Sunday, March 13, 2016

Consequentialism and Underdetermined Conditionals

A far future conditional is a claim about how the world be in the far future conditional on some contemporary event. 

Example: If I eat cheerios for breakfast, humanity will survive for at least one thousand years.

Far-future conditionals are very difficult to assess, and not only because they concern the far future. It is very hard to trace the chain of events that might connect a given action with its consequences. Even if we knew what the far future looked like, it would be very difficult to trace its development back to actions taken at the present. Though present actions can surely have significant ramifications in the far future, those ramifications are seldom easy to spot. 

Far future conditionals are probably also subject to underdetermination. The future effects of a contemporary event might depend on more than a crude characterization of that event. The details may matter in unpredictable ways. There might be a thousand different ways that I could eat cheerios -- different bowls I could select, different quantities of cheerios to eat, different patterns of cheerio spoonings to transport cheerios from bowl ot mouth -- that would all contribute in different ways to radically different futures. 

In this case, does it make sense to say that the future turned out any particular way because I ate cheerios (as opposed to, because I ate cheerios in this exact way?). Had I not eaten cheerios, would it make any sense to ask what would have happened in the far future had I instead chosen to eat cheerios?

Suppose not. Suppose that very few things we could do, crudely characterized, would have determinate effects in the far future. This creates a problem for consequentialism.

Consequentialists think we are obligated to do whatever of our available actions have the best consequences. Presumably, we can only be obligated to undertake those of our actions that it is in our power to undertake. Although hyperspecific actions might have determinate consequences, we can't undertake them. I can't control the precise movement of my body through the air. At some point, my subconscious brain takes over.

We can only choose to take actions that are fairly crudely characterized. If such actions do not have determinate effects in the far future, then their effects are indeterminate. If their effects are indeterminate, it is likely that it will be indeterminate which actions have the best effects. We might qualify that consequentialists should undertake those actions with the best determinate effects, but it seems like it is a mistake to simply ignore the indeterminate effects. However, if it is indeterminate which actions have the best effects, then it is indeterminate what we should do.

How problematic this is depends on how frequently our actions have massively indeterminate consequences. I suspect that the vast majority of actions do, because they seem to set off small chains of events, and because so many things we do are sensitive to very precise conditions. If you put away the cheerios a little further back in the shelf, you might take half a second more to reach for them next time. If it takes you half a second more to reach for them, you might be half a second behind in the day, which might cause you to stop a red light you otherwise would have made, and make you a minute late to work....

If this is right, then it is believable that in the vast majority of cases, there is no fact of the matter what we should do.

No comments:

Post a Comment