Part of the sequence: Rationality and Philosophy
Hitherto the people attracted to philosophy have been mostly those who loved the big generalizations, which were all wrong, so that few people with exact minds have taken up the subject.
Bertrand Russell
I've complained before that philosophy is a diseased discipline which spends far too much of its time debating definitions, ignoring relevant scientific results, and endlessly re-interpreting old dead guys who didn't know the slightest bit of 20th century science. Is that still the case?
You bet. There's some good philosophy out there, but much of it is bad enough to make CMU philosopher Clark Glymour suggest that on tight university budgets, philosophy departments could be defunded unless their work is useful to (cited by) scientists and engineers — just as his own work on causal Bayes nets is now widely used in artificial intelligence and other fields.
How did philosophy get this way? Russell's hypothesis is not too shabby. Check the syllabi of the undergraduate "intro to philosophy" classes at the world's top 5 U.S. philosophy departments — NYU, Rutgers, Princeton, Michigan Ann Arbor, and Harvard — and you'll find that they spend a lot of time with (1) old dead guys who were wrong about almost everything because they knew nothing of modern logic, probability theory, or science, and with (2) 20th century philosophers who were way too enamored with cogsci-ignorant armchair philosophy. (I say more about the reasons for philosophy's degenerate state here.)
As the CEO of a philosophy/math/compsci research institute, I think many philosophical problems are important. But the field of philosophy doesn't seem to be very good at answering them. What can we do?
Why, come up with better philosophical methods, of course!
Scientific methods have improved over time, and so can philosophical methods. Here is the first of my recommendations...
More Pearl and Kahneman, less Plato and Kant
Philosophical training should begin with the latest and greatest formal methods ("Pearl" for the probabilistic graphical models made famous in Pearl 1988), and the latest and greatest science ("Kahneman" for the science of human reasoning reviewed in Kahneman 2011). Beginning with Plato and Kant (and company), as most universities do today, both (1) filters for inexact thinkers, as Russell suggested, and (2) teaches people to have too much respect for failed philosophical methods that are out of touch with 20th century breakthroughs in math and science.
So, I recommend we teach young philosophy students:
more Bayesian rationality, heuristics and biases, & debiasing, | less | informal "critical thinking skills"; |
more mathematical logic & theory of computation, | less | term logic; |
more probability theory & Bayesian scientific method, | less | pre-1980 philosophy of science; |
more psychology of concepts & machine learning, | less | conceptual analysis; |
more formal epistemology & computational epistemology, | less | pre-1980 epistemology; |
more physics & cosmology, | less | pre-1980 metaphysics; |
more psychology of choice, | less | philosophy of free will; |
more moral psychology, decision theory, and game theory, | less | intuitionist moral philosophy; |
more cognitive psychology & cognitive neuroscience, | less | pre-1980 philosophy of mind; |
more linguistics & psycholinguistics, | less | pre-1980 philosophy of language; |
more neuroaesthetics, | less | aesthetics; |
more causal models & psychology of causal perception, | less | pre-1980 theories of causation. |
(In other words: train philosophy students like they do at CMU, but even "more so.")
So, my own "intro to philosophy" mega-course might be guided by the following core readings:
- Stanovich, Rationality and the Reflective Mind (2010)
- Hinman, Fundamentals of Mathematical Logic (2005)
- Russell & Norvig, Artificial Intelligence: A Modern Approach (3rd edition, 2009) — contains chapters which briefly introduce probability theory, probabilistic graphical models, computational decision theory and game theory, knowledge representation, machine learning, computational epistemology, and other useful subjects
- Sipser, Introduction to the Theory of Computation (3rd edition, 2012) — relevant to lots of philosophical problems, as discussed in Aaronson (2011)
- Howson & Urbach, Scientific Reasoning: The Bayesian Approach (3rd edition, 2005)
- Holyoak & Morrison (eds.), The Oxford Handbook of Thinking and Reasoning (2012) — contains chapters which briefly introduce the psychology of knowledge representation, concepts, categories, causal learning, explanation, argument, decision making, judgment heuristics, moral judgment, behavioral game theory, problem solving, creativity, and other useful subjects
- Dolan & Sharot (eds.), Neuroscience of Preference and Choice (2011)
- Krane, Modern Physics (3rd edition, 2012) — includes a brief introduction to cosmology
(There are many prerequisites to these, of course. I think philosophy should be a Highly Advanced subject of study that requires lots of prior training in maths and the sciences, like string theory but hopefully more productive.)
