Population. *Charles A. Weitz, tlvan G. Pawson, :j:M. Velma Weitz,. §Selwyn D. R. Lang, and §Ann Lang. [II]. IJ. *Department of Anthropology, Temple University, ...
Reprinted for private circulation from SoCIAL BIOLOGY Vol. 25, Fall 1978, pp. 17~195 Copyright © 1978 by the Society for the Study of Social Biology All rights reserved. Printed in U.S.A.
Cultural Factors Mfecting the Demographic Structure of a High Altitude Nepalese Population
[II]
IJ
*Charles A. Weitz, tlvan G. Pawson, :j:M. Velma Weitz, §Selwyn D. R. Lang, and §Ann Lang *Department of Anthropology, Temple University, Philadelphia, Pennsylvania; tG. W. Ho oper Foundation, University of California, San Francisco, California; +Philadelphia, Pennsylvania; and §Auckland, New Zealand
This study concerns the fertility of Sherpa and Tibetan women living at altitudes over 3,400 meters in Nepal. The average completed fertility (4. 77live births) and estimated crude birth rate (31 to 33 per 1,000) are low relative to low altitude peasant populations as well as to high altitude Andean peasants. Environmental phenomena (hypoxia, iodine deficiency) may be associated with retarded menarchial age and high infant mortality; but the major factors causing the low fertility appear to be cultural rather than environmental. Traditional ceremonial requirements delay the age at marriage until the mid or late twenties . Religious practices promote male and female celibacy. Migrant females and women married to migrant males report reduced fertility , probably because of poor nutrition and health care. Nonmigrant women living in villages that participate extensively in the cash economy have greater access to the growing market economy, health care, and education and report higher numbers of live births and fewer child deaths. ABSTRACT:
The inhabitants of the Dudh Kosi River Valley in northeastern Nepal, are generally referred to as Sherpas. The traditional homeland of the Sherpas can be divided into three areas: Solu, Pharak, and Khumbu (von Furer-Haimemdorf, 1964). Figure 1 shows the altitude and associated vegetation types of these areas. Most of the Khumbu population is concentrated in six major villages, all located above 3,400 meters and below the winter snow line (4,000 meters) . The population of these villages ranges between 250 and 800 inhabitants . Temporary settlements exist between 4,000 and 4,800 meters , but they are used only during the summer for agriculture and pasturage. Two large monasteries are located near the major villages, and a number of isolated homesteads are scattered throughout the area. In this region of severe climatic conditions and poor vegetation, the Sherpas
have traditionally depended upon a mixed herding and agricultural base, with trading activity carried out by a small percentage of the population. There is only one agricultural season, from early May to late August. Potatoes, along with some buckwheat and vegetables, are grown during this time. Although the potato is the dietary staple ofthe Sherpas, considerable time and effort are expended on the care of yaks and crossbred stock. They provide some meat and dairy products, but are primarily thought of as prestige items. Although most of the Khumbu's inhabitants share the same culture, differences in socioeconomic status exist within villages, and differences in acculturation exist between villages. These differences make it possible to examine whether the extreme environment or intragroup cuitural variation more significantly affects Khumbu demographic structure.
179
180
Weitz et al.
Social Biology
Nangpa-La Pass
6000m .
/frw./ SUMMER SNOW* LINE
5000m .
4000m.
Highest Summer Villages w1th Agriculture
3000m .
*
2000m .
