Tuesday, October 02, 2007

Time and Value by Oscar Horta

Time and Value
Oscar Horta
Faculty of Philosophy
University of Santiago de Compostela
Praza de Mazarelos
15782 Santiago de Compostela – Spain–

Can the value of some benefit or harm depend on the time of its occurrence (other things being the same)? Those who deny that this can be so assume the idea of the temporal neutrality of value, which can be expressed as follows:

(N) Other things being equal, the magnitude of the value which a certain benefit or harm has for its receptor is something in itself independent from the time at which it occurs.

Or else:

(›e)(›i) (e is a certain benefit or harm, and i is the one who is affected by e).
(›w)(›w’) (w is a world in which i is affected by e at a certain moment t and w’ is the world nearest to w in which e affects i not at t but at another certain moment t’)
The value of which i is a recipient in w = the value of which i is a recipient in w’
The value of e at t = the value of e at t’

This idea seems at first sight intuitively cogent. However, Derek Parfit has shown that our attitudes do not match it. He has pointed out that we have what he describes as a “bias towards the near”. We have a strong tendency to have higher concerns regarding the experiences that we may have closer in time to the moment at which we are. We display a preference for postponing negative experiences even when this will end up being worse for us in the long term, and we have an inverse tendency regarding positive experiences. On the other hand, as Michael Slote and Frances Kamm have pointed out, we also have a preference for what comes last (or, in other words, for inclines over declines). A story with a sad beginning and a fortunate ending seems to be happier than another one with a joyful start and a depressing conclusion, even if in the first one the period of joy is shorter. This seems paradoxical, since these two tendencies are at odds not only with temporal neutrality, but also with each other.
Besides, there is an even deeper way in which temporal neutrality is practically rejected, which has been also described by Parfit: we have what he calls a “bias towards the future”. We care more about the experiences which we will have in the future than about those we have already gone through in the past. This means that our appraisals, rather than being temporally neutral, do vary in accordance to the moment at which we are with respect to the benefits and harms we may experience. Explanations of this asymmetry in terms of the direction of causality or some physical arrow of time would be satisfactory if they were grounded on a sound ontology of time which gave the future some sort of priority over the past. However, none of the most accepted theories concerning the metaphysics of time grant this. Two of them are neutral regarding this: according to presentism, only the present exists; and eternalism maintains that all events occurring at some temporal point exist (so not only the present, but also the past and the future exist). For its part, the remaining leading theories, the growing block theory and the shrinking tree theory, actually defend an asymmetry towards the past, since according to them both the past and the present, but not the future, exist. Therefore, none of these theories provide the sort of the ontology of time which would be needed in order to support partiality towards the future. Such a theory should be described as a “futurist” or “shrinking block” one which considered that the present and the future, but not the past, exist. However, this theory has not received any support in the way in which the aforementioned have. Therefore, it seems that the appeal to the ontology of time not only does not provide a justification for the preference towards the future: it actually makes things more problematic for it. In the best scenario (that of presentism and eternalism) it dismisses it through an acceptance of temporal neutrality. And if we turn to the growing block and the shrinking tree theory, we can see that they actually contradict it by embracing a past-directed view, subsequently leading to a double and conflicting temporal asymmetry.
Things do not get any better to future-directedness if we consider what may the theories of the diachronic persistence of objects say on the matter. Three-dimensionalism can accomodote either this view or neutrality, whereas four-dimensionalism actually appears to support neutrality. Furthermore, a similar conclusion can be reached in the case of onto-semantic theories of time: neither the tensed nor the tenseless theory of time support future-directedness. If anything, it seems that the tenseless one is actually much more favorable to neutrality.
The peculiar conclusion to draw from here is that we value what we have a tendency to consider nonexistent over that which we are inclined to consider existent. This contradiction goes together with and is strongly reinforced –if not actually caused–by two contradictory facts concerning the way in which we cognitively relate to the past and future:
(i) We can know the past but not the future. The past, unlike the future, seems to be fixed and available to being recorded (rather than predicted). This clearly supports the ontological asymmetry, although it is not in itself a proof of it.
(ii) As Ingmar Persson has defended, we lack the capacity to make motivational representations of our past experiences which might resemble the ones we can make of our future ones (as a result of which we can have feelings such as fear concerning what may happen to us in the future but not towards what we have already passed through –when we do is because we fear that it happens again–). This supports the asymmetry concerning temporal value, albeit, again, it is not a sufficient proof of its appropriateness.
These conclusions are particularly puzzling since, even if we actually accept a priori the thesis of temporal neutrality, it seems hardly possible for us to get rid of the aforementioned preferences which deny it. But they are not only problematic on theoretical grounds, but also, obviously, on practical ones. This may not be so in the case of the preference for the future, given that, since we cannot change the past, our future-directed prudential concerns seem to be appropriate (although this is not necessarily the case of our future-directed ability to represent our experiences). But the preferences for the near and for the last often lead us to face significant practical dilemmas.