Does Culture Evolve (Fracchia & Lewontin, 1999)

Wesleyan University
Does Culture Evolve?

Author(s): Joseph Fracchia and R. C. Lewontin
Source: History and Theory, Vol. 38, No. 4, Theme Issue 38: The Return of Science:
Evolutionary Ideas and History (Dec., 1999), pp. 52-78
Published by: Wiley for Wesleyan University
Stable URL: http://www.jstor.org/stable/2678058 .
Accessed: 20/06/2014 23:42
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Wiley and Wesleyan University are collaborating with JSTOR to digitize, preserve and extend access to History
and Theory.
http://www.jstor.org
This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

All use subject to JSTOR Terms and Conditions
DOES CULTURE EVOLVE?
JOSEPHFRACCHIAAND R. C. LEWONTIN
ABSTRACT
The drive to describe culturalhistory as an evolutionaryprocess has two sources. One

from within social theory is partof the impetus to convert social studies into "social sci-
ences" providingthem with the status accordedto the naturalsciences. The other comes
from within biology and biological anthropologyin the belief thatthe theoryof evolution
must be universalin its applicationto all functionsof all living organisms.The social-sci-
entific theory of culturalevolution is pre-Darwinian,employing a developmentalmodel
of unfolding characterizedby intrinsic directionality,by definable stages that succeed
each other,and by some criterionof progress.It is arbitraryin its definitionsof progress,
and has had the political problem that a diachronicclaim of culturalprogress implies a
synchronic differentialvaluationof present-daycultures.The biological scheme creates
an isomorphismbetween the Darwinianmechanismof evolution and culturalhistory,pos-
tulatingrules of cultural"mutation,"culturalinheritanceand some mechanismof natural
selection among culturalalternatives.It uses simplistic ad hoc notions of individualaccul-
turationand of the differentialsurvivalandreproductionof culturalelements. It is unclear
what useful work is done by substitutingthe metaphorof evolution for history.
I. CULTURE,THE TWO CULTURES,AND HISTORY
In his well-known "Two Cultures" essay C. P. Snow reported a gap between the
literary and natural-scientific cultures. Acknowledging that "a good deal of the
scientific feeling" is shared by some of his "American sociological friends,"
Snow was well aware that there was a degree of artificiality in limiting the num-
ber of cultures to the "very dangerous" one of two. Yet, he based his binarist
decision largely on the cohesion of the natural-scientific and literary communi-
ties that made of them cultures "not only in an intellectual but also in an anthro-
pological sense."' The intellectual division of labor and the development of dis-
ciplinary languages certainly seem to substantiate his reference to two incom-
mensurate cultures. Anyone who has sat on a university committee reviewing
grant proposals from, and consisting of citizens of, each of the cultures must have
observed the pattern of who accuses whom of using jargon and be convinced that
at least the academic version of Snow's gap, that between the humanities and the
natural sciences, has widened into a seemingly unbridgeable abyss. It has
become commonplace that the two cultures have nothing in common.
Perhaps, however, too much has been made of this abyss. Members of the lit-
erary culture, and of the humanities in general, may be appalled at the thought of
1. C. P. Snow, The Two Culturesand a Second Look (Cambridge,Eng., 1964), 8-9.

DOES CULTUREEVOLVE? 53
scientists mucking around on cultural terrain and subjecting it to "scientific
analysis." But natural scientists seem more irritatedthan intimidated by the
apparentindependenceof humanculturefrom scientific study.And social scien-
tists expressing their discontent about being dangled over the abyss helped
promptSnow to take "A Second Look" and to acknowledge the "coming"of a
"third"social-scientificculturewith the potentialto "soften"the communication
difficulties between the other two.2 Culturalanthropologists,moreover, at least
those with a "scientific"ratherthan a "relativist"bent, could point to a long tra-
dition in their discipline of attemptingto bridge the abyss by subjectingculture
and its "evolution"to scientific study.
The idea that cultureevolves antedatedthe Darwiniantheory of organic evo-
lution and, indeed, HerbertSpencer argued in supportof Darwin that, after all,
everythingelse evolves.3 Of course, the validationof the theory of organicevo-
lution has in no way dependedon such argumentby generalization.It is Darwin-
ism thatbecame the theory of evolution, and, standingSpencer on his head, one
inspirationfor theories of culturalevolution since 1859. There has been a long
and bloody HundredYears War among cultural anthropologistsover whether
humanculturecan be said to evolve, a war in which the contendingpartiesalter-
nate in theirperiods of hegemony over the contestedterritory.That strugglehas,
in part,been a philosophicalconsequence of a diversity in the understandingof
what distinguishesan evolutionaryfrom a "merely"historicalprocess. In greater
part,however, it can only be understoodas a confrontationbetween the drive to
scientize the study of cultureand the political consequences that seemed to flow
from an evolutionaryunderstandingof culturalhistory.
Until the last decade of the nineteenthcentury,partly under the influence of
Darwinism, but also as an extension of pre-Darwinianprogressivistviews that
characterizeda triumphantindustrial capitalism, anthropologicaltheory was
built on an ideology of evolutionaryprogress.Lewis Henry Morgan'sconstrual
of the history of cultureas the progressfrom savagerythroughbarbarismto civ-
ilization was the model. In the 1890s Boas successfully challenged the racism
and imperialismthat seemed the inevitable consequences of Morgan'sprogres-
sivist views and set an anti-evolutionisttone that characterizedculturalanthro-
pology until afterthe Second WorldWar.Beginning with the celebrationin 1959
of the hundredthanniversaryof the publicationof the Origin of Species, there
was a demandfrom within anthropologyto reintroducean evolutionaryperspec-
tive into cultural history from which it had been purged by the Boasites, a
demandthat was later given collateralsupportby the developmentwithin biolo-
gy of sociobiological theories of human nature.But again the implication that
there were "higher"and "lower" stages of human culture, an implication that
seemed built into any evolutionarytheory,could not survive its political conse-
quences, and so by 1980 culturalanthropologyonce again returnedto its Boasian
model of culturalchange, culturaldifferentiation,and culturalhistory,but with-
out culturalevolution.
2. Ibid., 70.
3. HerbertSpencer,The Principles of Biology [1867] (New Yorkand London, 1914), 432-433.

54 JOSEPHFRACCHIAAND R. C. LEWONTIN
In his Preface to the manifesto of culturalevolution redivivus,Evolutionand

Culture,Leslie White bitterlyattackedthe Boas tradition,conflatingit with gen-
eral creationistanti-evolutionism:
The repudiationof evolutionism in the United States is not easily explained. Many non-
anthropologicalscientists find it incrediblethata man who has been hailed as "theworld's
greatest anthropologist". . . , namely Franz Boas, a man who was a member of the
National Academy of Sciences and President of the American Association for the
Advancement of Science, should have devoted himself assiduously and with vigor for
decades to this antiscientificand reactionarypursuit.4
But why does White insist, illogically and counterfactually,that a denial of

cultural evolution is anti-evolutionismtout court? There is a hint in the word
"antiscientific,"but all is explicitly revealed two pages later:"Thereturnto evo-
lutionism was, of course, inevitable if. . . science was to embrace cultural
anthropology.The concept of evolution has proved itself to be too fundamental
and fruitful to be ignored indefinitely by anything calling itself a science"
(emphasis added).5Thus, the demandfor a theory of culturalevolution is really
a demandthat culturalanthropologybe included in the grandtwentieth-century
movementto scientize all aspects of the study of society, to become validatedas
a partof "social science."The issue was particularlypressingfor culturalanthro-
pologists because they were engaged in an institutionalstruggle for supportof
their research and academic prestige with members of their own academic
departmentswho practicedthe undoubtedlyscientific activity of physical anthro-
pology.
But the demandfor a theory of culturalevolution also arose from among the
naturalsciences, particularlyamong evolutionarybiologists for whom the abili-
ty to explain all propertiesof all living organisms,using a common evolutionary
mechanism,is the ultimatetest of the validity of their science. Ever scornful of
what they acronymiouslydubbedthe SSSM (the "standardsocial science model"
based on Durkheim'saxiom), evolutionarybiologists doubtednot thatthe scien-
tific analysis and understandingof the place and evolution of culturein the life
history of Homo sapiens was properlythe province of studentsof humanevolu-
tion. The advent of culture was, after all, a biological adaptationand it must
thereforebe explicable by biological science. Yet a combinationof two inhibit-
ing factors kept the forays of evolutionarybiologists into the culturalrealm to a
minimumat least from the end of WorldWarII into the mid-1970s. These were:
the close link betweenbiologically based pseudoscientificsocial andculturalthe-
ories and genocide; and the lack of a properlycomprehensivetheory.This latter
problem,as most recent culturalevolutionists agree, was finally solved with the
concluding chapterof E. 0. Wilson's Sociobiology (1975) which provided the
impetusfor the latest roundof attemptsto subjecthumanhistory to evolutionary
explanation.There, Wilson sketched the certaintythat, as he put it a few years
4. Leslie White, Preface to Evolutionand Culture,ed. MarshallSahlins and Elman Service (Ann
Arbor, 1960), v.
5. Ibid., vii.

later in On Human Nature, the appropriateinstrumentfor closing the "famous

gap between the two cultures" is "general*sociobiology, which is simply the
extension of populationbiology andevolutionarytheoryto social organization."6
While ratheradamantabouttheir scientific right to explain not just the evolu-
tion of humanculturalcapacities, but also culturalevolution, biologists are also
ratheruneasy about their self-imposed obligation to do so. For they wager the
raison d'etre of science on establishing the validity of the principle of reduc-
tionism: in order for science to remaintenable, it must have universalexplana-
tory power; and this means "nesting"the human sciences in the great hierarchy
of sciences. If evolutionarybiology cannot explain humanculture,then perhaps
its explanationsof other phenomenaought to be reexamined.Intriguedby the
challenge, Wilson noted that reductionis "fearedand resented"7by too many in
the human sciences and, in a bold Napoleonic metaphor, he sniffed "a not
unpleasantwhiff of grapeshot"in the thoughtthat the applicabilityof sociobiol-
ogy to humanbeings is a battle on which hangs the fate of "conventionalevolu-
tionarytheory."8Thrilledby the challenge and inspiredby the apparentpotential
of the sociobiological synthesis, an increasingnumberof scientists attemptedto
build on Wilson's blueprintin orderto bridge the abyss and lay claim to the ter-
ritoryon the other side.
Some membersof the social sciences, those who preferredto be recognized as
bonafide scientists and notjust as membersof a "third"culture,were meanwhile
growing uneasy over the proliferationof opposing theories and models that had
apparentlybroughtthe productionof social-scientificknowledge to a standstill.
Such social scientists began to question their own SSSM and turnedincreasing-
ly to the new and seemingly infallible sociobiological synthesis for the models
and explanatorymechanismsthat would put their own disciplines on propersci-
entific footing. AlexanderRosenberg,for example, bemoans the inability of the
social sciences to live up to John StuartMill's hope for them, namely,to be based
on explanatorylaws. In a telling formulationhe claims that
the social sciences would be of only passing interest,only entertainingdiversions, like an
interestingnovel or an exciting film, unless they too stood the chance of leading to the
kind of technological achievementscharacteristicof naturalscience. For a social science
conceived as anything less practical in ultimate applicationwould simply not count as
knowledge, on my view. And if it does not count as knowledge,disputes about its meth-
ods and concepts are no more importantthanlearnedliterarycriticismor film reviews are
to our uninformedenjoymentof the books and movies we like.9
Rosenbergexpects this to be rectified as soon as the social sciences are treat-

ed as life sciences; and he optimisticallypredictsthatthe study of humanbehav-
ior, once set on a biological footing, "will admit of as much formally quantified
and mathematicaldescription as the most mathematicaleconomist could hope
6. E. 0. Wilson, On HumanNature (CambridgeMass., 1978), x.

7. Ibid., 13.
8. Ibid., 34.
9. Alexander Rosenberg, Sociobiology and the Preemptionof Social Science (Baltimore, 1980),
22-23.

for."Against all claims for their uniquenesshe insists that the traditionalsocial
sciences have been "superseded"by, and will only become truly scientific when
subsumedunder,sociobiology.10
More recently, anthropologistJohn Tooby and psychologist Leda Cosmides
have also chastised the social sciences for their "self-conscious stance of intel-
lectual autarky";their "disconnectionfrom the rest of science has left a hole in
the fabricof our organizedknowledge where the humansciences shouldbe." The
lack of progressin the social sciences has been caused by their"failureto explore
or accept their logical connections to the rest of the body of science-that is, to
causally locate their objects of study inside the larger network of scientific
knowledge."11
This desideratum is the cornerstone of the journal Politics and the Life
Sciences whose editors and contributorsinsist that the social sciences must be
nested within the life sciences. The hopes for a synthesis implicit in thejournal's
name were expressed by Richard Shelly Hartigan in a flattering review of
RichardD. Alexander's The Biology of Moral Systems (1987). Predictingmari-
tal bliss, Hartiganconfidently asserts that "the lengthy divorce of the natural
from the humansciences is aboutto end with reunion.Though the nuptialsmay
be delayed awhile, the partiesare at least getting to know each other again more
intimately."12The reunionconsists of articles devoted to the "Darwinian"expla-
nationof such topics as social alienation,the nucleararmsrace, the legal process,
social stratification,oral argumentin the supreme court, the relation between
humanintelligence and nationalpower, and even feminism.'3
These examples could be multiplied,but as this brief overview indicates, the
biggest engineeringprojectattemptingto bridgethe gap at least between the cul-
tures of the naturaland the human sciences over the last few decades has been
initiatedby naturalscientists, anxious perhapsabouthaving wageredtheirraison
d'etre on the success of their imperialistventure;and it has quickly drawn the
participationof those social scientists optimistic about overcoming their inferi-
ority complex and gaining respectabilityby groundingtheir own disciplines in
the natural sciences. The bridge itself is the concept of "culturalevolution"
whose scientific girders are the categories and explanatorylaws either directly
borrowedor derived from a narrowlyselectionist approachto the study of bio-
logical evolution.
10. Ibid., 4, 158.
11. Leda Cosmides and JohnTooby, "ThePsychological Foundationsof Culture,"in TheAdapted
Mind, ed. JeromeH. Barkow,Leda Cosmides, and John Tooby (Oxford, 1992), 22-23.
12. Richard Shelly Hartigan, "A Review of The Biology of Moral Systems" by Richard D.
Alexander,in Politics and the Life Sciences 7, no. 1 (1988), 96.
13. See for example the following essays, all from Politics and the Life Sciences: Elliot White,
"Self-selection and Social Life: The Neuropolitics of Alienation-The Trapped and the Over-
whelmed" (vol. 7, no. 1, 1989); John H. Beckstrom, "EvolutonaryJurisprudence:Prospects and
Limitationson the Use of Modern DarwinismThroughoutthe Legal Process" (vol. 9, no. 2, 1991);
Lee Ellis, "A Biosocial Theoryof Social StratificationDerived from the Conceptsof Pro/Anti-social-
ity and r/K Selection" (vol. 10, no. 1, 1991); Hames N. Schubertet al., "ObservingSupremeCourt
Oral Argument: A Biosocial Approach" (vol. 11, no. 1, 1992); Larry Arnhart, "Feminism,
Primatology,and Ethical Naturalism"(vol. 11, no. 2, 1992).

At the outset we must make clear what the issue of culturalevolution is not
about.First, there is no question that cultureas a phenomenonhas evolved from
the absence of culture as a consequence of biological change. Whether or not
otherprimateshave cultureon some definition,the insectivores, from which the
primates evolved, do not, so at some stage in biological evolution culture
appearedas a novelty. Second, no one challenges the evident fact thathumancul-
tureshave changed since the first appearanceof Homo sapiens, but not even the
most biologistic theory proposes that majorchanges within the phenomenonof
culture-say the inventionof an alphabetor of settled agriculture-was a conse-
quenceof genetic evolution of the humancentralnervoussystem. Humanculture
has had a history,but to say that cultureis a consequence of a historicalprocess
is not the same as saying that it evolves. What constitutes an evolutionary
process as opposed to a "merely"historicalone? What explanatorywork is done
by claiming that culturehas evolved?
Leslie White's cri de coeur accusing the Boasians of aligning themselves with
anti-evolutionistcreationismconfoundstwo quite differentissues. The mid-nine-
teenth-centurystruggle against evolution, mirroredin modern Christian cre-
ationism,was not over whetherthe succession of life forms from earliertimes to
the present has some law-like propertiesthat give some shape to that history.
Ratherit was, and remains,a denial that organismicforms have had a history at
all, that there has been significant change in species and that present-daylife
forms arosefrom othersquite unlike them. But no one denies thatculturehas had
a history,that industrialproductionarose from societies that were at a previous
time pastoralistand agricultural.Not even the most literal of fundamentalists
thinks that God createdthe motor car on the sixth day. Ironicallyit is a form of
traditional Christianity that simultaneously denies an intelligible history to
organic life as a whole, while asserting a directionalityto human history, the
ascent towardfinal redemptionfrom the depths of the Fall.
White's identificationof the struggleover culturalevolution with the struggle
over organic evolution, if it is more than a deliberatepiece of propagandain a
battle for academic legitimacy, is really a struggle over the natureof historical
processes.At base, it is meantto be a rejectionof the propositionthathumancul-
turalhistory is just one damn thing after another,claiming that, on the contrary,
thereis an underlyingnomotheticprocess. But in assertingthe claim thatculture
evolves White claimed more than what was necessary. History may indeed be
law-like in some sense, but does that make a historical process evolutionary?
There may be law-like constraintson historical change like Ibn Khaldun'srule
that "Bedouinscan gain control only over flat territory,"but we do not therefore
characterizethe Muqaddimahas providing an "evolutionary"theory of history,
any more thanHegel's thirdkind of history,the Philosophical,is claimed to be a
theory of evolution.'4
It might be assertedthatfor theoriesto qualify as evolutionarythey must con-
sist of more than mere constraintsand prohibitions;ratherthey must be charac-
14. Ibn Khaldun,The Muqaddimah(Princeton,1958), chapter2:24.

terized by generativelaws or mechanisms whose operationsproduce the actual

histories. But the Muqaddimahoffers laws of the origin, transformation,differ-
entiation, and eventual extinction of political formations:"Dynasties of wide
power and large royal authorityhave their origin in religion based either on
prophethoodor truthful propaganda";"The authority of the dynasty at first
expands to its limit and then is narroweddown in successive stages, until the
dynastydissolves and disappears";"Withregardto the amountof prosperityand
business activityin them, cities and towns differ in accordancewith the different
size of their population."15 These are not simply empiricalgeneralizations.Each
is derivedas the necessaryconsequenceof basic propertiesof humanmotivation,
just as the war of all againstall is derivedby Hobbes from the basic assumptions
that human beings are, by nature,self-expandingin their demands and that the
resourcesfor theirexpansionare limited. The ease with which the concept of the
"evolution of culture"has been employed in anthropologyand human evolu-
tionary biology finds no parallel in the discourse of contemporaryhistorians.
When Frangois Furet and Mona Ozouf write, in their Preface to A Critical
Dictionary of the FrenchRevolution,that "ignoringthe evolution of historiogra-
phy means overlooking an importantaspect of the event itself," they mean only
that historiographyhas changed, that is, that it has had a history.'6
It might be that "evolution"and "history"are meant to be separatedby ques-
tions of scale and grain.Modes of production,familial and other grouprelation-
ships, forms of political organization,levels of technology are seen as general
propertiesof humansocial existence. They are also "culture"and they are said to
"evolve" while spatio-temporallyindividualized sequences like the events in
Francefrom the Estates Generalto Thermidorare only instantiationsof classes
of culturalphenomena,schematathat are repeatedin differentplaces and at dif-
ferent times. So Leslie White makes the distinctionbetween the particularityof
micro (historical)events and the generality of macro (evolutionaryevents): "I
should like to call the temporalparticularizingprocess, in which events are con-
sidered significant in terms of their uniqueness and particularity,'history' and
call the temporalgeneralizingprocess which deals with the phenomenaas class-
es ratherthanparticularevents, 'evolution'."17 But if this is what is meantto dis-
criminateevolution from mere history,then the culturalevolutionistdepartsrad-
ically from theoriesof evolution of the physical world. For Darwinism,not only
organic life as a whole, but each species and each population in each species
evolves. The standardmodel of organic evolution begins with the evolutionary
forces that cause local populations to change over relatively short times, and
derives the evolution of individualspecies in time from changes in populations
that comprisethem. Moreover,in its usual reductionistform, evolutionarytheo-
15. Ibid., chapters3:3, 3:46, 3:11.

16. A Critical Dictionary of the French Revolution, ed. Frangois Furet and Mona Ozouf
(Cambridge,Mass., 1989), xvi.
17. Leslie White in Social and CulturalEvolution,volume III:Issues in Evolution,ed. S. Tax and
C. Callender(Chicago, 1960), Panel 5.

ry explains the evolution of life as a whole as a mechanicalconsequence of the

rise and fall of individual species. So why, if human culture evolves, has not
Bedouin cultureevolved, or the Middle East, or the state called SaudiArabia?
The attemptto differentiate"culturalevolution" from "history"brings us to
the edge of a different kind of abyss-one that is broader and older, though
obscuredby, the more visible one between the humanand naturalsciences. This
abyss cuts across establisheddisciplinaryboundaries,and separatesnomological
and historical modes of explanation. Civil wars always inflict the deepest
wounds. And the battles within the human sciences (between historiansempha-
sizing contingency and particularityand social scientists insisting on general
laws and models) and within the naturalsciences (between biologists who insist
on the contingency,the historicity,of evolution and those who view evolution as
a lawful process of selection and adaptation)are by virtueof the proximityof the
antagonistsfrequent,intense, and have perhapsthe longest lasting effects.
Snow's depiction of the abyss along disciplinary lines makes those battles
appear as perhaps bitter, but nevertheless only intradisciplinarysquabbles, as
merely differentperspectives on common problems. Yet, the cross-disciplinary
affinitiesof "historians"versus "scientists"are nowheremore evident thanin the
issue thatboth claim as their own: that which appearsto one group as "cultural
evolution,"to the other as "humanhistories."The ease, for example, with which
confirmedselectionists among evolutionarybiologists and those social scientists
similarlyconcernedwith explanatorylaws have found common cause in the con-
cept of culturalevolution indicates that on fundamentalontological and episte-
mological issues there is no abyss between them. That ease finds its counterpart
in the ease with which the two authorsof this essay, a historianand a geneticist,
agree on a historicalapproachto culturalchange. The differencesbetween these
two perspectives are incommensurable,not because of disciplinaryboundaries,
but because they involve different conceptions about the natureof "scientific"
inquiry,differentontological and epistemological assumptions,and accordingly
differentmodes of explanation.
Darwiniantheoristsof culturalevolution universallyagree that selection is the
explanatorylaw, the key to explaining all "evolutionary"or "historical"devel-
opments at any socioculturaland historicalcoordinates.In this way humanhis-
tory is reduced to a unitaryprocess, its complex dynamics to a rathersingular
logic, and the particularityof historicaltime is reducedto "emptyabstracttime"
(WalterBenjamin).18
18. In the introductionto a collectionof essays on Histoty and Evolution,ed. MatthewNitecki and
Doris Nitecki (AlbanyN.Y., 1992) MatthewNitecki statescategorically:"Thecommonelementof evo-
lutionarybiology and historyis the conceptof change over time"(6). Despite greatdifferencesin their
definitionsof the relationbetweenbiology andhistory,all authorsincludedin the volume (andprobably
all culturalevolutionists)shareNitecki'sdefinitionof historyas changeover time.This definitionallows
them to elevate historyto scientificstatusand to subjecthistoryto evolutionaryexplanation,as do, for
example,Boyd andRichersonin theircategoricalclaimthat"Darwiniantheoryis bothscientificandhis-
torical"(in Nitecki, 179-180). While thereare many problemswith defininghistorysimply as change
over time, we will only make two commentshere: this definitionalmost inevitablyresultsin treating
people living in societies in which changeis not the normas people withouthistory;andonce changeis
definedas a transhistoricalconstant,it is very likely, thoughnot logically necessary,thatthe next step
will be to seek a transhistoricalexplanatorylaw-which for culturalevolutionistsis thatof selection.

We begin with differentassumptionsabout historical objects and, according-

ly, abouthistoricaltime. We view historicalphenomenaas particularsembedded
in particularsocioculturalforms, each with its own systemic propertiesand dis-
crete logic of production and reproduction,its own dynamics of stasis and
change. Each socioculturalform thereforehas, to borrow an appropriatephrase
from Louis Althusser, its own time and history. Because every historical phe-
nomenon has its own particularlocus in a particularsocioculturalconstellation
with its own concreteand particulartime and history,there is no one transhistor-
ical law or generalitythat can explain the dynamicsof all historicalchange. Our
contention, therefore,is that culturalevolutionary theories have not been (nor
will be) able to meet even their own claims to explain the past and predict the
future.And this is because of the problematicassumptionsabout the natureof
cultureand the problematicconflationof historicaland evolutionaryprocesses.
II. THE FORMS OF EVOLUTIONARYTHEORY
Models of the evolution of phenomenaare traditionallymodels of the temporal

change in the natureof ensembles of elements. The individual elements in the
ensemble can be physical objects like organismsor stars or propertieslike size
or chemical composition or syntactic structure.So when we speak of the "evo-
lution of humanbeings" we mean a change in the composition of the ensemble
of physical individualsthatwe identify individuallyas human,but we can as well
consider the "evolution of Europeanpainting"as a change in the ensemble of
materials,techniques, subjects, and design principles that characterizethe pro-
ductionof thatart.Whetherit is physical objects or attributesor artifacts,it is not
any individualelement, but the composition of the ensemble that is at the center
of interest.
Evolutionarytheoriesas they have been constructedfor the physical world and
as they have been taken over into human social phenomena can be classified
accordingto two properties.First, they may be either transformationalor varia-
tional. In a transformationaltheory, the ensemble of elements changes in time
because each of the elements in the ensemble undergoesroughly the same secu-
lar change duringits individualhistory.That is, the evolution of the ensemble is
a result of the developmentalpatternof each individual. The transformational
model characterizedall evolutionarytheoriesuntil Darwin,and has remainedthe
model for the evolution of the physical universe since Kant and Laplace pro-
duced the NebularHypothesisfor the origin of the Solar System. The collection
of starsin the cosmos has been evolving because every staris individuallyunder-
going an aging process from its birth at the Big Bang, through a sequence of
nuclearreactionsuntil it exhausts its nuclearfuel and then collapses into a dead
mass. It is this model thatis embodiedin the very word "evolution,"an unfolding
or unrollingof a history that is alreadyimmanentin the object. It is a model of
evolution that takes as its cause the development(desarollo, Entwicklung),the
unrollingor unfoldingof the predeterminedfate of each element in the ensemble.

The alternative,invented by Darwin to explain organic evolution, is a varia-

tional evolutionaryscheme. In variationalevolution, the history of the ensemble
is not a consequenceof the uniformunfoldingof individuallife histories.Rather,
variationalevolution throughtime is a consequenceof variationamong members
of the ensemble at any instantof time. Differentindividualshave differentprop-
erties and the ensemble is characterizedby the collection of these propertiesand
their statistical distribution.The evolution of the ensemble occurs because the
differentindividualelements are eliminatedfrom the ensemble or increase their
numbersin the populationat differentrates. Thus, the statisticaldistributionof
propertieschanges as some types become more common and others die out.
Individualelements may indeed change duringtheirlifetime, but if they do, these
changes are in directionsunrelatedto the dynamic of the collection as a whole
and on a time scale much shorterthan the evolutionaryhistory of the group. So,
the developmentalchanges that characterizethe aging of every living organism
are not mirroredin the evolution of the species. Every humanbeing may become
grayerand more wrinkledwith age, but the species as whole has not become so
in 5 million years of evolution from its common ancestor with other primates.
Organicevolution is then a consequence of a twofold process: the productionof
some variationin propertiesamong individualelements followed by the differ-
ential survivalandpropagationof elements of differenttypes. Moreover,the pro-
duction of the variationis causally independentof its eventual fate in the popu-
lation.Thatis what is meantby the claim thatorganicevolution is based on "ran-
dom" variation.It is not that the changes in individualpropertiesare uncaused,
or the consequence of some force outside of normalphysical events. Ratherit is
thatthe forces of change internalto organisms,leading to the productionof vari-
ant individuals,are causally randomwith respectto the externalforces thatinflu-
ence the maintenanceand spreadof those variantsin the population.Many are
called, but few are chosen.
The invention of the variationalscheme for organic evolution, with its rigor-
ous separationof internaldevelopmentalforces from externalculling forces, is
the major epistemological break achieved by Darwin. All other evolutionary
schemes that had been postulated until the appearanceof the Origin in 1859,
whetherof the evolution of the cosmos, of organisms,of language, or of ideas,
were transformational.The Darwinianvariationalscheme, with its denial of the
causal role of individualdevelopmentalhistories was, in fact, a negationof evo-
lution as it had previouslybeen understood.The retentionof the term"evolution"
by Darwinists,while strippingit utterlyof its former structuralimplication,has
led to a considerableconfusion and ambiguity in subsequentargumentsabout
culturalevolution, for therehas been no agreementamong culturalevolutionists
aboutjust what sort of evolution they mean.
The choice of a transformational,developmentaltheory of evolution implies
propertiesof the process thatare not integralto, althoughthey may be presentin,
a variationaltheory:directionalityand staging. In an unfolding process the pos-
sibility of each successive transformationis dependenton the completion of a

previous step of transformationto provide the initial state for the next change. It
is not necessarythat the complete unfolding be predictablefrom the very origin
of the system because successive steps may be contingent.There may be more
than one local unfolding possible from a given state, and these alternativesmay
be chosen, contingenton variousexternalcircumstances.Transformationaltheo-
ries, nevertheless, usually assume a very restricted contingency, putting very
strong constraintson which states may succeed each other, and in what order.
Indeed the standard theory of embryonic development which provides a
metaphoricalbasis for developmentaltheories of evolution assumes that there is
one and only one possible succession of states.Thus, there is one direction,or at
most a few alternativepossible directions of change immanentin the natureof
the objects.Directionalitydoes not in itself imply thatchangeis monotoneor that
there is a repeatedcycling among states along some simple axis, yet again and
again transformationaltheoriestake the form of a "Lawof Increaseof . . .," com-
plexity, efficiency, control over resources or energy, of Progressitself. The task
of filling in the blanks we leave to later pages. A variationaltheory,in contrast,
does not have directionalitybuilt into it because the variationon which the sort-
ing process operatesis not intrinsicallydirectional,and changes in the statistical
distributionof types in the ensemble are assumedto be the consequenceof exter-
nal circumstancesthat are causally independentof the variation.Nevertheless,
one-way directionalityhas penetratedDarwinismby means of a claim aboutnat-
ural selection. If the differentialnumericalrepresentationof differenttypes in a
species occurs not by chance events of life and death,but because the properties
of some organismsconfer on them greaterability to survive and reproducein the
environmentin which they find themselves, might there not be some properties
thatwould confer a generaladvantageover most or all environments?Such prop-
erties, then, ought to increase across the broadsweep of organismsand over the
long durationof evolutionaryhistory,puttingaside any particularitiesof history.
So, for example, it has been claimed thatcomplexity has increasedduringorgan-
ic evolution, since complex organismsare supposed somehow to be able to sur-
vive betterthe vagaries of an uncertainworld. Unfortunatelyno agreementcan
be reachedon how to measurecomplexity independentof the explanatorywork
it is supposed to do. It is, in fact, characteristicof directionalitytheories that
organismsare first arrayedalong an axis from lower to higher and then a search
is institutedfor some propertythat can be arguedto show a similar ordering.
From directionalityit is only a short step to a theory of stages. Transfor-
mationaldevelopmentaltheories are usually describedas a movement from one
stage to the next in the sequence,from savageryto barbarismto civilization, from
artisanalproductionto competitive industrialcapitalism to monopoly capital.
Development begins by some triggering,startingthe process from its germ, but
there are thoughtto be a succession of orderedstages throughwhich each entity
must pass, the successful passage throughone stage being the conditionfor mov-
ing on to the next. Variationamong individualentities then arises because there
is some variation in the speed of these transitions, but primarily because of

arresteddevelopment, the failure to pass on to the next stage. Freudian and

Piagetiantheories are of this nature.It should be no surpriseto anthropologists
thattransformational evolutionarytheoriesof cultureidentifypresent-dayhunters
and gatherersas being in an arrestedstage of culturalevolution.
The second property that distinguishes among evolutionary schemes is the
mortalityof the individual objects in the ensemble. Members of the ensemble
may be either immortal,or at least have potential lifetimes that are of the same
orderas the ensemble as a whole, or they may be mortalor at least have lifetimes
significantlyshorterthanthe durationof the entire collection whose evolution is
to be explained.The lifetime of the materialuniverse is the same as the lifetime
of the longest lived of individualstars. Individualorganisms,on the otherhand,
invariablyhave their entrancesand their exits, but the species may persist. The
classificationof an evolutionarysystem as either mortalor immortalis indepen-
dent of whether it is transformationalor variationaland the constructionof an
evolutionary theory for a domain of phenomena-culture, for example-will
requiremodel assumptionsabout both of these properties.Two of the schemata
are illustratedby phenomena to which the concept of evolution is commonly
applied. Stellar evolution is a transformationalevolution of a system composed
of immortalobjects; organic evolution is variationaland its objects, individual
organisms,are mortal.Although we do not ordinarilythink of it in such terms,
an example of an evolutionaryprocess that is variational,but whose objects are
immortal,is any separationof a mixtureof physical materialsby sieving, as for
example in panningfor gold. The lighterparticles are washed away, leaving the
flakes of gold behind so that the concentrationof gold becomes greater and
greateras the process continues,yet the same bits of gold are presentat the end
of the process as at the beginning. Pre-Darwiniantheories of organic evolution
were transformational,the entire species evolving as a consequence of slow
directionalchanges in individualswho were, nevertheless,mortal.
The mortalityof the individualobjects in an evolutionaryprocess raises a fun-
damentalproblem,namely,how the changes in the compositionof the ensemble
thatoccur within the lifetime of short-livedelements are to be accumulatedover
the long-term evolution of the group. Whether the evolution is variationalor
transformationalthere must be some mechanismby which a new generationof
successors retainssome vestige of the changes that occurredin a previous time.
In the classical vulgarexample of Lamarckiantransformationalevolution, if the
ancestorsof giraffes slightly elongated their necks to reach up into trees, all the
effort would have been wasted, for after their deaths their offspring would need
to repeatthe process ab initio. Nor does the variationalscheme of Darwin solve
the problem.Were slightly longer-neckedvariantgiraffes to survive better or to
leave more offspringthan their short-neckedcompanions,and so enrich the pro-
portion of the longer variantin the species, no cumulative change would occur
over generationsunless the bias introducedby the sieving process in one gener-
ation were somehow felt in the compositionof the next. Thatis, it demandssome
mechanismof inheritanceof properties,in the broadestsense. Beyond the obser-

vation that offspring had some general resemblance to their parents, neither
Darwin nor Lamarckhad the benefit of a coherenttheory of inheritance,so they
had to content themselves with a variety of ad hoc notions about the passage of
characteristics,all of which had in common that the properties of individual
organismswere somehow directly influencedby the propertiesof their biologi-
cal parentsat the time of conception. Theorists of culturalevolution, conscious
of the need for a theory of inheritance,yet deprivedof any compelling evidence
for particularlaw-like mechanismsfor the transgenerationalpassage of cultural
change, are in a much more difficultposition, althoughthey do not seem to have
realizedit, because they do not even know whetheran actor-to-actor,not to speak
of a parent-to-offspring,model of the passage of culturehas any general applic-
ability.
III.A. THE PARADIGMSOF CULTURALEVOLUTIONARYTHEORY:

TRANSFORMATIONALTHEORIESOF CULTURALEVOLUTION
A remarkablefeatureof the historyof attemptsto createa theoryof culturalevo-

lution is the disjuncturebetween the powerfulimpetusgiven to those attemptsby
the triumphof Darwinism,and the form thatthose essays have takenuntil recent-
ly. Darwin's substitutionof the variationalscheme of evolution for a transforma-
tional one eliminatedthe need for the postulationof intrinsic directionalforces
drivingthe process of change and consequentlyavoided the need for a theory of
progress.If directionalityand its special variant,progress,are claimed to be fea-
turesof a variationalevolutionaryscheme, they must be importedby means of a
force not inherent in the variationalprocess itself. If there is directionality,it
must come from outside of organisms,as a claim, for example, about the nature
of environmentsand their histories. Differential reproductionand survival of
randomly generatedvariantscontains no intrinsic direction. Developmentalist,
transformationaltheories of evolution, in contrast, are directionalby necessity
because the motive mechanismis some form of unfolding of an alreadyimma-
nent program.
Beginning with EdwardBurnettTylor's Primitive Culture(1871) and Lewis
Henry Morgan's Ancient Society (1877), cultural evolutionary theory, called
forth by the historical phenomenon of Darwinism, ignored the structureof
Darwinian explanation, and remained transformationalfor nearly 100 years.
Nearly all of the theories of culturalevolution have had more in common with
Herbert Spencer's Progress: Its Law and Cause (1857) than with Darwin's
Origin. First, they have been dominatedby notions of progress and direction.
This accent on directionand progresshas even been used to characterizeorgan-
ic evolutionitself. In the most importantmanifestoof culturalevolutionism since
its revival afterthe Second WorldWar,Evolutionand Culture,MarshallSahlins
provides a diagramof the evolution, reproducedhere, not of culture, but of all
animal life. Superimposedon the upward trend along the axis of "Levels of
GeneralProgress,"identifiedby Sahlins as "generalevolution,"are minor diver-

sificationswithin a level of progress,symptomaticof "specificevolution"(mere

history,perhaps).19While diagrams like this were icons of nineteenth-century
evolutionism, notions of general progress in biology have been expunged from
currentdescriptionsof organic evolution. In the modernpracticeof reconstruct-
ing phylogeneticrelationships,the antonymof "primitive"is not "advanced,"but
"derived."
Second, given a commit-
// | ment to directionalityand
<
M>AMMI^AL~ progress, it then becomes
LEL
necessary to decide what
PROGRESS criteria should be used to
determine progress aside
BIRDS/ fromlateras againstearlier.
In theories of organic evo-
lution, recurrentattemptsto
i R/E
/TiL
REPTLE
P. I~y
. the notion of progress
use
have foundered on this
issue. It is clear from the
fossil record that there has
A IBANS been no increase in the
FISH duration of species since
the earliest record of multi-
cellular organisms. Nor
INVERTEBRATES would anyone be so foolish
as to predictthatvertebrates
will outlast the bacteria,
should a major catastrophe
overtakeall of life on earth.
PROTOZOA Increasingcomplexityhas
been a favorite of progres-
DIAGRAM I sivist theorists, both for
Diversity and progress among major lineages organic evolution and for
of animallife (schematizcd)
cultural and political struc-
tures, but there is no agreementamong physical scientists on how complexity is
to be measured and there is the recurrentdanger that it will be conveniently
defined,post hoc, to put Homo sapiens at the top. Sahlins dismisses that shibbo-
leth of bourgeois economic theory,efficiency, as a measureon the groundsthat
"an organism can be more efficient than another and yet remain less highly
developed."20By more highly developed he means having more parts and sub-
19. Evolutionand Cultureis the work of four authors,each of whom contributeda chapterto the
volume: Thomas Harding,David Kaplin, MarshallSahlins, and Elman Service. The most influential
of these has been Sahlins's "Evolution:Specific and General."Sahlins's approachto culturehas, of
course, evolved considerablysince 1960.
20. M. Sahlins, "Evolution: Specific and General," in Evolution and Cuture, ed. Sahlins and
Service, 20.

