The initial task facing the Committee on Graduate Education when it was created by the AHA Council in 2000 was "to obtain a clear sense of the national shape of doctoral education" for historians. 1 The Committee needed reliable and fairly detailed information about the present state of graduate training before it could offer any useful recommendations for the future. There was also a suspicion, abundantly confirmed in the course of this investigation, that most faculty members and graduate students have but a hazy familiarity with the workings of other graduate programs and that it would be a service to the profession to facilitate comparisons and a sharing of information and best practices. Unfortunately, the humanities in the United States suffer from a notable lack of formal and sustained data collection. 2

The Department of Education and the National Research Council (through its annual Survey of Earned Doctorates and its periodic rankings of research programs) do collect vital yet relatively narrow data about graduate education across the disciplines. Over the past two decades, the AHA has supplemented this with its own annual surveys of history departments and other data-collection initiatives. But there is a lot that we do not know about the professional preparation of historians.

The last time the AHA conducted a thorough review of graduate education was in the late 1950s. John L. Snell, a historian on leave from Tulane University while he served as research director for this effort, gathered all the data about historians and their training that he could get his hands on and then circulated a detailed questionnaire among history departments to fill in the gaps. The results of Snell's information-gathering were published in 1962 and in many cases have served as a basis for comparison with the current data in this report. 3 One can only go so far, however, in comparing discrete data points, and more than a few times we longed for an uninterrupted series of data to help shore up an analysis of some trend in the profession. A number of the recommendations in chapters 2 and 3 grew out of this frustration, and the Committee hopes that the AHA and individual graduate programs will do a better job of gathering consistent data in the years to come.

A Preliminary Sounding

In February 2001, the Committee distributed a brief, open-ended survey to the chair of every history department on the AHA mailing list (more than 630 in the United States and Canada). We invited the chairs to identify specific areas of concern in graduate education, regardless of whether their own departments trained graduate students. In just two weeks, the Committee received nearly a hundred responses, ranging in length from a few phrases to many careful paragraphs. Every part of the country and most types of colleges and universities were represented in the sample, from large departments that focus on graduate training to small departments at baccalaureate institutions that may not even have hired a history Ph.D. in recent years.

Perhaps not surprising, the respondents to this initial survey identified student funding and placement (or the general state of the academic job market) as the most pressing challenges in graduate education today, whether viewed from the perspective of their individual departments or of the profession as a whole. Nearly every response contained a variation on these two themes, alone or in counterpoint. Beyond the ubiquitous worries about funding and placement, the chairs identified a range of other concerns about graduate training, listed here in a rough order of the frequency with which they were mentioned: the length of time it takes most graduate students to complete a Ph.D.; the inadequate preparation of many graduate students for college teaching (a special concern for departments that had recently hired new junior faculty members); the increasing difficulty of recruiting top students; the struggle to build a "critical mass" of students for many graduate seminars; the challenge of providing students with both a sufficient breadth of historical knowledge and enough teaching experience for today's job market; overworked teaching assistants; the place of the master's degree in history education; the weak preparation of undergraduates, especially in foreign languages; the "growth in interdisciplinary scholarship"; and the "feeling among graduate students of being left out." 4

With this list of concerns in hand, and after reviewing other recent surveys by investigators in the field of higher education, the Committee's research director began to draft a detailed questionnaire on the state of doctoral education for historians. 5 The drafting phase involved extensive consultation and review by members of the Committee and the AHA staff, and the resulting questionnaire reflected four important decisions about the survey's shape and extent:

1. The survey would be distributed only to doctoral history programs in the United States, as it did not seem desirable (or even possible) to craft a single questionnaire that addressed the very different goals and procedures of doctoral and master's degree programs. 6

2. The questionnaires would be sent to directors of graduate education, as they are the faculty members in the best position to answer detailed questions about particular graduate programs.

Other surveys would subsequently be distributed to graduate students and to employers of public historians.

The questionnaire for the DGS's would include three distinct types of questions: (a) factual or statistical questions about the doctoral programs, which we expected to yield fairly precise and reliable quantitative data; (b) observational questions, where respondents would be asked to comment on the typical practices and attitudes of their colleagues and graduate students; and (c) matters of personal opinion. With these latter questions, the Committee was looking to gain a better understanding of the culture of the profession, as viewed from the particular perspective of senior faculty members who train new Ph.D.'s. (The Committee used other research methods to explore other perspectives on graduate education; the results have been incorporated throughout this report.)

In May 2001, the final version of the questionnaire was distributed to directors of graduate studies at 158 doctoral programs—virtually the entire population of doctorate-granting history departments in the nation (including a few institutions with two completely separate Ph.D. programs, one of them usually in the history of science). Appendix C contains a copy of the questionnaire text and the raw numerical results. In the end, usable questionnaires were returned from 105 programs, a large and representative sample of the population, whether judged by regional distribution, program size, academic reputation (as reflected in the NRC rankings), or public versus private institutions (see appendix B for a full list). The questionnaires were usually completed by the local DGS, though in a few cases they were completed by department chairs or other designated faculty members or staff.

Taken together, the replies provide a useful snapshot of doctoral education for historians at the start of a new millennium, as seen through the particular lens of the DGS. The snapshot can be summarized in six C's: C oncern, C omplexity, C hange, C areers in Transition, C omplacency, and (Mis) C ommunication between Faculty and Graduate Students. The analysis that follows is based on hard data from the completed survey forms but also on the varied opinions and judgments of the respondents. It draws upon the detailed experience of these faculty members while frankly noting the areas where they have little firm knowledge. The Committee on Graduate Education does not agree with all the opinions of the survey respondents; other historians, with different relationships to the process of graduate education, are also likely to disagree with some (perhaps many) of the opinions and conclusions in this chapter. For that reason, the survey results should be read in close conjunction with chapter 1 of this report, "We Historians," which draws upon a wider range of data sources.

The pages that follow rely heavily upon a statistical analysis of the survey data performed by Robert B. Townsend, assistant director of the American Historical Association for research and publications. Townsend also supervised the tabulation of the raw data, with the aid of AHA executive assistant and special projects coordinator Miriam Hauss. Without their continual assistance and advice, it would not have been possible to conduct the Committee's business, much less write this report.

Concern

The most striking evidence of historians' concern about the state of graduate education is the exceptional response rate to the Committee's survey. Presented with a forty-one-page questionnaire and little incentive but moral suasion to fill it out, 105 of our colleagues still took the time to complete the survey—an extraordinary response rate of two-thirds. (Three additional doctoral programs politely declined to respond, claiming they were too new to provide the Committee with any meaningful data.) In many cases it took hours, even days, to complete the questionnaire, and we want to thank the respondents for their efforts on behalf of the entire profession.