Absolute Upper Border of Plant Life
FIG . I.-Habitation zones of the Khumbu region in Nepal
MATERIALS AND METHODS
Two teams independently collected demographic data during 1970 and 1971. The first survey was undertaken by two New Zealand physicians working at a hospital in one village. Both physicians had resided in the area for about six months before the survey was initiated and were well known. With two assistants trusted by local inhabitants, they conducted a house-to-house survey in the fall of 1970. An effort was made to contact and enumerate all inhabitants of the Khumbu . A preliminary report of this survey has been published elsewhere (Lang and Lang, 1971). A second team, together with different assistants, resurveyed two-thirds of one village and about one-fourth of the other villages in the winter and spring of 1971. In both surveys, respondents were asked to relate information on age, village of birth, village of spouse's birth, occupation, and number of marriages. Females
were asked to relate age at menarche, age at first pregnancy, number of pregnancies, number of live births, number of miscarriages and stillbirths, number of neonatal, infant, and child deaths and the number, age, and sex of living children. The resurvey also gathered data on clan affiliation, geneology, and age at marriage. An attempt was also made to expand mortality information beyond the childhood years. However, a disinclination to relate the names of dead adults prevented these data from being systematically gathered. The determination of age was often problematical, both because age categories are defined in a non-Western manner and because of the low literacy level. However, formal Western-style education in local schools enabled most individuals under the age of 30 to calculate their ages in terms of a Western calendar with reasonable accuracy (i.e., within one or two years). The ages of older individuals were determined by recall of important events
Vol. 25, No. 3
Nepalese Population
in the area or by relating them to the more accurately determined ages of their children . In a few cases, ages could not be determined. The possibility of falsified information or poor recall was assessed by comparing the responses of interviewees to the two questionnaires . Data on age corresponded closely until about 35 years . Thereafter, the reporting of age is less precise , with a one- to four-year difference existing between the two surveys. In cases where discrepancies existed , data from the more complete survey were utilized . Reporting of other factors showed only minimal variation , except for the number of pregnancies , mis carriages , and stillbirths. Either these events were not culturally recognized and therefore not remembered by women or they were falsely related to one or both of the survey teams. It is therefore not possible to include these data in the analysis . Not every woman replied to all fertility questions . Of all women who reported their ages to be 15 years and older, 348 (3 7. 0 per cent) provided menarchial data, 294 (31. 2 per cent) related the age at which they first became pregnant, and 420 (44. 6 per cent) replied to questions on the number of live births , neonatal deaths, infant deaths , child deaths , and the number of living children . For analytical purposes , fertility data supplied by Khumbu women were organized by age group , residence, and socioeconomic status. Each of these, in turn, were broken down into categories designed to reflect the impact of mountaineering on the local economy. In recent years, mountaineering jobs have begun to replace participation in the traditional agrarian economy as the major source of wealth and prestige . The presence of several high mountains (including Mt. Everest) in the Khumbu region has attracted many climbers, particularly
181
since the American Everest expedition in 1963. Because there are no roads, goods and equipment must be carried in by porters . The extreme cold and altitude of the area deters hiring lowlanders and has provided the Sherpas with a virtual monopoly on the activity. It is not uncommon for a well-outfitted mountaineering expedition b hire as many as 250 Sherpas for a month at a time. In addition, a growing number of tourists hire Sherpa guides and porters to hike with them into the Everest area. The need for porters, mountaineers, and tourist guides provides Sherpa men and women with the opportunity to earn high wages and has introduced them to new goods, values , and attitudes. The presence of a substantial cash reserve in the Khumbu region has induced lowlanders to make journeys of several days to bring food and dry goods to a weekly market. The imported food has provided many Sherpas with a more varied diet. Considerable sums of money are also spent on Western clothes, kerosene, and radios. Not all Khumbu villages have shared equally in the new cash economy. Of the six major population centers in the Khumbu , three have disproportionately large shares of the tourist and mountaineering income. These three acculturated villages are clustered around the major trekking route to Mt. Everest just after it enters the Khumbu . In these villages, almost SO per cent of the males aged 15 and older work primarily as expedition porters and guides , while only 22 per cent work mainly as traditional farmerherders. The weekly market is held in one of these villages and is only one hour's walk away from the other two . Westerntrained physicians have been situated permanently in one of the acculturated villages since 1966 and in another since 1970. All have permanent schools. The other three villages are more iso-
182
Weitz et al.
