Tuesday, October 11, 2016

Extending evolutionary theory? - Samir Okasha

I will be attending the Royal Society Meeting on New trends in evolutionary biology: biological, philosophical and social science perspectives. I'll post each of the abstracts and ask for your help in deciding what question to pose to the speakers. Here's the abstract for Samir Okasha's talk on
Evolution and the metaphor of agency

It is striking that evolutionary biology often uses the language of intentional psychology to describe the behaviour of evolved organisms, their genes, and the process of natural selection that led to their evolution. Thus a cuckoo chick ‘deceives’ its host; a worker ant ‘prefers’ to tend the queen's eggs to those of other workers; a swallow ‘realises’ that winter is approaching and ‘wants’ to escape it; an imprinted gene ‘knows’ whether it was inherited paternally or maternally; and natural selection ‘chooses’ some phenotypes over others.

This intentional idiom is a symptom of a broader way of thinking about and modelling evolution, which I call ‘agential’. This involves treating evolved entities, paradigmatically individual organisms, as if they were agents trying to achieve a goal, namely maximisation of reproductive fitness (or some proxy). The use of rational choice models, originally intended to apply to deliberate human action, in an evolutionary context, is one symptom of agential thinking.

I offer a cautious defence of agential thinking in evolutionary biology. I argue that this mode of thinking does genuine intellectual work, and is not ‘idle metaphor’. The key point is that attributions of agency presuppose a ‘unity of purpose’. Therefore an evolved organism can only be treated as agent-like to the extent that its phenotypic traits have complementary rather than antagonistic functions, i.e. contribute to a single overall goal. Where this is not the case, e.g. because of unresolved intra-genomic conflict, the metaphor of agency ceases to be applicable.
Samir Okasha is a professor of philosophy at the University of Bristol (UK). He raises an interesting point but it's a very old point that has been discussed and resolved many years ago. I don't think this has anything to do with extending evolutionary theory. It doesn't even count as a "new trend in evolutionary biology" (the title of the conference).


  1. Okasha writes: "The key point is that attributions of agency presuppose a ‘unity of purpose’."

    I don't see why biology needs to limit 'agential thinking' to this case. Especially in the context of this conference which celebrates the putative wonders of 'multi-level selection'.

    Prince and Maynard-Smith long ago brought game-theoretic considerations into situations where there is conflict of purpose in biology, and the Swedish National Bank is regularly handing out Nobel prizes in economics to thinkers who investigate the consequences of multi-agent conflict and cooperation within firms and other collectives. Yes, Arrow and others have shown that multi-agent collectives cannot always be completely rational, but they can be usefully thought of as purposeful agents nonetheless.

  2. This comment has been removed by the author.

  3. Agree with Jim Menegay's critique of Samir Okasha's stated positions. Organisms are menageries - but you can still call them "agents".