Constitutive Instrumentality: a response to Lazar

Seth Lazar's forthcoming paper, 'Moral Status and Agent-Centred Options' contains some interesting objections to my 'Value Receptacles' paper.  Here's the first:

Chappell’s treatment of the separateness of persons has several weaknesses. First, what does it mean to value aggregate well-being non-instrumentally, while valuing the well-being of individuals only instrumentally? The view seems a straw man. Aggregate well-being is composed of the well-being of individuals. If aggregate well-being is a noninstrumental value, then individual well-being is a non-instrumental value, since aggregate well-being just is all the individual well-being taken together. Treating different people’s well-being as totally fungible is a conceptual mistake, hence not a charitable interpretation of the separateness of persons objection. 

This is incorrect.  An important upshot of my paper was that we need to recognize two very different kinds of instrumentality.  The most familiar kind is when one thing is a causal means to another, as (e.g.) in the case of money being useful for buying desired objects.  But we should also recognize the possibility of one thing being a constitutive means to another.  This is what's going on in the case of someone who most fundamentally cares about a kind of good in the aggregate, rather than having any basic, non-derivative concern for the particular instances that make up the aggregate.  Their concern for the instances is wholly derivative of their concern for the whole, in a way that makes them entirely indifferent to internal variation (e.g. in the identities of the instances) as long as it doesn't affect the overall value of the whole.

This is plainly conceptually coherent.  Indeed, it is the correct attitude to take towards goods that really are (fitting to regard as) fungible.  In my paper, I gave the example of being indifferent between a left-foot massage and a right-foot massage.  Either massage would equally well serve to instantiate one and the same final value (rather than being options that realize numerically distinct but equally weighty values).

Translated into the more precise lingo of fitting attitudes: It is fitting to regard certain kinds of (qualitatively similar) pleasures as fungible, i.e. to have a single basic desire for such pleasures in general, rather than to (bizarrely) have distinct basic desires for each possible such pleasure.  To answer Lazar's question, then: to have this sort of desire profile is what I mean by valuing the aggregate of a good "non-instrumentally, while valuing [its individual instances] only instrumentally."

There is a real psychological difference between whether the aggregate or the particulars are the focus of one's basic desire(s) for a good.  (As explained in my paper, this psychological difference is reflected in one's emotional responses to trade-offs involving particular goods: for example, whether one reacts with indifference or ambivalence to a forced choice between two equally valuable instances.)  Given that there are two (genuinely distinct) possible psychological responses in this vicinity, we can ask which, in any given case, is the fitting or correct response.  Should we care more fundamentally about the aggregate or about the instances that make it up?  Or, in other words, is the good in question fungible?  I think the correct answer to the question varies depending on the nature of the good in question.  Some (e.g. pleasures of a certain sort) plausibly are fungible.  Other goods (e.g. individual wellbeing) plausibly aren't.

These are substantive, coherent (albeit neglected) questions in normative ethics.  You can't just dismiss the very idea of fungibility as a "conceptual mistake"!

[For Lazar's second objection, see my follow-up post: 'Acts, Attitudes, and the Separateness of Persons'.]