The responses suggest pervasive institutional, departmental, and personal concerns about graduate training. In a majority of cases (51 percent), respondents told the Committee that their universities had begun to pay an increasing amount of attention to graduate education in the last five years (versus just 11 percent who saw less attention being paid). 7 Some attributed the new interest to a sincere effort on the part of administrators to improve graduate programs (e.g., "a greater commitment from [the] administration to graduate training"). Others attributed crasser motives to the deans, such as an "effort to gain [a] higher Carnegie rating" or the "need to make more money on graduate tuition." Nearly every DGS who noted a declining interest in graduate programs came from a public institution; this, plus several comments on the survey about the need to address state funding formulas, illustrates the split between public and private universities discussed in chapter 1.

History departments have also been looking more closely at their graduate programs. More than three-quarters (79 percent) of the departments in the survey had undertaken an internal evaluation of graduate programs in the previous five years or have definite plans for an internal evaluation in the next five years. The impetus for these internal evaluations varied, including pressure from state agencies; campus-wide strategic planning efforts; multicampus strategic planning; ongoing periodic assessments (typically conducted every five years, but sometimes on seven-, eight-, or even ten-year cycles); or as preparation for an external evaluation. 8 In at least three cases, the internal evaluations were prompted by a "large number of retirements and [the] hiring of young people"; in another half-dozen cases, evaluations were prompted by significant changes in the number or quality of graduate students being attracted to the program; and some departments just felt the "need to keep tabs on things."

Finally, many individual historians are concerned about the state of graduate training for the sake of their own students and for the sake of the discipline. This came through, among other places, in the dozens of comments that respondents appended to the doctoral survey questionnaires but also in the thoughtful participation by hundreds of historians (including graduate students) at open forums and site visits conducted by the Committee on Graduate Education (see appendix A). Although one DGS bluntly told the Committee that it was an "inauspicious time for improvements [in graduate training]," the more common view was that reform is overdue.

Complexity

On the basis of the survey results, it is impossible to describe a typical (much less a model) doctoral program in history. The one constant in the data is variability; in almost every particular of doctoral training, the programs presented a wide range of different sizes, shapes, and practices. The Committee's commitment to promoting changes in graduate education at the individual department level—rather than offering "one-size-fits-all" recommendations for reform—was reinforced by the complex picture of doctoral training education that emerged from the survey. Four areas of variability deserve particular emphasis: program size, applicant pools, student funding, and basic program requirements.

Program Size

The doctoral programs in this survey range in size from three to ninety full-time faculty members engaged in graduate instruction (with a median of twenty-five), and from 11 to 300 currently enrolled graduate students. 9 During the five years prior to the survey, most graduate faculties remained the same size (43 percent) or even grew larger (38 percent), while only 18 percent of the programs reported a reduction in size. Yet over that same period of time, the total number of doctoral students in history declined. The sharpest drop came from history departments in the top quartile of the NRC's 1993 reputational rankings, while unranked programs (many created in the 1990s) actually increased their number and share of graduate students (see figure 1.11, p. 30).

Applicants and Admissions

Between and 400 would-be students applied to each history doctoral program for the academic year that began in fall 2001 (with a mean of 75 and a median of just 38); between and 150 were offered admission (mean of 24, median of 19); between and all of the admitted candidates (but usually around half of them) also received some offer of financial aid. In the end, 1,176 students accepted offers of admission from the 102 doctoral programs that included this information on their surveys, for an average entering class of just 12 students (though ranging between and 50, with a median of 10). One thing that was not diverse about the incoming cohort was its demographic composition: only 62 of the new Ph.D. students reported in the survey were African American, 66 were Asian American, 65 were Latino/a, and 10 were Native American. Forty-six percent of the cohort in total was composed of female students, though this was probably not consistent across all ethnic categories. 10

If the number of applicants for graduate study in history has shrunk in the last decade (and perhaps become less diverse in the process), has this entailed a decline in quality as well? One DGS shared his personal observations that "nationwide, fewer students are applying to graduate school in history. The ones who do are generally apt and interested." But not all of his colleagues seemed to agree: only 37 percent reported that the applicants to their own departments had become better in the last five years, versus 21 percent who noticed a decline in quality and another 37 percent who saw no real change. In the majority of cases (54 percent), respondents also think it has become more difficult to recruit good students, while just 10 percent think that recruiting has gotten easier in the past five years. A notable trend in the data is that representatives of larger doctoral programs (as measured by enrollments and annual Ph.D. production) were much more likely to report a decline in quality and a rise in recruiting difficulties, which might indicate that smaller, more-specialized programs have become relatively more attractive to potential graduate students. 11

The Cost of a Ph.D.

In section III of the Committee's survey, we asked the directors of graduate studies to list how much it costs, per year, to pursue a Ph.D. in history at their school. Unfortunately, this was easier to ask than to answer, as every institution calculates tuition rates and other fees in its own particular way—and the resulting total may bear no relationship to what graduate students are actually expected to pay, especially at institutions that routinely offer tuition waivers as part of any financial aid packages. 12 At many (but not all) schools, the tuition fees decline or even disappear once a student passes his or her exams and becomes a Ph.D. candidate. That being said, respondents did offer the following data about the "full cost of tuition for a year of graduate study in [their] department[s]." The numbers in table 4.1 should not be considered fully reliable, both for the reasons mentioned above and for the fact that a large number of respondents were unable to provide an answer at all.

Seventy-nine percent of all entering doctoral students received at least some financial aid for the 2000-2001 academic year (75 percent at public institutions, 91 percent at private institutions), and for 89 percent of them, the financial aid package began with a partial or total tuition waiver. Beyond that, it is hard to describe a "typical" amount or pattern of financial aid or to make adequate comparisons among programs. Even determining the precise value of a student's aid package is difficult, as various institutions include (or do not include) tuition waivers, fee waivers, health insurance costs, housing subsidies, etc., in their calculations. The different cost of living in different cities also makes any comparison difficult. Nonetheless, we asked the DGS's to describe the range of outright awards (excluding loans and employment offers) that they were able to offer incoming students for the first year of doctoral study, with the results as listed in table 4.2.

