Some Alternative Conceptions of Fairness

There was a careful bit of word choice in my last post that could have easily been missed, so I want to shine a spotlight on it.

It wrote that fairness is an outcome or procedure that people generally have many and strong reasons to cause others to like.

I did not write that fairness is an outcome or procedure that everybody (or everybody in a particular culture) likes. Nor did I write that it is an outcome or procedure that the agent likes or that it maximizes utility (provides the most happiness, preference satisfaction, desire fulfillment, endorphin production, etc.).

I did not identify fairness with what everybody likes for a couple of reasons.

First, there is almost certainly no outcome or procedure that everybody likes. Nor is it the case that finding one person that dislikes an outcome outcome or procedure proves that it is unfair. Never do we here the argument, "Jimmy did not like the outcome; therefore, by this fact alone, we have proof that it is unfair." This is simply not what we are talking about when we talk about fairness.

Second, what people actually like or dislike (or think they like or dislike) may be substantially different from what they have reason to like or dislike. Their likes might be grounded on a false premise - an environment in which they were taught to dislike X "because God disapproves of X". Or perhaps a cultural approval or disapproval fit a particular time and a particular set of circumstances that no longer apply. For example, a community facing frequent famines might hold that it is "fair" to give women of child-bearing age an extra share. This reason to promote a particular distribution would disappear when the famines disappeared.

The objection to equating fairness with what the speaker likes springs from the fact that language is a public and interpersonal phenomenon. We invent terms for those things it is worthwhile to talk about. "I do not like X" is a useful piece of information - a piece of data that may be added to the data pool. However, when people get together to talk about what is fair or unfair it is foolish to think that the thing that concerns them and that is worth all this time and attendion is what a particular person - the speaker - does or does not like.

On the other hand, it is quite useful to talk about what people generally have many and strong reasons to cause people to like. Indeed, do not like fairness we are not dedicating our conversation to, "What you like." for a public conversation, the useful thing to talk about is what people generally have reason to cause others to like.

It simply does not follow from, "I don't like X" that you and everybody else - or, more accurately - that people generally have reason to jump up and do something about X that calling it "unfair" would imply. Useful or not, it does not carry those implications and is not what we are talking about when we talk about fairness.

On the other hand, it follows by definition that a form of distribution or a procedure that people generally have many and strong reasons to promote a disapproval of is something that people generally ought to promote a disapproval of. Flagging something as "unfair" in this sense tells people, "Here is a distribution or a procedure that you people out there - people other than me (or in addition to me) have many and strong reasons to respond to with condemnation." Of course, when this is not true, then the claim that the act was "unfair" can be challenged and, potentially, proved false.

Technically, and more accurately, calling something "unfair" means that there are reasons to condemn it. However, as a matter of fact, the only reasons that exist are reasons that people have. Consequently, only claims about reasons that people have are actually relevant to the fairness or unfairness that exists. Reasons that do not exist are not relevant.

Finally, I did not say that fairness is found in the rules or procedures that maximize utility, in part, because nobody is actually or even potentially concerned solely with maximizing utility. The motives that people have for promoting desires for or disapprovals of certain types of distributions or procedures are the various interests that people have. It is found in their aversion to pain, their concern for their children, their food preferences, their enjoyment of various types of activites, their love for their spouse, their hatred of the next door neighbor with a barking dog. There may be a desire to maximize utility somewhere in this soup, but it is only one ingredient among many constantly being pressured by all of these other concerns.

When is the last time you looked at a menu at a restaurant and made your selection based solely on which choice will maximize utility? Or went to the movie that would maximize overall utility? Did you choose your spouse based solely on the principle of utility maximization? And when you have sex, is it solely because having sex with that person at that time maximizes overall social utility and for no other reason? Do you buy Christmas gifts based solely on maximizing overall social utility?

To demand that people act only this one interest - the interest in maximizing utility - is folly. It cannot happen. And if it is the case that X cannot be done, then it cannot be the case that X should be done.

The second objection to this claim is that there is no reason - no physical manifestation in the world - that gives the desire to maximize utility priority over all others. It is one desire among many. It has no intrinsic merit that gives it a special claim for consideration that other interests lack.

The value of a desire to maximize utility is not determined by its intrinsic merit. It is determined by the degree to which such a desire tends to fulfill other desires. This gives others reason to promote such an interest. On this measure, a desire to maximize utility certainly has a lot to recommend it.

Using the same measure, we can look at other reasons people may have for promoting or inhibiting certain outcomes or procedures and discover that they are not very good reasons. Unlike the desire to maximize utility, some of these reasons are themselves reasons we have reason to condemn or inhibit. Using this standard, we can identify some potential reasons for promoting a particular outcome or procedure as poor reasons and discount them appropriately. Again, this discounting does not depend on their intrinsic merit, but by their tendency to thwart other desires. Intrinsic merit or demerit does not exist.

So, I did not write that fairness was a system of distribution or a procedure that everybody likes, or that the speaker likes, or that maximizes utility. The formulation I used was that fairness identifies forms of distribution or procedures that people generally have many and strong reasons to cause others to like.

To call a procedure "fair" - as in a "fair trial" - is to call attention to those features that give people generally reason to promote approval. A fair trial means that a person is spared punishment if they did nothing wrong, and that we punish only those for whom there are many and strong reasons to punish. It is a procedure that reveals relevant facts - and keeps irrelevant facts that might prejudice a jury and produce an unjust outcome hidden.

This, then, is how to go about determining whether a distribution or a procedure is fair. Determine whether people generally have many and strong reasons to promote a liking of that outcome or distribution. It is not enough to like it. It is not relevant that everybody likes it if their liking is grounded on false beliefs or cultural traditions meant for different circumstances. We are not looking for an outcome or procedure that maximizes utility. We are, instead, for an outcome or procedure that people generally have many and strong reasons to get others to like.