Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 381

22 Population and Society: Historical Trends and Future Prospects

CHARLES HIRSCHMAN

The statistics of population behavior in the mass dull the senses and elicit little response from are a dry topic treated in isolation, though they sociologists or the public at large. Yet, these possess their own fascination and rational struc- bare demographic facts reveal the absolutely ture. But they measure events, which are central to the life of men and women in all ages. Once amazing progress of humankind during the attention is turned outward from the events twentieth century. themselves to the social and economic environ- After a century with two world wars, the ment in which they occur, the appeal and Holocaust, a Cold War that legitimated the importance of demography is apparent. The theory of mutually assured destruction, the failed pressures of hard times and the opportunities of happier periods are reflected in historical demo- experiment of communist regimes, and the graphy like images in a camera obscura. The pic- spread of HIV/AIDS – to name only a few of the ture always needs interpretation and may lack twentieth-century horrors, it may seem counter- the polychrome fullness of historical reality but intuitive to speak of progress in modern times. it forms a clear and dependable outline to which In general, sociological accounts and theories color may be added as the population charac- teristics are related to their setting. (Wrigley, offer little respite from the general tendency to 1969: 28) point to the many failings of the modern world. As a discipline, sociology often focuses on the study of social problems, and there is a thin line between the sociological perspective and social INTRODUCTION criticism. Yet, a critical awareness of the many shortcomings of the contemporary world must The population of the world grew by 50 per be balanced with an historical awareness that cent from 1900 to 1950, and then increased most present-day societies are much less dan- by 200 per cent over the next 50 years to reach gerous, stratified and autocratic than the agrar- 6 billion just before the turn of the twenty-first ian world (and the early industrial world) of century. Even with fertility declining and a previous generations. slower rate of population growth in most Perhaps the best evidence of improvement in countries, the United Nations predicts a global the human condition is the demographic change population of about 9 billion in 2050 – eventually that has taken place over the past two-and-a-half stabilizing at about 10 billion by the end of the centuries. For all species, population growth is century. Large numbers, such as these, tend to the best measure of adaptation to local ecological Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 382

382 THE SAGE HANDBOOK OF SOCIOLOGY

niches. Throughout human history, the size of a organizational and technological means to apply local population was considered an index it, has been the engine that has allowed contem- of prosperity (Livi-Bacci, 1997: 1). At the eve of porary populations to far exceed any historical the first industrial revolution around 1750, the limits on subsistence. world’s population was about 770 million, reflect- Can this continue? Even with renewable ing an average growth rate of only slightly above sources of energy and materials, there would zero for the entire course of human history. be an eventual collision between continued From 1750 to 1950, the growth rate of the world’s population growth and a finite world. But the population ‘took off’, with an annual average collision may be averted because population rate of 0.7 per cent and reached a level of 2.5 growth has been slowing in recent decades, not billion by 1950 (Livi-Bacci, 1997: 31). Then, the from rising mortality, but because of voluntary world’s population literally exploded from 1950 controls on fertility. Fertility declines began to 2000, with a growth rate of over 1.8 per cent in several rapidly modernizing countries in the per year, and humanity now numbers more than late nineteenth century and spread to most 6 billion souls. Western countries in the early decades in the How did this happen? The most proximate twentieth century. Then quite unexpectedly, answer is that mortality began to decline in fertility declines began in several Asian coun- poor countries as well as rich ones with the tries in the 1960s and then spread to many control of infectious diseases through the dif- developing countries in Latin America and fusion of public heath programs and modern Africa in the last two decades of the century. If antibiotics. But there is a longer story that current trends continue, the world’s popula- includes the expansion of agricultural produc- tion will grow by only 50 per cent from 2000 tion and economic growth in many countries, to 2050 and then gradually level off at about the development of local and long-distance 10 billion. Although public policies may help communications and transportation systems, to speed the process, fertility transitions and a world system that included the World appear, in ways not fully understood, to be endo- Health Organization and international assis- genous to the process of socio-economic devel- tance to head off epidemics and famines. opment and improvements in the human Compared to what they should be or could condition. be, social and economic conditions around the Although much of the sociological interest in world are deplorable, but relative to past times, population trends relates to the consequences of the twentieth century, and especially the second current and future population trends (for half of it, looks pretty good. example, Can the world sustain a population Population growth is a two-edge sword. In of 10 billion? Do population growth and the short term, growth reflects good times, but density affect state formation and political if maintained for any period of time, increasing integration? What are the effects of population numbers put pressure on resources, especially growth and age composition on savings rates food production. In the world of nature, rapid and economic growth?), most demographic growth in the numbers of any species soon leads research has focused on the somewhat less to a population implosion, as the carrying capa- complicated questions of the determinants of city of the local environment is exceeded. Homo population trends, and in particular on mor- sapiens has had, on occasion, a few ways out tality, fertility, migration and urbanization. of the Malthusian trap of a population growing Demographers, most of whom are sociologists, faster than its subsistence. The first is migration, have accumulated a substantial reservoir of and much of human history has taken the empirical generalizations on all of these topics form of expansive societies spilling outwards to and have also generated several theoretical inter- nearby and distant lands. The second means is pretations of demographic transformations that technological and organizational change, which have accompanied the creation of the modern has allowed human societies to extract more world. This chapter is a preliminary overview subsistence and energy from natural resources. of some of the major research findings, empir- A spectacular growth in knowledge, and the ical generalizations and theoretical debates Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 383

POPULATION AND SOCIETY: HISTORICAL TRENDS AND FUTURE PROSPECTS 383

on the relationship between social change and Cavalli-Sforza, 1995: 122; Diamond, 1997: 341). population trends, with a primary focus on the This early population explosion, though modest twentieth century, which witnessed the most by modern standards, revealed the potential dramatic demographic changes in human reproductive capacity of human beings and the history. flexibility of human societies to respond to pop- ulation pressure by developing varied cultures adapted to local ecosystems. The most consequential response to popula- POPULATION GROWTH AND tion pressure in prehistory was the domestica- MIGRATION IN HUMAN HISTORY tion of plants and animals about 10,000 years ago (Harris, 1977: ch. 3). Although settled agri- Modern human beings (Homo sapiens) are cultural communities created the conditions the most recent branch of the hominids that for an expansion of population growth and the emerged in Africa around 100,000 years ago great civilizations of antiquity, the shift from (Cavalli-Sforza et al., 1994; Diamond, 1993). migratory foragers to farmers was probably The development of language (which was involuntary, and the last resort when populations dependent on the human ability to speak) had run out of all other options for survival. allowed human beings to accumulate and Even today, hunting and gathering populations share knowledge and to coordinate group generally resist efforts by states and missionaries actions in order to collect and hunt for food to settle them in permanent villages and adopt more effectively than any other large animal. farming. Agricultural populations generally work The ability to communicate through language longer hours and consume less than hunting- allowed humans to rise to the top of the food and-gathering populations, in addition to endur- chain and created the first population explo- ing the oppression from the ruling class that sion, or ‘great leap forward’,with early humans inevitably arises in societies with agriculture spreading to all corners of the globe, beginning (Harris, 1977; Boserup, 1981). around 50,000–60,000 years ago. Population pressures (perhaps accompanied McEvedy and Jones (1978: 14) estimate that by coercion from elites) led to the independent the average density of hunting and gathering ‘invention’ of agriculture in numerous places human populations was about 1 person per 10 from around 8500 to 3000 BC, including square kilometers, though it could have been Southwest Asia (the ‘Fertile Crescent’), China, somewhat higher in bountiful environments. New Guinea and multiple locations in the Even with modest levels of population growth, Americas and Africa (Diamond, 1997: 100). For most Paleolithic bands could have begun to a few generations, agriculture can provide a exceed the carrying capacity of their local envi- safety valve for population growth. For the same ronment within a few generations. Migration is unit of land, agricultural production can sup- probably the first human response to popula- port a much denser population than hunting tion pressure (Davis, 1974). As families moved and foraging. Agriculture also changes the con- to new areas, they had to learn how to adapt ditions affecting fertility and mortality dynam- to new climatic zones and to survive on differ- ics, including the motivation for children. ent flora and fauna. But human societies have Fertility levels in hunting and gathering proved extraordinarily adaptable, and within a societies were probably lower than in agricul- few tens of thousands of years, human settle- tural populations. Most pre-agricultural popu- ments had spread to most of the major regions lations are (were) migratory and mothers had of the world (Cavalli-Sforza et al., 1994; Davis, to carry small children for long distances. Since 1974; McNeill, 1984). Although archeological a woman could carry only a single child at a evidence is sketchy, human societies reached time, the optimal fertility pattern would be one Asia around 60,000 years ago, the Americas with long inter-birth intervals. Infanticide between 15,000 and 30,000 years ago, and finally was a common cultural practice in such popu- spread to some of the small Pacific Islands only lations if one birth followed too soon after within the last millennium (Cavalli-Sforza and another. The availability of edible plants and Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 384

