Questions on the basic ideas of utilitarianism are covered in a separate document. Common criticisms of the theory are also covered separately. This FAQ is a collection of matters arising, not covered in other areas.
Try The Internet Encyclopedia of Philosophy, or, better still (for many purposes), the more studious student can read his Autobiography. A listing of Mill resources is available on the links page.
Rule utilitarianism is suggested as a "variant" of utilitarianism which differs from act utilitarianism (ie. the original utilitarian system, as described by Jeremy Bentham - "act" is prefixed merely to help distinguish the two) in that it proposes we use rules for determining right and wrong, rather than the felicific calculus, and only the formation of rules being subject to the calculus. (See the Internet Encyclopedia of Philosophy, the Penguin Dictionary of Philosophy and also the works of RM Hare - e.g. "Moral Thinking" - for details.) It is now commonly believed that rule utilitarianism and act utilitarianism are not mutually exclusive as they have sometimes been held to be... since rules can be of arbitrary complexity, it would be theoretically possible to design a rule which gave the same results as felicific calculus in any specific situation.
The arguments in favour of rule utilitarianism (for example: reducing the time taken to determine the proper course of action; and reducing the likelihood that our judgement will be prejudiced by features in a specific case which do not exist in the general case - e.g. personal involvement) can be quite easily accommodated under a normal (act) utilitarian scheme - Jeremy Bentham certainly never advocated that we try to calculate the rightness or wrongness of every single act from first principles.
All utilitarians will re-use the results from previous calculations in those situations which they judge to be suitably similar: I suggest that the difference between an act utilitarian and a rule utilitarian is simply the frequency with which they judge these "suitably similar" situations to occur.
An act utilitarian would more inclined to treat each new situation as different from all the others and requiring a fresh calculation; a rule utilitarian will be more inclined to notice the similarities between the present case and previous ones, and therefore having a similar solution as those previous cases. The difference will in part be determined by the agent's estimation of the cost of applying the felicific calculus (e.g. in time, lost opportunities etc) against the benefits of doing so (e.g. the allowance for details specific to that case, where these aren't prejudicial to the agent).
A sensible act utilitarian will re-use the results of previous calculations where this seems optimal; a sensible rule-utilitarian will take care to ensure that his approach remains consequentialist; the difference, then, between an act-utilitarian and a rule utilitarian, may not be as large as is sometimes thought.
Utilitarians can believe or disbelieve in free will theory - the principle of utility is compatible with both views. That said, utilitarians generally tend to reject those theories of free will that are incompatible with determinism.
It is a common mistake to suppose that utilitarians have a fixed approach to controversial political issues. While utilitarians agree on what the criteria for good social policy are (being its conformability to utility), we freely and commonly disagree on which policies are actually useful. We know which goals we should aim for, but we are no better equipped than any others in politics to decide which policies will, in practice, help us achieve those goals. One may ask a utilitarian what makes people happy, but one would be unlikely to receive a better answer than can be given by the psychologist. One may ask a utilitarian about the national debt, but would be unlikely to receive a better answer than would be given from the economist. Plainly, knowing the goals that we wish society to achieve does not imply that we know in which direction we must travel to arrive there. But, of course, some directions seem more promising than others.
The scenario is usually presented something like this: a doctor has several patients who will die if they don't receive organ donations. Perhaps one needs replacement kidneys, another needs a replacement heart, or whatever. Now supplies of these organs have come short, and there are no spare organs. There are, however, a number of healthy people available who could be suitable donors, except that they are unwilling to sacrifice themselves to save the others. The number of people required to supply the organs is less than the number of lives can be saved by carrying out the transplants and, in this situation, it is suggested, utilitarianism supports killing some people to save the lives of those in need of replacement organs, since the harm of killing a few is supposed to be less than the harm of many dying. Does this follow?
Not necessarily. There are a whole host of side effects we would need to consider. For example, how could we pick a victim for our supplies, without generating fear and alarm in the community? If someone who goes to hospital with a minor complaint gets killed for his body parts by the doctors, would this not generate a fear of hospitals in the general populace... who would then refuse to enter one lest a similar situation occur again? And how, if we give doctors the power to decide who should die and who should live, do we stop doctors abusing their powers and becoming, for instance, extortionists? There is also the assumption that different people's lives necessarily have roughly equal worth... which is simply ridiculous from a utilitarian perspective. What if the two recipients are Hitler and Goring, and the forced donor is Martin Luther King?
Now, some may suggest, these concerns are irrelevant to the basic issue. The scenario can changed in the details to overcome these potential problems - it can be specified in the description of the scenario that the necessary assumptions hold. In that case where, on balance, the effects are the best if we kill someone, does utilitarianism support it? Of course it does. Can it ever be right to kill some to save others? Under any plausible system, I suggest, the answer is yes. The alternative is that we would let everyone die rather than kill a single individual, which is nonsensical if we assume that life is valuable.
Why is it that web sites and books on utilitarianism tend to focus on Bentham, Mill, and other long-dead philosophers? Are there no new ideas in utilitarianism?
I suggest the emphasis on Bentham and Mill occurs for a number of reasons:
None, so far as I know, who are famous for anything other than their philosophy. The most well-known utilitarians (not necessarily hedonistic) include
The main similarity between the two is that they are both usually hedonistic theories, ie. both concerned (solely) with pleasure and pain. The difference between the two, is whose interests (whose pleasure and pain) count when determining the rightness or wrongness of an action. An egoist would think that only the interests of the agent himself count, while the utilitarian would think that all interests count. So, for example, a utilitarian would believe it is wrong to do something that harms someone else, if the benefit (to the agent) is smaller than that harm. An egoist would think it always right to act in such a way as to benefit himself, even if this resulted in massive harms for everyone else. So an egoist would tend to agree with a utilitarian that, "what is good for an individual is that individual's happiness". But he would deny, "So what is good generally is happiness generally".
