Following is the English translation of: NAGASHIMA Yūki, "Kinsei kōki Kyōto ni okeru shōka hōkōnin no koyō to saiseisan." Jinbun chiri / Japanese Journal of Human Geography 67.1 (2015): 1–19.
*For citations, please use the original publication.
“Employment and Replacement of Merchants’ Servants in Early Modern Kyoto: A Case Study of the Endō Household,” by NAGASHIMA Yūki.
Translated by Wendy Matsumura
***
I. Introduction
(1) Objectives and previous research
This study explores the nature of the labor market in late Edo period Kyoto, labor migration to Kyoto, and the regional structure of Kyoto and its surrounding areas, based on the employment and reproduction[1] of merchant’s servants. Previous studies on such themes have been conducted mainly in the historical demography field that focus on “labor mobility,” which have utilized the documents, “Shūmon-aratame-chō.” While research on “mobility” within historical demography are limited,[2] there are studies that focus on the labor mobility of servants within the field of economic history that utilize the methods of historical demography.[3]
Hayami Akira, who has driven the field of historical demography in Japan, has clarified the impacts of temporary migration for work in service (dekasegi hōkō) of tenant farmers to the cities upon the reproduction of the population, as well as on mobility between strata, through his long-term behavioral observation studies of peasants from Nishijō village in Mino province.[4] His work was an important contribution to research in the fields of population transfer and labor mobility. For example, scholars such as Matsuura Akira,[5] who concluded that there was a status-based labor movement of labor from countryside to city in Hanakuma village, Settsu province, and Sasaki Yōichirō,[6] whose research revealed the entry of people from Takayama, Hida province during the late Edo period, continued his work. Following these case studies, research on relatively larger cities also began, with Hayami’s works[7] on Kyoto and Nara often being cited as examples.
The development of historical demography impacted the discipline of geography, which began to investigate regional structures using methods from historical demography. Those directly address the question of labor mobility include: Asanuma Misao’s work[8] that analyzes the inflows and outflows of servants to and from Kurome village, Echizen province; Tsuchida Ryōichi,[9] who investigates the entry of yearly and seasonal servants into Mikka-machi in Kōfu, Kai province; Mizoguchi Tsunetoshi,[10] whose work demonstrated that in the early modern period, multiple hamlets in Kai province were suppliers of yearly seasonal servants to Kōfu and Edo; and Asano Kiyotaka,[11] who has investigated the movement of servants from 54 villages in Echizen province. There are other studies that examine other forms of labor mobility, including movement due to marriage.[12]
Building on these studies, Saitō Osamu[13] argued that the bakumatsu labor market among merchant households developed a dual structure between those who were employed as front office (tana-omote) servants and domestic (oku), comprised of servant men (genan), servant women (gejo), and other miscellaneous workers. In addition to this dual structure, we see that from the late 18th century onward, when the merchant households of Edo and Osaka are compared, the former, with the exception of one front office, saw live-in servants from the countryside replaced by urban miscellaneous workers (short term/temporary hires). In the latter, all of the merchant households relied on long-term live-in servants. These were tendencies at both ends of the spectrum.[14]
Saitō’s work points to the relationship between the changes to the urban labor market and the movement of labor from the countryside. Since then, other works have followed including Takahashi Miyuki,[15] who has written about Kōriyama [16] from a similar framework, and Washizaki Shuntarō, who has written about Hachiōji as a rural town (zaigō-machi). The next important point of inquiry seems to be larger urban areas like Edo, Kyoto, and Osaka. This is because when thinking about the positioning of these major cities, which were centers of politics, the economy, and culture, research on them is relatively sparse.[17]
Of the three major cities, research on Kyoto is most abundant. In addition to Hayami,[18] whose research utilized the “Shūmon-aratame-chō,” Nakano Takashi,[19] Akiyama Kunizō & Nakamura Ken,[20] and Mary Louise Nagata[21] have taken Kyoto as their site of inquiry. They have utilized the various municipalities’ historical archives, and together, provide a picture of high degrees of fluidity among residents. We are also able to gather the municipalities with high and low rates of employment of servants, and from this, understand the heterogeneity of conditions within cities. Building on these works, Hamano Kiyoshi[22] examined the demographic indices of Kyoto city through the “Shūmon-aratame-chō,” and revealed that the proportion of the servant population in Kyoto during the bakumatsu period was 14%, and that this was quite a bit lower than the proportion in Osaka, which was over 30%. The major reason for this discrepancy was likely the fact that in the central regions where the front office (tana-omote) operations that relied on the employment of servants remained constant, while their numbers in surrounding areas fell. Therefore, from mid- to late Edo, the changes in Kyoto’s labor market resembled the pattern seen in Edo, but was not as dramatic as there, where we can observe the almost complete disappearance of servants employed in anywhere but the most centrally located municipalities. Further, Hamano[23] also concludes through his analysis of 26 Kyoto municipalities’ “Shūmon-aratame-chō” that the servants from other provinces comprised 24.5%, or 23.3% men and 26.8% women.
Building on this research on Kyoto, geography offers the following observations. First, little attention has been devoted to understanding the relationship between birthplaces, positions, and regional affiliations of servants. For example, what is the relationship between a servant’s birthplace and what specific tasks they took on in their site of employment? Further, while the general observation has been that a large proportion of servants working in Kyoto were from Kyoto, what was the internal distribution like? It is possible to deepen more spatially oriented perspectives to the existing research.
This kind of work that examines the labor market and labor mobility within early modern Kyoto is not meant to simply supplement research in historical demography, but can contribute meaningfully to geographic studies. The relationship between Kyoto and neighboring regions, and in turn, a better understanding of the regional structure, can be clarified. While geographic studies of early modern Kyoto have been conducted in a variety of ways, very few works exist on the relationship between nearby regions, labor mobility, and forms of non-labor market driven movement.[24] One reason for this paucity may be that the kinds of documents that have been discovered and made publicly available through edited volumes like Kyōto no rekishi[25] have been so voluminous that any kind of exhaustive analysis has been difficult.[26] Hamano’s aforementioned work of historical demography is a valued contribution in this regard, and this article also seeks to build upon that research from the field of geography.
This article analyzes the relationship between employment and birthplace of servants through a close examination of one merchant household. The archives of this merchant household have much more information about individual servants under their employ than the “Shūmon-aratame-chō,” making it easier to conduct a concrete analysis. Existing studies of servants of merchant households are limited, and include Mitsui Bunko[27] and Nishizaka Yasushi,[28] and Nakano[29] on the pharmaceutical wholesaler Yamatoya. These have focused on business history, urban social history, and dōzoku-dan studies, and have not paid attention to the labor market.
In the article’s second section, I focus on concrete employment of merchant household servants, and seek to draw some conclusions about 19th-century Kyoto’s relationship between the labor market, labor mobility, and surrounding regions.
(2) Archives and research methodology
The archival materials utilized in this piece come from the Endō (Kō) household archives (hereafter, Endo household archives) that are held in the Kyoto City Library of Historical Documents (Kyoto-shi rekishi shiryō-kan).[30] Within these archives, the main ones utilized here total 92 items of servant contracts (hōkō-nin ukejō)[31]: are the bakumatsu “Shūmon-aratame-chō” and the 65 items from 1805–1842 from the Main house (honke) (Yasaburō household), the 27 items from 1855–1864 from the Branch house (bunke) (Yasaku household). The servant contracts are personal guarantor contracts between employer and the guarantor,[32] and are useful for ascertaining the terms of employment.
