Indentured Prostitution in Imperial Japan: Credible Commitments in the Commercial Sex Industry > Communication | 한일갈등타파연대

Indentured Prostitution in Imperial Japan: Credible Commitments in th…

J. Mark Ramseyer

University of California, Los Angeles

---

In 12th century Toulouse, the public brothels split their profits with the local university (Shadwell). Not so in Japan. Japanese brothels never tried to buy academic support, and never had it.

Instead, academics have consistently criticized the brothels for "enslaving" peasant women. Indenture contracts were an important part of the tales they told: poor and unsophisticated, peas ants unwittingly accepted indenture contracts that the brothel owners used to reduce prostitutes to sexual slaves.

These tales cheat the prostitutes of their due-for they drastically under- state the resourcefulness that they could show, even in the direst situations.

Although prostitution was harsh work, most brothel owners were not able to manipulate indenture contracts to keep prostitutes at work indefinitely, and most prostitutes did not become slaves. Instead, licensed prostitutes generally enlisted under six-year indenture contracts.

They earned (what were for them) very high incomes. Many repaid their debts in three or four years and quit early. Most of the rest quit when their contracts expired.

Within this world, the indentures helped make the employment market itself possible for despite the promises of high incomes, a woman entering the industry for the first time could never be sure.

She knew she and her family suffered a loss in social status if she took the job, knew some brothel owners had an incentive to lie about the money she would make, and knew most owners would be able to invoke the courts more easily than she.

Pre- cisely because she could never be sure of the money, she found the indenture contract advantageous. Through the contract, the brothel owners could promise her total earnings large enough to offset a substantial part of her lost status, could make that promise credible by paying her in advance, and could shoulder the costs of invoking the legal system themselves.

The point is not that licensed prostitution and indentured servitude were necessarily "good for Japan"; neither is the point that, overall, peasant women benefited from the availability of a legal market in prostitution. As noted earlier, both issues are beyond the scope of this essay.

Instead, the point is more limited: Given the substantial stigma women incurred in entering the industry, many women hesitated to take jobs at brothels (and many parents hesitated to send their daughters to brothels) without some assurance that they would earn much higher wages than they could earn elsewhere. The indentured contracts offered that assurance.

첨부파일

첨부파일

J. Mark Ramseyer 1991.pdf (3.5M)4회 다운로드 | DATE : 2021-02-13 14:31:44

다음글韓日紛争を終わらせるための連帯

----

Vladimir Tikhonov

tlSpo5nhasorede  · 

Looks like 'Mitsubishi professor' Ramseyer made a long career as a freelance advocate for sex industry and indentured slavery. Here is a specimen of his 'academic writing' dating back to 1991. The full-text article is available on a neonationalist website -  http://historicaltruth.net/bbs/board.php?bo_table=52&wr_id=3

Basically our Mitsubishi legal expert assures his readers that prostitution was a well-regulated industry in pre-war Japan, that, once the indentured slaves there paid off their advance money they were free to move elsewhere, and that their slavery was profitable for them too, since, on average, they earned 179 percent of the mean factory female worker's wage. 

Of course, our Harvardian professor mentions the evidences of indentured slaves being cheated and detained/exploited for much longer than their contract duration only in passing, treats all accounts of trafficking and violence in brothels contradicting his 'theory' as idiosyncratic and untruthful and gives no thought to the amount of both physical and  psychological traumatization which the indentured slaves in the licensed brothels had to go through. 

What does he know about the real world around him? It probably never even occurred to him that the earnings of the indentured slaves in the brothels were looking relatively high compared to 'factory girls'' wages because the latter were on semi-starvation level. Have he ever talked directly with anybody who was sold, trafficked and exploited? Does he imagine the approximate amount of beatings and other forms of physical violence which usually accompany brothel slavery? 

In a normal world, Harvard should have apologized for employing 'teachers' of this kind, since advocacy of sex industry based on cherry-picked, de-contextualized 'evidences' won't qualify as serious scholarship. Otherwise, students heading for Harvard should think twice about what sort of institution they are going to study at.... 

https://www.jstor.org/stable/764879?seq=1

-----

Indentured Prostitution in Imperial Japan: Credible Commitments in the Commercial Sex Industry

J. Mark Ramseyer

Journal of Law, Economics, & Organization

Vol. 7, No. 1 (Spring, 1991), pp. 89-116 (28 pages)

Published By: Oxford University Press

https://www.jstor.org/stable/764879

Cite this Item

 ====

Indentured Prostitution in Imperial Japan: Credible Commitments in the Commercial

Sex Industry 

 Author(s): J. Mark Ramseyer 

Source: Journal of Law, Economics, & Organization, Spring, 1991, Vol. 7, No. 1 (Spring,

1991), pp. 89-116 Published by: Oxford University Press

 Stable URL: https://www.jstor.org/stable/764879

====

Indentured Prostitution in Imperial Japan: Credible Commitments in the Commercial Sex Industry

Author(s): J. Mark Ramseyer

Source: Journal of Law, Economics, & Organization, Spring, 1991, Vol. 7, No. 1 (Spring, 1991), pp. 89-116

Published by: Oxford University Press

Indentured Prostitution in Imperial Japan: Credible Commitments in the Commercial Sex Industry

J. Mark Ramseyer

University of California. Los Angeles

I'm glad we won

a flower for a dollar I'm mad we lost

a flower for a dollar ---Japanese children's song

1. Introduction

So interpreted, the song makes no more sense in Japanese' than in English. But in its first two lines are two puns: "win" also means "buy," and "flower" can mean "girl." Thus the song can also mean "I'm glad we bought a girl for a dollar." As in Mother Goose, the macabre is historical: Peasant families in pre-20th-century Japan [like peasants in medieval Europe (Engerman, 1973:44)] bought and sold children. "I'm glad my parents bought for me," the child sings, "a sister for a dollar."

I received helpful suggestions from far more than the usual number of generous readers: N. Abrams, W. Alford, P. Arenella, H. Baerwald, T. Bryant, W. Comanor, K. Crabb, R. Fretz, M. Fruin, S. Garon, C. Goldberg-Ambrose, W. Klein, L. Lynn, H. Motomura, M. Nakazato, G. Noble, H. Ooms, H. Patrick, I. Pug, M. Polinsky, R. Posner, A. and R. Ramseyer, B. Rasmusen, R. Romano, F. Rosenbiuth, A. Rosefl, R. Sander, G. Schwartz, H. Scogin, R. Smethurst, F. Upham, J. Wiley, 0. Williamson, N. Wyse, S. Yeazell, two anonymous referees, and participants in the UC Berkeley Workshop in Institutional Analysis, the UCLA Political Economy Workshop, the Yale Law, Economics, & Organization Workshop, the SSRC—UCLA Japanese Legal History Workshop, and the Pacific Roundtable on Industry, Society, and Management. I received financial assistance from the UCLA School of Law Dean's Fund and the UCLA Japan—U.S. Research and Exchange Program. In conducting the research, I received help from the library staffs at UC Berkeley, UCLA, Harvard—Yenching Institute, Hiroshima University, Hitotsubashi University, University of Michigan, Tohoku University, and Tokyo University. The views expressed in this paper are mine; I know for a fact that many people gracious enough to comment on earlier drafts of the article disagree with the approach.

I. "Katte ureshii/hana ichimonme/makete kuyashii/hana ichimonme."

C 1991 by Oxford University Press. All rights reserved. ISSN 8756-6222

 

If some premodern Japanese peasants sold their children, some are said to have indentured their daughters to brothels. At the same time, many women contracted to become prostitutes independently, and did so through long-term employment contracts. In this article, I examine a more modem version of these various contracts of the macabre: the employment arrangements among parents, prostitutes, and brothels in the commercial market for sexual services in early 20th-century Japan. More specifically, I examine the use of indentured prostitution agreements—contracts in which a prostitute agreed to work at a brothel for several years in exchange for a large advance on her future earn­ings. Prostitutes and brothels have used these indenture contracts in a variety of societies, and I use data from Japan to ask why.

Prostitution in pre-World War II Japan was a regulated industry, and it was a controversial industry. For precisely those reasons, government records and independent studies together yield an astonishingly candid set of empirical records. Accordingly, I first use those records to recover and reconstruct the contractual arrangements themselves. In doing so, I describe both the deals the parties cut and the way they enforced those deals.

Second, I study the most common hypothesis about these indentures—that the brothel owners offered the contracts because they could manipulate them in a way that tied the prostitutes to the brothel beyond the stated contractual term (because, in short, the owners could transform indentured servitude into debt peonage). I find no evidence for the hypothesis. Instead, I find that many prostitutes repaid their advance substantially before the end of their terms and quit early. Indentured prostitution was a harsh job, but it was a job women took for only a few years.

Third, I examine two additional hypotheses about the contracts: (i) that they enabled the brothels to control prostitutes, and (ii) that they (like the indenture contracts that brought thousands of Europeans to North America) provided credit to destitute peasants. Although these hypotheses may capture some facets of the Japanese prostitute indentures, I explain why they do not suffice.

Instead, I contend that prostitutes and brothels used the indenture contracts for a different set of reasons. First, the contracts enabled brothel owners to make credible their promises about future earnings to the women involved (and, sometimes, their parents) (Williamson, 1983, 1985). The owners found it advantageous to make these promises because the recruits were entering an industry where they would suffer a certain and substantial reputational loss, but where they would earn a wage about which they had little reliable infor­mation. Second, the contracts enabled the owners and prostitutes to allocate the costs of invoking the legal system to the party able to do so most cheaply. That brothels and prostitutes would create such arrangements suggests, ulti­mately, that scholars would do well not to let the brutality of prostitution blind them to the effective ways peasant women and men make the most of bad situations .2

2. See the books by Judith Walkowitz and Barbara Hobson (Chapter 4) for evidence of rational self-assertion by prostitutes in 19th-century England and the United States.

Given the politically charged nature of prostitution and indentured ser­vitude, more than the usual disclaimers may be in order. First, and most obviously, this article is a study in the history of the organization of an industry, and not a normative essay. I take no position in this article on whether societies should legalize prostitution.3

Second, for lack of data, I also leave unanswered one basic empirical question: How often did Japanese women choose to become prostitutes on their own, and how often did they do so under family pressure? On the one hand, prostitutes earned reasonably high incomes, and many women were poor. Some women independently chose the job simply because it paid so much. On the other, some women took the job under parental pressure. Although I find evidence that such pressure mattered less than often alleged, the evidence is too tentative to justify any conclusion.4 Other research does suggest that some peasant families share risk by pooling incomes (Popkin: 18­22, Rosenzweig and Stark), and some daughters probably joined the brothels as part—albeit a bitter part—of such a family insurance scheme. Parents or wealthy neighbors often did guarantee the prostitute's debt (though they could well have done so even when the woman chose to become a prostitute on her own).5 For expositional simplicity—but only for that purpose—I describe the employment arrangement as though the women involved consistently chose it themselves.

Third, I deliberately ignore most contemporaneous narrative accounts of Japanese prostitution. To date, almost all histories of the industry have relied on firsthand accounts of reformist journalists, abolitionists, and ex-prostitutes rescued by those abolitionists. Necessarily, the prostitutes who approached

1.      Compare, for example, Shrage and Pateman (prostitution should be banned) with 0km and Radin (some forms of prostitution should be allowed).

