Economics
Another conversation I had about a priori psychology
I previously discussed a conversation I had with a senior, well-known Austrian at a conference. Remarkably, I had a very similar fictional conversation this weekend with a fictional aproirist psychologist:
Apriorist Psychologist: I can deduce universal laws of human psychology from axiomatic first principles. So, for instance, I have determined that religion makes people unhappy.
Brennan: Whoa, really? But what is your response to empirical psychologists whose work seems to show that your principles and laws don’t explain or describe actual human behavior in the real world? For instance, empirical psychology consistently finds that frequent religious attendance is positively correlated with happiness and life satisfaction. Surely, if you want your theory to be a theory about actual people, not just about hypothetical people, you’ll need empirical observations to show that your theory has explanatory power with regard to those people. Otherwise, you’re just unpacking definitions, and you have to remain agnostic about the ability of your theory to explain real-world phenomena. You can’t intuit real-world causal connections or derive such connections from first-principles. A priori models can help, but you need empirical work for that.
Aproirist Psychologist: Ha! You probably think that because you didn’t read Person, Mind, and Society, by psychologist Eugene Hackman. Hackman is basically ignored or considered a crank by the academy, but that’s because the academy is corrupt.
Brennan: Well, I did read much of that book–it came highly recommended from one of my friends–but there wasn’t much there of interest to me, since everything I saw in it that seemed right was already established by someone else. But, at any rate, I don’t see how Hackman’s book could overcome my basic objection: How the actual world works cannot be determined a priori. We can develop models and theories a priori, but to use these theories to explain the world, we need observations. And, so, a priori psychologists cannot just dismiss empirical psychologists. I’m not saying that everything or anything the empirical psychologists say is right—I’m agnostic about that—I’m just saying that you can’t know a priori that they are wrong. You have to really engage with their work.
Apriorist Psychologist: What do you think of Hackman?
[At this point, I made a big mistake. I cringe when I recall how the conversation went.]
Brennan: Frankly, he seems like a hack–a second-rate at best psychologist who couldn’t cut it in the academy, and who decided instead to communicate his new work primarily to laypeople without a psychology background. [See! I’ve got to cut it out with this stuff!] Despite writing so much, he’s more of an activist than a scholar. But that shouldn’t distract you–even if my opinion of Hackman’s work is unfair, the question here is not about his work, but about whether you can do psychology a priori.
Apriorist Psychologist: Eugene Hackman is the greatest libertarian ever, and among the greatest psychologists ever. He’s super-duper famous on the Internet, and there’s an institute called the Stanley Milgram Institute* that is dedicated to Hackman’s work. Most of Hackman’s fans on the Internet haven’t heard of you, Brennan, so you should be careful about whom you call a hack.
Brennan: For the sake of argument, let’s agree that I totally suck at everything, and that I will never amount to anything. Let’s agree that I’m such a loser that my existence is fodder for the Problem of Evil. Regardless, you can’t intuit causal connections a priori, and so a priori psychology can’t tell us, by itself, how actual minds in the actual world actually work. Insofar as Hackman has a response to empirical psychologists, it’s either to show that they made a mistake in their reasoning from their empirical work (I don’t have to run an experiment to know that someone made bad inferences from his or her data), or because Hackman actually relies upon at least some empirical work.
Apriorist Psychologist: Well, did you read Hackman’s paper, “In Defense of Extreme Apriorism in Psychology?”
Brennan: Yeah, I did. In fact, let me read it again right now on my iphone. Okay, in that paper, he asserts that psychology can be based on two axioms or postulates, or maybe four depending on how you count them, and he then asserts that you can deduce everything else in psychology from that. But there’s absolutely nothing in here that would potentially convince someone who didn’t already agree. As far as I can see, it’s almost all assertion with no argument.
Apriorist Psychologist: A ha! Well, you must not see very well.
P.S.: I forgot, we talked a little more later. Heres the rest:
Apriorist Psychologist: You can’t do social scientific experiments the way you can do natural science experiments. Human beings react to being experimented on, and you can’t isolate variables quite so well.
Brennan: Well, sure, this is well known, and people work hard to overcome this as much as they can. But, at most, your point tells us we can’t do social science at all. It doesn’t tell us we can do social science a priori.
P.P.S.: I forgot, after this, my buddy Matt chimed in:
Matt: Not everyone at BHL dislikes Hackman. In fact, I can think of lots of nice things to say about him.
Brennan: Great, Matt, now I look extra mean.
*For the record, I’m a big fan of Stanley Milgram, and there’s at least a few people at the Milgram Institute who do good work in my view.
somethingthatneverhappened.txt
What tipped you off?
What do you mean a fictional conversation with a fictional psychologist? Do you mean that this conversation did not really happen, or do you mean that you have basically dramatized the conversation for summary purposes?
It’s fake.
Haha! Good show.
I don’t know what to say, except that I bought your introductory libertarianism book for someone. Had it shipped directly to their house, and she totally took a pic of it and put it on Instagram. Let’s just say it was a nice gift. Shipping was fast too! Ouuhhh Amazon.
(Irrelevant but you know you love it!)
Cool, thanks! I appreciate it.
Sometimes I can’t tell if the internet is making you miserable or if there is no one in the world having as much fun as you.
I’m leaning toward the latter.
Ding ding!
I’m afraid you have a point there. Brennan’s conduct is not easy to defend in this episode.
1) He’s engaging only the weakest part of his opposition, which is rightly considered an intellectual sin.
2) He’s way too enamored of his own success as a place-hunter and rent-seeker in academia. Now…that would be just an unseemly little character flaw, except that he tries to use it as evidence in the debate by saying “I publish here, therefore valid. He didn’t, therefore HACK!” That takes it from character to flaw to infuriating fallacy.
3) Correct me if I’m wrong, but the proposition “If economics is to teach us anything about the real world, it must be empirical” sure sounds kinda, you know, not empirical.
But to be fair, let me also say a few things in Brennan’s defense.
A) I don’t think he actually wants to silence his critics here. I’ve made him look ridiculous a few times, and he left my comments unmolested (if also unanswered!). One possibility to consider is that he’s just not much of a webmaster. But the fact remains… in the end, he DID let those posts and comments stand.
B) It’s not Brennan’s fault Rothbard made himself such an easy target, what with the starving babies and the whole polymathic overreach that seeks to reduce complicated questions down to simple, immutable rules. Guys like that are asking for trouble, and deserve to get it.
C) Just because Brennan is impressed with himself for the wrong reasons, doesn’t prevent us from being impressed with him for the right ones. The fact is, the guy is a good philosopher and an important libertarian who works right at the heart of what may be libertarianism’s most interesting project.
The very fact that you believe that his fake conversation says anything relevant about anything relevant proves that you don’t understand the main point of Milgram’s classic book “Logic and Psychology”, where he exhaustively describes the methodological differences between the two and their respective, complementary roles.
Yeah, that’s what I suspected.
Well. I don’t think Rothbard – I am sorry Hackman – would disagree with the empirical necessity of a priori concepts.
As far as I read Hackman, it’s ‘we have these a priori knowledge’, which we can use for empirical interpretation. For example, in Man, Economy & State – I am sorry, Person, Mind, & Society – Hackman makes it clear that although most of what he describes is a priori (infused with a few general empirical observations), the fact that the market is ‘better’ than the government is not a priori.
I think most academics (including Hackman) are on the same page, here. The point is to educate the internet consumers of apriori psychology on the relationship between the a priori and the empirical.
Nice! I chuckled heartily at the the PSS (Although I think it’s supposed to read “P.P.S.”, cuz the p stands for Post)
Good catch. Fixed.
I would be more impressed if you replied to In Defense of “Extreme Apriorism” specifically rather than constructing this vague, straw man argument. I could reply to it, so I suppose you could, but I don’t know that you could empirically, because I haven’t seen you do it.
I don’t see any argument in there to reply to.
I’ll help you.
“From the Fundamental Axiom is derived the truth that everyone tries always to maximize his utility. Contrary to Professor Hutchison, this law is not a disguised definition–that they maximize what they maximize. It is true that utility has no concrete content, because economics is concerned not with the content of a man’s ends, but with the fact that he has ends. And this fact, being deduced directly from the Action Axiom, is absolutely true.”
How does having an end imply maximizing anything? Why couldn’t an end involve minimizing something or seeking any sort of plateau (maximum, minimum or a saddle point) or even tossing dice?
