Vol. 12, No. 2, Ueda Shinya

Contents>> Vol. 12, No. 2

Family Structure in Early Modern Vietnam: A Case Study of Villages around Huế

Ueda Shinya*

*上田新也, Thang Long Institution of Cognition and Education Studies, Thang Long University, Nghiem Xuan Yem, Dai Kim, Hoang Mai Hanoi 11718, Viet Nam
e-mail: uedashinya_vnjp[at]yahoo.co.jp
https://orcid.org/0000-0003-3797-1399
Due to space limitations, Table 4 is opened on the following website: https://www.jstage.jst.go.jp/article/seas/12/2/12_239/_article/-char/en#supplimentary-materials-wrap

DOI: 10.20495/seas.12.2_239

Introduction

In present-day Vietnam, patrilineal kinship groups called dòng họ are widely dispersed. In recent years there have been some transformations and a decline of the family structure due to urbanization—for example, an increase in nuclear families and a declining birth rate—as well as some regional differences such as between the north and south. But for the Kinh, who are the main ethnic group in Vietnam, accounting for more than 90 percent of the country’s population, dòng họ is considered part of their “traditional culture.” However, from a historical viewpoint, there have been various arguments about the formation and transformation of the Kinh people’s patrilineal kinship groups.

The beginning of this argument was most likely a series of studies in Lê triều hình luật 黎朝刑律 (Lê code). Lê code, which contains numerous regulations unique to Vietnam though it is a law adopted from China, has attracted attention from early on as a symbol of the cultural and social uniqueness of premodern Vietnamese society. It drew the attention of researchers such as Makino Tatsumi (1930) and Yamamoto Tatsuro (1938) during the first half of the twentieth century, and since then researchers such as Katakura Minoru (1987), Yu Insun (1990), and Yao Takao (2020) have conducted further research. According to their studies, the original restrictions contained in Lê code, especially those related to the family system—such as marriage and inheritance—represented the Southeast Asian social situation at the time of enactment.

However, a new research trend has emerged in recent years. This consists of studies using local documents from the early modern period. Vietnamese economic reforms from the 1990s onward enabled foreign researchers to conduct field surveys, resulting in the rise of studies utilizing local documents. In particular, the new historical materials made it possible to examine local society and communities in ways that government compilations of historical documents could not. As a result, the study of premodern Vietnamese history since 2000 has made a clear distinction not only in terms of content but also in terms of historical materials,¹⁾ and research fused with anthropology using family genealogy and testaments is rapidly developing in early modern Vietnamese social history.²⁾ At a minimum, these studies indicate that Vietnam’s patrilineal kinship group, known as dòng họ, is not just a copy of China’s but has its own uniqueness. These new studies have led to a controversy among some researchers about the social status of patrilineal kinship groups and women’s social status in premodern Vietnamese society.³⁾

Thus, in recent years, debates have resurfaced in cultural anthropology and history over premodern dòng họ. However, the fundamental issue of why a patrilineal kinship group with fixed and closed membership like dòng họ developed in Southeast Asia, which is generally characterized by a “loosely structured society,” has not been extensively examined. Therefore, in previous research, the author (Ueda 2021a) took Thanh Phước village near Huế as an example and demonstrated how the village community gradually became closed due to the limits of agricultural development and population pressure, resulting in the establishment of social groups based on Confucianism. In another article, Ueda Shinya (2021c) examined the testaments and land cadastres for the nineteenth-century Red River Delta and established that “multi-household compounds” may have been transformed into patrilineal kinship groups by the generalization of patrilocal residence due to the popularization of Confucianism. The above studies did not cover the specifics of household composition and household division. Nevertheless, after collecting more village documents around Huế, the author discovered some historical sources recording detailed ownership of residential areas in eighteenth- and nineteenth-century villages. In this article, we will introduce the outline of these village documents and examine the family structure and household division at that time.

I Outline of Viên Bộ

I-1 Basic Bibliographic Information for Viên Bộ

First, the author would like to explain the viện bộ 園簿 (garden cadastre) that was used as the main historical source in this article. Simply described, the garden cadastre is a land register that records the landowners and areas in the residential area of the village for each parcel of land. As is generally well known, the Nguyễn Dynasty conducted nationwide surveys since the dynasty was founded, and the survey results are still in existence as a địa bạ 地簿 (land cadastre). However, residential areas comprising house and garden parcels were only recorded with their total area as thổ trạch viên trì 土宅園池 or thổ trạch viên cư 土宅園居 in the land cadastre. The main purpose behind compiling the land cadastre was to determine the tax amount for each village. Considering the administrative costs, it is understandable that there was no detailed survey of the living spaces that were not taxable under the tax regulations of the Nguyễn Dynasty.⁴⁾ However, when we consider the premodern family structure in early modern Vietnam, the lack of information on residences presents a barrier for analysis. In fact, in the previous studies using the land cadastre up to the present, only the amount of cultivated area for each individual was examined, and the type of family/household connection was not comprehensively considered.⁵⁾ However, because the “garden cadastre” used as the main historical source in this article records the detailed ownership status of each parcel of land in the residential area, we can use this to speculate on the actual living conditions and family ties in the village.

The garden cadastre is a type of historical material that is hardly known even in Vietnam, and the author obtained a total of only four documents photographed in two villages in the northern suburbs of Huế. At this point, it is unclear how many analogous historical materials exist, and we must await the results of future investigations. Prior to these analyses, I will introduce the bibliographic information and outline of the four garden cadastres.

The garden cadastres analyzed in this article were photographed in the village communal halls (đình) of Thanh Phước and An Thành villages.⁶⁾ Both settlements are located about 7–8 km north of the Imperial City of Huế, and they are only a few kilometers apart from each other. The basic bibliographic information of the garden cadastres photographed by the author is provided in Table 1.

 

 

In this article viên bộ is used to refer to the four village documents mentioned above, though the actual titles of each document vary slightly. This term applies not only to the title but also to the format and content, which are slightly different for each era and village. There is no unified format such as the Nguyen Dynasty’s land cadastre. For example, in Thanh Phước A and An Thành A, which were compiled in the late eighteenth century, landowners are listed in the form of a personal pronoun and personal name such as “Lão Phúc 老福 (Old Phúc),” “Chú Minh 注明 (Uncle Minh),” and “Mụ Lưu 媒畱 (Aunt Lưu).” Modern Vietnamese also uses various personal pronouns with given names depending on the person’s age, gender, and social status. The method of writing the personal names in the two garden cadastres is basically a chữ Nôm notation of “personal pronoun + given name (or alias).”⁷⁾ On the other hand, An Thành B, compiled in the mid-nineteenth century, is basically described in the form of “family name + middle name (tên đệm) + given name” with Chinese characters, such as “Trần Viết Tô 陳曰蘇” and “Trần Viết Lương 陳曰良.”

In addition, the garden cadastres of Thanh Phước A and Thanh Phước B not only document the owner, location, length, and area of each parcel of land but also record the usage status of each parcel, such as viện thổ 園土 and viện cư 園居. Viện thổ probably refers to a parcel of land meant only for a garden, while viện cư most likely refers to a parcel of land where a house is built in the garden. However, in the case of An Thành A and An Thành B, the owner, the length, and the area of all four sides are listed, which is the same as Thanh Phước. In addition, an amount of money is recorded. This is the tax levied on the parcel of land, as described later. However, it appears that both garden cadastres of An Thành are unconcerned about the usage status of each parcel, because An Thành A classifies all the pieces of land as Thổ trạch 土宅 (garden and residence). Conversely, An Thành B classifies all the land pieces as Thổ viên 土園 (garden) for the same residential area. That is, the garden cadastre of An Thành does not strictly distinguish between “garden” and “garden with house” in the document.

Instead, the garden cadastres of An Thành provide us with a more detailed transaction and inheritance history for each parcel of land. For example, the parcel of land owned by Uncle Bỉnh 注秉 listed at the beginning of An Thành A and the parcel of land owned by Trần Viết Lương 陳曰良 at the beginning of An Thành B clearly refer to the same parcel of land, because the length and area of all four sides are exactly the same. According to An Thành B, this parcel was purchased by Trần Viết Tô 陳曰蘇 and subsequently inherited by Trần Viết Lương. In other words, when we combine the information from these two sources, we can see that between 1796 and 1859 Trần Viết Tô purchased the parcel of land from “Uncle Bỉnh” and subsequently bequeathed it to his offspring Trần Viết Lương. Furthermore, when the ownership of the land was transferred through a transaction after creating the garden cadastre, a note to that effect is appended to the left and right.⁸⁾ By combining these descriptions, it is possible to trace the sales transactions and inheritance of a residential area from the end of the eighteenth century to the beginning of the twentieth.

I-2 Purpose of Compiling the Garden Cadastre

In this sub-section, we will briefly examine why the documents called “garden cadastre” were compiled in Thanh Phước and An Thành villages, and clarify the social background of the documents. To begin with, we introduce the regulations at the beginning of An Thành B.⁹⁾ The head of the garden cadastre describes the situation as follows:

We, the village officials of An Thành commune, An Thành canton, Quảng Điền District, Thừa Thiên Prefecture, gathered together and considered that roads, gardens, and houses always change their boundary over time. Thus, maintained according to the previous scale regulations, the village officials worked together to measure the area of each parcel of land in the residential area. Compared to the previous garden cadastre [An Thành A], there was an increase or decrease, but it was not due to an individual’s evil act. When editing the landowner, tax amount, and area of land, we rewrote or followed the description of the previous garden cadastre [An Thành A], but of course, we have no evil intention at all in those new registrations. Regarding the new and old regulations, we will clarify by adding an annotation to each entry. Keep this new garden cadastre for generations and make this an invariant rule. The details are described below.

1. Every year, the tax is 130 tiền per 1 sào as payment of land tax by coin. If another commune purchases the garden, the tax is 230 tiền as payment of land tax by coin. The regulation is stipulated here.

2. Hereafter, none of the landowners of the residential area shall be allowed to lend to other villages. If a person violates this rule, the punishment will be one pig, one tray of betel nuts, and liquor. The regulation is stipulated here.

3. If a person permits the lease of a garden to other people in the commune, the contract should be submitted to the village official of our commune, and his name should be recorded in the garden cadastre, and the contract will be transcribed and used as evidence. After that, he/she must pay 1 quan, betel nuts, and liquor. If a person makes a bad attempt and falsifies it, it will definitely be a felony.¹⁰⁾

As stated in the preamble, An Thành B was compiled in 1859 to update the contents of An Thành A, compiled in 1796. The first article that follows the preamble most clearly indicates the purpose behind compiling the garden cadastre. Because Article 1 stipulates that the owner of each parcel of land in the residential area pays tax according to the owned area, in the case of the residents of An Thành village, the tax rate is 130 coin/1 sào.¹¹⁾ The tax rate for residents of other villages is 230 coin/1 sào. This is the main reason why not only the land area but also the amount of money is recorded for each parcel in both garden cadastres of An Thành. In other words, the garden cadastre was compiled to collect taxes from residents according to the area owned in the residential area. It is probable that the garden cadastre was a kind of tax collection ledger compiled by the commune (xã 社). Another point to note is the restriction on landownership for outsiders. In Article 1 the tax rate is set high for outsiders, and Article 2 prohibits future sales of land to outsiders. An Thành A, which is difficult to read due to missing characters, also has tax regulations for village members at 300 tiền/1 sào and 500 tiền/1 sào for outsiders.¹²⁾ The second article of An Thành A is also difficult to read, but it states: “If a person from another commune buys the parcel of land in our commune and builds a house, . . . [missing some characters] . . . must not record it with the land cadastre of another commune (如有他社人造買在本社地分内土, 而結立家居 . . . 不得著入他社簿).”

It is presumed that the reason for preventing land purchase by an outsider was to avoid disputes caused by the purchased land being registered as the land of the other village.¹³⁾

However, if the main purpose of the garden cadastre was to collect taxes from the owner of the garden, a new question arises. As mentioned above, in the land cadastre of the Nguyễn Dynasty, the description of a residential area is very simple, because the residential area is tax-exempt. Why was such a tax collected in Thanh Phước and An Thành villages? Was it not a legitimate tax but a kind of “village membership fee” that each village collected voluntarily? To clarify these questions, we look at the land cadastres from the Nguyễn Lords period to the Nguyễn Dynasty period and observe that the two villages had a common feature regarding the legal treatment of residential areas in the national land system.

At first glance, in the case of Thanh Phước village—whose land cadastres from 1669 to 1814 still exist (Ueda 2019, 298–305)—we make a strange observation. In the land cadastres from the Nguyễn Lords and Tây Sơn periods, all of Thanh Phước’s village territories are recorded as public rice fields (công điền 公田) or official land (Quan thổ 官土). If this registration is accepted as the truth, it means that there were no residential areas in Thanh Phước village, only agricultural lands, which is obviously unlikely. Probably, the parcel called “Viên Trạch xứ 園宅處” (meaning “parcel of garden and house,” 37 mẫu 1 sào), which was registered as official land, was actually a residential area.¹⁴⁾ That is, in the case of Thanh Phước, from the Nguyễn Lords period to the Tây Sơn period, its residential area was registered as taxable “official land” under the land system, and thus the village had to collect money from the residents in some way to pay the land tax. This was the reason behind compiling the garden cadastre. However, “Viên Trạch xứ” was removed from the land cadastre of the Gia Long era in 1814; instead, a new description appeared as follows: “the former constructed area consisting of garden, house, village communal hall, market, shrine, and pond is taxed as rice field. 34 mẫu 4 sào 3 tước 3 tấc.” This means that even though residential areas were not taxable under the Nguyễn Dynasty tax system, land tax was still collected because the residential area of Thanh Phước was classified as “rice field” under the tax system. In other words, the tax system of the Nguyễn Lords was virtually inherited by the local administration of the Nguyễn Dynasty, who disregarded the tax system laid down by the Nguyễn Dynasty’s central court.

The residential area of An Thành village is also similar to that of Thanh Phước village. There is no residential area in the land cadastre of An Thành compiled in 1731 (An Thành-Đình40), which still exists in the village communal hall. However, two official lands of “Thượng Thôn xứ 上村處” and “Hạ Thôn xứ 下村處” (meaning “parcel of the upper village” and “parcel of the lower village”; total area about 12 mẫu) were registered. Probably, these parcels were residential areas. In the land cadastre of the Gia Long era in 1814 (An Thành-Đình36), instead of those parcels of land disappearing, there is a new description and note: “Public land 14 mẫu 8 sào 5 tước 4 tấc. (Note: Taxed from Tân Mùi year [1811?]. This commune has been building gardens and houses for some time).” Thus, even though the two parcels “Thượng Thôn xứ” and “Hạ Thôn xứ” had been residential areas since the Nguyễn Lords period, they were not tax-exempt “residential areas” under the land system, and the taxation for residential areas was transferred to the Nguyễn Dynasty.

Therefore, in the case of Thanh Phước and An Thành villages, the residential area was subject to taxation during the Nguyễn Lords period, and this was passed down even after the Nguyễn Dynasty was established. Thus, in both villages it was necessary to compile a detailed cadastre of residential areas in order to collect taxes from residents. As a result, the creation of garden cadastres and their updates continued long after the nineteenth century. Normally, the taxes of rice fields and plowed fields were paid in the form of products of those fields; however, in the case of these two villages, as they were actually places of residence, tax was paid in the form of money. Therefore, unlike the land cadastre, the garden cadastre is a semi-official document independently created by the commune (xã), which was the tail end of the administrative organization, to meet the demands of the state. This is why the format and description are slightly different for each village.

II Analysis by Personal Pronouns

II-1 Household Size in Thanh Phước Village

The average household size will be defined as a basis for analyzing family structure using the garden cadastres. The garden cadastres of Thanh Phước village are useful in this analysis. During the period of study, Thanh Phước village had two garden cadastres; however, Thanh Phước A is not very well preserved—about one-fifth of the total information is missing as the pages were eaten by moths. Thanh Phước B is a transcription of Thanh Phước A, but because Thanh Phước B is almost identical to the remaining part of Thanh Phước A, we can trust its description.¹⁵⁾

In Thanh Phước B the residential area is subdivided into 289 parcels of land, and each is recorded with its owner and area. What is important and advantageous while considering family structure is that these parcels of land are classified into viên thổ 園土 (garden) and viên cư 園居 (garden with house). When the utilizations of the parcels of land are arranged according to the descriptions in Thanh Phước B, “garden with house” accounts for just 100 parcels.¹⁶⁾ That is, as of 1772, there were 100 houses in the residential area of Thanh Phước. Meanwhile, per the census registers in 1786 (ĐTP45), the oldest population registration in Thanh Phước village recorded 162 males aged over 20, and the census registration in 1793 recorded 201 males (including 48 males aged 17 to 19). Judging from these censuses, it is estimated that Thanh Phước village at the end of the eighteenth century had around 160 adult males. Assuming that there was no significant change in the number of houses and population in the village between 1772 and 1786, we can presume that each house had about 1.6 adult males on average. Vietnamese premodern census registrations recorded only males who were subject to taxation and conscription, and no females were recorded at all. Therefore, if the village had roughly the same number of adult females, each house had about 3.2 adults on average. In other words, there were only three to four adults in each house, and it is not possible to assume a large household size. The typical household composition assumed from this figure consists of parents, one or two grandparents (grandfather/grandmother), and minor children. In the case of slightly older families, the household was made up of parents, one adult son/daughter, and minor children. That is, the general household in late-eighteenth-century Thanh Phước was a nuclear or lineal family. When a child came of age, he/she formed an independent household due to marriage and was separated from the parents’ household. At the very least, a large family system in which the patriarch had strong power is hard to imagine from these figures.

