Gary Yia Lee | Culture | Household and Marriage

Household and Marriage in a Thai Highland Society - Marriage and Residence Rules Among the

Hmong in Thailand

Originally published in the Journal of the Siam Society, 1988, no. 76, pp. 162 – 173

Contents

Acknowledgements

This paper resulted from research carried out in Thailand with financial support from the Wenner-Gren Foundation, New York. I greatly benefited from comments by Prof. W. R. Geddes, University of Sydney, Australia.

The Hmong have been referred to as semi-nomadic people or "migrants of the mountains" (Geddes, 1976), because they move houses or sometimes entire villages every ten to twenty years after stable residence in a particular place. The reasons for such migrations are often complex, ranging from lack of farming lands to social -economic factors such as the desire to join relatives in another settlement or the fear of sickness and diseases.

Residential Patterns

A few writers have suggested that the ideal Hmong settlement seems to be one consisting of members of one clan (Cooper,1984). However, this is not the case, even with the smaller villages of only a few households. Each village is usually dominated by a particular clan, but no village is completely occupied by any one clan. The one-clan village is to be expected if we follow the Hmong’s rule of patrilocal residence to the limit, since all newly married or couples with young families are supposed to live in the same household or settlement as the man’s father or male relatives. However, this rule is not always strictly enforced so that bilocal or neolocal residence is also prevalent with the result that members of a few clans are found living in one settlement.

In some cases, residential preference is decided by social bond with members of the clans from which the wives originate. Similar religious practice, personal conflicts with one’s own agnates, or business commitments may make it necessary for some Hmong to live in villages dominated by other clans. Residential patterns are, thus, determined not only by the search for new agricultural prospects but also by social- economic factors such as social members of their own clan or kinship groups, but will adopt matrilocal or neolocal residence when in conflict with their own kinsmen.

In their traditional hill environment, the Hmong tend to live in small groups of five to twenty households, and are rarely found in "big agglomerations" (Savina, 1925: 182). Houses are built in a random fashion on a village site without any sense of direction or order. There is no village square, no main street. There may be fence here and there around the village, but this is only to protect a garden or crops from village animals and does not serve as a defence barrier for the settlement. Mickey (1947) also does not mention any fortified walls around the Miao village she studied in Kweichow, southern China. Graham (1937: 22), on the other hand, says that "there were formerly many fortified places… where groups of houses were clustered together for protection". This is further confirmed by Lombard-Salmon (1972: 118) in her historical study of the Miao in Kweichow in the 18th century. Today, however, and as one French writer puts it, only scrub forms a natural enclosure for a Hmong village (Anonymous, 1952: 31). A Hmong settlement is a beautiful sight from the distance, but the houses often look like old ruins close- up.

What is striking about Hmong houses is that they are built at random, deliberately to conform to the Hmong beliefs in geomancy and supernatural forces. No two houses are in line or parallel to one another, even when all the buildings have their fronts facing the downward slope. Unlike many Green Hmong whose houses have only one door, the White Hmong houses have two doors and sometimes even three, depending on the people’s need (Chindarsi, 1976: 15; Geddes, 1976: 39; and Lam Tam, 1974: 60). They are variably constructed on earthen floors compressed using water from hewn boards to mud, depending on what is available. Some houses have timber shingles or thatches as roofing covers, and even tiles or corrugated iron for the better – off families.

Nearly all the Hmong houses are rectangular in shape, although some look almost square. There is always a veranda running part of the length of the house, generally used to store firewood. The side door is on the left of the house when looking from the front where another door also exists, although the latter is more often used for the performance of rituals. The door on the front side of the house is called "qhov rooj tag" (khor daung ta), and the door on the left side is referred to as "qhov rooj txuas" (khor daung txua). Which door is used for entering and leaving the house depends on how conveniently it is situated, as there is no sanction against using either door.

