Table of Contents 

PC.GIF     PC.GIF    

Alfred de Grazia:


Law and Behavior:
A Unified Approach to their Study

In seeking a position from which all political phenomena might be viewed, political scientists are frequently blocked by the intransigence of legal materials. Some scientists impatiently consign legal data to the age of alchemy. Most of the same people, probably for similar reasons, banish values as unfit company to behavioral science. Certainly those who attend strictly to behavior without concern for evaluational and legal contexts can answer questions that for long were buried beneath social and legal myth. Yet this primitive progress of political science should not continue indefinitely. New modes of assimilating legal and evaluative data are suggested as cures for naive behaviorism.

It is sometimes claimed that there is an inevitable conflict between the study of legal and behavioral data. This depressing theory grew out of a reaction to the juridical preoccupations of an inadequate political science of some decades ago. Consequently most recent advances in political science have come from without, a development that probably was inevitable. Those who have contributed most to political science have tended to approach their subjects with the prejudices of behavioral or psychoanalytic psychology.

Yet one does not have to reject the problems in order to reject the problems in order to reject the approach. Narrow behaviorism, although it may justify itself by real advances in several areas and by the confessed desuetude of ancient theories in other areas, encourages the persistence in legal and philosophical study of hostile and inept doctrines, and gives an inappropriate meaning to behaviorism. behaviorism, if the term has to be used at all, ought to mean science, that is, an attempt to give a parsimonious, generally agreeable, and factual statement of political relationships.

In the view of "behaviorism" or "political behavior," legal and philosophical problems stand on the same ground as problems of elections, political parties, administrative organization, public opinion and the other subjects of political science. Political science studies law in order to discover "behavioral" or "scientific" facts that cannot otherwise be learned. It studies values for the same reason. When legal facts are irrelevant to political inquiry, they should not be studied; judges, lawyers and other principals may perhaps be interested in them.


The law consists of directives to a certain grouping of persons to act in a specified manner lest they be compelled so to act or be punished by public officials. now whereas behavior is composed of all acts of individuals, one's study of behavior may be limited at any moment to the behavior described in a law. Thus, if a law directs men how to become candidates for office, one may take as hisown scope of behavioral study those persons within the purview of this law. The relevant behaviors then would be the actions of men interested in political candidacy.


Under certain conditions a perfect relationship might exist between law and behavior, as follows:


(1) LEGAL DIRECTIVE: to achieve A1 do X1

(2) IMPLIED STATE OF LAW All men who have achieved A1 have done X1





The conditions of the perfect correspondence are first that A1 (goal) and X1 (action) are defined identically, in all four categories of actions, and, second, that the legal directive is inescapable (i.e., enforcement is complete). Under these conditions, the law is practically behavior. On an empirical first level of science, to know the law would be to know the "science of behavior." And, to adapt this idea to a common worry in pedagogy, one is teaching behavioral science of a sort when one is teaching law. So long as the law rather completely describes behavior, one can teach the law as a model of behavior, using "model" as Max Weber uses the term, to designate an abstractive, heuristic device, rationalizing reality.


Under other, more usual conditions, the law and behavior differ and the law may be taken only as an indicator of behavior, having unknown validity.

(a) A1 (goal) and X1 (action) are not defined the same in all cases:

- inadvertently;

- because of vagueness in either category;

- because (1) and (2) are too detailed for the purposes for which (3) and (4) are defined;

- because different time and space dimensions are being used for (3) and (4).

(b) The legal directive is escapable.

(c) Additional behavioral statements must be made to describe more fully the legal action X1.If additional actions are called "Y", then the implied behavioral directive becomes: "to achieve A1, do X1 and Y1." (Since we are using the law here to define our scope of behavior, we ignore behavioral statements that cover actions a law does not cover.)

Therefore, we see that a perfect correspondence between a legal and behavioral statement may exist when definitions are the same and enforcement is perfect or nearly so. In such cases, law presents no problem in being studied as behavior. We also see that, more frequently, the law as a statement of behavior is insufficient: because the scientist often has reasons for defining an action differently; because the law is escapable and not enforced; and because behavioral statements are richer than the law in explaining what happens in a given case.


