² Hiroshi, Hazama, Nihon Rōmu Kanriihi Kenkyū (Studies in the History of Japanese Labor and Management Relations) (Tokyo: Daiamondosha, 1964);Google ScholarTaira, Koji, Economic Development and the Labor Market in Japan (New York: Columbia University Press, 1970);Google ScholarMarsh, Robert and Mannari, Hiroshi, “Lifetime Commitment in Japan: Roles, Norms, and Values,” American Journal of Sociology, 76, 5 (March, 1971). pp. 795–812;CrossRefGoogle ScholarCole, Robert, Japanese Blue Collar (Berkeley: University of California Press, 1971);Google Scholar and Evans, Robert, Jr., The Labor Economies of Japan and the United States (New York: Praeger Publishers, 1971).Google Scholar

³ The Representative Study was conducted in 1967 and used a two-stage random sample design to obtain a representative sample of university scholars of the rank of lecturer or above (assistants excluded); 805 scholars (56 per cent of the sample and closely matching the population in terms of type of university, rank, age, and field) returned usable questionnaires detailing their careers and attitudes concerning a wide variety of university activities. The Mobile Man Study of 1971 was directed to a randomly selected sample of 500 scholars who had in the previous five years joined any one of twenty-two universities selected to represent the full diversity of Japanese four-year universities; 220 finally returned the questionnaire for a corrected response rate of 44 per cent with poor cooperation at three of the selected institutions.

⁴ Sunao, Ogosc, “Furui Kcnkyūsei o Daha Scyo” (Let's Tear Down the Outdated Research System), Asahi Jānaru, VIII, No. 30 (July, 1966), p. 60.Google Scholar

⁵ For an interesting interpretive history of Japanese higher education, see Michio, Nagai, Higher Education in Japan: Its Take-off and Crash. Dusenbury, Jerry, trans. (Tokyo: University of Tokyo Press, 1971).Google Scholar The statistics for Figure 1 and the text are taken from Ministry of Education, Education in 1968–70: Japan (Tokyo: Ministry of Finance Printing Office, 1971)Google Scholar and Ministry of Education, Educational Standards in Japan (Tokyo: Ministry of Finance Printing Office, 1971).Google Scholar

⁶ Educational Standards, p. 35.

⁷ Two recent discussions in English of the private sector are Cummings, William K., “The Japanese Private University,” Minerva, XI, 3 (July, 1973), pp. 348–371;CrossRefGoogle Scholar and Pernpel, T. J., “The Politics of Enrollment Expansion in Japanese Universities,” Journal of Asian Studies, XXXIII, 1 (November, 1973). PP. 67–86.Google Scholar

⁸ This information was supplied by Tokoyama Tsuncsaburō, President of the Private School Promotion Foundation, in a private conversation. For additional evidence on the improvements by Japan's leading independent student of private university fiscal matters, see Ken, Ogata, “Shiritsu Daigaku Kyāshokuin no Nenkin Jyātai: 1972 Nendo Zemt Chāsa Kara” (The Annual incomes of Private University Staff: Based on a Survey by a Seminar), Keizai Shirin, Vol. 41, 2 (April, 1973), pp. 1–36.Google Scholar

⁹ This survey titled Daigafyt Seido no Kaikaku ni Kansuru Ankēto (Questionnaire on University Reform) was distributed in 1969 by the Central Council for Education to all of Japan's university professors. The return rate was only 23 per cent, but reasonably distributed by type of university and field. The results were supplied by the Ministry of Education.

¹⁰ Kokuritsu Daigaku Kyōkai Daigaku Unci Kyāgikai (Committee on University Administration of the Association of National Universities), Daigaku Mondai ni Kansuru Chōsa Kenkyū Hōkokutho (Report of Investigation on University Problems) (Tokyo: Kokuritsu Daigaku Kyōkai, June, 1971), pp. 16–17.Google Scholar

¹¹ Chūō Kyōiku Shingikai (Central Council for Education), Kyōik” Kaikaku no tame no Kihonteki Shisaku (Basic Policy for the Reform of the Educational System) (Tokyo: Monbushō, June, 1971), pp. 68–69.Google Scholar

