CURRENT ANTHROPOLOGY Volume 28, Number 2., April 1987 CI 1987 by The Wenner-Gren Foundation for Anthropological Research, all righlS reserved 0011-3204/8712802-0003$1.75

Ph.D., 19591 and, after teaching for two years at the University of New Mexico, returned to join its faculty in 1961. He was Visiting Fellow at Cambridge University in 1970-71. His research inter­ Research in ests are historical demography, social Structure, cognitive an­ thropology, the ethnography of Europe (especially the Balkansl, and quantitative analysis. He has published Power in lea: The So· Population cial History of a Peruvian Valley (BostOn: Little, Brown, 1969); Ritual Relations and Allernalive Social Structures in the Balkans {Englewood CWfs: Prentice·Hall, 1968}; with K. Wachter and and Culture: P. Laslett, Statistical Studies of Historical Social Structure (New York: Academic Press, 1978h with S. R. Johansson and C. Gins­ berg, 'r'fhe Value of Children during Industrialization" 1I0UIna] of An Evolutionary Family History 8:346-66); and "Short·term Demographic Fluctua­ tions in the Croatian Military Border of Austria" (European lour· 1 nal of Demography 1:265-90). Framework NANCY HOWELL is Post-doctoral Research Fellow in the Program in Organizations and Aging, Department of Sociology, Stanford University. She was born in 1938 and received her B.A from Bran­ deis University in 1963 and her Ph.D. from Harvard University in by E. A. Hammel 1968. She has taught at Princeton University (1969-72) and the University of Toronto 11972.-85). She is a fellow of the American N Association for the Advancement of Science and did this work and aney Howell while a fellow at the Center for Advanced Study in the Behavioral Sciences, Stanford, Calif. Her research intereSts are demographic anthropology and occupational health and safety in anthropology. Her first book was Search for an Abortionist (Chicago: University of Chicago Press, 1969), and she is the author of "Village Compo­ Theories on the relationship between population and social evolu· sition of a Paleodemographic Life Table: The Libben Site" (Ameri­ tion, such as those of Malthus, Marx, and Boserup, have disap­ can lournal of Physical Anthropology 59:2.63-69), "TowaId a pointed analysts who wish to use such theories as a basis for infer' Uniformitarian Theory of Human Paleodemography" (Journal of ence in unrecorded history, largely because of the unrealistic Human Evolution S:25-4ol, and Demography of the Dobe !Kung specificity of the hypotheses and the lack of adequate data in cate· lNew York: Academic Press, 1979). gories that help to clarify the application of theory to human pop­ ulation history. We propose that it is time to reformulate such a The present paper was submitted in final form 8 VII 86. theory, drawing upon a new level of sophistication in an­ thropological demography that pennits more complexity in the theory itself about alternative outcomes of stressful points in pop­ ulation history and the testing of hypotheses in more realistic Four observations motivate us to make the following contexts of continentwide examinations of populations influenc­ proposals on the place of population theories in an­ ing each other by their expansions and contractions. We antici· pate finding that population fission is a particularly interesting thropology. First, we see the fundamental demographic alternative response to population pressure when we are dealing events of birth, marriage, migration, and death as im· with relatively sparse populations and that the place of the given bued by all human societies with intense emotional and population, in the core or on the periphery of population growth moral significance. These events are central to the centers, is crucial in influencing the alternatives used. drama of human life, and all cultures treat them as im­ portant and interesting, worthy of elaboration and cere­ E. A. HAMMEL is Professor of Anthropology and Demography at the University of California, Berkeley (Berkeley, Calif. 9472.0, monial notice. U.S.A.}. Born in 1930, he was educated at Berkeley lA.B., 1951; Second, we are encouraged by the increases in techni­ cal quality of the demographic studies in anthropology that have been appearing in the past few years (Howell 1. Portions of thiS paper were delivered as the keynote section of a 1986aj. This higher-quality analysis and discussion, symposium on anthropological demography at the annual meet· ings of the American Anthropological Association, Washington, largely free from the ambiguities and errors that charac­ D.C., December 6,1985. We acknowledge the contributions of the terized demographic anthropology in the past, give us participants in that symposium, especially Jane Buikstra, Mark Co· optimism aboUl the expected yield from theoretical hen, Henry Harpending, Penn Handwerker, Ann Millard, Robert work in this area. Netting, Lucille Newman, Patricia Reining,. Susan Scrimshaw, and Third, we see the subject matter of demography as G_ William Skinner, both in the fonn of their own papers and discussions and in the comments they made, both prior to the cutting across and unifying the subdisciplines of an· meeting and at the meeting, on our remarks. Similarly, we thank thropology as few other topics do. A focus on demo­ George Collier, William Durham, James Spuhler, and Aram Yen­ graphic phenomena provides a common ground for ob­ goyan, with whom we consulted at an earlier stage in the arrange· servation, discussion, and intellectual integration across ment of the symposium. A version of this paper was also presented at the Stanford·Berkeley Colloquium in Population, San Francisco, the subfields. The simplicity of demographic concepts, December 10, 1985, and to seminar participants at Berkeley in the the possibility of unambiguous counting procedures, spring. We appreciate the advice and comments of participants, and the consequent ability to test hypotheses provide especially of Paul David, Ronald D. Lee, Nicholas Townsend, and opportunities for new research and new collaboration Kenneth Wachter. Our interest in this topic and especially in evo­ lutionary economics from a demographic perspective owes much between physical and biological anthropologists, archae­ to the inspiration, encouragement, and critical assessments of Lee ologists, and social and cultural anthropologists inter­ and Wachter, with whom we continue to differ on some points. ested in a range of questions.

r41 1421 CURRENT ANTHROPOLOGY Volume 28, Number 2, April r987

Finally, we see human demographic regimes as impor­ and a breeding population. From the time of its initial tant causes and effects in a theory of the coevolution of expansion, H. sapiens was neither. The stochastic pro­ the human species and the cultural equipment that it cesses leading to extinction or infinite expansion applied employs. Evolutionary theory has the potential to be to local, not global populations. The zero or near zero net central land integratingj to anthropological theory, but growth of the global population could have been an aver­ the specifications of evolutionary theory have been de­ age of widely differing local population growth rates. If terministic and teleologicat and its mechanisms for theories about response to resource constraints are to be mankind have often been vague. A demographic per­ applied within the framework of economic rationality spective may dispel some of this vagueness by focusing and decision making, or even within the framework of upon recognizable events in real human groups. The role random variation and selection, they must apply to pop­ of culture in human evolution has presented many ulations sensu stricto, not to aggregates of populations. difficulties fOI theory. A demographically based formu­ How to combine theoretical principles that operate at lation may integrare cultural with biological evolution the local level with concepts of global population and IBennett 1976; Durham 1978; Harris '977, 1985; Wilson resources is an important problem, and when in human 19751· history we have to begin to think about locaVglobal dif­ We fIeely admit that none of the ideas here are entirely ferences is another. How can we tum these ideas and the new, even when we cannot at the moment point to our basic notions of evolution itself to an understanding of predecessors. What we propose is not really a theory but the coevolution of human demographic regimes, the only a starting formulation, a partial specification, and species, and its cultures? an incitement, to ourselves and to colleagues, to join in construcring a philosophically acceptable extension of evolutionary theory, one in which demographic events Elements of the Theory are the central mechanism and leading indicators of the coevolution of bodies, minds, and societies. We cannot The main thrust of our argument is to place the develop­ emphasize too stronglYI however, that in our view lIeve­ ment of human demography and coevolved culture lutian" is not a direction but a set of interacting mecha­ squarely within the framework of biological evolution­ nisms resulting in the production of variation and its ary theory-evolutionary theory as modern biologists selection. know it, not as some others remember or imagine it. The two main points of that theory, simply conceived, are II) that variations-in genes, in individuals, and in cultural Global Population History: A Statement elements, large and small-appear as lIenorsH in the pro­ of the Question cess of transmission of elements of information from one generation to the next and 12) that some of these Our colleagues R. D. Lee and K. Wachter have in recent /lerrors" have more reproductive success than others in seminars and in forthcoming publications addressed particular environments. Their initial appearance is in­ themselves to problems of population homeostasis in dependent of their potential reproductive success, so premodern populations. At the base of these explora­ that this evolutionary scheme in its simplest form de­ tions is the question why human populations grew so pends only on random variation and selection, in the slowly over the enormously long span of prehistory. classic style of the Huxley synthesis. Wachter 119851 spells out that the theory of branching We expand on these basic notions in minor ways. A processes in statistics tells us that, no matter how slow a first modification is obvious: beyond the occurrence of rate of growth, the populationlsl characterized by it will, random variation, sentient goal-oriented animals can given enough time/ become either extinct or of infinite and do give cultural development a twist in which the size. The population of H. sapiens has done neither dur­ occurrence of "copying errors' I on any level can be inten· ing the period of observation, although it may have had tional and designed to be adaptive. IWhether they are time to do both. Cri tical to this question is the adaptive, and for how long and in what environments, is definition of the date and size of the "starting popula­ moot.IThat is, in humans the invariability of the statis­ tion. 1I Under some plausible assumptions, Wachter tical independence of variation and its selection is lost in shows, the inevitable outcomes of branching-process principle at least in part. Second, we propose that one theory may not have had time to emerge. Lee 119841 has response to reproductive success in the presence of maintained that the answer lies in the negative feedback ecological constraints is group fission and that fission mechanisms of Malthusian homeostasis operating on can act as an intensifier of evolutionary development, the population demand side and that at some point a multiplying the opportunities for change. Third, we pro­ threshold is crossed at which Boserupian endogenous pose that fission makes some kinds of cultural develop­ technological change operates on the resource supply ments, especially communicative ones, particularly side. adaptive and that these developments have substantially OUf response is that these are answers to the wrong changed the environment of demographic and social pro­ question. It seems to us misleadin& at least at this stage cesses. of our theoretical sophistication, to talk about the total We need to provide a few definitions and some population of a species unless that species is localized clarification of our assumptions. A population is a group HAMMEL AND HOWELL Population and Culture I r43 of self-replicating organisms of a single species, such alternatives unavailable to a population for any reason, that most of the breeding is done within the population. including lack of organization, are not alternatives for Another useful definition stresses the multigenerational them} and that the actions and decisions of other popula· nature of a population: it consists of the common de­ tions will frequently be major factors in the immediate scendants of a group of ancestors, and within it future situation of a particular population. descendants are produced and raised. The minimum size Finally} what do we mean by scarcity, given our desire of a population will be set by the ability of its members to formulate a framework that will work equally well for to provide mates for one another's offspring. The max­ protohominid bands and the "world system" of modem imum size will be set by the ability of its members to market capitalism (Hayden 1975)? In most cases, we ex­ mix freely with one another, so that an individual is pect that absolute declines in food or other material re­ about as likely to select a mate from one segment of the sources signal scarcity} but the level of need per person is population as from another. A population can be inter· subjective and variable. There is perhaps no level of nally differentiated, but its members necessarily share a plenty that would satisfy an elite focused upon con­ geographic base, characteristic responses to problems, sumption and perhaps hardly any level of scarcity thar language and culture, etc. Groups that violate these would fail to satisfy a Zen community. We assume that definitional criteria are special cases that may not be a population needs to meet some standard of per capita well described by our model. We do not assume that distribution of goods and services to its members. The populations are tighrly bounded or sealed-off units: the needs are met by the application of land, labor, capital} island isolate population is a rare, even if an important, and organization to the environmental resources avail· phenomenon in population history IBenoist r9731. We able to the population. We assume that there is a con­ expect that in every generation some members of the stant process of assessment and evaluation of the rela­ population will go outside, and others will come in, tion of population needs to resources. When the bringing their genes as well as their culture, their experi­ resources, compared with the absolute standards of the ence} their language} and their social ties with them. The people or with the standard of the recent past, for ex­ degree of transfer between populations per generation is ample, are in balance with the needs of the population, an empirical question complicating our need for an oper­ we anticipate the continuation of the population with­ ational definition of population. Adams and Kasakoff out pressure for change in any of the elements of the Ir975) suggest that 80% of breeding choices made within equation. (We do not assert that social} cultural, or tech­ a group is a useful and realistic leven if arbitraryI crite­ nological change will only occur when there is popula­ rion. One implication of these definitions is that a popu­ tion pressure/ but merely that population pressure in lation is frequently not coterminous either with the 10­ this sense will always produce a need for evaluation of cal living group (the population may be much largerl or change.) Population pressure need not be produced by with the society Ithe population may be much smallerl. population growth, although thar is one frequent path to The concept of "effective population size" IKimura and pressure. Others are reductions in productivity of the Crow 1963), difficult as it is to operationalize, is helpful technological base, a loss of land or other elements of in this context (see also Barth 1969) Hammel, Wachter, production} an invading army, or just a run of bad and McDaniel 1983, Wobst r9741. weather or bad luck. When there is a surplus of resources The assertion that a population "confronts Jl scarcity relative to population Iwhich can be caused by a decline and "responds" to it by making choices or selecting al­ in population as well as by an increase in resource avail­ ternatives is difficult. What does it mean to say that a ability), the surplus may be "spent" in the form of less unit as potentially amorphous and unorganized as a work} more leisure, more pleasure, or investment in the breeding population makes decisions or has alterna­ elements of production for future consumption. Alterna­ tives? The units involved in perceiving and deciding are tively, the population may respond to surplus by an in­ likely to be subunits of a population-individuals, crease in size or by some combination of increase in families, leaders, and, at some levels of population devel­ numbers las Malthus suggested long agol and increase opment, perhaps the government of a nation. Never­ in the size of individuals. theless we continue for reasons of theoretical Simplicity to focus on the level of the population, even when the spokesmen for and the obvious decision-makers in popu­ Responses to Population Pressure lations may not do so. We do not claim that populations "make history just Population history consists of the results of responses to as they wish/} to paraphrase Marx. We can assume that pressure on resources through a relatively small number actors will be rational, at least on the average and in the of choices available to populations. Such a history is long run, in the expectation that people prefer full bellies modified by stochastic variation, especially in small to empty ones, leisure to hard work, survival to death. populations. Through these responses and fluctuations} But we also recognize that rational actors may not per­ populations are created, maintained} and transformed ceive these issues, may have other agendas than demo­ into larger or smaller units, and organizations and ideas graphic ones, and may well not care about the unit of the and technological systems are created and transformed. breeding population. We recognize that populations can The direction of change observed in surviving popula­ only move from where they are at the moment, that tions has tended to be toward the larger and more com­ 1441 CURRENT ANTHROPOLOGY Volume 28, Number 2, April 1987

