short answer(1 paragraph each)

Population Council and Wiley are collaborating with JSTOR to digitize, preserve and extend access to Population and Development Review.

http://www.jstor.org

Fertility Transition in Thailand: A Qualitative Analysis Author(s): John Knodel, Napaporn Havanon and Anthony Pramualratana Source: Population and Development Review, Vol. 10, No. 2 (Jun., 1984), pp. 297-328 Published by: Population Council Stable URL: http://www.jstor.org/stable/1973084 Accessed: 24-01-2016 21:40 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

Fertility Transition in Thailand: A Qualitative Analysis

John Knodel Napaporn Havanon Anthony Pramualratana

The transformation of reproductive behavior involved in the transition from high to low levels of fertility represents one of the most fundamental changes in human history. Recent Thai experience in this respect is particularly remarkable because it has been rapid and pervasive and has been occurring while the large majority of the population is still rural and agrarian. Within the single decade between 1969 and 1979, marital fertility fell by about 40 percent. Contraceptive prevalence among married women of reproductive age increased from under 15 percent in 1969 to nearly 60 percent by 1981. Family size preferences declined substantially. Only a generation ago, married couples averaged six to seven live births and five to six living children by the end of childbearing. In 1981 about half of recently married women indicated a preference for two children and only one in ten wanted more than three. With respect to these changes, the small Muslim minority (about 5 percent of the population) and, to a lesser extent, Buddhists in the Southern region have lagged somewhat behind but most other identifiable groups have participated fully (Knodel et al., 1982; Kamnuansilpa et al., 1982 and 1983).

While there is considerable quantitative documentation of the ongoing fertility transition in Thailand, there has been little qualitative analysis of the perceptions of the people involved. ' We share the conviction that "most actors involved in major social changes know that something is happening, have some idea of the direction and shape of the changes, and have speculated-at least to themselves-about what is happening and why" (Caldwell et al., 1982a). With proper scrutiny and interpretation, the information, opinions, and attitudes expressed by the actors regarding the rapid decline in fertility can yield useful insights into the nature and causes of the process taking place.

Owing to the recency and magnitude of change in reproductive behavior in Thailand, an older generation whose members built their families under a pretransition fertility regime lives side-by-side in virtually every village and POPULATION AND DEVELOPMENT REVIEW 10, NO. 2 (JUNE 1984) 297

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

298 Fertility Transition in Thailand

town with a younger generation whose members are effectively limiting family size to two or three children. For researchers, this provides an unusual op- portunity for comparisons of pre- and post-transition generations. The present study attempts to do just this by analyzing qualitative information gathered through focus group sessions.

Research approach and study design

A focus group session consists of a group discussion in which a small number of participants (usually six to ten), under the guidance of a moderator, talk about topics of relevance to the particular research study. The informal group situation and open-ended nature of the questions are intended to encourage participants to comment on behavior and elaborate opinions to an extent that is difficult in more formal individual interview situations. Interactions among participants stimulate discussion. This group dynamic distinguishes focus group sessions from more conventional, individual in-depth interviews typical of ethnographic research. While the technique is widely used in marketing re- search, it has seldom been used in social research.2

The purpose of focus group sessions is not to provide statistically gen- eralizable quantitative data but rather to obtain qualitative information exposing underlying attitudes, opinions, and behavior patterns. In this sense, it is a complement rather than an alternative to the sample survey approach. For example, focus group sessions can serve to generate hypotheses to be tested by surveys on a broader population base; to probe findings from surveys that seem ambiguous or puzzling; or to investigate topics that do not lend themselves to a survey format.

Participants in a session are usually selected to be of roughly equal status and to have similar values relating to the topic under investigation. The rea- soning is that discussion of sensitive topics is facilitated if participants perceive one another as sharing the same perspective. Separate sessions with groups selected on different sets of criteria permit the in-depth exploration of views of persons with quite different characteristics. Typically the discussion is tape- recorded. The resulting transcripts serve as the basic data for analysis.

As with any research method, the focus group approach has a variety of limitations and potential problems. There is a risk that one or two participants may dominate the discussion. A group situation can also inhibit discussion if other members are perceived to be unsympathetic to a participant's views. Covert behavior or opinions disapproved by the community may also be less likely to be revealed in front of a group than in an individual interview unless they are perceived to be shared by others in the group. A skilled moderator can reduce these problems by bringing all or most participants into the dis- cussion and creating a permissive atmosphere in which a range of opinions can be expressed and any tendency to simply agree with others can be reduced.

In view of the qualitative nature of the data and the analysis, considerable subjective judgment is involved in interpreting the discussions. Statements

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 299

cannot always be taken at face value and often require interpretation based on the context in which they are made or on information available to the researcners from external sources. Nevertheless, the analyst is considerably closer to what has actually been said and its context than is the case when analyzing survey responses. Distinguishing between spontaneous statements and those made only after probing, or noting the extent to which a particular topic or question generates discussion, can help the analyst in making judgments about the meaning of what is being said in ways that are not possible with the survey approach. Moreover, as in the present project, when the analysts are present at the sessions as observers, moderators, or notetakers, they can have an additional "closeness" to the data that should aid in analysis and interpretation.

The use of focus group sessions is clearly a less ambitious approach to collecting qualitative data than is involved in either a full-fledged anthropol- ogical study or the "micro approach" to demographic investigation advocated by Caldwell and associates (1982a). However, since conducting focus group sessions does not require long-term residence and extensive participant obser- vation in the communities, a greater number of communities can be included in a given study and thus a broader coverage attained. Moreover, since tran- scripts are produced, verification of findings and reanalysis can, in principle, be undertaken by persons other than the original investigators. Such an en- deavor is not feasible with the field notes of the anthropologist or the practitioner of the micro approach to the study of demographic behavior.

Our study was designed to facilitate a comparison between older and younger generation Thais representing modal pre- and postfertility decline situations. Separate sessions were held for older and younger generations of each sex. Complete sets of four focus group sessions were held during late 1982 and early 1983 in five villages: one each in the Central, Northern, and Northeastern regions, and one each in a Buddhist and Thai-speaking Muslim village in the South (where Muslims are concentrated). In addition, all but an older generation men's session were held among rural migrants living at con- struction sites in Bangkok. There were 23 sessions in all.3 While the total sample is not statistically representative of the Thai population, each major region is included as well as one village of the Muslim minority and a group of rural migrants. The sample has a strong rural emphasis by intent, since our major interest is in the rapid reproductive changes of the rural majority. In four villages, resident anthropologists collaborated with us, providing back- ground information and helping in the selection of participants.

Three criteria were used to guide our recruitment of older generation participants: preference was given to those who were over age 50, had at least five children surviving childhood, and had no more than an elementary-level education. Since there is little evidence of rural fertility declining before the second half of the 1960s (Retherford et al., 1979), persons over 50 in late 1982 or early 1983, when the focus group sessions were conducted, would either have been close to the end of or had already passed out of their repro- ductive years by the start of the fertility transition. Concentrating on older persons with at least five children ensured a contrast with the reproductive

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

300 Fertility Transition in Thailand

experience typical of couples currently in the process of having families. Information from previous studies suggested that such older participants would be reasonably typical of the generation who bore their children just before the onset of the fertility transition in rural Thailand. For example, according to the National Longitudinal Survey in 1969, six to nine live births and five to eight living children were the most commonly reported numbers by rural women in their 40s. About half of all such women reported numbers in these ranges. Furthermore, only a very small minority of older, or even younger, rural Thais have progressed past an elementary education.

In recruiting younger generation participants, preference was given to married women under 30 and married men under 35 who wanted no more than three children and had not exceeded that number, and who had no more than an elementary education. In addition, to ensure a personal reference point against which they could contrast their own low fertility goals, they themselves were to have come from families of typical pretransition size (five or more children).

Some deviations from the preferred criteria were accepted to facilitate formation of a group. Moreover, due to occasional misunderstandings on the part of persons helping with the screening process, older generation participants were sometimes selected on the basis of the number of children born rather than the number of children surviving childhood.

Older generation participants were on the average 59 years of age (58 for women and 60 for men). Younger generation participants were 27 years old on average (26 for women and 28 for men). About three-fourths of the older generation participants reported having between five and nine children. Although it is not possible to distinguish in all cases those reporting children ever born from those reporting surviving children, in either case the numbers would be within the modal range for the generation completing childbearing just prior to the fertility transition. Among the younger generation participants, only one wanted more than three children and over half wanted no more than two. The groups for both generations were relatively homogeneous and typical in terms of educational background.

Since the selection criteria were not designed to yield groups represen- tative of the entire spectrum of reproductive experience, our sample differs in certain ways from the overall population. Perhaps most importantly, older generation Thais with fewer than five surviving children are largely excluded, and thus couples who experienced a high rate of child loss, those who may have practiced effective family limitation, and the subfecund are underrepre- sented. Nevertheless, since participants were all members of village commu- nities for all or most of their lives, they were likely to have been aware of the attitudes and behavior of others in the community regarding many of the dimensions covered in the discussions and presumably were able to act as informants about their generation's situation in general. Given our interest in gaining qualitative insights into the fertility transition process, we believe the advantages of a purposive design involving modal groups outweigh those of attempting to form a statistically representative sample.