But the behavioral scientist should not lose one advantage in gaining another. In deciding to operate according to the hypothesis that the law is almost always and at best a partial indicator of behavior, he should not ignore the problems raised by this very fact. For these problems are some of the most important in political science; they would include these questions, which are implied by the conditions stated above:

(1) Why is behavior sanctioned? Why does every law embrace a "tendency" to become a "natural" law?

(2) What social controls cause legal and behavioral statements to coincide?

(3) What forces and institutions tend to increase the gap between law and behavior?

(4) What ideology underlies and supports the law, when it departs from behavior?

(5) What social or individual utilities tend to maintain the discrepancy between law and behavior?

(6) What are the respective parts played by law (X1 + Y1) as determinants of A1?

In brief, the political scientist who discovers that the transfer of data between law and behavior is not perfect should not abandon the discrepancy as a proof of the irrelevancy of law, but rather accept it as an invitation to the study of crucial political problems. A law is a behavioral hypothesis, at the very least. It suggests that such actions as it describes may exist and it affirms that the actions are important to know about (helping us therefore to select our research problems), since some groups of men have devoted great energy to its enactment.

We translate law into behavioral terms in order to treat of actions excluded from the law or contrary to the law. This development can be seen in American political science where "realistic" political analysis has superseded "legalistic" analysis. We can assert again, however, that the succession has not been rationalized: it has often been an impulsive reaction against formalism and unrealism, without a logic or vocabulary.

Also, law must be translated in order to manipulate the behavioral materials represented by the law with propositions that are based on behavioral universals. Thus, if we wish to generalize about candidacy for public office among diverse cultures, we must select a behavioral concept of candidacy in order to include in our studies important cultures that have no laws of candidacy. As political science loses more of its Western European focus, the translation of law to behavior must move apace.


Law has been defined strictly here, at face value, for translation into behavior. We ought also to consider a common approach to the study of legal materials that goes beyond the face value of the law and mitigates some of the criticisms of legal study put forward by behavioral science. It is the practice of many courts and legal scholars to search out the origins of laws, the character of their enforcement, and the nature of their consequences when and if enforced; this they do in order to understand the behavior of men vis-a-vis the law. Many judges and scholars look behind the laws to their social setting, whether or not their legal philosophies grant or forbid such sociological license. If a case concerns the duty of a State legislature to reapportion legislative districts periodically, the court may well determine from other cases whether the judiciary is "allowed" (by custom, precedent, political circumstances, etc.) to order the legislature to cause such an apportionment. Or, if the behavior of people in an area within the scope of a law has altered from the time of the law's enactment (e.g., deciding that the early laws canoed be now enforced since they have long been out of effect).

Can political science be satisfied with this kind of legal study? In what senses does it obviate the need to translate law into behavior? If I may be allowed to generalize the juridical and research procedure just described, I should put it as follows:

(1) The legal directive used in the cases coming before the courts or juridical scholars takes one of the following forms:

(a) To achieve A1, do X1.

(b) To achieve U do X1 where U is an unknown goal.

(c) To achieve A1 do Q where Q is an unknown action.

(2) The "research" procedure of the courts and juridical scholars is as follows:

(a) In the case of (1a) above:

Expose the insufficiency of X1 to achieve A1. That is, employ the formula: To achieve A1 do X1 and y1 where y1 is non-legal behavior. This has already been treated above; and in effect, as we have said, it uses the law to provide the behavioral hypothesis (defining scope and nature of the behavior), and goes beyond the law to describe the relevant complementary behaviors. As we have pointed out, this is quite adequate for limited purposes, and especially for vocational instruction, but allows "law" to "choose" our political theories.

(b) In the cases of (1b) and (1c) above:

-A1 is missing because: the law is poorly defined, or ambiguous, or the researcher supplies his own value to the law;

-X1 is missing for the same reasons.

