Where Graduate Schools Fail

THE Atlantic

FOUNDED IN 1857

BY CHRISTOPHER JENCKS AND DAVID RIESMAN

IT HAS become a truism to say that the university is one of the major shapers of American culture. The bulk of the American intelligentsia now depends on universities for a livelihood, and virtually every would-be member of the upper middle class thinks he needs some university’s imprimatur, at least in the form of a B.A. and preferably in the form of a graduate professional degree as well. But while everyone now recognizes the central role of the university, not everyone is aware that within the American university the centers of influence are the graduate schools, especially those of arts and sciences.

The importance of the graduate schools is twofold. First, they have a virtual monopoly on permanent university appointments. Many people pass through the academic community in various roles, but almost nobody gets a permanent appointment or a voice in choosing other members of the community unless some academic department judges him competent to train graduate students. In addition, and perhaps even more important, almost nobody is allowed to become a teacher of American undergraduates, even at a nominally independent four-year college, unless he has been awarded a Ph.D. or at least an M.A. by a graduate department somewhere. Moreover, professors habitually try to remake the college where they teach in the image of the graduate school that trained them. The typical liberal arts undergraduate major is therefore more and more a mini-Ph.D. program, whose adequacy is more or less consciously judged by the success of its alumni in the nation’s leading graduate departments.

Considering how much criticism has been directed at this system of education, and considering how many of its problems can be traced to the graduate schools, one might expect them to be under continual attack. But this is not the case. Over the past decade a variety of government panels, task forces, and committees have met to discuss higher education. These meetings invariably begin from the premise that the graduate schools and the academic profession are in reasonably good shape, but that there are problems of quality control further down the academic line: in the undergraduate colleges, in the secondary schools, or wherever. The chief official concern about graduate education is quantitative, not qualitative: the supply of Ph.D.’s is less than the demand, it takes too long to get through, and so on.

There are several reasons for this complacency. First, whatever their present failings, the graduate schools are almost certainly better than they were in the past. They are also better than graduate schools in other countries; indeed, they are a mecca and a model for people all over the world. Perhaps most important, they are infinitely better off (though not necessarily more effective) than undergraduate colleges, secondary schools, or primary schools. They spend far more money per student than any lower-level institution. They pay higher faculty salaries, they give their faculty virtually complete freedom in deciding how to do their jobs, and the result is generally higher faculty morale than in most other schools. In a profession where talent and virtue are synonymous, they recruit and hold far more talented people than other educational institutions.

The fact that the graduate schools constitute a privileged elite, for which many are called and few are chosen, means that they must be judged by higher standards than other educational enterprises. How well is this elite doing its job?

There are two vantage points from which to answer this question. Either one views the graduate school as the center of American scholarship, research, and intellectual life, and judges its effect on this life; or one takes the graduate school as the training ground for American college teachers, and judges it in terms of their pedagogic competence. From either viewpoint the present state of graduate education leaves a good deal to be desired. Judging the alumni of American graduate schools as scholars, we would argue that they are impressive in their technical competence and their ability to do impersonal, objective, and quantitative work. These same men are not always equally impressive in areas where individual sensitivity is essential or where personal values determine the quality of their work. Judging the typical Ph.D. as a teacher — which about half of them are — we would argue that his professionalism as a scholar often alienates undergraduates, and that his lack of any specific training as a teacher or as a mentor of the young leads to unnecessarily poor performance in the classroom and to insensitivity in dealing with undergraduates.

THE growth of the graduate schools has been closely and causally related to the professionalization of the life of the mind. The more professionalized men become, the more they tend to work on questions which will “advance the discipline,” and the less they care about questions which might be of interest to politicians, intellectuals, businessmen, students, readers of the Atlantic, or laymen generally.

The main curb on professionalization of this kind is that laymen still have ultimate control over university budgets, and are much more willing to make money (especially federal money) available for work on problems of potential practical importance.

Yet lay pressure of this kind can hardly be said to have squeezed out “pure” research, directed at questions of interest only to academicians. On the contrary, such research flourishes as never before in history. Still, it is only when one turns from the nature of the questions men ask to the methods they use to answer them that the triumph of professionalism is beyond question. The Defense Department, for example, makes money selectively available to university scientists working in areas it deems important, and in this way has a considerable longterm effect on what gets investigated and what gets neglected. But it allows investigators almost complete freedom in deciding how to do their work, and this means that most men choose methods devised and respected by their colleagues.

