What papers and books have most influenced you as an ecologist? This isn’t the same as asking about the best papers and books you’ve ever read. I’ve read lots of brilliant papers which haven’t influenced me, for instance because they merely reinforced, rather than changed, how I look at the world, or because they were so different than anything I would ever produce myself that I could only admire them ‘from afar’ as it were. And I’m thinking of direct influences here, not indirect influences. We’ve all been indirectly influenced by Darwin’s Origin.
For me, the most influential papers have mostly been the ones that shape how I think about science in general, not the ones that shape my views on any particular topic. My biggest influences include:
- Bill Wimsatt’s 1987 book chapter on “False models as means to truer theories,” recently reprinted in his book Re-engineering Philosophy for Limited Beings. Wimsatt is a philosopher of science, but before that he trained as an engineer, and his philosophy reflects that training. He’s not an a priori theorist about how science ought to work in some hypothetical world, he’s an a posteriori theorist who wants to understand why scientists actually do what they do, why those approaches (sometimes) work, and why they (sometimes) fail. In this chapter, he lays out a taxonomy of all the different ways in which models can be false, and why their falsity is not merely an unfortunate reflection of the fact that we can never know the whole truth, but actually is essential to how they help us learn the truth. If you, like me, have always found Box’s famous remark that “All models are wrong, but some are useful,” to be unsatisfyingly question-begging (“Yes, but how are they useful?”), you should read Wimsatt. If you think that models are nothing but a source of falsifiable hypotheses to be rejected, or not rejected, by data, you should read Wimsatt. If you think that models aren’t useful at all, or at least aren’t essential, you should read Wimsatt. If you’re a scientist, you should read Wimsatt.
- Deborah Mayo’s Error and the Growth of Experimental Knowledge. Terrific book on the philosophy of statistics. If you want to know why I remain a frequentist, and why I am deeply uncomfortable both with those scientists who have become subjective Bayesians, and with those more numerous scientists who think of themselves as “pragmatists” who don’t need to bother themselves with what “probability” really means, read this book. Not always an easy read, and somewhat repetitive, but very thorough and written (like Wimsatt) with a strong eye towards understanding and justifying actual scientific practice.
- Frank 1997. The paper that reintroduced the Price equation to the world. I read this in grad school just because it was the lead article in that month’s issue of Evolution, and didn’t fully grasp it, though I could tell it was about something deep. Just goes to show that influence sometimes has a latent period. Because years later, when I read Loreau and Hector 2001 and their claim to be using the Price equation to solve a technical problem in biodiversity and ecosystem function, I thought “Hmm, the Price equation, I vaguely remember reading something about that, wonder what that’s all about.” So I went back and re-read Frank 1997. And then read it again. And again, this time working through the math line by line. And then I read some of George Price’s own stuff (including the next paper on my list). And eventually, over a period of about three years, and with several false dawns (including one which led me to embarrass myself in front of Alan Grafen), I understood the Price equation. And saw that you could do a lot with it in ecology, including such unexpected applications as this (I also saw that Loreau and Hector’s own attempt to apply it was problematic, but fixable). Indeed, this paper influenced me so much that some people now seem to think of me as “the Price equation guy,” as if I don’t think about anything else. So these days I actually actively avoid trying to think about the Price equation, because I don’t think it’s good to get too fixated on one idea (at least, it’s not good for me and my science). More deeply, thinking about the Price equation has shaped my thinking about a range of broader conceptual issues like the relationship between “mechanistic” and “phenomenological” models, and about how math helps us learn about the world.
- Price 1995. “The nature of selection,” Price’s final paper, published posthumously. If you’re struggling to wrap your head around the Price equation, read this, Price’s informal discussion of what “selection” is. Natural selection is merely one special case; “selection” is a much more general principle. A tremendous help to me.
- Ludwig Wittgenstein’s Philosophical Investigations. Not necessarily for the specifics of any of his philosophical positions, but more for his general approach: the questioning of assumptions, especially unrecognized ones (but not in a pointless, “What if we’re all just brains in vats?” way; Wittgenstein recognizes, as deeply as anyone, the importance of taking some things for granted), the importance of our choice of question and the way in which we frame it, the ways in which the very words we use can mislead us, the importance of not mistaking empirical questions for conceptual ones or vice-versa (while also recognizing that the empirical and conceptual shade into one another). Maybe above all, the notion that sometimes the right question to ask, in science as well as epistemology, is not “How do you know?” but “What do you mean?”
So if you think I’m full of bollocks, now you know what to avoid reading. 😉
This is a neat question that I think forces me to remember and think about all the things I’ve read. Two things came to mind after a bit of reflection. The first one is Gary William Flakes The computational beauty of nature . This really helped me think about how to understand ecological systems with computational methods. I read it a few years ago, but I still crack it open. It seems abstract at first, but I’ve found it to be very useful still. I opened it up for a brush up on the logistic map equation and help creating bifurcation plots for some simulations I’m working on.
Next I’d say Douglas Hofstadter’s Godel, Escher, Bach: An eternal golden braid . This is a pretty far sweeping book that has to do with the nature of intelligence and artificial intelligence. What stands out for me is that it was the first thing I read about how to think in a formal system where you have rules, theorems and proofs. If you’ve ever read the book at all you’l know it goes way beyond that, but that really stands out for me.
Godel, Escher, Bach? Man, I thought I was going far afield by naming the PI! 😉 I read it and loved it back in high school, but I can’t say it influenced me–it struck me as such a unique one-off. I’ll have to pick up a copy of the Flakes book.
Organization of Communities Past and Present: Gee and Giller
Pingback: 37 ways choice of words can mislead scientists « Oikos Blog
Your posts are all very stimulating, but I’m overwhelmed by the rate at which you place new stuff… I just don’t have time to think about them! Congrats for this!
It is very amazing to see that your taste is very similar to mine; I do agree that everything from Bill Wimsatt, George Price and Steve Frank is worth several readings by every biologist; and I’m personally very keen on Wittgenstein`s philosophy, and do think it has deep implications for ecology (material for a new post?) but if I’m to maintain my friendships, I wouldn’t suggest to read their books first! (probably the excellent biography of Wittgenstein`s universe by Ray Monk [“The duty of a genius”] is a good starting point to place his philosophy in context; and the movie/theatrical exercise/experimental work by Derek Jarman on Wittgenstein is a great visual and intellectual experience; I highly recommend both).
A more personal note: the paper by James Drake in 1991 in Am Nat (“Community-Assembly Mechanics and the Structure of an Experimental Species Ensemble”) is a masterpiece, in my opinion, on community assembly dynamics, showing the importance of e.g., contingency on community composition. Just the abstract is able to stimulate several new ideas and experiments. In my opinion, it is all intriguing that it has not attracted a greater attention, in particular in the light of the current mess of theoretical community ecology.
Pingback: Must-read: on ordinary language vs. scientific understanding « Oikos Blog
Pingback: Friday links: the internet is (mostly) in reruns this week | Dynamic Ecology
Pingback: Answers to reader questions, part 3: what we’d say to Congress, tropical vs. temperate systems, and more | Dynamic Ecology
Pingback: False models are useful BECAUSE they’re false | Dynamic Ecology
Pingback: Books that all ecology grad students should read | Dynamic Ecology