Under some definitions, being entirely self-centred, egoism actually fails to qualify as an "ethical" theory. Further information on egoism is available from the Internet Encyclopedia of Philosophy.
I would first like to point out that this question is very odd: utilitarianism is an ethical theory, as such it is either correct or it is incorrect. If it is correct, then it has no weaknesses. Never-the-less, the question is asked frequently enough to warrant it being answered here.
Tom Regan has suggested some (more-or-less) reasonable criteria for an ethical theory (see The Case For Animal Rights section 4.3), and against these we see that utilitarianism does very well indeed: It is consistent, it has ample scope (it covers all actions), and it is reasonably precise.
On the subject of agreement with intuitions, I can say that in terms of the core values, utilitarianism is in complete agreement with my intuitions - in my experience, happiness and hapiness alone is felt to be intrinsically good or worthwhile, and suffering alone is felt to be intrinsically bad or disvaluable.
In terms of right and wrong, there are a number of counter-intuitive implications, but I would reject the idea that our intuitions of right and wrong are reliable. For example, if our intuitions of what a person is are wrong - and they are, see Reasons and Persons by Derek Parfit - then why would we assume them to be correct on (e.g.) whether it is right or wrong to kill a person?
As my friend Dave remarks,
Science has comprehensively confounded "common-sense" in all empirical matters. Our traditional ethical intuitions, when wrapped in secular guise, are less susceptible to experimental challenge. It would be a piece of singular good fortune if the least testable aspects of common-sense folk-wisdom just happened to be the ones that could most be relied on.
There is a substantial body of scientific evidence that we are taught our "intuitions" of right and wrong, via the process of socialization and (Pavlovian) conditioning. That being the case, we should give up the claim to special insight on their behalf. Our intuitions on intrinsic value are perhaps more reliable than those on right and wrong, since we do not need to be taught to value happiness or disvalue suffering - rather the valuation of them is intrinsic to those experiences.
If, on the other hand, one believes that one's intuitions always correctly discriminate right from wrong, then one need not take the trouble to learn about utilitarianism (or indeed any other ethical theory) - so why even ask the question?
In short, my answer to the question is this:
Of course, many people disagree with utilitarianism. A collection of some of their arguments is available here.
Whether it is right to keep a promise depends not only on the effect this has on the happiness of the promiser and the promisee, but also the community at large. That is to say, it is generally worth keeping a promise because it can be very upsetting and distressing to be deceived by one who is trusted, and (if the community finds out about it) it can make society generally less trusting and less co-operative.
One must be very careful not to give one's own position favourable treatment - to not accept, as sufficient grounds for our breaking of a promise, grounds which we would not hold sufficient if someone else had given them for their breaking of a promise.
That said, there are some cases where utility is better served by the breaking rather than the keeping of a promise. For example, if (while in a drunken rage) I had promised to kill someone, that would probably not be a good promise to keep. And even the most carefully considered of commitments can sometimes go astray.
Utilitarianism does not accept that the "guilty" deserve "punishment" - it views punishment as a prima facie evil since it involves the infliction of harm. This harm can be justified if it has greater benefits in terms of maintaining order in the community, or similar, but the utilitarian position is that if punishment is not justified by utility then it is not justified at all.
Many people, usually critics, suggest that Mill held a very different view of utilitarianism to Bentham. For example, it is suggested that Mill was more inclined to rule-utilitarianism, and related concern for "rights". However, this only displays the ignorance of such commentators on the work of Bentham, who spent many years working out a system of laws (i.e. rules) by which people could live to maximize utility. It might also be suggested that Mill was for the extending of the vote to women, as he surely was, but so too was Bentham so no disagreement can be found here. (And in any case this would not be a difference in point of theory, only in estimation of the results of that theory - see the FAQ on politics.)
There does seem to be at least one point on which Mill considered his position to be different to Bentham's, and may have been correct, and this is on the concept of "quality" of pleasures and pains which Mill introduced in chapter2 of his essay. (I have commented on this concept of quality elsewhere.) Apart from this one point, I believe Mill was very much a standard Benthamite utilitarian.
Technically, according to utilitarianism, no acts are supererogatory - for each act, either you should do it or you should not do it. Even on those utilitarian definitions of "should", "ought", etc which admit of positive and/or negative degrees, there is not some class of acts that are "better than is morally required", for utilitarian morality always requires the best [2]. However, this should not be misinterpreted: on minor issues, the feeling of the agent may be all important in determining the value of the act. So Utility does in fact say, of many things, "Do it, if you like; and if you don't like it, don't do it." When this is recognized, it may be that the apparent appeal of some acts being supererogatory is substantially diminished.
[1] There is some evidence that Bentham was aware of the animal liberation issue. It seems likely, in that case, that he declined to clarify his position in this area to avoid his theory becoming subject to ridicule - society simply wasn't prepared to consider the idea in his day. I suggest it is for similar reasons that Bentham's considered views on homosexuality (he was against punishment for it, and saw little if any harm in it) were not published in his life-time. (Similarly, he was quite shy in his support for giving women the right to vote - his unpublished papers are clearer on this than those he published.)
[2] I take the view that so-called "satisficing utilitarianism" is not really a kind of utilitarianism, though it is a form of consequentialism.