Comments?
I'd love to hear.Date: 2003-04-30 09:07:24 PST
Fred wrote:
> “Rob� wrote:
> > Fred wrote:
> > >
But there couldn’t be a tooth-fairy and there is no sense in which
it could
> > > be true.
> >
> > If
there couldn’t be a tooth-fairy, then it’s false that
there is
> > one, not as you seem to want to say, neither
true nor false. Again,
> > what you’re saying about the
arbitrary breaks down.
>
> I’d agree with you that it
was false if some reason/evidence were presented
> for a
tooth-fairy which could be evaluated and then rejected. Even then
there
> is no way strictly to prove that a tooth-fairy doesn’t
exist. Where would you
> look? How would you go about it?
However that doesn’t mean it might or could
> exist and it is
not anyone’s responsibility to show that. Whoever wants to
>
assert the positive in this instance has the burden of proof.
I think you’re mixing up two different issues. One is whether something is possible or not. The other is a matter of burden of proof. I think people often have too simple an idea of who has or how to tell who has the burden of proof, but I don’t want to focus on that now. I shall assume that you are exactly right that the tooth-fairy-asserter has the burden of proof and that no one is called upon to show that the tooth-fairy does not exist, so long as no positive evidence for its existence is presented.
Returning to possibility, I think you’re trying to have things both ways. Again, if it is impossible that there is a tooth-fairy, then it is false that there is one. There’s no room for a claim of this kind to be impossible in a relevant sense but not false.
I think what leads to the problem is the (mis)definition of the possible (from Peikoff) as that for which there is (a) some, but (b) not much, evidence and (c) which does not contradict anything known. [Letters inserted for future reference.]
One of the problems with that definition is the inclusion of the phrase “but not much�. Consider what happens when you try to deny that something is possible (assert that it’s impossible) in that sense. You get something like this:
A claim is impossible if either (a) there is no evidence for it, or (b) it is not the case that there is only a small amount of evidence for it, or (c) it contradicts something known.
But that means – preposterously – that claims overwhelmingly supported by the evidence are impossible! A claim overwhelmingly supported qualifies as impossible under clause (b) and does not need to qualify under either of the other clauses, since they are linked only by ‘or’s.
So we do better to throw out clause (b). That leaves the impossible being something (a) there is no evidence for or (c) which contradicts something known. This still leads to problems. Consider the following two claims:
There are intelligent aliens within 10 light-years of earth.
There are intelligent aliens within 20 light-years of earth.
I take it that there is no evidence for either, so, by the definition, both are impossible. However, it is plain that the second is more probable than the first, since the second will be true in every case in which the first is, but includes eight times the volume. But surely, it is absurd to say that two claims can both be impossible but one more probable than the other!
I suggest that the definition needs to be trashed in its entirety. The best part of it is condition (c), but that can be derived, given a better definition of possibility, so it’s not needed, either.
What’s a better definition? We want something that is general – that covers different kinds of possibility. We also want it to smoothly handle “impossible�, since that’s just “not possible�, and “necessary�, since that’s equivalent to “not possibly not�.
Here’s my candidate: Something is possible if and only if it is not ruled out by a relevant set of constraints. What makes a set of constraints relevant is the kind of possibility in question. Thus,
· Something is logically possible if it can be stated consistently.
· Something is physically possible if it does not violate physical law.
· Something is physically possible under certain conditions if its occurrence (existence) under those condtions does not violate physical law.
· Something is legally possible if it does not violate any (statutory, common or constitutional) law.
· Something is epistemically possible if it is not incompatible with something we know.
And so on.
Notice that all of these admit of smooth extension to cover logical, legal, physical (etc.) impossibility and necessity.
What, now, does this have to do with arbitrary claims? On one level, not much. The fact that a claim is made without evidence does not automatically justify any claim about whether it’s possible or not. The tooth-fairy may be epistemically impossible because we know something about aerodynamics that makes it physically impossible for the small gossamer wings to support her body or physiologically impossible for the tiny brain in her tiny head to support human-level intelligence, but nevertheless she may be legally possible (no law against being a tooth-fairy) and logically possible, because she can be described without contradiction.
There are several general points to be
gleaned. One, a claim that something is possible is not a claim that
there is some evidence in its favor and hence does not have to be
denied simply because no evidence has been provided – nor does it
have to be denied in order to place the burden of proof on the
appropriate set of shoulders. Two, nobody gets a free ride:
possibility claims need justification, but so do impossibility and
necessity claims. Three, the correctness of a possibility claim may
depend on the kind of possibility in question.
Rob
_____
Rob
Bass
rhbass@gmail.com
http://oocities.com/amosapient
