The Household and the Legal System in Early Modern Japan

In Tokugawa Japan, the household (ie) served as the only viable container for business assets. Historian Ōtō Osamu offers one definition for the household: “An ie is an organizational entity that seeks to exist eternally, that bears a specific house name, house assets, and house line of business, and that is inherited generation after generation with the spiritual support of ancestor worship and its related rituals.”Footnote ¹⁴ By the middle of the Tokugawa period, the household became a powerful source of meaning in the lives of early modern people.

Significantly, household names did not need to correspond with any given individual, possessing a kind of legal personality.Footnote ¹⁵ For this reason, household names—and the status associated with them—also came to be treated as assets over the course of the Tokugawa period.Footnote ¹⁶ The fluid relationship between name and identity in early modern Japanese society enabled this set of practices. People took on different names in different settings and at different stages of life.Footnote ¹⁷ Yet the proper registration of household names served as one of the basic tools of Tokugawa governance.Footnote ¹⁸ Whether under the direct rule of the shogunate or in a domain, the warrior government exercised control over the commoner population by recording households into population and land registers.Footnote ¹⁹

But even these names did not necessarily match those of real people. Moreover, they masked a substantial degree of variation in actual household structure.Footnote ²⁰ Sometimes, multiple villagers paid land tax on property registered in the name of a long-dead individual.Footnote ²¹ Double registration of a single individual under different household names in different areas simultaneously—as a townsman ( chōnin ) in the city and a villager ( hyakushō ) in the country—also occurred, even though it violated the governing principles of the Tokugawa state.Footnote ²² Household names thus served as a flexible container for people and assets amenable to manipulation.Footnote ²³

Unlike in premodern Europe, early modern Japanese merchants could not register partnerships or gain official recognition for any business entity. For temporary business partnerships, like trading voyages, merchants made use of agreements similar to the commenda, with multiple passive investors and a single active partner engaging in trade or shipping, all contracting as individual households.Footnote ²⁴ Merchants made these agreements on a private basis without official registration or reference to a formal body of law.

For longer-term partnerships, however, Tokugawa traders needed a more durable container for assets. In response, they chose the only institution at hand: the household. A pioneering 1930 article by the historian Kanno Watarō introduced the case of “Ōmi-ya Sōbe,” an entirely fictive entity formed through a partnership of three merchants to run a fishery business in 1838.Footnote ²⁵ The discovery of other examples followed. Between 1741 and 1757, twenty-one different investors, holding a total of forty shares, formed the house “Nishikawa Denji” in order to trade dried sea slug from Matsumae, in modern-day Hokkaido, to Nagasaki.Footnote ²⁶ In Osaka, two merchants named Inamoto Riemon and Nishimura Jūrōbe partnered to form the “Inanishi Shōtarō” clothing store in 1813.Footnote ²⁷ Merchants continued to make use of similar entities into the modern period, with three major Osaka financiers banding together to found Yamahiroya-Chōbe in 1870, a “store name recognized as a member of the neighborhood ( chō ) without a flesh-and-blood head in existence.”Footnote ²⁸

On paper, these names looked like real merchant households, even though in reality no household head with such a name existed. Professional managers and employees staffed each store, with profits and losses split in accordance with the distribution of shares among the investing partners, who often did not even occupy the property on a regular basis.Footnote ²⁹

Not all Tokugawa merchants expanded through this model of fictive household partnerships. But all major trading houses took advantage of the fictive quality of household names to hold assets and employees within named entities that did not correspond to real people. Stores and offices, after all, did not exist as independent legal entities within Tokugawa society—they needed to be constituted as households.

Tokugawa merchants made use of the household in the absence of meaningful legal alternatives. While organizations like guilds, lending funds, and domain commodity offices did receive a form of official recognition from the authorities, ultimately, the households that made up the membership of these entities took responsibility for transactions, loans, and other forms of business.Footnote ³⁰

As a general rule, the Tokugawa authorities—whether the shogunate or the domains—demonstrated relative indifference to the development of business and commercial law. The samurai rulers did enforce private contracts, distinguishing between cases involving lending and credit transactions at interest with no collateral ( kane kuji , or kingin deiri ) and other types of loans (treated as hon kuji ). In Osaka, the commercial capital of Japan, the office of the town governor developed a legal framework that worked to favor creditors and sped up the process of contract enforcement for non-collateralized loans.Footnote ³¹ These provisions applied broadly over cross-border disputes in western Japan, with officials in Edo adopting similar rules only in 1843.Footnote ³²

But the authorities flatly refused to handle cases related to the settlement of accounts in partnerships, explicitly designating them as one of the primary categories of non-adjudicable issues (referred to as nakama-goto , or “in-group matters”).Footnote ³³ In other words, the law offered no tools for merchants looking to form businesses, nor did it give a helping hand to creditors looking to collect after the dissolution of a partnership.Footnote ³⁴

Even for the commercial cases that the samurai did claim authority over, they often lacked the interest or capacity to cope with the demand for legal intervention. The shogunal authorities in the major cities of Edo, Osaka, and Kyoto regularly issued orders for the private settlement of money disputes when overwhelmed by the number of cases.Footnote ³⁵ In 1718, the number of non-collateralized loan disputes handled by the town governor in Edo reached some 33,037 cases, a paltry number in comparison to some other contemporary societies.Footnote ³⁶ Given that the population of the city at the time numbered over one million, this caseload likely stood far below the actual number of debt-related disputes ongoing in the capital. Informal mechanisms for settlement remained the norm.Footnote ³⁷

