In 1959 CBS Reports presented the population explosion to viewers as a crisis whose ingredients were statistical. Conceptualizing quantification as a socially situated media practice, this article interrogates the history of the production and circulation of the demographic data and the iconic figures that brought the population crisis to life. The analysis exhumes the intertwined quantitative contingencies and affective associations in three demographic formations: the calibration of natural fertility and pregnancy risk that naturalized fertility as an excessive feature of “other” women's bodies; accounts of population dynamics haunted by racialized feminine figures of natural excess and cultural incompetence; and population projection procedures that elided mortality and amplified the effects of fertility on population growth and economic development. The resulting figures, deployed by US demographers in a concerted media campaign, provoked a mathematical panic about global futures that moved nations to intervene in women's reproductive lives.
On November 11, 1959, CBS Reports aired “The Population Explosion,” an hour-long special that presented an audience of nine million viewers with “the facts” of an impending world population crisis.1 In the opening sequence, host Howard K. Smith narrates the crisis as one in which “the volatile ingredients are statistical.” As the scene shows a hospital ward filled with Indian mothers and newborns, Smith says, “This is a portrait of a statistic: the number of births in the world is increasing. One hundred and ten million babies will be born this year, 208 every minute.” As the camera pans across a crowded cityscape in which vultures perch on telephone poles, he continues, “This is a portrait of another statistic: the death rate of the world is dropping sharply. Sixty-one million people will die this year, 116 every minute. That's a net gain of forty-nine million every year.” Bringing the statistics to life, Smith's litany attaches to one final figure. As the camera focuses on the face of a sari-clad woman sitting on the floor listening to a lecture, Smith says, “This too is a statistic: this woman is thirty years old. She has had five children. One is dead and the others are hungry. She is learning how not to have more babies.” Smith concludes: “Virtually everyone believes we must control this birth rate [my emphasis]. The controversy is only as to the method.”2 Having positioned Indian women's fertility as requiring control, the program describes the dire consequences of continued population growth and assesses India's ability to curtail it before it is too late.3 Throughout the hour, the “fact” of a “population explosion,” that it represents a “crisis,” and that India is its exemplar are never in question.4 Indeed, demographic figures of unprecedented population growth in the mid-twentieth century remain authoritative.
The statistical ingredients that framed the CBS report were the product of a small group of well-funded demographers and their patrons who fostered the production and global circulation of the population crisis.5 Following danah boyd and Kate Crawford's call for investigation into what drives aggregation algorithms that “extract and illustrate large-scale patterns of human behavior,” this article reexamines the history of the data that catalyzed that crisis.6 Conceptualizing statistical quantification as a socially situated media practice, it unravels the intertwined quantitative and affective contingencies inscribed in and by “mundane” demographic “tables, figures, charts, and equations.”7 Specifically, the analysis exhumes three gendered and racialized formations aligned in and by the midcentury global population imaginary: the demographic calibration of natural (that is, uncontrolled) fertility and pregnancy risk that configured women's bodies, especially those of women of color, as excessively fertile; demographic accounts of population dynamics that the racialized feminine figures of excessive fertility haunted; and population forecasting procedures that elided the effects of declining mortality and amplified the effects of fertility on population growth and economic development. Together these figurations stoked a mathematical panic about global futures that intensified scrutiny of and intervention into women's reproductive lives.
According to Michel Foucault, population can best be understood as one of the ruling instrumentalities of modern governmentality.8 Population figures, aggregated products of statistical regimes of knowledge, compose new objects, “matters of national importance.”9 Censuses quantify the numbers and types of bodies within the spatial and temporal worlds of nation-states and their colonies. Vital statistics aggregate “characteristic movements of life”—births, deaths, marriages—in annualized linear time.10 Through such state-based quantification practices, national populations acquire the character of independent entities, subjects whose character and dynamics are represented in charts and graphs. But concurring with Bruce Curtis, I argue that population is a “not an empirical entity.” Population figures are not simply inscriptions made about observable objects. They are “a way of organizing social observations.”11 All of the categories of bodies, places, and times in censuses are configured by the institutional needs and hegemonic commitments of the state. Similarly, vital statistics fit life events into predefined categories of administrative ledgers so that they can be monitored and modified for the health of the social body.12 In fact, the value of statistics, Thomas Malthus asserted in his famous An Essay on the Principle of Population (1798), is that they enable observation of the oscillations of life events among the poor masses, whose history is otherwise unrecorded.13 Counting people is thus a historically situated, politically inflected, interpretive act. It involves sorting them by kind in order both to quantify and to manage them. Furthermore, the algorithms and equations by which counts are aggregated involve additional interpretive acts, removing them from anything that might be called “raw” data. Nonetheless, the procedures required to produce and make sense of population data quickly disappear behind the solidity of the numeric figures. Their provisional relation to the social world is elided by the objective transparency accorded to numbers, as when Smith enumerates births and deaths per minute. In life, such events do not occur like clockwork.
