The Inferiority Complex of the Social Sciences
[On Freedom and Free Enterprise (1956)]
It is said and repeated over and over again that the social sciences are so very young, relatively speaking. Why is it that social scientists insist on this as a statement of fact, and why do they consider it worthwhile repeating?
The habit of not-so-very-young women of understating their age and emphasizing their youthfulness probably rests on the observation that, as a rule, younger women are regarded as more eligible, desirable and attractive, partly because from some point on beauty is a decreasing function of age, partly because inexperience and innocence are associated with youth and are highly valued by many men. This, however, is not a helpful analogy for us. Innocence, inexperience, beauty — these are surely not the attributes which social scientists wish to claim for their subjects as means of attracting more followers and admirers.
Another analogy may come closer to an explanation. Very young children are forgiven when they misbehave and do silly things. Perhaps social scientists wish to claim this privilege of childhood in order to secure the indulgence of the adult world, as if they were saying, "Pardon us for being so dumb, but we are still so very young." By implication they seem to promise, "Wait till we grow up, wait just a few hundred years, then you will see how smart we shall be." In any case, apparently, while they are children they should be accorded the privilege of being silly; after all, children do not know what they are doing.
The closest analogy, in my opinion, is the well-known apology of many people in games and in sports, trying to account for their awkwardness and clumsiness. If they admit that they are old practitioners of the game or sport, their poor performance may be attributed to lack of intelligence or talent; but for "novices" they are not doing so badly. Thus, "Excuse me, I am just a beginner," is an often-heard apology from participants in sports and games who have a feeling of inferiority. This is what is probably behind the social scientists' pronouncements emphasizing how young the social sciences really are: "Please do not think we are stupid; we are merely beginners."
Only those who feel that their accomplishments are unsatisfactory and inferior to those of others have a reason to point to the fact that they are relatively new at their business and thus should not be expected to be any better than they are. Whether or not they actually are poor performers is not of the essence: an inferiority complex may or may not be justified by some "objective" standards. It is the feeling of inferiority which makes the sufferers overapologetic, excessively aggressive, or looking for other sorts of compensations.
The trouble with the protestations by social scientists is that their story about their "young" science is not true. We have only to open our textbooks on the history of social theory, political science, or economics to find that we have no right to engage in that baby talk about being mere children, or in those novices' excuses of being mere beginners. Our subjects are as old as any; the scholars and writers in classical Greece had as much interest in problems of society as in problems of the physical world, and their achievements in the former are not less than those in the latter.
But the social science "youngsters" or "beginners" will quickly protest against my reference to our ancient predecessors and will proclaim, "What they did must not be called 'science'! Only recently has social thought become social science." Such pronouncements force me to return to the analogy of the "beginner" in sports. When I once heard the familiar "I am just a beginner" from a ski bunny whom I had seen snow-plowing many years before, I was impolite enough to remind her of it. But undaunted she said, "Oh, that does not count! That was not the right technique; you cannot call it skiing!"
This is precisely the line these perennial beginners, the social scientists, are trying to sell: "Oh, what all these people, long ago, were doing was not the right scientific method, you cannot call it Social Science!" I do not buy this line about the "right method" and want to warn against it. The old students of society used whatever method they believed was right and expedient, and they thought — 2,500 years ago; 2,000; 1,000; 200 years ago — that they had succeeded in acquiring more knowledge, and more accurate knowledge, about human action than the man in the street had. That should make them social scientists in no less "good standing" than anybody who uses the most fashionable methods of our day.
That the old scholars engaging in the study of society did not call themselves "social scientists" is surely irrelevant. Until recently their subjects were part of "moral philosophy," just as physics was part of "natural philosophy." The fact that Newton and his contemporaries considered his work as natural philosophy does not prevent us from calling him a physicist (although he also wrote much on philosophy and theology and believed that his contributions to these subjects were of major importance).