plex, but we do not assert a uniditectional system. Small might be classified under Alternative 2, or changes in societies are at greater risk of disappearance for competi­ distribution, which might be included under 3. tive and stochastic reasons and are less likely to be part These outcomes should be considered not as compet­ of a set of survivors, thus those tending toward small­ ing theories of population adjustment but rather as alter­ ness ate less likely to appear to us. While combinations native responses used or experienced by populations of responses are possible, in the most general terms we under varying conditions, with considerably different see four alternative responses to population pressure as consequences for the future of the population in ques­ explaining most of population history. In discussion of tion land perhaps of others, to the extent to which popu­ all of these points we must recognize complementary lations are an element in the reality confronted by other responses when population pressure is negative, al­ populations) within an evolutionary theory. (This view though we go into little detail. of complementarity, at least of Malthusian and Boseru­ 1. Fission. The population can expand geographically pian approaches, has also been propounded by Lee 11984] lor allemptto expand). The form that this takes will vary and Simon Ir9781, among others.j We see the task of depending upon whether the population has unoccupied demographic anthropology as specifying the conditions land available and with how large a membership its so­ of complementarity and testing the hypotheses derived cial organization can cope. When there is vacant land from them. and social organization is relatively inelastic, we expect expansion by fission. When the population is surrounded by other popula tions and there is no unoccupied land Initial Theoretical Considerations available, we look for forms of conflict with neighbors, and we expect to examine catefully the differing posi­ The names of three major theorists emerge in any dis­ tions of those who startterntorial conflict as opposed to cussion of the dynamics of population: Malthus, those who are trespassed upon and different positions for Boserup, and Marx. We take these three theories in their those who win and those who lose. When population simplest form as making these basic points: pressure is negative we may often expect fusion of social All three are characterized by some notion of the groups in response. short-term finite nature of resources for exploitation by 2. Limitation. The population can restrict fertility, in­ human populations, a notion of a carrying capacity of crease mortality, or otherwise modify itself to adjust the environment more or less complexly determined by population size to the available resources. This may in­ the qualities of that environment and by the percep­ volve deliberate changes or quite unconscious adjust­ tions, habits, technology, and social structure of the ex­ ments and second-order effects and may require techno­ ploiting groups. logical or cultural change or be strictly biological. Clear All three agree that humans, like other animal popula­ density-dependent population restriction does not work tions, are characterized by an inherent tendency to in­ invariably and automatically for human populations, crease toward that carrying capaCity as defined in the but this is not to say that it plays no role in population short term, in the absence of any customs or responses regulation IHowell 1980j. Malthus 11976 117981J called that would diminish that increase. these responses to population pressure lithe preventive In Malthus's 11976 117981J view, an approach too near checks. II When population pressure is negative and re­ the resource boundary would trigger misery in varying sources exceed demand, population may grow as a con­ combinations of the Four Horsemen of the Apocalypse. sequence of inherent excess fertility capability, or spe­ These positive checks on population could be avoided by cial measures may be taken to enhance fecundability, three mechanisms: adoption of methods of fertility con­ depress mortality, increase exposure to the risk of preg­ trol, prudential delay in marriage {either of these holding nancy, etc. down the increase in populationJ, and exogenous tech­ 3· Intensification. The population can increase density nological change that increased the resource base. He and adapt itsel1 to increased population size by intensifi­ regarded the first mechanism as a vice, tbe second as cation of food production (or whatever the scarce re­ morally appropriate, and the third as providing only sources may be). This response usually involves techno­ temporary surcease. This is, with apologies to the se­ logical change and is familiar as Boserup's 11965j view of quence of historical events, a Danvinian view, in which population adjustment. If population pressure is nega­ tbe appearance of technological solutions is not (neces­ tive, the population can fall back on less intensive sarilyj brought about or encouraged by the conditions of modes of production. Modifications of the practices of existence but exogenous in the economist's terms or distribution of the product of economic activity may random in those of the evolutionary biologist. Thus, al­ also be made, they may affect the resource balance of though technological change may relieve population segments of a population. pressure, its emergence is made no more likely by the 4· Decline. The population can restrict or reverse pop­ prior existence of such pressure. ulation growth by increases in misery, starvation, sick­ Boserup's (19651 view differs in proposing that techno­ ness, and war. Malthus called these factors lithe positive logical change emerges endogenously and in response to checks. II We would also include here decreases in con­ population pressure on resources, in principle infinitely sumption, although it might be difficult to distinguish extending the cycle of challenge and response. these empirically from shifts in consumption, which Marx's (r967 Ir86711 view, particularly in criticism of HAMMEL AND HOWELL Population and Culture I t4S

Malthus, was that Malthus misidentified the source of absence of some new technology or biological change the immiseration of the working class. It was not, for Isuch as developing the ability to digest cellulose), the Marx} the mere tendency of populations to increase to­ resource boundary of a given environment is fixed, the ward or beyond the carrying capacity of their environ­ ability of human populations to avoid such boundaries is ment but rather the maldistribution of the products of impressive. We call attention to the need for culturally their labor. appropriate and not merely external definitions of im­ Particularly in the light of the evolution of Malthus's miseration. thought as illustrated in the successive tefinements of Another difficulty lies in Malthus's oft-quoted asser­ his Essay, we see no reason these views should be re­ tion that population increases geometrically (exponen­ garded as necessarily incompatible. Political polemics tiallyl, while the food supply increases only arithmetic­ and rhetorical strategies aside, there is no barrier to ally (linearly). This notion is questionable in any context thinking of the resource base as defined by the environ­ but particularly before primary dependence on the con­ ment not only as culturally petceived and exploited but trol of fixed areas of land typical of most known hor­ also as socially distributed. Similarly, although history ticultural and agricultural societies. Food supplies are as we know its evidence has shown Malthus to be wrong populations of edibles that wax and wane just as geomet­ and Baserop right, in the sense that technological change rically as do their consumers. It is the territory suppon­ has in some sense continued to overcome resource scar­ ing the edibles that does not increase geometrically. If city, any definition of a finite universe would necessarily territorial limits are flexible, a population simply has to show the classical Malthus right, in principle, in the move enough to maintain some acceptable level of ex­ very long run. There are of course empirical questions to ploitation of renewable resources, whether these are be answered on the degree to which the positive and acorns, deer, or milpas. By contrast, if the area available preventive checks have operated; our point is only that is fixed, productivity can be increased only through within reasonably known history the inevitability of the some kind of technological change and probably only at positive checks and the rarity of technological rescue the expense of diminishing returns to labor. This obser­ seem less certain than Malthus painted them. The criti­ vation leads to the view that the problem posed by cal difference between the classic Malthusian and Malthus did not become important until population Boserupian views is whether the appearance of cultural densities became so great that mobility was restricted. variation is independent of its adaptive value-in this Our argument rests in part upon the recognition that context whether technological (or, more broadly, cul­ Malthus, Boserup, and Marx conceptualized the relation tural) change is exogenous. Even this difference is not 3 of population to resources as a brief encounter, with a necessary one in any broader theory of population and beginnin& a middle, and an end. Our extension of their resources} for there is nothing in Malthus insisting that argument attempts to take into account the continuous technological change be exogenous; his argument re­ nature of the confrontation of population and resources, quites only that it be, in the long run, futile. Com­ in which members of populations, generation after gen­ plementarily, there is nothing in Boserup that insists eration, experience and perceive surpluses and scarcities that technological amelioration is permanent; on the in resources relative to their numbers. We assume that contrary, the cycle of challenge and response extends biological and social responses to scarcities are transmit­ into the long run. How long that run will be, and what ted from one generation to the next, so that responses cultural considerations will intervene in tolerance of in­ which restrict population growth in one generation will creased densitiesl are unknown to us. likely do the same for subsequent generations. The prospect of broad theoretical unanimity should We stress that biological and cultural evolution pro­ not blind us to the existence of problematic areas. The vide alternative and sometimes complementary re­ first of these is the notion of carrying capacity or re­ sponses to population pressure. While it seems to be true source boundary, net of any technological innovation. A that responses have been increasingly social and cultural population lives in and simultaneously exploits a over time, there is no point at which biological evolu­ habitat, which consists of a territory and the resources, tionary responses end in the human record, and we especially food, contained in it. The perception of what should be cautious about assuming that cultural, social, is edible is not constant and immutable. Neither is the and technological responses cannot appear even at very perception of what is habitablel or marriageable, or ex­ early points in the hominid record. For instancel the ploitable in any other sense. Hierarchies of food prefer­ dates at which the origin of language and the control of ence are observable in animals and more elaborately de­ fire produced important consequences for population are veloped among human beings. Thus, part of the tesponse not known with any degree of certainty. to approaching the resource boundary is to eat some· Howell (t98 5b) has considered some of the elements thing different. This makes for a flexible resource bound­ of the culture of early hominids that are likely to have ary in twO senses. Firstl the eaters may move down the had considerable survival value and hence were particu­ hierarchy of food preferences. Second, they may alter the larly likely to be retained once established for whatever hierarchy of preferences. The first tactic allows popula­ reason. These elements include cooking, provisioning of tions to subsist at the cost of reducing culturally appro­ members of the group, especially children, beyond in­ priate satisfaction. The second device allows subsistence fancy/ and nursing of the sickl injured, and disabled of and recovers satisfaction. Although at some point, in the any age. She has argued that an automatic consequence 1