A typical house measures about 9 x 7 metres with the walls reaching 1.6 metres in height, and with the centre of the roof about 4 metres from the ground. Doors are usually 1.8 metres high. Bedrooms often have only door frames no more than half a metre wide, usually no with doors. The bedrooms are located alongside the front of the house, with the beds consisting of raised wooden boards about 0.5 metre above ground. Parents and young children sleep together in the big bedroom while adolescent and grown-up daughters share the second bedroom, and older sons or unmarried male family members occupy a third bedroom further away. A spare raised timber flooring serves as bed for guests and is sometimes found at one end of the house, near the granaries.

Rapoport (1969: 129) suggests that human settlements and house forms are primarily the physical expression of a people’s culture and way of life, even though climatic conditions and building materials or technology may influence where and how a house is constructed. House forms and location cannot be understood in terms of their descriptive value only, but must be related to the shared goals and life values or beliefs of the people living in them. To the Hmong, a house is not only a shelter, but also a place of worship where one’s ancestral cults are observed and protected from outside influences. It is the sanctuary which unites members of a household into an extended family and later confronts all married male members to form their own separate dwellings.

We will now turn to this process of Hmong household formation and dispersion through marriage and procreation.

Marriage and Family

As a rule, the Hmong observe clan exogamy in that two persons of the same clan or surname cannot marry each other. There are some exceptions as stated by de Beauclair (1970: 133) when "the partners do not descend from the same ancestor". This is true of the Vang clan in certain parts of Laos, and the Yang clan in West and Central Kweichow, China. Sometimes, individual deviation is found among clans who otherwise practice no clan endogamy. The violation of this general marriage rule seems to be rare, since it is not mentioned by Binney (1968), Geddes (1976) or Cooper (1976). Mickey (1947: 50) states that the Cowrie Shell Miao in China prohibit marriage between people "of exactly the same surname", while Graham (1937: 27) writes of the Chuan Miao that it is considered "a crime for two persons having the same family name to marry".

Given that the norm for marriage is clan exogamy, the onus is on Hmong young men to court only girls born into clans other than their own. For some groups of Hmong in China (where they are generally called Miao), " patrilineal cross-cousin marriage is said to be obligatory" (Chen and Wu, 1942: 20). Among the Magpie Hmong of southern Szechuan, cross-cousin marriage is favoured but is "by no means obligatory, and sex relations between unmarried cross-cousins are freely permitted" (Ruey, 1960: 146). Formerly, a young man could marry his father’s sister’s daughter as a matter of course; and if this right was not exercised the girl’s father must pay the boy’s parents for the right to marry her to another man. This seems true also of the Chuan Miao (Graham, op. cit.: 27). Such a practice, however, does not occur among the Hmong in Laos and Thailand.

In the past, marriage by capture was allowed and so was the betrothal of small children by parents who are friends or relatives by affinal ties, especially in the case of a brother’s son and sister’s daughter. However, such marriages are becoming the exception since nearly all parents have now begun to take more account of their children’s wishes rather than their own. It is today left to the sons and daughters to choose their own marriage partners so long as the latter are of acceptable personal and social standing. Parents will interfere mainly when a son or daughter decides to marry someone who is considered a bad risk such as an opium addict, a person of loose character or lazy disposition, a married man, a widowed or divorced woman, a spinster or a man whose male relatives have a reputation of using violence on their wives.

The most acceptable marriage process for a Hmong man begins with the courting of a girl, preferably one with an industrious nature. This must be done in the least conspicuous way, particularly towards the girl’s relatives. This means that young people are free to meet or court the opposite sex mainly in the evenings in the dark of the night after each day’s work. The procedure seems to vary from one region to another, and to some extent from one generation to the next. In the old days when many Hmong lived in isolated pioneering small villages, a young man would have to travel a few hours each evening by himself or with some friends before reaching the girl’s hamlet.

From time to time, such a group might take enough food supplies with them to stay for a few days with relatives in the girl’s village or in the bush to court her at night. Courting also took place when boys and girls spent time working in the neighbouring fields away from home when it was inconvenient to return to their own villages during intensive farming periods. Much of this tradition is still carried on by the present generation, except that young men today do not travel a long way each evening to see their girlfriends, thus having to return home the next morning for another day’s work on the family farms. Of course, some may say court in their own village individually or as a group.