One advantage of using these servant contracts is that they include the birthplace of both servant and guarantor, down to their residence to the municipality or village unit. For example, while many of the “Shumon Aratamecho” records the home temple (danna-dera) and the birthplace province, and it is possible to infer more details from that, we are not able to gain a complete understanding. In contrast, we are able to conduct a detailed analysis of the birthplace from the information provided by the servant contract. When thinking about the objective behind creating the servant contract, we have to keep in mind that when the employment relation is dissolved, it is easier for these to be discarded or lost. We cannot deny that this has happened to a certain degree for the Endō household’s records.[33] However, there are not very many merchant houses in Kyoto that have servant contracts in Kyoto, with the exception of the Mitsui household. Considering the archival limitations, I believe this to be a worthy project.
In addition to the servant contracts, I utilize a set of documents called “Mōshi-watashi no koto” that contain the names of the branch families (bekke)[34] where live-in servants were either allowed to commute to or start up. Despite the fact that a lot of front office servants were young people from the branch families,[35] servant contracts were not always created.[36] The “Mōshi-watashi no koto” is an excellent set of materials in order to ascertain the birthplaces of the people who became apprentices of the branch families of the Endō family. In addition, the “Matsuki Shōzaemon-ke monjo” (held at the Fukui Prefectural Archives) and a few other municipal and merchant archives will be used to understand the birthplaces of the servants.
The composition of the article is as follows. First, part II will provide an overview of Kyoto, the history of the Endō family’s business, and the composition of the family and their servants. Part III is an analysis of the duties of the servants that can be gathered from the servant contracts and their relation to birthplace or status. Part IV is a consideration of the branch family residential distribution, which is another route to the employment of servants. Further, it focuses on one of the servants who arrived at the branch family, and will examine their work experience till then, as well as their method of reproduction. Part V summarizes the perspectives gained throughout the piece and will make some concluding comments about the labor market and labor market as understood from the reproduction and employment of servants.
II. Summary of early modern Kyoto and the Endō family
(1) Overview of the region
The foundations of early modern Kyoto’s cityscape were laid by Hideyoshi’s Kyoto reconstruction. Through it, the construction of Jurakudai, Teramachi and Teranouchi, the enactment of the Tenshō land redistribution, the formation of Kugemachi, and the building of Odoi took place. Residential areas were segregated by status.[37] The inside of Odoi was considered inside of the city, while outside was also considered beyond the city boundary. Within the city boundaries, there was a division at the border of Nijo dori between Kamigyo and Shimogyō, and there was another division at East and West Hongwanji Temple at Jinaimachi.[38] Thereafter, with the urbanization of Jinaimachi around Higashi- and Nishi-Honganji and the area east of Kamo-gawa, these also joined with Kamigyō and Shimogyō. With this, the shape of early modern Kyoto took shape (see figure 1). If we turn our attention to the economy, Kyoto was at the center of national trade. Many daimyo were dependent upon Kyoto’s handicraft manufacturers like cloth wovens, and by the middle of the early modern period, the raw materials and semi-processed goods from the various domains were manufactured in Kyoto and sold nationally. This kind of regional formation had developed. This was not limited to Kyoto’s handicraft industry, but also led to the growth of wholesalers who handled the import and export of these products.[39]
The Endō family lived in Shimogyō’s Umetada-chō. Sanjō-dōri, which ran east to west along the city, ran through it. To the west was Karasuma-dōri, and to the east was Higashi-tōin- dōri. During the early modern period, Umetada-chō was under the “Shimogyō kochō minami ushitora jūni-chōgumi”[40] and with the second chōgumi reforms of 1869, it was placed under “Shimogyō yon-bangumi.” During the bakumatsu period, Umetada-chō had 32 ie-yashiki.[41] The Endō family possessed the first and second properties on the northern side from the west, as well as the ie-yashiki on the northeastern part of Sanjō-dōri and Karasuma. The Endō family’s 10th generation head of household Yasaburō served as the head of “Shimogyō yon-bangumi” called chū-doshiyori from 1869, and then from the following year, served as the Shimo-ōgumi’s (Shimogyō) ōdoshiyori joyaku. In these ways, he contributed a great deal to local administration.[42]
(2) The Endō Family and Family Business Operations
Previous research on the Endō Family comes from Nagata,[43] who took up part of the servant contracts. However, since that research did not engage in a clear overview of the family’s activities, this section will lay that out through a combination of interviews with the present head of family,[44] as well as historical and archival materials. An overview of the Endō family and its family business follows.
According to the current head of household and “Nisshō hyakunen-shi,”[45] the Endō family has its genealogical roots in the Ōmi Mikami domain’s Endō family. It began its business in the Jōō era (1652) in Kyoto as a wholesaler that handled linen, cotton, and silk produced in Ōmi. Many of the family heads succeeded the name, “Hiranoya Yasaburō.” There are no extant documents that can substantiate this founding history. However, according to the “Umeda-chō kokenjō utsushi” published in 1846, we learn that of the second yashiki from the west on the northern side, Hiranoya Rizaemon combined two properties that he purchased in 1657 and 1682 respectively, and since passed it down to his successors.[46] Further, in terms of the business, there are records that since the second half of the 18th century, they were in service to the imperial court and feudal lords.[47]
By the start of the Meiji period, the Endō family was dealing in bleached cloth, wadding, and kimono consulting.[48] The “bleached cloth” was linen and the “wadding” was silk. It seems like they had continued the same business as its founding. “Kimono consulting” referred to an intermediary business between the highly specialized dye manufacturer and the customer, the kimono seller, and was an occupation[49] that they began to participate in from the first half of the 19th century in service of feudal lords and the imperial court.[50]
In order to trace how the focus of the business changed over the years, we examine how they appear in the topography and shopping guides in publications like the “Shinshū Kyōto sōsho” and the “Shinsen Kyōto sōsho.” There we see that during the bakumatsu period, the Endō family name (Hiranoya Yasaburō, Endō Yasaburō) appear as “Ōmi wholesalers,”[51] whereas from the Meiji 10s onward, they are listed as “Kimono consultants.”[52] “Ōmi wholesalers” refers to their role as consignment wholesalers collecting Ōmi’s linen, following the public recognition of a merchant guild entrusted by the Shogunate in Kyoto called “Gōshū nuno-don’ya dōmoto shiire nakama” from the mid-18th century. From the bakumatsu period, the Endō family came to play a central role in the merchant guild and served as its head.[53] From this point onward, the early modern family business’s core was determined to be as a wholesaler of linen and silk, and once the merchant guild received its dissolution order in the Meiji period, they switched to kimono consulting.[54]
(3) The composition of the Endō family and servants
When we observe the Endō family composition during the bakumatsu period through the receipts of the “Shūmon-aratame-chō,”[55] we see that in the main family, Yakichi became head of family at the age of 22 in 1857, and thereafter, there is no change in the archival materials. In terms of the head of the branch family, as of 1840, the younger brother of the 9th generation head Yasaburō, Nisaburō had taken over. There are some documentary gaps after that, but when we are able to verify once again in 1853, Yakichi had become head of family at the age of 18, and from the following year till 1857, Yakichi’s mother, Nobu, and then from 1858, Yakichi’s younger brother Yasaku, had become head of family at the age of 20. These changes in the head of the branch family were due to the death of Nisaburō in 1848.[56] As a formality, Yakichi had to first become the branch family head. Thereafter, Nobu took over for Yakichi who succeeded as family head of the main branch, and then Yakichi’s younger brother Yasaku took over. In 1860, Yakichi married Chika, and in 1861 Yasaku married Toshi, and the latter had a daughter and a son.
