implications (1989), Monad to Man: The Concept of Progress in Evolutionary Biology (1997) and Taking Darwin Seriously: A Naturalistic Approach to Philosophy (1998).

Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives