I have some naïve ideas about utilitarianism. What should I read that might be related?

Question

I've only taken a few introductory philosophy classes, but I'm very interested in philosophy in general. Within ethics, (I think) I subscribe to utilitarianism in some form. I feel as though most (if not all) of the criticisms of utilitarianism point to problems that can be resolved via more careful calculations. For example, problems such as the utility monster arise because the calculation of utility that simply adds everything up is ridiculous. It seems very clear to me that the total utility is not the only factor that matters; the shape and spread of the distribution of utilities should also be considered.

The way I see it, it's very clear that some form of utilitarianism is the best way to do ethics. We should acknowledge this and move on to the more interesting question of how exactly we calculate the utilities.

The most relevant criticism is probably that utility is difficult to measure or define consistently. Avoiding difficult calculations seems necessary in any decision-making we do via utilitarianism.

We should estimate, to the best of our ability, the negative and positive consequences of our actions. This includes an estimate of the amount of time we need to do the calculations themselves.

Anyway, I'm rambling. I'd like to read more about attempts to make utilitarianism a practical tool for decision-making, but I don't know where to look. Mill, for one, doesn't say much about this kind of stuff. Are any authors who discuss utility-calculation-related ideas, such as taking into account the shape/spread of the distribution of utility or the time required to make decisions?

Answer 1

For a bare minimum, you need to understand Harsanyi's Theorem, many accounts of which are available on the web.

You can find Harsanyi's paper and several other foundational papers in the book "Utilitarianism and Beyond", edited by Amartya Sen and someone else whose name is escaping me.

Answer 2

You may be interested in the article on rule consequentialism over at the Stanford Encyclopedia of Philosophy: http://plato.stanford.edu/entries/consequentialism-rule/

Another interesting take is the modern trend towards negative utilitarianism, to work towards minimizing pain rather than maximizing pleasure. Popper addressed this in The Open Society and its Enemies. Peter Singer has also been basing much of his animal rights work on that idea.

Answer 3

One issue with utilitarian principles is a bullet in your question: the estimation of utility.

What do you do when the estimation of the utility has a very poor bound? What do you do if the best result you can come up with is "it's either the most important thing we've found in the world, or utterly devoid of utility, and we don't really get a better bound by spending more time on estimation?"

This becomes important around chaotic systems. Because of the tremendous sensitivity of chaotic systems to initial inputs, the utility of their output is tremendously difficult to predict without actually doing the action. This forces a utilitarian to seek to avoid chaotic systems in their life, because it isn't worth the time to estimate them.

Unfortunately for that utilitarian, life is notoriously chaotic. This means a utilitarian has to be very careful with how they talk about estimating utility, because a casual definition can quickly lead to a utilitarian being forced to divorce themselves from life because their utility is "maximized" that way.

Your discussion of shape/spread fits into this category of thought. Two major extremes occur. One defines a shape, and then follow it regardless of how unhappy they may feel about it. This leads to utility monsters. The other extreme is to define a goal for the shape, and extend the shape until those goals are achieved. This leads to paralysis in difficult to estimate scenarios.

The interesting middle ground is trying to define enough of a shape to avoid paralysis, enough goal to be flexible in difficult situations, and to be able to use partial information from your estimation as you go.