parts,more specializationof partsand more effective integrationand, subserving

these, the transformationof more total energy.Exactly how thatcashes out in the
greatprogressfrom fishes to reptiles in the diagramis not made clear. It is clear,
however, what work is done in the domain of culture. Industrialcapitalismcer-
tainly turns over more calories per capita than does the economy of the Yana-
mamo of the Orinocanrainforest, and almost any descriptionof a Europeanpoli-
ty of 1999 will show it to have more parts and subpartswith greaterspecializa-
tion than a fief in thirteenth-centuryEurope,althoughthe questionof the relative
integrationof feudal and bourgeois society as a whole can be debated.Nor can
this characterizationof an increasinglevel of culturalprogressbe attackedon the
grounds that some earlier cultures, say Athenian democracy, as most would
agree, were more progressive than Carolingianfeudalism. The combinationof
general and specific evolution allows for local exceptions, especially if cultures
in differentpartsof the world are undergoingindependentevolutionarytrajecto-
ries because accidentsof geographypreventany effective contact between them
or because catastrophichistorical events have left a culture without a sufficient
populationto sustain it. It is only the long sweep of humanculturalhistory that
is meant to be progressive.The problem with such a theory is that it is hard to
imagine any observationthat could not be rationalized.The mere numerosityof
the human species makes it impossible to returnto feudal agriculturalproduc-
tion, althougha global nuclearwar with a 95% mortalityrate might do the trick.
Wouldthat be an example of specific or general culturalevolution?
Third,transformationalevolution demandsa mechanism,or at the very least,
a set of empirical law-like regularitiesthat are characteristicof all times and
places, even if these cannotbe generatedfrom lower level mechanicalprinciples.
Transformationaltheoriesof culturalevolution, to the extent that they attemptto
generate putative trends from some lower level principles at all, usually do so
from middle level laws of the same ontological status as Ibn Khaldun'sgenera-
tive rules, ratherthan deriving them explicitly from propertiesof humanbeings
and their consequentinteractionsin assemblages, as Hobbes did. Evolution and
Culture provides a "Law of CulturalDominance" that assures that more ad-
vanced cultures will spread and replace the less advanced when they come in
contact, and a "Law of EvolutionaryPotential"that asserts that the more spe-
cialized and adaptedto local circumstancesa culture is, the less likely it is to
progressto a higher stage. Beyond appealingto the reasonablenotion that cul-
turesthatcontrolmore energy are likely to take over those that control less, pro-
vided they do not destroy themselves in the meantime,and the rathermore ide-
ological prejudicethatprogresscomes from struggle,no lower level mechanisms
are adducedthat generatethese laws.
Although transformationaltheories do not have carefully articulatedlower
level mechanismsproviding the mediation for the law-like higher level proper-
ties that are claimed, there is general agreementon elements that would go into
such a theory of mediation.Humanbeings have certainproperties:
1. They have great physical power to altertheir surroundingcircumstances;

2. They have self-reflexive consciousness so they can assess and react to their
own psychic states;
3. They can imagine and plan what does not yet exist, so they can invent nov-
elties;
4. They have a recursivelinguistic function that allows them to communicate
complex hypotheticalstructuresand causal assertions;
5. They are always born and develop psychically in group contexts.
These propertiesare sufficient to allow groups of human beings to generate a
varietyof artifacts,activities, and group relations,to decide how well these sat-
isfy theirphysical and psychic desires, to consciously plan and altertheir activi-
ties and beliefs, and to pass information about these activities and beliefs
between individuals and across generationalboundaries,and they generate the
possibility of coercing or convincing other groups to adoptparticularpatternsof
activity.
The problemwith this list of propertiesof humanbeings and the powers that
derive from them is that they contain no assertionsabout the natureof the trans-
formation of individual propertiesinto group propertiesand structures,or the
way in which individualsare transformedby the group, or the mannerin which
grouppropertieshave theirown dynamicrelationships.That is, thereis no social
theory or psychosocial theory.Of course, a completely atomistic and reduction-
ist evolutionarytheory would not require such a social theory,but no transfor-
mational theory of cultural evolution denies the relevance of social and psy-
chosocial causes. There is simply no agreementon what these are or how they
would generate the "laws" of directionalityand progress. It has remained for
variationaltheories of culturalevolution to play the reductionistgame.
III.B. THE PARADIGMSOF CULTURALEVOLUTIONARYTHEORY:

VARIATIONALTHEORIESOF CULTURALEVOLUTION
Variationalmodels for culturalevolution have appearedin the last twenty years

as a concomitantof the invention of sociobiology and its transformationinto
evolutionarypsychology. It was the intention of sociobiology to give an ortho-
dox Darwinianexplanationof the origin of majorfeaturesof humanculturelike
religion, warfare,family structure,and so on, as manifestationsof the higher
reproductiverate of individuals with certain behavioral properties,but not to
explain changes thathave occurredin the forms of those phenomenaduringthe
process of human history. Indeed, the chief evidence offered for the origin of
these featuresthroughbiological, genetic evolution was precisely that they were
universal.All human cultures have religion, all engage in warfare, and E. 0.
Wilson claimed thatmale dominationin society would persist indefinitely.2'The
ambition to extend classical Darwinism to the explanation of all aspects of
species life, including species social behavior,resultedin an immense populari-
ty of adaptiveevolutionarythinking in fields like economics, political science,
21 . E. 0. Wilson, "HumanDecency Is Animal,"New YorkTimesMagazine (October 12, 1975).

and psychology that were in search of more "scientific"explanatoryschemes.

One result of this intellectual fashion was, ironically, the creation of formal
Darwinianmodels of differentiationand temporalchange of social institutions,
but withoutthe biological genetic content of organicevolution. It is importantto
stressthatDarwiniantheoriesof the evolution of humanculturaldiversityin time
and space are emphaticallynot theoriesthatthis diversityis based in genetic dif-
ferences and thatgenetic evolution is at the base of the change from agricultural
to industrialsocieties, or the developmentof the centralizedstate. Instead,a vari-
ety of theories of culturalevolution have been createdthat are isomorphicwith
the skeletal structureof Darwinianevolutionarytheory,substitutingfor its vari-
ous concrete biological elements analogicalfeaturesfrom culture.
The skeletal structureof the Darwinianvariationalscheme for organic evolu-
tion consists of three assertions:
1. Individual organisms within populations vary from one another in their
characteristics.This variation arises from causes within organisms that are
orthogonalto their effects on the life of the organism(The Principleof Random
Variation).
2. Offspringresemble their parents(and other relatives) on the average more
than they resemble unrelatedorganisms(The Principleof Heredity).
3. Some organisms leave more offspring than others (The Principle of
DifferentialReproduction).The differentialreproductionmay be a direct causal
consequence of the characteristicsof the organism(naturalselection), or it may
be a statisticalvariationthatarises from purelyrandomdifferentialsurvival.This
latterpossibility is often ignored in vulgar expositions of Darwinianevolution,
and all changes are ascribedto naturalselection, but it is now certainthat a great
deal of evolution, especially molecularevolution, is a consequence of stochastic
variationsin reproduction.
If there is no variation among organisms, then even if different individuals
leave differentnumbersof offspring,nothingwill change. If therewere no hered-
ity of characteristics,then even if differentorganismsleft differentnumbersof
offspring,there would be no effect on the characteristicsof the next generation.
Finally, if differentorganisms all left exactly the same numberof offspring no
change would be expected in the composition of the population.In orderto pro-
duce a scheme of culturalevolution that is isomorphicwith the Darwinianvari-
ational structurethere must be analogs of its elements.
The productionof those analogs has occupied a great many people in a vari-
ety of disciplines over the last few decades. With so many competing models
produced,it is hardlysurprisingthatthereis a greatdeal of spiriteddebateamong
the authorsof the largeand expandingliteratureon culturalevolution.22But how-
22. On the basis of the two questionshe asks of each culturalevolutionarytheory,WilliamDurham

in Coevolution:Genes, Culture,and HumanDiversity (Stanford, 1991) is able thoroughlyto survey
the culturalevolutionistplain and, in so doing, he providesa sense of its paradigmaticunity.His ques-
tions are: "Is culturea second inheritancesystem? What are the best units to use in the study of cul-
turaltransmission?"(155). Based on the responses, he establishes in an approximatelyl] chronolog-
ical" order(155) a tripartitedivision of the culturalevolutionist terrain.The earliest theories of cul-

ever full of sound and fury, this debate is essentially an intramuralaffair. For
beneath all the differences in details, there is a paradigmaticunity among
Darwiniantheories of culturalevolution based on the assumptionthat cultural
evolution can and must be explained in terms isomorphicwith the three princi-
ples of Darwin's variationalscheme. Before they can proceed with that explana-
tion, however, culturalevolutionists undertakea clean-upproject, accomplished
throughsleights of conceptualhand,thatclears away anythingbetween the "bio-
logical" and the "cultural"thatmight have a constitutiveeffect in the production
and"evolution"of culturalforms.This entails firstof all the disappearanceof the
social or, at least, deprivingthe social of causal efficacy, and then the neutraliz-
ing of culture.
The easiest way to make society disappearis simply to dissolve it by defini-
tionalfiat into a mere population.E. 0. Wilson, for example, writes:"Whensoci-
eties are viewed strictly as populations, the relationshipbetween culture and
hereditycan be definedmore precisely."23RobertBoyd and PeterRichersonstate
rathercategorically that "culturalevolution, like genetic evolution in a sexual
species, is always a group or population phenomenon";and in a later work:
"because cultural change is a population process, it can be studied using
Darwinianmethods."24A more nuancedway of dissolving society into a collec-
tion of atomistic individualsis to create a choice between two extreme alterna-
tives. Melvin Konner, for example, correctly rejects the society-as-organism
metaphorby contrastingthe cell that is devoted "entirely to the survival and
reproductionof the organism"with "thepurposesof the individualhuman [that]
turalevolution tended to be "models without dual inheritance."These conceptualizeculture not "as

partof the phenotype";and they explain "phenotypicchange in humanpopulationsin terms of a sin-
gle fitness principle,namely,reproductivefitness in one of its guises" (155-156). Examples include:
both the strongerandweakerversionsof sociobiology, David Barash's"geneticdeterminism,"and the
"genes on a leash" model of Wilson's On HumanNature.The second type, "modelswith dual inher-
itance and traitunits,"also conceptualizesculture "as partof the phenotype"but views it "as a sec-
ond, nongeneticinheritancesystem whose units are defined as culturallyheritableaspects of pheno-
type. These units arerecognizedas having theirown measureof fitness within the culturalsystem [i.e.
'culturalfitness'].... The differentialtransmissionof traitsor behaviorswithin a populationconsti-
tutes culturalevolution"(156). Examples are: Cavalli-Sforza'sand Feldman's culturaltransmission
model, RichardAlexander's social learning model, Lumsden's and Wilson's gene-culturetransmis-
sion model, and his own early coevolution model. The thirdtype is "modelswith dual inheritanceand
ideationalunits."These also treatculture"as a separate'track'of informationalinheritance,"but they
focus not on phenotypicaltraits,but the differentialtransmissionof ideas, values, and beliefs in a pop-
ulation"(156). Examplesinclude:AlbertG. Keller's social selection model, H. RonaldPulliam's and
ChristopherDunford's "programmedlearning model," Boyd's and Richerson's "Darwinianculture
theory,"and Durham'sown book, Coevolution.The "evolution"of theories of culturalevolution has,
as Durham indicated, roughly followed the sequence of his types: from behavioral to (Clifford
Geertzian) ideational definitions of culture and from single inheritance,culture-on-a-genetic-leash
models to dual inheritancemodels. This evolutionresultedfrom dissatisfactionwith single inheritance
models for tying culture to a short genetic leash, and with behavioraldefinitions of culturefor their
unreliability,the impossibilityof knowing precisely which meme motivatesa given behavior.
23. Wilson, On HumanNature, 78.
24. RobertBoyd and PeterRicherson,Cultureand the EvolutionaryProcess (Chicago, 1985) 292;
and "How MicroevolutionaryProcesses Give Rise to History"in History and Evolution, ed. Nitecki
and Nitecki, 181.