384 THE SAGE HANDBOOK OF SOCIOLOGY

animals usually set an upper limit on the and extrapolations (Coale, 1974; Durand, population that could survive in a local envi- 1977; Livi-Bacci, 1997: 31; United Nations, 1973: ronment. If the population of a band exceeded ch. 2). From the origins of humankind around the carrying capacity of the local environment, 100,000 years ago to the dawn of agriculture the welfare of the community would be endan- (about 10,000 BC), the human population gered until some persons died or left the band. of the world increased to about 6 million. A stable population size was more adaptive for Although this figure may seem tiny, it included most hunting and gathering populations than tens of thousands of hunting and gathering a growing population. societies, whose presence had spread to virtually The higher level of fertility in agricultural all corners of Europe, Africa, Asia, Australia societies was fundamentally a response to the and the Americas. As more and more popula- higher incidence of mortality in the large tions became dependent on agriculture, human densely settled agrarian populations. This population densities increased in most regions: structural necessity (high fertility needed to the world’s population grew to approximately offset high mortality) was accompanied by a 250 million by the year 1 AD. About two-thirds variety of economic and social incentives for of this population lived in Asia (including the high levels of childbearing at the household Middle East, China and South Asia), but there level (Cavalli-Sforza and Cavalli-Sforza, 1995: were about 30 million in Europe, 25 million in 134). Children were a primary source of labor Africa and perhaps 12 million in the Americas. in agricultural communities. On farms, even The growth from 6 to 250 million over 10,000 small children can provide productive hands by years represents an average rate of less than watching over farm animals or doing routine 0.04 per cent per year. chores. An increase in the population of an Just as the world had probably reached its car- agricultural community could generally be rying capacity for hunting and gathering popu- absorbed through expansion as nearby lands lations at about 6 million (or 5–10 million) were cleared and brought into production. around 10,000 BC, agricultural populations The tendency of agrarian populations to probably oscillated with years of expansion spread to new areas, coincident with the dis- followed by contractions caused by famine, placement of hunting and gathering peoples, epidemics and wars. The estimated world pop- reveals the close parallels between demo- ulation growth rate over the next 1000 years graphic and social change. Analyses of the genetic of approximately zero does not mean that heritage of the contemporary populations of agricultural societies maintained a sustainable Europe reveals that outward migration of agri- balance with their environments. Indeed, culturalists from the Middle East gradually many early civilizations in the Middle East, populated most of Europe from about 9000 to Asia and the Americas experienced demographic 6000 years ago (Cavalli-Sforza and Cavalli- implosions as precarious irrigation systems Sforza, 1995: ch. 5). Although there were cer- or trade networks were destroyed because of tainly admixtures from both the indigenous invasions, plagues, or breakdowns in social Mesolithic hunter–gatherers and Neolithic systems. farmers, the higher densities of agricultural The population on all continents increased populations gave them a demographic edge from the year 1000 to 1750 at an average annual that probably led to military dominance and rate of about 0.01 per cent – about double greater contributions to the gene pool of sub- the estimated growth rate from 10,000 BC to sequent generations. This process has been the year 1 AD. Again, the historical trends are repeated in modern times with the expansion uneven, with decades or even centuries of of Europeans to Australia and the Americas, population decline followed by periods of Russians to Siberia, and Chinese to their southern expansion as ascendant empires developed frontiers. trade networks and lessened local warfare. Some Population data for ancient times are scholars have speculated that the diffusion of crude estimates based on limited information crops from the New World (potato, maize, Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 385

POPULATION AND SOCIETY: HISTORICAL TRENDS AND FUTURE PROSPECTS 385

manioc) contributed to the expanded food possibilities are present today than ever before production in Asia, Europe and Africa after 1500 in history. (Durand, 1977). European expansion, however, had deadly consequences for the peoples of the New World. The indigenous American peoples, POPULATION CHANGE IN THE estimated to number over 40 million in 1500, were TWENTIETH CENTURY decimated by European diseases and conquest. The next major turning point in world demographic history is the industrial revolu- Although world population size began to slope tion, which is usually dated with its origins upward in the eighteenth and nineteenth around 1750. It is more realistic to consider centuries, the major drama of world popula- industrialization as a process, which began in tion growth is a twentieth-century phenome- eighteenth-century England and spread through- non. The basic facts of twentieth-century out much of the world over the next two-and-a- population growth are presented in Table 22.1, half centuries. Indeed, the full weight of the with figures of the estimated population of the industrial and urban revolutions in Asia, Africa major world regions for 1900, 1950 and 2000, and Latin America is still in the future. The and the medium variant projections for 2050. breath-taking changes in scientific knowledge, The 1950, 2000 and 2050 figures are from the transportation and communications, economic most recent edition of the United Nations’s productivity and the structure and role of gov- World Population Prospects, and the estimates ernments have few precedents in human history. for 1900 are based on the medium value of the In turn, these social, economic and political estimated range reported by John Durand changes have transformed the conditions under (cited in United Nations, 1973: 21). which human communities live, with revolu- The three top panels in Table 22.1 show the tionary consequences for health and well-being. estimated populations, the percentage distri- Initially, the impact on human mortality was bution by major world regions, and the modest, with perhaps less frequent periods of estimated rates of annual population growth. crisis mortality in the nineteenth and early Trends in population growth over the twenti- twentieth centuries than in earlier eras. The eth century reveal the impact of varied social, major real gains in life expectancy occurred in economic and political changes across major the twentieth century, especially during the world regions as well as the diffusion of ideas second half of the century. and institutions designed to reduce mortality It is difficult to fully understand the enor- and morbidity. mity of the demographic consequences of the In 1900, Europe and North America were declines in mortality as the world’s population ascendant. Although comprising less than grew from 1 billion shortly after 1800 to 6 bil- one-third of the world’s population, European lion shortly before 2000. Livi-Bacci (1997: nations had, through their military might and 32–3), drawing upon the work by Bourgeois- industrial economies, colonized much of Asia Pichat, helps to convey these almost unfath- and Africa and organized a world economic omable magnitudes by interpreting current system that directed a disproportionate share population numbers as fractions of the human of the profits of global production (from plan- population that has ever lived on Earth. For tations, mines, trade and transportation) to example, the population alive in the year 2000 Europe. Demographically, Europe had experi- represents almost 8 per cent of the estimated enced a substantial recovery from the Black 82 billion humans ever born. If we were to Death of the fourteenth century, and population weight each birth by longevity, then the esti- growth rates accelerated with the developments mated person-years lived by those alive in 2000 in agriculture, industry and governance in the represent almost one-fifth of the person-years nineteenth century (Durand, 1977). lived since the origins of the species. Whatever In 1900, the greatest concentration of the the human potential for good or evil, more world’s population was in Asia, where more Calhoun-3267-22.qxd 5/24/20053:40PMPage386

Table 22.1 Estimates of population trends and medium variant projections of population size, fertility and mortality for the world and major world regions, 1900 to 2050 Population (000) Percentage distribution Ave. annual growth rate (%) 1900 1950 2000 2050 1900 1950 2000 2050 1900–50 1950–2000 2000–50 World 1608 2519 6057 9322 100 100 100 100 0.9 1.8 0.9 Africa 120 221 794 2000 7 9 13 21 1.2 2.6 1.8 Asia 915 1399 3672 5428 57 56 61 58 0.8 1.9 0.8 Europe 423 548 727 603 26 22 12 6 0.5 0.6 −0.4 Latin Am. and Caribbean 63 167 519 806 4 7 9 9 1.9 2.3 0.9 North America 81 172 314 437 5 7 5 5 1.5 1.2 0.7 Oceania 6 13 31 47 0 1 1 1 1.5 1.7 0.8

Total fertility rate Life expectancy at birth (both sexes) Estimated Projected Estimated Projected 1950–55 1995–2000 2045–50 1950–55 1995–2000 2045–50 World 5.01 2.82 2.15 46.5 65.0 76.0 Africa 6.71 5.27 2.39 37.8 51.4 69.5 Asia 5.88 2.70 2.08 41.3 65.8 77.1 Europe 2.66 1.41 1.81 65.7 73.2 80.8 Latin Am. and Caribbean 5.89 2.69 2.10 51.4 69.3 77.8 North America 3.47 2.00 2.08 68.9 76.7 82.7 Oceania 3.87 2.41 2.06 60.9 73.5 80.6 Source: United Nations, 1953: 11; 2001 Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 387