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 301

All sessions followed more or less the same discussion guidelines, which encompassed topics gleaned from what we judged to be the most promising hypotheses about fertility transition as embodied in the current literature and in our own thinking. Some modifications in the topics discussed were made over the course of the project in accordance with the experience gained from prior sessions. In constructing the guidelines, we considered the ordering of questions to be important. For example, we first asked an open-ended question about the reasons for the intergenerational fertility differences and only asked about particular factors afterwards to enable us to distinguish factors mentioned spontaneously from those cited after the moderator introduced them. We de- liberately deferred the discussion of birth control until after the discussion of reasons for the change in family size, since we wished to determine whether availability of contraceptive methods spontaneously arose as a reason for the intergenerational fertility differences.

The presentation of our findings is organized around issues that we believe are fundamental for an understanding of the motivations, means, and processes involved in the radical shift in reproductive patterns that is taking place. To illustrate our findings, we make occasional use of direct quotations from the focus group discussions. We have attempted to select quotations that are both reasonably typical and clear. Implied words are included in brackets where necessary to improve intelligibility.

Perception of fertility decline

Respondents of both generations are well aware of the changes in reproductive behavior and attitudes that have been taking place and identify these changes as a widespread phenomenon. Approval of the changes was virtually universal among the younger generation and predominated among the older participants.

Every day now they have fewer children. (older man, Northeast) Almost everyone wants few children, at most no more than three. When I say this, I mean those my age.... (younger man, Central) [Having fewer children nowadays] is good because making a living is different than before. (older Buddhist man, South) Few is good. The expense is lower. Money is hard to find now. (older woman, Northeast)

A small family size is widely acknowledged and accepted among both older and younger generations to be more adaptive to present-day conditions. When disapproval was expressed by older participants, it primarily reflected the judgment that younger couples were not exercising enough foresight about help and support from children later in life. Among those indicating disap- proval, many in reaction to a direct question saw advantages as well to the small family. Indeed, in most groups there was some recognition of both advantages and disadvantages of smaller family size, although most agreed that under present circumstances the advantages were predominant. Thus a

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

302 Fertility Transition in Thailand

deeply rooted large family ideal does not seem to have been an important impediment to rapid fertility decline among rural Thais. The small family norm has become imbedded in people's thinking and behavior and is now recognized as a new generation's way of life.

Societal change

A picture of interwoven social and economic changes affecting both the daily lives of Thai villagers and their future plans for themselves and their children emerged from the focus group discussions. The net result of these changes is that large families are now increasingly perceived as an economic burden. Extensive market penetration of the countryside, monetization of daily life, and changing consumer aspirations have led to a sense that the general cost of living has increased. Particularly salient in the increased cost of raising children is the cost of education. Providing children with more schooling is seen as the primary mechanism through which families can cope with limited availability of land and reduced prospects for themselves and, more impor- tantly, for their children, for making a satisfactory living through agricultural activity. At the same time, children are generally seen to be less helpful in doing household chores and in contributing to the family's economic activities, in large part as a result of increased schooling. However, expectations of support from children once they become adults clearly persist. Many of these changes are reinforced by the mass media, which are increasingly penetrating the countryside. To think in terms of monetary costs when contemplating raising children is now considered normal, and the view that the more children a couple has, the more costly it will be has become the accepted way of thinking.

The participants did not explicitly make all these connections themselves. This is undoubtedly due in part to the difference in the level of abstraction we seek as social scientists and the level of explanation that seems of direct relevance to them. Nevertheless, their comments go a long way toward sug- gesting this basic picture of the social and economic changes underlying the fertility transition in Thailand, which we now examine in detail.

The perceived rise in the cost of living and child rearing

Time and again, when participants were asked why the younger generation wants smaller families than the older generation had, they responded that nowadays everything is expensive and that earning a sufficient living is difficult. A multitude of examples were cited pointing out how much more costly vir- tually every item is today compared with the past when it either was far cheaper or was home-produced, home-grown, or available for free in nature. The increasing cost of raising children is seen as an important part of the rising cost of living, and although not stated explicitly, expenditures on children are likely to be competing with the expanding array of other expenses deemed

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 303

necessary to conduct a satisfactory life. Thus, bringing up children is viewed by both generations as a far greater economic burden now than in the past. Limiting family size is seen as an important means for the younger generation to keep their expenditures within reasonable bounds.

Before when things were not expensive, there were plenty of children. Now when things are expensive, there are fewer children. If you have many children, you need to spend a lot. If you have fewer children, you spend less. (older Buddhist woman, South) Right now everything is expensive, and there's also education for your children. Earning a living is also very hard. To have many children is to be poor. (younger Buddhist woman, South) The more the children, the more the expense; the fewer the children, the less the expense. Parents can't earn enough money to serve their children's needs. (older woman, Northeast)

Market penetration, monetization, and consumer aspirations

Objectively, it is unlikely that people are worse off economically now than they were in the past. Real per capita income has increased substantially over the last few decades. Most Thais probably live more comfortably in the sense that they are better fed, better clothed, and own more consumer durables nowadays than a few decades ago. However, young parents feel they need to spend more on their children than was true for the older generation because they believe their children need more education than in the past and because child care practices and views of the legitimate needs of children have changed.

The crucial process underlying the perceived increase in the general cost of living is the rapid expansion of the cash economy through widespread market penetration and the consequent monetization of daily life. The availability of durable goods and food products in rural areas has increased substantially through an expanding network of village shops and increased access to urban stores and markets due to expanded public transportation. This has led not only to substitution of store-bought items for those previously produced at home or gathered in nature but also to aspirations for goods previously un- available or unfamiliar to the average villager. Mass media penetration has surely contributed to a greater awareness of the availability of goods and to heightened consumer aspirations for parents and children. While only two decades ago radios were rare and television virtually unknown in the country- side, now radios are commonplace in rural households and there is hardly a village without at least one television set.4

The increased need for money both to cover the costs of raising children and to meet a couple's other needs and desires has undoubtedly made these costs far more salient. Moreover, the increasing availability and awareness of consumer items and the desire to acquire them create a situation in which having another child and purchasing goods may be perceived as alternatives.

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

304 Fertility Transition in Thailand

These days we need to spend money to raise children. We can't have many. (younger woman, North) Before, parents used rice to raise children. Now they use money. (younger woman, Bangkok construction site)

Participants were keenly aware of the extent to which monetization pervades their lives, frequently discussing the expenses that confront them just to meet their everyday needs. Moreover, the older generation groups made clear how rapid the change has been, identifying a variety of differences between just a generation ago and now. Participants no longer view themselves as self-sufficient home producers but rather as consumers in a cash economy. Numerous examples were cited: rope made from jute has been replaced by nylon rope; the palm bowl, an all-purpose receptacle, has been replaced by plastic, aluminum, stainless steel, and glass bowls. Formerly fish, fruit, and vegetables either were far less expensive or could be acquired without pur- chasing. Clothes and material that were commonly woven and sewn at home have now been replaced by factory-manufactured products.

Before, you made it at home but now you buy it in shops. (older Muslim man, South) Today we've got to buy everything. In the past we didn't have to spend money on these things. We didn't spend much money like nowadays. (older woman, North)

In interpreting the pervasive sense of economic hardship today, Easter- lin's concept of relative income, defined as the ratio of anticipated earnings to aspirations, may be relevant (Easterlin, 1980). Even if real income is higher now for the younger than it was for the older generation, higher consumer aspirations, higher anticipated costs of adequate child rearing, and the greater extent to which these require monetary payment result in a lower sense of relative income.

Medical costs and child care

The modernization of medical care is seen as contributing to the rising cost of living. While traditional and modern medicine still coexist in Thailand, the greater effectiveness of modem medicine is increasingly recognized by both the older and younger generations (Riley and Sermsri, 1974). The cost of modern treatment, however, is considerably higher than that of traditional services and often involves additional transportation costs. Even when the treatment is provided at subsidized prices at government health outlets, the cost can be significant to the average villager.

The increasing resort to modern treatment affects costs both for the parents' own medical care and for their children's, contributing to the perceived increased cost of child rearing. In addition, the increasingly common practice

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 305

of giving birth in a hospital or health center rather than at home has increased the cost of childbearing.

Some participants stated that nowadays people are more concerned about the health needs of their children and take better care of them than in earlier times. This may largely be due to an awareness of the availability of health care facilities and a conviction that parents can now exert a greater influence on their children's health than in the past. In several groups, however, there were suggestions that children are now the subject of more concern and at- tention, not only in terms of health needs but other needs as well, and that parents indulge their children more today than in the past. For example, a number of participants indicated that children were watched more carefully and that parents were more concerned about properly clothing a child. Some seemed to feel that the wishes and demands with which children confront parents are accorded more attention today. Remarks indicated that children now demand and apparently get small amounts of spending money, to buy candy and other treats, whereas previously such requests would have been considered out of place. Participants also commented that children are permitted to play more now. There were hints that parent-child relations may be shifting because, with the spread of schooling, a fairly young child often has more formal education than his parents, potentially undermining parental authority and increasing the legitimacy accorded children's demands and wishes.