The capacity of numbers to appear as transparently commensurate with things in the world suggests, I contend, that the production and circulation of population statistics can also be understood as a set of practices that mediate between administrative ledgers, popular imaginaries, and embodied practices. In his history of statistical thinking, Theodore Porter identified statistics as a communication strategy by which to “summarize a multitude of complex events” at a distance from the circumstances of their occurrence.14 That, as Marshall McLuhan recognized, is the power of numbers: they “create the effect of an icon or an inclusive compressed image.” Likening statistics to cave paintings, McLuhan argued that aggregation gives “man a new influx of primitive intuition and magically subconscious awareness.”15 In other words, the power of numbers is bound up with the affects they archive and animate. As Sara Ahmed argues, affect is “what sustains … the connection between ideas, values, and objects.” Repetitive associations of bodies, signs, and objects in publicly circulating texts “accumulate affective value” that aligns bodies and collectivities with structures of (good and bad) feeling. Such affective economies “shape the surfaces of bodies and worlds.”16 Ahmed's observations refer to linguistic signs. However, I contend that population figures work similarly. The production and circulation of population figures organizes bodies, signs, and numeric figures into specific epistemic and affective associations that shape an imagined vision of social life. As official accounts of the past and present, they inscribe histories, shape embodied practices in the present, and constrain possible futures. They are, therefore, ripe for cultural analysis.
In particular, vital statistics mediate biopolitical relationships and affective alignments of individuals, collectivities, the nation, the state, et cetera with life and death itself. They concretize the national imagination of who is a citizen, what their character is and should be, how they should and do live, and what futures await them. One of the first census products, the typical Belgian (man), was constructed by Lambert Adolphe Quetelet using arithmetic averages of body type and consumption patterns. Quetelet circulated this figure as the exemplar of national character.17 Through dynamic nominalism, the articulation and circulation of such national figures shape social norms and individuals’ behaviors as they strive to conform to them. Such striving influences future measurements, as reported practices shift closer to stated norms.18 Population figures can also shape international norms and the behavior of state actors. Global inventories of population (and economic) statistics index the hierarchy of nations based on the vitality of social bodies, marking their progress and failures relative to that of others.19
As ruling instruments, population statistics produce not just exemplary figures to emulate. “Bound up with the securing of social hierarchy,” repetitive associations of some bodies and figures also produce signs of threat and contamination, figures to fear.20 Crisis moments are particularly crucial sites for investigation of the affective economies sustained by numeric figures. In such moments, drawing on histories of articulation, the repetitions of threatening associations become “sticky,” producing “actionable objects,” “fetishes” that can catalyze counter-actions against the threat.21 At least since Malthus, the associated signs of overpopulation—excessive fertility figures, indexed miseries, the bodies of “other” women and character of “other” nations—compose one such sticky association. In the mid-twentieth century, the iconic power of population figures mediated the precarious present of domestic renegotiations of gender, global decolonization, and Cold War economic conflict, revitalizing the Malthusian account of poverty as a product of profligate reproduction.22 Repetition of those associations in population discourse and popular media such as CBS Reports encouraged bourgeois classes worldwide to achieve the two-child family required for zero population growth and spurred emerging nations to invest in population control to stave off catastrophe. By tracing the affective associations aligned in and by population figures through the specific sites of their production in population data and discourse, the analysis exhumes the structure of feeling that captured and enclosed birth and death in the population explosion.23 It thereby opens invaluable space to “learn” the “lessons” of the population crisis “otherwise.”24
NATURAL FERTILITY AND THE RISK OF PREGNANCY
As an example of the enclosure of “everyday/everynight” life configured by demographic figures, consider the calculation of natural and controlled fertility.25 In the 1920s, demographers worried about the impact birth control agitation would have on differential fertility rates between social classes, races, and nations. Subscribing to the theory of cultural lag then current within sociology, they feared that the more advanced, the more intelligent, the more modern—that is, the professional classes like themselves—would be more motivated and better able to use contraceptives effectively. However, no one had yet demonstrated scientifically that contraception actually prevented pregnancies that would otherwise have occurred. Nor had “reliable” estimates been made of its use and effectiveness among different social groups. In the early 1930s, two founders of American demography, Raymond Pearl and Frank Notestein, filled that knowledge gap by developing authoritative statistical measurement protocols for calibrating pregnancy risk, natural fertility, and competent contraceptive practice. Those measurement practices inscribed masculine anxieties about the procreative power of women's bodies and racialized anxieties about the incompetence of others.