POPULATION AND SOCIETY: HISTORICAL TRENDS AND FUTURE PROSPECTS 387

than 900 million people lived (double the size panel of Table 22.1. For most of the twentieth of Europe) – almost six in ten of all humanity. century, the European population was growing The large Asian population reflects the legacy at a rate only half or one-third of that in other of history; even in antiquity, the populations world regions. During the second half of the of China and India numbered in the tens of twentieth century, Europe grew at 0.6 per millions (Durand, 1977). Fertile river valleys, cent per annum while Asia, Africa and Latin complex irrigation systems, extensive trading America grew at almost 2 per cent per year or systems, and strong states created agrarian civ- even higher. The medium level UN population ilizations with extensive peasant populations. projections from 2000 to 2050 show that Over the centuries, these civilizations expanded Europe will experience negative growth, most and contracted with accordion-like waves of other regions will be below 1 per cent, and demographic growth and decline. Asia had African rates will decline to just below 2 per fallen behind the European economic and cent per year. technological advances in the eighteenth and Because of the large initial populations and nineteenth centuries, but Asian populations had, the very high rates of growth in Asia, Africa overall, continued to grow during this period and Latin America, the projected levels of (Durand, 1977). absolute numbers are staggering. In 1900, the At the margins, with collectively less than population of Africa was about one-quarter one-fifth of the world’s population in 1900, that of Europe; by 2000, the two regions had were Africa, Latin America and northern about the same demographic weight, and by America (this UN classification puts Mexico 2050, Africa will be three times larger than and Central America in Latin America). Europe. The projected African population of Although thickly populated in places, Africa 2 billion in 2050 will be considerably larger than did not have the densely settled agrarian the world’s population in 1900. The popula- empires of Asia nor the several centuries of tion of Asia in 2000 is already twice the world’s economic expansion that Europe had experi- population a century earlier and even exceeds enced. After being virtually depopulated by the world’s 1950 population by 50 per cent. European diseases and conquest, the Americas Not only will Europe and other ‘developed had been growing rapidly by migration and regions’ represent a smaller share of the world’s natural increase for several centuries, but only population, they are likely to be smaller in comprised less than 10 per cent of the world’s absolute terms. The United Nations medium population in 1900. projections, based on rather conservative crite- During the twentieth century, there were ria, predict that Europe will have 100 million two dominant demographic trends: unprece- fewer persons in 2050 than in 2000. The power dented rates of population growth and a shift- of compound interest (or of geometric increase ing of the balance of the world’s population in Malthus’s memorable phase) can lead to to the poorer regions of Africa, Latin America huge increases in population size in only a few and Asia. This second trend will accelerate decades. This power is a valuable trait when during the twenty-first century. The combined a society is recovering from a major demo- populations of Europe and northern America graphic catastrophe, but it can lead to trouble were 30 per cent of the world’s population in in a world of fixed resources. 1900, 29 per cent in 1950, 17 per cent in 2000 How will the world be different with more and will be only 11 per cent in 2050. Over this persons from the poorer regions and less from 150 year period, Asia will roughly hold its own, the contemporary advanced countries? This is with almost 60 per cent of the world’s population, an important topic, but one that has hardly been the African share will triple from 7 to 21 per addressed in sociological research. In general, cent and the Latin American proportion will sociology as a discipline has been primarily double to 9 per cent. This redistribution of the focused on the study of the United States and a world’s population is reflected in the differential few other industrial societies (Hirschman, 2003). rates of population growth in the right-most The fields of international and comparative Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 388

388 THE SAGE HANDBOOK OF SOCIOLOGY

sociology will become increasingly important, with life expectancy of about 80 in Europe and as sociologists begin to address questions about northern America, the high 70s in Latin America the full range of contemporary and historical and almost 70 in Africa. These optimistic projec- societies. tions assume that present trends will continue, The contemporary period of population and that there will be no major societal break- growth is unique, not only because of the huge downs or demographic catastrophes. The numbers involved, but also because the values of HIV/AIDS pandemic has had catastrophic the components of growth have changed. effects on mortality in some African countries, Historically, population growth occurred under but its overall impact on world population conditions of high fertility and moderately high growth is fairly modest, perhaps lowering the mortality. The mortality declines of the twentieth rate of world population growth from 1.5 to 1.4 century have no parallel in human history. In around the year 2000 (Caldwell, 2000: 117). the context of very low mortality, even moder- These declines in mortality have been ate levels of fertility have given rise to substan- accompanied by parallel trends in fertility over tial population growth. The bottom panels of the twentieth century. Table 22.1 shows Total Table 22.1 show estimates of mortality and fertil- Fertility Rates (TFR), an index of period fertil- ity by world regions for the middle (1950–5) and ity for a population, constructed to correspond late (1995–2000) twentieth century with projec- to the average number of births per woman tions to the mid-twenty-first century prepared at the completion of childbearing. At mid- by the United Nations. century, women in developed countries had The mid-twentieth-century estimate of life two to four children – these rates were actually expectancy for the world as a whole, about somewhat higher than those prevailing in the 46 years, is an average of two wildly divergent 1930s because of the post-Second World War patterns. In the developed regions of northern baby boom. From 1870 to 1930, most Western America and Europe, life expectancy was 65 countries experienced fertility transitions of years or higher, while average life expectancy 50 per cent or more – from five to six children was 51 years in Latin America, 41 years in Asia per couple to only two to three. Given that and only 38 years in Africa. The range of mor- mortality was falling even faster, most Western tality in the poorer regions of the world at mid- countries experienced rapid population growth century, from life expectancies in the high 30s during the first half of the twentieth century. to the low 50s, were comparable to those pre- In contrast to declining fertility in Europe vailing in Europe and areas of European settle- and northern America during the first half of ment around 1900 (United Nations, 1973: 116). the twentieth century, fertility remained high The progress in lowering mortality in the devel- in Africa, Latin America and Asia. In the oped world during the first half of the twentieth mid-1950s, TFRs in Asia and Latin America century was a harbinger of comparable progress averaged just below six children per couple and in the developing world in the second half of almost seven children per couple in Africa. the twentieth century. At century’s end, average During the second half of the century, espe- life expectancy in Latin America and Asia had cially after 1970, fertility declined everywhere risen to 65 or higher, and was only a few years in the world. In advanced countries, the baby lower than the life expectancies in Europe and boom of the 1950s gave way to below replace- northern America, which stood in the mid-70s. ment level fertility (replacement level fertility Although significant absolute progress was is defined as two surviving children per woman registered in raising life expectancy in Africa in the reproductive age range). In Asia and Latin to about 51 years by the late 1990s, a huge gap America, fertility declined by more than 50 per in mortality remains between Africa and devel- cent during the second half of the century, with oped countries and even relative to most of average fertility of less than three children per Asia and Latin America. couple in the late 1990s. The United Nations mortality projections for Only in Africa does aggregate fertility the mid-twenty-first century predict convergence remain high. The UN estimates in Table 22.1 Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 389

POPULATION AND SOCIETY: HISTORICAL TRENDS AND FUTURE PROSPECTS 389

show that African fertility declined from 6.7 to due to accidents and violence. Although there 5.3 births per woman from the 1950s to the are many problems and great tragedies in indus- 1990s. The United Nations projects an acceler- trial societies, the retreat of death from the ating pace of the African fertility decline in the young and middle ages is a social achievement early decades of the twenty-first century, and that is all too rarely acknowledged. a predicted TFR of 2.4 births per woman at The conquest of high death rates was not a mid-century. For the balance of the world, fer- smooth trend over the millennia or even during tility projections hover around the replacement the first century of the industrial era. Rather, level for the mid-twenty-first century. The one progress in lowering mortality is a product of exception is Europe, where current fertility is the socio-economic development, and especially around 1.4 births per woman, and the projec- of scientific advances, in the twentieth century. tions suggest a recovery to a TFR of 1.8. Since For example, the germ theory of disease, which projections typically try to extrapolate recent laid the groundwork for modern medicine, changes into the future, there is much more only originated with the scientific work of the uncertainty when it is assumed that a lower Pasteurs in the late nineteenth century, and it asymptote (replacement level fertility) is being was some decades before the medical establish- approached or a change in direction is likely. ment adopted procedures that reduced the The theories behind these global changes are spread of infectious diseases. But we are getting addressed in the following sections on mortality ahead of the story. To understand why and how and fertility transitions. mortality declined in the twentieth century, we need to understand what factors kept mortality high throughout most of human history and the conditions that allowed for modest progress THE RETREAT OF MORTALITY before the twentieth century. Early theories postulated that the expansion Prior to the past 100 years, high ‘normal’ levels of of the food supply (with agriculture and domes- mortality and periodic episodes of crisis mortal- ticated animals) led to larger populations ity were part of the fabric of every society. Good because of improved nutrition and lower levels times without war, famine or epidemics meant of mortality (Childe, 1951[1936]). Agriculture that only 2–4 per cent of a population would die did lead to larger populations and more dense in a year. When disaster struck, as it periodically settlements, but not because of improvements did, up to 10 per cent of a population might in health and lower mortality. Indeed, most perish in a single year. In societies with ‘normal’ research points to the conclusion that hunting high death rates, parents could expect to lose and gathering peoples were healthier and had one or more of their children in infancy, and longer lives than settled agricultural popu- marriages were typically broken by the death of lations (Davis, 1974; Livi-Bacci, 1997: 37–47). one spouse before old age was reached. Oppression and high taxes by ruling elites By the middle of the twentieth century, aver- meant that peasant populations probably had age death rates in industrial societies were rou- poorer and less varied diets than hunters and tinely less than 1 per cent per year, and most gatherers. But the primary reason for higher children could expect to reach maturity experi- mortality in settled agricultural populations was encing only the occasional death of an elderly the spread of infectious diseases. relative. Nowadays, it is taken for granted that Initial encounters between parasitic micro- most parents can expect to live to see all of their organisms and human populations often lead to children reach adulthood and to anticipate grand- epidemic diseases that kill off a high percent- parenthood. It was not always so. Only a few age of the exposed population. Over time, epi- generations ago, infectious diseases were an ever- demic diseases often become endemic as the present danger that took the lives of people at all population develops resistance and the micro- ages and in all social classes. At present, the rare organism ‘adapts’ to become less lethal to its deaths of children and young adults are primarily host. Although susceptible to epidemics, most Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 390

