“I would let the five die and not feel guilty about it, because I am not the cause of their deaths.”
A more charitable way of phrasing the consequentialist PoV here is that we care more about stopping the deaths than avoiding feelings of guilt. Yes, it’s true that on certain accounts of morality you can’t be held responsible for the deaths of five people in a Trolley Problem-esque scenario but the people will still be dead and consequentialism is the view that consequences trump all other considerations, like adherence to a deontological moral code, acting as a virtuous person would act, acting in a way that is perfecting of one’s teleological nature etc.
Whether or not to hold people responsible for certain actions, like everything else on consequentialism, is, for us, a matter of determining whether that would lead to the best consequences.
Now, having said that, in practice consequentialists will behave like deontologists, virtue ethicists etc. quite a lot of the time. The reason for this is that having everyone go around making individual act utilitarian calculations for every potentially moral decision would likely be catastrophic and consequentialists are committed to avoiding terrible consequences whenever possible. It is usually better to have rules that are held to be exceptionless (but which can be changed to some degree, as is exactly the case with law) and teach people to have certain virtuous qualities so that they won’t be constantly looking for loopholes in the rules and so on.
Does this mean that deontology, virtue ethics etc. are correct after all? No! Because it’s still all being justified on consequentialist grounds, which is how we decide which rules to have and what counts as a virtue etc. in the first place. They will be the rules and virtues that lead to the best real world consequences. Because the kind of philosophical scenarios under which it is morally correct to push a fat man off a bridge are carefully constructed to be as inconvenient as possible the rules and virtues that we use in practice will probably forbid pushing people off bridges or raising the kind of person who would do that. This will mean that once in a blue moon someone really will find themselves in such a scenario and that person will likely make the wrong decision. It will also mean that the majority of the time, people will be making the right decisions and the consequences, overall, will be better than they would be if we let people decide for themselves on a case-by-case basis whether murdering someone maximised utility or not.
Consequentialism will still have counter-intuitive results. It should! There’s no reason to think that our intuitions are an infallible guide to what’s right. However, the kind of consequentialism attacked in a lot of philosophical arguments is a pretty naive version and it would be much more productive for everyone if we focussed on the stronger versions.
I had noticed it and mistakenly attributed it to the sunk cost fallacy but on reflection it’s quite different from sunk costs. However, it was discovering and (as it turns out, incorrectly) generalising the sunk cost fallacy that alerted me to the effect and that genuinely helped me improve myself, so it’s a happy mistake.
One thing that helped me was learning to fear the words ‘might as well,’ as in, ‘I’ve already wasted most of the day so I might as well waste the rest of it,’ or ‘she’ll never go out with me so I might as well not bother asking her,’ and countless other examples. My way of dealing it is to mock my own thought processes (‘Yeah, things are really bad so let’s make them even worse. Nice plan, genius’) and switch to a more utilitarian way of thinking (‘A small chance of success is better than none,’ ‘Let’s try and squeeze as much utility out of this as possible’ etc.).
I hadn’t fully grasped the extent to which I was sabotaging my own life with that one, pernicious little error.