It is not by what name it was called, nor by what method was used, nor by what success was had from the point of view of posterity that we should judge whether a certain body of knowledge at some time past was "science." Knowledge is "scientific" if it is impartial, systematic, and more complete or more accurate than "popular" knowledge at the time. The fact that in the course of the last hundred years several writers have proposed rather narrow definitions of "science" — restricted in terms of particular subject matters or particular methods — and were allowed to get away with these restrictive definitions, has caused anguish to many social scientists.
If the restriction had always been in terms of subject matter and had excluded social phenomena once and for all, less serious harm would have followed — because the study of society could do nothing to "qualify" for the title of "science." But many of the restrictions were in terms of particular methods, and this created an ambition on the part of social scientists to earn the right to the honorific title by adopting as far as possible, and even farther, the methods that were elected as the definitional characteristics of "Science."
It is in terms of some of these restrictive definitions that the social sciences are deemed to be so very young. Those who insist that a science must be a system of deductions inferred from a small number of axioms or postulates will date the birth of economic science with the publication of Ricardo's Principles and will reject the scientific character of political science, sociology, and most other social disciplines.
Those who insist that a science must be exclusively based on a series of inductions from a large number of exact observations and precise measurements of objectively discerned phenomena will date the science of sociology as a rather recent creation and will reject the scientific character of economics, political science, and most other disciplines commonly counted among the social sciences. These are only two of a large number of definitional restrictions. When in a recent textbook on the methodology of social science the author states that "If we are honest we have to admit that the first century of social science has left us somewhere short of victory," we can infer that he proclaims Auguste Comte as the progenitor of social science and accepts his method of "positivism" as the essential criterion of "science."
Perhaps it ought to be said that there exists no method-oriented definition of science under which all parts and sections of physics, chemistry, biology, geology, and other generally recognized natural sciences could qualify as "sciences." Definitions of science which stress the theoretical system, the network of logically interrelated hypotheses using mental constructions of ideal exactness, undoubtedly exclude large parts of chemistry and biology. Definitions stressing repeatable experiments and verified predictions clearly exclude the parts of biology, geology, and cosmology which deal with the evolution of life, of the earth and of the universe. And even within physics — the discipline which is the science par excellence because most definitions of science were formulated with physics in mind as the model — the authorities are by no means agreed as to whether the deductive system or the inductive technique constitutes its scientific nature.
It would be interesting to catalogue the definitions of science proposed or adopted by writers in different fields or in specialized branches of larger disciplines. They all formulate the specific characteristics in such a way that their own kind of work would still qualify as "scientific," while they have little concern, if not undisguised scorn, for fellow workers in their own discipline, in cognate fields, or in fields with which they are entirely unfamiliar.
Many a scholar thus excluded from the honorary fraternity of "true scientists" suffers from severe frustrations and develops an inferiority complex, or aggravates the one he had to begin with. In defense against the humiliating "rejection" he either tries to change the definition of science by enlarging the extension of "scientific method" just enough to have his own particular working techniques covered or he adopts working techniques which, however unsuitable to the subject matter or problems under investigation, are safely approved, or can somehow be represented, as "scientific."
A mere enumeration of the subjects now customarily regarded as social sciences will suffice to make it clear that a demand that they follow the same methods (let alone, the same method) is entirely impractical, if not fantastic. The list includes sociology, cultural anthropology, social psychology, human geography, demography and population theory, ethnography and ethnology, political science, economics, history, international studies. This list is incomplete and overcomplete, depending on whether particular fields are granted "autonomy."
Moreover, it can easily be shown that many of the supposedly separate fields are largely interdependent. Finally, most of the subjects call for several approaches, descriptive, historical, statistical, and theoretical, which have to be skillfully integrated in the application to concrete problems. An insistence on the use of "the" scientific method for all would be nonsensical.
What is really meant by "the" scientific method? In its narrowest sense, scientific method is supposed to mean experimental method, or the demand that every proposition be "verified" by repeated laboratory experiments with strict controls of all conditions. In a wider sense, scientific method is supposed to mean statistical method, or the demand that every proposition be "verified" by numerous sets of statistical data relating to sufficiently comparable situations. If no wider extension of the definition is conceded and if no proposition is deemed "scientifically" acceptable unless it is confirmed by such scientific method — alas, only a minute fraction of all propositions about human action in society would be acceptable, and only the most insignificant propositions at that.
