12 Sep 2013

Mises on A Priori Reasoning

This comes from page 38 in the Scholar’s Edition of Human Action. I just had to paste this into a project I’m working on; thought it would be relevant:

All geometrical theorems are already implied in the axioms. The concept of a rectangular triangle already implies the theorem of Pythagoras. This theorem is a tautology, its deduction results in an analytic judgment. Nonetheless nobody would contend that geometry in general and the theorem of Pythagoras in particular do not enlarge our knowledge. Cognition from purely deductive reasoning is also creative and opens for our mind access to previously barred spheres. The significant task of aprioristic reasoning is on the one hand to bring into relief all that is implied in the categories, concepts, and premises and, on the other hand, to show what they do not imply. It is its vocation to render manifest and obvious what was hidden and unknown before.

In the concept of money all the theorems of monetary theory are already implied. The quantity theory does not add to our knowledge anything which is not virtually contained in the concept of money. It transforms, develops, and unfolds; it only analyzes and is therefore tautological like the theorem of Pythagoras in relation to the concept of the rectangular triangle. However, nobody would deny the cognitive value of the quantity theory. To a mind not enlightened by economic reasoning it remains unknown. A long line of abortive attempts to solve the problems concerned shows that it was certainly not easy to attain the present state of knowledge.

163 Responses to “Mises on A Priori Reasoning”

1. Daniil Gorbatenko says:

But was Mises really correct in this quote? I mean if you look at all the available proofs of Pythagoras theorem, you’ll see that none of them is constructed by looking at a single rectangular triangle. All of them involve using other geometric figures in addition to it. It would thus seem that no knowledge is analytic, not even the Pythagoras theorem.

• Major_Freedom says:

All geometric relations are a priori.

• Daniil Gorbatenko says:

Depends on what you mean by a priori. If you mean “something that is necessarily true” then I agree. If you mean “something that follows from a definition” then, as I think I’ve shown, say Pythagoras theorem isn’t a priori because it doesn’t follow logically from the definition of a rectangular triangle.

• Major_Freedom says:

The relation follows logically from the meaning of a right triangle.

• Daniil Gorbatenko says:

You seem to misunderstand what logical implication is. An example is modus ponens:

If X is a man, he is mortal. Socrates is a man. Therefore, he is mortal.

The Pythagorean theorem doesn’t follow like this from the definition of a rectangular triangle. You can see it yourself if you look at all the available proofs.

• Major_Freedom says:

There are logical forms other than that which you describe.

Pythagoras’ relation does follow logically. The fact you aren’t aware of it, doesn’t negate its existence.

I think you’re conflating what you don’t know, with what doesn’t exist.

• Daniil Gorbatenko says:

So can you describe how the Pythagorean theorem logically follows from the definition of a rectangular triangle or give a link to such a description?

• guest says:

Lord Keynes: Kant’s belief in the synthetic a priori is false, and we know this now given the empirical evidence in support of non-Euclidean geometry: this damns Kant’s claim that Euclidean geometry – the geometry of his day – was synthetic a priori …

(Note: I am unfamiliar with the debate for and against Euclidean geometry, and am not commenting on that debate, per se.)

This argument requires an appeal to the aprioristic reasoning he claims to reject.

He’s not saying there’s a data point that says “false”; He’s attempting to apply logic TO data. The logic is aprioristic.

As an aside, LK says he knows synthetic apriorism is false because Euclydian geometry isn’t an example of synthetic apriorism.

But it does not follow logically that synthetic apriorism is false from proof that what someone claimed was an example of it turned out not to be true; At best, all it could prove is that THAT particular thing was not an example of it.

At any rate, as Rothbard noted, Praxeology is aprioristic in the sense that matters for an understanding of economic laws:

In Defense of “Extreme Apriorism”
[WWW]http://mises.org/daily/5195/

Now the crucial question arises: How have we obtained the truth of this axiom? Is our knowledge a priori or empirical, “synthetic” or “analytic”? In a sense, such questions are a waste of time, because the all-important fact is that the axiom is self-evidently true, self-evident to a far greater and broader extent than the other postulates. For this axiom is true for all human beings, everywhere, at any time, and could not even conceivably be violated. In short, we may conceive of a world where resources are not varied, but not of one where human beings exist but do not act. We have seen that the other postulates, while “empirical,” are so obvious and acceptable that they can hardly be called “falsifiable” in the usual empiricist sense. How much more is this true of the axiom, which is not even conceivably falsifiable!

Positivists of all shades boggle at self-evident propositions. And yet, what is the vaunted “evidence” of the empiricists but the bringing of a hitherto obscure proposition into evident view? But some propositions need only to be stated to become at once evident to the self, and the action axiom is just such a proposition.

Whether we consider the action axiom “a priori” or “empirical” depends on our ultimate philosophical position. Professor Mises, in the neo-Kantian tradition, considers this axiom a law of thought and therefore a categorical truth a priori to all experience. My own epistemological position rests on Aristotle and St. Thomas rather than Kant, and hence I would interpret the proposition differently. I would consider the axiom a law of reality rather than a law of thought, and hence “empirical” rather than “a priori.” But it should be obvious that this type of “empiricism” is so out of step with modern empiricism that I may just as well continue to call it a priori for present purposes. For (1) it is a law of reality that is not conceivably falsifiable, and yet is empirically meaningful and true; (2) it rests on universal inner experience, and not simply on external experience, that is, its evidence is reflective rather than physical;[7] and (3) it is clearly a priori to complex historical events.[8]

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.[10]

• Major_Freedom says:

“But it does not follow logically that synthetic apriorism is false from proof that what someone claimed was an example of it turned out not to be true; At best, all it could prove is that THAT particular thing was not an example of it.”

Correct.

But even so, Euclidean geometry was not in any sense refuted or falsified by non-Euclidean geometry. After all, the very equipment and tools used by cosmologists to observe a non-Euclidean spacetime, are themselves built on Euclidean geometry!

• Keshav Srinivasan says:

I don’t think the equipment argument is very convincing. The tools are built based on the fact that space is locally approximately Euclidean.

• Lord Keynes says:

“But even so, Euclidean geometry was not in any sense refuted or falsified by non-Euclidean geometry.”

It was falsified as a necessarily true universal theory of real space known a priori, and true throughout the universe.

• Daniil Gorbatenko says:

Euclidian geometry applies if the space to the analysis of which it is applied is Euclidian. It is the same with praxeology. Praxeology applies to the subjects that have free minds. Since a free mind to which praxeology doesn’t apply is inconceivable, it’s meaningless to speak of such a mind. Whereas a non-Euclidean geometry is quite conceivable.

• Major_Freedom says:

“It was falsified as a necessarily true universal theory of real space known a priori, and true throughout the universe.”

No it wasn’t. Once again, Euclidean geometry cannot possibly refute itself, for that would presuppose its legitimacy.

When you say “falsified”, you are, whether you know it or not, claiming that it was falsified using Euclidean geometry. The equipment, tools, and the researcher’s treatment of the equipment and tools in relation to him- or herself, is Euclidean.

Euclidean geometry is, and was always true a priori. It still is true a priori. It is how our minds of necessity must work, and hence an actual fact of the universe.

The “universal” aspect of it is that it always and forever applies to action from actors such as ourselves wherever and whenever they exist.

Empirical non-Euclidean geometry is grounded on a priori Euclidean geometry.

Your claim that non-Euclidean geometry falsifies a priori Euclidean geometry defined as “true throughout the universe” is a straw man, because Kant never intended Euclidean geometry to be an explanation in the “dogmatic” sense of being true regardless of our minds, that is, true external to ourselves “out there”.

• Lord Keynes says:

“When you say “falsified”, you are, whether you know it or not, claiming that it was falsified using Euclidean geometry. The equipment, tools, and the researcher’s treatment of the equipment and tools in relation to him- or herself, is Euclidean.”

Irrelevant. Since we discovered that Euclidean geometry is a useful *approximation* of reality in a certain restricted domain, its usefulness there does not refute the empirical discovery that it is not a description of 3 dimensional curved spacetime.

” It is how our minds of necessity must work, and hence an actual fact of the universe.

The internal traits of the human mind and experience do not provide necessary knowledge of the external universe known a priori

” Kant never intended Euclidean geometry to be an explanation in the “dogmatic” sense of being true regardless of our minds, that is, true external to ourselves “out there”.

On the contrary, Kant held Euclidean geometry is a necessarily true theory of real space, not just some state of our minds.

• Major_Freedom says:

“Irrelevant. Since we discovered that Euclidean geometry is a useful *approximation* of reality in a certain restricted domain, its usefulness there does not refute the empirical discovery that it is not a description of 3 dimensional curved spacetime.”

It is not irrelevant, because if you concede that YOUR A PRIORI ASSUMPTIONS, i.e. Euclidean geometry, are “an approximation”, then you can’t be certain that all of spacetime is non-Euclidean.

In other words, by undercutting Euclidean geometry, you have undercut your own conclusion that is grounded on what you are undercutting.

This is highly “relevant”, because if you’re going to go around claiming what the universe is and is not, you must have a foundation that is, not a foundation that might be, or seems to be.

“The internal traits of the human mind and experience do not provide necessary knowledge of the external universe known a priori”

You need to read Mises and Hoppe again, because they have shown that yes, because thinking is itself an action, that because action is a part of the world, then anything a priori true about thinking IS also a priori true about the real world.

Action is the bridge that connects what is true for our thinking, and what is true for the world.

“On the contrary, Kant held Euclidean geometry is a necessarily true theory of real space, not just some state of our minds.”

False. You obviously have not read Kant either. Kant was ADAMENT that his a priori categories are only applicable to objects OF EXPERIENCE, that is, constrained to our minds.

• Lord Keynes says:

“because if you concede that YOUR A PRIORI ASSUMPTIONS, i.e. Euclidean geometry, are “an approximation”, then you can’t be certain that all of spacetime is non-Euclidean.”

I don’t concede that any “a priori” assumption demonstrates that Euclidean geometry is “an approximation” of reality. Only empirical evidence proves that.

You’re inventing a straw man argument.

• Major_Freedom says:

“I don’t concede that any “a priori” assumption demonstrates that Euclidean geometry is “an approximation” of reality. ”

You do by virtue of the fact that you utilize it when “approximating reality.” You can’t help but do it. It is necessary. It is why even cosmologists of non-Euclidean spacetime use it.

You don’t have to consciously accept X, before you can be shown to presuppose X.

It isn’t a straw man. You’re just in denial.

• Economic Freedom says:

>>>It was falsified as a necessarily true universal theory of real space known a priori, and true throughout the universe.

Was it asserted as a necessarily true universal theory of real space known a priori, and true throughout the universe? If so, by whom?

• Lord Keynes says:

Kant and many supporters of synthetic a priori knowledge.

• Major_Freedom says:

False. Kant held those a priori categories to only be relevant to objects of experience.

Kant did not hold them to be a “necessarily true universal theory of space time and throughout the entire universe”.

Kant himself wrote:

“But although this knowledge is limited to objects of experience, it is not therefore all derived from experience.” – Critique of Pure Reason, Section 2
Transcendental Deduction of the Pure Concepts of the Understanding, §27
Outcome of this Deduction of the Concepts of Understanding.

You’re so wrong it hurts.

• Matt Tanous says:

“Kant’s belief in the synthetic a priori is false, and we know this now given the empirical evidence in support of non-Euclidean geometry: this damns Kant’s claim that Euclidean geometry – the geometry of his day – was synthetic a priori”

Does not follow. Synthetic a priori indicates that the truth of a statement follows form a combination of its meaning and facts regarding the world. Non-Euclidean geometry is the same as Euclidean in this regard. They remain axiomatic systems with the same relations.

Geometry is a priori – Euclidean or not – as the geometric laws all are derived from the definition of the constraints of the system. It is synthetic in the sense that one must use observation to know which geometry applies to reality. This is synthetic a priori truth – that which LK denies because he absurdly postulates, as his initial premise, that if something is synthetic (initial premises relying on empirical observation for application to reality) they cannot be a priori (derived from the meaning of the premises alone). LK fails at epistemology.

• Bala says:

“LK fails at epistemology.”

Glad to see many people saying this….. 🙂

• Lord Keynes says:

“Synthetic a priori indicates that the truth of a statement follows form a combination of its meaning and facts regarding the world”

This is so badly and stupidly wrong, it suggests you do not understand even the basics of epistemology concepts.

A synthetic a priori statement is:

(1) not semantically analytic
(2) known without any recourse or appeal to empirical evidence and
(3) necessarily true.

• Major_Freedom says:

(2) is false.

Kant did not hold that synthetic a priori knowledge is “known without any recourse or appeal to empirical evidence.” He wrote:

“THERE can be no doubt that all our knowledge begins with experience. For how should our faculty of knowledge be awakened into action did not objects affecting our senses partly of themselves produce representations, partly arouse the activity of our understanding to compare these representations, and, by combining or separating them, work up the raw material of the sensible impressions into that knowledge of objects which is entitled experience? In the order of time, therefore, we have no knowledge antecedent to experience, and with experience all our knowledge begins.”

You are stupidly wrong. This is the very first paragraph of the Critique!

A priori knowledge is not 100% divorced from experience.

• Lord Keynes says:

A passage taken out of context.

Kant thought that the forms and categories are prior to experience.

And Kant did not think that a synthetic a priori statement was known at all by appeal to experience. If he did, it would not be known a priori.

• guest says:

This is saying what I was trying to share, before: That without experience, we could not know that we exist.

• Major_Freedom says:

“A passage taken out of context.”

Haha, no it is a passage that introduces Kant’s view and refutes your claim that he held that a prioriu knowledge is “known without any recourse or appeal to empirical evidence.”

You are wrong to claim there is NO recourse to empiricism. He believed all knowledge starts with experience!

“And Kant did not think that a synthetic a priori statement was known at all by appeal to experience. If he did, it would not be known a priori.”

The meaning of an “a priori” proposition is that it is not cognitively PROVEN by appeal to external experience. Not that it is totally unrelated to experience.

Without experience, Kant held that all knowledge would be impossible.

• guest says:

Game over.

As Rothbard noted, the Action Axiom is a priori in the sense that matters:

In Defense of “Extreme Apriorism”
[WWW]http://mises.org/daily/5195/

Now the crucial question arises: How have we obtained the truth of this axiom? Is our knowledge a priori or empirical, “synthetic” or “analytic”? In a sense, such questions are a waste of time, because the all-important fact is that the axiom is self-evidently true …

That’s really all that matters, here. You don’t have to empirically test the logical implications of the axiom to know that they’re true.

• Lord Keynes says:

And in any case, Kant thought that space (as Euclidean space) and time are the two “pure forms of intuition” and are both a priori.

• Major_Freedom says:

Yes, but he held that they are true a priori in being limited to objects of experience.

He never made any argument that you claimed he made, which is that the categories are true everywhere and throughout the universe, i.e. in the noumenal world as well as the phenomenal world.

“We cannot think an object save through categories; we cannot know an object so thought save through intuitions corresponding to these concepts. Now all our intuitions are sensible; and this knowledge, in so far as its object is given, is empirical. But empirical knowledge is experience.Consequently, there can be no a priori knowledge, except of objects of possible experience.”
and
“But although this knowledge is limited to objects of experience, it is not therefore all derived from experience.” – The Critique of Pure Reason, Deduction of the Pure Concepts of the Understanding, Section 2, Transcendental Deduction of the Pure Concepts of the Understanding, §27 Outcome of this Deduction of the Concepts of Understanding

2. Bharat says:

Haha, so many posts on methodology. Now I’m starting to think you lost a bet with Wenzel or something.

• Dan says:

I think he has just been trying to demonstrate how difficult it is for most people to comprehend what Mises was saying on this issue. The amount of people who haven’t seemed to grasp what Murphy’s been saying is a good case for not abandoning the “defeat them on their own terms” strategy.

• Daniel Kuehn says:

One troubling tendency I’ve seen, though, is to confuse “thinking there are more options than just Friedman on the one hand and Mises on the other” with “not understanding what Murphy’s been saying”.

• Dan says:

I’d have to see you explain what you believe Murphy’s position is before I’d accept that you understand it. Because I’ve seen you too many times say something along the lines of “Austrians are in favor of the unscientific approach to economics.” In fact, I thought you were just being intentionally obtuse at one point, which is why I said you were acting childish, but now I just think you haven’t fully grasped the Austrian position on this. I could be wrong, but I would need to see you accurately describe our view before I’d believe you get it. I need empirical proof.

• Daniel Kuehn says:

Well I’d need to see you explain his view before I’m convinced you’re in any position to pass judgement on the question.

Part of the problem is that if you’re working off of two definitions of science, you’re going to potentially get two answers to the question of whether Austrians advocate a scientific or unscientific view of economics even if you are on the same page on what it is that Austrians think.

What (some) Austrians refer to as economic history, which they distinguish from the science of economics, is in my view essential to the scientific credentials of the science of economics. So I am going to assess any separation of that endeavor as inherently de-scientizing.

The “some” is important. A whole lot of Austrians are not of this view, and that’s important to keep in mind in this conversation too.

• Daniil Gorbatenko says:

=What (some) Austrians refer to as economic history, which they distinguish from the science of economics, is in my view essential to the scientific credentials of the science of economics. So I am going to assess any separation of that endeavor as inherently de-scientizing.=

Would you say the same of physics and physical history? And if not how is economics different?

• Daniel Kuehn says:

The physical or natural analog is part of physical science/natural science. I understand them symmetrically. Physicists looking into the Big Bang are scientists. Paleontologists are scientists.

• Matt Tanous says:

“Physicists looking into the Big Bang are scientists.”

Not really. They’re forensic experts regarding the physical universe. The CSI tech running blood samples is not a scientist, is he?

• Daniel Kuehn says:

The CSI is not making general inferences about how the human body works. Someone doing very similar work to the CSI, but who is using the data to make inferences about the human body is called a medical researcher.

• Dan says:

“Well I’d need to see you explain his view before I’m convinced you’re in any position to pass judgement on the question.”

You don’t have to rely on my judgement, you can rely on Murphy’s. it’d be pretty simple to explain what you believe is his position and let him say whether you correctly explained it or not. That way you’d be able to see if you guys are just talking past each other.

• Daniil Gorbatenko says:

I think Bob never said that economic history isn’t a science. The main point Austrians make about theory and history is that you can only do history through the lens of a theory.

• Daniil Gorbatenko says:

Oops, sorry this reply was intended to Danel.

3. Brian Albrecht says:

Reading the passage, specifically the 2nd paragraph, I wonder- is it worth knowing all the logical deductions from Pythagoras? While, of course it would be nice, but is it worth the cost? Or it it better to skip steps and know the results?

Similarly, is it better for people to learn the quantity theory than the whole deduction from the action axiom? Yes, it might not improve the frontiers of knowledge, but most of us won’t be on the frontiers anyways. Maybe it is more important, for most people, to skip a few steps and learn the implications of the quantity theory. Of course, each person will have their own subjective “optimal knowledge” set.

Where am I wrong?

• Major_Freedom says:

“Or it it better to skip steps and know the results?”

Somebody has got to do the steps if another is to know the results without the steps.

4. Lord Keynes says:

Nonetheless nobody would contend that geometry in general and the theorem of Pythagoras in particular do not enlarge our knowledge.

By that Mises means provide necessary knowledge of the real world known a priori — or the Kantian synthetic a priori knowledge.

He is wrong. Euclidean geometry provides no epistemologically necessary knowledge of the real world, nor can we know if it is a true description of reality a priori. When asserted of the real world, it becomes synthetic a posteriori and contingently true.

Euclidean geometry is of course necessary a priori only when considered as pure geometry or a analytic a priori system. But it asserts nothing necessarily true of the real world.

http://socialdemocracy21stcentury.blogspot.com/2013/08/bob-murphy-all-at-sea-on-geometry-and.html

• Daniil Gorbatenko says:

LK, you conveniently forget the Aristotelian position in relation to both praxeology and mathematics. E.g. my first comment to this post, if correct, casts into doubt the idea that Euclidean geometry is tautological. But there doesn’t seem to be any doubt that Pythagoras theorem is correct and that it applies to the real world (albeit imperfectly).

Now, I’m not a mathematician, nor a philosopher of mathematics but there are serious thinkers who’ve been developing this (Aristotelian) view of mathematics. As an example and a very good exposition of the view see this http://web.maths.unsw.edu.au/~jim/irv.pdf

• Bob Murphy says:

Lord Keynes,

I was going to be snotty about it, but that would be unfair since many Misesians thought Mises was making synthetic a priori propositions. But, look at literally the sentence right before the one you quoted. Mises wrote:

This theorem is a tautology, its deduction results in an analytic judgment. Nonetheless nobody would contend that geometry in general and the theorem of Pythagoras in particular do not enlarge our knowledge.

So Mises is here saying that the Pythagorean theorem is an analytic a priori statement.

I agree though that he was wrong when he said, “Nonetheless nobody would contend…” because Lord Keynes contended that very thing in his comment.

• Bala says:

And I couldn’t repress my smile when I read the blog article and focused my attention on the clause you have now placed in bold 🙂

I love this series on methodology.

• Lord Keynes says:

Your interpretation of Mises is wrong.

(1) Even in the passage you quote selectively, the whole point is to deny that the tautological character of Euclidean geometrical theorems means that “it cannot add anything to our knowledge.” That is, Mises is implying that Euclidean geometry has the same status as synthetic a priori knowledge.

(2) Mises also thinks that praxeological theorems are necessarily and absolutely true and are known a priori but also yield necessary knowledge of the real world. That is nothing but Kantian synthetic a priori knowledge too.

See here:

http://socialdemocracy21stcentury.blogspot.com/2013/09/robert-murphy-gets-misess-epistemology.html

• Major_Freedom says:

(1) False. Mises held that even analytic a priori knowledge expands our knowledge.

(2) It’s not “nothing but” Kant, because Kant didn’t go as far as Mises when it comes to action.

• Lord Keynes says:

(1) Even if that were true, it demonstrates colossal ignorance of basic logical concepts on Mises’s part, and does not refute my point.

(2) It is still epistemologically in the same category as synthetic a priori knowledge, whatever the differences between Mises and Kant on human action.

• Major_Freedom says:

(1) No, it demonstrated that even a priori knowledge is not given to us nominally, the way apples and pears are, that we must think them through. Pythagoras relation is tought to people precisely because it isn’t self-evident. This is the very definition of expansion of our knowledge. You are collosally ignorant of logic if you can’t distinguish betweena priori propositions that are self-evident, and what has to be thought through.

(2) No, it is not the same, because Kant did not realize wht Mises realized about action.

• Lord Keynes says:

(1) Murphy says that Mises interprets the Pythagorean theorem is an analytic a priori statement.

On the contrary, Mises is saying it is synthetic a priori.

Your irrelevant ramblings do not refute my statement.

(2) So now you’re saying that the action axiom or praxeological theorems does not provide synthetic a priori knowledge? lol.

The debate is over. Praxeology is rubbish.

• Bob Murphy says:

LK wrote:

Murphy says that Mises interprets the Pythagorean theorem is an analytic a priori statement.

On the contrary, Mises is saying it is synthetic a priori.

LK, does it bother you that

(a) Mises never in Human Action uses the term “synthetic a priori”

and

(b) I have here quoted him as explicitly saying the Pythagorean theorem is an analytic statement?

I’m not sure what else you would need to see to admit you are wrong on this.

Maybe Mises SHOULD be classifying it as synthetic a priori, but he’s not.

• Lord Keynes says:

“LK, does it bother you that

(a) Mises never in Human Action uses the term “synthetic a priori””

Yes, he never uses the expression “synthetic a priori” directly, but instead uses synonymous expressions and circumlocutions.

And Mises does explicitly mention and defend “synthetic a priori” truth in The Ultimate Foundation of Economic Science: An Essay on Method (1962), p. 5.

Does this latter fact bother you?

(b) I have here quoted him as explicitly saying the Pythagorean theorem is an analytic statement?

No, Bob Murphy, you have selectively quoted him
and misinterpreted the meaning of this passage.

This would be clear if you had bothered to read my refutation of you here:

http://socialdemocracy21stcentury.blogspot.com/2013/09/robert-murphy-gets-misess-epistemology.html

• Bob Murphy says:

LK you are simply wrong. You can try whatever gymnastics you want; Mises is clearly saying the Pythagorean theorem is analytic, and your only real argument is to say he must be confused.

Look, I AGREE with you that a strong case can be made that Mises SHOULD be classifying praxeology as synthetic a priori; after all, that’s what Hoppe does, and I taught that to my students at Hillsdale when I did this stuff. I am making the modest point that Mises himself did NOT do so.

Your quotes from UFofES also don’t work. Mises is saying that the logical positivist claim “only experience can lead to synthetic statements” would itself be a synthetic a priori statement, if it were true, but it is not true (in the opinion of Mises, as he says). So, Mises nowhere in that passage says he believes that there are true synthetic a priori statements; he is merely doing jiu jitsu on the logical positivists when they think they can make a blanket declaration that no such statements could be true.

Again, I am acknowledging that it’s very subtle; had David Gordon not brought it to my attention, I would have gone to my grave thinking Mises classified praxeology as synthetic a priori. But, after David showed me what he was talking about, I now can see everything you’re offering as evidence, as showing that Mises didn’t think geometry or praxeology was synthetic a priori.

UPDATE: Actually LK I re-read that passage from the Ultimate Foundations and you’re right, that looks like Mises is saying he believes that there are true synthetic a priori propositions, because he seems to be saying the statement “there are no true synthetic a priori statements” is false. So, I am backing off my claim that Mises denied there could be such statements. However, it is still crystal clear to me that he thought the Pythagorean theorem was analytic, since he explicitly said so and you have given me no evidence to the contrary.

• Major_Freedom says:

(1) It is irrelevant what Murphy said and didn’t say when it comes to my responses to your false claims. At any rate, Murphy is right, and you are wrong. Mises did not hold that Pythagoras in synthetic a priori. He held it was analytic a priori.

(2) No, I am not saying that at all. I am saying that Mises’ action is on a higher epistemological plane than Kant, because not even Kant understood the profound implications of action. He only glimpsed them.

• Major_Freedom says:

Hahaha

Kicking sand in the sandbox doesn’t count, LK.

• Lord Keynes says:

“But, after David showed me what he was talking about, I now can see everything you’re offering as evidence, as showing that Mises didn’t think geometry or praxeology was synthetic a priori.”

If you really believe this, then you have demonstrated that Mises was one of the most confused and incompetent economists and methodologists ever.

For now Mises’s and your own claim that praxeology cannot be refuted by experience must collapse.

If praxeology is nothing but analytic a priori, then it is tautologous and asserts nothing necessarily true of the real world.

The only way you can demonstrate its truth is empirically (a posteriori).

• Bob Murphy says:

LK wrote:

If praxeology is nothing but analytic a priori, then it is tautologous and asserts nothing necessarily true of the real world.

LK, please, I beseech you, go re-read the first paragraph in the Mises quote from this very blog post. He is taking on your claim here, and saying it is wrong. Now maybe he’s wrong, but you have to stop acting like you’re doing Mises a favor by saying “he can’t possibly mean that.” He is clearly saying that a tautologous analytic statement enlarges our knowledge of the real world. If you think that makes him an idiot, OK fine, but you have to realize that’s what he is clearly saying.

• Lord Keynes says:

UPDATE: Actually LK I re-read that passage from the Ultimate Foundations and you’re right, etc.

So we’ve now established that Mises does indeed seem to be defending the existence of synthetic a priori in “Ultimate Foundations”.

It’s but a short step for you to read that original Mises quote as I’ve quoted it in full on my blog:

http://socialdemocracy21stcentury.blogspot.com/2013/09/robert-murphy-gets-misess-epistemology.html

In particular, see the last paragraph:

“The theorems attained by correct praxeological reasoning are not only perfectly certain and incontestable, like the correct mathematical theorems. They refer, moreover, with the full rigidity of their apodictic certainty and incontestability to the reality of action as it appears in life and history. Praxeology conveys exact and precise knowledge of real things.” (Mises 2008: 38).

The only way praxeology can do this is if it really is synthetic a priori.

• Bob Murphy says:

Hmm very interesting LK. I have to call in backup on this one.

• Lord Keynes says:

“He is clearly saying that a tautologous analytic statement enlarges our knowledge of the real world.”

If it has not been given an empirical hearing (as in applied geometry) and proven a posteriori, then
the only way a tautologous analytic statement can “enlarge our knowledge of the real world” is by being a synthetic a priori statement.

• Bob Murphy says:

Bob: “[Mises] is clearly saying that a tautologous analytic statement enlarges our knowledge of the real world.”

LK: If it has not been given an empirical hearing (as in applied geometry) and proven a posteriori, then
the only way a tautologous analytic statement can “enlarge our knowledge of the real world” is by being a synthetic a priori statement.

Is anyone else amused by the fact that Lord Keynes continually ignores his sensory observations–namely what his eyes show him Mises wrote–and clings to his a priori definitions of what analytic and synthetic mean?

• Lord Keynes says:

In case it’s unclear to you, I am saying Mises was mistaken and confused if he thought such a statement was analytic. Of course, it must be synthetic. He logically needs the epistemological category of synthetic a priori here.

This is exactly the point made by H. Albert, 1999. Between Social Science, Religion and Politics: Essays in Critical Rationalism. Rodopi, Amsterdam. pp. 131–132.

• Bala says:

Bob,

Get away before you hurt yourself.

“Of course, it must be synthetic.”

Why must it be if

“a tautologous analytic statement enlarges our knowledge of the real world”?

• Bala says:

Sorry Bob. That question was not for you.

• Guillermo Sanchez says:

LK: “In other words, the collapse of Euclidian geometry as synthetic a priori knowledge does not apply to the synthetic a priori status of praxeology!”

Yes! because “praxeology is not geometry”!!! Mises clearly says they are NOT the same thing.

LK: “This, if nothing else, is breathtaking in its pig-headed unwillingness to reconsider the epistemology status of praxeology given the fall of Euclidian geometry as the paradigmatic case of synthetic a prioriknowledge.”

I can rearrange your own words to apply them to other case in order to to show you how odd is your assertion. “In other words, the collapse of neo-keynesian as economic theory in the 70’s does not apply to the economic theory status of Keynesianism!” “This, if nothing else, is breathtaking in its pig-headed unwillingness to reconsider the economic theoretical status of keynesianism given the fall of neo-keynesianism as the paradigmatic case of macro-economic theory.”

The fall of neo-keynesianism in the 70’s as THE paradigmatic case of Keynesianism is not going to convince you that keynesianism is a big failure pseudo-theory. Because you will say that neo-keynesianism is NOT post-keynesianism. Well, in the same way praxeology is NOT geometry. You have 2 choices: a) you accept that neo-keynesianism = post-keynesianism and Mises is wrong or b) neo-keynesianism =/= post-keynesianism and Mises is right.

One of your biggest problem, and I have notice it a lot of times, is that you arbitrarily do not apply your austrian “criticisms” to your own keynesian beliefs.

• Lord Keynes says:

You misunderstand the argument:

(1) praxeology’s status as synthetic a priori must fall because the human action axiom is not synthetic a priori:

http://socialdemocracy21stcentury.blogspot.com/2013/07/what-is-epistemological-status-of.html

(2) If anyone ever claimed that all versions of Keynesianism are synthetic a priori, then your criticisms might have some merit. However, nobody ever has.

And on the contrary, the failure of neoclassical synthesis Keynesianism prompted many to careful reexamination of basic Keynesian theories.

The neoclassical synthesis Keynesianism was a bastardised version of Keynes’s theory; it was flawed because it rejected crucial elements of Keynes’s GT, in which stagflation can occur.

• RIchard Moss says:

I know plenty of people who use the findings of Euclidean geometry every day in the real world.

Tell you what – let’s do away with those findings right now since they and the real world have nothing to do with each other. Let’s see how far we get after that.

Perhaps we might then see how Euclidean geometry enlarged our knowledge.

• GeePonder says:

@Richard,

I just fell out of my chair. At least it used to be a chair.

• Lord Keynes says:

We only that Euclidean geometry is useful and an approximation of reality empirically, not a priori.

• Lord Keynes says:

” We only *KNOW* that”

• Richard Moss says:

So, that means it doesn’t enlarge our knowledge?

• Major_Freedom says:

Then why do you presuppose its validity before you even conduct empirical experiments? Why do you presume its validy in the course of experiments?

• Lord Keynes says:

Its validity is not presupposed: just because scientific objects might have been constructed with Euclidean geometry does not prove Euclidean geometry is a necessarily true universal theory of all space known a priori.

It is the case that is an *approximation* of reality under certain conditions. But we only know this a posteriori.

• Major_Freedom says:

“Its validity is not presupposed”

Yes it is. You are not using non-Euclidean geometry to prove or disprove Euclidean geometry.

You are using Euclid.

Anything you say of Euclid will affect what you say about non-Euclid.

• Lord Keynes says:

Your using “validity” in a sense different from me, and are committing nothing but classic fallacy of equivocation.

As for Hoppe’s lame argument about scientific tools:

http://socialdemocracy21stcentury.blogspot.com/2013/09/hoppe-on-euclidean-geometry-part-2.html

• Major_Freedom says:

“Your using “validity” in a sense different from me, and are committing nothing but classic fallacy of equivocation.”

It is not an “equivocation” fallacy on the part of A when B believes A is using a different definition than B.

I am using ths term the way it is commonly understood.

• marris says:

LK, I’m not sure what you mean by “epistemologically necessary,” or why you chose to insert those words into the sentence. Are you asserting that Euclidean geometry provides *no knowledge* of the real world? Because that’s obviously false. If I draw a triangle for you and let you empirically measure the first two angles A and B, is it some complete accident that your empirical measurements of the third angle will be around (180 – A – B)? Or is it that deductive Euclidean geometry is *closely applicable to* (or at least a good approximation for (our world)? Of course, there is nothing in geometry which says that the physical universe is Euclidean (or Einsteinian, etc), but saying that it says nothing about the world is bonkers.

• Bala says:

“but saying that it says nothing about the world is bonkers”

Bonkers is par for the course when you are dealing with LK.

• Lord Keynes says:

Euclidean geometry provides knowledge of real space in the sense that it is an approximation of it but only in a restricted domain.

We know it is useful in its proper limited domain a posteriori, not a priori.

You must distinguish (1) pure geometry (which analytic a priori) from (2) applied geometry ( which is synthetic a posteriori)

• Major_Freedom says:

“Euclidean geometry provides knowledge of real space in the sense that it is an approximation of it but only in a restricted domain.”

That is all Kant ever meant for it to be. Restricted to objects of experience.

This is precisely why he introduced the concept of the “noumenal” world. This refers to that which is beyond experience.

But your claiming, falsely, that Kant’s a priori categories of space and time apply to that world, when he specifically argued that they only apply to objects of experience.

• Economic Freedom says:

>>>When asserted of the real world, it becomes synthetic a posteriori and contingently true.

To take simple arithmetic as an example:

If we say “2 + 2 = 4”, that’s true without appeals to observation or experience (i.e., it’s aprior); and it’s true by necessity (i.e., denying the conclusions entails a contradition of one of the premises).

But according to you, if we say “2 apples + 2 apples = 4 apples”, that’s true only because we observe 4 apples today and because we also observed 4 apples yesterday, so we build up a kind of database of experiences on which to say that the statement is true (i.e., it’s a posteriori); and it’s true contingently (i.e., we can deny “4 apples” as a conclusion without entailing a contradiction).

• Lord Keynes says:

Even arithmetic must be separated into (1) pure arithmetic (which analytic a priori) and (2) applied arithmetic( which is synthetic a posteriori).

You cannot comprehend the argument being offered because you do not understand this difference.

If I have 2 real drops of water right now on my table and add 2 drops of water, how many drops are there?

Can you tell now a priori how many drops of water there are?

• Hoodnick says:

Then if sales is \$10 dollars and cost is \$5 dollars how do will fine out what profit is. Do we need to empirically verify what profit is or do we know a priori what it is. Does 10-5=5 analytic does 10-5 contain 5 or is it synthetic 10-5 does not contain 5. .

• Economic Freedom says:

>>>If I have 2 real drops of water right now on my table and add 2 drops of water, how many drops are there?

Fantastically insightful argument! So “one minute” is NOT “60 seconds” of time, each second added to the next! Because the seconds (a small unit) “coalesce” into a minute (a larger unit). I get it. I really do.

Try applying a little marginalist thinking by considering units. If the original unit was “the drop of volume X” then whether or not everything coalesces and becomes continuous, or retains its boundaries and becomes contiguous, is utterly irrelevant. YOU HAVE 4X DROPS OF VOLUME X. So 2 drops-of-volume-X plus 2 drops-of-volume-X equal 4 drops-of-volume-X. That 4 drops-volume-X might be defined as 1 drop-of-volume-Y is IRRELEVANT.

Duh. That’s why recipe books the world over say, “This cake requires 4 cups of sugar.” Guess what, genius? When you measure out “4 cups” of sugar, it looks to the eye as if it’s just one big amount of sugar, right? And yet . . . it’s really 4 units of sugar coalesced into one larger unit.

That should have taken you no more than 3 seconds of thought (coalesced into one unit of time or not; take your pick). But you were too busy impressing yourself, and too concerned with impressing others. To the extent you succeeded at the former is the extent to which you failed at the latter.

That you resort to grade-school rhetorical tricks regarding water drops to try to win an unwinnable argument shows rather clearly that you “cannot comprehend the argument being offered because you do not understand the difference.”

• Major_Freedom says:

“Euclidean geometry provides no epistemologically necessary knowledge of the real world,”

Sure it does. Every civil engineer builds real world objects that withstand the elements, precisely because Euclid is relevant to their task.

You’re wrong.

• Keshav Srinivasan says:

Well that’s a contingent fact about the world, isn’t it? What if humans lived at a length scale where the curvature of space was noticeable in daily life?

• RIchard Moss says:

Then they use the relationships established by non Euclidean geometry, relationships that are established through deductive reasoning and that do not require ‘testing’ to make sure they are ‘true’.

• Keshav Srinivasan says:

Yes, but isn’t the question of which system of geometry is most practical for civil engineering dependent on empirical testing?

• Ken B says:

They get to choose which completely different “necessary” truths to accept?

• Lord Keynes says:

“Sure it does. Every civil engineer builds real world objects that withstand the elements, precisely because Euclid is relevant to their task.”

Euclidean geometry provides knowledge of real space in the sense that it is an approximation of it but only in a restricted domain.

We know it is useful in its proper limited domain a posteriori, not a priori.

You — like the person above — cannot properly distinguish (1) pure geometry (which analytic a priori and necessarily true) from (2) applied geometry ( which is synthetic a posteriori and if true, only contingently true).

Can you know a priori that Euclidean geometry is a necessarily true universal theory of real space throughout the universe?

I like to see MF overturn about a hundred years of modern physics and general relativity.

• Richard Moss says:

I can make the distinction. That it ‘s premises are ‘contigent’ doesn’t means its conclusions don’t apply to the real world.

• Major_Freedom says:

“Euclidean geometry provides knowledge of real space in the sense that it is an approximation of it but only in a restricted domain.”

Necessarily true in the restricted domain of action, which is what Mises (and Kant to a degree) only intended to show.

“We know it is useful in its proper limited domain a posteriori, not a priori.”

No, it’s a priori, because you presuppose its validity when you conduct empirical experiments and make a posteriori judgments.

“You — like the person above — cannot properly distinguish (1) pure geometry (which analytic a priori and necessarily true) from (2) applied geometry ( which is synthetic a posteriori and if true, only contingently true).”

“Can you know a priori that Euclidean geometry is a necessarily true universal theory of real space throughout the universe?”

I never intended to claim that. I only intended to claim it is necessarily true about the universe constrained to action.

“I like to see MF overturn about a hundred years of modern physics and general relativity.”

I’d like to see you stop using Euclidean geometry to allegedly disprove Euclidean geometry.

• Lord Keynes says:

“Necessarily true in the restricted domain of action, which is what Mises (and Kant to a degree) only intended to show.”

Rubbish. Non-Euclidean geometry was developed only 20 years after Kant’s death, and only proved empirically to be the right theory of space in the 20th century.

Kant thought Euclidean geometry was a necessarily true theory of all space known a priori (Paul Guyer, The Cambridge Companion to Kant and Modern Philosophy, pp. 88-89), not “restricted [to a] domain of action”.

As for Mises he repeatedly says that Euclidean geometry provides real knowledge of external reality, not just some “restricted domain of action” or the internal sense data of humans.

• Major_Freedom says:

“Rubbish. Non-Euclidean geometry was developed only 20 years after Kant’s death, and only proved empirically to be the right theory of space in the 20th century.”

It was discovered using Euclid. This is crucial.

“Kant thought Euclidean geometry was a necessarily true theory of all space known a priori (Paul Guyer, The Cambridge Companion to Kant and Modern Philosophy, pp. 88-89), not “restricted [to a] domain of action”.”

No, he thought ALL his a priori categories are constrained to objects OF EXPERIENCE. You dont’ know Kant.

“As for Mises he repeatedly says that Euclidean geometry provides real knowledge of external reality, not just some “restricted domain of action” or the internal sense data of humans.”

It does, when you realize he made this argument constrained to ACTION.

• Major_Freedom says:

“We cannot think an object save through categories; we cannot know an object so thought save through intuitions corresponding to these concepts. Now all our intuitions are sensible; and this knowledge, in so far as its object is given, is empirical. But empirical knowledge is experience.Consequently, there can be no a priori knowledge, except of objects of possible experience.

and

“But although this knowledge is limited to objects of experience, it is not therefore all derived from experience.” – The Critique of Pure Reason, Deduction of the Pure Concepts of the Understanding, Section 2, Transcendental Deduction of the Pure Concepts of the Understanding, §27 Outcome of this Deduction of the Concepts of Understanding

• Lord Keynes says:

(1) Space isn’t a category according to Kant: it is an a priori “form of cognition / sensibility / pure intuition.”

(2) in any case, the world of the “objects of our experience” is “empirically real” for Kant as well as “transcendentally ideal”.

Space and the statements about it of Euclidean geometry are synthetic a priori and necessarily true of the objects of experience.

That is false. The objects of our experience are all ultimately subject to non-Euclidean space.

Euclidean geometry is merely an approximation at a certain level.

• Major_Freedom says:

(1) I call them categories. I don’t think I’m obligated to use the exact same terminology as Kant for this particular point.

(2) in any case, the world of the “objects of our experience” is “empirically real” for Kant as well as “transcendentally ideal”.

Transcendental IDEALISM, to Kant, does not mean the propositions are necessarily true for objects NOT of our experience, that is, “through the entire universe”.

Transcendental IDEALISM is still constrained to the subject and the subject’s experience.

All transcendental idealism means is that our experience of things is how they appear to us, not how those things are “in and of themselves.”

“Space and the statements about it of Euclidean geometry are synthetic a priori and necessarily true of the objects of experience.”

“That is false. The objects of our experience are all ultimately subject to non-Euclidean space.”

The point is that you were wrong to claim that Kant believed his Euclidean concepts are relevant for the entire universe. You were wrong to claim that they are not constrained to objects of experience only.

Now you’re evading that and claiming Kant was wrong.

Those are two different things.

You won’t even admit that you got Kant’s argument wrong.

“Euclidean geometry is merely an approximation at a certain level.”

Then so is non-Euclidean geometry, because non-Euclidean geometry is grounded on Euclidean geometry.

If all you have is an approximate measurement of equipment and tools, then whatever you measure with those tools will be an approximation as well.

But then the question remains: If non-Euclidean geometry is but an approximation, it means that you can’t claim the universe IS or IS NOT non-Euclidean.

You’d be doing what I claimed all along you’d be doing: You’d be thinking of both Euclid and non-Euclid as you act. In other words, Euclid doesn’t have any meaning unless non-Euclid is real and valid, and non-Euclid doesn’t have any meaning unless Euclid is real and valid.

In other words still, and like I said above, the discovery of non-Euclid did not “refute” or “falsify” Euclidean a priori conceptions. It hasn’t destroyed the necessary thought that is Euclidean.

• Lord Keynes says:

If your position is now that Euclidean geometry is only synthetic a priori of sense data and the Kantian world of phenomena, then your position is even less tenable.

Now you are committed like Kant to saying that we can never have knowledge of the thing-in-itself, which means that you cannot defend the position that synthetic a priori provides any necessary knowledge of external reality.

• Major_Freedom says:

So you admit now that Kant did not hold Euclidean geometry to be “true everywhere in the entire universe”, but only true for objects of experience?

He was very clear on this.

• Major_Freedom says:

So you admit now that Kant did not hold Euclidean geometry to be “true everywhere in the entire universe”, but only true for objects of experience?

He was very clear on this.

• Major_Freedom says:

“Now you are committed like Kant to saying that we can never have knowledge of the thing-in-itself, which means that you cannot defend the position that synthetic a priori provides any necessary knowledge of external reality.”

Actually I am not. This is precisely why I am not a Kantian, but rather, a Misesian.

Mises discovered that action is the bridge that enables us to connect the necessary forms of cognition a la Kant, with external reality of “things in themselves”, such that synthetic a priori epistemology is possible.

A superficial reading of Kant can admittedly lead to the conclusion that the necessary forms of cognition do not imply any truth of external reality; that it is idealism par excellence.

• Lord Keynes says:

“Mises discovered that action is the bridge that enables us to connect the necessary forms of cognition a la Kant, with external reality of “things in themselves”, such that synthetic a priori epistemology is possible.”

Is it your belief that the statements of Euclidean geometry provide necessary and universal truth about the space of the external universe known a priori?

• Ken B says:

MF “because non-Euclidean geometry is grounded on Euclidean geometry.”

No. This is flat incorrect.

• Major_Freedom says:

LK:

“Is it your belief that the statements of Euclidean geometry provide necessary and universal truth about the space of the external universe known a priori?”

Once again, do you still believe that Kant held that the space and time pure intuitions are restricted to objects of experience? That he did not, contrary to your original claim, hold that these intutions are necessarily true “everywhere in the entire universe”, i.e. including the noumenal world?

——————-

Ken B:

Actually it’s flat correct. The tools used by cosmologists to measure spacetime, the relationship between those tools and the researchers, is Euclidean.

Non-Euclidean geometry is meaningless unless Euclidean geometry exists, and vice versa.

• Mike T says:

Hey, look on the bright side. At least when debating Kant, you guys can keep it relatively civil:

http://thelibertarianrepublic.com/ironic-russian-man-shot-debate-kants-critique-pure-reason/#.UjdsJaPzhtp

• guest says:

We know it is useful in its proper limited domain a posteriori, not a priori.

Once you’ve tested, empirically, that such is the case, do you need to continually do tests to see if it’s still true?

(Again, I’m not asking about geometry, per se, here, if that makes sense.)

• Major_Freedom says:

“Once you’ve tested, empirically, that such is the case, do you need to continually do tests to see if it’s still true?”

Yes, because of the a priori synthetic argument that we humans are not capable of definitively knowing anything about reality with apodictic certainty, that all our knowledge is at best hypothetical only, and that all of our knowledge can at any time be falsified by future experience.

Hahaha, oh the irony…

• Tel says:

Once you’ve tested, empirically, that such is the case, do you need to continually do tests to see if it’s still true?

Yes if you believe in empiricism, because it is at best a working presumption that the future will operate by the same rules as the past. However, this is not as much of an obstruction as it might seem. Suppose a bridge is built based on Euclidean geometry, together with a knowledge of material strength (based on testing) and gravity. Every day the bridge does not fall down, we have successfully tested that the physical manifestation of geometry has not significantly changed. The day it does fall down we study closely to discover why.

Thus, our entire technological edifice is constantly re-testing our existing presumptions. So far as we know neither the bridge, nor gravity, nor the steel can learn from the past and adapt itself. Thus, so far we have a lot of evidence that these structures will remain solid.

However, in economics, people can learn from the past and adapt, and because new ideas are always appearing we cannot say that measurements we took 100 years ago are still relevant today.

• guest says:

Yes if you believe in empiricism, because it is at best a working presumption that the future will operate by the same rules as the past.

Isn’t the belief that the same rules will produce the same results a reliance on a priorism?

• Major_Freedom says:

Yes, because it is an assumption upon which empiricism is based, whether empiricists consciously recognize it or not.

The concepts “falsification” and “confirmation” imply constant causality rules in nature.

A theory/hypothesis proposed in the past, tested in the present, and claimed as being confirmed or falsified in the future, is the empiricist utilizing the a priori proposition that laws that are true in the past, must be true in the present and in the future.

• Tel says:

Isn’t the belief that the same rules will produce the same results a reliance on a priorism?

No, as I said it is a working presumption. So far it has worked well, and it constantly gets re-tested, so we will know quite quickly if things start to change.

• guest says:

You are looking at bridges not fall and thinking that you’re testing geometry, but this is merely exemplifying it.

(It also assumes that the rules you use are the only ones that can result in a persistently standing bridge – looking at it won’t tell you one way or the other.)

No, testing the bridge without reliance on a priorism would require proving the relevant math involved all over again, from scratch.

• Tel says:

Geometry as an abstract mathematical construct only needs testing for self-consistency. It is an abstract concept, it will forever be what it is.

Geometry as a means to build bridges is another thing entirely. This is Applied Mathematics, and the application is the important part. Empiricism is about testing in the real world, taking measurements, making predictions.

Testing the bridge never requires any a priori. Just walk across from one side to the other, that’s the test done. Handy thing about empirical tests is they can be so efficient, especially when the question you are testing is whether anything changed from last time around.

• guest says:

Just walk across from one side to the other, that’s the test done.

So, if the bridge falls, that means that the laws of geometry have changed?

• Tel says:

So, if the bridge falls, that means that the laws of geometry have changed?

Well something would have changed, obviously.

In the abstract sense Pythagoras’ Theorem cannot change because it is a made-up thing… but the application of that theorem to bridge building might now require revision, or there might have simply been an assumption that was missed first time round, or might have just been rust and lack of maintenance.

Testing is the easy part, understanding what happened is hard, but in large parts of science and engineering, not many new things do happen. That’s an empirical discovery in itself, not many bridges do fall down… so we can feel confident we got something right.

In comparison, if you look at economics you have the equivalent of bridge disasters once every few years, and everyone scurries around trying to explain it, then saying, “still trust us, we know something.” Mind you it is unfair to completely blame the Economists when so much Politics is now involved.

5. scineram says:

The theorems of finite characteristic 3 fields are also implied by the axioms, you will still never count 0,1,2,0,1,2,0,1…

6. Daniil Gorbatenko says:

2DK,

For some reason I again can’t reply to you in the thread so here goes.

=The physical or natural analog is part of physical science/natural science. I understand them symmetrically. Physicists looking into the Big Bang are scientists. Paleontologists are scientists.=

I think Bob never said that economic history isn’t a science. The main point Austrians make about theory and history is that you can only do history through the lens of a theory.

• Daniel Kuehn says:

It may be that I am reading into Bob’s earlier post because he never came out and made a clear statement. Let me put it this way:

I think that to be scientific you must have an interplay of theory and empirical verification (either predicting the future or explaining the past). If you do not have both, it is not scientific. The latter without the former is simply impossible. The former without the latter is philosophy.

If he agrees with that in all its particulars, fantastic. I know he doesn’t think empirical work is worthless. The question is, does he think it can discipline theory. He’s given indications that he thinks it cannot, but I could be misreading him.

• Bala says:

Daniel,

You should replace “I think that to be scientific ….” with “I think that to be scientistic….”

• Economic Freedom says:

>>>The former without the latter is philosophy.

Why philosophy and not, e.g. mathematics?

>>>The question is, does he think it can discipline theory.

Invent an example in which empirical observation or experience could “discipline” a purely theoretical insight such as “On a Euclidean plane the angles of a triangle must always add up to 180-degrees.”

• Daniel Kuehn says:

– Sure, that’s fine too.
– Empirical evidence on the impact of the minimum wage should discipline what theory we use to explain its effect

• Ken B says:

“Why philosophy and not, e.g. mathematics?”

Perhaps mathematics is just what we call philosophy when the reasoning is rigorous enough. This isn’t that radical an idea. Goedel believed he was doing philosophy when he proved his theorems.

