Bryan Caplan  

Bleeding-Heart Libertarianism and Its Critics: Preliminary Mediation

David Friedman on Bleeding Hea... My Thoughts on Technology and ...
Jason Brennan replies to the critics of bleeding-heart libertarianism, most notably David Friedman.  David feels like he's nailing jelly to a wall:

My complaint about the BHL, as may be obvious from the exchanges now going on, is that they insist that social justice ought to be part of libertarianism but are unwilling to tell us what it means. As far as I can judge by observations of usage, "social justice" means "ideas of justice that appeal to left wingers," and its practical implication is the rule that, with regard to any issue at all, the first question to ask is how it affects the poor.

The only reason I can see why libertarians would want to adopt that terminology is to appeal to leftish academics. Fraudulently.

Jason finds David "obtuse":

I think it's weird for David Friedman to say he doesn't know what we're talking about, since we've defined the term many times here. Let's define it again.

Here's a rather generic definition:

Social justice is a moral standard by which some people judge political and economic institutions. Advocates of social justice believe the moral justification of our institutions depends on how well these institutions serve the interests of the poor and least advantaged. The basic institutions of society must sufficiently benefit all, including the least advantaged and most vulnerable members of society.

I don't speak for David, but I understand why he remains dissatisfied.  Does "depends" mean "depends to some extent"?  Almost every moral theory says the same - including, as David points out, old-school utilitarianism.  Does "depends" mean "depends entirely"?  That seems implausibly absolutist - especially since "serving the interests of the poor and least advantaged" is (a) arguably supererogatory in the first place, and (b) dependent on how deserving the poor and least advantaged are.

Jason's last sentence is more specific - and sounds ominously close to the absolutist position.  When as careful a philosopher as Jason includes the clause "must sufficiently benefit all, including the least advantaged and most vulnerable members of society," the natural interpretation is "if institutions are not sufficiently beneficial for the least advantaged and most vulnerable members of society, those institutions are definitely not morally justified."

If Jason and David will permit me to play the role of mediator, here's where I'd start.

1. David, are you willing to grant that Jason and others have tried to tell you what "social justice" means?

2. Jason, are you willing to refine your definition for David?  In particular, are you saying anything stronger than a utilitarian would accept?

Comments and Sharing

COMMENTS (22 to date)
Yuri writes:

Good on you, Bryan.

J Storrs Hall writes:

There's another elephant walking through the holes in that definition. Who is to decide what "benefit" is? Current lefty practitioners seem to feel that the poor are benefited by dependency and enormous incentives to pathological behavior. I see no reason to believe that the social welfare state is anything other than a giant bird feeder, erected for the emotional benefit of the bleeding hearts (their term!) and NOT that of the poor at all. I would go so far as to claim that the practice of "social justice" is largely responsible for the condition of the underclass today.

And yet it is just as invalid to ask the poor themselves what they need. It'll be handouts every time. Nobody wants to be forced to make the hard choices, do the hard work, take on the hard responsibilities.

It's valid, even noble, for a libertarian to be concerned over the fate of the poor. But as Einstein said, solving the problem will require different thinking than that which created it.

Greg G writes:

The devil is always in the details. I don't see why this is a bigger problem for BHL than anyone else.

No one has been able to define "aggression" or "non-aggression" in a way that actually solves difficult problems. Everyone in a dispute usually simply claims the other is the aggressor.

And so we have libertarians who think non-aggression means abortion should be legal and other libertarians who think the state should force a woman to carry her rapist's baby to term.

We have right libertarians and left libertarians. We have anarchists and Republican presidential candidates all claiming to be libertarians defending the non-aggression principle.

But, by all means, lets pretend this definition issue is just a special problem for BHL.

Thucydides writes:

We should not forget Hayek's point that to talk of social justice is to make a kind of category error. If society were planned and designed, then it might be appropriate to talk in such terms. But given that society as it exists is the result of a myriad of unplanned voluntary interactions, it is nonsense to demand that it somehow meet an abstract standard of justice. In any case, nobody could ever agree on what that would look like. To talk of social justice is simply to talk of preferences appealing to the political left.

Javier writes:

It seems to me that Jason Brennan and John Tomasi have already sketched an answer to:

"are you willing to refine your definition for David? In particular, are you saying anything stronger than a utilitarian would accept?"

See their piece on neoclassical liberalism.