2.      Note that even though prostitutes could legally quit whenever they pleased (Section 3.2, infra), most worked until they had either repaid their debt or served their full term. From 1927 to 1929, for example, only about I percent of all licensed prostitutes quit their work without the brothel owner's consent (It6:21 3-4). Yet if their parents had signed the indenture contract (as they generally had), then the only cost of quitting was that they and their parents became liable on the note. And if the parents had forced them into prostitution, their parents would have taken the advance and the prostitutes themselves would have been left judgment proof. Effectively, then, the prostitutes' only real risk was that the brothel owner would seize their parents' assets. As a result, women forced into prostitution by abusive parents could have walked away from their jobs, left their parents to repay the debt, and faded into the anonymity of metropolitan Tokyo. Admittedly, often it may have been a dauntingly grim prospect. Yet if large numbers of parents coerced their daughters into prostitution against their will, one would expect a significant number of women regularly to abandon the job—and opt for that cold anonymity over the job they had learned to despise and their parents who had sold them into it. Almost none, however, ever did SO.

According to the Salvation Army, 31 percent of the women it "rescued" from jobs as inden­tured prostitutes in 1930 reported that they had been induced to take the jobs by their parents (Kakusei, 1931b).

3.   Note that under the prewar Civil Code, men and unmarried women could contract on their own if age 20 or older. [Mimpö (Civil Code):l 31. Men and women under the age of 20 could not contract without the Consent of their guardian (Mimpo: 4).

the abolitionists were among the most dissatisfied, while the women who uneventfully fulfilled their indenture terms wrote few books, articles, or even diaries about their experience. As a result, the extant narratives present enor­mous problems of sample bias. Issues of balance also plague quantitative records, to be sure. Yet because prewar Japanese prostitution was both regu­lated and controversial, both government agencies and abolitionists spent massive resources counting the people and money in the industry. Surprising­ly, perhaps, these quantitative records corroborate each other on most critical points. I use those records where they converge; where they do not, I note that divergence.

The article proceeds as follows. In Section 2, I outline both the conven­tional accounts of prewar Japanese prostitution and the basic contours of the industry. In Section 3, I trace the evolution of the regulatory scheme involved. In Sections 4.1-4.3, 1 ask how the indenture contracts actually worked, and, in Sections 4.4 and 4.5, I explore why the contracts were so common. The regulatory scheme did not survive the 1950s, however, and I trace its collapse in Section 5.

2. Scholars and Prostitutes

Most scholars explain prewar Japanese prostitution through what they consid­er "exploitative" economic growth. Consider social historian Mikiso Hane. Relying heavily on journalistic reports, Hane claims a "growing chasm" separated rich and poor in prewar Japan (34): "[T]he condition of the peasants remained pathetic in contrast with the growing well-being of bourgeois cap­italists," and peasants thus found themselves waging "a bitter struggle for survival" (31, 27). Within this impoverished world, some of "the most pitiful victims . . . were the young farm girls who were sold to brothels" (207).

Anthropologist Liza Dalby studied the licensed entertainer—prostitutes known as "geisha," and tells a similar story. The geisha, she writes, lived decent lives only compared with the lives ordinary licensed prostitutes (shogi) lived: "Dreadful as life was for [the geisha,} they were yet better off than the girls who were sent not to geisha houses but to brothels" (222). In part, that dreadfulness resulted from the way the geisha house used indenture contracts to tie the women to it. Like many scholars of prostitution elsewhere,6 Dalby suggests that the Japanese house owners manipulated the contracts to turn prostitution into debt peonage—and keep the women working beyond the original term. Trapped with "unscrupulous owners who charged the inmates exorbitant rates for room and board, intentionally keeping them in a state of dependence," the geisha worked in "virtual captivity" (221).

Fact or fiction, the stories sold newspapers and magazines, and contempo‑

1.                See, for example, Corbin (78) (France), Harsin (293) (France), O'Callaghan (13) (China), and Rosen (130) (United States).

2.                Note, however, that most women apparently received room and board free. See Section 4, infra.

rary reporters made the most of them. Journalists told of naive women tricked by usurious brothel owners into a life of vice, tied by a debt that increased by the month. It was thinly disguised slavery, they argued, and the government should ban it. Most modem scholars merely repeat these accounts. Even Japan's eminent legal sociologist Takeyoshi Kawashima concludes that the prostitutes were caught, through "the power of the patriarchal family sys­tem," in "slavery" (1950:89, 1955).

It was not all fiction. Many families were poor in prewar Japan. The international depression hit the country early and hard, and may have hit the countryside hardest of all.8 Given the alternatives, some peasant women did become prostitutes. Those who did so often contracted to stay at a brothel for several years. In exchange, they received much of their salary in advance.

Whatever virtues it lacked, the vice was a major industry. In 1924, Japan had 550 licensed red-light districts, 50,100 licensed prostitutes, and 11,500 licensed brothels. These brothels were substantial businesses, not street-corner pimps. Most typically, they served food and drink besides sex, and were large 15- to 16-room establishments that carried 4 to 7 prostitutes and 6 to 10 additional employees on staff. In addition, Japan had 77,100 licensed geisha. The number of unlicensed (and hence illegal) prostitutes is less clear, but one otherwise reliable observer placed the figure at about 50,000. With a population of 59.7 million, Japan apparently had I prostitute for every 350 people. In the city of Kyoto, the figure was 1 for every 150.9 By contrast, in the modern United States, scholars estimate the figure at 1 for every 650 to 900 (Symanski:10).

As harsh as prostitution was, the women who became prostitutes were not women with many attractive alternatives. Most licensed prostitutes were badly educated. Notwithstanding that all were at least 18 years old, half had fewer than five years education and 16 percent had not been to school at all'°—this at a time when 99 percent of all Japanese primary-school-age students (grades 1-6) were in school (Minami:19). Such unskilled and uneducated workers had few attractive choices in prewar Japan, for the country was poor. Consider one survey of 12- to 18-year-old night-school students: Among those with factory jobs, the modal student worked 10 hours a day; among those with commercial jobs, the modal student worked 12 hours (and that they found time for school suggests they worked shorter hours than most). Of all students, only 13 percent worked fewer than 28

1.      See the trend in farm household income in Table 1, infra. Scholars have often exaggerated this poverty, however, as Smethurst rightly points out in his excellent book.

2.      The figures are from Fukurni (26-28, 32, 50-56, 178). Fukutni estimates the total number of licensed and unlicensed prostitutes at 174,000 in the mid-1920s. Nakamura (222-23) places the total at 276,000. See also Kusama (14-26).

3.      See Fukumi (66-68) (Tokyo data), ItO (204), and Kusama (100-03). Licensed prostitutes were required by law to be at least 18 years old [Naimu shO rei (Home Ministry Order) No. 44, Oct. 2, 1990, § 11.

days per month." Prostitutes on average had 2.54 customers per night in 1924,12 and likewise worked about 28 nights per month. 13 Prostitution was sordid work. But the alternatives were not easy.

And prostitution did pay well. Consider Table 1: Rom 1926 to 1932, prostitute earnings averaged 179 percent of the mean female factory worker's wage,14 and 53 percent of the total mean income of farm households (an average of both landlords and tenants). 15 Oliver Williamson (1985:35-8) and Price Fishback (1986a, 1986b, 1989) recently discussed how American miners received higher wages when coal companies provided unsanitary or unsafe conditions; Clark Nardinelli showed how British parents received high­er wages for their children when a factory used corporal punishment. Much of the same logic applied in Japan. Japanese peasant women faced an array of options, among which a job as a prostitute was one of the most squalid and stigmatizing. In exchange for taking the job, they demanded—and received—high pay. According to one 1934 sample of female workers from a northern prefecture, young women who left home to work earned room and board plus a mean 884 yen per year as a licensed prostitute, 575 yen as a geisha, 518 yen

1.        Only 13 percent had more than two days off per month (Shakai, 1936:23-25). This is roughly consistent with Ohsato's data for the country as a whole: in 1926, workers in manufactur­ing industries worked 10.32 hours per day, 27.1 days per month. See Ohsato (58-61); see also Naikaku tOkei (109, 122) (similar figures).

2.        From 1922 to 1932, the mean ranged from a low of 1.71 customers per night in 1930 to a high of 3.04 in 1923 (when some of the records were lost in the great Tokyo earthquake). The 11-year running average was 2.10 [Keishi cho (1933:96)]. Other estimates confirm these figures [see Maeda v. Yanadani. 841 1-l6ritsu shimbun 21, 22 (Tokyo Ct. App. Nov. 14, 1912) (70-100 customers per month); Uemura (1929:492-501) (similar figures for urban prostitutes); and Kusa-ma (220-1) (2.54 customers per night in Tokyo, 1924)1. On the other hand, the figures for the city of Osaka for 1913-15 ranged from 0.72 customers per night to 0.78 [see Uernura (1918:33­4)]. By contrast, Corbin (81) cites studies of the French licensed quarters indicating four to eight or more clients per night; see Harsin (283) (even higher figure for France).

3.        In 1924, customers made 4.20 million visits to 4989 licensed prostitutes in Tokyo. At 2.54 customers per prostitute per working night in 1924, that leads to 331 working days per prostitute [Keishi chfl (1933:96-8)].

4.        During this period, mean annual wages for all factory workers were 1926, 554 yen; 1927, 632 yen; 1928, 657 yen; 1929, 666 yen; 1930, 650 yen; 1931, 605 yen; and 1932, 580 yen (see Ohsato:60, 69). Female factory workers made 30 to 40 percent of the male wage, though observers argued that much of that difference disappeared when adjustments were made for age, work experience, and tenure on the job [ROdflsho (1952:14-7)].

The Table I figures on prostitute income are based on data the government required brothel owners to submit, and reliability is an obvious issue. Nonetheless, although the data were widely reported in the prewar studies of the industry, I know of no abolitionists who complained that the figures were inaccurate. To the contrary, abolitionists themselves used income figures that were quite close. See, for example, Ito (229-30). Their quarrel was not with the amount of income, but with whether, given that income, prostitutes could live within it. See Section 4.4, infra. These figures represent prostitute income before any repayment of the indenture principal.

11.  Most prostitutes, of course, came from households with below-average incomes. Note that in 1926-27, farm households had a mean monthly income of 96.2 yen, land-owning households had income of 112.5 yen, and tenant families had an income of 79.2 yen. See Naikaku tOkei (352).

