The University of Chicago Department of Economics was dissatisfied with its procedures for appraising and approving dissertation projects in late 1948 and a committe was formed to make recommendations with Milton Friedman as its chairperson. Here I post T.W. Schultz’s official memo naming the members of the committee and Milton Friedman’s initial memo to the committee clearly signalling his intention of having a major rethink about what a Ph.D. thesis is supposed to be about.
___________________________
[Memo #1, 10 Dec 1948]
THE UNIVERSITY OF CHICAGO
Date December 10, 1948
To Mr. Friedman, Mr. Blough, Mr. Marschak, Department Economics
Mr. Johnson
From T. W. Schultz Department Economics
The faculty of the Department of economics authorized a committee to prepare a memorandum setting forth the problem of students’ Ph.D. outlines and the procedure to be followed by the Department in appraising and approving Ph.D. thesis projects, including the type of outlines and supporting materials that a student should submit to the Department for its use when it passes upon the petition to admission to candidacy.
May I ask you to serve as members of this committee with Professor Friedman acting as chairman?
The report should be directed to the Department to be circulated well in advance of the departmental meeting in which it is to be considered.
___________________________
[Memo #2, early 1949]
[Undated, written sometime after the Schultz memo of December 10, 1948 and the Friedman memo of May 23, 1949 that followed two meetings of this Committee which had taken place.]
TO: R. Blough, J. Marschak, G. Johnson
FROM: Milton Friedman
SUBJECT: Committee on Ph.D. Thesis Outlines and Requirements
The purpose of this memorandum is to provide a basis for discussion by our committee, of which I am chairman. I have been derelict in my duty in not having prepared it much earlier, or not having called a meeting earlier.
-
Our Assignment
To refresh our memories, I quote from Mr. Schultz’s note establishing the committee: “To prepare a memorandum setting forth the problem of students’ Ph.D. outlines and the procedure to be followed by the Department in appraising and approving Ph.D. thesis projects, including the type of outlines and supporting materials that a student should submit to the Department for its use when it passes upon the petition to admission to candidacy.”
Interpreted literally, this assignment would limit us to the steps up to and including admission to candidacy, and would exclude consideration of the characteristics of the thesis itself and the criteria used in its acceptance. Since it seems to me the earlier stages cannot properly be judged except in terms of the desired end product, I suggest that, at least in our own discussions, we interpret the assignment more broadly to include all problems associated with the thesis requirement.
-
Present Procedure
a. Admission to candidacy. As I understand it, we have no very formalized procedure or requirements. Students typically discuss possible thesis topics with one or more faculty members, construct outlines of the projected thesis, ordinarily get the reaction of one or more faculty members to it, revise it accordingly, and then formally submit the thesis topic and outline to the Department for approval and admission to candidacy. The submitted outline is occasionally extremely detailed, occasionally very general, and is sometimes accompanied by a general statement of objective and purpose, sources of material for the thesis, etc.
b. Thesis requirements. Aside from the general and vague requirement that the thesis be an “original contribution to knowledge”, we have, so far as I know, no concrete standards for theses. Among ourselves, we have frequently expressed the view that short theses of high quality were desirable and to be promoted, and have bemoaned the tendency on the part of students to prepare lengthy, pedestrian, theses. It is my feeling, however, that the students themselves think of the thesis in terms of a full-length book, and feel that quantity is an important requirement.
The procedure for guidance of theses is informal and vague. The student ordinarily consults separately with the members of his committee as he feels the need to do so.
-
The Immediate Reason for a Committee
The immediate occasion for the appointment of a committee to consider the problem is primarily the feeling of frustration and incompetence we all feel when we are required to consider thesis topics and outlines and to approve admission to candidacy. The topics are often, if not typically, vague and broad, the outlines have the appearance of being “dreamed up” along rather formal lines in order to get approval rather than of being really working outlines providing a pattern for work or a real prediction of the final organization of the thesis. We are typically reduced to approving or disapproving the tesis larely on the basis of our knowledge of the ability of the student submitting the outline rather than on the merits of the project itself.
-
The More Fundamental Problem
It seems to me that the dissatisfaction with the procedure of approving admission to candidacy reflects a more basic problem—the function of the thesis in the education of the students and the best means of accomplishing that function. I feel that we will make more progress on our particular assignment by considering afresh the general problem.
It is my own feeling that nothing has done so much in the United States to degrade standards of research in economics as the Ph. D. dissertation in its existing form. (These comments do not apply in any special sense to Chicago—indeed, it seems to me that our record in this respect is outstandingly good). The standard which has, in principle, been set for the dissertation is that it be a major piece of work making an original contribution to the field, the model being a book of substantial magnitude. The usual graduate student, expected to begin his dissertation after two years of graduate work and supposed to be able to complete I in another year, is not at all prepared to do a piece of work of this character or quality in the time allotted. He does not have enough background in the field, or broad enough experience, and even if he had, he could hardly complete the dissertation in one year. Equally important, even if the student could do it, faculty advisers would find it impossible to supervise properly more than one or two studies of this magnitude and scope. Proper supervision would mean applying to the work the standards they would apply to their own work; it would mean repeated and detailed consultations with the student, word-by-word and sentence-by-sentence criticism of drafts of pieces of the thesis and of the entire thesis, some independent checking on the student’s work, etc.
