¹ Throughout this essay, all Japanese names appear with the last name first.

² For the major contributions in the debate, see: Miura, Hiroyuki, Hōseishi no kenkyū (A study on the history of the legal system) (Tokyo: Iwanami Shoten, 1919), 834–909Google Scholar; Shiba Kentaro's two-part essay, “Shōji kumiai toshite no za no kigen” (The origins of the za as commercial associations), Shigaku Zasshi, vol. 37, no. 1 (1926), 20–46Google Scholar and vol. 37, no. 4 (1926), 341–361; Moto'o, Endo “Shokunin no soshiki toshite no za ichi kōsatsu” (An examination of the za as the organization of craftsmen), Shakai Keizai Shigaku, vol. 3, no. 2 (1933), 115–143Google Scholar; Tokuzō, Fukuda in his “Za ni tsuite no zakken” (Sundry views on the za) which appeared in two parts in Kokumin Keizai Zasshi, vol. 10, no. 6 (1912), 889–903Google Scholar and vol. 11, no. 1 (1911), 19–36. A critical review of the debate, not included in this essay because of its length, is available from the author on request.

Four general works on the za were useful in writing this essay: Hiraizumi, Tōrn, “Za kanken” (“My views on the za”), Shigaku Zasshi, vol. 28, no. 12 (1917), 1271–1278Google Scholar; Toyoda, Takeshi, “Za o meguru ronsō no tenkai” (The development of the debate on the the za), Rekishigaku Kenkyū, vol. 1, no. 4 (1934), 291–297Google Scholar; Toyoda, Takeshi, Toshi oyobi za no hattatsu (The development of cities and the za) (Tokyo: Chūō-Kōronsha, 1948)Google Scholar; and Akamatsu, Toshihide, “za ni tsuite” (On the za), Shirin, vol. 37, no. 1 (1954), 1–25Google Scholar.

³ For example, Wakita, Haruko, Nihon chūsei shōgyō hattatsushi no kenkyū (A study on the development of commerce in medieval Japan) (Tokyo: Ochanomizu Shobō, 1969), 235–252Google Scholar, though the author does not offer an explicit definition, these pages clearly indicate that she had in mind a definition approximating the one offered here.

⁴ Some specialists of the period read tomobe as shinabe. As is the case with scores of words from this period, no one can authoritatively say how this designation of skilled craftsmen was pronounced. This essay chose tomobe following Kokushi Kenkyushitsu, Bungaku-bu, Kyoto daigaku (The Research Center for National History, Department of Literature, University of Kyoto), ed., Nihonshi jiten (A dictionary of Japanese history) (Tokyo: Sōgenshinsha, 1962), 837Google Scholar. This dictionary also contains a succinct description of the tomobe on page 374. For a useful and detailed description of the tomobe, see: Takigawa, Seijirō, Ritsuryō jidai no nōmin seikatsu (The life of peasants during the Bitsuryō period) (Tokyo: Tōkō Shoin, 1969), 18–22Google Scholar. This is a reissue of the book published originally in 1926.

⁵ Takeshi, Toyodo, A History of Pre-Meiji Commerce in Japan (Tokyo: Kokusai Bunka Shinkōkai, 1969), 101Google Scholar.

⁶ Many Japanese writers refer to temples and shrines collectively as jisha or simply as temples while meaning both. This is done in the quotation taken from Miyamoto below in this paragraph in the text. Thus, throughout this essay when meaning temples and shrines, we mention only the temples. This has the advantage, especially later in the essay, of allowing reference to those who sanctioned the za as “the court nobles and the temples” instead of repeating “the court nobles, the temples, and the shrines.”

⁷ Miyamoto, Mataji, Nihon shōgyōshi (A history of Japanese commerce) (Tokyo: Ryūginsha, 1943), 76Google Scholar.

