The Hoover Institute on War, Revolution, and Peace is sponsoring a major series of studies into the nature of the ruling groups in the Soviet Union, Nazi Germany, and the democratic countries. The project is part of a larger research undertaking known as RADIR—Revolution and Development of International Relations. Here, RICHARD H. S. CROSSMAN—as a member of Parliament himself a fairly recent accretion to Britain’s ruling group—looks into four volumes so far published: The World Revolution of Our Time, by Harold D. Lasswell, a general introduction to the RADIR studies; The Comparative Study of Elites, by Harold D. Lasswell, Daniel Lerner and C. Easton Rothwell; The Politburo, by George K. Schueller; The Nazi Elite, by Daniel Lerner (Stanford University Press). 

_____________

Why have the social sciences progressed so little during the last hundred years? In the 1850’s it was confidently believed that the methods of natural science could be rapidly applied to the study of society; having learned to predict, and so to control, the development of his physical environment, man would soon be able to apply the same techniques in the mastery of human relations too. The “association of ideas” was regarded as a natural law of psychology as simple and all-pervasive as Newton’s law of gravitation; and it was assumed that social behavior would soon be regulated according to Jeremy Bentham’s “felicific calculus”—with each man maximizing pleasure and minimizing pain, and each free society harmonizing interests and so producing the greatest good of the greatest number. It seemed so obvious that once these laws had been understood and permitted free play the social engineers would be able to construct a rational and free society with the same assurance with which the railway engineers were constructing their railway systems.

Alas! These visions—popularized by such philosophers as Auguste Comte and Herbert Spencer—were based on superficial and quite fallacious analogies between Newtonian physics and psychology. As their inadequacy became clear, they were replaced, towards the end of the century, by another set of analogies—equally superficial and fallacious—between biology and sociology, stimulated by Darwin’s Origin of Species. This pseudo-science envisaged history as a struggle of individuals and social groups for survival, and concluded that any nation or social system which survived had thereby demonstrated its virtue. In answer to the challenge of Darwinism—the philosophy of a militant Free Enterprise—Karl Marx produced his counter social science on behalf of a militant Proletariat. Borrowing and turning upside down the Hegelian dialectic, he sought to extrapolate—that is, calculate future relations on the basis of past ones—and so to predict social evolution by fitting it into the straitjacket of class war. Social relations, which had previously been viewed as the rational behavior of self-interested individuals, were now to be explained “scientifically” in terms of irrational economic forces-just as later on, at the beginning of this century, Freudian psychology was to find the “scientific” explanation of individual behavior in the workings of an irrational unconscious.

So far, in the development of social science, it had always been tacitly assumed that there must be some substance (the individual mind, or economic class, or the unconscious) underlying human society and corresponding to the matter or atoms which formed the unanalyzable basis of 19th-century physical science. But when the physicists began to question the existence of basic “matter,” and probability replaced natural law in physics and chemistry, the fashion in social science suffered a corresponding change. Statistics, based on “samples,” replaced the crude laws of psychological association, and the sociologist began to evolve quantitative techniques—such as the public opinion poll—as his correlative to the new nuclear physics.

My contention, indeed, is that ever since Aristotle started on the job the social scientist’s attempts to classify societies or political institutions have been imitated from botany and zoology, and his “social laws” have been little better than adaptations of the natural science prevalent in his age. The natural scientist demonstrates that his discoveries are true by showing that, within limits, he can use them to predict the future: indeed, a scientific law means a statistical prediction, verified by observation. In an effort to rate as respectably “scientific,” the social scientist borrows the current categories of natural science and applies them to human relations. Unlike the natural scientist, however, he has usually been unable to verify his predictions. What he succeeds too often in doing is to express in complicated statistical terms what had already been observed by the non-scientific historian.