are wedded to the survival and reproductionof the society only transientlyand
skeptically."But he overdrawsthe consequencesof this obvious insight and con-
cludes that evolution "has designed the individual with a full complement of
independenceand a canny ability to subvert,or at least try to subvert,the pur-
poses of society to its own. Every time a humanbeing gets fed up with his or her
society or churchor club or even family, and voluntarilychanges affiliation,we
have anotherfactual disproof of the centralmetaphorof social and political sci-
ence."25Here he assumes thatthe repudiationof the obviously false metaphorof
society as organismis a justificationfor an equally obviously false atomisticindi-
vidualism that renderssociety a mere population.
However accomplished,the dissolution of societies into populationsor, as in
more nuancedapproaches,the reductionof differentialsocial power to the status
of a subordinatevariable,26precludes the possibility that social systems might
have propertiesuniqueto them as organizedsystems, thatis, that social relations
might be characterizedby structuresof unequal power that affect individual
social behaviorand the fitness of culturaltraits.This dissolution means, in turn,
that social hierarchyand inequalityare explained as just the consequence of the
differentialculturalfitness of individualsor of the culturaltraitsthey bear,rather
than, say, as a consequenceof antagonisticand exploitative social relations.27
25. Melvin Konner,The Tangled Wing:Biological Constraintson the Human Spirit (New York,
1982), 414.
26. Unawareof the implicationsof theirreductionof societies to populations,Boyd and Richerson,
muchto theirsurprise,foundthemselvescriticizedby David Rindos("TheEvolutionof the Capacityfor
Culture:Sociobiology,Structuralism, andCulturalSelectionism,"CurrentAnthropology27 [1986], 315-
316) andWilliamDurham(Coevolution,179ff.) for not havingadequatelyaddressedthe social. In their
directresponseto Rindos(includedin Rindos,327), Boyd andRichersonclaim, correctly,thatthey spent
an entirechapterof theirCultureand the EvolutionaryProcess on "thescale of humansocial organiza-
tion,"implying,incorrectly,thattherewiththe matterwas resolved.Thatchapterfirstdevelops a taxono-
my of biases (direct,indirect,and frequencydependent)and then constructsmodels to analyzehow the
frequencyof these biases affects the transmissionof culture.Thoughsuch biases certainlyaffect social
behavior,theiroriginsandpersistencearenowherediscussed.Consequently,they end up explaininghow
social biases affect individualchoice by transformingcliches into explanatoryprinciples:"When in
Rome,do as the Romansdo"becomesthe law of "frequency-dependent bias"(286) and"keepingup with
the Joneses"the law of "indirectbias"(287). The questionsof whetherall Romansdo as some Romans
do or of whetherkeepingup with the Jonesesmakessense in societies not basedon commodityproduc-
tion andexchangeare crucialquestionsthatdisappearin theirbiases.
Durhammakes perhapsthe most concertedeffort to considerasymmetriesof social power and the
"imposition"of group values on individual"choice"(Coevolution,198-199). He identifies "reference
groups"withina given population,therebyacknowledging"thesimple fact thatculturalevolutionis an
intrinsicallypoliticalprocess"(211). Because he does not ask the essentialquestionsof why particular
"referencegroups"exist andwhatis the distinctlyanddiscretesocial logic behindparticularasymmetries
in grouppower,Durhamcan only treatany particularset of referencegroupsand social asymmetriesof
poweras arbitraryandsubordinatevariablefactorsaffectingindividualchoice, ratherthanas constitutive
factorsof social and culturalformsand their"evolution."
27. The most perspicaciouscriticof theoriesthatreducesocieties to populationswas KarlMarx.The
hallmarkof politicaleconomy and the sourceof its errors,Marx argued,was that it took as its starting
point the populationwithouthaving determinedthe componentsof the populations,its "subgroups"or
classes and the logic of theirinternalrelations.Such an approachwould producenot "a rich totalityof
many determinationsand relations,"but "ever thinnerabstractions"and "a chaotic conceptionof the
whole"(Marx-EngelsReader,ed. RobertTucker[New York, 1978], 237). Or as he latersummarizedit
more succinctly:societyey does not consist of individuals,but expressesthe sum of interrelations,the
relationswithin which these individualsstand"(247). The analysis of a society reveals much aboutits
population,but the converseis not necessarilytrue.

Having taken the crucial preliminarystep of dissolving society, the next step
is, perhaps surprisingly,to neutralizeculture as well. In order to qualify as an
instanceof a variationaltheoryof evolution, culturemust be provento consist of
isolatable, individualentities, and to be only the sum of its parts. It is thus nec-
essary to refute any and all claims thatcultureshave unique and discreteproper-
ties and a system-specificlogic thatrequirethem to be analyzedeach on its own
terms. This is sometimes done by definitionalfiat aimed at anothersuperorgan-
ismic strawman. E. 0. Wilson, for example, insists that "culturesare not super-
organismsthatevolve by theirown dynamics."Culture,concursJeromeBarkow,
"is not a 'thing,' not a concrete, tangible object. It isn't a cause of anything.To
describebehaviouras 'cultural'tells us only that the action and its meaning are
sharedand not a matterof individualidiosyncrasy."28
The definitionalfiats thatposited population-likemodels of culturereceived at
least two slight challenges. Discontentwith an excessively atomisticview of cul-
ture, Bernardo Bernardi, for example, constructs a constellation of "anthro-
pemes" consisting of "ethnemes,"themselves subdivided into "idioethnemes"
and "socioethnemes";and Martin Stuart-Foxdivides memes into mentemes.29
Though these attemptsappearto reject the notion of isolated, individualmemes
and to aim at systematiccomplexity, they fall short.Tellingly, in suggesting the
division of the meme into mentemes, Stuart-Foxquite consciously attemptedto
constructa categorialanalogywith modernlinguistic terminology.But he did not
follow up this overtureand consider Saussure's fundamentalinsight on which
modem linguistics is based, namely that meaning is system-specific, that each
term (sign) acquiresits historically-specificmeaning by virtue of its place with-
in a discreteset of differentialrelations.By neglecting this insight, attemptssuch
as Stuart-Fox'sand Bernardi'sfocus only on the aggregateratherthan the sys-
temic. Only additive in method, they treatmemes as aggregatesof smallerenti-
ties, as cultural molecules composed of cultural atoms-which effects only a
slight displacementof their ontological individualism,reproducingit at the level
of compounds.
Coevolutionistshave also made overturesto the systemic characterof culture
by removing it from a tight genetic leash and insisting that cultureevolves rela-
tively autonomouslyon its own culturaltrack. But regardlessof the numberof
evolutionarytracksadvocated,all theories of culturalevolution pay only lip ser-
vice to the complexity of culture:because they persistin treatingcultureas mere-
ly the sum total of individual cultural units at a given stage in the selection
process, as a kind of "stateof the 'memes"' at a given point in time, they deny
cultureany system-specificcharacteristics;and this, in turn,allows all culturesto
be explained according to the same (transhistoricaland therefore ahistorical)
selectionist logic.
28. E. 0. Wilson, On Human Nature, 78; Jerome Barkow, Darwin, Sex, and Status: Biological
Approachesto Mind and Culture(Toronto,1989), 142.
29. Bernardo Bernardi, "The Concept of Culture:A New Presentation"in The Concepts and
Dynamics of Culture,ed. BernardoBernardi(The Hague, 1977); Martin Stuart-Fox,"The Unit of
Replicationin Socio-culturalEvolution,"Journalof Social and Biological Structures9 (1986), 67-90.

