Wednesday, February 23, 2011

If you think it's been too long since I've substantially ruffled some Austrian/libertarian feathers...

...then you're in luck.

Astrophysicist and science educator Neil DeGrasse Tyson is on record saying that he much prefers the movie Deep Impact to Armageddon, because the former gets a lot more of the facts right. I have to disagree with Tyson - Armageddon has long been a favorite of mine. If I wanted factual accuracy, I wouldn't go to the movies, damn it*! But certainly part of the reason I like Armageddon better must be that I would prefer we use a lot of really expensive (tax-payer funded) equipment to send Bruce Willis up there to blow it up, rather than have it hit (as it did in Deep Impact). Many libertarians disagree apparently (although some qualify their support for the preservation of the human race), and Brad DeLong called them out on it here. Bob Murphy responded, and DeLong took that up too, here.

This was my comment on DeLong's second post:

"Your last line [that the Sabbath was made for man, not man for the Sabbath] encapsulates the whole controversy. Libertarians in general and Austrians specifically enshrine what they understand to be deductive logic (although its a conspicuously anemic sort of deductive logic that often eschews math) at the center of their assessment of any question. If you start where they do - axiomatically defining what a "right" is and defining "immorality" as being the violation of that right - then both the Volokh and Murphy version of the argument follows naturally.

The problem is not the logic itself - that logic is simple enough. The problem is their decision to apply that logic to this circumstance. Deductive logic is a tool - we have long since passed the point where we can claim that it deserves any deeper epistemological priority. Is it really true that taxing to avoid an asteroid is immoral? Well, if you start with the right assumptions, yes. But who cares about that? You don't care. I don't care. As you say, reasonable people don't care. But modern libertarians who swear by a few choice axioms that are easy to logically manipulate into a doctrine of anarcho-capitalist orthodoxy have infused a great deal of meaning to it all. Logic is not a tool for them, as most careful thinkers know it should be. For them it is a path, not to truth contingent on assumptions, but to some deeper absolute truth.

So to expand on your concluding thought, deductive logic is a tool for humanity, and not the other way around. "Rights" as the libertarians conceive of them are nonsense on stilts, and any strict deductions from that are at real risk of being nonsense as well.

Brad, Keynes had it right eighty years ago when he wrote of Hayek that he provided an "example of how, starting with a mistake, a remorseless logician can end up in Bedlam". Hayek wasn't always guilty of this - he was a good thinker. Modern Hayekians aren't always guilty of this either - many of them are decent thinkers. But they seem to have a tendency of ending up in Bedlam for precisely the same reasons that Hayek did back then

Bob Murphy was disappointed I didn't defend him (he and I get along decently enough, although we disagree on a lot), but as I pointed out to him - I do confirm in my comment what Bob said in the Mises piece of his that I linked above. I say that the Volokh/Murphy logic on the question is quite simple to grasp, and given certain assumptions they're exactly right. Maybe Murphy wouldn't have needed to write his Mises Institute response if DeLong said this more clearly - "the logic isn't the problem, the problem is that you think that logical chain has any relevance". Murphy points out that Brad also misinterpreted a side-point of Volokh's, and Murphy seems right about that, but I think Murphy completely misses the point that what's most aggravating is that there is any equivocation at all (from the Volokh line of reasoning or the Murphy line of reasoning, which are slightly different) on this question.

The whole problem with the Volokh/Murphy logic here is their fundamental assumptions about rights, of course. Rights are social constructions - they are threats of violence from a well-defined group of people that allow individuals to appropriate and use goods and services. A right is essentially three hundred million people saying "we will beat the crap out of anyone that tries to take this property from this person". In that sense, society is certainly being coercive (and government is too to the extent that it legally recognizes property rights). The difference between that and gangs asserting ownership of territory is only one of etiquette and tradition. Rights are not some Platonic ideal floating around that are innate. They are functional, of course, but they are not innate. As Bentham pointed out, to pretend otherwise is "non-sense on stilts".

The problem with Murphy's position is that he accepts social (and in some cases even government, perhaps - although state recognition is not essential) coercion when he accepts his axiom of rights, but he rejects social coercion in other circumstances. Why? I'm not quite sure. But this is the problem we're dealing with. Libertarians make assumptions about rights that are romantic, idealistic, and unreasonable. Taking this romantic view of property rights is highly functional under most circumstances, which is why I and most others in the classical liberal tradition nod in agreement and swell with pride when we read natural-rights-type proclamations (and even slip into natural-rights lingo ourselves). But that doesn't make it accurate. The only reason for privileging property rights in the way that libertarians do that I can think of is if you accept a sort of "I called it first" ethics. Rights established by society are the status quo ante, so they get special priority in ethical consideration of things like taxes to save the human race from extinction**. So the logic is right, but the assumptions are wrong. Now, Volokh and Murphy's critics usually don't say what I just said, so do I have warrant to defend them with this argument? I certainly can make my own case on the basis of this argument. But I think I can also extend it to others who rebel against Volokh and Murphy's conclusions because to a large extent I think this is intuitive. Intuition and tacit knowledge is under-rated in this society, which worships the expert. People understand the idea of rights, but there are certain applications of that idea where they feel that something is fundamentally wrong. There's an intuitive sense that rights are just a functional fiction which is why people consistently apply it in a lot of circumstances but find rights-talk funny in others. What libertarians identify as an inconsistency in logic is actually (I would wager) a great deal of intuitive insight about the appropriate applications of that logic. Commenter sanjait had a good thought on DeLong's post:

"I'll add that it is continuously remarkable to me how little introspection some people engage in, despite the fact that their puritanical libertarianism leads them so often into places that are both absurd and deeply unpopular. When a system of thought leaves you frequently out in the woods alone, shouldn't you question it's universality?"

I think it's an admonition that Murphy and Volokh (and others) should take to heart. Maybe what they say sounds natural to them at this point, but they should probably interrogate a little deeper about why it sounds so wrong to so many other people. If your arguments are really so self-evident, why is it that so few people agree with them. People like Thomas Jefferson also made claims about self-evidence, and the fact that more and more people agree with Jefferson on those points every year is evidence that he was probably right to assert it in the first place. When you continually get resistance from very thoughtful people, rather than running through the relatively easy logic one more time (a la Murphy's Mises post), maybe you should instead ask questions like "are these assumptions about rights reasonable?", or "am I applying the logic appropriately?". The popularity of a belief is not a guarantor of its accuracy, but it oughta at least inform the questions you ask.

For more on my thoughts on property rights, and where libertarians go wrong you can read this, this, and this is sort of relevant. For me on libertarians and logic you can read this. For me on libertarians and bad axiomatic claims you can read this, this.

The other Austrian/libertarian feather I ruffled was Jonathan Catalan's, although after reading it there's not all that much to say. He suggests I should not conflated Austrians and liberatians (check). He thinks I should be cognizant of the fact that there's a huge variety of Austrians (check). He thinks I should recognize that there's a lot of common ground between Austrian economics and mainstream economics (check - I always manage to get Steve Horwitz's blood to boil when I say this on Coordination Problem). Finally, that praxeology is different from other sorts of logic (check). So in a lot of the beginning, I don't have much to disagree with. I would have thought my reasons for mentioning both groups would have been obvious, namely that (1.) Murphy is both a libertarian and an Austrian and he was the subject of Brad's post, (2.) most Austrians are libertarians, (3.) it's possible that most libertarians commenting on blogs these days are at least sympathetic to Austrian economics, if not Austrians themselves, and of course (4.) both share a proclivity for tight little deductive catechisms. To recognize these points is not to say that they are the same thing.
Jonathan gets himself into more of a mire when he gets into praxeology and the action axiom. Nobody that I'm aware of would really disagree with Austrians on the action axiom. That's part of why they like it so much. It is truly self-evident. Jonathan himself proclaims that Austrianism is based on this "single axiom". I don't know if this is true or not, strictly speaking, but whether it is or not it's not exactly something you should boast about. For a long time humans didn't take as axiomatic the real number line we know today, which includes negative numbers. We didn't have an axiomatic understanding of the number zero for a while. Needless to say, math was substantially stunted when we omitted these axioms. Yes, these things were "true" contingent on the axioms, but if you went around life thinking that other math wasn't "true" because it couldn't be derived from this narrow set of axioms, you'd be in a lot of trouble. That's what Austrians often do with their very limited set of assumptions. So how do we arbitrate between the Austrian conclusions from their single axiom (let's assume they have all their ducks in a row in the derivation exercises) and other understandings of society that accept other axioms or even derive insights from observation? Well, there is no common basis for arbitration, so you essentially figure out which paradigm is more useful for understanding the world. I do not hesitate at all to say that the Austrians have failed in this comparison. Even if some Austrians disagree with me on that, they certainly can't take their derivations from a single axiom and rule out all other claims because it's not consistent with that axiom. Calculus wasn't consistent with the limited set of axioms that the ancients worked with - do the ancients have a case against calculus? Of course not. That's not how logic works. Logic is an extended chain of claims that we use as a cognitive tool to bring order to different scenarios. Newtonian logic "worked" to get us to the moon. Newtonian logic works given its own assumptions. That doesn't prove Einstein is wrong.
This leads to another point that Jonathan raises - haven't I been generally supportive of Austrian business cycle theory as at least a component of modern macroeconomics? Haven't I said the distortion of the capital structure seems to make good sense and probably plays a role just like other processes that contribute to the business cycle? Well certainly I have. Jonathan asks why I endorse Keynes's critique of Hayek, then. I endorse it because I myself have noted Hayek's tendency to rule out conclusions because of the narrow assumptions that he starts with. Just because Hayek highlighted what seems to be a meaningful process doesn't mean that he was right to dismiss others. Keynes himself noted the roundaboutness argument, after all. Jonathan also suggests Hayek corrects a problem with Keynes on capital theory. I need a little more background on what he means by this, and perhaps I can comment later.
In conclusion, rights are social tools. Logic is an intellectual tool. Asteroids headed for earth are a BFD. Bruce Willis is a badass.

* - one big exception that has always bothered me in one of the final scenes is where people all over the globe - America, France, India, China, etc - are all looking up at the exploding asteroid and it's day-time in each of these places.

** - another option for priveleging property rights is available for those that want to toy around with Lockean labor theory of value thinking. I think this is dangerous. Locke may be right in his analysis of the person himself, but to take him any farther than that (i.e. - to actual goods and the proceeds from those goods), requires some really troubling assumptions in value theory that most philosophers and economists have long since rejected.


  1. Am I expected to read that? Seems rather lengthy.

  2. I did briefly dip down into it:

    "As Bentham pointed out, to pretend otherwise is 'non-sense on stilts'."

    (a) Bentham could be wrong.

    (b) Bentham was famously stupid on all sorts of things.

    I'd say that it is clear that rights are more than mere social constructions; that there is a lot of biology behind them (particularly re: epigenetics).

  3. You and I seem to differ on the extent to which social construction as a process is tied to our biology, otherwise you wouldn't frame it is "mere" social constructions, and then submit biology as a sort of counter-argument.

  4. "Rights are social constructions - they are threats of violence from a well-defined group of people that allow individuals to appropriate and use goods and services."

    What you are describing is more minarchist than anarcho-capitalist, where Murphy hangs his hat. The anarcho-capitalist would expel the violator and allow him to suffer the consequences of being outside of protection. Actual examples of this occur today in rural Somalia.

  5. So much talk about a fictional comet strike that will never happen.

    If you've ever seen the discussions between Bala and Lord Keynes on the Krugman In Wonderland blog, they will always make up some make fictional situation that will be as extreme as possible to illustrate their moral argument. It always MUST involve starving people or putting a gun to someone's head.

    Why do people discuss morality in a way that they begin by setting aside all facts and start in a vacuum? I much prefer some more specific past and present day situations. Should Churchill have extradited Polish dissidents back to Poland, where they could be arrested or shot dead,.etc,.etc.

  6. You amazingly continue to miss the point concerning Austrian praxeology. The axiom of human action is an underlying common factor in all economic activity. It was the factor which Mises (and earlier, Robbins, and many other economists, including Knight, et. al.) believed was at the root of economics, or what economics basically studied once you deaggregated macro theory. So, if all economics is is a series of actions committed to by humans, by employing means to achieve certain ends, it follows that you can deduce these theories from that action axiom.

    Thus, using an absolutely true statement (the action axiom), Mises deduced the remainder of his theory.

    If A is absolutely true, and B is logically deduced from A, then B is absolutely true, and so on.

    Regarding new axioms or different axioms, that's all good, but you haven't shown there to be any other axiom to consider. You haven't even shown understanding of the idea of praxeology, or the idea of why an axiom is considered in the first place, or the work Austrians put into guaranteeing the accuracy of their deductions. You commit the same mistake you accuse libertarians of making. But, I mean, it's not surprising, because you've made this mistake over and over again in most of your work (especially blog posts), which is why a lot of your work is just wrong.

  7. If you ever are under the suspicion that I have done or said anything meriting the descriptor "amazing", you are in all likelihood misinterpreting me!

    I'm really not sure what the problem is here. All I'm saying is that yes, you can derive the Austrian school from an axiom, but why should I care what you can derive from one axiom? Do you care about a mathematics that is derived from a restrictive set of axioms? Does that have inherent value to you, such that you would ally with that mathematics not just in claiming that it is right - but in actually having some allegiance to it in its disputes with other mathematics derived either from other axioms or from observation?

    You seem to be making a relatively trivial claim for the Austrian school here. All I'm adding is that that seems true as far as it goes, but I'm forced to shrug my shoulders and respond "so?".

    1. It's no disproof of other claims, as you seem to agree in this comment.

    2. It's a relatively limited set of claims, and

    3. A huge variety of far more useful claims can be derived potentially from other axioms, but certainly from other non-axiomatic assumptions, and from observation.

    4. It's a set of claims which, if sloppy practitioners slip into thinking offers a counter-argument for other theories, can be extedned to make false claims (i.e. - that Hayek has refuted Keynes, etc.).

    I'm simply making these additional critiques, while agreeing with you that within the context of the action axiom, derivations from the action axiom are true.

    In other words, nothing you continue to repeat here seems to me to change my reaction of "meh". Are you just frustrated that that's my reaction to it all? I don't get it.

    Now - how does all this relate to asteroids and libertarianism. That point is one that I think you are overthinking. It's relevant is purely associational. The crowd of Austrians and the crowd of libertarians all have a (for lack of a better word) cultural tendency to rely too heavily on neat, clean, tight, deductive chains of claims. On top of that, Murphy is both an Austrian and libertarian. That's the only reason why I mentioned it in passing in my DeLong comment. It was purely an associational/group-culture statement.

  8. Daniel,

    "All I'm saying is that yes, you can derive the Austrian school from an axiom, but why should I care what you can derive from one axiom?"

    This is a basic epistemological question, and I'm not sure that you've really paid a sufficient amount of attention to the subject. This is evident to your response to a recent blog post of mine, where I questioned where exactly the profession was in terms of methodology and epistemology. You commented that economics is now a "mature science", where that issue is settled. The issue has actually not been settled, and what illustrates this fact the best is that economic theory is "all over the place" and without any real backbone to unify it all.

    The epistemological debate in economics has always revolved around what exactly economics deals with. At one point many thought it dealt with the acquisition of wealth, or maybe with wealth in general. Then, it went on to concepts such as the idea of the action of exchange. In the early 20th century there was a movement towards the notion of "human action". This was an attempt to explain economics by isolating a lowest common denominator in all economic theory/fact.

    Mises, et. al., isolated this lcd as being the fact that humans act purposefully. They have certain ends, and they employ means to fulfill these ends.

    Given that all economic relationships stem from this lcd, it follows that you can deductively come to these relationships (and establish theories based around these relationships) by starting from that axiom, which is the lowest common denominator.

    This is important, because it allows a greater degree of accuracy when developing economic theory. Otherwise, you are either practicing positivism (which is a debate for another time) or you are deducing theory from premises which are not previously tied to any axiom, and so their accuracy is far more questionable.

    How does this tie in asteroids and libertarianism? It doesn't. My point is that you made a false comparison based on your lack of understanding of what exactly you are comparing. The fact that Murphy is both an Austrian and a libertarian has nothing to do with anything, as he doesn't mix methodology, and his methodology in either case is not unique to him.

    My other point is that Austrian methodology is clearly more rigorous than libertarian methodology, because specific claims can be checked against prior premises, which themselves can be checked against prior premises, until you reach the axiom of purposeful human action. In other words, you can check the logical rigor of Austrian economics, and the logic reflects objective reality, not subjective belief (which is what libertarianism oftentimes reflects).

    "It was purely an associational/group-culture statement. "

    Yes, I'm saying your statement was grossly mistaken.

    Btw, what irks me is not that your response is "meh". It's that your response is "meh", but you continue to comment as if you really knew what you were talking about.

  9. Other comments:

    1. No, you're right, all theories must be tested on their logical rigor.

    2. It's not a limited set of claims at all. It's one claim which encompasses all sciences related to sociology, since all these sciences revolve around purposeful human action (see Max Weber).

  10. 1. Settled in the sense that we have converged on an epistemology and methodology that we move forward with. Not settled in the sense that we have plumbed the depths of epistemology and all come out satisfied an in agreement. You are looking for philosophical fulfillment. I am saying we have philosophical common-ground so that we can proceed with scientific work.