Unlike many parts of early modern Europe, Tokugawa Japan never saw the establishment of merchant-run courts for commercial disputes nor anything equivalent to a debtors’ prison.Footnote ³⁸ Despite some creditor-friendly innovations in legal arrangements and a willingness to hear cross-border cases, by and large the authorities favored debtors, settling disputes through long-term repayment schemes at no interest after imposing costly legal fees.Footnote ³⁹ Outside of Osaka, officials in most of the country showed little real interest in helping merchants enforce private contracts.Footnote ⁴⁰ For this reason, many merchants opted to avoid legal entanglements entirely, effectively renouncing the right to reclaim debts in some cases.Footnote ⁴¹

In early modern Japan, the formal legal infrastructure to support the formation and perpetuation of large-scale businesses did not exist. A rich literature suggests that such institutional limitations would impose higher transaction costs on businesses and hamper investment—an insight with which many Tokugawa merchants would no doubt agree.Footnote ⁴² At the same time, the Tokugawa period saw the emergence of major merchant houses that accumulated vast sums of money and spread across the country. Why could this occur? To answer this puzzle, the next two sections look at how the Nakai Genzaemon and other merchants used the household to structure complex enterprises.

The Nakai and the Fictive Household Partnership Model

On the surface, the Nakai Genzaemon looked like a family firm (Table 1). Each store bore the last name of Nakai or used the Nakai’s hometown of Hino as a house name ( yagō ).Footnote ⁴³ For instance, the Nakai’s largest store in the northeastern castle town of Sendai took the name Nakai Shinsaburō, while the house’s outpost in Osaka went by Nakai Iwanosuke. Members of the Nakai lineage, naturally, did make up part of the broader Nakai business network. Cadet branches (known as bunke ) founded by the younger sons of the first Nakai head ran largely independent businesses in Kyoto and Ōtawara (in modern-day Tochigi Prefecture), even though they remained deeply entwined with the affairs of the main house.Footnote ⁴⁴ Successful managers could also, as in other major Tokugawa merchant houses, set up dependent franchise households ( bekke ) with the help of start-up funding from the Nakai. They became, in effect, collateral members of the house’s lineage and subject to the patriarchal authority of the head of the house. Some franchisees chose to continue working for the Nakai as senior managers. Together, the main house in Hino, the cadet branch households, and the franchises made up what scholars call a lineage group ( dōzokudan , a neologism not used during the Edo period).Footnote ⁴⁵ Most research on Tokugawa merchant houses tends to center the lineage group as fundamental to business organization.

Table 1.

The Nakai’s stores organized by type

Table 1. Long description

The table is organized into five main groups by store type, each with rows for individual stores. Columns are, from left to right: Type, Store name(s), Location, Date founded, Date closed. The first group, Cadet House Branch (bunke), lists two sets of stores: Omi-ya Shoji/Nakai-ya Shoji/Yoshino-ya Shoji/Nakai Shojiemon in Kyoto (founded 1788, closed 1872) and Omi-ya Gensaburo/Omi-ya Gennosuke/Nakai Gensaburo in Otawara (1749–1914). The second group, Direct Main House Control, includes 14 stores such as Hino-ya Shinsaburo/Nakai Shinsaburo in Sendai (1769–1890s), Omi-ya Hikotaro in Fushimi (1769–1788), Ito-ya Hikozaemon in Ushirono (1769–c.1800), and others in Sōma, Oshitate, Onomichi, Edo, Ishinomaki, Tendō, Osaka, Kitsuki, Karasu, Kyoto, and Osaka, with founding dates from 1769 to 1863 and closing dates from 1799 to 1938. The third group, Franchisee (bekke) under Direct Control, lists Hino-ya Jihe in Nagoya (1813–1846). The fourth group, Branch Store (edadana), includes 12 stores such as Izumi-ya, Juichi-ya Mozaemon, and Hino-ya in Otawara (founded 1749, 1757, 1779), Shichi-dana/Nakai Genshiro/Hino-ya Genshiro in Sendai (1796–1871), and others in Sendai, Minato, Sōma, Nagoya, Onomichi, with many closing dates unknown. The fifth group, Franchise Households/Senior Managers (bekke), lists 11 stores including Hino-ya Sahe in Sendai (unknown–after 1855), Hino-ya Gohe (unknown), Hino-ya Zenbe (likely Hino), Hino-ya Hachibe in Sendai, Hino-ya Heibe and Hino-ya Chōbe in Hino, Hino-ya Mokube (likely Hino), Hino-ya Kakube, Uhe, Shirōbe in Sendai, and Hino-ya Hikobe in Ishinomaki, with most founding dates unknown and closing dates after 1855 or 1860. The final group, Branch Store & Franchise, lists six stores: Hino-ya Gihe, Rinbe, Tōbe, Tahe, Ginbe in Sendai, and Kishimoto-ya Manbe in Ishinomaki, with founding dates from 1797 to 1851 and closing dates mostly unknown or after 1805/1878. Each row provides the store name(s), location, founding year, and closing year or status.

Source: Based on Hino chōshi hensan iinkai, ed. Ōmi Hino no rekishi , vol. 7. and Egashira, Ōmi shōnin Nakai-ke no kenkyū , 382, 984-986. Bekke location information drawn from Nakai Genzaemon-ke monjo, #8584, #8580.

Yet relatives of the Nakai—whether flesh-and-blood ones like the cadet branches or those connected by lineage like the franchisees—neither owned nor operated the core parts of the house’s business that employed around 100 people and held over 250,000 gold ryō in assets by the end of the Tokugawa period.Footnote ⁴⁶ Stretching across Sendai, Osaka, Nagoya, Sōma, Tendō, Ishinomaki, Karasu, and later Kyoto, the Nakai’s major stores extended their reach from western to eastern Japan, enabling the cross-country trade in spun cotton and used clothing that provided the motor for the house’s capital accumulation (Figure 1).

Figure 1.

A map of Nakai’s main stores during the early nineteenth century.

The lineage group model cannot explain how and why the Nakai could establish and maintain this network of stores. But the Nakai’s stores—like Nakai Shinsaburō in Sendai to Nakai Iwanosuke in Osaka—looked on paper just like family members. If Nakai Shinsaburō in Sendai was not, then, a relative of the main house Nakai Genzaemon, who (or what) exactly was he?

It turns out that Nakai Shinsaburō, much like the other major stores operated by the Nakai, was neither a member of the Nakai house nor the name of a real person at all. The first Nakai Genzaemon (also named Ryōsuke or Mitsutake), the founder of the Nakai and head of the main house in Hino, established Nakai Shinsaburō as a partnership with four other merchants in 1769 via a private agreement between the parties. He set out 3,375 ryō in gold, making him the primary investor. Two merchants invested 500 ryō apiece, with the other two each putting 312.5 ryō toward the venture.Footnote ⁴⁷ The four together served as passive partners in the business.

Called “prospective funds” ( mōshōkin ), this initial investment by the founding partners entitled them to receive regular interest payments as well as dividends from the store. Each partner received shares ( kuchi ) in proportion to their investment, and split profits and losses accordingly. In Sendai, profit distributions to the outside partners stopped in 1782, when the Nakai transformed the four merchants’ initial investment into interest-bearing deposits.Footnote ⁴⁸ Significantly, what tied Shinsaburō to the Nakai’s headquarters in Hino was not a familial relationship but instead an obligation to regularly pay interest and distribute profits in proportion to the main house’s investment. Just like in the examples mentioned in the previous section, the Nakai main house worked with non-kin to establish a fictive household partnership.

Nor did kin ties play much of a role in staffing the business. The Sendai store employed some sixty people by the 1850s and operated under the leadership of professional managers.Footnote ⁴⁹ For this reason, some staff referred to it and the other core parts of the Nakai’s enterprise as “manager-held” ( shihainin-mochi ) stores.Footnote ⁵⁰ Even the main house of Nakai Genzaemon in Hino, though formally under the Nakai head, maintained a staff of top non-kin managers who coordinated operations among the different stores and operated under the Genzaemon name.Footnote ⁵¹ Nakai family members did not control the management of the enterprise.

Both at the stores and in headquarters, the principle of collective decision-making among staff stood paramount. A Nakai store code from the early nineteenth century decreed: “Regarding all store affairs, no matter how insignificant, the main members [of the store] should gather and without reservation share their opinions, deliberate, and decide on the issue at hand.”Footnote ⁵² All major decisions, like the appointment of managers, involved the top staff working in consultation with the head, who could not make unilateral decisions.Footnote ⁵³

Like in all other large-scale Tokugawa merchant houses, a clear separation existed between household and business accounts as well, leaving no overlap between the expenditures of Nakai family members and the accounting at each store.Footnote ⁵⁴ Unlike other Tokugawa merchant houses, however, where sometimes over 90% of employees shared a common origin, the Nakai’s staff came in significant numbers from the areas they did business, with some 43% of all hires between 1836 and 1862 originating from outside the Nakai’s home province of Ōmi.Footnote ⁵⁵

“Prospective funds,” not kinship or same-place ties, linked stores to the Nakai main house. Stores made regular interest payments and profit distributions to the headquarters in Hino.Footnote ⁵⁶ Like some other Tokugawa merchant houses, the Nakai calculated profits after interest on their investment, tying staff bonuses to profits. The main house and other investors often chose to reinvest their profits back into the stores as “‘new’ prospective funds,” further ratcheting up performance targets.Footnote ⁵⁷ In the late eighteenth century, staff at the Sendai store complained about how the Nakai treated funds from the main house and other stores exactly the same as investment from outside merchants. Clerks protested that “no other store in the Sendai area is charged interest” in this way.Footnote ⁵⁸ The charging of interest on the movement of money “internally” within the Nakai network surprised contemporaries.

This impersonal treatment extended to all business relationships between the stores, which dealt with each other as independent entities. When the Sendai store bought used clothing from the Osaka store, for instance, it would appear on the books as a debt for Sendai and a credit for Osaka. No exceptions were made even for minor expenditures. An 1823 set of regulations addressed to the Osaka store noted that “regarding various costs associated with guests, costs for hosting clients of the main house should be treated as loans to the main house, while those for clients of the Sendai store should be accounted as loans to the Sendai store.”Footnote ⁵⁹

To ensure that stores fulfilled their obligations, managers from the main house, as well as the Nakai head, conducted annual checks of the account books submitted to headquarters by each store.Footnote ⁶⁰ The head of the house performed semi-frequent inspections of each store in person as well, traveling hundreds of miles at a time.Footnote ⁶¹ Employees were also obligated to travel to headquarters in Hino after a fixed number of years for performance reviews.Footnote ⁶² Managers sent reams of letters and monthly reports to the main office, with a given year of letters from a single store often running to the hundreds of pages.

As the scale of the their operations expanded, the Nakai increasingly eschewed the need for outside partners—or real people—entirely.Footnote ⁶³ Instead, they set up new branches as partnerships between the main house and already existing fictive household-stores. The extensive use of this practice sets the Nakai apart from the partnerships discussed in the previous section.

