Wednesday, October 31, 2007

Von Mises.

At the request of Ron in the comments over hyar, I've looked over some material from Ludwig von Mises. My responses are below; basically, I look at the claims I see von Mises making, then offer my critique. This originally came up in the context of a discussion about whether state redistribution of income is a good thing, or even possible.

Science and Value

I see von Mises making a couple of weird, and rather disconnected, claims here. Some do make sense, but, on the whole, this is really showing its age. (Original publication date is 1933.)

He claims that you can do economics and other sciences without making "value judgements". Heterdox epistemology and philosophy of science has pretty much destroyed this view. Saying that neutrality between different perspectives on, say, moral values is itself an expression of a value -- an epistemic one, as well as, possibly, a metaethical one. You're saying it's better or right to be neutral. Value goes all the way down in judging and reasoning, and there's really no escaping it. After all, how can you reason at all without any idea of what counts as a good reason? (To be fair, moral value may have been von Mises' actual target. If it was, then he owes an argument as to why moral value is singled out.)

He also claims that social organization requires private ownership. This is sort of an odd thing to say. I don't disagree nearly as harshly as above, but I think he's suffering from limited perspective here. Maybe it's most charitable to interpret him as saying there must be some private ownership. After all, there are systems where some things are held in common (say, a grazing area for livestock) and they seem to work just fine.

He assumes a position of individualism. That is, the view that the primary unit of concern in economics and "the science of human action" is the individual person. This is fair enough; he's certainly not alone. What's not good, though, is the lack of argument for the view. There is a competing picture in the philosophical tradition that sees social groups as coming first, and individuals emerging only in context. You can see this in Plato quite plainly, but it's also at work in Aristotle to some extent, and was revived more recently in the communitarian critiques of liberalism. The point is, you have to take seriously the opposition and not assume it's obvious that the primary unit of concern is the individual. I, for one, am not sure it is. After all, it's hard to really define oneself without reference to others; most of what makes me me requires a particular social context.

He makes some frankly outdated claims about the reach and scope of science, claiming that its conceptualizations are not universal and unified, and that it can't tell us about things like the mind. I'm sympathetic to the latter, to some extent, but that's a strong minority view. Again, von Mises owes us an argument. Philosophers of science (some, anyway) would object that science can give us a unified picture of the world, through a coherent set of laws. Philosophers of mind (again, some) would object that science tells us the basis of the mind and, again, by a set of reductive laws, we can connect the two together. The whole idea of laws doesn't really seem to be on his radar.

He then talks at length about "intoxicated vision" and the inability of science to provide something "mystical". It's really just noise, as far as I can tell. The best sense I can make of it is that his personal aesthetic sense is such that he finds scientific discoveries austere. Well, that's his problem. Some, like me, find the structure of laws and concepts inherently beautiful. That von Mises apparently didn't is a personal thing of his.

I do agree with a distinction he draws between science and what he calls "metaphysics", though. It seems correct that the two have different methods and different contents. What's sort of weird, though, is the way he wants to split them apart and say that they're not closely intertwined. They of course are. How could they not be? If you do science assuming that you're investigating an objective world, your conclusions and even methods will very likely differ from those of someone who thinks he's just investigating a world in his mind. The first guy will probably do some experiments and tests, and the second guy would sit in his armchair and try to discover the fundamental shape of the universe. (This is, of course, little more than a crude gloss on the empiricism/rationalism business.)

The Impossibility of Economic Calculation Under Socialism (.pdf)

Here, von Mises makes a fairly familiar argument against the planned economy. He points out the difficulties inherent in planning even a simple project centrally, largely based on the inability to reduce decision-making to a mathematical/mechanical process. That's sort of a weird reason to reject planned economies (how about because they don't, in fact, realize their own goals very well?), but it seems fair enough.

What's really bizarre, though, is that he never takes the next step in the argument and consider a market-based socialism. That is, a system that takes seriously the (empirically-confirmed) claim that markets are an amazingly effective tool for social organization, and yet regulates this system in such a way that people tend to make economic decisions that benefit the collective, i.e., society as a whole. That's quite realizable, because it avoids the error he's critiquing. That is, this market socialism seperates the means from the end. The end is something like a more egalitarian, fairer, etc. society, and the means is a carefully-regulated market. So, the disagreement between the market socialist and a liberal or libertarian is a disagreement on (1) the end and (2) the nature of the regulations. (It's worth noting, though, that only an extreme libertarian would object to any regulations. Even Ayn Rand would put at least something in place, e.g., Murder, Inc. is not allowed to sell its assassination services.) So, there's really no argument here against limited redistribution of wealth, say. Spot corrections and manipulations of a market's operation are quite different from the large-scale, detailed planning that von Mises is targetting.

On the whole, von Mises seems to view the central question as strictly dichotomous: either you're in favour of totally free markets, or you're in favour of planned economies. The fact that there's a lot of ground in between those positions doesn't seem to occur to him, and that robs the argument of a lot of its force.

The Starting Point of Praxeological Thinking

This piece is a fairly unreflective way of stating the idea that humans act consciously. Von Mises calls this "a priori" and "self-evident"; but, as is clear from the number of hard determinists running around, it's not a priori or self-evident for everyone. For what it's worth, these are not the same thing; an analytic truth ("a bachelor is an unmarried man") can be a priori and yet not self-evident. Of course, this may be a non-philosophical use of the term.

Given the compatibilist and libertarian (free-will style, not political-style) defenses floating around in the literature, even people who believe humans act consciously don't think it's self-evident. (And they may not think it's a priori; an a posteriori proof of, say, compatibilism seems at least possible to me.)

While von Mises is, of course, entitled to assume for the sake of his project that humans can act for conscious ends, he really needed to take seriously objections from opposing views. Part of the problem is, I suspect, the age of the text (published in 1962). Davidson hadn't published "Actions, Reasons and Causes" (1963), and thus the sea-change in approaches to action was still some distance away.

Some Preliminary Observations
Concerning Praxeology Instead of an Introduction


For this, there is no excuse. Von Mises is trying to argue for a very hard sort of objectivism about the external world. External objects are real and impinge upon our senses. He's writing post-Kant, though, so he should know that the way external objects impinge upon our senses may be hidden from us so completely that we can only ever known the objects as they seem, not as they really are. (And, if he doesn't, why's he trying to do epistemology and metaphysics?)

And there's also problems with the idea that whatever theoretical objects we later concoct (his example is germs, we could also use subatomic particles) are necessarily real things, because they explain what we were always able to do. That assumes, without argument, that the explanations will continue to hold in all cases. They may not. After all, classical mechanics works just fine -- until it doesn't. So, the world as classical mechanics describes it, strictly speaking, is not real. It's just "real enough" for everyday purposes. For all we know (and this is always a live possibility), relativity (the replacement for classical mechanics, at some extremes) may also collapse in some case that, as yet, we have not experienced. It's pretty much a truism in phil of science that (per Quine) theory is underdetermined by data; that is, you can always concoct a different theory that equally well explains the data in front of you. It also follows from this that, while future data may decide a current dispute between theories, there will always be another theory that equally well accounts for even this future data.

7 comments:

undergroundman said...

I have drawing the energy to comment when I'm not high, but one thing stood out at me:
This piece is a fairly unreflective way of stating the idea that humans act consciously. Von Mises calls this "a priori" and "self-evident"; but, as is clear from the number of hard determinists running around, it's not a priori or self-evident for everyone.

That humans act consciously is not at all inconsistent with hard determinism. That should be obvious -- humans act consciously (sometimes), but those conscious actions are still completely caused by prior conditions. Humans are in many cases aware of the causal forces driving their actions, which gives them some ability to preempt that causal chain -- but of course, the extent to which they are aware and actually able to execute based on that awareness is causally determined as well.

He points out the difficulties inherent in planning even a simple project centrally, largely based on the inability to reduce decision-making to a mathematical/mechanical process. That's sort of a weird reason to reject planned economies (how about because they don't, in fact, realize their own goals very well?), but it seems fair enough.

The difficulty of planning correctly is one of major causes for planned economies not reaching its goals very well.

Of course, there are lots of ways to argue against planned economies, not the least the fact that they tend to empower tyrannical governments, who may reach their destructive goals.


On the whole, von Mises seems to view the central question as strictly dichotomous: either you're in favour of totally free markets, or you're in favour of planned economies. The fact that there's a lot of ground in between those positions doesn't seem to occur to him, and that robs the argument of a lot of its force.


I haven't read a lot of Mises but I think you're right. It seems that its hard for a lot of people to grasp that the market-based economics can be tweaked with certain laws, taxes, ect. to be a lot more effective. Requiring information to be distributed is one great way to overcome information problems in the market, for example. The "third wave" in pollution regulation is requiring the information to be made public. (The first was fines, the second was taxes.)

He's writing post-Kant, though, so he should know that the way external objects impinge upon our senses may be hidden from us so completely that we can only ever known the objects as they seem, not as they really are. (And, if he doesn't, why's he trying to do epistemology and metaphysics?)

Von Mises is writing a practical book. The practical implications of Kant's a strictly skeptical interpretation of Kantian metaphysics seems almost nil, besides just a wariness of trusting the senses. But ultimately we have to trust the senses and dig as deep as possible into reality with science.

It's pretty much a truism in phil of science that (per Quine) theory is underdetermined by data

I don't care whether Quine said it, I don't buy it. There's plausible and implausible theories when you look at the theories holistically. You haven't made an argument backing up Quine's argument here, either, and from what I've read, the argument doesn't really make sense.

What you may want to keep in mind is that von Mises is arguing in the 30s against totalitarian planned economies. To say that he's arguing against market-based socialism might be a straw man. Pigou was a contemporary of von Mises, and it would be interesting to see if von Mises ever commented on Pigou's work. Here there a quote which suggests that von Mises was, as you say, opposed to all government intervention based on a slippery slope argument.

I read von Mises' book Bureaucracy. It was fairly simple but I liked that he repeated over and over that basic economics, unlike some other skills, is a necessary skill for citizenship -- because it allows you to analyze policy.

ADHR said...

That humans act consciously is not at all inconsistent with hard determinism. That should be obvious -- humans act consciously (sometimes), but those conscious actions are still completely caused by prior conditions. Humans are in many cases aware of the causal forces driving their actions, which gives them some ability to preempt that causal chain -- but of course, the extent to which they are aware and actually able to execute based on that awareness is causally determined as well.

My fault. "Consciously" is his word, and I read it more broadly than this. If it's just the idea that we're aware of what we're doing, then it seems like a truism.

The difficulty of planning correctly is one of major causes for planned economies not reaching its goals very well.

Of course, there are lots of ways to argue against planned economies, not the least the fact that they tend to empower tyrannical governments, who may reach their destructive goals.


Mm. I still think the more direct argument is that the planned economy doesn't do what it wants to do very well. At the end of the day, that's devastating. Planned economies can't even pass their own test for success. Why the planned economy fails is sort of a different issue.

I haven't read a lot of Mises but I think you're right. It seems that its hard for a lot of people to grasp that the market-based economics can be tweaked with certain laws, taxes, ect. to be a lot more effective. Requiring information to be distributed is one great way to overcome information problems in the market, for example. The "third wave" in pollution regulation is requiring the information to be made public. (The first was fines, the second was taxes.)

That's certainly one option. There's lots of mechanisms available, in principle, to manipulate the results of market operations. Given that, there has to be a real discussion about what the results of the market should be. I don't see it in von Mises.

Von Mises is writing a practical book. The practical implications of Kant's a strictly skeptical interpretation of Kantian metaphysics seems almost nil, besides just a wariness of trusting the senses. But ultimately we have to trust the senses and dig as deep as possible into reality with science.

Fair enough, but why not just say that instead of trying to do bad epistemology and uninformed metaphysics? That is, why not just say "as scientists, we have to trust our senses and see where doing so leads us"?

It's pretty much a truism in phil of science that (per Quine) theory is underdetermined by data

I don't care whether Quine said it, I don't buy it. There's plausible and implausible theories when you look at the theories holistically. You haven't made an argument backing up Quine's argument here, either, and from what I've read, the argument doesn't really make sense.


You're not quite seeing the point still. Quine doesn't say that any theories go. What Quine says is that, in principle, you can always come up with an equally good theory to explain any data set. If this wasn't possible, science would never advance. When there are two equally good competing theories, science (amongst other things) can investigate their predictions and, where they diverge, see which prediction actually obtains, and thus get an improved theory. But, whichever theory is thus vindicated, it, too, can have an equally good competitor. And so the process endlessly repeats. Thus, the data never forces us into a particular theory.

Furthermore, the data never actually forces us to revise a theory. When we're confronted with an observation that seems to violate a theory, according to holists like Quine, we always have the option of rejecting the observation as somehow defective. After all, we have a whole toolbox full of ways to reject observations, and any of them can be invoked to defend a theory that we're loath to give up. Giving up a theory is a choice that we make. It may be a choice we should make, but it's nonetheless a choice.

What you may want to keep in mind is that von Mises is arguing in the 30s against totalitarian planned economies. To say that he's arguing against market-based socialism might be a straw man. Pigou was a contemporary of von Mises, and it would be interesting to see if von Mises ever commented on Pigou's work. Here there a quote which suggests that von Mises was, as you say, opposed to all government intervention based on a slippery slope argument.

Oh, I know the totalitarian planned economies are his major target. But it seems clear to me that there's conceptual ground between totalitarian planned economies and unregulated free markets. Market socialism falls in the middle as does, say, Rawlsian justice-as-fairness. While these are not contemporary ideas for von Mises, it should have occurred to him that he was dealing with extremes, and at least gestured at the middle ground between them.

I read von Mises' book Bureaucracy. It was fairly simple but I liked that he repeated over and over that basic economics, unlike some other skills, is a necessary skill for citizenship -- because it allows you to analyze policy.

I think we've had that discussion before. But, yes, some economics should be required (why is it not in high school?) as well as some basic philosophy, and so on. The core educational requirements of secondary and postsecondary education, although supposedly designed to create good citizens, generally don't do any such thing. I'm not sure they ever really did. (This is not to say that something like English literature should be sacrificed in favour of economics. But perhaps we should teach a little less lit and a little more of something else. To say nothing of the poor way sciences, hard and soft, are taught at the high school level.)

undergroundman said...

You're not quite seeing the point still. Quine doesn't say that any theories go. What Quine says is that, in principle, you can always come up with an equally good theory to explain any data set. If this wasn't possible, science would never advance. When there are two equally good competing theories, science (amongst other things) can investigate their predictions and, where they diverge, see which prediction actually obtains, and thus get an improved theory.

I think I get it and disagree with it still. I see humans to a large extent (especially when doing science and coming up with theories) as data-interpreting machines. The fact that we have theories which seem right to us and ones that don't proves that there are theories which are better than others. On the other hand, there are theories which are equally possible -- which is what I thought that Quine's argument was. You could be a sophisticated bot created by an intelligent machine. That's a theory which explains the data that I've gathered, but it doesn't explain it as well as the alternative theory that you are just a philosopher, when the big picture is analyzed. I love examples, as always...you want to offer some examples of these "just as good" yet not used theories?

This is not to say that something like English literature should be sacrificed in favour of economics.

English Lit., like so many other things, is best pursued in one's own time with one's own interest. I learned a lot of nonsense in high-school. Didn't pay attention in math, took all biology and no chemistry, got droned at in history, and was read to in English. High-school could use a lot of improvement -- and that would probably require federal standards.

I never took economic classes in high-school, but from what I've heard they don't emphasize economic concepts such as opportunity costs, declining marginal returns, ect. as much they do how to save and such. Which is sad.

ADHR said...

The issue isn't really equal possibility. More equal coherence with the overall theoretical structure by which we understand the world. Keep in mind that Quine's a holist in this regard. So, when a theory seems right to you, says Quine, that's because of your underlying epistemic and semantic theories. If there were a severe conflict between those underlying theories and something, you always have a choice: accept the new thing and reject your underlying theories, or accept the underlying theories and reject the new thing.

The point, ultimately, is that data alone doesn't do it. There has to be something that comes from you involved. Which is really just Kant all over again.

As for English lit, I tend to think it's worth doing as long as it's taught correctly. It can make you a better writer, and it can also make you a better citizen by exposing you to alternate perspectives. Often, though, it turns into some turgid over-analysis of some obscure poem.

I'm not opposed to teaching basic finance in high school. Lord knows it has to be taught somewhere. But not under the guise of economics.

Ron said...

just a note to let you know that I apreciate your response to this and I'm still thinking--and still crazy busy...

undergroundman said...

As for English lit, I tend to think it's worth doing as long as it's taught correctly. It can make you a better writer, and it can also make you a better citizen by exposing you to alternate perspectives. Often, though, it turns into some turgid over-analysis of some obscure poem.

It's funny how we talk about education as functional -- it can make you a better citizen, better writer, more creative, ect. Are things like English literature not good in their own right? :p

I wonder what Nietzsche would think, since he seemed to think that appreciating art for its own sake was so important (life is art, really) -- but at the same time seemed rather oriented towards functionality and improvement. I don't know. Haven't had the time (heart, really) to read much lately. Been sickly as usual.

ADHR said...

I think English literature is good in its own right; I'm just not sure that's a good justification for including it in public schooling.

So, while it may be right that we should appreciate art for its own sake, I'm not sure it's right that the public should fund one's attempts to appreciate art. (As compared to helping one develop the ability to appreciate art, which may be justifiable. See earlier re: being a better citizen, etc. ;) )