Sunday, November 30, 2014

How I learned to stop worrying and love moral relativism

This is partly autobiographical. I used to spend a lot of time thinking about metaethical issues and arguing for a kind of moral realism that would be incompatible with moral relativism, but I no longer worry so much about those kinds of issues. You probably don’t care about Swan intellectual autobiography, but I’ll outline an argument that gave me a push.

I’ll start with a definition. Many associate moral relativism with the view that no moral directives are justified or true. They then worry that moral relativists must think that any moral directive is as good or plausible as any other and so anything goes. But this is mistaken. Moral relativists think moral directives can be justified, but the justification of moral directives is relative to the beliefs, values or commitments of a group of people. This means that moral directives aren’t objective in the sense of applying regardless of people’s beliefs, values or commitments. Their justification, and the legitimacy of holding people to them, depends on their beliefs, values and commitments.

Here’s the argument:
1. M is a distinctively moral directive only if it provides reasons for acting.
2. R is a genuine reason for action for an agent only if it is capable of motivating that agent.
3. Therefore, M is a moral directive only if it is capable of motivating an agent.
Premise 1 is a statement of internalism about morality and reasons. It is a conceptual claim about the semantics of moral directives: claims of morality are essentially normative in the sense that there’s a necessary connection between them and practical reason. Moral claims employ terms that are evaluative, action-guiding and prescriptive. Of course, many non-moral claims, like judgments of etiquette, aesthetics, and prudence, have this semantic feature, as well. Some philosophers have exploited this fact in an attempt undermine morality/reasons internalism. But morality is supposed to be distinctive from etiquette, aesthetics, and prudence in the categoricity or authoritativeness of the claim. Something that’s a genuine M has ‘practical clout’ or ‘oomph’, such that someone who said ‘I know it’s M, but I don’t really care about M’ is making some kind of mistake. A moral directive adverts to reasons or considerations that cannot be legitimately shrugged off in this facile way.

This view about the connection between the requirements of practical reason and the requirements of morality is a species of moral rationalism. Moral rationalists usually say that the requirements of morality are practically decisive, but I only say that claims of morality purport to provide reasons for action that have greater deliberative significance or oomph than other practical directives. This version of the thesis avoids, on the one hand, the suggestion that morality has absolute weight in practical deliberations, and, on the other hand, the implication that morality is merely a system of hypothetical imperatives.

Premise 2 is a statement of existence internalism about the connection between reasons and motives. It is very often associated with Neo-Humean theories of motivation. Some consideration provides an agent with a genuine normative reason for action only if it is capable of playing a motivational role in her deliberations. This is just what it means for some consideration to be a reason for the agent. This statement of internalism, then, is a thesis about what has to be true in order for a reason statement truly to apply to an agent. It must connect up with things the agent cares about, or which are deliberatively accessible. This means that externalists about reasons and motivation are wrong to think that an agent has a reason for action when the proffered considerations are deliberatively inaccessible. An agent cannot sensibly be said to possess a reason that is deliberatively inaccessible to her, and so it cannot be a reason for her. The externalist could say, “Tuff. We’ll still apply the reason statement to her,” even if the proffered considerations are deliberatively inaccessible from her point of view. But that makes it sound less like the kind of thing that we should see as an authoritative directive and more like the kind of thing that would be an authoritarian directive.

The conclusion in 3 is a thesis about how to identify the contours of genuine moral directives. Understood this way, premise 2 is presenting answers to the questions raised by premise 1. Moral directives essentially claim that some agent has a significantly weighty reason to act (or avoid acting) in some way. They purport to direct others authoritatively. What, if anything, could possibly justify these claims? Premise 2 suggests the response that moral claims directed to an agent are appropriately justified in terms of considerations that are reasons for her. It follows, then, that moral directives have authoritative normativity when (and because) they are grounded in her rational and evaluative commitments; the authority that grounds these claims is her own.

In other words, the connection between normative reasons and motivation (in 2), and so the connection between morality and motivation (in 3), is read in such a way that an appropriate connection to motivation is what makes it the case that R is a genuine reason for action and M is a genuine moral directive. This is moral relativism. The most obvious objection is that moral relativism just gets the contours of morality badly wrong. Doing the moral thing cannot be so easy – like shooting an arrow, drawing a circle around where it lands and calling it a ‘bulls-eye.'