Pearl, a leading American protégé of Karl Pearson's biometrics, developed the first statistical measure of pregnancy risk in order to obtain a “clear-cut” measure of the extent “statistically” to which contraception was “actually practised” in the United States and to gauge its “quantitative effectiveness” in reducing the aggregate pregnancy rates of different groups.26 The first measure was a fairly straightforward calculation of the percentage of the sample that had ever used contraceptives (29.55 percent). It was a more complicated matter to gauge the extent to which the pregnancy rates were reduced. It required some means for comparing pregnancy rates of different groups using different contraceptives for different lengths of time. A standard measurement of the duration of risk was needed for statistical comparison of the multiple temporalities and materialities of contraceptive practice. Pearl began with the assumption that between puberty and menopause a woman was “exposed to the risk of becoming pregnant when she is more or less regularly indulging in sexual intercourse, as in the married state.”27 But the occurrence of marital intercourse was not an easily predictable or observable event, and therefore was not useful for calibrating a standardized risk metric. Instead, to get nearer to “the real biological roots of the matter,” his protocol relied on a different risk event: ovulation. As he put it, “What we want as a measure of pregnancy-rate is the answer … to the following question: What proportion of all the ovulation experiences during the period of observation resulted in the fertilization of an ovum and pregnancy?”28 Biologists had authoritatively described the rhythms of the human menstrual cycle in the late 1920s. So the regular occurrence of ovulation could simply be assumed, making it statistically useful as a biologically regular metric of risk. In his view, ovulation as the risk event was “sufficiently close to the actuality” to warrant its use.29 It made the mathematics more manageable. And it had the additional advantage of multiplying potential data points, because it provided thirteen distinct risk episodes per case, per year.30
Pearl's model of risk also made the gendered sexual politics more manageable. By indexing pregnancy risk to ovulation, his model disconnected it from the social and emotional processes and temporalities that produced it. Demographic quantification rendered illegible the conflicts about marital sex and the specter of childbirth death that were central to women's accounts of reproductive risk.31 There are no continuous intertwined dangers of having or avoiding marital sex in Pearl's calibration of risk. In fact, heterosexual practices are not pertinent to statistical calculation of the risk of pregnancy. Instead pregnancy risk, converted into the biological abstraction, resulted from natural rhythms of women's bodies. Pearl's statistical procedures enclosed pregnancy risk in a conceptual space in which the sexual politics of reproductive risk and its management had no place.
Frank Notestein used Pearl's methods in his own investigation of contraceptive effectiveness among US birth control clinic patients. His analysis relied on two data points. “(1) The number of pregnancies experienced” while using contraceptives, and “(2) the number of pregnancies that would have been experienced had no contraception been practiced.” Again, the number of pregnancies was a straightforward calculation. The tricky part was estimating the number of pregnancies that would otherwise have occurred. No standard measurement yet existed of “uncontrolled fertility.” So Notestein produced it. He started by figuring out the sample's actual pregnancy rate during cycles in which no contraception was used. From that observed rate, he used Pearl's model of risk to extrapolate an aggregate expected lifetime pregnancy rate represented “dramatically … as if it were that of an ‘average woman’ who made no attempt at contraception throughout her married life” (my emphasis). His estimate of uncontrolled fertility—fourteen pregnancies between twenty and forty-five years of age—“set up a control by which the effectiveness of contraceptive practice could be measured” in terms of the reduction in “the risk of pregnancy.” That is, the difference between the actual number of pregnancies and the expected number was, he asserted, a measure of a contraceptive's effectiveness.32 The measurements of controlled and uncontrolled fertility were thus co-configured. The assessment of controlled fertility was calibrated as a measure of deviation from, reduction in, a statistically configured aggregate measure of uncontrolled fertility.
The “as-if” average woman who never practiced contraception is, however, an imaginary figure. She is a materialization of the calculation methodology, which aggregated atomized months of nonuse within the sample and rendered them as a single life span. Like Pearl, Notestein decontextualized pregnancy risk from the social and affective relations in which it occurred. Marital sex and its avoidance were again not the basis of risk. Ovulation in the absence of contraceptive practice was. The months of nonuse on which the “as-if” figure was calculated were clustered in the period between marriage and a first pregnancy, a time period which both Pearl's and Notestein's data suggested involved high levels of sexual activity. Notestein's estimate of “uncontrolled fertility” thus extrapolated from but ignored the sexual practices of newly married American couples. Again, gendered heterosexual relations were irrelevant to the calibration of pregnancy risk. Instead it was conceptualized as a purely biological potentiality of women's bodies.
