Population Growth and

Earth's Human Carrying Capacity

Joel E. Cohen*

Earth's capacity to support people is determined both by natural con-

straints and by human choices concerning economics, environment, cul-
ture (including values and politics), and demography. Human carrying
capacity is therefore dynamic and uncertain. The element of human
choice is not captured by ecological notions of carrying capacity that
are appropriate for nonhuman populations.
Scientific uncertainty about whether and how the earth will support
its projected human population has led to public controversy: Will
humankind live amid scarcity or abundance or a mixture of both? 1

The Past and Some Possible Futures

Over the past two thousand years, the annual rate of increase of global
population has grown about fifty-fold from an average of 0.04 percent

*Joel E. Cohen is the Abby Rockefeller Mauze Professor of Populations at Rockefeller

University and Columbia University. This essay contains excerpts from "Population
Growth and Earth's Human Carrying Capacity," Science 269 (July 21, 1995): 341--46;
and from "How Many People Can the Earth Support?" The Sciences 35, no. 6 (Novem-
ber-December 1995): 18-23. Copyrights 1995 American Association for the Advance-
ment of Science, reprinted by permission of the American Association for the Advance-
ment of Science; and 1995 Joel E. Cohen. The author acknowledges with thanks U.S.
National Science Foundation grant BSR92-07293 and the hospitality of Mr. and Mrs.
William T. Golden, Jesse H. Ausubel, Danny J. Boggs, Griffith M. Feeney, Richard B. Gal-
lagher, Shiro Horiuchi, and Robert M. May, who reviewed and improved previous drafts.

55
56 Part II. Science Population Growth and Earth's Human Carrying Capacity 57

100 for social, political, and economic activities. 8 The resulting frictions are
evident in all parts of the world.
...-..
(/')
As of 1999, the world has about 6 billion people. The population
c would double in forty-nine years if it continued to grow at its present
.Q
1.4 percent per year, though that is not likely. The population of less
:.0
-.._.... developed regions is growing at 1. 7 percent per year, while that of more
c
0 developed regions is growing at 0.1 percent per year. 9
~ The future of the human population, like the future of its economies,
::I
0. environments, and cultures, is highly unpredictable. The United Nations
0
0... regularly publishes projections with a range from high to low (figure 2).
In 1992, its high projection assumed that the worldwide average num-
ber of children born to a woman during her lifetime at current birth
0.1
0 500 1000 1500 2000 2500 rates (the total fertility rate, or TFR ) would fa ll to 2.5 children per
Year woman in the t\venty-first century; in that scenario, the population
would grow to 12.5 billion by 2050.1 Its 1992low projection assumed
Figure I. Recent world population history A.D. 1- 1990 (solid line) 37 and
that the worldwide average TFR would fall to 1.7 children per woman;
1992 population projections of the United Nations10 from 1990 to 2150.
in that case, the population would peak at 7.8 billion in 2050 before
Population growth was faster than exponential from about 1400 to 1970.
beginning to decline.
Asterisks, dashes, and dots indicate high, medium, and low projections,
There is much more uncertainty about the demographic future than
respectively.
such projections suggest. 11 At the high end, the TFR in the less devel-
oped countries today, excluding China, is about 3.8 children per
per year between A.D. 1 and 1650 to its all-time peak of 2.1 percent per woman; that region includes 3.5 billion people. Unless fertility in the
year around 1965- 70.2 The growth rate has since declined haltingly to less developed countries falls substantially, global fertility could exceed
about 1.4 percent per year (figure 1).3 that assumed in the UN's high projection. At the low end, the average
H uman influence on the planet has increased faster than the human woman in Germany now has about 1.3 children, and in Italy and Spain
population. For example, while the human population more than
quadrupled from 1860 to 1991, human use of inanimate energy in-
100r-------------------------~
creased from 1 billion megawatt-hours per year to 93 billion megawatt-
~ 9
hours per year (figure 2). In the minds of many, human action is linked ::J
>. 80
to an unprecedented litany of environmental problems, some of which e>
Q) 70
affect human well-being directly. 4 As more humans contract the viruses c
Q) 60
and other pathogens of previously remote forests and grasslands, dense -g 50
urban populations and global travel increase opportu nities for infec- en
c 40
tions to spread .5 The \vild beasts of this century and the next are micro- 0
~ 30
bial, not carnivorous.