During the period of 1857 and 1867 when the main and branch family “Shūmon-aratame-chō” exists, the Endō family had between 20 and 30 servants. (see table 1) That figure fell to 18 in 1865, and the reason is thought to be the effects of the Great Fire of Genji during the Hamaguri Rebellion. During normal years, there were between 21–26 male servants, and between 5–8 female servants. If we take a look at their “Provincial Origins” (table 2), we see that around half were from Yamashiro. In actuality, many of them were likely from Kyoto city, but it is not determinable through the “Shūmon-aratame-chō.” If we analyze these figures by gender, we see that there were many men from Yamashiro, and women came more from Ōmi. Further, if we turn to tables 3 and 4 that summarize the age at the start of service to age at end of service, we see that during the bakumatsu period, the male servants were all employed from their later teen years. Their suitability was assessed in a relatively short period of time. Thereafter, we see that the men were dismissed at each stage, and by their early 30s, they were no longer in service. That being said, those who lasted in service until 30 were finally permitted to have their own branch family[57] and they either opened up their own shop or they commuted to the Endō family. On the other hand, the start and end date of service for women is concentrated in their 20s. Most of them end their service within one year.
The servant composition of the Endō family reveals that the male servants in the bakumatsu period were all people who engaged in the business operations department.[58] Many of them came from Yamashiro, and were likely from Kyoto city. On the other hand, the women were largely in their 20s and from Ōmi province.
III. Consideration of the servants’ birthplace
(1) Types of servants and the formation of ukejō
This section utilizes the servant contracts in order to understand the birthplaces of the servants by examining: 1. the official duties of servants; and 2. the status of servants. In this analysis we are assuming that the servant’s birthplace is the residence of their parents.
First, on the official duties of servants, they can be divided into two. The servants on the business operations side, and the servants on the domestic affairs (kaji) side. In terms of their spatial division inside of the family, this would be “tana-omote” and “oku.” Further, if we divide these into more detailed occupational categories, we get the following: in the “tana-omote,” there are tedai/decchi (shop clerk), and in “oku,” there are genan,[59] gejo, and menoto. Tedai are people who worked under the direction of the manager (shihainin) and the head clerk (bantō) did the work of stocking and sales. Prior to the Genpuku era, they were apprentices (unlike decchi) who received direction from the tedai and performed miscellaneous jobs for the head clerk.[60] After serving as decchi and then tedai for a set number of years, they would become independent and set up their own household as a branch family, and either commute from there or have their own store. In contrast, genan and gejo were exclusively there as subordinates.
Servant contracts record information like birthplace (province, town, and village), their parents, religious denomination, their ancestral temple, the name of their guarantor, and the contract term. In general, what was recorded in the servant contract varied depending on if they were “tana-omote” or “oku.”[61] However, since women servants were without exception, domestic servants in “oku,” what needs elaboration is the differences within the male servants; that is, between those classified as “tanao-mote” and “oku,” or tedai/decchi or genan. In the case of the Endō family, we are able to make distinctions in the following manner. First, there is a difference in the contract length. For the tedai and decchi, the contract term is listed as 15 or 10 years, whereas there is no designation for genan. The second difference is in the guarantor security agreement. For the servants who work in “tana-omote,” the guarantor takes on the legal obligation in the following event: “if they engage in business in secret from the master, which leads to a loss, wasted money, and other failures to capture business opportunities.” In contrast, for the genan, the only thing written in as part of the guarantor’s legal obligation is the case of “ketsuraku,” or elopement. This reflects the differences in the kind of work that the tedai/decchi and the genan performed for the family. If we look at the equivalent for women, the servant contract for gejo is almost identical to that of the genan. On the other hand, the servant contract for the menoto includes a clear payment account, and there is a note for how to deal with the “lack of milk.”
From this analysis, we can conclude that the important occupational categories to keep in mind are A, tedai (decchi) within the “tana-omote,” and the B, genan in the “oku,” as well as C, gejo.[62] Since the number of menoto are extremely small, we will include them in our analysis with the gejo.
With regards to the status of servants, we will classify them according to X (those with surnames), Y (those with house names [yagō]), and Z (peasants). These are determined based on the names of their parents listed on the servant contract. We know that people labeled X occupied a relatively high status within early modern society, where the recognition of commoner surnames was restricted. It is a bit of a stretch to say that the servant contracts are official documents, but at the very least, in the Endō family archives, only those who had publicly permitted surnames used surnames.[63] The people classified as Y are people who had a yago in front of their names, and a majority of artisans and merchants had this. Z was for people who had “peasant” in front of their given names, and did not have any surname or yagō listed. These people were not permitted to have surnames and did not belong to a yagō, and were likely of relatively low status.
(2) Duties and origins
In our overview (table 5) of the duties and origins of the servants in order to compare it to earlier studies, we see that they are largely divided into 8 regions within Kyoto and the Northern Kinki region and the Japan Sea side. The furthest away were people from Kaga and Etchū. We did not see people who lived in places like Settsu, Kawachi, or Yamato that were adjacent to Yamashiro. Servants came mainly from regions in Kyoto and north of it, as well as from regions east of Kyoto. In the “Shūmon-aratame-chō hikae,” we see that there was only one servant from southern Kinki, from Izumi (table 2). Earlier research confirms this, with most of the servants in bakumatsu era Kyoto coming from northern Kinki and the Hokuriku region, and the relative absence of people from southern Kinki.[64] When it comes to southern Kinki, we see households that had opportunities to move to Osaka. Further, the region around Nara[65] and Osaka’s Dōjima within a 5 ri radius of Osaka are seen as the “Osaka hinterland” that supply dependents to.[66] In this way, Osaka is quite influential. The places from which the Endō family servants were drawn conformed to this general tendency. Whereas some earlier research argues that the eastern edge of movement to Kyoto was the Nobi region, the sources utilized in this paper did not see anyone from the Nobi region,[67] and its eastern boundary stopped at people from Nagahama in Ōmi province.
It is difficult to determine the exact causes of this formation since historical, economic, and societal factors are all intertwined, but I would like to explain them a bit here. First, the provinces that we take up here all have a strong economic relationship to Kyoto. As already stated, since the middle of the early modern period, Kyoto had been importing commodities, raw materials, and semi-processed goods from various provinces, and had been exporting these throughout the country after turning them into manufactured goods. Provincial specialty wholesalers that were in Kyoto to handle the import of these goods numbered 7 from Kanto, 29 from the Japan Sea coast provinces from Echigo to Tango, 14 from Ōmi, and 3 from Mino.[68] We are able to see a tendency of servants from these regions. Further, it is well known that since the 18th century, in Kyoto, the position of Nishijin, which was Kyoto’s silk producing region, was threatened by the development of other regions’ textile industries. That is, until then, the raw materials and semi-processed goods had been sent to Kyoto, but thereafter, finished products manufactured by regional silk came to be sent into Kyoto and then sold off to other regions. In concrete terms, the import of regional silk to Kyoto in 1756 is majority “Kanto silk,” and in second place is “Kaga silk,” and third is “Tango silk crepe.” From this, we can see that the links between the provinces along the coast of the Japan Sea and Kyoto remain strong even as the imports changed from raw materials and semi-processed goods to manufactured products. Further, the provinces that are along the way from Kyoto to Hokuriku and the Echigo regions like Ōmi and Tanba also have a strong link to Kyoto, the former in its hemp cloth wovens and the latter in its lumber called “Tanba-zai.”[69] In addition to these economic links, the origins of the servants were impacted by the large cities, with Osaka to the west, Nagoya to the east, and Edo further away.
