There is a point about desert that Fred Feldman and Brad Skow get wrong in their SEP entry on desert. They distinguish between desert and entitlement (their terms; other terms would be ‘pre-institutional desert’ and ‘institutional desert’). There should be a threefold distinction here, not a twofold distinction.
If a convention or institution specifies that, if someone makes a certain mark, then they are to be rewarded or honored in a certain way, Feldman and Skow say there is entitlement, not desert (they want to reserve the term ‘desert’ for pre-institutional desert). However, it is both true that the winner of the footrace deserves (or is entitled to) the specified prize for winning and that the person who is fastest (but failed to win due to happenstance) deserves to have won. The former is institution-dependent in an obvious way, but so is the latter. Neither makes sense except relative to the conventions of a contest (and similarly for other conventions or institutions). We might call the first procedural or specified-reward institutional desert and the second substantive institutional desert with respect to what is meant to be (but might not always accurately be) measured by the making of the mark. (The basic point here is not original to me: Scanlon makes a version of this point in his 2013 article, “Giving Desert Its Due.”)
Both procedural and substantive institutional desert should be contrasted with natural or genuinely convention- and institution-independent desert. Like Frankie deserving not to be wronged by Genu. Or Genu deserving to be condemned for his wrongdoing. (However, matters are less clear when it comes to things like punishment and rectification: at least arguably, these are “deep” cultural artifacts and hence count as institutional forms of desert, though perhaps forms of desert with very strong normative backing.)
Institutional desert can have various degrees of positive or negative normative backing (or be neutral). Zippy wins the “murdering people” game by murdering the most people in the allotted time (thus deserving any specified prizes or honors), but perhaps Pinhead was the most murderous contestant and failed to win only because he had the bad luck of his favorite murder implement being broken. Each of these matters of persons having or getting something (of benefit to them) is valuable (impersonally valuable), I think, only in the specialized sense of “value in or relative to the practice or convention or institution.” This is independent of whether the institution itself is good, bad or neutral. Both Zippy getting the prize and Pinhead winning demand impersonal moral approval only from within the perspective of the practice (its procedures, its aims).
I take Sher’s implied explanation (of procedural institutional desert — he explains away rather than explaining substantive institutional desert) to appeal to the obligation of veracity (in that giving the prize includes expressing that the person has made the mark) and the obligation of fidelity (we have agreed to give you the prize or distinction if you make the specified mark). (See Sher, Ch. 7 Desert.) The step he does not make explicit is this: obligations being adhered to is an impersonally valuable thing. This, I think, is plausible (e.g., we should approve of and promote people not wronging each other).
On this interpretation of Sher, Zippy deserving his honors or rewards comes to it being genuinely valuable that he gets them (because his getting them is an instance of people meeting their veracity and fidelity obligations). In my view, by contrast, what matters for desert is that, for any arbitrary practice, relative to being seriously committing to it and participating in it (or the like), it is appropriate that one approve of the practice going as it is meant to. But this conditional appropriateness in approval is not of the right sort to generate things being genuinely valuable (only valuable relative to the practice, as indicated).
There are three good reasons weighing against Sher’s view. First, even if relevant parties are prima facie (or in a respect) obligated to give Zippy his rewards or honors (due to veracity, fidelity), what they are actually obligated to do (or obligated to do, all-in) is put a stop to the murdering-people contest! If so, then what they are doing in giving Zippy the honors and rewards is wrong — and so impersonally, genuinely bad. Second, Zippy getting his reward would still be a good thing (in the relevant sense of ‘good’) if, due to some kind of chance event or execution error — but not due to anyone meeting their obligation — he ended up with the prize. This suggests that Zippy getting the rewards and honors is good (in some sense) but not due to obligations being met. Third, as mentioned, Sher’s account has to explain away, rather than explain and justify, Pinhead deserving to win the murder game. But he does. These are, I think, pretty good reasons to prefer my impersonal-value-based approach to institutional desert to Sher’s similar approach.