After the University of Chicago Economics Department’s Committee on Ph.D. Outlines and Requirements (Blough, Friedman, D.G. Johnson and Marschak) met twice, Milton Friedman, the chair of the committee, circulated a five page summary of the committee’s deliberations. This summary along with brief comments by Blough and Marschak are included in this posting.
Core sentences: “The standard should be a first-rate journal article, not a full-length book.” “The student scheduled to report at any meeting [of the thesis seminar] should prepare a written report sufficiently in advance of the meeting to permit duplication, and circulation among all faculty members and all student participants in the seminar. He might then begin the discussion with an introductory summary taking not more than, say, five minutes. The rest of the time would be devoted to critical discussion.”
___________________________
[MEMO #3, 23 MAY 1949]
THE UNIVERSITY OF CHICAGO
Date May 23, 1949
To R. Blough, J. Marschak and Department Economics
D.G. Johnson
From M. Friedman Department Economics
In re: Tentative Agreements reached by committee on Ph.D. outlines and requirements.
[p. 1] In the two meetings our committee has held so far we have concentrated on two main problems: (1) the standards to be applied to a thesis; (2) methods for getting more effective supervision, direction, and criticism of a thesis. The purpose of this memorandum is to summarize the tentative agreements reached on these points.
-
Standards to be applied to a thesis.
It was our feeling that the existing (implicit) standards for a thesis are both too high and too low: too high ex ante and too low ex post. It was agreed that we should seek to stimulate shorter, better organized, and better written theses than those ordinarily submitted. The problems here are first, to avoid simply reducing length without improving quality; second, to enforce the standard and make it a part of the mores of the Department.
In order to accomplish these purposes it was agreed (a) that a statement should be prepared in the role of the thesis for distribution to candidates; (b) that every thesis should be required to have a central core not to exceed roughly 15,000 words.
(a) Role of the thesis
The thesis, in our view, is to be viewed primarily as part of the training of the economist, not as a means of securing additions to knowledge. Any additions to knowledge is a welcome by-product, not a major objective. Up to the point at which he writes a thesis, the student has been concerned primarily with absorbing substantive material, acquiring tools and becoming familiar with techniques of analysis. He has only incidentally applied [p. 2] these techniques. Equally important, he has had little occasion to acquire absolute standards of quality; most of his written work has been of a “one-short” variety involving doing his best once and then being through with it. He has not had the experience of re-doing a thing again and again until it is satisfactory in an absolute sense and not merely the best he can do in an hour or a week.
The role of the thesis is to round out the student’s education by remedying these deficiencies. More specifically it should:
(1) Give the student training in research by “doing” and instill in him absolute standards of quality in research.
(2) Deepen 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 process, he should think through the material he has been subjected to and make it his own.
These objectives affect both the choice of topic and the character of the thesis. The topic should be chosen from the point of view less of novelty or importance than of the contribution it can make to the student’s education; the opportunity it offers for improving and expanding his capacities. As a general matter, this suggests topics sufficiently narrow and specific to permit the student to do a thorough and exhaustive piece of work in the time available. It argues against broad general topics in which maturity and judgment are the prime requisites.
To accomplish these objectives, the final thesis should satisfy exceedingly high standards of quality; this is far more important than quantity. As a regular matter, it should be expected that numerous re-writings of the thesis will be required, that an absolute standard of excellence rather than a labor-theory of value will be applied. This means that at least the central core of the thesis must be relatively brief. The standard should be a first-rate journal article, not a full-length book.
[p. 3] It should be emphasized that this objective is unlikely to be attained if the students sets out from the beginning the objective of writing not more than, say, 50 pages. A final acceptable thesis containing 50 pages will ordinarily require the writing of several hundred pages in the process. Indeed, it is frequently easier to write 300 pages on a topic than to write 50 pages of high quality, and the 300 pages will frequently be a preliminary step in getting to the 50 pages.
(b) The scale of the thesis
It was agreed to recommend that every thesis should be required to contain a central core of not more than roughly 15,000 words. This central core is to contain an integrated development of the topic and to be self contained. It may, however, be supplemented by such documentary evidence as is required to support it in the form of supporting appendices.
The central core should, in general, not give much space to the character of the problem, earlier work on the problem, and the like; these belong in the supporting appendices if anywhere. It should concentrate on the original material developed by the writer. It must contain an economic analysis of the problem tackled, not simply summarize date, report views, or describe events. In this context, of course, economic analysis is to be interpreted broadly, not as synonymous with technical economic theory.
One further reason for keeping theses to this scale is the desirability of having every member of the faculty read every thesis and vote for or against its approval. This is not at present feasible but might become so if the scale of the thesis were restricted.
-
Methods for getting more effective supervision, direction, and criticism of theses.