Once the courting has been done long enough (from a few days to a few months as the case may be), the girl may agree to marry the young man who will then have to ask permission for the marriage from his parents. This is necessary because the parents or guardians have to help pay part or all of the bride- price and wedding costs. If the girl consents to the marriage, her parent’s permission does not have to be obtained beforehand. A mediator is used to negotiate with the girl’s parents only when she herself has not agreed to the marriage or when the prospective couple do not know each other well enough. Today, parents are reluctant to force their daughters to marry and will try first to persuade them to agree to a marriage, because the parents wish to avoid being blamed in case the marriage prove unsuccessful. The groom and his relatives are also apt to treat the bride and her parents with respect if the latter do not consent to her marriage too readily.

If a girl is willing to marry, the man will take her to his home quietly, then send a messenger to inform her parents. If the man does not live too far away, the girl’s mother may go there to claim her back and may even use violence on her and her intended husband to show her displeasure. This verbal and physical abuse has to be accepted without retaliation, and has to be manifested even when the girl’s parents secretly approve of her match in order to demonstrate their reluctance to hand over their daughter so that her husband will take better care of her, knowing how highly they valued her. The man and his relatives will, on their part, lavish verbal promises or money gifts on the mother and in the end, she will return home without her daughter to await the day when the marriage will be celebrated.

Another variation of this procedure is for the man and a handful of male relatives to "abduct" the girl at a pre- arranged place, often with her full knowledge and consent. She will then scream for help, and her mother will come to her rescue, again full of verbal abuse and brandishing a stick. If the daughter indicates that she is unwilling to be carried off for marriage, the mother will rain blows upon her abductors and ask her for release. On the other hand, if the girl shows willingness to go with the men, the mother’s blows will be on her for being too eager to get married.

It should be noted that at this stage of the marriage process, no male relatives of the girl are involved in her so- called rescue from her husband-to-be and his helpers. They have no roles to play until the wedding when they take full charge of all negotiations and tasks related to it. Abduction is still deemed preferable to elopement, even when the girl has no objection to marrying her boyfriend, because elopement is worthy only of those girls without self- esteem or respect for their family members. Abduction is also regarded as a face- saving protection for both the girl and her family should the marriage fail, as she will then be able to say that she was uncertain about the prospect all along and her family, too, can claim that they were not responsible for the failure.

Once the "abduction" or elopement has occurred, the young man’s parents send a message to the girl’s relatives asking for a convenient date to celebrate the wedding. The wedding itself is a costly procedure, consisting of: (a) the bride price (b) fines (c) miscellaneous expenses for pigs, food and alcohol. Thus the Hmong wedding is never less than $500 (US) in overall expenses, and can run up to $1500 (US) in the case of the richer or more difficult families. For this reason, few young Hmong are able to pay for their wedding immediately. Sometimes, the date set by the bride’s parents may not be convenient to the groom if he and his relatives have not found enough money on time to cover the wedding expenses, and a new date may have to be arranged.

The wedding ceremony is too elaborate to describe in detail here. It is sufficient to say that it comprises one to two days of negotiations and feasting, firstly at the groom’s house, then at the bride’s house and again back to the groom’s residence. At least one medium- sized pig and four chickens are slaughtered for meals and rituals at the groom’s house, and two large pigs as well as two chickens for the bride’s relatives. Many gallons of rice alcohol are also needed. The bride’s parents on their part have to kill a large pig to feed those who helped them during the wedding feast either with negotiations or other duties. Each of those helpers, both on the groom’s side and on the side of the bride’s relative, has to be paid in silver coins. These payments and the cost of food items can amount to $150 to $200 which may have to be outlaid in cash to buy them if the two parties do not have them. The bride- price, which some Hmong see as the nurturing charge, ranges from $400 to $800. The fines paid to make up for the past grievances suffered by members of the girl’s clan in the hands of the groom’s clansmen may claim from $5 to $100, depending on how serious or how many are these wrong- doings.

