Tuesday, May 10, 2005

Future Truths

FQI has an interesting post up on whether future contingent propositions (e.g. "tomorrow I will rob a bank") have truth values. He argues that they do, and I agree, though I don't think his arguments work.

For example, he responds to the "correspondence argument" by saying:
Truth as correspondence claims that in order for a statement to be true, it must correspond with the truth-maker it describes. Moreover, since future contingent propositions correspond to future events, it follows that [statements can be true even if corresponding to non-present existents.]

But his opponent is surely not going to grant that there is (yet) any determinate future event for the statement to correspond to. The future hasn't been decided yet, therefore neither have the truth values of future-directed statements. FQI hasn't satisfactorily responded to this argument. (As a determinist, I deny that "the future hasn't been decided yet", but I don't think FQI wants to grant this. As such, I think he is inconsistent in holding that future contingents already have truth values.)

Next, FQI presents a couple of positive arguments in favour of his position:
(a) The same facts that make present and past facts true, makes future facts true...

(b) If future contingents are not true or false because they do not presently exist, then the same follows for past events.

But (a) faces the same problem as above, in that his opponent will simply say that there are no determinate future facts (yet). What makes past- and present-directed statements true is that they correspond to determinate facts about how things are (and have been). There are no such determinate facts about the future (or so it is argued). This also shows how to respond to (b), since there are determinate facts about the past, which statements may correspond to.

Lastly, and most interesting, FQI argues that if a future-directed proposition P has no truth value, then this leads to the absurd result that "P or not-P" is not true (though not false either). Unfortunately, FQI then invalidly infers that the truth of both disjuncts follows from the truth of neither, and claims that this result is "ridiculous". Dismissing this illogic, we're left guessing as to whether there are any real reasons to consider the result absurd.

But let me propose one: we can know that "P or not-P", for any future-directed statement P. And if we can know it, then it must be true. Then, according to classical logic, either P is true or not-P is true (i.e. P is false). Thus P has a truth value after all.

I don't think this is a knock-down argument though, because I think our opponent should reject classical logic when dealing with future-directed statements. After all, it is a determinate fact that "P or not-P", even if it is neither determinate that P, nor determinate that not-P. So classical logic is mistaken: the truth of a disjunction does not guarantee the truth of one or other disjunct.

FQI concludes:
Does [the truth of future contingents] mean that fatalism is true? No, because the truths about the future are contingent truths, not necessary ones. To say there are facts about the future is merely to assert a tautology: "What will be, will be." Affirming that future contingents are either true or false is not a problem. This becomes a problem only if future contingents cease to be contingent and become necessarily true or necessarily false.

I'm not sure we can escape fatalism so easily. After all, we're not just saying "what will be, will be", but adding real content, e.g. "that tomorrow I will rob a bank, will be." That seems rather more serious. Even if this is merely 'contingent' in that it's false in other possible worlds, there's no avoiding the fact that it is [ex hypothesi] true in this world, and it is not in our power to change this. As I wrote in my free will essay:
Suppose the proposition Q: ‘Tomorrow, S will perform act A’ is true. Then we cannot (categorically) do anything which would entail the falsity of Q, because it is not false, and we surely cannot (categorically) bring about a logically inconsistent state of affairs.

I go on to suggest that this overly restrictive result gives us good reason to reject the categorical analysis of 'could', and adopt the compatibilist's conditional analysis instead.

But I digress. My point is that if future-directed propositions have truth values already, then the future facts must already be determined. That is, it commits you to a form of fatalism. Though, as I argue here, it's not a particularly serious form (and shouldn't be confused with notions of 'destiny' which deny us causal powers. Follow the link for details). Nonetheless, I'm not sure that FQI would want to accept this result.

P.S. Funnily enough, that logic post I linked to in my previous entry also has a relevant section on 'future truth', wherein I suggest an alternative interpretation:
A better way to understand future truths might be to say that as a matter of convention, we will call a statement about the future 'true' if (in retrospect) it turns out to be accurate. The truth itself is in no way embedded in the present, however.

That would allow us to avoid fatalism.


Post a Comment

Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)