Kingsley Davis

Social Structure and Fertility: An Analytic Framework

A striking feature of underdeveloped areas is that virtually all of them exhibit a much higher fertility than do urban-industrial societies. This welldocumented but insufficiently analyzed fact is known to be connected with profound differences in social organization as between the two types of society, and is therefore significant for the comparative sociology of reproduction. The clarity and importance of the contrast, however, should not be allowed to obscure the equally important fact that underdeveloped areas themselves differ markedly in social organization, and that these differences appear to bring about variations in fertility. Though the demographic statistics of backward regions have generally been so poor as to place in doubt the validity of reported differences, there are cases in which the evidence is reliable (e. g., as between Puerto Rico and Jamaica, or Arab Palestine and Ceylon). Of equal interest are the cases in which societies with differing social organization have the same level of fertility, for they may reach this common result by quite different institutional mechanisms. All told, ample opportunity exists for the comparative analysis of social structure as it affects fertility. In view of the bearing of future population trends on economic development, the pursuit of such analysis has a practical as well as a theoretical significance.

The World Demographic Transition

IEWED in long-run perspective, Vthe growth of the earth’s population has been like a long, thin powder fuse that burns slowly and haltingly until it finally reaches the charge and then explodes. For a million or more years our species increased with infinitesimal slowness, flourishing temporarily in some areas, hardly getting started at all in others. Throughout at least 99 per cent of its history it remained extremely sparse. Sustenance was obtained by hunting, fishing, and gathering, which required huge areas for few people, sometimes as much as 200 square miles per person.’ Not until the beginning of the Neolithic era, some eight to seventeen thousand years ago, when agriculture, domestication of animals, pottery, and textiles were invented, did greater density become possible. After that time cultural evolution moved at a faster pace, for eventually metallurgy and writing were invented and agriculture and transport improved; but still the world’s population, as distinct from that of particular areas, grew so slowly as to seem sta-

Urbanization and the Development of Pre-Industrial Areas

The process of urbanization, known to be intimately associated with economic development, deserves close attention if we are to understand the recent and future mechanisms of change in pre-industrial areas. Yet up to the present our comparative knowledge of cities and of urbanization is slight, particularly for underdeveloped areas. Considerable interest was shown in the subject toward the end of the last century in the works of Levasseur, Meuriot, and Adna Weber but since then, except for isolated cases such as Pirenne and Mark Jefferson, there has been little work done in the comparative analysis of cities and urbanization.

Final Note on a Case of Extreme Isolation

Anna, an extremely isolated girl described in 1940, died in 1942. By the time of her death she had made considerable progress, but she never achieved normality. Her slowness is probably explained by long isolation, poor training, and mental deficiency. Comparison with another case, a girl found in Ohio at the same age and under similar circumstances, suggests that Anna was deficient, and that, at least for some individuals, extreme isolation up to age six does not permanently impair socialization.

Extreme Social Isolation of a Child

A girl of more than five years was discovered incarcerated in an upstairs room. She had apparently been there since babyhood and was physically malnourished and apathetic as well as mentally blank. Taken first to county home, then to a foster-home, and finally to a school for detective children, she improved very slowly. She is still a virtually unsocialized creature, manifesting many parallels with other cases of isolated children and bearing out the Cooley-Mead-Dewey-Faris theory of socialization.

THE PATTERN OF MORTALITY CHANGE IN LATIN AMERICA

Using 69 new life tables recently made by Arriaga for Latin American countries by stable-population methods, the authors examine the mortality trends for more countries and more periods of history than have previously been available for analysis. For the late nineteenth and early twentieth centuries, the new tables yield a substantially lower life-expectancy than that shown by previously published life tables; for recent decades the difference is smaller, though in the same direction. As a consequence, the new tables show a speed of mortality decline in Latin America greater than the speed hitherto assumed. When the trend is analyzed in terms of economic development, it appears that the decline was extremely slow in the more backward Latin American countries until around 1930, whereas in the more advanced countries of the region, a more rapid decline had set in before that. After 1930, however, in both groups of countries the pace of decline was faster than ever, and it was virtually the same for both groups, suggesting that after that date public health measures were exerting a strong influence independently of local economic development. This result is confirmed by comparison with the past history of now developed countries; the mortality decline in Latin America after 1930 was much faster than it was historically at the same level in the industrial countries. As compared with other underdeveloped countries today, the unprecedented decline of mortality in Latin America is typical. In most underdeveloped countries, whether in Latin America or elsewhere, mortality change seems increasingly independent of economic improvement and more dependent on the importation of preventive medicine and public health from the industrial countries.