To summarize, these early contraceptive effectiveness studies involved a set of quantifications that mediated the social world of women's reproductive lives and the virtual world of aggregate population dynamics. Individual women provided information about sequences of their marriages, contraceptive practices, pregnancies both accidental and planned, and (illegal) abortions and miscarriages, which, disconnected from the flow of life and coded into the predefined boxes of the researchers’ forms, became separate data points in statistical analyses of aggregate trends. The women who peopled the demographic landscape were products of the measurement protocols required to make the equations work, rather than flesh-and-blood embodied women who negotiated reproductive life as it was lived. Their aggregated practices were calibrated against the “as-if” woman who represented the frightening excess of women's nature, eliding the gendered negotiations of managing pregnancy risk in hetero-patriarchal marriage.
This “as-if” figure became the standard demographic estimate of “natural fertility,” and she would come to haunt the midcentury population imaginary.33 She confirmed Malthusian logic that “wherever … there is liberty, the power of increase” produces “superabundant effects.”34 Unlike Malthus, however, for twentieth-century demographers, the natural excesses of their fertility required women to become competent contraceptors. To Pearl, contraceptive use rates signified “the degree of general enlightenment in the population.” He defined competent users as those social groups who “did their contraception intelligently, precisely, and ‘effectively.’”35 They did not experience contraceptive failures (that is, unintended pregnancies). In his analysis of differential contraceptive practices, working-class and racialized others had higher proportions of what he deemed to be incompetent practices.36 Pearl's commitment to the eugenic ideas that spurred early twentieth-century statistical innovations is clear in his harsh judgement of women who experienced contraceptive failure. They suffered, he concluded, from “a combination of bad luck and bad management, actually most of the ‘bad luck’ was probably really bad management.” Such qualities were “matters of character, bred in the bone.”37 Notestein's analysis also defined competence as 100 percent success in preventing unwanted pregnancies. Notestein correlated contraceptive effectiveness to character as measured by personal appearance (careless and slovenly, average, or neat and precise).38 Both studies ignored evidence that contraceptive practices reflected differential access to effective methods.39 Instead, they attributed the group differences to characterological flaws associated with race and class identities. Thus their analysis of contraceptive competence confirmed their faith and pride in the (prim and proper) white professional middle classes, while their disdain for and anxiety about incompetent contraceptors adhered to the bodies of “other” women.
The “as-if” woman of natural fertility and her incompetent sisters peopled the midcentury population imaginary through Notestein's articulation of demographic transition theory.40 A social eugenicist, Notestein deployed a standard narrative of the evolution of human societies from primitive to modern.41 The demographic transition refers to the shift from high death and birth rates to low rates said to accompany the advance of civilization. That is, the theory held that in the past and in “primitive” cultures, fertility was necessarily high. It had to be in order to match death rates that were inevitably high because, Notestein assumed, “societies with low levels of technical skill are inevitably poor, ill-housed, ill-clothed, ill-fed and subject to the uncontrolled ravages of disease.” Thus “the very existence of such populations in the face of the toll of heavy mortality proves that the birth rates are high” and must always have been so.42 According to the theory, the transition to low mortality and fertility was triggered by the social progress and growing prosperity of the modern era.
Malthus's England and Wales provided the standard case. There, according to Notestein, in the middle of the seventeenth century, mortality rates began to decline as a result of governmental and economic innovations that improved sanitation, transportation, communication, and security. Mortality declined first because it was more readily affected by such public efforts. Fertility decline lagged because it depended on the actions of individuals guided by religious customs, the “heritage of past ages,” which, residing in the hearts of individuals, were deemed to be less responsive to external changes.43 Fertility would only begin to fall when a threshold of economic development was achieved that made the cost of raising children greater than their productive value to households.44 But the theory was written with men as the subjects of the transition. It was only when societies began to see “man as the master of his own destiny” that “deliberate control of fertility” became “reasonable and desirable.”45 The theory also assumed the subordination of women in past and “primitive” cultures. In this frame, women enter transition theory only as wives through whom men's reproductive decisions are realized. Modern men, of course, “let” their wives control fertility.46
The graph on the left side of figure 1 represents the standard demographic transition as rendered by the US State Department at the height of the population crisis. The vertical axis is events per one thousand population; the horizontal axis represents historical time. The lower of the two lines represents mortality change; the top line represents fertility change.47 Fertility begins to fall later, after the decline in mortality is well established, but approaches the low level of mortality in the present of the graph. The gap between the lines represents the surge in growth during the transitional phase, which is contained by the trajectory of the fertility line.48 The population problem of the mid-twentieth century, depicted by the graph on the right in figure 1, was that the normal progress of the demographic transition had been disrupted in “less developed countries.” That disruption threatened to produce continuous, unprecedented growth. The trouble occurred because those nations had “unwittingly” benefited from importation of Western mortality control technologies during colonialism. However, Notestein argued, “In enlightened colonial regimes there had been considerable protection of native customs, and religions, and social organization.” Thus, with declines in mortality colonial populations began to grow, but “unlike the situation in the West,” the requisite economic changes that would automatically lead to lower fertility had not occurred.49 These abnormal circumstances produced a “Frankenstein,” a population with the potential for explosive population growth.50 In figure 1, the threat of such circumstances is represented by the lines that do not converge but extend indefinitely, creating a gaping maw that stalks the future.51
Such large and rapid growth, demographers asserted, would undercut the efforts of newly independent nations like India to modernize their economies and raise living standards. Without higher standards of living, the changes that provoked reduced fertility were unlikely. Thus, without intervention, potentially catastrophic growth would continue unabated, raising the specter of the return of older Malthusian miseries. Obscured by the Malthusian threats were the destructive impacts of colonialism and the continuing disadvantaged position of former colonies in the global economy. Instead, the demographic transition assuaged anxieties about the growing population of racialized others with the promise of population control aligned to US foreign aid practices that championed capitalist modernization.