    2. re: "This is important, because it allows a greater degree of accuracy when developing economic theory". No. It does not guarantee a greater degree of accuracy. It guarantees that you are generating more "true" claims. There is no guarantee at all that you are generating claims that allow you to do better at accurately describe the world. A rough theory not derived from axioms, but honed through observation could definitely help you understand the world "more accurately", even if there is no guarantee that the claim is some ultimate truth.

    3. re: "My point is that you made a false comparison based on your lack of understanding of what exactly you are comparing." - and you are far too quick to accuse people of accusing others of not understanding a point just because they aren't talking about it in the same way you're interested in talking about it (ie - you want to talk about this from an epistemological perspective whereas I started by talking about it in a more sociological context).

    4. re: "It's that your response is "meh", but you continue to comment as if you really knew what you were talking about." I would take this slight more seriously if you ever got around to demonstrating some fundamental misunderstanding of mine. I used "axiom" too broadly on Murphy's blog? Sure - I'll admit I did that. But it doesn't change the point at all, and this is what you always seem to miss. You always want to go off on an epistemological tangent in a discussion that isn't even fundamentally about epistemology (or if it is, is about me questioning the value epistemological preoccupations).

  11. I'm going off on an epistemological tangent precisely to illustrate to you the differences between Austrian thought process and libertarian thought process. You continue to evade my actual accusation, which is that you improperly compared the two. You still have to address that particular point.

    1. We do not have sufficient philosophical common ground, obviously.

    2. If human action reflects reality, and therefore all theory stemming from that reflects on the reality of human action, then it follows that the theory developed will reflect on real action. If A, then B. Whether or not the theory is applicable to history depends if A happened or not, but that's what history is about.

    Of course you need observation. Nobody claimed otherwise. But, you are confusing yourself here.

    "Does the theory apply" is a question of observation. Austrians agree. What Austrians disagree with is developing theory based on observation.

  12. It's entirely unclear to me why you think you need to demonstrate those differences to me. Are you under the impression that I was unaware of the differences you outlined here?

    Jonathan - really!

    You oughta know I know the assumptions/axioms that feed into each are different, that the derivations highlighted be each are different, etc. etc. All this time that you've been repeating those differences, have you been thinking that was unknown to me?

    My point is that despite these differences they share something in common as well - a retarding reliance on logical constructions (sometimes under suspect assumptions, sometimes not). I don't understand why you think my highlighting of this commonality implies an ignorance of the differences!

    re: "What Austrians disagree with is developing theory based on observation."

    That and developing theories from non-axiomatic assumptions (although to be clear, self-identified Austrians inevitably do both some of the time - but that's not the "Austrian methodology").

  13. Jonathan,

    As a follower of Mises, can you tell me: Do you think that praxeology successfully limits all its deductions to this single "human action" axiom, and it alone? Put differently, do you mean to say that there are no secondary axioms (assumptions) or subsidiary propositions involved in deducing any of the major themes that characterise this particular (i.e. Misean) strain of Austrian economics?

    Not trying to be funny, I'm literally curious.

  14. Stickman,

    Yes. That is correct. Rothbard, writing later, chooses to introduce some elementary empirical observations (that can still be squared in Austrian methodology) in his work. But Mises, as far as I can tell, does not.

    All conclusions are logically derived from the single, irrefutable action axiom. What I think Daniel doesn't quite understand is that if there are corollary or conflicting economic theories, they would be equally discoverable under the action axiom. It is by no means an axiom exclusive to "Austrian conclusions," but Austrian conclusions are those which derive from the axiom.

  15. "What I think Daniel doesn't quite understand is that if there are corollary or conflicting economic theories, they would be equally discoverable under the action axiom."

    This is not necessarily true. Why would you necessarily expect other valid theories to discoverable under this axiom? Or am I misunderstanding the claim, in which case could you reword it?

  16. If can agree, in principle, that laws of economics can be discovered by logical deduction from the action axiom, it follows that true Keynesian or Neoclassical conclusions could be found using the Austrian methodology. This is what, I believe, Jon has done with regards to liquidity trap theory.

    I was trying to argue that all economics stems from the truth that humans act. If the paradox of thrift is true, it should be as discoverable under the action axiom as originary interest or ABCT. The axiom does not yield just Austrian theories, but the opposite: Austrian theories are those conclusions derivable from the action axiom.

  17. A proof or deduction is a sequence of propositions or formulas, such that each one is either an axiom or a modus ponens of two previous ones. Therefore you cannot derive anything from just one axiom.


All anonymous comments will be deleted. Consistent pseudonyms are fine.