Moral Progress & Moral Philosophers
A post by Eric Schliesser (Amsterdam) at Digressions & Impressions, responding to an essay by Daniel Kelly and Evan Westra (Purdue) at Aeon that was mentioned recently in the Heap of Links, raises questions about moral progress and the role that the work of moral philosophers plays in it.
[Giorgione, “La Tempesta”]
What set me off on this rant is this sentence: “changing the social world for the better will very often mean changing some old, harmful norms and replacing them with better ones”…. Let’s stipulate there is a germ of truth here. But we don’t consciously norm-engineer our way to a morally improved social world primarily or even very often; the drivers of moral progress are not the moral elect; rather collective action, prophecy (religious and philosophic), institutional change, and the unintended side-effects of long-range patterns of behavior.
You can read Kelly and Westra’s essay here and Schliesser’s response here.
Discussion welcome.
What happens if we reverse the direction of change—from good established norms to potentially bad new norms? Is affective friction in this case something we should heed or ignore? Doesn’t moral progress seek to instill a disposition to react to future violations precisely in the ways that are said to be unreliable?
Yes, affective friction is a generalized response to norm change, good or bad. Sometimes it will “get it right,” but only by coincidence. We think resistance to new, bad norms should be grounded in reflective evaluations of their contents, not just in the fact that we may find them annoying.
Your second point is interesting. I think it points to a persistent way in which our norm psychology makes moral progress hard. Our norm psychology is dumb and doesn’t care about morality, but aligning social norms with our moral reasons is an important part of making the world a more just place. There will probably always be a tension here.
Thanks for your response, Evan. (I liked the piece and am on board with the general project FWIW). I guess my concern is how malleable we can make our norm psychology. In my view, if we want “good” norms to stick, we have to rely on our norm psychology. There is no way around it. But in doing so we open ourselves to making future change (whether good or bad) more difficult, for the reasons you highlight. There seems to be a trade-off. I hope you’re right that reflection and curiosity can go some way towards mitigating the concern (we should not just rely on norm psychology).
One question for Kelly & Westra: do norms necessarily make it harder for us to engage in real moral reflection, on your account? If so, to what extent should we be trying to reduce the centrality of norms in our social life? It seems that the purpose of norms is to make our social lives more smooth, our engagement with them more “fluent.” I think it’s clear from your article that you think that such smoothness and fluency are goods, but presumably ones that can be outweighed by other goods. Presumably, there are more and less “norm-centric” kinds of social life. And if we were to make life less norm-centric, the harms that you see coming from misinterpreting affective friction would be lessened.
It’s not clear whether you’re advocating for a moral progress that effectively comes down to finding the right norms (after which affective friction would be an effective moral indicator), or moral progress toward a society in which we are simply more reflective and (presumably) more morally experimental, and so just constitutionally more resistant to taking norms and their accompanying affective friction as serious moral indicators.
Thanks for your question, Jordan. We think that norms are a pretty ubiquitous feature of human social life, and that they’re a big part of how we successfully interact with one another. So in that sense we think of these basic facts about our norm psychology as fixed obstacles that need to be navigated around. That said, there is variability across cultures when it comes to the number of norms in an community and how strictly they are enforced. Michele Gelfand’s work on this is really nice: she calls it cultural tightness and looseness. We weren’t thinking of our proposal as a way of promoting cultural looseness, though we do think that affective friction is probably worse in tighter cultures. So maybe that’s something to explore.
Overall, we’re really speaking to two audiences: first, everyday people experiencing affective friction, and second, people who see themselves as trying to promote norm change. We think the first group of people would benefit from greater reflectiveness in response to feelings of irritation, and the second would benefit from pursuing strategies for norm change that don’t exacerbate affective friction.
>and the second would benefit from pursuing strategies for norm change that don’t exacerbate affective friction
This is a point that I probably missed in your original essay and that I find very helpful as someone who considers himself to have fairly progressive values but finds himself experiencing extreme affective friction in the face of obviously counterproductive attempts at norm change by other progressives.