II-2 Classification by Personal Pronouns

As mentioned in the previous section, the owners of each parcel of land in the garden cadastres of the eighteenth century were recorded by the colloquial notation of “personal pronoun + personal name” rather than the official family name and given name. More than one hundred landowners are recorded in the garden cadastre with various personal pronouns. Most of the Vietnamese personal pronouns are derived from the words used to indicate relationships between members of a family. For example, ông (meaning “grandfather”) is used as the personal pronoun for a man who is considerably older than the speaker, and chú (meaning “younger brother of father/mother”) is used as the personal pronoun for a man who is considerably older than the speaker but younger than the speaker’s father.

However, in Vietnamese the personal pronoun is variable due to the relative age relationship and hierarchical relationship between speakers. For example, if an eighty-year-old man and a 75-year-old man are having a conversation, the older man can call the younger one em, which means “younger brother/sister.” Therefore, the personal pronoun and the actual age do not always match. According to such rules of grammar, if the editor of the garden cadastre was a very young person, most of the landowners would be recorded by the personal pronoun for elder people; and conversely, if the editor of the garden cadastre was an aged person, most of the landowners would be recorded by the personal pronoun for younger people.

Although the garden cadastre was not an official document of the Nguyễn Dynasty like the land cadastre, it was a public document used by the village to collect land taxes from residents. Probably, the use of personal pronouns, which could change greatly depending on the personal relationship between the editor and the landowner, was not common. Rather, it is assumed that the use of personal pronouns in the garden cadastre was determined according to the relationship between the village and the landowner, or by the social status of the landowner in the village. Even in modern Vietnamese, personal pronouns are often determined by hierarchical relationships and social status, regardless of age—for example, the boss at work is called anh/chị, which means “elder brother/sister.” Unfortunately, there is no description in the garden cadastre of an objective standard for the use of personal pronouns. We have to rely on analogies from the modern Vietnamese language for many of these usages. Table 2 summarizes the usage of each personal pronoun and the landownership of each personal pronoun group in Thanh Phước and An Thành.¹⁷⁾

 

 

II-3 Men’s Landownership

We can understand the cycle of land acquisition and dissolution in the village to some extent by examining each generation. According to >Table 2, basically, the owned area per person tends to gradually increase from the young to the elderly in both Thanh Phước and An Thành villages. The average owned area in Thanh Phước and An Thành is nearly twice as large, because the garden cadastre of Thanh Phước was not only a registered residential area called Nội viên thổ 内園土 (“Inner garden”), where “garden” and “garden with house” were mixed, but also was registered as an upland field called Ngoài viên thổ 外園土 (“Outer garden”), which was located along the Hương River, north of the settlement. The per capita owned area in the residential area called “Inner garden” is not very different from that of An Thành village.

Furthermore, in both villages the number of people in the “Cha” group, which supposedly refers to men aged between forty and fifty years, is much smaller than that of the “Lão” and “Chú” groups. This is probably because many of the official groups comprised men between the ages of forty and fifty. In the case of Thanh Phước village, many people who belong to the official group use the personal pronoun Viên, and in the case of An Thành, official titles are used instead of personal pronouns, although it is assumed that they actually refer to the same individual.¹⁸⁾ In the administrative documents of Thanh Phước during the Nguyễn Lords and Tây Sơn periods, village officials are often referred to as Viên chức 員職. In the Nguyễn Lords period, tax collectors with various names were abundant as government posts were on sale, and they were collectively called Giám tô viên 監租員. However, in the first half of the eighteenth century tax collectors were so numerous that their number was regulated according to the population of each administrative unit.¹⁹⁾ Probably, the “Official” groups in Thanh Phước and An Thành villages (Table 2) of the eighteenth century made up a terminal organization for tax collection, and in reality its members were wealthy peasants in charge of tax collection in the village.

We may infer the family cycle at that time to some extent from the garden books of both villages. In both villages, the areas owned by the elderly group were the largest; the area owned by Thanh Phước’s Ông/Bác group was particularly large. However, there was a large disparity in the owned area between the five people in the group. Three men—Ông thủ Hiến 翁首憲 (six parcels of land, total 5,688.3 m²), Ông Lam 翁藍 (four parcels of land, total 3,074.4 m²), and Ông giáo Uyên 翁教淵 (four parcels of land, total 2,751.2 m²)—owned a fairly large area. The area owned by the two remaining men—Ông thủ Công Điển 翁首公典 (one parcel of land, 537.3 m²) and Bác cai Phù 博該扶 (one parcel of land, 509 m²)—was extremely small. We can assume that Công Điển and Phù kept only what was essential for their existence and passed on the rest to their children during their lifetime.²⁰⁾ As a result, there was a large disparity depending on whether the person was retired or still in the Ông/Bác group. Nevertheless, the Thằng group clearly owned a small area in both villages. They were a relatively new household that had become independent from an elderly household through property division and had not yet accumulated sufficient property. However, as the number of families increased, they purchased new parcels of land and gradually increased their holdings, forming a middle-aged group called Chú.

As mentioned above, Thanh Phước B records each parcel of land by classifying it into “garden” or “garden with house,” and we can see that villagers’ property normally comprised a parcel of a garden with a house and other parcels of garden. However, according to Thanh Phước B, there were a certain number of people who owned only gardens but no houses in the village. For instance, the Chú group included ten people who owned only a garden but no house. On the other hand, four members of Thanh Phước’s Chú group owned two houses. This was an unusual pattern of land possession not seen in other groups in Thanh Phước village.