~ 20
Along with human population, the inequality in the distribution of
global income has grown in recent decades. 6 In 1992, 15 percent of
d: 10
~~~~~~~~
people in the world's richest countries enjoyed 79 percent of the world's 1880 1900 1920 1940 1960 1980 2000

income.? Economic contrasts are compounded by cultural ones. In Year

every continent, in giant city-systems, people increasingly come into Figure 2. Inanimate energy use from all sources from 1860 to 1991: aggregate
direct contact with others who vary in culture, language, religion, val- (solid line with asterisks)38 and per person (dashed line). Global population
ues, ethnicity, and socially defined race, and who share the same space size is indicated by the solid line.
58 Part II. Science Population Growth and Earth's Human Carrying Capacity 59

A historical survey of estimated limits is no proof that limits lie in this

OOOOl~illillillillillilliTSITSITSITSITSITSITSITS:ml
1

w range. It is merely a warning that the human population is entering a

c zone where limits on the human carrying capacity of earth have been
J
~ t= :. m mmmm~m'l
1000
anticipated and may be encountered.
.100
~ 100 ................................................................. .
8. ~~~~~~~~~~~~~~~~~~~~~!~~~~~~~~~~~~~~~~~~~~:~:::~:~:~:~~~~~:~::::~:~::::~::::::::::::::::::~:i!!i
..................................................................................................
Methods of Estimating Human Carrying Capacity
E
::J !~
Estimates of earth's maximum supportable human population are
E
~
E
10
: : : : ~ :.

:~ : : : ~~:!~~-:i : \

made with one of six methods, apart from those that are categorical
assertions without data. First, several geographers divided earth's land
..............................................

1+-~--~--~~--~--~~ into regions, assumed a maximum supportable population density in

1650 1700 1750 1800 1850 1900 1950 2000
each region, multiplied each assumed maximal population density by
year the area of the corresponding region, and summed over all regions to
Figure 3. Estimates of how many people earth can support, by the data at get a maximum supportable population of earth. The assumed maxi-
which the estimate was made. When an author gave range of estimates or mum regional population densities were treated as static and were not
indicated only an upper bound, the highest number is plotted here.39 selected by an objective procedure.
Second, some analysts fitted mathematical curves to historical popu-
lation sizes and extrapolated them into the future.l 5 As the causal fac-
1.2. Fertility could fall well below that assumed in the UN's low pro- tors responsible for changes in birth rates and death rates were and are
jection. not well understood, there has been little scientific basis for the selec-
Can the earth support the people projected for 2050? If so, at what tion of the fitted curves.
levels of living? In 1679, Antony van Leeuwenhoek (1632-1723) esti- Third, many studies focused on a single assumed constraint on pop-
mated that the maximum number of people the earth can support is ulation size, without checking whether some other factors might inter-
13.4 billion. 12 Many more estimates of how many people the earth can vene before the assumed constraint comes into play. The single factor
support followed (figure 3 ). 13 The estimates have varied from < 1 bil- most often selected as a likely constraint is food.l 6 In 1925, the German
lion to >1,000 billion. Estimates published in 1994 alone ranged from geographer Albrecht Penck stated a simple formula that has been
< 3 billion to 44 billion. 14 Since 1679, there has been no clear increas- widely used: 17
ing or decreasing trend in the estimated upper bounds. The scatter population that can be fed =
among the estimates has increased with the passage of time. This grow- food supply I individual food requirement. [1]
ing divergence is the opposite of the progressive convergence that
would ideally occur when a constant of nature is measured. Such esti- This apparently objective formula can lead to extremely different esti-
mates deserve the same profound skepticism as population projections. mates of maximum supportable population because it depends on esti-
They depend sensitively on assumptions about future natural con- mates of the food supply and of individual requirements. The food sup-
straints and human choices. ply depends on areas to be planted and watered, choice of cultivars,
Many authors give both a low estimate and a high estimate. Consid- yields, losses to pests and waste, cultural definitions of what constitutes
ering only the highest number given when an author stated a range, and acceptable food, and random fluctuations of weather. Individual
including all single or point estimates, the median of sixty-five upper requirements depend on the calories and protein consumed directly, as
bounds on human population was 12 billion. If the lowest number well as on nutrients used as animal fodder. 18 Besides food, other factors
given is used when an author stated a range of estimates, and all point proposed as sole constraints on human numbers include energy, bio-
estimates are included otherwise, the median was 7.7 billion. This range logically accessible nitrogen, phosphorus, fresh water, light, soil, space,
of low to high medians, 7. 7-12 billion, is very close to the range of low diseases, waste disposal, nonfuel minerals, forests, biological diversity,
and high UN projections for the population in 2050: 7.8-12.5 billion. and climatic change.
 60 Part II. Science Population Growth and Earth's Human Carrying Capacity 61

Fourth, several authors reduced multiple requirements to the amount assumes no interactions among the inputs; independence among limit-
of some single factor. For example, in 1978 Eyre reduced requirements ing factors is not generally observed. (For example, equation [3]
for food, paper, timber, and other forest products to the area of land neglects the possibility that changes in the water supply may affect the
required to grow them. 19 Other factors that cannot be reduced to an food supply through irrigation.) Liebig's law assumes that adaptive
area of land, such as water or energy, are sometimes recognized indi- responses will not alter requirements or resources during the time span
rectly as constraints on the extent or productivity of cultivable land. of interest; economic history (including the inventions of agriculture
The authors who combined different constraints into a single resource and industry) and biological history (including the rise of mutant infec-
assumed that their chosen resource intervened as a constraint before tions and the evolution of resistance to pesticides and drugs) are full of
any other factor. such adaptive responses.
Fifth, several authors treated population size as constrained by mul- Sixth and finally, several authors treated population size as con-
tiple independent factors. For example, Westing, in 1981, estimated the strained by multiple interdependent factors and described the interde-
constraints on population imposed independently by total land area, pendence in system models. System models are large sets of difference

cultivated land area, forest land area, cereals, and wood. 2 Constraints equations (deterministic or stochastic), which are usually solved numer-
from multiple independent resources are easily combined formally. For ically on a computer. System models of human population and other
example, if one assumes, in addition to a food constraint, a water con- variables have often embodied relationships and assumptions that were
straint neither mechanistically derived nor quantitatively tested. 24
population that can be watered = The first five methods are deterministic and static. They make no
water supply I individual water requirement [2] allowances for changes in exogenous or endogenous variables or in
functional relations among variables. While a probabilistic measure
and if both constraints [1] and [2] must be satisfied independently, then of human carrying capacity has been developed for local populations
in the Amazon, 25 no probabilistic approach to global human carrying
population that can be fed and watered = capacity has been developed. Yet stochastic variability affects local
minimum of {food supply I individual food requirement and global human populations through weather, epidemics, accidents,
or water supply I individual water requirement}. [3] crop diseases and pests, volcanic eruptions, the El Nino Southern
Oscillation in the Pacific Ocean, genetic variability in viruses and
other microbes, and international financial and political arrange-
ments. Stochastic models of human carrying capacity would make it
This formula is an example of the law of the minimum proposed by the possible to address questions that deterministic models cannot, such
German agricultural chemist Justus Freiherr von Liebig (1803-73). 21 as: conditional on all the assumptions that go into any measure of
Liebig's law of the minimum asserts that, under steady-state conditions, human carrying capacity, what level of population could be main-
the population size of a species is constrained by whatever resource is tained ninety-five years in one hundred in spite of anticipated vari-
in shortest supply. 22 Liebig's law has serious limitations when it is used ability?26
to estimate the carrying capacity of any population. If different compo- Some have urged that individual nations or regions estimate their
nents of a population have heterogeneous requirements, aggregated human carrying capacity separately. 27 While specific resources such as
estimates of carrying capacity based on a single formula will not be mineral deposits can be defined region by region, the knowledge,
accurate; different portions of the global human population are likely energy, and technology required to exploit local resources often depend
to have heterogeneous requirements. In addition, Liebig's law does not on other regions; the positive and negative effects of resource develop-
apply when limiting factors fluctuate, because different factors may be ment commonly cross national borders. Human carrying capacity can-
constraining at different times; an average over time may be misleading. not be defined for a nation independently of other regions if that nation
Liebig's law assumes that the carrying capacity is strictly proportional trades with others and shares the global resources of the atmosphere,
to the limiting factor (within the range where that factor is limiting); oceans, climate, and biodiversity.
