Famous Errors





1st Problem introduction

2nd Gold backing - a requirement for monetary stability?

3rd Advantage of one always disadvantage of the other (mercantilism)

4th Only work creates value?

5th Saturation thesis

6th Do deterministic processes exist?

7th Is growth necessary for full employment?

8th Purchasing power theory

9th Quotaism - suitable for avoiding discrimination?

10th Will of the people only fulfilled with proportional representation?

11th Budget deficit automatically a burden on the future generation?

12th Taxes on profits cannot be passed on?




1st Problem introduction




1st Introduction

2nd Teaching history as a history of errors

3rd Statements about reality versus approaches

4th Theories versus thought models

5th Different statement types

6th Problems with metaphysical statements

7th Singular statements

8th General statements

9th The demand for general validity

10th Human freedom and general validity




1st Introduction


The following lecture will deal with famous mistakes made by scientists. 'To err is human', as is commonly said, and this certainly applies to scientists as well. Generally, we must assume that human beings as well as human actions in particular are imperfect, this is true in every respect, both with regard to human abilities as well as their moral behaviour, furthermore for all important societal systems, for technology, economy, politics, even for cultural systems, which also include the sciences.


In this lecture, I would like to limit myself to the field of science. Here, we also must assume that in the course of history, theories have been developed again and again, which have proved to be errors in retrospect. However, when we take a closer look at the significance of these errors, we realise that this imperfection of scientists refers first and foremost to the individual scientists themselves and less to the entire scientific system. The statement that, in the course of the history of scientific doctrines statements have been formulated again and again, which in retrospect have proved to be wrong, therefore applies first to the scientist who, as a kind of pioneer, was the first to formulate a hypothesis about an earthly event.


In fact, the individual scientist exhibits all the characteristics of human incapacity. He can simply make a mistake in the formulation of a thesis, in this case he is firmly convinced that he has pointed out a new connection, but nevertheless this statement can prove to be wrong in retrospect, simply because in the formulation of this thesis certain variables were not recognised, perhaps could not even be recognised, which have an influence on the connection expressed.


Scientific statements generally have the following format: Whenever the variable x is given, then the occurrence of the variable y is to be expected. However, it must almost always be assumed that the variable x only triggers the event y if a number of other variables are also given, but sometimes other variables are not given at that time. Since our knowledge of reality is always imperfect, we can never be completely sure whether we already know all the relevant constraints and whether the statement is therefore true in each and every case.


Furthermore, it must be considered that individual factors such as origin, inclination or approach can very well affect the view of the individual scientist, so that certain statements of individual scientists can be untrue already due to these individual imperfections.


If, despite this statement, the sciences have made enormous progress in their efforts to gain knowledge of reality in the course of their history, i.e. if very many statements of science have proven to be true, then this is because science represents a system in which a division of labour is involved and the results can be achieved primarily because a large number of scientists have always been involved in the formulation of a theory.


A theory generally begins with making observations (perhaps by pure chance) and thereby observing certain regularities, precisely in the sense that a certain event x results in another event y. A hypothesis is then formulated from these observations. And this means: it appears that this correlation of effects exists. At this stage, however, no new theory has been born yet; numerous enquiries are needed to confirm or reject this hypothesis. Only then, when several empirical enquiries, which have been carried out in accordance with very specific criteria, come to the conclusion that it has not been possible to disprove this hypothesis, it is said that this statement can provisionally be regarded as confirmed, as a kind of theory; only provisionally because we can never be sure whether all factors relevant to an interdependence have really been taken into account in the experiments.


Therefore, it must always be expected that one or the other theory is not true despite prior correctly conducted empirical investigations. However, while we can assume that individual imperfections in the formulation of a hypothesis can contribute to one or the other hypothesis being formulated incorrectly in a first step it must be said in the context of empirical verification that human weaknesses are more likely to contribute to the recognition of the falsity of a statement. Just the fact that, for example, a researcher B envies researcher A's success can contribute to B's efforts to refute the hypothesis formulated by A through empirical verification. And the more intensively he pursues this refutation, the more likely it is that, if an empirical refutation has not been successful, it can be established that the hypothesis formulated by A probably does indeed correspond to the truth.