II-4 Women’s Landownership

As seen in Table 2, there is a clear economic disparity between men and women in both communities. In the case of Thanh Phước, women’s ownership is clearly lower than men’s, and the average area of the Mụ group is almost identical to that of the Thằng group. Moreover, in the case of Thanh Phước the number of listed people is 97 men and 22 women, whereas in the case of An Thành the number of listed people is 37 men and 5 women (excluding those in the “Other” category). Assuming that the actual male and female populations in the villages were almost equal suggests that most women did not own any real estate in the residential area, and the women listed in Table 2 must have belonged to a relatively affluent hierarchy within the village. Assuming that there were many women who were not registered in the garden cadastre, we can assume that the actual economic gender gap was much larger than the data in Table 2. However, from the viewpoint of family structure, of particular note is the high homeownership rate for women. This suggests that, as compared to men, women’s real estate ownership was more likely to be concentrated in the house than in the garden. What makes this fact even more unique is that, as mentioned in the introduction, the kinship group currently called dòng họ is a patrilineal kinship group based on patrilocal residence. If it was common for a wife to live in her husband’s house, why did a woman need to own her house? In addition, who lived in that house?

The first possibility we should consider is that there may have been a certain number of matrilocal residences in late-eighteenth-century Thanh Phước village. A detailed examination of the land registration in the garden cadastre reveals some traces of matrilocal residence. To understand this, it is necessary to explain the common name in Vietnamese. As mentioned earlier, in daily life people often call each other by the “personal pronoun + personal name.” However, in reality, not only the personal pronoun but also the personal name changes. For example, when there is an elderly couple A (male) and B (female), the wife may sometimes be referred to as bà A using her husband’s given name. If we translate this term into English, it means “Mr. A’s old woman.”²¹⁾ Even in Thanh Phước B, which was strongly influenced by colloquial Vietnamese, such common names were adopted mainly by relatively elderly couples. For example, Thanh Phước B records two men named Tiền thủ Duyệt 前首悦²²⁾ and Ông giáo Uyên 翁教淵. Correspondingly, there are records of two women named Mụ thủ Duyệt 媒首悦 and Mụ giáo Uyên 媒教淵. Personal pronouns such as thủ 首 and giáo 教 were used by only a very limited number of people in Thanh Phước, so it is unlikely that the given names matched by chance. It can be considered that the couples Tiền thủ Duyệt and Mụ thủ Duyệt, and Ông giáo Uyên and Mụ giáo Uyên, were almost certainly in a marital relationship. Thus, there are seven pairs who can be inferred to be married couples in Thanh Phước B. Table 3 shows details of these couples.

 

 

In three of the seven couples in Table 3 (No. 2, No. 5, and No. 7), there was a high likelihood of matrilocal residency because the husbands owned several parcels of “garden” but did not own a “garden with house.” Each of their wives owned a parcel of “garden with house.” In other words, the couple owned one house, and the owner of that house was the wife. It is natural to think that they had a matrilocal marriage. In the case of Thanh Phước village, only seven pairs are clearly presumed to have a marital relationship. However, if three of the seven couples were in matrilocal residences, then a few of the female-owned houses also should be suspected of having a matrilocal couple living in them.

However, such matrilocal residence cannot adequately explain a woman’s ownership of a house. This is because, as depicted in Table 3, there are many cases where one person owns multiple houses—as in the case of Lão Tài. Table 3 also suggests that in many situations, such as the No. 1, No. 4, and No. 6 couples, a couple owned multiple houses. These cases demonstrate that female homeownership cannot be explained completely by matrilocal residence alone. As mentioned earlier, the average household size in Thanh Phước was not very large; and it is presumed that adult children lived separately from their parents’ households, establishing their own households at a relatively early stage. When a newlywed couple did not have sufficient financial resources, they most likely constructed and lived in their new dwelling on the land of the husband’s or wife’s parents. As the children’s households built on the land of their parents were registered in the name of their parents, many aged/middle-aged couples owned multiple houses. Consequently, the house ownership ratio of young couples was reduced in the garden cadastre. Assuming this, it is possible to explain the homeownership rate and household size in the garden cadastre of Thanh Phước without contradiction.

III Property Inheritance in An Thành Village

III-1 Garden Cadastre of An Thành and Family Genealogy of Trần Viết Lineage

As previously stated, An Thành village has two garden cadastres: An Thành A, written by someone with the personal pronoun chữ Nôm in 1796; and An Thành B, written by someone with a full name made up of Chinese characters in 1859. Neither of the garden cadastres distinguishes between “garden with house” and “garden,” and we cannot examine homeownership as in the case of Thanh Phước village. However, An Thành’s garden cadastres have two major merits for us. First, An Thành village has two editions of the garden cadastre, one compiled in 1796 and the other in 1859, allowing us to trace the transition of landownership over a long period of time. Second, the Trần Viết lineage 陳曰族 living in An Thành village has a detailed family genealogy through which we can compare the landowner recorded in the garden cadastres with the person recorded in the family genealogy. This makes it possible to infer to some extent what kind of property inheritance took place within the lineage.²³⁾

Table 4 is a genealogical tree based on the family genealogy of the Trần Viết lineage. As established from this, there is extremely scarce information on the generations before Trần Viết An 陳曰安. The given names from the founder to the fifth generation are unknown, and all of them just add the honorific title đại lang 大郎 to the family name Trần 陳. The sixth generation has only the common nickname Nghiễn 硯, and this individual’s official given name is unknown. Subsequently, the full names of 24 individuals are enumerated in the family genealogy without noting their mutual genealogy. It is presumed that the individuals belong to four or five generations, though we cannot ascertain any details. In other words, by the time this genealogy was compiled in 1870, the details of people before Trần Viết An were already unknown.²⁴⁾

 

 

As for the generations after Trần Viết An, the years of birth and death are recorded for some people. From these figures, assuming that one generation spans 25 to 30 years, it is estimated that Trần Viết An lived around the last half of the seventeenth century. Trần Viết Nghi 陳曰儀, the grandson of Trần Viết An, had as many as 11 sons, and the Trần Viết lineage is divided roughly into three branches comprising the second son Trần Viết Tô 陳曰蘇, the eighth son Trần Viết Giao 陳曰膠, and the 11th son Trần Viết Hà 陳曰何. In the family genealogy, all the descendants of Trần Viết Hà have the middle name (tên đệm) Viết 曰; but in reality they all seem to be the same in the garden cadastre using the middle name Đức 德, as mentioned below. Furthermore, according to An Thành B, there were other lineages in An Thành village with the same family name (but different tên đệm), such as Trần Được 陳得 and Trần Văn 陳文. However, the blood relationship between the Trần Viết lineage and these same family-name lineages is unknown because the latter do not appear in the family genealogy of the Trần Viết lineage. It is presumed that the Trần Được and Trần Văn lineages were branches of the Trần Viết lineage before the generations of Trần Viết An in the sixteenth century.

