My Traverse through Columbia

I came to Columbia resolving to give the educational system one last chance. It had failed, I felt, through high school and the several colleges I had attended. My teachers had been engaged in transmitting information, but none (except for two at Purdue) had been interested in me, in what I might do with the information they had imparted.

From nearly the beginning at Columbia I felt a difference. I sensed that some faculty members had a personal (that is, selfish) interest in some of their students. They seemed to be interested in those students in a way I had never felt since the ninth grade: their interest seemed tinged with the interest that parents have in their children. If asked to explain that now, I would say it is because, as children do for parents, graduate students help bring professors closer to immortality, extending their influence beyond their own life span. In the then large sociology department at Columbia this interest was in conflict with another, the faculty's desire to protect itself from graduate-student demands, so that for many of my fellow students graduate school did not have this special—perhaps essential—quality. For me, from nearly the beginning, it did.

I moved into an apartment at 111th Street and Broadway and sold my 1947 Chevrolet to a couple of Puerto Ricans, hoping they would not notice an ominous sound in the differential. I entered Columbia still a naive boy from Seven Mile, Ohio, and Herndon, Kentucky, with training as a chemical engineer and work experience as a chemist. I was not, however, in great awe of sociology. I had read enough by that time to have discovered that the density of ideas was not comparable to that in a book on physical chemistry. I wanted to progress rapidly, to skip background courses and proceed with the more advanced ones. As for statistics, I had had no courses but had read Quinn McNemar's statistical text over the summer so that I could enter the regular course on sociological statistics without preliminaries. But I decided after the first session to opt out of the course and take my statistics along with the statistics students in the mathematical statistics department. (It was the first of several courses I took in that department. My mathematical background from chemical engineering was strong enough to enable me to take courses in statistics that my sociology colleagues could not.)

As I have already indicated, Merton's theory course was riveting, a conversion experience for those of us eager for conversion. But my interest in coursework was largely confined to Merton's course and to the mathematical statistics courses. I did learn something about methods from Herbert Hyman and Charles Wright's research methods sequence. But with rare exceptions I have never found lecture courses particularly appealing as a medium for learning. Ernest Nagel's course in philosophy of science, which I sat in on, unregistered, was one such exception. The class always continued for ten or twenty minutes after its time with a group of students clustered around him, asking questions and arguing points. I stood on the periphery, verbally inadequate and unsure of myself, listening to the interchange and wanting to interrupt but remaining mute.

Seminars were another matter altogether. Some of the seminars were run by two faculty members, and the interchange was far more instructive than the lectures. In seminars we were induced to perform, to write and present papers, to show what we could do. The Lazarsfeld-Merton seminar in bureaucracy had both these components, as did the Lazarsfeld-Nagel seminar in mathematical sociology.

I sat in on a number of courses and seminars outside the department, stimulated by the extraordinary range of intellectual activity. One was Abram Kardiner's seminar, which I attended while I was reading his

Psychological Frontiers of Society; a second was Greenberg's course on linguistics; and a third was George Stigler's course in microeconomics. In an informal course on operations research I found an interesting mixture of statistics, mathematics, and an engineer's approach to problem solving. And I took a course in industrial engineering from Albert Rubenstein in which we engaged in a kind of sociology of formal organizations with an emphasis on patterns of communication.

So much for coursework. In my second semester I learned that Lipset and Trow were looking for interviewers for the ITU project, which was just going into the field. I applied for a job interviewing printers and was hired. I learned New York this way, riding the subways to the World-Telegram, the Sun, the Times, the Herald Tribune, the Daily News, the job shops, and printers' homes in Queens, Brooklyn, and the Bronx. I also learned interviewing. When the officials of the union were to be interviewed, Trow kept Joseph Greenblum and me on, and the three of us did those interviews. Then I helped in sending out mail questionnaires to get panel data on New York printers' voting preferences in the international union election.