The result is naturally a compromise. Faculty advisers do not provide the supervision and critical guidance required, they do not and cannot be expected to go over manuscripts in great detail and require that it be rewritten repeatedly until it meets a high standard. Even aside from the time and effort required, competition prevents such a course of action. The Ph.D. is something of a trade-union card, competition from other schools and the fair treatment of our own graduates requires that they be able to get one on terms that are not intolerably stiffer than those at other institutions. The result is that the theses all of us accept are typically pretty poor products, poorly organized, and full of poor grammar and writing, to say nothing of bad economics and analysis. The student who has a dissertation of this type accepted not only fails to get the training in economic research the thesis should provide, he also goes away, at least to some extent, with the idea that this is the kind of work that is done in economics and that is acceptable and respectable. In latter years, he is not unlikely to produce a flood of additional work no higher in quality than his original effort and even more useless since it does not even provide a trade-union card.
-
The Role of the Thesis in the Education of the Student
There are a number of different functions that can be assigned to the thesis in the educational process:
(a) To give the student training in research by “doing” and some feeling for standards of quality in research.
(b) To sharpen the student’s knowledge of the techniques and subject matter he has acquired in course work by requiring him to apply what he has learned to a particular problem in the belief that in the process he will be forced to think through the material he has been subjected to and make it his own.
(c) To establish habits of work and some feeling for research, in the hope thereby of stimulating him to do work on his own in latter years.
(d) To give him the unquestionably important experience of carrying through to completion a major piece of work.
The thesis might also be viewed, not solely as a part of the educational process, but also as a means of advancing knowledge in economics. I am myself inclined to give this little or no weight. At the stage at which students are not now expected to write their theses, not one student in a hundred is capable of making a “real” contribution to knowledge. Any contribution to knowledge ought in my view to be considered a welcome by-product, not a major objective.
Of the objectives listed, only the first two seem to me capable of accomplishment, with the present general standards about the stage in his career at which the student is expected to write his thesis and the time he is expected to devote to it. The last two, and particularly (d) would require something of a revolution of these standards.
-
Possible Solutions of the Fundamental Problem
There seem to me only two directions in which one can proceed to solve the fundamental problem if one takes as given roughly the present student-faculty ratio.
(a) One approach would be to restrict the Ph.D. degree to many fewer persons and to make it mean something very different from what it now means. As I understand it, this is more or less the approach followed in the Scandinavian countries where the Ph.D. is ordinarily no granted except for a major piece of work done by a man ten, fifteen, or more years after he has begun his professional career. This approach, while promising and desirable if it could be followed, does not seem to me feasible. It consists essentially in saying that one ought to establish a more advanced degree than the present Ph.D. It still leaves the problem of an intermediate degree like our present Ph.D., which would be a mark of certification that an individual is ready to begin his scientific career. It seems hardly possible for one school to do so or to overturn our established custom that a thesis is part of the attainments certified to by such a degree.
(b) The other alternative that seems to me to be open is to make the professed standard of the Ph.D. more modest while raising the attained standard. Instead of a book, the standard would be a journal article. In a way, this does not involve any change, since I do not believe there is anything in our present rules which would prevent us from accepting the equivalent of a journal article as a thesis. However, unless we explicitly make an effort to change our standard and to set a different standard for our students, I doubt very much that they, or we ourselves, will depart from the standard of a book.
What I have in mind is that we should emphasize that the requirement for the Ph.D. would be satisfied by a piece of work not to exceed a specified number of pages in length and of a quality suitable for publication in a professional journal—whether actually published or not is immaterial. The emphasis should be on quality of performance, not on quantity. The expectation would be that the faculty advisers could really go over a piece of moderate length in great detail, that they could if necessary require it to be rewritten any number of times without imposing too great a hardship on the student. It could further be expected that a larger number of members of the faculty would be led to read the thesis before final acceptance, and that in this way higher standards of quality would both be imposed and actually effected. I should be inclined myself to set something like fifty double-spaced typewritten pages as the absolute maximum limit on the size of any dissertation.
-
The Problem of the Thesis Outline
If we were to follow the line just suggested, it seems to me we could appropriately require higher standards in the thesis outline itself. Instead of the present brief and formal statement, we could require something of the order of a brief term-paper. This paper could be expected to contain three items as a minimum: (1) A brief statement of the problem; (2) a succinct but reasonably comprehensive summary of existing literature on the problem; (3) a fairly precise statement of the particular respects in which the student expects to extend or supplement the existing literature. Whether it contained an outline of the present form seems to me immaterial. In addition there ought to be a flat prohibition on any attempts to “justify” the topic in terms of its path-breaking importance for economic science. If we set the training of students as the primary objective, topics should be judged primarily in terms of the training the student will get, only secondarily in terms of their importance to economics.
It should be expected that the student will in general have gone over this statement with some faculty member and have gotten tentative clearance from him.
This is not a very specific recommendation, and I am hopeful that something better will come from the other members of the committee.
Beyond admission to candidacy, there are a number of additional possibilities we should investigate. I mention them only briefly.
(a) There seems to me considerable merit in the suggestion that has been made by Koopmans that the committee as a whole should meet with the candidate shortly after admission to candidacy so that there can be a meeting of minds on the direction his work should take.
(b) I have the feeling that much could be gained by getting the students to help one another by criticism and discussion. This would be valuable training both for the critic and the criticized. Could we set up some sort of a seminar for students writing their theses? In such a seminar, a student would be expected to submit something in written form, duplicated so that the other members have copies in advance. Some students now get the benefit of such discussion through the Cowles Commission and Agricultural Economics groups. Ought we to extend it to all? Or are informal groupings really more effective?
Source: Hoover Institution Archives. Milton Friedman Papers, Box 79, Folder 5 “University of Chicago Minutes, Ph.D. Thesis Committee”.
Image Source: Clipping from a photograph from Hoover Institution Archives (Milton Friedman Papers Box 115) in online Wall Street Journal (18 Oct 2012): Dalibor Rohac’s review of The Great Persuasion by Angus Burgin.