Here is a quote:

Neoclassical liberals are not material egalitarians, but instead... sufficientarians, and/or prioritarians.... Sufficientarianism is the thesis that all people should have enough to lead minimally decent lives. Neoclassical liberals advocate market democratic institutions in part because they believe these institutions will tend to satisfy this condition of material adequacy. Prioritarianism is the doctrine that when considering changes to current institutions, all things being equal, we are required to give more weight to the worst off members of society than the better off members.
So the answer to the question is: BHLs are typically sufficientarians or prioritarians. But it seems that they could in principle also be utilitarians.

Although this part of the view seems clear enough, I think there are still unanswered questions. Here is one. Are BHLs making claims about what would be morally best? Or are they making claims about what people are morally required to do? For instance, most libertarians probably believe that it would be morally better if a free market benefited the poor and that this would count in favor of free markets. But how does this fact link up to anyone's obligations? I can think it is morally good that, say, there are more people in the world (like Bryan does) without thinking that I am morally required to have more children. So far, it seems that BHLs have spent little time answering the question about obligation. Sure, it would be good if free markets benefited the poor. But who has the responsibility to ensure that free markets do benefit the poor?

David Friedman writes:

"David, are you willing to grant that Jason and others have tried to tell you what "social justice" means?"

No. I can't tell if they are deliberately avoiding doing so, sufficiently unclear about their own ideas not to realize that what they are offering so far is mush, or honestly offering an explanation that I am unable to follow. The longer the process goes on, the more plausible the first alternative becomes.

Greg: The problem isn't that the BHL folk can't give a precise answer, it is that, so far, they give no answer at all. On one reading, their definition of "social justice" is consistent with views I've argued for the past forty years or so, on another it's some sort of blurred version of the original Rawls, a position I find intellectually indefensible. They won't say--at least one of them deliberately refused to say in the main discussion in response to an explicit question--whether "the poor" have a special moral status in their view, or merely count along with everyone else.

I asked:

"Do they want to argue that we must give greater moral weight the least well-off or do they want to argue that we should treat everyone with equal moral weight, which includes the least-well off?”"

Matt responded:

"This is a fine and important distinction for philosophers to make. But it’s worth noting that for most of the real world problems that the classical liberals were concerned about, it is a distinction without a difference."

And in four hundred words thereafter he never answered my question.

If the distinction between libertarianism, utilitarianism and Rawlsianism doesn't matter, why make a point of being for "social justice" instead of just "justice?" If it does matter, why won't he say which position he holds?

David Friedman writes:

In response to Javier's quote:

"Neoclassical liberals are not material egalitarians, but instead... sufficientarians, and/or prioritarians.... Sufficientarianism is the thesis that all people should have enough to lead minimally decent lives. "

"Minimally decent lives" is mush. Somewhere, one of them included health care in his definition thereof. By that definition, not a single human being prior to 1900 lived a minimally decent life, since nothing was available that would satisfy that requirement in the view of a modern.

And it's mush which, taken seriously, has unreasonable implications. Suppose that, due to my undeserved misfortune, the only way I can lead a "minimally decent life" requires other people to give up a hundred billion dollars a year worth of resources that would otherwise go to their purposes. Are they required to do so? If they don't, is the society unjust?

To avoid such problems, you need something like a watered down egalitarianism, where what I am entitled to depends on the resources other people have. But they don't seem willing to come out and say so, at least not in the main discussion—I haven't read everything the BHL folk have written.

"Prioritarianism is the doctrine that when considering changes to current institutions, all things being equal, we are required to give more weight to the worst off members of society than the better off members."

But, at least in the main discussion, they won't say whether they are prioritarians. And the quoted definition leaves open the question of whether, if they are, they are giving more weight in any sense beyond that in which utilitarianism already does--when, and only when, it follows from declining marginal utility of income. And if so, why.

It's worth remembering that, as I pointed out some time back in the main discussion, Harsanyi came up with the original position about twenty years before Rawls did, and correctly worked out the implication, which was utilitarianism.

Javier writes:

I appreciate David Friedman's response. Let me try to defend BHLs from some of his criticisms.

Friedman claims that sufficientarianism is implausible in part because it is hard to define what an adequate threshold would be. But maybe it is not impossible. At least, I don't know how we can rule this possibility out a priori.