390 THE SAGE HANDBOOK OF SOCIOLOGY

hunting and gathering bands were simply century. Although vaccination was slow to too small in population numbers to sustain spread throughout the world, the deadly scourge endemic diseases. Moreover, frequent move- of smallpox was dramatically reduced by 1900. ment from place to place provided some pro- The most influential interpretation of the tection from the environmental conditions overall decline in the nineteenth century is that favored breeding grounds for parasitic the improvement in nutrition hypothesis by microorganisms. Thomas McKeown (1976). Based on detailed Infectious diseases were much more likely to study of the decline in mortality in England spread when large numbers of humans (and and Wales of specific infectious diseases from animals) were living in close proximity. Water the mid-nineteenth to the mid-twentieth cen- sources in densely settled communities in pre- tury, McKeown shows that most of the declines modern times, especially cities, were generally preceded any effective medical treatment. polluted because pathogens in human waste Indeed, the mortality declines of many infec- contaminated rivers, streams and backyard wells. tious diseases occurred before there was any The development of urban sanitary systems scientific identification of the pathogen and and modest improvements in the living condi- understanding of transmission. McKeown tions of urban workers in the nineteenth century argues that the expansion in agricultural and were the essential first steps in lowering the livestock production, and especially trans- spread of communicable diseases in cities. One portation networks, allowed populations to of the few exceptions to the demographic sink- improve their diets and therefore develop holes that characterized premodern cities was greater resistance to disease. Critics have raised Japan. Cities in premodern Japan were much questions about McKeown’s data on the expan- larger (and healthier) than cities in Europe sion of food supplies and have suggested other because human waste was systematically col- possible reasons for the declines in mortality in lected for its value as fertilizer in Japanese agri- the nineteenth century, including the shift to culture (Hanley, 1987). cotton clothing, which allowed for more fre- Life expectancies in premodern times ranged quent washing of clothes and related improve- from the mid-20s to the mid-30s. Although ments in personal hygiene, including the there were systematic patterns in mortality widespread use of soap (Razzell, 1974). across time and between societies, there is little The magnitude of these early declines in evidence of long-term trends before the nine- mortality was relatively modest compared to teenth century. William McNeill (1976) has the much more dramatic declines in the twen- argued that the prevalence of plagues (the most tieth century. For example, the average gains common form of crisis mortality in agrarian in life expectancy in Western nations in the societies) was highly correlated with the rise nineteenth century were about 10 years of life and fall of empires and other epoch-making (from the high 30s to the high 40s), while the events in history. Since there was no scientific gains in the twentieth century averaged about understanding for the occurrence and spread 30 years of life (to the high 70s) (Livi-Bacci, of disease, effective means of control and super- 1997: 121). Moreover, the declines in mortality stition were often confounded. Cultures and spread beyond the industrial world to Latin religious teachings that encouraged care for America, Asia and Africa in the decades after the sick, frequent bathing and basic sanitation the Second World War. The gains in life could have had consequential impacts on expectancy in the developing world during the mortality and population growth. second half of the twentieth century rival those There appear to have been moderate registered in the developed world during the declines in mortality in Europe in the eighteenth first half of the century. and nineteenth centuries that preceded the Although some details remain in dispute, impact of any health and medical interven- the major cause for the worldwide progress in tions. Perhaps the one consequential interven- lowering mortality is the scientific knowledge tion was the vaccination against smallpox that has advanced medical science, public health developed by Jenner in the late eighteenth programs, and changes in behavior. Samuel Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 391

POPULATION AND SOCIETY: HISTORICAL TRENDS AND FUTURE PROSPECTS 391

Preston (1976: ch. 4) decomposed changes in in the United States declined by 30 per cent life expectancy from the 1930s to 1960s for all from 1940 to 1954, and almost all of the change countries in the world for which adequate mor- is attributable to declines in deaths due to tality data were available. Less than 20 per cent infectious diseases (Crimmins, 1981: 236). of the average world-wide progress in life After the conquest of infectious diseases in expectancy from the 1930s to the 1960s could industrial countries, it was expected that there be accounted for by increases in income. The would be only modest progress in lowering balance, more than 80 per cent of the gains in mortality because chronic conditions that pri- life expectancy, is due to other factors, such as marily affected the elderly (such as heart dis- modern medicines and vaccinations that were ease and cancer) were much less susceptible to widespread in the 1960s, but could not have the tools of modern medicine. Indeed, during been purchased for any price in the 1930s. the 1950s and 1960s there was a plateau in mor- The lack of scientific knowledge to control tality levels, especially among adults (Crimmins, mortality in past times is best illustrated by the 1981: 236). During the 1970s and 1980s, how- very modest differentials in infant and child ever, there were significant gains in life expec- mortality by social class in the United States tancy, including among persons above age 65 in 1900 (Preston and Haines, 1991: ch. 3). The (Crimmins et al., 1997). The reasons for this children of physicians experienced death rates continued progress are not completely under- higher than those of farmers (Preston and stood. New medications have been effective in Haines, 1991: 184–9). The major correlate of the treatment of high blood pressure and other child mortality around 1900 was urban resi- conditions that affect the elderly. There have dence. Although incomes were generally lower also been important, though not universal, in rural areas, the less dense environments of changes in lifestyles, such as reduced smoking rural areas provided a modicum of protection and increased exercise. Although some of the from the transmission of communicable air- earlier gains in life expectancy among the elderly borne and waterborne diseases that were ubiq- appeared to be primarily to increase the ‘years uitous in cities. Within the first few decades of living with a disability’, the latest evidence for the twentieth century, however, the mortality the 1980s suggests that most of the additional rates of children in urban middle class families years of added life are without disability dropped substantially, although the major break- (Crimmins et al., 1997). throughs in curative medicines did not appear After the Second World War, modern medi- until the 1940s and 1950s. Knowledge, derived cines and public health programs, often spon- on the germ theory of disease, allowed advan- sored by the World Health Organization and taged families to practice a number of sanitary other international organizations, spread around measures, such as pasteurizing milk and steril- the globe. The impact, in the words of Kingsley izing bottles, washing hands before preparing Davis (1956), was ‘an amazing decline in mor- meals and not allowing foods to become cont- tality’ in Asia, Latin America and Africa. Based aminated, which dramatically improved the on the summary measures reported in Table 22.1 survival chances of their children (Preston and (United Nations estimates from the mid-1950s Haines, 1991: 209). to the mid-1990s), life expectancy increased In the 1930s and the 1940s, sulfa drugs and by more than one-third in Latin America and antibiotics were discovered, which were fol- Africa, and by almost 70 per cent in Asia. The lowed by the development of other ‘miracle interventions responsible for most of these drugs’ to combat infectious diseases. For the dramatic declines in mortality are immuniza- first time in human history, medical science tions, antibiotics, insecticides, and improved was able to cure many illnesses quickly and rel- sources of purified water and sewage disposal atively inexpensively. These medical break- (Preston, 1980: 300). throughs were accompanied by a wide range of In the 1980s, there were claims of a slowdown preventive inoculations for infants and young in mortality progress in the developing world, children. The impact on health conditions and but the evidence seems to point to continued mortality has been incredible. The death rate declines in mortality (Hill and Pebley, 1989). Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 392

392 THE SAGE HANDBOOK OF SOCIOLOGY

There remain, however, wide gaps in health and FERTILITY TRANSITIONS mortality between the developing world and Western industrial countries, although the gap is One of the major turning points in human even wider between many of the rapidly mod- history, though its significance was only recog- ernizing countries in Asia and the very poor nized decades later, was the sustained decline regions of Africa and Latin America. Poverty and in human fertility that began in a number of malnutrition remain basic obstacles to improv- Western European countries and North America ing health in many regions in of the world. There in the late nineteenth century that eventuated in are also a number of major new health risks in replacement level fertility (about two surviving many countries, including the epidemic spread children per couple) within a span of 50–70 of HIV/AIDS and the breakdown in the health years. Although fertility levels had fluctuated care infrastructure in the former Soviet Union. throughout human history as resources waxed Estimates of those infected with HIV and and waned, the range of variation was from future projections vary widely. Bongaarts moderately high to very high levels, say from (1996: 22) relies on WHO estimates to report a four to eight children per couple. Motivations cumulative total of around 18.5 million adults for high levels of childbearing were part of the infected by HIV worldwide from the mid- core culture in all human societies. Not only was 1970s to 1995. Caldwell (2000: 117) estimates high fertility an essential requirement for soci- that by the end of 1999, 50 million persons will etal survival in the face of high mortality (and have died of AIDS or will be infected and will periodic waves of crisis mortality), but social likely die. The incidence and impact of AIDS institutions and cultural patterns were organized is particularly severe in sub-Saharan Africa, to support and encourage high levels of child- and especially in East and South Africa. In bearing within marriages. The sustained decline Zimbabwe, where the HIV prevalence rate in fertility that began in the late nineteenth exceeds 25 per cent (Caldwell, 2000: 118), adult century was not only a demographic revolution, mortality rose dramatically, with the likelihood but also the beginnings of a profound redefini- of death from age 15 to 60 rising from 0.20 in tion of the family and the adult roles of women 1982 to 0.50 in 1997 for females, and from 0.31 and men. Modern societies are still in the process to 0.65 for males (Feeney, 2001: 779). For sub- of adapting old (and creating new) institutions Saharan Africa as a whole, it is estimated that and gender roles in the wake of the transition to AIDS mortality raised the crude death rate low fertility. (CDR) by about a point (from 13.8 to 14.8) in About a hundred years after the beginnings 1995 and is projected to have a somewhat larger of fertility declines in Western Europe and impact in 2005 (raising the CDR from 11.2 North America, a similar process began in the to 12.7) (Bongaarts, 1996: 38). developing countries of Asia, Latin America Although there has not yet been a medical and Africa. This second wave of fertility transi- breakthrough to cure or prevent HIV infections, tions began in the late 1960s and early 1970s in the incidence of HIV/AIDS has been dramatically a few Asian countries and small island soci- reduced through behavioral changes (such as eties. By the 1990s, signs of declining fertility using condoms, needle exchanges) in a number had reached almost every part of the globe, of countries. Thailand has been an exemplary including areas of persistently high fertility in model of how government public health efforts South Asia and sub-Saharan Africa. Although and media attention can reduce the spread of these fertility transitions are still in process, the HIV/AIDS (Hanenberg et al., 1994). The pre- end is in sight. Replacement level fertility valence of HIV infection continued to rise in has been achieved in some East and Southeast sub-Saharan Africa in the last decade of the Asian countries in the 1980s and 1990s, and twentieth century, and there appears to be the United Nations predicts (medium variant) insufficient political will in many African coun- that almost all developing countries will reach tries to launch public health campaigns to arrest replacement level fertility by the middle of the the spread of HIV/AIDS (Caldwell, 2000). century (United Nations, 2001). Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 393