1461 CURRENT ANTHROPOLOGY Volume 28, Number 2, April 1987 of the development of these cultural traits would be an ous density. These innovations are of two kinds: one is increase in the frequency of survival of some adults past the other side of the Malthusian problem, namely, some the reproductive ages. These cultural traits would trans­ mechanism for the reduction of fertility, whether pru­ form a primate group from one resembling living groups dential or vicious (that is, a demand-side solutionI. The of gorillas and chimpanzees to something we would rec­ second is some mechanism for increase in resources ognize as human, but it is not at all clear that it was (that is, a supply-side solution). These latter are all more these and not some other cultural traits that trans­ or less Boserupian in principle, but they can include any­ formed the human species. thing from walking a little further each day or conquer­ Which responses to scarcity are elicited and which re­ ing one's neighbors to inventing new tools and tech­ warded and/or Ifpunished" by the environment can have niques. But the most interesting outcome would be important consequences for the population. The popula­ fission. If there is not enough to eat, some people go off tion itself may be transformed into a larger one, at a new to find more, and they mayor may not come back. All exploitative equilibrium, or may be restricted to its old animals range more widely in the face of diminishing size and exploitative techniques. Cultural conservatism, food supplies, and people tend to do so in familial groups or integrity if you will, makes cultural systems slow to that can form the core of new populations. respond to perceived scarcity. Nevertheless we take the What are the consequences of fission' The first is that view that in the long run population size and the cul­ the daughter groups are smaller than the parent group. tural corpus within a given environment are usefully As such they are more subject to the vagaries of chancej seen as having some long-term homeostatic equilibrium their demographic rates become unstable simply by vir­ point. Failure to achieve the particular point appropriate tue of their smaller size, so that they are more likely to to a given environment and given cultural corpus should begin precipitous declines or sharp accelerations in size. continue pressures in its direction or result in its modifi­ If groups decline in size, they can of course become ex­ cation through either biological or cultural change. tinct, thereby creating gaps in the network of communi­ cation that increase physical and cultural differentiation among those that survive. They can also rejoin the par­ The Fission Alternative ent groups or combine with some other group. Any such fusion is facilitated by the ability to communicate, espe­ The fission alternative is particularly interesting be­ cially if the separation has lasted more than, say, a gener­ cause we regard generational replication as deeply con­ ation or two. The development of cultural symbols, lan­ servative and insulated against change. Although cul­ guage in particular, facilitates either recognition and tural change is not the goal of the process of adjustment reincorporation or diplomacy and fusion. Symbols are to scarcity, it is often a consequence, and one of the most thus hedges against extinction (and speciationI. Lan­ interesting from the point of view of the long-term ob­ guage is important for more than intragroup interaction, server. We argue that cultural change is more likely in and the groups that developed it had a better chance of many ways in a small population than in a large one, not disappearing as victims of mere random fluctuation more likely in a population that is new in its niche than in their demographic rates. Kinship systems, permitting in one that is well established, and more likely to come differential claims on the resources of others, are a spe­ from a larger group of replicates than from a smaller cial form of symbolic system facilitating these pro­ group. Hence group fission provides more trials, more cesses. Social anthropology in its functionalist form has recombinations of cultures and population structure/ a lot to say about this and has had since McLennan and more combinations of cultures and niches, all of (18651. which will increase the probability of cultural change. A second consequence of fission is that smaller size None of this is to deny countervailing forces that facili­ combined with geographical separation makes it harder tate change in larger and denser social systems. to find mates, which condition may contribute to the A local breeding population somewhere along our line reduction of growth or, alternatively, increase the proba­ of ascent Ifor instance, see Chagnon 19801 might, under bility of cultural change in rules of mate selection or the the vagaries of existence, decline in size, in which case development of kinship nomenclatural systems ensur­ its members might disperse to join other similar groups, ing at least one incumbent in each named kinship role or its remnant might be absorbed in some other group, or (Hammel 19601. it might become extinct. None of these outcomes is par­ A third consequence of fission is that daughter soci­ ticularly interesting in an evolutionary sense, although eties will not be perfect copies of the parent society. Not absorption dampens the process of differentiation and for a very long time (if ever) has every adult in a human speciation, while extinction accelerates the develop­ group known exactly as much as every other. On fission, ment of lacunae that separate species. If such a popula­ some knowledge may be lost. It may of course be re­ tion were instead to increase in size and press against its gained by reinvention, but this may be as unlikely as a resource boundary, what might its responses be? repeat mutation recovering a lost biological characteris­ It might exhaust its resources, then decline in size and tic. (Since people began to think systematically about take one of the paths JUSt noted. Alternatively, it might problem-solving algorithms, the probability of repetition develop some innovation that would permit it to exist at of identical solutions has been enhanced.) Lost knowl­ the new, higher population density or revert to the previ­ edge may be regained by borrowing or fusion with other HAMMEL AND HOWELL Population and Culture I 147 groups, but these processes will involve some of the others and because selection pressure is greater at (or same mechanisms for diplomacy. Most interesting, new toward) the core. This is a density effect but a different solutions for old problems may emerge. Some of these kind from some others that may also playa role. For may work worse than earlier solutions, but some may example, even if the number of intelligent persons per work better. In fact, some may overcome the resource capita in a population does not increase, the number of limitations that earlier forced fission. One can think of a these per unit area is greater in denser populations. model in which economies of scale in large populations Thus, increased density may increase the rate of inven­ produce new ideas in segments of that population, ideas tion. Further, the chance that a new idea will take hold that can be implemented only with difficulry in the face is enhanced by density, since the chances of transmis­ of traditional conservatism and interest-group conflict sion to others are increased. There are, of course, but are then liberated for experimentation in small countervailing density-dependent resistances to change. daughter groups. For example, it is reasonable to speculate that increases The upshot of this is that fission does two important in density accompanied by increase in hierarchy and in­ things. It increases the survival value of fundamental terest-group formation may increase cultural conserva­ human characteristics such as intelligence and language, tism, time lost in diplomacy, and a variety of other kinds because these characteristics permit amelioration of iso­ of makework. Parkinson Ir97 51 and Murphy IBlock 19771 lation and hedge against downside demographic tisk. need to be included as sources of major theoretical en­ Further, it permits daughter groups a new chance at lightenment along with Malthus, Marx, Boserup, and breaking the resource barrier with a somewhat different Weber I(947). combination of biological and cultural resources. Repro­ There are a number of speculative consequences of ductive success that pushes societies against the gradual expansion and higher densities at the cores of Malthusian boundary is productive in an evolutionary areas beyond the idea of higher variability and innova­ sense not only because someone occasionally has a tive cycling at the core. The first of these is the develop­ bright idea but because a possibly common response, ment, through the encounter with new ecological condi­ fission, multiplies the number of semi-independent non­ tions as the periphery expands, of new cores and thus the identical trials for success. basis for regional systems. The second is the establish­ ment of differential access to the same resources within a given area, through hierarchicahzation or other differ­ Core-Periphery Population Development entiation of the social structure (such as ethnic or caste group specialization). The third, related to the last, is Some important differentiations in this hypothetical conquest. Under any of these, trade or exchange, equal or process must have taken place even at the beginning. unequal, may operate. In all of these circumstances the Imagine, for convenience, an ecologically homogeneous resource boundary limit remains the same but is over­ plain of vast extent, occupied at its center by an imper­ come for a larger population through some kind of differ­ fectly but reasonably well-adapted species which ex­ entiated redistributive scheme. Of course, stratification pands against its local resource boundary. It continually need not lead to increased efficiency. It can lead to an divides, like a colony on a petri dish, and daughter increase in the standard of living for the elite, at the groups surround it like rings on a pond. Suppose that expense of the mass, with no net change in overall per daughter groups on the periphery experience no changes capita consumption ltotal resources per capital. At the in technique. Indeed, all that the people on the periphery same time, consolidation of political control can raise have to do is walk centrifugally and breed. At the core, it the prnductivity of labor and decrease the wage, for a net is different because there is no place to walk. The envi­ increase in product, as a result of simple exploitation. ronments in which these otherwise identical or ran­ Differentiated social systems create new kinds of niches domly differing groups live on this ecologically homoge­ within which pressures for different kinds of demo­ neous surface become different in a density-dependent graphic behavior may operate. The conditions are laid way. There will always have been a difference in ease of down under which human reproduction, in the pursuit movement from the periphery toward the core as op­ of perceived self-interest, leads to higher or lower fertil­ posed to outward from the core and through the periph­ ity in different subgroups of the same population or at ery if there is any density gradient along a radius from least the same society. Other consequences are differen­ the core. tial mortality, nuptiality, and migration. Variability becomes important toward the center in One of the most important consequences of the pro­ two ways: first, it allows failures to emerge and go out of cess of group fission, as here conceptualized, has been business, making room for other groups. Second, it al­ the emergence of elaborate symbolic codes for identifica­ lows better solutions to emerge by the processes already tion and communication. These have more or less over­ mentioned and these to spread to other groups through come isolation, less earlier and more later. One direct communicative mechanisms. Under these circum­ consequence of increased communication was an in­ stances the rate of culture growth will be higher at the creased probability of epidemic disease. Increases in den­ center (or, if there is a density gradient, just closer to it ) sity following on successful technological innovations than on (or towardJ the periphery simply because some also facilitated the spread of epidemic disease, as did the variations have better adaptive consequences than maintenance of and close proximity to dense popula­ 1481 CURRENT ANTHROPOLOGY Volume 28, Number 2, April 1987 tions of animal hosts acting as reservoirs for some dis­ oped among many primate species. The social delay in eases. Cattle carrying tuberculosis and plague-carrying male maturation has no necessary effect on the repro­ fleas living on rats eating stored grain are examples. ductive potential of a group. The time difference in phys­ Mortality patterns since the development of increased iological maturation between male and female would densities and enhanced communication may be very dif­ not have had any necessary effect either if it had oc­ ferent from those that prevailed during most of human curred through slower development of males. If it had history. This makes the use of modem life tables very occurred through accelerated development of females, it problematic outside modem contexts ILovejoy et al. would have added reproductive potential, given no 19851. change in life expectancy or timing of menopause (Ham­ But the most important effect of increased com­ mel 1976, Lancaster 1978). Is this kind of change at the municative ability has been the emergence of mass cul­ bottom of the dryopithecine expansion? An adaptation ture, a process that has been going on for longer than we in which females mature early in times of food plenty complain about and perhaps from time immemorial. and more slowly during periods of scarcity would be a Currently, starving children in Ethiopia are featured particularly convenient arrangement, as is suggested by "live" on the evening news, and rock groups organize Frisch and McArthur 119741, and the resolution of that systematic intervention into the monality process of a question by empirical research will provide important geographically and culturally distant population. Any insights into prehistory (Scott and Johnson r985). Solid increase in communicative efficiency damps the isola­ empirical studies of primate populations are still scarce tion whose very nature gave rise to the intensifying ef· but are an accepted goal of the field (Burton and Saw­ feet of multiple tries at ecological solutions. Indeed, we chuk 1981, Cutler 1976, Dolhinow 1984, Hamburg and now approach the situation in which questions about McCown 1979, Hrdy '977, Sade et al. 19851- the global human population make good sense. Disrup­ tions in one part of the earth may impinge immediately PALEODEMOGRAPHY on all of it, today, whereas only a short time ago they would never do so immediately and might only rarely do The empirical bases for answering many of the questions so in the long run. Economies of scale in the develop­ we want to ask about prehistory will necessarily come ment and distribution of innovations that overcome re­ from paleodemography, a field which is going through source barriers are now surely more important than the extensive revision (see Acsadi and Nemeskeri 1974, Boc­ independent invention of solutions in local groups. The quet-Appel and Masset 1982, Hassan r981, Howell r976, multiple tiny world systems of the earliest human be­ Petersen r975, Van Gerven and Armelagos 19831. Many ings that were ultimately eclipsed by the more ambi­ of the issues in this currently lively field are method­ tious partial world systems of early empires are now ological, scholars finding sufficient fault with methods approaching coalescence into a larger one, so that it is used in the past that the usefulness of many earlier appropriate that demographic phenomena are a topic for findings has been questioned. With a new generation of concern in the forum of the United Nations. Many fac­ scholars asking more specific questions and treating tors, from the Green Revolution to organizations like their available data bases with more rigor, a new genera­ Oxfam, decrease the importance of locality. The mecha­ tion of studies has emerged (Buikst1a and Konigsberg nisms of evolution, driven by demographic process, have 1985)_ transcended themselves as processes of competition be­ Results from empirical studies based upon the analy­ tween allopatric segments of the species, insofar as the sis of archeological and skeletal data will have to await species becomes communicatively homogeneous. While resolution of pressing methodological questions IWeiss Malthus and Boserup could have been right in the long 19731. New techniques for aging and sexing skeletons or short run for local societies, they are now more likely (Lovejoy et al. 19851. new methods of data collection and right in the long or ShOll run for all of mankind together. data reduction that facilitate an increase in com­ Marx's conclusion that the fundamental basis of modern pleteness and reliability of the empirical base (Lovejoy demographic differentiation is social and based in distri­ 19711. and new investigations of the distribution of bution of resources is brought closer to being the whole causes of deaths in isolated populations all promise to truth in just the same measure that the world is one. contribute to the value of paleodemography for the kind of population reconstruction we are considering here. Indeed, it is the increasing promise of valuable data from Research Possibilities for Theory Building paleodemography that encourages us to believe that the kind of ambitious model proposed here will eventually PRIMATE POPULATION STUDIES find empirical clarification (Mann 19681. Special topics One of the attributes of Old World primates, and thus in paleodemography, such as the effects of status, infec­ probably of our ancestors, is that females mature sexu­ tions, and diet on the skeletal population (Buikstra and ally and otherwise earlier than males. Generally, and Mielke 1985), are of particular interest, even if eventual given premodern mortality schedules, more of an aver­ changes in the methods of aging and sexing skeletal ma­ age females life-span is reproductive than of an average terials require reanalysis of some of the raw data. Conse­ male's. This difference is increased in any social system quences of cultural traits for paleopopulations have been that delays the access of physically mature males to re­ almost entirely ignored (but see Hammel, Wachter, and productive females, and some such systems have devel­ McDaniel 1983, Isaac 1978). HAMMEL AND HOWELL Population and Culture 1149

