Marriage and Divorce in the Edo Period

TAKAGI Tadashi

Under the Japanese legal system of the Edo period (1600–1868), a clear distinction existed between the provisions concerning family matters that applied to the samurai class and those that applied to commoners. In some respects the provisions were in fact completely different. One example is the case of gyakuenkon, marriage to the widow of one’s elder brother. For samurai this practice was strictly forbidden; an elder brother outranked his younger brothers, and by extension so did his widow. But for commoners the same practice was allowed—and it was even considered mandatory in some regions.

One sign of the common nature of this practice is that it even became the subject of a senryû, a form of satirical verse made up of 17 syllables, like haiku, but without the seasonal allusion required in the latter. These short verses became very popular from around the middle of the Edo period. The poem in question goes, “Goke no ki wa hotoke no otôto iya na koto.” Literally translated, it simply says, “The widow’s mind is averse to the younger brother of the deceased.” This refers indirectly to the fact that circumstances often forced a widow to marry her brother-in-law against her own will.

Given the sometimes diametrically opposite provisions of Edo-period marriage law for the samurai and common people, as seen from the example of gyakuenkon, both should be addressed as separate subjects. But here, given the constraints of space, I will focus on the economic aspects of marriage and divorce among commoners and will refer only to the dowries of samurai and briefly in passing to other topics relating to them.

One further introductory point: If a marriage continued until the death of one of the partners, then the subsequent handling of the couple’s money and property was not an issue between husband and wife (though there might naturally be inheritance issues between the surviving parent and the couple’s children). The only significant economic issue between husband and wife in the Edo period was that of the dowry, particularly in the case of divorce. In other words, the economics of Edo marriage are largely the economics of Edo divorce. This will thus be the main topic I address below.

MARRIAGE AND DOWRIES

In the case of marriages among the samurai, success or failure in concluding a union often depended on the amount of the dowry. (This was also true of adoptive marriages.) Particularly from the middle of the Edo period on, the practice of marrying with an eye to the bride’s dowry became widespread among members of the warrior class ranging from high-ranking daimyô to lowly retainers. Since the size of the dowry was the key determinant in arranging a marriage, the marriages of members of daimyô houses often depended on the finances of the domains they ruled. If the domain was in difficult financial straits, it would generally be impossible to come up with dowries sufficient to allow the daimyô’s daughters to marry into families of equal rank.

Few records remain concerning the handling of these dowries in the case of divorce, but we do have a small number of materials relating to divorces by low-ranking samurai. These indicate that in principle the dowry was supposed to be returned to the wife’s family. In practice, it had often already been spent, leaving no option but to pay it back by borrowing money. And if the divorced husband wanted to remarry without carrying the earlier debt over into the new marriage, he needed to find a bride whose dowry was greater than the amount he had borrowed. So it is no exaggeration to say that the amount of the dowry was the crucial factor in the selection of marriage partners among samurai families. Discussions of a possible union started out with negotiations over the dowry, and they generally proceeded with no reference to the wishes of the individuals to be married.

With regard to commoners, we find some receipts for dowries (or betrothal money) among the remaining documents from the Edo period. However, I have never encountered any historical materials suggesting that the amount was a key determinant of marriage. The reason some of the receipts remain is that they were needed in case of divorce. I will return to this point later.

DIVORCE IN LAW AND IN PRACTICE

Among the samurai, family matters were subject to control by the lord of the domain; marriages and divorces could take place only with his permission. Among commoners, by contrast, a man could issue a letter of divorce to his wife, and if the wife accepted this, the divorce would be complete. The common term for the Edo letter of divorce is mikudari-han, meaning “three lines and a half,” a reference to the short conventional form of these documents. The popular image is of men who could divorce their mates at will with a simple “rejection slip” and of tearful women returning to their parents’ home carrying one of these short letters of divorce. And it is true that under the legal code of the shogunate, the letter of divorce could only be issued by the husband to the wife, a fact that has led to the view that divorce was a unilateral male prerogative. This is related to the view of Edo-period Japan as a feudal society where women’s status was vastly inferior to men’s.

