Wednesday, August 31, 2016

The Death of Desire Utilitarianism

Friends, readers, lurkers, lend me your ears, for I have come not to praise desire utilitarianism, but to bury it.

Desire utilitarianism was the name I originally gave to the theory that I defended here. A few people talked me into changing the name to Desirism because too many critics wrongly assumed that "desire utilitarianism" meant "desire fulfillment act utilitarianism." They were tired of explaining to people that this is not the case.

It turns out that they were more right than they knew (or that they let on).

Desirism is not a utilitarian theory.

Let me explain . . .

There are several utilitarian theories - identified to the different answers they gave to two questions.

(1) What is the proper object of moral evaluation?

Act utilitarianism says that the proper object of evaluation is acts. The right act is the act that maximizes utility. It's leading competitor on this measure has been an interpretation of John Stuart Mill's utilitarianism called "rule utilitarianism". The right act is the act that conforms to the best rules, and the best rules are those rules that maximize utility. (Note: I hold that this interpretation of Mill is incorrect.)

The difference is that act-utilitarianism will command an agent to kill an innocent person if it produces more overall utility. The rule utilitarian would argue against killing an innocent person because the rule, "Do not kill innocent people" maximizes utility.

(2) What is "utility"?

Hedonistic utilitarians say that people should maximize pleasure and minimize pain. Other utilitarians say to maximize happiness and minimize unhappiness. Preference utilitarians say to maximize preference satisfaction.

A preference satisfaction act utilitarian says, "Do that act that satisfies the most preferences." Hedonistic rule utilitarianism says to perform that act that conforms to the best rules, and the best rules are the rules that produce the most pleasure and the least pain.

Desire Utilitarianism.

Desire utilitarianism said that the right act is the act that conforms to the best desires, and the best desires are desires are those that produced the most utility.

People kept confusing this with a thesis that said that the right act was the act that produced the most desire fulfillment. Desire fulfillment act utilitarianism says that if killing an innocent person fulfilled a lot of desires, then he should be killed. Desire utilitarianism would argue against killing the innocent person because a general aversion to killing innocent people would maximize utility.

Defenders grew tired of defending desire utilitarianism against counter examples aiming to describe cases where fulfilling the most and strongest desires (e.g., desires of those who wanted to have a child tortured) were objectionable. Because a desire to torture children is not a good desire, desire utilitarianism does not recommend torturing children regardless of the number of people desiring it.

Desirism is not Desire Utilitarianism

Desirism quit being a utilitarian theory when I realized that there is no utility to be maximized. When one evaluates desires, desires are not, in fact, being evaluated according to their ability to bring about the most desire fulfillment. Desires were being evaluated according to the degree to which they fit in with other desires.

A common case that I use to illustrate desirism imagines a person (Alph) with one desire - a desire that the planet Pandora B exists. He has a button in front of him that will cause the planet Pandora B to come into existence. It will utterly destroy Alph at the same moment. Pandora B, we will assume, is a planet that has no being capable of having desires.

The state that would be created if Alph pushes the button is a state in which no desire fulfillment exists. In fact, it is a state in which no desires exist - let alone a desire that is fulfilled. Alph, the only creature with desires, ceases to exist so there will be no desire fulfillment in the state after the button is pushed.

Strictly speaking, a desire utilitarianism account that says to create as much desire fulfillment as possible would have to conclude that pushing the button produces a state with no value. Consequently, it concludes that there is no reason to bring about such a state. This problem applies regardless of whether we make acts the object of evaluation (the right acts are those that produce the most desire fulfillment) or desires (the best desires are those that produce the most desire fulfillment). Making desires the primary object of evaluation does not protect it from this conclusion.

On a desirism account, Alph's desire that Pandora B exists is a reason to realize a state in which Pandora B exists. The state that would exist upon pushing the button would be a state in which Pandora B exists. Therefore, Alph has a reason to push the button. Alph is not pushing the button to create a state where desire fulfillment exists. Alph is pushing the button to create a state in which Pandora B exists. That is what he wants. That is what he has a reason to do.

We can see the same point in another common example that I use to describe desirism. In this account, there is one person (Alph) with one desire (to gather stones). Alph lives on a planet where there is a limited number of stones. We introduce another person to this world (Betty) and we tell Alph that he can either give Betty a desire to gather stones or a desire to scatter stones.

We can tell from Alph's desire to gather stones that he has a reason to give Betty a desire to scatter stones. With Betty spending her time scattering stones, Alph can work full time on gathering stones. Here, too, I would like to note that Alph is not doing this to maximize desire fulfillment. Alph has no desire to maximize desire fulfillment. Alph only has a desire to gather stones, and he gives Betty a desire to scatter stones for no reason other than it allows him to spend more time realizing a state in which "I am gathering stones" is true.


So, Desire Utilitarianism was a nice theory while it lasted. However, it did not last. It carried with it too much dead weight from traditional utilitarian theories. It took me a while - and several discussions with people I could not hope to name - for me to come to realize that the theory does not, in fact, fit the utilitarian mold. There is no state of "utility" to be maximized.


Martin Freedman said...

Finally and what I long ago argued for! ��

Great to see you blogging regularly again and that you are finally doing your PhD.

Unknown said...

Isn't this the same as the kin selection vs individual selection (kin selection being wrong, individual selection being right)?
I mean, it's not about maximizing, it's about surviving.

Unknown said...

Yes, thank you for official acknowledging the confusion in the terms.

Adil Zeshan said...

"There is no state of "utility" to be maximized."


This is a very welcome development. Nice one, Alonzo.

Alonzo Fyfe said...

Yes, Faithlessgod, I acknowledge that you were one of the people who had been trying to convince me otherwise many years ago. It took me a while. You were right.

Unknown said...

Actually, I think you haven't gone for enough. I've mentioned this before, but I think it should be called the "General Theory of Morality". It serves the same purpose for morality as the GTR does for gravity. It is the basis of many special cases as well, such as divine command, well being-based, and happiness-based models. Those models take one desire and make it the standard. Desirism gives no such special "reference frame".

I suspect that you would think the name too grandiose. But if so, then GTR needs to be changed to "Gravitism". Einstein wasn't "Einstein" when he released his theory, so arguing that he had more clout fails as well.

Also, this is not to shower you with praise or worship. It is intended to maintain consistency and to set the correct tone when presenting the theory. Too many times I run into issues of "if this theory doesn't do X, it isn't a moral theory". I think this name would help communicate that it is unlike any other theory so far presented and in fact is an encompassing theory.

I'm probably over-reaching a bit, but I think you get the idea.

Just my 2¢.