Physicists realized that they must deal with these complexities (see classical Legrangian dynamics and Feynman’s path integral formulation of quantum mechanics). Economics is not physics, because human action is more complex, right?
Introspection is empirical, not a priori.
This is just one of the reasons Rothbard goes wrong.
At some point in a week or two, maybe I’ll do a line-by-line explanation of why Rothbard’s article is mistaken.
Right. I just said that. Rothbard says it too.
He writes, “Actually, despite the ‘extreme a priori’ label, praxeology contains one Fundamental Axiom–the axiom of action–which may be called a priori, and a few subsidiary postulates which are actually empirical.”
The Fundamental Axiom, as Rothbard formulates it, is laughably vague. “Humans act”, he says, and he then specifies “action” as “seeking ends through means”. I’ll buy “humans act” as an indisputable truth, and it’s not quite a tautology, if we ignore the dead and including breathing among human actions, and one can define “means” and “ends” so that disputing “humans seek ends through means” also seems pointless.
On other hand, “canines act” seems equally indisputable for the same reasons, and Rothbard doesn’t seem to derive immutable laws of canine social organization, so he must derive these laws from the few subsidiary postulates, which are actually empirical.
But as I try to argue above, Rothbard doesn’t really extend his Fundamental Axiom with a few additional assumptions. His development of economics is an endless stream of additional assumptions. I have no problem with that, but I wouldn’t call his conclusions “absolutely true”, if only because I have extremely high standards for “truth” and will call almost nothing “true”.
No Austrian is saying that introspection is apriori. It is purposeful action. It is a means of discovering what is necessarily presupposed by “experience” in of itself. If you actually read Mises– you might know this.
ps. I bet you do no such thing as a line by line refutation of Rothbard. You don’t have it in you.
I gotta say, I’m with Brennan on this. I read that article, and I just don’t see how it can be called a “defense.” If anything, it’s a clarification of what a priorism is (except that it leaves a lot of things totally unclear–as you pointed out, “humans act” is laughably vague). Anyway, from my perspective as a mathematician, I find his point of view on economics bizarre. He seems to want a field of “pure” economics just like there is a field of “pure” mathematics, a purely theoretical field which could discover theorem about the relationships between actors. But it seems like there’s a nice home for that already–namely, mathematics.
What *really* bothers me about the article is this statement: “Mises demonstrates that this science asserts that laissez-faire policy leads to peace and higher standards of living for all, while statism leads to conflict and lower living standards.”
Full disclosure: I have not read Mises. But if Mises thinks that this is kind of statement (that laissez-faire leads to peace and higher living standards) can be deduced a priori from self-evident axioms *without empirical testing,* then I have to think there’s something fatally flawed with Mises’s philosophy.
Jameson,
Allow me to give you a brief explanation of Mises. Mises and his followers didn’t just create a nice theory and never use it to analyze the real world. Nor did Mises simply create theory without reference to the real world.
Mises felt there was only one axiom which was apriori (which even that is a point of contention with ‘Austrians’ such as myself). Everything else had to be supported by empirical observations and linked together by sound logic. For example, I don’t see how you can ’empirically test’ what the nature of interest is. You need a whole lot of qualitative observations, reflection, and the use of sound logical analysis. The analysis is much more qualitative than you may be used to (but no less empirical). It was certainly a shock to me as my double major in economics and finance had led me to think that ‘hard’ data was the only way to do economics.
Mises also did explicitly write (and I can’t find the quote at the moment) that theory must be developed first to then analyze specific historical events. If the theory is unable to explain these events either your theory is wrong and you must find where you made a mistake or you have overlooked data. But a ‘correct’ (where no mistakes have been made) theory doesn’t need ’empirical testing’ of specific historical events to prove its validity. Obviously, to be absolutely sure that your theory is correct, analysis of specific historical events is required. I think most sane ‘Austrians’ recognize this.
Where you may feel uncomfortable is with the fact that ‘Austrians’ reject econometrics and ’empirical’ testing as is being now performed by the large majority of economists. The two major reasons are 1) because the methods used ignore extremely important qualitative data which can’t be modelled (such as the passing of time) and 2) there is simply no way to prove causation given the multitude of variables which cannot be isolated.
“The two major reasons are 1) because the methods used ignore extremely important qualitative data which can’t be modelled (such as the passing of time) and 2) there is simply no way to prove causation given the multitude of variables which cannot be isolated.”
This seems like something you could say only if you know very little about the use of statistical methods in the social sciences.
So you’re saying that the quant models can effectively incorporate the heterogeneous nature of capital and the passage of time in the production process for example? Please enlighten me because I distinctly remember making all kinds of simplifying assumptions in my models in university precisely because these factors were impossible to model mathematically.
A model doesn’t need to represent something 1:1 in order to produce useful results or offer insight. I have no idea why being unable to incorporate ‘the heterogeneity of capital’ or ‘the passage of time’ would prevent coming up with useful models.
I also have no idea why it would be impossible to model the passage of time, or heterogeneity.
That’s certainly a reasonable presentation of Mises. It doesn’t excuse Rothbard’s statement from his essay defending apriorism.
I have no problem with critiques of empirical methods in economics. I think Hayek’s Nobel Prize speech, “The Pretence of Knowledge,” is a rather compelling critique (but it certainly doesn’t defend apriorism). But I am not an expert in economics; what competence I have comes either from a layman’s reading or a little bit of my mathematical research.
I’m not against Brennan on this as much as I wish he’d actually address Rothbard rather than caricaturing him. A caricature doesn’t persuade anyone who isn’t already persuaded.
I’m also a mathematician, but I realized long about that mathematics is not about discovering truth. It’s about constructing logically consistent systems. A consistent system can be a useful artifact, but a consistent system need not assert any empirical truth, and I’m not convinced that empirical truth must be logically consistent with any formal system either.
“… but I realized long about that mathematics is not about discovering truth.” It seems you have decided the glass is half full of nothing.
Suppose you are a Platonist. Then you should should believe that maths tells you true things about the World of Forms. Suppose you are not a Platonist, then you believe that maths tells you what is and is not a logically consistent fairy-story about the World of Forms. The claims of logical consistency are still true or not.
You seem to disbelieve Platonism, and yet believe that there is no truth but Platonic truth. Thus you see no truth. You are lost in a maze of tautological untruth!
No. I only use “true” very cautiously. My preferences are not truths, except insofar as they are truly my preferences.
If the world of forms is the world that math tells me true things about, then I’m a Platonist, but the world of forms is also what I call “theoretical fiction” above.
I can say things that are “true”, in a sense, about Sherlock Holmes, like “Holmes never wore a deerstalker cap”, but in a stricter sense, nothing is true about Sherlock Holmes.
I don’t ordinarily use “true” to describe statements like “2+2=4”. I use “tautological” or “analytic”. The difference is a matter of semantics.
I suppose I’d be more lost if I confused tautologies with empirical truth.
What’s wrong with calling tautologies true? I mean we would be in real trouble if tautologies were false. If I were being absurdly strict, I would admit nothing *but* tautologies under the standard of truth.
But you are quite right that the truths of maths are not truths about the world. Which in some ways makes mathematical truths less real. And yet. And yet. The fairy stories we tell about numbers, are the most solid facts we know. Bar none. And even though they are all tautologies, many of them are far from obvious.
A tautology is an assumption or a linguistic convention. “An animal can move while a plant cannot” is true only in the sense that we agree to use the words “animal” and “plant” this way.
If we do not carefully distinguish this variety of analytic “truth” from the sort of truth from which we may reach conclusions about reality, we can conclude that a Venus Fly Trap either does not move or is not a plant.
Ahh, I see, your later sentence pulls back from your extremist position, and limit maths to not stating “emperical truth”. True enough.
What extremist position? I’ve never claimed that mathematics states empirical truth.
The extremist position was that mathematics states no truth at all. You became reasonable when you limited it to empirical truth. (In the last sentence “it” means the non-existence), obviously.
Ouch, did an ant just bite me?
At any rate, theres a sense in which the stuff I am writing is very dumb. I’m writing unobjectionable third week of introduction to philosophy of science stuff in response to crazy views a bunch of people with little background in philosophy of science hold.
“I’m writing unobjectionable third week of introduction to philosophy of
science stuff in response to crazy views a bunch of people with little
background in philosophy of science hold.”