What I kept thinking about while reading your essay was that it isn’t necessarily, or even oftentimes, the case that the irritation people feel comes from being confronted with a new norm, it comes from how they’re being confronted. Most of the cases that come to mind are cases where the person with new norms clearly presumes to be the other person’s moral superior. They make it clear by how to talk to the other person. They simply dictate to the person what’s right and wrong, implying that they’re in a position to do so. That’s what people find irritating more often than not. I’ve had great success in opening and even changing more old-fashioned people’s minds by posing a new norm as a topic of open-ended conversation where I just lay out reasons and different ways of framing an issue that my interlocutor can think about without feeling judged. It feels to me like most fellow progressives don’t even try to do something like this but prefer to act smug and condescending.
Thanks for articulating this. Hard agree.
(Though I’m not sure how progressive my values are; I’ve lost all sense of what political labels mean.)
In light of Eric Winsberg’s guest post, I’ll add another thing I and others find not just irritating but corrosive: that the self-described moral know-betters aim to institutionalize and centralize their diktat in ways that insulate it from democratic pressure.
(I’m not presuming you agree with my addition. I just took your comment as an occasion to add it.)
Thanks Kitty. I don’t think we address this in much depth in the Aeon piece, but in the talk/paper version (in prep), we talk about how affective friction can grow into outright backlash once you layer in certain intergroup dynamics. This is especially the case when a norm spreads unevenly in the population, such that some subgroups adopt it try to enforce it upon other groups that have yet to do so. When you get that kind of intergroup dynamic, we think that norm enforcement can trigger psychological responses to intergroup threat that exacerbate affective friction and make people double down on existing norms. In other words, sometimes even a good norm change can feel like an attempt at domination when it’s coming from a perceived outgroup. That’s a huge obstacle for would-be norm changers. Our one suggestion for how to address it is to look for ways of assimilating a new norm to shared identities so that those intergroup threat responses aren’t triggered.
Thanks, Evan, that’s really helpful. I’ll read up on Gelfand.
The passage that sets off Schliesser is: “changing the social world for the better will very often mean changing some old, harmful norms and replacing them with better ones.”
He objects: “But we don’t consciously norm-engineer our way to a morally improved social world primarily or even very often…”
This objection seems like a non-sequitur. After all: surely old harmful norms may be replaced with better ones by mechanisms other than conscious norm-engineering. Indeed: even if moral improvement to the social world mostly occurs as a side-effect of other changes, I would expect that such improvement would *often* consist in improvements to social norms.
Or: does Schliesser think social norms are themselves causally inert?
In the context of the essay he’s responding to, he sees the essay’s authors, Kelly and Westra, as supporting the kind of norm-engineering approach that he rejects. I read him as saying that we can agree with the point about moral progress but that it doesn’t imply, and we shouldn’t support, Kelly and Westra’s specific conclusions about which methods help us replace harmful norms with better ones.
That is clearly his overall position. But in the final paragraph, it really seems like he presupposes that the triggering quote (“changing the the social world for the better will very often mean…”) entails the problematic idea about the causes of moral improvement.
In fact, immediately after the triggering quote, he writes: “It’s the very same commitment that, back in the day, annoyed me about MacAskill’s narrative.” (His main criticism of MacAskill’s narrative is that, on Schliesser’s account, change to social rules is *not* in individual hands.)
This seems to be the persistent philosophical debate between cultural conservatism (institutions, practices, traditions and power relations) and moral progress. Isn’t this what was reflected in the debate between Gadamer and Habermas? And Gadamer provided a kind of resolution in his theory of expanded horizons. This would say that there is no thoroughly rational path to moral progress. Affective friction is inevitable and without analytic reduction. Moral progress requires hermeneutic engagement and communication. A reaching in to the other to understand them as an other. This isn’t solvable by a moral vanguard or elite except to the extent that such “vanguard” can engage in such communication under the condition that it could be wrong. A wild card in such dialogue though is the extent which power plays in and potentially distorts the dialogue. Power needs to be a part of the discussion therefore. What role does power play in the measure of affective friction?Does power obviate the possibility of moral progress? How does it limit the range of discussion of normative horizons and moral potential?