Table 1. Mean Income of Female Factory Workers, Farm Households, and Licensed Tokyo Prostitutes

A             B             C

Factory      Farm        Licensed

Workers-, Householdsb    Prostitutes c

Year      (yen)        (yen)         (yen)      C/A     C/B

 

1926

312

1433

641

2.05

0.447

1927

320

1183

658

2.06

0.556

1928

322

1361

660

205

0.484

1929

320

1201

554

173

0.461

1930

289

810

430

1.49

0.531

1931

260

552

406

1.56

0.736

1932

245

644

388

1.58

0.602

 

Note: Prostitutes also received tree room and board; many factory workers did not.

aDaity wages from RadOshO fupn shönen kyoku, 1952. Fujin fodrl no fitsolO frhe Reality of Female Labor], p. 14. Tokyo: RociOshO; number of work days per month from Ohsato. Katsuma. 1966. Meiji i/cO honpO sluqo keizai tOkei (Principal Economic Statistics for our Nation since the Meiji Period], p. 60. Tokyo: Bank of Japan. b5Omu chO tOkei kyoku. 1987. Nihon diSki tOkei sOran [General Long-term Statistics for Japan], Vol. 4, Table 18-5-a. Tokyo: SOmu chO.

c(KeiShi chS) sSkan kanbS bunsho ka. 1933. S/iOwa nana nen keishi chO tOkel iChi ippan (An Outline of Police Agency Statistics for /9321, p. 96. Tokyo: (Keishi chol sOkan kanbO bunsho ka.

as a bar maid (a job that often involved prostitution), 210 yen as a waitress (also often involving some sex), and 130 yen in other jobs. 16 Ultimately, many more women looked for a position as a licensed prostitute than located one: from 1920 to 1927, only 62 percent of the women applying for work as licensed prostitutes in Tokyo found jobs (ChUO shokugy6:381-2, Kusama:27-30, 36).

Through its licensing scheme, the government effectively created three overlapping hierarchical tiers in the industry: licensed geisha, licensed pros­titutes, and unlicensed prostitutes. The geisha were licensed as entertainers, and could not legally market their sexual services. They entered the industry with better educational backgrounds than the licensed prostitutes, and (in theory) took extensive training in singing, dancing, and witty repartee. 17 Although they sang, danced, and chatted for money, most (by one 1930 study,

1.        Akita prefecture. The figures for Miyagi prefecture were 315 yen for a licensed prostitute, 337 yen for a geisha, 187 for a bar maid, 132 for a waitress, 107 for a factory worker, and 78 yen for a maid or child-care worker. Most of these employers probably provided room and board in addition to these wages [Shakai (1935:160-1)). As the sampling techniques are not clear from these documents, these data should be treated with more caution than the aggregate amounts reported in Table 1—a point confirmed by the divergence between the figures for Akita and Miyagi prefectures. For evidence of high prostitute incomes in the West, see Rosen (147-8) and Mustang.

2.        See Dalby. By one study, 92 percent of the geisha had completed at least primary school. Of the licensed prostitutes, only 42 percent had completed at least 5 of the 6 primary school years, and 16 percent had had no schooling at all [see Fukumi (66-8; 216-7)].

about 80 percent) also traded sexual services for money (Kusama:5, 20; Fukumi:234).

Much below the highest grade geisha came the licensed and unlicensed prostitutes, women with few pretenses about any extrasexual skills. For sever­al reasons, customers preferred the licensed prostitutes to the unlicensed. First, the licensed women probably invested in a reputation for providing high-quality service. They could sell sex legally, and worked for a finn that could market their services legally. As a result, they could safely make such investments. By contrast, unlicensed prostitutes worked under the constant threat that the police would interrupt their careers and close their employer. 18

Second, the licensed prostitutes were healthier. Licensed brothels at least screened their recruits for contagious diseases, had weekly medical inspec­tions, and maintained special arrangements with designated clinics. Although the medical records are problematic, contemporary studies did consistently show a much lower incidence of venereal disease among the licensed pros­titutes than among the unlicensed. 19

Finally, customers apparently considered the licensed prostitutes physically more attractive. First, many of the unlicensed prostitutes were women no licensed brothel would hire (Kusama:37). Second, many unlicensed pros­titutes were older. Although customers seem to have preferred women in their late teens and early 20s, the unlicensed sector included a larger percentage of women in their late 20s and early 30s than did the licensed sector (Fukumi:59, 144).

Whatever the reasons, prices reflected this consumer preference for the licensed prostitutes over the unlicensed. In the late 1920s, the "special" (highest) grade licensed prostitute in the Yoshiwara district of Tokyo charged 14-17 yen for a full night's assignation, and the fifth (lowest) grade licensed prostitute charged 6 yen for a full night. By contrast, the unlicensed pros­titutes in the Tamanoi and Kameido areas charged 3-5 yen for a night (Kusama:230-1, 242).

1.        See Klein and Leffler. On the prohibition of unlicensed prostitution, see Gyosei shikkö ho (Administrative Enforcement Act), Law No. 84 of June 1, 1900, § 3; Naimu shö rd (Home Ministry Order) No. 16 of Sept. 29, 1908, § 1.

2.        See Keishi chO (1933:143) and Uemura (1918:47). In 1932, these inspections of licensed prostitutes in Tokyo resulted in findings of disease in 3.22 percent of the cases [Keishi chO (1933:143)]. Of the positive diagnoses, 41.7 percent were of gonorrhea, 26.2 percent of chan-croid, and 7.6 percent of syphilis [Keishi chO (1933:144)]. Inspections of unlicensed prostitutes resulted in findings of disease in 9.7 percent of the cases [Keishi chO (1933:144)). Other statistics confirm this difference between licensed and unlicensed infection rates. In a 1927 government study, for example, 32 percent of the unlicensed prostitutes had venereal disease, but only 2.1 percent of the licensed prostitutes did (Kusama:288, 291). A third study found an even larger difference: 2.8 percent of the licensed prostitutes had a venereal or other communicable disease (in 1924), compared to over 40 percent of the unlicensed prostitutes (in 1925-26) (Fukumi:93, 168-9). See also ChUO shokugyo (433-35) (similar statistics). Note that the abolitionists also reported these low venereal disease rates for licensed prostitutes. See Kakusei (193] a) (infection rate of 1.82 percent).

3. The Regulation of Sex

3.1 The Decrees

Prostitutes used indenture arrangements that were subject to a 19th-century regulatory scheme. The American commodore Matthew Perry had forced open Japanese ports in 1853, and the Western nations had soon imposed on a confused Japanese government various "unequal" treaties. Determined to eliminate these agreements, government leaders set out to adopt at least the trappings of Western culture and law.

In 1872, their chance arrived. A Peruvian ship named the Maria Luz sailed into Yokohama harbor for repairs with 231 Chinese coolies hired under eight-year indenture contracts.20 One jumped overboard and swam to a nearby English gunship. Unsure what to do, the English captain sent him to the local English consulate. The consulate in turn contacted the Japanese Foreign Min­istry, and the Ministry called a local judge.

The time seemed perfect to prove Japanese sophistication. The judge did his best, and declared that international trafficking in human beings violated public international law. When Peru objected, the imperial Russian court intervened as arbitrator and backed the Japanese. But the ploy backfired. In the middle of the proceedings, the lawyer for the Peruvian ship declared that the Japanese kept their own slaves. Japanese brothels were full, he insisted, of indentured women who were no freer than the coolies.

Embarrassed, the Japanese government immediately liberated all inden­tured prostitutes. The women could return home, and courts would entertain no suits on their debt.2' To implement the decree, the Ministry of Justice issued its own regulation:

Indenture contracts rob people of their rights and reduce them to horses and cattle. As one cannot demand that horses and cattle repay their debts, neither can one demand that prostitutes and geisha repay their indenture amounts [Shihdshö tatsu (Justice Ministry Circular) No. 22, § 2, Oct. 9, 1872].

The regulation became infamous for its metaphor, but the point at least was clear: women could leave the brothels and keep the money the brothel owners had advanced them. Newspapers reported prostitutes loading their effects on carts and rickshaws, and leaving the brothels in droves (Nakamura:174).

But the 1872 decree did not ban prostitution. It merely voided the indenture contracts. By the end of the year, the city of Osaka had adopted a "room rental" licensing scheme to make clear that prostitution was still legal, and other cities soon followed suit. The prostitute was a licensed "independent contractor" under this system, and the brothel was a "room rental service."

1.                For more elaborate accounts, as well as complete versions of the court and arbitration decisions, see ItO (107-32); see also Stewart (Chapters 7 and 8) and Gardiner (Chapter 1).

2.                DajOkan fukoku (Cabinet Proclamation) No. 295, § 4, Oct. 2, 1872. The principal na­tional decrees and regulations relating to prostitution are reproduced as a legal appendix in Yamamoto (747-67).

By 1875, the national government even decided to permit indentures again. Sales of human beings were illegal, it explained, and mortgages of humans were no better. Yet it followed those bans with a cryptic distinction—arrange-ments where a debtor repaid a debt through a fixed-period labor contract were legal.22 Once again, indenture contracts were valid.

3.2 The Courts

For all this earlier confusion, by the turn of the century the courts made clear the legal status of the indenture contracts.23 In 1896, the Supreme Court announced part of that legal status—the personal service contract itself was void. Iki Kushi had apparently agreed to work as a prostitute for several years, but now wanted to quit. The Court let her do so. No matter what the financial relations between the parties might be, it held, the personal service contract is an independent agreement. It cited the 1872 decree banning indentures (but not the 1875 decree allowing them) and held:

We see no evidence that the promisor [prostitute] intended to abandon her personal freedom. Even had she so intended, she had no capacity to make such a binding agree­ment. Hence, the promisee [brothel] can enforce this contract only in ways it can enforce any other contract—in ways that do not bind the person of the promisor. [Musashino v. Kushi, 2-3 Daihan minroku 50, 52 (S. Cf. Mar. It, 1896)]

Four years later, the Supreme Court explained how a disgruntled prostitute could quit—although she could not leave the brothel unless the owner added the brothel's seal to her petition to de-register herself as a licensed prostitute, she could force the owner to do so. 24 Rita Itai had agreed to work for 30 months, but wanted to quit early. The court cited the 1872 law, and noted its 1896 opinion. It had already held such long-term service contracts unenforce­able, it declared. If the woman needed the brothel owner's seal to quit, she could compel the owner to affix it (Itai v. Yamada, 6-2 Daihan minroku at 83-84).

In refusing to enforce the personal service agreement, the Supreme Court distinguished the accompanying loan agreement. Never mind that it voided the personal service agreement by citing the 1872 decree—which explicitly declared the loan unenforceable. The Court effectively transformed the in­denture contract into two severable contracts: an employment contract and a loan contract. The former it held void, and the latter valid.

1.                DajOkan fukoku (Cabinet Proclamation) No. 128, Aug. 14, 1875; see Nakamura (171-8) and Yamamoto. In 1900, the government adopted the "room rental" licensing scheme on a national scale. See Naimu shö rei (Home Ministry Order) No. 44, Oct. 2, 1900; Keishi chö rd (Police Office Order) No. 37, Sept. 6 1900, discussed in Yamamoto (372-80).

2.                On the case law governing the indentures, see Nishimura, Nömi, Kawashima (1955), Wagatsuma (1923, 1955), and Yonekura (1985); on the turn-of-the-century court system, see

Ramseyer.

1.                Itai v. Yamada, 6-2 Daihan minroku 81 (S. Cf. Feb. 23, 1900); accord Fujiwara v. Kondo,

translated and reprinted in Murphy (140) (Nagoya D.C. May 21, 1900); Ohashi v. Suzuki, translated and reprinted in Murphy (143) (Nagoya D.C. June 11, 1900).