lated and are less directly involved in the cash economy and its social consequences. In these villages, only 21 per cent of the males aged 15 and older work primarily as expedition porters and guides, while 55 per cent engage mainly in farming and herding. Theseunacculturated villages are located at a greater distance from the weekly market (up to a day's journey), and most inhabitants have less cash to spend. Some of these villages do not have permanent schools, and none has a resident physician. In cases of medical emergency, a doctor from one of the acculturated villages may be summoned. The two Buddhist monasteries exist on the periphery of Sherpa economic activity. The inhabitants are supported primarily by donations, although some of the younger monks and nuns occasionally serve as porters. Because of the small sample of female respondents from the two monasteries (21), only two residence categories are used: acculturated villages and unacculturated villages. Over 90 per cent of all Khumbu women, age 15 and over, live in one ofthese two village types. Age categories were designed (1) to orient the period of female fertility (roughly ages 17 to 45) relative to the development of mountaineering, and (2) to be broad enough to minimize errors associated with age recall. Mountaineering expeditions have been a major feature of Khumbu life since the early 1960's. Consequently, most women age 50 and older would have ended their childbearing years before the major expeditions began. Cultural changes associated with mountaineering occurred after the beginning and before the completion of the fertility of women age 30 through 49. Most women age 29 and younger would have entered their fertile periods after these changes began. On the basis of the resurvey informa-
Social Biology
tion, we felt that two socioeconomic categories could be distinguished, using marital and migration information. The first, referred to as Socioeconomic Category 1, includes women, regardless of their village of birth, married to nonmigrant men and unmarried women living in the village of their birth. Marriage is a gradual process. It involves several ritual feasts and gift exchanges, culminating in a final ceremony called zendi (von-Furer-Haimendorf, 1964). This ritual tends to occur after the groom can persuade his father to provide the material necessary for building a house, land for agricultural production, and yaks for developing a herd. Traditionally, this happens when the couple is in their mid or late twenties, although it may even be later. When a man decides to yield part of his property to a son, that son remains in the village to enjoy the material benefits and status associated with the inheritance, as well as to participate in the important social relationships that tie him to his family. Although there is a preference for village endogamy, the Sherpas also practice a system of clan exogamy that sometimes makes it necessary for men to seek spouses in nearby villages. Under these circumstances, a woman may receive a dowry of land or goods in the village of her birth, but will move to the village of her husband and subsequently become economically and ritually a member of his family. Because the great majority of nonmigrant males possess property, and because their spouses (or unmarried daughters) have direct access to it, women in Socioeconomic Category 1 have a disproportionately large share of the Khumbu's material benefits. The second category, referred to as Socioeconomic Category 2, includes women, regardless of their village of birth, married
Vol. 25, No. 3
Nepalese Population
to migrant men and unmarried women who are not living in the village of their birth. In some cases, a man may move to another village to marry a woman who has no brothers and who consequently will inherit her father's property. Such marriages are relatively rare, however. More commonly, men (sometimes with their wives) move into a village where no family connections exist. Under these circumstances, access to land, animals, and material items is restricted to what migrants can afford to purchase. Often, this is very little. Most of the Khumbu's residents were born in the area and reside in the village of their birth or one nearby. About 10 per cent, however, were born in Tibet, and another 2-3 per cent were born in other regions of Nepal. Most Tibetan residents immigrated to the Khumbu in 1959 and were among over 6,000 migrants camping in the area at that time (von Furer-Haimendorf, 1964). In 1970 and 1971, only 287 Tibetans remained. They comprise
183
the bulk of Socioeconomic Category 2. Of the unmarried women in this category, almost 60 per cent are Tibetans . Of the married women in this category, most of whom are Tibetan themselves, 65 per cent have Tibetan spouses and another 8 per cent are married to men born in another part of Nepal, mostly the Solu or Pharak areas. Most Tibetans, regardless of their sex, occupy low positions in the Khumbu social hierarchy and maintain cultural patterns that are often distinguishable from those of larger Sherpa society. Tibetans tend to participate in the cash economy, rather than in the traditional Sherpa agrarian pursuits, and therefore reside primarily in acculturated villages . Many Tibetans in the acculturated villages are expedition guides or work at crafts they sell locally. A large number of Tibetans are also associated in some manner with local village Gompas (Buddhist temples) . Because the great majority of migrant males possess little property and because Tibetan migrants of both sexes are subject to social discrimination, women in Socio-
TABLE 1 AGE DISTRIBUTION OF POPULATION BY ETHNICITY , VILLAGE TYPE , AND SEX A CCU LTU RATE D
UNACCU L TURATED
VILLAG ES
VILLAG ES
A GE AND
M ONAST ERIES
E TH N I C
Males
Femal es
Males
Females
Age 0--14 Sherpas . . . ..... . ..... • . Tibetans .... .. .. . .... . .
237 2
231 2
157 3
199 0
7 5
9 0
840 12
Age 15-29 Sherpas ..... ... ...... . . Tibetans ... .. . . ... ... . .
156 16
156 13
132 0
164 10
16 2
2 5
626 46
Age 30--49 Sherpas ... . .... .. . . . . . . Tibetans ...•....•. .. . . •
108 35
123 28
130 8
147 11
25 10
9 14
542 106
Age SO+ Sherpas . ............ . .. Tibetans .......... . ... .