¹² The American figures arc from unpublished data supplied by Talcott Parsons and Gerald Piatt in connection with their ongoing study of the American academic profession. The British figures arc from Halsey, A. H. and Trow, M. A.. The British Academics (Cambridge, Mass.: Harvard University Press, 1971), pp. 225ff. The reader may have observed that British elite scholars are more mobile than “average” scholars, American elite scholars arc about as mobile as “average” scholars, and Japanese elite scholars are less mobile than “average” Japanese scholars. Given different age distributions, definitions of universes, and measurement errors these differences may not reflect the real situation. However, they arc plausible when we consider the different structures of the respective marketplaces, a task touched on later in this paper.CrossRefGoogle Scholar

¹³ Calculated from Zenkoku Daigaku Shokuin-roku (Tokyo: Kōjunsha, 1970).Google Scholar

¹⁴ Halsey and Trow, op. cit. It should be noted that the Japanese sample docs not include the lowest ranking staff (assistants) whereas the English sample docs include assistant lecturers and “others” (a term that is not defined but probably refers to some low-ranking tutors) and a 28 per cent number of those trained elsewhere who work at Oxbridge occupy these two low rank positions whereas only 11 per cent of those trained at Oxbridge and now employed at Oxbridge hold these positions.

¹⁵ Berelson, Bernard, Graduate Education in the United States (New York; McGraw-Hill Book Company, Inc., 1960), pp. 115–116.Google Scholar

¹⁶ Examples of the traditional approach to the academic marketplace arc Sunao, Ogose, “Furui Kenkyŭsei o Daha Seyo,” op. cit., 53–66;Google ScholarMichiya, Shinbori, Ninon No Daigaku Kyōyū Shijō (The Japanese Academic Marketplace), (Tokyo: Tōyōkan Shuppansha, 1964);Google Scholar and Chie, Nakane, Japanese Society (London: Wiedenfeld and Nicol-son, 1970).Google Scholar We have tried to distill the arguments of these and other researchers in a coherent synthesis: no doubt, specific features of each of the respective arguments may be somewhat misrepresented, but hopefully the spirit of these arguments is reflected in our synthesis. The classic example in English of the traditional approach applied to blue collar markets is Abegglen, James C., The Japanese Factory (New York: Free Press, 1958).Google Scholar

¹⁷ We borrow this constellation of Japan's traditional values from Robert N. Bellah and from Nakanc Chie, op. cit., whom Bellah relies on; see Bellah, , “Continuity and Change in Japanese Society,” in Barber, Bernard and Inkeles, Alex. eds., Stability and Social Change (Boston: Little, Brown, and Co., 1971), especially page 382.Google Scholar

¹⁸ Nakane, op, cit., p. 133.

¹⁹ Shinbori, op. cit., provides us with a fascinating statistical account of this process. On page 94 he presents a table showing the regional differentiation of different academic cliques—Tokyo in the Kanto area, Kyotō in the Kansai, Hiroshima in Chūgoku and Shikoku, Kyūshu in Kyūshu, Hokkaidō in Hokkaidō, and Tōhoku in Tōhoku.

²⁰ In particular. Berghe, Pierre Van den, Academic Gamemanship: How to Make a Ph.D. Pay (New York: Abelard-Schulman, 1970);Google Scholar also see Cornford, F. M., Microcosmographia Academical Being a Guide for the Young Academic Politician (Chicago: University of Chicago Press, 1945).Google Scholar

²¹ Caplow, Theodore and McGee, Recce J., The Academic Marketplace (Garden City, N.J.: Anchor Books, 1965), p. 117.Google Scholar

²² Cummings, Marketplace, p. 309.

²³ On France, Clark, Terry N. and Clark, Priscilla P., “Le Patron et son cercle; clef de l'Univcrsité française,” Revue Française de Sociologie, XII (1971), pp. 19–39.CrossRefGoogle ScholarOn Italy, communication from Clark, Burton R. as well as Burn, Barbara B., The Emerging System of Higher Education in Italy. Conference Report No. 1 (New York: Interna- tional Council for Educational Development 1973).Google ScholarOn Belgium, see Fox, Renee C., “Medical Scientists in a Chateau,” Science, Vol. 136 (1962), pp. 476–83.CrossRefGoogle ScholarOn the U.S., Gouldner, Akin, “Cosmopolitans and Locals: Toward an Analysis of Latent Social Roles—II,” Administrative Science Quarterly, II (March, 1958), pp. 444–80Google Scholar and Caplow and Mc-Gee, op, cit., pp. 168–69.