Fictive households functioned as contractual partners, investing, buying, selling, borrowing, and lending as independent legal entities. For example, in 1783, the main house in Hino and Nakai Shinsaburō in Sendai joined forces to establish Ōmi-ya Genzaemon in Sōma, modern-day Fukushima. Out of an initial “prospective fund” investment of 1,000 ryō lent at a monthly 0.6% simple interest rate, the main house, Nakai Genzaemon, held nine shares, while Shinsaburō owned ten, with one additional share reserved for the manager of the store “… to be used as a bonus for hard work; however, losses will be divided by 19 shares.”Footnote ⁶⁴ Even though the Nakai divided the “prospective funds” into shares, actual interest payments went directly to the main house alone, while profits after interest remained in store coffers. Partnerships between fictive households could generate new entities without recourse to outside capital or lineage ties.

Other stores came under the Nakai’s direct control through investment in other merchant houses or in their own franchisees. Hino-ya Seiichirō, the Nakai’s store in Tendō (in modern-day Yamagata Prefecture), at first belonged to another merchant from Hino, Uemura Chōemon, and went under a different name. Many Nakai lineage members lent money to the store before its full takeover, with the Nakai’s Kyoto cadet branch, Jukei, the wife of the first Nakai head, his daughter Fumi, and Riyo, the daughter of the founder of the Ōtawara cadet branch, all holding interest-bearing deposits at the store.Footnote ⁶⁵ But during the early 1800s, when the Uemura experienced severe financial difficulties and their head came down with a serious illness, the Nakai offered to buy them out, transforming their Tendō branch into a “member store” ( nakama-dana ) of the Nakai in 1804.Footnote ⁶⁶ From 1816 onward, the store went by the name Hino-ya Seiichirō. At first, Chōemon’s son Yoichi—under the name Denbe—served as manager over a staff of around eight people.Footnote ⁶⁷ By 1836 at the latest, the Nakai appointed their own manager to the store.Footnote ⁶⁸ Between an initial round of 2,000 ryō in “prospective funds” in 1806 from the main house and a “new” round of 3,000 ryō invested by the Sendai store in 1849, the store became much like the other purely fictive entities within the Nakai’s enterprise, bound through ties of debt and regular reporting requirements.Footnote ⁶⁹ In this way, the Nakai could hollow out “real” household-based family firms and reshape them into fictive, dependent entities underneath their control.

The Nakai’s Nagoya store, in contrast, began as a franchise named Hino-ya Jihe under the ownership of Kawara Jihe, a long-standing Nakai employee.Footnote ⁷⁰ Founded in 1813, Jihe began sending his books for inspection to the main house at least by 1816, and also submitted regular reports to the main house.Footnote ⁷¹ As the region’s spun cotton became vital for the Nakai, the main house poured “prospective funds” into Jihe’s operation—some 2,000 ryō in 1835—transforming it into a de facto procurement center for the business.Footnote ⁷² Nagoya, as the main supplier for spun cotton, also coordinated with the Sendai store, the Nakai’s main sales hub, on purchases. In effect, Jihe continued his work as a professional non-kin manager for the Nakai, under a store in his own name but tied to the Nakai main house through the same debt relations that connected the enterprise’s other stores.

Stores also set up sub-branches ( edadana ). Sometimes, a franchisee helmed these sub-branches under a different name, like when former manager of the Sōma store Hino-ya Gihe founded the Sendai store sub-branch “Ōmi-ya Sadashirō.”Footnote ⁷³ Another similar case involved the franchisee Hino-ya Rinbe opening “Hino-ya Ichitarō” in Sendai, with Rinbe offering the nonexistent Shinsaburō “deep thanks for financing this store in the same way as a store of your own.”Footnote ⁷⁴

Sub-branch formation did not depend on franchisees, however. The Sendai store acted as the founder of the Nakai’s store in Ishinomaki, called Hino-ya Genzaemon, which then founded a sub-branch in nearby Minato, also called Hino-ya Genzaemon.Footnote ⁷⁵ Such sub-branches were subject to strict accounting checks from the management of their “parent” store—in this case, Nakai Shinsaburō—and their line of business could not overlap or compete with that of other Nakai stores in the same area. As an investor in the Tendō and Sōma stores, Shinsaburō also monitored their books and activities. Given its control over the Tendō, Ishinomaki, and Sōma stores, the Sendai store acted as a local headquarters for the entire northeastern region, with significant leeway over business decisions on the ground.

As the above examples suggest, even real lineage group members could choose to establish a fictive household entity underneath the direct supervision and control of a larger Nakai store instead of taking on business in their own names. Under the conventional understanding of the lineage group, Gihe and Rinbe ought to have simply run franchise stores as themselves. But the fictive households Sadashirō and Ichitarō functioned as useful vessels for store assets, goods, and staff formally separate from Gihe and Rinbe, allowing, in theory, other Nakai staff members to occupy them as needed.

The household form enabled a significant degree of abstraction. Real people with actual lineage or blood ties were not necessary to create new stores. Even when a new store was founded by a family or lineage group member, it could soon take on a life of its own. When Nakai Iwanosuke, the grandson of the founder of the Nakai Genzaemon, “established” the Osaka store under his own name at the age of five in 1806, a store clerk ( tedai ) served as his legal representative, placing Iwanosuke’s seal on the moving documents ( teraokuri shōmon ).Footnote ⁷⁶ Twenty-one years later, clerks continued to cite Iwanosuke’s youth when using his seal on a 1827 population register ( shūmon aratame-chō ). Iwanosuke did not need to be present for “Nakai Iwanosuke” to remain on the books as a registered household. Whether or not the real Iwanosuke actually moved to Osaka in 1806 is unclear, but his name certainly did. Later registers in 1829 and 1835 changed the excuse for the absence of Iwanosuke’s seal from youth—increasingly implausible—to sickness. In fact, Iwanosuke had died in 1807, just one year after founding the store. His death was no obstacle to business. The nonexistent Nakai Iwanosuke continued to pay interest on the “prospective funds” deposited by the main house year after year—over 20,000 ryō between 1806 and 1845.Footnote ⁷⁷ Much like a corporate name, the household name could continue to exist without the need for a real household head to bear it.