POPULATION AND SOCIETY: HISTORICAL TRENDS AND FUTURE PROSPECTS 393

Much of demographic science has been that could be linked with urbanization and devoted to documenting and measuring the industrialization were considered interchange- scope, character and pace of fertility transitions. able and equally valid as predictors of the tran- The pace of fertility transitions is affected by sition from high to low fertility. Although there the tempo and quantity of children born in were many potential hypotheses about specific reproductive unions: couples typically have two aspects of social change, such as the changing decades (or more) of potential childbearing, cost of children in rural and urban environ- and births can be accelerated or postponed ments, these were rarely differentiated from within the childbearing career. Translating these the broader story about industrialization and life cycle patterns of childbearing into aggregate urbanization. The empirical evaluations of the temporal patterns of fertility creates consider- theory were also quite general, typically with able analytical complexity that has stimulated parallel time series on indicators of modern- substantial efforts to formulate mathematical ization and declining fertility. Although there models that capture the significant aspects of were many empirical studies of fertility during fertility transitions. Even more complex and the 1950s and 1960s, often with anomalous daunting are the efforts to explain historical and patterns, there was no challenge to the theory contemporary fertility transitions in terms of of the demographic transition. the social, economic and cultural changes. In the 1970s and 1980s, two important devel- The earliest efforts to account for fertility opments in demographic research shattered declines in the West by Warren Thompson the hegemony of demographic transition theory. (1929), Kingsley Davis (1945) and Frank The first was surprising findings from the Notestein (1945, 1953) created what is gener- Princeton European Fertility Project, initially ally labelled as the theory of the demographic noted in an article by Knodel and van de Walle transition. Although sometimes glossed as simple (1979) and later developed in the volume by theory, which posits that modernization or Coale and Watkins (1986). Although the socioeconomic development leads to declining European Fertility Project was envisaged as an fertility, the original accounts of the theory spec- empirical test of demographic transition theory ified a variety of causal mechanisms, including on its home ground, the results showed that the the declining role of the family in economic pace of fertility decline across provinces and organization, the independence of women from regions of Europe was only modestly correlated traditional roles, and the shift to rationality with the socio-economic variables that played spurred by popular education. such a prominent role in the standard account Up until the 1970s, the theory of the demo- of demographic transition theory. Instead, the graphic transition was widely accepted by patterns and pace of fertility decline appeared to specialists and was widely presented in intro- be associated with regions that shared common ductory textbooks and beyond through stylized languages and culture. graphs and an interpretation of declining fer- The second challenge to demographic tility in response to the forces of industrializa- transition theory came from the comparative tion and urbanization. These processes had syntheses of results from the World Fertility occurred in many Western countries during Survey (WFS) project (Cleland and Hobcraft, the nineteenth and twentieth centuries and 1985, Cleland and Scott, 1987). The WFS were presumed to be on the horizon of many project consisted of a series of cross-sectional developing countries. Relative to most other studies of individual-level correlates of fertil- branches of the social sciences, the theory of ity behaviors and attitudes and contraceptive the demographic transition represented one practice in developing countries around the of the more successful efforts to explain social globe. Although these studies showed that, in behavior and social change. general, fertility was correlated in the expected The generality and universal scope of direction with female education, urban resi- demographic transition theory were, however, dence and other socio-economic variables, the sometimes a liability. Almost any indicators relationships were often modest and many Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 394

394 THE SAGE HANDBOOK OF SOCIOLOGY

exceptions could be found. Following on these The impact of public intervention, and family findings and research of Lesthaeghe (1983; planning programs, on fertility trends continues Lesthaeghe and Surkyn, 1988), John Cleland to be debated. The conventional wisdom, initially and Chris Wilson (1987) wrote a bold essay proposed in the classic study by Freedman and that directly challenged demographic transi- Berelson (1976), is that the combination of tion theory and suggested that an alternative effective family planning efforts and a favorable model of culture and fertility, labeled ideational socio-economic setting produce conditions theory, would be a more appropriate theoretical most likely to lead to lowered fertility. Sorting framework than demographic transition theory. out the independent and joint effects of setting Ideational theory holds that cultural values and policy has been remarkably elusive. The ini- have long-term consequences on human fertility tiation of family planning programs tends to be and are only slowly (and partially) eroded by endogenous to the process of development socio-economic changes. itself, and it is difficult to obtain independent Karen Mason (1997) cogently argued that empirical assessments of each. Successful gov- much of the debate on the causes of fertility ernments tend to have effective public pro- transitions is largely over variations in the grams, including well-managed family planning proximate conditions that influence the timing programs. Within countries, family planning of fertility declines, and that there is consider- clinics are not distributed randomly, but are able agreement over the long-term historical typically placed in areas of high fertility. Thus, factors, especially mortality decline, that have the bivariate association between proximity to led to fertility transitions. The portrayal of family planning services and fertility is usually demographic transition theory as a universal positive. The results of more complex multivari- model of modernization and fertility decline is ate models are heavily dependent on initial too general and vague (Hirschman, 1994; Kirk, assumptions and the analytical formulations. 1996), but there is a considerable body of evi- Several studies show only modest effects of fam- dence that socio-economic development has ily planning programs (Gertler and Molyneaux, been more influential in shaping historical 1994; Molyneaux and Gertler, 2000; Prichett, (Friedlander et al., 1991; Lee et al., 1994) and 1994), but others have reported more positive contemporary (Bongaarts and Watkins, 1996) assessments (Ahlburg and Diamond, 1996; fertility declines than many critics have Bongaarts et al., 1990; Tsui, 2001). acknowledged. There are, of course, consider- The end of fertility transitions was never able variations in the timing of the onset and defined beyond the general expectation that low the pace of fertility declines across popula- fertility would approach the replacement level tions, and across groups and regions within (around two children per couple) within some populations, which are often associated with modest range of fluctuation. This has generally linguistic and cultural factors. In particular, the been the case in the United States, when the total diffusion of knowledge of fertility control gen- fertility rate (births per woman) dropped a little erally follows cultural boundaries. The influ- below 2 in the mid-1970s, and then rose slightly ences of socio-economic and ideational factors to around 2 births per woman in the 1990s. In are complementary, not opposing, elements of Europe, however, fertility continued its down- an integrated theory of fertility change. ward descent, and by the late 1990s fertility was Fertility, and population growth more gener- well below the replacement level and showing no ally, responds to societal pressures that sign of changing (Frejka and Ross, 2001.) In threaten the survival and well-being of human some eastern and southern European countries, communities (Davis, 1963; Wilson and Airey, average fertility appears to be approaching one 1999). Socio-economic development is surely a child per couple. One school of thought holds major force influencing demographic processes that this is a temporary phenomenon, driven pri- in modern times, but it is not the only source marily by poor economic conditions and a tem- of population pressures that may generate porary rise in the average age of childbearing demographic, technological and social change (Bongaarts, 2001). Fertility expectations data in varied circumstances. show that most couples still desire to have two Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 395

POPULATION AND SOCIETY: HISTORICAL TRENDS AND FUTURE PROSPECTS 395

children. Other observers believe that the costs may appear to be almost continuous in many of childbearing (social and economic) are so historical accounts, there were undoubtedly long high in modern industrial societies that below stretches of normalcy over the long sweep of replacement fertility is likely to remain indefi- history. During such periods, the overwhelming nitely with the prospect of declining population majority of most peoples lived out their lives size in industrial societies (Chesnais, 2001). within a very narrow radius from their place of birth. This was especially true in agricultural societies, where investments in land and the sea- sonality of production tied the bulk of the pop- MIGRATION ulation to semi-permanent rural communities that spanned generations, with few internally The massive movements of peoples across the generated sources of social or cultural change. globe, including refugees fleeing war and political In every traditional society, there have instability, dispossessed peasants entering the always been groups that were prone to migra- urban proletariat and international migrants tion, including merchants, soldiers, and others seeking their fortune in distant lands, have given in the retinue of elites, and those belonging to rise to the claim of the distinctly modern ‘Age of deviant or persecuted ideologies. And in many Migration’ (Castles and Miller, 1998). Although traditional societies, there was usually a ‘float- the scope and volume of contemporary migra- ing population’ – persons who did not have tion are probably unique, human history is claims to land or social position. Depending replete with major waves of migration to nearby on the times and conditions, these people and distant regions (Davis, 1974; McNeill, 1984). served as a class of expendable labor who often The portrait of immobile communities joined the military expeditions or who sought wedded to the land and local villages over the refuge by settling the frontier. millennia, only to be disrupted by modern The modern era is distinctive because of higher civilizations and global capitalism is, at best, a levels of long-distance migration, which occur very partial view. Local communities did per- during periods of normalcy as well as during peri- sist over long stretches of human history, but ods of political turbulence. Beginning during the the peoples who inhabited them were con- eighteenth century, and accelerating during nected to frontier regions and cities through the nineteenth and twentieth, a series of techno- periodic waves of out-migration in times of logical, economic, social and demographic population surplus. Moreover, local villages forces have contributed to much higher levels and communities were periodically subjected of migration, both local and long-distance. to conquest that led to displacement, enslave- These movements are not just responses to cri- ment, or absorption into dominant popula- sis (though this condition continues in many tions. Patterns of uneven population growth, parts of the globe), but have become part of the environmental changes and technological dif- routine fabric of modern societies. ferences between populations inevitably made One of the most important distinctions migration, voluntary and involuntary, a recur- between the premodern and modern eras was rent feature throughout history. the widening geographic imbalance between The image of immobile populations and the demographic pressures and economic oppor- historical record of periodic large-scale popula- tunities, and the consequent awareness and tion movements can perhaps be reconciled with ability of persons to respond through migra- the distinction between two types of social tion. The acceleration of population growth, settings: normalcy and periods of political tur- especially in the nineteenth and twentieth bulence. Most of the moments of significant centuries, created immense strains in many human migration – especially of large numbers oversaturated agrarian societies, which were of people moving long distances – occurred becoming increasingly commercialized. From during periods of ecological catastrophe, famine, East Asia to Europe, successive cohorts of peas- revolution, war, or during political or economic ants faced not only poverty, which was part breakdowns. Although moments of turbulence of the normal order, but also the collapse (or Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 396