Needless to say, all sorts of additional concessions are proposed in order to accommodate other kinds of scientific inquiry. But there is no epistemologically defensible borderline short of the widest meaning of scientific method, defined in the Encyclopedia Britannica as "any mode of investigation by which impartial and systematic knowledge is acquired." Such largess would give away any pretensions by which one scholar may assert superiority over another on grounds of the purity and sanctity of his method; it would remove any need for feelings of guilt or inferiority on the part of scholars who ably and diligently add to our store of knowledge by inquiries which are neither experimental, nor statistical, nor quantitative, nor of predictive usefulness. But this largess in the meaning of scientific method is not widely accepted and we must continue to labor under the restrictive definitions and to bear the consequences of the inferiority complex of the social sciences.
These consequences or manifestations of the inferiority complex of the social sciences are chiefly in the form of scientistic compensations. Some of them are old and may yield to treatment; for some more recently observed forms no cures have as yet been developed. Some, though satisfactorily described have not even been given technical names, and I shall have to propose nomenclature. Although there are probably several more, we shall deal here only with the following: (1) Historicism, (2) Institutionalism, (3) Holism, (4) Behaviorism, (5) Operationism, (6) Metromania, (7) Predictionism, (8) Prescriptionism, (9) Mathematosis, and (10) Experimentomania. Needless to say, most of the afflicted will not recognize their attitudes as aberrations in any sense, but will insist that they, and they alone, have the right insights and all others are "unscientific."
Before I attempt to formulate the briefest possible statements of the symptoms and manifestations of these conditions, it may be well, in order to avoid even temporary misunderstandings, to anticipate here in the form of examples some explanations that will later be given in greater detail. A historian need not be a historicist — indeed, few historians are — and, moreover, even a fanatic historicist may be an excellent historian. Scholars engaged in social statistics, quantitative economics, econometrics, mathematical economics, or mathematical analysis in the other social sciences — however exclusively their interests may be in quantitative and numerical research and analysis — may be far removed from the attitudes characterized as metromania and mathematosis; and even some who are afflicted may produce useful results. Thus, their work is not in question here. What I find unhealthy in the ten listed attitudes or beliefs is, above all, the attempt to urge certain methods on others in the name of "science" and to disparage the research of others, not perhaps because their arguments or findings are fallacious, self-contradictory, or contradicted by evidence, but because they fail to employ the method claimed to be the only "scientific" one.
Historicism insists on the accumulation of historical facts as the only legitimate beginning and as the sole basis of social research; on the prohibition of the use of theory in the interpretation of past events, though sometimes admitting that theories might eventually be distilled from large masses of historical data; but the validity (not merely applicability) of any such theories will be strictly limited as to time and place. What laboratory experiments are to the natural sciences historical research is to the social sciences: just as the experimental method is required in the study of nature, the "historical method" is required in the study of society and makes it "scientific." Pure theory is useless speculation, sheer metaphysics; history is the scientific method of the social sciences.
Institutionalism, sharing with historicism the view that social theory cannot be general theory and is neither "perpetual" nor "cosmopolitan," holds that human attitudes, objectives, and organizations — all called "institutions" — are subject to human control and, hence, must not be taken as fundamental assumptions in the analysis of human action; instead, social sciences must concentrate on factual descriptions of the institutions and their evolution; thus they will be based on facts rather than on speculation and preconceptions.
Holism (derived from "the whole" rather than "the holy") takes several forms; one insisting on the notion that the whole is prior (logically and historically) to its parts and that, therefore, the study of society must start with the "social wholes" or collectives — the nation, the community, the market, etc. — rather than with the individual and some of his motivations and actions; another insisting that different aspects of human action should not be separated in analysis, but that social conduct and organization should be studied realistically and "as a whole." To start with the individual and to isolate particular aspects of his actions is held to be unrealistic speculation, whereas the observation of the undissected whole will permit scientific social research.