_____________

These reflections are reinforced by a reading of the first four volumes of the Hoover Institute Studies, which have been given the resounding title of the RADIR Project (“Revolution and Development of International Relations”). Mr. Daniel Lerner is the Director of Research, and Professor Harold Lasswell the presiding genius of this very considerable enterprise, which is financed by the Carnegie Corporation of New York. This is how Mr. Lerner and Professor Lasswell describe their objective: “The object of the project on Revolution and the Development of International Relations (RADIR) is to describe and explain the ‘world revolution of our time.’ More specifically, its purpose is to make clear the major trends of social change throughout the world within the past six decades, to estimate their direction, intensity, and tempo, and to seek explanations of their significance. Because of our concern with a developing world community in which democratic values shall prevail, we seek to describe these changes in relation to four principal goal values: shared power, shared respect, shared well-being, and shared safety. It is with these values as compass points that we attempt to locate the direction of change and to help reveal the significance of the vast revolution that is reshaping our contemporary world.”

No one can object to a project with such an edifying aim. Mr. Lerner and Professor Lasswell are staunch democrats, “in step with ideal values of the American tradition and with the progressive ideologies of our epoch.” Their hope is that “the historic trend away from caste societies will continue until the free man’s commonwealth is achieved on a global stage.” This hope, as professional social scientists, they find it necessary to describe as a “developmental construct.” A “developmental construct,” we are told, is the characterization of “a possible sequence of events running from a selected cross-section of the past to a cross-section of the future.” (Some of us who are old-fashioned used to call this a guess about the future. But never mind.) Mr. Lerner and Professor Lasswell recognize that they are faced with a rival “developmental construct,” namely that “. . . the direction of history is reversing itself, and moves towards the restoring of caste. More specifically: Assuming that the world crisis of insecurity continues, power and other values will be further concentrated in a few hands in the name of providing for the common defense. As the world is bipolarized between the United States and the Soviet Union, perpetual crisis favors the loss of freedom, and the eventual consolidation of garrison-police states. . . . In the end, if the process is carried to the logical (thinkable) conclusion, the leaders of the garrison-police state will constitute the top layer of the new caste system.”

Apparently the aim of the RADIR Project is to apply the latest methods of social science in order to discover which of these “developmental constructs” (or guesses) will be confirmed by history. Shall we advance towards a free man’s global commonwealth, inspired by the values of the American tradition, or will the world be polarized into contending police states?

Reading the first fruits of their research, I am ready to enunciate a law about social science. The verifiability of any research which purports to extrapolate and predict social developments is in inverse proportion to its social importance. It is quite possible, by means of consumer research, for instance, to state scientifically whether Americans (or Englishmen) prefer their chocolates in a black or white box; and the success of this prediction can be measured by the graph of the sales achieved by rival firms. It is also possible to study electoral preferences in the same way, and any party boss is well advised to use these market research techniques in selecting his merchandise for the political market, though he may let himself in for unpleasant surprises at times. Of course, no one can deny that statistical methods can be used, within narrow limits, to predict developments in human behavior. They are useful auxiliary instruments in the task of political, as of commercial, salesmanship. What they can never become without disaster is a guide either to statesmanship or to the writing of history.

_____________

Let Me state the reason why. Professor Lass-well and Mr. Lerner are anxious to serve the cause of American democracy by putting social science at its service in the struggle for a free society. Their aim, presumably, is to improve the quality of American statesmanship by enabling Americans to understand the “revolution in which they are all living” and permitting them to foresee the trend of history. But, in dealing with broad historical trends, statistical methods are at least as likely to confuse as to clarify. And for two reasons. In the first place, the assumptions on which their extrapolations are based are usually not neutral scientific hypotheses but value judgments, or, to put it more bluntly, prejudices. Try as hard as we may, we cannot make history “scientific.” And, if we pretend that we are doing so, we either cloak our prejudices in scientific language and so blunt our vision, as the Stalinites have done in Russia; or else we do succeed in remaining strictly scientific, in which case our researches degenerate into an insignificant jumble of statistical tables that offer the unfortunate politician or historian who looks to them for advice an infinite variety of possibilities to choose from, without any standard by which to choose.