Table 4 illustrates that the An Thành village of the eighteenth and nineteenth centuries had the typical high birth and death rates. For example, Trần Viết Giao 陳曰膠, a fourth-generation man from Trần Viết An, had four sons (and six daughters) from his first and second wives, but only one son got married and left a grandson. The other three sons probably died prematurely, because the family genealogy gives no description of their wives or children. Similarly, Trần Viết Nho 陳曰儒, a fifth-generation descendant of Trần Viết An, had 13 sons (and six daughters) from four women: his first and second wives, and two concubines. However, only five sons got married and left grandsons.²⁵⁾ For the sons who died early without leaving any offspring, the age of death is unknown because only the date of death is recorded. However, at the time there was only a 25 percent to 50 percent probability that a son would marry and leave a grandson; that is, if a man wanted to ensure male descendants, he would need to have at least three or four sons. If there were about the same number of daughters as sons, three or four sons and three or four daughters would be the standard number of children at the time. As it was difficult for a single woman to give birth to this many children, polygamy was very common in the village. In fact, the family genealogy has recorded many cases that are presumed to be divorce due to sterility and remarriage due to maternal mortality. Trần Viết’s lineage consisted of members who were not particularly wealthy but became influential bureaucrats and literati. Most men in the lineage had wives and a few concubines.

However, because the average number of adults per house in the previous section was three to four people, we cannot presume a large household. It is probable that the ownership of numerous houses by men or couples, and the ownership of houses by women, were due to the fact that each wife/concubine and child owned their own house depending on the polygamy scenario. In the case of landed gentry during the early twentieth century in northern Vietnam, each wife and concubine owned her own home (Luong 1989, 748–755). Probably, Thanh Phước and An Thành had many separate households for women similar to Luong’s case study of the landed gentry.

III-2 Inheritance in An Thành Village

Polygamy, as stated earlier, is likely to have resulted in several separate establishments for concubines and mistresses in An Thành village. What implications did this have for property inheritance? The best way to understand the situation is by looking at the inheritance of property in the generations from Trần Viết Tô 陳曰蘇 to his grandchildren in Table 4. The time span of these generations ranges from the end of the eighteenth century to the first half of the nineteenth, when An Thành A and An Thành B were compiled. Therefore, we can easily trace the family property inheritance by comparing the garden cadastre of An Thành with the family genealogy of the Trần Viết lineage. For example, according to An Thành A, the 70th parcel of land was owned by a person called Xã chính Diệu 社政曜 in 1796. This is the same person as Trần Viết Nghi 陳曰儀 in Table 4, because according to the family genealogy of the Trần Viết lineage, Trần Viết Nghi had another name: Diệu 曜. After that, in An Thành B, compiled in 1859, the same parcel of land was written as “Trần Viết Hân 陳曰欣 inherited by the late secretary Trần Viết Tô 陳曰蘇.”²⁶⁾ According to the family tree of the Trần Viết lineage, Trần Viết Hân was a grandson of Trần Viết Tô. Thus, from the end of the eighteenth century to the first half of the nineteenth, it can be observed that this parcel of land was passed down through the male bloodline from Trần Viết Nghi to Trần Viết Hân. Table 5 summarizes the land inheritance from Trần Viết Nghi to his great-grandchildren based on the two garden cadastres of An Thành.

 

 

According to Table 5, Trần Viết Nghi owned three parcels of land in 1796, two of which were inherited by his second son, Trần Viết Tô (ostensibly the eldest surviving son, because the eldest son had passed away early). It is unknown what happened next, but the remaining parcel was owned by a man named Trần Được Chính 陳得正 in 1859.²⁷⁾ Among the sons of Trần Viết Nghi, the eighth son, Trần Viết Giao 陳文膠, tenth son, Trần Viết Hạc 陳曰鶴, and 11th son, Trần Viết Hà 陳曰何, also left offspring; however, no trace of them can be found in the garden cadastre.²⁸⁾ Trần Viết Tô bought six parcels of land in addition to the property inherited from his father. According to the family genealogy, Trần Viết Tô had at least four sons, but three of them died young. As a result, it is presumed that all his parcels of land were inherited by his second son, Trần Viết Nho. Trần Viết Nho had 13 sons and six daughters from his former wife 元配, second wife 再配, concubine 側室, and mistress 小妾. Of his children, the fourth son, Trần Viết Lương 陳曰良, fifth son, Trần Viết Long 陳曰隆, sixth son, Trần Viết Hân, seventh son, Trần Viết Hoan 陳曰歡, eighth son, Trần Viết Gian 陳曰閒, and 13th son, Trần Viết Cẩn 陳曰謹, grew to adulthood and left male descendants. Of these six sons of Trần Viết Nho, the five other than Trần Viết Hoan are listed in An Thanh B. In terms of the boys’ inheritance, their father, Trần Viết Nho, gave preferential treatment to the eldest son of his former wife, divided the share of inheritance evenly among the sons of his second wife, and gave the son of his mistress about half the inheritance of the sons of the second wife.²⁹⁾

Polygamy was widely practiced in An Thành village, as mentioned above, to ensure male offspring due to the high birth but high death rates. In many cases, a husband would have as many sons as possible with his legal wife. When it became difficult for the latter to give birth due to her age, the husband would set up concubines to have more sons. As a result, it was not uncommon for a man to have two or three wives and concubines and ten or more children. However, that does not mean everybody lived under one roof. Considering the average household size in Thanh Phước, it may be presumed that each wife/concubine had one house, where she formed a semi-independent household with her own minor children. In fact, Thanh Phước A has some cases where a middle-aged man owned two houses, and the parcel of the second house is marked with “Allow Aunt X to live.” Probably this meant that his concubine or mistress lived in the secondary house owned by him or his legitimate wife. This assumption may explain to some extent the ownership of multiple houses by middle-aged people and the ownership of a house by a woman.