52 37
79 31
92 15
127 7
10 4
10 6
370 100
Age undeterminable Sherpas . . . . .. ...... .• . . Tibetans . .. .... . .... . . .
0 0
0 0
3 6
0 0
0 10
4 23
Total . . .... .. .. . .......
643
537
674
79
65
2,669
671
Males
Females
GROUP
TOTAL
184
Weitz et al.
economic Category 2 have limited access to the Khumbu's material benefits. RESULTS DISTRIBUTION OF THE POPULATION
Table 1 shows the residence distribution of Tibetans and Sherpas broken down by age and sex. It is apparent that most Tibetans have chosen to live in one of the acculturated villages, although a high per-
Social Biology
centage of Tibetan women (35 out of 150, or 23.3 per cent) live in one of the two monasteries . Figure 2 shows the age and sex distribution of Sherpas and Tibetans by five-year age groups. More females than males are present in most age groups. The major exception occurs among children under one year of age. The sex ratio among infants is 118 males per 100 females,
1 8o- 84 1 •
75-79 .
-
70-74 -
-
65-69 60 - 64 55-59 50 - 54 45-49 40 - 44
(/)
0:::
35-39
z
30-34
w
20-24
25-29
(!)
eluding and excluding the responses of never-married nullip arous women. In Socioeconomic Category 1, the average number of reported live births is much
greater when women at risk are excluded than when they are included . In Socioeconomic Category 2, however, the higher percentage of women at risk contributes to
190
Weitz et al.
Social Biology
TABLE 6 LIVE BIRTHS BY AGE A D SOCIOECONOMIC CATEGORY INCLUDING AND EXCLUDING NEVER-MARRIED NULLIPAROUS WOMEN LIVE BIRTH S
Socioeconomic Category I
Socioeconomic Category 2
AGE
No.
x
so
No.
x
so
t-VALUE
Age 29 and younger Including never-married and nulliparous women ....... Excluding never-married and nulliparous women .........
91
0.89
1.27
30
1.10
1.30
-0.74
40
2.03
1.19
19
1.74
1.24
0.63
116
4.11
2.60
52
3.33
2.36
I. 93*
106
4.50
2.37
52
3.33
2.36
2.93t
61
5.51
3.66
43
4.37
2. 75
1.81 *
54
6.22
3.34
42
4.47
2.69
2.84t
Age 30 through 49 Including never-married and nulliparous women Excluding never-married and nulliparous women .... Age 50 and over Including never-married and nulliparous women ....... Excluding never-married and nulliparous women ... .. .... • Significant at the 0.05 level of probability . t Significant at th e 0.01 level of probability .
a much smaller difference . Nevertheless, risk in Socioeconomic Category 2, regardwhen only the responses of women at risk less of age. were compared, differences in the average EFFECTS OF SOCIOECONOMIC CATEGORY number of live births between the two soAND RESIDENCE ON MORTALITY cioeconomic categories increased. Women at risk in Socioeconomic Category 1 reTable 7 shows the percentage of infant ported more live births than did women at deaths, child deaths, and surviving chilTABLE 7 PER CENT INFANT AND CHILD DEATHS ARRANGED BY AGE OF MOTHER, RESIDENCE , AND SOCIOECONOMIC CATEGORY A CCU LTURATED VILLAGES
Socioecono mi c Category I
UNACCULTURATED VILLAGES
Socioeconomi c Category 2
Socioeconomic Category 1
Socioeconomic Category 2 CHI-
Per
AGE
No.
Cent
No.
Per Cent
Age 29 and younger Infant deaths ......... Child deaths . . . ......... Surviving children . .. ....
48 48 48
15.6 4.4 77.8
19 19 19
15.4 0 84.6
43 43 43
16.7 5.6 77.7
11 II
14.3 0 85.7
Age 30 through 49 Infant deaths ........... Child deaths ..... . .. . . . . Surviving children.
66 66 66
17.0 8.0 75.0
35 35 35
IS. 2 8.0 76.8
so so so
14.8 15.9 69.3
17 17 17
14 .3 11.4 74.3
0 .271 3.884* 0.419
Age SO and over Infant deaths .... ....... Child deaths ............ Surviving children .. . .. . .
26 26 26
16.2 14.9 65.0
31 31 31
16.4 15.7 63.0
35 35 35
13 .2 25.3 57 .2
12 12 12
16.1 23 .2 59.0
0.475 4 .292* 0.548
*
Signifi cant at 0.05 level of probability.