Once students are equipped with some of the latest math and science, then let them tackle The Big Questions. I bet they'd get farther than those raised on Plato and Kant instead.
You might also let them read 20th century analytic philosophy at that point — hopefully their training will have inoculated them from picking up bad thinking habits.
Previous post: Philosophy Needs to Trust Your Rationality Even Though It Shouldn't
Provocative article. I agree that philosophers should be reading Pearl and Kahneman. I even agree that philosophers should spend more time with Pearl and Kahneman (and lots of other contemporary thinkers) than they do with Plato and Kant. But then, that pretty much describes my own graduate training in philosophy. And it describes the graduate training (at a very different school) received by many of the students in the department where I now teach. I recognize that my experience may be unusual, but I wonder if philosophy and philosophical training really are the way you think they are.
Bearing in mind that my own experiences may be quite unusual, I present some musings on the article nonetheless:
(1) You seem to think that philosophical training involves a lot of Aristotelian ideas (see your entries for "pre-1980 theories of causation" and "term logic"). In my philosophical education, including as an undergraduate, I took two courses that were explicitly concerned with Aristotle. Both of them were explicitly labeled as "history of philosophy" courses. Students are sometimes taught bits of Aristotelian (and Medieval) syllogistic, but those ideas are never, so far as I know, the main things taught in logic (as opposed to history) courses. In the freshman-level logic course that I teach, we build a natural deduction system up through first-order logic (with identity), plus a bit of simplified axiomatic set theory (extensionality, an axiom for the empty set instead of the axiom of comprehension, pairing, union, and power set), and a bit of probability theory for finite sample spaces (since I'm not allowed to assume that freshmen have had calculus). We cover Aristotle's logic in less than one lecture, as a note on categorical sentences when we get to first-order logic. And really, we only do that because it is useful to see that "Some Ss are Ps" is the negation of "No Ss are Ps," before thinking about how to solve probability problems like finding the probability of at least one six in three tosses of a fair die. Critical thinking courses are almost always service courses directed at non-philosophers.
(2) You seem to think that philosophers do a lot of conceptual analysis, rather than empirical work. In my own philosophy education, I was told that conceptual analysis does not work and that with perhaps the exception of Tarski's analysis of logical consequence, there have been no successful conceptual analyses of philosophically interesting concepts. Moreover, I had several classes -- classes where the concern was with how people think (either in general or about specific things) -- where we paid attention to contemporary psychology, cognitive science, and neuroscience. In fact, restricting attention to material assigned in philosophy classes I have taken, you would find more Kahneman and Tversky than you would Plato or Kant. And you would also find a lot of other psychologists and cognitive scientists, including Gopnik, Cheng, Penn, Povinelli, Sloman, Wolff, Marr, Gibson, Damasio, and so on and so forth. Graduate students in my department are generally distrustful of their own intuitions and look for empirical ways to get at concepts (when they even care about concepts). For example, one excellent student in my department, Zach Horne, has been thinking a bit about the analysis of knowledge (which is by no means the central problem in contemporary epistemology), but he's attacking the problem via experiments involving semantic integration. And I've done my own experimental work on the analysis of knowledge, though the experiments were not as clever.
(3) You seem to think that philosophy before 1980 (why that date??) is not sufficiently connected to actual science to be worth reading, and that this is mostly what philosophers read. Both are, I think, incorrect claims.