III-3 Landownership by Women

As we saw in the previous sub-section, it was not uncommon for a man to have more than ten children in the Thanh Phước and An Thành of the eighteenth and nineteenth centuries. Naturally, about half the children were female, but there is little evidence that daughters inherited their parents’ property in either the family genealogy of Trần Viết or the garden cadastre of An Thành. For example, in the case of An Thành B, which describes in detail the history of parcels of land, many parcels were recorded as “Bequeathed from A to B” or “Purchased by A and bequeathed to B” with both the decedent and the heir being men. Most of the lands owned by women were marked only as “Purchased by A.” This indicates that, in principle, women were not included as heirs in the inheritance of parents’ real estate.³⁰⁾ However, there are many cases in which women conducted land transactions. In this section, we will introduce some cases of land transactions to clarify landownership by women.

◆ Case 1: No. 6 parcel of land in An Thành village
According to An Thành B, No. 6 parcel of land in An Thành village was owned by Trần Đại Điền 陳大田 in 1859. However, in 1877 it was sold by a woman named Đỗ Thị Điền 杜氏田 to a woman named Lê Thị Viện 黎氏院. Subsequently, in 1882, Lê Thị Viện 黎氏院 and her son Lê Văn Khoa 黎文科 resold this land to Phan Văn Thiệp 潘文渉. As mentioned earlier, since it was customary for a woman to be called by her husband’s name, Đỗ Thị Điền was probably Trần Đại Điền’s wife. It is speculated that Đỗ Thị Điền became a seller because Trần Đại Điền died between 1859 and 1877. Lê Thị Viện bought the land from Đỗ Thị Điền on her own, but the land was resold under the names of Lê Thị Viện and her son five years later.

◆ Case 2: No. 7 parcel of land in An Thành village
According to An Thành B, No. 7 parcel of land in An Thành village was purchased by the late Đỗ Thị Nghĩa. However, Đỗ Đăng Sĩ 杜登仕 sold the land to a woman named Nguyễn Thị Mãi 阮氏買 in 1859.³¹⁾ The relationship between Đỗ Thị Nghĩa and Đỗ Đăng Sĩ is unknown, but there is a high possibility that they were a parent and child.³²⁾ Nine years later (1868), Nguyễn Thị Mãi alienated the land for a woman named Nguyễn Thị Hám 阮氏憾. In 1872, Nguyễn Thị Hám resold the land to a woman named Thị Vẫn 氏隕 (her family name is unknown) and her son named Trần Văn Bị 陳文備. Nguyễn Thị Mãi is unknown, because there is no historical material, but Nguyễn Thị Hám was the second wife of Trần Việt Hà 陳曰何 (the 11th son of Trần Viết Nghi) in Table 4. His former wife, Nguyễn Thị Tùng 阮氏叢, had given birth to only a son and a daughter; and after that, his second wife, Nguyễn Thị Hám, gave birth to seven sons and five daughters. Since the first wife probably died after giving birth to only two children, Trần Viết Hà remarried Nguyễn Thị Hám. Judging from the phonology, it is presumed that a woman named Thị Vẫn 氏隕 and her son named Trần Văn Bị 陳文備 in An Thành B are the very same people: Nguyễn Thị Văn 阮氏聞 (wife of Trần Viết Hà’s second son) and Trần Văn Bị 陳曰被 (Trần Viết Hà’s legitimate grandson) in Table 4.

The transaction between Nguyễn Thị Hám and Nguyễn Thị Văn involved complex family circumstances. In the family genealogy of the Trần Viết lineage, although Trần Viết Mỹ 陳曰羙 (Trần Viết Hà’s eldest son) had neither a wife nor a concubine, he had one son named Trần Viết Bị. Instead, Trần Viết Bị is annotated with “Change mother改母.” Presumably, Trần Viết Bị’s parents died young for some reason, and he was brought up by his uncle Trần Viết Kim’s wife, Nguyễn Thị Văn. As a result, Trần Viết Bị’s position in the Trần Viết Hà branch was subtle because Trần Viết Bị was a direct descendant of Trần Viết Hà in the genealogy. However, he was brought up in the household of his father’s stepbrother Trần Viết Kim. Thus, although Trần Viết Bị was a grandson in the direct bloodline of Trần Viết Hà, it was difficult for Trần Viết Kim to give him preferential treatment in the inheritance of property like Trần Viết Nho and his sons in the previous section. Therefore, the grandparents prepared the land for their grandson Trần Viết Bị separately from the land to be inherited by their sons, and the adoptive mother purchased it on the condition that it would be inherited by him after adulthood. This line of thought can easily explain a series of transactions.

◆ Case 3: No. 26 parcel of land in An Thành
As of 1859, No. 26 parcel of land in An Thành was only marked as “purchased by the late Nguyễn Xuân Bế 前阮春閉.” Its inhabitants in 1859 are unclear. This parcel was bought in 1866 by a man named Lê Văn Viện 黎文院, and then in 1888 it was sold by three people—“Ngô Thị Hóa 呉氏化 and her sons Lê Khoa 黎科 and Lê Cường 黎強”—to the Trần lineage. It was probably sold by Lê Văn Viện’s wife and sons after his death.

