Protolanguage Was Symbolic

Protolanguage evolved perhaps as much as two million years ago, in the context of stone tool use, but was already used for complex symbolic actions the University of California (Berkeley)’s anthropologist and linguist Terrence Deacon told his audience in the final session of the first day of the conference on protolanguage in Torun, Poland.  Deacon was not present, but presented his material over the Internet and took questions via Skype’s online telephone system.

Deacon began by rejecting the argument that language is a relatively recent development tracing to the upper Paleolithic period when the first art and technological diversity appeared. That idea rests on the notion of a sudden grand mutation in which one of many possible universal grammars appeared. Deacon favors a “prolonged co-evolution” of vocal-motor adaptations, cognitive adaptations, and social adaptations.

As evidence supporting his position Deacon pointed to the robustness of language. If language came from a big mutational leap 100 to 150 thousand years ago, we should see only a small number of genes involved in its production. These genes would be easily mutated, producing birth defects with very specific language deficits such as an inability to insert clauses within clauses.  Instead there are many genetic correlates of language and language acquisition follows a predictable pattern in cultures around the world.  Language is well integrated with our other cognitive abilities. These patterns all fit with an ancient and extended evolution.

This robustness leads Deacon to assume that symbolic communication began with the rise of stone tools for cutting meat, about 2.4 to 1.8 million years ago. Readers of Deacon’s book, The Symbolic Species, will know that the term “symbolic communication” is not used casually. Communication, Deacon teaches, can be iconic, indexical, or symbolic. Ionic communication is illustrative, for example making a chopping gesture is iconic, showing what an action looks like. Indexical communication depends on the larger context. Pointing at something gets its meaning from the object of the pointing. Symbolic communication lacks an intrinsic link to its reference. Pointing or mimicking will not do. Rituals can be symbolic. A blood-brother ritual in which blood flows and mixes can be powerful, but its meaning cannot be indicated by pointing or showing.

Deacon disagrees with the many linguists who insist that grammar is a system of rules for organizing a series of arbitrary signs. Grammar, he says, adds indexical signs to “ground” the symbols. Thus, “He married her in that house,” grounds the symbolic action married with a whole range of indexical signs (he, her, in, that).

This view of grammar has implications for protolanguage, which is commonly thought of as language without an organizing set of rules. In Deacon’s view protolanguage is language that is not fully grounded. To make up for the absence, ritual has to play a larger role in grounding the symbol. We still see much of this sort of thing in greetings, ceremonies, trading, and threats.

Deacon believes that originally, protolanguage was limited to pantomime and pointing with only stereotypical sounds. The basis for this position is the familiar one that apes have no voluntary control over their vocalizations but do make voluntary gestures. Thus, they would have first used gestures and only later transferred these signs to vocalizations. Why transfer them? Deacon suggests it is easier to learn a vocalization than a hand sign. Hand signs are reversed for the looker. If we face one another and I see you wave your right hand, the hand you wave is on my left. In order to imitate you and wave my right hand, I have to make a transformation, translating your action on my left as a right side action. Vocalizations, however, require no such transformation.

It was a clever argument, but it is not clear how many people were persuaded by it. First, deaf children seem to learn hand signs without the transformation problem. Second, it is not immediately clear that evolving voluntary control over one’s vocal apparatus is easier than evolving an intuitive transformation.

Although protolanguage is symbolic, it is still rooted in the here and now. To move beyond that, to talk of absent things, protolanguage had to become a full language by evolving syntax. The syntax serves the indexical functions when there is nothing to point to.

Deacon also discussed at length the symbolic need that he believes created the pressure for protolanguage. This was the need to establish a way of males hunting together but mating separately. I won’t go into that argument here because readers of his book will be familiar with it, and the argument is so detailed that to give it justice would require a separate report.

Comments

[uzza]

Is there any consensus on what protolanguage is? Bickerton has offered a half-dozen putative examples, all of which are different and have been shown to possess more structure than claimed. In the absence of a definition, what can we really say about it?
--------------------------
BLOGGER: Science often talks about things that are poorly defined. The hope is to study a phenomenon and come to understand it. Thus, the theory of evolution is older than Darwin, talk of gravity predated Newton, etc. Science is not math, which is indeed definition dependent.

[JoseAngel]

It's not clear to me that syntax is a requirement, or even a decisive factor, in de-localizing reference. Well perhaps I should get hold of Deacon's text and that would make it clearer... I agree of course that syntax allows meanings to have a degree of complexity and clarity they cannot have without it. But absent reference? Not so clear to me. E.g., miming the sound of eating, in cartoon protolanguage, "yummy yummy", you are "referring" albeit in a vague sense to absent food, maybe even "signifying" it. But there is no syntax there, only a primitive lexis and the relevant situation.
------------------
BLOGGER: The issue of what's decisive in de-localizing/displacement is discussed in another presentation reported here under the heading "Simulation Suggests Steps to Speaking of the Unseen."

[Stanly]

This idea belong to Breskin. “Triad method for studying the anthropology of language and art” published in Russian language by Institute of Philosophy of Russian Academy of Science journal “Epistemology & Philosophy of Science” 2, v.16, 2008, 119-132, Moscow. ISSN 1811-833X T.Deacon must red this paper in Russian.

http://www.russian.slavica.org/article9068.html
-------------------------
BLOGGER: The citation you give is dated 2008, but Deacon's book is 10 years older than that.