HUNTER-GATHERER POPULATIONS after the origin of cultivation as a regular basis of life, we certainly see increased size of social units, the develop­ During the past decade the field of demographic studies ment of much larger political units and much larger of hunter-gatheret societies has incteased from almost communities than previously (SpooneI 1972). Indeed, nonexistent to the point that a few hunter-gatherer soci­ this historical period produced urbanization, increases eties have been well described, a few have been well in the scale of trade, differentiation of urban and ruIal srudied and partially described, and rich matetials are social structure, development of status groups, and dif· anticipated to be available during the next decade. One ferences in access to the means of production that had of the results of these studies has been an appreciation of not been regularly seen before ISkinner I977}. It is not the complexity of the population structure of even the clear, however, whether the distribution of the size of smallest and simplest groups. Observations which have breeding populations changed very much during this pe­ emerged as central include the extent to which hunter­ riod (Netting r98rl. With food production, societies be­ gatherer peoples restrict their population growth and the came Iegularly much largeI than breeding populations, extent to which the restriction is conscious and deliber­ and populations tended to lose their geographical integ­ ate. The low fertility adaptation observed among the rity. Holding of property by kinship units and eventually !Kung (Howell r9791 seems to be common among by individuals complicated social structure and motives hunter-gatherers (Hastings 19781, but there is much to of action. The extension of contact between populations learn about its frequency and the mechanisms by which and even between societies by trade, travet and regular it is established and maintained. We need to know communication similarly complicated the "locus" of whether the mortality patterns of hunter-gatherers tend culture. to be distinctive from those of peasant populations, per­ haps because of differences in population density and exposure to infectious diseases. DEMOGRAPHY OF NATIONAL-LEVEL INDUSTRIALIZED The increasing body of studies also serves to warn us SOCIETIES ISchrire 19841 not to oversimplify the definition of The transformation of demographic regimes known as hunter-gatherer societies or to exaggerate the differences the "demographic transition" started some 250 years ago between them and food producers ITestart 19841. Many and has been proceeding apace ever since. It remains hunter-gatherer societies have tried food production to unclear whether any of the societies involved have some extent at times in the past. Many regularly trade "completed" the transition in any real sense, but it is collected or hunted products with food producers, so obvious that many societies are still changing rapidly that their occupational specialty resembles wage wotk (Teitelbaum 19751. The form of the transition has been a or cash cropping in some respects, and few if any have drastic reduction in mortality, from typical average life­ been completely isolated either from other populations spans of 30-50 yeaIs to spans of 60-80 years. In many or from non-hunter-gatherers. The niches occupied by cases this change has been well under way before fertil­ humer-gatheIers vary widely, from the Arctic to the ity starts to decline through the adoption of contracep­ tropics, in both obvious and subtle ways ICashdan t9831. tion within marriage. Typically the decline in fertility is Srudies of hunter-gatherer groups, contemporary and from some 6-8 children per couple to 2 01 fewer children historical, that include the dynamics of the group over a per couple (on the averagel· period of time as long as a generation are needed. To a The trend toward extreme differentiation within and standard demographic reconstitution we need to add between populations that has characterized population quantitative assessment of diet, energetics, decision history for over 10/000 years tended to be reversed in making, contact with outsiders, and so on. some interesting respects during the industrial period. We note generally an increase in the permeability of pop­ ulation boundaries, perhaps as a consequence of the DEMOGRAPHY OF FOOD PRODUCERS ready access of individuals to cultures other than the one The "origin" of food production by agriculture and herd­ they are born into through mass education, mass cul­ ing around 10,000 years ago is obviously a central event ture, and mass communication. In some ways, industri­ in any account of population history (Cavalli-Sforza alized populations resemble hunter-gatherer populations '983, Clark and Brandt 1984, Cohen '977, Cohen and more than peasant or agricultural populations in respect Armelagos 1984l, but we continue to urge caution about of individualism, control over self and family, and free­ overgeneralizing. The model of Ammerman and Cavalli­ dom to try cultural varieties if available. Sforza's 119841 continentwide framework for incorpora­ tion of data over centuries is a major source of our op­ THE RESEARCH AGENDA timism that the time is ripe for developing a model of responses to population pressure lsee also Sanders and To construct an operationalizable theory of the coevolu­ Price r9681. Similar attempts to look at the disrribution tion of human biology and culture through demographic of population and cultural traits continentwide over gen­ functioning, the research community may wish to orga­ erations and centuries may now be in order. nize available research results in new ways. Following With the development of food-producing societies, we Ammerman and Cavalli-Sforza (19841 in their exem­ find the phenomenon of populations growing stably over plary study of the spread of agriculture in Europe, we see substantial periods of time. During the few centuries a need to examine the results of studies flOm a range of 150 I CURRENT ANTHROPOLOGY Volume 28, Number 2, April r987

disciplines on an organizing canvas as large as a conti· DON E. DUMOND nent, with observations arranged by centuries, including Department of Anthropology, University of Oregon. not only a talget population isolated hom the rest of the Eugene, Ore. 97403, U.S.A. I x 86 world but all populations in the population of popula­ tions as they influence one another, provide reservoirs of Aspects of the presentation of Hammel and Howell variation and sources of change, and create the environ­ seem hauntingly familiar to me-in particular the pto­ ment in which they and others live. jection of human responses to perceived population pres­ We expect that databanks organizing observations sure and the notion that those responses to scarcity are from a range of scholars in a range of specialist subfields instrumental in societal evolution Ie!. Dumond 19721. will be necessary to construct and test the kind of theory Rather than pursue parallels, however, I limit this state­ we have in mind. We anticipate that simulation, both ment to two diffeting but articulating points. One iden­ microsimulation and macrosimulation, will provide a tifies a seeming paradox within the presentation; the framework within which observations can be used to other suggests an alternative conceptualization of a estimate the parametets and check the implications of change-generating mechanism that can be used in more observations, permitting both reconsideration and occa­ situations than can the notion of population pressure sional correction of the empirical base and the produc­ and that may contribute toward resolution of the tion of null hypotheses for the continent and time period paradox. used as base and predictive hypotheses for other places, I. Hammel and Howell open with a persuasive state­ other periods. We anticipate that such simulations, al­ ment that the dynamic of population change lies in indi­ ways simplifications of reality, may become dauntingly vidual breeding units rather than in populations writ complicated when the range of questions considered large and that the focus of analysis must be these smaller here is admitted to them, just as investigators continu­ units of the whole. But their ensuing relates only to such ally find that detailed srudy of particular societies be­ units that grow or otherwise experience scarcity, leading comes more complex when more is known about them. one to conclude that only those contribute to evolution Nevertheless/ the construction of such a theoretical in any sense. If this be so, what is the advantage of focus­ framework would seem to be possible and timely on the ing upon the smaller units rather than the larger/ since basis of the state of the field at the present time. on the whole and through time the latter have grown­ thus showing the only characteristic that is important? It is only if the contribution to evolution of the smaller units cannot be predicted from that of the larger that the smaller need be studied. The argument must be, rather, that contributions to evolution are also made by popula­ Comments tions that undergo fission or otherwise change without increase in density and before stimuli other than per­ ceived scarcity. ELIANE S. AZEVEDO 2. The confrontation of scarcity by means of altered Laboratorio de Gemltica Medica, Hospital Prof. Edgard behavior is a subclass of phenomena that may be termed Santos, Universidade Federal da Bahia. Salvador, the exploitation of differenti.l opportunity. While some Bahia. Brazil. 23 x 86 such opportunities may be mitigating paths taken in the face of a deteriorating present situation (scarcityl, others The integration of biological and cultural evolution is a may be seized in circumstances in which there is no challenge to the human intellect limited by the curtent deterioration of existing conditions and no perception of stage of development of science. However, one has to scarcity. begin to approach this problem, since the isolated under­ For example, some fissioning, migratory move­ standing of cultural and biological evolutions is unsatis­ ments-the drive of ancestral native peoples across factory. The selection of demography as capable of unify­ northern Canada around 2000 B.C. and again around A.D. ing the subdisciplines of anthropology and untangling IOOO, the expansion of Euro·Americans within North the multiple knots of coevolution seems to result more America in the 19th century-can be seen as generated from the prevailing influences of great thinkers IMalthus not by scarcity but by the promise of new and richer [976, Marx 19671 than hom its inherent power. Mote­ resources to be gained with less effort in the out beyond. over, while the high technical quality of the data from But those movements occurred only as demographics demographic studies today assures the fidelity of the in­ supplied population sufficient for geographic expansion formation, it neither contributes to a better understand­ (without increase in densityI with an accustomed pro­ ing of the past not guarantees that demography is the ductive system, thus permitting unsettled areas to be right vehicle for us to ride on the roads of evolution. treated as a familiar, if drastically enlarged, ecological Only the results of the possible research for theory niche. building will settle the matter. Looking ahead, I would It is also likely that recent examples of subsistence in­ venture the opinion that the data compiled will add to tensification and expansion are related as often to normal science IKuhn 1978) but will be far from revolu­ money and markets as to scarcity at home. The in­ tionary with regard to our knowledge of evolution. tensification of rice culture in the region of Thailand HAMMEL AND HOWELL Population and Culture 1151

