Ramachandran's synesthetic bootstrapping theory of language origins

I'm currently reading a book of neurologist and neurosurgeon Vilayanur Ramachandran's fascinating 2003 Reith Lectures on "The Emerging Mind". It's all well worth reading for anyone interested in human consciousness, but one section in particular has stood out for me so far—the one where Ramachandran outlines his "new synesthetic bootstrapping theory of language origins".

Synesthesia is a brain condition where sensory information normally processed exclusively by one area of the brain is additionally processed by another brain area or areas. A well known example is the way that some people perceive particular colours when they see numbers (red for "1", blue for "2", for example). Indeed, it seems that artists' skill in expressing metaphors of reality could be informed by an unusually strong aptitude for synesthesia.

Ramachandran postulates that human language may have evolved through just such cross-domain remapping of sensory information in the fusiform gyrus area of the brain. He describes a chain of three such mappings:

1) A non-arbitrary synesthetic correspondence between visual object shape (represented in visual centres) and sound contours represented in the auditory cortex.

For example, people across cultures tend to associate a word like "booba" with a rounded shape and "kiki" with a pointed one—a kind of innate visual onomatopoeia.

2) Cross-domain mapping (perhaps involving the arcuate fasciculus) between sound contours and motor neurons in Broca's area (mediated, perhaps, by mirror neurons).

For instance, we make a small shape with our lips to say "tiny" but big shape to say "enormous".

... Such mimicry indicates a pre-existing bias to systematically map certain visual shapes on to certain "sounds" represented in the motor maps in the Broca's area.

3) Motor to motor mappings (synkinesia) caused by links between hand gestures and tongue, lip and mouth movements in the Penfield motor homunculus.

... A system of non-verbal communication would have been important to our ancestral hominids unable to engage in loud vocalisation when hunting

Acting together, these three [cross-mappings] have a synergistic bootstrapping effect—an avalanche culminating in the emergence of a primitive language.

The author goes on to ponder the origins of the hierachical structure of syntax (imagining it could come from the sub-assembly technique of early hominids' tool use) and the role of mirror neurons in developing shared vocabulary in infants by unifying the cognitive processes for word perception and production.

Mirror neurons must be establishing (1) volitional motor [commands]; (2) felt lip and tongue position; (3) the seen image of someone else's lips and tongue; and (4) the heard phoneme.

The thread running through the author's theory is the notion of language emerging from the cross-mapping or repurposing of existing brain functions. And beyond providing us with a far deeper understanding of our own brains, I have a strong feeling that a modular and emergent approach to language acquisition could pay huge dividends in the search for solutions for both intelligent digital networks and machine-based language processing...

Technorati Tags: identity, language_acquisition, neurology