Errors have occurred in the history of doctrines in all areas of science. Even the exact natural sciences have not been spared these shortcomings. For example, in ancient times, some scientists and philosophers assumed a geocentric view of the world, i.e. the conviction that the sun revolved around the earth, whereas today we assume since Copernicus that the earth revolves around the sun like the other solar planets. In this lecture, however, we want to deliberately limit ourselves to scientific statements formulated within the humanities (i.e. the human sciences). And here again critically examine primarily, but not exclusively, economic theories in the narrower sense.


2nd Teaching history as a history of errors


In my lecture on the history of economic teaching, I mentioned at the beginning that a history of teaching can be structured according to different criteria. I had decided there to place the respective basic statements of scientific schools at the centre of this lecture. However, another possible point of view from which to scrutinise the teaching history is to list the errors of thinking of famous economists. This teaching history is then based on the motto: Nobody is perfect, even very famous and recognised researchers have made serious mistakes already.


So, it was mainly Ernst Wagemann who published a paper in 1951 entitled 'Famous Errors of Thought in National Economics' and collected some important erroneous developments in national economics. The table of contents reveals in which theoretical works the author denounces an erroneous development.


Thus, we learn, among other things, about "Old and new perspective in the doctrine of value and money (the sin register of individualistic thinking)", about " Conceptual and real order (the sin register of monistic thinking)", "About the scope of absolutist doctrines (the sin register of absolutist thinking)" and finally about "Quantitative analysis (the sin register of out-of-measure/unmeasured thinking)".         


Such an approach may well have some positive side effects, by highlighting the fact that everyone can make mistakes. And if this must be stated even for the greats and for "scientific popes", the making of a few mistakes by lesser-known researchers should not already mean a death sentence for the scientific careers of newcomers.


Nevertheless, it contributes more to confusion than to an understanding of the development of economic theory if one wanted to begin the study of the history of doctrine by first learning about the mistakes made by the researchers. It is much better to start with the actual doctrinal structure of the individual schools of thought and then, for the sake of greater depth, to learn about the individual errors in the course of the history of doctrine. For these reasons, I have not oriented my history of teaching precisely by means of the actually formulated errors. This does not mean, however, that it can still be appealing to critically examine individual errors in a second step after the individual scientific schools of thought have been presented in their main features in a first step.



3rd Statements about reality versus approaches


Also in this case, I will not follow in the footsteps of Wagemann, who critically examined individual lines of thought in economic theory and obviously branded certain approaches as errors in reasoning. However, approaches can never ever be classified as true or untrue. Approaches are methods that we use to verify the truth of certain interdependencies, but they themselves do just not consist of statements about reality.


Approaches can be suitable or less suitable for recognising the truth of certain statements, but they themselves cannot be true or untrue. It should also be clear that approaches can serve quite different goals and that therefore the question of suitability can only ever be assessed with regard to a specific goal; there is no general suitability or unsuitability of a particular approach.


Let us take the example of rose-coloured glasses. Of course, these glasses are not suitable for recognising colours; all objects viewed with such glasses appear to the observer as pinkish red, regardless of what colour these objects radiate actually. But it is quite conceivable that these glasses allow certain other structures of the considered object to be recognised, indeed under certain circumstances the very fact that the different colours are not recognised, i.e. that the main focus is not on recognising colours, can contribute to the fact that one can concentrate all the better on the other features and therefore also recognise them more sharply. Nor need the purpose of the glasses be to recognise new features of the object being viewed; for example, there are glasses that are simply intended to give the viewer the pleasure of making certain objects in nature, such as meadows, appear all the more radiant and in a saturated green, which, with these glasses, makes the green appear lusher than it really is.



4th Theories versus thought models


This lecture is therefore about the truth of scientific statements and not about the appropriateness of certain approaches. However, it can be admitted that models of thought have sometimes been developed in science which have been falsely passed off as statements about reality. For example, a dispute over methods within economics arose from the rather different uses of the word 'theory'. Some scholars understand the term 'theory' in the sense of a thinking tool. Although a thinking tool makes it possible to handle social reality, however, it must not be confused with actual knowledge of factual connections.


The theory in the sense of a model of thought is a method of expanding our knowledge, but it is not this knowledge itself. Such a theory is limited to arranging already known facts and drawing further conclusions from given assumptions. According to this understanding, finding new factual connections is the responsibility of sociologists, not of economic theorists. Economic theory could not verify the validity of the assumptions on its own. Leaving aside logical errors in thinking, such a theory cannot be wrong or right at all, but at most up-to-date and expedient.


These widespread theses of neoclassical theory were attacked by the neopositivists at about the middle of the last century. Hans Albert, for example, said that economic theory had degenerated into model Platonism. One could only speak of a genuine theory if its statements referred to factual connections that were already verified in reality or were at least verifiable. The neoclassical theory, by contrast, presented doctrines already as proven truths, which were only logically derived from hypothetically presumed assumptions and still needed empirical verification. 


Furthermore, neoclassicism worked with assumptions that had already been refuted on the basis of past empirical studies. Finally, in an attempt to defend classical positions, neoclassicism had increasingly taken refuge in tautologies, which, although logically consistent always correct, had therefore solidified into an empirically insubstantial empty formula.



5th Different statement types


Thus, let us deal in the following with the statements about reality and not with the methods by which these statements were obtained. Here, quite different types of statements can be distinguished, and we will see that the possibilities and limits of scientific knowledge are different for the individual types of statements and that therefore the question whether a true or untrue statement has been made must also be examined in different ways. Let us begin with a systematic overview of the multitude of possible types of statements. The following diagram shows the connection between the individual possible types of statements:




First, we divide the possible statements into normative and explicative statements. Normative sentences say something that is desired, they represent an evaluation. Explicative sentences, in contrast, refer to explanatory statements, they show what is and what is not, and what should be.


The explicative statements can be further subdivided into logical and factual statements. Logical statements result from purely intellectually derived connections, while factual statements refer to actual (or assumed) causes and effects.


The factual statements can furthermore refer to metaphysical or to empirical correlations. We always speak of empirical statements when these theorems have been derived from observation, whereas all hypotheses that escape observation must be classified as metaphysical. For example, I can possibly observe that a certain person died on a certain day and in a certain place. But whether the dead person continues to exist in some spiritual form as a soul is beyond human observation and all statements on such events are therefore of a metaphysical nature.


Both, the empirical and the metaphysical statements can be further subdivided into singular and general statements. Statements that refer to singular events are characterised by the fact that they occur in a specific time and place and are limited to a single object or a section of reality. General statements, on the other hand, refer to all objects of a certain class regardless of time and space.


Let us consider initially the meaning of normative statements. Jeremy Bentham had already recognised that it is not possible to derive normative statements from factual statements alone. To get from factual statements to evaluative statements, at least one normative premise is always required.


Then it was particularly Max Weber who pointed out that the scientist should refrain from value judgements, since the ultimate fundamental values could neither be proven nor disproven scientifically. Different conclusions were then derived from this postulate. Hans Albert interpreted Max Weber's demand in such a way that the scientist had to limit himself to the analysis of the possible. He had to show which possible solutions to political problems are given, but not to promote one of these possibilities.


In contrast, Gerhard Weisser, Hans Albert's teacher, had advocated the view that the scientist could and should very well make political recommendations, but he should always begin his political recommendations by revealing the value premises on which his statements are based. It is precisely the scientist who is called upon to formulate economic policy goals, since he could recognise the correctness of individual policy plans much better than laymen in economics could.


The modern welfare theory, on the other hand, took a third path in the question of evaluating economic correlations: here, in a first step, it is asked for the fundamental values that are accepted by everyone or at least by the majority of people. In a second step, normative conclusions are drawn from these fundamental premises together with further factual statements.


Max Weber himself had indeed taken an evaluative stand on political issues of the day at different times. In doing so, he expressed that the principle of freedom of value judgement postulated by him required by no means that the scientist should refrain from any value judgement. What is important for Max Weber is merely the demand that in connection with statements by a scientist it had to be clear always whether the scientist was reporting on factual correlations or subjecting them to a political evaluation. For the question of whether a statement about reality corresponds to the truth, the scientist is an expert, and his scientific judgement thus generally has a greater claim to truth than statements by untrained laymen. However, when a scientist evaluates a political measure or a political condition, he expresses himself as a free citizen whose statement does not deserve a higher claim to truth than the political evaluation of any other citizen.


So next we distinguish between logical and factual statements. The mind is capable of recognising logical contradictions. It is true that we must reckon with the possibility that in everyday life we repeatedly encounter logical contradictions and do not even recognise them as such in the beginning. However, if we examine two statements with scientific meticulousness to see if they are free of contradictions, we succeed in uncovering a logical contradiction actually almost always.