Increasing educational costs and aspirations

While both generations identified a variety of reasons for the increasing cost of raising children, expenses associated with schooling were probably most consistently mentioned and considered the most significant. Most groups men- tioned spontaneously that the cost of education played an important part in the younger generation's desire for smaller families. Everyone seemed to believe that education is more important today than in the past and that only by having fewer children can a couple afford the appropriate education for each.

[People want fewer children] because now things are expensive and because they want their children to get a higher education. If there are many children you cannot send them to school, but if you have only one or two children, you can manage. Nowadays education must come first. (younger Muslim woman, South) We can afford our children's education if we've got just a few. But if we have more it would be a big burden. (younger woman, North)

The perceived financial burden of educating children is compounded today by the feeling that all children, whether sons or daughters, should be given an opportunity to acquire formal education. In the past, sons were clearly favored, but sex differentials in educational attainment have diminished con- siderably over recent decades.

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

306 Fertility Transition in Thailand

The increased need for education is seen as emanating both from an increasing need to find nonagricultural employment and from the attraction of jobs that require some education as a way to improve one's standard of living. Due to diminishing opportunities to make a satisfactory living off the land, discussed more fully below, education has come to be seen as a substitute for land.

Having higher education is the same as owning ricefields. Without education there's no job. (older woman, Northeast) Some people do not have land for their children so they want them to get education, to be civil servants, to be a soldier or policeman, things like that. (younger man, Central)

Schools are now more accessible and available than before. The educational infrastructure in Thailand expanded considerably over recent decades. Com- pulsory education has been increased from grade four to six. Furthermore, schools, especially beyond the primary level, are more accessible through improved transportation. Thus as education has come to be viewed as more desirable, increasing numbers of rural Thais also see education for their children as more readily within their grasp.

Higher educational aspirations are interwoven with hopes for upward social mobility so that the children can live a more comfortable life and will be in a better position to provide financial support for their parents later in life. Participants generally perceived farming (in which most are engaged themselves) as an arduous and precarious occupation, dependent on circum- stances beyond human control and yielding uncertain and unstable income. Many indicated that they did not want their children to experience the same difficulties as they had. Younger participants most frequently mentioned gov- ernment service occupations as most desirable for their children. Such jobs are seen as secure, providing steady income, social welfare benefits, and prestige. Children who obtain such jobs will live comfortably themselves and be able to help their parents as well, especially through financial contributions.

We are in difficult times as farmers. I don't want my children to do this type of work like me. I want my children to have knowledge, to do work sitting in a chair like other people. (younger Buddhist woman, South) Children who have an education will sometimes make their parents comfortable, but those without an education will depend on their parents. Those with education will use their education to earn a living. (older Muslim woman, South) To be a government employee is good both for them and us. (younger man, Central)

In most villages there are already examples of farmers' children who have been successful in attaining the types of positions many parents wish for their own children, thus making such aspirations seem realizable. Mass media

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 307

also fuel such hopes by providing abundant examples of relatively affluent urban lifestyles along with stories of upward social mobility.

The focus group discussions support the point made recently by Caldwell (1980) that at the onset of fertility transition, the changing educational level of children is a more important stimulus to fertility reduction than the changing educational level of the parents. Not only is the adult rural population fairly homogeneous with respect to low educational attainment, but educational dif- ferentials in fertility and contraceptive practice are weak (Knodel et al., 1982). Thus it seems unlikely that changes in the parents' education have played as important a role in encouraging a smaller family size as changes in the perceived need for and cost of education for children.

Land availability and concern over inheritance

Many participants expressed a feeling that land is now limited and the pro- ductivity of farming is not sufficient to support many children. Indeed, while the amount of land under cultivation has increased over recent decades, it has not kept pace with population growth, and by now most usable agricultural land in Thailand has been occupied except in parts of the South (Whitaker, 1981: 127). Participants often raised the problem of limited land availability spontaneously when discussing why couples nowadays want fewer children and why there was an increased need to provide children with education. Conversely, plentiful land in the past was often given as a reason why the older generation had large families.

Many participants stated that their present landholdings are inadequate. While in former times holdings could be expanded simply by clearing forests, this is no longer seen as possible. Instead the primary means of acquiring more land is through purchase, but for most participants prices are prohibitive. The problem of land availability was particularly salient when people discussed inheritance and the difficulty of providing sufficient funds for their children's future.

Traditionally land is seen as the main asset parents pass on to children. Inability to acquire additional land has forced parents to reassess not only the need for education but also the way to subdivide the family's land among their children. The tradition of dividing the land equally threatens the viability of individual holdings. Moreover, parents are also concerned with the problems their children or grandchildren will have to confront when the land must be subdivided once again. Many felt that unless family size is limited, their children will be unable to make a living from the land. Participants also mentioned that their control over other natural resources important to their sustenance, such as fisheries and forests, was declining.

I don't have enough land. It's not like earlier times. There are more people now so I don't want to have many children. (younger woman, Northeast) If people had many children [nowadays], they would starve. They got very little

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

308 Fertility Transition in Thailand

[land] from their parents because there were four or five children. They could get only small shares and if there were more children it would be terrible. (younger woman, North)

Contributions of children to household economic activities

So far we have dealt only with the effects of recent social change on the costs of children. Also important are changes in the benefits provided by children, since it is the net balance of costs and benefits-broadly defined to include both economic and noneconomic dimensions-that determines family size preferences. For Thailand, it is useful to distinguish between contributions children make while they are still young and support provided once the children become adults. While participants largely agreed that the help children provide while growing up has declined, expectations of support for parents by adult children apparently remain substantial. We discuss contributions of children while they are growing up in the present section and turn to the support provided by adult children in the subsequent section.

In agricultural countries such as Thailand, virtually all members of the household except infants contribute some labor both to productive activities and to household chores. Older focus group participants confirmed that in their youth they helped their parents plant rice, take care of buffaloes, fetch water, cook food, and perform other work that contributed to the economic sustenance of the household. When asked how much children help their parents nowadays compared with the past, the consensus in most groups, both younger and older, was that children currently help less, in part because children now spend more time at school, study more, and play a lot more than previously. Moreover, the content of the schooling is not seen as useful for teaching children how to do needed chores. Although parents complain about the decline in the help children give to them, they seem to accept it as part of the cost of educating children today.

[Nowadays, children] do not help very much because they all go to school. Only the parents are left at home. (younger Buddhist woman, South) [Children back then] worked in the field, in the garden, did everything. Whatever the parents did, the children would do the same. But today they only study, they don't help their parents. Only study and the parents have to earn money to support them. (older woman, North)

Participants rarely related the decline in help from young children to the fact that people now want to have fewer children. Unlike education, which is seen as involving direct and continued monetary cost and thus as an obvious economic burden for the parents, the decline in children's help, which is often a matter of degree, is seen more as an inconvenience than as a serious hardship. Moreover, the need for child labor may be declining as farming technology

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 309

changes. Nevertheless, parents' perception of children is changing from a view of their taking an active share in the division of family labor to being recipients of the family's resources.

Support to parents by adult children

The expectation that children, when adults and economically active, will pro- vide comfort and support to their parents, particularly when parents are too old to work or care for themselves, is shared by virtually all segments of Thai society. This support is viewed as repayment to the parents for having borne, cared for, and raised the child. It is a tradition deeply rooted in the secular culture and firmly linked to the broad normative structure. In one sense, it epitomizes the patron-client model of social relations thought by many to be the pervasive mode of social interaction in Thailand (Kaplan, 1981). On another level, helping support parents, both economically and affectively, is viewed as a way a person can accumulate "merit, " and thus has religious significance. 5

Parent repayment in Thailand is a process that traditionally begins when children are old enough to be useful and continues through the child's adult years as long as the parents live. It is the later stage of the process, however, when parents are no longer able to work or care for their own needs, that seems to be most important. Because of this, and because the declining con- tribution made by children prior to adulthood contrasts with the persistence and shifting forms of support to parents provided after children achieve adult- hood, we focus our discussion of parent repayment on contributions made in the children's adult years.

Focus group participants of both generations agreed that parents expected and were entitled to support from their adult children as repayment for, or as an obligation created by, the care parents provided children during infancy and childhood. The discussions made clear that repayment encompassed both eco- nomic and noneconomic dimensions. Contributions of money, material goods, and labor assistance to the parents' economic activities were cited as well as noneconomic labor services and moral support.

