On the Misuse of Reason and Science

11.12.2019

The following paper by Professer Hardy Bouillon* was presented at the IV. ECAEF/CEPROM Conference on ‘Towards a Viable Alternative: Markets and Entrepreneurship to Protect the Environment’, 10 December 2019. Initiated by the European Center for Austrian Economics Foundation based in the Principality of Liechtenstein, this academic conference series is dedicated to the eminent late French scholar Jacques Rueff. The co-operation with CEPROM (Le Centre d’Etudes Prospectives pour Monaco) was highly appreciated.

On the Misuse of Reason and Science

by Hardy Bouillon

Abstract | The current environmental debate suffers under serious misuses of both, reason and science. The paper sketches telling examples to highlight some of the most worrisome that impact the well-being of human society. While the misuse of reason is particularly visible in the ignorance of the subjective character of reason, its counterpart in science becomes prominent in the perversion of basic methodological rules in the natural sciences. To show this, we will focus on two phenomena: first, the hubris marking the claim that private vices and public benefits can be identified in a scientific and objective way; second, the improvident use of methodological rules that akin to Friedman’s concept of positive economics.

Introduction | The concern of this paper is mainly methodological. This being said, it should be made clear from the outset what methodologists, or philosophers of science, can and cannot contribute to the topics discussed in this conference. So, let us start with a few pre-emptive remarks!

Philosophers of science are not judges who simply know (or even know better than others) what is scientifically true in any field of knowledge where the problem of theory preference comes up as a result of rival theories contradicting each other. Philosophers of science are nothing of that sort. How could they? In most cases, they are not scientists themselves, they are meta-scientists. They do not possess the knowledge of the competing scientists, i.e., they do not know better what those know. Therefore, in no way do they qualify for the job as chief umpire (1).

Given the high degree of complexities and manifold interrelations among the causal factors in most sciences, a philosopher of science cannot even assess whether a proposed test to settle the dispute among rivalling theories really qualifies as test, or which of the alternative tests is to be preferred (2). Neither can he, for the same reasons, conclude whether any, some or all of the competing theories provide causal explanations or correlations only when shedding light on the phenomena put up for debate. Finally, he cannot – as little as everybody else – decide questions of theory preference by applying any majority rule. Even in cases where allegedly 97% of the experts favor one theory to its competitors, one cannot conclude that the majority is right and the minority is wrong (or vice versa). Similar things hold to reliability issues. A scientific theory is neither true nor false because its adocates have a higher (or lower) degree of reliability or reputation within their scientific community than rivalling colleagues.

Bearing this in mind, it appears that the philosopher of science is a referee who can tell both teams how to play, but cannot intervene when any of the team players plays foul (3).

This being said, what can philosophers of science do, and what can they do in order to correct the above image? A few things are quite obvious and basic, namely explorations on the methodological criteria theories have to fulfill in order to be justly called scientific. For instance, following Popper, theories should be formulated as falsifiable universal statements, namely by allowing for predictions that can be tested independently of the theory. Should the test lead to predictions that contradict the predictions of the theory – by theory we mean theoretical systems, consisting of one or more universal hypotheses and boundary conditions – and should the scientist value the test less problematic than the theory in question, then he has to rate the theory as falsified. That is what methodology tells him. Of course, falsifications are fallible too, hence each falsification holds valid only as long as the abovementioned problem ratio stays positive – i.e., the theory (short for theoretical system) is viewed less problematic than the test statement (4).

Logically, the following takes place. A theoretical system ‘TS’ leads to prediction ‘P’. A corresponding test leads to a test statement implying non-P (¬P). Logically, we can infer that ‘TS’ is false (5). Formally put:
TS ├ P
¬P
—————-
¬TS or: ¬(H1, H2, …C1, C2 …) via modus tollens (6).

The question is: Where to put the blame? Neither logic nor methodology can tell. All we can say is that at least one element out of TS is false (given ¬P is less problematized). It is up to the scientist to decide how to go on.