Also, in both cases, the researcher may add complementary behaviors (y) to his scope of study. In these cases, the comment made above regarding (1a) applies. In additions arising from the probability that the researcher is introducing personal preferences (not of the law nor of a general character) to his study. This is appropriate political science in the case of U, if the preferences are made clear and the distinction made that the law is ambiguous or that the substitute preference is being inserted. In the case of Q, however, the researcher is allowed by the rules of scientific procedure only to estimate how much of Q is X1 and how much Y1 because the solution of Q is a scientific problem. (It is not permissible, we may add, to enlarge or diminish the scope of the law portion of Q, i.e. X1, by preference, because though X1 may be unknown or vague, it is nevertheless a factual problem, not an evaluational one.

To sum up the several preceding paragraphs, therefore, it is believed that those more sophisticated renderings of the law that probe into behavior, represented by the work of such men as Holmes, Cardozo, Brandeis, and Pound, are subject to the conditions of research that were originally stated. They are complex versions of the translation problems treated earlier, and Introduce no essentially new problems. The logic of research into law and behavior remains essentially simple. Legal materials do not involve a third language; they are soluble in the languages of axiology and sociology.


In extending a proposition of political science to cover larger units of time and space, one finds less and less use for law as meaningful statement. It is understandable that anthropologists have scarcely any use for the rubrics of statutory or documentary legality. This writer discovered in 1954 that the Human Relations Area Files of Yale University, an extensive cross-indexing of materials on the world's cultures, was of little help to his comparative study of election-systems because of present gaps between the Western specialized election vocabulary and the diffuse, unmodified systems of selecting public officers that prevail over much of the globe. In moving outward in time and space, the laws, whether of election or of other subjects, lose their concrete behavioral referent. The greater the distance the more profitable it is to study laws as ideational conglomerations, i.e., as ideologies or institutions. We can intelligibly compare the election laws of France and Germany or those of the United States and the Soviet Union in order principally to compare ideologies, or the tendency of clusters of laws of similar type to have some behavioral effect. In discussing a single jurisdiction, we can study the laws in greater detail as synopses of decisions, evidences of concern, determinations of moral impulses, and hurdles or barriers for specified behaviors.

Extensions of the generalizations of science tend to produce the following effects upon legal generalization, as contrasted to applied behavioral science.




To achieve A1, do X1 (e.g., to become a candidate for election in district I, one must file a prescribed petition of candidacy with a certain number of signatures).



To achieve A1, do X1+Y1 (e.g., to become a candidate for election in district I, one must file a prescribed petition of candidacy with a certain number of signatures, and one must bring witnesses with him to the clerk's office to assure that the petition is duly registered, admitted, and unaltered by the political opposition).



To achieve A1, A2... An, do X (e.g., to become a candidate for election in county I, one must file a petition).



To achieve A1, A2 ...An do X+Y (e.g., to become a candidate for election in county I, one must file a petition and take steps to prevent the petition being ignored or voided by actions of the political opposition).



To achieve A, do X (e.g., to become a candidate for election one must fulfill existing laws of candidacy).



To achieve A, do X + Y (where X=O, to achieve A do Y). (E.g., to become a candidate for election one must have fulfilled any existing laws of candidacy and must have access to or power over the officials who control admissions. More simply, one must have access to or power over the officials who control candidacy).

These examples indicate that, as the scope of one's generalization increase, it is increasingly less useful to consider the law qua law. Rather, one tends to incorporate the legal directive into the behavioral one. The same process occurs in the case of the implied state of lawfulness and the behavioral statement from observation.

We should also add that recording the very existence and extent of laws over time and space may be useful in generalizing about the ideology of a group of societies or historical periods. In certain connections, the "law-making compulsion" is an important fact about cultures. Ordinarily, however, one is not confined to a particular functional branch of law (e.g., electoral law) to demonstrate a characteristic "legalism" of a culture or group of cultures. The researcher may select some example of the laws or some branch of law whose development is closely related to the ideological facet he wishes to describe.

PC.GIF     PC.GIF    

 Table of Contents