The most frequently mentioned expectation was assistance during ill- ness-provision of physical care and psychological comfort, primarily, but also financial help in purchasing medicine and paying for medical services. Children are also expected to be responsible for arranging and paying for funeral ceremonies, often a significant expense in rural Thailand, and for "making religious merit" on behalf of deceased parents. Other common examples in- cluded provision of food and clothes, donations of money on a regular or irregular basis, social visits, help around the house, and help with cooking. The form of repayment is seen to vary with the stage in the parents' as well as in the adult children's life cycles, balancing need and ability. Parents an- ticipate a reduction in monetary and material contributions once adult children start to raise families of their own.

The expectation of help and support later in life is perceived as a fun- damental reason for having children.

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

310 Fertility Transition in Thailand

If my children don't come [to look after me] when I am old, I don't know why I had children. (younger woman, Northeast) I hope for everything from my children but I don't know if they will do it all for us . . . to help us in work, get money for us, help the family, things like this. (younger man, Northeast)

Some participants stressed the need for several children in anticipation that some will turn out to be undependable.

While expectation of old-age support from children clearly persists, it is more difficult to assess whether the extent of support has changed, in part because the form of support appears to be changing. Many participants observed that monetary remittances from wage labor or nonagricultural jobs outside the village are replacing help in activities requiring labor. Thus, parent repayment is becoming monetized along with other aspects of villagers' lives. Both gen- erations appear to be aware of this change. While some see this as an im- provement and even as an increase in the extent of support provided to parents by adult children, this viewpoint is not universal.

The monetization of support is tied in with many of the other changes discussed. The increasing scarcity of farm land or the greater rewards of wage earning jobs mean more children seek their livelihood away from home. It is often difficult to leave a job to return to the home village when labor is needed, such as when rice is being planted or harvested. Older parents are faced with the same monetization of daily life as everyone else but may have greater difficulty earning cash incomes of their own. Hence the shift from labor as- sistance to cash remittances may be convenient for both parents and children. Moreover, many elderly parents probably have access to both forms of support, since often some children remain home or near the village while their siblings have nonagricultural jobs elsewhere.

Back then [children] helped with labor. Now they help with expenses. (older Buddhist man, South) Those who stay at home help us cook. Those who are far away send us money. (older woman, North) In the past children only helped growing ric...e . [Today] they both grow rice and make money for us. [When I was young], I didn't know where to go. After the rice growing season, we stayed at home. Now after growing rice, [young people] go to earn some money. It's better than before. (older woman, Northeast)

We were particularly interested in how participants reconciled the continuing anticipation of dependence on children for old-age support with the younger generation's desire for a smaller number of children. Old and young participants alike were evenly divided as to whether or not having few children threatened security in old age.

The most common view was that a tradeoff is involved: parents today opt for small families because they judge the additional benefits to be gained

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 311

later in life from many children not worth the hardship of raising them. For some, there is no choice to be made since under current social and economic circumstances having many children is viewed as simply out of the question.

I want many children after they have grown up. But right now I want only a few because I have no time to raise them. (younger Muslim woman, South) In principle, having a lot of children will make you comfortable when you are old, but when they are young it will be difficult to find the money to raise them. (younger man, North)

One view among participants who did not feel that having few children jeopardized old-age support was that support in later years does not depend on the number of children but rather on their upbringing. A small number, if properly raised, should still be sufficient to provide parents with security. There were also participants who believed that the relationship between number of children and economic support later in life has changed and that a simple association between more children and greater economic security in old age no longer holds. Rather, they saw a relationship between the amount of ed- ucation children receive and the ability to support parents, especially in terms of cash remittances. Since the cost of providing an adequate education to many children would be prohibitive, the economic prospects of both children and parents are improved by limiting family size to a few.

[Support in old age] depends on how you raise and bring up your children. To have a small number of children but to raise them well is better than to have many children but not to raise them well. (younger Muslim man, South) [Parents will still be supported well] because education has progressed. Now children get a higher education. Repayment of parents will go according to the level of education. . . . Two children with education can repay parents better than ten [without education]. (older man, Northeast) [Children should study a lot.] That's good. They'll be able to find jobs. We'll be able to rely on them. There'll be a future for them. (older woman, North)

Reproductive decision-making

Family size as a matter of choice

For the pretransition generations, family size was generally not thought of as a matter of choice but rather the product of a natural process over which the couple had minimal control. This is clearly reflected in many comments by participants that there were no effective means of birth control in the past (see below) and that children continued to come until the end of the couple's natural reproductive span. Some older participants indicated they consciously wanted many children, and others seemed at least accepting of large families even if they did not actively seek them. There were also substantial numbers of older

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

312 Fertility Transition in Thailand

participants who indicated they had thought about limiting family size but generally felt helpless to stop childbearing.

We didn't think about how many [children] we should have. [My husband] felt sorry for me when I had lots of children but no one knew what to do. (older Buddhist woman, South) In the past there was no choice. Just let it be. (older woman, North) We never thought about [how many children to have] so [my husband and I] never disagreed. (older woman, Central)

For couples having families nowadays, the situation is very different. Virtually everyone considers the number of children to be a matter of deliberate choice. Most young husbands and wives start married life with the idea that they want only a small family, and they are likely to settle on a precise number by the time one or two children are born. Thus in the course of the fertility transition, family size has emerged as a matter of choice. Among younger participants, allowing family size to result from uncontrolled fertility is as remote an idea as effectively limiting family size to a chosen number was for the older participants just a few decades ago.

The locus of reproductive decision-making

Recent discussions of the props supporting stable high fertility in parts of the Third World place considerable emphasis on the fact that in a number of societies the responsibility for reproductive decisions does not lie only or even primarily with the couple themselves but extends to a wider kin group, par- ticularly the husband's parents. The influence they exercise is thought to be generally pronatalist in accordance with their own interests (e.g., Caldwell, 1982: 117-118). In Thailand this is emphatically not the case.

Childbearing decisions in Thailand are generally defined as being pri- marily and even exclusively the responsibility of the couple themselves. Focus group participants of both older and younger generations were largely in agree- ment on this matter. When asked whether they sought advice from anyone about how many children to have, younger participants rarely mentioned par- ents. They either stressed that the couple decides themselves or mentioned friends and siblings. Moreover, when they discussed such matters with others, they were simply seeking friendly advice and not being given imperatives from persons with deeply felt vested interests. The older generation also rarely mentioned discussing reproductive matters with their own parents. When asked whether they had advised their own children on such matters, many indicated they had not. Those who did typically gave advice more anti- than pronatalist, encouraging the use of contraception and advocating few children. The two generations agreed that such advice is considered nonbinding by parents as well as children.

The husband and wife consult only each other. Nobody else should have an opinion. Other people don't help raise the children. (younger woman, Northeast)

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 313

It's not related to anyone else. The husband and wife decide together. (older man, INortneast) We don't talk with anyone else. If we want two, we get two.... We wouldn't listen to what other people advise. We listen to ourselves. (younger woman, North) [I advised them] not to have many children because not only they but we will also be in difficulty. (older man, Northeast)

These focus group results are consistent with a 1975 rural survey which found that the vast majority of husbands and wives had not received advice from other family members about the number of children to have. Moreover, among the few who did receive advice, a substantial proportion said they did not follow it (Deemar Company Ltd., 1975). Pronatalist influence of the older generation is considered to be a critical barrier to fertility decline in many societies. Its absence in Thailand contributes to a cultural setting permissive of rapid reproductive change. Female autonomy and decision-making

Another important feature of the Thai cultural setting that facilitated rapid fertility decline is the relatively high degree of female autonomy, extending into the area of reproductive decisions. While observers may hold differing views about women's status in Thai society, most feel that Thai women act more independently in many spheres than women in most other developing world societies.

Most participants believed that reproductive decisions were a matter to be decided upon jointly by husband and wife; however, since younger partic- ipants often shared the same reproductive goals as their spouses, the question of whether the husband or wife should have the ultimate word was sometimes seen as irrelevant.

You must discuss [the number of children] between husband and wife. (younger Muslim man, South) Both sides should be satisfied, the husband and the wife. . . . [The number of children] can usually be agreed upon. (younger man, Northeast)

When participants did not stress the joint nature of reproductive deci- sions, they indicated that they had the dominant influence or ultimate say themselves. Men usually held that the husband, as head of the household and the primary provider for the family, should be the ultimate determiner. In contrast, women stressed wives had more at stake, inasmuch as they experience the pain of giving birth and most of the burden of child rearing. Some men agreed with this view. Moreover, since most contraceptive use in Thailand relies upon female methods, contraception is indeed under women's direct control.6

[The wife decides on the number of children because] after we give birth, we are responsible for housework, raising children. The men . . . are always away

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

314 Fertility Transition in Thailand

from the house. We have to raise the children and the buffaloes. (younger woman, Northeast) The woman is the person who thinks [about birth control] first and tells [her husband] so he will know. . . . If he wants three and she wants two children, from what I see the wife prevails. (younger woman, Central)

The relative autonomy of Thai women also means that they are fully exposed to the socioeconomic changes discussed above and have equal access to the communication and transportation networks that have penetrated the countryside. Hence their attitudes toward family size and reproductive behavior are likely to fully reflect the social changes under way.