◆ Case 4: No. 9 parcel of land in An Thành
As of 1859, No. 9 parcel of land in An Thành was owned by Corporal Trần Đức Bài 伍長陳德排. In 1877, two people—“Thị cai Bài and Trần Đức Phượng 氏該排, 陳德”—sold it to a woman named “Lãnh binh Viện’s legal wife 領兵院正室.”³³⁾ It may be presumed that Trần Đức Bài and Trần Đức Phượng were the same as Trần Viết Bài 陳曰排 (third son of Trần Viết Hà) and Trần Viết Phượng 陳曰鳳 (eldest son of Trần Viết Bài) in Table 4.³⁴⁾ “Thị cai Bài” is the same as Hoàng Thị Hợi 黄氏亥 (Trần Viết Bài’s second wife).³⁵⁾ It is supposed that after the death of Trần Đức Bài, his wife and eldest son sold part of his bequest in the name of the mother and child.

◆ Case 5: No. 84 parcel and No. 94 parcel in An Thành village
As of 1859, No. 84 parcel in An Thành village was owned by a man named Đỗ Ngọc Sô 杜玉芻 and No. 94 parcel by a man named Đỗ Ngọc Tại 杜玉在. However, both of these parcels were sold to Nguyễn Tất Vãn 阮必挽 by a woman named Đỗ Thị Lý 杜氏𥚃 on the same date of the seventh lunar month in 1863. This transaction is a fairly peculiar case because, judging from their family name, it seems that the woman named Đỗ Thị Lý inherited these parcels from two men of the Đỗ Ngọc lineage. With reference to An Thanh A, it seems that these parcels were passed down the Đỗ Ngọc lineage for generations as they were purchased by two men named Đỗ Ngọc Thiện 杜玉善 and Đỗ Ngọc Can 杜玉乾 before the eighteenth century. This transaction should be considered not as a sale of Đỗ Thị Lý’s personal property, but as a sale of some of the assets of the Đỗ Ngọc lineage on behalf of Đỗ Thị Lý.

There are not many women-only transactions based on the examples of land transactions involving women. In many cases, land purchased in the name of the husband was sold by his wife after his death, or purchased in the name of the mother instead of her minor child. In both cases, the seller’s name was often a joint name with a son or grandson rather than a woman’s alone. Basically, when a woman became the entity of a land transaction, there were some circumstances in which a man could not become the entity of the transaction. However, such a situation was temporary or transitional, and it disappeared when the male heir became an adult. It is presumed that women’s ownership of a house and joint ownership by a woman and a man in the garden cadastre are the results of the registration of such transitional states.

Conclusion

In this article, we examined the family structure based on historical materials called viện bộ (garden cadastre) of Thanh Phước and An Thành villages in the northern suburbs of Huế. From those examinations, the general family structure in the rural areas around Huế is inferred to have been a two- or three-generation patrilineal family. In family genealogies, it is not uncommon for one man to have a wife and mistresses along with ten or more children; however, in this case they did not all live under one roof. One of the reasons for this is that sons other than the eldest one were separated from the father’s household during their youth due to marriage or other reasons, and they established independent households. Real estate was inherited by sons—with a slight preference given to the eldest son, who inherited ancestral rituals of the lineage; females were not granted the right to inherit their parents’ estate.

Although it is impossible to say with certainty since the analysis in this article is based only on residential areas and does not include other forms of real estate such as rice fields, Thanh Phước and An Thành villagers were patrilineal kinship groups under the influence of Confucianism.³⁶⁾ However, there remain aspects that cannot be fully explained by Confucianism alone. For example, in Thanh Phước and An Thành villages, concubines and mistresses formed separate households from that of the husband and his wife. It was often the case in the Confucian cultural area that after the death of the husband, his widow became the head of the household because her sons were minors (Washio 2018, 380–382). Especially in the Ming and Qing periods in China, a widow was required to remain chaste and loyal to her husband even after his death, and widow remarriage was severely criticized.³⁷⁾ It is conceivable that Thanh Phước and An Thành villages also had such widowed households. In addition to such households, concubines and mistresses formed separate households with their own children from the husband’s household, and as a result the husband visited his concubine or mistress rather than living with her. In other words, it is assumed that the situation was a “duolocal marriage.” If we understand that this situation was a remnant of the matrilineal family structure, as Luong Van Hy (1989) insists, then it makes sense that there were many matrilocal residence marriages in Thanh Phước and An Thành villages.

However, we should be careful not to overemphasize the high social status of women in eighteenth- and nineteenth-century Vietnam. An inspection of the garden cadastre shows that women were not granted the right to inherit their parents’ property. Furthermore, land transactions by women were conducted mostly when their husbands had died and their sons were still minors. In other words, the status of women as “female heads of household” was basically only transitional, until their sons came of age, and females’ property was eventually inherited by the male descendants along the paternal bloodline. If we focus only on the household of the concubine and mistress, it is certainly possible to consider it as a “matrilocal marriage.” However, considering the inheritance of property and ancestral rituals, the household clearly belonged to the husband’s patrilineal kinship group.

This article only examines the family structure of two villages in the suburbs of Huế. Even including the case study of the former Hà Tây Province by the author (Ueda 2021b), it must be said that there are too few cases to reveal the overview of the Kinh family structure and patrilineal kinship groups in the premodern period. More examples will have to be developed in the future. At the very least, what can be said from some case studies is that Vietnamese gender history, for example women’s social status and property inheritance (Miyazawa 2016; Tran 2018), is clearly inadequate to elucidate household composition and household division within kinship groups, which are prerequisites for Miyazawa’s (2016) and Tran’s (2018) analyses. Microanalytical subjects such as household and marriage have not been adequately studied due to archival constraints. However, as seen in this and the author’s previous article (Ueda 2021b), it is now possible to examine the eighteenth and nineteenth centuries by combining the land registries of the Nguyễn Dynasty period and village documents. Newly discovered village documents will bring about new possibilities and enable the study of new aspects of Vietnamese history.

Accepted: September 1, 2022

Acknowledgments

The author is grateful for fiscal support from the Japan Society for the Promotion of Science (JSPS, Research project number: 21H00577) and the National Foundation for Science and Technology Development, Vietnam (NAFOSTED, Research project number: 601.99–2020.302).