I sort of accepted rule-utilitarianism as my system for evaluating morals as there are many problems with strict utilitarianism but at the same time utility still seems to be the most logical factor in determining what's right and wrong, so rule-utilitarianism seems to be the middle-ground.
However, I've found one moral conclusion through rule-utilitarianism that seems to be wrong. It doesn't really attack the system as a whole, but rather the practical application of it.
So let's say I’m drafted to an unethical military and I break the law by refusing to serve, I would still say that's morally right since the outcome would be good.
Rebuttal: But if everyone now starts to do that because they believe that’s morally right, the country would have no protection against its enemies and that would start a war in which many people would die in the process, causing a lot more harm than there would be if people had served.
Rebuttal: I would say that’s irrelevant to the argument because the only thing that I need to consider when evaluating the morality of my decision is my decision’s outcome and nothing else.
Rebuttal: Would you then be ok with one guy choosing to litter once because it saves him the hassle of finding a bin? Since littering once wouldn’t do any significant harm to the world and since he does gain happiness from not having to find a bin then technically the net-gain of happiness would be increased.
Rebuttal: Yes, I would be fine with him doing that once.
Rebuttal: Would you then be ok with a society where everyone decided to litter once because it’s not morally wrong?
Rebuttal: No, I would impose a rule that would maximize the utility of everyone by not allowing people to litter.
Rebuttal: Wouldn’t that rule be the same as the rule of mandatory service? So therefore, you should always serve in a mandatory military, even if the military is highly unethical E.G. the Nazi military.
That conclusion seems wrong even though it seems to make logical sense within a rule-utilitarian framework. Ideally, we'll know for certain which rule is best to maximize utility and then we'll be able to accurately answer this question but since we can't realisticly measure that, we're left with this conclusion which appears to be false.