“Proof” of the claim that high fertility would undercut development efforts came in an influential 1958 study published by Ansley Coale and Edgar M. Hoover entitled Population Growth and Economic Development in Low-Income Countries: A Case Study of India's Prospects, the stated aim of which was “to attempt to give as concrete an answer as possible to a specific question: what difference would it make in economic terms if the birth rate, instead of remaining unchanged, should be cut drastically in one generation.” To answer this question Coale, the demographer, and Hoover, the economist, built elaborate comparative projections of population and economic growth for India from 1956 to 1986. They chose India because it represented “an urgent exemplar” of the problem: a large and growing population, low income levels, and a “traditional” culture.52 In addition, unlike other low-income nations where “scanty data” was a problem, there was a long history of population statistics compiled by British colonial administrators.53 Thus the demographic history of India was said to be “well known,” making it possible to calibrate its current population patterns and project its future trends.54 It is important to note that demographic history was compiled by colonial administrators within an imperial discourse that linked recurrent famines and epidemics to the nature of Indian culture and climate and not, as Indian nationalists argued, to the violence and destruction wrought by British colonialism.55
Coale and Hoover took up the earlier census data as transparent accounts of past vital events from which it was possible to project future trends. Based on their statistical projections, they concluded that there would be a “large advantage attaching to an early reduction in fertility.” In fact, they quantified the magnitude of that advantage quite precisely, calculating that the per capita income of India “would attain a level about 40% higher by 1986 with reduced fertility than with continued high fertility.”56 Those figures represent a dramatic economic effect of reduced fertility. But again, it is a product of the elaborate procedures that produced them. Interrogating those procedures, the following discussion shows that the perilous futures represented in their projections elided the effects of falling mortality and thus amplified the effects of fertility. In the opening pages of the book, Coale and Hoover do acknowledge that their projections resulted from “the specific implications of assumptions about population change” that configured them.57 Yet through the more than two hundred pages of technical narrative, written in leaden prose, those assumptions are eclipsed by the precisely drawn vectors of their graphs of impending crisis (figs. 2 and 3).
Because Indian vital registries were incomplete, they relied on census figures for their population data. However, censuses do not capture information on births and deaths directly. They merely offer a snapshot of the types and distribution of bodies in a landscape in ten-year intervals. Therefore, all the variables used in their population projections had to be configured through procedures to convert decennial census data into estimates of annual mortality, fertility, and growth rates.58 Fertility rates were the last figure calculated in their conversion sequence and thus inscribed the sum of negotiated contingencies that produced the data used in the projections.
