A journey implies a starting point, which, I should add, is much easier to describe when it is not over. Mine began when I returned to civilian life and Princeton University in the spring of 1946, a freshly minted liberal-leftist veteran of World War II (I had grown up in comfortable Republican surroundings) who graduated with a major in political science and then pursued a Master's degree in sociology at Columbia University with Robert S. Lynd who (with his wife Helen) had written the classic Middletown studies. Attracted to a kind of shapeless Marxism that rejected any attempt to marry capitalism to democracy, I was especially drawn to Lynd's central thesis that in a capitalist society “real” power is located in its economic spinal column, and that to view political power as independent of economic power was to be entrapped by a convenient fiction.

It is said that an ideologue is someone with a comprehensive vision of society as a whole with pretensions to consistency. If this is a fair (albeit incomplete) characterization, then I would have to describe myself—when I arrived at the University of California at Berkeley in 1949 to begin my Ph.D. work in political science—as one who had an ideological vision limited by an inability to recognize that there was more to ideology than economics. I had become, in two words, a “terrible simplifier.” Little did I know that my education would take a new turn and start all over again.

Nostalgia, as we know, is not all it's cracked up to be. But the four years I spent in the department of political science at Berkeley more than half a century ago remain among the more rewarding of my academic life. Everything, of course, was very different then. Berkeley itself, though always vibrant and alive, was a pale shadow of the “free republic” it would become. With the exception of the demonstrations in 1950–1951 protesting the loyalty oath imposed on the faculty, the campus for the most part was relatively quiet and peaceful. The political science department (at least by today's standards) was small, with a faculty of around 25 and perhaps a dozen or two Ph.D. students who paid $35 per semester ($150 if they were out-of-state). In many respects we were a “community” anchored in old South Hall, from which the department long ago moved as the faculty grew to its present size of approximately 50 members and the number of Ph.D. students total over 100, virtually all of whom (I note enviably) receive some sort of financial stipend.

Inasmuch as my primary field of concentration was American Government and Politics, I took several of the traditional courses—“Political Parties” with Joseph Harris, “Public Opinion” with Eric Belquist, and a seminar in “American Politics” with Peter Odegard. It was here that I was confronted with (for me) a new question: Is the American political system a valuable instrument of democratic government? In my work with Lynd, I had no interest in democracy as a purely political concept of making governmental decisions (I was concerned only with the “undemocracy” of capitalism). Now I was beginning to examine how decisions were made rather than exclusively their content. But Odegard's primary interest was in having us evaluate the claim that the American party system is the linchpin of democratic government. His own position was clear and unambiguous: The American party system, he argued, was inadequate to ensure responsible politics, the development of coherent governmental programs, and majority rule. While we were expected to have read earlier proposals for “party government” (e.g., Woodrow Wilson's Congressional Government (1913) and E. E. Schattschneider's Party Government (1942), we discussed at length the strongest and most recent case for the “model” of party government, Toward A More Responsible Two-Party System, authored by the Committee on Political Parties of the American Political Science Association.

I found myself in disagreement with Odegard. The more I read about the ideal of party government, the less it seemed to fit the American political scene. I knew that prominent political scientists had long accepted an idealized version of the British two-party system as a model for democratic government. But I also knew, as professor Norton Long, a dissenter in the ranks, observed, it is at once the revered example of party government and the outstanding exception to its general practice throughout the world. I remember giving an oral report in which I outlined the theoretical foundations of my opposition to party government, including the belief that our political parties arose from and reflected the cultural, social, and legal context of the society in which they operate and that it was a mistake to conceive of party government and responsibility as simply a matter of organization. I thought I had made a persuasive argument, but after much debate it was voted down by a majority of my seminar colleagues. Odegard was pleased with the vigorous discussion—and the vote.

I do not know precisely when the Ph.D. requirements in political science at Berkeley were revised, but in the late 1940s and early 1950s one had to offer five separate fields (for which there were five all-day written examinations), two foreign languages, and an outside minor (in my case, political sociology). Over a period of three years I took courses and seminars in comparative governments, public administration (as it was still called then), international law, and political theory. Although there were no so-called behaviorists in the department—they would come later—we all took a seminar in “The Scope and Methods of Political Science” in which we studied (in the language of the catalog description) “methods to the development of a science of politics.” The focus was on which kinds of political science questions were amenable to the touch of “scientific” analysis. This, of course, is an old debate that has not been resolved to everyone's satisfaction by the “behavioral revolution” in the social sciences, notwithstanding the fact that some of its earliest disciples seemed to think it would be.