Of the six marriages studied in Khun Wang, Thailand, in 1977 by Lee (1981: 46), one was finalised in the sense that the full bride- price was paid at a wedding ceremony, which took place within a month of the bride going to live with the groom. Two other wedding ceremonies were carried out, but these were for marriages contracted two to three years earlier. The remaining three couples only had a small preliminary ceremony with the bride’s parents at which an agreement was made to postpone the wedding either indefinitely or to a later date. Postponement usually means that the full wedding will never take place and the bride- price never be fully paid.

There is no sanction against such couples living together as husband and wife, or raising a family in a similar fashion to those who have gone through the prescribed ceremonies. The reason for not formalising the marriage may be that the husband and his relatives were too poor to afford it at the time, but often no pressure is put on them later by the wife’s parents, or her relatives after they are dead. The failure to formalise marriage, however, has certain social implications for the husband. He has no claim on his unpaid wife’s children should she divorce him, nor can his relatives keep her and her children in case of her being widowed and remarrying outside the husband’s kin group.

There are other reasons apart from poverty why bride- price may not have been paid. The match may be disapproved by the boy’s parents and the young couples are left to their own devices. In such a situation, the groom may have to do service by staying with his wife’s family and working for them until he can afford the wedding costs. In normal circumstances, however, a couple usually depends on the man’s parents to pay for both the bride- price and the wedding expenses, they will accept this as a project for their household, provided: (a) the prospective bride has approved the marriage (b) The groom is a loyal and productive member of the household, and (c) the amount of the bride- wealth required by the bride’ s family is considered reasonable (Barney, 1970: 157).

When the man’s parents have paid his bride- price, the couple will live in his father’s house to profit from his advice regarding marriage problems, to show their gratitude by staying with him and helping with all household tasks (Chindarsi, op. cit.: 77). They remain there until they can set up a house for themselves separately, often when they have two or three children. Bernatzik (1970: 43) states that this occurs when the husband is thirty years of age. However, this is not true in all cases. Segmentation from the parental household can take place any time, depending on how crowded the father’ s house is and whether there are married sons living in it.

It is not easy to determine age correctly as the Hmong do not, as a rule, keep written records of their vital statistics. Only declared statements can be obtained, usually in connection with certain important personal or national events used as chronological landmarks. This is no longer true of the younger generations, of course, because parents are now compelled by law to register the births of their children or deaths with local authorities of the countries where they now live.

Based on such time reckoning, Lee (op. cit.: 48) discovered that of the ninety-three married persons he surveyed, seventy-five persons or 80.6 per cent married between the ages of fifteen and twenty-one. Only four persons or 4.3 per cent were twenty-six years or older when they first married while for the remainder the age of marriage could be as early as thirteen and as late as twenty-five. Women tend to marry at an earlier age than men, either through their own will or by obligation such as coercion and capture. This is not to suggest that all married women are younger than their husbands since the age gap between some couples can be as much as twenty-two years, the older being usually the husbands who married more than one wife or who remarried after divorce or widowhood. Geddes (op. cit.: 80-81) says that among the Green Hmong " boys usually married when fifteen or sixteen years old and girls at about the age of twenty years". According to him, this is because a Hmong man often marries more than one wife; and the older the first wife, the more respect she is likely to receive from her husband and his other later younger wives. Lee (op. cit.: 49) finds that this generalisation does not apply to the White Hmong he surveyed at Kun Wang in Chiangmai, Thailand. There are only a few couples with the wife older than the husband, but this situation has arisen from factors other than deliberate design to conform with any polygynous pattern. A close analysis of other Hmong settlements will probably reveal that as a rule the husbands are older than their wives when they marry for the first time.

Of the four weddings witnessed by Mickey (op. cit.: 50) in Kweichow, the ages of the couples were: boy thirteen, girl twelve; boy fourteen, girl thirteen; boy sixteen, girl twenty; boy and girl both eighteen. In each case, betrothal was initiated some years before marriage. According to Graham (op. cit.: 35), the Hmong of Szechuan often marry at an early age through arrangements made by their parents so that "most children are married by the time they are twelve".