Coale and Hoover began by calculating growth between each census, smoothed out across the decade in average annual increments. Next they estimated changes in the size of age cohorts between the 1941 and 1951 censuses to estimate death rates. That is, the difference between the size of the population of zero- to five-year-olds in 1941 and that of ten- to fifteen-year-olds in 1951 was taken to represent deaths in that age cohort over the decade. Doing this for all ages in five-year increments, they built a plausible life table (a measure of the risk of death by age).59 Using the life table, they estimated age-specific death rates. From those, they worked backward to estimate birth rates. That is, if migration is ignored, growth is a result of the proportional relationship of deaths to births. Given an age-specific pattern of death rates, it was possible to estimate the level of general fertility that would be required to produce the intercensual population growth they calculated. In turn, they used that figure to estimate annual age-specific birth rates. With age-specific rate estimates, Coale and Hoover were able to use the formal mathematical equations of stable population theory to calculate a figure representing the inherent rate of increase and project growth into the future.60
In negotiating the numerous data gaps and contingencies of converting the census data, they selected what seemed to them to be the most plausible figures. Plausibility was predicated on the demographic transition, a theory that secured colonial difference to the cultures and bodies of others. Consider their inferences about the adjusted age distribution in their projections: An age distribution results from the dynamic interaction of fertility and mortality. Changes in the age distribution can result from changes in fertility, mortality, or both. Coale and Hoover noted that the shift in the ratio of children to adults in the 1951 Indian census was “consistent” with either “a fairly substantial decline in infant mortality offset by a slight drop in fertility” or “a moderate decline in infant mortality.” While they noted that “no published version of the theory of demographic transition states precisely what conditions are essential for fertility decline,” they concluded that “fertility declines have been non-existent or only moderate” in India. To conclude otherwise would suggest that Indian fertility rates had begun to decline at an earlier point of economic development than had occurred in Europe, a possibility that they felt was unlikely.61 Their projections, therefore, aligned with the arrogant cultural logic of the theory their analysis was meant to substantiate.
The ambiguities in the data and the contingencies of the calculations are not at all apparent in summary graphs of population or income growth. Instead, the contingencies represented in figure 2 are of future growth, the broken lines representing the three projected trends in future fertility. The high-fertility projection, the top line, represents the figure for no change in fertility across thirty years between 1956 and 1986; it continues an upward ascent unabated, nearly reaching the upper limit of the graph. That rate is the “present” rate they reconstructed from the estimation procedure described above. The low-fertility projection, the bottom line, assumes a 50 percent decline in fertility by 1986. Its upward ascent continues but intersects the vertical axis of the graph at a substantially lower point than the high-fertility projection. There is still growth, but it does not threaten the very limits of the graph. The middle line, appearing in more tentative dots than the broken lines of the other two, assumes a fertility decline that starts later but still reaches 50 percent by 1986. The difference between the top and bottom lines is the basis for their conclusion that a 40 percent higher per capita income would result from drastically reduced fertility. They picked the 50 percent fertility decline not because it was plausible—they make clear that it is not realistic—but because a “large contrast” would “make it easier to bring out the nature of the economic effect of alternative population courses.”62 So their projections set up an unattainable metric of fertility change to show the effects of present fertility rates on the future, amplifying both the peril of the present moment and pessimism about India's ability to cope with it.
Not depicted in the graph is another rate change critical to their mathematical model: a 50 percent decline in the mortality rate between 1956 and 1976. That is, although fertility is represented as the single important factor in their analysis, the population growth they projected hinged on a continuous steep decline in mortality. By itself a substantial decline in the mortality rate will produce substantial growth regardless of fertility patterns. In the high-fertility projection, the projected growth is entirely due to the assumed decline in mortality. Also note that, following transition theory, they assume mortality falls faster than fertility, which means that even if the fertility rate drops in the same proportion as mortality, population growth will continue long after fertility rates decline. This is represented in the middle case of lagging fertility decline, which demonstrated that the sooner fertility began to decline, the better. But even in the low-fertility projection, where mortality and fertility begin to fall at the same time, substantial growth is still projected because mortality falls more quickly (50 percent in twenty years versus thirty years).63 Perhaps because they held the mortality rate change constant across the model, they saw no need to represent it in the contingencies of the graph. Whatever their reasons, mortality disappears from view. Its absence amplifies the effect of fertility on the future. In consequence, growth and fertility become conflated. In the fetish object of population control, growth and fertility are commensurate. The elision of declining mortality realigns the narrative of population change with the Malthusian hazard of natural procreative excess and pessimism about the ability of human reason to contain it.
Figure 3, the most important chart in the book, depicts the projected difference fertility makes on per capita income. The solid lines in the graph represent national income. The broken lines represent the capita, in this case labeled adult consumers. The gap between them visually represents the income difference based on fertility. The mortality rate change is again not represented in the graph. It graphs only the effect of fertility, this time on economic well-being. The gap marking the income difference between high and low fertility is dramatic visual confirmation of the “advantage” low fertility affords the nation and the individual. Not only is the national income higher, but the amount per person is so large it envelops the high fertility income measure. Although both start at the same point, the gap widens annually throughout the projection period, confirming the personal and national risk of continued high fertility. Unlike the fertility graph, the lines of the income graph end well short of the right-hand boundary of the graph. Thus the capacious space beyond them suggests the possibility of even greater prosperity, if population growth can be contained.