An important effect of this professionalization has been the academicians’ attempt to transcend individual subjective experience and find “objective” truths which all members of a given discipline can accept. Graduate schools train students to judge the importance of a particular piece of work by its effect on mature professionals, not by its relevance to their own intellectual, personal, or political commitments. Some disciplines have enough collective wisdom to justify this abdication of individual judgment and public responsibility, but many do not. Very few disciplines in the social sciences or the humanities can claim an agenda which is reliably constructive or cumulative. The researcher who lets his colleagues shape his research interests is therefore in serious danger of devoting himself to problems of no consequence to anyone. Almost all the really distinguished social scientists and humanists of the past century have resisted such pressure and have worked on questions which they felt had genuine significance for their own personal lives, for laymen outside the university, or for both.

In the humanities, the tension between the objective and the subjective often takes the form of conflict between scholars and practitioners. Most English professors, for example, are quite defensive vis-à-vis writers and poets. They try to establish themselves as a different breed with different standards and objectives. Reading poetry and writing it are said to be totally different activities. But while reading poetry is not the same thing as writing it, and the same men will not reach the top in these two fields, the two activities are related to each other. A graduate training program for literary scholars ought to recognize this and get the prospective scholar to try his hand at the art which he will be criticizing. Many would not perform very well, but that is not the point. Nobody should get a Ph.D. in English who has not tried to write a sonnet.

The problems of connecting theory to practice and professional objectivity to personal subjectivity are as central in the social sciences as in the humanities. Today a man can become a political scientist without ever having engaged in political activity of any sort. In principle, this is entirely reasonable, for there are shy and reclusive men whose imaginations can grasp and decipher events they dare not encounter face to face. But the problem is to widen and multiply the roads toward graduate certification instead of narrowing them, and to encourage students to go beyond their well-defined limitations. Very few of today’s Ph.D. programs do this. In an earlier era, when graduate training was less well financed by fellowships and research assistantships, many would-be sociologists had to support themselves by sporadic schoolteaching, social work, professional interviewing, and so on. Now they can move directly from college through graduate school, doing their thesis on someone else’s data, and can avoid almost all contact with nonacademic people. Similarly, we know of no psychology department that requires its Ph.D. candidates to undergo self-analysis, Freudian or otherwise; on the contrary, many nonclinical psychologists, unable to distinguish between sentimentality and sensibility, regard an interest in people as a liability in a prospective psychologist (especially a woman). In anthropology, “field work” in an alien culture is still regarded as having educational value above and beyond the data collected, but few other disciplines have comparable requirements. This has remained true despite the increasing availability of secondhand data, and is a tribute to the anthropologists’ awareness that socialization of apprentices depends on what they have done as well as what they have read.

The graduate schools are full of gifted students dying of boredom, primarily because they have not been able to link up the professional skills they are supposed to acquire with any genuine personal or social commitment. Yet any attempt in this direction quickly confronts the problem of requirements. If an English department requires Anglo-Saxon and merely permits the writing of poetry, or if the political science department requires a course in theory but merely permits active involvement in politics, students quickly get the message that Anglo-Saxon or political theory is part of their professional identity, while poetry or political activism is marginal. If, on the other hand, departments were to require poetry or political involvement, this would also pose problems.

The present system of teaching assistantships, for example, does not give the student the impression that his elders care whether he teaches well or badly, but at least it gives him time and a legitimate excuse for attempting to teach if he has an inner impulse to do so. Most other “activist” impulses do not get even this limited reinforcement. A sociology student cannot get credit for union-organizing in the South or for selling textbooks to school systems, even though either of these activities would usually teach him more than a regular course. He can, it is true, write a paper about the people he has observed. But what he has actually done with these “outsiders” is of no interest to his department and gets him no closer to a teaching job; he is rewarded only for what he can verbalize, not for what he has become.