With respect to the Hmong of Thailand, it is worth quoting Chindarsi (op. cit.: 71-72) when he writes that; "The Hmong marry between the age of fifteen and eighteen years. One of the factors contributing to such an early marriage is the need for (more) people to work in the fields and in the house…. The father of a girl sees that there will not be too wide a gap between the ages of his daughter and prospective son-in-law. If the girl is older than the boy, he may marry again, hence (girls) avoid such marriages".

It does appear that the Hmong exhibit ambiguity about polygamy (a man marrying more than one wife). On the one hand, some men approve of it if the husband is rich and the wives can get along with one another "like sisters" so that hardly any conflicts exist between them (which is rare in real life). However, in the case of poor men, its practice is often frowned upon, particularly by girls of marriageable age and their mothers. On the whole, it is tolerated if it is done through family obligations as in the case of the levirate (when a widowed woman is taken as an additional wife by the younger brother of the woman’s dead husband to take care of her and her children and to keep the latter within the family). Polygyny may also occur in cases where the first wife is barren or unable to perform her duties effectively for various reasons.

Thus, polygyny does not seem to be a factor causing the age difference between married partners. Generally, husbands are older than their wives. Only in a minority of cases is the first wife older than her husband (such as when a young son is persuaded to marry an older wife because she is industrious or of good character and his parents do not want to lose her to another man). This is, however, becoming rare, probably because marriage today is the result of romantic courtship rather than parental arrangements as in the old days. Age is not a major concern so long as both the bride and groom are deemed ready for matrimony and socially suitable to each other.

Another contributing factor to the age gap in Hmong marriages is the requirements for a substantial bride-price which often delays the age of marriage in times of economic decline, particularly for the men (Kunstadter, 1983: 35 and 39). Most young men will wait until their parents have at least some savings to pay for part of the bride-price and wedding costs. This delay can be averted if the groom agrees to do service by joining the bride’s family and work off the bride-price, or by incurring long term debts with the bride’s parents and remaining with his own relatives.

Today, some Hmong parents will not ask for a bride-price, because they do not want to be seen as "selling" their daughters off. Strictly speaking, the so-called bride-price is really a levy or a "nurturing charge" (nqi mis nqi hno – literally, a fee for mother’s milk) from the groom to the bride’s parents in recognition for the many years of love and care they have given her only to see her go to make a life for herself with him. This "nurturing gift" is found in many other Asian societies, including the Thai and Lao, among whom it is known as "sinsot". Anthropologists have incorrectly referred to this as bride-price for want of a better term, from the Hmong word "nqi poj niam" which refers to the total costs of getting a wife, including wedding expenses and the "nqi mis nqi hno". The term "nqi taub hau" is sometimes used, but this is a term which should not strictly be used in regard to weddings as it mainly refers to compensation money for a death through someone else’s fault or negligence.

The costs of marriage aside, whether the man will later have other wives is for him and sometimes his first wife to decide, depending on the quality of their relationship and other socio-economic considerations. It may have been the case that in the past some Hmong parents might have persuaded one of their young sons to marry a much older girl so as to get her economic contributions to the household, knowing that he would likely later acquire another wife of his own or younger gage group. Such matchmaking, however, is no longer acceptable, as it today considered exploitative and too calculating for a happy marriage. It is now left to the young man to choose his own wife. There is usually no regular age discrepancy favouring polygyny. In general, the relative ages of married couples in Hmong society are similar to those of most other human societies.

The Household

As has been indicated, a young couple does not normally form their own household in the first few years of marriage. Because only the groom’s parents or guardian can have accumulated enough money to pay for the wedding costs, a newly married couple is expected to remain in the parental household to render service until such a time as they have two or three children when they may then move out on their own. By then, other sons in the household will have been married and have had children so that the household becomes too overcrowded for them to remain together. Moreover, friction between some of the wives or disagreement over the allocation of labour are likely to divide the household so that one or more couples will have to establish their separate living quarters.

The factors that lead to actual separation are not so much an ideology of neolocal residence, as strains that develop inside a large household. This applies equally to polygynous households which often become segmented once the family members have increased in number and the children of the older wives have grown up enough to be able to support themselves and their mothers. They will then build a house for themselves, usually not far from the father and his household if all is well between them. This pattern is found particularly with the White Hmong.