From within this complex (though still underdeveloped) snapshot of doctoral funding, two clear trends emerge from the survey data. Many doctoral students in history are receiving more financial support than they would have received just a few years ago, although it's still not enough. On the positive side, most history doctoral programs (86 percent, based on ninety-three responses) now promise their incoming students multiple years of funding—three-and-a-half years of funding on average, with at least two dozen programs offering a full five or even six years of guaranteed support, assuming steady progress toward a degree. 13 Sixty percent of the respondents reported that their programs had made significant changes in financial aid policies over the past five years, usually in the direction of increased funding. Meanwhile, anecdotal evidence (including scattered comments on the survey forms) indicates that a number of departments have also moved in the direction of equal funding for all admitted students as a way to remove invidious comparisons among them; far fewer students are being admitted without aid than in the past. Graduate students in many programs also have access to additional funding for foreign language training (43 percent), dissertation research/travel (92 percent), travel to conferences (95 percent), and even for training in multimedia or other new technologies (17 percent).

Despite this good news, however, only 42 percent of the respondents describe the typical funding package offered by their program as "adequate," given the local cost of living; just 4 percent thought that the typical financial aid package would fully support a single parent and child; and not one thought that it could fully support a married graduate student with a family. As a result, most respondents seemed to acknowledge that graduate students were relying on debt to make ends meet—though they would not venture a guess as to how much debt the average student was carrying. 14 Faculty members are clearly dissatisfied with the amount of financial support they can offer their doctoral students (indeed, 59 percent of respondents describe their colleagues as being dissatisfied) but probably less dissatisfied than the students themselves.

Academic Requirements

In order to earn a Ph.D. in history in the United States, a graduate student is required to complete somewhere between four and twenty-seven courses (with an average load of fourteen), or between 30 and 135 credits (with an average of 61.5), depending on the school and the accounting system in place there. This total might—or might not—include the "placeholder" credits that some schools (especially public institutions) require to maintain residency or formal enrollment. At 79 percent of the graduate programs, the required load will include at least one course that every doctoral student is expected to take (usually a seminar in research methods and/or historiography, but see survey question VI.3 and table 4.3 for the range of other required courses).

To fill out the rest of their coursework, graduate students in the 2000-2001 academic year were able to choose from at least 7 course offerings and as many as 164, though the average number of offerings per program was under 50 (and even that figure was inflated by the many undergraduate lecture courses that also happened to be open to graduate students, especially at large public universities). These graduate courses came in a variety of formats, from large lectures to small research seminars to independent tutorials (see survey questions VI.1 and VI.7). Independent courses are clearly on the rise as graduate programs shrink in size and can no longer attract enough students for specialized historiographic seminars. As one DGS explained, his department does not "keep a count of individually offered reading and research courses" because there are too many. Testimony from the Committee's site visits and open forums suggests that, in many cases, these courses are being offered on an informal basis, which means that the instructors receive no teaching credit for them. 15 For their part, many students would also prefer a more structured and reliable curriculum of graduate seminars, rather than having to "make do" (in the words of one concerned student) with the vagaries of independent reading courses.

Their coursework complete, graduate students at different institutions will then go on to face very different examination procedures before they can start writing their dissertations. They might take their general exams—a.k.a. "comprehensive" or "qualifying" exams, among several other aliases—as early as their third semester (especially if they come to the program with a master's degree in hand) or as late as their fifth year, though the typical timing is some point during a student's third year. Different programs require different numbers of exams, from one to four "major" fields (the mode is one) and zero to three "minor" fields (the mode is two). Every possible combination of written and/or oral exams is employed by some department, though minor fields are less likely to be subject to a written exam; a growing number of programs have also begun to include a formal defense of dissertation proposals as part of the examination process.

Despite the variety of formats, a few things are generally true about graduate examinations across the discipline. Minor fields are considered teaching fields, not research fields (only 18 percent of the survey respondents said they expected students to be able to conduct original research in these fields). In most departments, the structure of graduate exams—and perhaps their rationale as well—has remained unchanged and unexamined for years (see survey question VI.18), though the intellectual breadth and historiographic content of the exams have undergone a dramatic shift, as will be discussed in the next part of this analysis. Perhaps as a result, a significant number of faculty members and students are less than fully satisfied with current examination procedures. 16 One survey respondent summarized his colleagues' concerns about the exams in their own department this way: "too bureaucratic, too many regulations; too lax or too severe requirements; too much or too little DGS discretion"; another added that the "format of comprehensive exams may be too rigid and demanding." These were typical comments. Whenever the Committee discussed examinations with graduate students, the students pointed to the enigmatic purpose and daunting nature of the exams as a source of unnecessary anxiety.

Although we have focused here on the variations among doctoral programs, this should not distract us from the common patterns and shared endeavor of training new historians. Graduate programs can look surprisingly different on the surface, but how many of the variations actually reflect more fundamental differences? And how do the variations affect the lives of graduate students? On these questions, the survey data are inconclusive. Many different approaches can achieve the same ends, which is a valuable thing, but too narrow a focus on local practices can make it difficult to see the complex whole of graduate training in the discipline. Whenever we talk about general trends in this report, some doctoral program(s) will not fit the pattern. But it is important to describe the general trends—which point in the directions of both change and stasis—as a starting place for local comparisons and reforms.

Change

At the very end of the survey (question XIII.5), we asked respondents to comment on the most significant changes that they had seen in the graduate training of historians over the past decade or so. Their answers to this question underscore six broad changes (in no particular order of precedence):

1. more emphasis on teaching skills and pedagogy;
2. more-sophisticated approaches to historical analysis, from both faculty and students;
3. the development of new research fields (sometimes gathered under the broad headings of "social history" or "cultural history");
4. the growth of interdisciplinary, transnational, and comparative work;
5. the mounting pressure on graduate students for early professionalization, together with rising expectations about the quantity (and quality) of the students' written work; and
6. greater use of electronic resources for research and teaching.

Some historians have always been concerned that graduate students do not receive enough systematic training as teachers, so it comes as a welcome sign that departments are paying more attention to this aspect of professional development. A rising number of doctoral programs now offer their students a formal introduction to pedagogy, but history departments "could [still] do more to integrate teaching content and techniques into required coursework" (to quote the DGS from a large state university). Just a quarter of the doctoral programs in the survey, for example, would accept graduate research projects (such as seminar papers) that have an explicitly pedagogical emphasis, and fewer still (10 percent) would accept a dissertation with a pedagogical emphasis; smaller and newer doctoral programs (i.e., those that have opened their doors since 1971) are the most receptive to such innovations.

