¹ These revisions are encompassed in the following statutes: Horei lensho (hereafter H.Z.), Dajōkan Ordinance 117, 02 5, 1869Google Scholar; H.Z., Dajōkan Ordinance 203, Nov. 30, 1875; H.Z., Imperial Ordinance 17, July 22, 1878; H.Z., Dajōkan Ordinance 32, July 15, 1878.

² Ibid., arts. 1–4.

³ H.Z., Dajōkan Ordinance 35, Aug. 3, 1878.

⁴ H.Z., Dajōkan Ordinance 32, July 25, 1878, art. 3.

⁵ H.Z., Dajōkan Ordinance 1, Jan. 4, 1884. The system created by this statute was to remain as the basis of the retirement-pension system through-out the entire pre-Occupation period. Changes were made in 1890 and 1923 but they did not radically effect the basic structure. See H.Z., Diet Law 43, June 20, 1890 and H.Z., Diet Law 48, Apr. 13, 1923.

⁶ The development of the examination system can be pursued by examination of the following statutes: H.Z., Imperial Ordinances 37, 38, Cabinet Ordinances 8, 19, 20, July 23, 1887; H.Z., Imperial Ordinance 57, Nov. 5, 1887; H.Z., Imperial Ordinance 58, Nov. 7, 1887; H.Z., Cabinet Ordinance 25, Dec. 1, 1887; H.Z., Imperial Ordinances 63, 64, Dec. 24, 1887; H.Z., Cabinet Ordinance 28, Dec. 28, 1887.

A major revision of the system occurred in 1893–94, one which eliminated many of the failings of the orginal structure. This revision is encompassed in the following statutes: H.Z., Immade perial Ordinances 126, 183, 187, 197, Oct. 31, 1893; H.Z., Imperial Ordinance 54, May 24, 1894; H.Z., Cabinet Ordinance 2, May 7, 1894; H.Z., Foreign Ministry Ordinance 7, June 22, 1894. Other less important revisions occurred in 1918, 1929, and 1941. For these see: H.Z., Imperial Ordinance 7, Jan. 18, 1918; H.Z., Imperial Ordinance 15, Mar. 28, 1929; H.Z., Imperial Ordiperial nance 1, Jan. 6, 1941.

⁷ H.Z., Imperial Ordinances 61, 62, 63, Mar. 28, 1899.

⁸ For the purposes of this analysis, all the statutes published in the definitive Horei Zensho (Collected Statutes) between 1867 and 1920 were analyzed in terms of their assignment of operations the prefectural level of the bureaucracy. Statutes operacontaining such assignments were then examined on the basis of a number of variables. Briefly, these variables may be summed up by the following categories: (1) the number of categories of operations; (2) area of the categories of operations in terms of the following outputs usually associated with governmental structures—(a) rule application (administration), (b) rule adjudication (judiciary), (c) rule-making (decision-making) functions; (3) increases and decreases in categories of operations.

In this analysis categories of operations was defined as a cluster of quantitatively defined operations performed for a formally defined purpose. Rule application operations were defined as the performance of clusters of operations specified and determined by legislation and/or ordinance from the central ministerial offices and/or the Diet. Rule adjudication operations were defined as clusters of operations resulting in legitimate operations which are unspecified by legislation and/or ordinance and are related specifically to settlement of disputes over the implementation of specified and quantified legislation and/or ordinances. Ruletions; making was defined as specified and quantified clusters of operations resulting in the creation of new legitimate clusters of specified and quantified operations hitherto unspecified by existing legislation and/or ordinances.

The terminal date of the analysis was selected primarily because by this period the Japanese political structure was relatively stabilized. At the same time it represented a date by which Japan was emerging as an industrialized society and as a world power.

⁹ Tatsukichi, Minobe, Nippon gyoseiho, (Tokyo, 1936), I. 422–23.Google Scholar

¹⁰ These adjudicative functions were vested in the fu-ken sanjikai or prefectural council which was an administrative office. H.Z., Diet Law 35, May 17, 1890.

¹¹ Determination of the actual norms of recruitment and career is based on analysis of the social educational and career backgrounds of a simple random sample of 215 (25%) prefectural governors who were appointed to this office between 1868 and 1945. Prefectural governors were chosen for this analysis on the basis of several considerations: (1) to provide continuity with the first aspect of the analysis; (2) the prefectural governor was in the highest ranks of the civil bureaucracy and thus may be viewed as representing the upper civil service as a whole with regard to the norms and behavior of recruitment and career. The analysis is limited to a group of upper civil servants since the bureaucracy here is viewed as an in deciding and implementing policy. It thus “makes sense to think primarily of groups of advanced rank …” Fritz Morstein Marx, “The Higher Civil Service as an Action Group in Western Political Development,” in Joseph LaPalombara (ed.), Bureaucracy and Political Development, p. 63Google Scholar. Upper civil service is defined here as those holding either sonin, chokunin, or shinin rank in the pre-1945 civil bureaucracy. These ranks were legally designated as the upper civil service.

