I think I disagree with your interpretation on every one of these points!
1: If you're assuming the reason Bob enjoys his crimes is because he's lowering the utility of his victims, you need to make that assumption much more explicit. If Bob only committed his crimes because it makes him feel good, but doesn't feel better knowing that the victim is unhappy because he has committed them, then he's only in the same position as anyone else who wants something that's good for emself but hurts another person. For example, if I want a coat made of panda-bear fur, then this upsets panda-lovers and (if we allow animals to have utility) pandas, but it doesn't allow anyone to disregard my desire - it just means that I might not get it satisfied if the panda-lovers turn out to have more clout.
Even if Bob explicitly enjoys lowering utility, I still find it sketchy to dismiss his desire as an axiom of the system. In a counterfactual world where all desires are completely fixed and non-self-modifiable, Bob's desires should be taken into account (and probably rejected when it turns out they hurt other people more than they help him). In a world more like our own, where desires become more common and intense if there's a possibility of fulfilling them, these desires should be dismissed for game theoretic reasons to discourage any more such desires from ever forming, not just disregarded.
1.5: I can't tell whether you're just acknowledging the existence of punishment, or something different. Yes, we should punish child molesters as a game theoretic action to discourage child molestation in the future. And this punishment has to involve negative utility. But punishment should be specific and limited. If the child molester has already gone to jail, then ey gets the same weight in our calculations as anyone else. And if we do punish em, we do it not because we're using a system that says we can disregard eir utility on general principle, but because a greater good (discouraging future child molestation) overrides it.
2: Here I would say the villagers don't have a preference against the girl living, they have a preference against witchcraft. We take all of their preferences into account, but because drowning the girl doesn't reduce witchcraft, their preferences have no bearing on whether we save the girl or not. So since there's massive disutility (death) and minimal utility (they get small utility from not having to waste time worrying about witchcraft, but they don't get what they were hoping for which is less witchcraft occuring), it's a net loss and we save the girl. In the ritual dunking case, there's no disutility, and the utility is that they stop worrying about witchcraft, so it's a net gain and we let them do it.
(this makes more sense if we assign numbers: say having less witchcraft is +50, killing a girl is -25, not having to worry about witchcraft so much is +5, and ritually dunking is -1. These assignments make more sense if you imagine asking the villagers "If this girl wasn't a witch, would you want her to continue living?" - their answer is the appropriate preference)
I don't think it's ever okay to literally discount utility. I think it's often okay to let one source of utility counterbalance and override another source of utility, and that sometimes that first source may be a game theoretic precommitment to ignore someone's utility, but you don't stick ignoring someone's utility into the theory directly.
Pure curiosity, in response to the whole non-discounted utility argument.
In a case similar to the beginning to Kill Bill - orderly selling sex with a comatose, brain dead woman - how does your utilitarian calculator come out? Assume (unlike the movie) she is in fact completely gone. Do you simply bite the bullet and agree it's a positive outcome?
I'm planning a top-level post (probably two or three or more) on when agent utility should not be part of utilitarian calculations - which seems to be an interesting and controversial topic given some recent posts. I'm looking for additional ideas, and particularly counterarguments. Also hunting for article titles. The series would look something like the following - noting that obviously this summary does not have much room for nuance or background argument. I'm assuming moral antirealism, with the selection of utilitarianism as an implemented moral system.
Intro - Utilitarianism has serious, fundamental measurement problems, and sometimes substantially contradicts our intuitions. One solution is to say our intuitions are wrong - this isn't quite right (i.e. a morality can't be "wrong") unless our intuitions are internally inconsistent, which I do not think is the problem. This is particularly problematic because agents (especially with high self modification capacities) may face socially undesirable incentives. I argue that a better solution is to ignore or discount the utility of certain agents in certain circumstances. This better fits general moral intuitions. (There remains a debate as to whether Morality A might be better than Morality B when Morality B better matches our general intuitions - I don't want to get into this, as I'm not sure there's a non-circular meaning of "better" as applied to morality that does not relate to moral intuitions.)
1 -First, expressly anti-utilitarian utility can be disregarded. Most of the cases of this are fairly simple and bright-line. No matter how much Bob enjoys raping people, the utility he derives from doing so is irrelevant unless he drinks the utilitarian Koolaid and only, for example, engages in rape fantasies (in which case his utility is counted - the issue is not that his desire is bad, it's that his actions are). This gets into some slight line-drawing problems with, for example, utility derived from competition (as one may delight in defeating people - this probably survives, however, particularly since it is all consensual).
1.5 - The above point is also related to the issue of discounting the future utility of such persons; I'm trying to figure out if it belongs in this sequence. The example I plan to use (which makes pretty much the entire point) is as follows. You have some chocolate ice cream you have to give away. You can give it to a small child and a person who has just brutally beaten and molested that child. The child kinda likes chocolate ice cream; vanilla is his favorite flavor, but chocolate's OK. The adult absolutely, totally loves chocolate ice cream; it's his favorite food in the world. I, personally, give the kid the ice cream, and I think so does well over 90% of the general population. On the other hand, if the adult were simply someone who had an interest in molesting children, but scrupulously never acted on it, I would not discount his utility so cheerfully. This may simply belong as a separate post on its own on the utility value of punishment. I'd be interested in feedback on it.
2 -Finally, and trickiest, is the problem of utility conditioned on false beliefs. Take two examples: an african village stoning a child to death because they think she's a witch who has made it stop raining, and the same village curing that witch-hood by ritually dunking her in holy water (or by some other innocuous procedure). In the former case, there's massive disutility that occurs because people will think it will solve a problem that it won't (I'm also a little unclear on what it would mean for the utility of the many to "outweigh" the utility of the one, but that's an issue I'll address in the intro article). In the latter, there's minimal disutility (maybe even positive utility), even though there's the same impotence. The best answer seems to be that utility conditioned on false beliefs should be ignored to the extent that it is conditioned on false beliefs. Many people (myself included) celebrate religious holidays with no belief whatsoever in the underlying religion - there is substantial value in the gathering of family and community. Similarly, there is some value to the gathering of the community in both village cases; in the murder it doesn't outweigh the costs, in the baptism it very well might.
3 - (tentative) How this approach coincides with the unweighted approach in the long term. Basically, if we ignore certain kinds of utility, we will encourage agents to pursue other kinds of utility (if you can't burn witches to improve your harvest, perhaps you'll learn how to rotate crops better). The utility they pursue is likely to be of only somewhat lower value to them (or higher value in some cases, if they're imperfect, i.e. human). However, it will be of non-negative value to others. Thus, a policy-maker employing adjusted utilitarianism is likely to obtain better outcomes from an unweighted perspective. I'm not sure this point is correct or cogent.
I'm aware at least some of this is against lesswrong canon. I'm curious as to if people have counterarguments, objections, counterexamples, or general feedback on whether this would be a desirable series to spell out.