In the last four decades, the intellectual map of the entire discipline has been redrawn by new subjects, new theories, and new approaches. The important news from the survey is that graduate education reflects these historiographic transformations while tenaciously preserving older subjects, theories, and approaches. For every respondent who enthused that "the range of questions that are asked and the sorts of historical analysis undertaken have broadened notably," there was another who "[did] not think training has greatly changed." Likewise, for every DGS who applauded a graduate program that was "somewhat more comparative and transnational [than in the past], though not as much as I think [it] should be, given the way the profession is moving," a colleague was ready to counter that "our innate professional conservatism has kept the [students'] exams ... from keeping up [with some of the latest trends]—i.e., we still demand competence along traditional lines."

Consider the range of courses available to history graduate students in the 2000-2001 academic year (table 4.3). A higher percentage of departments offer courses in women's (or gender) history than any other field, closely followed by social and cultural history. Newer fields, such as environmental history and the history of sexuality, also make a strong showing on the list—but not as strong as the traditional fields of political, diplomatic, or intellectual history and not much stronger than legal or military history. What the list fails to capture is a significant shift away from European history and toward other regions of the world, as well as the growing popularity of modern U.S. history.

This becomes clearer in the comments that respondents included about the recent changes in course offerings, which typically describe "more topic- or theme-based courses" and "new courses ... in World Areas," while "post-medieval, pre-1900 [European history courses] are in sharp decline." (The author of this last comment added, "I expect these developments to accelerate over the next five years.")

The same shift was reported in regard to graduate examination fields, where "American history has markedly increased its following, European [fields] of all periods have been used less," and "comparative, thematic, and global" exam fields have been on the rise. Indeed, more than half the respondents reported a noticeable shift during the past decade in the historical interests of both graduate students and faculty members, as reflected in the selection of graduate examination fields, with especially sharp increases in comparative and thematic fields, as opposed to fields defined strictly by chronology and/or region (see survey question VI.16). In a related development, as many as a fifth of the doctoral programs in the survey have made a serious commitment to some variant of world history (including global history, the Atlantic world, borderlands history, and the African Diaspora), beyond merely offering a graduate course or two in the area. 17 World history is now being offered as a minor field, a major field, and a research topic for dissertations. At Northeastern and Florida International Universities—two institutions visited by the Committee—the entire graduate programs have been redirected toward a transnational approach, and they are not the only history departments to be engaged in such experiments.

Professionalism is a good thing, but the excessive professionalization of some doctoral students was among the negative changes in graduate education noted by the survey respondents. It was also the theme of a recent presidential essay by Lynn Hunt in the pages of Perspectives. 18 This topic has already been addressed at length in chapter 3, so only a few additional comments are necessary here. Asked to compare the professional identities of today's graduate students with those of their counterparts a decade before, many survey respondents seemed to be at a loss; 48 percent could not or did not answer the question, by itself a troubling indicator of possible miscommunication between students and faculty. But 26 percent did identify a new attitude among today's graduate students, who were described by one DGS as "aggressively professional" (other comments contained similar language). This was attributed to the job market and the related pressure for graduate students to publish their work early and often. A large majority of respondents (78 percent) reported that students in their department are aware of the rising expectations of the publishing record required for hiring and promotion, an impression that was confirmed during the Committee's site visits and other conversations with graduate students. According to the directors of graduate studies, the rising expectations "affect ... [the] choice of dissertation topics," and can make graduate students "anxious to begin publishing early" (or, conversely, "increase their determination to get something published")—but it is not clear that pressure makes dissertations any better 19 or broadens the horizons of the discipline. 20 Indeed, as one graduate student warned the Committee, the pressure to publish can just as easily "lead ... to a situation where [dissertation] topics that might have been more original in design become more conventional in order to survive."

Finally, the growing attention to electronic resources for research and teaching at the graduate level that the survey respondents noted in question XIII.6 is just partly borne out by the rest of the survey data. Of course, historians have been active users of computer technology for many years, but doctoral programs have not embraced the new media with much enthusiasm. Half the departments in the survey maintain their own computer or multimedia labs, and the students generally have adequate access to other technology resources at the local institution. Barely a fifth of the programs in the survey offer their graduate students any formal training in digital, Web-based, or multimedia research techniques, however. Thirty-one percent of the programs do offer some formal introduction to teaching with technology, and half the respondents also report that training in these skills is available somewhere else on campus. Yet only 44 percent report that their graduate students have a regular opportunity to test their technology skills in an actual classroom. In 68 percent of the departments, graduate students never have an opportunity to engage in any Web-based or distance education (see survey question IX.6)—but this statistic tells us nothing about the actual number of graduate students (or even faculty members) teaching via these new media. Relatively few history departments will even allow graduate students to submit their own coursework in any multimedia format (see survey question VI.8).

Careers in Transition

Professional historians do not face the same universe of career opportunities today as their counterparts did in the past. Indeed, fewer than half the respondents to our survey (41 percent) believe that today's graduate students will follow career paths as historians roughly similar to their own, while 44 percent believe that their students will certainly follow different paths. 21 Eighty-six percent think that graduate students and faculty share a roughly similar set of expectations about the career challenges that new history Ph.D.'s will face in the next few decades. At the same time, only a fifth of the respondents say they have observed a significant change in the career expectations of doctoral students over the past decade, while another fifth (with some substantial overlap) say they have observed a significant change in the career expectations that faculty members have for their students (see survey questions X.20-X.22).

What should we make of these apparently contradictory responses? Taking the survey data as a whole, there is a clear mismatch between career expectations, the actual experience of graduate students on the job market(s), and the preparation they receive for careers in transition. A similar slippage was revealed in the recent survey of graduate students conducted by the Wisconsin Center for Education Research, in which history students expressed an overwhelming desire to become professors at research-oriented universities while recognizing that such positions are scarce and worrying that their preparation for these jobs was incomplete. 22

Let us begin with the actual positions that new Ph.D.'s obtained during the 2000-2001 academic year (table 4.4). It was, by all appearances, a good season for job seekers: an average of ten graduate students per program were on the market that year, and eight found and accepted positions (limiting our analysis to the ninety-two doctoral programs included in the table). 23 In almost every case, the DGS's reported that their students exclusively or primarily looked for academic jobs (see survey question IX.2B), but the placement data do not seem to bear this out. About a quarter of the students found jobs at research universities; about half found jobs at institutions that stress teaching; a few became postdoctoral fellows; and the rest took positions outside the academy. (The final group was probably undercounted, as many students pursue nonacademic jobs on their own, outside their departments' placement mechanisms.) According to the respondents, this distribution of new jobs was fairly typical of placement patterns over the past decade. 24

This placement pattern contrasts sharply with the stated aims of most doctoral programs when it comes to career preparation. Given a choice, survey respondents were much more likely to describe the Ph.D. offered by their program as a "research degree" (45 percent) or a "research-based teaching degree" (41 percent), instead of a "teaching degree" ( percent) or a "general-purpose degree, preparing the holder for a variety of professional settings" (13 percent). Yet many students are launching careers that will emphasize teaching and other skills that go beyond the relatively narrow limits of historical research. The highest priority of most programs is training their students for positions at research universities or comprehensive colleges; training students to teach at community colleges is at best a moderate priority; training students for other sorts of professional careers as historians—in secondary schools or the public sector, for example—is often no priority at all (see table 4.5). It is not clear whether these institutional priorities match the intended career paths of the students themselves.