The second limitation on the contribution of social science to the study of human behavior is, of course, the nature of free will. What distinguishes human from natural evolution is precisely the ability of man to change society by defying the observed patterns of behavior. If we seek to defy the laws of our physical environment by jumping off a cliff or by building a house in defiance of the observed behavior of bricks and steel, we suffer the predicted consequences. But the advances towards freedom have been achieved by defying the trends of history. The social mutations which have enlarged our freedom of choice have nearly always been the jump in the dark of an outstanding genius or a new, dynamic social group—acts of will, in despite of observed social laws. Thus a social science which lays large claims to being “scientific” is always in danger of straitjacketing the human will it seeks to liberate.

National Socialism had a “scientific” theory of race, Communism has one of class and social evolution; and in both Germany and Russia there grew up a breed of social scientists ready to give elaborate scientific justifications for these inroads on human freedom. The free man’s response to such scientific ideologies is not a counter-ideology, or a counter social science, but a denunciation of all ideologies as such, and an exposure of every attempt to confine human will power and social change within predictable patterns.

Suppose, for instance, that Professor Lasswell and Mr. Lerner, after two or three years more of sociological research, come to the conclusion (which has already been reached by Mr. Toynbee) that the trend of history is against the free society and in favor of totalitarianism. Will that affect in any way the American determination to fight for the free society’? Of course not. Men fight better against odds, as Mr. Churchill discovered in 1940, when, by all the laws of statistics, he would have been well advised to surrender to Hitler. To show us what is statistically likely to happen can neither teach us what we ought to do nor yet reveal to us the “laws” of social change. It was the complete defiance by the early Christians of the trend of history which finally enabled Christianity to conquer the Roman Empire. It was because its missionaries were prepared to lose their lives that they saved their faith from extinction. Those who let the trend of history influence their value judgments form the vast majority of the human race—but they contribute nothing to the enlargement of freedom. That enlargement is accomplished by the tiny minority who confute predictions based on observed patterns of behavior by a belief in the impossible.

_____________

We Can express this critique of the claims of social science to study revolution in yet another way. The patterns of behavior of a stable, unchanging society (a primitive, savage community, for instance) can be observed, classified, and even broken down into quantitative sociological laws. But what differentiates civilized from primitive society is precisely the fact that its observed patterns are constantly being changed by acts of human will. To seek for the “laws” which determine social change (as Marx certainly did, and as the authors of the Hoover Studies seem inclined to do) is to fetter the future in the chains of the past. For the sociologist can only predict the unknown future insofar as he assumes that it is determined by the known past, i.e. so long as it contains no new mutations caused by acts of free will.

Of course, for many purposes we are entitled to make this assumption. Even within a civilized, changing community—the United States, for instance—there will be countless types of behavior that repeat themselves in almost identical form. We are creatures of habit and custom, and our freedom is rooted in usage. The social scientist can successfully study these established patterns of behavior—the net reproductive rate of a country, for instance; its rate of capital investment; or even (where democracy has become a habit) its voting patterns. What the social scientist can never do—except at the risk of making himself look extremely silly, or else falling into totalitarian habits of thought—is to explain the laws of change or, in the words of Professor Lasswell, “the revolution in which we live.” For if this revolution is a genuine revolution, then it is one of those mutations which will change the framework of our existence and the categories of our thinking, and so invalidate the past as a guide to the future.

Let me give one example to illustrate what I mean. Looking back now on the Russian Revolution, it is easy for the social scientist and the historian to treat the conquest of power by the Bolsheviks as something “inevitable.” With the hindsight of history we see the process which led from the Kerensky revolution to the establishment of the Stalinite dictatorship as a chain of cause and effect. The present seems to determine and so to explain the past. But this is the optical illusion of hindsight. As Trotsky himself pointed out in a famous passage in his History, without the genius of Lenin the Bolsheviks could never have succeeded. Here was one of those completely unpredictable mutations in human history caused by a man of genius. A sociologist who studied the Russian situation in 1917, and tried to extrapolate the past into the future, could have predicted at least fifteen possible developments of the Kerensky revolution, and assessed their varying degrees of probability. But the last thing which any reasonable and well-informed person could have foretold was what actually happened. There was nothing inevitable about Lenin’s success, and, since there was nothing inevitable, there was nothing predictable by social science.