For example, on an advance of 300 yen, Raku Uchiumi apprenticed herself as a geisha in a hot-springs town. When the restaurant transferred her to another restaurant for 660 yen, she apparently decided she had negotiated a bad deal and tried to quit. The Court of Appeals held her five-year, five-month service contract void, but suggested that her loan contract was still valid [Uchiumi v. Takeda, 29 Höritsu shimbun 12 (Hiroshima Ct. App., Oct. 30, 1900)]. The Supreme Court affirmed:

If a person contracts to live with another and perform services, the contract is not necessarily void as a restriction on personal freedom. After all, those who agree to work always incur some restraints. Yet a contract of the sort at bar, that restricts the promisor for five years and five months, violates the public order and good morals. By contrast, the promise [to repay the advance] is not invalid. It is not invalid even though the promisee cannot enforce it directly either under law or in fact, since to do so would bodily restrict the promisor. [Uchiumi v. Takeda, 59 Horitsu shimbun 9, 10 (S. Ct. Oct. 10, 1901)]

The Supreme Court repeated the distinction two years later. Prostitution is not work blessed "in the sutras," it explained, but it is work licensed by law. 25 Accordingly, a prostitute may quit her job if she likes, but that which she has borrowed she is bound by law to return.

Courts occasionally held to the contrary. On the one hand, some courts enforced the personal service contract. For example, a district court refused to allow one 16 year old who had enlisted at a geisha house for six years to quit. "For a sixteen-year-old who must learn the arts of the geisha," it wrote, "six years is not necessarily long." As the length and other terms were reasonable, the house could enforce the contract [ItO v. Yoshida, 2115 HOritsu shimbun 5, 6 (Yokohama D.C. Feb. 7, 1923)].

On the other hand, some courts held even the loan contract void. Although they generally held the loan and service agreements severable, they did not always do so. At root, severability was a question of fact. For example, Yoshie Murakami entered a geisha house on a 10-year contract. Part way through her term she quit, apparently because the house would not train her. When the house sued her on the note, the court sided with her. Her loan and personal service agreements were inseparable, it explained. As the service agreement was void, so was the loan: she could quit and keep the money.26 And when a lower court held (as it did here) that the two agreements were not severable, the Supreme Court often deferred.27

1.           Okuma v. Watanabe, 8-2 Daihan minroku 18, 21 (S. Ct. Feb. 6, 1902); see also Hama v. Watanabe, 116 Hflritsu shimbun 10 (Tokyo Ct. App. Nov. 21, 1902).

2.           Murakami v. Izumi, 1986 Höritsu shimbun 7 (Miyagi Ct. App. Apr. 22, 1922). Loan

contracts were also held void in Ito v. Ito, 21 Daihan minroku 1718, 1726 (S. Ct. Oct. 18, 1915); SanjO v. Oki, 852 HOritsu shimbun 19 (Tokyo Ct. App. Oct. 11, 1912); Yamashita v. R45/ca, 947 Hflritsu shimbun 26, 27 (Osaka D.C., no date, journal issue of June 30, 1914); and Ono v. Ueda, 408 Hontsu shimbun 7, 8 (Osaka D.C. Oct. 20, 1906).

25.    Yamashita v. Ma, 21 Daihan minroku 905 (S. Ct. June 7, 1915); Murakami v. Izumi, 27 Daihan minroku 1774 (S. Ct. Sept. 29, 1921).

Nonetheless, the drafters of the time apparently knew which loan agree­ments courts would enforce and which agreements they would not, for most indenture contracts seem to have contained loan agreements the courts en­forced. Consider the case of Osumi Umezu.28 She had contracted to work as a licensed prostitute and received a 650 yen advance. Before repaying the advance, she decided to leave. The Court let her go. Long-term personal service contracts were void, and if Osumi wanted to quit, she could quit. Her loan, though, remained valid. If she quit, she and her guarantors still owed the amount she had borrowed.

Similarly, Hani Kiyomizu had agreed to work as a geisha and received 2300 yen in advance, but had died part way through the contract. When the house sued her father as guarantor, the court held for the house: "When Kiyomizu Ham dies, the debt is no longer repayable through her earnings. Nonetheless, the contractual liability continues." The agreements were sever­able, and even if she were gone, the note and guarantee were still valid [Kiyomizu v. Takeuchi, 3336 Höritsu shimbun 10, 10 (S. Ct. Oct. 23, 1931)].

4. The Indenture Contracts

In large part, modem scholars misunderstand prewar Japanese prostitution because they mischaracterize the prewar indenture contracts.29 Most scholars describe the contracts as debt peonage: Under the indenture agreements, prostitutes worked at a brothel until they repaid their loans. Brothel owners and recruiters enticed them into the industry by claiming that repayment would be fast. In fact, they lied. Once a woman enlisted, they charged her usurious rates on the loan and exorbitant prices for necessities. Through the scam, they kept her perpetually in debt, and trapped at the brothel for as long as she would sell. Effectively, as one historian put it, "the licensed prostitute ended her life as a sexual slave.1130

No doubt some brothel owners did manipulate contractual terms, and no doubt some did tie prostitutes to the brothel for as long as customers would buy their services.31 Yet the industry-wide records instead show that cases

1.                Umezu v. Abe, 2884 HOritsu shimbun 5,6 (S. Ct. May 12, 1928). Other cases reaching the same result include Shibuya v. Yokoyama, 4355 Höritsu shimbun 7, 9 (S. Cc. Nov. 22, 1938); Haneda v. Matsumoto, 2272 Höritsu shimbun 19 (S. Ct. Apr. 1, 1924); Mon v. Oshita. 1808 HOiitsu shimbun 11 (S. Ct. Oct. 30, 1920); Wazamaki v. Haneda, 10 Daihan minroku 1687, 1691 (S. Ct. Dec. 26, 1904); and Maeda v. Yanadani, 841 HOritsu shimbun 21 (Tokyo Ct. App. Nov. 14, 1912).

2.                For examples of actual contracts, see Ito (221-8), SaicO, Kusaina (170-204); and ChOO shokugyo (392-400).

3.                See Murakami (50); similar accounts appear in, for example, Yamamoto (391-2), Yonekura (59:41), and Yoshijni (31-36). To "prove" this account, abolitionists relied on surveys of those prostitutes who quite in violation of their contracts—not a random sample. See, for example, ItO (301-7).

4.                For examples of contracts where the women did work until they had fully repaid the loan, see ItO (227) and Kusama (170-78). See also, for example, Okuma v. Watanabe, 8-2 Daihan minroku 18, 20 (S. Cc. Feb. 6, 1902); Ito (230-1) (earnings and expenses of one insolvent

where the brothel owner kept the prostitute tied to the brothel were the exception. Consider first the geisha indentures, then the licensed prostitutes.

4.1 Geisha Indentures

Indentured geisha generally contracted to work only for a set number of years—the mode being three.32 Because they were often young when they apprenticed themselves [a quarter were under 18 (Fukumi:208-9)], their par­ents routinely demanded their net expected earnings at the outset—a mean amount in 1925 of 959 yen.33 In exchange, the house trained the women to sing and dance, paid their expenses, gave them a small allowance (generally 10 to 20 percent of their earnings), but kept all other revenues they generated.

As a result, the economics of geisha indentures were simple: if a woman raised less than her advance payment plus room and board and any pocket money over the indenture term, the house bore the loss; if she raised more, the house kept the extra. In effect, the contract ensured (a) that no one (including the geisha house) would defraud the woman of her earnings, (b) that she would receive food and shelter, however modest, and (c) that she would be free to leave when the initial term expired .34 To parents sending their children to the city alone for several years, such security mattered.

4.2 Prostitute Indentures

Because licensed prostitutes were older (by law they were at least 18 [Naimu shö rei (Home Ministry Order) No. 44 of Oct 2. 1900, § 1], they negotiated different contracts. They agreed to work for a maximum number of years, but kept track of their earnings and retained the right to quit early if they were successful. In the mid-1920s, the maximum time for which they agreed to work was usually six years.35 In exchange, they too received an advance of

prostitute over a four-month period); Ito (301-7) (abolitionist data finding that various samples of prostitutes had an increase in debt); and ChuO shokugyo (428-32) (high expenses).

1.                  Of 6603 indentured geisha studied in 1925, 1.5 percent had one-year contracts, 4.6 percent had two-year contracts, 29.0 percent had three-year contracts, 28.5 percent had four-year con­tracts, 13.9 percent had five-year contracts, and 22.5 percent had contracts of six years or more. See ChflO shokugyo (414) and Kusama (214-5).

2.                  See ChUO shokugyo (412-3) (note, too, that a second study found a mean of 955 yen); see also Kusama (205-6). The mode was 1000 to 1200 yen [Kusama (215)].

3.                  This contract was known as "rnarukakae," and was the contract most common among geisha who sold sex (Kusama:5). Approximately 60 percent of all Tokyo geisha used this type of contract (Kusama:5). For descriptions of these and other types of geisha contracts, see Fukumi (237-43), Higuchi (45-50), and Kusama (2-5).

Occasionally, a girl was "adopted" into the geisha house in exchange for a payment to the parents. Courts routinely declared such adoptions void. See, for example, Thkayama v. Taka-yama, 907 HOritsu shimbun 24 (Tokyo Ct. App. Oct. 9, 1913); Kato v. Katô, 514 HOritsu shimbun 11 (Tokyo Ct. App. July 7, 1908); and Ito v. ItO, 802 HOritsu shimbun 17 (Tokyo D.C. July 5, 1912).

32.       See ChuO shokugyo (414-15) (79.4 percent of licensed prostitutes had six-year inden­tures) and Kusama (211) (73.0 percent of licensed prostitutes had six-year indentures). This was apparently the longest contract allowed them. See Ito (220).

their net earnings—with the women who signed the longer-term contracts receiving the larger advances (Kusama:208-9). In the mid-1920s, the mean advance was 1194 yen.36

In calculating the prostitute's pay, the brothel first deducted its own fee. Generally, that fee varied from 67 to 75 percent of the prostitute's gross billings. The prostitute received the remaining 25 to 33 percent of the gross. She usually applied 60 percent of this amount toward the principal on her loan, and kept the rest for current expenses. In most cases, she did not explicitly pay interest, for most contracts did not contain an interest charge. Instead, the brothels apparently applied an implicit interest rate to her advance in calculating her share of the gross earnings.

Under the standard contract for licensed prostitutes, the prostitute could fulfill her part of the bargain either by repaying the principal or by serving the maximum contractual term. If (through the 60 percent she applied toward the loan) she repaid the loan principal before the end of her term, she could quit. If at the end of her term she had not yet repaid the amount, she could still quit. In the meantime, most brothels did not charge her additional room and board.37

4.3 Contractual Enforcement

Such was not just the way the parties usually drafted the contracts, it was also the way they usually implemented them. With assumptions drawn from mod­em illegal sex markets, scholars generally assume that brothel owners in legal sex markets use violence and coercion and ignore contractual niceties. And some abolitionists did recount anecdotes of hooligans and corrupt owners.38 Yet the industry-wide data instead suggest that the owners regularly kept their part of the bargains.

Consider first some simple finances.39 In 1925, consumers made 3.74 million visits to the 5159 licensed prostitutes in Tokyo. Aside from charges

1.                See Fukumi (70). The mode was 1000-1200 yen. See also ChüO shokugyo (412-5) (citing two studies: one with a mean of 1222 yen for 1925, one with a mean of 1018 yen) and Kusama (206) (same). Where it had been customary to deduct 200-300 yen for clothing, bedding,

miscellaneous household effects, and the placement fee, the industry eventually shifted to a system in which the parent or prostitute received the entire indenture amount, but then borrowed

an additional amount to cover those expenses (Kusama:213). On typical charges for these mate­rials, see Kusama (260-3). The mean charge by placement agencies in 1926 was 8.5 percent of the advance [ChOfl shokugyo (400)].