strictly linear responses are not generally observed. 23 Liebig's law Some ecologists and others claim that the ecological concept of car-
62 Part II. Science Population Growth and Earth's Human Carrying Capacity 63

rying capacity provides special insight into the question of how many centuries because he did not foresee how much people can expand the
people the earth can support. In basic and applied ecology, the capacity human carrying capacity of earth, including but not limited to food
for carrying nonhuman species has been defined in at least nine signifi- production. To examine whether Malthus will continue to be wrong,
cantly different ways, none adequate for humans. 28 Human carrying economists, demographers, and system analysts have constructed
capacity depends both on natural constraints, which are not fully models in which population growth drives technological change,
understood, and on individual and collective choices. How many peo- which permits further population growth. 30
ple the earth can support depends in part on how many will wear cot- These models illuminate the earth's human carrying capacity. First,
ton and how many polyester; on how many will eat meat and how the statement that "every human being represents hands to work, and
many bean sprouts; on how many will want parks and how many will not just another mouth to feed" does not specify the cultural, envi-
want parking lots. These choices will change in time and so will the ronmental, and economic resources available to make additional
number of people the earth can support. hands productive, and therefore does not specify by how much the
The deceptively simple question "How many people can the earth additional hands can increase (or decrease) human carrying capacity;
support?" hides a host of thorny issues: How many people with what yet the quantitative relation between an increment in population and
fashions, tastes, and values? How many people at what average level of an increment in carrying capacity is crucial to the future trajectory of
material well-being? With what distribution of material well-being? both the population and the carrying capacity. Second, the historical
With what technology? With what domestic and international political record of faster-than-exponential population growth, accompanied by
institutions? With what domestic and international economic and demo- an immense improvement in average well-being, is logically consistent
graphic arrangements? In what physical, chemical, and biological envi- with many alternative futures, including a continued expansion of
ronments? With what variability or stability? With what risk or robust- population and carrying capacity, or a sigmoidal tapering off of the
ness? What standards of personal liberty will people choose? growth in population size and carrying capacity, or oscillations
How many people for how long? Human carrying capacity depends (damped or periodic), or chaotic fluctuations, or overshoot and col-
strongly on the time horizon people choose for planning. The popula- lapse. Third, to believe that no ceiling to population size or carrying
tion that the earth can support at a given level of well-being for twenty capacity is imminent entails believing that nothing in the near future
years may differ substantially from the population that can be sup- will stop people from increasing the earth's ability to satisfy their
ported for one hundred or one thousand years. wants by more than, or at least as much as, they consume. The mod-
els focus attention on, and provide a framework in which to interpret,
quantitative empirical studies of the relation between rapid population
Mathematical Cartoons growth and changing human carrying capacity.
If a current global human carrying capacity could be defined as a
statistical indicator, there would be no reason to expect that indica-
tor to be static. In 1798, Thomas Robert Malthus (1766-1834) Issues for the Future
described a dynamic relation between human population size and Three valuable approaches have been advocated to ease future trade-
human carrying capacity: "The happiness of a country does not de- offs among population, economic well-being, environmental quality,
pend, absolutely, upon its poverty or its riches, upon its youth or its and cultural values. Each of these approaches is probably necessary, but
age, upon its being thinly or fully inhabited, but upon the rapidity is not sufficient by itself, to alleviate the economic, environmental, and
with which it is increasing, upon the degree in which the yearly cultural problems described above.
increase of food approaches to the yearly increase of an unrestricted The "bigger pie" school says: develop more technology. 31 The "fewer
population." 29 Malthus opposed the optimism of the Marquis de forks" school says: slow or stop population growth. 32 In September
Condorcet (1743-94), who saw the human mind as capable of remov- 1994 at the UN population conference in Cairo, several approaches to
ing all obstacles to human progress. Malthus predicted wrongly that slowing population growth by lowering fertility were advocated and dis-
the population growth rate would always promptly win a race against puted. They included promoting modern contraceptives; promoting eco-
the rate of growth of food. Malthus has been wrong for nearly two nomic development; improving the survival of infants and children;
64 Part II. Science Population Growth and Earth's Human Carrying Capacity 65

improving the status of women; educating men; and various combina- no. 9 [September 1960]: 195-204) to 330 million (J. Durand, Population
tions. Unfortunately, there appears to be no believable information and Development Review 3, no. 3 [1977]: 253-96).
to show which approach will lower a country's fertility rate the most, 3. 1998 World Population Data Sheet (Washington, D.C.: Population Refer-
now or a decade from now, per dollar spent. In some developing coun- ence Bureau, 1998).
tries such as Indonesia, family planning programs interact with educa- 4. P. Demeny in Resources, Environment, and Population: Present Knowl-
edge, Future Options K. Davis and M. S. Bernstam, eds. (New York:
tional, cultural and, economic improvements to lower fertility by more
Oxford University Press, 1991), p. 416, gives a grim list of such environ-
than the sum of their inferred separate effects. 33 Some unanswered ques- mental problems: "loss of topsoil, desertification, deforestation, toxic poi-
tions are: how soon will global fertility fall? by what means? at whose soning of drinking water, oceanic pollution, shrinking wetlands, overgraz-
expense? ing, species loss, shortage of firewood, exhaustion of oil reserves and of
The "better manners" school says: improve the terms under which various mineral resources, siltation in rivers and estuaries, encroachment
people interact (for example, by defining property rights to open-access of human habitat on arable land, dropping water tables, erosion of the
resources; by removing economic irrationalities; and by improving ozone layer, loss of wilderness areas, global warming, rising sea levels,
governance). 34 When individuals use the environment as a source or a nuclear wastes, acid rain."