_____________

Let us now turn from a study of philosophical presuppositions to an examination of the special techniques which the directors of the Hoover Institute Studies have invented in order to examine the Revolution of Our Time. All these booklets lay special emphasis on statistical sampling methods, and three of them are exclusively concerned with the study of elites. “The elite concept,” they write, “fills a blank in the language of science and policy.” Again, “Changes in the elite structure are indexes of revolution.” And again, “Elite studies are the heart of research into the world revolution of our time, since an essential criterion of revolution is change in elite structure.” And, finally, revolution is defined as “rapid and extensive change in the composition and the vocabulary of the ruling few.” This hypothesis was first adopted by Pareto in his famous theory of the circulation of elites, and popularized by James Burnham in his Managerial Revolution. It is a natural reaction against Marx’s economic determinism to assume that the leadership causes the revolution instead of being thrown up by its class forces; and this reaction is justified to a large extent by the growth of the power of the state during the 20th century. At a time when control of the instruments of coercion and persuasion can give total power, it is futile to describe the state as “the executive committee of the bourgeoisie,” and to suggest that power is the monopoly of one social class. The theory of elites is, from this point of view, merely the return to the traditional view that the science of politics is the study of power.

The novelty of Pareto’s work—further developed in the Hoover Institute Studies—is the hypothesis that the inner law of revolution is to be discovered by studying the social composition of ruling groups or elites. Instead of observing the aims and purposes of a regime, or analyzing the economic forces behind it, we should concentrate on a study of its social structure: who are its rulers, from what classes are they drawn, what are their occupations, what wealth do they possess? The obvious convenience for the social scientist of this hypothesis is that it permits him to apply his favorite techniques to the study of a revolution. The social structure of an elite can be quantified by scientific sampling: we can say what percentage of the American, or Russian, or Nazi elite were workers or engineers. But this advantage is outweighed by the distortion of fact which is introduced. To begin with, the tacit and illegitimate assumption has to be made that in all modern states (in the United States and Britain, as well as in the USSR and Nazi Germany) there exists an elite to be sampled and quantified. An essentially totalitarian concept is applied universally to all communities with a distortion as baleful as that caused by the Marxian doctrine of class. Just who, in America or Britain, is to be defined as “elite”? If social scientists would substitute “ruling class” for “elite,” this distortion would be too obvious to be overlooked.

_____________

But let us see if Professor Lasswell’s bold assertions can be justified by studies of particular revolutions. So far, we have a study of the Nazi elite by Mr. Lerner, and of the Politburo by Mr. Schueller. Can we claim that either of these two examples of “elite study” adds to our understanding of the “revolution of our time”?

Mr. Lerner candidly explains to us his method of research in his opening pages. He has taken the Nazi Fuehrerlexikon, the Who’s Who of Nazism, published in 1934, and subjected the 1,600 biographies it contains to various kinds of statistical analysis. In certain cases he has compared the results he obtains with comparable data on the membership of the whole Nazi party and of the whole German population. On the basis of this he provides us with eighty-four statistical tables.