Per
No.
Cent
Per
SQUARE
No.
Cent
VALUE
0.193 Not calc. 0 .681
11
Vol. 25, No. 3
Nepalese Population
191
here and by Pawson (1976), is considerably later than that of high-altitude Andean women (Cruz-Coke et al. , 1966). Recall problems may exist, and the frequency of women reporting fifteen and twenty years as their age at menarche is quite high . However, the existence of late menarchial age is supported by responses of younger women, who are more likely to remember the occurrence of menarche accurately, as well as by the results of probit and radiographic analyses carried out in an independent sample (Pawson, 197 6) . The cause of this extremely late menarchial age is presently unclear. The data presented here suggest that it is unaffected by the economic, nutritional, or health care differences associated with residence or socioeconomic category. Since these factors are associated with variable menarche atlow altitude (Petersen, 197 5), their lack of effect at high altitude may suggest the action of a more universal phenomenon . Noting that the menarchial age of low altitude Tibetans is equally late, Pawson (1976) posits a genetic basis. Ibbertson et al. (1971) have demonstrated that goiter and cretinism are endemic among the Sherpas and that both are associated with hypothyroidism. These conditions are known to retard sexual maturaDISCUSSION tion (Whipple, 1966) as well as physical FERTILITY growth and development (Greene, 197 3). The fertility of Sherpa women is some- Hypoxia, too, may operate to retard what low compared with peasant popula- menarche (Frisancho and Baker, 1970; tions at low altitude (Henry, 1961; Hoff, 1968) as part of its general effect on Weiner, 1964) as well as with other high- the pattern of growth and development at altitude groups (Hoff and Abelson, 1976). high altitude . The average completed fertility among A two- or three-year retardation in the women over 50 is fewer than five live average age at menarche may only parbirths , and the estimated crude birth rate tially explain the low fertility of Sherpa (31 to 33 per 1,000) is considerably below women. Of greater importance may be the that reported at high altitudes in South social requirements placed on the families America (between 46 and 56 per 1,000) by of betrothed couples. Traditionally, these Baker and Dutt (1972). requirements involved a series of gift exOne contributory factor may be the late changes and ceremonies that delayed maronset of menarche. The average age at riage until the mid or late twenties. menarche reported for the Sherpa, both Among the two older age groups, resurvey dren arranged by age of mother, socioeconomic category, and residence. Differences are analyzed by means of a chisquare test. In all three age groups , no statistically significant residence or socioeconomic related differences exist in the percentage of infant deaths . In the two older age groups, the percentage of child deaths reported by women from unacculturated villages is nearly twice as high as that reported by women from acculturated villages . Differences are statistically significant at the 0. 05 level of probability. Among children of women aged 29 and under, infant and child deaths are more common in Socioeconomic Category 1. However, differences between the two socioeconomic categories are not statistically significant. Among women in the two older age groups, a slightly higher percentage of surviving children is reported in acculturated villages. However, differences are not statistically significant. Among women in the youngest age group, survivorship is slightly higher among children of women in Socioeconomic Category 2. However, here too , the differences are not statistically significant.
192
Weitz et al.
