Fully addressing the question…
Fully addressing the question of thick and thin morality would take a lot more space and time than I’ve got at the moment, of course, so instead I’ll focus on one specific issue and beg off the rest until later: the question of “rationally defensible” and “rationally indefensible” lifestyles:
It is asking far too much of people to demand that their preferences must be ‘rationally justifiable’ in order to count as moral, indeed I don’t think that such a standard can even be coherent. We accept readily enough that when choosing sleeping partners we can be as ‘irrational’ or as ‘bigoted’ as we like in choosing whom to let enter our bodies or whose bodies we choose to enter. By extention I think the same applies to our non-corporeal property. So if someone wishes to open a nightclub for gay men only or a country club for WASP’s only or establish a town only for Welsh speaking Hindus nothing unlibertarian nor, I contend, even immoral is committed thereby.
Well, but there are a lot of things that need to be unpacked here.
Firstly, are we asking whether you’re morally obligated to have a rational justification for forming friendships (or sexual liasons, or …), or whether you’re morally obligated to have a rational justification for treating Smith and Jones differently with respect to X given that Smith is your friend (or lover, or…) and Jones is not? I think the second question is actually the only one I broached, not the first; but I think most of the suspicion towards the notion of “rational justification” in these spheres gets whatever plausibility it has from the first. All I urged above is that “Smith is my friend and Jones isn’t” is a rational justification for treating Smith better than you treat Jones with regard to some things (although not, importantly, with regard to coercive violence), but that “Smith happens to have been born in the same country as I was but Jones wasn’t” is not, and that it’s wrong to treat Jones worse than you treat Smith without any good reason to do so. That leaves open the question of what to say about the grounds you might or might not have for having become Jones’s friend but not Smith’s in the first place. You might say, “look, you said that treating Jones worse than Smith without a rational justification is immoral, so doesn’t that mean you also need to come up with some rational justification for becoming Smith’s friend rather than Jones’s if you want to maintain your position?” Well, only if becoming Smith’s friend but not becoming Jones’s is treating Jones worse than you treat Smith. I don’t think it is, unless you can articulate some way in which becoming Smith’s friend but not Jones’s is in some way (e.g.) unfair to Smith. (Which it might be—if, for example I become Smith’s friend because he flatters me but despise Jones because he honestly criticizes me. But I think that forming friendships like that is pretty clearly ethically objectionable.) The fact that there’s a certain amount of arbitrariness involved in the formation of many of our ties doesn’t entail that by forming those ties we are treating anyone any better or worse than anyone else.
Secondly, what do the notions of “rational justification” and “rationally indefensible” come out to? To be sure, I left that open in giving my own definition—which makes for a pretty thin definition of bigotry, I guess! But it’s important to note that when I described bigotry as “rationally unjustifiable” group-preferences, I was using the terms in such a way as to qualify bigotry as a vice term. (I think it’s part and parcel of how we use the word “bigot” that bigotry is eo ipso a bad thing to practice.) If that’s how it’s being used, then it’s important that you can draw a distinction between conduct that’s merely arbitrary and conduct that’s irrational. Doing A instead of B even though you haven’t got any reason to pick A over B isn’t the same as doing A instead of B when you’ve got a reason not to pick A over B. Merely arbitrary decisions can often be perfectly ethical—it may be that the choice is just not one in which ethics makes a ruling. But actively irrational decisions can’t be; as such they entail overruling ethical constraints. My claim is that bigotry (by definition) involves group preferences that are actively irrational, not just arbitrary. Someone’s merely arbitrary group-preferences may be signs of partisanship, or boosterism; they may even (in some cases) be signs of a silly disposition. But they’re not of themselves signs of bigotry. Bigotry as I defined the term is bad not because it’s arbitrary, but because it hurts people; not because it’s independent of reason but because it’s contrary to it. (This attaches to the question about forming, say, Welsh/Hindu-only towns. The question is, first, whether that exclusiveness is reasoned or arbitrary; and then, second, if it’s arbitrary, whether it’s merely arbitrary, or whether it exists in a social context where it’s actively harmful. In this respect there’s a pretty distinct moral difference between, say, Welsh/Hindu-only towns, on the one hand, and white seperatist institutions on the other: there may be reasons to think that a Welsh/Hindu-only town is silly, but there are also good reasons to think that it’s harmless to non-Welsh/Hindus. But that’s not at all clearly the case with whites-only towns (or, say, whites-only boycott campaigns) in a society with a history of violent, pervasive, and intensely damaging white supremacism.
This is also connected to the question of forming friendships and sexual liasons, incidentally: part of the issue here is the need to distinguish whether “Your decision to X is/is not rationally justifiable” means “Your decision to X has/has not got an articulate rational justification,” or “Your decision to X is rationally corrigible but is/is not undefeated by any other considerations.” Note that I left this entirely open in the definition; the only proviso is that a decision that flunks at least the second test (i.e., a decision that’s rationally corrigible and in fact should be constrained for other reasons) is immoral. Whether the stronger claim that a decision on this or that particular topic that flunks the first test is also therefore vicious.
(You might ask: “Well, what about decisions that the first test flunks, and that the second test can’t be applied to, because they aren’t made on rationally corrigible grounds?” The answer is that nothing that could qualify as a decision is rationally incorrigible. Those would not be decisions at all, but tics or fits.)