Fertility regulation among the older generation

Previous research has established that the major proximate determinant of the decline in fertility in Thailand is the increased use of contraception and perhaps abortion (Knodel et al., 1982). Thus understanding the extent of knowledge and the nature of attitudes and practices concerning birth control in the past helps us understand the reproductive changes currently under way.

Birth control

Many statements were made by both younger and older participants that no deliberate birth control was practiced in earlier times. These comments usually arose spontaneously in response to an open-ended question on why the number of children differs between the two generations.

Back in our time it was unavoidable to have many [children]. There was no birth control or contraception. You kept on having children until menopause. (older Buddhist woman, South) In the past, children came until they stopped naturally. (older woman, Northeast)

Reproductive behavior of the older generation as described by focus group participants approximates what demographers term natural fertility (see Bongaarts and Menken, 1983; Knodel, 1983). Moreover, until recently, po- tentially effective means of birth control appear to have been unknown or considered inappropriate or unacceptable-indeed virtually unthinkable- within the cultural context. There is little reason to believe the situation differed for previous generations. There were some exceptions. Occasional reference was made to abortion by massage and, among Muslim participants, to with- drawal, both potentially effective means of birth control. Most commonly mentioned, however, was the use of herbal medicines that, as participants acknowledged, were unlikely to be effective. Such medicines were cited by participants in every village as well as by Bangkok construction workers. Also mentioned as ineffective were certain postpartum rituals intended to delay the

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 315

next pregnancy. In general, there appears to have been little or no store of knowledge of efiective traditional birth control upnon whic co"pe in th past could have drawn had they been interested in limiting or delaying births.

There was also near consensus that abstinence was not used as a means of either spacing births or terminating childbearing in earlier times. All groups believed that a brief period of postpartum abstinence should be honored for various reasons, particularly out of consideration for the health of the mother and because of the belief that resumption of intercourse shortly after childbirth is unsanitary. The period of postpartum sexual restraint believed appropriate was quite short, commonly one or two months. There is no evidence that this practice differed much between the older and younger generations.

Only rarely was abstinence spontaneously mentioned as a way that births could be controlled, and when specifically asked whether couples ever practiced abstinence, participants expressed disbelief that couples could do so for any length of time. Women often mentioned that husbands would be unwilling to forgo sexual relations.

We also probed to discover whether there was a normative basis for terminal abstinence not directly connected to intentional birth limitation. There was clearly no general feeling that sex relations are inappropriate for older people or that there is any link between reaching some stage of the family life cycle, such as becoming a parent-in-law or a grandparent, and cessation of sexual intercourse, as is apparently the case in some cultures.7 While many participants acknowledged that the frequency of intercourse dropped off with age and that some elderly couples may stop completely, this was explained in terms of loss of interest, rather than normative proscriptions. Far more common were comments that sex continued until very old ages, at least on an occasional basis, largely due to continued interest by the husband.

If you stay together as a couple, I can assure you it is difficult to stop. (older Muslim woman, South) Women are not interested in it. But men are still interested. [They continue having sex] until death. (older woman, North)

The lack of normative prohibition of sexual relations during later stages of the reproductive span has also been noted by at least one anthropologist (Mougne, 1981: 95) and accords well with the observation that, in the past, relatively high fertility persisted at older childbearing ages (Thomlinson, 1971: 69-70; Prasithrathsin, 1976: 67-68).

Previous survey research has indicated that withdrawal is generally not known or practiced in Thailand, except in the South where it appears to be familiar to substantial proportions of Buddhists and Muslims alike (Kamnuan- silpa et al., 1983; Porapakkham et al., 1983). In the Muslim village, partic- ipants of both sexes and generations spontaneously mentioned that withdrawal has long been known and practiced. Withdrawal, referred to by a local expres- sion, was usually discussed in the context of the more general concept of

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

316 Fertility Transition in Thailand

"controlling by yourself," which, for some participants at least, also encom- passed douching, the use of herbal medicines, and perhaps abstinence. Thus in the Muslim village, potentially effective methods of contraception were traditionally known, although, as some Muslim participants stated, not to everyone. If the Muslims in our sample are typical of Thai Muslims, this may help explain why prior to the recent fertility decline, Muslim women averaged fewer children ever born by the end of childbearing than did Buddhist women, despite an earlier age at first marriage (Goldstein, 1970; Jones and Soonthorn- thum, 1971).

In no session other than those held in the Muslim village was withdrawal spontaneously mentioned, although in the Southern Buddhist village a number of participants acknowledged having heard of withdrawal in response to a direct question. The extent of awareness and practice, however, seems to be substantially less than in the case of the Muslims. Outside the South, only rarely did a participant indicate having heard of withdrawal. Many participants were incredulous that anyone would use such a technique.

Induced abortion is illegal in Thailand and is considered by the vast majority to be contrary to the principles of Buddhism. Although participants in every session knew of abortion, especially by a traditional massage tech- nique, they usually stated either that it was uncommon in the past or that it was reserved for pregnancies resulting from extramarital or premarital sex relations. The dangers of abortion, especially by traditional methods, were stressed.

Although Western social scientists often recognize that cultural differ- ences affect the receptivity to modern methods of birth control, they rarely acknowledge possible cultural differences in receptivity to traditional methods, assuming instead that effective folk methods are both universally known and acceptable and would be used to reduce fertility if motivation were sufficient. Our results suggest that neither abstinence nor withdrawal was thought of as a means of fertility control in most of Thailand, and when described as such they were generally thought to be objectionable. The fact that withdrawal is acceptable among Southern Muslim villagers simply underscores the role that cultural differences can play. While cultural barriers to the use of specific contraceptive methods are not immutable, they may affect acceptability of traditional methods to such an extent that they are in effect unavailable, as appears to be the case in Thailand.

Age at marriage

Family size depends not only on the extent to which births are controlled within marriage but also on the age of entry into marriage (or some equivalent re- productive union). Most Thai women marry in their early 20s and have done so for at least several generations, making age at marriage moderately late by regional standards. Age at marriage for women has increased somewhat over the last few decades, contributing modestly to the fertility decline. More im- portantly, delayed marriage helped keep fertility in the past well below the theoretical maximum (Knodel et al., 1982).

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 317

Attitudes about the appropriate age to start marriage as revealed in the focus group sessions have been analyzed in detail elsewhere and indicate that participants make little connection between age at marriage and the number of children eventually born (Pramualratana et al., 1983). Age at marriage was linked to fertility in the conscious thoughts of participants only inasmuch as it was contingent on readiness to start childbearing and to assume responsibility for child rearing. Thus it is unlikely that the older generation viewed delaying marriage as a deliberate way of controlling fertility.

Child transfer

In populations where birth control is not practiced, an alternative and potentially effective way to control family size is to transfer children from families in an unfavorable position to raise them to families who are better situated to do so. The focus group discussions make clear that such transfers of children occurred throughout Thailand at least until recently. Participants in almost all sessions cited examples, sometimes from their own family's experience. Most believed child transfer has declined substantially or no longer exists because couples are now able to control the number of births they have.

The transfer of children as described by participants commonly involved a poor couple with many children giving up one or more children to a couple, often relatives, who were childless or had few children. The receiving couple were also typically said to be in a better position to provide for the transferred children and, if childless, to be in need of a child to care for them in old age and to inherit their property. Other frequently mentioned circumstances leading to child transfer were nonmarital births and the absence of a spouse due to marital dissolution through death, divorce, or desertion.

Child transfer thus appears to have been an acceptable mechanism for dealing with unwanted children in Thailand before methods of deliberate birth control became widely known and acceptable. Although we cannot judge the frequency of the practice from the focus group discussion, analyses of life histories of married women from a recent survey in Bangkok and a Central village suggest that it occurred only on a modest scale (Knodel et al., 1982: 122-123). -Its importance is mainly as a symptom of a broader unrealized demand for birth control in the past, a topic we now address.

Latent demand for birth control

One important factor underlying the suddenness and rapidity of fertility decline in rural Thailand appears to be the existence of a receptivity or latent demand for effective and acceptable birth control prior to the widespread availability of modem contraception. The frequent references by older participants to attempts to limit and space births through the use of traditional but ineffective means, particularly herbal medicines, is convincing testimony of such a latent demand. Recent anthropological studies have also noted an absence of effective birth control in the past and unsuccessful attempts to control fertility through folk methods (Lauro, 1979; Mougne, 1981).

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

318 Fertility Transition in Thailand

Back in our time, people also did not want to have many children. They took the traditional medicine but it didn't work. (older Buddhist woman, South)

In an attempt to assess latent demand in the past, older participants were asked whether they would have had fewer children if modern methods of birth control had been available. Not all participants were of like mind. Some indicated they had wanted large families; however, the more common opinion, sometimes expressed spontaneously in connection with prior discussion, was that they would have had fewer children had modem birth control methods been available.

I stopped having children before learning about birth control. If I had known about it, I wouldn't have had as many children as I have now. (older woman, Bangkok construction site)

Three, four, and occasionally five children were the numbers older par- ticipants most frequently indicated they would have liked to have had. While this is fewer than the number they actually had, not surprisingly it is still more than opted for by the present younger generation. The main reason why this latent demand did not translate itself into reduced fertility, in the participants' views, was their own lack of awareness of acceptable and effective means of fertility control.