Next, we will investigate the relationship between the tasks and the status of the servants (table 5 and figure 2). First, if we look at A, tedai, we see that the 28 tedai were classified either under “with surname” or “with yagō.” There were none that were classified as “peasant.” This indicates that generally speaking, it was either impossible, or extremely rare for a person classified as a “peasant” to get hired as a tedai. If we look at the region of origin, we see that they were drawn from Kyoto, Yamashiro, Ōmi, Wakasa, and Tanba. Of these, the ones classified as “with surname” were from Yamashiro and Wakasa, meaning that they were from notable regional families. With regards to the category “with yagō,” we see that the majority of people from Kyoto fell under that category, and we also see it in people from Ōmi and Tanba. We cannot see any distinct characteristics here. If we focus on the people from within Kyoto, we see that there is a conspicuous leaning toward Shimogyō. One of the reasons why many people were employed from Shimogyō’s southern region can be explained through the Endō family business. It was one of the families that held monopoly rights to the intake of Ōmi linen through an organization called “Gōshū nuno-don’ya dōmoto shiire nakama.” 26 of the 27 members in 1853 were residents of Shimogyō.[70] They were quite unified, in that when their interests were threatened, they got together and protested at the magistrate’s office. These relationships were valued, and we can guess from this that the business relationships of the Endō family were quite strong throughout Shimogyō.[71] It is thought that they also had a wide range of business relationships in Kamigyō, but there is a high possibility that many of the personal ties they relied upon to employ tedai came from their guild members in Shimogyō. We also see quite a few people from nearby Umetada-chō. As we consider in the next section, there were a lot of branch families of the Endō family around Umetada-chō (figure 3), so their personal ties were quite widespread through these connections with the branch families. Through these various personal ties, hires were drawn not only from Kyoto, but from other regions as well.
Even though the number of servants who were hired as B, genan was quite small, I will conduct an analysis of their origins next. The distribution of origin of servants who were classified as B was along the Japan Sea coast, and none are recorded from the Kyoto or Ōmi regions. In terms of their status, they were only drawn from those “with yagō” and “peasants.” Further, all five of the genan who were from Tanba, Wakasa, Echigo, and Tajima respectively were named “Shinpachi,” but we see that their start date of employment was quite close to each other.[72] As it is not thinkable that they would give servants who came under contract at around the same time the same name, we can conclude that the genan were cycled in and out at relatively short intervals. We can conclude that the genan were employed through connections with boarding. This type of employment of genan was pretty integrated by origin and status, as Mitsui Echigoya Kyō-honten’s example shows.[73]
Despite the fact that they entered into service from quite far away, their terms were relatively short. What we can garner from this is that despite the fact that their terms were short, they did not have trouble finding a new place of employment. That is, despite the fact that genan did not have the personal ties to become tedai, they may have drifted into the large city of Kyoto where there were strong ties to their hometowns in search of employment opportunities. The origins and status of the genan and tedai are not the same, and we see the clear existence of different employment structures.
Finally, if we turn our attention to C, gejo’s origins, we see that Kyoto or regions close to Kyoto like Yamashiro, and Ōmi are prevalent. If we first look within Kyoto, we see that they have a similar pattern as tedai, in that many of them came from Shimogyō. However, we also see that there were a lot of genan from regions that were within 100km of Kyoto. Of this, Ōmi played an outsized role in supplying them with women’s labor power.[74] In addition to the close relationship between Kyoto and Ōmi, one reason for this in the case of the Endō family may be its position as an Ōmi cloth wholesaler.
Women servants from merchant houses at that time went into service as part of temporary migration for work or as part of their preparations for marriage. It is difficult to distinguish between those things in the archival materials, but especially for women who are in their early 20s, the latter reason most likely figured in many cases.
According to tables 3 and 4, the women servants to the Endō family began their service at the age of 23.9 years, which was clearly higher than previous research that examined 16 chō and Shimogyō’s Seidō-chō in Kyoto city.[75] If we look at research from the agrarian villages, we see that Noshu, which was the eastern border of movement into Kyoto, the average age of women with experience in service began their service around 14.4 years, and continued their service for 14.3 years.[76] The average amount of time that gejyo were employed by the Endo family was 1.2 years, and was 1.78 years in the case of Seidō-chō,[77] and many of the women servants moved from place to place to be employed as servants. We can conjecture from this that the gejo of the Endō family, many of whom were older, likely worked at other merchant houses before being employed by the Endō family. It is not clear why this was the case, but a high degree of skill and experience may have been required by the Endō family even for gejo, since they were a major establishment.
If we only analyze the gejo based on their status, we see that there is no major tendency we can identify with the exception of those from Kyoto, where a large number were from merchant and artisan yagō. In other words, for gejo who were from regions other than Kyoto, we see that they were “with surnames,” “with yagō,” and “peasants,” which is a different tendency from what we saw in the case of tedai and genan. What is also notable here is that among the gejo of the Endō family were people who were classified as “kami jochū” who accompanied the headmistress of the household, and the “shimo jochū,” who were in charge of all domestic and miscellaneous affairs.[78] If we think about their status from this perspective, we might conclude that the former were from families mainly “with surnames,” while the latter were mainly those from families of “peasants.” The reason for this is that those from families “with surnames” were of relatively high status, whereas there were not very many well-off families of “peasants.”
From the above, we see that the servants who were employed by the Endō family were divided into tedai, genan, and gejo, and based on these different duties, had a different birthplace and status profile. This suggests that the labor market in Kyoto was quite specialized based on duties.
IV. Branch family residences and their life courses
(1) Overview of branch families
Of the tedai/decchi, this section examines the origins of the shitei (apprentices) of branch families, the people that we did not take up in the previous chapter. We will focus on one concrete figure from the branch family and trace how they became a servant, how they founded the branch family, and how their children came to become servants themselves.
The “Mōshi-watashi no koto”[79] of the ninth month of 1828 that we utilize this chapter has the signature and seal of the people of the Endō family branch family, and was added to every time that a new branch family was founded or when a new successor to the headship of the family was approved. The newest head of family was Jūkichi, who succeeded in the first month of 1920. This document was updated until the end of the branch family system.[80]
If we organize the succession records and the branch family records of the “Mōshi-watashi no koto,” we can count 51 people over 93 years. The branch family takes on the family name; for example, the name “Shinbei” is continued from the first to the fifth (though the fifth is called “Shinpei”). This is the same with the other branch families, but the successor is not limited to a biological child, but could be a relative or a tedai who became the adopted child of the Endō family. However, many of the branch families with sons sent them as decchi servants to the main house. In this way, we can see that the hiring of servants from the branch family meant employment of the closest person within the network of merchant houses.
(2) The residences of the branch families
Of the branch families of the Endō family in Kyoto, we know the locations of 17. We will examine their locations within Kyoto up to the end of the Meiji period. The numbers are not that large compared to the total number of branch families, but if we consider that in many cases, branch families inherited family estates, we may be able to ascertain some kind of trend.
When thinking about the residences of the branch families, we have to also think about the family estates of the Endō family. Even when looking at other families, servants who were granted branch families first resided in the family estate of the main branch,[81] and the family estates of the main branch proscribed to some extent what distributions were granted to branch families. Here, we will examine the land title documents of 1886 in order to think through the ownership conditions of the family estates.
The distribution of the chō within Kyoto city that had Endō family-owned family estates and the chō where the branch families reside is represented in Figure 3. We see that many of the residences of the branch family were near Umetada-chō, and many of these overlapped with where the Endō family had its family estates. In particular, in Tearaimizu-chō (Karasuma-dōri Takoyakushi-sagaru), the family estate of the Endō family and the branch family residences overlapped.[82] On the one hand, there are also branch family residences that were slightly far away from that, including in Sanbongi-chō (Higashi-tōin-dōri Marutamachi sagaru), Tōjiji-chō (Yanagi-no-banba Nijō sagaru), Karigane-chō (Takatsuji-fuya-chō Nishi iru). What these have in common is that they were branch families during the late Meiji period.[83] We do not consider temporal changes to the areas in which the branch families resided here, but what we can see is that the early modern branch families resided in places within walking distance of Umetada-chō, and many of their children became servants for the main branch. As I noted in part III, section (2), these branch families were used as contacts when the main branch employed tedai from places outside of the branch families themselves.