Although community colleges are a growing area of employment for historians, just 15 percent of the doctoral programs offer their students any formal introduction to history teaching at the community college level (while just 5 percent offer any exposure to history teaching at the K-12 level). Yet most respondents report that new Ph.D.'s from their department are at least "adequately prepared" to assume positions at community colleges and secondary schools (see survey question IX.22). The majority (54 percent) also report that graduate students are better prepared for college teaching when they enter the academic job market today than they would have been a decade ago—despite the fact that many programs continue to rely on informal "apprenticeship" arrangements as the primary way to train graduate students as teachers (see survey question IX.13).

Teaching is a formal degree requirement at just 15 percent of the history doctoral programs in our survey, though more often it becomes a de facto requirement because teaching assistantships are a mandatory part of financial aid packages. Most students do teach at some point in their graduate careers. To judge from the comments on the survey, history departments have received the message that their graduate students need better teaching skills to prosper on the academic job market. One respondent noted, with more than a hint of frustration, that "the entire culture of our department is aimed toward research and training for research," while several of his colleagues suggested that their own programs "could do much more in the way of training and pedagogy ... especially on using technology and collaborative modes of teaching and learning." But the surveys included many more indications of recent and positive changes in the training of future college instructors, at least partly in response to the market pressures for better teachers. In the five years prior to the survey, half the departments made some change in how they train graduate students as teachers, with smaller and less-prestigious doctoral programs the most likely to have implemented improvements. 25 Various programs now supplement the personal supervision by an advisor or other faculty mentor(s) with required seminars in college teaching (36 percent), voluntary seminars (22 percent), teaching workshops initiated by the history department (51 percent), or workshops run by other parts of the university (55 percent). Most institutions (71 percent) also have a "teaching center," though they seem to be underused by history departments (see survey question IX.14). As the DGS at a small state university summed up these important trends, "We've come a long away recently in training to teach, but we should and will be more intensive and purposeful in the act."

Unfortunately, the preparation of graduate students for careers outside of the academy has lagged far behind these efforts to enhance the students' teaching skills. One reason, as discussed elsewhere in this report, is that nonacademic careers have not been highly valued by some members of the profession. A related reason is ignorance: more than a third of respondents felt that their faculty colleagues were unaware (or just somewhat aware) of the range of professional opportunities for historians outside the academy—and less aware, certainly, than the students themselves. In many departments, graduate students are expected to learn about professional opportunities outside the academy on their own and to independently prepare for nonacademic interviews (see table 4.6). Very few programs could boast on the survey of recent initiatives to improve the career prospects of graduate students seeking nonacademic employment (see survey question XI.14). Survey respondents were uncertain, too, about the effectiveness of university-wide career offices in helping their history graduate students find nonacademic employment. 26 As the DGS at a major research university candidly reported, "I expect we will be more conscious of this market [in the future]."

Complacency

The counterpoints to concern and change are complacency and a satisfaction with the present state of graduate education. There is ample evidence of these in the survey of doctoral programs. In the five years prior to the survey, the majority of history departments declined to make any significant changes in their existing foreign language requirements, the number and/or character of their graduate examination fields, their dissertation requirements, or their procedures for advising graduate students. 27 Barely half made an effort to reduce the time it takes to earn a doctoral degree, even though the relatively long time to degree for historians has been a perennial complaint, especially of deans (see survey question VI.27). 28 Even in those areas of graduate education that experienced the widest changes in recent years—recruitment, graduate student funding, pedagogical training, new course offerings, etc.—a sizable minority of departments (from 27 percent to 43 percent) stuck with their existing procedures and policies (see survey questions II.11, III.20, IV.6, VI.6, and IX.26).

Five years is a short period of time, and no one should conclude from these survey data alone that we historians are persistently or systematically resistant to change—especially when the signs of change and continuity are so closely balanced. In many cases, "professional conservatism" is the most effective approach to graduate education. And yet, when the evidence from this disciplinary snapshot is laid beside other descriptions of graduate education from the past, continuity can look more disturbing. For example, in 1940 the Dartmouth historian A. Howard Meneely presented a paper at the annual meeting of the American Historical Association in which he described the leading contemporary concerns about graduate education: "One hears it argued that there is excessive specialization, overemphasis on research, too much of the belt-line method and too little supervision in training graduate students, a failure to equip them properly for the kind[s] of teaching they must do, and an insufficient effort to limit the production of degree holders to somewhere near the demands of the academic market." 29 A remarkably similar set of concerns can be found in the 1962 report by Perkins and Snell, and the present report as well. Why do these widely acknowledged problems persist?

Both optimistic and pessimistic conclusions can be drawn from what the survey respondents told the Committee about levels of satisfaction in their various departments. A plurality of respondents concluded that, on the whole, their colleagues on the faculty were "somewhat satisfied" about graduate course offerings, examination procedures, the professional development that doctoral students received in the department, and the quality of the advising/mentoring that their students received; another 10 percent to 25 percent described their history department colleagues as being "very satisfied" with these aspects of graduate training. The levels of satisfaction that the survey respondents ascribed to the graduate students in their departments were just as high. Indeed, there was a strong correlation, across the board, between reported satisfaction levels for students and faculty at each institution—which invites the speculation that some respondents projected faculty sentiments onto the students in the absence of more direct knowledge. 30

Without dispute, many historians (including graduate students) are happy with the present state of graduate education. 31 But the levels of satisfaction suggested by these survey responses seem exaggerated and certainly out of step with the clear signs of dissatisfaction captured in other parts of the survey and reinforced during the Committee's face-to-face conversations with historians around the country. In part, we blame this on a set of poorly worded questions; at the very least, the questionnaire should have offered a working definition of "satisfaction." The questionnaire also asked the respondents to summarize the opinions of two highly diverse groups—always a difficult task. Finally, historians have commendable loyalties, which led most respondents to rank their own departments more highly than the profession as a whole. For example, 44 percent of the respondents claimed that morale in their own departments has improved during the past decade (versus 28 percent who noted a decline in morale and another 20 percent who perceived no change), while only 9 percent thought that morale has improved across the profession.