With society and culturereducedto mere aggregatesand deprivedof any sys-

temic and system-specific characteristics,the ground is preparedfor the con-
structionof a scheme of culturalevolution thatis isomorphicwith the Darwinian
variationalstructure.This, as mentionedabove, requiresthe constructionof cul-
tural analogs to the three fundamentalprinciples of the Darwinian variational
scheme.
First, a decision has to be made about the Principle of Random Variation,
about the identity of the objects that have variation, heredity, and differential
reproduction.Are these objects individualhumanbeings who are the bearersof
differentculturalcharacteristicsand who pass on those characteristicsto other
humanbeings by variousmeans of social and psychological communication,and
who have differentialnumbersof cultural"offspring"?This is the approachgen-
erally favoredby those focusing on behaviorand definingculturalin behavioral-
ist terms. Or are they the characteristicsthemselves with propertiesof heredity
and differential reproduction?This is the more common approach in recent
years, especially among the "coevolutionists"who have taken an "ideational"
view of cultureusing so-called "trait-based"models of the evolutionaryprocess.
An example of the formeris Cavalli-Sforza'sand Feldman's theory of cultural
transmission,while Dawkins's "memes"are an example of the latter.30
Eitherway, a fundamentalproblemresultsfrom the assumptionthatthese cul-
turalunits, say the idea of monotheism,or the periphrastic"do,"somehow spread
or disappearin humanpopulations,namely: no theory of culturalevolution has
providedthe elementarypropertiesof these abstractunits. Presumablythey are
mortaland so need rules of heredity.But, for a variationaltheory,it must be pos-
sible the count up the numberof times each variantis represented.What is the
equivalentfor memes of the numberof gene copies in a population?Perhapsit
is the numberof individualhumanbeings who embody them, but then the death
of a human carriermeans the loss of a meme copy and so memes do, after all,
have the problemof heredity.A majorproblemof creatinga variationaltheoryof
culturalevolution is thatthe task of buildinga detailedisomorphismhas not been
taken seriously enough.
Once the individualunits are settled upon, little time is spent determiningthe
sources of variationin those units, the "culturalanalogs of the forces of natural
selection, mutation,and drift that drive genetic evolution."'3'Following a quick
definitional determinationof the sources of variation-randomness and drift,
selection, and perhapsthe additionof a uniquely culturalsource such as inten-
tionality-the next step is to find the culturalanalogsto the Principleof Heredity.
Most culturalevolutionists simply accept as given that cultureis a system of
heredity or at least of unidirectionaltransmission.Boyd and Richerson state
axiomaticallythat"Darwinianmethodsare applicableto culturebecause culture,
30. L. L. Cavalli-Sforza and M. W. Feldman, Cultural Transmission and Evolution: A
QuantitativeApproach(Princeton,981); RichardDawkins, The Selfish Gene (New York, 1976).
31. See Cavalli-Sforza and Feldman, Cultural Transmission and Evolution, 10; Boyd and
Richerson,Cultureand the Evolutionat-yProcess, 8ff, and "How MicroevolutionaryProcesses Give
Rise to History,"182.

like genes, is informationthat is transmittedfrom one individual to another"
(emphasisadded). In a lateressay they turninheritanceinto the definingcharac-
teristic of cultural evolutionary theory: "The idea that unifies the Darwinian
approachis thatcultureconstitutesa system of inheritance";and aftera brief dis-
cussion thatmoves from inheritancethroughthe "population-levelproperties"of
culturethatmakes it "similar... to gene pools," they conclude that"becausecul-
tural change is a populationprocess, it can be studied using Darwinian meth-
ods."32
To be sure, however, Boyd and Richerson spoke a bit too inclusively. While
some cultural evolutionists use "inheritance"and "transmission"interchange-
ably,othersare uneasy aboutthe genetic and parentalovertonesof "inheritance"
and prefer "transmission."But both terms refer to a process of descent that
occurs in the same unidirectionalmannerbetween an active donor and a passive
recipient.The semantic advantageof "transmission"is that it drops the genetic
connotationalbaggage of "inheritance"while preservingthe portrayalof cultur-
al change as a unidirectionalprocess of descent with modificationand selection.
Whetherconceptualizedas "heredity"or "transmission,"however, the prob-
lematic issue is that both terms requirethe establishmentof some laws of the
heredity of units or their characteristicsif human individuals are the units. We
then requirethe details of the passage of cultureto new individuals,by analogy
with the Mendelianmechanismof the passage of genetic informationfrom par-
ent to offspring by way of DNA. In making this analogy,however, the biologi-
cal model implies constraintsthat have not been apparentto culturalevolution-
ists. We say that parents"transmit"their genes (or at least copies of their genes)
to their offspring, so models of cultural evolution begin with models of the
"transmission"of culturaltraitsfrom one set of actorsto othersby analogy with
the transmissionof genes. Parentsmay transmittraitsto their children,or teach-
ers to their pupils, or siblings and other peers to each other by a variety of sim-
ple rules. The outcomes of evolutionary models of this kind turn out to be
extremely sensitive to the postulatedrules of transmission,and since there is no
firmbasis on which to choose the rules, almost anythingis possible. But thereis
a deeperproblem.Is culture"transmitted"at all? An alternativemodel, one that
accords better with the actual experience of acculturation,is that culture is not
"transmitted" but "acquired."Acculturationoccurs througha process of constant
immersion of each person in a sea of culturalphenomena, smells, tastes, pos-
tures, the appearanceof buildings, the rise and fall of spoken utterances.But if
the passage of culturecannotbe containedin a simple model of transmission,but
requires a complex mode of acquisition from family, social class, institutions,
communicationsmedia, the work place, the streets, then all hope of a coherent
theory of culturalevolution seems to disappear.Of course, it was simplerin the
Neolithic, but there was still the family, the band, the legends, the artifacts,the
naturalenvironment.
32. Peter J. Richerson and Robert Boyd, "A Darwinian Theory for the Evolution of Symbolic
CulturalTraits,"in The Relevance of Culture,ed. MorrisFreilich (New York, 1989), 121; Boyd and
Richerson,"How MicroevolutionaryProcesses Give Rise to History,"181.

Some dissenters present serious challenges to the inheritance/transmission

model even thoughthey remainfaithful to its explanatoryprinciple.MartinDaly
questions the value of the inheritancemodel because he finds no culturalanalog
to the gene, because cultural traits "are not immutable" like genetic traits,
because cultural "transmissionneed not be replicative,"because the recipients
are not "simply vessels to be filled," and because "social influence"makes the
processes of culturalchange less regularthan is implied by the term "transmis-
sion."33Though Daly and others raise perfectly legitimate and very important
questions about inheritance and transmission analogies, they deprive their
insights of real force by still maintainingthatculturalchange is a process thatcan
and must be explained in terms isomorphic with "the evolutionary model of
man."34
This assumptionbrings us to the third analogical element in theories of cul-
turalevolution, the Principleof DifferentialReproduction.Whetherthey define
the units as culturalatoms or culturalmolecules, whetherthey speak of cultural
change as inheritance, or of transmission to passive recipients or to active
acquisitors,they all insist that culturalchange is a process of descent with mod-
ification;and as such it has all the attributesof a variationalevolutionaryprocess
eligible for Darwinian,thatis, selectionist explanation.To all culturalevolution-
ists may be extended that which MartinStuart-Foxsaid of himself, namely that
they "takefor granted (a) the scientific statusof the synthetictheoryof evolution
and (b) thatthis theory provides the most likely model on which to base a theory
of culturalevolution"(emphasis added).35
However, the forces that cause the differentialpassage of cultureacross gen-
erations and between groups seem not to be encompassed by the reductionist
model in which individualactorshave more culturaloffspringby virtue of their
persuasivenessor power or the appeal of their ideas, or in which memes some-
how outcompete others through their superior utility or psychic resonance.
Atomistic models based on the characteristicsof individualhumans or individ-
ual memes can be made, but they appearas formal structureswith no possibility
of testing their claim to reality. How are we to explain the disappearanceof
Germanand French as the languages of internationalscientific discourse, and
their universalreplacementby English without terms like "Nazi persecutionof
Jews," "industrialoutput,""militarypower in the Cold War,"or "gross national
product."That is, no variationaltheory of culturalchange can be adequateif it
attemptsto create a formal isomorphismwith Darwinistindividualism.
Historical,political, social, and economic phenomena,in short, must be dis-
mantledin orderto be molded into the raw materialfor selectionist theories of
33. MartinDaly, "SomeCaveatsaboutCulturalTransmissionModels,"HumanEcology 10 (1982),
402-404.
34. Ibid., 406. See also David Hull, "TheNaked Meme," in Learning,Development,and Culture:
Essay in EvolutionaryEpistemology,ed. H. C. Plotkin (Chichester,Eng., 1982); H. Kaufman,"The
NaturalHistory of HumanOrganizations,"Administrationand Society 7 (1975); TimothyGoldsmith,
TheBiological Roots of HumanNature: Forging LinksBetween Evolutionand Behavior (New York,
1991).
35. Stuart-Fox,"The Unit of Replicationin Socio-CulturalEvolution,"68.

culturalevolution. This is effected throughthe dissolutionof social systems with

structuralasymmetriesof power into individuals;and throughthe reductionof
culturalsystems to eclectic aggregatesof differentiallyreproducedmemes. This
dualprocess stripshistoricalphenomenaof theirsocioculturalparticularity.Once
transformedin this way, they may be subjected to nomological explanationas
individualinstances of the exogenous, because transhistorical,law of selection.
Even the recognitiongiven by William Durhamand others to the systemic char-
acter of culture and to the possibility that social asymmetriesof power might
affect culturaltransmissionand fitness are drainedof content by the fundamen-
tal assumptionsof the culturalevolutionistparadigm:the definitionof cultureas
an aggregateof individual,heritableunits and the selectionist explanationof its
evolution.And in these assumptionslies the self-validatingcircularityof cultur-
al evolutionarytheories: selectionist explanationrequires individual, heritable
units of culture;and reductionof cultureto an aggregateof such units rendersit
susceptible to selectionist explanation-whose scientific status had been taken
for grantedfrom the very beginning.
As its etymology suggests, any "theory"is a way of looking at the world, and
what one sees is that which is visible throughone's particularset of theoretical
lenses. Culturalevolutionarytheories, however, base (and wager) their claim to
breakthroughall theoreticalbiases and to attainscientific statuson their verifia-
bility, their ability to postdict past and predictfutureculturalevolution. If, with
the emergenceof the hegemony of the physical sciences, the cornerstoneof a sci-
entific theory has been the eliminationof the historical,its touchstonehas been
its predictivecapacity-a matterthatculturalevolutionists addresswith increas-
ing confidence.
We have alreadyencounteredAlexanderRosenberg'soptimism about the use
of mathematicalmodels in the new sociobiologicallybased social sciences andhis
confidencein theirpredictivecapacities.36The same optimismis prevalentamong
the contributorsto Politics and the Life Sciences who are convinced thatthe pre-
dictive powers of the new evolutionarypolitical science will renderit capableof
informingpolicy decisions. Certainthat Darwinianmodels of culturalevolution
can produce"a useful retrodictionof ethnography,"Lumsden and Wilson were
somewhatcircumspect,anticipatingonly predictionsof "shorttermchangesin the
forms of ethnographicdistributions."Nevertheless,they remained-and Wilson
has become ever more-optimistic that "the history of our own era can be
explainedmoredeeply andmorerigorouslywith the aid of biological theory,"and
thatthis approachmightenableus to look "downthe world-tubeof possible future
histories."37 Similarly,Boyd and Richersonquickly overcametheirinitial caution
to assertthat "Darwinianmodels can make useful predictions."38
36. Rosenberg,Sociobiology and the Preemptionof Social Science, 151.