treated by Hanks (1972) came about as a response not to ical framework, requires that we explicitly identify vari­ internal pressure but to the presence of a rice market in ables the optimization of which can account for the Bangkok. And yet it occurred only through redistribu­ differential reproductive success of one or another tion of the population and (evolutionary) change in or­ biological and/or behavioral characteristic. ganization, and the whole could be described by Hanks With respect to the second question, is change in as a Boserup-like linkage of population and agricultural constraint specification a random (historically unique) regime. This is "scarcity" in an economic sense but not process, or is it nonrandom (perhaps even directional), in a personal, perceptual one. determined by prior demographic, economic, or environ­ In these cases the demographic conditions may be mental states? To phrase the issue in these terms means more permissive than coercive, but they are crucial even that, in the theory to which Hammel and Howell direct without the driving mechanism of increased population our attention, and as is implicit in much of their discus­ density. With this perspective the end sought by Ham­ sion, demographic phenomena are unlikely to be the mel and Howell should still be served. central mechanism of change. Almost certainly, that "mechanismll will be ecological interdependencies. These interdependencies will include but will not con­ w. PENN HANDWERKER sist solely of demographic phenomena, and they will Program in Anthropology, Humboldt State University, reflect not merely relationships among environmental, Arcata, Calif. 9552r, U.S.A. 14 x 86 technological, and demographic parameters but also power differentials within or between populations (see, Hammel and Howell propose extending evolutionary e.g., Handwerker '987, n.d.j. theory in ways such that "demographic events are the But what of innovation and individually unique de­ central mechanism and leading indicators of the coevo­ cision making, which contribute to the creation of lution of bodies, minds, and societies." I applaud and phenomena that have never before existed? To what ex­ agree both with the thrust of their argument and with tent (if at all) do such phenomena determine macro­ nearly all of the specifics. I wish to emphasize a point level changes? Can we develop explicit, formal models that remains largely implicit in their presentation: the that identify macro-level determinants of micro-level behavioral and biological change for which such a theory changes the effect of which is to alter macro-level con­ must account entails change in system states (see Hand­ straints? werker I986b:14-171. Finally, of course, evaluating these and other possibili­ To date, anthropologists (like sociologists and econo­ ties requires that we clarify interdependencies between mists) have largely been content to analyze particular biological and cultural phenomena and between biolog­ system states, and perhaps to contrast different system ical and cultural evolution. What, specifically, are the states (e.g., traditional and modern societies or bands, linkages between our genome and cultural behavior, and tribes, chiefdoms, and states). We have focused on mi­ how do they operate? How, specifically, does selection cro-level investigations and have produced explanations operate, at what organizational levels, and with what that take the following form: (r) If people make assump­ effects? Answering these questions implies a major shift tions All A2I .. . An (21 within the set of constraints ell in research strategy. Not only, as Hammel and Howell C" Co. (3) they conform to behavioral pattern B. suggest, might we organize research or information dif­ Quantum improvements in our understanding of demo­ ferently, but we need to develop explicit models of the graphic phenomena have not been forthcoming because manner in which phenomena at different organizational micro-level a:nalyses, whether relying on formal optimi­ levels articulate over time. zation models or informal descriptions of cultural con­ text and decision making, cannot address the most im­ portant issues: (1) on what assumption or assumptions HENRY HARPENDING (consciously or unconsciously) do people generate be­ Department of Anthropology, Pennsylvania State havior, and why? and (lj how and why do changes in University, University Park, Pa. 16802, U.S.A. 16 x 86 constraint specifications come about? Because changes in system states come about only when there are This article is an excellent synthesis and statement of changes in assumptions, constraint specifications, or prospects for anthropological demography. I do have both, the central task of the theory building to which some reservations, but the following criticisms should Hammel and Howell direct our attention must be to not be taken to deny my general admiration for and answer these two questions. agreement with it. With respect to the first question, the most common First, Hammel and Howell state early on that specifi· model assumption is that people are optimizing crea­ cations of evolutionary theory for humans have been tures. But what do we optimize, and why? Is cultural/ "deterministic and teleological./I But the incorporation behavioral diversity explained by the optimization of a of stochastic models into evolutionary theory is an old wide range of variables or by the optimization of one or a accomplishment and widely acknowledged by evolu­ small range of variables in the context of constraints tionists as necessary to explain and understand gene dis­ that are unique to the life history of the individual or tributions. Evolutionists may write deterministic equa­ population? To construct a theory, rather than an analyt- tions, but they all understand how stochastic processes L

1521 CURRENT ANTHROPOLOGY Volume 28, Number 2, April 1987 modulate their results. Evolutionary theory, as I under­ nomic! and environmental variables. The crucial con­ stand it, is haldly evel teleological, although I would ceptual problem therefore is how to combine the study algue that it should be when applied to an olganism with of demographic, technological, economic, and environ­ fOIesight and intelligence. mental variables into a coherent research strategy. The Second, their use of Itpopulation pressure" is trou· authors do not concern themselves in any general sys­ bling, since their discussion seems to indicate that they tematic sense with this problem, although they do dis­ would include anything that induces cultUIe change. cuss some aspects of the interrelationship of demo­ But, if population p,eSSUIe is effectively synonymous graphic processes with technological, economic, and with culture change, we are left with words chasing each environmental factors as well as some demographic in­ other in circular definitions. They might write a hy­ terrelationships with political and symbolic-ideological pothetical dynamic equation and identify the term or systems. The most glaring symptom of the authors! par­ terms that encapsulate what they mean by pressure. adigmatic ambiguity is their failure to deal with the con­ Third, the meaningful definition of a population cept of mode of production lin either the Marxist or the seems to me to be a hopeless task, except in the case of cultural materialist sensei and with the various efforts isolated islands. The 80% rule is albitrary, and if we pick that have been made to demonstrate the causal priority an individual and draw an 80% circle around him, then of modes of production and modes of reproduction (the we are faced with the uncomfortable arrangement that demo-techno-econo-environmental infrastructure) in re­ everyone belongs to a different population. An empirical lation to the structural and superstructural elements consequence of this difficulty is that anthropological de­ of sociocultural systems (Harris '979, Ross and Harris mographers sometimes can be more confident of the 19871. Another expression of the same theoretical short­ numerators of their calculated rates than of the de­ coming is their choice of breeding populations as the nominators. basic aggregate of human actors upon which and through Finally, talking about populations doing things leads which the processes of biological and cultural evolution subtly to thinking that there are regularities about the operate. While certain demographic and genetic prob­ behavior of populations. There are at least two problems lems are best approached through breeding population with this. In sciences with laws reflecting large numbers isolates, the integration of diverse anthropological inter­ of interactions at a lower level such as Boyle's law of ests cannot be achieved on such a basis. Sociocultural gases, the particles are simple equivalent replicates of systems have seldom coincided with breeding popula­ each other or nearly so. Humans are all different, they tions. Throughout most of prehistory! human breeding compete with each other, and they evolve. OUf grounds populations extended beyond the boundaries of sociocul­ for faith that such particles will behave in replicable tural systems, while with the development of the state, ways in the aggregate are scant. Further, the implicit sociocultural systems came to embrace several different assumption in talk about population adaptation is that breeding populations. If the goal of nomothetic an­ food is the prime scarce resource driving cultural evolu­ thropology is to account for the divergent and conver­ tion. In fact there is debate in anthropology and in evolu­ gent evolution of sociocultural systems, it is the sociocul­ tionary biology in general about the relative salience of tural system-people plus infrastructure, structure! and food and sex as driving concerns. Insofar as Marx was superstructure-which must be the basic bounded ag­ talking about consequences of male competition, he is gregate and not breeding populations! whose basic demo­ given short shrift in favor of Malthus and Boserup. graphic rates cannot be understood apart from the entire infrastructural conjunction and the latter's feedbacks with structural and superstructural factors. Incidentally, MARVIN HARRIS there is no evidence that the bounding of empirical Department of Anthropology, University of Florida, sociocultural systems presents any greater methodologi­ 1350 GPA, Gainesville, Fla. 3261 I, U.S.A. 2t x 86 cal difficulties than the bounding of human breeding populations, especially if Adams and Kasakoff's 80% The authors argue for the importance of the study of rate of inbreeding is to be the operational definition. demography for anthropological theory. In my view, de­ Turning to specific population·related issues, the au­ mographics should be treated in introductory textbooks thors provide some useful summaries and research agen­ on a par with more traditional subjects such as economy, das. It is regrettable, however, that their overview of domestic organization, or religion IHarris 1987:89-1051. anthropological demography does not reflect a growing Despite the shared enthusiasm for anthropological de­ realization that basic demographic rates Ifertility, sex mography, I cannot express much enthusiasm for the ratios, mortality, and population growthI and their ideo­ authors' proposal to make demographic studies the basis logical accompaniments throughout prehistory and into for integrating theories of biological and cultural evo­ modern times can best be understood in terms of the lution. No one would disagree that "demographic variable costs and benefits of rearing children in regimes" Isicl are "important causes and effects" in bio· sociocultural systems characterized by specific infra­ logical evolutionary processes. But as the authors, im­ structural and structural conditions (Ross and Harris plicitly at least, acknowledge, patterns of demographic 19871· This formulation largely supersedes much of the growth are caused by and affect other basic sociocultural authors' discussion of the role of scarcity as an explana­ processes involVing, for starters, technological, eco­ tion of demographic rates, prOVides currencies by which HAMMEL AND HOWELL Population and Culture 1153

cultural selection can be distinguished from natural to genetic replication without mutation. However, even selection, and thereby links demographic theory with the most stable cultural traditions represent outcomes theories that account for the divergent and convergent of dynamic, context-dependent layers of cultural cre­ uajectories of sociocultural evolution IBirdsall 1983; ation and transmission, unlike the programmatic repli­ Caine 1977, 1983; Caldwell 1982, 1983; Coontz 1957; cation of genes. Handwelker 1986a; Harris 1984; McKee 1984; Mamdani A second instance further illustrates confusion from 1973; Millel 1981, 1983; Nag and Kak 1984; Nag, White, failing to recognize the metaphoric aspect of cultural and Peet 1978; Scheper-Hughes 1984; Scrimshaw 1983; evolution when placed in a biological evolutionary Skinner 1985; Sussman 1972; Tilakaratne 1978; Vining framework. The authors predict more rapid culture 1985; White 1973, 1982; Wood 1982). The failure of the change in populations that are smaller and newer to authors to consider this growing corpus of facts and the­ their environments, a characterization seemingly draw­ ories suggests that we still have a long way to go before ing on Wright's work le.g., 19781 on random genetic drift demographic anthropology gets its own act together and and selection. While many researchers note the place of can present itself as a model for integrating our fractious culture change, we wonder whether anyone's sense of it profession. is clear enough to substantiate cross-cultural compari­ sons. Existing measures of culture change are few, none as precise as measures of change in gene frequency. Cer­ ANN v. MILLARD AND KIM A. MCBRIDE tainly a phenomenon of nuclear winter, produced by Department of Anthropology, Michigan State state-level decisions, would bring about the most rapid University, East Lansing, Mich. 48824, U.S.A. 21 x 86 culture change that the world has seen, if not extinction li.e., biological evolutionary change). Hammel and Howell's papel is a valiant effoIt to deal The general approach advocated by Hammel and How­ with the biological and cultural aspects of demography, ell, the use of demographic analysis in examining the "to place the development of human demoglaphy and coevolution of biology and culture} certainly merits fur­ coevolved culture squalely within the framework of ther attention. The task should be approached carefully, biological evolutionary theory." Our leaction is to hope since, as the instances above illustrate, theoretical that this article marks a recovery from the polarization borrowing from biology can occur in disregard of impor­ stimulated by Wilson's Sociobiology 119751. a breach tant differences between organic evolution and culture that has stifled discussion by members of junior cohorts change. It is an inescapable thought that the metaphor of who were not loyal to the sociobiologists' camp. Ap­ "cultural evolution" has singular power in our times plause is due Hammel and Howell also for setting the because of the ascendancy of science in the latter half of record straight in regard to the population as the biologi­ the 20th century. As others have noted, biological evolu­ cally evolving entity but not the decision-making unit. tion is a widespread commonsense model in social sci­ Finally, their treatment of the concept of "resource ence le.g., Giddens 19841. Approaches by population ge­ boundary" is instructive and should prove useful in de­ neticists that show some awareness of the resulting mographic anthropology. problems range from Fisher 119301 and Haldane 119321 to From the perspective of a biological anthropologist, more recent work including Williams and Williams the major difficulty in Hammel and Howell's proposal Ir974), Kaplan 119761. Reid 11976), and Cavalli-Sforza and lies in their approach to culture change. Their formula­ his coworkers ICavalli-Sforza and Feldman 1981). fn de­ tion does nOt clearly recognize that cultural evolution mography, the work of Howelllr9791 herself is a sterling becomes a metaphor when placed "within the frame­ analysis of population dynamics in small groups and can work of biological evolutionary theory." For example, serve as a model for anthropological approaches. The the concept of Ifcopying errors," while appropriate for issues presented by Hammel and Howell in population characterizing mutation, misconstrues culture change. dynamics and culture change are intriguing and remain Humans seem to be incapable of the exact cultural repli­ formidable challenges worthy of more attention at this cation implied by "copying,II and cultural variation does early stage of development of demographic anthropol­ not inherently signify "errors. 1I In fact, variation is as ogy. fundamental to culture as "copyingll is to genetics. Sup­ POrt for this statement is widely available, ranging from studies of language acquisition-the first sentence ut­ FRANCISCO M. SALZANO tered is not a copy of someone else's sentence-to eth­ Departamento de Gemltica, lnstituto de Biociencias, nographic observations. For example, in central Mexico, Universidade Federal do Rio Grande do Sul, Caixa one of the authors of this comment IMillardl observed Postal r953, 9000r Porto Alegre, RS, Brazil. I IX 86 that each woman's tortillas were regarded as distinctive, and women also noted that their tortillas varied from In principle I am basically in agreement with the ideas day to day, but none were seen as incorrect. Thus, char­ expressed by Hammel and Howell; my comments, acterization as "copying errors" misconstrues cultural therefore, are going to be brief, their main purpose being variation in fundamental ways. The other side of the to call attention to a large number of papers that are simile of cultural variation as "copying errors" is the being published in this area on South American Indian authors' concept of cultural stability, seen as analogous populations. 1541 CURRENT ANTHROPOLOGY Volume 28, Number 2, April r987

