Introduction to Social Science

1. The foundation of political economy and of every social science is psychology.

A day may come when we shall be able to deduce the laws of social science from the principles of psychology, in the same way that some day, perhaps, the principles of the constitution of matter will give us, by deduction, all the laws of physics and chemistry.

But we are still very far from this state of affairs, and we must adopt another approach. We have to start from certain empirical principles to explain the phenomena of sociology, as well as those of physics and chemistry.

At a later date, psychology, by extending the chain of its deductions, and sociology, by going back to principles that are more and more general, may combine and form a deductive science, but these hopes are still far from being realized.

2. To put some order into the infinite variety of human actions that we must study, it will be useful to classify them according to certain types. Two of these at once suggest themselves.

A well-brought-up man enters a drawing room; he takes off his hat, says a few words, and makes certain gestures. If we ask him why, all he will be able to tell us is: it’s the custom. He behaves in the same way in certain matters of far greater importance. If he is Catholic and goes to Mass, he will perform certain actions “because this is what is done.” He will justify many other actions of his by saying that this is what moral principles require.

But let us imagine this same individual in his office, engaged in buying a large quantity of wheat. He will no longer say that he is acting in a certain way because such is the custom, but his purchase of wheat will be the final outcome of a process of logical reasoning that is based on certain experimental data. If the data should happen to change, so would the conclusion, and he might refrain from buying, or he might even sell wheat instead of buying it.

3. We can thus, by abstraction, distinguish between () nonlogical actions and () logical actions.

I say “by abstraction” because in real actions the categories are nearly always mixed, and an action may to a large extent be nonlogical and to a small extent logical, or conversely.

For example, a speculator’s actions on the Stock Exchange are certainly logical; but they depend, even if only in small part, on the individual’s character, and because of this they are also in part nonlogical. It is a known fact that some individuals are generally bullish; others are more bearish.

It should be noted, moreover, that nonlogical does not mean illogical; a nonlogical action might have been the best that could be found, given observed facts and logic, for adjusting the means to the end; but this adjustment was obtained by a procedure other than that of logical reasoning.

It is known, for instance, that the cells in a honeycomb end in the form of a pyramid, and that with a minimum surface, hence with the minimum outlay of wax, they achieve the maximum volume, that is, they can hold the largest quantity of honey.

Nobody, however, supposes that this is so because the bees have solved a maximum problem by syllogisms and mathematics; it is obviously a nonlogical action, although the means are perfectly adjusted to the end, and consequently the action is far from being illogical. The same observation could be made about a great many other actions, which are usually called instinctive, either by men or by animals.

4. It should be added that man has a very marked tendency to consider as logical actions that are nonlogical. It is because of a similar tendency that man animates and personifies certain material objects and phenomena. Both these tendencies are found in everyday language, which, in retaining traces of the sentiments that prevailed when it was being formed, personifies things and facts, and presents them as the outcome of a logical will.

5. This tendency to consider nonlogical actions as logical lessens and turns into a tendency—which is equally mistaken—to consider the relations among phenomena as having solely the form of relations of cause and effect, whereas the relations among social phenomena are much more often relations of mutual dependence.1 Let us note in passing that causal relations are much easier to study than relations of mutual dependence. Ordinary logic is usually adequate for the former; while, for the latter one often has to resort to a special form of reasoning: i.e., mathematical reasoning.2

6. Let A be a real fact and B another real fact, which have a causal relation between them, or one of mutual dependence. This we shall call an objective relation. To this relation there corresponds, in man’s mind, another relation, A B , which is strictly speaking a relation between two concepts in the human mind, whereas AB was a relation between two things.

This relation A B we shall call subjective.

If we find in the minds of men in a given society a certain relation, A B , we may investigate the following things: (·) What the nature is of this subjective relation— whether the terms A , B have a precise meaning, whether there is or there is not a logical link between them. (‚)

What objective relation, AB, corresponds to the subjective one, A B. („) How this subjective relation, A B , arose and how it was determined. (‰) In what way the relation AB was transformed into the relation A B . (Â) What effect the existence of these subjective relations A B has upon society—whether they correspond to something objective, AB, or are entirely imaginary.

When A B corresponds to AB, the two phenomena develop in a parallel way; when the former becomes somewhat complex, it is called a theory. It is considered as true (I,) when A B corresponds to AB throughout its course, that is, when theory agrees with experience. There is not and cannot be any other criterion of the truth of a theory.

The same facts may, moreover, be explained by an infinite number of theories, all of them equally true, because all of them reproduce the facts to be explained. This is what Poincaré meant by saying that when a phenomenon admits of one mechanical explanation it admits of an infinite number of them.

More generally, it may be observed that establishing a theory amounts in a sense to fitting a curve through a certain number of determinate points. An infinite number of curves can satisfy this condition.3

7. We have already remarked (I, ) that we cannot know any natural phenomenon in all its particulars; consequently, the relation A B will always be incomplete in comparison with the relation AB—if for no other reason than this, that it will never be possible for these relations to completely coincide; the subjective phenomenon will never be a strictly faithful copy of the objective phenomenon.

8. But these phenomena can diverge from one another for quite different reasons.

Whereas, for the scientist who studies natural facts experimentally in his laboratory, the subjective phenomenon approaches the objective phenomenon as closely as possible; for the man who is affected by feeling and passion, the subjective phenomenon may diverge from the objective one to the point of having nothing in common with it.

9. It should be noted that the objective phenomenon appears to our minds only in the form of a subjective phenomenon, so that it is the latter rather than the former which is the cause of human actions. To be able to influence human actions, the objective phenomenon first has to be transformed into a subjective one; 4 hence the great impor- tance to sociology of the study of subjective phenomena and their relation to objective phenomena.

Relations among subjective phenomena are rarely a faithful copy of the relations among the corresponding objective phenomena. The following difference is very often observed. Under the influence of circumstances, some people perform certain actions, P, . . . , Q; afterwards, upon reasoning about them they discover, or believe they have discovered, a principle common to P, . . . , Q; they then imagine that they have derived P, . . . , Q as a logical consequence of this principle. In reality, P, . . . , Q are not the con- sequence of the principle, but it is the principle which is the consequence of P, . . . , Q. Admittedly, once the principle is established, actions R, . . . , S, which are deduced from it, follow; thus, the disputed proposition is only partly false.

The laws of language provide us with a good example. Grammar did not precede, but followed, the formation of words; nevertheless, once established, grammatical rules give rise to certain forms, which have been added to the existing forms. To sum up, we may group the actions P, . . . , Q and R, . . . , S into two sets: the first, P, . . . , Q which is the most numerous and the most important, is preexistent to the principle which seems to govern these actions; the second, R, . . . , S , which is secondary and often of very little importance, is the consequence of the principle; or, in other words, it is an indirect consequence of the same causes which directly produced P, . . . , Q. 10. The phenomena A and B of § do not always correspond to the real phenomena A, B; it very often happens that A or B , or even both, do not correspond to anything