Let us take an example. We divide the income recipients into two classes: the class of wage earners and the remaining class of profit earners. Those who do not belong to the class of wage earners are ex definitione counted among the group of profit earners. If I now assume that the share of wage earners (the wage share) in the total income of both classes increases, then necessarily the share of profit earners (the profit share) must decrease by the same amount by which the wage share increases. Thus, the statement: 'the wage share rises' is in logical contradiction to the statement that at the same time the profit share rises or even just remains constant.


This conclusion, however, does not represent any real gain in knowledge about the reality of this world. The proven contradiction refers solely to the system of concepts that we humans have formed. In our example, we have defined the terms of the wage and profit ratio in such a way that, for reasons of definition alone, the profit ratio cannot increase if the wage ratio increases.


Of course, this realisation does not mean that definitions constitute something superfluous in the context of the cognitive process. We are dependent on our conceptual apparatus, without which we would not be able to gain any knowledge about factual correlations. Only by initially forming contradiction-free concepts we are at all able to make unambiguous statements about the real interrelations in this world, which other people can also verify.


Above all, without unambiguous terms that have the same meaning for all people, we would not be able to communicate our observations to other people at all; each individual would be left to his or her own devices in the knowledge process and would not be able to draw on the sum of knowledge previously gained by others. Scientific progress would not be possible. Nevertheless, it remains the case that we have not yet gained any knowledge through a contradiction-free concept formation alone; we have only gained a precondition for gaining at all knowledge by observations.


6th Problems with metaphysical statements


Let us turn now to the problem of those statements that refer to factual connections. Here we distinguish between empirical and metaphysical relations. Everything that we can observe here on earth or even in the universe refers to empirical relations, but what eludes our observation has a metaphysical background. Thus, we can observe what people do or how natural events develop, but we are not able to gain knowledge about what happens after our death, whether a part of us, our soul, continues to exist or whether all individual existence ends with earthly death. 


Nor can we observe whether there is a God who created the world and thus also human beings and, if this is the case, in what way God did this. These are metaphysical connections, i.e. connections that lie behind the physical or earthly realm. Precisely because we cannot observe these connections, a belief is needed to answer these questions that does not arise from scientific knowledge - gained with the help of the mind. 


Now Ludwig Wittgenstein expressed the opinion that one should remain silent about what one cannot prove. Although I largely agree with this view for the narrower field of science, this statement certainly does not apply to metaphysical questions. Only when we are clear about the question of whether there is a God and a life after death, and whether we will be judged after death according to our behaviour on this earth, then we can also determine how we should arrange our lives and how we should behave towards our fellow human beings. Without a determination of these questions about the meaning of life, we will likewise not find a convincing answer as to whether man is really the measure of all things and is therefore also allowed to do everything that benefits him, or whether moral commandments are to be observed, which apply regardless of the interest of the individual human being.


Now it is in turn repeatedly claimed that even without religious ties, it could be clearly demonstrated that the human being benefits its own well-being if it adheres to the moral commandments. To arrive at the ultimate basic values of humanity, there would be no need for religious faith. 


These atheistic attempts of justification have never really convinced me. Of course, it is true that it benefits my individual interest if all others around me behave in such a way that they observe the moral commandments. It is even true that the willingness of others to observe these moral commandments towards me depends on whether they can determine that I myself behave towards them in accordance with these commandments. 


From these considerations, however, the conclusion cannot be drawn that it benefits the individual in every case if he himself adheres to moral commandments. This would only be the case if all my actions were known to the respective others. But precisely this cannot be expected for two reasons. Firstly, the greater one's power, the greater the possibility of keeping one's own immoral behaviour secret from the public. For many leading elites, it is true that they have entered into informal bonds already in their education, e.g. in elitist boarding schools, which are often stronger than the official rules even in later years. 


Because of this bond, behaviour that must be classified as immoral is often covered up, and it is prevented from being made known even when official knowledge of immoral behaviour should actually lead to a complaint and thus to criminal prosecution. It is significant that when one day the misconduct of a leading person does become known, the reproach from the other leading elites does not so much culminate in the fact that the other person had committed this misconduct, but that he was incapable of keeping this misconduct secret.


Secondly, with the power of the individual also increases the possibility of forcing the others, especially the subordinates, to behave in a way that benefits the interests of the powerful, even if it is clear to these others that the powerful person does not adhere to these moral commandments towards the inferiors.