One final measurement crucial to Coale and Hoover's analysis warrants mention. For the economic measure they use per capita income, which was developed at the end of World War II specifically to assess the level of economic development in the Global South. This measure recast what were largely non-monetized economies in terms of US dollars per year available to the average citizen. With it, in 1948 the World Bank for the first time determined that two thirds of the world was poor.64 Of course, per capita income is a measure enormously sensitive to population figures. It amplifies the impact of population change on income: the larger the figure of total population, the lower the average income per citizen. However, Coale and Hoover's use of the measure is important in another, less obvious way. It helped to shift the meaning of “overpopulation” from a measure of the density of bodies in the landscape, the older Malthusian measure, to a measure of the density of bodies in the market.65 Thus the same measurement by which the World Bank defined two thirds of the world as poor, defined two thirds of the world as overpopulated as well. The population statistics of this case study confirmed that decolonizing nations such as India were too large and growing too fast as well as poor, and would likely remain so without intervention. Thus the older Malthusian threats attached to the bodies of other women—misery, war, and famine—still loomed over the future. But mortality decline was no longer the marker of progress, as it was for Malthus. Fertility rates were now the index of modernity. Demographic representations of the “as-if” woman's reproductive excess and decolonizing nations’ incapacity to contain it stoked a mathematical panic and recommended state-based family planning and capitalist development as the solution to world poverty.
CENTERS OF CALCULATION
Demographic statistics of crisis were the first mediation that configured midcentury population knowledge. But further mediations were required to invest the crisis in the popular imaginary and reproductive practice, in which US demographers also played a substantial role. CBS Reports's statistical framing of the population crisis aired at the end of a decadelong effort by demographers to disseminate their reading of the facts. The Princeton Office of Population Research and the Population Council facilitated the construction and dissemination of demographic data that represented population growth in the Global South as the most significant danger to world peace and prosperity. Coale and Hoover's analysis is a prime example of the research they supported and disseminated.67 The case for population control was also explicitly laid out in a popular 1958 book by Frederick Osborn, the Population Council's first president and a renowned social eugenicist. Entitled Population: An International Dilemma, the book compiled results from the Population Council's two-year investigation into the “present conditions” and appropriate “lines of action” by which educated elites and governments might affect “the scale and tempo” of population growth, which it said “could well be the decisive factor in the race between progress and catastrophe.”68
In Osborn's view, the principal impediment to rapidly reducing birth rates was that those rates were the aggregate result of the mass of individuals acting on personal, not national considerations. So a transition from high to low fertility required “stimulation” of “latent” personal motivations. In council discussions of the problem, staff noted that, aside from the “delicacy of interference” in the “most intimate domestic affairs” that advocacy of family planning involved, there was also the “hazard of rousing intense anti-colonial and racial consciousness.” Because “what could be more dangerous material for Anti-American propaganda than the idea that rich, white Americans want to restrict the growth of colored Asian and African peoples.”69 Thus, the council sought the best approach to influence national population policies and family limitation practices in decolonizing nations without stirring up racial resentments. In developing that approach, the council drew on and contributed to communication theory, meeting with experts who outlined methods for influencing elites, intellectuals, and mass publics.70
Drawing on innovation diffusion theory, which (consistent with cultural lag theory) held that new ideas diffused out and down from intellectual and public elites, the experts recommended that the initial focus should be on identifying and persuading indigenous intellectuals and opinion leaders to see population as a national problem that required their immediate attention.71 Specific steps included “plant[ing]” articles on population matters in the press directed to elite audiences and translating and publishing “cheap editions” of population treatises. Once they had cultivated a group of leading “indigenous professionals,” they should be encouraged “to write articles …; make speeches to professional and intellectual groups; and to promote discussion in teachers’ meetings and comparable forums … and on radio and TV.”72 Council staff concluded that given “the sensitivity of this field,” it would be best to build their efforts around population facts, which the research they sponsored produced. In their view, there was no need “to convince if our case is as good as we think it is. Local leadership is bound to reach the same conclusions we would.”73 Relying on the “cool idiom of numbers,” they would not need “to propagandize a solution.”74 Once local leaders were aware of the facts, they would understand the problem “not as a theory but as a nightmare” and would be moved to act to secure national well-being through population control programs.75
After local leaders were aligned with the cause of population control, then it would be time for a mass campaign led by indigenous sponsors. With what they assumed would be limited literacy among the “Third World” masses, the communications experts consulted by the council recommended that mass campaigns use popular forms such as comic-book-style publications and wall posters to convey the personal value of limiting family size. In particular they recommended “A Tale of Two Families” to extend the reach of the income effect calibrated in Coale and Hoover's projections. Contrasting images of a well-fed and well-housed small family enjoying leisure and the harried parents, dire circumstances, and hungry children of the large family, the tale promised happiness through the small family standard.76