As I said, I’m not so worried anymore about this implication, but I can say more in the comments. What’s attractive to me about relativism is that it secures the distinctive authoritative status of moral judgments without authoritarian bossing around.

Kyle Swan
Department of Philosophy
Sacramento State


  1. Thanks for the interesting post, Kyle. Let me provide a bit of intellectual autobiography too: I've never got the pull of Premise 2, the neo-Humean view of existence internalism. Much of our ordinary thinking about reasons seems to be externalist, attributing reasons to people regardless of whether they care about the relevant considerations. A person who beats his wife but doesn't at all care about her well-being still seems to have a perfectly good reason for not treating her that way. Bernard Williams worries that's just brow-beating, as I suspect you would, but I don't at all.

    In fact, this kind of thinking is so much a part of our ordinary discourse that some, like Richard Joyce, think it's a non-negotiable constraint on the concept of a moral reason. So he of course is lead to think there just aren't moral facts at all (error theory) since he thinks there aren't any such "external" reasons that morality presupposes. I don't really see why we have to treat it as a necessary condition on the concept of a moral reason, since I'm open to thinking we could discover internalism is true. However, getting back to intellectual autobiography, I just don't get the pull of the conclusion that the wife-beater lacks a reason. If anything, it sounds like some serious bullet biting to say he doesn't have a reason to stop beating his wife.

    Of course, Williams and similar theorists argue that the wife-beater can't have a reason because it can't motivate him. Motivation, they say, only arises from a relevant desire. But I don't find any good reasons for believing this Humean view of motivation. I think rational people can be motivated to do what they come to recognize as right, even if they lack any desire to do it. So, if the wife-beater were rational, the relevant considerations would motivate him if he were to recognize that her well-being matters. Fancy arguments are given on both sides of such debates, but even if it were a stalemate, I'd fall back on what I'd think should be the default assumption: we can have reasons to do things even if we don't antecedently care about the issues at hand. I'd personally need a damn good argument to throw out this intuitive belief.

    1. Gee way to go Kyle! I spend all semester trying to convince my students not to be moral relativists and then you deliver this early Christmas present!
      For what it's worth, I also have a problem with premise 2. First notice that what ever set of moral beliefs a particular group holds, they can provide reasons for why those beliefs are true. For those reasons not to apply in other groups in similar contexts seems bizarre (unless they are the product of some kind of explicit contact-making). Nevertheless, cultural moral differences do seem to exist. But, here comes my objection to premise 2, people should be motivated to do what's morally right and if they aren't, then they are probably misinformed about what is morally right or they are a morally bad, perhaps psychopathic, person. The unrepentant wife-beater is either woefully misinformed, lacking in normal neurological development, or has a very evil wife and had to regularly resort to self-defense.
      Getting autobiographical now, I really hope I'm right. :)

  2. Thanks, Josh. I can think of a few things that might help:

    1. Yes, premise 2 is very often associated with Neo-Humean theories of motivation because of the way Williams defended it, but as stated premise 2 is broad enough that it is amenable with Christine Korsgaard’s Kantian-inspired conception of practical reason (and her qualification that the agent is rational in some Kantian sense). I’ve put it in terms of the agent’s beliefs, desires and commitments. The argument doesn’t depend on Williams’s Neo-Humean desire theory.
    2. So, still less do reasons for action depend on the agent’s actual desires. It could be true that someone desires to do something that they actually don’t have normative reason to do because of a variety of other beliefs, desires and commitments they have that make it true that they shouldn’t act on the occurrent desire. I think this is true about wife-beaters. They have and endorse projects (for Williams) or practical identities (for Korsgaard) and, for that matter, have even made agreements that are incompatible with wife-beating. A rule against it is justified from their point of view. I am worried about moralistic bullying, brow-beating or bluffing, but this isn't that.
    3. More generally, I think it will (almost?) always turn out true that a set of rules that protect basic agency rights can be grounded in people’s beliefs, desires, values and commitments. It will (almost?) always turn out true that people will have sufficient reason to go along with rules like that because of their beliefs, desires, values and other commitments.
    4. You can easily come up with counter-examples to this. Hobbes’s “foole” comes to mind. A few of your counter-examples might even be real and not just hypothetical cases. “Other people? Phhft. My interaction with them isn’t based on mutual benefit. Considerations like that are just incapable of moving me. My practical identity is quirky like that.” In the rare cases we encounter people like this the argument suggests that we can’t be in moral community with them and so we can’t exercise genuine moral authority with respect to them. But I don’t think that means we don’t have the moral authority to protect ourselves from such people.