It is therefore to the benefit of the functioning of a peaceful society if all people - including the powerful - adhere to certain moral commandments and if this attitude has arisen from the conviction of a moral responsibility and does not automatically arise only from the fact that, under certain conditions, good conduct towards others also benefits one's own interest. So much for the significance of metaphysical statements.


A metaphysical problem also exists when in ancient times and medieval times the opinion was expressed that the existence of God could be unequivocally proven by scientific methods alone. Since scientific knowledge of factual connections is only ever possible to the extent that the claimed connections can be observed, and since it is not possible to observe the existence of God unambiguously, there is also no possibility of proving the existence of God unambiguously by scientific means alone. The assertion that one can prove God's existence scientifically is therefore clearly false. There is only the one possibility, that one considers the existence of God to be true by means of an act of faith.


For just the same reasons, however, the assertion of some atheists that one can disprove the existence of God with scientific means alone is also clearly false, because this would also require the fundamental possibility of observing God with human senses. Thus, the thesis of an atheist that there was no God is just as much an act of faith as the faith of a Christian, which cannot be decided by scientific methods alone.



7th Singular statements


The statements about empirically observable events can - as already indicated - be further subdivided according to whether they are statements about singular or general events. We always speak of singular statements when the event refers to a concrete single occurrence, i.e. when one can determine the place, time and the respective subject or object of this occurrence. If, for example, I state that a Mr XY performed a very specific action on a specific day and at a specific hour in a specific place, I have made a singular statement.


However, if I make certain connections for a larger group of objects or subjects, then we speak of general statements. General statements can always be expressed in the form: Whenever x is given, y is also present, or respectively, for all x applies, if x is present, then y also occurs. Let us give an example here as well. Let us consider the general hypothesis of business theory, according to which entrepreneurs try to maximise their profits under certain conditions (such as full competition). 


We do not refer this statement to an individual entrepreneur, nor do we restrict this statement to a certain period of time or to a limited area but claim that this statement applies in principle to all entrepreneurs, provided that certain conditions are given. Strictly speaking, the profit maximisation thesis states that all entrepreneurs under conditions of full competition show this behaviour, regardless of time and space and also regardless of which good an entrepreneur offers. 


However, most hypotheses in economics are so-called statistical correlations and not exact laws. Like all human sciences, economics is not capable of formulating laws in the sense of strict laws that apply to each individual object since the objects to be dealt with are so complex. The hypotheses formulated in economics are merely intended to express the fact that a certain event on a considerable scale usually triggers another event.


Let us first look at the problem of singular statements in more detail. In principle, we can assume that singular statements can be clearly confirmed or, if necessary, also be refuted. Singular events can be observed and for this very reason can be checked for their truthfulness with the help of the human mind and the human senses. When I saw with my own eyes that Mr Müller from Hagen gave a speech about the breeding of wolfhounds in front of a selected audience on 02.03.2011, I can justifiably claim that this statement could be verified precisely on the basis of my own observation. 


Individual events are - as far as scientific endeavours are concerned - first and foremost the subject of historical investigations, these occurrences then often lie many years and centuries in the past, there is often a lack of any written records in investigations of events of very early times, so that here the observation is limited even to a special extent. Witnesses of history cannot be interviewed themselves, the historians are dependent on the in many cases few written reports and must - without being able to ask the witnesses clarifying questions - check the correctness of the reports on the basis of manifold plausibility considerations.


These considerations apply, of course, especially to prehistoric epochs, for which almost every form of written account is missing. For these epochs, science is dependent on archaeological results. And it is precisely in this field of archaeology that considerable progress has been made in recent decades. While in the first years of its development archaeology still had to limit itself to proving the existence of certain people and cultures on the basis of certain excavations, in recent decades archaeology has succeeded in delivering downright astonishing results with regard to determining the age, but also the ways of life and the circumstances that brought about the collapse of ancient cultures.


Let us state in conclusion: In principle, singular statements can indeed be verified and also falsified. In practice, however, there are numerous difficulties that make neither a clear falsification nor a verification possible. We must therefore assume that a considerable part of the singular statements can actually not be unambiguously verified as to their truth content.



8th General statements


With the exception of the historical sciences, the main focus of scientific work is on formulating and verifying general statements. Here it also applies to an even greater extent than it does to singular statements that the human mind is only to a limited extent able to unambiguously verify such statements.