Lee (op. cit.: 54) find that fifteen or half of the thirty households in Khun Wang referred to previously are comprised of nuclear families while the remaining fifteen are extended families. This suggests that the Hmong appear to prefer these two types of households to those complicated by polygyny. This household pattern appears to be similar elsewhere in other Hmong settlements in Thailand. At Meto, Chiangmai, for instance, only twenty six of the sixty-five households have some form of polygyny, fifteen are polygynous simple households, five polygynous extended, four extended polygynous, and two polygynous extended polygynous (Geddes, op. cit.: 124). Mark (1967: 57) also notes that the nuclear family is the predominant household form with the Magpie Hmong in China; stem families are next in frequency; and five percent of the households have extended joint families. Binney (1968: 257, 269A and 273) finds that nineteen of the forty-nine White Hmong households in two different settlements in Chiangmai are of the nuclear type, twenty-six are extended, and only four are polygynous.

The number of persons in a household usually ranges from six to eight persons. This seems common among the Hmong, as the Meto figure for the Green Hmong is eight per household (Geddes, op. cit.: 110). In the three Green Hmong villages studied by Lemoine (1972: 39) in Laos, the average number of household inhabitants is 6.8, 7.9 and 8.1 respectively, giving an overall figure of 7.6.

Due to the prevalence of extended families in Hmong society, household size appears to be bigger than family size, a factor seen by Geddes (op. cit.: 128) and Kunstadter (op. cit.: 38) as important in the Hmong opium economy since the merging of manpower from two or more families into one single unit result in increased agricultural production and cash income. The figures also attest to most Hmong’s desire and ability to maintain cohesive social groupings at the household level. The only factors limiting the growth of household size are the lack of money to pay for additional wives to join the group, and the unsuccessful control by a household head over the disagreements or personal activities of married sons who may move out to live on their own.

It is obvious that by Western standards, the Hmong household is rather large, from six to eleven persons overall, and Keen (1978: 210) even found an average of fourteen in Tak province, Thailand. The largest households in Khun Wang have four to six families with between sixteen and twenty-three persons (Lee, op. cit.: 56). In the nearby settlement of Mae Wak, there is a household of forty-four persons and five families, probably one of the biggest in existence. However, on a family population basis, the average number of persons rages from 4.1 to 8 by clan with a total average of 4.9. This seems to suggest the Hmong family is not very different from that of many other groups of people. However, the household tends to be larger because of the Hmong extended family system.

Conclusion

Although some Hmong in Thailand as well as in China, Laos and Vietnam live close to or on lowlands, the majority of them live in mountainous areas, often at altitudes above 3,000 feet. This is where they have traditionally been located. Generally, they build their houses at the foot of a hill or on mountain slopes which have running water and fields for crop growing. The houses always face the downward slope with one door on the front and another on one of the side walls.

Almost all the houses are simply furnished, with only essential farm tools and household utensils maintained in different places. The buildings are not in line with one another bit because of lack of suitable terrain and deliberate design. The orientation of the house is influenced by the topography of its site as well as by religious and cultural determinants. This settlement pattern also reflects the need of clan affiliates or relatives to stay close to one another for mutual protection and assistance, although a few household heads achieve these objectives by settling among relatives of their wives.

Marriage is by clan exogamy and often takes place following a period of courtship. Some arranged marriages and a few unions by capture or coercion still occur, but the mutual consent of the principals is now more considered. Weddings are expensive, and many young men cannot afford them. Even among those who have married for many years, many them still have not formalised their marriage owing to the lack of resources to meet the wedding expenses and the bride-price.

The majority of first marriages occur among the sixteen to eighteen age group, although the age of marriage in some cases does vary between thirteen and thirty-five years. In general, the husbands are older than the wives. There are more households with nuclear families than other types, despite a prevalence of extended families. The average number of persons per household is 8.4 with an average of six persons per family. Notwithstanding the occurrence of some polygynous marriages, Hmong family and household size does not appear generally to be greatly different from that found in the majority of human societies.

References