36.      See Keishi chfl (1933:96) (aggregate earnings data), Fukumi (97-9, 220) (contractual terms), Okubo (246-7) (same), Kusama (283) (same), Ito (229) (allocation of earnings), and Fukumi (115-6) (expenses).

See, for example, Murphy; see also de Becker (186-8) (fraudulent bookkeeping). Note, however, that by 1928 even some abolitionists were writing that it had become quite simple for licensed prostitutes to quit with both unserved time and unpaid debts. They noted the risk of hooligans only in the case of unlicensed brothels. See Ishijima.

All figures are from Keishi chO (1933:96, 98); see Kusama (227-8) (same).

for food and drink, they spent 11.1 million yen. Of this amount, the pros­titutes kept 31 percent, or 3.4 million yen-655 yen per prostitute. Under the usual arrangement, each prostitute would have paid 60 percent of this amount (393 yen) toward the loan principal and kept the rest (262 yen) for personal expenses. With an initial loan of 1194 yen, the average prostitute would have repaid her principal in 3.03 years. Nor should she have found it hard to live on 262 yen-21.8 yen per month. After all, she received room and board free. Young industrial workers earned modal monthly wages of less than 2 yen if they also received room and board, and 15-16 yen per month if not;40 adult factory workers (generally not receiving room and board) received a mean monthly wage in 1925 of 47 yen.4'

Second, consider the incidence of after-acquired debt among prostitutes.42 Modem scholars routinely claim that brothel owners kept their prostitutes tied to the brothel by increasing their debts. And as of 1925, 92 percent of the 5000-odd Tokyo licensed prostitutes did have some outstanding debt to the brothel besides their initial advance. Yet 37 percent had less than 200 yen outstanding, and an additional 19 percent had only 200-400 yen outstanding. Recall that the prostitutes had initially borrowed about 1200 yen. Only half of them had outstanding additional debts of at least one-third that amount, and only 5 percent had outstanding debts of at least an additional 1000.

More to the point, the longer a prostitute had worked at a brothel, the smaller the chance she had after-acquired debt outstanding. Consider again the 5000-odd Tokyo licensed prostitutes. Among those in the first year of their contract, 1484 had some after-acquired debt. Among those in their third year, only 703 had such debt, and among those in their sixth, only 84. In other words, once they had adjusted to their new job, they learned to save and repaid their debts. Most did borrow money beyond their original advance, but they borrowed it early in their career and repaid it fast.

Third, consider Table 2, the age distribution of licensed prostitutes in Tokyo. By law, a woman could not become a licensed prostitute until age 18, and most (though not all) prostitutes began work at a brothel between the ages

1.           See Shakai (1936:53). The figures are for 12-18 year olds, and date from 1935. For purposes of comparison, note that the average daily factory wage (adults, both sexes) in 1925 was 1.75 yen, and in 1935 was 1.88 yen. See Ohsato (68).

2.           See Ohsato (60,68) (27 days at 1.746 yen per day; 1925). In 1926-27, working class Japanese families spent 40 percent of their expenses on food and 16 percent on housing. See Naikaku tökei (353); see also Seiji keizai (7) (similar data for mid-1930s). In 1926, men earned 2.35 yen per day at factory work, while women earned 0.96 yen per day. See Rödöshd (1952:14); see also Naikaku tökei (130) (2.10 yen for men, 0.88 yen for women; 1924).

3.           The data are from Fukunii (122-3); see also ChUO shokugyo (433) and Kusania (280). Much the same is true of the geisha. According to one 1926 study of 2554 Tokyo geisha with after-acquired debt, 63 percent had borrowed less than 200 yen, and an additional 22 percent had borrowed 200-400 yen. Only 7 percent had borrowed at least an additional 1000 yen. See Kusama (258-9).

 

104 The Journal of Law, Economics. & Organization, V7 Ni Table 2. Licensed Prostitutes in Tokyo, 1925

 

Age

Number

Age

Number

18-20

1104

26

330

21

737

27

254

22

632

28-29

306

23

631

30-34

185

24

515

35-39

29

25

423

40—

6

 

Source: Fukumi, Takao, 1926. Ta/to ni okew ba/in no kenkyU [A Study of Prostitution in the Capital), pp. 58-9. Tokyo: 1-fakubunkan.

of 18 and 21.43 Yet beyond age 21, the number of prostitutes working falls steadily. Were brothels manipulating debts to keep their prostitutes employed beyond their initial six-year terms, the number of licensed prostitutes working in each age bracket should have held constant into the late 20s. Instead, Table 2 shows a steady decline.44 The table does show different age cohorts of women in 1925, rather than a history of one cohort. Because the number of licensed prostitutes in Tokyo stayed relatively stable (5188 in 1916, 5144 in 1925), however, these figures should approximate the attrition in a given age cohort over time (Fukumi:45).

Other data make the point more directly. In the early 1920s, of 42,400 licensed prostitutes studied, 11,400 (27 percent) had worked less than one year, 16,200 (38 percent) were in their second or third year, 10,400 (25 percent) were in their fourth or fifth year, 3100 (7 percent) were in their sixth or seventh year, and 1300 (3 percent) had worked more than seven years. The mean number of prostitutes in their fourth or fifth year (5200), in other words, was less than half those in their first year (11,400).45 Similarly, consider the entry and exit from the industry. During 1922, 18,800 women registered as

1.        See Naimu shfl rei (Home Ministry Order) No. 44 of Oct. 2, 1900, § 1 (minimum age). Fukumi, the source for Table 2, does not break out the age distribution of the 18-20 year olds. During one 10-month period in 1925-26, however, 494 women registered as licensed prostitutes in the Shin-Yoshiwara district of Tokyo. Of these women, 24 percent were 18 years old, 17 percent were 19, 13 percent were 20, 14 percent were 21, 7 percent were 22, 6 percent were 23, 5 percent were 24, 5 percent were 25, and 10 percent were over 25 [Uemura (1929:545-6); see Kusama:122-3]. Using these enlistment rates and assuming no retirements before age 20, we can (very roughly) estimate the composition of the 1104 women of ages 18-20 in Table 2 as age 18, 223; age 19, 380; and age 20, 501.

2.        Other sources confirm this data. See Kusama (136-8). The notion that prostitution was a transitional job women took for only a few years in their early 20s is consistent with studies of prostitution in other societies. See Walkowitz (19) and Hobson (86-7). It is also Consistent with the experience of Japanese women in other jobs. Saxonhouse (98), for example, writes that women in prewar Japanese textile factories stayed on the job only an average of two years.

3.        See Ito (208-11). In another study of 5734 licensed prostitutes, 29 percent were in their first year, 36 percent were in their second or third year, 26 percent in their fourth or fifth year, 8 percent were in their sixth or seventh year, and 2 percent had been there longer than 7 years (Kusama:281).

prostitutes and 18,300 women deregistered. Of the work force of some 50,000, about one-third turned over each year.46

The basic point, therefore, is that the brothel owners did not, on any wide scale, manipulate the indenture contracts to keep prostitutes tied to the brothel significantly beyond the term of the initial contract.47 Women enlisted for a limited number of years with the understanding that they could quit early if they worked hard. Many did. Most of the rest apparently quit when their term expired.48

To be sure, because prostitutes incurred a fixed cost (a reputational loss) in entering the industry, one might have expected many to reenlist. Once they had worked as a prostitute, their fixed cost was sunk. If their earnings stayed constant, many should have found it profitable to stay in the industry. Pros­titute revenues, however, fell with age (Kusama:207). As a result, apparently only those few who earned the most money and found the job least offensive considered it worthwhile to reenlist. Most of the rest seem to have considered the work not worth the new lower wages—for most seem to have quit.

4.4 Control and Credit

In addition to debt peonage, modern scholars raise two hypotheses to explain why brothel owners and prostitutes used multiyear indenture contracts: that the contracts enabled the brothels to control the prostitutes, and that they enabled peasants without credit to borrow. Nonetheless, although the con­tracts did provide credit and often did constrain the prostitutes, neither thesis very adequately explains why licensed brothels and prostitutes used them so pervasively.

4.4.1 Control. if brothel owners used the indenture contracts to control pros­titutes (e.g., Kawashima, 1951), they did not choose a very straightforward way of doing so—or even the most effective. Suppose A deposits a large sum

1.                  See Yamamoto (388). Ito (211-3) uses data primarily from 1923, but the data is less reliable because of the way he haphazardly includes some data from other years. He finds that 13,500 registered in one year and 11,000 deregistered. Uemura (62, 184-7) finds that, in Osaka, one-third of the new registrations were not genuinely new recruits but rather, inter alia, prostitutes moving from one city to another or reenlisting with a new indenture contract. If, following Uemura, one assumes that 1/3 of Yamamoto's 18,800 registrants were not genuinely new registrants, then the annual industry turnover rate would be about ¼. In general, only about 1 percent of the licensed prostitutes quit their jobs with outstanding debt (11:6:213-4).

2.                  We should not be entirely surprised that the brothels appear to have exploited prostitutes less than generally alleged. After all, brothels were repeat players in a competitive market. They could more cheaply recruit if they developed a reputation for treating their prostitutes relatively fairly. Some even worked to reassure their recruits by using standard form contracts approved by the local police station. See Shibatani v. Yokoyama, 4355 HOritsu shimbun 7, 8 IS. Ct. Nov. 22, 1938).

3.                  In a study of licensed Osaka prostitutes, Uemura (62, 184-7) found that less than ½ of the 18,800 annual registrants were reenlistments—but how much less than ½ is unclear from his data. Other studies suggest that a very few licensed prostitutes did work as unlicensed prostitutes when they had fulfilled their contractual terms (Fukumi:147).

of money with B. B can now constrain A, for B has A's cash. By paying the bond, A makes itself vulnerable to B; A does not thereby gain control over B. Consider banks. Banks concurrently lend money to borrowers and exercise control over the borrower's business. They bargain for the control, though, because they have lent the money and become vulnerable. They do not lend the money because they thereby gain control. Likewise, when a brothel paid a prostitute several years' wages in advance, it put itself in a position where she could force it to keep its part of the bargain—not vice versa .49

More basically, if the brothel owners had wanted to control the prostitutes, they had other far more straightforward and powerful ways of doing so. Most obviously, they could have forced the prostitute to deposit money (e.g., part of her earnings) with the brothel, rather than do what they did—deposit their own money with the prostitute. In demanding that the prostitute post such a bond, they would have had plenty of precedents. For example, some textile factories of the period routinely withheld a portion of their employees' wages. If and when the employee fulfilled the terms of her contract, she received the withheld amount. If she reneged, she forfeited it (Murakarni:135). Similarly, American indentured servants often received a large lump-sum payment called "freedom dues" at the end of their indenture term (Engerman, 1986:268-9). By paying the servant those dues only if he or she satisfactorily completed the contract, the employer gave the servant an incentive not to quit or shirk. Were Japanese brothel owners determined to control their em­ployees, they could easily have used such a system. In fact, they did not.