But I'll put this point aside. Many BHLs are prioritarians, not sufficientarians. John Tomasi in his recently book explicitly says that he endorses prioritarianism. See Free Market Fairness, chapter 6. Prioritarianism says that it is in itself better to benefit the worse off, which utilitarianism would deny (this is different from the idea of diminishing marginal utility). This is also not a version of egalitarianism. Prioritarianism only focuses on the absolute holdings or well-being of the poor and not their relative position.

There are different versions of prioritarianism. Some of them are quite implausible, like the difference principle. As Friedman points out elsewhere, the difference principle has absurd implications. But other less extreme versions of prioritarianism are more plausible. I would recommend Derek Parfit's influential discussion of prioritarianism for more on this.

R Richard Schweitzer writes:

Gee! It's sad to read all this intellect spent on the rhetorical use of a term. Even the reference back to Rawls. . . .

Nozick was on to the course Rawls was taking for a philosophical affirmation of what Hayek referred to as constructivism. Nozick laid out the roads roughly available for course choosers. Unfortunately, but for good reasons, he never returned to pave them.

Rawls, like many, many others, was convinced (or so it seems) that the concepts of what is “just” or results in “justice” within a social order develops in its members from the way in which that social order comes to have a particular structure. [Here, refer to Hayek, et al. on spontaneous order]

Thus, if another set of concepts of “just” and “justice” which is appealing because it might provide “treatments” for the imperfections of the relationships in the social orders we know, the way forward is to consider constructing a social order on a tabula rasa format with the objective of producing that different set of concepts in its members. [Now why can that not be congruent with spontaneous order?]

Some may now hear the faint wind-chimes melody of “the general Will,” that dirge of individual freedom.

Tracy W writes:
Advocates of social justice believe the moral justification of our institutions depends on how well these institutions serve the interests of the poor and least advantaged.

I don't think this makes sense, at least as an absolutist rule.

It has always struck me that the really disadvantaged in society are those who are both poor and have serious mental problems (such as paranoid schizophrenia). I would certainly prefer to be poor and reasonably sane to being poor and being wrongly convinced that everyone is trying to kill me, and indeed I'm not certain that I wouldn't prefer being poor and reasonably sane to being rich and being wrongly convinced that everyone is trying to kill me.
Do therefore institutions like marriage, or an independent central bank, really depend on their justifications for whether they serve the interests of the poor and least advantaged of society? Or are they justified if they benefit most people, even if a poverty-stricken paranoid is incapable of contracting a marriage and can't handle any amount of money, regardless of what the inflation rate is?
Now, perhaps you can run an argument that marriage and independent central banks, by improving the ability of the slightly-more advantaged to help the most mentally ill, do advantage the least-disadvantaged. But then you could run that argument about anything that is generally beneficial to society, so why introduce the "social justice" rule?

Matt Skene writes:

I'm not too familiar with BHL, but I find the following view intuitively compelling:

(1) There are certain conditions where people's need is sufficiently great and the availability of alternative options sufficiently scarce that it is permissible for someone to steal in order to provide for those need. (Think of classic examples like stealing bread to feed your starving family, stealing medicine for your dying wife, etc.)

(2) Societies that fail to provide adequate alternatives to theft as a means of meeting one's basic needs are unjust societies.

(3) Many standard libertarian views advocate societies that remove current social welfare programs that are used to meet needs in these sorts of circumstances without providing viable alternatives to satisfy those needs.

(4) So, many standard libertarian views advocate unjust societies.

On this account, social justice consists in providing viable methods for meeting the sorts of needs that would, in the absence of those methods, make criminal action morally permissible as a means of meeting them.

This obviously doesn't provide a clear answer in every case, since whether or not theft is permissible in certain cases is highly debatable, but it seems to provide non-mushy content to "minimally decent lives" and therefore likely captures much of what BHLs are after. It also seems to provide a clear battle-ground for discussions of what needs to be done to create a just society, and captures both the social justice intuition that a just society protects people from dire circumstances and the libertarian intuition that government action is typically immoral, and therefore requires strong overriding considerations to be justified.

Tracy W writes:

Matt Skene:
I think the obvious response from libertarians is that they also support efforts to increase the ability of society to provide adequate alternatives to theft (eg by freeing up immigration, reducing taxes, eliminating occupation licensing).

