Thursday 29 November 2012

Structural Realism

The New Scientist 'Big Idea' piece on Structural Realism ("What truly exists? Structure as a route to the real" Eric Scerri, New Scientist 24th November 2012, pp30-31) is signposted on the front cover with
Every thing must go. Why theories of reality don't need stuff
As presented inside, though, it's not quite what I thought.  It's not information as reality (such as presented by Vedral), but Structural Realism. Structural Realism, as it is descibed in the New Scientist article, is an alternative to scientific realism or anti-realism. The argument is that scientific realism has trouble coping when theories change:
so many past theories and theorised entities have come and gone (remember ether or phlogiston?), why should we ever regard any of them as "real"?
Structural realism gives up on the reality of entities but argues for the reality of structure
In 1989, John Worrall, a philosopher of science at the London School of Economics, published the paper "Structural Realism: The Best of Both Worlds?" in the journal Dialectica. In it, he outlined structural realism, an approach he traced back to French mathematician Henri Poincaré, among others. For Worrall, what survives when scientific theories change is not so much the content (entities) as the underlying mathematical structure (form)
While I'm all in favour of championing the importance of form vis a vis the material, superficially I don't see how theories of form should be expected to be any more enduring than theories of entities.  I don't see how we can ever go beyond accepting the provisionality of knowledge, whether it is about entities or structures.  Is it not the case that what physicists are doing at the moment is thrashing about to find mathematical models that account for measurements - finding maths that accounts for things observed in the experiments on the large hadron collider or and cosmological measurements?


Eric Scerri said...

Thanks for your comment on my New Scientist article. I never claimed that theories concerned with "form" rather than entities were in any way immunized against refutation. The claim is that while an entity can be definitely shown not to 'exist' some mathematical aspects tend to survive even following a theory change.

eric scerri

David Chapman said...

Thanks for your comment, Eric, and thanks for the thought-provoking article in New Scientist. Perhaps I was reading too much into what you were saying.