Before commenting on Mr. Lerner’s conclusions, I must, in fairness, quote a judgment of Professor Lasswell on the use of the statistical method. “It is sometimes falsely assumed that statistical modes of description and correlation are supposed to eliminate other ways of characterizing current or historic happenings, such as the interpretative essay. We have suggested that this assumption is mistaken. . . . It is not a question of eliminating significance but of enlarging relevance.” This is a perfectly fair test, but I fear that if we apply it to Mr. Lerner, he fails abjectly. For one reader, at least, he did nothing to “enlarge relevance.” I could find no single result of his computations which could not be learned from any competent history of National Socialism. Indeed, there is scarcely one of his eighty-four tables which is intelligible without an interpretative essay that would, at the same time, render it unnecessary. On page 28 Mr. Lerner solemnly writes: “Professional intellectuals live, occupationally, by their output of symbols in spoken and written form. Since we are here interested in demonstrating only the degree of professionalism as intellectuals, not the quality of thought, we adduce as a relevant indicator the amount of printed symbolic output among propagandists and administrators.”

And this is followed by:

Table 21. Number of Publications

We really do not need an elaborate statistical sample in order to discover that Nazi propagandists published more books and articles than Nazi administrators. Table 56, to take another example, proves to us that German army officers listed in the Fuehrerlexikon are likely to be better born than German policemen, a fact which any student of German history must be aware of. Fortunately, Mr. Lerner knows a lot about contemporary Germany and, in between his statistical tables, he inserts some paragraphs of interpretative essay. These paragraphs would have been all the better if he had dispensed with his samples altogether and accepted as his starting point that the character of Adolf Hitler had a lot more to do with the Nazi revolution than the backgrounds of all the “little Hitlers” who formed his elite.

In The Politburo, on the other hand, Mr. Schueller has provided us with a useful source book. But the reason why it is useful is that it is not perceptibly “scientific.” Since there have been only twenty-seven members of the Politburo since it began, the sampling method is obviously inappropriate. What we are therefore given are quite useful canned biographies of these twenty-seven men, from which Mr. Schueller is able to draw some quite interesting conclusions. In particular, he contrasts the characteristics of the Old Bolsheviks, who carried out the revolution, with those of the new type of Stalinite administrator now taking their place. The Politburo is a solid, rather dully written notebook, such as any old-fashioned historian might put together for his private use before he got down to the serious job of writing a book. I have a horrible suspicion that Mr. Schueller feels that he would not have been a really scientific social scientist if, after roughing this material out, he had put it under his desk and composed a historical essay. This is a pity. A furnished house is really more useful than some scaffolding, bricks, and mortar, along with a pile of fitments and furniture. A historical essay is none the worse because the research that has to precede it has been tidied away for the pleasure of the reader. Certainly no historian would end with the sentence Mr. Schueller has selected to conclude his study: “In general, it may be said that while there are some similarities between the leadership of Soviet Russia and that of the Western world, the dissimilarities are much greater and more important.”

Would Professor Lasswell seriously maintain that this portentous platitude “enlarges relevance”?

_____________

I Must regretfully conclude that the whole attempt to explain the German and Russian revolutions by elite sampling seems to me to be a lapse into pseudo-science more ridiculous than anything perpetrated by Marx’s successors or imitators. Marx himself provided for at least two generations a magnificent, if distorting, framework for social studies. His economic analysis of history was brilliantly relevant to the age of expanding capitalism in which he lived; and, if it now blinds the student more than it illuminates, that is not surprising, considering the date of Das Katpital. I can foresee no such future for the Hoover Studies.

In the first place, the assumption on which they are based—that the background of those who wield power is a determining factor in revolution—is a mere variant of vulgar Marxism and suffers from the same defect, that it neglects the factor of individual genius and human will power. In the second place, the method of sampling seems to me singularly ill-suited to an inquiry into totalitarian elites, whose members, by definition, are likely to be persons of exceptional, if unpleasant, character. The sheep in any state can perhaps be studied in the mass and their behavior quantitatively assessed. Surely the shepherds need individual observation.

Strange, that a group of Americans, sincerely concerned to defend the dignity of the individual, should believe that we can learn something relevant about the nature of world revolution by mechanically submitting the canned biographies of Who’s Who to statistical analysis. That was not the sense in which our epoch was once called “the century of the common man.”

_____________