A virtue theory of global justice? (random thoughts)

One of the more fashionable areas in the amorphous boundary between political philosophy and political theory right now is global justice… the idea being that our duties of domestic justice and global justice are different, and it’s about time we sorted out the latter. I confess that I find most of this literature utterly boring, because I have very strong cosmopolitan intuitions, and so endless dithering about the global institutional landscape, and conflicts between global and domestic justice, and the duty of assistance, and all that stuff just make me want to point to Peter Singer and say “go listen to that guy some more!” And it’s not that I agree with Singer’s approach (as a non-utilitarian, I’m more of a Beitz kind of a boy), but that the intuitions to which he appeals are so compelling that I don’t understand why those who work on global justice can do so without seriously confronting them instead of appealing to the way Rawls, then Nagel, poisoned the well with all this stuff about institutions and basic structures as the proper subject of justice claims.

But perhaps people would take cosmopolitanism more seriously if it were grounded in something other than Singerian utilitarianism or Beitz and Pogge versions of Rawls that claim there’s a global basic structure (as if that’s what matters).

I don’t think I’m the person to write it, because I have too many other projects on my plate, and also because I don’t know if I swallow virtue ethics (though if anyone wants to co-author something and has a very long timeframe…), but it seems like someone ought to work on grounding cosmopolitanism in a virtue theory.

Would the phronimos prefer his fellow citizens to the starving abroad? I think not.


30 Responses to “A virtue theory of global justice? (random thoughts)”

  1. Matt Says:

    Most virtue ethics types also tend to lean towards communitarianism, and communitarians, of course, tend to be pretty anti-cosmopolitan. So, I’m skeptical that there are lots of people working this way. That’s not to say it can’t be done, but it would be swimming against an awful lot of the literature on virtue ethics and how it works.

  2. ben wolfson Says:

    The Aristotelian phronimos probably would.

  3. Paul Gowder Says:

    True enough guys… but I wonder to what extent that’s just an artifact of the way in which virtue theory was developed. There isn’t an essential communitarian property of virtue theories.

  4. Mike Says:

    Cosmopolitanism, to me, is intellectual hubris to the maximum. And futile.

    What can do we do? Imagine we give a bunch of money to an sub-Saharan African country. Some warlord will take it. Some research has been done showing that giving money is actually a waste.

    So we go to a country that needs justice and do what? Impose our Western values on them?

    When Ann Coulter said we should go to Iraq and convert everyone to Christianity, people flipped. Yet that’s a form of cosmopolitanism. And, indeed, while Christianity ain’t great; it seems better than the Islam they’ve got going on there – at least if you’re a woman.

    There isn’t an essential communitarian property of virtue theories.

    Eh….. Yeah, there sort of is. Virture is relevant to the culture you’re in. Aristotle taught that we learn virtue through observation and practice. So we observe those in our communities. In a global, wired world, does that apply? I suppose that can be your new paradigm, if no one else has created it.

    I know there was quite a bit of lit showing why the connection is strong. I’ll see what I can find in Google.

  5. Mike Says:

    LOL a me, but yes, I am going to quote Wikipedia. That’s only because it reminds me of what I read in more serious sources:

    “Other proponents of virtue theory, notably Alasdair MacIntyre, respond to this objection by arguing that any account of the virtues must indeed be generated out of the community in which those virtues are to be practiced: the very word ‘ethics’ implies ‘ethos’. That is to say that the virtues are, and necessarily must be, grounded in a particular time and place.”

    Now, I haven’t read seriously into these issues since 2001 – well before the world got as small as it is today.

    So there might be a Universal Virtues movement in virtue ethics. I don’t know. I do know that what I quoted was the view when I thought about these issues.

  6. Daniel Goldberg Says:

    What Matt said. Virtue ethics also trend heavily towards the particular and the local — one reason virtue ethicists and care ethicists get alone famously — which means that the idea that we owe equal duties to strangers as we do to intimates is not likely to get much traction with a virtue ethicist, IMO.

    I’m both very attracted to virtue ethics and very attracted to social justice, as you well know, but I am not enthusiastic about either deontological theories or rights talk as adequate theoretical groundings for either.

  7. Paul Gowder Says:

    Mike, the claim that virtues are generated by a community does not entail the claim that the virtuous are only or primarily interested in those communities. A virtuous citizen of a specific community might, in instantiating the virtues to which she was raised, be disposed to aid foreigners.

