Symbolic Species [Paperback]

After several years, I found only one other book which
covers the same essential territory, Derek Bickerton's
"Language and Species," which was published in 1990,
fully eight years before Deacon's book. Deacon includes
Bickerton's book in his bibliography, and even refers
to it in his discussion of creole and pidgin languages.

But there is apparently something deeply strange going
on here. We have two books on the evolution of language,
one written by Deacon (who is basically a biologist,
an evolutionary anthropologist, and a polymath) and the
other by Bickerton, who is a linguist and a polymath.
Since the subject is the evolution of language, in
theory the linguist might have an advantage, especially
since Deacon apparently has no linguistic training at
all.

Yet Deacon manages to ignore all of Bickerton's most
important points! I'll single out the one I find most
important here.

Describing the evolution of language as a
System Of Communication is fraught with problems. As
Deacon points out, there are apparently no "simple
languages" -- there are only animal calls and the hugely
different phenomenon of human speech. This indeed was
the key problem which caused Deacon to begin researching
his book -- a child asked him why animals do not have
simple languages.

Yet Bickerton has the answer to these puzzles!! The
smooth and orderly evolutionary process is clearly
visible if we understand first that language is a
System Of Representation before it is a System Of
Communication. All animals with nervous systems and
senses have systems for representing the external world,
with a clearly visible evolutionary path from Venus
Flytraps through cockroaches, fish, frogs, dogs, chimps
and humans. As this Primary Representational System (PRS)
becomes richer and more informative, the animals so
endowed become progressively more ready to develop
"protolanguage" -- the earliest form of language,
clearly visible in pidgins, two-year-old humans, and
the "speech" of chimpanzees. The step from protolanguage
to language has only been taken by us, and it involved
_syntax_ as the defining characteristic.

How could Deacon possibly have ignored all of these
extremely intuitive insights? One is forced to conclude
that (a) Deacon didn't actually read Bickerton
(b) Deacon holds these ideas in utter contempt, or
(c) Deacon is behaving like a territorial academic,
ignoring anything which is "Not Invented Here."
Unfortunately, alternative (b) seems unlikely, since
Deacon did put Bickerton in his bibliography. Thus
we are apparently left with (a) and (c) as our
unappetizing choices.

Academic power politics may well be involved, too.
Deacon comes from the Harvard axis of academic power,
and he has loads of highly prestigious backers.
Bickerton is only a guy from the University of
Hawaii, and therefore a "lightweight."

But there is, to my mind, no question at all (!) of
which book is more helpful -- indeed revolutionary.
Anyone interested in the subject will want both books,
but Bickerton is clearly the man who bears the palm
away.

This three-fold enterprise depends on the neuroscience results discussed in Part Two of "The Symbolic Species". For example, Figure 7.8 draws our attention to the idea that prefrontal cortex is disproportionately large in the human brain. Deacon suggests that changes in the relative sizes of brain regions during human evolution is a mechanism for adaptations that allow humans to better perform language tasks. Figure 8.3 pictorially illustrates an evolutionary trend in anatomical connections towards more direct cerebral cortex control over the motor neurons that are involved in vocalizations. These examples illustrate the fact that Deacon's theory of brain/language co-evolution is heavily dependent on comparative studies of brain anatomy. Deacon tries to convince us that the major anatomical changes during human brain evolution are the precise types of changes in an ape brain that would facilitate human language behavior. According to Deacon's theory, early humans started using language as a social innovation and then the human brain changed so as to make it easier to use human language. The fact that human social interactions are a huge part of the human environment guarantees that there has to be some truth in Deacon's theory, but is it just part of a larger story?

A specific issue that Deacon touches on is the fact that non-human apes are able to learn the basics of human language simply by being exposed to a social environment where human language is being used. Why do non-human apes learn the basics of language rapidly and then stop developing more sophisticated language behavior just at the developmental stage where human children are taking off with a huge vocabulary and increasingly complex syntax? The best that Deacon's theory can suggest is that humans, unlike chimps, have had 2 million years of language use and subsequent brain evolution in response to selective pressure for larger brain regions that aid in symbolic thought. I agree that it would be astounding if certain brain regions such as the adult human prefrontal cortex is not more useful for human language tasks than is the chimp prefrontal cortex, but is this really the most important thing we need to know about the relationship between brains and language?

Is there another way of looking at the difference between human and chimp brains? One that might better inform us about the functional differences between human and chimp brains that give humans superior language skills? Deacon mentions an alternative in Chapter 6, "...the rate of human brain maturation...is prolonged compared to other primates..." In fact, most human brain growth happens after birth while most chimp brain growth happens before birth. What does this have to do with language behavior?

Perhaps everything. Why DO humans have big brains? Even though Deacon correctly points out the fact that, in the case of brains, bigger does not mean better, his whole theory ends up depending on the idea that by making some brain regions bigger, you get an ape that is better at learning human language. Deacon tries to gloss over this contradiction by assuring us that his theory is really making use of a powerful mechanism for evolving a more language-competent brain, the mechanism of "parcellation", which he claims can mechanistically explain data such as those given in Figure 8.3. Can parcellation really do all the explanatory work that his theory demands or is there a need for additional mechanisms?

Why DO humans have big brains? What if big human brains are just a side effect of some other more important aspect of brain physiology? What if larger human brain size is just a side effect of evolutionary selection for prolonged synaptic plasticity during human childhood? Maybe if we could alter a few genes in bonobos so as to prolong postnatal brain growth in certain bonobo brain regions like the anterior cingulate cortex, just maybe we would give bonobos a longer window for developing sophisticated language skills.

There is a whole tradition within neuroscience that started with behavioral studies of associative learning and led to studies of the cellular and molecular mechanisms of learning and memory. This branch of neuroscience research is almost completely ignored by Deacon. We have to wonder if Deacon's focus on neuroanatomy has provided him with a limited data set which paints his theory of brain/language co-evolution into a corner.

So my advice is that people who are interested in language should read Deacon's book, but recognize the limitations of his perspective. In the next few decades the rest of the story of how brains make human language behavior possible will come rolling in. Deacon has provided us with a working model of how to apply this hard-won knowledge of the brain to our understanding of human language, but Deacon's is just an early pass at this kind of empirically-anchored theoretical neurolingustics. Much more is yet to come. Even scientists should heed Wittgenstein's warning not to be too quick to formulate grand theories of language while so much data remains to be collected.