sink and when they have additional children, their actions have conse- 5. ]. Lederberg, journal of the American Medical Association 260, no. 5
quences for others. Economists call "externalities" the consequences (1988): 684-85; S. S. Morse, ed., Emerging Viruses (New York: Oxford
that fall on people who are not directly involved in a particular action. University Press, 1993); R. M. Anderson and R. M. May, Infectious Dis-
eases of Humans: Dynamics and Control (Oxford, England: Oxford Uni-
That individuals neglect negative externalities when they use the envi-
versity Press, 1991), chap. 23.
ronment has been called "the tragedy of the commons" 35 ; that individ-
6. In 1960, the richest countries, with 20 percent of world population, earned
uals neglect negative externalities when they have children has been 70.2 percent of global income, while the poorest countries, with 20 per-
called "the second tragedy of the commons." 36 The balance of positive cent of world population, earned 2.3 percent of global income. Thus, the
and negative externalities in private decisions about fertility and use of ratio of income per person between the top fifth and the bottom fifth was
the environment depends on circumstances. The balance is most fiercely 31:1. In 1970, that ratio was 32:1; in 1980, 45:1; in 1991, 61:1. In U.S.
debated when persuasive scientific evidence is least available. Whatever dollars, the absolute gap between the top fifth and the bottom fifth rose
the balance, the neglect by individuals of the negative externalities of from $1,864 in 1960 to $15,149 in 1989 (United Nations Development
childbearing biases fertility upward compared to the level of aggregate Programme, Human Development Report 1992 [New York: Oxford Uni-
fertility that those same individuals would be likely to choose if they versity Press, 1992], p. 36, Human Development Report 1994, p. 63).
could act in concert or if there were a market in the externalities of 7. P. Demeny, Population and Development, International Conference on
childbearing. Voluntary social action could change the incentives to Population and Development 1994 (Liege, Belgium: International Union
for the Scientific Study of Population, 1994); these numbers are based on
which individuals respond in their choices concerning childbearing and
World Bank estimates.
use of the environment. 8. From 1950 to 1995, the world's urban population increased more than
3.5-fold, from 0.74 billion to 2.6 billion, and from 29 percent to 45 per-
cent of the total population (United Nations, World Urbanization
Notes Prospects: The 1992 Revision, ST/ESNSER.N136 [New York: United
1. L. R. Brown and H. Kane, Full House: Reassessing the Earth's Population
Nations, 1993), pp. 74-75, 82-83.
Carrying Capacity (New York: W.W. Norton, 1994); F. Duchin and G. 9. 1998 World Population Data Sheet; World Population 1994 (New York:
Lange, The Future of the Environment: Ecological Economics and Tech-
United Nations, 1994).
nological Change (New York: Oxford University Press, 1994); N. Myers
10. United Nations, Department of International Economic and Social Affairs,
and J. L. Simon, Scarcity or Abundance? A Debate on the Environment
Long-Range World Population Projections: Two Centuries of Population
(New York: W.W. Norton, 1994).
Growth 1950-2150, ST/ESNSER.N125 (New York: United Nations,
2. M. Livi-Bacci, A Concise History of World Population, trans. Carl Ipsen
1992).
(Oxford, England: Blackwell, 1992). Estimates of global population size at
11. Systematic retrospective analyses of past population projections indicate
A.D. 1 vary from 133 million (E. S. Deevey, Jr., Scientific American 203,
that more confidence has been attached to projections than was justified
66 Part II. Science Population Growth and Earth's Human Carrying Capacity 67

by their eventual predictive accuracy; M. A. Stoto, Journal of the Ameri- (1971): 23-49, modified the logistic curve to allow for faster than expo-
can Statistical Association 78 (1983): 13-20; N. Keyfitz, Population nential growth followed by leveling off; they fitted their curve to past
Change and Social Policy (Cambridge, Mass.: Abt Books, 1982), chap. 13. global population sizes and predicted an asymptote around 50 billion peo-