396 THE SAGE HANDBOOK OF SOCIOLOGY

weakening) of traditional feudal or semi-feudal observed that the ‘barbarian ideal of an ethnically economies and the moral order that tied peas- homogeneous nation is incompatible with ants to the land. The breakdown of traditional the normal population dynamics of civiliza- patron–client social institutions eliminated tion’, nationalist leaders considered open the safety nets in rural economies and acceler- borders to be inimical to the construction of ated out-migration to urban places or to dis- the ‘nation-state’. tant lands. Another factor that probably helped to turn Although most migration responses in the tide against an open system of interna- modern times are from rural villages to nearby tional migration was the increasing rate of urban areas (or rural frontiers), the spread of population growth. For most of the nineteenth long-distance communication and the cheapen- century, migration had been a necessity because ing of transportation costs meant that major of the high mortality in cities and a general waves of migration could develop on a global shortage of labor to settle the frontier and to scale. During the nineteenth century, a large work in the factories of the new industrial share of the surplus population that was being age. With declining levels of mortality in the shed by rural economies in Europe was absorbed early twentieth century, most countries were into the rural frontiers of America. In the later generally able to meet their labor needs from decades of the nineteenth and the opening natural increase. In such circumstances, nation- decades of the twentieth centuries, the dynamic alist and racial ideologies were, perhaps, given economies of the New World continued to be a a freer hand to legitimate restrictive immigra- major destination for redundant peasant labor tion policies. from Europe and to a lesser extent from Asia. The The nationalist impulse, which sought to limit migration to the New World was monumental, and control international migration, waned both in its demographic size and the diversity of during the last few decades of the twentieth its origins. For the 75 year period from the mid- century. The United States adopted a less nineteenth century to the end of the first quarter restrictive policy of immigration in the 1960s of the twentieth century, almost 50 million and there have been comparable developments Europeans migrated to the United States alone in other countries. In the early 1970s, Australia (Massey, 1988). Global patterns of labor migra- ended its ‘White Australia’ policy and allowed tion also developed in response to colonial needs significant numbers of Asians to immigrate. In for cheap labor in plantations, mining and other the early 1990s, the countries of the European extractive industries in Asia and Africa. Economic Community loosened restrictions Just as long-distance migration across inter- on inter-state migration. Citizens of any national borders was becoming a character- country in the EEC (now the European Union, istic feature of the emerging modern world, with 25 member states) can move to any other the door began to close. Modern states, legiti- country and are free to seek employment or mated by the potent ideology of nationalism, attend schooling on equal terms with natives began to issue passports and regulate who of the member state. entered their countries (Torpey, 2000). During These common patterns in a number of the nineteenth and twentieth centuries, the countries and regions are suggestive of under- political map of the world was transformed lying shifts in political, economic and demo- with the dissolution of empires, the rise and graphic forces. The contradictions between fall of imperialism, and the emergence of many tightly regulated international borders and the new states. In many cases, state formation was modern world economy are becoming increas- based on a claim of nationalism, which usually ing clear. The first sign was ‘labor demand’ in implied an ethnic homeland or a sense of industrial countries that could not be met belonging to a common people. Although there by domestic supply, at least not at the wages is no simple one-to-one correlation between offered. Employers found it more desirable to nationalism and the tightening up of inter- import labor from abroad than to raise wages national boundaries to migration, the logic was or to mechanize production. The demand similar. Although William McNeill (1984: 17) for ‘cheaper immigrant labor’ is evident in Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 397

POPULATION AND SOCIETY: HISTORICAL TRENDS AND FUTURE PROSPECTS 397

many sectors (agriculture, manufacturing, reasons for pushes and pulls are exogenous to construction, repair services, restaurants and the theory (Portes and Rumbaut, 1996: ch. 8). child care) in most industrial countries, Individuals, at least some individuals, do including a growing number of rapidly grow- respond to the uneven spatial distribution of ing developing countries. The increasingly opportunities by moving from one location global international economy seems to create to another, but this begs the question of why recurrent needs for labor greater than that there is an imbalance between the availability available from domestic population growth. of workers and the availability of economic The demand for immigrant labor is not opportunities across countries and across restricted to unskilled manual labor. The United regions within countries. States and other industrial countries have In a very ambitious essay, Wilber Zelinksy encountered a shortage of scientific and engi- (1971) proposed a theory of the mobility neering workers, particularly in the high-tech transition to parallel the theory of the demo- sector. This demand has been met, in part, graphic transition. According to Zelinsky, the by allowing many talented foreign students in pace of mobility increased from premodern American universities to convert their student to modernizing societies with the increasing visas to immigrant status. There has also been differentiation of geographical areas and shift a gradual shift over the past few decades to from agriculture to industry. As geographical more open immigration policies for a variety areas become more similar to one another in of reasons – refugees, agricultural workers, fully modern societies, Zelinsky predicts the pace ‘illegal’ immigrants with long residences in the of permanent migration will decline, but that country, people in countries that have few kin- circular migration would increase. Although ship ties to American citizens, and workers in Zelinsky has captured some important insights, high demand by US employers. The prejudices the theory of the mobility transition has not against immigrants and nativist fears have not become part of the theoretical corpus of modern entirely disappeared, but their open expression demography. Without an underlying biological has been sharply reduced in many modern model (as with mortality and fertility) and industrial societies. These changes in econom- with wide inter-societal variation in economic ics, demography, labor demand and ideology and political conditions (transportation sys- contributed to a much freer flow of interna- tems, labor demand, government policies, tional labor migration in the late twentieth and etc.), it has been impossible to formulate an early twenty-first centuries. abstract model of expected changes in migra- At the core of most theories of migration is tion levels and patterns over the course of an economic model which posits that people socio-economic development. move from places of labor surplus (low wages Between the narrow economic approach and high unemployment) to areas of labor and Zelinsky’s attempt at a grand theoretical scarcity (low unemployment and high wages). synthesis, there are a number of alternative Social attractions (and dis-attractions) are theoretical approaches to the study of historical added to economic motives as the central ele- and contemporary patterns of migration. But ments of the ‘push–pull model’ presented in with sharp differences in disciplinary approaches Everett Lee’s (1966) theory of migration. Lee and limited data sources on migration, the framed his theory as an update of Ravenstein’s general tendency is for fragmentation of the statement on the ‘laws of migration’,which was field into different research communities that first presented in 1885. In addition to noting espouse independent theories. However, in a the forces attracting and repelling migrants, Lee comprehensive review of the research litera- and Ravenstein provide a very helpful summary ture, Massey and his colleagues (1998) argue of the many other regularities of the types of that the major theories in the field of interna- migration and the characteristics of migrants tional migration are not mutually exclusive. In in different situations. contrast to the standard practice of ignoring or The primary limitation of most push–pull denigrating the utility of theoretical perspec- (economic) theories of migration is that the tives beyond one’s own field, they find empirical Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 398