Human fertility in India.

Indias population problem might be solved if her traditionally high fertility could be reduced. The present analysis assesses the possibility of such a reduction in the next two or three decades. The birth rate in the past shows no definite downward trend. The rural-urban differential shows no increase in the gap between city and country. A study of religious and caste differentials according to social status, occupation, and literacy shows no displacement of institutional by deliberate controls. No downward trend is imminent under present conditions.

Illegitimacy and the Social Structure

There are two approaches to the study of illegitimacy: the social welfare approach and the sociological. The latter tries to understand the institutional norms which define certain births as illegitimate and to explain, in terms of the functional and structural connections of these norms with the rest of society, why illegitimacy occurs in spite of the norms, and why the illegitimate child and its parents have an inferior social status. This type of analysis reveals the basic defects in existent proposals for the elimination or the diminution of illegitimacy, and indicates that the simple measures that would be necessary for abolition will never be taken so long as the reproductive institutions of society are familial in character.

Influences Affecting Fertility in Urban and Rural Latin America

Preliminary results of recent surveys on fertility in rural and small urban areas in Latin America are compared with the results of previous surveys which concentrated on large Latin American cities. The large urban areas are Buenos Aires Mexico City Rio de Janeiro Bogota Caracas Panama and San Jose Costa Rica. The rural-small urban areas are Cauquenes-Chanco and Mostazal Chile; Cartagena and Neira Colombia; and Guelavia-Teotitlan and Pabellon Mexico. Although there are large differences in fertility among urban areas and to a lesser extent among rural-small urban areas there is a significant difference between urban and rural-small urban fertility as measured by average number of live births. No definite conclusions can be drawn between nuptial patterns and fertility levels. At best it can be said that fertility tends to be higher for consensually marrieds in cities with a higher proportion of common law marriages. There is a clearcut difference between the education level of urban and rural-small urban women and this shows up in fertility levels with illiterates having the highest fertility. The largest difference occurs at the completed primary level; differences for secondary and higher levels are much less marked and in some cities there is even a slight upturn at the university level. In urban areas fertility tends to decrease as the occupational status of the husband increases. In rural-small urban areas agricultural workers particularly lower level ones tend to have more children. In the cities there is a trend toward higher fertility among nonworking women; in the rural-small urban areas no clearcut differences in fertility exist perhaps because work in these areas is often done in the home and strongly linked to traditional modes of life. In the urban areas there is a much greater incidence of contraceptive use and to some extent this runs parallel with lower fertility. However the data does not distinguish between the effectiveness of different methods or the regularity with which they are used. The use of contraception increases with educational level.

India and Pakistan: The Demography of Partition

THE advent of independence in the Indian subcontinent caught the experts by surprise. There is consequently an extreme paucity of literature on the Union of India and Pakistan, and what does exist is characterized more by vagueness than by exactitude. Demographic data, for example, are scanty indeed. The governments of the Indian Union and of Pakistan have had little time to gather and publish statistics. If they take censuses in i95i, as they plan to do, information will be forthcoming. In the meantime one must depend on the I93i and I94I censuses, whose usefulness as sources of information concerning the two countries considered separately is limited by four circumstances. (i) Publication of the I94I census had to be curtailed drastically because of the war. Such volumes as have been published (representing only a small fraction of the data collected) are but thin wraiths compared to the large and comprehensive ones published for each previous census in a magnificent series beginning in I872. (2) In the award separating Pakistan from the Union of India, the boundary cut through certain districts. Since in the census volumes the statistics are not generally cross-tabulated for areas smaller than districts, the numbers and kinds of people living on each side of the boundary in the divided districts have to be estimated. (3) Certain of the princely states have until recently been in dispute, and Kashmir still is in dispute. (4) Use of the I94T census to deduce current population characteristics in India and Pakistan is affected by the mass migration which occurred after partition. Since accurate statistics on this migration could not be compiled, the roughest sort of estimates must be utilized. The main purpose of the present article is to give the results of computations that have not been made before. In most cases the conclusions are not new but are simply old ones stated more precisely. Some of the calculations support results reached by other quantitative investigators. A few suggest new conclusions heretofore unsuspected.