12. A. van Leeuwenhoek, Collected Letters (Amsterdam: Swets and Zeitlinger, ple.
1948), Letter 43, April 25, 1679, vol. 3, pp. 4-35. Leeuwenhoek multi- 16. H. Brown, The Challenge of Man's Future: An Inquiry Concerning the
plied his estimate of the population of Holland (1 million people) by his Condition of Man During the Years That Lie Ahead (New York: Viking,
estimate of the ratio of the earth's habitable land area to Holland's area 1954); H. Brown, J. Bonner, and]. Weir, The Next Hundred Years: Man's
(13,385). Natural and Technological Resources (New York: Viking, 1957); C. Clark
13. ]. E. Cohen, How Many People Can the Earth Support? (New York: W.W. in Nature 181 (May 3, 1958): 1235-36, reprinted in W. Y. Davis, ed., Read-
Norton, 1995). ings in Human Population Ecology (Englewood Cliffs, N.J.: Prentice-Hall,
14. V. Smil, Population and Development Review 20 (June 1994): 255-92, 1971), pp. 101-06; M. Cepede, F. Houtart, and L. Grond, Population and
estimated 10-11 billion; D. Pimentel, R. Harman, M. Pacenza,]. Pecarsky, Food (New York: Sheed & Ward, 1964); W. R. Schmitt in Annals of the
and M. Pimentel, Population and Environment 15, no. 5 (May 1994): New York Academy of Science 118, no. 17 (1965): 645-718; H. Lieth in
347-69, estimated< 3 billion; P. E. Waggoner "How Much Land Can Ten Human Ecology 1, no. 4 (1973 ): 303-32; K. Blaxter, People, Food and
Billion People Spare for Nature?" in Task Force Report No. 121 (Ames, Resources (Cambridge, England: Cambridge University Press, 1986); P.
Iowa: Council for Agricultural Science and Technology, February 1994) Buringh, H. D. ]. van Heemst, and G. ]. Staring, Computation of the
estimated at least 10 billion; Brown and Kane, Full House, p. 202, esti- Absolute Maximum Food Production of the World (Wageningen, The
mated that a projected world grain harvest of 2.1 billion tons in 2030 Netherlands: Agricultural Press, 1975); P. Buringh and H. D.]. van Heemst,
could feed 2.5 billion people at the U.S. consumption level of 800 kilo- An Estimation of World Food Production Based on Labour-Oriented Agri-
grams per year per person, or just over 10 billion people at the Indian level culture (Wageningen, The Netherlands: Agricultural Press, 1977); G. M.
of consumption of 200 kilograms per year per person; Wetenschappelijke Higgins, A. H. L. Kassam, L. Naiken, G. Fischer, and M. M. Shah, Poten-
Raad voor het Regeringsbeleid [Scientific Council for the Dutch Govern- tial Population Supporting Capacities of Lands in the Developing World:
ment], Duurzame risicos: een blijvend gegeven [Sustainable risks: an Technical Report of Project INT/75/P13, "Land Resources for Populations
enduring given] (Den Haag, The Netherlands: Sdu Uitgeverij Plantijn- of the Future," FPA/INT/513 (Rome: Food and Agricultural Organization
straat, 1994), p. 9, estimated 11-44 billion people, depending on the sce- of the United Nations, 1983); Robert S. Chen et al., eds., The Hunger
nano. Report: 1990 (HR-90-1, Alan Shawn Feinstein World Hunger Program,
15. For example, R. Pearl and L.]. Reed, in R. Pearl, ed., Studies in Human Brown University, June 1990); S. R. Millman et al., The Hunger Report:
Biology (Baltimore: Williams and Wilkins, 1924 ), chap. 25, p. 632, fitted Update 1991 (HR-91-1, Alan Shawn Feinstein World Hunger Program,
a logistic curve to past world population sizes and confidently estimated a Brown University, April1991); Brown and Kane, Full House, p. 31.
maximum world population of 2 billion. The world's population passed 2 It is remarkable that food continues to be viewed as a limiting constraint
billion around 1930. Undeterred, R. Pearl and S. Gould, Human Biology on population size even though, globally, the countries with the lowest fer-
8, no. 3 (1936): 399-419, again used the logistic curve to project 2.645 bil- tility and the lowest population growth rates are among those where food
lion people as an ultimate limit to be nearly approached by the end of the is most abundant(]. Mayer, Daedalus 93:3 (1964): 830-44.