²⁴ In all fairness, we must admit that we may be going further in interpreting Shinbori's analysis than he might accept. He may not view these structures as a supplement to bolster the traditional explanation, but rather as the core of that explanation.

²⁵ On the process of adopting the chair in Japan, sec Tcrasaki Masao “‘Kozasei’ no Rekishitcki Kenkyū Josetsu—Nihon no Baai (1)” (Historical Review of the Japanese Chair System (1)), Daigaku Ronshū. 1 (1972). PP. 1–10 For a discussion of the chair in Western universities, see Ben-David, Joseph, “Universities and Academic Systems in Modern Societies,” European journal of Sociology 111 (1962), pp. 45–84.CrossRefGoogle Scholar

²⁹ These differences axe made clear in Michiya Shinbori, “Comparative Study of Career Patterns of College Professors,” pp. 284–296.

²⁷ The only serious historical investigation of the Japanese chair is Tcrasaki, op. cit. Our comments arc based on interviews and inferences from our reading of several of the standard histories of Japanese higher education; of particular interest was Toshiaki, Okubo, Nihon no Daigaku (The Universities of Japan). (Tokyo: Sōgcnsha, 1943).Google Scholar The question of why these structures were adopted still remains to be investigated. It is important to recognize (hat not all universities arc organized around the chair; indeed nearly two-thirds are organized around the more flexible course system. See Ikuo., Amano “Kokuritsu Daigaku” (National Universities), in Yoshihiro, Shimizu, ed., Nihon no Kōtō Kyōiku (Japanese Higher Education) (Tokyo: Daiichi Hōki Shuppan Kabushiki Kaisha, 1968).Google Scholar

²⁸ For some evidence on this point, see Nakayama, Shigeru, “The Role Played by Universities in Scientific and Technological Development in Japan,” Journal of World Hittory, IX (1965), pp. 340–62.Google Scholar

²⁹ An important statement of the labor market approach is Rottenberg, Simon, “On Choice in Labor Markets,” Industrial and Labor Relations Review, IX, 2, (January, 1956), pp. 183–199.CrossRefGoogle Scholar The labor market approach has not been applied to the Japanese academic marketplace, though traces of it can be found in policy statements by educational leaders; for example, see Ichirō, Katō et al. , “Daigaku Kyokan no Taigfi Mondai” (The Problem of the Salaries of University Teachers), Jurisuto, No. 356 (Oct. 15, 1966), pp. 78–92.Google Scholar The related approach of social exchange has been used by Harumi Btfu in “Power in ihc Great White Tower,” paper read in “The Ethnography or Power: Oceania and Asia” section of the American Association for Advancement of Science meetings. San Francisco, February 25, 1974. The labor market approach has been used in several attempts to explain behavior in the American academic marketplace; see Brown, David G., The Mobile Pro-lessors (Washington, D.C.: The American Council on Education, 1967)Google Scholar or Marshall., Howard D.The Mobility of College Faculties (New York: Pageant Press, Inc., 1964).Google Scholar

³⁰ In this and the following section, we will report several empirical generalizations which arc documented in Cummings, Marketplace, especially Chaps. 3 and 4.

³¹ The discrepancy by a comparison of relative prestige and income of occupations is illustrated in Shigeki Nishihira, “Le Prestige Social des Dif-ferentes Professions,” Revue Francaise de Sociologie, IX (1968), 555.Google Scholar

³² Degree of openness is a relative matter. The OECD examiners of Japanese education outlined two Western models of “Open” recruitment and concluded that Japanese universities relied on neither of these, but rather used a closed procedure of “automatic succession within the chair unit” (p. 85). The examiners were speaking of the ideal procedures found in the respective systems, and we feel they overstressed the closedness of the Japanese case. However, the basic contrast along an open— closed dimension seems reasonable. Reviews of National Policies for Education; Japan (Paris: OECD, 1971).Google Scholar

³³ An adjustment to the staffing problem is the part-time teacher arrangement, especially common in private institutions. Indeed, in some private institutions well over one-third of all classes are taught by part-time teachers. Interestingly, these part-time teachers also have tenure in the sense that once the employing institution hires a teacher it cannot sever its formal tic. But the employer can reduce the number of courses it asks a part-timer to teach (down to none) and the pay which is computed according to the number of hours in the classroom.