More than 50 years later I look back on my Ph.D. studies as a series of steps in trying to construct a framework within which I could judge the value of a democratic political system both empirically (of which more later) and normatively. It is an effort that continues to this day and has strong roots in my work in political theory with Norman Jacobson, one of the most gifted and thought-provoking teachers I encountered at Berkeley, who made me realize the value of political ideas as a serious contribution not only to my interest in American politics but to more disciplined thinking about how political and moral principals merge into each other (and when they need to be kept separate). For example, I had read Plato's Republic as an undergraduate, but it was Jacobson who introduced me to Plato's conception of the state as an organism, a metaphor for a living creature in the sense that the individual cannot possibly be self-sufficient and can only become a full moral being within the context of the state. This organic theory of the community and the individual's place in it is in direct conflict with America's idea of individual equality as a matter of principle, derived from a “consent theory” assuring that the individual does not surrender to the state his or her judgment of what is right and wrong. I was now discovering, in effect, the philosophical grounds on which to reach conclusions about the nature of the state and about the rights of individuals.

In Jacobson's two-semester graduate seminar in American Political Thought (a course not commonly offered today), I became interested in the Transcendental movement, whose leaders represented an attack on the values of rationalism and skepticism that had grown up in the Age of Reason and who favored a return to democracy seen as “faith.” Thus Emerson discarded politics and said men must turn to God, who dwells in each individual soul. Democracy to Emerson was not majority rule, but rather the ability of each individual to judge, always and everywhere, on the basis of his or her own individual conscience. The true majority is the majority of one—he who has seen God.

I was troubled by the metaphysical notion expressed by both Emerson and Henry David Thoreau that each person's soul and conscience are the only true test of what is politically right and wrong and the only valid guide to political action. Having transformed conscience into an inviolable political principal, Thoreau could draw conclusions as opposite as day and night. He could champion the cause of political quietism by withdrawing to Walden Pond, or lead the call for violent action in defense of Abolitionism. In short, anyone who believes that moral justification is simply a matter of conscience can easily justify both passive resistance and civil disobedience and, for that matter, almost anything at all.

I had made an important discovery, one that would shape my way of thinking in the years to come. Like many dedicated moralists, Thoreau was perfectly willing to sacrifice the democratic process before his own special gods. Acting on moral precepts which, to him, were always self-evident, he struck out in whatever direction they led him. Each man, he felt, could determine for himself what is right and just. Thoreau was only dimly aware that his “transcendental individualism” could have little practical application and that, when translated into politics, individual conscience was not always a sufficient guide. As he commented during the early “unpolitical” stage of his life, “I came into the world not chiefly to make this a good place to live in, but to live in it, be it good or bad.”

Although concentrating in political science, my years at Berkeley also included courses and seminars in political sociology with Seymour Martin Lipset and in sociological theory with Reinhold Bendix. These seminars led me to examine various developments in the social sciences that changed my concept of politics and the role it plays in society. It was my work with Bendix, for example, that first introduced me to the place of self-interest in democratic social and political life. His major point was presented in the form of a proposition: When a person emphasizes self-interest as a central value and factor in political, economic, and social matters, he automatically commits himself to the view that uncertainty and deliberation play an important part in social and political action. Self-interest, which an individual was believed to be able to judge for himself (based, in turn, on a belief in human reason and perfectibility) was seen as an important determinant of individual conduct.

After an extensive discussion of various theories of self-interest from Adam Smith to Rousseau, Bendix examined how the concept of self-interest and the potentiality of the individual's ability to reason has been severely challenged in our time, most notably by many of the disciples of Marx who reduced self-interest to class interest. This meant (to engage in oversimplification) that the individual is less important than the class or group to which he belongs and that his ideas cannot be studied apart from his actions and what they reveal about his group or class affiliations. (Bendix insisted this was a vulgarization of Marx himself, who, in the tradition of the Age of Enlightenment, was concerned with the individual and his perfectibility.)

It did not take me long to realize how the theoretical implications of an emphasis on self-interest called into question my determinist assumptions about the dominating control of economic over political power in our society. As I have indicated, I was taking graduate courses and seminars in American politics, where we were studying Congress and the legislative process, how our political parties operate, why people vote the way they do, and other aspects of political life. However, I was now also interested in a different kind of question: Does politics itself command justification for its study as a legitimate subject of inquiry and consideration? I began to see that it does, but for reasons I had not appreciated as an undergraduate or in my work with Lynd. I was approaching politics in a new way—that is, of recognizing that political decisions are the end result of con-flicts and strategies that define politics as this is generally understood in democratic societies. But this would be true only if politics is not treated as an epiphenomenonon, or, to put it more succinctly, if men's ideas and actions are not thought to be determined, for example (in the Marxist view), simply by their position in the class system. Politics, in short, could be separated from economics in a capitalist system not because economics is unrelated to people's lives, but because the economic piper does not always call the political tune.