According to the Meiji 2 cadastral map, the two branch families that set up shop on their own had their own family estates.[84] In order to engage in this kind of independent branch family business, they needed a place where they could take advantage of the brand reputation, which meant being in close proximity to the main branch. To provide one example, the aforementioned “Shinbei” operated lodgings in Senshōji-chō (Takoyakushi Higashi-tōin Higashi iru). Many travelers stayed there, and included amongst the lodgers were likely the clients of the main branch. That is, people who had come to Kyoto in order to do business with the Endō family main branch stayed at the lodgings operated by Shinbei.[85]
As seen above, many of the branch families, which were suppliers of front office tedai and decchi resided in close proximity to the main branch. This was because they were affected by the distribution of the family estates owned by the main branch. Further, even in the case where they owned their own family estate and had set up shop on their own, their ties to the main branch were still quite intimate. The branch family was the closest in terms of personal connections to the main branch, and not only did the former send tedai and decchi to the main branches, further hiring was also done through these connections.
(3) Biography of Hiranoya Yasubei
This section will focus on one tedai. Through the details of how he entered into service, started a branch family, and died, will consider the reproduction process of servants. I use two types of family trees that are kept in the Matsuki Shōzaemon family archives held at the Fukui Prefectural Archives as source materials. One of the family trees, (hereafter, family tree A)[86] was created during the early modern period, and continued until 1843. The other family tree (hereafter, family tree B) was created by reorganizing and revising family tree A, and has records until around 1972.
Yasubei’s real name was Jūgorō, and started in service with the Endō family from 1830. According to the servant contract, his parent was Matsuki Shōzaemon of Shindō-mura in Kaminaka-gun, Jakushū, and his guarantor was Tangoya Yahei of Karasuma-dōri Sanjo-sagaru-chō (Manjūya-chō). Shindō-mura was located along the Kurihan-kaidō road that connected Ōmi province’s Imazu and Wakasa province’s Obama. The Matsuki family, which was Jūgorō’s natal home, was a prominent family in the region, and often served the role of village headman.[87]
His father, the 10th generation head of the Matsuki family was the father of 11 children and also welcomed 1 adopted son (Table 6). Of this total of 11 biological children and 1 adopted child, 8 were men and 4 were women. Let us first turn to the circumstances of Jūgorō’s brothers and sisters.
First, 3 of his 4 children, Kameko (no. 3), Tsuruko (no. 4), and Yukiko (no. 5) experiences being in service in Kyoto. Kameko and Tsuruko married into families in Wakasa’s Onyū-gun, Nishizu village, and Obama, respectively, following their 1-2 years in service and a temporary return home. Yukiko married into the same family as Tsuruko as the second wife following the latter’s death. Only Hamako (no. 9) married into a family in Obama without any experience in service. With the exception of Hamako, therefore, all of his sisters went into service for a short period of time in what appears as training for marriage before marrying into families within Wakasa province.
If we turn to the circumstances of Jūgorō’s brothers, we see that their paths can be divided into 3:1. Founding of main family or succession of main family or branch family; 2. Adopted into another family; 3. Service. There are 3 people who fall into the first category. Matsujirō (no. 1) succeeded the headship, Shigezaburō (no. 7) succeeded the branch family, and Sadaemon (adoptee) who revived and succeeded the branch family. Those who fall into the second category are Takejirō (no. 2) who entered into a family in Obama following several years in service in Kyoto but had his marriage dissolved, and as we elaborate later, went back into service in Kyoto, Komanosuke (no. 6) who was adopted into a family in Shindō-mura but had his marriage dissolved and was adopted into another family in a nearby village, and Tomojirō (no. 11) who was adopted into a family in Obama but had his marriage dissolved, returned to his natal home and passed away at the age of 44. Those who fall into the third category of service are Takejirō and Jūgorō (no. 10). As we stated above, following Takejirō’s dissolution of marriage he went back to Kyoto and entered into service for Tangoya Yahei. He then succeeded the family following the death of the head of family. Jūgorō entered into service in the Endō family and worked there until he was granted a branch family.
Compared to his brothers who were granted opportunities to succeed branch families or enter into other houses as adopted sons, Jūgorō was never granted that opportunity. It is unclear why that was the case, but Jūgorō entered into the Endō family as a servant in order to try to advance through his own hard work. He was hired by the Endō family at the age of 13 and was given the name Jūzō. It appears as though his guarantor, named Tangoya Yahei was his brother Takejirō.[88] Thus, Jūzō’s employment as a tedai in the front office was made possible through the regional bonds that the Endō family had with the Tangoya Yahei family.
We can compile the following biography of Jūzō’s life following that from the family tree B, the servant contract, and the “Mōshi-watashi no koto.” At his coming of age, Jūzō was granted the name Yasubei, was granted his own branch family in 1848 at the age of 31 through “chōnen-kan no kinrō” (long years of hard work). In terms of the time that it took for him to be granted a branch family, 19 years seems consistent with the average, but it is clear that he did endure a rigorous selection process during this time. Following the establishment of the branch family, Yasugei resided in Sanmonji-chō (Higashi-tōin-dōri Sanjō sagaru) (Figure 3), which was adjacent to Umedata-chō and married a woman named Satoko who was from Mimasaka province’s Tsuyama. Sanmonji-chō was where his brother Tangoya Yahei ultimately set up his residence, and it may have been the case that they either lived together or resided in a plot he rented from Tangoya Yahei. Yasubei continued to work in service for the Endō family as a commuting branch family, but spent a long time in Mito, as the Endō Family was “the kimono supplier to the Mito,” but “a physical misfortune befell the master while at the Mito compound,” so he had to remain in Mito for a long period of time. Upon returning to Kyoto, he fell ill. He recovered from his illness but after returning to his hometown to escape the worsening safety conditions in Kyoto during the bakumatsu era and to recover further, he died in his natal home on September 28, 1962. According to family tree B, the main branch responded by “building a stele in his honor at Kurodani Konkai-kōmyōji temple and conducting a memorial service.” Further, Kyoto’s Kurodani Konkai-kōmyōji temple provided the burial grounds for members of the Endō family main branch, as well as a “branch family gravesite” and over 20 tombstones from what is thought to be branch family members surrounding it.[89] Yasubei’s “stele” was likely built in the same manner, but the current location is unknown.
According the family tree B, the aftermath of Hiranoya Yaseubei’s death is recorded as “no children, discontinued.” However, on the son of the 12th Matsuki Shōzaemon (this would be Yasubei’s grand nephew), Yasunosuke, it is noted, “first took on the family name of Kyoto’s Hiranoya Yasubei, but following the family’s discontinuation, became adopted successor to the current gun’s Nishizu-mura, Matsumi Sukedayū.” Yasunosuke was called to Yasubei’s family since he had no sons, but following the latter’s return home, the former had to make a decision about his future.
From the above, Yasubei found employment through the regional bonds between the Endō family and his brother, Tangoya Yahei, and after spending close to 20 years in service, was permitted to found his own branch family. Yasubei resided near the main family, and one of the hopes was that he would be able to send his own children into service to the main family. However, as the conditions of Yasubei falling ill and having no children of his own show, this was not a ready-made path. The field of historical demography has shown that the long years in service in Osaka’s merchant houses actually drove up the marriage age and brought forth a lowering of birth rates, leading to a suppression on the macro level of population maintenance abilities of the city.[90] Accurate birth rates for Kyoto are not available, but in the Endō family’s case, the age at which one would be able to marry, as well as the age that they would receive permission to set up branch families was in the late 30s at earliest. We took up just one example in this section, but if we reflect upon it, we cannot conclude that Yasubei was exceptional in not being able to have a son by birth after marriage. In these cases, a relative of the branch family, or some other direct or indirect personal connections were used to hire a tedai, or they adopted someone into their family.
Conclusion
This piece considered the Endō family’s servant employment and reproduction practices in order to better understand Kyoto’s labor market and labor mobility. From the examples presented here, we are able to understand the conditions under which they were employed and socially reproduced, and to gain a clearer sense of regional relationships.