  3. Kyle, I'm inclined to agree with you, though I'm suspicious that part of the reason you are at peace with MR so understood is that it doesn't necessarily commit you to any particularly controversial theses. For example, we can’t, on the basis of what you say in the post, stick you with the view that there are or ever have been moral communities of humans here on earth in which slavery or the subjugation of women or genocide is morally permissible. Maybe they just didn’t think about it enough. Though feel free to cop to something like that... I'll just be standing over here.

    But there is something else that is interesting here about the motivation condition. I appreciate the simplicity with which you stated it initially, but it seems clear from your elaboration of premise 2 and your reply to Josh that what you really mean by motivation is "rational motivation". Premise 2, in other words, would better read:

    R is a genuine reason for action for an agent only if it is capable of rationally motivating that agent.

    But then the question arises, why is a directive D not motivating in the relevant sense if it is simply a part of an agent's cognitive architecture, something that causes D-like behavior even though she has no reflective deliberative access to it at all? Or, alternatively, if D is just one of a set of directives chiseled on a stone tablet believed by the agent to be of divine origin? Of course, I think I understand why. Your concept of a moral agent is that of an autonomous self-legislating one, and reflective grasp of the reasons that motivate us is just what such a being requires.

    But then the question is how do we arrive at that conception of a moral agent? It is, I believe, not a universal conception, but one that is specific to WEIRD cultures. I would suggest that we have hammered out this view of moral agency partly on the basis of our own moral values. In other words, it is our belief in the importance of respecting individual rights and responsibility that makes that sense of moral agency particularly compelling to us.

    So if that is true then moral relativism may not be true in other cultures.

  4. Many thanks for the reply, Kyle! I see, so you want to say wife-beaters *typically* have a reason to treat their partners better but that's only because they *typically* have a relevant belief, desire, or commitment that's incompatible with wife-beating (or they would have such attitudes if they drew out the logical implications of their current attitudes). But that does mean you'll have to bite the bullet on the right sort of case, namely one in which the wife-beater has no such beliefs, desires, etc. Suppose he doesn't believe he ought to be kind, doesn't care about others, doesn't plan to interact with them productively, etc. In such a case, you still have to say he has no reason to treat his wife better, right? That doesn't seem to me like any less of a bullet to bite. But that's just me!

    The only way I can see avoiding this consequence is if you say something like: Well, the wife-beater will always have a reason not to beat because all rational people necessarily have a commitment to being rational. But then I'd worry that the view might not deserve the name "relativism." The idea would then be that moral truths are "relative" merely because the reasons they provide only apply to us if they can motivate us if we're fully rational. But that's a kind of "relativity" that is presumably present in truths that seem paradigmatically objective or non-relative. For example, the truth of the Pythagorean Theorem is "relative" in the same weak sense: I have reason to believe it only if I'd be inclined to believe it if fully rational.

    So it seems there's a dilemma: either
    (a) bite the bullet and really say that wife-beaters don't have reasons to stop beating if refraining fails to fits with their own mental states, or
    (b) avoid this consequence at the expense of embracing non-relativism.

  5. Randy, some folks go that way in order to avoid moral relativism. They impose idealized constraints to ensure that everyone has been cleansed of reasons to pursue the quirky or the ‘bad’. I find these proposals, according to which practical rationality will deliver convergence on the ‘correct’ WEIRD morality, which turns out to be, despite appearances to the contrary, deliberatively accessible from any starting point, to be implausibly optimistic.

    But I didn’t mean anything that strong or idealized. Just that there is often some deliberative distance to travel from people’s actual, occurrent beliefs and desires and what they have genuine normative reason to do given their fundamental commitments. I don’t think those commitments, though, in order to be their true, fundamental commitments, have to meet some substantive, WEIRD test. I had something in mind more like coherence and consistency in reasoning concerning beliefs that are minimally epistemically rational.

