If this is what morality is, then it is not difficult to see why we should prefer simple, fixed rules to case-by-case calculations. First, for morality to work as a social system we need others to be predictable. If we cannot be sure whether someone might decide to kill us tomorrow in order to save others, we can never be sure that we are safe from anyone. We can have no faith in a justice system that allows the odd innocent to be punished in order to deter those who might otherwise harm even more. So although having a fixed rule that we should never harm the innocent might sometimes result in more innocent people being harmed, on balance the price we pay for that is much less than the cost of uncertainty. From a social point of view, the predictability and reliability of moral behaviour are much more important than getting it right from some abstract, absolute perspective.
Related to this is the problem that we really don’t want to live among relativists. To be clear, relativism is not the common-sense view that what is right or wrong always depends at least to a certain extent on the particular circumstances. It is the view that there is no shared standard of right and wrong, and what might be right for you could be wrong for me, depending on the day, our mood or the weather. When we are relativists in that sense, we really cannot trust anyone to stick to any kind of principle at all. We might legitimately worry that a lot of flexibility is too dangerously close to relativism to be admirable.
Another reason not to trust people who are very intellectual in their moral thinking is that they can very easily simply be rationalisers, using their brain power to justify whatever they want to believe. Evidence from psychology supports this, suggesting that most thinking we do about moral choices is an after-the-event rationalisation of whatever immediate, intuitive impulse we have.
It might seem a troubling thought for anyone who favours a more reflective ethics that in practice, ethics is rooted much more in feeling than in thinking, but there is good reason for this. The fundamental impulse to treat others well derives from a kind of empathy, not obedience to authority or a rational principle. For sure, we ought to use our reason to check whether our impulses are misleading us, as they undoubtedly often do. But in daily life, it makes perfect sense to trust the person of generosity and good heart more than the professor of abstract intelligence.
In that sense, the research has it back to front. The people we most trust are not absolutists, but people who don’t think too much about moral principles at all and stick to common sense. And morality is a form of common sense: the sense we have in common of what we all owe to each other. It can only work if we refuse to make ad hoc exceptions, no matter how intellectually justified they appear to be.