On April 1, the Hague Convention on the Civil Aspects of International Child Abduction went into force in Japan, as did the necessary implementing legislation. Having already written about this legislation in a prior column, I won’t revisit the subject here.
While Japan’s accession is a welcome step forward, I wonder how it will actually pan out. Implementation may be frustrated by some basic features of Japanese family law that can be neatly described mathematically: 1+1=1.
I would love to take credit for this wonderfully expressive formula, but it comes from a Japanese lawyer I met a few months ago, who used it to explain his own doubts about the convention’s viability here: “To Westerners, marriage means 1+1=2. But in Japan it equals 1.” It made perfect sense to me, but perhaps I should explain.
The traditional Japanese family structure was the ie, or house. This was an extended family that might encompass four generations of kin and several married couples sharing the same surname and even residence. The head of the house — typically the eldest male — had great authority over other members. Junior members needed his consent to marry or establish their own households. If the house was an economic entity — a farm, shop or other family enterprise — the head would also manage the business.
Primogeniture was the rule, with the eldest (legitimate) son inheriting the house, its property and the status of head. Marriage, childbirth and adoption were the means by which the house was perpetuated. While these were events that involved individual members of the house, they were ancillary to the greater collective: One plus one plus any number of additional ones still equaled one — a single ie.
The ie system was embodied in the Civil Code adopted at the end of the 19th century. It was feudalistic and discriminatory: Children born out of wedlock and women were disfavored, as were younger brothers. The head of house was a legal, heritable status that entailed both power and responsibilities, including a duty to support members of the household as well as control of family property.
The ie system was also a convenient tool for identifying, controlling and governing the Japanese people, which brings us to the koseki (family register) system. By requiring each ie to be registered, the government could implement policy through its head. The house rather than the individual was the smallest social unit directly subject to governance, with the head of the house being formally responsible for paying taxes, facilitating conscription and implementing other government programs.
The koseki system gave a weak central government leverage to govern nationally. The trade-off was noninterference within the family, with the head of the house accorded broad autonomy to govern it as he saw fit. This might include both corporal punishment and the use of junior members as a form of economic asset.
The register was also significant for commerce: It would show who was authorized to dispose of family assets, as well as an individual’s status within a particular house. This information could have financial significance. For example, an eldest son would be a lower credit risk since he could be expected to inherit the family property. The family register became a public document, a state of affairs that continued into the 1970s.
Elements of the ie system were fundamentally inconsistent with the postwar Constitution, which contains both a general egalitarian mandate and a clause specifically requiring gender equality and respect for the individual in family law. Amendments to the Civil Code and the koseki system were unavoidable.
The Americans governing occupied Japan had officially indicated that the Japanese were free to make such amendments to these laws as they thought appropriate so long as basic constitutional requirements were satisfied, which meant gender equality, marriage based on free will and the elimination of head-of-house status. This was not a clear mandate to completely excise the ie system from the Civil Code. Nevertheless, the Japanese drafting team apparently decided that such an approach would likely be viewed most favorably by the Americans.
However, what the Japanese did try to do was surreptitiously preserve elements of the ie system in case there was a desire to revive it after the Occupation (the subject was debated in the 1950s but nothing came of it). Their attempts focused on the Family Register Act rather than the Civil Code. The former being a mostly administrative statute that implemented the latter, the Japanese calculated that the Americans would stop paying attention once they were happy with the amendments to the Civil Code. They were wrong: Having been pleased to see the ie system formally excised from the Civil Code, the American authorities both noticed and objected to efforts to preserve it in the family register system.
In light of the individualistic principles of the new Constitution, the Americans actually advocated introducing a system that registered individuals rather than families. Using classic bureaucratic arguments (including insufficient paper!), the Japanese side held out for a family-based system. The Americans conceded but adamantly opposed any system that would enable three or more generations to appear in the same register, as this would have smacked of the ie system.
The result is the current koseki system, a compromise based not on individuals but on nuclear families: married couples and their children sharing the same surname. If a Japanese man and woman marry, they must establish a new register: 1+1=1. If they have a child it appears in that register: 1+1+1=1. If an unmarried Japanese woman has a child, she must establish a new register: 1+1=1. The same applies if she marries a foreigner and has a child (in family register math, non-Japanese equal zero). Japanese persons appearing in the same register are supposed to share the same surname; it’s part of the equation. If a woman gets divorced, she can revert to her parents’ register, but only if no children are involved.