twenty-first century. That population size was surpassed before 1955. On 17. A. Penck, Zeitschrift fur Geopolitik 2 (1925): 330-48; Sitzungsberichte
a logarithmic scale of population, the logistic curve is concave, while the der Preussischen Akademie der Wissenschaften 22 (1924): 242-57. The
observed trajectory of global population size was convex until about 1970. formula was used but not stated explicitly in 1917 by G. H. Knibbs, The
The failures of Pearl's logistic projections and the usefulness of Alfred ]. Mathematical Theory of Population, of Its Character and Fluctuations,
Lotka's theory of population growth and age-composition (Theorie analy- and of the Factors Which Influence Them, Appendix A (Melbourne: Cen-
tique des associations biologiques, II Analyse demographique avec appli- sus of the Commonwealth of Australia, Minister of State for Home and
cation particuliere a l'espece humaine [Paris, Hermann: 1939]) led demog- Territories; McCarron, Bird & Co.), vol. 1, p. 455.
raphers to abandon studying the absolute size of populations in favor of 18. In 1972, domestic animals were fed 41 percent of all grain consumed; in
studying population structure and change. Since World War II, estimates 1992, 37 percent (Brown and Kane) Full House, p. 67; World Resources
of the earth's human carrying capacity have been published almost exclu- Institute, World Resources 1994-95 [New York: Oxford University Press,
sively by nondemographers. Demography, like economics, still lacks a 1994], p. 296).
working theory of scale. In another example of curve fitting, A. L. Austin 19. S. R. Eyre, The Real Wealth of Nations (New York: St. Martin's Press;
and ]. W. Brewer, Technological Forecasting and Social Change 3, no. 1 London: Edward Arnold, 1978).
68 Part II. Science Population Growth and Earth's Human Carrying Capacity 69

20. A. H. Westing, Environmental Conservation 8, no. 3 (1981): 177-83. ference on Environment and Development, held in Rio de Janeiro in June
21. ]. F. von Liebig, Principles of Agricultural Chemistry (New York: John 1992 (Rio Earth Summit).
Wiley, 1855); German edition: Die Grundsatze der Agriculturchemie 28. B. Zaba and I. Scoones in Environment and Population Change, B. Zaba
(Braunschweig: F. Vieweg und Sohn). Also see D. L. DeAngelis, Dynamics and J. Clarke, eds. (Liege, Belgium: Ordina Editions, 1994), pp. 197-219;
of Nutrient Cycling and Food Webs (London: Chapman and Hall, 1992), H. R. Pulliam and N. M. Haddad, Bulletin of the Ecological Society of
pp. 38-45, 228. America 75 (1994): 141-57; J. E. Cohen, How Many People Can the
22. Liebig's law extends to any number of independent constraints. When pop- Earth Support? chap. 12.
ulation on the left side of the formula is replaced by production, the for- 29. T. R. Malthus, An Essay on the Principle of Population, complete 1st ed.
mula is known in economic theory as the Walras-Leontief-Harrod-Domar (1798) and partial 7th ed. (1872) reprinted in Gertrude Himmelfarb, ed.,
production function. On Population (New York: Modern Library, 1960), chap. 7, p. 51. P.
23. V. Smil, Population and Development Review 17, no. 4 (1991): 569-601, Demeny, in Population and Resources in Western Intellectual Traditions,
pp. 586, 597, reported that in the 1980s, nitrogen applied in Zhejiang and M.S. Teitelbaum and J. M. Winter, eds. (New York: Population Council,
Shandong provinces of China increased rice yields by amounts that were 1989), p. 232, generalized Malthus's view to incorporate all aspects of eco-
only 50-60 percent as large as the increments from an additional kg/ha nomic output, not just food: "Posed in the simplest terms, the economics
applied in the 1960s. of population reduces to a race between two rates of growth: that of pop-
24. Examples of system models are: J. W. Forrester, World Dynamics (Cam- ulation and that of economic output."
bridge, MA: Wright-Allen Press, 1971); D. Meadows, D. L. Meadows, J. 30. F. L. Pryor and S. B. Maurer in Journal Development Economics 10
Randers, and William W. Behrens III, The Limits to Growth (New York: (1982): 325-53; R. D. Lee in The State of Population Theory: Forward
New American Library, 1972); M. Mesarovic and E. Peste!, Mankind at from Malthus, D. Coleman and R. Schofield, eds. (Oxford: Basil Black-
the Turning Point (New York: E. P. Dutton and Reader's Digest Press, well, 1986), pp. 96-130; R. D. Lee in Mathematical Population Studies l,
1974); P. House and E. R. Williams, The Carrying Capacity of a Nation: no. 3 (1988): 265-88; R. D. Lee, Population (Paris) 47, no. 6 (1992):
Growth and the Quality of Life (Lexington, Mass.: D. C. Heath, 1975);]. 1533-54; R. D. Lee in Explorations in Economic History 30 (1993): 1-30;
Gever, R. Kaufmann, D. Skole, and C. Vorosmarty, Beyond Oil: The M. Kremer in Quarterly Journal of Economics 108, no. 3 (1993):
Threat to Food and Fuel in the Coming Decades, 3rd ed. (Niwot: Univer- 681-716.
