Analysis Question

I’m hunting for a certain type of case and could use some help. I’m looking for a biconditional with something like the form

    For all x, x is F iff blah blah x blah blah

that is BOTH correct as an analysis of Fness AND plausibly true for other reasons. In other words, this would need to be a biconditional of which we think: (i) it’s a correct analysis, and (ii) were it not a correct analysis, it’d still be true.

Any ideas?


11 Responses to Analysis Question

  1. Jonathan says:

    Hmm. It’s not obvious what it means for condition (ii) to be met. Correct analyses are usually thought to be necessary truths; it’s not clear what’s intended when you say that you’re looking something that would be true even if it weren’t a correct analysis.

    (Given S4 and the necessity of analytic truths, analytic truths are necessarily necessary, so counterfactuals with antecedents denying the necessity of analytic truths are counterpossibles, which are usually thought to be trivially true.)

    Can you say a bit more about what you’re looking for? Why aren’t all correct analyses examples?

  2. dtlocke says:

    Hi Jon. I don’t think Steve intended the subjunctive. (ii) should probably read: ‘even if it is not a correct analysis, we still have reason to think it is true’. Of course, philosophers also have trouble making sense of indicative conditionals with necessarily false antecedents. But most people seem to do it just fine.

  3. Jonathan says:

    If that’s really what’s intended, I don’t see why most analyses wouldn’t be like this.

    x is a square iff x is an equalateral, equiangular quadralateral.

    That looks like a good analysis to me, but even if it’s not, I’m pretty sure it’s at least true.

  4. Steve C. says:

    Hi Jonathan, thanks for the replies. I posted quickly and should have framed (ii) as an indicative, as Dustin said.

    Your response is interesting. I suppose it shouldn’t be surprising that most of us will continue to have the intuition that a biconditional like the one you’ve given is true, even when supposing that it’s not a correct analysis. Can you say why you think it’s true, even if not a correct analysis? (I worry that the lingering intuition might be driven by the conviction that it is the right analysis.)

  5. Andrew says:

    Hi Steve,

    What about the epsilon-delta definition of continuity? It’s standardly taken to be an analysis of continuity these days, but before it was, people would presumably have thought that all continuous functions satisfied the condition. (I’m sure there are better examples, I just had ‘analysis’, of the other sort, going through my mind when I read the post.)

  6. Jonathan says:

    Why would I think the square biconditional is true, even if I didn’t think it was a proper analysis? I guess it depends on why I reject it as an analysis. One reason someone might do that is if she didn’t believe in analyses; she thought, maybe for Quinean reasons, that there is no such thing as a correct analysis. Such a person would still, presumably, think that squares are equalatral equiangular quadralaterals.

    Heck, you could believe the generalization on empirical grounds if you really wanted to. (I think Mill had a wacky super-empiricist view like that.) Go count and measure the sides and angles of a whole bunch of squares; as you find more and more correlations between the right-hand side and the left-hand side of that biconditional, you gain inductive empirical support for the generalization. (The same way you know that all emeralds are green.) You could definitely do this without thinking it’s a correct analysis.

    Another way someone might believe the biconditional universal is by testimony. Maybe most school children are like this. If you don’t really understand what it means to be an analysis, or the difference between an analysis and a merely necessarily true generalization, you might believe this claim is the latter, but not believe that it’s the former.

  7. Steve C. says:

    Andrew, thanks for the example. I had analyses of folk concepts in mind, but your example is suggestive. Perhaps it is natural for analysis-acceptance to come in two stages: first, coming to think that the biconditional is true (while being agnostic as to whether it’s a good analysis), and then coming to think it is a good analysis. And if so, I suppose that conviction of its truth may have a certain degree of independence from our conviction of its status qua analysis. (Perhaps this is what Jonathan was getting at.)

  8. Andrew says:

    Continuity of a line isn’t a folk concept!? What do you mean by a folk concept then? I’m confused.

  9. Steve C. says:

    Jonathan, thanks for clarifying. That’s very helpful.

  10. Alex S. says:

    Hi Steve,

    Here’s another not-so-folksy example: the Church-Turing thesis (roughly, the claim that the computable functions are all and only those computable by a Turing machine) may or may not be an analysis. But either way, many people think we have good reason to think it’s necessarily true.

    Here’s a slightly more folksy example: x has n kilograms mass iff x does such-and-such in the appropriate circumstances and the laws are such-and-such. Again, it’s controversial whether that’s an analysis–whether mass just is some dispositional property–but arguably you can spell it out in such a way that it comes out necessarily true (keep adding conjuncts for different law-circumstance-behavior patterns across worlds).

  11. Steve C. says:

    Thanks for the examples, Alex.

%d bloggers like this: