Hoppe Praxeology The Austrian Method Seminar Austrian Economics (2005) Lecture 6 of 11 (extended)

Um I I want to talk about um the Austrian method um because more than anything else it is um the method that distinguishes Austrian economics from um what we call mainstream um mainstream economics um and um uh in order to understand somehow uh the results that we come up with, it is also necessary to see how we come up with um with these results. Um the question that I want to address is um is economics um a science that we might call an axiomatic deductive science that starts with uh firm starting points and then deduces certain conclusions from these starting points and provided that uh there are no errors made in the process of deduction uh arrives then at conclusions that we can call apodictically true uh apodictic being a term that Mises um uh uses frequently um or is uh um economics uh a science that is called an empirical uh science um that uh uh gives us hypothetical propositions that require continual testing in order to find out whether they are true or not or to phrase the whole thing in a slightly different way, is economics a science more akin to applied logic? Um Logic being a discipline where we obviously do not test our propositions but just grasp them and um would need other logical arguments in order to find out whether we made errors in our logical reasoning or is economics more a science akin to uh physics, for instance? Um and I want to show that the answer to this question whether it is more akin to logic or more akin to uh physics has uh highly important uh implications. Um If we look at the uh the present situation in uh in economics, then we can say that uh except for the Austrians, um all other economic schools believe that economics is a science of the type that we have also uh in physics. Only the Austrians believe that economics is a science um akin to uh applied um applied logic. When I say that uh all others but economists think that economic propositions are hypotheses that need to be tested, I'm not actually saying that all uh mainstream economists actually practice what they preach. Um They preach in their books that it is economic propositions are hypothesis that need to be tested, but frequently, especially among the better economist, the better mainstream economist, in their actual practice, they frequently contradict their own methodological prescriptions. It is just the minor figure, so to speak, in mainstream economics that take their own methodological prescriptions very very seriously. Now, saying that the Austrians with their view that economics is more akin to logic are a minority, I should add that this is a relatively new phenomenon. Until about the 1950s, um the view that the Austrians now hold, um um alone, used to be the standard view that was shared by a great majority of of the economics profession. For many, many years, the most important book on economic methodology was a book written by um Lionel Robbins, um the nature and significance of economic science. It was initially published in the early 1930s, and Lionel Robbins, who was the chairman of the economics department at the London School of Economics, and was also responsible for recruiting Hayek to come from Vienna to London, Lionel Robbins had uh, as a young man frequently visited Vienna and participated in in the famous Mises seminar that he held in Vienna and was very much influenced by the methodological views that that Mises propounded. You can still this book Nature and Significance of Economic Science is still in print. Um, and if you look at the index for instance, you will find that there's no other person mentioned more frequently in the name index than than the name of >> [snorts] >> Mises. And if you go back into the 19th century, you will find that um actually most economists in the 19th century shared the view that nowadays is only propounded by the Austrians that economics is some sort of a sophisticated common sense logical type discipline. Um, you can find that that Jean Baptiste Say for instance or Nassau Senior or John Keynes. All of these 19th century authors held views similar to what nowadays only the Austrians believe. Um Now this new modern view about the nature of economics as an empirical discipline that formulates hypotheses that require testing. Um, this new view um is due essentially to the influence of um a philosophical school of thinking um that is called the positivists or uh the logical um empiricists. Um logical empiricists indicating that they uh considered themselves to be some sort of the modern um development of uh the traditional empiricist philosophers such as uh David Hume and uh and John Locke. Um and uh this school the logical empiricists empiricists were also called positivists uh originated uh also in Vienna at the time when Mises um was uh active in uh in Vienna. It is frequently even called the Vienna school um not at the time when it originated. At that time uh it was called the Schlick the Schlick Kreis the uh or the Schlick circle um named after a philosopher who taught at that time at the University of Vienna Moritz Moritz Schlick. Um and this circle was a circle that competed so to speak with the Mises the Mises circle. Um actually had a smaller number of uh um followers than the Mises circle uh itself had um and um to give you just some uh some names that were participants in this circle, either direct participants or on the outskirts of the circle, besides Moritz Schlick, Rudolf Carnap, Hans Reichenbach, Gustav Hempel, Otto Neurath, and also on the on the fringes of the circle, Karl Karl Popper. Another also on the fringes of the circle was Ludwig von Mises' brother, Richard von Mises, who was a famous mathematician, the founder of the frequency interpretation of of probability theory, a a man who ended his career in the United States at at the University of of Harvard. And is this the influence of [snorts] the Vienna Circle was rather moderate in in Germany, but similar to the fate that affected most of the members of Mises' own circle, most of these Vienna School philosophers emigrated to the United States in in the 1930s, many of these people were Jewish, and it was in particular during their during the time in that they spent in the United States that their philosophy became, so to speak, the dominant academic philosophy. This is no longer true today. The high point of the influence of the positivist philosophers is long over, but within outside outside philosophy itself, within various disciplines such as sociology and economics and so forth, their views are still the dominant views about the proper scientific procedure. Um and I want to explain to you first what their view, the view of the positivist, is about the nature of science and the proper scientific procedure, and I want to take the most popular version of this positivist philosophy, namely the version that has been formulated by Karl Popper. Karl Popper does not refer to himself as a positivist because there were small differences in his philosophy as compared with of other major proponents of the school. Karl Popper referred to his version of this philosophy as critical rationalism. That seems to be somewhat of a misnomer because he is clearly not a rationalist in the traditional sense. What has happened here is something similar to what has happened to the word liberalism. As you all know, liberalism in the European sense means somebody who is in favor of a relatively free markets and little government interference. And in the United States, of course, liberalism has assumed the meaning of being a social democrat and something similar is true actually about this term critical rationalism. What Popper's critical rationalism amounts to is pretty much the opposite of that what traditional rationalists proposed. So he tries to fool the public in the same way as the public is being fooled by Americans using the term liberalism. But in any case, so what is Popper's view about the appropriate procedure in in the sciences? His view is that there exist only two types of legitimate scientific propositions. The first type of legitimate scientific propositions are what he calls empirical propositions. Empirical propositions contain knowledge about the real world. And empirical propositions are characterized by the fact that they must, according to Popper, at least in principle, be falsifiable. That is, experience is necessary in order to show whether this statement can be accepted temporarily or whether we have to reject a certain a certain proposition. And importantly, Popper's view and the view of all positivists is to say that there is nothing about the real world, there's nothing about real phenomena that we can know with certainty, that we can know apodictically. All knowledge about the real world is hypothetical knowledge. And the second type of legitimate scientific propositions, according to Popper, are so-called analytical propositions. Analytical propositions are true by definition. Um They are statements like bachelor means unmarried man. Um Analytical propositions, according to the positivist and according to Popper, do not tell us anything about the real world. They only tell us something about how we use certain words. Um They are tautological information, just definition um definitional type of uh um information. And they Popper, for instance, um that's a difference to with respect to the old um empiricists, um they believe that logic, for instance, only produces analytical propositions, mathematics only proposes uh presents analytical uh knowledge and so forth. The traditional view of logic and mathematics was that they say something about the basic structure of reality. That is that they have some empirical content. The logical positivist would deny this. According to them, mathematics and logic say nothing about the real world. The Aristotelian view for instance was that logic of course does tell us something about the basic basic structure of of reality. And in addition, I should emphasize as an implication of this view, there are only two legitimate types of proposition, namely empirical testable and analytical non-testable but not saying anything about the real world. As a further implication of that is also their conviction that normative propositions, that is you should do or this is right or this is wrong, that normative propositions have no cognitive content at all. Normative propositions are just more like uh fooey uh yuck. Uh that is expressions of emotions and and feelings, but uh the question is that true or false uh is something that does not even appear uh when we talk about propositions of this of this kind. Um and in addition to this distinction that they make with respect to analytical and empirical propositions their view then of what the objective of sciences is is the following. Uh science consists in nothing else but um giving us explanations, and explanations have the same structure as predictions. And all explanations and predictions uh always have the same structure of if-then statements. Um, if such and such is the case, then such and such um will will follow. Um, all explanations have this this if-then uh structure, which is structurally the same, so to speak, as um as the structure of um of predictions. And what we do then with these explanations is um we test these um explanations. So, we have, let's say, if A uh then then B. Um, so we observe does A occur? And then we find out did B indeed follow as we hypothesis hypothesized? Uh or does B not follow? Um, now if B does follow, then we speak of in Popper's word of a confirmation uh of our hypothesis. But a confirmation, this is important to recognize, a confirmation of a hypothesis is not the same thing as a verification. That is, a confirmation does not show us that this is really true. Uh it only tells us that so far it has not been found to be false. Um, for the following reason, um A and the term B are universals. They apply to an infinite number of instances of A, and they apply to an infinite number of instances of the phenomenon B. Um so, even if you find in one instance that A was indeed followed by B, this does not show that for all instances of A, uh B will indeed follow. That's why we only can say confirmed, meaning so far we have not found it to be false. The other result would be, okay, we observe A to occur, but B does not um follow A as we hypothesis hypothesis hypothesized. In this case, we would say, uh this is a falsification. Okay. Now, again, it is important to realize what falsification means and what it doesn't mean. Um falsification in Popper's view also does not mean that we have now uh shown once and for all that there exists no relationship between A and B as it was hypothesized. It only shows that the relationship between A and B cannot be as simple as it was uh formulated in the initial hypothesis. It might well be that if we control for some other variable, Z or so, uh that we then do find that A was in fact followed or is in fact followed by um by B. Um uh So, we might well maintain uh our hypothesis or the core of our hypothesis that A and B are related except they cannot be quite as simply related as the initial hypothesis stated. But if we control for something else, then we might still find out indeed they are in some way related as we initially hypothesized. Um, so neither a confirmation is so to speak a final result that we have reached. We cannot stop with that. Nor is a falsification in any way a final stopping stopping point. It might just indicate that we have to reformulate our our hypothesis. Um, I should also say as another tenant of the positivist philosophy, >> [snorts] >> they believe that there exist only two types of definitions. Um, there exist what they call um, ostensive definitions. That is we point to something and say this is green or this is a tree or this is a boat. Uh, that is pointing to something and explaining what the meaning of a certain word is by pointing to it. And the second type of legitimate definition are what they call stipulative definitions. That is just defining again one word by by another. Um, again would be men in English means the same thing as man in German. Um Now, before I come to a refutation of this um positivist view uh about the nature of science, I want to explain to you uh the relativistic implications um of the whole um uh positivistic program. First, again let me remind you of the fact that um that normative propositions uh are not cognitive propositions at all. So, they take some sort of relativistic position when it comes to what is right, what is good, what is wrong. Those are things alleged about which we can make allegedly no judgment um what's whatsoever. Um um from which it follows that the positivists adopt some sort of program like uh social uh social engineering. Karl Popper actually uses the term uh piecemeal um social engineering. Um >> [snorts] >> Now um I want to um I want to look at some elementary economic propositions and um and raise the question whether these um elementary economic propositions are in fact hypotheses as they must be according to um to the positivists, according to Popper. Um take for instance a statement like like this. Uh in every voluntary exchange betw- between two individuals uh both exchange partners must at the outset of the exchange expect to benefit from the exchange, otherwise they would not go through with the exchange. Uh in every voluntary exchange [clears throat] the goods or services that are being exchanged must have unequal value in the eyes of the two exchange partners, that is to say, whatever I give away must have in my view less lesser value than whatever I acquire in turn, which must have in my view a higher value than what I give away, and uh the two exchange partners must have opposite preference orders. Whatever I consider as having a higher value must be considered uh as something by the other exchange partner as having a comparatively lower value. Now, is this a hypothesis? Uh how would we even go about testing this? Is there any way that we could ever find that this might be false? How would uh an experience look like that would show us that this is false? >> [snorts] >> Uh Now you immediately realize that this is somewhat difficult to imagine how this should be a hypothesis and not something that once we hear it formulated recognized as being necessarily true and saying something about reality as well. Namely, saying something about the phenomenon of voluntary exchanges. Um, we'll take another example. Um, if we raise the minimum wage per hour to $1 million tomorrow, um, tomorrow, uh, and enforce this to the hilt, um, uh, then, uh, mass unemployment will result. Now, is this a hypothesis that requires any type of testing? Or does that not simply follow from the fact that $1 million is significantly more than what is currently paid for wages, and that by outlawing certain exchanges between people, uh, from being conducted that are below $1 million, there will be plenty of people, employers, who will not employ other people, and employees who will not get any employment. Is that a hypothesis or is that something that we can know ahead of ever trying this sort of thing out? Um, or take this example. Um, if we, uh, quadruple the supply of money from one day to the next, um, and the consumer goods and producer goods in existence remain what they were before, uh, that there will be a drastic increase in the level of prices. Um is that a hypo- hypothesis that requires any type of testing, or is that something that follows simply um from the way that we describe the situation? Uh that follows simply from what we know about money, what we know about consumer goods, what we know about producer goods, and so forth. Um one last example. Um under socialism, there exists no private ownership in factors of production. All factors of production are owned by one entity, namely the state. If there is no private property in factors of production, then there exist no prices for factors of production. And if there exist no prices for factors of production, then not then we cannot engage in cost accounting. That is, we cannot compare the input prices with the output prices that we get for our product, and accordingly we cannot determine whether we used our input factors efficiently or inefficiently, whether we produced wastefully or not wastefully. Again, is that something that is only hypothetically true, or doesn't that follow automatically, so to speak, simply from the from the terms that that we um that we use, um and don't we know the result of this without trying it out, or do we have to try it out? Um now, I take it for granted that when you hear these examples, that the first impression that you have is all these statements say something about real phenomena. They say something about the world as it is. But, on the other hand, they are not falsifiable by experience. We can recognize them as necessarily true. Um, as something that simply follows from the way that we described the situation. And this, as I indicated early on, this of course was a view that economist until the 1950s held. Almost every economist thought, "This is ridiculous to consider these things to be hypotheses that require any type of testing." But, according to the positivist, all of these statements must be hypotheses. Now, if they are only hypotheses, then the following must be possible. That is, we can deny these statements. That is, formulate the opposite of these statements, and we would not, on the face of it, say something that is nonsensical. If we formulate the opposite, that is just as legitimate a hypothesis as the hypothesis as the statements that I just proposed. For instance, going back to the statements, um, in a voluntary exchange, um, uh, not uh, both partners do not expect to gain from the exchange. Uh uh, the preference order is the same for both individuals. Um and things are being exchanged because they have equal value. Now, this should be just as a legitimate hypothesis as the one statement that I formulated before. Um Now, but on the face of it, you realize, of course, that this is a nonsense statement. Um but according to positivist, this is just as legitimate as the other one. We would only have to try it out and then find out which one of these two versions is right. Uh that is they are only prospective winners, not prospective losers. Uh there must be opposite preference orders, or there must be no opposite preference orders. Uh goods are exchanged because they have unequal value, or they have equal value. Allegedly, only a test would tell us which one of these two incompatible hypotheses is correct. And now, again, I'm asking try to design a test that would just decide between these two incompatible interpretations, between these two incompatible hypotheses. I I have racked my brains for years and have not come up with a test that could possibly decide this sort of matter. Um or take take this example. So, if it is just a hypothesis that we if you increase the minimum wage to $1 million, unemployment will result, we can just formulate the opposite. The opposite would be if we increase the minimum wage to $1 million, more employment will result. Um that's just as legitimate a hypothesis as the other one. We would have to try it out in order to find out whether one is right, that is, more unemployment will result, or less unemployment, more employment will result. Um, or the example with socialism. Um, it would be just as legitimate to formulate the as a hypothesis to the uh calculation under socialism is possible as to say that calculation under socialism as is impossible. Both are legitimate hypotheses. Uh, and we would have to try out socialism in order to find out which one of these hypotheses is right and which one is wrong. Um, now all of this, again, as I said before, um, strikes me as being um, strikes me as uh, as absolutely absurd. Do we indeed have to introduce socialism first in order to then decide uh, whether calculation under socialism is possible or not, or don't we know that ahead of time? Doesn't that simply follow from the way that we defined what socialism is? Do we indeed have to just first try out and set a minimum wage at $1 million in order in order to know whether unemployment will increase or decrease? And even if we find out, for instance, that, okay, we set the minimum wage at $1 million, $1 million, and we find out that unemployment did increase. What a positivist then can still do, recall what I said about falsification, is say, "Yes, the hypothesis as I initially formulated, cannot be true. But if I control, let's say, for Z, which is the weather, um then we might find out that indeed, if you increase uh the minimum wage to $1 million, uh employment will actually increase. And if you control for Z and it still doesn't come out right, then you say, "Okay, but that was only because we have not controlled variable B." >> [clears throat] >> Um that is um the president of the United States or something like that has to give a has to give a a speech every Sunday, um and then we will see that uh employment will actually go up. Um and if that is not the case, then we just have variable C that must be controlled. That is, we never have any reason to give up any of our hardcore hypotheses, so to speak. We always have excuses that we can come up with in order to show why the results did not come out the way they were. Um to give you another example of this type, you say, when people just pointed out why socialism doesn't work, you know, quite well, but what what sort of excuses they always found out? Uh what sort of excuses they always gave? It was, "Yeah, socialism didn't work because Stalin was a guy who was in charge. But imagine we would just uh we would take that guy out and and put Trotsky in charge. Um then the results will come out right." And so, then you put Tros- Trotsky in charge and it still doesn't come out, and then you say, "But how about Bukharin?" Um if if he would be in charge, then the result would be all right. And if Bukharin doesn't work out all right, then you say, "It just has to be Hopper." Um if he does it, um then of course it will come out all right. And if that still doesn't come out all right, I can give you many many other names that we can put uh put in there. So there is never any reason for these people to give up any of their hypothesis ever, because you have an unlimited reservoir um of um um of excuses. Um so again, to emphasize, um intuitively, the statement that these people come up with, that these are just hypotheses, make no sense whatsoever. Most people on the street, if you would tell them this sort of stuff, they would think that you are from the loony bin, that you think that these sorts of things are just hypotheses, and the opposite might be just as a legitimate statement as a statement that I initially uh formulated. Now now let me come to to some sort of more rigorous uh refutation of this uh entire program. So far I have remained more or less on the level of uh of intuition. Um >> [snorts] >> now the more rigorous refutation goes something like this. For that we use what is called self-referential arguments, which is a standard technique in um in philosophy. Um so we ask the positivists who say there exist only empirical statements and analytical statements, we ask them back, Now, what sort of statement is this statement that all statements are empirical statements or analytical statements? Uh according to their own doctrine, there are only two answers that we can give. The first answer is this statement is itself an empirical statement. Okay? Well, then we can say, "Okay, that's just a hypothesis, right? Why should I accept that?" Uh I can just formulate the opposite hypothesis just as well. Uh you have not given me any reason why I should believe what you are telling me. Okay? The other alternative that they can propose is that this is just an analytical statement. But if it's just an analytical statement, then I then I can say, "Well, then you just simply defined things like that." Uh you know, then I can of course define my I I I can define exactly the opposite as well. Uh again, you have not given me any reason to believe that this is the right procedure uh uh that we should adopt in uh in sciences that there are only these two types of statements. And then there exists, of course, the third possibility that they have, and that would to would be to say uh this statement that all statements are empirical or they are analytical, this statement is neither an empirical statement, that is neither hypothetical, nor is just a definition. It is just something that is true about uh reality, and it is at the same time a non-hypothetical truth, which is precisely what Austrians claim can be done. That is, they claim there can be statements about reality that are non-hypothetically true. Um but then of obviously, this doctrine would be internally inconsistent. Um or take uh take the second statement of the uh of the logical positivists of the Popperites. They say, um all explanations um are uh if-then statements, and these if-then statements are hypothetical statements. Um Now, then we have to ask, what is the status of this explanation of what an explanation is? Um and again, we have uh we have only two two answers that we can give according to their own doctrine. The first answer would be just is a just is just a hypothesis of of what uh uh explanations look like. And then we would have to say, so what? Uh And then we can formulate other hypotheses just as well. Or they can say, this is simply what we define an explanation to be. And then I can say yeah, but then then I'm free, of course, to define it in any other way also. Again, you have not given me any type of uh convincing argument to believe what what you are what you are saying. The same thing with definitions. If all definitions are uh either ostensive or stipulative, then we have to ask him, yeah, but what but what is the definition of definition? Uh uh Is that a stipulative definition, or is it an ostensive definition? It is obviously neither one. This must then be what we call a real definition, a definition that defines what definitions are in a correct way. And it is neither a stipulate neither defining one thing by another word nor is it an ostensive definition just pointing out such and such is a definition. So, you see the entire doctrine is riddled with with self-contradictions. Now, I want to show in the final step of the argument that um that this procedure that Popper and his followers advocate that we should adopt in the social sciences or in all sciences whether we accept it in the natural sciences might be one thing, but in any case it cannot possibly work um in um in the social sciences. Um Uh for this, let me just indicate briefly how they proceed how they uh suggest that we should proceed in the sciences. So, we formulate a hypothesis and I use this uh use some simple regression equation. So, Y is the variable that we want to explain, and we explain variable Y by uh variable X, and here we have [snorts] uh some regression uh constants plus some uh plus some error term. Um and um so, then we collect data, and then we find uh then we uh get some sort of estimates for these types of uh constants, and then we take a second set of data, and we retest this this hypothesis, and the results can be uh either we confirm this with a second uh uh experiment or we falsify it and we come up we formulate now a different hypothesis that might be uh why will be explained not only by X but some additional variable Z and so forth. So, this we uh react so to speak to the experience of having falsified our initial hypothesis by reformulating um a new a new hypothesis. Now, ask yourselves this question. Um why is it that we can Well, what is necessary in order for us to say we have in the second experiment confirmed the first hypothesis? Um why don't we just simply say, "Well, yeah, in the second experiment we have gotten pretty much the same results as in the first experiment." Um to say we have confirmed it is more than simply saying the second time we have observed the same thing as we have in the second instance we have observed pretty much the same thing as we have observed um in the first instance. Um to say this is just a repetition means, "Well, yeah, things happen to be the same the same way." But, that does not mean there's any confirmation. Um in order to say this is a confirmation we must make the assumption that the nature of things actually remains the same in the course of time. Uh, and because of that, we say it is a confirmation. Um, the same thing with falsification. Um, we could say, "Okay, the next time we observe something different than we observed the first time." Um, so nature of things changes. Sometimes things come out that way, sometimes they come out that way. Um, why do we say the second experience falsified the first one? Um, why do we say because we did not observe the second time exactly the same thing that we observed the first time that that shows something that there was something wrong with the initial hypothesis. And the answer is again, we can only do that if we assume that the nature of things remains remains the same, and because the nature of things remains the same, then the same thing would have to show up again over and over, and if it doesn't, then there was something wrong with our hypothesis. But if we think that nature of things can change, then we just have a result. So, first we observed this, then we observed that. So what? Uh, that's not a falsification. Just one time it was this way, and one time it was that way. So, we must make an assumption here in order to say confirmation or falsification. And the assumption that we must make, um, is that there exists, so to speak, uh, prediction constants that are time invariantly true, that do not change in the course, um, of time. Um, and the question that we now have to address is, uh, can this assumption, that is the nature of things does not change. There are time invariantly operating constants that we can use in order to predict certain results. Whether we can make this assumption as soon as this Y is a human action. And there I would venture to say that to make this assumption there exist time invariantly operating causes on the basis of which we can predict certain phenomena, that this assumption cannot be made as soon as it as soon as we are dealing with human action for the following for the following reason. Yeah. Namely because all actions are influenced by by human knowledge. What we do or don't do depends on what we know about this or that, what objects they are and so forth. And it is impossible for us to predict our own future states of knowledge. And that you can see this, you can see by just looking the what Popperites tell us to do. What they tell us to do is you should operate in this way. Formulate a hypothesis, see is it confirmed? Is it falsified? Then formulate a new hypothesis and so forth. But do the Popperites know ahead of conducting this experiment whether this result will result or this result will result and we answer then with this hypothesis as this new hypothesis and the answer is of course no the proper rights do not know what the outcome of all of this is that's the very purpose of engaging in research namely to find out something that we do not yet know. If we would know the end point of all of this so to speak there would be no purpose to ever engage in any type of research. Um that we engage in this type of research confirm falsify try again falsify again and so forth indicates that we ourselves do not know what we will know at the very end of this ongoing ever ongoing research project. That is to say nothing else but uh we do not know how and in which way we ourselves will be changed as a result of conducting this research. But as I said our knowledge that we have about the world does of course influence how we ourselves act. If we cannot predict however what our future states of knowledge will be then we also cannot predict what our future actions will look like. The entire idea of being able to predict future actions in the same way as we can predict whatever the future stages of the moon or the future stages of the sun is just absolutely absurd. Every researcher must admit by going through these stages that that cannot be true for himself. That there cannot be prediction constants that allow him to predict his own future states of knowledge. He can only know what his states of knowledge are once he already has them, but he cannot predict them ahead of time. Otherwise, why do this sort of thing? So, we reach then two conclusions. The first is in so far as knowledge and the changes in our knowledge can influence our actions those things cannot be in any scientific way be predicted. That is what uh um That is so to speak speculative what we do. That's what entrepreneurs what entrepreneurs do, predicting so to speak what people will do in the future and so forth. It is not a science that can be learned according to formulas. Um that is an art. Um And the second thing that we uh that we can conclude from this is we do know, however, something that is true about action as such. In so far as action cannot be influenced by changes in knowledge. We have learned, for instance, about actions that actions cannot be predicted based on prediction constants. That is a statement that is true about all actions. And now I want to give you an example to illustrate these two Yeah, these two branches. There's things that we cannot predict or where prediction is just an art and because those things can be influenced by changes in knowledge and statements that we can make about actions as such that cannot be influenced by knowledge where we have uh apodictic knowledge that is non-hypothetical that we can just recognize as being necessarily true under all circumstances whatsoever. And for that we just as the example let me just simply choose uh the law of marginal utility. Um I think we have not really talked about that law. Um the most fundamental more or less the most fundamental law in economics that says that as a supply of a homogeneous good increases uh the marginal utility decreases and vice versa as the supply of a homogeneous good decreases the marginal utility increases. Just very briefly how we come up with this law it is the first unit of a good whatever that good is will always be used in order to satisfy the most highly ranking desire of all desires that can be satisfied with the help of this unit by definition. Um if I have two units then the second unit will be used in order to satisfy the second most highly ranking desire that can be satisfied with a unit such as this. The third one will be used in order to satisfy the third most highly ranking unit desire that can be satisfied with a unit such as this. You realize that these are so to speak logical that this this logical law uh not a psychological law a praxeological law um >> [snorts] >> that cannot be falsified in any way. That is true throughout. We cannot possibly imagine any experience that we could ever encounter that could invalidate this type of law. Now this law on the other hand does not help us at all to say what is the most highly ranking goal that you have uh once you have one unit of an orange for instance. Uh or what is your most highly ranking goal that you would satisfy if you would have one unit of an orange in your supply. Or what your most highly ranking uh desire would be that can be satisfied with a unit of an orange. It does not help us one bit to predict what his second most highly ranking goal would be that he will satisfy um with the help of an orange. It does not help us one bit to say whether what he considers his most highly ranking goal that can be satisfied with the first unit of an orange will be the same most highly ranking goal tomorrow or in three days or in four days. So it does not help us in any way to predict any concrete action that you will do. Nonetheless, it is true that every person at any given point in time will use the first unit of a good in order to satisfy whatever he considers to be his most highly ranking goal that can be satisfied with such a unit. Um So, the first aspect is that we cannot predict what these goals will be. Those are things that can be influenced by changes in knowledge. I find out I can do something with oranges that I didn't know until yesterday. Um and accordingly I put it to a new use. Um uh That is influential but that can be influenced by knowledge and accordingly we cannot make predictions about this. Their predictions are speculative. Um But the other part that everyone will use the first unit in order to satisfy whatever is his most highly ranking goal that can be satisfied with a unit of such, that can never be influenced by whatever we might learn. And because of that we can say this is apodictically true. Um can never be falsified. Uh does not need to be tested at all. And this is precisely the status that economic laws have. They are statements that say something about the real world. And they are at the same time non-hypothetically true. It is indeed absurd wanting to test these types of propositions. Uh what the positivists do is frequently things like um You a statement, let's a ball cannot be red and non-red all over at the same time. That's a logical truth. Um What they do is, well, yeah, they look at the ball and say, "Yes, indeed, this ball is not red and non-red all over at the same place, but maybe we should just finance an expedition to Australia and see if we find balls that are red and non-red all over at that place." Uh of course requiring use huge amounts of travel travel expenses to do this sort of stuff in order then ultimately to show only what we could know in in advance. Uh or people who Yeah, who try to prove uh the uh um uh the theorem of uh Pythagoras uh by measuring uh sides and uh and angles not recognizing that the uh theorem of Pythagoras can be deductively proven. Um and then again measuring measuring measuring and uh using a huge amount of uh resources to do it and come up ultimately with a result that can be shown uh by purely deductive by purely deductive reasoning. Um Most of what you find in uh in contemporary economics is either uh if it is decently correct, they they show something that you can already know in advance that water is running downward, something like that. And uh and sometimes um they also just prove with empirical methods that water is running upward and that the theorem of Pythagoras is wrong and that balls can be non-red and red all over at the same time. That is either showing something that is trivially true that for which we don't need any empirical investigations or presenting ourselves with with results that are on the face of it absurdities. Uh thank you very much. I overdrew my time here by much I have the feeling. UM THANKS ANYWAY. >> [applause] >> WE HAVE A 5-MINUTE BREAK IF THERE'S NO QUESTIONS. >> WELL, you made a strong case for a superior Austrian methodology. I just wonder how come there's no Austrian theory of finance? All theory of finance and insurance is based on neoclassical theories about general equilibrium about uncertainty and risk and time and how we define that. And I work in the financial sector and we make tons of money based on these theories. We work on based on theories based on some very abstract mathematical concepts which have absolutely no correspondence to the real world like martingale measures and expectations on the martingale measures which is the basics of all modern finance. And this is a completely fictitious concept which I understand nobody in in Austrian economics like. I mean, how come there's no Austrian theory of finance? And I mean, if your methodology is so much better, then why don't we make tons of money based on an Austrian theory of finance and insurance? >> Okay, the answer to that one is rather rather simple. Um as far as insurance is concerned, I think there's no principal difference. We use the same uh perhaps the same views on probability theory. We use the uh frequent frequency theory interpretation of Richard von Mises. Those are the theories that are used in actuarial business. There's no difference. As far as finance as far as finance is concerned, uh I'm telling you there are lots of people who don't use any of your models make far more money than you do. Uh there are plenty of Austrians uh who do not claim that they have any uh mathematical models uh available to them, who are very humble in terms of their predictive abilities, who make tons more money than you do. Um Bill Gates makes far more money than you do, and he certainly doesn't use any mathematical models whatsoever. Um so it is just an illusion to think because some people make money uh using certain techniques that shows the superiority of these techniques. Um most of most of the businessmen don't use any models whatsoever uh and make oodles more money than all of these investment advisors that sit that uh that are out there and uh and and pretend that their models um provide them some uh some basic insight about uh about the future. We know that some of the most sophisticated people in this regard uh who did these huge won Nobel Prizes for their alleged predictive capabilities, uh built these huge hedge funds, uh went broke, totally broke, and had to be bailed out by the American Central Bank because uh they threatened, so to speak, the stability of the entire financial system. That's the capability of mathematical models when it comes to this. I think You know, there exist There exist thousands of investors. They use all sorts of different techniques. There exist people who just do funda- Look at what product do they produce? Is this a good good product? What type of clients might they have? They are just as successful or more successful than people who study charts and curves and all the rest of it. So, I I think all of these um financial uh financial in um uh wizards, um they should ask themselves uh if they are so smart, uh why why aren't they the richest people on Earth? They are not the richest people on Earth. They are most people uh who rank among the richest people on Earth never ever use any of these models whatsoever. >> That's not quite an answer to the my question about how come there's no Austrian theory of finance? I mean, if you have a superior methodology, you would also be able to develop a superior theory. And then, of course, Bill Gates and a lot of people make money without finance theories. I mean, they have some other >> [clears throat] >> abilities. So, um that's not that's not the point here. The point is, if you have a better methodology, why don't you make a better theory? >> No, it is just we are more humble. We know we we we know better we know better We know better No, our methodology says for instance, that there are no equations that you can formulate that allow you to predict for instance, the movements of the stock market. >> This is not about prediction. This is simply about pricing. I mean pricing an existing product. I mean pricing an option if I say >> And this circle was a circle that competed so to speak with the Mises the Mises circle. Um, actually had a smaller number of uh um followers than the Mises circle uh itself had. Um, and um to give you just some uh some names that were participants in this circle either direct participants or on the outskirts um of the circle um uh besides Moritz Schlick uh Rudolf Carnap um Hans Reichenbach uh Gustav Hempel uh Otto Neurath and also on the on the fringes of the circle uh Karl uh Karl Popper. Uh another member also on the fringes of the circle was uh Ludwig von Mises's uh brother Richard von Mises who was a famous mathematician um the founder of the frequency interpretation of uh of probability uh theory of a man who ended his career in the United States at at the University of of Harvard. Um and um uh this the influence of the Vienna Circle was rather moderate in um in Germany, um but similar to the fate that affected most of the members of Mises' own circle, um uh most of these Vienna School philosophers um emigrated to the United States um in um in the 1930s. Many of these people were uh Jewish and it was in particular during their during the time um in that they spent in the United States that um they um I I want to talk about the Austrian method um because more than anything else it is um the method that distinguishes Austrian economics from um what we call mainstream um mainstream economics. Um and um uh in order to understand somehow uh the results that we come up with, it is also necessary to see how we come up with um with these results. Um the question that I want to address is um is economics um a science that we might call an axiomatic deductive science that starts with uh firm starting points and then deduces certain conclusions from these starting points and provided that uh there are no errors made in the process of deduction uh arrives then at conclusions that we can call apodictically true uh apodictic being a term that Mises um uh uses frequently um or is uh economics uh a science that is called an empirical uh science um that uh uh gives us hypothetical propositions that require continual testing in order to find out whether they are true or not or to phrase the whole thing in a slightly different way is economics a science more akin to applied logic um logic being a discipline where we obviously do not test our propositions but just grasp them and uh uh common sense uh logical type uh discipline um you can find that that uh um uh Jean-Baptiste Say for instance or uh Nassau Senior or um John Cairns um all of these 19th century authors held views similar to what nowadays only the Austrians believe. Um Now this new modern view about the nature of um economics uh as an empirical discipline that formulates hypotheses that require testing, um this new view um is due essentially to the influence of um a philosophical school of thinking um that is called the positivists or uh the logical um empiricists. Um Logical empiricists indicating that they uh considered themselves to be some sort of the modern um uh development of uh the traditional empiricist philosophers such as uh David Hume and uh and John Locke. Um and uh this school the logical empiricists empiricists were also called positivists uh originated uh also in Vienna at the time when Mises um was uh active in uh in Vienna. It is frequently even called the Vienna Um Not at the time when it originated. At that time uh it was called the Schlick the Schlick Kreis the uh or the Schlick Circle um named after a philosopher who taught at that time at the University of Vienna, Moritz uh Moritz Schlick. Um and he I should add that this is a relatively new phenomenon. Um until about the 1950s, um the view that the Austrians now uh hold um uh alone uh used to be the standard view that was shared by uh uh a great majority of uh of the economics profession. Uh for many, many years, the most important book on economic methodology was a book written by uh Lionel Robbins, uh the uh uh Nature and Significance of Economic Science. It was initially published in the early uh 1930s, and uh Lionel Robbins was the chairman of the economics department at the London School of Economics, um and uh was also responsible for recruiting Hayek to come from Vienna uh to London. Uh Lionel Robbins had uh uh as a young man frequently visited Vienna and participated in um in uh the famous Mises seminar that he held uh uh in Vienna and was very much influenced by the methodological views that um uh that Mises propounded. Uh you can still uh this uh book Nature and Significance of Economic Science is still in print, um and if you look at uh the index, for instance, you will find that there's uh no other person uh mentioned more frequently in the name index than uh than the name of >> [snorts and clears throat] >> uh Mises. And if you go back into the 19th century, um you will find that um um actually most economists in the 19th century shared the view that nowadays is only propounded by the Austrians that economics is some sort of um, yes, sophisticated um, would need other logical arguments in order to find out whether we made errors in our logical reasoning or is economics more a science akin to physics, for instance. Um, and I want to show that the answer to this question whether it is more akin to logic or more akin to physics has highly important implications. Um, if we look at the the present situation in in economics, then we can say that uh, except for the Austrians um all other economic schools believe that economics is a science of the type that we have also in physics. Only the Austrians believe that economics is a science akin to applied applied logic. When I say that uh, all others but economists think that economic propositions are hypotheses that need to be tested, I'm not actually saying that all mainstream economists actually practice what they preach. Um they preach in their books that it is economic propositions are hypotheses that need to be tested, but frequently, especially among the better economists, the better mainstream economists, in their actual practice, they frequently contradict their own methodological prescriptions. It is just the minor figure, so to speak, in mainstream economics that take their own methodological prescriptions very very seriously. Now, saying that the Austrians, with their view that economics is more akin to logic, are a minority