    Also, cosmopolitans today recognize that implementing our cosmopolitan ideals is a complicated task, not just handing out sacks of money. That doesn’t mean it’s not worth doing.

    As for the cultural imperalism question, I think it is overblown. Much of it can be separated from issues of distributive justice: we need not tell a society to stop being Muslim to help it feed its poor. We might need to tell it to stop oppressing people (including women — see the research on the inverse relationship between women’s rights and famine) so badly, but how much damage would that really do? It isn’t “kill them and convert them to christianity” territory by a long shot. I think we tend to overestimate the extent to which elements of a culture are bound together, as if it’s impossible to be a Muslim without oppressing women.

  8. Paul Gowder Says:

    Daniel, but why should that be the case? That is, why should the virtuous disposition to care for others be localized that way?

  9. Daniel Goldberg Says:

    Because virtue ethics is all about practices, and practices are inevitably local. We may try to have an influence on global events, conditions, and ideas through our actions, but it seems implausible to me to argue that we generally can do so by any means other than acting through local, particular communities, and networks.

    Even if one emulates Paul Farmer and travels to Haiti and Russia out of a deep commitment to local justice, the acts Farmer performs are still deeply local and particular — this time local and particular to the Haitian and Russian communities he is moving through.

    Does a virtuous disposition logically HAVE to be localized that way? I don’t know, and I don’t think I’d need to defend such a claim. My claim was simply that virtue ethicists tend to be vastly more interested in these local, particular practices. So was W, for that matter, which is what makes his move from the Tractatus to the PI so fascinating; he abandoned his search for an epistemic Esperanto, and instead grounded his notions of meaning in irreducibly local, particular practices and forms of life.

  10. Matt Says:

    You might also want to think about differing degrees of cosmopolitanism and universalism. The idea that all duties apply equally to everyone is pretty implausible and not one that anyone seems to actually hold in real life, so the question then is which duties are universal. I’m slowly reading a book by a german philosopher, Otfried Hoffe, _Kant’s Cosmopolitan Theory of Law and Peace_ that has some interesting discussion on this. (It’s translated in a stilted, overly-literal way that makes reading it a bit unpleasant, though.) Perhaps most relevant to you would be the discussion he has in the start, “Aristotle instead of Kant?” where he argues that there is a universalist core to Aristotle’s ethics. (He favors Kant in the end, and I can’t say I found his discussion extremely clear, but there might be something there to work with.) You might also look at David Miller’s new book, _National Responsibility and Global Justice_. It’s a good book in general but has an interesting discussion about which duties are generally universal and which are more local and why. I don’t think his particular account is right, and it’s more Hegelian than Aristotelian, but it might be worth looking at to see how he distinguishes local from universal duties and where you disagree with that and then see if you could work that into the framework you wanted. This is very much unlikely to get you anything like the position of a global utilitarianism as far as duties, though.

  11. Mike Says:

    we need not tell a society to stop being Muslim to help it feed its poor.

    How about sending women to school?

    What about societies where there are caste systems?

    But let’s get more basic: You want to feed the poor. Sounds great in theory.

    In order to feed the poor, you need logistics. Logistics is what separates theory from practice.

    Who is going to hand out the food?

    If you consider the full logical implications of that question, you’d see why so many of us would prefer to keep our efforts local.

    You think we can just show up and start handing out food?

    And things get even more complicated once you go beyond food and start talking about jobs, education, etc.

    Hell, look at prisons. Let’s give prisoners access to education. Sounds great!

    If you’re a prisoner, try enrolling in a class. See what the gangs do to you.

    Idealized notions that we should all agree on quickly get torn to shreds once you consider their logistical implications.

  12. Paul Gowder Says:

    Daniel: Aah. I see where we differ. I see virtue ethics as being about dispositions rather than actions, so it seems perfectly sensible to say that a virtuous person is disposed to aid everyone without regard to nationality etc. while still acting from that disposition on a local level. The material for that is already in our moral culture and folk wisdom (a fact that’s surely highly relevant to virtue ethicists) – think of “think globally, act locally.” While permitting lots of practical localism, this would clear out a bunch of superstition about special regard to co-nationals, as well as demand that local action be taken with regard to the interests of non-local people. (A cosmopolitan virtuous citizen, for example, would doubtless be disposed to buy fair trade coffee.)