Is this widely held view of Edo society an accurate one? I assert that it is not, and I would suggest that the legal treatment of divorce among commoners offers clear support for this assertion. Under the law of the shogunate, it was provided that a woman who remarried without having received a letter of divorce was to be returned to her parents’ home with her head shaved. But I think that people considering the Edo period have paid too much attention to this penalty, while tending to overlook the penalty for a man who remarried without having given his former wife a divorce letter, namely, that he was to be banished. In this light, it seems that the husband’s “right” to issue a letter of divorce was actually more like an obligation. And since a man who attempted to remarry without some proof that he had issued a letter of divorce might be the target of an objection from his former wife, some men had their newly divorced wives provide them receipts (kaeri issatsu) for the divorce letter. Furthermore, there were some cases where a wife with a misbehaving husband would have him give her a letter of divorce that she could keep and use later if he misbehaved again. These were called sakiwatashi rienjô, or “advance letters of divorce,” and sometimes they were actually drawn up when a couple married and were given to the bride’s parents, containing a promise by the groom that, should he fail to provide adequately for their daughter, they could take her back and have her marry somebody else. To judge from arrangements like this, it would appear that divorces among commoners in this period generally involved discussions between the two families. In other words, they were what researchers now call jukudan rikon, or negotiated divorces, and it is inaccurate to think that husbands could simply send their wives packing whenever they wished.

Another point is that even if the wife was to blame for the failure of the marriage, the letter of divorce did not refer to her responsibility. This was because such a reference would have interfered with the smooth completion of the divorce process. In most cases the letter did not mention the cause for divorce, and even if it did, it was only in abstract terms. This reflects both a willingness to consider the woman’s position and the fact that divorce in Japan has since Edo times been taken to represent the irremediable collapse of the relationship. Regardless of cause, a marriage that had failed was settled through divorce.

Looking at typical letters of divorce from this period, we commonly find the phrase “warera katte ni tsuki,” or “of my own will.” What this is saying, in effect, is, “I am divorcing you at my convenience; you are not responsible.” Traditionally this phrase has been interpreted to indicate that the husband was divorcing his wife at his own whim, but this interpretation is incorrect.

The number of divorces has been increasing in Japan recently, but the rate of divorce was much higher back in the Edo period, both for samurai and for common people. According to a recent quantitative study of 100 daimyô houses and 100 families of direct retainers to the shôgun, their total divorce rate was 11.23% (of all marriages) and their remarriage rate was 58.65%. The divorce rate was higher for daimyô than for retainers, and the remarriage rate for the former was about 10 percentage points higher than for the latter. Among the daimyô, the highest divorce rate was seen among those whose domains’ rice production assessments were the lowest—less than 50,000 koku (a koku being roughly 180 liters)—and the divorce rate declined in inverse proportion to the koku level. Divorce was naturally less common among the lords of the largest (and wealthiest) domains because if one of these daimyô wanted to separate from his wife, he could move her to another residence and take a concubine instead.

Statistics like the above are based on official records submitted to the shogunate, but it seems likely that these records do not tell the full story. In principle divorce was frowned upon; it was certainly not encouraged. So there is a strong possibility that not all the cases were reported. In one specific case for which we have documentation, a person actually divorced five times, but only three of the divorces were officially reported. This suggests that divorce was even more common among the samurai than the historical records indicate.

We have no way of determining the divorce rate for commoners. As of 1883, the fifteenth year of the Meiji era, which was the first year for which national divorce statistics were recorded, the figure was 3.39 per 1,000 population (compared to 2.30 in 2002). This high level is said to have represented the legacy of Edo-period practices, lending credence to the idea that the divorce rate was relatively high for commoners as well. And we have documents for cases in which people divorced and remarried as many as five or six times. At least we can assert with confidence that divorce was more frequent then than now.

THE FINANCES OF DIVORCE

In the remainder of this essay I will concentrate on divorce cases among commoners, focusing particularly on the financial aspects.

One key principle of Edo-period divorces was what I call the “principle of the return of the dowry.” A wife’s assets can be considered to have fallen into five categories: (1) goods in the wife’s name, (2) dowry money, (3) dowry real estate, (4) real estate in the wife’s name, and (5) money in the wife’s name. The final two items would continue to belong to the wife regardless of a divorce. As for the first three, shogunate law stipulated that dowry money and real estate (such as fields and houses) belonged to the husband after the couple married, but the wife’s movable property, such as clothing and furniture, belonged to her and would not become her husband’s property through marriage.