First, the majority of tedai that were employed by the Endō family were from Kyoto’s Shimogyō, and approximately 25–30% were from places other than Kyoto (Yamashiro province, not including Kyoto city, and other provinces). Like general large houses’ employment systems, the tedai employed by the Endō family had to undergo a rigorous selection system until they were able to have their own branch families. Thus, there were not enough people who could provide employment from the branch families alone, so the Endō family relied on its own, as well as the branch families’ merchant relationships in order to draw people from various places inside and outside of Kyoto who were of the appropriate age, origin, and capacity as tedai. Kyoto’s Shimogyō happened to provide the largest numbers, but there were other cases, as Hiranoya Yasubei’s example reveals. The servants who were employed from other provinces also then mobilized their own personal connections to reproduce servants from other provinces. The influence of Kyoto’s economic unification loomed large, as did the effects of general trends of large urban areas. However, due to the fact that employment from within Kyoto city was a priority, and because employment as tedai from other places had to fulfil certain status requirements suitable to the merchant family, the absolute numbers of people from other provinces was not very large.
This kind of employment system was generalized from the second half of the 18th century, and led to the contraction of employment of peasants of a low status from surrounding regions.[91] The genan of the Endō family that we traced in this piece did not have the kind of personal connections or background that would make them eligible to become tedai, but came from areas like the Japan Sea coast seeking employment opportunities in a large city. They eventually ended up being employed by the Endō family after spending short periods of time in service elsewhere. However, from the bakumatsu period, and in particular, between 1843 and 1853, genan as a category disappeared from the Endō family’s employment records. One data point that helps shed light on this is the results of a Kyoto occupational survey of 1872 that reveals that approximately 8%, or 4,242 household units out of 55,186 (a figure that excludes peasant households) were classified as “day laborers.”[92]
On the other hand, if we look at gejo, in addition to people from Kyoto, there were many from Ōmi and surrounding provinces and regions. Even in the bakumatsu period, the number of people hired was steady. One characteristic of the Endō family’s gejo was that their average age was higher than that of other Kyoto merchant houses. One large difference between the gejo and genan who were employed by the Endō family was that even though the former was for a short term, their service continued. What they had in common was that the people from outside of Kyoto tended to move from one merchant house to another. Therefore, the existence of word-of-mouth and of boarding were indispensable for merchant houses to be able to employ new gejo. This applies to shimo jochū (kitchen maid), but for kami jochū (head female servant), we can conjecture that their reproduction was done through personal connections between merchant houses. While the latter numbers are relatively small, more research is required to gain a better sense of their reproduction.
We can conclude that the way that the reproduction of the Endō family’s servants was done in an integrated manner, and thus, is fairly consistent with Saitō’s argument of a two-tiered urban labor market structure. We see the tedai, who were reproduced through the personal connections and branch families of the merchant houses, and the genan and gejo whose employment was more fluid. However, when looking at the Endō family’s example, there were some differences between genan and gejo in terms of where they came from and their tasks. Further, if we read the birthplaces of tedai, genan, and gejo comprehensively, we see that they came from Kyoto city, and the Northern Kinki and the Japan Sea coast regions, which is consistent with the previous research by Hamano. However, the mode of reproduction differed based on their duties/type of employment, and there is a diverse range of personal connections between Kyoto and the surrounding regions. This indicates that in order to more concretely understand the movement of people as laborers from surrounding regions to Kyoto, more study is required around the conditions of movement, types and quality of service, and so on.
This piece engaged in a geographical study at the scale of the merchant house. Historical demographic research has illuminated labor mobility and population transfer at a broader scale using the “Shūmon-aratame-chō.” In order to utilize those past studies from a different perspective, it is advantageous to rely on materials like the servant contract. However, when taking an approach like the present one, we have to come face-to-face with what is generalizable from the scale of one family business. This is nevertheless one method that can be useful, when tied to historical demographic research, to understand the employment of merchant house servants in relation to the labor market, labor mobility, and urban society. This is also important in understanding regional structures from a geographical perspective, and more work is needed to develop this.
Endnotes
[1] In this text, “reproduction (saiseisan)” will be used to discuss the employment of new servants once the term of employment of one has concluded, in order to maintain the merchant household’s activities.
[2] Since changes to population come from births, deaths, and mobility, mobility is one part of research from historical demography. Kinoshita Futoshi, Kindaika izen no Nihon no jinnkō to kazoku (Minerva Shobō, 2002), 132.
[3] Takahashi Miyuki, “Idō no rekishi jinkōgaku,” in Hayami Akira, Kitō Hiroshi, and Tomobe Kenichi, eds., Rekishi jinkōgaku no furontia (Toyo Keizai Shinposha, 2001), 193–213.
[4] The “city as antlion thesis” forwarded by Hayami is that due to the drawing of population from the tenant farmer strata in agrarian villages through temporary servant migration into the early modern cities that have high death rates, this also impacted the population reproduction and class mobility in the countryside. The drawing of servants from the tenant farmer strata in Nishijō village and the resulting decline in population there brought about the reduction of birth rates due to the increase in marriage age even for those who returned, due to the deaths or stay in the cities. Due to these direct and indirect reasons, in Nishijō village, the reproduction rate of the population was obstructed, and there were many within the tenant farmer strata whose families became extinct. However, there was alamo movement from the upper to lower strata through the founding of branch families by the landlord strata or movement between strata, and the village’s social structure saw major transformations. In these ways, this thesis includes population indices like birth and deaths, in addition to movement. Hayami Akira, Kinsei Nōbi chihō no jinkō, keizai, shakai (Sōbunsha, 1992), 255–284. (However, this was first articulated in Hayami Akira and Uchida Nobuko, “Kinsei nōson no kōdō tsuiseki chōsa,” Tokugawa rinseishi kenkyūjo kenkyū kiyō, 1971 nendo (1972): 217–256.
[5] Matsuura Akira, “Kinsei kōki rōdō idō no ichi-keitai,” Shakai keizaishigaku, 38–6 (1973): 50–74.
[6] However, there were a lot of instances of movement by family or through marriage when speaking of movement to Takamatsu. See Sasaki Yōichirō, “Edo jidai toshi jinkō iji nōryoku ni tsuite,” Shakai keizaishi gakkai eds., Atarashii Edo jidaishi-zō o motomete (Toyo Keizai Shinposha, 1977), 135–152; Sasaki Yōichirō, “Edo jidai no ichi toshi ni okeru jinkō idō ni tsuite,” Chiba Daigaku Hōkei Kenkyū, 9 (1990): 41–87.
[7] Hayami Akira, “Kyoto machi-kata no Shūmon-aratame-chō,” Tokugawa rinseishi kenkyūjo kenkyū kiyō, 1980 nendo, (1981): 502–541; Hayam Akira, “Kinsei Nara Higashimukikita-machi no rekishi jinkōgaku,” Nihon kenkyū, 3 (1990): 11–33.
[8] Asanuma Misao, “Kinsei ni okeru Echizen Kurome-mura no jinkō dōtai,” Rekishi chiri-gaku kiyō, 12 (1970): 231–256.
[9] Tsuchida Ryōichi, “Kinsei Kōfu Mikka-machi no jinkō keitai,” Jinbun chiri, 31(6) (1979): 71–83.
[10] Mizuguchi Tsunetoshi, “Kinsei Kai-no-kuni ni okeru hōkōnin no idō ni kansuru kenkyū,” Jinbun chiri, 33(6) (1981): 483–506.
[11] Asano Kiyotaka, “Hōkōnin no idō kara mita sonraku-kan no ketsugō kankei,” Jinbun ronkyū, 36(2) (1986): 61– 86.