    If I understand your case above, I’m fine with thinking D is fine (justified in light of that group’s interests, values, beliefs, etc.). You were probably thrown off by what I said in 3 in response to Josh about agency. But minimal agency is different than some full-bodied account of rational autonomy.

    Let’s have D be something like “Let’s not send our children to schools where they teach WEIRD stuff.” D is understood as fundamental to a group’s identity (or necessary to protect fundamental aspects of that identity). They live here in America and the government wants to force the group to send their children to a public school to learn WEIRD stuff to promote certain aspects of these children’s autonomy, but the group sees “jet-fueled” autonomy as destructive of their tradition-oriented community. Here, it seems to me that this community is nonetheless committed to the value of minimal agency. Their argument against the legitimacy of the coercive educational policy when applied to them actually relies on the value of agency since, in demanding to be exempt from the education policy, they are asserting the right to be left alone to organize their lives in line with certain prized values that they reflectively endorse. It also seems to me that this commitment to minimal agency would make it illegitimate for the group to act on a desire to coercively prevent someone from leaving the community and why a state could legitimately impose a coercive policy protecting people’s right of exit from it. From a first-person point of view everyone is committed to wanting to be left alone to act in accordance with his or her best light. We all presume the legitimacy of this. This claim derives from the generic features of human agency as such, e.g., deliberation and choice, so everyone has the same grounds for claiming it. Everyone values their own agency, wants others to respect it, and so is plausibly committed to respecting the agency of others. It isn’t something distinctive to WEIRD cultures.

    1. Kyle, that's a stand-up and be counted answer, thanks. I don't know what I think about your last paragraph. It seems to me that our best shot at a conceptual truth about morality is that it is fundamentally about cooperation, which is inherently about sacrificing our autonomy for other benefits, such as sleeping better at night. So it seems to me that moral systems can run the gamut from minimal to more or less complete sacrifice of autonomy, in the latter case with minimal agency being fully accounted for by reference to the needs of a functioning society.

      You (almost) appeal to Brandeis' phrase "the right to be let alone" which is interesting in that he represented this as a relatively new right based on rising expectations and living standards. And he's right, right? Certainly it had no place in tribal morality. I mean, yeah, if Og and Grog are out on a hunt and Grog keeps flicking Og's ear for entertainment Og won't like it. But that doesn't mean there is anything moral about it. Chief Reginald may not like it because they are compromising the hunt, and thus the tribe. That's a moral consideration.

    2. I agree, but they would reflectively endorse these cooperative rules and various sacrifices of their autonomy. Doing so is necessary to have decent social life, which is necessary to advance many interests and projects they have. That's where the authority of the rules comes from.

    3. So, are you saying that moral relativism is true, but all individuals in all societies (with obvious exceptions for crazies and incompetents) in fact endorse substantive fundamentally moral directives relating to respect for personal autonomy sufficient, at least, to counter my original suggestion that your argument from MR presupposes moral values not shared by all? (I am not suggesting this is incoherent, just checking.)

      If you were ok with that, then it would mean that your argument for MR actually depends on assumption that some moral directives/values are universal. Which is kind of cool.

    4. Yes! Or, the way I was defending premise 2 depends on something like the (near) universal reflective endorseability of values or commitments related to agency.

  6. Josh, I guess I was mainly trying to show that the argument is able to avoid the conclusion that people had the moral authority to do nasty things like beat their wives. I tried to do that in points 3 and 4 in my response to you. Maybe it is a significant bullet to bite that, regrettably, it turns out that not everyone has reason, a priori, to avoid doing things that I think are awful (where, again, their actually having a reason is distinct from whether I think it would be ok to prevent them from doing it). But I’m not sure. It is certainly more a morality without romance.

  7. Kyle, Thanks for a thought-provoking article. I’m inclined to agree with Randy’s observations. Your argument seems to presuppose that some sort of claim about the value of individual autonomy is one that every person ought to endorse. (If your assumption is that every person does in fact endorse it, I suspect this is empirically false.)