That summer Lipset went to teach at the Free University in Berlin, and Trow went to help C. Wright Mills build a cabin on an island in one of the Great Lakes. The data were in, and they left me alone with it. At the end of the summer I had written up an analysis that I could show to them. Lipset read it, then asked me if I would like to be a coauthor of the book they were writing on the ITU. I said yes.

Thus began an enormously productive and instructive experience. We talked, wrote, tore each others' drafts apart and rewrote them. Lipset provided the intellectual framework and background for the study in the first chapters, and from there we shaped the analysis in extended discussions. Trow and I spent hours on the counter-sorters in the Fayerweather basement, pored over cross-tabulations together and separately, and in our extended arguments over data analysis, we taught each other. There were no long lines of students to curtail our interaction with Lipset. Nor was it all data analysis. We reread Michels, we read and argued over Mannheim, the mass-society theorists, Scheler, Emil Lederer, Hannah Arendt, and Selznick's The Organizational Weapon . I have coauthored books, but the interaction during the academic years 1952–53 and 1953–54 was unlike any other.

In the fall of 1953 Lazarsfeld asked me if I would work, for pay, on the panel project. I had to decide between that and continuing work without pay on the ITU project. I had a tuition fellowship and could

manage without the extra money, so I said no and stayed with Lipset and Trow.

Lazarsfeld had three large projects, the Behavioral Models Project, the Panel Project, and the Concepts and Indices Project. In all three I was involved with Lazarsfeld but never was employed by him. Following the pattern of the Rashevsky paper, he paid me for products, not hours. One might say I worked as an independent contractor. It may have been this distance that made me less bitter at Lazarsfeld than those who had been closer. I stayed on the periphery of the Behavioral Models Project, never employed by it but attending seminars and, in the end, writing a report for the project on mathematical models of small groups. For the Concepts and Indices Project I wrote papers analyzing economists' use of concepts and analyzing economic indices (learning along the way more economics), and gained admission to the summer seminar at Dartmouth on concepts and indices. I was greatly interested in methods of panel analysis and wrote a paper attempting to extend those methods (which led some years later to my Introduction to Mathematical Sociology ). That paper got me into another Dartmouth summer seminar, on methods of panel analysis. These summer seminars were extraordinarily important to my socialization in sociology, for I watched social scientists in action, debating points, arguing methods, developing ideas. I began to see myself on the same plane, emerging as a professional among professionals.

After two and a half years at Columbia I had reached an apex of status, prestige, and deference that I have never approached in my subsequent career. I then took my oral qualifying examination for admission to Ph.D. candidacy. I had prepared for the exam over the summer and had a rich and well-organized set of ideas and information about social theory and theorists. I remember the first question of the examination, a statistical question asked by Howard Raiffa, which I bungled miserably. I remember nothing else except that when it was over after two hours, I knew I had failed. I waited outside the examination room; Lipset summarized the examination when he came out and said to me, "If we all had not known before the exam that you would pass, you would have failed." The others were equally grim-faced as they left the examination room.

I trudged disconsolately down Amsterdam Avenue toward my 123rd Street apartment but was hailed by shouts from Sidney Morgenbesser and others from an office window across Amsterdam, a Bureau office I shared with Morgenbesser, with whom I was working on Lazarsfeld's Concepts and Indices Project. My friends had prepared a party in celebration of my passing my orals, but I had decided to skip it. They would

not let me go home, though, despite my agony. So I went to the party and shared my troubles with my friends. Soon the members of my examination committee, including Merton, came over to wish me well and have a drink in what was a muted, but relieved, celebration. The incident was, I am sure, not forgotten (and may have been responsible for my not getting the job teaching Lee Wiggins's statistics course a year and a half later), but it was put aside and perhaps attributed to my general oral ineptitude. I went on to other things—the ITU analysis, my small-group monograph for the Behavioral Models Project, and my paper on properties of groups for the Concepts and Indices Project.