[12] Recent studies in areas not including labor mobility include: Kawaguchi Hiroshi, “Edo jidai kōki no Kinki chihō ni okeru kon’in idō,” in Kurosu Satomi ed., Rekishi jinkōgaku kara mita kekkon, rikon, saikon” (Reitaku Daigaku Shuppankai, 2012), 140–170; Mizoguchi Tsunetoshi, “Kinsei Yakushima ni okeru setai kōzō to ‘tsumadoi-kon,’” Mizoguchi Tsunetoshi, Abe Yasuhisa, eds., Rekishi to kankyo (Hana Shoin, 2012), 88–111.
[13] Saitō Osamu, Edo to Ōsaka (NTT Shuppan, 2002), 71–136. This is the revised and expanded version of Shōka no sekai, uradana no sekai (Riburopōto, 1987) but we have relied on the former.
[14] Saitō also stated that in Edo, the population scale was maintained through the marriages and family formation of miscellaneous workers in Edo through conclusions he drew from demography, and that in Osaka, the increase in the age of marriage due to the length of live-in servitude in Osaka obstructed this population scale. See above, p. 13.
[15] Takahashi Miyuki, Zaigō-machi no rekishi jinkō-gaku (Minerva Shobō, 2005), 153–234.
[16] Washizaki Shuntarō, “Tenpō-ki Hachiōji Yokoyama shuku no jinkō idō,” Mita gakkai zasshi, (92)3, (1999): 594– 626.
[17] There are only 10 or so volumes of the “Shūmon-aratame-chō” discovered on Edo due to earthquakes, fire damage, and so on. In Osaka, there are some municipalities where they are intact, but it is difficult to consult them due to what is recorded in them. Hamano Kiyoshi, Kinsei Kyoto no rekishi jinkōgaku-teki kenkyū (Keiō Gijuku Daigaku Shuppankai: 2007), 14–16.
[18] Hayami (1981).
[19] Nakano Takashi, Shōka dōzoku-dan no kenkyū (Miraisha, 1964), 194–315.
[20] Akiyama Kunizō and Nakamura Ken, Kyoto “machi” no kenkyū (Hōsei Daigaku Shuppan-kyoku: 1975), 292–299.
[21] M. L. Nagata, “Migration and Networks in Early Modern Kyoto,” International Review of Social History, 47(2), (2002): 243–259.
[22] Hamano (2007), 230–231.
[23] Hamano Kiyoshi, “Bakumatsu Kyōto e no chiri-teki idō pataan,” Kansai Daigaku keizai ronshū, 60(2-3), (2010): 1–13.
[24] For a morphological approach, see: Ashikaga Kenryo, Chukinsei toshi no rekishi chiri (Chijin Shobō, 1984), 11– 74. For research on the relationship with urban structure and social structure, see Yamachika Hiroyoshi, “Kinsei kōki no Kyōto ni okeru jisha keidai no kōgyōchi-ka,” Jinbun chiri, 43(5) (1991): 25-45; Honda Ken’ichi, “Kinsei kōki no toshi sairei ni okeru kūkan Kōzō,” Jinbun chiri, 64(1) (2012): 1–18. Works that discuss the way that famous places and scenic sites were portrayed, see Tsukamoto Akihiro, “Kinsei Kyōto no meisho an’nai-ki ni egakareta ba no kūkan-teki bunpu to sono rekishi-teki hensen,” GIS riron to ōyō, 14(2) (2006): 41–52; Hasegawa Shōgo, “‘Miyako meisho zue’ ni miru 18-seiki Kyōto no meisho kūkan to sono hyōshō,” Jinbun chiri, 62(4) (2010): 60–77.
[25] Kyoto-shi, ed., Kyōto no rekishi (10 vols) (Gakugei Shorin, 1968–1976).
[26] Hamano states the following for research on historical demography, but the reason is the same as the reason for the relative paucity of research concerning people’s movement in geography. Hamano (2007), 15–16.
[27] Mitsui Bunko, ed., Mitsui jitsugyoshi, Honpen dai-ikkan (Mitsui Bunko, 1980), 242–254.
[28] Nishizaka Yasushi, Mitsui Echigoya hōkō-nin no kenkyū (Tokyo Daigaku Shuppankai, 2006).
[29] Nakano (1964): 475–507.
[30] The Endō family archives is a large archival collection that contains 756 materials. It is mainly comprised of tanaoroshi-cho, contracts of loans and debts, and servant contracts. Kyōto-shi ed., Shiryō Kyōto no rekishi, dai 9-kan, Nakagyō-ku (Heibonsha, 1995), 96–97.
[31] The main and branch families are connected. In actuality it is appropriate to think of them as a single family.
[32] The guarantor confirms the identity of the servant, and has a blood or other kinship relationship to them. The original shape of this figure is someone, like a local official or elite with the ability to economically vouch for them. Shimojū Kiyoshi, “Hōkō-nin ukejō ni tsuite,” Takizawa Takeshi ed., Ronshū: Chū-kinsei no shiryō to hōhō (Tōkyōdō Shuppan, 1991), 496–527.
[33] Of the servant contracts, there do not seem to be any missing entries for the branch family parts (27 items between 1855 and 1864), as it takes the form of a register of accounts. On the other hand, the records for the main family (65 items between 1805 and 1842) are less complete. The records from 1805–1824 have 12 items total, but there are many that are missing. As for 1827–1842, these exist continuously, and are in relatively good condition.
[34] To be precise, branch families fall into the following three types: 1. The apprentice is rewarded for long years of loyal service, and is permitted to commute from a separate location. Their descendants also go on to be in service to the main family; 2. They receive treatment like a branch family based on either the main family or the servant’s needs and are permitted to commute; 3. The branch family becomes one of the same business colleagues, and is given the main family’s customers or is permitted to operate within their business district as an independent business. Adachi Masao, Shinise no kakun to kagyō keiei (Hiroike Gakuin Jigyōbu, 1974), 331.
[35] Saitō (2002), 130.
[36] For example, in Meiji 15 (1882), the Kōnoike, Mitsui, Hirooka, and Tonomura families did not have contracts for servants that they hired from the branch family. Maki Hidemasa, Koyō no rekishi (Kōbundō, 1977), 138.
[37] Sugimori Tetsuya, Kinsei Kyōto no toshi to shakai (Tokyo Daigaku Shuppankai, 2008), 19–50.
[38] Akiyama Kunizo, Kinsei Kyōto chōgumi hattatsu-shi (Hōsei Daigaku Shuppan-kyoku, 1980), 86–87.
[39] Kyoto-shi ed., Kyoto no rekishi, dai 5-kan (Gakugei Shorin, 1972), 30–31 and 510–513.
[40] Usui Kosaburō, “Kyōto bōmoku-shi gekan no yon,” Noma Koshin kanshū, Shinshū Kyōto sōsho dai 20-kan (Rinsen Shoten, 1970), 110–111.
[41] Hanazono Daigaku rekishi hakubutsu-kan, ed., Umetada-chō Ie-yashiki ezu mokuroku (Hanazono Daigaku rekishi hakubutsu-kan, 2011).
[42] Nisshō hyakunen-shi henshū iinkai, ed., Nisshō hyakunen-shi (Nisshō hyakunen-shi henshū iinkai, 1971), 155–159.
[43] Nagata, M. L., Labor Contracts and Labor Relations in Early Modern Central Japan (Routledge, [2005] 2011), 32 and 40.
[44] Conducted interview with the current head of the Endō family on September 27, 2012.
[45] Nisshō hyakunen-shi henshū iinkai, ed., 156.
[46] In addition, there are records of a purchase of a property right to the west in 1788. In 1825, two properties were purchased under the name of Hiranoya Nisaburō (or Jinzaburō) (branch family). Dōshisha Daigaku jinbun kagaku kenkyūjo, Hosotsuji family archives, G-4.
[47] Kyōto-shi rekishi shiryō-kan, ed., Endō (Kō) family archives 638 “Gokenjō owata goyō-chō.”
[48] Nisshō hyakunen-shi henshū iinkai, ed., 155–159.
