It’s not that scientists are immune to having silly ideas and/or falling prey to invalid arguments. But they are all subject to possible experimental disproof of their predictions and, perhaps as a result of having to often admit they are wrong, their community is also reasonably rigorous about noticing and reaching consensus about rejecting invalid arguments (even when they seem to make correct predictions).
Philosophers, on the other hand, perhaps due to not having the benefit of frequent exposure to irrefutable correction, seem unable to reach a consensus rejecting even the most egregious examples of silly invalid argument (despite advertising themselves to the world as the go-to experts on checking validity).
A case in point is the claim that Infinite Dust Specks Are Worse Than One Torture.
The account linked to above repeats a couple of major errors that are common in pseudo-mathematical analyses of ethical problems.
One is the assumption that there is an actual ordered quantity that reflects the net well being or pain level of an individual. But this is almost surely not the case. It is quite possible and even seems most likely that our feeling about an experience has several independent components and that our ranking of experiences suffers from the same kind of non-transitivity as is known to lead to “paradoxical” effects in multi-candidate elections where it is indeed possible that the electorate can, in two way races prefer A to B and B to C but still prefer C to A. And it is also possible that both our relative weightings of the criteria and our rankings according to each criterion are not constant and are affected by many things such as our recent experience and even just hormones and brain chemistry.
Another is the fact that many philosophical analyses of ethical problems ignore the existence of empathy and so talk as if it is possible to torture one individual without causing emotional pain to others who know about it or at least as if for any torture that happens there is no chance that it may become widely known about. But it is easy to construct models in which the potential suffering caused by knowledge of the torture is included in the analysis in such a way that, for everyone, the expected gain from elimination of the torture of one other is much greater than the pain of a dust mote.
Source: Infinite Dust Specks Are Worse Than One Torture (by Bentham’s Bulldog on Substack)