  13. Paul Gowder Says:

    Matt: Thanks. Hoff’s book sounds particularly interesting, despite the usual problems with translations from german. It’s now on my reading list (albeit pretty far down, since I have active projects that need lots of attention).

    How much new ground does Miller cover? I confess to having a very negative reaction to his work on global justice in general (particularly the business about people being unable to develop attachments to foreigners sufficient to satisfy justice claims, which just seems to me to be inappropriately naturalizing the current political arrangements of nation-states), though if anything that suggests that I ought to be more rather than less disposed to read the latest.

  14. Paul Gowder Says:

    Mike: “X is difficult” rarely allows us to infer “so we don’t have an obligation to try to do X.” I agree with you that delivering global distributive justice would be extremely difficult. Most cosmopolitans (with the exception of Pogge) agree with you on this. But that just suggests that we should work hard to figure out how much we can do, and how we can best do it.

  15. Daniel Goldberg Says:

    I don’t necessarily disagree with what you say here, though to see some arguments why virtuous dispositions themselves ought to be local rather than universal, you might take a look at some of the care ethicists’ work, in particular Virginia Held (BTW, her new book is reviewed in the latest Ethics). She argues that the problem with the rationalist (whether Kantian or consequentialist) notion that all moral agents should be treated equally is precisely that it means the mother on the about-to-crash-bus ought to value the well-being of each of the perfect strangers as much as she values the well-being of her own child.

    This is not a criticism about whether the utilitarian can get the right answer (they typically can), but a much more subtle and powerful critique of the meta-ethics itself which posits that intimates do not have any greater moral pull on us than perfect strangers. While Held is speaking of acts, I’m reasonably confident her argument also applies to dispositions, such that by a similar chain of reasoning she would claim that a virtuous agent ought to be disposed to prioritize the needs of intimates over the needs of strangers.

    All of this tends to capture Matt’s sentence: “The idea that all duties apply equally to everyone is pretty implausible and not one that anyone seems to actually hold in real life, so the question then is which duties are universal.”

    Believe it or not, I discuss exactly these issues in a paper coming out in Pierce Law Review sometime soon, which addresses a rights-based justification for universal health care.

    I’m also assuming you’re aware of Peter Unger’s 1996 book on this topic?

  16. Paul Gowder Says:

    Daniel: on the Held line of argument, I think many move too quickly from the relationships like mother-child to relationships like citizen-cocitizen. Part of the reason I object to a lot of the local duties literature is that it seems to naturalize some very contingent social relationships. There’s a necessity (though not an immutable one) to the mother-child relationship, to the genetically determined (and good) emotional connections, etc., that simply doesn’t exist in nearly as strong a sense between cocitizens.

    Haven’t read Peter Unger’s book, but been meaning to…

  17. Mike Says:

    Mike: “X is difficult” rarely allows us to infer “so we don’t have an obligation to try to do X.”

    I agree totally.

    Yet “X ain’t going to work” does allow us to infer that, “We shouldn’t do this,” or, “This would be wasteful.”

    Some studies on aid to Africa has shown that less than 10% of aid actually makes it to the people who need it. The other 90% is lost through entropy, waste, or payoffs to war lords.

    In a sense, this discussion reminds me why I didn’t pursue political philosophy. The tough stuff isn’t being done in class rooms. The really tough issues are the logistical ones.

    Theory guys (not talking about you, just the field more generally) fancy themselves quite smart and above the fray. Yet theory is actually quite easy to do. You sit around, make some arguments, intimidate some people, form some alliances, and then you’re a big star. Great. Now what?

    “Feed the poor: It’s the right thing to do.” That’s easy. That’s not even an overly controversial theory.

    It’s getting the food to the poor that is the problem.

    So, to me, a theory is not even valid unless it’s workable.

    Now, I’m not some guy who hates knowledge. There is something to be said for knowledge for the sake of knowledge. Or theory for the sake of theory. That doesn’t work here, though.

    Political theory or ethical theory isn’t quantum mechanics or metaphysics. Political theory is supposed to be about concrete situations – ultimately, people. So if a theory does actually work, then it’s not actually valid.

  18. Paul Gowder Says:

    Mike: You sit around, make some arguments, intimidate some people, form some alliances, and then you’re a big star.

