The Subjectivity of Value

For some reason, the past few weeks have found me embroiled in more debates about value theory than I can remember in the past year. Accordingly, I figured I'd post something on the subject as a starting point for those debates, so that I don't have to repeat my entire view in every conversation.

The problems people seem to be wrestling with are twofold. First, people have been arguing that because value is subjective, we can't say anything about how much value something has except by saying how much someone values it. Second, people have been claiming that because there is no acceptable way to objectively measure utility, it is impossible to coherently make claims about utility which compare utility between one individual and another. I want to address both issues, but this post will only discuss the former; I'll deal with the latter another time [Update: here].

To begin, I want to make clear that I don't deny that utility is subjective. But what does it mean to say this? One uncontroversial, but relatively weak, way of interpreting this is to say that without people (or other valuers) to value things, nothing would have value. While this shouldn't offend anyone, it also doesn't tell us very much. It only means that value must be value to someone; the cake is not valuable, it is valuable to people. If there were a cake on a planet with nothing else on it, so that no sentient being would ever come across the cake or even know that it existed, it would seem odd to attribute any value to the cake.

But it's clear that this isn't what people mean when they say that value is subjective. It seems like what people are saying is that something becomes valuable because someone values it. George Reisman might seem to have embodied this sort of view when he said, "...the starting point both of goods-character and of the value of goods is within us--within human beings--and radiates outward from us to external things, establishing...goods-character and value..." It also might seem to have been present in Mises' thought, when he said, "Judgments of value are voluntaristic. They express feelings, tastes, or preferences of the individual who utters them. With regard to them there cannot be any question of truth and falsity. They are ultimate and not subject to any proof or evidence."

Mises continued, "What the theorem of the subjectivity of valuation means is that there is no standard available which would enable us to reject any ultimate judgment of value as wrong, false, or erroneous in the way we can reject an existential proposition as manifestly false. It is vain to argue about ultimate judgments of value as we argue about the truth or falsity of an existential proposition." It is on the back of statements like these that people argue that we are unable to posit value in anything apart from the value placed on it by some individual.

But this immediately leads to some problematic conclusions. For one, it means that if we cannot say that it would be good for the alcoholic to avoid taking another drink, or that the person in an abusive relationship would be better off leaving his partner, unless they believed that to be the case. If it is the nature of value that it does not exist except as it is placed on certain things by people, then it would be impossible to be mistaken about the value of something. Clearly the alcoholic values the drink, and the abused partner values his relationship, but there seems to be a sense in which we want to say that the alcoholic shouldn't value the drink, and the abused partner should leave.

The problem arises from the fact that we can talk about something's being "valuable" in two ways. One is positive: "I value X; X is valuable." The other is normative: "X would help me to achieve my ends; X is valuable." When my opponents talk about the subjectivity of value, they slip into the former kind of thinking. That is, they take the view that what is desired is the same as what is desirable. But it is my contention that what is desirable, in discussing the value of some object, is what ought to be desired, given the ends of the agent in question. So while I agree that ultimate ends can't be disputed (if you want to be a devout Christian, I can't tell you that you're wrong), I can dispute the means you choose for obtaining your ultimate ends.

This is played out in Mises' discussion when he says, "The characteristic mark of ultimate ends is that they depend entirely on each individual's personal and subjective judgment, which cannot be examined, measured, still less corrected by any other person. Each individual is the only and final arbiter in matters concerning his own satisfaction and happiness." He continues, "Means are judged and appreciated according to their ability to produce definite effects. While judgments of value are personal, subjective, and final, judgments about means are essentially inferences drawn from factual propositions concerning the power of the means in question to produce definite effects. About the power of a means to produce a definite effect there can be dissension and dispute between men. For the evaluation of ultimate ends there is no interpersonal standard available." Accepting this view of value, it seems clear that we can indeed make claims about how valuable something is as a means for achieving some end, even though we can't actually say that the end in question is valuable.

In passing, I want to preempt an obvious objection. In practice, I (and I think most people) do tend to assume that individuals have certain ultimate ends, and that under this assumption, I can largely ignore the issue of the subjectivity of ends. For example, I don't tell the alcoholic, "If you desire the sort of life I find that most people do, you shouldn't have that drink." I simply say, "You shouldn't have that drink." If I were to discover that the alcoholic actually thought that his purpose on the Earth was to explore the effects of alcoholism, producing knowledge for himself and the rest of humanity in the process, then I couldn't criticize his choice of drinking in the same way. But recognizing this possibility doesn't preclude me from assuming that they alcoholic is making a poor decision (in the absence of evidence to the contrary). That I could be wrong because of a faulty assumption, doesn't prove that the position outlined here is wrong.

One thing that I should address, but I won't, is the issue of justice as it relates to value. It's fully possible that someone could do something which would actually produce the effect of promoting her ends, but we would still want to say that there's something wrong with it. Mises seemed to think that justice could be completely explained in utilitarian, contractarian terms, saying, "The ultimate yardstick of justice is conduciveness to the preservation of social cooperation. Conduct suited to preserve social cooperation is just, conduct detrimental to the preservation of society is unjust. There cannot be any question of organizing society according to the postulates of an arbitrary preconceived idea of justice. The problem is to organize society for the best possible realization of those ends which men want to attain social cooperation. Social utility is the only standard of justice. It is the sole guide of legislation." I completely disagree with this assertion, and I am supported in this by a considerable tradition in philosophy. But how this relates to the discussion of value is beyond the scope of this post. I think that what I've said so far is good enough for my purposes.