The resulting system retains anachronisms that continue into the 21st century. Being rooted in marriage and surnames, it discriminates between children depending on whether they were born in or out of wedlock. Furthermore, since one purpose of the system is to identify family relationships so that government agencies and others can confirm who is responsible for whom, it is designed to minimize ambiguity. Parental authority over children is tightly linked to this system, with the Civil Code vesting it in mothers of children born out of wedlock, jointly in both parents during marriage, and in only one parent after divorce. Under this system the locus of parental authority should always be clear from the family registry.
The system is also laissez-faire. Most changes made to the family register are consensual and can be carried out with limited government interference — adoptions, dissolutions of adoptions, even many divorces. So long as a form indicating compliance is submitted, the authorities will accept it and the koseki will be amended to reflect the new status. About 90 percent of divorces, including many involving children, are made in this way, with no governmental oversight of custody arrangements. Even when parties can’t agree, the primary role of courts in the minority of cases in which they become involved is to convince parties to reach some consensual solution rather than to find facts or apply law. Since most resolutions are agreed to by the parties themselves (with or without court intervention), enforcement is something of an afterthought.
A Japanese lawyer recently related to me a consultation he had had with a young foreign man who had been living in his home country with his Japanese wife and their child. The wife took the child back to Japan for the summer and asked him to sign what she said was a school application. He could not read Japanese but signed. His wife and child never returned. When he tracked them down here, he discovered that he had signed a consensual divorce form awarding his (ex-)wife parental authority. Not only that, but his wife had since remarried and her new husband had adopted the child as his own (a common practice in Japan; otherwise the horrible anomaly of 1+1+1=2 — a man not sharing the same surname as or having parental authority over a child in his home — might arise).
This was essentially the same scenario that the other lawyer had enlightened me with using his formula, but it seems to be one that nobody implementing the Hague Convention here seems to have considered. Fundamentally, the Convention treats the parents and child as individuals, while Japanese family law still treats the family as a single unit and doesn’t handle fractions very well.
There are thus some very large gaps to bridge between the convention on the one hand and the Civil Code-family register combination on the other. The former does not distinguish based on the marital status of parents, their nationality or that of the child and is concerned primarily with the best interests of children. The latter is fundamentally rooted in marriage, the marital status of parents and Japanese nationality, all as they are reflected in the family register system. Furthermore, formal family law is largely unconcerned with the best interests of children because that would involve treating them as whole numbers and make consensual resolutions harder to achieve.
Finally, the convention is concerned with place: the child’s habitual residence. The family register system is not: Japanese nationals can register divorces, marriages or even adoptions (of other Japanese people) in their family register from abroad. Whether these transactions are valid in their country of residence is questionable, but the possibility of conflict seems obvious.
With the convention routinely and incorrectly described in Japan as being about “international divorces,” the process of bridging these gaps probably has a way to go. Certainly nobody here seems to have anticipated the possibility that cases seeking return under the Convention might involve families comprised entirely of Japanese nationals.
Of course, case resolutions will be where the rubber meets the road. The government hopes that most cases can be resolved amicably through mediation, whether through the courts or other organizations. (Full disclosure: I am registered as a mediator/arbitrator candidate with the Osaka Bar Association’s dispute resolution center.) Paradoxically, however, it is mediation and a focus on consensual results that contributed to Japan being an abduction haven in the first place — by allowing courts to remain involved while doing nothing affirmative for long enough that the child was settled in his or her new environment.
If mediation and the judicial process just end up being part of an exercise in “convincing foreign father to let the kids stay in Japan” (as one lawyer explained the need for mediation in such cases to me), then the whole process of Japan joining the dozens of other countries already party to the Hague Convention might add nothing to Japan’s family law equation.
Colin P. A. Jones is a professor at Doshisha Law School in Kyoto. Law of the Land appears on the third Thursday of the month. Comments and ideas: firstname.lastname@example.org