398 THE SAGE HANDBOOK OF SOCIOLOGY

support for a variety of hypotheses, ranging Although demographic theories do not from neoclassical economics to world systems provide precise predictions of the threshold theory (Massey et al., 1998). This work suggests values of social conditions needed to affect demo- that the efforts at developing a comprehensive graphic change, there is a general consensus on theory of migration must seek to integrate the the factors that have shaped modern demo- leading ideas from different fields. graphic history (Mason, 1997; Preston, 1976). The single most important finding from the The growth of scientific knowledge, and the empirical literature, which is frequently redis- accompanying changes in production, health covered by researchers in different disciplines care and personal behavior, have led to much and areas, is the salience of collective forces on lower levels of mortality. And declines in mortal- individual decisions to migrate. There are some ity and changes in the role of the family have led self-directed persons – pioneers – who weigh to lower incentives for childbearing in modern the economic and social costs of migration and societies. Wide variations in demographic trajec- then set out on their journeys to distant and tories in different places and occasional reversals unknown lands. Much more numerous, how- do not negate the overall account. ever, are the followers, whose decisions to Even though absolute population attributes migrate are buoyed by the family and friends can only be predicted within wide bands of who have gone before. Return migrants at the uncertainty, there are a number of new demo- point of origin can provide information and graphic realities on the horizon. Demographic encouragement, and advance financial support momentum and the slower pace of demogra- for the journey. Even more important are phic transitions in some poorer regions in the friends and family at the destination site, who world will contribute to the current trend of an can provide temporary housing and sustenance increasing fraction of the world’s population as well as assistance in finding employment. living in Africa and Asia. Unless there are sig- This cumulative character of migration (Massey nificant increases in international migration et al., 1998) leads to chain patterns of migration to Europe and North America, the absolute, that link origin and destination communities, as well as the relative, size of the populations often over long distances and time periods. in industrial Western countries will shrink. Although current levels of economic welfare around the world are certain to change, the likely imbalance between larger numbers of people LOOKING AHEAD IN living in the poorer regions of the world and THE TWENTY-FIRST CENTURY fewer persons living in the traditional wealthier regions is likely to contribute to increasing polit- The major demographic trend of the twentieth ical tensions. Of course, it is not just relative century was the rapid expansion of the world’s numbers of persons in poor and rich countries population fueled by unprecedented declines that shape political discontent, but also the in mortality in both developed and developing awareness of differential levels of consumption countries. At the turn of the twenty-first and economic opportunities in countries at century, population growth was close to zero different levels of development. in most industrial countries (below zero in The long-term consequences of a population most of Europe) as a result of fertility transi- of 6 billion, rising to 10 billion, on the world’s tions that began more than a century ago. The resources and ecosystem, is largely unknown. global fertility transition spread to Asia, Latin The dramatic increase in population numbers America and Africa during the last third of the during the twentieth century appeared to have twentieth century. Although population growth only modest effects because scientific applica- rates remain high in most developing coun- tions increased agricultural and industrial pro- tries, continued declines in fertility and the duction while increasing utilization of the gradual shift to an older age structure will lead world’s supply of fossil fuels. But rising living to the cessation of world population growth standards (and consumption) of an increasing later in this century. fraction of the world’s population will add Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 399

POPULATION AND SOCIETY: HISTORICAL TRENDS AND FUTURE PROSPECTS 399

new pressures on the world’s supply of natural with the fraction over 65 rising from 5–6 per resources and food production. Perhaps, renew- cent in 2000 to about 18 per cent in 2050. The able resources and new energy supplies can be percentage of elderly in Africa will rise only developed to create a sustainable world econ- slightly from 2000 to 2050 because of the much omy and ecosystem, but this positive scenario slower pace of fertility decline. assumes rapid technological progress and a The rising share of the elderly (and declin- world political environment that can balance ing fraction of children) will have profound long and short-term demands as well as the effects on the economic, social and cultural interests of societies at different levels of devel- fabric of future human societies. One of the opment. If significant global climate change is defining attributes of modern welfare states is produced by human activities, the transition to the transfer of income from the working age a sustainable world ecosystem with 10 billion population to the dependent elderly, thereby people expecting high levels of consumption relieving individual families of the historic will be all the more difficult. burden of caring for elderly relatives. As the As noted earlier, many of the historical con- ratio of elderly to the working age population straints on long-distance migration were eased increases, the average cost (tax) per working in the last decades of the twentieth century. adult will increase, putting additional pressures Traditional immigration policies are residues on the fiscal system of modern welfare states. of the first half of the twentieth century, when It is possible to imagine rosy scenarios including regulated borders were a hallmark of the modern rapid economic growth, a healthier elderly popu- statecraft of nation-building states. Restrictive lation and delayed retirement that might reduce border policies originated in Europe and then the welfare state’s fiscal burden of supporting spread around the globe, including the tradi- an increasing share of dependent older popu- tional immigrant-receiving societies in the New lation. However, other factors, including the World and even to former colonial societies in increasing costs of health care and pharmaceu- Asia and Africa. tical drugs, the weakening of familial inter- These policies ‘worked’ because domestic generational support obligations (Lye et al., 1995), population growth in most countries was suffi- and the increasing political power of a larger cient to meet labor demand. Although rapid elderly population may work in the opposite population growth created immense pressures direction. The impact of population aging may in many labor surplus countries, there were few be even more acute in developing countries places that needed additional labor or allowed that have much weaker institutional infra- open migration. Passport controls were expen- structures and the health of the elderly popu- sive and irksome to many, but they became lation may be much more precarious (Palloni, accepted as normal features of modern states. 2001: 55–61). Over the past few decades of the twentieth The impact of an aging population on many century, however, strains in the system of tight social institutions may be equally significant. immigration policies were beginning to show. In general, older persons tend to be risk-averse In addition to larger numbers of people, and less willing to vote for taxes for education twenty-first-century human societies will also and other governmental programs that benefit have to adapt to a much older age composition children and younger persons. The consump- than ever before in human history. Population tion needs of the elderly will create different eco- aging is primarily determined by fertility decline, nomic demands and perhaps the pace of social and secondarily by increases in longevity. The and cultural change will slow as markets and very low levels of fertility and mortality in indus- institutions shift their priorities. Older persons trial countries have led to population fractions have been accorded high status and deference in of about 13–14 per cent above age 65 in the year traditional family systems in Asia and elsewhere, 2000, rising to almost 29 per cent in Europe and but these cultural values arose in demographic 21 per cent in North America by 2050 (National settings where the elderly have been relatively Research Council, 2001: 32). The changes in rare. The obligations of caring for the relatively Latin America and Asia will be even more rapid, few elderly were generally shared by a large Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 400

400 THE SAGE HANDBOOK OF SOCIOLOGY

number of adult children and their families. It NOTE remains to be seen whether these family values stay intact in settings with many more older rel- I thank Stewart Tolnay, Bussarawan Teerawichitchainan atives and when the cultural distance between and Susan Wierzbicki for their thoughtful comments on aging parents and their adult children has been an earlier draft. stretched by rapid social change. Family patterns have already begun to change with declining numbers of children REFERENCES and more elderly. If children are not antici- pated (or desired), there are fewer incentives for young couples to enter into formal mar- Ahlburg, Dennis A. and Diamond, Ian (1996) ‘Evaluating the impact of family planning pro- riages. It is increasingly common for young grams’, in Dennis A. Ahlburg, Allen C. Kelley and adults to live together in a common residence, Karen Oppenheim Mason (eds), The Impact of either as an alternative to, or as a prelude Population Growth on Well Being in Developing to marriage (Bumpass, 1990; Cherlin, 1992). Countries. Berlin: Springer. p. 299–336. Indeed, age at marriage has increased in the Bongaarts, John (1996) ‘Global trends in AIDS United States, although the average age of mortality’, Population and Development Review, cohabitation has not. Declining adult mortality 22: 21–45. during the twentieth century lengthened the Bongaarts, John (2001) ‘Fertility and reproductive average duration of married life with fewer preferences in post-transitional societies’, in marriages broken by spousal death at younger Rodolfo A. Bulatao and John B. Casterline (eds), ages. On the other hand, marriages have become Global Fertility Transition. Supplement to Vol. 27 of Population and Development Review. pp. 260–81. more unstable with the recent rise in the risk of Bongaarts, John, Parker Mauldin, W. and Phillips, divorce in many Western countries, especially in James F. (1990) ‘The demographic impact of family the decades after the Second World War. The planning programs’. Studies in Family Planning, increase in the divorce rate may be partially 21: 299–310. affected by demographic factors (longer marital Bongaarts, John and Watkins, Susan C. (1996) ‘Social duration, fewer children in marriages), but the interactions and contemporary fertility transitions’, long-term secular rise in divorce is probably Population and Development Review, 22: 639–82. because modern societies allow individuals Boserup, Ester (1981) Population and Technological greater freedom to act on individual preferences Change. Chicago: University of Chicago Press. (Preston and McDonald, 1979). Although indi- Bumpass, Larry (1990) ‘What’s happening to the vidual marriages may be more fragile, most family? Interactions between demographic and institutional change’, Demography, 27: 483–98. divorced persons remarry, suggesting that dis- Caldwell, John C. (2000) ‘Rethinking the African AIDS satisfaction with a particular partner does not epidemic’, Population and Development Review, signify any declining interest in marriage as a 26: 117–35. social institution. These changes in marriage and Castles, Stephen and Miller, Mark J. (1998) The Age family life are likely to continue in the coming of Migration: International Population Movements decades and probably spread to countries that in the Modern World, 2nd edn. New York: The are in the early stages of industrial development. Guilford Press. As the opening quotation by Anthony Wrigley Cavalli-Sforza, L. Luca, Menozzi, Paolo and Piazza, suggests, the demographic picture always needs Alberto (1994) The History of Geography and interpretation and may lack the polychrome full- Human Genes, abridged edn. Princeton, NJ: ness of historical reality, but even in black and Princeton Unviersity Press. Cavalli-Sforza, Luigi Luca and Cavalli-Sforza, white, demographic patterns and social demo- Francesco (1995) The Great Human Diasporas: The graphic analyses remain a central prism to History of Diversity and Evolution. Cambridge, MA: perceive and to understand social change and Perseus Books. human welfare. This is the reason why the socio- Cherlin, Andrew (1992) Marriage, Divorce, Remarriage, logical imagination must always remain tethered revised and enlarged edn. Cambridge, MA: Harvard to its demographic anchor. University Press. Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 401

