Commonsense and Social Science

Claudio Gutierrez

A fundamental point frequently made in the writings on methodology of economics is the assertion that this science, and social science in general, is no more than an elaboration of the content and categories that one finds in commonsense knowledge. This argument is usually employed in order to support the concomitant assertion that social sciences and natural sciences have methods completely different from each other even at the logical level. On the other hand, one also encounters writers in sharp opposition to these views maintaining, on the contrary, that commonsense categories must be superseded by more abstract ones in order for a particular science to exist at all, and, as a corollary, that there is one, and only one, homogeneous method of science. I side in this controversy with the former authors, although I do not think that the continuity between social science and commonsense is sufficient ground for maintaining a fundamental diversity in the methods of the social and the natural sciences. My view on the matter of how the two types of sciences are to be distinguished, and also related, is to be made clear in the sequel. I consider entirely gratuitous and almost mythical the notion that the rise of a scientific conception must be understood as a rupture with the historical past and a creation of a self-contained, wholly new, interpretation of reality. On the contrary, I believe that here, as in the case of values, previous opinions are always fundamental.

The problem of the advance of science has always been a problem of reformation of opinion, of clarification of accepted knowledge which we all, even scientists, have in a systematic form. On the other hand, it is not clear where to draw the line between common and "exceptional" sense or knowledge, since the knowledge in the possession of intellectual elites has always been a source of utmost importance for the formation of common knowledge –knowledge of the common people. Even more importantly, if one is to treat this problem rigorously one should better avoid the use of such commonsense terms as "science" and "commonsense." One should rather, in this context talk about formalism and non formal or unformalized knowledge, even if these terms do not correspond completely in their extension of meaning with the former. If one does this, then one can easily see that not only the historical origin of sciences, but also the regulation of its formal apparatus and the application of its formal systems to concrete cases must be controlled by unformalized, even ultimately unformalizable thought. The attempt to formalize rules of application, for example, is bound to incur infinite regression. NOTE 1 That this regression does not in fact present itself in the actual functioning of science is what we refer to when we speak of a professional setting or a professional atmosphere in the practice of science. (FRIEDMAN 53, p. 25).

Over-preoccupation with formalism and the making of sharp distinctions between commonsense and science, is no doubt based on sound concerns and on the realization of an important problem. We have to have some criteria to distinguish between authentic science and "metaphysics"; we must be able to defend science against the pseudo scientist (HUTCHISON 38, p. 13). Nevertheless, the only real protection against the pseudo scientist turns out to be reliance on organized science and on the sense of responsibility of the professional scientist. Other safeguards are at best insufficient (POLANYI 64, pp. 53, 203-207). At worst, over-preoccupation with formal guarantees tends to degenerate into a new brand of obscurantism which could eventually arrest the development of science. Objective tests are not enough to defend science against intruders. How are the tests themselves to be tested? The important question is whether the rules are to be applied always in the same way by the professional people. Consensus is of the essence of science –as much as it is of the essence of truth and of value (KNIGHT 51, p. 118; KNIGHT 56, pp. 153-155).

In summation: One could profitably reinterpret the meaning of "commonsense" as signifying non formal responsible thinking, and the meaning of "science" formal responsible thinking. Of course, the professional reality of science is more than pure formalism, and real commonsense is more than purely non formal thought. But the polarization is illuminating in that it shows more clearly what the issues involved are. It is interesting to note that this polar distinction makes of commonsense (identified as non formal thought) the governing level, because it supplies the conditions for the good operation of (formal) science. Compare this scheme with the claim sometimes made as to the correspondence of commonsense with an object language and of science with the respective meta-language. NOTE 2 This distinction of levels can, no doubt, also be made, but it certainly cuts across the boundaries between science and commonsense. As a matter of fact, commonsense itself moves in different levels of discourse, as the ample treasury of popular wisdom about the object of the social sciences (people) might very well demonstrate. There are "commonsense object-language" and "commonsense meta-language" as much as there are different levels of language within formal science itself. Here again it seems that we come to the same conclusion: Not the distinction between science and common sense, but rather the distinction between formal and non formal, even unformalizable, thought or knowledge is fundamental. And this is only an ideal or abstract distinction. In practice formalism is absolutely incapable of existence without the full simultaneous cooperation of non formal intellectual powers. In the face of this, we cannot look upon "objectivity" as the easy solution for our restlessness and need for security. We can only receive the reassurance which comes from responsibility in the use of formal and analytic devices by professional people. This responsibility, coupled with a careful development and application of formalism, is what I want to call the professional sense, an essential defining characteristic of a scientist.

Closely related to the problem of the relationships between science and common sense is the problem of the role of teleology in the explanation of the phenomena of the social sciences. Conflicting views are also encountered in this area. I will concentrate here on a significant pair of them, represented by the following revealing quotations:

Under a wide range of circumstances individual [business] firms behave as if they were seeking rationally to maximize their expected returns....
............
[Confidence in the maximization-of-returns hypothesis is justified, however, by the fact that] unless the behavior of businessmen in some way or other, approximated behavior consistent with the maximization of returns, it seems unlikely that they would remain in business for long....


Milton Friedman


* * *

The characteristic feature of man is... that he consciously acts. Man is Homo agens, the acting animal.

All–apart from zoology–that has ever been scientifically stated to distinguish man from nonhuman mammals is implied in the proposition: man acts. To act means: to strive after ends, that is, to choose a goal and to resort to means in order to attain the goal sought. (Italics mine.)