(Mis)communication between Faculty and Graduate Students

Students and faculty members do not always see eye-to-eye when it comes to graduate education, and no survey was required to establish that fact. Whenever the Committee met with historians, someone would raise the point that students and faculty bring different expectations to the process and have different understandings of the goals and procedures of graduate training. In most cases, we were told, these differences are resolved through effective student-advisor relations (and the Committee believes this is true). But advising can also be fraught with miscommunication. As one DGS pointed out on the survey, "Some faculty may feel that students don't listen as carefully as they would like, but what else is new?" For their part, history graduate students have been just as blunt in assessing the current state of advising: "[Some] students expect ... more attention from their advisors than is appropriate or necessary," wrote one Ph.D. candidate at a prominent history department after reviewing the preliminary results of our survey, and some "advisors treat ... their students too much like children." Another student wondered whether "tenured professors could ever be provided with seminars on how to advise graduate students effectively. Many of them cannot do it in a sane, reasonable way." 32

History departments usually rely on multiple methods to announce the formal requirements of their doctoral programs, from printed handbooks and Web sites to orientation meetings and informal chats with the director of graduate studies (see survey question IV.5). Unfortunately, that does not mean that everyone is equally conversant with the details. "In spite of the guidelines ... we provide," one DGS noted on his questionnaire, "graduate students often claim they don't understand the requirements" (and the Committee heard variations on this complaint during several of its site visits). Yet, overall, the respondents felt that graduate students were usually or always familiar with the various program requirements. They could not say the same for faculty members: nearly a third of the respondents reported that their own colleagues rarely or at best only sometimes had a clear understanding of the requirements (see survey questions IV.3 and IV.4). In fact, to judge from the comments on the survey, most departments have at least one or two members who are quite badly informed about program requirements. One DGS acknowledged that "some faculty really don't understand the system very well—[the] students actually have a much better grasp of the regulations and expectations than faculty do." At another school, the "graduate students complain that faculty members are insufficiently well informed about the program to advise them adequately." Perhaps for this very reason, graduate students routinely informed the Committee during its site visits that they rely more heavily on members of the department's staff than on members of the faculty when it comes to navigating the local regulations.

Well informed or not, individual advisors seem to be the dominant source of information for most history graduate students. In the survey, the Committee invited directors of graduate studies to consider a list of different activities that contribute to the professional training of historians, ranging from formal instructional activities to the informal transmission of "craft knowledge" about the historical profession. For each activity, they were asked to indicate the "most important source of information and assistance for the graduate students in your department. These should be the most important sources in practice, not necessarily the sources suggested in departmental guidelines or other official statements." The responses are summarized in table 4.6. 33

There are some striking patterns to this data, starting with the commanding role ascribed to the individual advisors: in more than half the cases (fifteen of twenty-nine activities), the bulk of survey respondents felt that a graduate student's own advisor was the most important source of his or her information. This was true for some of the most straightforward activities on the list (such as selecting graduate courses) and for some of the least tangible (such as developing "survival skills" for the profession). These responses may not be surprising, given the strong tradition in graduate education of mentoring as apprenticeship. But one should be concerned when graduate students come to rely too much on a single source of information, especially when "advising varies significantly from one professor to the next," as the DGS from a medium-size public institution reminded the Committee. "Some are energetic and involved; others allow the students relative autonomy or prefer to remain less involved. Some are just plain lazy." (A number of other comments echoed this sentiment.)

For more than a few respondents, we suspect, "advisor" was simply a default answer to the question being posed: If graduate students are picking up the information somewhere, why not from their advisors? Nor were the DGS's shy about suggesting that graduate students are on their own when it comes to some important aspects of professional training. For six activities on the list, a plurality of respondents listed the "graduate student's own initiative" as the primary source of information. These included "keeping up with the current historical literature" but also "learning about the history, mission, and purposes of higher education" (which should be vital for any future faculty member), "learning about nonacademic job openings," and "following public discussions about the role of history in [American] society."

Finally, just a handful of respondents ever listed the "other graduate students" as the leading source of information for any aspect of professional training. This seems out of line with the actual experience of graduate students, who tend to rely heavily on the support and advice of their peers—whether individually or through collective endeavors such as graduate student organizations, student-run conferences, dissertation writers' groups, even unions—to navigate the challenges of graduate school (though, admittedly, we have not collected any hard evidence to support this conclusion). A majority of the survey respondents (70 percent) reported that faculty members in their departments always have "a clear sense of what the graduate students expect from them as teachers and advisors." This does not seem credible, considering the other data from the survey presented in this chapter and the anecdotal evidence gathered by the Committee while talking to historians at all stages of their careers. Instead, we take it as a further sign that students and faculty members could do a better job of communicating to each other their goals and expectations for graduate training.

The communication between students and faculty members may well be impeded by the relatively limited role that graduate students are permitted to play in the operations of some departments. Most history departments (88 percent) invite graduate students to serve on faculty search committees, and a smaller percentage (64 percent) invite them to serve on other faculty committees. At the same time, 21 percent of the survey respondents indicated that graduate students in their departments are rarely or never allowed to play an active role in program decisions that directly affect them, and 35 percent indicated that graduate students are only sometimes allowed to play an active role. Interesting to note, the history departments with large doctoral programs and those at public institutions—except in the West, for some reason—are the most open to student participation in departmental affairs.

After reviewing the preliminary results from the Committee on Graduate Education's survey of doctoral programs, a recent Ph.D. in American history wrote that he "found [the data] quite interesting ... [but] I think many of the responses are vague and incomplete. I write this not as a criticism, but because I realize that you ha[d] a difficult task: history graduate programs are small and do not [always] have specific data on the questions you pose." Two lapses in departmental attention are notable, because they touch on areas that graduate students consider especially vital: debt and attrition. 34 As mentioned before, very few departments are well informed about the debt burden of their students, and there is little evidence that the issue is seriously considered by the faculty (see survey question III.12).