For this reason, multiple people could make use of the same name. In 1834, when the fourth Nakai Genzaemon took on the headship of the merchant enterprise, he also inherited the name of the Sendai store, Nakai Shinsaburō. At that time, the Nakai’s manager in Sendai and another affiliate of the store, who falsely claimed to be Shinsaburō’s “relative,” submitted a document verifying the inheritance of Nakai Shinsaburō to domain officials in Sendai on his behalf, even signing the name of Shinsaburō on the letter. Succession, an act of some importance that would seemingly require the presence of the heir, could be done by proxy. The manager claimed that “because (the new) Shinsaburō is not present, matters have been entrusted to me, the store manager, in his stead,” but this was far from an exceptional case.Footnote ⁷⁸ Managers regularly used the name of the store as their own, both in official correspondence and even in letters sent within the Nakai network. A real person with the exclusive right to use the name Nakai Shinsaburō never existed.

Despite the importance of accurate household registration to early modern governance, merchants manipulated household names with surprising ease. In his 1839 diary, the fourth generation head of the Nakai noted that when the merchant house rented a property in the Kiya-machi neighborhood of Kyoto from someone named Wakamatsu-ya Seiji, they “set the provisional name of the head of house as Hino-ya Shinsaburō”—the same name as the Sendai store.Footnote ⁷⁹ Other examples abound in the Nakai records. As part of a messy legal case involving the settlement of debts at the Sōma store at its closure, the Nakai decided to create a new household named Ōmi-ya Genzaemon in the Motohama neighborhood of Edo. In his records, the Nakai head wrote that “we will claim officially, for the sake of the authorities, that it is the house of a direct clothing trader…a store from Hino in Gōshū operated by Ōmi-ya Genzaemon, managed by Mokube, and with Chūbe and Kōkichi on staff,” but “in fact, no one will live there.”Footnote ⁸⁰ With the help of some bribes to a local official, they established the new Genzaemon without issue. As with the case of the sub-branches, establishing a new household with a name distinct from that of its founder proved relatively straightforward.

There was nothing inherently problematic about the name of a house not aligning with that of an individual—this was, in fact, one of the key features of the early modern household. On occasion, however, the Nakai needed to realign legal fiction with reality. In 1850, the lord of the Sendai domain toured the north of his realm and summoned all of the elderly residents in the area to reward them for their longevity.Footnote ⁸¹ Domain officials included the name of the Nakai’s store in Ishinomaki, also Nakai Genzaemon, on the list of elderly residents. Per the local population register, Nakai Genzaemon, a villager in Motomachi, Ishinomaki, had turned eighty-eight years old, making “him” a candidate for the summons.Footnote ⁸² But the head of the Nakai enterprise at the time was only forty-eight. Unfortunately, the Nakai did not employ any eighty-eight-year-old managers who might have been able to stand in as Genzaemon of Ishinomaki, either.Footnote ⁸³

The Nakai needed to erase the existence of the elderly Genzaemon from the books, or else they would end up in the uncomfortable situation of an unexpectedly youthful man getting summoned by the authorities. Staff issued a request to the samurai district governor ( kōribugyō ) to allow “Genzaemon’s” fifteen-year-old adopted son, “Genkichi,” to inherit the house and remove “Genzaemon” from the population register.Footnote ⁸⁴ By paying three ryō —a mixture of administrative processing fees, gifts, and bribes—to the samurai officials and some local guarantors, the “Genzaemon on the land and population registers” ( jindaka-chō Genzaemon ) was eliminated without issue. On paper, fictive households aged, died, and passed on their assets.Footnote ⁸⁵ With the help of local guarantors and the liberal payment of fees, merchant houses could maintain fictive households across multiple generations.

Why bother with this conceit at all? What explains this fixation on setting up fictive household entities? For one, this practice allowed the Nakai to set up stores in the absence of preexisting kinship ties. They did not need a relative or senior manager to found a new store on their behalf—they could simply invent a household from scratch. Second, according to the business historians Uemura Masahiro and Yasuoka Shigeaki, Tokugawa merchants used fictive store names to facilitate asset shielding.Footnote ⁸⁶ Merchants could ward off claims on their total capital by dividing up their assets between different named entities. Considering the cost and difficulty of collecting debts across the various borders that split up the Tokugawa polity, setting up different entities across jurisdictions made the protection of assets much easier for merchants.Footnote ⁸⁷ On the surface, independent households had no obligations to each other. Nakai Shinsaburō in Sendai, as a separate household from Nakai Genzaemon in Hino, did not bear responsibility for the latter’s debts—or vice versa. Nor could the lord of the Sendai domain, for instance, force donations or extract service from Genzaemon.

By enabling both the division and movement of assets across different entities, sometimes in different jurisdictions, the fictive household partnership model in effect allowed merchants to “offshore” their assets. The use of different names could even protect merchants from the consequences of the most serious crimes. In one example introduced by Yasuoka, all of the stores of a major Kyoto merchant house managed to escape punishment when the house’s head committed a series of murders in town, simply because they operated under different names.Footnote ⁸⁸ He concludes that “in practice, a system of limited liability was realized through the features of the societal structure of Edo-period Japan.”Footnote ⁸⁹

Despite the existence of partnership agreements that laid out the division of profits and losses among partners, the existing literature on Tokugawa merchants does not offer any concrete examples demonstrating that investing partners only bore liability for their share of the total capital in a partnership—meaning that true limited liability did not exist.Footnote ⁹⁰ Even across the vast array of store diaries and other documents that make up the Nakai archive, legal cases crop up infrequently, and the Nakai invariably appear as creditors rather than debtors. As made clear in section one, the samurai authorities did not consider disputes over the settlement of accounts in a partnership worthy of official consideration, so few legal sources survive. How liability worked in the case of bankruptcy for fictive household entities remains a mystery.