POPULATION AND SOCIETY: HISTORICAL TRENDS AND FUTURE PROSPECTS 401

Chesnais, Jean-Claude (2001) ‘Comment: A march Freedman, Ronald and Berelson, Bernard (1976) to population recession’, in Rodolfo A. Bulatao ‘The record of family planning programs’, Studies and John B. Casterline (eds), Global Fertility in Family Planning, 7 (1): 1–40. Transition. Supplement to Vol. 27 of Population Frejka, Thomas and Ross, John (2001) ‘Paths to sub- and Development Review. pp. 255–9. replacement fertility: the empirical evidence’, in Childe, V. Gordon (1951[1936]) Man Makes Global Fertility Transition, ed. Rodolfo A. Bulatao Himself. New York: New American Library. and John B. Casterline. Supplement to Vol. 27 of Cleland, John and Hobcraft, John (eds) (1985) Population and Development Review. pp. 213–54. Reproductive Change in Developing Countries: Friedlander, Dov, Schellekens, Jona and Ben-Moshe, Insights from the World Fertility Survey.Oxford: Eliahu (1991) ‘The transition from high to low Oxford University Press. marital fertility: cultural or socioeconomic deter- Cleland, John and Scott, Christopher (1987) The minants’, Economic Development and Cultural World Fertility Survey: An Assessment.New York: Change, 39: 331–51. Oxford University Press. Gertler, Paul J. and Molyneaux, John W. (1994) Cleland, John and Wilson, Chris (1987) ‘Demand ‘How economic development and family planning theories of the fertility transition: an iconoclastic programs combined to reduce Indonesian fertility’, view’, Population Studies, 41: 5–30. Demography, 31: 33–63. Coale, Ansley J. (1974) ‘The History of Human Hanenberg, R.S., Rojanapithayakorn, W., Kunasol, Population’,in The Human Population (A Scientific P. and Sokal, D. (1994) ‘Impact of Thailand’s American Book). San Francisco: W.H. Freeman. HIV-control programme as indicated by the pp. 15–25. decline of sexually transmitted diseases’, Lancet, Coale, Ansley J. and Watkins, Susan Cotts (1986) 344: 243–5. The Decline of Fertility in Europe. Princeton, NJ: Hanley, Susan (1987) ‘Urban sanitation in preindus- Princeton University Press. trial Japan’, Journal of Interdisciplinary History, Crimmins, Eileen M. (1981) ‘The changing pattern of 18: 1–26. American mortality decline, 1940–77, and its impli- Harris, Marvin (1977) Cannibals and Kings: The cations for the future’, Population and Development Origins of Culture. New York: Random House. Review, 7: 229–54. Hill, Kenneth and Pebley, Anne R. (1989) ‘Child Crimmins, Eileen M., Saito, Yasuhiko and Ingegneri, mortality in the developing world’, Population and Dominique (1997) ‘Trends in disability-free life Development Review, 15: 657–87. expectancy in the United States, 1970–90’, Population Hirschman, Charles (1994) ‘Why fertility changes’, and Development Review, 23: 555–72. Annual Review of Sociology, 20: 203–33. Davis, Kingsley (1945) ‘The world demographic tran- Hirschman, Charles (2003) ‘The development of the sition’, Annals of the American Academy of Political social sciences prior to globalization and some and Social Science, 237: 1–11. thoughts on the future’, Southeast Asian Studies Davis, Kingsley (1956) ‘The amazing decline of mor- for the Twenty-First Century: Pacific Perspectives, tality in underdeveloped areas’, American Economic ed. Anthony Reid. Temple, AZ: Program for Review, 46: 305–18. Southeast Asian Studies, Arizona State University. Davis, Kingsley (1963) ‘The theory of change pp. 157–78. and response in modern demographic history’, Kirk, Dudley (1996) ‘Demographic transition Population Index, 29: 345–66. theory’, Population Studies, 50: 361–87. Davis, Kingsley (1974) ‘The Migrations of Human Knodel, John and van de Walle, Etienne (1979) Populations’,in The Human Population (A Scientific ‘Lessons from the past: policy implications of American Book). San Francisco: W.H. Freeman. historical population studies’, Population and pp 53–65. Development Review, 5: 217–46. Diamond, Jared (1993) The Third Chimpanzee. Lee, Everett (1966) ‘A theory of migration’, Demo- New York: Harper Perennial Library. graphy, 3: 47–57. Diamond, Jared (1997) Guns, Germs, and Steel: The Lee, Ronald D., Galloway, Patrick R. and Hammel, Fates of Human Societies.New York:W.W.Norton. Eugene A. (1994) ‘Fertility decline in Prussia: esti- Durand, John D. (1977) ‘Historical estimates of world mating influences on supply, demand, and degree population: an evaluation’, Population and Develop- of control’, Demography, 31: 347–73. ment Review, 3 (3): 253–96. Lesthaeghe, Ron (1983) ‘A century of demographic Feeney, Griffith (2001) ‘The impact of HIV/AIDS and cultural change in Western Europe: an explo- on adult mortality in Zimbabwe’, Population and ration of underlying dimensions’, Population and Development Review, 27: 771–80. Development Review, 9: 411–35. Calhoun-3267-22.qxd 5/24/2005 3:40 PM Page 402

402 THE SAGE HANDBOOK OF SOCIOLOGY

Lesthaeghe, Ron and Surkyn, Johan (1988) ‘Cultural Preston, Samuel H. (1976) Mortality Patterns in dynamics and economic theories of fertility change’, National Populations. New York: Academic Press. Population and Development Review, 14: 1–45. Preston, Samuel H. (1980) ‘Causes and conse- Livi-Bacci, Massimo (1997) A Concise History of World quences of mortality declines in less developed Population, 2nd edn. Malden, MA: Blackwell. countries during the twentieth century’,in Richard, Lye, Diane N., Kelpinger, Daniel H., Hyle, Patricia Easterlin (ed.), Population and Economic Change and Nelson, Anjanette (1995) ‘Childhood living in Developing Countries. Chicago: University of arrangements and adult children’s relations with Chicago Press. pp. 289–360. their parents’, Demography, 32: 261–80. Preston, Samuel H. and Haines, Michael R. (1991) Mason, Karen (1997) ‘Explaining fertility transitions’, Fatal Years: Child Mortality in Late Nineteenth Demography, 34: 443–54. Century. Princeton, NJ: Princeton University Press. Massey, Douglas S. (1988) ‘Economic development Preston, Samuel H. and McDonald, John (1979) and international migration in comparative ‘The incidence of divorce within cohorts of perspective’, Population and Development Review, American marriages contracted since the Civil 14: 383–413. War’, Demography, 16: 1–26. Massey, Douglas S., Arnago, Joaquin, Hugo, Graeme, Pritchett, Lant H. (1994) ‘Desired fertility and the Kouaouci, Ali, Pellegrino, Adela and Taylor, J. impact of population policies’, Population and Edward (1998) Worlds in Motion: Understanding Development Review, 20: 1–55. International Migration at the End of the Razzell, P.E. (1974) ‘An interpretation of the modern Millennium. Oxford: The Clarendon Press. rise of population in Europe: a critique’, Population McEvedy, Colin and Jones, Richard (1978) Atlas of Studies, 28: 5–17. World Population History. New York: Penguin. Thompson, Warren (1929) ‘Population’, American McKeown, Thomas (1976) The Modern Rise of Journal of Sociology, 34: 959–75. Population. New York: Academic Press. Torpey, John (2000) The Invention of the Passport: McNeill, William H. (1976) Plagues and Peoples. Surveillance, Citizenship, and the State. Cambridge: Garden City, NY: Anchor. Cambridge University Press. McNeill, William H. (1984) ‘Human migration in Tsui, Amy (2001) ‘Population policies, family historical perspective’, Population and Development planning programs, and fertility: the record’, Review, 10: 1–18. in Rodolfo A. Bulatao and John B. Casterline Molyneaux, John and Gertler, Paul (2000) ‘The (eds), Global Fertility Transition. Supplement to impact of targeted family planning programs Vol. 27 of Population and Development Review. in Indonesia’, in C.Y. Cyrus Chu and Ronald Lee pp. 184–204. (eds), Population and Economic Change in East Asia. United Nations (1953) The Determinants and Supplement to vol. 26 of Population and Develop- Consequences of Population Trends.New York: ment Review. pp. 61–85. United Nations. National Research Council (2001) Preparing for An United Nations (1973) The Determinants and Aging World: The Case for Cross National Research. Consequences of Population Trends: New Summary Washington, DC: National Academy Press. of Findings of Demographic, Economic, and Social Notestein, Frank W. (1945) ‘Population – the long Factor. New York: United Nations. view’, in T.W. Schultz. ed. Food for the World, United Nations (2001) World Population Prospects: Chicago: University of Chicago Press. pp. 36–57. The 2000 Revision. Volume 1: Comprehensive Tables. Notestein, Frank W. (1953) ‘Economic problems of New York: United Nations. population change’, in Proceedings of the Eighth Wilson, Chris and Airey, Pauline (1999) ‘How can a International Conference of Agricultural Economists. homeostatic perspective enhance demographic New York: Oxford University Press. pp. 13–31. transition theory?’, Population Studies, 53: 117–28. Palloni, Alberto (2001) ‘Living arrangements of Wrigley, E.A. (1969) Population and History. older persons’, Population Bulletin of the United New York: McGraw–Hill. Nations, Special Issue No. 42/43: 54–110. Zelinksy, Wilber (1971) ‘The hypothesis of the mobility Portes, Alejandro and Rumbaut, Ruben G. (1996) transition’, Geographical Review, 61: 219–49. Immigrant America: A Portrait, 2nd edn. Berkeley, CA: University of California Press.