Answers to hypothetical questions intended to assess latent demand must be interpreted cautiously. The older generation's perception of how they would have reacted earlier is likely to be influenced both by the socioeconomic change that has occurred since then and by their awareness of the younger generation's widespread use of contraception and acceptance of a small family size norm. Nevertheless, the combination of their present opinions and the frequent men- tion of past frustrated attempts to limit childbearing suggests the existence of a substantial latent demand for birth control prior to the start of the sustained fertility decline.

The focus group discussions also provide insight into the forces behind the latent demand. The socioeconomic change that makes large families dis- advantageous for the younger generation was already under way when the older generation was having children. In addition, there are strains inherent in bearing children and raising large families. These include time consumed by child rearing, particularly the mother's time, that could otherwise be spent doing household tasks and contributing economically. Such opportunity costs would be felt by both husband and wife, especially in a society such as rural Thailand where most farming is done by nuclear family units rather than as a corporate family enterprise.

When a woman has an infant she can only do work in the kitchen. (older Buddhist man, South)

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 319

It's very good to work together [in the rice fields]. But when we have an infant, we have to work alone while the wife looks after the baby. (older man, North)

In addition, the pain, suffering, and health hazards associated with child- birth, particularly frequent childbirth, and the general burden of caring for and rearing children, are costs borne predominantly by the mother and of concern to her. Such costs are not products of social and economic changes but rather aspects of reproductive life that hold for both the older and younger generations. It seems plausible that they played a role in generating demand for birth control in the past.8

If there had been [birth control] like today, women at that time would have had few children also. Some feared the pain of bearing babies, some of taking care of [many] children. I wouldn't have had so many children like this. Probably only four. (older woman, North)

Fertility regulation among the younger generation

In sharp contrast to their parents' generation, the current younger generation of married couples has virtually universal knowledge of effective means of birth control and shows little hesitation in adopting them in order to limit family size to a few children. This picture is well documented by survey research (Knodel et al., 1982) and confirmed by our focus group sessions.

In all groups, when participants were asked why the younger generation has smaller families than the older generation, they cited differences in the availability of modern contraceptive methods, often in connection with dif- ferences in socioeconomic conditions affecting the desire for children.

Nowadays young people can get the number they want. Now they can control. (older Buddhist man, South) Nowadays there are pills, so we have a smaller number [of children]. Back then there was no pill so they had many children. (younger man, North)

The upsurge of contraceptive use in Thailand coincides with the devel- opment of an active national family planning program that is the source of contraceptive supplies and services for the overwhelming majority of rural users (Knodel et al., 1982). Operating primarily through existing health stations and hospitals, the program introduced modern methods into all rural areas within a relatively short time span between the late 1960s and early 1970s (Rosenfield et al., 1982). Most older participants indicated they first learned about birth control through the official program. In contrast, the younger generation frequently mentioned learning about birth control through friends or relatives or through observing the practice of other villagers or their traveling to get supplies or service.

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

320 Fertility Transition in Thailand

We learned about [birth control] from inside the village. Most of the time people will talk about it and tell others. (younger woman, Central)

The contrast between the two generations illustrates the process by which modem birth control practice is transformed from an innovation emanating from outside the community to an established, institutionalized behavior pattern learned from persons inside the community. That process can clearly occur more rapidly today than it could have in the past in the now developed countries. Mass media, particularly the radio, had already permeated the countryside in Thailand by the early 1970s and spread information to villages almost instan- taneously. Moreover, compared with withdrawal, abstinence, and secretive abortion, which played the major roles during early stages of sustained fertility decline in the West, the use of modern methods is more visible to others in the community. Users openly go to obtain services and, in the case of oral contraceptives, may be observed taking a daily dose or possessing a packet, as was occasionally mentioned by younger generation participants.

Survey data also show, and the focus group participants confirm that virtually all contraceptive use today is of modern methods-except among the Muslims and, perhaps, Southern Buddhists (Kamnuansilpa et al., 1983). De- spite the powerful forces of social change leading to widespread deliberate limitation of births, no general upsurge occurred in the use of traditional methods, with the possible but unknown exception of massage abortion. In view of the low acceptability of traditional methods, it seems improbable that overall prevalence would have risen to the present level solely through their use had modern methods been unavailable, and even less likely that fertility would have fallen as fast, despite the considerable socioeconomic changes that have led to the desire for small families. Whether, in the absence of the national program, the limited number of commercial outlets providing modern contra- ception to the rural population would have expanded sufficiently to generate a level of prevalence equivalent to the current one is unknown. Obvious con- straints would have existed, however, because most modern methods require either trained medical personnel or licensed drugstores for their provision, and because the monetary costs of contraception affect levels of use among the rural poor (Knodel et al., 1984). Moreover, although increases in induced abortion might have been greater, its illegality, unacceptability in a moral sense, and the perceived health risks would probably have been serious con- straints to its widespread practice.

The importance of the national program lay not only in the provision of contraception but also in increasing awareness of the possibility of controlling family size through effective and acceptable means. Information dissemination by the program was probably of great significance. It occurred through activities explicitly intended to provide information and, indirectly, through a demon- stration effect set in motion by users who obtained their method at a program outlet. Moreover, the program attempted to spread messages legitimizing pref- erences for small family sizes.

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 321

Child mortality reduction

Child mortality has declined substantially in Thailand, especially since World War II (National Research Council, 1980). Most focus group participants were aware that nowadays there are fewer child deaths than before. Many mentioned the greater availability of hospitals, health centers, doctors, and modem med- icine as a reason for the improvement. Younger participants expressed only minimal concern that their children would die of disease or sickness and seemed to worry more about accidents such as drowning or motor vehicle mishaps.

Nowadays few children die. It is different from the earlier times when there were a lot. (younger Muslim man, South) I don't think children will die; I think that they will grow up. I am not afraid of death due to diseases but rather due to car accidents. There are many cars. (younger woman, Northeast)

In view of the greater concern with manmade dangers than with disease, at least in the villages we studied, further reductions in infant and child mortality through public health or medical means seem unlikely to change the perception of survival chances significantly, and thus unlikely to have a substantial effect on fertility preferences in the future.

A few participants spontaneously mentioned that in the past it was nec- essary to have more children than today in order to insure against the greater chance of a child dying. When asked directly about this, however, many simply denied that this was something they had thought about previously. Others felt couples should consider this but were uncertain whether couples actually did.

You should not think about [having an extra child in case one might die] because it is like condemning your living child. (older Muslim woman, South) Like people said, only a two-ply rope was not enough in the past. They wanted three, four, or five-ply rope. If one or two of the plies were torn, there would still be some left. This is why they wanted more children, because children died a lot. (older man, North)

It is difficult to draw firm conclusions from the focus group discussions about the role mortality reduction played in bringing about the sharp decline in fertility in Thailand. While participants recognized the improvement in survival chances, they rarely mentioned this among reasons that younger cou- ples desire only a few children today. This does not necessarily mean that it was an insignificant factor. The fact that the loss of children is not a frequent occurrence-even when child mortality is relatively high-may reduce the saliency of mortality improvement to participants, compared with many other changes for which the effects are more obvious on a daily basis. Moreover, the selection criteria for older participants favored those whose experience with child mortality was favorable. A reasonable interpretation would be to consider

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

322 Fertility Transition in Thailand

the improvement in child survival chances as facilitating the reduction in fertility rather than precipitating it.

Synthesis and conclusions

Based on our focus group sessions and previous research, we believe an ad- equate explanation of Thailand's fertility transition, including its timing and pace, involves four major components, all of which interacted to result in an abrupt and rapid shift in reproductive patterns. First, rapid and fundamental social and economic changes have been taking place, causing couples to view large numbers of children as a burden with which they are either unable or unwilling to cope. Second, the cultural setting is relatively conducive to the acceptance of deliberate fertility regulation and limitation of family size as adaptations to changing circumstances. Third, there was a latent demand for effective and acceptable means to control fertility among a sizable proportion of couples before fertility began to decline and before modern contraception was widely available. Finally, organized efforts to provide modern contracep- tive methods, especially through the government's national family planning program, resulted in a massive increase in awareness of and accessibility to effective and acceptable means of fertility regulation.

Fundamental social and economic changes, under way in Thailand for some time, are responsible in part for the latent demand for fertility control among the older generation, and far more so for the current desire for small families among younger couples. A number of interrelated processes are in- volved. It is thus difficult and perhaps misleading to attempt to assess the relative importance of any one shift divorced from the larger total process of change. From the participants' viewpoint, the increased need for and cost of education for children appeared to be particularly salient in creating pressure to have only a few children. At the same time, expectations of parent repayment persist, especially in the form of providing comfort and support during the parents' later years. Many participants viewed a reduction in the support they expect to receive as a necessary price for avoiding the hardship of raising many children. Some, however, felt more was to be gained from rearing few, better educated children than from having many, less educated ones.

