Nov 14, 2023·edited Nov 15, 2023Liked by Richard Y Chappell
A big idea:
It seems to me that once you get away from utilitarianism, it's almost inevitable that you're gonna end up being moral particularist to some degree.
So far, moral-particularist theories were basically intractable to "analyze". But, in principle, AI might eventually offer tools to explicitly represent (at least approximations to) ultra-complex moral-particularist theories. How would such models be trained? I guess using experimental philosophy questionnaires to elicit people's intuitions.
The tech is not there yet, but could it ever get there? I'd like to hear from the Wittgenstein-Anscombe-inspired particularists.
Regarding 2 (as well as 3 and 5), I'd be interested to hear more how your ideas relate to the ideas of "conequentialization" of moral theories and "scalar ethics".
I'm generally suspicious of the "consequentializing" project insofar as it involves collapsing the distinction between consequentialism and deontology. I think the latter is a significant distinction!
Telic ethics (as in #2) is meant to be compatible with maintaining the traditional distinction. Roughly speaking, we might think of deontologists as having some distinctively "deontologically"-flavoured fundamental goals (e.g. to avoid violating rights in each instance), whereas I think consequentialist goals are more apt to be characterized without essential reference to moral concepts (instead invoking broader normative concepts like that of well-being). That's not a strict analysis or anything, just a rough first pass.
I say a bit more about scalar ethics if you follow the 'Deontic Pluralism' link. I think it's (broadly speaking) the right way for consequentialists to go, at least insofar as our *reasons for action* are concerned. But attention to fittingness, and especially blameworthiness, can give us the resources to make some more "binary" distinctions. I just don't think we should focus *excessively* on the resulting deontic binary (as that minimal standard seems antithetical to the sort of moral ambition urged in #2).
A big idea:
It seems to me that once you get away from utilitarianism, it's almost inevitable that you're gonna end up being moral particularist to some degree.
So far, moral-particularist theories were basically intractable to "analyze". But, in principle, AI might eventually offer tools to explicitly represent (at least approximations to) ultra-complex moral-particularist theories. How would such models be trained? I guess using experimental philosophy questionnaires to elicit people's intuitions.
The tech is not there yet, but could it ever get there? I'd like to hear from the Wittgenstein-Anscombe-inspired particularists.
e.g. https://academic.oup.com/edited-volume/43987/chapter-abstract/371424801?redirectedFrom=fulltext
Regarding 2 (as well as 3 and 5), I'd be interested to hear more how your ideas relate to the ideas of "conequentialization" of moral theories and "scalar ethics".
I'm generally suspicious of the "consequentializing" project insofar as it involves collapsing the distinction between consequentialism and deontology. I think the latter is a significant distinction!
Telic ethics (as in #2) is meant to be compatible with maintaining the traditional distinction. Roughly speaking, we might think of deontologists as having some distinctively "deontologically"-flavoured fundamental goals (e.g. to avoid violating rights in each instance), whereas I think consequentialist goals are more apt to be characterized without essential reference to moral concepts (instead invoking broader normative concepts like that of well-being). That's not a strict analysis or anything, just a rough first pass.
I say a bit more about scalar ethics if you follow the 'Deontic Pluralism' link. I think it's (broadly speaking) the right way for consequentialists to go, at least insofar as our *reasons for action* are concerned. But attention to fittingness, and especially blameworthiness, can give us the resources to make some more "binary" distinctions. I just don't think we should focus *excessively* on the resulting deontic binary (as that minimal standard seems antithetical to the sort of moral ambition urged in #2).