³⁴ The obvious reason for this undifferentiated picture of an ideal working place is that most young scholars were trained at former imperial universities by professors who likewise were trained at these institutions. All of the imperial universities were designed according to a common plan and with a common Germanic view of academic work; furthermore these former imperial universities where most scholars spent their formative years had all of the desired attributes—the best research conditions, prestige, and location.

Private institutions train only a small proportion of graduate students, but we do find greater variation in the preference orders of these students. A greater proportion give high preference ranking to such attributes as cordiality of colleagues, tradition, and student spirit which we have included in the category of social atmosphere. Nevertheless, the overall preference order among private university professors is not very difTerent from that of national university professors simply because the vast majority were trained at national universities.

³⁵ The one obvious exception is the distinguished professor of a national university who moves amakudari style to a private university at the zenith of his career and after he has served long enough to gain the full retirement benefits. It is of interest that moves of this kind are less than 10 per cent of a)! moves each year, despite the claims of some observers that these post-retirement moves constitute virtually the only form of mobility in Japanese academia.

³⁶ As our sample ranges in age from several scholars in (heir late twenties to a few over seventy, some will wonder if the temporal comparisons in the text are valid. What wc concentrate on in the comparisons are experiences common to all scholars whether young or old—i.e., securing their first job and then the first two status changes after that. Nearly all of the scholars in our sample were old enough to have had that much career change. The youngest in our sample tend not to have more than two or three status changes, and our argument does not depend on what happens in the advanced stages of the careers of scholars in either the prewar or postwar period.

³⁷ We also investigated the careers of elite scholars sampled from the 1937 Jinji Kōshinroku. Comparisons of their careers with those of the postwar sample reported in Table 2 above indicated that the prewar elite scholars had been less mobile. As earlier noted, the conditions affecting the careers of elite scholars are different than those affecting ordinary scholars. In particular, elite scholars in the prewar period were less dispensable at their places of employment as they were fewer in number and had much greater responsibility in education and administration. While the general demand for academic talent was stronger in that period (greater elasticity as well), the elite scholars had to forego opportunities due to their moral obligations to their employing institutions; see Cummings, Marketplace, pp. 168ff.

³⁸ The statistics are reported in Cummings, Marketplace, pp. 367ff. The five other institutions following the University of Tokyo are Kyoto. Tōhoku, Tokyo Kyōiku. Kyūshū, and Hokkaidō.

³⁹ See Footnote 19.

⁴⁰ Despite two reviews of the evidence, there is no clear finding on the consequences of inbreeding in Japan; see Shinbori, Michiya, “The Academic Marketplace in Japan,” The Developing Economies, VII, 4 (December, 1969), pp.Google Scholar 637ff.; and Cum-mings, Marketplace, pp. 340–347. Lowell L. Hargcns and Grant M. Farr after a careful analysis applying controls for prestige of current depart-ment and year of Ph.D. conclude that inbred scientists “tend to be slightly less productive in terms of quantity and quality of publications than their non-inbred colleagues” in “An Examination of Recent Hypotheses About Institutional Inbreeding,” American Journal of Sociology. Vol. 78, No. 6 (May, 1973). P. 1393.Google Scholar

⁴¹ Cummings, Marketplace, p. 154.

⁴² We used the following pieces of information to construct a quantitative index of “quality”: whether the institution had trained a large number of scholars or not, whether it had university status before World War II or not, whether it has a graduate school or not, whether it has affiliated research institutes or not, its student-teacher ratio, and finally the extent to which it attracts students from across the nation. Details arc presented in Appendix 3 of Cummings, Marketplace.