Caldwell (1976) has argued that sustained fertility decline is set into motion when the net intergenerational wealth flow, broadly defined, is reversed from a situation favoring parents to one favoring children. Whether and when such a reversal occurred in Thailand is difficult to determine from the qualitative and limited nature of our focus group data. Our results do make clear that the perceived (and undoubtedly real) monetary costs of raising children have sig- nificantly increased, while some benefits have decreased; such findings are consistent with a possible reversal of the wealth flow. Nevertheless, it remains to be determined whether the net flow was from children to parents for the older generation. Moreover, the views of some participants that, in the current socioeconomic setting, parents will gain higher returns from fewer, better

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 323

educated children suggest that even if the wealth flow remained upward, a basis for reducing fertility would still exist.

In any society, the impact of social and economic change on reproductive behavior is mediated through the cultural setting. In important ways, Thai culture is conducive to the limitation of family size and the adoption of birth control as ways to adjust to changing socioeconomic circumstances. Several pronatalist props or barriers to fertility decline thought to characterize many cultures of the Third World are notably absent in Thailand, at least during the period spanned by the two generations we studied. In particular, reproductive decisions are defined as the responsibility of the couples themselves rather than as a matter over which parents and kin exert considerable influence. Moreover, older generation parents generally support the small family size goals of the younger generation. Such a pattern conforms to the prevailing expectation that each conjugal unit will be largely responsible for the support of their own children (Pramualratana et al., 1983). Even during the initial period following marriage when a couple often lives with parents, usually the wife's, the younger couple may contribute more to the household's upkeep than the older couple (Yoddumnern, 1983). This contrasts with the situation thought to be typical elsewhere in the developing world whereby coresidence frees the younger couple from direct economic responsibility for rearing children (Davis, 1955).

In addition, the considerable degree of female autonomy in Thailand has undoubtedly contributed to the extent and pace of fertility decline. Women have considerable influence over birth control practice and family size, enabling them to effectively take into account their own stake as the bearers and rearers of children. Indeed, most birth control methods practiced in Thailand depend on women's initiative. Moreover, their autonomy in many other spheres of life exposes them as fully to the forces of social change as men.

Buddhism as practiced in Thailand also appears to pose no major barriers to the reduction of family size or to the use of contraception. Only abortion is proscribed on religious grounds. Indeed, religion was almost never mentioned in the focus group sessions in connection with discussions of family size or fertility control. More generally, Buddhism as practiced in Thailand emphasizes the primacy of individual action and responsibility (Pfanner and Ingersoll, 1962: 353; Phillips, 1967: 363-364), thus contributing to the overall flexibility and tolerance associated with Thai culture. In such a cultural setting, modern tastes, attitudes, and behavior, including changing reproductive patterns, can take hold with relative ease.9 While culture is also subject to change, the traits discussed, at least in their broad outline, show considerable consistency between the two generations covered by the present study.

According to comments by older participants as well as survey evidence, effective means of birth control were neither widely known nor widely practiced in rural Thailand until recently. An exception may be massage abortion, a potentially effective method but one probably limited to extreme circumstances. Withdrawal and lengthy abstinence are largely alien concepts in Thai culture, except among Thai Muslims and some Southern Buddhists. The existence of

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

324 Fertility Transition in Thailand

a previous latent demand for fertility regulation, however, may be deduced from comments to this effect made by older focus group participants and from their reports of having tried unsuccessfully to control births through ineffective means, particularly traditional herbal medicines. These attempts preconditioned couples in a sense for the acceptance of modem, effective methods. Such latent demand arose out of concerns that couples, and particularly women, had about the burdens inherent in childbearing and child rearing, as well as from some of the same social forces that became more pronounced during the recent fertility decline.

Organized efforts to provide contraception throughout the country met with immediate success largely because of the existence of a latent demand. The family planning program began in the mid-1960s, and by 1972 had more or less penetrated the entire countryside through local health stations (Rosen- field et al., 1982). Information dissemination spread awareness of birth control methods, legitimized concern over large numbers of children, and reinforced desires for fertility control. The timing of the fertility decline broadly coincides with the development of the national family planning program, which is the major supplier of contraceptive methods to rural Thais (Knodel et al., 1982). While some reduction in rural fertility may have preceded organized family planning efforts, it is unlikely that the same massive and rapid increase in awareness and practice of effective means of birth control could have occurred in the absence of such efforts.

The two most dynamic components of our explanation for Thailand's fertility transition are the set of fundamental social changes that have been taking place and the effect of the family planning program. Neither can be understood properly in isolation from the other. The socioeconomic changes played an essential role in creating the initial and continued receptivity to limitation of births. The organized efforts to promote and provide contraception have facilitated, and in some localities helped initiate, the widespread use of birth control. In our view, it is the interaction between these two forces, both operating within a cultural setting conducive to reproductive change, that has resulted in the rapid and extensive decline of fertility. Under such circum- stances, the conventional debate over the relative importance of family planning programs versus development in bringing about a fertility reduction seems to pose an inappropriate question.

There is little evidence at present to suggest that the forces of socioec- onomic change underlying the decline in fertility are spent. Most of these changes are continuing and may well intensify in the future. For example, the penetration of the countryside by television, undoubtedly the most potent of all the mass media, is still only at an incipient stage but is proceeding rapidly. Moreover, the family planning program continues to be active, and access to modem, effective means of birth control is increasing. Travel to existing outlets is easier and new outlets are being created or old ones upgraded to provide a wider variety of methods. The focus group sessions also revealed an almost universal desire for spacing births, only recently being realized to any sub-

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 325

stantial degree through contraceptive use (Knodel et al., 1982: 114-119). Thus there is still sufficient momentum to lead to further decreases in desired family size, increases in contraceptive practice, and reductions in fertility.

Many of the socioeconomic forces discussed in connection with the ongoing fertility transition in Thailand are operating in varying degrees through- out the Third World. Further, other countries have also mounted organized efforts to promote and provide contraception. Given these commonalities, it is interesting to consider the extent to which the dynamics of the Thai fertility transition are relevant for understanding reproductive behavior elsewhere. As we have stressed, the factors underlying these dynamics are mediated through a cultural setting unique to Thailand. Moreover, latent demand for fertility control may have been unusually far-reaching in Thailand. Nevertheless, de- spite social structural and cultural differences, similar socioeconomic forces seem to underlie the recent fertility declines in an area of South India being intensively studied by Caldwell and colleagues (Caldwell et al., 1982b, c, and d). In the Indian case, as in Thailand, socioeconomic change has generated demand for smaller families, and organized family planning efforts have pro- moted awareness of and increased the availability of modern means of con- traception (in the Indian case, primarily sterilization). There too, the Muslim minority is not fully participating in the reproductive changes, pointing again to the importance of culture as a mediating factor. The cultural milieu of Hindus in South India and Buddhists in Thailand may have more in common with each other than with many societies in the Middle East, Africa, or Latin America. A challenging task for future research will be to determine more precisely which particular constellations of cultural traits promote and which hinder a fertility response to the combined forces of socioeconomic change and the organized efforts to promote fertility control that are increasingly characterizing most of the developing world.

Notes

The research on which this paper is based was supported by a grant from the International Research Awards Program on the Determi- nants of Fertility. The awards program is sup- ported by USAID funds. Nibhon Debavalya and Aphichat Chamratrithirong served as co- principal investigators. Kritaya Archavanitkul, Worrawiman Chaiarut, and Rachanee Kalay- akhunavuth were of considerable help at dif- ferent stages of the work. The authors gratefully acknowledge the help of the following resident anthropologists for assisting in arranging focus group sessions in their villages and for offering comments on the sessions: Fern Ingersoll, Jas- per Ingersoll, Chai Podhisita, Michael Polio- dakis, and Bencha Yoddumnern. Peter Weldon and Sanchai Anumanrajathon of the Deemar

Company Ltd., Bangkok, generously provided advice on conducting focus group sessions.

1 Two recent anthropological studies by Lauro (1979) and Mougne (1981) are notable exceptions to this generalization.

2 A special issue of Studies in Family Planning (12, no. 12, December 1981, Part 1) describes some of these efforts and provides a general discussion of the technique. A fuller discussion is provided in a longer version of this study, issued as Research Report No. 83- 44 of the Population Studies Center, University of Michigan.

3 Construction site residents were chosen because of their rural background. No older men's session was held because of lack of male

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

326 Fertility Transition in Thailand

candidates meeting our age selection criteria. Some older women selected were not workers themselves but were living with their adult children, taking care of grandchildren. De- tailed descriptions of all the sample sites and the characteristics of the participants as well as the discussion guidelines are provided in the fuller report referred to in note 2.

4 1980 census results indicate that 86 per- cent of rural households have a radio, up from 64 percent in 1970 (National Statistical Office, 1982: Table 5). Unpublished results from a 1983 sample of 64 villages in rural Thailand surveyed by the Institute of Population Studies, Chulalongkorn University, found only two vil- lages with no television sets.