Behaviorism insists on confining social sciences (as well as psychology per se) to the establishment of regularities in the physical behavior of man under strictly controlled conditions. All interpretation of human action on the basis of introspective insights or in terms of mental constructions, postulating the existence of motivations or preferences, is rejected as speculative; in order to be scientifically sound, research must be restricted to objectively discernible facts, observable and describable in physical terms.
Operationism (or operationalism) insists on the exclusive use of so-called operational concepts in scientific discourse; that is, all concepts must be defined in terms of operations, chiefly physical operations of the scientific observers. Mental constructs without operational counterparts — idealized concepts — are either rejected outright or only temporarily admitted on the expectation that they will soon be replaced by operational concepts. "Conceivably operational" concepts are sometimes, in exceptional cases and only grudgingly, condoned for want of "practically operational" concepts. As a concession it was (somewhat inconsistently) proposed to admit "mental operations" besides physical operations, but this was not widely accepted since it would open the door to metaphysical speculation.
Metromania, stemming from a fixation on the dogma that "science is measurement," takes the form of attempts to measure everything however faintly connected with the subject under investigation and to imagine the resulting figures to be relevant, and of urgent claims that any proposition not amenable to quantitative verification be rejected as "unscientific." The questions of the stability of computed numerical relations and of their historical relativity are usually ignored and ever-new statistical figures for different or longer time intervals are produced in order to devise "corrected" parameters or coefficients "explaining" the measured magnitudes of social reality.
Predictionism, impressed by the success of natural scientists in predicting the outcome of controlled laboratory experiments, sees the sole purpose and justification of scientific inquiry in the formulation of propositions instrumental in successful predictions of events in the real world, including the social world in which only few relevant factors can be controlled or even reliably ascertained, let alone measured. Generalizations of merely explanatory, not predictive, usefulness are rejected as speculative.
Prescriptionism insists, in emulation of the great practical achievements of the physical sciences, on practical usefulness of the findings of research in the social sciences; it demands their use in devising improved social institutions and, especially, in economic organization that satisfies the needs of mankind substantially better than the present one; embracing the dogma "savoir pour prévoir pour pouvoir" it denounces pure theory as apology of the status quo and, in the name of "science," calls for action to carry out the prescriptions. These are usually for social control of economic life either on the basis of "scientific socialism" or by governmental planning and interventions.
Mathematosis is the urge, incited by admiration of the paramount use of mathematics in the physical sciences, to employ higher mathematics in expressing propositions that could equally well be expressed in ordinary language. Purely "literary" arguments are scorned, and ideas or problems not reducible to mathematical formulation are suspected of being "metaphysical" or "pseudo problems."
Experimentomania combines the firm conviction that practical experiments alone are "scientific" with the illusion that social research will eventually be "solidly" founded on practical experiments under strictest controls; all present research techniques are regarded as preparations for eventual experimental research, and research problems are invented that are immediately amenable to laboratory techniques even if they are of little relevance to any hypotheses significant in the systems thus far employed in the various social sciences.
All these attitudes, beliefs, and ambitions use the flag of "true science" as a means for gaining support and allegiance and for combating the nonbelievers. Their own method is the best — not perhaps because it has proved particularly fruitful and has yielded results not obtained by other methods — but because it is the only "truly scientific" one. All other methods ought to be rejected — not perhaps because they have not been instrumental in producing or confirming knowledge or insights — but because they are "not scientific."
There is at least one other notion that the described attitudes, beliefs, ambitions have in common. The social scientists who display them are apparently ashamed of the one thing that really distinguishes social sciences from natural sciences, namely, the fact that the student of human action is himself an acting human being and therefore has at his command a source of knowledge unavailable to the student of the phenomena of nature. The student of atoms, electrons, magnetic fields, enzymes, genes, etc., is himself none of these things and has no immediate experience of them, whereas the student of human thinking and acting is a thinking and acting human being and knows a good deal about the subject of his inquiries before he starts inquiring.