Each new requirement is a sieve, eliminating a few more promising people from academic life, people who could, once in the field, contribute to the diversity of its potential aims. The problem here is seldom that the people eliminated cannot meet the requirements, but that they will not. Many requirements threaten or irritate prospective students, some of whom decide a Ph.D. is just not worth it. There are brilliant students who choose one graduate department over another, or even one field of study over another, because it requires no German, for example. This is not because they have serious language blocks —though a few do. But many students, even at twenty-two, are skittish about learning anything really new. They have to be led to this by indirection. If the requirement strikes them as irrelevant and silly, their anxiety can be turned into righteous anger, and their laziness into a virtue. The same thing can happen with other requirements, such as “field experience” of various kinds.

Under these circumstances it is tempting to urge that all requirements be abolished. Yet we fear this would encourage students to play only from strength. What is needed is a system sufficiently structured so that students must try to learn new things and master new skills, but sufficiently flexible so that those with real blocks know they can get through if they have other valuable skills.

In all of this the graduate schools ought to be trying to put their students in closer touch with themselves, to develop individual sensibility, and to help them find work relevant to their personal experience as well as to the preoccupations of their profession. But few graduate schools are even trying to move in this direction. Most still think their primary problem is to ensure professionalism and train men to do replicable research.

THE graduate schools are not only hostile to nonacademic modes of learning; they are also too rigid in their grouping of academic skills and expertise. American scholarship has been noteworthy for its ability to cut across disciplinary boundaries and bring together men with different sorts of knowledge to work on a single problem. But the graduate schools have been conspicuously slow to follow this lead and allow students to look at problems rather than disciplines. A discipline is nothing more than an administrative category. The various subdisciplines within biology or history or psychology, for example, have only the most limited intellectual relationship to one another, and the same is true in every other field. They are grouped together mainly because the men working in them went through the same sort of graduate programs and have some residual feeling of common identity. Many research projects regroup subdisciplines in ways that cut across departmental lines, and many individual researchers become expert in subjects nominally outside their discipline. New journals are founded every day to fill the interstices between disciplines and encourage cross-disciplinary contact and fertilization. But the instructional program remains almost untouched. Faculty who want to teach subjects outside their department’s traditional boundaries often find this difficult, and graduate students who want to pursue a pattern of studies that does not fall under conventional departmental definitions are likely to run into trouble.

The issue here is not specialization versus generalization, but whether one way of aggregating specialized skills is better than another. Historians, for example, all know some historiography and a little philosophy, something about medieval Europe and something about pre-industrial America, a bit of French and a bit of German. This may not be a bad combination. But would it not be better if there were also men who, for instance, combined knowledge of pre-industrial American history with knowledge of demography or economic development or comparative government?

We need all sorts of combinations, and we get relatively few of them. Still, a large academic community needs some sort of formal subgroupings around which individuals can cluster. The departments fill this need; that is the source of their power. If they were abolished, something would have to be put in their place. Interdepartmental committees or groupings, such as the Committee on Human Development at Chicago, or the Department of Social Relations at Harvard, or a number of area studies programs, have the temporary advantage of starting off less hallowed by tradition and more subject to criticism and ad hoc modifications. But over the long haul there is no reason to think these new combinations have any significant advantage over the old ones; all are somewhat arbitrary, and all tend eventually to become houses of worship and power as well as work. Abolishing old departments and establishing new ones should, therefore, be a continuing process, like Jefferson’s revolution, every twenty years.

Whatever the character of the departments, however, there is a good deal to be said for curtailing their present powers and distributing them to smaller groups. The real unit of intellectual work is the subdiscipline, which usually has only one or two representatives on a small campus and seldom more than half a dozen even on a big one. These subdisciplines are evanescent, arising and disappearing over the lifetime of a single faculty generation. They fertilize one another in unpredictable ways, and the curriculum should be sufficiently flexible to accommodate such promiscuity. The most obvious way to do this is through ad hoc faculty committees, which set requirements for a particular student’s graduate work that are adapted to his special interests. Each student would be free to seek out faculty members with interests he judges relevant to his own, and if they approved his program, he would be free to pursue it.