⁸ An excellent discussion of the sanjo is found in Hayashiya, Tatsusaburo, “Sanjo: sono hassei to tenkai” (Sanjo: their origins and development), Shirin, vol. 37, no. 6, (1954), 511–536Google Scholar. As the sanjo could only be discussed briefly here, interested readers are strongly recommended to read this excellent article.

⁹ The imperial household had nearly a dozen groups especially to supply its needs, which are referred to variously according to their functions and the status of the persons involved.

¹⁰ One of the Seiren-in monjo (document) included as No. 4655 in Takeuchi, Rizō, ed., Heian ibun (The extant documents of the Heian period) (Tokyo: Tokyo-dō, 1964), IX, 3637Google Scholar. This source was first referred to in Akamatsu op. cit., 11–12 and then discussed by Wakita, op. cit., 210. Since Miura suggested in 1919 that written evidence of the za could be found no earlier than the end of the thirteenth century, many scholars attempted to find a komonjo (classic document) proving that the oldest za existed before that. Many were successful, and scores of earlier komonjo were unearthed. As of now, the author is convinced, after examining numerous secondary sources and relatively easily obtainable original sources, that the Seiren-in monjo referred to above is the oldest. Miura, op. cit., 868–869.

¹¹ What the extant records do not make clear is whether a pre-existing za-like organization came under the protection of the temple of whether a new group was organized de novo under the temple.

¹² Wakita, op. cit., 218.

¹³ Heian ibun, No. 1373. See: Wakita, op. cit., 184 and 221.

¹⁴ Kodama, Kōta and Takeshi, Toyoda, Ryūtsūahi (A history of commerce) (Tokyo: Yamakawa Shuppansha, 1969), 79Google Scholar.

¹⁵ Wakita, op. cit., 172 and 210.

¹⁶ Sasaki, Ginya, Chūsei no shōgyō (Medieval Commerce) (Tokyo: Shibundō, 1966), 143Google Scholar.

¹⁷ Endo, op. cit., 135.

¹⁸ This could very well be an excessively conservative estimate. There is, however, no way to know how many za existed.

¹⁹ The diary referred to is Sanchōki, written by Sanjō Nagakane. See: Shiba, op. cit., 21.

²⁰ Takeshi, Toyoda, “Shifu Gayochō-za no kenkyū” (A study on the za of the Shifu litter-bearers), Sigaku Zasshi, vol. 45, no. 1 (1934), 15–77Google Scholar. The observations contained in this paragraph in the text are from pages 20–21 of that article. In addition to being a detailed and well-documented study on the za of the litter-bearers, the article, written as the za-ronsō was ending, is a seminal work which includes an attempt to classify the za (by the relationships between the sponsors and the za members, and by the nature of the emerging monopoly rights), and an excellent description of the development of other za against the background of the economic changes over time. The “metal products” refer to imono and smiths to kajishi.

²¹ Takenaka, Yasukazu and Kawakami, Tadashi, Nihon shōgyōshi (A history of commerce in Japan) (Kyoto: Minerva Shobō, 1965), 39–10Google Scholar.

²² Descriptions contained in this paragraph are based principally on: Ono Hitoshi “Abura-shōnin to shite no Oyamazaki jinin” (The Oyamazaki jinin as oil merchants), Shakai Keizai Shigaku, vol. 1, no. 4 (1931), 717–719Google Scholar. Many works on the za contain references to this well-known za.

²³ For the political background, see: Duus, Peter, Feudalism in Japan, (New York, 1969), 43–56Google Scholar.

²⁴ This paragraph is, in effect, a drastically abbreviated economic interpretation of the complex political and economic factors which brought about the Kamamura Bakufu. Though the origins of the myōden are being debated among Japanese scholars, the fact that historical evidence referring to the myōden began to increase beginning in the mid-eleventh century is not disputed. A myōden was a paddy named after a myōshu, who was the head of a large household and who supervised the cultivation of the paddy by members of his household and/or by the cultivators who had initially commended their land to the shōen to which the myōshu paid dues on their behalf. These cultivators then worked under his supervision. It perhaps is accurate to say that the shiki (rights to benefit from the output of the land) of myōshu became better established because, during this period, the power of the shōen-owners was being neutralized by that of the rising samurai class. Though basically a politically-oriented analysis, a concise and able discussion on the emergence of the myōden and the economic position of the myōshu is found in Nagahara, Keiji, Chūsei no shakai (The Medieval Society) (Tokyo: Iwanami Shoten, 1968), 83–103Google Scholar.