The close and unbreakable link between prescientific everyday knowledge and scientific knowledge about the subject matter of social sciences is both an aid and a burden. It is an aid in that it furnishes the social scientist with an initial stock of experiences, working hypotheses, and interpretations of fundamental importance. It is a burden in that it saddles him with the obligation to work with constructs that are understandable to him and his fellow men in terms of their everyday experiences; that is to say, he is under the obligation to make his scientific constructs correspond in all relevant respects to the constructs that are used in everyday life in the commonsense interpretation of our fellow men's actions.
Social scientists laboring under the inferiority complex they have developed under the frustrating notion that the methods of the natural sciences are the only truly scientific ones refuse both to recognize the "obligation" and to take advantage of the "aid" just mentioned. They mistake the prescription of scientific "objectivity" for a proscription of "subjectivism" — confusing "subjective" in the sense of impartial with "subjective" in the sense of cognizant of inner experiences.
But we must also guard against a possible misunderstanding: that we do not respect the positive and constructive values in the described attitudes, convictions, and ambitions; such values should be recognized. Thus we must be sure not to confuse historians with historicists, nor to discount the value of good historical work merely because its author happened to cling to historicist views aggressively critical of all theoretical analysis.
We must not underestimate the importance of descriptive work on the institutional features of our social organization, even if its author is a firm believer in institutionalist methodology and should be deadly opposed to all general theory. We should admit that the holists' fervor for integrated studies, though often destructive in their rejection of isolating abstraction, may at times result in the discovery of data and the development of promising hypotheses. We must acknowledge that behaviorists have done good work and have come out with significant findings, even if their campaign against introspection and speculative reasoning about intervening variables probably has obstructed progress in the social sciences more than a little.
Although it is true that the attempts of the operationalists to ban pure constructs has had obscurantist effects, we must grant that they have been successful in developing a number of statistically operational concepts as useful counterparts for pure constructs and thus have contributed much to our stock of factual information. We must not take all specialists in social statistics, quantitative economics, or econometrics for metromaniacs; moreover, while some metromaniac may have wasted money on piling up mountains of stultifying statistics, and may have misdirected some of our best talents, his enthusiasm for empirical work has probably been productive also of useful quantitative studies, for which he deserves credit regardless of the damage done by his preaching about his exclusive scientific method.
The predictionists are of course perfectly right in encouraging the formulation of generalizations useful for prediction and testable by the success of predictions based on them, and we must thank them for such encouragement, despite the gratuitous and harmful disparagement of purely explanatory hypotheses. The prescriptionists have frequently turned the attention of the social analyst to practical problems of immediate urgency when the latter was preoccupied with spinning hypotheses of remote applicability; for this they must be given credit even if most of the time their zeal has badly messed up theoretical analysis as well as practical policymaking. We should be careful not to regard every mathematical analyst as a mathematotic; and even the latter should be thanked for having contributed to substantial improvements in the mathematical training of social scientists, useful for a better selection of talents and also for greater elegance of exposition. Perhaps there is also something good to say about the achievements of the social-science experimentomaniacs, though I have not yet been able to find anything.
In brief, good historical and institutional studies, interesting holistic hypotheses and behavioristic research, the development of operational concepts, improved quantitative-empirical research, encouragement of attempts to predict and to test, attention to the practical problems of the day, and better training in mathematics — all these are highly desirable things in the social sciences. What is harmful is the attitude of snubbing, disparaging, excommunicating, or prohibiting the working habits of others and of preaching a methodology that implies that they are inferior in scientific workmanship.
Good "scientific method" must not proscribe any technique of inquiry deemed useful by an honest and experienced scholar. The aggressiveness and restrictiveness of the various methodological beliefs which social scientists have developed — in subconscious attempts to compensate for their feelings of inferiority vis-à-vis the alleged "true scientist" — are deplorable. Attempts to establish a monopoly for one method, to use moral suasion and public defamation to exclude others, produce harmful restraints of research and analysis, seriously retarding their progress.