Thanks to the Nakai’s takeover of a store in Karasu (modern-day Mie Prefecture) following a failed partnership by three merchants in 1839, however, their records do offer some clues about how partnership dissolution ended up in practice.Footnote ⁹¹ The three partners, two being relatives of the Nakai and funded by the merchant house, with the other being unrelated, had established a fictive household named Ōta-ya Uhe to operate a sake brewery. Despite an uneven split in shares among the three partners, it seems that in practice their debt—held under their three individual names, not the fictive name of the store—was divided equally. As part of the dissolution process, one partner sold off his house and belongings to repay debts incurred on behalf of the store, entrusting his assets to creditors in a Tokugawa equivalent to filing for bankruptcy ( bunsan ).Footnote ⁹² Following extensive negotiations, the Nakai main house took on much of this partner’s debt in 1841, with the final settlement process requiring years.Footnote ⁹³ Since the samurai authorities left the settlement of liability among partners up to the parties involved, no hard and fast rules existed for determining the final obligations borne by individual partners. In the end, the total debt burden did not appear to exceed the stated value of the partners’ shares in the original partnership agreement.

Due to norms surrounding repayment and the limited interest of authorities in enforcement, partners could often escape the full weight of their obligations, even if, in theory, they might bear unlimited liability. Given that the partners did not take on their debts as the fictive household Ōta-ya Uhe, probably because they could only raise funds based on their own creditworthiness as individual merchants, the Karasu case does not offer direct insight into how the division of debt might have worked for the Nakai’s stores. The use of a fictive name would likely further complicate debt collection for creditors and thus limit risk.

By making the stores entirely fictive entities, merchant houses also eliminated issues related to succession or the development of spin-off enterprises. The sudden death of a cadet branch head might risk the continuity of a key line of business. A stubborn relative could resist the main house’s authority. Fictive households solved such problems.

Though the circumstances of their formation or eventual envelopment into the core Nakai enterprise differ, the stores discussed in this section all shared a common bond to the Nakai main house, one mediated not through blood but debt relations and the regular monitoring of accounts. Just as the Sendai or Sōma stores paid interest to the main house, so too did the Tendō, Nagoya, Ishinomaki, and other stores. Managers and the head of the main house checked books, took part in an active schedule of on-site visits, and exchanged a flurry of letters with each of these fictive households. All of this worked to bind together a large commercial enterprise.

This use of fictive household stores allowed the Nakai to distribute risk and expand their enterprise in the absence of direct lineal or kin relationships. Other Tokugawa merchant houses also made use of fictive household names, but as the next section shows, the most famous examples of early modern Japanese business structured themselves through the lineage group rather than the fictive household partnership model.

The Lineage Group Model

While the Nakai expanded through fictive entities that stacked one on top of the other, the most prominent merchant houses of the Tokugawa period relied instead on the lineage group. Younger, noninheriting sons could set up cadet branches that ran stores linked to a headquarters run by the main house. Successful former managers could found semi-independent franchise households. Together, the lineage group formed a larger business linked by common lineal membership. Conventionally, scholars have taken the lineage model as the norm, with the partnerships formed by the Nakai and other merchants often viewed as important outliers that anticipated the introduction of the stock corporation.Footnote ⁹⁴

The way merchant houses made use of lineage ties, however, differed substantially. For the Kōnoike, one of the most powerful financial houses of the period, cadet branches acted as passive investors in the main house, combining their capital to issue massive loans to lords.Footnote ⁹⁵ Though founded earlier than the Nakai, the Kōnoike developed this structure during the eighteenth century as a decline in interest rates pushed the house to ban less competitive branches from unprofitable lending and consolidate their capital. At the same time, by splitting up the same loans among lineage members, they also distributed the risk of default.Footnote ⁹⁶

The Mitsui, the wealthiest and most prominent business of the period, also transformed the lineage group into passive investors. But they eschewed the idea of central control under the main house. Instead, the nine houses that made up the Mitsui lineage group pooled their total assets under the collective control of an unlimited liability organization, known as the ōmoto-kata , run by a combination of top managers and core kin household heads.Footnote ⁹⁷ This executive board formed in 1710, overseeing nine main stores, with an additional seven underneath them. Stores received a start-up investment ( moto-date ) in exchange for regular fixed payments ( kōnō ) to the board.Footnote ⁹⁸ While the terms differ, this system strongly resembles the Nakai’s use of “prospective funds.” Each store also bore a distinct name, though many of the major stores went by Echigo-ya Hachirōemon or Hachirōbe.Footnote ⁹⁹ The board determined the appointment of store managers and top officers. At their peak, the Mitsui employed over a thousand people, with their store in Edo alone housing some 342 employees.Footnote ¹⁰⁰

Most significantly, no permanent link tied together specific lineage group members with specific stores.Footnote ¹⁰¹ Any member could inherit any store, meaning that stores, though constituted as independent, named households, worked like assets. Just like the Nakai, the Mitsui made use of the fictive nature of household names to set up stores run by non-kin professional staff. Yet unlike the Nakai—and like the Kōnoike—they sought to consolidate their assets into a single entity rather than distribute risk across multiple fictive household partnerships.