The first point I want to make is that l as is mentioned enumerate and elucidate the many assumptions about by Ihe aUlhols, the hunter-gatherers/food producers "local breeding population" used so often in studies over dichotomy is an oversimplification. At least in my can· time and space. Their paper complements a proposal by tinent, hunting and gathering is always supplemented by Paul Baker to recognize human population biology as a some form of agriculture, and this should be kept in viable transdiscipline; he observes that lithe wedding of mind in any extrapolation from extant to prehistoric demographic and traditional population genetic thea· populations. Secondly, in my opinion the category "food ries offers new perspectives on evolutionary processes" producers" is too inclusive. The population structure of (Baker 1982:2121. Both papers synthesize methods, de· pastoralists probably differs in important ways from that emphasize disciplinary boundaries, and suspect attrac­ of agriculturalists. The type of land exploitation is basi· tive theories which may have Ilexceeded the extant data cally different in the two cases, and pastoralists may he base for testing" (Baker 1982:212j. Both emphasize the more mobile, which would place them in an inter­ study of local populations for comparative purposes and mediate category between groups engaged mostly in therefore recall the empirical value of inductive case hunting-gathering or in agriculture. Finally, another studies moving slowly towards theory Ithough some­ point of similarity between industrialized and hunter­ times not fast enough for the theoristsI. gatherer populations is their fertility patterns. Hunter­ The reader may get the impression that population gatherers controlled their family size in a much more fission is the usual alternative to adaptive pressure. It is rigorous way than agriculturalists, since numerous chil­ listed first in the "Responses to Population Pressurell dren would he a source of embarrassment in wanderings and then discussed under its own heading, flThe Fission in the jungle bur a valuable help in agriculture. Alternative." As the ideological artifacts of the concept Population structure studies among South American of Iflinear progress fl in Euro-American culturef growth indians have been manifold, and a review of them is be­ and development dominate population thinking in an­ ing presented elsewhere (Salzano and Callegari-jacques thropology. Expanding populations are easier to study in 1986). Relatively extensive information is available the field or in the archaeological record because, epis· about their basic demographic parameters (age and sex temologically, we know presence better than absence. distribution, marriage patterns, mobility, fertility, and But Hammel and Howell balance this notionf because mortalityi, and this empirical information has been they /fda not assert a unidirectional system." They ap­ compared with the results expected under various mod­ preciate that population decline and extinction are pos­ els. Attempts at improving these demographic data have sible, if not likelYf events of some evolutionary conse­ also been made, using such procedures as urine tests for quence (c!. Gallaher and Padfield 19801. Their research pregnancy and physical examinations to check the reli­ agenda for core·periphery development should add the ability of reproductive histories or the application of se· historical study of marginal communities on the edge of rial censuses INeel and Weiss 1975; Black et al. 1978; early industrial centers; there are instructive examples Gage, Dyke, and Riviere r984a,bl. Questions abour the from Europe, especially in the United Kingdom (Cohen genetic effects of fission in general or the differences 1982f Prattis 1979f Stevenson 1984)- If we are to under­ between random and lineal linvolving genetically re­ stand the dynamic present in the sense of Hammel and lated individualsl fission have been considered by Howellf we might strive to measure the carrying capaci­ Thompson 119791 and Smouse, Vitzthum, and Neel ties of nation-states (Maserang 19761 and evaluate the 119811. Computer simulations have also been used to demography of local habits and history within the estimate the probability of survival of a neutral muta­ emerging world community (Sheets and Kelly 19871. tion (Li, Neel, and Rothman 19781 or the inbreeding coefficients for alleles present in the founding popula· tion of American Indians ISpielman, Neel, and Li r9771. Thus multidisciplinary studies of the type proposed by Hammel and Howell are already being conducted among Reply Amerindians.

E. A. HAMMEL AND NANCY HOWELL JOHN W. SHEETS Berkeley, Calif., U.S.A. 18 XI 86 Department of History and Anthropology/Museum, Central Missouri State University, Wallensburg, Mo. We are grateful to the commentators for their responsesf 64093-5060, U.S.A. 6 x 86 which help us identify the areas where we were unclear or excessively terse and those where we simply disagree. When viewed as a paradigm to be tested by much more In our response we expand on some of these. We point

