Recently, Meg emailed Brian and I the following question:
Hi Brian and Jeremy,
You’ve both written about philosophy of science, and those posts always make me realize that I have not read nearly enough of that. I’ve read Platt and Chamberlin, but not Lakatos and Mayo who you both favor.
I’m wondering if:
1. you have general recommendations on Readings in the Philosophy of Science 101, and
2. if there might be interest in a blog post along these lines, since I’m sure lots of other folks are clueless about this, too.
Brian and I replied at sufficient length that we decided that we could just publish edited versions of our replies as a blog post.*
Our replies take for granted that you want to read some philosophy of science. If you’re not convinced of that, you should read this old post first.
Ooh, good question! But one which, embarrassingly, I’m not sure I’m a good person to answer, because my own reading has been quite haphazard. I’ve found what little I’ve read to be useful. But I’m not sure if the stuff I happen to have read and found useful would necessarily be a great starting point for anyone else.
If I was going to pick one philosophy paper that I’d recommend to any scientist, it’d be Wimsatt 1987, “False models as means to truer theories”. It’s a book chapter in Neutral Models in Biology. I’ve linked to it and summarized it in an old post on how false models are useful because they’re false. I read it as a grad student and it really opened my eyes. And it’s totally non-technical and accessible, any scientist can read it easily. Wimsatt also has a book called Re-Engineering Philosophy for Limited Beings that lays out his overall point of view. But it’s aimed more at philosophers than scientists.
That classic Richard Levins piece on “the strategy of model building in population biology” is something everyone should read. And philosophers like Steve Orzack, Elliott Sober, and Michael Weisberg have written papers commenting on the philosophy of science implicit in Levins’ piece (see here and here for instance).
I’d like to read some Karl Popper. It’s embarrassing that I still only know him n-th hand. I think reading him is kind of the philosophy of science equivalent of reading Darwin’s Origin. From what I hear, he’s different (and more sophisticated) than the sort of garbled cartoon version scientists know.
I’ll let Brian talk about Lakatos since he’s read him and I haven’t.
I guess besides Wimsatt’s piece, I’d probably suggest philosophers who are writing about some bit of science that interests you and that you know a bit about. I find that sort of stuff both more interesting and easier to follow than “general” philosophy of science that’s either example-free, or that focuses on examples about which I know nothing and on issues that I as a scientist have no reason to care about. So I’ve read a decent amount of philosophy of evolutionary biology. Folks like Elliott Sober, Samir Okasha, and Peter Godfrey-Smith on issues like the levels of selection. Sober’s 1993 book The Nature of Selection is a classic that set the agenda for that whole subfield of philosophy of science. I like that stuff a lot, because it’s an area where there are very tight links (or maybe very fuzzy boundaries) between philosophical and scientific issues. But if you’re not into the levels of selection or whether selection is a “force” or whatever, it might not grab you.
Philosophers of science haven’t paid much attention to ecology, unfortunately, in large part because to outsiders ecology seems like a mess with no conceptual core. But Chris Eliot, who’s commented on the blog, is a philosopher of science who writes about ecology. He’s published several philosophy papers about community ecology. I read Cooper’s The Science of the Struggle for Existence, which is all about ecology, but had mixed feelings. But apparently philosophers liked the book better.
Bob Holt suggested Deborah Mayo’s error statistics book to me in grad school. I loved it and it had a big influence on me, but I’m not sure it’s for everyone. It’s a long book and repetitive at times. And she spends a lot of time attacking a philosophy of statistics (subjective Bayesianism) that’s never gotten much traction among either professional scientists or practicing statisticians as far as I know. But if you wanted something on philosophy of statistics, that’d be my suggestion. (UPDATE: Deborah Mayo herself has been kind of enough to comment and make clear something I should’ve made clearer here, namely that her attacks on subjective Bayesianism are only a part of her much larger project to develop a philosophy of statistics and experimentation. That broader project, not just the bits about subjective Bayesianism, is what influenced me and why I recommend the book. See this old post where I had a go at summarizing and extolling the virtues of Deborah Mayo’s work.) And the link above goes to her website, where you can also find papers presenting condensed versions of her ideas (which lose something compared to her book, I think). Or you could have a look at a couple of volumes she’s edited or written chapters for, one of which was edited by ecologists: Error and Inference, and The Nature of Scientific Evidence.
