Reading over this post, I think the main thrust of my argument is that you should be able to see what is wrong with act utilitarianism if you have a sociological imagination. What if everyone justified their decisions in such a way without the guardrails imposed by considerations of human rights nor the signposts from various traditional values such as honesty, fairness, and so forth (that we can yet evaluate and reevaluate if they look like they might cause harm in some cases or indeed in every case for some of the more controversial traditions)? Even if done perfectly, it can have bad results. And if done imperfectly, it can be horrible. So we can say that in the long term, act utilitarianism seems like it fails based on its own foundational principle. It is not the best way to pursue utility.
It would still seem that every time you make an exception to a rule, you're acknowledging that an act calculation has taken precedence and the rule was inadequate.
An act calculation might give rise to a reevaluation of a rule, but it has hardly 'taken precedence'. E.g. abortion: it is better to say that anyone should be able to have an abortion than to evaluate each abortion for its effect on utility and allow or disallow on that basis.
That is not to say that one should not consider the consequences of their actions, only that they needn't be reconsidering the morality of theft every time they visit a relative.
It's really not very difficult or time consuming to think about the consequences of one's actions.
It is if you're operating from first principles every time and want to do it honestly and optimally. Act utilitarianism demands not just that you consider consequences but that you do the best thing as far as you can tell. Ironically, that perfectionism itself makes it more prone to serious error; there are situations that are rather uncontroversially acceptable. But what if, by sacrificing some principle or value or even person(s), you might do better than just acceptable? What if the "hard choice" is really the right thing to do? The rule utilitarian will tend to reject these unless given a very good reason not to do so, whereas the act utilitarian will be tempted-- especially if the 'sacrifice' is not something important to them but the benefit is. The result may be an atrocity.
That is not to say rule utilitarians are not capable of the same, but they have some handy signposts warning them of the danger ("Maybe don't do atrocities, it's probably going to be bad"). The act utilitarian only has the abstract principle of utility which
oughtto lead them in the right direction, you would think, but easily may not because it is to be considered in its purest (and vaguest) form. If only because humans are bad at math, I shouldn't want to live in a society in which everyone is an act utilitarian. Which means I shouldn't be one either because it is correct to guard against hubris.
I fail to see how all these pitfalls aren't magnified, rather than mitigated, by a rule approach.
Act utilitarians can find themselves justifying torture, for example, in certain specific cases, like to get information that will stop some plot to bomb lots of people. Rule utilitarians are more likely to notice that being able to do torture effectively enough that it would reliably give the desired result in such a case-- getting the information required to stop the bomb plot-- requires expertise. And that expertise requires practice. And that practice means living in a society which tortures people somewhat regularly. Which is probably not the best approach to maximizing happiness and so forth-- just a guess.
Now, an honest act utilitarian with perfect information and making no errors is unlikely to justify torture in the specific case in which it is not already normalized (but keep in mind that any of those three can be false). But if it is already normalized? Then the downside of continuing that normalization is less and the upside of employing it in the least objectionable cases can keep it normalized. Act utilitarianism tends to reinforce the status quo because the status quo is generally speaking stable. And moving away from that stability has costs; barriers to exit, in the economic jargon. No matter what that status quo is. An act utilitarian will not meaningfully oppose tyranny until that tyranny is already defeated.
Say you live in a village. The village is terrorized by a dragon. It periodically eats the villagers or burns a building. Individually, none of the villagers can solve the problem of the dragon-- adventurism is not effective. Collectively, they may be able to stop the dragon in some way. But they are disunited and afraid and the dragon seems to target the houses of those that are not as obsequious. At any given time, the villagers have a number of options on how to spend their time over and above doing what they need to survive. For example, they can try to cheer up the families of those who have lost someone to the dragon. They can work hard and produce food and have lots of sex in order to have babies to replace those lost to the dragon (this might otherwise be phrased as "they can be good livestock"). They can pretend that there is nothing to be done about the dragon except learning how best to cope with it. They could also try organizing themselves to solve the problem of the dragon-- this will not have any benefits to utility until the problem is actually solved; the dragon decisively dealt with in some manner. Such organizing is hard. It involves many steps and risks; getting enough people on board, figuring out what to do once they are. This takes time which could be spent doing anything else. And then there is the possibility of reprisal. The utility function for these villagers is going to have local maxima in which the effects of the dragon's violence are mitigated but the dragon itself remains unchallenged; attempting to make progress on solving the problem comes at the expense of utility and will only have benefits assuming a particular series of events follows in which others also make choices which move away from the local maxima of the utility function. So no proper act utilitarian calculation is going to tell the villagers to organize themselves to solve the problem of the dragon; they will instead be directed to the best way in the short term to suffer its periodic killing and destruction but never to stop it. The status quo is maintained.
But calculation is also prone to error. And power will justify itself using whatever tools are available. So you will see justifications of the kind, "yes, imperialism has some downsides but now India has railroads, so who can say really whether the British Empire was a good or bad thing?" Mill vacillated on his opinion of the colonial administration of India; but he was at least capable of thinking about it. Act utilitarians-- once it has started-- will take it as a given. Should it end? Well, gosh, that would be terribly disruptive. Rule utilitarians have a much easier time justifying human rights and avoiding (or at least recognizing) certain moral quandaries because in a funny way they have more time and reason to think about the finer details and grander possibilities. Or to put it another way, and quoting Mill in
On Liberty they can "regard utility as the ultimate appeal on all ethical questions; but ... utility in the largest sense, grounded on the permanent interests of man as a progressive being" rather than a neat calculation for each individual case uninformed by the rest of them.
Rule utilitarians have the categorical imperative (and relevant modifications of it) in their toolkit. What would the world be like if everyone acted this way? What would the world be like if well-meaning people all acted in this way? Both of these questions are irrelevant to the act utilitarian. "What if we had a radically different economic system?" is not a question that should even occur to someone who is really an act utilitarian. Why dedicate any resources to pursuing that when you can donate all your money to Oxfam and volunteering at the soup kitchen? Do both? But that is bad to the extent that you're wasting time and other resources on revolutionary self-education and organizing which, let's face it, is probably not going to come to fruition any sooner or later (or at all) based on the actions of just
you. But if everyone thinks that way, nothing will ever change-- at least, not in that direction.
The capacity to ignore the complexity or subtlety of an action is far greater.
Not really. Rule utilitarianism has a greater capacity to consider the impact of second order effects; it is not impossible for act utilitarians to consider these, but they always have the out of "but what if we just do it this once and never again?" Which is fraught with danger-- danger that, by definition, the act utilitarian has excluded from their analysis. What if everyone confronted with roughly this situation responded by doing X is a question that can have a much different answer than what if, in this situation, I do X. What is hard to measure in the individual case-- and thereby easy to disregard-- can be easier to conceive of in the aggregate.
A rule utilitarian approach tends to be informed by secondary principles that are well founded and uncontroversial; such as the benefits of honesty, fairness, etc., and tends to lack the audacity of an act utilitarian approach in which you are attempting to predict the future for every single action with an eye toward not just acceptable outcomes but optimal ones.
