Elliot Sober on the present and future of philosophy of biology

Back in Sept. I was fortunate to be able to attend a philosophy of science “summit” at the University of Calgary, with talks by a bunch of the world’s top philosophers of science. I thought I’d share my notes from Elliot Sober’s talk, on the present and future of philosophy of biology. As I’m sure most of you know, Sober is a top philosopher of evolutionary biology, his book The Nature of Selection is a classic. I found his talk very interesting for several reasons. He talked about the state of philosophy of biology and its place within philosophy more broadly. I always have an anthropological interest in hearing about how people see the state of their own fields. He had a lot of advice about how to do philosophy of science, much of which encouraged philosophers to engage in scientific debates. And he made some passing remarks on how scientists in various fields perceive philosophers (apparently we ecologists are unusually receptive to philosophical input!) I don’t know enough about philosophy to evaluate all of Sober’s remarks, but I enjoyed mulling them over.

My notes follow. I did the best I could, but obviously any errors or omissions are mine.*

*************************

Philosophy of biology today seems to have less and less connection to the rest of philosophy, and seems to have little to contribute to science itself. Talking about science is science journalism; it’s not the same as contributing to science. Worried that philosophy departments will stop hiring philosophers of biology.

Philosophers seem to think that philosophy of science, and philosophy of biology, are now less central to philosophy than was the case 20-30 years ago. Why?

Public controversies about biology which had philosophical elements (e.g., sociobiology) used to have a high public profile. Not so much anymore. Gifted popularizers of biology also used to talk about philosophical issues (Dawkins, Gould, Lewontin). Again, not so much anymore.

Sociologist Kieran Healy (aside from me: hey, I’ve heard of him, I read his blog!) has done citation analyses of changes in philosophy, rankings of philosophy depts., centrality of different disciplines. Philosophy of science is not central to philosophy (though not peripheral either).

Biology is relatively hospitable to philosophy of science. Half-joking: 99% of physicists think philosophy is bullshit, it’s only 95% for biologists. (aside from me: Wonder what the number is for ecologists? I bet it’s fairly low, which maybe that means ecology is especially fertile ground for philosophy of science? But in that case, why does ecology seem to get much less attention from philosophers than evolutionary biology? Presumably because evolution has an agreed-on core set of ideas and questions that give philosophers a handle to latch onto? Whereas ecology is kind of a mess, so that it’s hard for outsiders to develop a road map of the field and figure out where they might profitably contribute philosophical insight?)

Overspecialization and the “regionalist turn” in philosophy of science (Jean Gayon–the view that nothing of interest in philosophy of science can be done except in within-discipline work). Reason to doubt this—methods of reasoning and inference are not subject-matter specific. Philosophers of biology sometimes unaware that their questions had been addressed in general philosophy of science. Unfortunate because a good trick for developing your career is to use well-developed ideas from one area to solve problems in another area (aside from me: Yup! That’s not just good advice for philosophers, it’s good advice for anyone. That’s what a good chunk of my own career consists of, anyway–shamelessly stealing ideas from one area and applying them to a different area: applying the Price equation to ecology, applying modern coexistence theory to the IDH, half the blog posts I write. Or think of neutral theory in ecology, or MaxEnt, or using ideas from economics to understand resource trade mutualisms…)

Philosophers seem to be retreating from making normative statements about the practice of science. Describing science without critiquing it. We critique creationists, why not scientists? Scientists make normative judgements of one another’s methods, so why can’t philosophers do so too? Or think of statistics—gives normative advice on how to proceed, given epistemic goals and empirical facts.

Clarifying a concept or the logic of a line of argument is clearly recognized as “real” philosophy by philosophers outside of philosophy of science. That’s not just purely descriptive work.

Retreat from normativity also due to influence of history of science on philosophy of science? Historians think of normative judgements as anachronistic and hubristic.

Rational reconstruction of historical scientific arguments—is this legit? Helpful? If the scientists themselves didn’t use the logical and mathematical tools used in your reconstruction, isn’t that anachronistic? (aside from me: Deborah Mayo would say yes. She calls rational reconstructions–e.g., trying to show that a piece of pre-Bayesian scientific reasoning was “really” Bayesian–“painting by numbers”. Just because you can come up with a paint by numbers picture of the Mona Lisa, in what sense does that let you understand how the Mona Lisa was painted? But Sober likes the approach and uses it himself.)

One way for philosophers to make normative contributions without fear—find a scientific controversy and engage in it. Of course need to identify controversies that do have a philosophical component. Which means you need to reject or at least not take too seriously Quine’s claim that philosophy is continuous with science. (aside from me: speaking as a scientist, this is great advice. There are a lot of scientific controversies that are really philosophical, but that aren’t recognized by the participating scientists as philosophical. Or even if they are, the scientists lack the philosophical expertise to properly resolve them. If any philosophers are reading this and want some suggestions for ecological topics that could use some proper philosophical attention, drop me a line!)

Another way to be interestingly normative—find a proposition scientists accept uncritically, and identify scientifically interesting conditions under which it would be true or false. Note that these conditions need not be highly probable. For instance, think of Felsenstein’s demonstration that cladistics parsimony is statistically inconsistent. Turns out that parsimony does in fact make implicit substantive assumptions.

Practical tips to get scientists to pay attention—collaborate with scientists and publish in scientific journals. (aside from me: many of the philosophers I admire have done this. Sober himself. Deborah Mayo. William Wimsatt. Samir Okasha. It would be cool if someone like Chris Eliot were to publish some philosophy of ecology in an ecology journal. I’ll bet Oikos would take philosophy of ecology in its Forum section, and the right paper might fly as a Synthesis & Perspectives piece in Ecology. Ideas in Ecology & Evolution would go for it too, though they’re less widely-read.)

Normative problems are often discarded as dead ends when the question could be revised in a fruitful way. Example: best Bayesian definition of degree of confirmation? Change the question to “how do Bayesian and frequentist accounts of testing differ and which is better under which circumstances?” The latter question is of practical relevance for science—journals have policies on what statistics are appropriate. What’s the right criterion for empirical significance? Change the question to “What does it mean for a theory to be testable, relative to background info?” Does Ockham’s razor depend on the assumptions that nature is simple? Change the question to “Does cladistics parsimony work only when evolution is parsimonious?”

All research programs experience diminishing returns. But when it happens with normative problems, that doesn’t mean you should stop asking normative questions, you just need fresh ones.

After the talk, the discussion was kicked off by some designated commenters, some of whom were scientists (aside from me: I thought this was an interesting way to structure the symposium; I’d be curious to see a similar structure tried at an ESA symposium). Ford Doolittle remarked that molecular biologists and genomicists often don’t even realize they have a philosophy. Which leads to disputes that are thought to be empirical, but are not (think of the debate over ENCODE and whether 80% of the genome is ‘functional’). Another example: the debate over whether there’s a tree of life is really a debate over “what do you mean by ‘tree’”? As opposed to evolutionary biologists and ecologists, who are open to philosophy. To reach biomedical and molecular types, Doolittle suggested that philosophers will need to publish in Nature and Science and PNAS, presumably in collaboration with biologists.

*Sorry, no time for real posts at the moment, we’re all swamped. But my grant deadline will soon be past and I’m now done with my teaching for this term, so normal service from me will resume shortly.