This system transformed over the course of the eighteenth century. The executive board dissolved itself over succession disputes in 1774, splitting the enterprise into three branches run by three different groups of houses within the lineage group, which had grown to number eleven houses in total.Footnote ¹⁰² Designed to help preserve the Mitsui’s assets against the risk of break-up between members of the lineage group, the executive board system still suffered from the fact that it depended on amity among the cadet branches to survive. In contrast, the depersonalization of the Nakai model—the thorough separation of real lineage ties from the bulk of the business—allowed the house to solve the problems posed by succession.

When facing the public, however, the Mitsui took on the single name, that of Mitsui Hachirōemon. In their service to the shogunate and their home domain of Kishū, the Mitsui insisted that “… unlike other townsfolk, [we] are bound by the house law passed down from our ancestors into a single body united as one.”Footnote ¹⁰³ Since Hachirōemon, they argued, is “a name of particular heft and importance, if it were punished with a heavy crime, not only all of the stores and residences in the three cities, but truly all of the total assets ( shinjō ) of the Mitsui would be destroyed.”Footnote ¹⁰⁴ The Nakai, in contrast, maintained a strict separation in identity between their different stores precisely to distribute risk and limit the threat of a complete collapse of the enterprise.

In fact, the discovery of the house’s split by the shogunate, despite the Mitsui’s insistence on being “of one body,” led to the forced exile of a top member of the house and the reintegration of the executive board system in 1797.Footnote ¹⁰⁵ Under the new system, local managers gained more power at the expense of the board, and the lineage group grew increasingly distant from the actual workings of the business. After a century-long period of trial and error, the Mitsui managed to eliminate the risks posed by succession and the problems of coordinating among lineage group members, both typical causes for the collapse of other merchant houses.

Their new unity, however, depended in part on the coercive power of the shogunate and their home domain of Kii, which both valued the stability of a unified Mitsui lineage given their role in providing financial services to the authorities. The samurai thus not only accepted the fiction of a single, all-encompassing Hachirōemon—they preferred it. Notably, the authorities did not regard the executive board as an entity in its own right. Like all other businesses in early modern Japan, the Mitsui needed to take the form, however obviously fictitious, of a named household.

Most scholars view “lineage group-based combinations” like the Mitsui as the typical mode of ownership for major Japanese merchant houses.Footnote ¹⁰⁶ Yasuoka Shigeaki argues that lineage groups exercised “collective ownership” ( sōyū ) over large-scale enterprises much in the same way that villages maintained ownership over commons.Footnote ¹⁰⁷ Lineage members received dividends and other forms of support from the group, though they could not dispose of or divide their shares in the collective enterprise like shareholders in a modern stock company.Footnote ¹⁰⁸ This form of ownership, while not recognized under early modern or modern Japanese law, aptly characterizes the internal logic of many large Tokugawa merchant houses, in which the collective “house” laid claim to all of the business assets underneath its authority.Footnote ¹⁰⁹ Such lineage group-based organizations tied together by this logic of collective ownership remained a powerful force in the Japanese economy well into the modern period.

Though founded in different times and under different circumstances, the Mitsui, the Nakai, and the Kōnoike all developed their solutions to the problem of how to run a large-scale business over the course of the mid- to late eighteenth century, a time of increasing scale in markets and competition.Footnote ¹¹⁰ The ability of the Nakai to extend their reach into local markets across Japan in some respects reflects this development. Given the Mitsui’s close financial ties with the shogunate, and the fact that they largely worked within the three shogunally controlled cities of Edo, Osaka, and Kyoto, they likely could not benefit as readily from the division of their assets into separate households. Well-studied large merchant houses like the Kōnoike and Sumitomo also limited their operations to the same cities. They had no place to hide.

That these houses are often taken as paradigmatic cases of large-scale Tokugawa business organization has led to a narrowing of our perspective on the nature of business in early modern Japan. Their organization offers little help in explaining how merchants could develop large-scale business networks that spanned multiple jurisdictions. In contrast, the Nakai can offer better insights into how early modern businesses solved the problems of distance and scale.

The Nakai stand out not just for the scope of their operations, but also for the aggressive deployment of fictive household entities. They did not limit themselves to a single partnership masked as a household, like the examples mentioned in section one. They even made their own fictive household entities parties to partnerships, incestuously reproducing fictive household-store entities much like how many businesses today stack limited liability corporations one on top of the other. This unusual mode of business organization shows the extent of what was possible for enterprises within the institutional context of early modern Japan.

Tokugawa Businesses in Comparative Perspective

The use of the household form in the absence of other formal or legal vehicles for the organization of business assets stands out as the unique feature of large-scale businesses in early modern Japan.Footnote ¹¹¹ According to a long tradition of scholarship, the absence of corporate law or even alternative modes for business organization ought to have inhibited the formation of big businesses.Footnote ¹¹² Limited liability partnerships and the joint stock corporation, many argued, not only provided decisive advantages to businesses but also emerged as part of a grand shift from personal to impersonal modes of economic relations and from rule by status to contract.Footnote ¹¹³ One of the most recent exponents of the traditional view regarding the superiority of the joint stock corporation, Ron Harris, lays out seven attributes that made the European business corporation distinct: longevity, collective decision-making, joint-stock equity finance, investment lock-in, transferability of shares, protection from state expropriation, and asset partitioning (both asset shielding and limited liability).Footnote ¹¹⁴

Measuring Tokugawa business on the yardstick of this ideal-type image of the European joint-stock corporation can help clarify what distinguished the development of business organization in early modern Japan. Tokugawa merchant houses enjoyed the benefits of longevity, collective decision-making, and asset partitioning. The household, much like a corporation, waqf, or entail, was eternal. Tokugawa merchant houses hired non-kin professional managers and even adopted talented non-kin as heirs, avoiding the pitfalls of succession so commonly seen in family firms, and even developed formal collective decision-making structures.Footnote ¹¹⁵ Splitting up assets among different households worked to limit risk and shield assets. But in the absence of private foreign trade, major colonial projects, or industrial enterprise, businesses during the Tokugawa period did not require the high levels of capital that only joint-stock corporations could provide. Merchant houses in Tokugawa Japan developed some of the advantages of the modern corporation without recourse to legal or formal institutional innovations. Yet they failed to develop joint-stock equity finance or transferability in the absence of any real use for these features.