²⁵ There were exceptions, of course, and the shōen in areas distant from the capital suffered incursions of the local power more frequently and earlier in time than those nearer the capital.

²⁶ A clear and analytically cogent exposition of the relationship between an increased market size and increases in productive efficiency (specialization, economies of scale, reductions in costs of transactions, technological progress, etc.) is found in Young, Allyn, “Increasing Returns and Economic Progress,” Economic Journal, XXXVIII, no. 152 (December, 1928), 527–540CrossRefGoogle Scholar. This classic article also has the advantage of being highly readable and relatively free from economists' jargon.

²⁷ One could also say that the za emerged because the costs of organizing and maintaining them became, about this time, smaller than the potential gains to be realized by the parties concerned. Some readers may find the recent advances made in the theory of institutional changes (based on economic theory) by Lance Davis and Douglass North useful in analyzing the emergence and growth of the za. See Davis, and North, , Institutional Change and American Economic Growth (Cambridge, 1971), 1–79CrossRefGoogle Scholar.

²⁸ See, for example, Keiji, Nagahara, ed., Nihon keizai-shi (An economic history of Japan ) (Tokyo: Yūhikaku, 1970), 30Google Scholar.

²⁹ Takeuchi, Rizō and Mori, Katsumi, Nihonshi gaisetsu (A general history of Japan) (Tokyo: Hanawa Shobō, 1970), the fourthedition, 76–81Google Scholar.

³⁰ Enjoying the privileges of fuyu-funyū (no taxation and no entry) from the Bakufu and its agents, the temples maintained their own police force, which had considerable power.

³¹ The za, not being literal monopolists who could price their products at will (facing a totally inelastic demand), would have increased output and caused the price to decline. As this happened, the consumer would have gained what economists call “consumers' surplus.”

³² The power of the Ashikaga Bakufu was such that it was not until 1386 that the Bakufu was able to try persons associated with temples in the Bakufu courts. The Bakufu began to collect taxes from money-lenders and pawn-brokers in Kyoto only after 1393. Wakita, op. cit., 277.

³³ Takenaka and Kawakami, op. cit., 38.

³⁴ Toyoda, “A study on the za of the Shifu litter-bearers,” 24.

³⁵ Ibid., 25–26.

³⁶ The list is not intended to be exhaustive.

³⁷ Toyoda, “A study on the za of the Shifu litter-bearers,” 24.

³⁸ Sasaki, op. cit., 144–46.

³⁹ Though nearly all the writers on the pre-Tokugawa za make references to this wellknown za, and though their descriptions were useful, the discussions in the text rely principally on Ono's article cited in footnote 22 above.

⁴⁰ Ibid., 724–25.

⁴¹ Wakita, op. cit., 269–273.

⁴² Sasaki, op. cit., 59–82, and Toyoda, Takeshi, Chūsei nihon shōgyōshi no kenkyū (A study on the commercial history of medieval Japan) (Tokyo: Iwanami Shoten, 1934), 89–117Google Scholar.

⁴³ The use of coins as a medium of exchange increased rapidly during the Kamakura period. By the end of the period, rice and cloth were rarely used for this purpose. Kobata, Atsushi, Nihon kahei ryūtsūshi (A history of money circulation in Japan) (Tokyo: Tōkō Shoin, 1969), 44–46Google Scholar (This is a reissue of the book published originally in 1933); and Toyoda, A study on the commerical history, 86. The costs of transactions in the context of the text refer to the costs of exchanging a bundle of commodities to be sold for a bundle of desired commodities.