To put a finer point on Schliesser’s post: It seems like Kelly and Westra argue that, in most instances in which there’s moral progress, there’s a high probability of encountering the eye-roll heuristic, and hastily conclude from this that given one is experiencing the eye-roll heuristic, there’s a good chance it’s in response to moral progress and so should be ignored. To make such an argument it doesn’t do to just pick out examples where the connection holds, you really need to provide some evidence that most of the proposed changes were positive, and not fall for the selection bias of only attending to the changes that ended up being successfully adopted. If most silly proposed changes don’t get adopted *because* they encounter eye-rolls, and thus don’t spread much, then KW would be making the exact opposite recommendation that they should.
Kelly and Westra obviously know a lot about norms, so it seems particularly surprising how they managed to leave out the whole idea that a lot of norms have evolved over time through cumulative culture and can provide important effects that are not always well-understood by either individual norm-followers or outsiders, which is why history is replete with attempts to intervene in and ‘improve’ other cultures (whether by governments, missionaries, or colonialists, even reflective ones) that had negative unintended side-effects. Henrich’s “Cultural evolution is smarter than we are” seems like a heuristic that anyone speaking authoritatively about norms really owes it to their audiences to at least have on the table.
Thanks for your comment, Adam. To your first point, our position is that people are likely to experience affective friction whenever there is an incongruity between the norms they’ve internalized and those they confront in their social environment. We also think that affective friction is a product of the same psychological processes that allow norm-guided behavior to proceed smoothly and fluently when a person’s internalized norms are in sync with those of the social context they’re moving through. So we expect affective friction in the face of new norms — whether or not those new norms are improvements, whether accepting them would count as progress or regress. We argue that unpleasant as it feels, that feeling can invite, but never justifies a dismissive eye-roll, and so should be set aside by anyone trying to make a serious assessment of whether the norm is question should be adopted or not.
To your second point, we take the set of ideas you point to seriously. It is one of several things we had in mind in our brief discussion in the essay of the idea that affective friction functions as something like a normative immune system. But we agree there is much more to say here, and we hope to explore it in the longer version of the paper we’re working on. For now: even if one accepts that the process of cultural evolution can and has produced all kinds of complex cultural adaptations that provide benefits that the humans who use them don’t fully understand, what follows for the issues we raise in the essay remains unclear. I’d — uncontroversially, I think? — reject the inference from the claim that some trait X evolved to the claim that trait X is morally good. Certainly when the kind of evolution in question is the OG version i.e. genetic evolution by natural selection. Fitness enhancing and morally good aren’t the same thing. I’m still inclined to think that holds, and so to be suspicious of that same inference, when the kind of evolution in question is cultural evolution. But even if the cultural evolution version of it can be shored up with some extra steps, evolutionary processes can get stuck in local maxima. We can still use our critical capacities to evaluate the traditions and institutions and individual norms we inherit from the past, which, even if the product of a massively complex and pretty “smart” process of cumulative cultural evolution, still fall short of ideal in many of ways, including many morally relevant ones. Lots more to say on this front, though. Stay tuned.
Thanks Dan, I agree the heuristic ‘x is an operative norm, therefore x is good’ exhibits a certain flaws with many counterexamples and objections. But for the reasons Dan G outlines below it seems there’s opposite kinds of flaws (perhaps from overcorrecting to the first heuristic) with ‘Here are these ways in which new norm x seems better, therefore it probably would be better’, which we also need to take steps to avoid, in part by acknowledging that there’s many examples of this reasoning going badly, and that such flaws are more likely to manifest when one has the mindset of ‘let’s just use our critical capacities to sort the good norms from the bad’. Of course we can’t avoid using our capacities to make assessments entirely, but just trying to be careful isn’t enough to minimise the risks, one has to actively consider how often other people proposing new norms were in a similar epistemic situation to us and turned out to be overconfident, and what steps they would have needed to take to avoid making such mistakes.
Really interesting piece and response, thanks for sharing.
I want to stick up a bit for an idea in the Schliesser piece: “not all affective friction or feelings of irritation should be a cue for further reflection. This really depends on context. Reflection is time-consuming and costly, and involves lots of opportunity costs.”