Yet, just as a family can be more oppressive than a college, a relatively small ad hoc doctoral committee can sometimes be more oppressive than a larger and more impersonal department. In a department, every faculty member asks for limited fealty; he wants graduate students to take his course and answer his questions on the general examinations, but he cannot ask more because other faculty members are around to protect their students from “unreasonable” and “excessive” demands. Only one professor, the candidate’s thesis supervisor, is likely to ask for a more basic commitment. A committee system, on the other hand, largely eliminates the protections provided by departmental traditions and formal rules, putting the student entirely at the mercy of his chosen mentors. For this reason we would not urge the complete abolition of departmental programs, but only the creation of alternative routes, worked out on an individual basis between the candidate and a committee of his choice.

In addition, whenever a certain number of faculty members — perhaps five—decide there is a need for a new kind of doctoral program based on a new combination of specialties, they should be free to establish it on their own, setting whatever requirements they think appropriate for the students who enroll. Sociologists interested in professionalization and occupational subcultures might, for instance, join with economists interested in manpower problems and psychologists interested in vocational choice to establish a new program. Or anthropologists, political scientists, and economists concerned with economic development might unite to devise a new set of requirements for those who shared their interests.

Such a system would, of course, be anarchic, and subject to some abuse. It would also lead to the proliferation of courses and degrees. But then these same complaints could be made against the present individualistic system of research initiatives and support, which seems to be working considerably better than the graduate training system. Without some such assault on the regulatory powers of the collective faculty and of individual departments, the graduate curriculum is certain to remain as rigid as any other arrangement that must placate a multitude of competing interest groups.

Thus far we have been talking about simultaneous combinations of subdisciplines in a single doctoral program. It is also important, however, to facilitate sequential combinations, so that students can do their graduate work in a different field from the one in which they did their undergraduate work. This is less common than it should be, in part because undergraduates choose a field that appeals to them and then want to continue in it because it still interests them. But in part it is because they believe, or actually find, that they cannot get into a graduate department unless they have already done undergraduate work in the same area.

One of the most useful things universities, foundations, or the federal government could do to enliven graduate training — and also to free undergraduates from the shadow of their professional futures — would be to provide financial support and make academic arrangements for students who wanted an extra “transitional year” between completing their B.A.’s and entering a graduate department. A student can prepare himself for firstyear graduate work in almost any realm, from medicine to Far Eastern studies, in a single year. At present, however, this year is extremely hard to come by, for it does not fit any established program, and nobody (including parents) is particularly interested in paying for it.

THESE suggestions are obviously not a blueprint for revolution in the graduate schools. The problems are enormously difficult, the alternatives all filled with obvious and not-so-obvious pitfalls, and the results extremely difficult to measure or even judge impressionistically. Innovation must proceed on a case-by-case, place-by-place basis, drawing strength from the particular combination of scholars that happens to assemble itself on a particular campus at a particular moment. The general direction of change should be toward a more elastic mixture of theory and practice, demanding a wider range of skills from the student and rewarding a wider range of competencies. Graduate schools must allow and even ask their students to take more initiative, exercise more responsibility, and make more moral and political choices, both within academic contexts and, more important, outside them. Only if they are asked to do these things and judged accordingly will they come to believe that these activities are part of their set of professional roles. Judging graduate students’ moral and political responsibility is, of course, a dangerous business, and not one that graduate schools only recently freed from church control and Victorian standards will adopt readily. But that revolution has been largely won, and it is now time for the graduate schools to abandon the pretense of social and moral neutrality and to try to develop a responsible position in keeping with their contemporary role. Only if they see themselves and their profession in these terms will they set out to develop such skills among the next generation of undergraduates.

A more open, permissive, flexible system would carry with it many evident risks: it would usually be less satisfactory for the uncertain or listless student who really wants to be taken charge of; it could prolong the self-indulgence of those whose undergraduate training had been slipshod; and given the present relative scarcity of Ph.D.’s (in terms of institutions needing them for accreditation or pride), it could certify people with a call neither for teaching nor for scholarship or research. Still, a few major universities with restricted enrollment at the graduate level could experiment with different styles. Some could remain highly structured and encourage such traditional academic virtues as clarity of thought, skepticism, and precision. Others, however, ought to try putting more emphasis on other values: self-knowledge, compassion, and so forth.

We are convinced that those departments that put more responsibility on their individual faculty and students would attract many of the ablest students now coming out of college and many of the ablest young faculty. Most of these people now go faute de mieux to the big-name departments, for they cannot see significant differences between one place and another except in terms of the research reputations of their faculties.