Under shogunate law, it was stipulated that a husband could freely divorce his wife and that the divorce could not be challenged from the wife’s side, provided only that the husband returned the wife’s goods (movable property) and the dowry money and real estate. But this final proviso, which I call the principle of the return of the dowry, was in fact the most significant point of the divorce law. A senryû from the period highlights this requirement, speaking the words of a husband who starts to write a letter of divorce and then remembers that he has pawned some of his wife’s dowry goods: “Sarijô o kaku uchi shichi o uke ni yari,” meaning, “While writing the writ [of divorce], I sent for return [of my wife’s things] from the pawnshop.”

Documents from the period include references to a case where dowry property was lost not at the husband’s responsibility but because of a fire that spread from a neighbor’s house; upon divorcing his wife, the husband paid a substantial cash solatium (15 ryô) to compensate for this loss. We also find the case of a woman called Kiwa who remarried without having received a letter of divorce but even so had her dowry chest of 45 items returned to her. So it seems that the principle of return of dowry property was faithfully carried out.

A second key principle was that the party seeking the divorce had the duty to pay. When a husband divorced his wife, he was obligated to return the dowry. But if the wife “took her leave” (that is, sought divorce), the law provided that the dowry should be handled through an agreement reached between the two sides. So the dowry was not automatically returned in such a case; in practice, the wife would probably have to renounce her claim on it.

Since the husband had to return the dowry upon divorce, if he had already used it up and could not return it, then he could not divorce his wife. In this sense the dowry served as “collateral against divorce,” that is, it functioned to keep the husband from divorcing his wife arbitrarily. One senryû goes, “Akikaze o fusegu jisan no kinbyôbu.” This means, “Autumn winds, blocked by the gilded folding screen of the dowry”; here the gilded folding screen blocking the wind is being likened to the feeling of the husband who has grown to hate his wife but cannot divorce her because he has used up her dowry. Another one goes, “Sarieeru mono ka to, o-Kane nikui koto,” meaning, “‘How can I leave?’ the hateful o-Kane jeers.” Here we take it that the husband has told his wife, whose name is Kane, that he wants to divorce her, but she has shot back with the taunt that (having spent her dowry) he cannot make her go. The word okane also means “money”; to the husband, both the wife and the money she brought as her dowry have now become hateful.

In one recorded case, a maker of precious-metal lacquer for the shôgun sought a waiver of the return of a huge dowry (100 ryô). In another case, the husband provided farm fields in place of the dowry money; he had to go to this extreme in order to meet his obligation to return the dowry. In other words, the party seeking the divorce had to accept an economic disadvantage. The operative principle here, to reiterate, was the principle that the one seeking the divorce was obliged to pay.

Another senryû laments: “Hyaku no uchi no jûryô nuketa yome o tori / saru toki wa kyûjûryô de wa sumanu nari,” or “Ten of the hundred ryô were taken out when I took my bride, but ninety ryô will not suffice when she departs.” Apparently the matchmaker took a 10% commission out of the original dowry before the husband received it, but in order to divorce her, he must return the full amount. Incidentally, above I referred to dowry receipts; though some of these remain, we have not yet found any documentary evidence for this sort of matchmaker’s commission.

What if there had not been a dowry? In such a case, if the couple divorced, the partner seeking the divorce would pay some sort of solatium. Here again we see the working of the second principle that I cited above, namely, that the divorce-seeking party must pay.

Let us consider some cases in which the husband paid a solatium. Though Edo-period Japan is generally thought to have been a society of low status for women, it was not considered socially acceptable for a husband to decide on a new wife and divorce his present one (which shows that this was not in fact a society of low female status); such cases were referred to as fujitsu rien, or “bad-faith divorce.” In one surviving letter of divorce for such a case, we find that the husband starts the document with the declaration, “I apologize,” and he offers a solatium of two ryô and two bu. In another case, where the husband sought a divorce for the selfish reason that he no longer cared for his wife, he offered her a substantial settlement of 50 ryô.

