Illusion of Purity

The claim that there is such a thing as “pure libertarianism” should puzzle you. What is the definition of “pure libertarianism?” Do we find that definition in Rothbard? Nozick? Hayek? Mises? Rand? Which Rothbard? Nobody’s views are intertemporally consistent. If they were, our views wouldn’t evolve. The answer can only boil down to a definition which equals exactly what you believe “pure libertarianism” is, because you are going to choose between alternatives based on what you agree with. That seems a bit shallow.

Within some range, “pure libertarianism” will inevitably be impure. Few libertarians would doubt Roderick Long’s libertarian credentials, even though he prefers virtue ethics over a certain brand of deontology. Okay, Long avoids the term “capitalism” (according to his Wikipedia entry), so maybe Long is not a “pure libertarian.” But, no serious academic holds a identical set of beliefs with another. There is always some degree of reasonable disagreement. You cannot define a “pure libertarianism” without defining yourself as irrelevant, because the only “pure libertarian” will be you.

If you accept that there is a range of beliefs that can be considered libertarian, how do we determine that range of “reasonableness” (is that a word?)? Mises — except that government nonsense, but it’s okay, because it’s Mises1 —, plus Rothbard, plus Hoppe? Do we exclude Hayek, Buchanan, Friedman,2 Nozick, Gaus (?),…? Surely, you cannot define the range by what is generally agreeable to you, because that defeats the purpose of having a range of reasonable disagreement. You would have a difficult time arguing that Stiglitz’ political views ought to be considered libertarian. Hayek, though? How could the views of one of the most influential academics on libertarian thought not be considered reasonable in the debate on what libertarianism is? It would, of course, be just as absurd the other way around. How could Rothbard’s views not be considered libertarian?

The point I’m making is that any determinant definition of “pure libertarianism” is arbitrary, in that it requires the person writing the definition to choose based on her own judgment. Where is this authority derived from? What if there is some chance that she’s wrong? There is no way to even know whether we believe in “pure libertarianism,” because “pure libertarianism” is impossible to define with certainty. The concept is empty.

So far, most of you may agree with me. But, I don’t want you to agree with me. I want to give you a headache. I am going to take another step forward, making the strong claim that libertarianism as a political philosophy is too exclusive to be useful.

The limits to what is libertarianism will always be arbitrary. But, if we leave those limits up for debate, I think that debate is easier to solve than the one over “pure libertarianism.” At least, we have a fuzzy border around libertarianism. For example, it’s clear that Krugman is not a libertarian. But, Buchanan is — at least, his ideas can directly inform libertarianism. But, here, libertarianism defines only a narrow group of reasonable beliefs over the right institutions of governance. There are reasonable non-libertarian beliefs, as well.

If a group of people join into a division-of-labor, they will have to develop an arbitration process (whether planned or spontaneously) to settle disputes. There will be several orders of rules. Personal defense associations, for example, will appeal to a higher order of rules to settle inter-association disputes. If you are an anarchist who believes that people will settle into communities with more-or-less consistent political views, you should still recognize that people will want to associate with others outside of their community — there are gains from trade by extending the division of labor. People of different communities may come into disputes against each other, and there will need to be a process by which these disputes — where there are two conflicting set of rules — are solved. Let’s call the highest order of rules, and the process by which these rules emerge, the meta-process.

How do rules emerge out of the meta-process. Maybe a dictator is the final word of the law. That doesn’t seem very appealing. Imposing Rothbardian deontology would be dictatorship, because most people would want a completely different set of meta-rules. Further, we have no way of knowing if Rothbardian deontology is right, and we want the meta-process to at least evolve over time, so that its rules can improve. Because we know that there is a range of reasonable opinion on what kind of rules there should be, the process of rule-making should consider a plurality of opinion.

The people governed by the meta-rules will include libertarians, but libertarians are a minority of the total population. Reasonable debate would need to include reasonable non-libertarian beliefs. The meta-process would be a liberal one. It’s hard for me to accept the possibility of both agreeing with the public reason approach and believe in a libertarian political philosophy. Public reason is extremely attractive, because it is consistent with heterogeneity of beliefs and pluralism (i.e. liberalism); maybe a libertarian political philosophy does not make much sense. Libertarianism can’t even describe the ideal society, because if societies include non-libertarians then the rules, in a voluntary society, will be, at least in part, non-libertarian. Public reason leads to compromise, not unanimous consent in favor of one of the extremes.

The libertarian has some preference set on meta-rules. But, you can’t advocate an overarching philosophy of a just political process that requires either belief homogeneity or coercion of the non-believers. Well, I suppose you could. But, calling that philosophy one of “liberty” should raise eyebrows. Libertarianism, then, is not useful at all for the purpose of helping us achieve a free society. But, (public reason) liberalism can. I agree that this might lead us to other uncomfortable conclusions, but that’s no reason to reject my argument. Of course, the catch in believing that there is some probability (0 < p < 1) that a belief is wrong is, well…that your beliefs can be wrong. So, prove me wrong!



  1. I guess Mises’ “partisan biases” don’t matter.
  2. Both Milton and David. Pure libertarianism might require two points. You earn one point for being an anarchist, which David is. You earn another point for being an Austrian. Sorry David. And Milton might as well be Stalin.

8 thoughts on “Illusion of Purity

  1. RobertRoddis

    Of more concern is the inability or refusal of the “non-purists” to meticulously identify the prior instances where strict adherence and deference to private property and contracts failed resulting in an alleged “market failure” that allegedly requires violent intervention. I do not think that analysis is as difficult as some people suggest.

  2. Ryan Long

    Have you heard of Sorites paradox, otherwise called “the paradox of the heap?”