sity Press of Colorado, 1991); A. J. Gilbert and L. C. Braat, eds., Model- 31. J. H. Ausubel in The Sciences (New York Academy of Sciences) 33, no. 6
ling for Population and Sustainable Development (London: Routledge, (1993): 14-19; P. E. Waggoner, "How Much Land Can Ten Billion People
1991); W. Lutz, ed., Population-Development-Environment: Understand- Spare for Nature?"
ing Their Interactions in Mauritius (Heidelberg: Springer-Verlag, 1994). 32. J. Bongaarts in Science 263 (1994): 771-76.
Critiques of systems models are: C. Kaysen in Foreign Affairs 50, no. 4 33. P. J. Gertler and J. W. Molyneaux in Demography 31, no. 1 (1994): 33-63.
(July 1972): 660-68; H. S. D. Cole, C. Freeman, M. Jahoda, and K. L. R. 34. R. Repetto, ed., The Global Possible: Resources, Development, and the
Pavitt, eds., Models of Doom: A Critique of The Limits to Growth, with New Century (New Haven, Conn.: Yale University Press, 1985); D. W.
a reply by the authors of The Limits to Growth (New York: Universe Pearce and J. J. Warford, World Without End: Economics, Environment,
Books, 1973); W. D. Nordhaus, Economic Journal 83, 332 (December and Sustainable Development (New York: Oxford University Press, 1993).
1973): 1156-83; E. van de Walle, in Science 189 (1975): 1077-78; D. 35. G. Hardin in Science 162 (1968): 1243-48.
Berlinski, On Systems Analysis (Cambridge, Mass.: MIT Press, 1976); P. 36. R. D. Lee in Kingsley Davis and Mikhail S. Bernstam, eds., Resources,
R. Ehrlich, A. H. Ehrlich, and J. P. Holdren, Ecoscience: Population, Environment, and Population: Present KriOwledge, Future Options (New
Resources, Environment (San Francisco: W.H. Freeman, 1977), pp. York: Oxford University Press, 1991), pp. 315-22. Earlier discussions of
730-33. the negative externalities of childbearing include H. F. Dorn in Science
25. P. M. Fearnside, Human Carrying Capacity of the Brazilian Rainforest 135, no. 3500 (1962): 283-90; P. E. Sorenson inN. Hinrichs, ed., Popu-
(New York: Columbia University Press, 1986). lation, Environment and People (New York: McGraw-Hill, 1971), pp.
26. In many regions, the average amount of fresh water available annually is 113-21; P. Dasgupta and R.]. Willis in D. G. Johnson and R. D. Lee, eds.,
more than twice the amount of water that can be counted on ninety-five Population Growth and Economic Development: Issues and Evidence
years in one hundred; (P.P. Rogers in The Global Possible: Resources, (Madison: University of Wisconsin Press, 1987), pp. 631-702.
Development, and the New Century, R. Repetto, ed. (New Haven: Yale 37. C. McEvedy and R. Jones, Atlas of World Population History (New York:
University Press, 1985), p. 294. Viking Penguin, 1978); J.-N. Biraben in Population (Paris) 34, no. 1
27. Paragraph 5.23 of Agenda 21, final document of the United Nations Con- (1979): 13-25. See also Durand, Population and Development Review.
70 Part II. Science

38. C. M. Cipolla, Economic History of World Population, 6th ed. (Har-

mondsworth, England: Penguin, 1974), pp. 56-9; World Resources Insti-
tute, World Resources 1994-95 (New York: Oxford University Press,
1994), pp. 332-34.
39. Data are from note 13, Cohen, How Many People. The estimate by J. H.
Fremlin in New Scientist 24, no. 415 (October 29, 1964}: 285- 87, would
be off the scale and is omined.
 CONSUMPTION,
POPULATION,
-AND-
SUSTAINABILITY
Perspectives from
Science and Religion

Edited by Audrey R. Chapman,

Rodney L. Petersen, and Barbara Smith-Moran

ISLAND PRESS
Washington, D.C. Covelo, California