Wednesday, October 24, 2012

Consequences in Time

M. Oreste Fiocco has a curious paper, 'Consequentialism and the World in Time', forthcoming in Ratio.  It contains a number of arguments that strike me as very confused.

One central argument may be characterized as follows:
(1) Consequentialism is committed to "temporal homogeneity" (i.e. Eternalism)
(2) The truth or falsity of consequentialism is non-contingent
(3) Temporal homogeneity is contingent
So, Consequentialism is false.

I'm unsure of the third premise -- the alternative of a "moving now" is arguably incoherent.  But let's focus on the first, which I think is clearly false.  Oreste Fiocco offers two arguments for the first premise, i.e. that Consequentialism -- by decreeing that the deontic status of an act depends upon future consequences -- is incompatible with temporal heterogeneity (whether presentism, growing block theory, etc.).

(i) First, he objects to the idea that future events can "change the past", by making some act have a property (e.g. rightness) that it did not have while present.  This seems to be an example of what I've previously called a "derivative objection", in that it's really an objection to the natural property of having such-and-such consequences, rather than to the normative theory that the property of rightness holds in virtue of that natural property.  After all, even non-consequentialists should think that there is such a natural property, with just the feature that MOF objects to.  So this isn't an objection to consequentialism per se.

It also seems to be clearly unobjectionable in its own right.  There's nothing mysterious about "changing the past" in relational or extrinsic respects.  Whenever someone asserts a future contingent ("I will wake up tomorrow"), the future will end up determining whether that assertion has the property of being true.  That's because the truth value of the assertion is not intrinsic to the moment of assertion: it depends upon subsequent events.  What I call "forward-looking properties" like having such-and-such consequences and being objectively right are similarly relational, in that they depend upon the intrinsic natures of other times in addition to the time of action.  So there's no mystery here.  It's not like we're changing the distribution of matter that obtains at some past moment, or other such temporally intrinsic features!

(ii) MOF's second objection is that if the future is unsettled, then (consequentialism implies) at no time can an act have a settled deontic status, since it remains an open possibility that some overwhelmingly good or bad effect is yet to come about.  But this rests on a mistake: it's possible that two diverging possible timelines might at some future point re-converge, such that events at all subsequent times are unaffected by which of the previously diverging paths was actualised (i.e., whether or not you performed some act).  Weaker still, two diverging timelines might reach a point after which they are evaluatively convergent (even if still physically divergent), say because a cosmological "big crunch" or "big freeze" has rendered life physically impossible.

On the other hand, if it turns out that life continues indefinitely, and the morally relevant consequences of choosing to perform some act A (or not) are truly never settled, then it just seems accurate to describe this as a world in which the objective rightness of A is never settled. (Again, no-one should think it objectionable that there are forward-looking natural properties of actions, e.g. maximising happiness, that have this feature.)  So this doesn't seem like a problem, either.

(iii) As a more basic objection to MOF's paper, it's worth flagging that not all Consequentialists appeal to future facts.  Expected-Value (or "Subjective") Consequentialists instead hold that the deontic status of an act depends on the agent's evidence regarding the probabilities of various outcomes at the time of acting.  This is temporally intrinsic, so none of MOF's arguments can even get off the ground against this form of the view.

Evaluating possibilia: There's one final argument from the paper that I want to address here.  MOF claims that "the smallest amount of real, actual goodness is better than any amount of merely possible good". So if (as eternalists claim) the actual future is real, it follows that our actual choices are always right: they yield actual value, whereas alternative choices yield merely possible value.  But it makes a mockery of morality for acts to be right simply in virtue of having been actually performed.

Of course, this argument rests on a conceptual confusion.  This is seen most easily by noting that "good" just means "desirable", and an "amount" of value is just another way of talking about the degree to which it is desirable (i.e. the strength of desire that's warranted).  Now it's just clearly false that an actual state of affairs must be more desirable than its alternatives, simply in virtue of its actuality.  Special cases of partiality aside, modal status makes no difference to the desirability of a state of affairs.  If it's a good state of affairs, then that's just to say that it's fitting to desire [that it be actual].  Whether or not the state of affairs actually obtains is generally not relevant to how desirable it is.  Hence, modal status is generally not relevant to how good the state of affairs is.  It's analytic that a better (more desirable) state of affairs has a greater amount of value than a worse (less desirable) state of affairs, even if the latter is actual and the former is not.  Any assertion to the contrary reveals a lack of understanding of what it means for one state of affairs to be of greater value than another.

There is of course a trivial sense in which "actual value" is worth more than "merely possible value", namely, that we should want a valuable state of affairs to become actualised rather than remaining merely possible.  But when assessing actions for whether they "maximize value", this is not the relevant sense.  We are not asking which action actually has the property of possessing the most realized (as opposed to merely possible) value.  That would collapse into the question of which action is actually performed, which is of course not the relevant question for practical deliberation!  No, we are simply wondering which action leads to the most desirable outcome -- the outcome we should most desire to be actual, not just whatever outcome is actual.


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.)