I finished my dissertation in the fall of 1954 and submitted it in December. The defense was in January. But matters were not so simple. Students did not finish so quickly at Columbia. Certainly if I had done my dissertation under Merton, I would not have finished then. No more likely would have been a quick completion under Lazarsfeld. Even under Lipset alone, matters would not have been so simple. But in the spring of 1954, I met with Merton, and he asked me about my dissertation plans. I said I didn't know; I had none. He asked why not the ITU study? Did I have chapters that were largely my own and formed a coherent whole? I said I thought they did. So, armed with this proposal from Merton, I went to Lipset and reported the discussion. He agreed to the dissertation. He might have done so anyway, or he might not have—for he was a young assistant professor, and the ITU study was his first major work after his own dissertation. Then a graduate student proposes to carve a dissertation out of a portion of it. Even though I was the principal author of the chapters I proposed to use, and even though many of the ideas in them were mine, they nevertheless had resulted from long discussions among the three of us and extended critiques from Trow and Lipset. Because they were intended from the start as potential chapters of the book rather than as chapters of my dissertation, Lipset gave them far more attention than dissertation chapters ordinarily receive.

The waters of Columbia were difficult to navigate, at least with any speed. It helped to have a sponsor interested in one's well-being. But it helped more to have two. Still more did it help to have three. Lazarsfeld was my mentor, but I did not do my dissertation under him and was not beholden to him. Lipset was the faculty member under whom I worked most continuously, but the structure of Columbia demanded that he be attentive to Merton's or Lazarfeld's wishes. Merton was less close than either of the others to me but was ready and able to serve as my protector.

I finished my dissertation quickly in the fall of 1954, making the necessary modifications to form dissertation chapters from what had been book chapters and adding one analytical chapter for good measure. But I needed work that fall to support myself. I was finished with the department, and it was finished with me. The Bureau would be, if anything was to be, my home for the academic year 1954–55. Charles Glock rescued me by offering a monograph on community conflicts from the Twentieth Century Fund, a task someone else had botched. I worked on that project, producing Community Conflict, which the fund did not like very much but nevertheless was published by Jeremiah Kaplan's Free Press. (Of everything I have written, this small, awkwardly published monograph has probably sold the greatest number of copies.)

In the winter, in part as an outgrowth of the traditions seminar, a drug company, Pfizer, gave a grant to study the way doctors introduce new drugs into their practices. That project occupied me during the spring (though I wrote most of my portion of the report during one tense, and intense, week in which my wife and I were for other reasons not speaking). Meanwhile, it was not clear what I would do the next year. I had no job offers. Herbert Simon at Carnegie Tech had offered me a job the previous year, but Lazarsfeld had induced me to remain at Columbia. Now there was nothing. Lee Wiggins was to be gone (at the Center for Advanced Study in the Behavioral Sciences—its second year), and I hoped to teach his statistics course, a part-time job that would be enough for support. The department, however, decided I was not a sufficiently able teacher.

My time at Columbia was approaching an end. One day I stopped to see Lipset to discuss details of the Union Democracy manuscript, and he introduced me to his new research assistant, Immanuel Wallerstein. He was a bright, energetic, and self-confident member of a new cohort of graduate students. Through his eyes I saw myself no longer as a member of the elite among graduate students but as a former graduate student, outside that system, in limbo, Maurice Stein of 1955.

My destination for the next year was still unclear. I was preparing myself for another year at the Bureau, with hopes of getting funding for my new project, a study of adolescents in high schools. Then, at the beginning of the summer, I received an invitation from the Center for Advanced Study (occasioned by the last-minute defection of a more deserving scholar but also due in part to Lazarsfeld's and Merton's influence). Late that summer I set off for California with a two-week stop along the way at Carnegie Tech in Pittsburgh, at Herb Simon's

invitation. (My wife and son had left earlier, stopping to visit her family in Indiana.) I packed the station wagon with all my belongings and stayed overnight at Philip Ennis's apartment on Ninety-sixth Street. The next morning I found my car plundered; the clothes I had left in the front seat were gone. I didn't much care. I was a journeyman sociologist, off to Pittsburgh and to the Center.