data revealed that the average age at first marriage was 24.6 years, about eight years after the average age at menarche reported by these women . While pregnancies can and do occur before marriage with little or no social stigma, the majority seem to occur afterwards. Only a few of the nonmarried women surveyed here (primarily those in the youngest age group) reported having been pregnant. However, resurvey data revealed that the average age at first pregnancy (23 . 7 years) among women in the two older age groups was earlier than their average age at marriage (24.6 years) . Consequently, prenuptial sex may have been a factor increasing the fertility span somewhat. However, the majority oflive births still occurred after formal marriage. Information from women in the oldest category, those who have completed their fertility, indicates that less than 10 per cent of all live births occurred prior to age 25. On the other hand , almost 50 per cent of all live births occurred when these women were between the ages of 30 and 39 (Weitz , 1973). Since contraceptive methods are not used (Lang and Lang, 1971), premarital sexual encounters may not have been as frequent as others have suggested , even among the two older age groups. It is not unlikely that betrothed couples were often separated when they participated in their respective family's economic activities (herding, farming, trading). Since sexual access is not continuous until after formalized marriage , it seems reasonable to conclude that the apparent retardation in the age of peak pregnancy risk among older Sherpa females is a consequence of the late age at which formalized marriage occurred . The earlier age at first pregnancy reported by younger women, particularly those in Socioeconomic Category 2, probably reflects social changes resulting directly from the eco-
Social Biology
nomic impact of mountaineering and tourism. Sexual access is increased because the cash economy makes it easier to afford or circumvent the traditional expenses and requirements of marriage and neolocal residence. Many young couples, for example, no longer own animals or possess land . The cash they earn by trekking is used to purchase or rent a house, to buy food at the weekly market, and to acquire prestige at local ceremonies. While the late age at marriage seems to have delayed the onset of pregnancy among older women, it alone does not explain low Sherpa fertility. Among groups where the average age at marriage is not too different from that of the Sherpa, and where little or no sexual activity occurs before marriage , the average completed fertility is higher than that of the Sherpa-between five and seven live births (Eaton and Mayer, 1953; Friedl and Ellis , 1976) . To an important extent, low Sherpa fertility seems to be a consequence of the large number of never-married and nulliparous women in the K.humbu. This is particularly true among women past menarche and under age 30. Over 50 per cent of all women in this category who were born in the six main villages reported being both never-married and nulliparous. It is less the case among older women in these villages, only 6.6 per cent of whom reported being both never-married and nulliparous. The frequency of nulliparous women residing in the two monasteries seemed particularly high, regardless of age. Most of these women are nuns and claim to be celibate. Unfortunately, it was impossible to substantiate these claims, or to assess how their life style affects their pregnancy risk. Part of the explanation for low Sherpa fertility may also lie in intragroup cultural differences associated with residence and socioeconomic categories . Among the two
Vol. 25, No. 3
Nepalese Population
oldest age groups, women in Socioeconomic Category 2 report an average of one to two fewer live births than do women in Socioeconomic Category 1, even though their average ages at first pregnancy are similar. Only in the youngest group is there an absence of intragroup fertility dif:ierences . The low fertility of older women in Socioeconomic Category 2 could be related to underreporting . Many of the women in this group are Tibetans who , for social reasons, may have incorrectly related the number of live births. However, comparable figure s were obtained by both surveys, which was not the case for antenatal mortality data, where inaccurate reporting was clearly a factor. In addition, the number of living children, independently enumerated by the two surveys, was also low. Finally, data on infant and child mortality corresponded with information provided on live births and living children . For these reasons, it was concluded that underreporting was not a major factor associated with the low fertility of migrant women . Intragroup differences may also be related to varying patterns of sexual access. Migrant men , for example , are more likely to be engaged in long-distance trade than are nonmigrant men , thereby possibly affecting the parity of women in Socioeconomic Category 2. For the most part, however , separation periods are probably not too different between the two socioeconomic categories. Migrant men who engage in the cash economy by portering or trekking may work with their wives, just as do nonmigrant men . If couples do not work together, separations are just as long among migrants as among nonmigrants . Many migrant men serve as artisans and only rarely leave their wives . An alternative explanation may be the low socioeconomic status of migrants and its effect on the nutrition and health care
193
of pregnant women. Tibetan migrants, in particular, are reluctant to seek Western medical assistance. Among those migrants who do not participate in the cash economy, access to food and goods in the local market is limited. At low altitude, inadequate nutrition and poor health are known to retard intrauterine growth and development (Newman, 1975) and may reduce fertility by increasing antenatal mortality (Frisch, 19 75, 19 77). Similar conditions may occur at high altitude as well . Residence pattern appears to affect the number of live births only among women aged 30 to 49. The reasons may be associated with the cash economy that began to develop about ten years before this study took place . At that time, most women in the oldest age group would have ended their childbearing years . Any effect of acculturation on the fertility of this group, therefore , would have been minimal. All the youngest women began their fertile periods following the introduction of the cash economy. The absence of village-related fertility differences in this group is probably associated with an increasing participation in the cash economy, regardless of residence pattern. However, most women aged 30 to 49 were entering their most fertile period when the cash economy was first introduced . The degree to which members of this age group participated in the cash economy depended upon residence pattern . Residents of the three acculturated villages filled most of the early trekking jobs, while those Jiving in the three unacculturated villages continued to work in the traditional manner. Even today, fewer residents of the unacculturated villages engage in the cash economy. Of those that do , most are under age 30. Since participation in the cash economy bears directly on access to nutritional supplements available in the market and to health care, particularly drugs , fertility of women in
194
Weitz et al.