⁴⁴ That is, the costs enforcing the agreement were made low enough by the use of the coins to make such an agreement feasible, i.e., profitable.

⁴⁵ See footnote 31 above. The existence of substitutes has the effect of making the demand for a commodity more elastic, i.e., a small percentage increase in the price of the commodity reduced the demand by more than the same percentage.

⁴⁶ For an excellent overview of these developments, see: Yasuda, Motohisa, Nihon shōenshi gaisetsu (A general history of the Japanese shōen) (Tokyo: Yamakawa Shuppansha, 1957)Google Scholar.

⁴⁷ While many authors provide useful examinations of this za, the best and most detailed is: Ono, Hitosi, “Kitano kōji za ni tsuite” (On the Kitano sake-malt za), Kokushigaku, vol. 11. no. 5 (1932), 1–27Google Scholar.

⁴⁸ One of the difficulties faced by Ono in interpreting some passages in the original source, the Kitano jinja monjo, was due to his failure to examine the economic logic of the situation; ibid., 16–17. While Ono saw the sake-makers' actions as efforts to eliminate “the unnaturalness” of the situation, i.e., being forced to buy the malt when they themselves could make it, the fact of the matter simply was that the total costs of buying an increased amount of the malt at a monopolistic price exceeded the fixed costs necessary to build the malt-chamber plus the variable costs required to produce the required malt.

⁴⁹ The observations and inferences contained in this paragraph, the author believes, are well established among Japanese scholars.

⁵⁰ Ono, “On the Kitano sake-malt za” 21.

⁵¹ Many examples of these changing patterns of payments are found in Wakita, op. cit., 371–388 and Sasaki, op. cit., 187, among various other sources.

⁵² Ono, “The Oyamazaki jinin as the oil merchants,” 739.

⁵³ Kyoto came under the control of Oda Nobunaga in September 1568, which many Japanese history text books date as the end of the sengoku period.

⁵⁴ This example has been cited often by Japanese scholars. A detailed description is found in Ono, “The Oyamazaki jinin as oil merchants,” 738–39. For a good discussion of the gōyōshōnin in general, see: Toyoda, A study on the commercial history, 353–368.

⁵⁵ Sasaki, op. cit., 164. The merchants of the Omi region, known as the Gōshu-shōnin, have a long history dating from the Kamakura period. Though this essay did not discuss the za organized by these merchants except for the Oyamazaki-za, readers interested in the za activities and commerce in general during the Ashikaga and Sengoku periods are referred to: Egashira, Tsuneharu, Gōshu-shōnin (The merchants of Gōshu), (Tokyo: Shibun-dō, 1965), 7–18Google Scholar.

⁵⁶ Kodama and Toyoda, A History of Commerce, 97.

⁵⁷ Miyamoto, op. cit., 118.

⁵⁸ The za were not only monopolists of the commodities bought by the peasants, but also monopsonists (single buyers), at times, of the products sold by the peasants. Because the sengoku daimyo demanded taxes in cash at least in part (and the proportion of cash payment tended to rise over time), more and more peasants were becoming involved with the market economy partly to obtain cash for tax purposes. As long as the za controlled the purchase of cash crops, peasants had to be satisfied with the prices set by the za, for the crops. To the extent the monopsonistic prices were below competitive prices, the peasants suffered “exploitation.” While several Japanese scholars comment on this point as a partial reason for the rakuichi-rakuza policy, this is most explicitly stated by Horie, Yasuzō in his Nihon keizuishi gatyō (An outline of Japanese economic history) (Tokyo: Hirano Shobō, 1956), 127Google Scholar.

⁵⁹ This point is made forcefully in Toyoda, Takeshi, “Kinsei shotō ni okeru rakuichi rakuza no igi” (The significance of the rakuichi-rakuza during the late sixteenth century, Rekishigaku Kenkyū, II, no. 2 (1934), 2–16Google Scholar. The title of the article was translated to convey what Toyoda meant by “kinsei shotō” a literal translation is “at the very beginning of the modern period.”