5 "Merit" refers to the Thai belief-de- rived from the Hindu-Buddhist doctrine of Karma-that all one's good conduct (merit) brings positive rewards while one's evil con- duct brings negative rewards, both in the pres- ent life and in subsequent incarnations (Pfanner and Ingersoll, 1962: 352-353).

6 The shortfall in some countries between predicted demand for contraception based on surveys and actual subsequent use once family planning programs were established has been attributed in part to the assumption inherent in survey questions that reproductive decisions were primarily controlled by the couple, es- pecially the wife (Warwick 1982: 115-117). For Thailand, not only does this assumption

seem valid but surveys showing considerable built-up demand proved to be quite good pre- dictors of the subsequent massive surge in con- traceptive use.

7 Such normative proscriptions regarding intercourse are thought to be widespread in some African and South Asian populations (Nag, 1983; Knodel, 1983; and Tan, 1983).

8 Caldwell et al. (1982b and 1982c) also note "problems peculiar to women," includ- ing lost work time, that enhance receptivity to sterilization and the limitation of family size in South India.

9 The influence of religion on the repro- ductive behavior of Thai Muslims is likely to be a very different matter. Although religious reasons for opposing small families or contra- ception did not surface in our study, the Mus- lim village we studied is probably atypical of the majority of such villages since it was Thai- rather than Malay-speaking. In contrast, recent focus group research conducted by a marketing firm on the acceptability of modern contracep- tion among Malay-speaking Muslims in the South found considerable opposition based on religious grounds (Deemar Company Ltd., 1983). While the influence of cultural differ- ences between Thai Buddhists and Thai Mus- lims is of considerable academic interest, it is of minor demographic significance nationally since Muslims constitute only 5 percent of Thailand's population.

References

Bongaarts, John, and Jane Menken. 1983. "The supply of children: A critical essay," in Determinants of Fertility in Developing Countries, vol. 1, Supply and Demand for Children, ed. R. A. Bulatao and R. D. Lee. New York: Academic Press.

Caldwell, John C. 1976. "Toward a restatement of demographic transition theory." Population and Development Review 2, nos. 3/4 (September/December): 321-366. . 1980. "Mass education as a determinant of the timing of fertility decline."

Population and Development Review 6, no. 2 (June): 225-255. 1982. Theory of Fertility Decline. New York: Academic Press.

,P. H. Reddy, and P. Caldwell. 1982a. "The micro approach in demographic investigation: Toward a methodology." Social Science Research Council and the Population Council, conference on innovative approaches to research on demographic and socioeconomic change in Latin America through the study of family and household, August 1982. , P. H. Reddy, and P. Caldwell. 1982b. "The causes of demographic change

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

J. Knodel / N. Havanon / A. Pramualratana 327

in rural South India: A micro approach." Population and Development Review 8, no. 4 (December): 689-727. , P. H. Reddy, and P. Caldwell. 1982c. "The determinants of fertility decline

in India." Unpublished manuscript. =- P. H. Reddy, and P. Caldwell. 1982d. "In search of a theory of fertility decline for India and Sri Lanka." Unpublished manuscript.

Davis, Kingsley. 1955. "Institutional patterns favoring high fertility in underdeveloped areas." Eugenics Quarterly 2: 33-39.

Deemar Company Limited. 1975. Rural Population Survey, Thailand 1975. Volume 1. A paper prepared for Westinghouse Population Center, Westinghouse Health Systems. Bangkok. . 1983. "Family planning-South: Qualitative report." A report prepared for

the United Nations Fund for Population Activities. Bangkok. Easterlin, Richard A. 1980. Birth and Fortune. New York: Basic Books. Goldstein, Sidney. 1970. "Religious fertility differentials in Thailand 1960." Popu-

lation Studies 24: 325-337. Jones, Gavin, and Yanee Soonthornthum. 1971. Fertility and Contraception in the

Rural South of Thailand. Bangkok: Manpower Planning Division, National Eco- nomic Development Board.

Kamnuansilpa, Peerasit, Aphichat Chamratrithirong, and John Knodel. 1982. "Thai- land's reproductive revolution: An update." International Family Planning Perspectives 8, no. 2 (June): 51-56. , Aphichat Chamratrithirong, and John Knodel. 1983. "Family planning and

fertility in the south of Thailand with a special emphasis on religious differentials: An analysis of data from the 1981 Contraceptive Prevalence Survey." Contra- ceptive Prevalence Surveys Further Analysis Report. Columbia, Maryland: West- inghouse Health Systems.

Kaplan, Irving. 1981. "The society and its environment," in Thailand: A Country Study, ed. Frederico Bunge. Washington, D.C.: U.S. Government Printing Office.

Knodel, John. 1983. "Natural fertility: Age patterns, levels and trends," in Deter- minants of Fertility in Developing Countries, Vol. 1, Supply and Demand for Children, ed. R. A. Bulatao and R. D. Lee. New York: Academic Press. , Anthony Bennett, and Suthorn Panyadilok. 1984. "Providing pills free: Does

it make a difference? Thailand's experience with a free pill policy. " Forthcoming in International Family Planning Perspectives. , Aphichat Chamratrithirong, Napaporn Chayovan, and Nibhon Debavalya.

1982. Fertility in Thailand: Trends, Differentials, and Proximate Determinants. Washington, D.C.: National Academy of Sciences.

Lauro, Donald. 1979. "The demography of a Thai village." Unpublished Ph.D. dissertation, Research School of Social Sciences, Australian National University, Canberra, Australia.

Mougne, Christine. 1981. "The social and economic correlates of demographic change in a Northern Thai community." Unpublished Ph.D. dissertation, School of Oriental and African Studies, University of London.

Nag, Moni. 1983. "The impact of sociocultural factors on breastfeeding and sexual behavior," in Determinants of Fertility in Developing Countries, Vol. 1, Supply and Demand for Children, ed. R. A. Bulatao and R. D. Lee. New York: Academic Press.

National Research Council. 1980. Fertility and Mortality Changes in Thailand, 1950-

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions

328 Fertility Transition in Thailand

1975. Panel on Thailand, Committee on Population and Demography. Wash- ington, D.C.: National Academy of Sciences.

National Statistical Office. 1982. Key Statistics of Thailand 1982. Bangkok: National Statistical Office.

Pfanner, David and Jasper Ingersoll. 1962. "Theravada Buddhism and village eco- nomic behavior." Journal of Asian Studies 3: 341-361.

Phillips, H. P. 1967. "Social contact vs. social promise in a Siamese village," in Peasant Society, ed. J. M. Potter, M. N. Diaz, and G. M. Foster. Boston: Little, Brown & Co.

Porapakkham, Yawarat, Somjai Pramanpol, and John Knodel. 1983. "Maternal and child health and family planning: Comparative study of Thai Buddhists and Thai Moslems, Jana District, Songkla." Paper presented at UNFPA Symposium on Fertility, Family Planning and Development Issues of Population in the South of Thailand, Hat Yai, 16-18 March, 1983.

Pramualratana, Anthony, Napaporn Havanon, and John Knodel. 1983. "Exploring the normative basis for age at marriage in Thailand: An example from focus group research in Thailand." Population Studies Center, University of Michigan, Research Report no. 83-50.

Prasithrathsin, Suchart. 1976. "Trends and differentials in fertility," in Population of Thailand, Economic and Social Commissions in Asia and the Pacific, Country Monograph Series, no. 3: 62-80.

Retherford, Robert D., Chintana Pejaranonda, Lee-Jay Cho, Apichat Chamratrithirong, and Fred Arnold. 1979. Own-Children Estimates of Fertility for Thailand Based on the 1970 Census. Papers of the East-West Institute, no. 63. Honolulu: East- West Population Institute.

Riley, James, and Santhat Sermsri. 1974. The Variegated Thai Medical System as a Context for Birth Control Services. Bangkok: Institute for Population and Social Research, Mahidol University.

Rosenfield, A. G., A. Bennett, Somsok Varakamin, and D. Lauro. 1982. "Thailand's family planning program: An Asian success story." International Family Plan- ning Perspectives 8, no. 2: 43-51.

Tan, Jee-Peng. 1983. "Marital fertility at older ages in Nepal, Bangladesh and Sri Lanka." Population Studies 37: 433-444.

Thomlinson, Ralph 1971. Thailand's Population: Facts, Trends, Problems, and Pol- icies. Bangkok: Thai Watana Panich Press.

Warwick, Donald P. 1982. Bitter Pills. New York: Cambridge University Press. Whitaker, Donald P. 1981. "The economy," in Thailand: A Country Study, ed.

Frederica Bunge. Washington, D.C.: U.S. Government Printing Office. Yoddumnern, Bencha. 1983. "Preliminary report on fertility behavior in a Northern

Thai village: Determinants and consequences." Unpublished manuscript pre- pared for the World Health Organization.

This content downloaded from 128.97.27.20 on Sun, 24 Jan 2016 21:40:14 UTC All use subject to JSTOR Terms and Conditions