No one likes to talk about attrition, either. Section V of the survey asked the respondents to describe the patterns of attrition (and, implicitly, retention) in their doctoral programs. Establishing the rate of attrition turned out to be more complicated than anyone imagined. In part, it was because attrition is hard to define: students do not always make a clean break from graduate school but instead take a hiatus (or begin to work on a dissertation) and then, gradually or abruptly, move on to another phase of their lives. In part, it was because most departments do not collect detailed information about attrition. As a result, the complicated grid that was provided on the questionnaire (survey question V.3) did not yield any statistically reliable data on the attrition of history graduate students. At their most revealing, the collected data suggest that students are more likely to depart early from a doctoral program (after the first year of classes or just before or after completing their graduate examinations) than later (once they begin the dissertation process).

Subjectively, the respondents predicted that 65 percent of entering students would eventually complete a Ph.D., with more optimistic predictions coming from larger programs, private institutions, and departments in the top quartile of the National Research Council reputational rankings. Sixty-eight percent felt that female graduate students were just about as likely as their male counterparts to leave the program before completing a doctoral degree (4 percent said female students were more likely to leave, 10 percent said less likely). Forty-seven percent felt that minority graduate students were about as likely as other students to leave the program before completing a doctoral degree (while 14 percent said minority students were more likely to leave and 15 percent said less likely). The respondents also predicted an average time to degree, for all entering doctoral students, of 6.9 years—which is well below the current median of 9 years for a Ph.D. in history. 35

In addition to asking how many students leave their graduate studies before completing a Ph.D., we asked the DGS's to explain why those students leave. Presented with a list of potential reasons for student attrition, they were invited to check as many reasons as they thought applied to the doctoral students in their own particular program and then to indicate the single most important reason (survey question V.9). The responses are summarized in table 4.7. Disappointingly, these results place the burden of attrition largely on the shoulders of graduate students. Almost no DGS reported that departing students were dissatisfied with the intellectual content of the historical training they received or even with the quality of faculty advising. Instead, they suggested that most departing students are motivated by an inability (or unwillingness) to pursue historical work at the doctoral level, or by the excessive financial burdens of graduate school, or by a sober assessment of the job market, or for undetermined family or personal reasons.

The heavy reliance on this final category ("personal reasons") by the survey respondents is especially troubling. Was it just a convenient alternative to "We don't know"? How can graduate programs know that students are leaving for personal reasons if they do not ask? Most departments (66 percent) rely on informal reporting to gather information about student attrition, though 45 percent do track attrition in a systematic fashion. Only 10 percent routinely conduct exit interviews with departing students. On the positive side, nearly half the departments in the survey say they have taken additional steps in the last few years to retain graduate students (see survey question V.10). The most frequent changes were better funding and a closer attention to mentoring arrangements, though many respondents also noted general revisions to the graduate programs—such as revamping field requirements—that had the secondary effect of reducing student attrition. Although the graduate students who reviewed these data for the Committee agreed that "students leave for financial or personal reasons," they emphasized that attrition is a complex process, sometimes made more painful by the asymmetrical relations of power between students and faculty members.

Conclusion

At a 1950 conference on the state of graduate training in history, John Krout of Columbia University wondered aloud whether "the people who come to us saying that they have finally made up their minds to work in the field of history know why they want to work in that field and what they want to do after they have worked in it at the graduate level. So many of them ... don't seem to know what they are preparing themselves for, except in a very vague fashion." 36 That is still true, though Krout should have added that many history departments are equally vague when it comes to articulating the disciplinary, professional, and occupational goals of graduate training. The bulk of this report (chapters 2 and 3) is a series of recommendations for improving the graduate education of historians. The initial challenge is for individual history departments to clarify their goals when it comes to advanced training in the discipline: What should faculty, students, institutions, the historical profession, and even society expect from a graduate program? To clarify these goals, historians need to know more about the current state of the discipline and the national shape of doctoral education; they also need more effective and systematic measures of how well any given program meets these goals. The Committee's survey of doctoral programs, together with the other data presented in chapter 1, can only be a start in that direction.

1. Thomas Bender, "AHA Launches Study of Graduate Education," Perspectives 38:6 (September 2000): 12.

2 . Robert M. Solow et al., Making the Humanities Count: The Importance of Data (Cambridge, Mass.: American Academy of Arts and Sciences, 2002).

3. . Perkins and Snell, Education of Historians in the United States. An even earlier model for our surveys was William B. Hesseltine and Louis Kaplan, "Doctors of Philosophy of History: A Statistical Study," American Historical Review 47:4 (July 1942): 765-800.

4. These two paragraphs are adapted from Philip M. Katz, "CGE's E-mail Survey Focuses on Challenges in Graduate Education," Perspectives 39:4 (April 2001): 11-15, which includes additional details about the results.

5. These models were especially helpful in drafting the AHA Survey of Doctoral Programs, in some cases by suggesting specific questions and/or phrasing: Robert M. Diamond and Peter J. Gray, 1997 National Survey of Teaching Assistants (Syracuse, N.Y.: Center for Instructional Development, 1998); Golde and Dore, At Cross Purposes; the survey questionnaire devised for the "Graduate Programs Climate Study" at the University of Iowa, spring 2001 (University of Iowa Graduate College and the Women in Science and Engineering Committee, chaired by Chris Brus); Lovitts and Nelson, "Hidden Crisis in Graduate Education," 44-50; Maresi Nerad and Joseph Cerny, "From Rumors to Facts, Career Outcomes of English Ph.D.'s: Results from the 'Ph.D.'s—Ten Years Later' Study," CGS Communicator 32:7 (fall 1999): 1-12; "The 2000 National Doctoral Program Survey" conducted by the National Association of Graduate-Professional Students and available at ; as well as the sources listed in note 3."

6. This decision should not be taken as sign that we consider the master's degree any less important that the doctorate. The AHA plans to issue a supplemental report on professional training at the master's level in the near future.

7. . These responses contained an interesting regional variation, with just 33 percent of respondents in the West saying that graduate education now receives more attention, compared to 57 percent in the Midwest and 75 percent in the Southwest; other regions did not vary much from the mean.

8. . Fifty-four percent of the history departments—most of them at public institutions—also had an external evaluation in the five years prior to the survey, usually by a state or regional accrediting board.

9. . The average enrollment was eighty students in doctoral programs at public institutions and eighty-eight students in doctoral programs at private institutions (without religious affiliation). However, these figures are not fully reliable, as some respondents included master's degree students in separate degree programs as part of their count. Because the formal definition of "enrolled" varies considerably from school to school, and usually excludes some students who are absent from the official rolls but still actively pursuing their degrees, the survey probably undercounts the number of graduate students.