4.4.2 Credit. The notion that prostitutes would use the indenture contracts to obtain credit makes more sense. Because of the international depression, the bad harvests, and the collapse of the textile market, perhaps prewar Japanese peasants had little to eat. With no assets to transfer, they had nothing to sell. With no assets to pledge, they had no way to borrow. Unable to raise cash otherwise, women sold themselves and parents sold their daughters. In other words, perhaps peasants used the indenture contracts to overcome failures in the credit market. On its face, the hypothesis works. Moreover, if true, it would explain the contracts in a way consistent with indenture contracts outside of prostitution.50

Although the brothels no doubt did give credit to many peasants who

1.        Some contracts provided penalties that applied when the prostitute quit early, but courts often refused to enforce the penalty clauses. In some cases, the courts even considered the clauses reason to invalidate the entire debt contract. See, for example, Murakami v. Izumi, 28 Saihan minroku 1774, 1780-81 (S. Ct. Sept. 29, 1921).

2.        See, for example, Cloud and Galenson, Emer, Fèeny, Galenson. Grubb (1985, 1988), and Popkin (54). Note, however, that the need for credible commitments (Section 4.5, infra) could also help explain the use of indenture contracts in international migration. The worker incurred a high fixed cost in relocating to a new country, and the recruiter had much better information (and an incentive to lie) about the opportunities available there. As a result, the recruiter could have been paying the worker a large part of his or her wages in advance in order to make the promises about the benefits of emigration credible.

wanted it,5' failures in the credit market cannot explain why brothels and prostitutes used multiyear indenture contracts so pervasively. First, as noted above, courts refused to enforce long-term personal service agreements. As a result, the indenture contract could not substitute for other ways (e.g., collat­eral, guarantees) of enforcing loans, and thus could not have alleviated any basic credit-market failure. Suppose a brothel recruited a woman through a 1200 yen indenture, and two months later found that she wanted to quit. Although it could sue on the note, it could not legally force her to work. Because of the judicial refusal to enforce personal service contracts, in other words, prostitute indentures were little safer than unsecured loans. Conse­quently, brothel owners tried to recruit only women from families with assets they could seize (Murphy: 119, 132); families without the assets necessary to obtain a loan from other sources had little reason to expect that their daughters could find work as indentured prostitutes. When (in 1896) the Supreme Court voided the personal service contracts, it apparently prevented the very poorest women from using the indentures to raise cash.52

Second, were peasants demanding the long-term contracts because they needed large loans, they would have demanded them of other employers besides the licensed brothels and geisha houses. Notwithstanding this, the multiyear indentures were rare outside of the brothels and houses. Take, for example, some of the largest employers of women in prewar Japan---the textile companies. Textile firms employed a total of 751,000 women in 1925, and each factory employed a mean of 69 workers (Fukuoka:55). By contrast, the modal brothel had only six prostitutes on staff, and the modal geisha house had only one geisha.53 As a result, a textile firm should have been able to

1.        When registering as prostitutes at the local police office, women were required to give their reasons for doing so, and almost all wrote poverty. The point is meaningless, however, for if they wrote anything else, the police refused their application to register. See ChOö shokugyo (390) and Kusama (32-3). Nonetheless, most prostitutes apparently did come from lower-class backgrounds (see Kusama:47-75).

2.        To be sure, abolitionist writers occasionally claimed that hooligans and corrupt police used violence to prevent women from quitting in violation of their contracts. See, for example, ltd (308). Yet note several points. First, the evidence suggests that brothel owners did not manipulate the indenture contracts to keep women at the brothel for the full six years, but instead let many women quit after two or three years. At least when women had repaid their initial debt, the brothel owners seem to have followed the letter of the law. See Section 4.3, supra. Second, if brothel owners could successfully enforce indenture contracts through these illegal means, then unlicensed brothels (with dubious access to the judicial system) should have been offering them as well. However, in fact, unlicensed brothels often did not offer indentures, and generally offered much smaller amounts than the licensed brothels when they offered any at all. See ChUd shokugyo (413), Keishi chd (1935:509-10), and Kusama (216-7). Third, other abolitionists reported that by the 1920s it had become quite simple for licensed prostitutes to quit in violation of their contractual terms. These abolitionists noted the risk of hooligans only in the case of unlicensed (and hence illegal) prostitutes. See Jshijima.

3.        See Fukumi (50-1, 203-4) (Tokyo data). Four Tokyo brothels had one prostitute; the two largest had 21 each; 66 percent of the brothels had from four to eight prostitutes (Fukumi:50— 1). Only two Tokyo geisha houses had more than 10 geisha (Fukumi:203-4).

diversify credit risks more effectively than a brothel or geisha house. In turn, peasants interested in exchanging several years' worth of labor for credit should have found textile factories more receptive than brothels. Yet such was not the case. Instead, to the extent that the factories offered indentures, they apparently offered only relatively short contractual terms and relatively small advances—enough money to pay the workers' travel and readjustment costs, perhaps, but probably not much more.

True, a woman indentured at a brothel should have received a larger ad­vance than a woman indentured elsewhere for the same length of time, if only because sex paid better than most other manual labor. Yet that fact alone does not explain why the multiyear indentures were so common at the brothels and so rare elsewhere. High wages in the sex industry did not induce most poor women to become prostitutes; neither, therefore, should large loans have induced most credit-short women to become prostitutes. In other words, if credit had been driving the multiyear indentures, a broad range of employers should have offered them. Some credit-short women would then have chosen prostitution because of the large advances. But many others would have opted for smaller loans from employers with less demeaning jobs, just as most poor women avoided prostitution despite the high wages. Notwithstanding all this, multiyear indenture contracts were nearly universal among licensed pros­titutes and common among geisha; among factory workers, they were not.54

4.5 Credible Commitments

Apparently, neither the hypothesis that brothels wanted power over prostitutes nor the hypothesis that many peasants needed credit satisfactorily explains why brothels aid prostitutes so often used multiyear indenture contracts. Consider instead, therefore, the possibility that they negotiated the contracts for reasons relating to (a) the need for "credible commitments" [Williamson (1983, 1985:Chapters 7 and 8); see Lee and Png], and (b) the efficient alloca­tion of legal costs.

4.5.1 The Contracting Problem. A woman entering the sexual services in­dustry faced several problems relevant here. First, in entering the industry, she

54. Hane (175), Nishimura (1026), and Sievers (63) suggest that some factories did use some indentures. Yet other empirical data suggest that multiyear indenture contracts were uncommon among industrial workers. First, of 1077 12- to 18-year-old workers (in a cross section of industries) surveyed in 1935, only 59 were aware of any indentures, and only one (the highest) had an indenture of 400-500 yen. The male indentures were all for under 100 yen. See Shakai (1936:45, 49), Fukuoka (40), and Suehiro (4). Second, Tamura (38-9) reports that of all workers working under indenture contracts in 1951-52,63 percent went into pure prostitution; 8.1 percent became maids at hotels, brothels, bars, and geisha houses; 8 percent went into agriculture; 3.8 percent became factory workers; 3 percent became geisha; 3 percent became waitresses or bar maids; and 2.9 percent became child-care workers. See also Kanzaki (99) and Rodöshö (1953:74, Appendix 74). Third, one early-1930s survey of indentured women found that the mean indenture for licensed prostitutes was 900 yen; for geisha, 800 yen; for bar maids, 400 yen; and for factory workers, 130 yen [Shakai (1935:159)]. Fourth, Fruin carefully documents a steady shortening in contracts for indentured labor for several denturies up to the Meiji period.

brought substantial stigma upon herself and her family. Because of this stig‑

ma, she would take the job only if she could expect to earn a total amount that at least compensated her for this "reputational loss." To be sure, one can

exaggerate the loss. All else equal, the women for whom the loss was least

severe should have chosen the job most often. And one study (of 300 licensed prostitutes who quit in violation of their employment contracts—not a random

sample) did suggest that ex-prostitutes were not necessarily social outcasts: 29 percent returned to their parents, 12 percent took factor jobs, 5 percent found office work, and most of the rest took other "respectable" jobs like nurses' aid positions.55 Nonetheless, most contemporaneous sources suggest instead that prostitutes skirted the margins of respectability if licensed geisha, and aban­doned all claims to it if not. Quoting a Japanese source, one astute (and otherwise relatively nonmoralistic) turn-of-the-century Tokyo observer cap­tures what seems the general consensus: an ex-prostitute "has violated the virtue of chastity, wasted the flower of her youth in vicious living, and as she is unaccustomed to attend to the proper duties of women her future prospects are anything but cheerful and reassuring" (de Becker:248).

Second, the would-be prostitute had little reliable information about how much money she would earn in the industry. Given that the stigma attached when she entered the industry, moreover, she had no cheap way to discover her market "price." And given the uncertainty of the aging process, even if she knew her current price she had no way of knowing how rapidly it would decline over time.

Third, the would-be prostitute could not diversify her human capital. She was her own most valuable asset, yet that asset was not one whose risks she could easily diversify.

The brothel owners faced very different problems. Theirs were not prob­lems of information or diversification. First, they had much better information about the sexual services market than a recruit. In most cases, therefore, they could estimate what a recruit would earn more accurately than the recruit herself could. Second, they could diversify their investments. When they invested in a prostitute's human capital (by buying her services for several years), they could eliminate much of the associated risk by simultaneously contracting with several prostitutes.

Instead, the brothel owners' problem was one of credibility. Although they hoped to recruit young women by promising them money, they could not easily make their promises about future earnings credible. For although they had better information about the potential recruit's earning capacity than the recruit herself, they also had an incentive to lie. They were asking her to suffer a certain and significant reputational loss, in exchange for income that was risky at best and about which they had considerable incentive to exaggerate. To be sure, her investments were not specific to the brothel—but neither does

55. Ito (494-5). In his fine study, Garon (19) discusses evidence finding that large numbers

(42 percent) of ex-prostitutes returned home to their parents and that large numbers (7h) eventually married.

the point matter: once a brothel owner had enticed a recruit into the industry by exaggerating her market earning potential, she could not improve her position by switching to another brothel.

4.5.2 Piece-Rate and Flat-Rate Contracts. These problems of informational asymmetry and promissory credibility foreclosed a piece-rate (productivity-based) contract between the brothel owner and the recruit. The contract would have solved one basic problem in the industry: how to motivate a recruit to work hard at what was a fundamentally squalid job. Unfortunately, it did not give the recruit the assurance she needed. She needed to know—before she entered the industry—that her earning as a prostitute would exceed her earn­ings elsewhere by an amount large enough to offset the stigma and squalor involved. That, however, was something the piece-rate contract could not assure her.

A flat daily wage did little to improve matters. Most obviously, it gave the prostitute an incentive to work less hard. But it also gave the brothel owner an incentive to fire workers who generated revenues less than their contractual daily wage, and to discharge even relatively productive workers during low demand periods (or to threaten to discharge them unless they agreed to a wage cut). Given that a prostitute could expect the revenue she generated to decline over time (Kusama:207), the problem was particularly acute—for she had to worry that the owner might fire her before she had had a chance to recoup her initial reputational loss.