Social Biology
considerably higher in unacculturated villages, with no apparent effect attributable to differences in socioeconomic category. Village differences in access to nutritional MORTALITY supplements available in the local market The relatively high percentage of infant and poor health care again suggest themdeaths (15 .8 per cent of all live births) is selves as possible explanations. comparable to that reported at high altitude in Peru (Baker and Dutt, 1972) and CONCLUSIONS is apparently unaffected by differences in the age of the mother, residence pattern, Like other highland Nepalese groups or socioeconomic category. Nutritional (Goldstein, 1976), the population structure and health care differences, comparable of the Sherpa and Tibetan inhabitants of to those related to patterns of residence the Khumbu seems affected more by culand socioeconomic category among the ture than by features of the physical enviSherpa, are known to influence the infant ronment. The low fertility of the populamortality of low altitude populations tion is due, in part, to the interaction be(Petersen, 1975). Their apparent lack of tween social factors that delay the age at influence on infant deaths in the Khumbu marriage and those that promote nulliparmay suggest the action of some other fac- ity. Both function to maintain a large pool of women whose pregnancy risk is very tor, perhaps hypoxia. High infant mortality is characteristic of low. Even among women of marriageable high altitude groups (Alers, 1965; Cruz- age, voluntary celibacy (as Buddhist nuns) Coke et al., 1966; Hoff, 1968; Mazess, is common enough to significantly reduce 1965; Spector, 1971) . However, as Baker the reproductive capacity of the populaand Dutt (1972) point out, comparably tion. Furthermore, socioeconomic and reshigh infant mortality rates also exist idence factors may affect access to nutriamong many low altitude peasant popula- tion and/or medical care . Both are related tions. Like other peasant groups, inade- to intragroup fertility differences, and soquate diet and poor health care were and cioeconomic category is also related to difto some extent probably still are common ferences in child mortality. among the Sherpa. Continuous WesternOnly menarche and infant mortality style medical services were initiated only seem unaffected by socioeconomic and/or five years before our study took place. residence category. Retarded menarchial While the impact of Western medicine has age and high infant mortality are also rebeen appreciable, particularly in the ac- ported among high altitude Andean popuculturated villages, various childhood lations and therefore may be influenced by exanthemas as well as tuberculosis and hypoxia. However, it is equally likely that pneumonia are still common (Lang and they could be affected by endemic goiter Lang, 1971; McKinnon, 1968), and goiter and cretinism, by hypothyroidism, or by is endemic. These diseases are pervasive, the general level of nutrition and health occurring in all villages and in both socio- care. It is possible that the effect of hypoxia economic categories. Consequently, their effect on infant mortality may be equal to may be too subtle to detect with data presented here and perhaps may elude deor even greater than that of hypoxia. Among the two older groups, reported mographic techniques altogether. The child deaths (between ages 1 and 12) are primary influence of hypoxia probably ocunacculturated villages could have been affected in a manner similar to that described for migrants.
Vol. 25, No. 3
Nepalese Population
curs before birth, a period for which demographic information is rarely comp lete. Studies directed toward understanding intrauterine development at high altitude may therefore offer the best opportunity for assessing the possible influence of hypoxia on fertility. Postnatal events at high altitude in Nepal seem more directly related to cultural factors .
195
ACKNOWLEDGMENTS
We wish to express our appreciation to Sir Edmund Hillary for his support and encouragement of the initial survey. We wish also to thank the Sherpa and Tibetan inhabitants of the Khumbu region of Nepal for their cooperation . Financial assistance for the resurvey study was provided by the National Institute of General Medical Sciences, Training Grant No . 5-TOI-GM0-1748.
REFERENCES ALERS, J . 0. 1965 . Population and development in a Peruvian community. J . Inti. Amer. Stud. 7 :423428.