TEACHING is not a profession in the way that research is. There is no guild within which successful teaching leads to greater prestige and influence than mediocre teaching, nor any professional training program that develops pedagogic skills in a systematic way. Very little is known about which teaching strategies work with which students. Under these circumstances it is hardly surprising that a great deal of teaching at both the graduate and undergraduate level is ineffective.

The reasons for this are apparent. It is easier to judge graduate students and assistant professors on the basis of papers they turn out than to judge them on the basis of their interaction with other people. A graduate student’s performance in comprehensive examinations is said to provide “hard” evidence of his competence, while visits to his classes provide only “soft” evidence. A professor’s book can be evaluated in “objective” terms, whereas his course syllabi, lectures, and examination questions can be valued only “subjectively.” The adjectives suggest the prejudices of the profession even though they throw little light on the actual criteria for judgment. These prejudices are not simply a matter of valuing “research” over “teaching.” A tenure committee member, for example, who argues for a particular man because of his catalytic qualities in helping his colleagues do better research also finds himself opposed by an ideology which insists men be judged entirely in terms of their own paper output rather than in terms of their effect on others.

In discussing these matters many critics talk about research “versus” teaching. We have found no evidence, however, that the two are antagonistic. Teachers cannot remain stimulating unless they also continue to learn, and while this learning may not focus on small, manageable “research problems,” it is “research” by any reasonable definition. When a teacher stops doing it, he begins to repeat himself and eventually loses touch with both the young and the world around him.

Research in this general sense does not, of course, necessarily lead to publication, but this is its most common result. Publication is the only way a man can communicate with a significant number of colleagues or other adults. Those who do not publish usually feel they have not learned anything worth communicating to adults. This means they have not learned much worth communicating to the young either. There are, of course, exceptions: men who keep learning but cannot bring themselves to write. Some have unrealistically high standards regarding what deserves publication. Some know no journal which is interested in the kinds of problems that interest them. Some are simply afraid of exposing themselves to their colleagues’ criticism, even though their ideas could in fact withstand such scrutiny. Some of these men are constantly learning, and some of them are brilliant talkers and teachers. Still, there are not many brilliant teachers who never publish.

There are, on the other hand, many good scholars who do a conspicuously and unnecessarily bad job in the classroom. They know that bad teaching is not penalized in any formal way. They have only a limited amount of time and energy, and they know that in terms of professional standing and personal advancement it makes more sense to throw these into research than teaching. This is particularly true in the early years of an academic career, not only because the feverish competition for talent puts an enormous premium on judging young men and giving them tenure early in order to recruit or hold them, but also because the young faculty member himself wants to establish his adequacy in a field before committing his life to it.

The real problem is to marry the two enterprises, research and teaching. Contrary to a good deal that is written by defenders of the status quo, this is precisely what the present system fails to do. Teaching is often adjusted to the exigencies of research, but research is almost never shaped by the experience of teaching. We have almost never encountered a professor, for example, who said he was working on a particular research problem because year after year his students showed an interest in it. Involved here is not only the understandable fear of one’s own showmanship or of sycophancy toward the young, but actual ignorance of what even the most sensitive undergraduates are interested in. Professors do listen to the questions their already socialized graduate students and postdoctoral fellows raise, but most undergraduate courses are so constructed as to provide almost no feedback directly to the professor. Even if they did, it would seldom occur to a professor that the questions raised by bright yet only half-socialized students might be important enough so that, if the answers are not known, an effort should be made to discover them.

THE relationship between graduate and undergraduate teaching is equally unsatisfactory. Many undergraduate courses at large universities are taught by one or two senior faculty and a group of graduate assistants. The issues raised in these courses are — or should be — of fundamental importance to graduate students as well as undergraduates. Yet we know no university where the staff of such courses habitually meets to discuss the intellectual questions being raised and initiate research in areas where it would be helpful. (Chicago is the closest approximation.) A serious effort along these lines would probably require constituting the staff of large undergraduate courses as a graduate seminar, to meet weekly for discussion of the books to be read by undergraduates and for the preparation and discussion of seminar papers that would also be delivered as course lectures.