research, the paper by Hammel and Howell successfully outf as in the original, that our presentation is far from challenges us with "a philosophically acceptable exten­ being a theory bur is only a beginning. We address the sion of evolutionary theory, one in which demographic comments of each reviewer in turn and conclude with events are the central mechanism/' and "a framework an attempt to clarify our perception of evolution. that will work equally well for protohominid bands and Azevedo: We hold out no hope of revolution in an­ the 'world system' of modern market capitalism." They thropology. Some gradual reform would be enough for HAMMEL AND HOWELL Population and Culture I ISS us. Azevedo is betting that our approach will not be wry observation that the complexity of the theory neces­ productive but does not convince us we should not try. sary to approach these problems is a function of the com· Dumond: We acknowledged that our arguments were plexity of the social organization of the society in the synthetic and did our best to absorb and cite the relevant context of which it is applied, and the amount and qual­ literature. Hence we are gratified that Dumond finds our ity of the data necessary to test the theory are currently argument Ifhauntingly familiar" and trust that the same also a direct function of that complexity. This means could be said by the authors of many other papers not that we have the worst data for the simplest theory. cited. The paper that he cites is close to our account in Because we were thinking about a theory that could re­ spirit but not in substance. late demographic to evolutionary approaches, we quite The query is why we should study small units. Our clearly swung toward the hunter-gatherer pole, and we distinction is between populations as breeding popula­ wonder, in spite of our enthusiasm, whether we can ever tions and the global human population, not between find a set of data sufficiently good to test a theory that is large and small populations as such. Readers will recog­ sufficiently simple to allow agreement about its princi­ nize this as the same distinction suggested years ago by ples. We just hope so and are prepared to try. Steward (19531 between multilinear and unilinear or Harpending: We agree that statements about the evo­ universal evolution and an insistence on the history of lution of H. sapiens and of culture, especially the latter, actors rather than the history of categories. We also state take scant advantage of modem concepts of evolution. explicitly that population pressure is not the only root of This is our complaint (although politely disguisedj. It is changej it is simply one to which we wish to draw atten­ not modem evolutionary theory we criticize but most tion. anthropologists' and especially social-cultural anthro­ In respect of Dumond's second point, we would agree pologists' use of it in arguing either for or against evolu­ that perception of opportunity is a potential cause of tionary approaches Isee also our response to Millard and culture change. Indeed, under our broad definition of McBrideI. "scarcity" it is exactly the perception of insufficiency We define pressure as an unfavorable change in the that is the driving force. This point is not new in popula­ ratio of perceived supply to perceived demand. This is tion theory, not to mention social theory, and was ex­ not the same as culture change, although an exogenous plicitly recognized by Malthus in the later editions of his shift upward in perception of demand/ for example/ Essay, as we note. If that perception is generated by a redefinition of culturally acceptable consumption, cultural propensity to maximize, the search for opportu­ would generate pressure in our sense by shifting the de­ nity can become pervasive and satisfaction can be only mand/supply ratio. ephemeral, as we again note. We would not agree that Of course the 80% endogamy boundary is arbitrary, the expansion of Europeans to and within North Amer­ and we say so. One really needs density equations for the ica was unrelated to scarcity, at least scarcity in our continuous case or good historical or ethnographic evi­ terms. It was resource insufficiency that sent many dence of discontinuities to permit empirical definition Europeans to North America in the first place­ of boundaries. The situation is no different from trying sometimes insufficiency at a minimal subsistence level to detect the boundary between species as speciation is Ithe Irish!. sometimes insufficiency with respect to de­ occurring in time and space. [n general the approach to sired wealth (some English, some Dutch, some Ger­ measurement of endogamy and thus of social boundaries mansL and sometimes a political insufficiency in respect is no different from the measurement of resources or of established religion (various Protestant sects, al­ interactions of any kind and requires a probabilistic ap­ though the action of economic forces in these move­ proach, typified by the demographers' notion of rates ments ought not to be denied too soon). related to a population at risk of an event. The technical We also question whether the perception of insuffi­ complexity of measuring endogamy is illustrated by ciency may nOt have been a factor in the expansion of Romney 119711. Amerindians across northern Canada or indeed into the We disagree with Harpending on the implications of New World. The choice for a hunter-gatherer group is his example from physical science and with Millard and whether local resources are good enough or whether the McBride on the making of children and of tortillas. If we hunting is better or easier further on. Even if the desire is could observe molecules of a gas with sufficient resolu­ to generate the same amount of food for less work and tion, we might discover that they were not perfectly more leisure/ so that what is scarce locally is leisure, not identical either but just sufficiently identical for some food/ increases in human population density or de­ purposes. We know that the behaviors of individual creases in resource density can inspire migration. In gen­ molecules are nOt identical and cannot be predicted eas­ eral, we cannot agree that "economic scarcity" is not ily-that is exactly the rationale for statistical mechan­ also personal and perceptualj if anything/ we insist in ics. We cannot predict with much certainty where an our presentation that definitions of scarcity be culturally individual molecule will be in a cloud of gas or which based and not externally imposed. ones Maxwell's demon will let in or out, but we can talk Handwerker: We recognize the cogency of Handwer­ about the average of these events and positions. Simi­ ker's commentS on changes in system states and agree larly/ we can often predict with impressive accuracy the that we did nOt carry the theoretical formulation to a number of persons who will marry or be born or die at formal recognition of those relationships. We make the some imminent date, even though we could predict who 1;61 CURRENT ANTHROPOLOGY Volume 28, Number 2, April r987 they would be only with the greatest of difficulty. Fur­ the moment. Insisting on the complexity of cultural ther, although we could predict the average production phenomena seems unnecessary, and assuming the sim­ of children and tortillas, we would no more expect the plicity of biological phenomena may be a mistake. individual producers of children ro be identical in their We dispute Millard and McBride's claim that the first reproductive behavior than we would expect them to be sentence uttered by a child is not a copy of someone identical in their productive behavior. JUSt as one tortilla else's sentence, in that the first sentence uttered by a is not exactly like the next, so also one child is not child is in important senses exactly a copy of someone exactly like the next, and culture is not different from else's sentence, else it would not be intelligible. It is biology in this respect. precisely the achievement of grammaticality as cultur­ Finally, we did not restrict our definition of scarcity to ally defined that permits the first sentence to be recog­ food but explicitly included what was marriageable as an nized as a sentence. We are also beginning to discover example of a resource. that there are biological underpinnings to all language Harris: In general we could not agree more about the and thus to all sentences/ consequently to all first sen­ importance of the entire range of linkages from natural­ tences. It is just a mistake to think of identity in terms resource infrastructure to the moral economy and that are too simple and direct lsee also our comments in thought we had said that. We are simply suggesting a response to Harpending, abovel· demographic focus as a useful avenue of approach. Un­ Similarly/ we take exception to the absolute distinc­ like many other phenomena of anthropological concern} tion between stable and variant. We made no such dis· problems in the area of demography often have definable tinction/ nor do we think our discussion implies the and measurable outcome variables. necessity of it. Even if at some observational level some A social and cultural theory about demographic be­ elements of a complex whole Iculture or genomej can be havior faces the same necessity to incorporate a wide said to have been perfectly copied, it is unlikely that all and complex range of causal forces as any theory about elements of any complex whole are ever perfectly human behavior. The integration of anthropology that copied. l"Unlikely" does not mean impossible but only we envision is an integration not of disputing theoretical less likely than most other outcomes.1 The proportion of factions (for example} Marxists} materialists} symbolists} imperfect to perfect copies, thus the error rate in copying and the rest of the ragbagl, as Harris describes it, but of in our terms, is a measure of variation and thus of one areas of substantive concern, namely, ethnology, ar­ factor in the potential for evolutionary change. chaeology, biology, as Kroeber might have described it. Regarding the relationship between group size and Again we acknowledge that this intent swung us not so speed of change, we point to both the accelerating and subtly toward the hunter-gatherer pole and the simplest the decelerating effects of small and of large size. We kinds of theory, and the poorest bodies of data, even if speculate that small sizes intensify random drift effects these bodies of data are improving. Notice land see also and do so with feedback from stickiness in the marriage Handwerker's and Harpending's comments and our re­ market and that the absence of multiple and well­ sponse) that scarcity in our terms must be defined by the developed interest groups might impede innovation less. moral economy. We include here cultural perception of We also draw attention contrariwise to the importance the cOSts and value of children, of intergenerational of higher population densities in multiplying the num­ transfers and especially old-age security and accession to ber of potential innovators within a single society and in resources, and other systems of evaluation of COStS and increasing the speed of transmission of innovations. Mil­ benefits as approached in the fertility theories of Becker, lard and McBride's introduction of nuclear winter into Easterlin, Caldwell, Handwerker, and others. this discussion confounds the problem of the production Clearly we disagree with Harris about the best units, and transmission of variation with that of the conse­ concepts, and research strategies, and we look forward to quences of adoption of some variants. We hold no brief his forthcoming book with Ross to see how he spells out for runaway political arrogance as a useful survival tool. his preferred alternatives. We could not/ on the other Salzano: We are grateful for the additional informa­ hand, agree more with his conclusion that we still have a tion on South American populations, some of which had long way to go, but a journey of a thousand miles ... escaped our notice. Millard and McBride: Millard and McBride encumber Sheets: Again we are grateful for references of which the notion of error with excessive semantic freight and we were unaware. In respect of expanding the scope of approach the analogy between cultural and biological our concern/ it is particularly important to note that we evolution at a level which is too substantive and can only study those social or cultural unitS which have insufficiently formal and processual. survived population processes to some point in time. An "error" in our sense is defined formally as a failure This means that any synchronic or cross-sectional study to achieve perfect replication. We do not claim that vari­ must deal with a censored data setj the nonsurvivors are ation is not important in cultural evolution/ nor did we missing/ and any comparisons are defective. It is only seek to judge whether it was more important in cultural with longitudinal data that we can begin to understand than in biological evolution. Our claim is only that vari· the consequences of different responses to the same or ation is the source of raw materal for evolution and that similar events. Hence our theory building is explicitly failure to replicate is the mechanism that produces the longitudinal and geographically based, rather than cross­ variation. That lormal equivalence is enough lor us at sectional and typological, although we acknowledge that HAMMEL AND HOWELL Population and Culture' 157 some excellent work has been done in cross-sectional BENNETT, J. w. 1976. The ecological transition: Cul­ and typological (or isolatedl case studies. tural anthropology and human adaptation. New Evolution: In conclusion/ we should comment on York: Wiley. some perceptions of evolution as applied to culture B END IS T, J. 1973. IfGenetics of isolate populations/II in change that we are trying to avoid. The difficulty with Methods and theories of anthropological genetics. the model of evolutionary theory applied to cultural phe­ Edited by M. Crawford and P. Workman. Albuquer­ nomena is that it is mostly wrong. It is usually Spence­ que: University of New Mexico Press. rian and not even Darwinian but pre-Darwinian. We BIRDSALL/ NANCY. 1983. Fertility and economic might in fact whimsically regard it as a vestigial sur­ change in eighteenth- and nineteenth-century vival. Where the concept of evolution is at least Darwin~ Europe: A comment. Population and Development ian/ it is mechanistic/ nonstochastic/ monocausal, and Review 9:II 1-23.IMHI insufficiently linked to mechanisms of variation. In­ BLACK, F. L., F. D. PINHEIRO, O. OLIVA/ W. J, deed, Darwin did not have the information necessary to HIERHOLZER/ R. V. LEE/ ,. E. BRILLER/ AND v. A. speak to the origins of variation/ only to its selection and RICHARDS. 1978. Birth and survival patterns in numer­ the origin of species. Even now/ the mechanisms of the ically unstable proto-agricultural societies in the origin of biological variation grow daily more complex as Brazilian Amazon. Medical Anthropology 2:95-127. knowledge increases/ and the issues of endogeneity [FMSj raised crudely by Lamarck continue to be difficult, espe­ BLOCK, A. '977. Murphy's law and other reasons why cially at the molecular level. We distinguish between things go wrong. New York: Price Stern. endogeneity and exogeneity but do not insist on one or BOCQUET-APPEL/ J,/ AND C. MASSET. 1982. Farewell the other as fundamental principles. to paleodemography. Tournal of Human Evolution We make no case for directionality but only for the 11:3 21 -33. identification of mechanisms that can be tested for their BOSERUP, E. 1965. Conditions of agricultural growth. effects. In particular we draw no explicit analogies be­ Chicago: Aldine. tween the substance of biology and the substance of so­ BUIKSTRA/ J. E., AND L. W. KONIGSBERG. 1985. cial science but only point to formal mechanisms that Paleodemography: Critiques and controversies. should be operative in any system of interacting forces, American Anthropologist 87:3,6-33. whatever their empirical realization (provided that they BUIKSTRA, J. E., AND ,. H. MIELKE. 1985. "Demog­ are indeed systems). Neither do we insist on closed lists raphy, diet, and health/" in The analysis of prehis­ of operative mechanisms or the relative strength of their toric diets. Edited by R. I. Gilbert, )r., and J. H. effects. We call attention to the complexity of the ecol­ Mielke. New York: Academic Press. ogy of population dynamics, including therein a spec­ BURTON, F. D., AND L. A. SAWCHUK. 1981. Demog­ trum of constraints and accelerators from the raw natu­ raphy of Macaca sylvanus of Gibraltar. Primates ral resource base through cultural perception of the 15:27 I -78. environment, the political structure/ and the moral CAINE/ MEAD. 1977. The economic activities of chil­ economy, but we do not insist on primacies/ only on the dren in a village in Bangladesh. Population and De­ necessity of paying heed. velopment Review 3:20r-27. [MHI ---. t983. Fertility as an adjustment to risk. Popula­ tion and Development Review 9:688-702. IMHI CALDWELL, JOHN. 1982. Theory of fertility decline. New York: Academic Press. [MHI References Cited ---. 1983. 'IDirect costs and benefits of children/' in Determinants of fertility in developing countries, vol. ACSADI, G., AND J. NEMESKERI. 1974. History of hu­ I, Supply and demand for children. Edited by R. man life span and mortality. Budapest: Academiai Bulatao and R. Lee, pp. 458-93. New York: Academic Kiad6. Press. [MHI ADAMS/ T. W., AND A. B. KASAKOFF. 1975. UFactors CASHDAN, E. 1983. Territoriality among human for­ underlying endogamous group size// in Population agers: Ecological models and an application to four and social organization. Edited by Moni Nag. The Bushman groups. CURRENT ANTHROPOLOGY Hague: Mouton. 24:47-66 . AMMERMAN, A. 1./ AND L. 1. CAVALLI-SFORZA. 1984. CAVALLI-SFORZA/ L. 1. 1983. "The transition to ag­ The Neolithic transition and the genetics of popula­ riculture and some of its consequences/' in How hu­ tions in Europe. Princeton: Princeton University mans adapt. Edited by D. Ortner. Washington, D.C.: Press. Smithsonian Institution. BAKER. P. T. 1982. Human population biology: A viable CAVALLI-SFORZA, 1. L./ AND M. W. FELDMAN. 1981. transdisciplinary science. Human Biology 54:203-20. Cultural transmission and evolution: A quantitative [Jwsj approach. Princeton: Princeton University Press. BARTH, F. 1969. Ethnic groups and boundaries: The so­ [AVM, KAMI cial organization of cultwe differences. Boston: CHAGNON/ N. A. 1980. "Mate competition favoring Little, Brown. close kin and village fissioning among the Yano· 1581 CURRENT ANTHROPOLOGY Volume 28, Number 2, April r987

marna Indians/' in Evolutionary biology and human ---.1976. liThe matrilateral implications of struc­ social behavior: An anthropological perspective. tural cross cousin marriage/" Demographic an­ Edited by N. Chagnon and W. lIons. North Scituate, thropology: Quantitative approaches. Edited by E. Mass.: Duxbury Press. Zubrow_ Albuquerque: University of New Mexico CLARK, J. 0./ AND S. BRANDT. Editors. ]984. From Press. hunters to farmers: Causes and consequences of food HAMMEL, E. A., K. WACHTER, AND C. MCDANIEL. production in Africa. Berkeley: University of Califor­ 1983. "Vice in the Villafranchian," in Genealogical nia Press. demography. Edited by B. Dyke and W. Morrill. New COHEN, A. P. Editor. 1982. Belonging: Identity and so­ York: Academic Press. cial organization in British rural cultures. Manches­ HANDWERKER/ W. PENN. 1986a. ilIntroduction,1I in ter: University of Manchester Press. Culture and reproduction: Reconstructing the demo­ COHEN, M. N. 1977. The food crisis in prehistory: graphic paradigm. Edited by W. Penn Handwerker. Overpopulation and the origins of agriculture. ew Boulder: Westview Press. IMHI Haven: Yale University Press. ---. 1986b. "Culture and reproduction: Exploring COHEN, M. N., AND G. }. ARMELAGOS. 1984. micro/macro linkages," in Culture and reproduction: PaIeopathology at the origins of agriculture. Orlando: An anthropological critique of demographic transi­ Academic Press. tion theory. Edited by W_ Penn Handwerker, pp. 1­ COONTZ, s. 1957. Population theories and the eco­ 27. Boulder: Westview Pless.lwPH] nomic interpretation. London: Routledge and Kegan ---.1987. Fiscal corruption and the mOlal economy Paul. [MHj of resource acquisition. Studies in Economic An­ CUTLER, R. G. 1976. Evolution of longevity in pri­ thropOlogy 9. In press. !WPHj mates.Journal of Human Evolution 5: I 69-202. ---. n.d. To chart my own course: Fertility transi­ DOLHINOW. P. 1984. "The Primates: Age, behavior, tion on Barbados. 1950-1980. MS. [WPH] and evolution/' in Age and anthropological theory. HANKS, LUCIEN M. 1972. Rice and man: Agricultural Edited by J. Keith and D. Kertzer. Ithaca: Cornell Uni­ ecology in Southeast Asia. Chicago: Aldine-Ather­ versity Press. ton·lowl DUMOND, DON E. 1972. I'Population growth and po­ HARPENDING, H. 1976. "Regional variation in !Kung litical centralization," in Population growth: An­ populations," in Kalahari hunter-gatherers. Edited by thropological implications. Edited by B. Spooner, pp. R. Lee and I. DeVore. Cambridge: Harvard University 286-310. Cambridge: M.I.T. Press. [owl Press. DURHAM, w. H. 1978. "The coevolution of human HARRIS, M. 1977. Cannibals and kings. New York: biology and culture," in Human behavior and adap­ Random House. tation. Edited by . Blurton Jones and V. Reynolds. ---. 1979. Cultural materialism. New YOlk: Ran­ (Symposia of the Society for the Study of Human Be­ dom House. [MHI havior 18.) New York: Taylor and hanois. ---. 1984. Animal capture and Yanomamo warfare: FISHER, R. A. 1930. The genetical theory of natural Retrospect and new evidence. Journal of Anthro­ selection. Oxford: Clarendon Press. [AVM, KAMj pological Research 40:183-201. [MHj FRISCH, R., AND J. W. MCARTHUR. 1974. Menstrual ---.1985. 4th edition. Culture, people, nature. New cycles: Fatness as a determinant of minimum weight York: Crowell. for height necessary for their maintenance or onset. --_ 1987. 2d edition. Cultural anthropology_ New Science 185:949-51. YOlk: Harper and Row. IMHj GAGE, T. B., B. DYKE, AND P. G. RIVIERE. 1984a. Es­ HASSAN, F. 1981. Demographic archeology. New timating mortality from two censuses: An applica­ York: Academic Press. tion to the Trio of Surinam. Human Biology 56:489­ HASTINGS, A. 1978. Evolutionarily stable strategies 501. [FMS] and the evolution of life history strategies. J. Den­ --. 1984b. The population dynamics and fertility of sity-dependent models. Journal of Theoretical Biology the Trio of Surinam: An application of a two-census 75:527-3 6. method. Human Biology 56:691-701. [FMSI HA YOEN, B. 1975. The carrying capacity dilemma: An GALLAHER, A., JR., AND H. PADFIELD. Editors. 1980. alternative approach. American Antiquity 40: II-I 6. The dying community. Albuquerque: University of HOWELL, N. 1976. "Toward a uniformitarian theory of New Mexico Press.!Jwsl human paleodemography," in Demographic evolu­ GIDDENS, ANTHONY. 1984. The constitution of soci­ tion of human populations. Edited by R. H. Wald and ety. Berkeley: University of California Press. [AVM, K. M. Weiss. London: Academic Press. KAM] --. 1979. Demography of the Dobe !Kung. New HALDANE, J. B. s. 1932. The causes of evolution. New York: Academic Press. York: Harper and Row. !AVM, KAMI ---. 1980. I/Demographic behavior of hunter­ HAMBURG, D. A., AND E. R. MCCOWN. 1979. The gatherers: Evidence for density-dependent popula­ great apes. Menlo Park: Benjamin/Cummings. tion cantrall II in Demographic behavior. Edited HAM Mn, E. A. 1960. Some models for the analysis of by T. Burch.IAAAS Symposium 45-1 Boulder: West­ marriage-section systems. Oceania 3t: 14-30. view Press. HAMMEL AND HOWELL Population and Culture I 159