BAKER, P. T., and J . S. DUTT . 1972. Demographic variables as measures of biological adaptation: A case study of high altitude human populations . In G. A. Harrison and A. J. Boyce (eds .), The structure of human populations. Clarendon Press , Oxford. COALE, A. J., and P. DEMENY. 1966. Regional model life tables and stable populations. Princeton Univ . Press, Princeton , N.J. CRUZ-COKE, R., A. P . CRISTOFFANINI, M. ASPILLAGA, and F. BlANCAN!. 1966. Evolutionary forces in human populations in an environmental gradient in Arica, Chile. Hum . Bioi. 38:42 1-438 . DAVIS, K. 1951. The population of India and Pakistan . Princeton Univ. Press , Princeton, N.J. EATON , J . W., and A. J. MAYER. 1953. The social biology of very high fertility among the Hutterites. The demography of a unique population . Hum. Bioi. 25:206-264. FRIEDL, J., and W . S. ELLIS. 1976. Celibacy, late marriage and potential mates in a Swiss isolate. Hum . Bioi. 48:2 3-35. FRISANCHO, A. R., and P. T. Baker. 1970. Altitude and growth: A study of patterns of physical growth of a high altitude Peruvian Quechua population . Amer. J. Phys . Anthrop . 32:279-292 . FRISCH , R. 1975 . Demographic implications of the biological determinants of female fertility. Soc. Bioi. 22 : 17-22. - - - . 197 7. Population, food intake, and fertility. Science 199:22-30. GOLDSTEIN , M . 1976 . Fraternal polyandry and fertility in a high Himalayan valley in Northwest Nepal. Hum. Ecol. 4 :223-2 33. GREENE , L. S. 1973. Physi cal growth and development, neurologi cal maturation and behavioral functioning in two E~uadorian Andean communities in which goiter is endemic. 1. Outline of the problem of endemic goiter and cretinism. Physical growth and neurological maturation in the adult population of La Esperanta. Amer. J. Phys. Anthrop. 38:119-134 . HENRY , L. 1961. Some data on natural fertility. Eugen. Quart. 8 :81-91. HOFF, C. J . 1968. Reproduction and viability in a highland Peruvian Indian population . In High altitude adaptation in a Peruvian community. Occasional Papers in Anthropology, No. 1. Department
of Anthropology, Pennsylvania State University, University Park. HOFF , C. J., and A. E. ABELSON. 1976 . Fertility. In P . T . Baker and M . A. Little (eds.), Man in the Andes: A multidisciplinary study of high altitude Quechua. Dowden , Hutchinson, and Ross, Inc. , Stroudsburg, Pa. IBBERTSON, H. K., M. PERL, J. Mc KINNON , J . M . TARB , T . LIM, and M. B. GILL . 1971. Endemic cretinism in Nepal . In B . S. Hetsel and P. 0. Pharoah (eds.), Endemic cretinism. Institute of Human Biology, Papua, New Guinea. LANG , S. D . R. , and A. LANG. 1971. The Kunde hospital and a demographic survey of the Upper K.humbu, epal. New Zealand Med. J. 74: 1-8. LI CHTY , J . A., R . Y. TrNG, P . D . BRUNS , and E . DYAR . 1957 . Studies of babies born at high altitudes. Amer. J . Dis . Child. 93:666-678 . MAzEss, R. B . 1965 . Neonatal mortality and altitude in Peru . Amer. J. Phys. Anthrop. 23:209213.
Mc KINNON , J . R. 1968. Health problems of K.humbu in Nepal: The work of the Kunde hospital. New Zealand Med . J. 67: 140--143 . NEWMAN, M . T. 1975 . Nutritional adaptation in man . In A. Damon (ed.), Physiological anthropology. Oxford Univ. Press, New York . PAWSON , I. G. 1976. Growth and development in high altitude populations: A review of Ethiopian Peruvian and Nepalese studies. Proc. Roy. Soc., London (B.) 194:83-98. PETERSEN, W . 1975 . Population. Macmillan , New York . SPECTOR, R . M . 1971. Mortality characteristics of a high altitude Peruvian population. M. A. thesis , Pennsylvania State University, University Park. VON F URER-HAIMENDORF , C. 1964. The Sherpas of Nepal: Buddhist hi ghlanders . Oxford Book Co. , Calcutta. WEINER , J . S. 1964 . Human Ecology. Itt G. A. Harri son, J. S. Weiner, N. A. Barnicott, and J . M . Tanner (eds.), Human biology. Oxford Univ. Press, New York. WEISS , K. M . 1973 . Demographic models for anthropology . Society for American Archaeology, Memoirs, No. 27. Washington , D.C . WEITZ , C. A. 1973. Factors affecting the demographic structure of a high altitude Nepalese population . Paper presented at the 72nd annual meeting of the American Anthropological Association, New Orleans .
