Tuesday, July 28, 2009

The Possibility of Impossible Cultures
Imagine Finnegan's Wake written in an impossible language. I imagine many people who've tried to read it have already concluded that it was written in an impossible language. But in fact what makes the book so nerve-wracking is that it was written in syntactically recognizable English, but with a slew of invented names, places and verbs occupying the place of more familiar subjects, predicates and adjectives, and no punctuation anywhere in sight.

No, what I mean is: imagine a language that, for example, mandates placing a particular word in a fixed position in the sentence, no matter when it is used. Or a language in which a statement of fact can be converted into a question by reversing the order of the words. (What kind of logic would follow from such a language?)

Marc Hauser, professor in the Departments of Psychology, Human Evolutionary Biology and Organismic and Evolutionary Biology at Harvard has a fascinating article (registration required) in the July 9th issue of Nature on the contraints placed on human cultural evolution. Or rather the possibility of getting around what have been observed to be the constraints on cultural evolution.

The article is detailed and I don't want to just cut and paste chunks of it here. But I do want to look briefly at Hauser's discussion of the sharp differences between animals and humans when it comes to various reflections of intelligence, and how it has been assumed that these differences between animals and our own species amount to a matter of degree, and not kind. Brain science has a lot to say about this.

The simple use of tools, for example, was long considered a significant distinction between us and other animals, until research of primates and other species showed that chimpanzees make use of natural items, like branches, for use as tools, as do some species of birds. But making use of tools for one thing, and making use of them for several different things...is something else.
Although anthropologists disagree about the timing of the human cultural revolution... many researchers point to fundamental changes starting some 800,000 years ago in the Early Palaeolithic, with a crescendo of change at around 45,000–40,000 years ago in the Late Palaeolithic. This period is associated with the generation of symbols (mathematical, artistic and ritualistic), controlled fire for use in cooking and other forms of environmental transformation, and tools with multiple components and functions (for example, tools used for expressing both aggression and music). Given that this interval of several thousand years is barely noticeable on an evolutionary timescale, and that such cultural expressions emerged rapidly, the parallel with the Cambrian is striking: that is, something similar to a genetic revolution must have occurred during this period, providing humans with an unprecedented set of capacities for generating novel cultural expressions in language, morality, music and technology.
The human brain, Hauser goes on, changed from a system with a high degree of modularity with few interfaces to one with 'numerous promiscuous and combinatorially creative interfaces.' These interfaces are what bestowed on humans a set of abilities to generate novel cultural expressions in language, morality, music and technology.

At this point in the article, Hauser really brings the distinction into a fine relief (to borrow an art term) by pointing out the limits of other animals' intelligence.
Although many vertebrates have evolved brains with reciprocal connections or loops between different cortical areas (for example, basal ganglia to the cortex and back), these loops are restricted to particular functions....At the most general level, it is clear that the motor systems of all animals must involve recursive operations to allow organisms to take a discrete set of motor options and generate a vast range of functionally meaningful motor acts or sequences in novel environments. For example, whether an organism flies or runs, its legs must repeatedly lift and fall or its wings must repeatedly beat. However, because an organism's habitat and climate is constantly changing, the iterative or recursive rule of cycling through leg lifts or beating the wings must be flexible so that the animal's response can vary in response to environmental change.

That said, the recursive properties of the motor system seem to be locked into motor function in all animals but humans. For example, in striking contrast to the recursive operations in human language, with its unrestricted use of different content or classes of words, the looping circuitry that is necessary for song acquisition in songbirds only supports singing and, in some cases, mimicry of other biological and non-biological sounds. This circuitry is not, however, used when they acquire the calls that constitute their repertoire more generally, including the sounds used in social interactions, food discovery and alarm calls.

Another example of generative computation comes from the domain of artefacts, in particular the creation and diversity of human tools. Unlike many of our simplest tools, such as the pencil, animal tools consist of a single material, never include more than one functional component, are typically dispensed after their first use and are never used for functions other than the original one. The first two features reveal that, unlike human tools, the representation of animal tools is not combinatorial. A pencil can combine four materials (graphite, wood, metal and rubber) to create four functions (graphite for writing, wood for holding the graphite, metal for attaching the rubber to the wood, and rubber for erasing). Moreover, each material can be used for a variety of other functions: for example, rubber can be a component of chewing gum. As experiments reveal, if a young child is asked what she can do with a pencil other than write, she will immediately offer such functions as holding up her hair, puncturing a plastic cover and poking a friend...Only humans think of artefacts as being designed for a particular function but, as a result of promiscuous interfaces, entertain many other possible functions. [emphasis mine]

I realize this may not be news exactly, but I like how Hauser has drawn such careful attention to these distinctions, which I think do tend to be overlooked in general science writing on evolution. It isn't just a matter of degree.

His article goes on to discuss what kinds of research might more clearly map out, for lack of a better term, the blind spots in human cultural evolution, what kinds of cultural expressions have not become evident either because they are impossible for us to evolve, given the constraints on our evolution, or because they would be so complicated as to not survive and take root.


Lab Rat said...

Intresting! I've been so wrapped up in Synthetic biology lately that i haven't been looking at many things human or cultural fora while. It was nice to read through this and get a perspective slightly larger than the 1o micrometer one I've been focussed on lately :D

TheOFloinn said...

Use of tools?