The council organized its activities according to these communication strategies. It cultivated opinion leaders through John D. Rockefeller III's and Frederick Osborn's elite social network.77 For example, it gave advice to members of the President's Committee to Study the US Military Assistance Program—commonly known as the Draper Committee after its chair, General William Draper. The committee's report was the first to recommend that the US government include family planning as part of its foreign aid program.78 In emerging nations, such as India, Taiwan, and Korea, to name three specific cases, the Population Council cultivated politicians, social scientists, and physicians through technical assistance, grants, and demographic research centers, often in conjunction with the United Nations. But of greatest significance, the council sponsored graduate fellowships that enabled scholars from decolonizing nations to complete PhD training in demography at US universities. In the course of its first twenty-five years, the program supported nearly one thousand fellows, more than seven hundred of whom were from nations of the Global South. When their education was complete, fellows generally secured positions in the burgeoning statistical bureaucracies and family-planning programs of their home nations. Through the fellowships and their role in UN census standard setting, US demographers and their patrons thus were able to greatly extend the reach of “research created here” in globally interconnected centers of calculation.79
In the popular media, instances such as the CBS Reports episode, which cited the Draper Report, modeled the council's approach by exposing a “dangerous statistical reality” for which family planning was the only reasonable solution. With the power of television, CBS Reports brought home the statistical crisis secured to images of distant neo-Malthusian misery and want, affectively aligning its American audience with the population crisis and control narrative. It is but one example of the burgeoning coverage of the population crisis located in distant cultures and the bodies of other women, a looming threat nonetheless.80
The council also provided technical assistance to population control programs in globally interconnected family planning programs designed to bring down aggregate rates as quickly as possible. The council even championed the development and deployment of the IUD, “birth control for a nation.”81 Along with clinical distribution of contraceptive devices, national programs engaged in mass education campaigns that followed communication experts’ advice. Into the 1970s, India's state-sponsored population control programs regularly featured wall posters and billboards displaying “colorful graphics and minimal text” juxtaposing two-child small families amid desirable consumer goods alongside large families shoeless and clothed in rags, labeling the small families as “happiness all the way” and large families as “problems all the way.”82 With these techniques, the mediation of reproductive lives accomplished in population statistics were translated into narratives and images that sutured individual well-being to national well-being. The numbers that guaranteed the narratives of explosive growth were first brought to life in administrative ledgers, then circulated in the popular media and instantiated in family planning programs, shaping the bodies and worlds of the midcentury.
Returning to the iconic power of numbers, the question of how to evaluate the midcentury population projections often turns on whether the numbers proved to be accurate. That is, did the global population growth rate explode? Were the dire predictions realized? But whether or not demographic figures accurately predicted the future course of growth is, in my view, the wrong question.83 It diverts our attention from the cultural work accomplished by the population figures. That is, the point of midcentury population projections was never to be right about the future in the future. The point was to influence policy and practice in the midcentury present, so that the present could shape particular futures. In this sense, the demographic figures worked. They made a fetish of fertility rates that moved nations to act in the present, and thus shaped futures. By the mid-1960s, the Lyndon B. Johnson administration made population control programs a condition of US foreign aid. In the 1970s, India's population control campaign assumed a “war-footing,” initiating mass sterilization.84 At the same time, effective contraceptive practice became and (remains) a mark of competence for “modern women” everywhere, as women both took up the practice of contraception or were compelled to. And by the 1990s, aggregate fertility rates were declining, securing man-made demographic transitions worldwide. The power of demographic numbers did not rest in their eventual accuracy, but on an affective-epistemic economy that, mediating between administrative ledgers, policy, and popular discourse, reanimated older cultural figures of misery, want, and excess that threatened the progress of nations.
To be clear, I am not calling for an abandonment of population statistics as always already contaminated knowledge. Rather, I hope to have shown that the “complex methodological processes that underlie the analysis of data” involve more than quantification.85 As historically situated mediation practices, the methodological processes underlying data are always inflected by the situations in which they are enacted. There are always gaps and contingencies in how and what the data represent that must be negotiated. In such instances, affect can bridge the gap, resolve contingency, and otherwise sustain the linkage of bodies, figures, and what counts as good data and solutions. In the initial articulation and production of crisis, affect may matter most, materializing figures with densely associated cultural affects and effects. Attention to the epistemic and affective economies that sustain such situations and moments opens space to ask different questions of the data. Such questions—about the statistical probabilities, political contingencies, and affective associations that bring statistics to life—can be asked about the interpretive acts that constitute settled facts in all their guises.