4.5.3 Fixed-Term Contracts and Sign-on Bonuses. Because prostitutes in­curred a high initial fixed cost, brothel owners could recruit them effectively only if they credibly promised to employ them long enough for the prostitutes to recoup a significant part of that fixed cost. The owners could recruit effectively, in short, only if they promised the prostitutes not just a minimum daily wage, but also a minimum total wage. To do so, they had to promise the prostitutes a minimum employment term besides a minimum wage. Even with such a guaranteed term, though, the problem of credibility remained. To promise credibly that the owner would employ a recruit for a given period of time, the owner had to be able to promise that he or she would keep the recruit employed even if her revenues fell permanently below her contractual wage. Such a promise is not one most owners would have found easy to make credible.

A sign-on bonus could have alleviated some of the problems of credibility. The brothel owner might have paid the prostitute enough money up-front to offset a substantial portion of her reputational loss. That ploy, however, would now have given the prostitute an incentive to pocket the money, quit, and move to a rival brothel.56 What the brothel and prostitute needed was a

56. Note, however, that the geisha houses did organize a central clearinghouse through which they enforced on each other an agreement not to hire any geisha who reneged on her contract with

another geisha house (ChM shokugy6:392).

 

contract that credibly promised a minimum total compensation package—but did so without simultaneously giving the prostitute an incentive to quit.

4,5.4 Indenture Contracts. The indenture contract (described in Section 4.2) mitigated many of these problems. Under the contract, the owners gave the prostitute a minimum daily wage: the indenture advance plus room and board, divided by the maximum indenture term. They gave her an incentive to work hard: the chance to quit early. They gave her a minimum total compensation package: the indenture amount plus room and board. And they gave her a disincentive to quit or move elsewhere: the requirement that she repay that part of her advance she had not yet earned.

That much the parties could have accomplished with an unindentured contract that combined a minimum daily wage with a performance-based bonus and a minimum contractual term. Yet the indenture contract differed from such a hypothetical contract in one critical way: under the indenture contract, the brothel owners paid the prostitute a substantial portion of her earnings up front, and thereby allocated the burden of invoking the legal system in a dispute to themselves.

And in agreeing to bear the cost of invoking the legal system in a dispute, the brothel owners helped make their promises credible. For credibility can derive from the assignment of legal costs. If a dispute arose under the hypo­thetical minimum-wage—bonus—minimum-term contract (e.g., if the brothel owner reneged on the promised wage), the prostitute had to invoke the legal apparatus to recover her promised wages—for the brothel owner had not yet paid her the total promised compensation. By contrast, if a dispute arose under the indenture contract, the brothel owner had to invoke the legal sys-tem—for the owner had already paid her three- to six-years' wages. Effective­ly, the brothel owner's willingness to advance the prostitute her promised earnings made the promise to pay those earnings credible. Were the legal system free, such a loan would not add credibility. But legal systems are not free. In Japan as here, possession is nine parts of the law. Because most peasants would have found the legal system foreign and intimidating, the most straightforward way to make them a credible promise was to advance them cash on the barrel.

Note that the indenture contract not only helped make the brothel owner's promises credible, it also allocated the cost of invoking the legal system effi­ciently. Recall that parties generally bargain in the shadow of the law.57 As a result, if a dispute occurred, either the owner or the prostitute would have invoked (or threatened to invoke) the legal system. Hence, the efficient deal was the one that placed that burden on the party able to invoke the system most

57. See Mnookin and Kornhauser, and Ramseyer and Nakazato (1989). Note that prostitutes would have received information about their legal rights not just by word of mouth from other prostitutes, but also from the abolitionists. During most of the period from 1890 to 1940, a wide variety of abolitionists worked hard to encourage prostitutes to quit. One detailed description of the process appears in Murphy; accounts also regularly appeared in such abolitionist journals as Kakncei.

cheaply. 58 As a repeat player, most brothel owners could afford to invest in the information and legal talent necessary to manipulate the judicial apparatus. By contrast, most prostitutes and their parents had little education or sophistica­tion, and as one-shot players were less likely to find it cost-effective to learn how to use the legal system. Given these cost asymmetries, the parties did what Coase predicts: they placed the transactions costs (here, the cost of marshaling the legal system) on the party that could bear them most cheaply. Under the contracts, if the prostitute tried to renege, the owner had to recover his or her money. If the owner tried to renege, the prostitute already had the cash.59

1.             Coda

By 1957, this long-standing legal market for commercial sex was gone. Douglas MacArthur and his advisers had begun the change in 1945, when they ordered the Japanese government to cancel the prewar licensing sys-tem.60 Ten years later, the Japanese Supreme Court continued the process by declaring the loans accompanying indentured prostitution contracts void [Fujita v. Okazaki, 9 Saihan minshU 1616 (S. Ct. Oct. 7, 1955)]. Since 1896, indentured prostitutes had legally been able to abandon their job whenever they pleased. Under the prewar cases, however, they and their guarantors were liable on the debt if they did. Now, indentured prostitutes could abandon their debts as well. The final blow followed quickly. Within a year of the Supreme Court decision, the abolitionists convinced the Japanese Diet to ban prostitution [Baishun bOshi hO (Prostitution Prevention Act), Law No. 118 of May 24, 1956]. Prostitution still continues, but only in illegal form.

1.            Conclusion

In 12th-century Toulouse, the public brothels split their profits with the local university (Shadwell). Not so in Japan. Japanese brothels never tried to buy academic support, and never had it. Instead, academics have consistently criticized the brothels for "enslaving" peasant women. Indenture contracts were an important part of the tales they told: poor and unsophisticated, peas­ants unwittingly accepted indenture contracts that the brothel owners used to reduce prostitutes to sexual slaves.

These tales cheat the prostitutes of their due—for they drastically under‑

1.                Even if litigation rarely ensued, imposing the litigation burden on the party able to invoke the courts most cheaply was still efficient, because it lessened the possible scope of ex post opportunism in the negotiations over a dispute.

2.                1 located only three suits by prostitutes (all geisha) against their houses for nonpayment of earnings. I found seven cases in which parents or guarantors complained about the house's (or brothel's) seizure of their assets (based on claims that the prostitute had quit in violation of her contract), and another 12 cases where a house or brothel sued the prostitute or her parents or guardians for nonpayment on her debt. Reported cases are not, of course, a representative sample.

3.                See Oppler (158). Naimu shö rei (Home Ministry Order) No. 3, Feb. 2, 1946; Naimu shö keiho kyokucho köanhatsu kö (Home Ministry Police Office Chief Public Safety Promulgation) No. 9, Feb. 2, 1946; Choku rei (Imperial Order) No. 9, Jan. 15, 1947.

state the resourcefulness that they could show, even in the direst situations. Although prostitution was harsh work, most brothel owners were not able to manipulate indenture contracts to keep prostitutes at work indefinitely, and most prostitutes did not become slaves. Instead, licensed prostitutes generally enlisted under six-year indenture contracts. They earned (what were for them) very high incomes. Many repaid their debts in three or four years and quit early. Most of the rest quit when their contracts expired.

Within this world, the indentures helped make the employment market itself possible—for despite the promises of high incomes, a woman entering the industry for the first time could never be sure. She knew she and her family suffered a loss in social status if she took the job, knew some brothel owners had an incentive to lie about the money she would make, and knew most owners would be able to invoke the courts more easily than she. Pre­cisely because she could never be sure of the money, she found the indenture contract advantageous. Through the contract, the brothel owners could prom­ise her total earnings large enough to offset a substantial part of her lost status, could make that promise credible by paying her in advance, and could shoul­der the costs of invoking the legal system themselves.

The point is not that licensed prostitution and indentured servitude were necessarily "good for Japan"; neither is the point that, overall, peasant wom­en benefited from the availability of a legal market in prostitution. As noted earlier, both issues are beyond the scope of this essay. Instead, the point is more limited: Given the substantial stigma women incurred in entering the industry, many women hesitated to take jobs at brothels (and many parents hesitated to send their daughters to brothels) without some assurance that they would earn much higher wages than they could earn elsewhere. The inden­tured contracts offered that assurance.

References

Chüã shokugyã shOkai jimukyoku. 1926. Geishogi shakufr shokaigyo ni kansuru chosa (An Investigation into the Placement Industry for Geisha. Prostitutes, and Bar Maids]. Reprinted in Ken'ichi Taniguchi, ed. 1971. Kindai minshu no kiroku [A Report of the Modern Populace], Vol. 3. Tokyo: Shin jinbutsu Orai sha.

Cloud, Patricia, and David W. Galenson. 1987. "Chinese Immigration and Contract Labor in the Late Nineteenth Century," 24 Explorations in Economic History 22-42.

Coase, R. H. 1960. "The Problem of Social Cost," 3 Journal of Law and Economics 1-44. Corbin, Alain. 1990. Women for Hire: Prostitutes and Sexuality in France after 1850. Cambridge: Harvard University Press.

Dalby, Liza Crihfield. 1985. Geisha. New York: Random House.

de Becker, Joseph Ernest. 1899. The Nightiess City, or the "History of the Yoshiwara Yükwaku." Yokohama: Z. P. Maruya and Co.

Emer, P. C., ed. 1986. Colonialism and Migration: Indentured Labour Before and After Slavery. Dordrecht: Martinus Nijhoff Publishers.

Engerman, Stanley L. 1973. "Some Considerations Relating to Property Rights in Man," 33 Journal of Economic History 43-65.

 1986. "Servants to Slaves to Servants: Contract Labour and European Expansion," in

P. C. Emer, ed., supra, pp. 263-94.

Feeny, David. 1989. "The Decline of Property Rights in Man in Thailand, 1800-1913," 49 Journal of Economic History 285-96.

Fishback, Price V. 1986a. "Did Miners 'Owe their Souls to the Company Store'? Theory and Evidence from the Early 1900s," 46 Journal of Economic History 1011-29.

- 1986b. "Workplace Safety during the Progressive Era: Fatal Accidents in Bituminous

Coal Mining, 1912-1923," 23 Explorations in Economic History 269-98.

,and Dieter Lauszus. 1989. "The Quality of Services in Company Towns: Sanitation in

Coal Towns During the 1920s," 49 Journal of Economic History 125-44.

Fruin, William Mark. 1973. "Labor Migration in Nineteenth-Century Japan: A Study Based on

Echizen Han." Ph.D. Dissertation, Stanford University. Ann Arbor: University Microfilms. Fukumi, Takao. 1928. Teizo ni okeru bai'in no kenkyu [A Study of Prostitution in the Capital].

Tokyo: Hakubunkan.

Fukuoka chihO shokugyo shbkai jimu kyoku. 1928. Dekasegi jokö ni kansuru chôsa [A Survey of Female Industrial Workers Employed away from Home]. Fukuoka: Fukuoka chihö shokugyO shbkai jimu kyoku.

Galenson, David W. 1984. "The Rise and Fall of Indentured Servitude in the Americas: An Economic Analysis," 44 Journal of Economic History 1-26.

Gardiner, C. Harvey. 1975. The Japanese and Peru, 1873-1973. Albuquerque: University of New Mexico Press.

Garon, Sheldon. 1991. "The World's Oldest Debate?: Prostitution and the State in Imperial

Japan, 1868-1945," unpublished manuscript, Princeton University, Department of History. Grubb, Farley. 1985. "The Incidence of Servitude in Trans-Atlantic Migration, 1771-1804," 22

Explorations in Economic History 316-38.