⁶⁰ Toyoda, A study on the commercial history, 346–48.

⁶¹ Though there is no doubt that the abolition of numerous tolls was the most significant factor contributing to the reduction in the prices of these commodities, it is not possible to calculate its precise impact on each commodity. For information on the gradually increasing importance of the revenue from tolls in the total income of the court nobles, see Wakita, op. cit., 358–369, which contains a detailed case study. The most egregious examples of excessive numbers of toll-gates established were the Yodo River bank, a major route of the period between Kyoto and Osaka, which had 380, and an eight-mile strech of road leading to the Ise Shrine which had sixty. Hiranuma, Toshio, Kinsei jiin monezenchō no kenkyū (A study on the temple towns during the Tokugawa period) (Tokyo: Waseda University Press, 1957), 19–20Google Scholar. For a good monographic treatment of the toll-gates of medieval Japan, see: Aida, Jirō, Chūsei no sekisho (Toll-gates in medieval Japan) (Tokyo: Sebō Shobō, 1943)Google Scholar.

⁶² See, for example, Ono, Hitoshi, Kinsei jōkamachi no kenkyū (A study on the castle towns during the post-sengoku period) (Tokyo: Iwanami Shoten, 1929)Google Scholar.

⁶³ Many examples are cited in Toyoda, A study on the commercial history, 335–339.

⁶⁴ The first kenchi was carried out in 1500 and the first katana-gari in 1514. Sugiyama, op. cit., 221, and John W. Hall, “Rules by Status in Tokugawa Japan: An Inquiry into the Dynamics of Early Tokugawa Society,” mimeo, a working paper, dated January 1973, 7.

⁶⁵ Though there were scores of well-known and rich gōyōshōnin, the total volume of their trade accounted for only a fraction of the rapidly increasing commerce. See: Toyoda, Takeshi, Nihon shōninshi (A history of Japanese merchants) (Tokyo: Tokyo-dō, 1949), 197–218Google Scholar.

⁶⁶ That is, income maximizing or in an economic sense.

⁶⁷ Thrupp, Sylvia, “The Gilds,” in Postan, M. M. et al. , eds., The Cambridge Economic History of Europe (Cambridge, 1965) III, 243Google Scholar.

⁶⁸ Hall, John W., “The Castle Town and Japan's Modern Urbanization,” in Hall, J. W. and Jansen, M., eds., Studies in the Institutional History of Early Modern Japan (Princeton, 1968), 171–74Google Scholar.

⁶⁹ Ono, “The Oyamazaki jinin as oil merchants,” 732–33.

⁷⁰ How rapidly each za began to specialize in retailing and ceased to be involved in the production (often, the growing) and processing of commodities varied widely, by commodity and region. Paper, salt, cloth which required special skills in weaving, and lamp-oil (for which economies of scale in the extracting process were significant) were some of the commodities in which the za member were retailers from an early stage.

⁷¹ As Akamatsu lamented, historical evidence on the za of skilled craftsmen is scarce and even when it is to be found, as for carpenters and others engaged in the construction of temples or meeting the needs of the imperial household, the records seem to indicate that these craftsmen had no bona fide organization functioning as a guild. Thus, Endō went so far as to argue that, in Japan, no counterpart of the European craft-guild existed. As Japanese scholars realize, this is a topic which requires much more research. See: Akamatsu, op. cit., and Endō, op. cit., and numerous passing comments on the subject in various sources cited in this essay.

⁷² Various categories of craftsmen, including those in the tomobe, were classified as senmin (base or inferior persons) as opposed to ryōmin (good or free persons) during the Ritsuryō period. Many scholars consider the sanjo, where many of the skilled craftsmen congregated, to have been the origin of the later buraku — the residential areas of those persons classified as eta (outcasts), i.e., below all the four social classes of the Tokugawa period. Hayashida, op. cit., discusses all this in detail.

⁷³ Thrapp, “The Gilds,” passim.