With the rise of the Age of Enlightenment, the triumph of modern science began which, freed from the limitations imposed by the Catholic Church on scientific inquiry, accumulated year after year new knowledge about almost all areas of the world.


At the beginning of the modern era, efforts were made initially within the framework of rationalism to gain knowledge solely by logical deductions. This was followed by empiricism, which attempted to gain knowledge not by deduction but by induction, i.e. by observation and generalisation of these observations. In this context, one also spoke of positivism, in order to indicate that science had to limit itself to the discovery of factual connections, to what was experientially given, and that it had to deal solely with ''is'', as Bentham had already recognised, and that it was therefore never possible to conclude directly from ''is'' to ''ought''.


Finally, within the framework of neopositivism, also called critical rationalism, it was recognised that laws in the sense of general statements cannot be definitively verified at all due to the limitations of the human mind. The human mind is only capable of falsifying general statements that were previously considered valid, but not of verifying them.


For example, if I formulate the statement 'All entrepreneurs maximise profits' based on observations, I can falsify this statement if I succeed in observing, strictly speaking, only one entrepreneur to whom this statement does not apply. For if I can find such an entrepreneur, then this statement does not apply to all entrepreneurs, because I have found one or more entrepreneurs who do not follow this law.


Scientific progress therefore spreads not so much by accumulating more and more knowledge or statements, i.e. that our construct of scientific statements becomes larger and larger, but rather by weeding out statements that were previously considered true and some of which were even in logical contradiction to each other, i.e. that, in other words, although perhaps fewer statements are considered true in total than before, the truth content of the remaining statements has increased.


The reason why it is not possible at all to unambiguously verify factual connections is that we can only gain our knowledge through observation, but that we are not able to check all objects of a statement for their truth content. Purely theoretically, it might still be possible in principle to include all objects that exist currently in our observation. However, general statements are valid independently of time and space. If we declare a general statement to be generally valid, then these laws would also have to apply in the past. However, this is exactly what we generally cannot do, we cannot assume that such observations have been made since the beginning of time and that - if they have been made - records of these observations have been made.


But even if we could make such complete observations for the present and the past, we do not know and cannot check whether these regularities will also be valid in the future. We therefore do not know whether the relationships discovered today or in the past are also valid for the future and thus independent of space and time.


De facto, we can always - even in the present - only make observations for a vanishingly small section of the objects to be investigated, they are essentially limited to our time and mostly also to a few spatial areas, so that we must always reckon with the possibility that an observed regularity is not generally valid, because for practical reasons - perhaps only because we lack the financial means for a complete examination - we have not been able to observe all objects.


But why is it not enough that we make certain observations on a few objects, why can't we assume that what we have observed on an individual case is correct for all similar cases? If there really are exact regularities, a correlation that has been proven for a certain object should actually be valid for all similar cases and therefore it should be sufficient to have proven this correlation in one example through observation.


However, at the beginning of a scientific study it is not certain whether the observations are based on exact laws; the cases observed in a few examples could have occurred purely by chance. For example, it has been found that in areas where an above-average birth rate was observed, an above-average number of storks nested at the same time.


However, it was not possible to conclude from this that the fairy tale according to which children are brought by storks corresponded to reality. Either this coincidence of high birth rates and high numbers of stork nesting sites was really purely coincidental, or both variables correlate with a third variable, which increases both the birth rate and the number of storks.


The answer still lies in the fact that the factual connections in reality are not so simple that a certain process x always results in a certain event y. If the real processes were really so simple, we could indeed assume that a few observations are sufficient to declare a general statement as verified.


In reality, the natural interrelationships are very complex, regularities express themselves in the fact that a certain event x only triggers another event y with certainty if a whole set of further conditions z1, z2, … zn is present at the same time or, under certain circumstances, is just not present, i.e. that the presence of a certain condition z just invalidates the occurrence of this regularity (from x follows y).


Thus, if we now find in a series of observations that event y was always triggered in response to event x (i.e. in all observations), we cannot conclude from these observations that this will be the case at all places or at all times. After all, we must reckon with the possibility that this regularity could only be observed because an additional condition z was given, which, however, was not known and which therefore could not be included in the set of conditions. We must therefore reckon with the possibility that in the future this regularity will no longer occur at all, either once or in other places, because this additional condition will then not be fulfilled any more.