The population of governors was determined on the basis of: (1) all those who were appointed by the central government as governors of officially designated prefectures (ken) m tne period May 1868 to Sep. 1945; (2) the exclusion of the 273 daimyō or domain lords who retained their positions until their domains were transformed and/or amalgamated into new prefectures by the end of 1871. The incumbents were determined by use of the following sources: Tsunekichi, Ijiri, Rekidai kenkanroku (Tokyo, 1925)Google Scholar; Insatsukyoku, Naikaku, Shokanroku (Tokyo, 1886–1945)Google Scholar. The biographical data were derived from a large number of sources of which the major and most representative are: Heibonsha (eds.), Dai jinmei jiten (Tokyo, 1957–58) 10 vols.; Eikichi, Igarashi, Taishō jinmei jiten (Tokyo, 1914)Google Scholar; Hensankai, Ishia Shiryo, Gendai kazoku fūyō (Tokyo, 1929)Google Scholar; Shimbunsha, Osaka Mainichi (eds.), Gendai jinmeiroku (Osaka, 1926–1935)Google Scholar, annual; Kōshinjo, Jinji (eds.), Jinji kōshinroku (Tokyo, 1903, 1908, 1911, 1915, 1918, 1921, 1925, 1928, 1931, 1934, 1937, 1939, 1941, 1943, 1948, 1951, 1953, 1955, 1957, 1959)Google Scholar; Shimbunsha, Asahi (eds.), Asahi nenkan (Osaka, 1920–1945, annualGoogle Scholar; Nobuhisa, Kaneko, Hokkaido jinmei jisho (Sapporo, 1923)Google Scholar; Hideichi, Noyori (ed.), Meiji Taishōshi: jinbutsu hen (Tokyo, 1930), vols. 13–15Google Scholar; Takeo, Okamoto (ed.), Shiga ken jinbutsushi (Otsu: Shiga, 1930)Google Scholar; Gensuke, Matsuda (ed.), Bōchō jinshi hatten kan (Yamaguchi, 1932)Google Scholar. In addition to these a large number of other regional biographical dictionaries, prefectural histories and biographies were utilized.

¹² For the details of this conflict see Ike, Nobutaka, The Beginnings of Political Democracy in Japan (Baltimore, 1950)Google Scholar, chap. 5.

¹³ Awareness of this problem is reflected in the concern of the bureaucratic leaders over the question of a constitution and how to define and allocate the decision and rule-making functions in the period after 1868 but especially in the period from 1873 to 1881. It is interesting to note that the debate within the bureaucracy over the kind of constitutional structure to be adopted reached a climax in March, 1881 with the submission to the emperor of Okuma Shigenobu's memorandum on constitutional government. I believe that this development was directly related to the crisis of decision-making which had evolved out of the changes in bureaucratic structure and function. For a review of these events see Akita, George, The Foundation of Constitutional Government in Modern Japan: 1868–1900. (Cambridge, Mass., 1967)CrossRefGoogle Scholar, chaps. I and III.

¹⁴ Between 1868 and 1900, of those holding positions of vice-minister and above or their equivalents 88% were recruited from the samurai class. Silberman, Bernard S., Ministers of Modernization: Elite Mobility in the Meiji Restoration, 1868–73. Tucson, Arizona, 1964), pp. 118–20, 49–55.Google Scholar

¹⁵ For a fuller examination of this problem in relation to decision-making see Silberman, Bernard S., “Bureaucratic Development and the Structure of Decision-Making in the Meiji Period: The Case of the Genrō,” Journal of Asian Studies, XXVII, 1 (11 1967), 81–94.CrossRefGoogle Scholar

¹⁶ The emergence of the Jiyūminken or People's Rights Movement of the 1870's indicated very clearly that nonbureaucrats could not be trusted to foundasupport all the goals the bureaucracy had agreed upon. The Jiyūminken party leaders wanted less centralization of decision-making and balked at the immediate fiscal consequences of the government's long range support of industrial development. Ike, The Origins…, pp. 101–10Google Scholar. The fear of the bureaucratic leaders that sharing power with the political parties would undermine the agreed goals is clearly described in Akita, Foundasupporttions… pp. 58–66.Google Scholar

¹⁷ Silberman, Bernard S., Ministers of Modernization …, pp. 56–108.Google Scholar

¹⁸ Kuroda Kiyotaka (1840–1900), Matsukata Masayoshi (1835–1924), Saigo Tsugumichi (1843–1902), Oyama Iwao (1842–1916, Ito Hirobumi (1841–1909), Inoue Kaoru (1836–1915), Yamagata Aritomo (1838–1922). Saionji Kimmochi (1849–1940) and Katsura Tarō (1848–1913) are usually considered as later additions to the group.

¹⁹ As quoted in Hirota, Kurihara, Hakushaku ltō Miyoji (Tokyo, 1940), p. 353.Google Scholar

²⁰ As quoted in Kyoku, Okurashō Insatsu, Gikai seido nanajūnenshi: kenseishi gaikan (Tokyo, 1963), p. 34.Google Scholar

²¹ Riggs, Fred W., “Bureaucrats and Political Development, A Paradoxical View,” in LaPalombara, Joseph (ed.), Bureaucracy and Political Development (Princeton, 1963), pp. 122–30.Google Scholar