10. . Based on incomplete demographic information provided by ninety-two of the doctoral programs in the survey. The ethnic and racial categories analyzed in this paragraph are the same ones provided on the survey questionnaire.

11. . At the same time, 83 percent of those who felt it was becoming easier to recruit also felt that the quality of applicants is on the rise.

12. . The Chronicle of Higher Education also discovered this while conducting a survey of graduate stipends at research universities. See Scott Smallwood, "Stipends Are Key in Competition to Land Top Graduate Students," The Chronicle of Higher Education, September 28, 2001, A24.

13. . The larger doctoral programs (i.e., those that awarded more than the average number of history Ph.D.'s in 2000) are able to offer their new students more secure funding: an average of 4.1 years compared to 3 years at smaller programs. On average, the private institutions are able to offer a full year of additional guaranteed support (4.2 versus 3.2 years for history departments at public institutions).

14. . Only 3 percent of programs in the survey maintain any systematic information about debt levels, and just eleven individual respondents were willing to offer an estimate for the average amount of debt accrued by their graduate students. Nonetheless, 38 percent felt that students in the program usually or sometimes incur excessive debts on the way to completing their Ph.D.'s.

15. . A typical comment from another DGS was, "We would like to have more formal graduate courses and [fewer] independent reading courses. But we have so few graduate students taking courses that we can't support them at half of two-course loads."

16. . In response to survey questions VI.19 and VI.20, respondents reported that 80 percent of their colleagues and 75 percent of their students were either "very satisfied" or "somewhat satisfied" by the current approach to examinations. But when the respondents were asked to comment on their colleagues' main concerns about the formal procedures of graduate instruction (question IV.4), exams were by far the most commonly cited cause for complaint.

17. . It is hard to offer a more precise count of programs with a commitment to world history because we did not ask any specific question on the survey about the presence of a world history field or "track" at each program; the estimate here is extrapolated from the responses to a number of different questions. A suggestive but incomplete list of graduate programs in world history is available at< http://www.worldhistorynetwork.org >.

18. . Lynn Hunt, "Has Professionalization Gone Too Far," Perspectives 40:2 (February 2002): 5-7.

19. . When it came to assessing the quality of recent student research, the survey respondents were not above a little chauvinism: 46 percent felt that the dissertations being produced by students in their own departments were stronger today than they were a decade ago, but only 14 percent felt that the general run of history dissertations had become stronger over the same time period (see survey question VI.24).

20. Most respondents felt that historians are living in a time of average or diminished historical creativity rather than a time of exceptionally imaginative scholarship (see survey question XIII.1).

      21. A striking trend appears when these responses are cross-tabulated by program rankings, size, and age. After deducting the nonresponse, fully two-thirds of respondents from the top-ranked, largest, and oldest programs expected their students to have similar career paths to their own, while only 30 to 40 percent of historians in newer, smaller, or less prestigious programs thought so. This has important implications for how different kinds of doctoral programs will train their students for the job market, not all of which are reflected in the survey data.

      22. . Golde and Dore, At Cross Purposes.

23. See also Robert B. Townsend, "Job Market Report 2001: Openings Booming ... but for How Long?" Perspectives 39:9 (December 2001): 7-12.

24. See survey question X.5. Again, counting just the ninety-two respondents who provided complete placement data, forty-six thought that the 2000-2001 academic year was typical, and twenty-five thought that it was not (though, notably, most of them thought it was a better year than average, especially for placements at research institutions); the rest did not know or did not respond.

25. The highest proportion of changes (75 percent) was found in the group of history doctoral programs ranked in the bottom quartile of the NRC's 1993 reputational rankings; here, as elsewhere, the higher-ranked departments have been more complacent.

26. Eighty-one percent of survey respondents reported that their institution has some kind of "career services office," more often designed to help graduate students find nonacademic than academic jobs; but only 14 percent of the respondents at schools with career offices knew for sure that history graduate students had found the offices to be a useful resource (versus 36 percent who reported that students did not find them useful and a full half who could not or did not answer this question).

27. See survey questions VI.9B, VI.18, VI.26, and VII.9. To be fair, most departments did make a change in at least one of these aspects of graduate education in the previous five years.

28. See, for example, Theodore Ziolkowski, "The Ph.D. Squid," American Scholar 59:2 (spring 1978): 177-95.

29. A. Howard Meneely, "Graduate Training in History," Social Education 5:1 (January 1941): 31-36, reprinted in part in the September 2002 issue of Perspectives. W. Stull Holt, a close observer of graduate education and the historical profession for decades, concluded in 1979 that "many, undoubtedly most, of the operations, thoughts and beliefs of the historical profession today are identical or closely similar to those I met and shared half a century ago." Holt, "Si jeunesse savait; si veillesse pouvait," Clio 8:2 (1979): 273.

30. See survey questions VI.4-VI.5, VI.19-VI.20, VII.3-VII.4, VII.7-VII.8, and X.11-X.12. In all cases, the representatives of larger doctoral programs tended to report somewhat higher levels of satisfaction for both students and faculty.

31. See Lillian Guerra's account of an open forum with graduate students at the 2000 annual meeting of the AHA, "'Dumb Enough to Want to Get a History Ph.D.': Views from the Trenches of Graduate Education," Perspectives 39:6 (September 2001): 31-33. Although the students had numerous and specific complaints about graduate training and disagreed with many of their own advisors' assumptions about the state of graduate education, they were "for the most part happy to be historians." This Committee's independent conversations with graduate students have pointed to the same conclusions.

32. See Lillian Guerra's account of an open forum with graduate students at the 2000 annual meeting of the AHA, "'Dumb Enough to Want to Get a History Ph.D.': Views from the Trenches of Graduate Education," Perspectives 39:6 (September 2001): 31-33. Although the students had numerous and specific complaints about graduate training and disagreed with many of their own advisors' assumptions about the state of graduate education, they were "for the most part happy to be historians." This Committee's independent conversations with graduate students have pointed to the same conclusions.

33. In spring 2002, the Committee invited all graduate student members of the AHA to comment on the preliminary results of the doctoral program survey via "An Open Letter to Graduate Students," Perspectives 40:5 (May 2002): 27 (also distributed by e-mail). The comments in this paragraph come from their responses.

34. An earlier version of the analysis presented here appeared as Philip M. Katz, "Sources of Information for Doctoral Students in History," Perspectives 39:9 (December 2001): 14-16.

35. We would like to thank the Committee for Graduate Students, which represents the interests of the AHA's graduate student members, for underscoring the importance of these issues at several points during our investigation.

36. Hoffer et al., Doctorate Recipients, 2000 .