    A difficulty I have with your argument probably just reflects my naivety or lack of familiarity with some of the recent literature relating to this topic. But I don’t really “get” the first premise. You say:
    M is a distinctively moral directive only if it provides reasons for acting.
    OK. So suppose I say to someone:
    “You ought not to kill that man.”
    Or “You ought not to have sex with that woman if she doesn’t want it.”
    Or, “You ought not to boil that lobster alive.”
    On the face of it these look like moral directives. If they aren’t, then I’m not sure what you mean by the term “moral directive.”
    None of them include a reason why they should be obeyed, so in that sense they are moral directives that don’t provide reasons for acting. Presumably, then, by “provide reasons for acting” you mean one of these:
    a) reasons for following the directive could be given
    b) reasons for following the directive could be given, and these reasons will in fact motivate the person the directive is addressed to
    c) reasons for following the directive could be given, and these reasons ought to motivate the person the directive is addressed to
    My sense is that you mean either (b) or (c), but I’m not quite sure which. The problem with (b) is this. If I say, “Don’t boil that lobster—doing so will cause it pain.” And you say, “Pah! Lobsters can’t feel pain” or “Pah! Lobsters don’t matter.” then obviously my reason doesn’t motivate you. Yet I can’t see why that should lead to the conclusion that what I said was not a moral directive.
    Option c) amounts to saying that the person addressed should ditch some of their beliefs and embrace some of mine (the speaker). I don’t have a problem with this sort of explicit ethnocentrism . But I don’t know well it sits with your premise 2:
    “R is a genuine reason for acting only if it is capable of motivating the agent”
    The exprression "genuine reason" looks normative; but the rest of the premise looks descriptive rather than normative. It suggests that if my reasons leave the addressee cold, they aren’t “genuine reasons”.
    I suppose at bottom, I’m just confused over the attempt to forge conceptual links between whether something is a “moral directive” and the speaker’s ability either to give reasons for it or his ability to motivate the person addressed. Jonathan Haidt notes that most people when asked why it’s wrong for siblings to have protected sex just once as an experiment, are “dumbfounded.” They just say,” It’s wrong!” Is that a reason? If not, is the statement, “Don't commit incest” not a moral directive?

    Emrys Westacott

  8. Thanks, Emrys. About your first question I'll just repeat what I said in response to Randy. Part of the argument depends on the (near) universal convergence on values or commitments related to agency. I would be surprised by evidence that shows it's false that people want others to refrain from interfering with them in the pursuit of their projects and ideals.

    Your second question about premise 1 spills over into your doubt about premise 2. But, first, 1 is conceptual. Here's Darwall's statement of it: "If S morally ought to do A, then necessarily there is a reason for S to do A consisting either in the fact that S morally ought so to act, or in considerations that ground that fact." Whether or not something is a moral directive or not, then, doesn't depend on anyone's ability to *give* reasons for it or their ability to motivate anyone. It depends whether there are reasons (with significant deliberative weight) to act that way.

    But *what makes it true* that there are reasons (distinctively strong ones, at that) for S to do A? 2 offers an account. Here's the Stanford Encyclopedia summary of Williams:

    "the internal reasons thesis is not the view that, unless I actually have a given motive M, I cannot have an internal reason corresponding to M. The view is rather that I will have no internal reason unless either (a) I actually have a given motivation M in my “subjective motivational set” (“my S”: 1981: 102), or (b) I could come to have M by following “a sound deliberative route” (MSH 35) from the beliefs and motivations that I do actually have—that is, a way of reasoning that builds conservatively on what I already believe and care about."

    So above, in response to one of Randy's question, I noted that there is often some deliberative distance to travel between people’s actual, occurrent beliefs and desires and what they have genuine normative reason to do given their fundamental commitments. And there's definitely room for debate about how far it's ok to go along this deliberative route from someone's actual motivations and still legitimately say they still have the reason, but (according to the view) certainly if there's no such route it's not true to say that they have the reason.

    So maybe think of 2 as a kind of justificatory test for a putatively moral directive. You could issue a directive using moral language where the directive doesn't pass that test. You're then mistaken -- the person you direct the claim at doesn't have that kind of reason to act as you say. In cases where you know that, you're just bluffing -- using moral language in a subtly coercive way to get the person to act the way you want her to.