Thursday, December 11, 2008

The virus of language

There's this great book out there that I encourage everyone to read: Manin's "Mathematics as metaphor". OK, if you can read the Russian version go for it! It contains a lot more then the English version published by the AMS. In fact to be precise, the AMS version in turn has a few things that are not in the Russian version, but you'll miss some of the great stuff like the piece on the Strugatsky brothers and Visotsky, the poems, and all the parts involving cultural life in the Moscow 1960s ... it's a real pity that's not available in the English text. I'll probably return some other time to comment on the various parts of this book, but right now I want to point especially to the parts that have to do with linguistics and psycholinguistics. These, as I understand, originated from a rich interaction between mathematicians, linguists and philologists that happened in academic circles in Moscow and more specifically a seminar on psycholinguistics organized by the author for a period of time there. This interest in linguistics manifests itself, in this selection of writings, in two different directions, one pointing towards mathematical logic (and if you haven't read the author's book on mathematical logic please do so now, oh no, better wait in fact: there's a new edition on its way... don't miss it) and another pointing towards the early developments of language, and the trickster figure of Jungian psychology. I won't say more and leave it to the enjoyment of the reader to discover more. I prefer here to let my thoughts be carried along a chain of associations that were somehow triggered by reading some of these essays and that I try to reconstruct here in some approximate form.

By far the weirdest idea about the origin of language emerged from the beat literature. The provocative concept of "language as a virus" was most famously voiced in the "Nova Trilogy" of William S. Burroughs. Perhaps his most intriguing and challenging literary production, this cycle of three "science fiction" novels (to be read in arbitrary order) is based on the idea that language is indeed a sort of viral infection imposed on the human species by a form of alien invaders from the Crab Nebula, as a way to enslave and control us. The virus of language can be fought be destroying and deconstructing language, which is what the experimental form of writing used to compose the three novels ("The Soft Machine", "Nova Express", and "The Ticket that Exploded") effectively does. The breaking up of the linguistic structure, achieved via the techniques of "cut-up" and "fold-in", that is, starting with an ordinary text with its ordinary linear narrative and linguistic structure, one cuts it up in short pieces consisting of a few words at a time and reassembles them in a random order, or else one assembles different texts via a shuffle operation, consisting of again cutting up each piece in short bits and mixing them by alternating those in different texts while keeping the ordering within each text fixed (that's what one calls a shuffle product in mathematics).

This breaking up of language is also a breaking up of the linear structure of time. This is evident in the narrative of Burroughs novels, where the same events occur simultaneously in all three novels and the chain of cause and effect is altered beyond recognition. The result provokes the readers into a deep reflection upon the relation between language and our perception of thermodynamic time and between the notion of causality, time-like curves and linguistic structures. Is language intimately tied up to time perception? Can one exist without the other? One can argue that some form of time perception clearly exists in not language producing animals, but what seems more difficult to argue about is whether there is a possibility of language and linguistic structure disjoint from temporality. At least the Burroughs experiments seem to claim it impossible.

Some might remember the Laurie Anderson's piece "Language is a Virus from Outer Space", inspired by the Burroughs trilogy. If not, it's here on You Tube:
Laurie Anderson: Language is a Virus from Outer Space

The idea of breaking up the linear structure of language
by "fracturing time", used in Burroughs' cut-up technique,
may have a parallel in the context of the psychedelic
experience of the US 1960s counterculture, where it
was argued that the liberation from the ego achieved
through the semi-mystical experiences induced by the
use of LSD or magic mushrooms happened through, among
other psychic phenomena, the breaking up of language
and of the perception of time.

The classic text of the counterculture era on
drug induced states of consciousness was "the
psychedelic experience", the highly controvertial
text inspired by a reading of the Tibetan Book
of the Dead as a guidebook for the use of mind
altering drugs and hallucinogenics, produced by
a trio of Harvard psychologists, Leary, Metzner,
and Alpert, turned gurus of the new psychedelic
generation. The unconditional enthusiasm and
almost religious fervor that emanate from this
book are in stark contrast with the much more
measured and interesting analysis of similar
experiences reported in Huxley's "Doors of
Perception". Huxley manages to present the results of
his self-experimenting with magic mushrooms not
with the preaching tone of the zealous convert,
but with the detached observations of a curious
and unprejudiced mind. He does not meddle with
the (non-psychedelic) culture of ancient Tibet,
dubiously mixing it up with Latin American cults
like Leary does, nor he pretends to have found the
way to illumination for all. However, in his description
of the mental states induced by the assumption of
hallucinogenic drugs, one also finds traces of
both the breaking down of the usual linear time
perception, and of the simultaneous breaking down
of language as the main method of analyzing and
organizing thought and experience.

Breaking down language into building blocks,
regardless of the fanciful idea of the virus from
outer space, is part of the baggage of linguistic
theory. More so in certain schools and circles
of linguistics than in others. I've been through a
few linguistics books recently where, in one form
or another, the idea of the fragmentation of
language into elementary atoms comes to play
a significant role.

"From molecule to metaphor" is the more neuroscience oriented of the lot. It works on three levels: biology, meaning studies of neurons and brain activity; simulation, based on computer programs aimed at studying models of language formation in the brain, and linguistics and psychology. The question is the emergence of the syntactic and grammatical structure of language in brain activity. A long term task, though several interesting observations are collected in this essay that make it worth paying attention to.

Given my own general cultural upbringing, except for an in depth five years study of comparative philology of the classical Indo-European languages, and especially of ancient Greek, my exposure to linguistics was mostly through the school of Noam Chomsky: parameters, transformative grammar, naturally. The idea of breaking down language in basic building blocks, responsible for correlations of certain types of grammatical structures across various groups of languages, is indeed central to that approach to linguistics, though highly contested by other schools. There is a fairly well known book that aims at popularizing the Chomskian approach and the notion of parameters in linguistics, aptly called "The atoms of language". It is pretty well written though it suffers from all the typical drawbacks of popularization, namely overemphasizing one point of view, using metaphors of dubious relevance (the whole comparison with the history of chemistry is too far fetched and mostly irrelevant to understanding the linguistic notions the book is describing). It has the nice feature of drawing plenty of examples from the large pool of native American languages, although once more some of the examples that the author draws "for effect" from exotic Australian aboriginal languages could just as easily be constructed in, say, Latin. In any case, the book is very entertaining and goes some way into presenting the main ideas of the Chomsky approach. I have to say though, that if I want to read about this particular brand of linguistics, I prefer to go directly for Chomsky's "Aspects of the theory of syntax" as I did many years ago rather than for the clever popularization of "Atoms of language", pleasant reading as the latter may be.

On a ground somewhere in between lies this other pretty interesting book, "Lingua ex Machina", another of the usual cool MIT press publications. It discusses, to some extent, the difficult evolutionary implications of Chomsky's idea of universal grammar. The point is to identify plausible evolutionary scenarios, partly grounded in the Darwinian idea of "conversion of functions" that may support glottogenesis in a Chomskian perspective. It is an intriguing book, better or at least more stimulating in my opinion than the other two I briefly reviewed here above. Especially the third possibility they discuss, the "corticocortical coherence" scenario, which would imply a threshold transition, is intriguing. In a series of recent books trying to integrate neuroscience and linguistics, this one still stands out as a particularly cleverly crafted effort.