A different sort of case is the 1859 divorce of one Kôno Gonbei (commoners did not ordinarily have family names; this was evidently a well-to-do farmer). His wife Tai was unfaithful, and she had borne him no children. They had been married 17 years, and Tai was 41. Upon divorcing her, he returned her substantial dowry property, which filled two great chests, two tansu (chests of drawers), two tsuridai (carrying racks), one utility cabinet, and one lacquer box. On top of that, despite her unfaithfulness, he paid her a solatium of five ryô. He was evidently eager to divorce her, either because he was uncommonly troubled by her marital transgressions or because he feared that word of them would spread, and so he paid her a settlement despite the fact that she was to blame.

Then there are the cases in which the wife paid the husband. A woman called Tsugi badly wanted to divorce her husband, but she had no money to pay him, so she took a job as a farm servant and used the advance she received for that work to give him one ryô and two bu as a divorce payment. Then there was Mise, who had been working as an indentured maid at a highway inn. A man had paid off her indenture and married her, but she decided that she did not care for him and wanted a divorce. Since the divorce was her idea, she needed to pay her husband. And the amount required was 22 ryô, this being what her husband had paid the inn for her freedom. She did her best to raise the money, but she could not come up with the full amount, and so she sold herself into indentured service at a different inn and used the proceeds to pay the divorce settlement. Both these women were extremely poor, but they got the divorces they wanted by going into service and paying off their husbands. This indicates how thoroughly the principle of payment by the divorce-seeking party was applied.

If the wife’s side were well off, the wife could secure a divorce either by waiving the return of the dowry or by paying a solatium. But this was of course financially disadvantageous. Hence the senryû, “Nigete kite tsumaru mono ka to, sato no haha,” meaning “‘It’s a waste to come running home,’ the bride’s mother tells her.” The mother, we can imagine, is telling her daughter that she is right to want to break up with that husband of hers, but that she should have him seek the divorce so that she can bring back her dowry or a solatium.

WAS EDO A “CONTRACT SOCIETY”?

Above I mentioned “advance letters of divorce.” What was the social context in which documents of this sort were transferred? If a couple remained happily married, then there would be no need to use the letter of divorce. But if the marriage failed because of financial hardship or other causes, the wife, having already received this document, could achieve a divorce simply by returning to her parents’ home; there was no need for messy talks between the two sides. In other words, people sought to avoid potential conflicts by reaching this sort of agreement in advance and putting it in writing.

Let me also refer to the cases of marriage by a man into a woman’s family. Up to now I have been discussing the conventional type of marriage by women into men’s families, which is thought to have accounted for 80% of the total. But the remaining 20% followed this reverse pattern, called yôshi engumi, or adoptive marriage, in which a couple married and the groom was adopted into the bride’s family, generally because the family had no sons (or only a young son or sons). In this case the adoption took legal precedence over the marriage, and the right to dissolve the relationship rested with the bride’s father. If he decided he did not care for his son-in-law, he could cut the adoptive tie, and this would automatically end the marriage, even if the couple were happy with each other. This arrangement, under which property is inherited on a matrilinear basis, is an unusual one; other than in Japan, it is found only in the Pyrenees region of France, parts of Scandinavia, and parts of India.

In such marriages, the groom came to live with the bride, who continued to live in the home where she grew up. Since the marriage could not be expected to succeed if the bride behaved selfishly under the protection of her doting parents, there was a need to place some sort of restraint on her whims. This led to the practice of warding off a breakup by deciding in advance on how the family assets and money would be divided in the case of a future conflict (divorce). A bond document would be drawn up starting with a declaration that all of the family’s assets, such as its house and fields, were to become the groom’s property, followed by the terms for division of property in case of a divorce sought by the son-in-law and in case of a divorce sought by the married daughter. Here again we see the “divorce-seeking party pays” principle at work. If the son-in-law was unfaithful to his wife, misbehaved, or asked for a divorce, he would forfeit his own dowry and depart, though in exceptional cases the dowry might be returned. If the bride’s side sought the divorce, they would provide a considerable share of their assets. In documented cases we find grooms receiving a third or a half of the family’s fields (or certain specific plots) and portions of its money.

These “divorce contracts” that were concluded in advance were intended to prevent potential future conflicts. This was a practice that grew out of the wisdom of common people accumulated over the ages. And when conflicts actually occurred, these documents offered an effective means of settling the matter.

Translated from an original article written for Japan Echo.

© 2003 Japan Echo Inc.