Preference Satisfaction vs. Desire Fulfillment
Among those who write about ethics, there are many who defend the idea that “preference satisfaction” has intrinsic value. This thesis is very close to my own, which is that “desire fulfillment” is the mark of value. Yet, there are distinct differences. Here, I would like to look at similarities and differences.
The Roots of “Preference Satisfaction”.
The idea that value is rooted in preference satisfaction arose from problems with the claim that humans always sought their own pleasure over pain, or happiness over unhappiness. Theorists recognized that there were things we prefer over others even though they did not effect our balance of pain over pleasure, or happiness over unhappiness.
G.E. Moore provided one example when he asked readers to consider two possible universes — one in which a beautiful world exists, and one in which a feted and rotten world exists. Readers were asked to assume that nobody would actually experience each world. Yet, we would still prefer that the beautiful world exist.
Other examples include questions like, “Would you care if somebody spread lies and rumors about you, even if you never found out about it and it never affected you in any way?”
Or, “Would you rather be thought of as a brilliant scientist now and, in the future, as a charasmatic and convincing hack in the future after you died, or thought of as a charismatic and convincing hack now and later, after you died, have people realize that your claims were brilliant and ahead of your time?”
Or, “Would you rather have your child live a healthy and happy life where you thought he was being mercilessly tortured, or would you rather have your child be mercilessly tortured where you thought he was having a healthy and happy life?”
All of these examples identify cases where our preferences do not reflect what makes us happy. Though, clearly, we prefer being happy to being unhappy — all else being equal. However, in many cases where all else is not equal, we will give up happiness for something else that we prefer.
Common Ground and Divergence
Both “preference satisfaction” and “desire fulfillment” handle these counter-examples to the idea that happiness is the sole value. “Preference satisfaction” gives weight to whatever situation we prefer. In the case of the parent who prefers that their child be healthy and happy to being happy, the “preference satisfaction” theorist gives this extra weight -- simply because it is the stronger preference.
But what is “preference satisfaction”? The concept is vague, amorphous, and ill-defined. And, what happens if a person has a “preference” for living in a society without Jews, or for torturing young children? If preference satisfaction has intrinsic value, then the satisfaction of these preferences also has intrinsic value. These situations are “good”, it seems, if one accepts “preference satisfaction” as having intrinsic value.
“Desire fulfillment”, in contrast, has a rather specific definition.
Desires are mental states -- members of one of two families of mental states that affect all behavior. The other family of mental states is “belief”.
Beliefs and desires are propositional attitudes — attitudes towards a proposition. I will use the letter ‘P’ to stand for any proposition.
A proposition is a statement capable of being true or false, such as “My child is healthy,” or “I am regarded as a brilliant scientist”, or “I am happy.”
A “belief that ‘P’” is an attitude that the proposition ‘P’ is true. So, if I believe that my child is healthy, then I have an attitude that the proposition “my child is happy” is true.
A “desire that ‘P’” is an attitude that the proposition ‘P’ is to be made or kept true. So, if I desire that I am regarded as a brilliant scientist, then I am disposed to make or keep the proposition, “I am regarded as a brilliant scientist true.
A desire is fulfilled in any state of affairs in which ‘P’ is true. So, my desire to be regarded as a brilliant scientist would be fulfilled in any state of affairs in which “I am regarded as a brilliant scientist” is made or kept true. So, we can answer questions like:
(1) What is a desire? (Answer: An mental attitude towards the proposition ‘P’ where the proposition is to be made or kept true.)
(2) What is it to have a desire fulfilled? (Answer: for a desire that ‘P’, it is to have a state where ‘P’ is made or kept true.)
(3) What is the relationship between desire fulfillment and intentional action? (Answer: A desire that ‘P’ motivates an agent, to a degree proportional to the strength of the desire, to make or keep ‘P’ true.) Intrinsic Value
Many “preference satisfaction” theorists hold that preference satisfaction has intrinsic value. So, the satisfaction of any preference has value merely in virtue of the fact that it is the satisfaction of a preference.
“Desire fulfillment” theory, in contrast, holds that there is no such thing as intrinsic value.