    Do I have to intimidate some people to get a good career in this thing? Oh goodie. I do so enjoy intimidating people.

    In all seriousness though, there’s a division of labor. Nobody can be expected to answer every question, and the questions about what one ought to do are often just as difficult (often much more difficult) than how it is possible to do it. Why is it an objection to work on the first that work is also needed on the second?

    (As for the relationship between practicality and correctness of a normative theory, I can’t do better than to recommend some of Gerry Cohen’s work, starting, perhaps, with “Facts and Principles.” And also to say that one disease of our social thought in general, which political theory can help cure, is the belief that our existing social institutions and relationships and ideas and dispositions, which are barriers to achieving justice, are immutable.)

  19. Mike Says:

    Why is it an objection to work on the first that work is also needed on the second?

    We’re talking past each other. :)

    I’m saying that if there is no good answer to the second question, then there is no good answer to the first. Or, if you can’t make the first happen, there’s no point in even discussing the first.

    Here is how I look at it…. I’m a libertarian, as you know. I know a lot of libertarians. We don’t eat babies for breakfast, or laugh at poor people. (We leave that for the liberals to do.)

    If government actually worked, most of us would be more Rawlsian. I’m telling you: We just would be. We don’t trust the government, though, and thus any redistributionist theory fails.

    Take something like the bailouts. The government gave $20 billion to Wall Street. %15 billion went to rich executives who made large campaign contributions.

    Much of government is like that. My taxes go to “feed the poor.” Yet people set up non-profits, give themselves and their friends nice salaries, give money to people like Charles Rangle…. All of that happens before any poor people get fed.

    So does it even make sense to talk about John Rawls in a society with a corrupt government? What good is a political theory if it cannot be a reality?

    I realize that the theory-practice dichotomy is a false one. How could we talk about a “corrupt government” if we didn’t have some theoretical model for what a just government was. Though…. I don’t imagine that, except at the margins, it wouldn’t take much theory to agree what a non-corrupt government would look like. We could all agree, I’d hope, that linking campaign contributions to government contracts is corrupt and undesirable.

    Anyhow, not trashing your entire discipline. I’m too afraid of you to that! ;)

    But I do think we too often have these hotly debated discussions on theory that, in reality, would not actually do anything. It just seems sort of bizarre to debate something that will never happen.

    It’s fun….. Fun like talking about which super power you’d like to have. Or if you’d rob banks or spy on women in dressing rooms if you were invisible.

    Yet it’d odd that in a discipline that should be focused on the real world, such fanciful discussions dominate.

  20. Paul Gowder Says:

    I don’t think we’re quite talking past one another… my Gerry Cohen plug was meant as a denial of “if there is no good answer to the second question, then there is no good answer to the first,” based on his argument for that denial. (Believe it or not, political theorists and political philosophers have spilled a lot of ink arguing over just this claim.)

    Rawls is actually the last person who ought to be criticized for this, as much of the point of his work was to suggest what a realistic utopia would look like, emphasis on the realistic.

    Even if all current work in this area is unrealistic, though, it still serves two valuable purposes. One: it offers a regulative ideal to which we can move toward. Knowing what the best society would look like might help us move closer toward a good society. Two: again, it helps us see the contingency of social relationships and practices. I disagree that “government is like” anything — government right now, in this country, functions a certain (corrupt) way. But we make a mistake if we reject theories about how government ought to be based on the claim that all governments are necessarily corrupt. (Edit: and once we recognize that, we can work toward making ours less corrupt.)

  21. Paul Gowder Says:

    Here’s another reading recommendation: this thread on Public Reason, which includes thoughts on this question from several quite smart people, plus one ninny, namely me. :-)

  22. Daniel Goldberg Says:


    Believe it or not, I agree with you, but naturalizing the relationships is not a necessaary move, and it doesn’t invalidate the care ethicists’ claims. Such relationships could legitimately be both contingent and generate powerful normative claims in any given situation.

    Whether or not all mothers and children have wonderful relationships is a separate inquiry from whether, as a meta-ethical notion, it makes good sense to require ex cathedra that moral agents rank order their duties to strangers just as they do to intimates.

    (FWIW, one of the basic lessons of a familiarity with disability studies is the notion that even well-intentioned and loving families can end up doing some pretty messed up things to their kin).