With respect to the first claim, there is lots of philosophical work before 1980 that is both closely engaged with contemporaneous science and amazingly useful to read. Take a look at Carnap's article on "Testability and Meaning," or his book on The Logical Foundations of Probability. Read through Reichenbach's book on The Direction of Time. These books definitely repay close reading. All of Russell's work was written before 1980 -- since he died in 1970! Wittgenstein's later work is enormously useful for preventing unnecessary disputes about words, but it was written before 1980. This shouldn't be surprising. After all, lots of scientific, mathematical, and statistical work from before 1980 is well worth reading today. Lots of the heuristics and biases literature from the '70s is still great to read. Savage's Foundations of Statistics is definitely worth reading today. As is lots of material from de Finetti, Good, Turing, Wright, Neyman, Simon, and many others. Feynman's The Character of Physical Law was a lecture series delivered in 1960. Is it past its expiration date? It's not the place to go for cutting edge physics, but I would highly recommend it as reading for an undergraduate. I might assign a chunk of it in my undergraduate philosophy of science course next semester. (Unless you convince me it's a really, really bad idea.) Why think that philosophical work ages worse than scientific work?
With respect to the second claim, you might be right with respect to undergraduate education. On the other hand, undergraduate physics education isn't a whole lot better (if any), is it? But with respect to graduate training, it seems to me that if one is interested in contemporary problems, rather than caring about the history of ideas, one reads primarily contemporary philosophers. In a typical philosophy course on causation, I would guess you read more of David Lewis than anyone. But that's not so bad, since Lewis' ideas are very closely connected to Pearl on the one hand and the dominant approaches to causal inference in statistics on the other. The syllabus and reading lists for the graduate seminar on causation that I am just wrapping up teaching are here, in case you want to see the way I approach teaching the topic. I'll just note that in my smallish seminar (about eight people -- six enrolled for credit) two people are writing on decision theory, two are writing on how to use causal Bayes nets to do counterfactual reasoning, and one is writing on the contextual unanimity requirement in probabilistic accounts of causation. Only one person is doing what might be considered an historical project.
Rather than giving a very artificial cut-off date, it seems to me we ought to be reading good philosophy from whenever it comes. Sometimes, that will mean reading old-but-good work from Bacon or Boole or (yes) Kant or Peirce or Carnap. And that is okay.
(4) You seem to endorse Glymour's recommendation that philosophy departments be judged based on the external funding they pull in. On the other hand, you say there should be less philosophical work (or training at least) on free will. As I pointed out the first time you mentioned Glymour's manifesto, there is more than a little tension here, since work on free will (which you and I and probably Glymour don't care about) does get external funding. (In any event, this is more than a little odd, since it typically isn't the way funding of university departments works in the humanities, anyway, where most funding is tied to teaching rather than to research and where most salaries are pathetically small in comparison with STEM counterparts.) Where I really agree with Glymour is in thinking that philosophy departments ought to be shelter for iconoclasts. But in that case, philosophy should be understood to be the discipline that houses the weirdos. We should then keep a look-out for good ideas coming from philosophy, since those rare gems are often worth quite a lot, but we also shouldn't panic when the discipline looks like it's run by a bunch of weirdos. In fact, I think this is pretty close to being exactly what contemporary philosophy actually is as a discipline.
I'm sure I could say a lot more, but this comment is already excessively long. Perhaps the take-away should be this. Set aside the question of how philosophy is taught now. I am receptive to teaching philosophy in a better way. I want the best minds to be studying and doing philosophy. (And if I can't get that, then I would at least like the best minds to see that there is value in doing philosophy even if they decide to spend their effort elsewhere.) If I can pull in the best people by learning and teaching more artificial intelligence or statistics or whatever, I'm game. I teach a lot of that now, but even if I didn't, I hope I would be more interested in inspiring people to learn and think and push civilization forward than in business as usual.
EDIT: I guess markdown language didn't like my numbering scheme. (I really wish we had a preview window for comments.)
You did indeed have an unusual philosophical training. In fact, the head of your dissertation committee was a co-author with Glymour on the work that Pearl built on with Causality.
Not really. Term logic is my only mention of Aristotle, and I know that philosophy departments focus on first-order logic and not term logic these days. Your training was not unusual in this matter. First-order logic ... (read more)