Monday, February 27, 2012

Introduction, Parts I-III

Running behind, unsurprisingly.  Here's a response I drafted Saturday to parts I-III.

In another aborted internet discussion of philosophy once upon a time between Carleton alums, with a philosophy Ph.D. student leading a discussion of Freud's Civilization and Its Discontents, I wrote a response to the first chapter that effectively consisted of, "okay, I get general idea X, but that seems obvious, and all of these specifics seem wrong."  I can't even remember what it was, the leader of the discussion responded that basically I had conceded the major idea Freud was arguing for as his major premise, and the concepts I had taken issue with were effectively asides or illustrative anecdotes.  In short, as I remember him putting it, it's now Freud's world and we just live in it.

In reading Kant's introduction, I feel like I'm in much the same situation.  The thrust of the first three sections seems to be laying out a rubric for what can be considered a priori knowledge (or that which we know in advance of experience), and arguing it into a very small corner.  Perhaps I should have started with Plato instead of Kant, as it appears that Plato is the foil with which Kant is jousting here, but my issue is that Kant doesn't go far enough.  In short, he lays out certain principles such as gravity, which I would generally term scientific laws, and tries to carve out a small space for a priori knowledge along with what he terms the perfection of mathematics.  My issue is that he carves out any space for a priori knowledge at all.  I'm resisting the temptation to launch into my speculations about the epistemological origins of mathematics and its contingent nature because I've just started this project, but I'm not sure I believe that any knowledge can be considered a priori in the sense that Kant describes it.  All that said, with my minimal background in philosophy, I suspect that I'm simply going along the road that Kant mapped out, and if that thought is true, I've already given Kant the victory over his primary targets in this space.


  1. Kant's description of the scientific process in his introductions worries me a bit. I think either he didn't have much exposure to lab science, or lab science has changed a lot since his day. He says of natural philosophers that "they learned that reason only perceives that which it produces after its own design; that it must not be content to follow, as it were, in the leading-strings of nature, but must proceed in advance with principles of judgement according to its unvarying laws and compel nature to reply to its questions. For accidental observations, made according to no preconceived plan, cannot be united under a necessary law."

    My dim reading of popular science suggests that accidental observations are quite often involved in significant scientific advances. And my equally dim knowledge of math history suggests that sometimes a practical mathematical result / process stands and continues to work, even when the theoretical framework under it shifts significantly.

    I'm picking at Kant in the intro for this because he's saying that metaphysics should be more like science and math if it wants to get ahead, and yet his conception of how math and science really work seems off.

  2. Also, I'm looking forward to digging in to this issue of how Kant grounds a priori knowledge with you more. I'm not convinced by his example/argument in section II any more than you are, and I suspect he's relying on a body of pre-existing philosophical work and people's comfort level with that work.

  3. My most recent post gets into this somewhat, but I really do think some of this is Kant entering debates of the time and using the terminology of the time. In some ways, I guess I keep expecting Kant to be a pure empiricist for some reason, even though I know he's not. His purpose isn't to hammer everything that's "enlightened" down to empirical grounds; rather he's trying to locate a space for "science" (by which I think he generally means structured forms of inquiry, not the more limited definition of the term as we use it today) that isn't tied at all points to empiricism. We're still in the introduction, so we can't expect this to be fully fleshed out yet, but I think what he's driving after is that while knowledge originates from experience, at some point it enters an abstract realm restricted to the mind. To be a total geek about it, take science fiction for instance. There is no way that stories about aliens and interstellar travel can be derived from experience, and yet we find some sci-fi stories believable and others not. The knowledge (or intuition) of what is credible comes from our experiences, but it is being extrapolated to realms far beyond our experiences.

    For your point on science, I had the same reaction initially, but the more I think about it, the more I think Kant has a point. Yes, science should always endeavor to be guided by empiricism, but as the critical science studies folks show us over and over again, scientists tend to find what they're looking for. That's a very damning way to say what Kant is getting after in the first place here, which is that the goal of the scientist isn't to just observe and observe and observe some more and then report those observations. It's to take theory -- or hypotheses, if you will -- and see which ones the data fits. I'm remembering a story I read in a very good biography of Darwin, from his early years (taken from his memoirs, I think), where he was out doing geology with a friend and found a fossil in a gravel pit that by all the theory shouldn't have been there. Excitedly, he reported it to his mentor, who told him it basically to disregard it. When Darwin protested, the mentor told him that believing that a fossil of that age belonged there would involve throwing out everything that they'd known about the geology of the region to accomodate one fossil. in short, the best explanation was that it had to have been dropped by another geologist, or blown by the wind, or something.

    To this day, scientists throw out outliers by the bushelful, and in Kant's terminology, that's because they're asserting what is, in the short term at least, a priori knowledge about the system on the data. It's the only way we can actually do science -- to try to impose some sort of structured knowledge that arises from our theories of how things could work onto our experiences. Science is, I would say, largely the work of situating, grounding, checking, verifying, and differentiating theories based on empirical observation. Yes, breakthroughs happen when you have outliers that you can't make go away, but you first try. I can still hear Craig Heinke telling me that "when physicists find neutrinos in their equations or observations, they work like beavers to get rid of them." That's because yes, it's possible that neutrinos actually exist, but they screw up an awful lot of stuff if they do.