Family succession through adoption in the Chosun Dynasty

Abstract

Using a unique data source of genealogies of upper-status families, called Bulcheonwye families, we assess how the extent of family succession through adoption changed over five centuries from 1450 to 1949 in Korea. Our analysis shows the continued increase in the share of adopted sons among total family successors up to the end of the 19th century when three out of ten family successors were adopted. The trend of the increasing role of adoption is closely related to the declining number of sons per family, suggesting that not only the rising influence of Confucian culture but also demographic changes increased the demand for adoption. Finally, our comparison provides evidence that the likelihood of achieving high social status was significantly higher among adopted sons than biological ones, suggesting that the socioeconomic potential of adopted sons could be an important factor for adoption decision.

Keywords:

• Family Succession
• Adoption
• Fertility
• Bulcheonwye genealogies
• Chosun Dynasty

Acknowledgements

We thank Ilgon Ryu for his archival work to locate Bulcheonwye genealogies. Mathew Creighton provided excellent research assistance.

Notes

1 However, also note that according to a study by Wolf and Huang (1980), diverse forms of adoption including adoption of sons-in-law and adoption of daughters-in-law existed during the last 19th century to the early 20th century in China.

2 Some researchers use Chosŏn, Choson, or Joseon instead of Chosun. We use Chosun throughout the paper.

3 Note that the dominant practice in Korea was to adopt a male adult, while in China adoption of an infant or a young kid was not uncommon (Feng & Lee, 1998).

4 For instance, the genealogy of Hapche’on Lee clan, which has been widely studied by historians, was first published in 1761 and was not updated until 1843 (Son, 2006). Although the third and fourth editions were updated with shorter intervals, it is notable that it took 82years for the second edition to be published.

5 We use “reproduction” rather than “fertility” because with our data source, genealogy, we do not know the exact number of children ever born but most likely the number of children who survived enough to be recorded in genealogy. See 4. Trends in adoption for a more detailed discussion about this difference.

6 For a small number of cases for which birth years were not available, we could impute those using information on father's birth year. Because we group individuals into 50-year periods, our imputation based on father's birth year does not cause a problem.

7 Among a total of 1293 family successors, 17 individuals were not the eldest sons. In the early to mid Chosen before the family continuation through the eldest son was firmly institutionalized, the family line could be carried over through the line of the second son or even the youngest son when the eldest son did not have an heir and did not adopt a son for family succession. Indeed, all of our 17 successors who were not the eldest son lived in 15th–17th centuries. We included those 17 cases into the category of “non-adopted.”

8 One may ask a question whether the declining number of sons is an artifact caused by change in the way in which sons were recorded in our Bulcheonwye genealogies rather than reflects real demographic change. We are not aware of any studies that suggest some significant changes over time in the pattern of recording sons. It is difficult to imagine that somehow sons were increasingly omitted from genealogies. Some studies examined population changes throughout the mid-to-late Chosun and they showed a trend that is somewhat inconsistent with the continued decline in the number of sons available shown in Fig. 1 (Kwon & Shin, 1977; Pang, 1975). However, due to the nature and unreliability of data used to examine population trends, it is hard to derive a consistent conclusion on the population trend during this period.

9 Among the total of 211 individuals with high status, 48 passed a sate exam and obtained a government office job, 5 passed a state exam but did not have a government office job, and the remaining 163 obtained a government job without passing a state exam.

10 For the analysis, we include 62 cases for which birth years were not available or were before 1450. Therefore, the total sample size for the analysis is 1462.

11 Because our genealogies provide status information only of adopting fathers (not biological fathers) for adopted sons, “fathers” here mean biological fathers for biological sons but adopting fathers for adopted sons.