[49] Ozawa Fumio, Kyōto-shi ni okeru shōkanshū (Shihōshō chōsa-ka, 1937), 40–47.
[50] This is based on an interview with the current family head. Further, at the top of the ledger of the dye shop (branch family), whose records begin in 1818, there is an entry that says, “this store began this year.” Kyōto-shi rekishi shiryō-kan, ed., Endō (Kō) family archives 589.
[51] Noma Kōshin kanshū, Shinsen Kyōto sōsho dai 2-kan (Rinsen Shoten, 1986), 169. Noma Kōshin kanshū, Shinsen Kyōto sōsho dai 8-kan (Rinsen Shoten, 1987), 49.
[52] Noma Kōshin kanshū, Shinsen Kyōto sōsho dai 3-kan (Rinsen Shoten, 1987), 117. Noma Kōshin kanshū, Shinsen Kyōto sōsho dai 6-kan (Rinsen Shoten, 1985), 369.
[53] Kagawa Takayuki, Kinsei Mitsui keieishi no kenkyū (Yoshikawa Kobunkan, 1985), 419–436.
[54] Endō Yasaburō’s name can be found among the 9 founders of the “Kyōto gofuku shikkai dōgyō kumiai” in 1909. Takahashi Shinroku, Zōho kyōzome no hiketsu (Kyōto Shoin, [1934]1973), 116–117.
[55] The “Shūmon-aratame-chō” receipts include only the people who are within the family and there are no records of people from other families. The ones in existence are from 1857–67 for the main family, and 1840, 1843, 1853–66 for the branch family. However, for 1867, in the records of the main family, the family and servants of the branch family are listed. Those without annotation here are from the organization of these “Shūmon-aratame-chō” receipts. Kyōto-shi rekishi shiryokan, ed., Endō (Kō) family archives 466, 624, 625.
[56] Kyōto-shi rekishi shiryō-kan, ed., Hosotsuji family archives 104–2, “Tōchō shoki.”
[57] Kyōto-shi rekishi shiryō-kan, ed., Endō (Kō) family archives 637, “Mōshi-watashi no koto.”
[58] On male domestic servants, the only entry we see is in the branch family’s 1843 records, “genin Tōhachi 37 years old.”
[59] As note 58 stated, the male domestic servants in the oku (back of the house) are recorded as “genin” in the Endō
family’s “Shūmon-aratame-chō,” but here, we refer to them all as “genan” in order to remain consistent with prior research.
[60] Adachi (1974), 449–450.
[61] Generally speaking, the servant contracts between apprentices and genan have the following differences: 1. apprentices are paid while genan are not; and 2. the master will burden the clothing costs of apprentices but these are not supplied to the genan; the apprentice has a written guarantee with regards to debts, but the genan does not. These differences reflect the fact that the apprentice is hired with the objective of preparing them for business while the genan is simply a labor power transaction. See Maki (1977), 150–151.
[62] Hayami also has tried to analyze the three types, apprentice, genin, and gejo based on the “Shūmon-aratame-chō.” However, the “genin” recorded in the “Shūmon-aratame-chō” in Shimogyō’s Shijō-tachiuri-nakano-chō have not been distinguished between the apprentice in the front of the house and the genan in the back. In other words, the likelihood that many of the people who should be placed in the genealogy of apprentice are included in the category of genan is high. See Hayami (1981).
[63] The private archives of the Endō family have the surname “Endō,” but the servant contracts without exception contain the yagō of “Hiranoya.” From this, we can determine that those who have the surname in the servant contracts of the Endō family are people who have been permitted to use surnames.
[64] Hamano (2010).
[65] Hayamo (1990).
[66] Kawaguchi (2012).
[67] Hayami (1981) and Hamano (2010).
[68] Kyōto-shi, ed. (1972).
[69] Kyōto-shi ed., Kyoto no rekishi, dai 6-kan (Gakugei Shorin, 1973), 241–249.
[70] Kyōto-shi rekishi shiryō-kan, Nishimura (Dai) family archives H26, “Gōshū nuno-don’ya dōmoto shiire nakama namae chōmen utsushi.”
[71] Kagawa (1985).
[72] They were hired in the ninth month of 1829, the third and the ninth months of 1831, and the seventh and the ninth months of 1832.
[73] In the case of Echigoya, some of the genan were hired following 10 years or more of long employment, and after this, a pathway to promotion opened up. See Nishizaka (2006), 51–90.
[74] Historical demography has also pointed out that there were a lot of women servants who were from Ōmi in Kyoto city. Hayami (1981) and Hamano (2010).
[75] According to Hamano, within the 16 chō of Kyoto city, the age of the highest number of genan was 17 years. Further, in Shimogyō and Seidō-chō, the women servants’ average starting age was 21.0 years. See Hamano (2007), 156–157 and 186–187.
[76] Hayami (1992).
[77] The average employment years of the Endō family is calculated based on table 4. In Seidō-chō, refer to Hamano (2007), 187.
[78] This is based on an interview with the current family head.
[79] Kyōto-shi rekishi shiryō-kan, ed., Endō (Kō) family archives 637, “Mōshi-watashi no koto.”
[80] According to the current family head, since the Meiji period, they expanded their market to the Kanto region and the majority of their sales came from there. However, due to the Great Kanto Earthquake of 1923 and the 1929 great depression, they suffered losses. As a result, they dissolved their formal relationship with the branch family and contracted their business.
[81] Nakano (1964).
[82] The Endo family owned rental housing in Tearaimizu-chō. Kyōto-shi rekishi shiryō-kan, ed., Tearaimizu-chō archives 3.
[83] It has been pointed out that even in the case of the pharmaceuticals business, the Yamatoya Chūhachi family’s branch families in Gojo Kawahigashi, a widespread distribution of residences took place from the late Meiji period. See Hamano (2007), 731.
[84] Kyōto-shi rekishi shiryō-kan, ed., Umetada-chō archives 11, “Meiji 2-nen tsuchinoto-mi Jūgatsu Shimogyō yon-bangumi.”
[85] Based on an interview with the current family head.
[86] Fukuoka-ken monjo-kan, Matsuki Shōzaemon family archives 2, 4. Note that “Shōzaemon” in the records and the kanji used here are slightly different.
[87] The third generation Matsuki Shōzaemon was the head of the “Wakasa-no-kuni Obama hanryō Jyōō gannen ikki,” which took place in 1652 demanding a decline in the soybean tribute. In the end, this demands were admitted in exchange for his crucifixion. Thereafter, the third generation Matsuki Shōzaemon has been known as the “Martyr of Wakasa.” Kokushi daijiten henshū iinkai, ed., Kokushi daijiten, dai 14-kan (Yoshikawa Kōbunkan, 1993), 843.
[88] The residence of Tangoya Yahei as indicated in the servant contract is recorded as Karasuma-dōri Sanjō-sagaru-chō (Manjūya-chō) but the family tree A lists it as Karasuma-dōri Sanjō-araru-chō (Banochō) and family tree B lists it as Higashi-tōin-dōri Sanjō-sagaru-chō (Sanmonji-chō). Despite these discrepancies, these three chō were all adjacent to Umetada-chō, where the Endō family was. If we consider the mobility of the residents of Kyoto, it could be the case that Tangoya Yahei lived in Manjūya-chō during the time of Jūzō’s employment, and then moved to Banochō and Sanmonji-chō.
[89] This is based on my interview with the current family head. In terms of the determination criteria for the 20 or so branch family gravestones, there are three: (1) that it is near the grave of the main family; (2) that they have Shōju-ken cloister at the Konkai-kōmyōji tacchū as their home temple (danna-dera), just like the main family; (3) and that there is a Hiranoya yagō.
[90] Saitō (2002), 164.
[91] Saito (2002), 119.
[92] Kyōto-shi, ed., Kyoto no rekishi, dai 7-kan (Gakugei Shorin, 1974), 170.