If an agent has a desire that ‘P’, then any state in which ‘P’ is made or kept true has value for that agent. However, the state in which “agent desires that ‘P’ and ‘P’ is true” has no value, except insofar as it fulfills yet other desires.
Let us imagine two universes. Universe (1) contains a universe in which there is an agent with a desire that ‘P’, and ‘P’ is true. Universe (2) contains no such agent, yet ‘P’ is true. Which universe is better?
The “intrinsic value” theory would state that Universe (1) is better because it contains preference satisfaction (or desire fulfillment) and Universe (2) does not.
However, in denying the existence of intrinsic value, I am stating that Universe (1) has more value than Universe (2) only to the degree that somebody actually desires that Universe (1) exists. We may prefer Universes (1) to Universe (2), but that does not mean that one has more intrinsic merit than the other.
In fact, even our hypothetical agent in this example would not necessarily have a reason to select Universe (1) over Universe (2). Unless we also give Agent a desire that he exists, or a desire for something that requires his existence (that is, a desire that ‘P’ where ‘P’ implies ‘agent exists’), then both universes will fulfill the agent’s desires equally well. He would be indifferent to the two options.
Bad Preferences and Desires
“Preference satisfaction” and “desire fulfillment” theories both have to deal with the problem of good and bad preferences or desires, such as the desire to torture children.
For “preference satisfaction” theories, particularly those that hold that preference satisfaction has intrinsic value, this is a problem. If preference satisfaction has intrinsic value, then the satisfaction of the preference for torturing a child has intrinsic value. “Desire fulfillment” -- particularly that type that holds that there is no intrinsic value, and value consists solely in ‘being such as to fulfill a desire’ has an answer to the problem of bad desires.
Desires, like everything else, has value only in terms of their ability to fulfill or thwart (other) desires. The desire to torture children clearly has the potential to thwart a great many desires — in particular, the desires of the person who would be tortured. On the other hand, the desire to torture children fulfills few, if any, other desires.
It is important to note here that I am not looking at the act of torturing a child. Such an act may well fulfill a desire to torture a child, or a desire that the child be tortured. Here, I am looking at the desire to torture a child, and looking at the (other) desires this desire fulfills or thwarts.
We can look at it this way:
What if nobody had a desire to torture a child? In this case, no child (and nobody who cares about children) would have those desires thwarted. Nor would a prohibition on torturing children negatively affect anybody, because nobody would desire that which was being prohibited.
However, if some people had a desire to torture a child, then some desires would be thwarted. Either the desires of the child being tortured (and those who care about children) would be thwarted, or the desires of the person wanting to torture children would be thwarted. What if everybody had a desire to torture children? There would still be desires thwarted, namely those of the children and those who care about children.
The first option, where nobody has a desire to torture children, is the only option where no desires are thwarted. The desire to children thwarts other desires, but the desire itself fulfills no other desires. This would classify it as a bad desire.
There is no intrinsic value in torturing a child, nor is there intrinsic value in satisfying a preference (or fulfilling a desire) to torture a child. Torturing a child clearly has value to the person who desires to torture children, but the desire to torture children itself has only negative value — it only thwarts desires.
Conclusion
“Preference satisfaction” theories can be modified so that they say everything that I have written about desire fulfillment theories. In fact, some may want to claim that I have mischaracterized preference satisfaction theories and that they are exactly like desire fulfillment theories.
I do not think this is accurate, but I see no reason to dispute it, because it does not identify any problem. This response does not identify any problem with desire fulfillment theories, it is only a dispute over which name to use in talking about them.
On the issue of which term to use, we still have reason to prefer “desire fulfillment”. The idea that beliefs and desires are propositional attitudes, that a “desire that ‘P’” is an attitude that the proposition ‘P’ is to be made or kept true, the way in which beliefs and desires form intentions which lead to intentional action, are all a part of an evolving theory of the mind. “Desire fulfillment” theories can tie directly into the work being done in that field with no need to translate or explain any differences.
Languages are invented for convenience, and it would certainly be more convenient to have a theory of value that tied in directly to a theory of the mind. It would certainly be more convenient to translate “good” as “is such as to fulfill the desires in question”, and leave behind the question of how to translate desires into preferences.