In fact, the Nakai’s fictive household partnership model closely resembled the partnership system of the Medici, the most famous of the northern Italian merchant houses. Each office underneath the Medici main house existed as a separate legal entity, each a partnership with its own name, capital, and books.Footnote ¹¹⁶ They dealt with each other as outsiders, and the use of different names, as with the Nakai, helped shield assets. In contrast, earlier major banking firms like the Bardi, Peruzzi, and Acciauoli failed due to an over-centralization of the business under the name of a single family that acted as the leading partner. These firms resembled the Mitsui in their commitment to acting under one name, pointing to how a diverse range of institutional and organizational solutions to the common problems of business can coexist within the same historical setting.

According to Raymond de Roover, the Medici’s decentralized organization prefigured the modern holding company. Writing around the same time, Egashira Tsuneharu made the same claim about the Nakai, likening the house to a German Konzern or large-scale conglomerate.Footnote ¹¹⁷ The idea that the development of business in early modern Japan paralleled that of contemporary Europe appealed to Egashira and many of his contemporaries, who also argued for the indigenous invention of key techniques like double-entry bookkeeping in premodern Japan.Footnote ¹¹⁸

That some of the most iconic firms of medieval northern Italy, long seen as one of the birthplaces of capitalism thanks to their superior forms of business organization and accounting techniques, shared their basic structure with merchant houses some hundreds of years later and oceans away reminds us that the convergent development of solutions to common problems of business need not depend on the adaptation of similar institutional forms like the partnership or corporation.

Recent scholarship on business organization in early modern and modern Europe has challenged the significance of formal institutional innovations to the actual functioning of the economy. Francesca Trivellato writes that the limited partnership, long deemed a crucial development in the history of business organization, remained quite rare in early modern Tuscany, showing that fewer than fourteen limited partnerships were registered on average per year from the mid-fifteenth to the beginning of the nineteenth centuries.Footnote ¹¹⁹ This suggests that even if formal institutional alternatives to the household existed in Tokugawa society, their adoption might not have been as widespread as their on-paper advantages would lead us to believe.

When looked at from the perspective of the business history of early modern China, the informality of Tokugawa business organization seems far less unusual. Like in early modern Japan, Ming and Qing merchants split up stores, entities, and rights into shares, creating businesses that could outlast a single generation in the absence of a formalized system of corporate law.Footnote ¹²⁰ As Madeleine Zelin shows in her work on the large-scale salt enterprises of Zigong, merchants in China could form large-scale businesses using lineage trusts, set up as “shareholding corporations” that perpetuated themselves over the generations with shares split among a lineage group.Footnote ¹²¹ Like Tokugawa merchant houses, they mostly recruited non-kin employees. Qing merchants also set up fake people—not fictive households—and even used the names of gods to act as owners of brokerage licenses, with the actual ownership split among several people.Footnote ¹²² Their structure and reliance on the lineage group offer one point of comparison with houses like the Mitsui. In both cases, merchants made use of deceptively familial forms to create more impersonal, complex business organizations. But unlike the Nakai and other Tokugawa merchants, Qing trusts lacked both owner and entity shielding, whereas the creation of fictive household entities allowed for the easier protection of assets.

Differences also extended to naming practices. Unlike in Japan, merchants in China could establish named corporate entities. George Qiao shows how Shanxi merchants set up partnerships with specific enterprise names that operated autonomously in the marketplace “despite the lack of corresponding legal stipulations in the Qing code.”Footnote ¹²³ Businesspeople used the term “mark and name” ( zihao ) to refer to the firm, and the very act of setting a name for a business became a crucial step in receiving common recognition for a new enterprise. In contrast, the flexibility of the early modern Japanese household—its inherent “firm-ness”—actually disincentivized the development of even privately recognized named partnerships that did not take the household form.Footnote ¹²⁴

Part of the reason for this adaptation of informal means to corporate ends lies in the nature of the relationship between the state and the economy in both societies. Late imperial China’s basic posture toward business was one of non-intervention, with Meng Zhang even describing the Qing state as essentially laissez-faire.Footnote ¹²⁵ Few would go as far in describing the relationship between the samurai and the economy in early modern Japan. The shogunate and the domains recognized guilds, established monopolies, exacted forced loans and donations, and set up an array of complex financial schemes in partnership with merchants. Yet like the Qing, they stayed aloof from the question of business registration and generally preferred to leave commercial disputes up to merchants.

Scholars like Egashira, de Roover, and many others since have liked to draw comparisons between large-scale premodern businesses and their modern counterparts to emphasize the dynamic, progressive nature of their subjects. But neither the Medici, the Nakai, nor the Qing lineage trusts were on the path to the modern corporation. Any similarities show instead just how effective premodern, informal mechanisms for the creation of enterprises were. Their success shows how the creation of new institutions was not always necessary or desirable. Existing institutional forms did the job just fine.