A Tension
- Posted by Andrew Bailey on Monday, December 11, 2006 at 11:01 AM
| 
2 Comments |
Good, careful philosophical work often yields quite complex results. The detail found in these results can get in the way of one's ability to see their truth, however, and this is unfortunate. Furthermore, there are features in the analytic philosopher's "philosophy by counterexample" toolbox that almost inevitably bring out this tension.
Philosophers will often begin their work by stating a principle that seems true on reflection, a principle that has a sort of glossy sheen to it, a luminous self-evidence. The object is then to test this principle with counterexamples. Some counterexamples suggest that principle as first stated was false, so a refinement is needed. If the philosopher in question is particularly clever, the principle can be clarified, refined, and caveated to do all the work (as a premise in an argument, say) that it was tasked to do while still avoiding the counterexamples in question.
It's struck me that something rather important might get lost in this process, though: the luminous self-evidence that inspired us to affirm the principle in the first place.
Here's an example, one dealing with the analysis of a concept. Suppose we were wondering about the soundness of an argument for the compatibility of free will and determinism. And a crucial premise of our argument was an analysis of free will, something along the lines of "Someone is free with respect to the fact that p if and only if x is able to make it false that p" (suppose we needed the biconditional for the argument to be sound). There are a lot of things to unpack in this analysis, notably "able" and "make it false." We might describe ability in terms of counterfactuals. But such an account might face the problem of finkish dispositions, so a few extra conditions would have to be added to avoid those counterexamples. Similarly, with "make it false" we might have a causal reading of the locution, or perhaps one involving counterfactuals, or a little bit of both. Counterexamples will threaten all such formulations, no doubt, so a few nips and tucks will be necessary here and there to ensure that our analysis succeeds.
If we are thinking clearly and carefully, our analysis of "someone is free with respect to the fact that p" will become very complicated very fast. And there's something I wonder at this point in the dialectic: is there any reason anymore to think that the analysis is true? Suppose that it seemed obviously correct to us at first that someone was "free with respect to p" just in case she could make p false (a reasonable starting point, I think). After unpacking and clarifying the concepts at play here, however, it may not be nearly so obvious.
If our confidence in the refined analysis is weaker than our confidence in its original statement, it may not have sufficient epistemic "oomph" to serve as a premise in an argument. Now, there's an obvious sense in which the refined analysis does have justification, and lots of it. It has been tried by fire, and subject to intense scrutiny. That it doesn't seem subject to counterexample even after being under the microscope seems to count in its favor. This much is true, but I suspect many want more for their premises. We do not merely want them to not seem false, we want them to seem true. And a premise that is so long and complicated such that no one can hold all parts of the premise before her mind at the same time is one that may lack this feature. Put differently, it's hard to have intuitions about propositions we don't understand.
It seems we are caught between the horns of a dilemma, then. Either our analysis seems true at first glance but is subject to counterexamples on further reflection, or it is refined to the point that it is no longer clear whether it is true anymore, even though it avoids the counterexamples. Neither horn of this dilemma is particularly nice.
2 Comments:
the metaphysician at 10:38 PM said... There's also the "one's man's modus ponens is another man's modus tollens" problem. :)
Noumena at 2:59 PM said... I don't think obviousness has any particularly high epistemic value. Or, if it does, it really shouldn't. As anyone who knows a little bit about Critical Theory (or even just the history of philosophy) will tell you, the last 3,000 years of Western thought has been pretty consistently based on substantive assumptions that `everyone' (often referring to the white guys with enough luxury time to sit around thinking all day) take to be obvious until one day no-one thinks they're really true at all.
Intuitions are a good place to start because they're the only place to start. But how are we going to refine and revise our ideas about what's the case without changing them in any way?