In addition, philosophers of science can reflect on methodological differences due to differences in the nature of the sciences, for instance methodological differences among physical and social sciences. Pointing to such differences, they can judge whether a theorist has a proper understanding of the methodological rules in his field; whether a natural scientist follows the rules of the social sciences or (vice versa) a social scientist follows the rules applicable to natural sciences only.

Likewise, a philosopher of science can observe and/or comment when concepts of rationality applied in the social sciences are used in natural sciences and vice versa. Equally basic, a philosopher of science can judge on the nature of scientific recommendations. That is to say, he can tell whether a scientist comes up with a hypothetical imperative (7) only or goes beyond.

Bearing this in mind, we have to correct our first impression of the almost useless referee. A philosopher of science cannot intervene in each and every case when a scientist plays foul, but in a few specific and important cases he can. Hence, let us look at some cases where the methodological referee might be useful in ecological debates. For that purpose let us keep in mind what science is about. It is a process of gaining new knowledge and replacing suboptimal knowledge by improved knowledge via critical assessment of given theories. In other words, we are taking about falsification.

The methodological point of view

In developing his falsificationist methodology, Karl Popper mainly addressed cases in which falsifiable theories are falsified (or corroborated) by tests and subsequent test statements viewed less problematic than the theory itself. His model8 works equally well for testing a single theory or for testing two or more competing theories, provided they allow for opposing predictions (9). In both cases the test (respectively test statement) fulfills a clear purpose, namely it indicates whether a claimed causality between cause and effect exists or not. Corroboration (usually) means it does, falsification means it does not. Of course, both judgments are based on a tertium non datur, i.e. on the fact that the existence of a third option can be excluded. (It may so happen that the test outcome is caused by a (yet) unknown causality.)

How important the tertium non datur proviso is, can be seen particularly well when a third option is given but ignored in the reconstruction of the case (i.e., as if it was not given). This becomes evident when we turn to a charming example once introduced by Wesley Salmon (10).

His example runs as follows:
1. Every man who regularly takes contraceptives will not become pregnant. (H1)
2. John is a man and regularly takes oral contraceptives. (C1)
John is not becoming pregnant. (P)

The pun of the example is obvious: It does not matter whether or not John takes oral contraceptives. We know that. Being male, he cannot become pregnant anyway. 1 and 2 are irrelevant for the truth status of the prediction. Although the tertium is not present in the logical reconstruction, we know that it exists. Hence, tertium datur – a third option of explaining the result is given. As a consequence, we know that John not becoming pregnant does not corroborate the thesis that taking contraceptives has caused his infertility.

This has consequences, not only for John and his family planning, but for another case also often happening in science, namely that competing theories allow for the same predictions, rather than for contradicting ones. Such a case is given in the current global warming debate, a case we are simplifying a bit for illustrative purposes. Under such simplifying conditions, we can argue that at least three competing theories exist, all claiming to explain rises in global mean surface temperature (GMST), all using causal links that are possible in reality. The first blames human carbon emissions, the second bovine methane, and the third solar activities. Now let us assume that predicted increases are measured, leaving aside the controversies regarding the fact itself, methods of measuring, actual size of increases and the percentage each of the suggested sources allegedly causes.

What does the measured increase (or test, for that matter) show? It is possible that the result corroborates one, some, all, or even none of the competing theories. The last option could be the case, given there is a yet unknown tertium datur (11). In any case, the result is by no means apt to give a definite answer. All we can conclude is that none of the rival theories is falsified by the “test”.

Let us make things easier. Let us assume that there is no unknown tertium datur. If so, the situation is comparable to a serial killer case. Assume the inspector caught three suspects. On the basis of the circumstantial evidence, he concludes that suspect A is the murderer. The court shares his view and put the accused into jail. Once he is in prison, the series stops, no more victims. What are we to conclude? Of course, it is possible that the right guy is now behind barred windows. But it is also possible that one of the remaining suspects has been the killer. Happy of not being convicted, he stops his series.

