Thursday, June 28, 2007

Pessimistic Induction and Two Fallacies

For those interested in certain philosophy of science issues (e.g. whether the success of a scientific theory necessarily indicates it's true, as one might argue regarding the theory of evolution), the article "Pessimistic Induction and Two Fallacies" (PDF) by Prof. Juha Saatsi might be worth reading.

The Pessimistic Induction from falsity of past theories forms a perennial argument against scientific realism. This paper considers and rebuts two recent arguments -- due to Lewis (2001) and Lange (2002) -- to the conclusion that the argument from Pessimistic Induction (in its best known form) is actually fallacious. With this I want to re-establish the dignity of the Pessimistic Induction by calling to mind the basic objective of the argument, and hence restore the propriety of the realist program of responding to PMI by undermining one or another of its premises.


  1. Any evolutionist -- and any sensible person -- can simply concede Saatsi's rather narrow point here, and then offer the "typical realist response" that Saatsi himself is "personally very optimistic about" (1089). Namely, deny premise (3) by "by describing (usually via careful case studies) some theoretical elements solely responsible for the successfulness of past theories in a way that renders these theories continuous with otherwise incompatible current theories, and hence candidates of approximate truth in some suitable, restricted sense" (1089). That would salvage the realist project, as Saatsi himself admits.

    So, if one can make a good case for denying (3), then there's nothing wrong with the evolutionist (or anyone else) from using theory-success as a reliable test for truth. In short, defending the validity of the Pessimistic Meta-Induction doesn't ensure its soundness.

    Just to unpack a bit the point of denying (3). We need to separate the sheep from the goats when we say that past theories were false. This may be true, but they can all go wrong for different reasons, and to different degrees. Contemporary scientific realists are typically also fallibilists who hold that well-confirmed scientific theories most likely still leave out some truths, and include some falsehoods. Nevertheless, some theories are more *truthlike* than others, in that they represent reality more accurately. (Here, "truthlike" is what Saatsi is getting at above with his notion of "approximate truth" [which is quite different from the oxymoron 'relative truth,' BTW].) Popper's version of truthlikeness was fairly quantitative: a theory is more truthlike than another theory to the extent it has more truths and fewer falsehoods than other theories. Popper's proposal failed for various technical reasons, but in its place others have defended different accounts, most notable being similarity approaches in terms of possible worlds and distance measures on them.

    In any event, it seems to me that, intuitively, we understand the concept of one theory being closer to the truth than another, even if (strictly speaking) both theories are false (because one or more of their conjuncts are false). Newton's theory of motion is (strictly speaking) false, but it is still extremely successful within a particular range of middle-sized and middle-velocity objects. Most realists would argue that its central theses approximate the Einsteinian truth much more than some other rivals we could dream up.

    Perhaps an intuitive illustration will help. My phone rings. My theory is that my wife is calling to tell me to pick up bread at the store. Unbeknownst to me, the woman who is calling *was* my wife, and she is indeed calling to tell me to pick up bread, but unfortunately five minutes ago she divorced me at the local city office. Still, my theory as to why the phone rang was *pretty close* to the truth (indeed, I even picked out the right causal agent; I just misdescribed her). Contrast that with the theory that the phone is ringing because gamma rays have randomly flown in from Alpha Centauri and struck the invisible sensor plate on my phone. Not only is that theory false, it's *much further* from the truth than the "wife is calling" theory. So it would be false to say that all past false theories are equally untrue, and therefore equally unreliable as a guide to building contemporary theories. Likewise, although all obsolete scientific theories have been falsified, some were pretty close to the truth, and contemporary theories build on their (proximate) success.

    Full disclosure: I had Marc Lange (one of Saatsi's main targets in this article) for undergraduate philosophy of science.

  2. Thanks, Prof. Welty, for stopping by and leaving a comment. It's always a pleasure to read what you write. The only downside is you don't comment nearly enough as I wish you would! :-)

  3. Sorry to hear about your marital difficulties, Greg, but if your wife is the sort of woman who would divorce you and immediately afterwards tell you to pick up bread at the store then you may well be better off without her. ;)

  4. James,

    What can I say? She's very shrewd that way. It's that honest self-reliance on her part that keeps our family afloat :-)