 1988. "The Auction of Redemptioner Servants, Philadelphia, 1771-1804: An Eco‑

nomic Analysis," 48 Journal of Economic History 583-603.

Hane, Mikiso. 1982. Peasants, Rebels, and Outcasies: The Underside of Modern Japan. New York: Pantheon Books.

Harsin, Jill. 1985. Policing PrOstitution in Nineteenth-Century Paris. Princeton, N.J.: Princeton University Press.

Higuchi, Monta. 1921. Geisha teisugaku [Geisha Philosophy]. Tokyo: Jitsugaku kan. Hobson, Barbara Meil. 1987. Uneasy Virtue: The Politics of Prostitution and the American Reform Tradition. New York: Basic Books.

Ishijima, Kamejiro. 1928. "Chikagoro no jiyu haigyo [Recent Free Cessation]," June Kakusei 24.

ItO, Hidekichi. 1931. KOtöka no kanojo no seikatsu (The Lives of Women under the Red Lamps]. Tokyo: JitsugyO no nihon sha. Reprinted in 1982. Tokyo: Fuji shuppan.

Kakusei. 1931a. "Shogi karyUbyo cho [A Survey of Venereal Disease Among Licensed Pros­titutes]," March Kakusei 20-I.

 193 lb. "ShOws gonen kyUseigun no haishO undO [The 1930 Activities of the Salvation

Army in Furtherance of the Abolition of Licensed Prostitution]," June Kakusei 29-31.

Kanzaki, Kiyoshi. 1953. Musume o uru machi [Towns that Sell Daughters]. Tokyo: ShinkO shuppan sha.

Kawashima, Takeyoshi. 1950. "Jinshin baibai no rekishiteki seikaku [The Historical Character of the Purchase and Sale of Humans]." Reprinted in Kawashima, Takeyoshi. 1982. Kawashima Takeyoshi chosaku shi (The Collected Works of Takeyoshi KawashimaJ, Vol. 1, pp. 80-96. Tokyo: Iwanami shoten.

                 1951. "Jinshin baibai no hOritsu kankei (1) [The Legal Status of the Purchase and Sale of

Humans: 1]," 68 HOritsu kyokai zasshi 699-712.

 1955. "Jinshin baibai keiyaku no hOteki koryoku [The Legal Effectiveness of Contracts

for the Purchase and Sale of Humans] ," Sept. Horitsu jihô 72.

[Keishi chO] sOkan kanbO bunsho ka. 1933. Shöwa nana nen keishi chO rokei ichi ippan (An Outline of Police Agency Statistics for 19321. Tokyo: SOkan kanbO bunsho ka.

Keishi chO. 1935. Tamanoi Kameido sekkyakufu honseki narabi zenshaku kin shirabe [A Survey of the Indentures and Hometowns of Tamanoi and Kameido Hostesses]. Reprinted in Nihon (1980), infra.

Klein, Benjamin, and Keith B. Leffler. 1981. "The Role of Market Forces in Assuring Contrac‑

tual Performance," 89 Journal of Political Economy 615-41.

Kusama, Yasoo. 1930. Jokya to baishO ft (Waitresses and Prostitutes). Tokyo: Hanjin sha.

KyOkOka no Töhoku nOson (fukkoku ban) (Reprint: The Northeast Farm Villages during the

Panic). 1984. Toyko: Funi shuppan.

Lee, Tom K., and I. P. L. Prig. 1990. "The Role of Installment Payments in Contracts for

Services," 21 Rand Journal of Economics 83-99.

Minami, Rydshin. 1986. The Economic Development of Japan: A Quantitative Study.

Houndmills, U.K.: The Macmillan Press, Ltd.

Mnookin, Robert H., and Lewis Kornhauser. 1979. "Bargaining in the Shadow of the Law: The

Case of Divorce," 88 Yale Law Journal 950-97.

Murakami, Nobuhiko. 1971, Part II; 1972, Part III. Meiji josei s/ri (A History of Meiji Women).

Tokyo: Riron sha.

Murphy, U. G. 1909. The Social Evil in Japan, 4th ed. Tokyo: Kyobunkan.

Mustang Ranch, Inc. 1989. Common Stock Prospectus.

Nailcaku tOkei kyoku. 1930. ROdO tOkei yoran (Outline of Labor Statistics). Tokyo: Thkyo tOkel

kyokai.

Nakamura, Saburo. 1954. Nihon baishun torishimari kO (A Treatise on the Regulation of Japa‑

nese Prostitution). Tokyo: Nihon flizoku kenkyO kai.

Nardinelli, Clark. 1982. "Corporal Punishment and Children's Wages in Nineteenth Century

Britain," 19 Explorations in Economic History 283-95.

Nihon fun mondai shiryO shdsei [Collected Materials on Japanese Women's Issues]. 1980.

Tokyo: Domesu shuppan.

Nishimura, Nobuo. 1939. "Zenshakkin keiyaku ni tsuite [Regarding Indenture Contracts]," 7

MinshO hO zasshi 418-34, 1022-37.

NOmi, Yoshihisa. 1980. "Case Comment," 97 Hogaku kyOkai zasshi 577-85.

O'Callaghan, Sean. 1968. The Yellow Slave Trade: A Survey of the Traffic in Women and

Children in the East. London: Anthony Blond.

Ohsato, Katsuma, ed. 1966. Meiji ikO honpo shuyO keizai zOkei (Principal Economic Statistics for

our Nation since the Meiji Period). Tokyo: Bank of Japan.

Oki, Susan Moller. 1990. "Feminism, the Individual, and Contract Theory," 100 Ethics 658‑

69.

Okubo, Hanayuki. 1906. KagaifOzoku s/ri [A Record of the Customs of the Red-light District].

Tokyo: Ryubun kan. Reprinted in 1983. Tokyo: Nihon tosho sentaa.

Oppler, Alfred C. 1976. Legal Reform in Occupied Japan. Princeton, N.J.: Princeton University

Press.                        -

Pateman, Carole. 1983. "Defending Prostitution: Charges against Ericsson," 93 Ethics 561-5. Popkin, Samuel L. 1979. The Rational Peasant: The Political Economy of Rural Society in Vietnam. Berkeley: University of California Press.

Radin, Margaret Jane. 1987. "Market Inalienability," 100 Harvard Law Review 1849-937. Ramseyer, J. Mark. 1989. "Water Law in Imperial Japan: Public Goods, Private Claims, and Legal Convergence," 18 Journal of Legal Studies 51-77.

_  and Minoru Nakazato. 1989. "The Rational Litigant: Settlement Amounts and Verdict

Rates in Japan," 18 Journal of Legal Studies 263-90.

ROddshO fujin shdnen kyoku. 1952. Fujin rOdO nojitsujO [The Reality of Female Labor]. Tokyo: ROdöshö.

_  1953. NenshOsha no tokushu koyO kankO [Special Employment Customs of Minors].

Tokyo: RoddshO.

Rosen, Ruth. 1982. The Last Sisterhood: Prostitution in America, 1900-1918. Baltimore: Johns Hopkins University Press.

Rosenzweig, Mark R., and Oded Stark. 1989. "Consumption Smoothing, Migration, and Mar­riage: Evidence from Rural India," 97 Journal of Political Economy 905-26.

SaitO, Naoshi. 1930. "YOjo yUgei jin no uchimaku [The Inside Story of Young Female Enter­tainers]," 135 Jikei 67-9.

Saxonhouse, Gary R. 1976. "Country Girls and Communication Among Competitors in the

Japanese Cotton-Spinning Industry," in Hugh Patrick, ed., Japanese industrialization and its Social Consequences, pp. 97-125. Berkeley: University of California Press.

Seiji keizai kenkyU sho. 1951. Nômin to rödô s/ia no seikatsu su{jun no hendö [Changes in the Living Standards of Farmers and Workers]. Tokyo: Seiji keizai kenkyü sho.

Shadwell, Arthur. 1911. "Prostitution," in Encyclopaedia Britannica, 11th ed., Vol. 22, pp. 457-64. New York: Encyclopaedia Britannica Co.

Shakaijigyo kenkyujo. 1936. ShürO shOnen shöjo rôdO jijo chösa [Survey of Working Conditions of Working Boys and Girls]. Tokyo: ChUö shakai jigyo kyokai.

Shakai kyoku. 1935. Tahoku chihã nôson hihei jokyo (dai ni hen) [Poverty in the Northeast Region Farm Villages), 2nd ed. Reprinted in K9ok6ka (1984), supra.

Shrage, Laurie. 1989. "Should Feminists Oppose Prostitution?," 99 Ethics 347-61.

Sievers, Sharon L. 1983. Flowers in Salt: The Beginnings of Feminist Consciousness in Modern Japan. Stanford, Cal.: Stanford University Press.

Smethurst, Richard J. 1986. Agricultural Development and Tenancy Disputes in Japan, 1870­1940. Princeton, N.J.: Princeton University Press.

SOmu chO tOkei kyoku. 1987. Ni/ion choki tokei sdran (General Long-term Statistics for Japan], Vol. 4. Tokyo: SOmu chö.

Stewart, Watt. 1951. Chinese Bondage in Peru: A History of the Chinese Coolie in Peru, 1949­1874. Durham, N.C.: Duke University Press.

Suehiro, ltsutarO. 1931. "Hanrei wo töshite mita jinshin baibai Inc Purchase and Sale of Humans, Viewed through Case Law]," Sept. HOrisu jiho 3-5.

Symanski, Richard. 1981. The immoral Landscape: Female Prostitution in Western Societies. Toronto: Butterworth and Co.

Tamura, Gorö. 1956. "Zenshakkin mukö no hanketsu ni tSuite [Regarding the Case Holding Indentures Void]," 63 Hogaku shimpö 436-63.

Uemura, Yukitada. 1918. Urare yuku onna [Sold Women]. Tokyo: Daitö kaku. Reprinted in 1982. Kindai fujin mondai meichosaku shü zokuhen [A Collection of Famous Authors on Women's issues-Continued Series], Vol. 5. Tokyo: Nihon tosho sentaa.

___  1929. Ni/ion yüri shi [A History of the Japanese Pleasure Quarters]. Tokyo: Shun yo dO.

Wagatsuma, Sakac. 1923. "Hanrei yori miru 'kO no chitsujo zenxyO no flizoku' ['Public Order and Good Customs,' as Seen in the Case Law)," 41 HOgaku kyokai sass/i 904-46.

 1955. "Zenshakkin mukO no hanketsu [Case Holding Indenture Contracts Void]," 93

Jurisuto 23-5.

Walkowitz, Judith R. 1980. Prostitution and Victorian Society: Women, Class, and the State. Cambridge: Cambridge University Press.

Williamson, Oliver E. 1983. "Credible Commitments: Using Hostages to Support Exchange," 73 American Economic Review 519-40.

1985. The Economic Institutions of Capitalism. New York: The Free Press.

Yamamoto, Shun'ichi. 1983. Ni/on kOshO s/i [A History of Licensed Prostitution in Japan]. Tokyo: ChOO hOki shuppan.

Yonekura, Akira. 1985. "HOritsu kOi [Legal Acts]," 59 Hogaku kyoshitsu 30-44; 60:28-42; 61:118-33; 62:30-45.

Yoshimi, ShOko. 1984. Baishd no shakai s/if [A Social History of Prostitution]. Tokyo: YOsankaku.