Peter Railton's Naturalistic Reductionism - Part 4 - Value Realism

On his way to establishing moral realism, Peter Railton first worked to establish value realism. ("Moral Realism", The Philosophical Review, Vol. 95, No. 2 (Apr., 1986), pp 163-207.)

Value realism has two requirements:

(1) independence: it exists and has certain determinate features independent of whether we think it exists or has those features, independent, even, of whether we have good reason to think this;

(2) feedback: it is such-and we are such-that we are able to interact with it, and this interaction exerts the relevant sort of shaping influence or control upon our perceptions, thought, and action.

The first matter relates to a distinction that I have been making that confuses a lot of debate about objective values.

It is a distinction between whether value claims can be true in a world where beliefs and desires did not exist, versus whether value claims can be true independent of whether or not people believe or want them to be true.

Railton uses the phrase, "whether we think it exists or has those features" when he discusses the independence of value. Thus, he is using the second type of objectivity described above. This is the same type of objectivity that scientists use, and allows for values to be as real as anything in science.

With this in mind, Railton begins the search for real values in the agent's subjective interests:

Consider first the notion of someone's subjective interests - his wants or desires, conscious or unconscious. Subjective interest can be seen as a secondary quality, akin to taste. For me to take a subjective interest in something is to say that it has a positive valence for me, that is, that in ordinary circumstances it excites a positive attitude or inclination (not necessarily conscious) in me. 

Railton does not like this idea because desire can be misinformed. Consequently, rather than talking about an agent's actual desires, he instead wants to place value in an agent's fully informed and rational self could choose his desires. More specifically, he imagines a fully informed an rational self deciding what desires he would want the less informed and less perfectly rational self to have.

Whenever a philosopher begins to talk about informed desires, I question whether that philosopher has carefully considered the distinction between desires-as-ends and desires-as-means.

Desires-as-means are bundles of desires-as-ends and beliefs. The beliefs can be true of false, so rationality and true and complete beliefs are relevant.

However, as Railton himself said earlier in the article, rationality is not applicable to ultimate ends.

Consider Railton's case of a vacationer.

Railton describes a vacationer, Lonnie, suffering from dehydration. Lonnie wants a glass of milk. However, the hard-to-digest milk will only upset his stomach and make him feel worse. The fully informed and rational Lonnie would want this Lonnie to desire water instead, or some clear soda. That will cure the dehydration and allow Lonnie to feel better.

Here, we seem to be talking about instrumental desire. The desire for something to drink is a means to fulfill other desires - mostly, the desire to not feel so poorly. A desire-as-means for water is better than the desire-as-means for milk these other desires. Drinking water will fulfill the relevant desires, while drinking milk will thwart them. Desires-as-means can be informed by reason and an awareness of the relevant facts in this way.

We could take the desire for milk as an end-desire. That is to say, our agent simply likes milk - and will drink milk for its own sake. To the degree that drinking milk is an end-desire, it is immune to reason. The agent will continue to want milk even if he is aware of all of the facts. He may discover that he has more reason to refrain from drinking milk at this moment, but this has no effect on the desire.

A person with a particularly strong desire to drink alcohol, take drugs, eat chocolate, or gamble knows all too well that true beliefs and reason have no effect on the desire. Avoiding the harmful activity requires having more and stronger desires to put against it. And, since future desires do not have backwards causation, those more and stronger reasons have to be found in the present.

However, Railton then goes on to describe how the agent can come to acquire a new end-desire through what amounts to a system of biological reward and punishment.

For example, suppose that Lonnie gives in to his craving and drinks the milk. Soon afterwards, he feels much worse. Still unable to identify the source of his malaise and still in the grips of a desire for the familiar, his attention is caught by a green-and-red sign in the window of a small shop he is moping past: "7-Up," it says. He rushes inside and buys a bottle. Although it is lukewarm, he drinks it eagerly. "Mmm," he thinks, "I'll have another." He buys a second bottle, and drains it to the bottom. By now he has had his fill of tepid soda, and carries on. Within a few hours, his mood is improving. When he passes the store again on the way back to his hotel, his pleasant association with drinking 7-Up leads him to buy some more and carry it along with him. That night, in the dim solitude of his room, he finds the soda's reassuringly familiar taste consoling, and so downs another few bottles before finally finding sleep. When he wakes up the next morning, he feels very much better. To make a dull story short: the next time Lonnie is laid low abroad, he may have some conscious or unconscious, reasoned or superstitious, tendency to seek out 7-Up. Unable to find that, he might seek something quite like it, say, a local lime-flavored soda, or perhaps even the agua mineral con gaz he had previously scorned. Over time, as Lonnie travels more and suffers similar malaise, he regularly drinks clearish liquids and regularly feels better, eventually developing an actual desire for such liquids-and an aversion to other drinks, such as milk-in such circumstances.

What began as a desire-as-means becomes something desired as an end.

At the same time, we may also tell a story about how drinking milk, resulting in fee,Inge of poor health, results in Lonnie coming to acquire an aversion to drinking milk, at least when traveling. Even under conditions where he is not dehydrated, he comes to desire clear sodas and to have an aversion to drinking milk.

Here we have an example of value that fits both of Railton's requirements for value realism.

Requirement 1: Independence. The effects of drinking milk or water on the fulfillment of the agent's desires is independent of the agent's beliefs. That one will cause him to feel better and the other worse are objective facts - true in the world. People who deny these facts are wrong.

Requirement 2: Causal interaction. An agent is able to interact with these objects - or, more precisely, their desire-fulfilling and desire-thwarting properties in ways that bring about relevant changes in the agent's attitudes and actions. That which has positive value becomes that which the agent pursues and that which has negative value becomes that which the agent avoids.

Values, then, are real.

Desirism states that desires are real, they exist as propositional attitudes that take the form, "Agent desires that P", and whether P is true in some state of affairs S is a matter of empirical fact. All of the parts of value are real, so the whole is real. Indeed, we cannot explain or predict the motion of real-world events without appealing and postulating the existence of desires as providing motivational reasons that give value to states of affairs.

The next question, then, is whether moral value is a type of real value.