While Kelly and Westra are right, I think, that the “eye-roll heuristic” is far infallible, I think they go too far in treating it as useless. I think that, in addition to the point about cultural evolution that Kelly raises in the comments here, there’s a bit more one can say about why a default skepticism about new norms–a default that can be overcome when the case is good enough!–is reasonable.
First, lots of norms are solutions to coordination problems, where it’s more important that everybody be on the same page than that we identify the optimal equilibrium. Think about stuff driving on the left vs. right. I can imagine a case that one of those norms is better than the other–most people are righties, and it’s easy for me to believe that this creates an asymmetry between the norms (I have no idea in which direction). So suppose it would be marginally better if we all drove on the left. Still, there are serious transition costs associated with getting from here to there–it’s easy to think that if we tried to make the switch, everybody who learned to drive under the old norm would be a much more dangerous driver for a long time, either because they sometimes forget and drive the old way, or because they’re just worse at driving in the new way–in a way that would (I think) justify a dismissive, eye-rolling response to somebody who proposed that we make the switch. It could easily be the case that long before we’d get to the point where people were driving well enough to justify the switch, human drivers were replaced by autonomous vehicles.
I think probably a lot of norms are a bit like this, though less dramatic. Think about norms governing social interactions. First, there’s clearly lots of convention here; in different times and different places, very different behaviors can signal respect, hostility, etc. Second, they serve important functions–maybe not avoiding car crashes, but avoiding conflict/violence based on misunderstanding or insult. Third, while it’s extremely plausible that the norms we’ve got aren’t optimal (that would be Panglossian) it’s also likely that (a) they’re better than a lot of alternatives (think of times/places when duels or honor killings were much more common, and (b) they work better at averting conflict/violence when widely accepted (like the driving norms).
So like with other complex machines, most tweaks that you can make to norms will make them worse.
Now if the filter for proposed tweaks (ie, norm revisions) were very strong, so that only extremely promising tweaks were proposed, then maybe there would be no value to a default skepticism about new norms. But it seems to me that’s not the case. First, a slightly cynical reason. It’s flattering to think of oneself as a norm entrepreneur. People get social credit for coming up with novel ideas. In our microcosm of philosophy, if you want to publish a paper, much better that it provide interesting (even if ultimately uncompelling) reasons for a change than that it rehash familiar justifications for the status quo. For what it’s worth, it seems to me that’s exactly as it should be when it comes to publication norms. But if proposed changes are rewarded for novelty independently of their promise, and most novel changes are bad, that’s some reason to think most proposed norm revisions will be bad. Just like knowing that somebody has defended some novel metaphysical view in a prestigious journal is an extremely weak reason to think the view is true, the fact that somebody is proposing a novel norm is an extremely weak reason to think it would be good to collectively adopt that norm.
Even absent those cynical reasons, it seems to me that in general we’re not so great at seeing untended negative consequences of the norms we propose–if somebody’s proposing a novel norm, it’s likely that they’ve thought more about the easy-to-imagine upsides than the unobvious downsides. (That strikes me as a special case of the kind of thinking that often leads political military leaders to overly optimistic assessments of how well wars will go; you typically imagine the best case, intended scenario for your proposed policy.)
Lastly, and perhaps the point I’m least confident of, it seems to me that the “eye-roll” heuristic isn’t totally undiscriminating among new norms. Rather, it seems to me that it especially targets new norms that seem too “clever”, as if designed to show off the qualities (virtue, creativity) of the norm entrepreneurs who propose them.
So I think that adds up to a case not to lament the existence of the “eye-roll” heuristic, and for why norm entrepreneurs should willingly accept the burden of proof in showing that their proposed norms are promising enough to overcome a default skepticism.
“changing some old, harmful norms and replacing them with better ones” But at what point do we or can we discover that tradition, both faith and intellecutal, have embedded ‘harmful norms’ into the cultural fabric” And who will define what they are replaced with and has the authority to effect such positive change, when that existing paradigm is exposed as error?