---. 1986. Demographic anthropology. Annual Re­1978. An anthropological approach to the study of the views in Anthropology 15. economic value of children in [ava and Nepal. CUR­ HRDY, s. B. 1977. The langurs of Abu: Female and RENT ANTHROPOLOGY 19:293-306. [MH! male strategies of reproduction. Cambridge: Harvard NEEL, J. v., AND K. M. WEISS. 1975. Thegeneticstruc· University Press. ture of a tribal population, the Yanornarna Indians. ISAAC, G. 1978. The food sharing behavior of pIOtohu­ 12. Biodemographic studies. American lournal of man hominids. Scientific American 238:90-108. Physical Anthropology 42:25-5 I. [FMS] KAPLAN, BERNICE A. Editor. 1976. Anthropological NETTING, R. 1981. Balancing on an Alp. Cambridge: studies of human fertility. DetIOit: Wayne State Uni­ Cambridge University Press. versity Press.IAvM/ KAM] PARKINSON, C. N. [975. Parkinson's law and other KIMURA, R./ AND J. CROW. 1963. The measurement of studies in administration. New York: Ballantine. effective population number. Evolution 17:279-88. PETERSEN, W. 1975. A demographer's view of prehis­ KUHN, T. s. 1978. The structure of scientific revolu­toric demography. CURRENT ANTHROPOLOGY tions. Chicago: University of Chicago Press. [ESAj t6:227-45· LANe A ST ER} J. B. 1978. Carrying and sharing in human PRATTIS, I. 1979. The survival of communities: A the­ evolution. Human Nature 1:83-89. oretical perspective. CURRENT ANTHROPOLOGY LEE, R. D. 1984. Malthus and Boserup: A dynamic syn­20:361-75· \TWS! thesis. Working Papers of the PIOglam in Population REID, RUSSEL M. 1976. "Effects of consanguineous Research, Institute of International Studies, Univer­ marriage and inbreeding on couple fertility and off­ sity of California, Berkeley. spring mortality in rural Sri Lanka," in Anthropolog­ LI, F. H. F., J. V. NEEt, AND E. D. ROTHMAN. 1978. A ical studies of human fertility. Edited by Bernice A. second study of the survival of a neutral mutant in a Kaplan, pp. 139-46. Detroit: Wayne State University simulated Amerindian population. American Natu­ Press·IAWM, RAMI ralist II2:83-96.IFMsl ROMNEY, A. KIMBALL. 1971. "Measuring endogamy," LOVEIOY, C. O. 1971. Methods for the detection of cen­ in Explorations in mathematical anthropology. sus errOl in paleodemoglaphy. American An­Edited by Paul Kay, pp. 191-213. Cambridge: M.LT. thropologist 73:101-9. Press. ---,1gB!. The origin of man. Science lII:34I-SO. ROSS, E., AND M. HARRIS. 1987. Death, sex, and fertil­ LOVEJOY, C. 0./ et al. 1985. Multifactorial determina­ ity: Population regulation in pre-industrial and de­ tion of skeletal age at death. American lournal of veloping societies. New York: Columbia University Physical Anthropology 68: 1-14. Press·IMH) MC KEE, L. 1984. Sexual differentials in survivolship SADE, D. s., B. D. CHAPKO-SADE, J. M. SCHNEIDER, and customary treatment of infants and children. S. S. ROBERTS, AND J. T. RICHTSMEIER. 1985. Basic Medical Anthropology 8(21:9I-108.!MH] demographic observations on free-ranging rhesus MC LENNAN, I. F. 1865. Primitive marriage. Edinburgh: monkeys. New Haven: Human Relations Area Files. A. C. Black. SALZANO, F. M., AND S. M. CALLEGARI-JACQUES. MAL TH US, T. R. r976. An essay on the principle of 1986. South American Indians: A case study in evo­ population as it affects the future improvement of so­lution. Oxford: Oxford University Press. In press. ciety. London: ]. Johnson. IFMs! MAMDANI, M. 1973. The myth of population control: SANDERS, w. T., AND B. J. PRICE. 1968. Mesoamerica: Family, caste, and class in an Indian village. New The evolution of a civilization. New York: Random York: Monthly Review Press.IMHJ House. MANN, A. r968. Thepaleodemographic aspects of the SCHEPER-HUGHES, N. 1984. Infant mortality and in­ South African australopithecines. Univelsity of fant care: Cultural and economic constraints on nur­ Pennsylvania Publications in AnrhIOpology I. turing in onheast Brazil. Social Science and MA RX, K. 1967. Capital: A critique of political econ­ Medicine 19:535-46. [MHI omy. Edited by F. Engels. New York: [nternational SCHRIRE, c. Editor. 1984. Past and present in hunter­ Publishers. gatherer studies. New York: Academic Press. MASERANG, c. H. 1976. Factors affecting carrying SCOTT, E. C., AND F. E. JOHNSON. 1985. Science, nu­ capacities of nation-states.lournal of Anthropolog­ trition, fat, and policy: Tests of the critical-fatness ical Research 32:255-75. Irws] hypothesis. CURRENT ANTHROPOLOGY 26:463-75. MILLER} BARBARA. 198!. The endangered sex. Ithaca: SCRIMSHAW, SUSAN. 1983. "Infanticide as deliberate Cornell University Press. [MH! fertility control," in Determinants of fertility in de­ ---. 1983. Daughter neglect, women's work, and veloping countries, vol. 2, Fertility regulation and in­ kinship: Pakistan and Bangladesh compared. Medical stitutional influences. Edited by R. Bulatao and R. Anthropology 8!21:r09-26.IMH! Lee, pp. 245-66. New York: Academic Press.IMHI NAG, MONI, AND N. KAK. 1984. Demographic transi­ SHEETS, J. \V., AND K. E. KELLY. 1987. Depopulation tion in the Punjab village. Population and Develop­and household dynamics in Colonsay IInner Heb­ ment Review IO:661-78.IMH! ridesl. Internationallournal of Sociology and Social NAG, MONI, BENJAMIN WHITE, AND ROBERT PEET. Policy. In press. [rws] 160 I CURRENT ANTHROPOLOGY Volume 28, Number 2, April 1987

SIMON, J. 1978. An inregration of the invenrion-pull tion and development: High and low fertility in and population-push theories of economic-demo­ poorer countries. Edited by G. Hawthorn, pp. 186-97. graphic history. Research in Population Economics London: Frank Casso [MHI 1:165-87· VAN GERVEN, P., AND G. ARMELAGOS. 1983. SKINNER, G. w. '977. The city in late imperial China. "Farewell to paleodemography": Rumors of its death Stanford: Stanford University Press. have been greatly exaggerated. TournaI of Human ---.1985. Gender and power in Japanese families: Evolution 12:353-60. Consequences for reproductive behavior and longev­ VINING, DANIEL. 1985. Social versus reproductive SliC­ ity. Paper read at the 84th annual meeting of the cess: The central theoretical problem of human so­

American Anthropological Associationl Washington/ ciobiology. Behavioral and Brain Sciences 9:167­ D.C., December 4-8. [MH] 216. [MHI SMOUSE, P. E., V. VITZTHUM/ AND ~. V. NEEL. 198r. WACHTER, K. 1985. Mathematical requirements for The impact of random and lineal fission on the ge­ homeostasis in prehistory. Paper presented to the netic divergence of small human groups: A case Stanlord-Berkeley Colloquium in Demography, Oc­ study among the Yanomama. Genetics 98: '79-97. tober. IFMs] WEBER, M. '947. The theory of social and economic or­ SPIELMAN, R. S.! ,. V. NEEL, AND F. H. F. LI. 1977. ganization. New York: Oxford University Press. Inbreeding estimation from population data: Models, WEISS, K. M. '973. Demographic models for anthropol­ procedures, and implications. Genetics 85:355-71. ogy. Memoirs of the Society for American Archeol­ [FMSj ogy 27. SPOONER, B. r972. Population growth: Anthropolog­WHITE, BENJAMIN. 1973. Demand for labor and popu­ ical implications. Cambridge: M.LT. Press. lation growth in colonial Java. Human Ecology STEVENSON, D. Editor. 1984. Population trends in the 1:217-36. [MH] Highlands and Islands. Northern Scotland 6:1-70. ---. 1982. Child labor and population growth in [Jwsl rural Asia. Development and Change 13:587-610. STEWARD/ JULIAN H. 1953. "Evolution and process/I IMHI

in Anthropology today. Edited by A. L. Kroeber, pp. WILLIAMS, L. A' I AND B. J. WILLIAMS. 1974. A re­ 313-26. Chicago: University of Chicago Press. examination of the heritability of fertility in the Brit­ SUSSMAN, R. 1972. Child transport, family size, and in­ ish peerage. Social Biology 21:225-31. [AWM, KAM] crease in human population during the Neolithic. WILSON, E. O. '975. Sociobiology: The new synthesis. CURRENT ANTHROPOLOGY 12:258-59. [MH] Cambridge: Belknap Press. TEITELBAUM, M. s. 1975. Population dynamics and WOBST, M. 1974. Boundary conditions for Paleolithic the state of demographic transition theory. Science social systems: A simulation approach. American 188:420- 23. Antiquity 39:147-77. TESTART, A. 1984. On food storage among hunter­ WOOD, C. 1982. The political economy of infant mor­ gatherers. CURRENT ANTHROPOLOGY 2T523-30. tality in Sao Paulo, Brazil. International Tournai of THOMPSON, E. A. 1979. Fission models of population Health Services 12:2r5-29. [MH] variability. Genetics 93'479-95. IFMSI WRIGHT, SEWALL. 1978. Evolution and the genetics of TILAKARATNE, M. w. 1978. IfEconomic change, social populations. 4 vols. Chicago: University of Chicago differenriaIion, and fertility: Aluthgana," in Popula­Press. [AVM, KAMI