• Economic Freedom says:

>>>>Perhaps mathematics is just what we call philosophy when the reasoning is rigorous enough. This isn’t that radical an idea. Goedel believed he was doing philosophy when he proved his theorems.

Ah, yes, of course. Nice bit of rigorous reasoning there. Congratulations. Would you claim your last post was an instance of philosophy, mathematics, or something else?

• Ken B says:

You don’t have to like it EF, but the main reason Godel’s theorem is considered mathematics is because of its rigor and expression NOT because of its topic.

Your question to DK seemed to suggest you understood that “the former without the latter” could also decribe mathematics, as indeed it can, but I guess I misread you as understanding that and you were just snarking at DK. I’ll know better next time.

• Daniil Gorbatenko says:

Daniel, two considerations.

1) It depends on what you mean by an interplay. The interplay suggested by many Austrians, regardless of how they view the epistemological status of basic economic propositions and theories logically deduced from them, is that we look carefully at history, see there the referents of concepts like money, banks, credit expansion, etc. If we find them and the theory doesn’t postulate a necessary connection (which, I think, the ABCT doesn’t), we don’t look at the data to see if it confirms our theory (in this case the ABCT). We look at the data to see whether a relevant set of historical events contained an Austrian cycle and how much (very roughly, not in a measurable sense) of the observed changes the Austrian cycle (if it’s present) explains.

2) We can’t, however, do what you seem to imply we can do with economic history, i.e. make some kind of semi-precise mathematical description of what was going on. Not even because the system is very complicated but because there are no mathematical relations in it.

7. GeePonder says:

[Not a response to anyone, though this blog will make it look like it]

“If one accepts the terminology of logical positivism and especially also that of Popper, a theory or hypothesis is “unscientific” if in principle it cannot be refuted by experience. Consequently, all a priori theories, including mathematics and praxeology, are “unscientific.” This is merely a verbal quibble. No serious man wastes his time in discussing such a terminological question. Praxeology and economics will retain their paramount significance for human life and action however people may classify and describe them.”

Mises, Ludwig von (2010-12-08). The Ultimate Foundation of Economic Science (LvMI) (Kindle Locations 1134-1136). Ludwig von Mises Institute. Kindle Edition.

• Bala says:

I agree, more particularly with what you have said 🙂

8. Bala says:

Bob,

I am sure you would have this in mind, but I thought saying it can’t hurt. One issue still remains to be addressed = the claim that analytic propositions do not tell/inform us about the real world. LK keeps coming with that rather irksome notion and then claims “victory” every time. It would be good to see you destroy him on that.

In case you already have, do direct me to it (and advance apologies for saying it is not addressed).

There is also this additional (minor) point about Daniel and his insistence that only that which makes verifiable/falsifiable predictions is science. Addressing this positivist hijacking of the methodology of science is, IMO, an important issue.

• Bala says:

I mention the 2nd point because it appears that The Ministry of Truth has been working overtime.

• Keshav Srinivasan says:

Well, lots of people think mathematics is analytic a priori (although Kurt Godel and Steve Landsburg consider it synthetic a priori), and mathematics is a vital tool in our understanding of the world.

9. Keshav Srinivasan says:

Bob, do you agree with Mises that praxeology is analytic a priori? Or do you agree with the later Austrians like Hoppe that it’s synthetic a priori?

• Bob Murphy says:

Keshav,

I’m not sure at this point that Mises did think praxeology is analytic. He definitely thought geometry was, and so I had just assumed he thought praxeology was too. But on this very thread several quotations from Ultimate Foundations have been brought to my attention, that suggest he either (a) thought praxeology was synthetic or (b) just cared that it was a priori, and didn’t care whether we called it analytic versus synthetic.

In any event, I think it’s synthetic, but only because I learned the term to mean “tells us more about reality than what we originally knew from the definitions / axioms.” But by that criterion, I think geometry is synthetic too. So if I’m disagreeing with Mises, it’s not because of how we view praxeology, it’s because of how we’re defining analytic vs. synthetic.

• Keshav Srinivasan says:

Bob, if you think geometry gives you more than what the axioms, how would you rebut Mises’ argument to the contrary, I.e. that propositions like the Pythagorean theorem are already contained in the axioms?

• Bob Murphy says:

Keshav, how would you rebut Mises’ claim that the Pythagorean theorem enlarges our knowledge of the world? At this point we’re quibbling over semantics. Mises and I both agree that to deduce economic laws, everything you “need” was right there in the action axiom. (Some results rely on auxiliary assumptions.) You don’t need to go observe anything to refine your understanding of reality. And yet we also agree that economists have done a lot of important work to uncover just what was “in” the action axiom all along.

So I triply agree with Mises that the important thing is to understand that praxeology is a priori, but whether you call the above “analytic” or “synthetic” is not as big a deal. There is more to praxeology than just saying “a bachelor isn’t married.”

• Lord Keynes says:

“Mises and I both agree that to deduce economic laws, everything you “need” was right there in the action axiom. (Some results rely on auxiliary assumptions.) “

That is logically impossible.

Once you require auxiliary assumptions, everything you “need” is clearly not in the action axiom.

You require a vast set of synthetic and empirical assumptions and premises to make your deductions work.

• guest says:

Once you require auxiliary assumptions, everything you “need” is clearly not in the action axiom.

You’re not citing an empirical test to butress this claim because it is of an a priori nature.

… Wait just a minute:

That is logically impossible.

THIS claim didn’t require an empirical test, and is of an a priori nature.

• Lord Keynes says:

“THIS claim didn’t require an empirical test, and is of an a priori nature.”

Correct: it’s analytic a priori.

You ignorant indeed if you think empiricists do not accept the meaningful nature of analytic a priori logical and mathematical statements.

• guest says:

You ignorant indeed if you think empiricists do not accept the meaningful nature of analytic a priori logical and mathematical statements.

But that’s our whole point; That’s all we’re saying.

We’re saying the Action Axiom IMPLIES certain things; And, up to certain points, Austrian Economics requires no empirical tests. But those points are few, and beyond those points there are no more empirical tests required:

In Defense of “Extreme Apriorism”
[WWW]http://mises.org/daily/5195/

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.

Setting aside the fundamental axiom for a moment, the empirical postulates are: (a) small in number, and (b) so broadly based as to be hardly “empirical” in the empiricist sense of the term.

What are these propositions? We may consider them in decreasing order of their generality: (1) the most fundamental — variety of resources, both natural and human; from this follows directly the division of labor, the market, etc.; (2) less important, that leisure is a consumer good. These are actually the only postulates needed. Two other postulates simply introduce limiting subdivisions into the analysis. Thus, economics can deductively elaborate from the fundamental axiom and postulates 1 and 2 (actually, only postulate 1 is necessary) an analysis of Crusoe economics, of barter, and of a monetary economy. All these elaborated laws are absolutely true.

When we analyze the economics of indirect exchange, therefore, we make the simple and obvious limiting condition (postulate 3) that indirect exchanges are being made.

The fourth — and by far the least fundamental — postulate for a theory of the market is the one which Professors Hutchison and Machlup consider crucial — that firms always aim at maximization of their money profits. As will become clearer when I treat the fundamental axiom below, this assumption is by no means a necessary part of economic theory. From our axiom is derived this absolute truth: that every firm aims always at maximizing its psychic profit. This may or may not involve maximizing its money profit. Often it may not, and no praxeologist would deny this fact.

Now the crucial question arises: How have we obtained the truth of this axiom? Is our knowledge a priori or empirical, “synthetic” or “analytic”? In a sense, such questions are a waste of time, because the all-important fact is that the axiom is self-evidently true, self-evident to a far greater and broader extent than the other postulates. For this axiom is true for all human beings, everywhere, at any time, and could not even conceivably be violated. In short, we may conceive of a world where resources are not varied, but not of one where human beings exist but do not act. We have seen that the other postulates, while “empirical,” are so obvious and acceptable that they can hardly be called “falsifiable” in the usual empiricist sense. How much more is this true of the axiom, which is not even conceivably falsifiable!

• Lord Keynes says:

“But by that criterion, I think geometry is synthetic too. So if I’m disagreeing with Mises, it’s not because of how we view praxeology, it’s because of how we’re defining analytic vs. synthetic.”

I presume you can see that this implies that Mises — if he really did think Euclidean geometry was analytic — was confused about the definition of “analytic” — precisely as Hans Albert (Albert, H. 1999. Between Social Science, Religion and Politics: Essays in Critical Rationalism. Rodopi, Amsterdam. pp. 131–132) has argued.

10. joe says:

How does this Mises quote relate to presuppositionalism and Christian Reconstruction? Is it fair to say that Austrian Economics is based on a Biblical Axiom?

• guest says:

No, not as such.