You could also try reading scientists who write about their own philosophy of science, like the Levins piece. But I have to say that, having read some philosophy by proper philosophers, I find the philosophy of science that scientists write to be a pretty mixed bag (and that includes my own tentative attempts at philosophy on the blog). Robert Peters’ A Critique for Ecology is poor philosophy, I think–it’s worth reading, but more as a provocation. There are various ecosystem ecologists who’ve written philosophical things that I just haven’t found very compelling for whatever reason, but I know other ecologists like that stuff. There are several chapters in Resetarits & Bernardo’s Experimental Ecology in which ecologists like Earl Werner and Peter Morin talk about the philosophy behind their research programs. I don’t know that that counts as “philosophy of science”, exactly, but it’s close. As I’ve mentioned on the blog, Dennis Chitty’s Do Lemmings Commit Suicide? is a really interesting read. It’s Chitty’s memoir of his career studying small mammal population cycles. Any population ecologist would be totally into it. And Chitty’s pretty clear about his philosophy of science and how it shaped his work–and also about how he failed to answer the questions he spent his career trying to answer. It’s interesting to think about whether he’d have done different work and maybe made more progress if he’d had a different philosophy of science.
Here’s the link I give my stat students for reading Lakatos: http://www.lse.ac.uk/philosophy/lakatos-4/. You can actually listen to him lecture if you prefer (although the lectures are also transcribed). The “Science and Pseudoscience” lecture which was for the general public is an especially good introduction. Essential features of Lakatos are: 1) all theories are falsified sometimes (direct slam on his PhD adviser, Popper). 2) To propose a theory you individually have to have a core idea that you assume to be true. 3) So what distinguishes successful theories are that they make novel, risky predictions that pan out. 4) Over time, other scientists will vote with their feet and choose research programs that are sucessfully making new predictions instead of post hoc explaining away inconsistencies (so there is a strong social element, much like Kuhn but more realistic in my opinion).
Quinn & Dunham (1983 AmNat On Hypothesis Testing in Ecology and Evolution ), is an interesting critique of Platt and whether it can apply to ecology.
The Golem is a very quick read popular book that picks apart the idea that science has a special formula and highlights just how messy science is and reinterprets many classic stories used by philosophers of science to show how things were fudged, misrepresented,etc. I once guest lectured for several weeks in a “Methods of Science” for high school teachers class that used this as a text. Even though its goal is subversive, it teaches a lot of core ideas by explaining them before trashing them.
It is hard to go too far and escape Kuhn’s very short readable book on The Structure of Scientific Revolutions.
Bacon’s Novum Organum (online lots of places) is a pretty short and interesting read. It is the backlash to the idea that science was 100% deduction that lasted for ~2000 years after Euclid (go look at Newton’s Principia sometime – everything is theorem, proof, theorem, proof). Seriously. Newton barely mentions fitting empirical reality. Even though Lakatos will cite him 300 years later for making a bold prediction in predicting the existence of Neptune to explain deviations between empirical reality and theory.
What passes for philosophy of science in philosophy departments is pretty far removed from what practicing scientists care or think about. So most “introduction to philosophy of science” is going to be dry and irrelevant. Two exceptions:
Theory and Reality an Introduction to the Philosophy of Science (haven’t read it but after a quick scan of Amazon this is what I would buy – indeed I”m tempted to buy it – although the reviews say it gives a harsh treatment of Lakatos). The Philosophy of Science Encyclopedia ($500 nobody is going to order this but lots of libraries have it in their reference section – I have used this fairly often although obviously its a reference and not an introduction)
Lots of great thoughts from the two of you, as I expected! I think I read Do Lemmings Commit Suicide as a grad student, but don’t remember it very well, which suggests I should reread it. Thinking about this more prompted me to go take the binder where I have all my notes from the Quantitative Methods in Ecology and Evolution course (taught by my advisor, Alan Tessier) off my book shelf. Alan devoted the second lecture to the philosophical framework for doing ecology and evolution. Looking back through my notes, the paper we read for this class was Quinn and Dunham’s On hypothesis testing in ecology and evolution. I’d forgotten about that paper, but, based on a quick skim, I really ought to reread it! Alan contrasted this with Loehle’s Hypothesis testing in ecology: Psychological aspects and the importance of theory maturation. That also looks well worth reading.
So, I’m back to feeling like I need to set aside more time to read! We’ve compiled quite a reading list here! I’d love to hear suggestions from readers, too.
UPDATE: The comment thread has a bunch of great suggestions, from both scientists and philosophers of science. As with many of our posts, you’re really missing out if you don’t read the comments.
*We are nothing if not