That is to say, even if all emitters of carbon dioxide would stop or drastically reduce their emissions, whereupon temperature would decrease, we could not conclude which of the three theories has been corroborated. Likewise, if we would stop or drastically reduce husbandry, whereupon temperature would decrease, we could not conclude which of the three theories has been corroborated.
Interestingly, all those who believe that none of the two theories claiming anthropogenic global warming (AGL, for short) is true should have a scientific interest in falsifying those theories. This appears to be quite ironic, for the best “test” would be to reduce or stop cattle raising and/or industrial carbon dioxide (12).

Certainly, the search for truth is the main ideal of each scientist. And the appropriate hypothetical imperative a scientist could give here would be: “If you want to falsify the two mentioned theories, then you have to test them by reducing or stopping cattle raising and/or industrial carbon dioxide.”
However, neither advocates nor repudiators of AGL-theories would seriously ask to cover the costs of such a test for reasons of scientific curiosity. Advocates ask to cover the costs for a different reason (13). They believe that the costs will be compensated by the expected gains of “saving the planet”. Repudiators, not sharing such expectations, refuse the test as useless and by far too costly. To them, reductions in carbon emissions are a waste of resources and as useful as oral contraceptives distributed for free to each juvenile and grown up male.

Such we face a very unusual situation in science. The regular case is that scientists are rather reluctant to expose their theories to hard testing, in particular after having invested a lot into them, fearing losses in reputation, future funding, etc. Their critics, interested in their own reputation and economic well-being, work hard to test their opponents’ theories and hope to falsify them. In the climate debate it is the other way round. The critics are not interested in testing, while the advocates are (14). Of course, there is an explanation to this apparent pervert situation.

The concepts of rationality differ. In the natural sciences the objects are physical entities and data regarding those objects are more or less objectively given and also relatively stable over time. In the social sciences the objects are human beings and the data are, as Hayek has put it, “opinions of those whose actions produce the object of the social scientist. … beliefs … we cannot directly observe in the minds of the people but which we can recognize from what they do and say merely because we have ourselves a mind similar to theirs.” (15) Scientists are not physical objects. They are human beings and behave as all other humans. That is to say, they follow their natural drives within universal boundary conditions; i.e., facing limited resources, they have to economize on their means in order to pursue their individual competing goals.

Against this background, it does not come as a surprise to us that advocates and repudiators have something in common. Although reasoning in different directions, they argue in terms of efficiency (saving planet, saving wealth). By doing so, they appeal no longer to the mind set of the natural scientist, but to that of the economist or political economist. Of course, it is perfectly legitimate to change places. But when you change from natural sciences to social sciences, then you face different methodological rules.

What this means in practice becomes evident when we look at what governments do and cause when picking up the results of scientists. They do not pick up on hypothetical imperatives scientists can give, but on political recommendations some scientists give ignoring the methodological limits of their field of knowledge, or on political recommendations they read into the results of some AGL-advocates.
Doing so, they misuse both science and reason.

Changing places and abuses of reason

Let us take Germany for example. Among the producers of territorial fossil fuel CO2 emissions worldwide in 2017, based on the share of global CO2 emissions, Germany ranks 6th (see table). It has a share of 2.21% and plans some drastic reductions, say about 1/3. Should this come true, the overall reduction of global CO2 emissions would be 0.7%.
Largest producers of territorial fossil fuel CO2 emissions worldwide in 2017, based on their share of global CO2 emissions

If we would live in a world guided only by the rules of natural sciences and filled with individuals whose rationality would be akin to that of physical objects in terms of reacting always in a given predetermined way and thus following universal laws, how then would we interpret such government behavior? Probably we would say something like this: The German government reduces the global CO2 emissions by 0.7%. By doing so, they react to the thesis that the current level of CO2 emissions is detrimental to good living conditions on earth. They reduce the part which is under their control and hope this will have an impact on global climate.16 If science should show that 0.7% does not matter at all, then the German government would view their initial hypothesis as being falsified and thereafter would resign from their reduction plans. If they would not do so, their behavior would be irrational …

Continue reading ->
On the Misuse of Reason and Science (.docx)

* Hardy Bouillon is Professor of Philosophy at the University of Trier, Germany, and Research Director at the Swiss Management Centre University, Austria