  23. Jacob T. Levy Says:

    “but it seems like someone ought to work on grounding cosmopolitanism in a virtue theory. ”

    Paul, do you have a particular reason not to think that this is what Martha Nussbaum has been up to for some time now? She denies that “virtue ethics” is a meaningfully distinct category, but I think that to those who think it *is* a distinct category, she qualifies as a virtue theorist– right?

  24. Paul Gowder Says:

    Jacob, I’ve always thought of Nussbaum as more of a perfectionist than a virtue theorist. That’s what I get out of the capabilities approach in general, at least. (I’m not super-familiar, beyond skimming a couple of papers, with how she applies it specifically to global justice.)

    It’s never been very clear to me what the border between those two categories is, though my working notion is something like “a perfectionist thinks that the ends of normative theory are at least in part bringing about some qualities in people, while a virtue theorist thinks that normative theory operates at least in part by asking what people with some qualities would do.”

    Though Nussbaum has denied it, I also think that the capabilities approach, unlike virtue ethics ordinarily conceived but like perfectionism ordinarily conceived, is a form of consequentialism — or at least has a very consequentialist flavor.

  25. Jacob T. Levy Says:

    Hm. I may be color-blind in that part of the spectrum; I can’t see that distinction at all.

  26. Matt Says:

    As least sometimes Sen seems to see his version of the capabilities approach as a type of consequentialism, if I recall correctly. I’m not really sure how that’s supposed to work since of course if the capabilities are our primary goal it can’t be a maximizing form of consequentialism, (can’t maximize for multiple values, after all) but if it’s not maximizing I’m not sure I understand it. So, I think maybe it’s supposed to be part of an indirect utilitarianism- we aim at the capabilities as the best way to maximize utility since if we aimed at utility directly we’d not maximize it. I’m not very sure about any of this, perhaps least of all if this really is, or ever was, Sen’s position, as it seemed to me to be at one point.

  27. Paul Gowder Says:

    Yeah, it’s definitely not maximizing. What I take from it is kind of a satisficing consequentialism (is there such a thing? there is now!) — the idea that there’s a value of each of N items that is the bare minimum for human dignity, and the theory directs us to achieve that value for each item, for each person.

    The reason Nussbaum disclaims consequentialism in the paper linked in my previous comment is that the capabilities approach isn’t a comprehensive moral theory meant (like, utilitarianism, say) to answer all moral questions:

    I do call it “outcome-oriented” in order to contrast the view with the Rawlsian procedural approach to justice, but it is quite important that the view is not a form of consequentialism, since consequentialism, as philosophers standardly define it, is a comprehensive ethical doctrine that proposes an overall test for all ethical choices.

    But is that what we’d need to know to answer the question? It seems to me that we can describe a partial theory like the capabilities approach, that is meant only to answer a limited set of questions, as consequentialist to the extent that it’s characterized by the standard consequentialist move of merging the right to the good and saying that the object is to bring about some amount of the good, rather than, say, characterized by side-constraints as a deontological theory would be.

    (Although I suppose it could be plugged into a deontological theory as well — one might say that we treat people as ends only to the extent they have the capabilities for functionings specified by the capabilities approach. Some of the stuff that Nussbaum says about dignity seems to want to go there. But it still seems like a consequentialist part of a deontological theory, much like the consequentialist part of Kant’s theory that says that we ought to promote the happiness of others.)

  28. Jacob T. Levy Says:

    Consequentialism of a particular sort (non-maximizing, and with plural incommensurable consequences) seems right to me about the capabilities approach broadly. But Nussbaum’s cosmopolitanism in particular seems to me somewhere in the perfectionist/ virtue spectrum (as I said, I’m not sure I can identify distinctions on that spectrum).

  29. Paul Gowder Says:

    Hmm… I obviously need to read more of her global justice stuff. In what respects does it differ from Just Another Application of the Capabilities Approach Broadly (TM)?

  30. Kony Says:

    Berkeley law/grad student here (jurisp. & social policy), endeavoring to cobble together my favorite theories into a dissertation topic. As I Googled various combinations of them, this blog post came up. I don’t know if you ended up forgetting about this idea, but I hope you don’t mind if I decide to run with it… :)

    Anyway, I am happy to see that at least one other person has had similar normative/intellectual impulses.

Leave a Comment