But this may represent what it is you’re getting at:

Science: Philosophy’s Handmaiden
[WWW]http://www.str.org/articles/science-philosophy-s-handmaiden

I don’t think it’s unreasonable for science to use empirical evidence. I don’t think it’s unreasonable for it to operate within the framework of scientific law. What I think is unreasonable, is the claim that you can’t trust knowledge unless it is demonstrated to be true by empirical fact. That is an underlying supposition that’s throughout the entire letter. My point is that certain things must be in place that are not scientific and they must be true before you can even begin to practice science. In the piece Sagan and Scientism, I mentioned that you can’t even prove math by science.

Both Austrian Economics and Religion have to contend with a view of science that has simply defined non-empirical knowledge out of existence (while somehow ignoring that the Scientific Method, itself, is philosophy and not science – otherwise science would be circular reasoning):

Billions and Billions of Demons by Richard Lewontin
[WWW]http://www.drjbloom.com/Public%20files/Lewontin_Review.htm

Our willingness to accept scientific claims that are against common sense is the key to an understanding of the real struggle between science and the supernatural. We take the side of science in spite of the patent absurdity of some of its constructs, in spite of its failure to fulfill many of its extravagant promises of health and life, in spite of the tolerance of the scientific community for unsubstantiated just-so stories, because we have a prior commitment, a commitment to materialism. It is not that the methods and institutions of science somehow compel us to accept a material explanation of the phenomenal world, but, on the contrary, that we are forced by our a priori adherence to material causes to create an apparatus of investigation and a set of concepts that produce material explanations, no matter how counter-intuitive, no matter how mystifying to the uninitiated. Moreover, that materialism is absolute, for we cannot allow a Divine Foot in the door. The eminent Kant scholar Lewis Beck used to say that anyone who could believe in God could believe in anything. To appeal to an omnipotent deity is to allow that at any moment the regularities of nature may be ruptured, that miracles may happen.

11. GeePonder says:

Start over with more space. Bob, you gave in too soon. Mises was merely using jiujitsu as you stated at first. He noted that he didn’t believe the statement was true, but that was simply an interjection. He was pointing out an objection to the statement: only experience can lead to synthetic propositions by stating that that statement is in itself a synthetic a priori statement.
———————————-
The essence of logical positivism is to deny the cognitive value of a priori knowledge by pointing out that all a priori propositions are merely analytic. They do not provide new information, but are merely verbal or tautological, asserting what has already been implied in the definitions and premises. Only experience can lead to synthetic propositions. There is an obvious objection against this doctrine, viz., that this proposition that there are no synthetic a priori propositions is in itself a—as the present writer thinks, false—synthetic a priori proposition, for it can manifestly not be established by experience.

Mises, Luwig von (2010-12-08). The Ultimate Foundation of Economic Science (LvMI) (Kindle Locations 183-187). Ludwig von Mises Institute. Kindle Edition.

• GeePonder says:

He was merely pointing out an objection. Not arguing the opposite.

• Bob Murphy says:

GeePonder hang on a second. Look at Mises statement:

There is an obvious objection against this doctrine, viz., that this proposition that there are no synthetic a priori propositions is in itself a — as the present writer thinks, false — synthetic a priori proposition, for it can manifestly not be established by experience.

So the relevant proposition isn’t, “Only experience can produce synthetic propositions.” Rather, the following is the proposition that Mises claims is false:

P: “There are no synthetic a priori propositions.”

So if P is false, what can we conclude?

~P: “There exists at least one synthetic a priori proposition.”

Now, the really subtle thing is, maybe Mises still believes that there are no TRUE synthetic a priori propositions. I.e. we have already established that there are FALSE synthetic a priori propositions, namely P above.

That’s part of the problem here, is that some writers like Hoppe are very precise and put in “true” when talking about their belief in synthetic a priori propositions, but in this context with Mises, I’m not sure if he means to put in “true” and is omitting it for brevity, or if he really is just being a kung fu master and walking the knife edge with the logical positivists.

• GeePonder says:

Yes, thanks Bob for the reply and clearing that up for me.

I am now curious whether ~P is synthetic a priori: There exists at least one (true) synthetic a priori proposition. And if so, does it verify itself? LOL a little.

12. GeePonder says:

True, he side-stepped the issue, as he points out later in UFofES. He was not interested in the answer.
————————
Praxeology is a priori. All its theorems are products of deductive reasoning that starts from the category of action. The questions whether the judgments of praxeology are to be called analytic or synthetic and whether or not its procedure is to be qualified as “merely” tautological are of verbal interest only.

Mises, Luwig von (2010-12-08). The Ultimate Foundation of Economic Science (LvMI) (Kindle Locations 753-756). Ludwig von Mises Institute. Kindle Edition.

13. hanz says:

So, praxeology doesn’t produce knowlegde. The results of praxeology are nothing more than tautologies, e.g. analytic propositions.

But its concepts are *means* to acquire knowledge. It’s like an instrument to look at the world data, to make them meaningful at all. Without analytic propositions/concepts you were blind. That’s Kant’s famous “Anschauungen ohne Begriffe sind blind.”

that’s at least my approach to praxeology.

14. Keshav Srinivasan says:

Bob, let me ask you this: do you believe that if praxeological reasoning was specified rigorously enough, with numbered axioms, step-by-step proofs, etc., then a computer program could mindlessly churn out the theorems of praxeology? Or do you think that there is some quality unique to humans, like insight or intuition, that is necessary for praxeological reasoning.

The reason I ask is that one of Kant’s main arguments for geometry being synthetic is that Euclid’s proofs sometimes relied on the reader’s visual intuition concerning geometric figures. But years later, Tarski and Hilbert created more rigorous versions Euclid’s system, eliminating such appeals to intuitions and diagrams.

• Bob Murphy says:

Keshav that is a fascinating issue. If I had to guess I would say no, the computer couldn’t do praxeology, because some of the “obvious” deductive moves involve things that you couldn’t specify in advance. But I would probably want to study things in geometry first to see why Kant’s intuition failed there.

• Major_Freedom says:

Where is the proof that Kant’s geometry failed?

• Keshav Srinivasan says:

Bob, how can you call it logical deduction if you can’t specify the rules of logic you’re allowed to use in advance? Or do you think that there is something in praxeological reasoning that goes beyond logical deduction, like intuition? Steve Landsburg, for instance, thinks that our knowledge of numbers comes from extra-sensory intuition rather than just deduction.

As far as Kant goes, Euclid’s Elements was considered the epitome of rigor in the ancient world. So when Kant found leaps of logic in Euclid’s reasoning, he attributed them not to any failings on Euclid’s part, because who could accuse Euclid of being unrigorous? Instead, he concluded that if Euclid couldn’t make geometric proofs a purely logical endeavor, then it can’t be done: there must be some essential intuitive element in geometrical reasoning.

But it turned out that while Euclid was very rigorous, he wasn’t as rigorous as he could have been. Consider his very first proposition:
http://aleph0.clarku.edu/~djoyce/java/elements/bookI/propI1.html
He proves the existence of an equilateral triangle, by taking the circle with center A passing through B, and the circle with center B passing through A, and marks their intersection as C. But he failed to show that point C exists, I.e. that the two circles intersect in the first place. Kant interpreted this as evidence that you couldn’t prove the theorem unless you used your visual intuition, but then later Tarski, Hilbert, Birkoff, and others made much more rigorous versions of Euclid’s system, to the point where computer programs can now easily prove all of Euclid’s theorems, without any use of visual intuition.

• guest says:

… to the point where computer programs can now easily prove all of Euclid’s theorems, without any use of visual intuition.

Let’s remember that the AND, OR, NOR, XOR, et al, circuits are only representations of laws of logic, and are only meaningful because we are the ones who poured the meaning into them to begin with.

And the 1’s and 0’s are meanings we assign to the states of some kind of medium.

So, just because a computer is doing it, doesn’t mean the work is empirical in nature.

• Keshav Srinivasan says:

I think you misunderstood what I was saying. I was discussing Kant’s argument that geometry is synthetic a priori as opposed to analytic a priori. I wasn’t trying to use the fact that computers can do it to argue that it’s empirical.

• Major_Freedom says:

Keshav,

Praxeology allows only one logic, so Murphy likely believed it wasn’t necessary to specify the rules.

• Keshav Srinivasan says:

Then what did Bob mean by “the computer couldn’t do praxeology, because some of the “obvious” deductive moves involve things that you couldn’t specify in advance.”?

• Tel says:

Suppose I come up with an encoding (e.g. ASCII) that converts a number into a block of text. So since I can easily write an algorithm to find the next number (by adding 1) I can therefore write every possible block of text (including all of those never yet written). With suitable encodings I could cover every photograph, every piece of music, and also every other human language (e.g. Unicode).

So now, since I have described the enumeration of all possible creative works, I hereby claim Copyright on all the output of humanity, forever.

• guest says:

You bastard.

There: I believe that output is yours.
😀

• Major_Freedom says:

Do you work for Monsanto?