You also omit, in your point 1, that if necessity compels theft, it also is morally obligatory to repay what the person stole. You also talk about stealing products, but you don't talk about stealing time - does a husband have a right to hold a surgeon to gunpoint to get them to perform surgery on a dying wife? And if not, then why is it permissible to steal a pharmacists' time as embodied in a medicine?

Matt Skene writes:


I realize that most libertarians support those efforts. However, the view that we should do something about these issues is compatible with two very different views. I was assuming the main point of debate between BHL and standard libertarianism had to do with necessary conditions for a just society. Although both parties should support non-governmental alternatives to providing for these needs, a view that takes justice to include a requirement to provide for them will think the establishment of those alternatives is a prerequisite for eliminating governmental solutions, while one who viewed such alternatives as a desirable, but not a required aspect of a just society would think we should get rid of the governmental programs even if we don't have those alternatives in place. A difference of opinion on this amongst libertarians would justify a debate over whether or not social justice should be part of libertarianism. Since these intuitions are part of common sense morality, it does so independently of a desire to appease leftish academics. I gathered that that was what David Freidman was asking for: an account of social justice that provides a motivating reason for libertarians to make it an integral part of their view. That's what I was trying to provide on behalf of BHLs.

As for your question about time, I guess I don't see why it would be any different. When people call out "is there a doctor in the house!?!" during an emergency, it is expected that any doctor will come forward and do what they can. If the doctor absolutely refuses to help despite being the only one who can, I see absolutely no problem with putting a gun to his head and forcing him to. Coercion is rarely justified, largely because situations with no viable alternatives rarely occur, but when it is the only possible means of providing for extreme needs, it becomes permissible, just like theft does.

Dan Carroll writes:

While I don't have time to get out my dictionary to figure out what all of these labels mean, I'll offer my two cents (if you find it is worth three cents, then you have enjoyed consumer surplus).

As a pragmatist, I believe that most institutions, including governments and markets, are organized for the benefit of the rich and the powerful. As such, most institutions distort and provent the market ideals proposed by libertarians. Thus, the distinction between government and markets is far less clear than it first appears. Therefore, there are individuals who are left out through no fault of their own. (While some may disagree, I generally believe that losing the genetic or epigentic lottery constitutes no fault, as well as unjust circumstances - up to a point.) "Social justice" seeks to rectify those disparities through an implicit recognition that no institution will be perfect, and in fact most are damaged beyond repair, because that is the nature of human society - the rich will exploit because they can.

But then, I'm a pragmatist, not an ideologue.

R Richard Schweitzer writes:

@Dan Carroll-

To you, as a pragmatist, consider:

Institutions arise or develop from civil (non-governmental) instrumentalities established to pursue objects determined by individual or groups within a "society." E.g., those who sought to establish means for learning by children and further learning by others. Their efforts created the schools, for which they engaged others to carry out their objectives. Those objectives generated capital needs and schools were co-opted into the political system. There the schools became an "educational system," and by that institutionalization those previously engaged to satisfy the objectives of others, have come to be the ones who set the objectives. That is the course of institutionalization, including governments - when those whose functions were to attain the objectives of others take over and determine the objectives of the organization.

Tracy W writes:

Matt Skene - Yikes! You'd not only put a gun to the doctor's head, but you'd have no problems in doing so?

I think our moral intuitions are massively totally different. Out of curiousity, in this hypothetical situation, how do you think the doctor should be compensated for the coercion they've suffered?

BTW, the difference between your scenario and calling out "Is there a doctor in the house?" is that the doctor is not obliged to respond, or responding, is not being forced to do his job. (Note, my brother's life was likely saved after a bad cycling accident by a passing off-duty doctor who kept his airways open until the ambulance got there, we are deeply grateful to him even though we don't know who he is.)

Matt Skene writes:

I don't understand your intuitions. Perhaps some other cases could help explain where we differ.

Case 1: You are in a wheelchair, and you come across a child drowning in a shallow pond. You are unable to save the child because of your condition. The guy standing next to you could easily walk into the pond and save the child from drowning, but is such a moral scumbag that he refuses to do so. You have a gun. Is it permissible to threaten to shoot the guy if he doesn't help voluntarily?

Case 2: Someone is choking to death in front of you. You don't know the heimlich maneuver, but you know the guy standing next to you does. He's such a scumbag, though, that he doesn't want to bother using it to help the person choking to death. You have a gun. Is it permissible to threaten the person if he refuses to help?