Ludwig Von Mises

The first quotation seems to be a very puzzling piece of philosophy of science because expressing a circumstantial rather than a logical argument; hypotheses have, according to that argument, the nature of supreme judges of businessmanship because the falsifying instances will destroy themselves in the process of falsifying the theory. A parallel type of methodological interpretation will make of driving an automobile, or of any other practical skill based on definite "laws of nature," a very queer experience. Confidence in the rules of good driving would be justified because unless the behavior of drivers approximated behavior consistent with those rules it would seem unlikely that they would remain alive for long. In this way, one could hope, the followers of the rules will be vindicated (!). Mises' solution would not fare very well either. Under rigid interpretation of conscious teleology one would have to assume that the driver makes complicated calculations of ends and means in the fraction of a second necessary for correct response to road stimuli.

One could still save Friedman's position by understanding his contention as saying that "as-if" hypotheses express only rules of correct behavior (KAUFMANN 58, p. 217). But this will not eliminate the teleology implicit in the normative prescription. One could also save Mises' by saying that strictly conscious teleological reasoning is only a prototype of actual decisions. I, for my part, find it easier and more convincing to think that the truth is not in either side. In my opinion, both businessmen and drivers are always making heuristic guesses as to the technique of conducting business or cars, rather than strict calculations or sheer random divination. They develop, in their daily personal contact with the trade, an habitual knowledge that we precisely call "art" or "skill." The heuristic framework which they dwell in for making their appraisals supposes teleology, of course, because they intend to guess right, but there is neither clear awareness of the steps involved, nor minute arithmetic of ends and means. They only try to guess right. Contrary to Friedman, one has to accept that teleology is allpervading, purpose being connatural to human thought. But, contrary to Mises, one has to accept trial-and-error and the instauration of good habits as equally important.

But a complete analysis of the role of teleology in social sciences will have to go deeper. It is indispensable to distinguish between two different, although complementary, teleological aspects. On the one hand we have teleology as a dimension of the explanation; on the other hand, we have teleology as an element of the subject-matter. Teleology of the explanation corresponds to the purpose of the investigator and has to do with the role of subjectivity on the foundations of science, with the operation of non formal powers of thought. Teleology of the subject matter corresponds to "purpose" as an essential category for the understanding of behavior. It is subjectivity in the object of thought rather than in the mind of the scientist. For clarity's sake, let us refer to the first aspect as teleology₁ or subjectivity₁ and to the second aspect as teleology₂ or subjectivity₂. Teleology₁ relates to the fact that the scientist, as the person that he is, has purposes, knowledge, desires, passions; all of it, of course, determinant in the way he conducts his scientific activity, even contributing to define him as a professional scientist. Teleology₂ relates rather to the fact that the subject-matter which social scientists study, i.e., people, do also have purposes, beliefs, passions, etc., so that the corresponding terms –"purpose," "belief," "passion"– are part of the language the scientist must use. Those terms are best interpreted as theoretical terms, as for instance the term "force" is interpreted in the physical sciences: it is impossible to render them directly as descriptions of pure sense data. If one does this, that is, if one takes teleology₂ as a case of theoretical terminology, then the difference between the social and the physical sciences appears less dramatic. Both have to deal with some content –theoretical terms– that they do not fully understand, at least from the point of view of strictly empirical observation, and both have much the same way of dealing with such content "that they do not understand." They treat it in a formal way, applying to them the same kind of logical manipulation and rules of inference that they apply to terms not directly derived from sense experience.

It is interesting to witness the different approaches of writers on methodology according to their personal intellectual inclination for teleology₁. Let us compare the following two quotations of Knight, and Simon:

We never succeed entirely in eliminating consciousness from our ideas of material things.... Mechanics has not been able to do without the notion of force, though whatever force is, more than motion, is a fact of consciousness.... We interpret the behavior of the most material thing by to some degree putting ourselves in its place....


(KNIGHT 51, p. 120)

* * *


To say that it is often convenient to use the term "purpose" in social science means no more and no less than to say that it is often convenient to use the term "force" in physics. A physical force is no more directly observable than a social purpose.... To bar from natural or social science all terms that do not refer to directly observable events is to confuse deduction with induction. All that we require of the language of science is that its propositions, whether directly observable or not, permit us to deduce other propositions that are directly observable and hence testable. Propositions about "purposes" and "desires" do permit such predictions and tests, and hence are not different from propositions in the natural sciences....


(SIMON 50, p. 409)

There is here a fundamental asymmetry that we could express in schematic form thus: (1) The teleology₁ approach –dialectical, synthetical– is not very fruitful at close range, but is self-validating and necessary in the last analysis. (2) The teleology₂ approach –positive, analytical– is in the end self-contradictory, but very useful at close range and for particular purposes. The former is not fruitful at close range because nobody wants to be saying only things which are directly implied by the way we set up our vocabulary. One does not want to be always pronouncing tautologies. It is unavoidable in the last analysis because all language is ultimately circular –it is always developed through its own application (as we shall see more fully in the next chapter). The latter is in the end self-contradictory because nobody can expect to know anything without being the one who is there to know about it (non-subjective knowledge would be a flat impossibility). It is useful, nevertheless and very much so, at limited range (within dependent hypotheses) because all logical inferences proceed by making essential, although transient, use of inconsistencies. NOTE 3 One can feel in all this the unmistakable scent of the two ultimate requisites of all systematism: the principle of completeness and the principle of consistency. In the autonomy of the two approaches in particular, one can sense an epistemological version of the metamathematical idea which teaches that (sophisticated enough) formal systems cannot be proved to be both consistent and complete (GÖDEL 31). More of that later. For the time being, let us note that we have come to a conclusion very similar to the one reached in the last section: Formalism is useful and indispensable, but its functioning necessitates the complementary full exercise of non formal, ultimately unformalizable, powers of thought. NOTE 4