Saturday, 23 September 2006

Interactional Expertise (Blagging Your Way in Academia)

A story which appeared in Nature a few months ago has tickled my ribs. Basically, a sociologist of science hung out with Gravity Wave physicists for a 30 years and picked up enough of the lingo that his responses to seven questions on gravitational wave physics was indistinguishable from answers from real gravity wave physicists. Some have called it a reverse-Sokal hoax. The sociologists involved in the experiment have likened it to a Turing test. I find that the findings of the experiment resonate with my experiences as a research scientist with fingers in half-a-dozen pies (like geometric phase, quantum channels, cryptography, quantum dots, ion traps, and control theory). My interactions with specialists in several fields are very like those of the sociologist and the GW physicists, trying to pass myself off as someone conversant with the specialist terminology and concepts. I find myself now doubting my own competence as a physicist, maybe I've just blagged my way to where I am?

One may try to use the results of the sociology experiment (Is this ironic, to use the techniques and methodology of the subject under study?) to reinforce the claim that all of science is simply social construction, that scientific truths have no greater claim to validity that any other socially conventions, and other notions of linguistic hegomonies. However, we need to look at the actual study and how it fits in with the wider framework of their sociological research. They distinguish between different grades of expertise and in particular, the sociologists do not claim to have imitated "contributory expertise", the ability to do research, merely the ability to appear to speak as if one had the ability.

I myself am philosophically in the functionalist camp. If someone does all of the things that a gravitational wave physicist does, then functionally they are a gravitational wave physicist. A full Turing test would probe the full technical competence of the subject. As it stood, the questions asked in the test were conceptual rather than technical or claculational so did not probe all of the functional aspects of being a gravity wave physicist. A movie set of submarine may look like a sub but can't submerge. I am reminded of the great mathematician, Ramanujan, who exemplified the total opposite of interactional expertise. He was totally self-taught, not even having talked to another professional mathematician before he was brought to Cambridge by Hardy, yet even today his results are still filtering into mainstream mathematics.

You can't spend 30 years with gravity wave physicists and not pick up some of the physics. The important part of the responses were that they were technically correct. Even though the sociologist couldn't do calculations in linearized general relativity, he had absorbed enough physics to answer the questions correctly. This is what happens when you go slightly out of your speciality, you may not be able to be perfectly conversant with the theoretical techniques but may know enough to understand the general concepts involved. But a thorough scientific training should allow one to bootstrap what one already knows to learn the new field. I greatly admire S. Chandresekhar who changed his specialism six times in his career, writing monographs on each field at the conclusion of each episode.

So what? Science, contrary to what some radical cultural relativists would say, is not about making up stories about the world and merely talking in the right terminology and giving the expected answers demanded by a imperialistic and fascist orthodoxy, but is about doing, probing nature and contributing to a tightly constrained framework of observation, experiment and theory.