As conditions for the occurrence of a general statement, only those conditions can be named of which we at least assume that they are necessary for this regularity. Conversely, we must also reckon with the fact that in the future this regularity will no longer occur because a certain, but so far unknown event did not occur, but this event prevents the regularity to be examined. In other words, we must assume that the occurrence of new, previously non-existent events prevents event x from causing another event y, as was previously the case.


This does not mean, however, that if a certain regularity has proven itself in thousands and thousands of cases, a single or even a small number of falsifications is sufficient to completely abandon this hypothesis. After all, the falsification only means that the connection formulated in this general statement is somewhat more complicated than it was assumed originally. We must reformulate this statement and restrict it to certain additional events.



9th The demand for general validity


Let us now take a closer look at the general statements about factual connections. Every science strives to arrive at statements that are as generally valid as possible. The more general a statement is, the greater is its scope of application. But when does a scientific statement correspond to this principle? We consider a statement to be generally valid if the thesis under discussion applies to all objects of a well-defined class. For example, the sentence "All entrepreneurs strive for profit" would satisfy our criterion, provided that this assertion corresponds to reality.


However, we must not expect to be able to fulfil this ideal always. But here too, the principle of 'all or nothing' does not apply. There are different degrees of approximation to this postulate. Even if we cannot always arrive at generally valid statements, we must nevertheless strive to come as close as possible to this ideal. If we can state that most entrepreneurs strive for profit, then this proposition corresponds better to the scientific claims than if we could only claim this of some entrepreneurs.


Sometimes this demand is combined with a further claim. While up to now we have only demanded that all, or as far as possible all, objects of a well-defined class have a certain characteristic, we can furthermore strive to extend our statements to ever larger classes. The sentence: "All entrepreneurs strive for profit" would in this sense be less generally valid than the sentence: "All economic people try to maximise their utility." Entrepreneurs form only a section of the larger class of people and profit maximisation is again only a section of the larger class of utility maximisation.


However, the demand for general validity comes at a high price. The greater the generality is, the greater is the degree of abstraction. If we restrict ourselves to a relatively small class of objects, we will generally find a wealth of common features. If, on the other hand, we consider an ever larger class of objects, fewer characteristics are likely to apply to all the objects in this class. Many things can be said about the behaviour of entrepreneurs. However, if we ask about the behaviour of all people, we will generally be able to identify far fewer common traits.


Here, science faces a conflict. We expect a theory not only to be generally valid, but also to have the lowest possible degree of abstraction. Both demands are in conflict with each other. To the extent that we meet one demand, we move away from the other. This conflict can only be resolved by asking both for the generally valid properties as well as for the specific differences from class to class.



10th Human freedom and general validity


Up to now, we have deliberately left aside one question. Can the demand for general validity be reconciled with human freedom? Are there any generally valid laws at all in the social sphere? Doesn't this demand assume a deterministic social philosophy?


First of all, we can state: In the field of economics, there is a whole series of scientific but socially relevant laws with which an economic theory must deal because they influence the solution of economic problems. Let us think of the law of diminishing returns, which provides information on how much production factors are needed to produce a certain quantity of goods. These correlations are primarily of a technical nature. Nevertheless, the law of diminishing returns falls within the field of interest of economic theory because entrepreneurial decisions also depend, among other things, on the course of this technical data.


There is a second point to consider. The demand for general validity only contradicts human freedom if we relate our statements to individuals. It would indeed be problematic if we wanted to assert a very specific behaviour for each individual person. No matter how much the closer circumstances suggest a certain action, we have no absolute certainty that a certain person will not yet decide differently.


The main interest of economic theory, however, is not individual behaviour but rather group behaviour. How Mr Müller or Mr Maier reacts to a price change is of little interest to economic theory. More important is the question of how the total demand or the total supply changes with a price increase. Statements about this do apply with much greater certainty. As long as only the number of considered cases is large enough, we can expect that random deviations from a general trend will largely compensate each other.


A third point should be noted. The empirical social sciences are not primarily concerned with human behaviour itself, but with the societal structure that influences this behaviour. The social sciences provide information on the extent to which the structure of society restricts the scope of action of individuals. Generally, people are left with a more or less great freedom of choice. However, there are also situations in which the social constraints become so strong that very specific behaviour must be expected. Here we are allowed to speak of societal regularities. An entrepreneur who is exposed to strong competition can only survive as an entrepreneur in the long run if he takes advantage of all possible cases of cost reduction as well.