Our chief recommendation on this topic is that there be established a thesis seminar. This seminar should be attended as a regular matter by all students writing theses in residence, by as many faculty members as can find [p. 4] it possible to attend, and, in any event, by the faculty members on the thesis committee of the student reporting at a particular session. Ideally, some one or more faculty members should have direct responsibility for it as part of his teaching load.
The student scheduled to report at any meeting should prepare a written report sufficiently in advance of the meeting to permit duplication, and circulation among all faculty members and all student participants in the seminar. He might then begin the discussion with an introductory summary taking not more than, say, five minutes. The rest of the time would be devoted to critical discussion.
It might be expected that a student would ordinarily appear before the seminar twice: once early in his work for a discussion of the topic and its possibilities; once, toward the end, for a discussion of his results.
This thesis seminar might be integrated with two other steps in the thesis procedure with which there is at present some dissatisfaction: (a) admission to candidacy, (b) the final examination.
The first appearance of the student before the seminar, and the paper prepared for that purpose, might also be used as a basis for deciding on admission to candidacy. At present, it is the general feeling that we have inadequate evidence on which to judge suggested theses. The suggested change in the scale of the thesis opens up the possibility that more time can be spent in the preparatory stages and more can be asked for from the student in the way of supporting evidence. Something of the scale of a term paper I perhaps not too much to ask.
Dissatisfaction with the final examination arises from a different source. The exam is in fact a pure formality, in view of the stage at which it comes. Candidates are in practice almost never failed at that stage. Yet the candidate is not told that it is a pure formality; he regards it as a crucial and important test. The entire procedure has an element of sadism about it.
[p. 5] In place of dispensing with the final exam, might the second appearance of a candidate before the thesis seminar take its place, not in the sense of an occasion for final approval of the candidate, but in the sense of a public exhibition, as it were, testifying to the candidate’s stage of development. Final approval of the thesis would be based on the decision of the thesis committee plus a poll of the entire faculty.
___________________________
[MEMO #4, 6 June 1949]
THE UNIVERSITY OF CHICAGO
Date June 6, 1949
To (R. Blough Department of Economics
(M. Friedman
(D. G. Johnson
From J. Marschak Department of Economics
In re: Ph.D. Theses.
M. Friedman’s draft of May 23 seems to express well the views of the Committee, with the following amendments suggested:
p. 3, par. 1. I propose to cancel this paragraph. To write 300 pages, later to be condensed to 50, is one possible method, but neither the most frequent nor a particularly commendable one. I think this should be left to the students and to their immediate advisers.
p. 3, par. 3, sentence 3. I suggest (suggested insertion underlined): “It must contain a precise statement of the problem and its economic analysis…” It is often unclear what the thesis writer proposes to prove.
p. 4, par. 3. I suggest (suggested insertion underlined):
“…once, early in his work for a discussion of the topic and its possibilities, on the basis of a brief circulated report (on the scale of a term paper); once, toward the end, for a discussion of his results, on the basis of a more detailed report and possibly of the draft of the ‘central core’ of the thesis itself.”
(signed)
Jacob Marschak
JM/fs
___________________________
[MEMO #5, 9 June 1949]
THE UNIVERSITY OF CHICAGO
Date June 9, 1949
To M. Friedman Department Economics
D.G. Johnson
J. Marschak
From Roy Blough Department Economics
In re: Ph.D. Theses.
Milton Friedman’s draft of May 23 seems to be a correct reflection of the views of the Committee. I am in agreement with Mr. Marschak’s suggestions of June 6.
The following afterthoughts are presented for discussion:
- I suggest eliminating the reference to sadism in the final examination. I do not so interpret the examination and, in any event, think the reference is a reflection on our colleagues and predecessors.
- The reference to frequent rewriting of the theses seems to put the emphasis on writing style. While this is important, there are other factors such as precise formulation of purpose, imaginative use of techniques, and logical organization.
- I suggest eliminating any implied criticism of the length of theses except as a criticism of verbosity. The kinds of theses vary so widely that comments which are applicable to one kind of subject would not apply to others. This is not any criticism of the recommendation for a central core for every thesis.
- One danger in suggesting that theses should approach articles rather than books in size is that some substitute will then have to be found as evidence that the student has done the equivalent of a year’s work, in general, on the dissertation. It would be most unfortunate if a few glib students should write plausible theses in a month or two and have them accepted. These students would not receive the research training which the thesis is intended to give, and the effects on the morale of other students might be disastrous.
Source: Hoover Institution Archives. Milton Friedman Papers, Box 79, Folder 5 “University of Chicago Minutes, Ph.D. Thesis Committee”.
Image Source: Roy Blough photo from University of Chicago Photographic Archive, apf1-00758, Special Collections Research Center, University of Chicago Library.