37. CharlesLumsden and E. 0. Wilson, Genes, Mind and Culture:The CoevolutionaryProcess
(Cambridge,Mass., 1981); 358, 360, 362. See also Wilson, Consilience (Cambridge,Mass., 1998).
38. Boyd and Richerson, Cultureand the EvolutionaryProcess, 25; "How MicroprocessesGive
Rise to History,"203.

Thoughthey wager the validity of theirtheories on their predictivecapacities,

theoristsof culturalevolutionrig the explanatorygame in a varietyof ways. One
is by covering all bets. This can be done by playing with probabilisticexplana-
tions. In the gamblinghall, probabilitiesonly provide the odds, but probabilistic
predictionsof culturalevolution are guaranteedwinners, since they encompass
all possibilities. Because, for example, of our evolved capacity to reason we
could be soberly advancingdown the road towardswisdom, courage, and com-
passion;or because of our innatecapacityfor aggressionwe could be headin'for
nucleararmageddon-or anythingin between. Or it can be done by constructing
a historicalanalog to randomdriftin theoriesof biological evolution-the catch-
all explanationof that which cannot be subsumedunderselection.
A second way to rig the game is with postdictive readjustment.The cultural
evolutionist, like the economist, is "anexpert who will know tomorrowwhy the
things he predictedyesterdaydidn'thappentoday."39The gambler'slosses might
be recoupedin a later game, but cannot be undone. But in culturalevolutionary
explanation and prediction, the game may be replayed indefinitely until the
model is successfully readjusted. Combined with probabilistic explanations,
postdictive readjustmentrendersthe model invulnerableby disarmingits weak-
nesses.
The irony here is thatthe constantrecourseto postdictivereadjustmentsbrings
the science of culturalevolution into the neighborhoodof "justplain history"-
almost. The differenceis that the faith in the scientific statusof the law of selec-
tion erects a third safeguardfor theories of culturalevolution. This belief pre-
cludes as "not scientific"any non-evolutionary,thatis, historical,explanationof
culturalchange. But because culturalevolutionarytheories are based on a uni-
tary,transhistoricalprinciple,they produceexplanationsthat are too broadto be
eitherfalsifiable or explanatory.
Historians,culturalevolutionistsargue,are too close to the fray,and theirtime
scales too short-which leads them into all kinds of unimportantdetours and
false starts that appear to the historical eye as enterprises of great pith and
moment.To gain properperspective,therefore,culturalevolutionists drawback,
occasionally indulgingin imaginaryspace travel, in orderto attaina sufficiently
distantviewpointfrom which to view the humanspecies as one amongmany and
to avoid the "anthropocentrism" that would exempt culture(a biological adapta-
tion) from biological explanation.But distancecan also be deceiving.
Fromtheirdistantviewpoint culturalevolutionistswillingly see only the broad
patternsof culturalevolution, and ignore the inconvenientand contingentdetails
of history that do not fit into those patterns.This conscious oversight produces
theories of culturalevolution that are explicitly or implicitly progressivist:since
cultureis a successful andcumulativeadaptationthatbreaksfree of naturalselec-
tion, the more culture, the better for human welfare and survival. This linear
logic points to the contemporaryWest with the most advancedlevel of science
39. LaurenceJ. Peter,Peter's Quotations:Ideasfor Our Times(New York, 1977), 477.

and technology (the ultimate cultural adaptationsinsuring human welfare and

survival) as the currentpinnacle of culturalevolution. But the road to modern
Westerncivilization has taken a series of abruptand thoroughlyunpredictable
turns.Whatgeneraltheoryof culturalevolution could postdictthe collapse of the
Roman Empire and the "DarkAges"? Or the emergence on a distantfrontierof
the Eurasianlandmassof a new geo-culturalentity, a "continent"called Europe?
Or thatin a very brief historicaltime span this new culturewould overtakemuch
more advancedAsian culturesand establishitself as the most powerfuland dom-
inant in the world, with one of its tiny "populations,"the English, having
acquiredan empire on which the sun never set? But the result of all those unpre-
dictable turns, the late modern West, which should be the pinnacle of cultural
evolution,has been the epitome of barbarism(which only a small groupoffin de
siecle artistsand intellectuals,membersof the "literaryculture,"suspected).
From their distant viewpoint, cultural evolutionists may ignore acts*of bar-
barism in Westernhistory like the genocide of Native Americans or the Nazi
Holocaust as just specks of dust on the plain of history,momentaryaberrations
irrelevantto the question of culturalevolution. Alternatively,they may subject
both to the same explanatoryprincipleas just two examples of humanaggression
explainedthroughsome selectionist variationor combinationof inclusive fitness,
innate aggression,the stress of overpopulation,and/orthe need for Lebensraum.
But to explain the character,causes, and consequences of these two forms of
genocide according to the same transhistoricalprinciple would lead to a gross
misunderstandingof each and would tell us little about their historically and
politically significant differences. Such an approach, for example, is far too
broad either to postdict the success of Nazism or to predict the ongoing conse-
quences of the Nazi period, of the historicalmemorythatcontinues to affect sig-
nificantlythe history not only of Germanyand Europe, but also of the Middle
East. Whether they forcibly subsume disparatehistorical phenomena under a
transhistoricalexplanatoryprinciple or write off as mere contingencies histori-
cally significantevents that cannot be so subsumed,culturalevolutionarytheo-
ries cannot answerthe many crucial questions pertainingto the particularity,the
uniqueness,of all historicalphenomena.In failing to live up to their own claims
to be able to explain history, including that of our own era, "more deeply and
more rigorously,"culturalevolutionarytheories also fail to live up to their fur-
ther claim to explain history more "usefully"-to explain Nazism, for example,
with sufficientprecision to prevent its reoccurrenceand to develop appropriate
policies to deal with its consequences.
It is thereforeno use to fall back on yet anothersafeguard,the claim that the
field is still young, the models are still being built, and one day.... The problem
is more serious than "not yet enough time." Culturalevolutionarytheories are
carefully constructed,logically consistent, and very neat. Their neatness, how-
ever, is achieved eitherby dismissing as inessentialto culturalevolution the con-
tingencies that are so essential to historical change or by subsuming them to a
single transhistoricalprinciple of explanation. But this formulaic treatmentis

fully inappropriateto the labyrinthinepathways,the contingent complexity, the

many nuances,and generalmessiness of history.And it results in linearexplana-
tions thatapproachclosely enough to historyto allow the distantobserverto mis-
take proximity for causality.These analyticallines, however, are actually false
tangents-briefly nearing,but never touching, the contoursof history.
We conclude, finally,by returningto the question of whetherany useful work
is done by consideringculturalevolution as distinct from the history of human
societies. Transformational theoriesof culturalevolutionhave the virtuethatthey
at least provide a frameworkof generalitywith which to give humanlong-term
history the semblance of intelligibility. But the search for intelligibility should
not be confused with the search for actual process. There is no end of ways to
make history seem orderly.Variationalisomorphismswith Darwinianevolution
suffer from the inverse problem. Ratherthan being so flexible as to accommo-
date any historical sequence, they are too rigid in structureto be even plausible.
They attemptto mimic, for no reason beyond the desire to appearscientific, a
theoryfrom anotherdomain,a theorywhose structureis anchoredin the concrete
particularitiesof the phenomenathat gave rise to it.
University of Oregon (JOSEPHFRACCHIA)

Harvard University (R. C. LEWONTIN)


Does Culture Evolve (Fracchia & Lewontin, 1999)

Uploaded by

Copyright:

Available Formats

Does Culture Evolve (Fracchia & Lewontin, 1999)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Does Culture Evolve (Fracchia & Lewontin, 1999)

Uploaded by

Copyright:

Available Formats

Wesleyan University

Does Culture Evolve?

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

The drive to describe culturalhistory as an evolutionaryprocess has two sources. One

I. CULTURE,THE TWO CULTURES,AND HISTORY

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

In his Preface to the manifesto of culturalevolution redivivus,Evolutionand

But why does White insist, illogically and counterfactually,that a denial of

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

later in On Human Nature, the appropriateinstrumentfor closing the "famous

Rosenbergexpects this to be rectified as soon as the social sciences are treat-

6. E. 0. Wilson, On HumanNature (CambridgeMass., 1978), x.

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

terized by generativelaws or mechanisms whose operationsproduce the actual

15. Ibid., chapters3:3, 3:46, 3:11.

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

ry explains the evolution of life as a whole as a mechanicalconsequence of the

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

We begin with differentassumptionsabout historical objects and, according-

II. THE FORMS OF EVOLUTIONARYTHEORY

Models of the evolution of phenomenaare traditionallymodels of the temporal

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

The alternative,invented by Darwin to explain organic evolution, is a varia-

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

arresteddevelopment, the failure to pass on to the next stage. Freudian and

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

III.A. THE PARADIGMSOF CULTURALEVOLUTIONARYTHEORY:

A remarkablefeatureof the historyof attemptsto createa theoryof culturalevo-

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

sificationswithin a level of progress,symptomaticof "specificevolution"(mere

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

parts,more specializationof partsand more effective integrationand, subserving

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

III.B. THE PARADIGMSOF CULTURALEVOLUTIONARYTHEORY:

Variationalmodels for culturalevolution have appearedin the last twenty years

21 . E. 0. Wilson, "HumanDecency Is Animal,"New YorkTimesMagazine (October 12, 1975).

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

and psychology that were in search of more "scientific"explanatoryschemes.

22. On the basis of the two questionshe asks of each culturalevolutionarytheory,WilliamDurham

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

turalevolution tended to be "models without dual inheritance."These conceptualizeculture not "as

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

With society and culturereducedto mere aggregatesand deprivedof any sys-

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

Some dissenters present serious challenges to the inheritance/transmission

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

culturalevolution. This is effected throughthe dissolutionof social systems with

36. Rosenberg,Sociobiology and the Preemptionof Social Science, 151.

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

Thoughthey wager the validity of theirtheories on their predictivecapacities,

39. LaurenceJ. Peter,Peter's Quotations:Ideasfor Our Times(New York, 1977), 477.

This content downloaded from 185.44.78.113 on Fri, 20 Jun 2014 23:42:14 PM

and technology (the ultimate cultural adaptationsinsuring human welfare and