If you find this kind of work rewarding, you might consider writing something about anthropogenic climate change. You’ve basically summarized >90% of the “debate.”
Let me guess, the crazy folk are the ones who question a science based upon flawed computer models, deliberately sabotaged data, browbeating and a phony consensus.
No the crazy folks are those whose confirmation bias precludes them from examining things in a disinterested manner, because if they were to find out they were wrong about anthropogenic climate change, their “principles” have no way of correcting it. So they must be right a priori.
You are like creationists pointing out a few phony fossils here and there, ignoring the rest of the data. You guys need your own institutes and journals because you can’t produce the quality work that gets you into real journals. Pathetic in my opinion, but have your fun.
Make good arguments, present good evidence, and you’ll be taken seriously. With respect to science and academia, mainstream is mainstream for a good reason, for the most part. Cranks.
That most non-mainstream arguments are not good does not imply that most mainstream arguments are good.
Who is they?
Can you point me to someone who argued anthropogenic climate change is wrong a priori?
Btw, I’m not a skeptic about global warming, though I certainly disagree with most proposed solutions, and am skeptic about prophecies of armageddon.
What is pathetic is how many people including yourself have closed your minds entirely. It is not incumbent upon me to publish an article because I am not the one making extreme claims. I am only pointing out that good science is always about skepticism and debate. Not, as I said, browbeating (what you just engaged in), consensus, (and not even a real one at that) and falsified data, (and there has been a lot of that). BTW, My mind is NOT made up yet, but you will never convince me with the kind of crap I have seen so far.
This blog is pretty awesome.
I started reading this and though, “wow, not only was Gene Hackman a great actor, he was also an important, if controversial and misunderstood psychologist. Who knew? He was even neater than I thought!” But then, Amazon soon let me know that there is no such book as _Person, Mind, and Society_, and soon I felt almost like a did when I learned there is no Santa Claus.
I’m having to write a term paper for my senior seminar class on Democratic Theory (…using Walzer’s Spheres of Justice, Rawl’s TJ, Dahl, Dworkin, Gutmann, and other leftist social justice types) and the Stanley Milgram Institute has been woefully unhelpful, for the first time in my college career, in pointing me in the right direction on replying to these arguments. The first time I’ve the word “Pluralism” used by a libertarian has been through the BHL crowd… Kinda off topic now, but just know that your point about the need to engage academia rather than unpack definitions from outside their argument is well received.
– Frustrated student
you used ‘religion makes people unhappy’ in place of ‘humans act using scarce means to achieve their ends’ in an attempt to make a SWPL passive-agressive snipe at praxeology.
you are either dishonest or stupid. take your pick.
You could probably do to lighten up on the rudeness, and really, there’s almost never a good excuse to use SWPL on a libertarian website. Nothing is more SWPL than libertarianism itself.
That said, you have a point. Brennan’s analogy is a poor one. The claim “Religion makes people unhappy” does not match the form of any a priori Austrian claim (that I can think of at the moment).
It’s just too obviously rigged to give him a empirical comeback. And what’s more, it doesn’t tell us whether there’s an empirical comeback to something like “voluntary exchange is ex ante beneficial to both parties”. To complete the analogy he would need to find something like that from psychology.
See my comment on Brennan’s behavior in reply to Jason Becker, if you want. I am calling this as I see it. Brennan does not need any more benefit of the doubt.
You: ” Frankly, he seems like a hack–a second-rate at best psychologist who couldn’t cut it in the academy…”
Him: “Hackman is basically ignored or considered a crank by the academy, but that’s because the academy is corrupt.”
The trouble is Jason, on this point your imaginary opponent is right and you are at least half-wrong.
It may be that Hackman/Rothbard was a hack, but that is clearly NOT the reason for his lack of academic respectability.
The proof of this is that academia frequently tolerates and indeed celebrates obvious hacks when they arrive at the right conclusions. Remember that Sokal, when preparing his hoax, knew that a key ingredient in selling it would be what he called “a pastiche of left-wing cant”. He understand that standards of scholarship would be easily set aside if he simply sounded like a political friend.
So whether he was a hack or not, Hackman’s/Rothbard’s position in the academy was and still is DEFINED by his political conclusions, not by his skill or lack of skill as a scholar.