    Basically, you’re criticizing libertarians for putting forth libertarianism as a Sorites paradox. If there’s “one drop of statism” in it, then it’s not “pure libertarianism.” It’s a “whole heap,” and if you remove one grain of sand from it, then it’s not libertarianism! As you say in your post, I agree, and you’re right.

    But then you spend the rest of your post committing the same fallacy from the other direction. You say that libertarianism isn’t useful because it can’t be defined in a useful way. Only grains of sand are relevant, not heaps, because heaps can’t be defined. But of course if you add some critical number of grains then you turn around one day and find that you’re looking at a heap.

    That is to say that the boundary lines of any thought process are fuzzy (or subjective), but once you move away from the boundary lines, you’re looking at libertarianism. To call that “not useful” or “too exclusive” is just silly. It’s a mode of thinking – it’s useful to anyone who finds it useful. You can opt to stop thinking of it as useful, but that would mean that your beliefs are just as subjective as everyone else’s.

  3. JCatalan

    @RobertRoddis:disqus: I haven’t seen you around here in a while! I was starting to fear that maybe I turned you off of my blog. But, I need reasonable minds like yours to help me be skeptical of my skepticism! For what it’s worth, I think your point is valid. If there are such things as public goods, name one. I can name candidates, but I’m not sure I can name one where the debate is settled. I do think there are market failures, and that these should be evidence (they are inevitable; markets aren’t perfect). That doesn’t justify the use of institutions of governance to supply that good/service, but at least it does show that it’s conceivable that the opportunity cost of pursuing an end through (a liberal) government is lower than pursuing it on the market. I don’t want to make the case for big government. I want to make the case that what the government should be involved in is an empirical question, and it is contingent on historical factors.

    @ryan_long:disqus: If my response misses the point, let me know. I’m not saying libertarianism is useless in general, only as a political philosophy. We can’t advocate a political structure, and claim that it embraces liberty, that incorporates only a relatively narrow set of beliefs on what that political structure will look like. At least, this is true if we want the benefits of a larger division-of-labor. Libertarianism is not useless as a way of categorizing a certain set of preferences. But, are these preferences really a political philosophy? If they are, they can only be one (small) part of that philosophy.

    1. JCatalan

      By the way, Ryan, knowing that you like to blog about physical training, I just wanted to pass this link along to a friend of mine’s blog on that subject (or a similar one, I don’t know the nuances of physical training): Bringas Athletics.

  4. Seth MacLeod

    Warning: Unorganized response incoming.

    Regarding Hayek, I think it makes more sense to refer to him in terms of classical liberalism. Regarding Long, this is what he says about his views:

    On the one hand, I’m committed to libertarianism in a fairly standard sense: self-ownership, the non-aggression principle, Lockean homesteading, private property, and free markets. On the other hand, I’m committed to a fairly standard set of traditionally leftist concerns, including opposition to such social evils as worker exploitation, plutocratic privilege, racism, homophobia, gender inequality, militarism, environmental degradation, and the prison-industrial complex. (Call them all “oppression” for short.)

    I don’t think either the libertarian or the leftist commitments need to be watered down in order to accommodate the other; on the contrary, I think they strengthen each other. I see these two sets of commitments as related both causally and conceptually.

    Some people use the term ‘libertarian’ to refer to ‘liberty-oriented’ belief systems. Sometimes this is useful, for example when a modern libertarian is referencing someone from one or two hundred years ago and wants to highlight that liberty was highly valued to this person. But this broad use can be problematic because there really are significant differences between classical liberals and individualist anarchists from the 1800s, and referring to them both as libertarians is not only bizarre but also blurs the very significant differences between them. And they are hardly the only liberty-oriented belief systems.

    Or another bizarre example would be including the Objectivists as libertarians. Generally speaking, Rand’s Objectivist philosophy is incredibly liberty-oriented, but Rand was very hostile to libertarianism — I don’t know if modern Objectivists are as well or if they have abandoned that viewpoint.

    I do see some merit in using libertarian in a broader sense, as it can be useful to contrast it against belief systems that are not liberty-oriented. But if you use it too broadly, it loses the ability to distinguish between classical liberals, Objectivists, (actual?) libertarians, etc.

    1. JCatalan

      I agree with you re: Hayek. In fact, I think he would sympathize with me on preferring the label “liberal” to describe his political philosophy; maybe libertarian to describe some of his normative views on optimal governance. Re: Long, I was mostly being facetious.

      Fwiw, I try not to call myself a libertarian. And, it’s not because I want to distance myself from libertarianism or because I want to distinguish between what I believe in and libertarianism. I am highly sympathetic to libertarianism, even anarchism. But, ultimately, I am a liberal, because I believe political institutions should be built around the opinions of all reasonable people, including non-libertarians. And, I can conceive of a situation where I might see it as worthwhile to allow the supply of a good/service through government, even if I can’t come up with any specific, 100 percent certain examples myself. Finally, I might see it as worthwhile to allow for the public provision of a good not because I think that’s how it’s best supplied,but because I use that as a way of compromising to get concessions out of those who hold heterogeneous beliefs. That’s why I like to talk about liberal, or inclusive, political institutions as being “exchange based,” rather than “extraction based.”

      1. Seth MacLeod

        Well, I don’t have a problem with governing institutions per se; it’s state government that I have a problem with. When governing institutions are built upon might makes right, I would say that it no longer is being organized by reasonable people.

  5. Roman P.

    I am cynical enough to propose that you are wrong: the exclusiveness of libertarianism is not a bug, it’s a feature. Just as they say democracy is a rule by democrats, libertarianism is ultimately about securing liberty for libertarians.

    And that’s actually a pretty solid position.


Leave a Reply

Your email address will not be published. Required fields are marked *