Staff-taught courses that hold regular meetings to discuss the intellectual substance of a course can also discuss and evaluate its effect on the students. If professors are sufficiently secure so that they can encourage criticism of their performance, and if graduate students are open enough and interested enough to visit one another’s classes, the beginnings of a clinical training program are in hand.

Suppose a cadre of committed faculty and graduate students could be assembled and an effective clinical training program for undergraduate teachers worked out. How should such a program relate to the traditional doctoral program of lectures, seminars, reading courses, comprehensive examinations, and dissertations? Many critics of doctoral programs have argued over the years that while these programs do an adequate job of research training, a different program is needed to train college teachers, perhaps leading toward a new degree. Such proposals must grapple with two issues: who would enroll, and what would they be taught? On the first score the answer is clear: very few able students would enroll. Only about half of those now earning Ph.D.’s take academic jobs. Perhaps half of those who take academic jobs are at universities where research and graduate teaching require as much time as undergraduate work. A program aimed strictly at training teachers could thus hope for no more than a quarter of the present doctoral market, plus an indeterminate number of would-be teachers who now drop out of doctoral programs because these do not fit their needs or interests. But even this is optimistic. A teaching doctorate would have less status and attract fewer talented students than one aimed at training scholars. Its graduates would have difficulty getting good jobs, even in colleges that claim not to be concerned with whether their faculty do research. This would scare away the abler students, because they would not want to settle for a degree that kept many academic doors closed to them.

Even if prospective college teachers could be siphoned out of traditional doctoral programs into special teacher-training alternatives, this would probably be a mistake on intellectual grounds. Teachers need at least as much expertise and technical competence as researchers, and almost anything that has a defensible place in a doctoral program for scholars can also be defended in one for teachers. The whole spirit of proposals for a separate teaching degree is probably self-defeating.

No real progress will ever be made as long as teaching is seen as a soft option for those who cannot make it in research. It is actually a “hard” option for those who find a career exclusively in research insufficiently challenging or excessively routinized. A teacher needs to know as much as his research colleagues know, and more. He needs the scholarly competence Ph.D. programs claim to develop, but he also needs expertise in working with late adolescents, both in the classroom and outside it. In this context the present pattern of preparing college teachers, far from looking too extended or too rigorous, looks both too hasty and undemanding.

It is often argued that doctoral training already takes too long. But this is because so many doctoral students drop out at some point, and so many others work only part-time for their degrees. The average Ph.D. spends less than four full-time years getting a degree, which hardly seems excessive. What is now needed to supplement this scholarly training is a program of supervised teaching internships for newly minted Ph.D.’s. Some existing fellowship programs try to achieve this by attaching novices to senior faculty members, who are expected to help induct the novice into teaching. Some universities also have counseling bureaus or the like which will help individual faculty members improve their teaching if they ask. But these are small, little-known programs, and they are clearly not comparable to internships, which should be required of every starting teacher. Such internships would have to pay competitive salaries, and they would have to be viewed basically as a modification of the present working conditions of new assistant professors. These men would be involved in at least one staff course of the kind we have already described, where both the general problems of teaching and their individual strengths and weaknesses would be subject to critical discussion.

Is there the slightest chance that any of the reforms proposed in this article will be adopted? History is not encouraging. Similar proposals have been made many times over the past two generations, and the graduate schools have remained untouched. It is tempting to argue that the academic profession cannot delay reform any longer, since it now faces an unprecedented wave of student discontent. This discontent, while not a direct threat to the graduate schools, seems to bring legislative reprisals against those campuses where graduate education is concentrated, and budget cuts hurt the graduate schools along with everyone else.

Unfortunately, however, improvement in the graduate schools is unlikely to reduce the level of student protest significantly. Today’s students are not protesting the frequently impersonal, inhumane quality of their professors’ research, or even the casualness and occasional incompetence of their professors’ teaching. They have almost all been bored in class since they were six, and very few even entertain the idea that this is unnecessary. Their anger and resentment focus on other problems, for which remedies are more obviously available: poverty, racism, the war in Vietnam, or even restrictive dormitory hours for girls. The reforms we are discussing will not resolve these issues, and they will not buy peace in the intensifying war between the generations. If they come at all it will be because the academic profession, or at least a handful of entrepreneurs within it, decides that reform is in its own self-interest. Today that is only a hope, not a prediction.