Hayek and Friedman seem to have had some modicum of success within the academy.
The academy is what the academy is, rightly or wrongly. Part of being a scholar is successfully placing your scholarship within some relevant context. If you come with the approach of “everything you believe, from your priors to your methods to your conclusions, is fatally flawed and here’s why,” you may have a hard time being successful academic. The reasons for this are numerous. Perhaps that is not ideologically pure enough for some, but it is what it is.
Libertarians sometimes have this rather annoying habit of purposefully pursuing the role of the outsider while simultaneously complaining that everyone is ignoring them.
“Hayek and Friedman seem to have had some modicum of success within the academy.”
Yeah, and of course Chinese men are not short because of Yao Ming’s notable success in being really tall.
These facts remain:
1) Most academics are not libertarians, and indeed far from it.
2) Many libertarians find the academy – at least the non-STEM parts – to be silly, inhospitable, corrupt, decrepit, filled to bursting with filthy lucre, etc.
Sounds like a variation on the old joke, “the food at that restaurant is terrible and even worse, I can never get a table.”
You can be a left-wing polemicist and a successful academic in a way that you cannot be a right-wing polemicist and a successful academic. I guess it’s not particularly fair, but… oh well. I can’t say that I’ve ever cared much about what’s fair.
The good news is that there is a relatively easy way around this: don’t be a polemicist.
That won’t work, I’m afraid. Here’s why:
When you have an academy that is dominated to the tune of 90% or 95% by left wingers, that means whole fields will tend to be dominated by dogma (a powerful enough tendency even without politics, by the way).
Inevitably, the dogma in those fields will sometimes clash with reality, and insist on the denial of plain facts.
And when THAT happens, even a non-polemical academic can become a heretic just by clumsily speaking the truth.
This happened a lot in the 1990s. Many of the most notable PC witch hunt victims were intellectually honest old leftists who said obviously true things like: “there is no evidence that the wheel was invented in sub-saharan Africa”, etc.
You see what I’m saying? The dogmatic leftism of the academy mostly just hurts libertarians by leaving us out. It hurts left-wingers by making them ridiculous and irrelevant, and every so often, by arbitrarily ruining an individual career.
Come to think…maybe we have the better end of it.
“Inevitably, the dogma in those fields will sometimes clash with reality, and insist on the denial of plain facts.”
And that’s when you go to work for the Heritage Foundation.
I guess the distinction between the two camps is that a left-wing intellectual is in an environment in the academy where they may be tempted to be dogmatic (as are right-wing intellectuals), while the libertarian in the private sector (who has few left-wing counterparts) is hired to be a shrill political dogmatist, and may in the service of their paymasters sometimes be tempted to attempt to be intellectual.
Rothbard himself says that his “few” postulates about the nature of human action are empirical. How does he observe the truth of these postulates if not introspectively? How does he know that human action seeks to maximize this thing he calls “utility”?
I think rothbard uses “empirical” in a more aristotelian sense, than in, say, a humean sense.
I don’t think Frege’s third realm is “observable”. Yet, I agree with him that the truth of “2 + 2 = 4” is definitely publicly acessible to anyone.
Praxeology, since it is about conceptual truths of human action, is equally open in the same way.
The truth of “2 + 2 = 4” follows tautologically from the definition of the symbols ‘2’, ‘4’, ‘+’ and ‘=’. The numerical abstraction is useful in a variety of contexts, even in formal theories of human action, but human beings and their interaction are far more complex than this abstraction or any other formal system.
Yes and no. Yes, it’s a conceptual truth, so the statement is true by virtue of the meaning of the statement alone. No, it’s not a tautology, at least not in the sense of an empty truth, or a truth by convention, since its meaning does not derive from some kind of convention, it derives from the application of its concepts to reality.
Yes, mathematics does not capture the conceptual truths about human action. Praxeology does.
What does “meaning of the statement” mean? Do you refer to correspondence with some observation here?
Again, to be more specific, what does it mean to say that human beings seek to maximize utility? Rothbard calls this conclusion a logical inference from his Fundamental Axiom of human action.
I’m with you, Brennan. I caught a lot of flak for, well, being a bit critical of Hackman myself. It’s pretty cultish out there.