I think the answer in both cases is obviously "yes." In fact, I think it's not only permissible to do so, it might be obligatory to do so. Emergency situations require those in a unique situation to help at no risk to themselves to do so, and if they refuse, you are entitled to force them to do so.

If you aren't convinced, consider the following scenario:

Case 3: You are walking past an elementary school with hundreds of children in it. You see a bomb with a timer on it set to go off in a few minutes. The guy you are with is a bomb expert and could easily diffuse the bomb well before it goes off. You wouldn't know what you are doing, though, so you can't safely diffuse it. The bomb expert refuses to help for no good reason. Should you threaten him to force him to diffuse the bomb?

If you think the prohibition against coercive force is so strong that it requires you to let a building full of school kids die, I'm not sure what I can say to you. If you don't, however, then you must think that some emergencies permit rights violations, including the threat of violence in order to avoid terrible consequences. Where the line is for those consequences is certainly debatable, but I think that the threat of immediate, needless, undeserved death in pretty much any case creates a justification for such coercion. Since I rarely have a problem doing what I know to be right, I wouldn't have any problem threatening people in any of these situations. To put it another way, if I'd been by that doctor who helped your bother and he had refused to help, I would have threatened him. And if I hadn't done so, I think that not only would you have a right to think he had done something terribly wrong, but you'd be right to think that I had, also.

Part of being a minimally decent human being is being willing to help in an emergency where you are needed and you can help at little comparative cost to yourself. People who aren't willing to be minimally decent human beings in such situations have lost their rights not to be coerced in those situations, especially if there is no other option.

As for compensation, I don't think the doctor deserves any. You don't deserve compensation for being a minimally decent person, and you certainly don't deserve any for the consequences of failing to be one. In fact, I think if he has to be forced to help in such a situation, he likely deserves to lose his medical license and to be reviled by the community for his indifference. Labor isn't sacrosanct. In some situations, you just have to pitch in and shouldn't expect any compensation for doing so.

Scott Scheule writes:


On a tangent, these sort of examples can be used to disprove any rights claim. The form of the example is typically: terrible consequence C will ensue if you don't violate someone's right R. Then it's just a question of ratcheting up the terrible consequence to such an extent (planet gets destroyed, a billion children get tortured for a thousand years, etc.) that one's intuitions favor violating the right.

Steven Kopits writes:

As I understand it, the libertarian (liberal) concept of justice is narrow, namely, that property rights are vested in the individual and that all individuals enjoy equal rights, which are essentially negative in nature. For libertarians, justice is blind; there is no "social" justice as a conceptual matter. Being poor doesn't get you special treatment (nor, conversely, does it allow you to be discriminated against.)

Now, problems arise in the case of endowments, whether of wealth, beauty, intelligence or any other attribute. We would wish these issues away, but they have a habit of re-asserting themselves.

Take, for example, education. The benefits of education are largely appropriable, that is, they accrue to the person receiving the education. Thus, they are materially a private good. However, the state provides universal primary and secondary school education. Should we do away with this? After all, the benefit is going to the student. If someone can't afford education, well, tough luck. They should have chosen better parents.

Or do we tax property owners (heavily) to ensure that all have some rudimentary education? If we do so, I might argue this is closer to a lefty definition of "social justice". We are compensating for issues associated with variability of initial endowments of intelligence, wealth, talent and parentage. We are burdening the intelligent and industrious and lucky with taxes to pay for the poor, mediocre, and the unmotivated. Surely this is not a libertarian policy.

So, as libertarians, what do we do?

Scott Scheule writes:


I suggest we do what everyone else does and pick some point on the continuum between 1. Assuming people are entitled to none of their chance endowments and 2. Assuming people are entitled to all of their chance endowments. And we aim for that. Both extremes are uncomfortable, but there's no principled way to pick a point in between. What more can we do?

The only other option is what Mises once called "balls to the wall" libertarianism, which involves assuming people are entitled to absolutely everything they're born with, from good (or bad) looks, to rich (or poor) parents, to being born in a country that's going to be above water in ten years (or the Maldives).

Matt Skene writes:


That's right. Examples like this seem to show that there are no absolute rights. At some point, consequences begin to justify rights violations. My overall suggestion for this debate was that this is the point where one can draw the "minimally decent lives" line that David Friedman called "mush."

Scott Scheule writes:


Quite right. Perhaps David finds your suggestion illuminating.

Comments for this entry have been closed
Return to top