By this route I was back again to Bendix's initial assumption, namely, that self-interest as a concept (and a commitment) has many important implications, but not because it implies that self-interest alone can or should explain human conduct. That would overshoot the mark. Self-interest is useful, rather, because it designates all kinds of conduct as a “maximizing” and “calculating” effort by the individual irrespective of the efficiency or success of that effort. Take the case of a black woman physician employed by a state Department of Health. As a doctor she has views that impel her in one direction. As a woman, she may react in another way. As an African American she has still another set of reactions, and as a government employee perhaps still another. However, in spite of the strains and possible dilemmas she may face, she will pursue her own self-interest as she interprets and calculates it, always seeking ways to maximize the advantages of her position.

That, in a word, is politics. Not only did I no longer regard politics as largely subterfuge or superstructure, but I was now beginning to see that what I had once dismissed as a sham game of compromise and vacillation was the very process by which a democracy resolves its problems of conflict and integration with a minimum of coercion and a maximum amount of agreement on shared values.

But as I continued to question who has the controlling power in America— or, to put the question more concretely, “Who prevails in the decision-making that determines policy outcomes?”—I was distancing myself even further from Lynd's assertion that the business class was the dominant force in our society. However, I was pursuing this line of inquiry for a very different reason. I was now interested in seeing if Lynd's claims of who wielded power (as well as the claims of C. Wright Mills in his The Power Elite and of others studying “community power” from a similar perspective) could be substantiated by meeting certain empirical tests. This was a new endeavor for me. Not too long ago I had accepted assumptions about business power (I recall not only Lynd's work but Robert Brady's Business As A System Of Power) as valid, without ever considering if they were untested or testable assumptions. Now I was asking if these assertions were, in fact, verifiable. Did they refer to events in the real world? Were they supported by corroborating evidence—and was it also possible to adduce discorroborating evidence to test the correctness of these assertions?

The first person to introduce me to this kind of thinking and discourse was Seymour Martin Lipset, who was just beginning his distinguished career and all of whose classes and seminar I took while at Berkeley. It was from Lipset I learned that the purpose of good analysis is not simply to present arguments for or against a given proposition but to indicate the procedures for testing the proposition or hypothesis without prejudging it one way or another. There may be several such procedures and they may give dissimilar results, in which case a competent observer will be led to other procedures to inquire further into the matter under investigation. This sort of activity involved a process of constant inquiry, repeated checks and counter-checks—in short, a quest for truth independent of one's preconceived notions. If the goal would never be completely attainable, at the very least the methodology was likely to produce more convincing results.

In my work with Lipset, along with reading the pluralist literature on group politics (e.g., Arthur Bentley's The Process of Government and David Truman's The Governmental Process), I was able to re-enforce my understanding that politics involved much more than class, wealth, and social standing or one's position in the social structure. I never doubted that the pattern of social stratification was of crucial importance. But I was now asking a more fundamental question: Do general theories of social stratification throw much light on the forces that determine the exercise of political power? It was becoming increasingly clear to me that the answer was no, essentially because these theories fail to grasp not only the indeterminate relationship between class position and class power, but when or whether political factors will influence the major forces in the struggle for power.

At one time, my way of looking at power was to focus on the social background of the power holders, an approach based in part on the belief that actions flow directly and logically from the narrowly defined self-interests of those who have power. Presumably, for example, if one knows which groups those with power belong to, then one would automatically know which groups will benefit from their decisions. But Lipset offered a different approach that stressed access to power. It assumed that the actions and decisions of those in power are determined by a complex calculation of what might be the consequences of their decisions— and to the extent that the predictable reaction of any group or individual to a decision will affect the results of that decision, the group or individual has access to the decision-making process. Seen in this way, it is possible for the composition of the decision-makers to remain fairly constant, yet for the power structure of a society to change when the groups having access to power change. Clearly the increase in power of organized labor that began in the mid-1930s did not mean that business could no longer play a major role in the power system. Finding it now more useful to treat power in terms of who has access, it was no longer possible for me to accept the “scarcity theory” of power, which assumed that an increase in power for one group must necessarily occur at the expense of another.

I am frequently appalled at how little I knew when I arrived at UC Berkeley in 1949, an entering Ph.D. candidate with far more answers than questions who had yet to learn that “to doubt is the only way to approach anything worth believing in” (the words are those of the late scientist Edward Teller). Since then, the only constant has been change, including, of course, change in the field of political science. More personally, however, I have in mind the change in the way I now try to think about and analyze political questions or matters of public policy. I no longer, for example, limit my associations to people who agree with me. That may (or may not) work as an electoral strategy for politicians and political parties who believe they must first appeal to their core supporters and “true believers” and then reach out to the independents and the undecided. But it was my work at Berkeley that taught me how restricting and unrewarding it was simply to preach to the converted. At the level of ideas and quality of thought, I grew to prefer the intellectual nourishment that comes from testing my understanding against the countervailing understandings of others.

I have often been asked if, in retrospect, I would take my Ph.D. at UC Berkeley again. The answer has always been yes, especially when I remember that those were the years that fully awakened me to the pleasures and challenges that constitute the never-ending journey of the mind.