Theory of Cultural and Social Selection
Theory of Cultural and Social Selection
T h e T h eory of C u lt u r a l
a n d so c i a l se l ect ion
W. G . Ru nc i m a n
Trinity College,
Cambridge
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore,
São Paulo, Delhi, Dubai, Tokyo
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521199513
© W. G. Runciman 2009
v
vi Contents
Telling it like it felt like 205
References 225
Index 244
Preface
The aim of this book is to bring the concepts and methods of current
evolutionary theory to bear on the agenda of comparative sociology – that
is, on the study of the underlying process by which the human cultures
and societies documented in the historical, ethnographic, and archaeo-
logical record come to be of the different kinds that they are. Its prin-
cipal contention is that collective human behaviour-patterns should be
analysed as the outwardly observable expression of information affecting
phenotype transmitted at three separate but interacting levels of heritable
variation and competitive selection – biological, cultural, and social. A
neo-Darwinian approach of this kind is by now commonplace in many
areas of the more specialized behavioural sciences. But its potential con-
tribution to sociology has hardly begun to be realized.
Since a preliminary formulation of a recognizably selectionist sociology
already underlay the trilogy on social theory which I published between
1983 and 1997, this volume might be thought to be no more than a post-
script to views which I have already put into print. But my Treatise on
Social Theory was, as I now recognize, little more than an introductory
exposition of its theme. Since it was published, I have not only become
increasingly aware of its deficiencies and increasingly indebted to the
many other authors on whose contributions to selectionist theory it is now
possible to draw, but increasingly conscious of how much still remains to
be done before sociology can be said to have moved decisively beyond
the agenda set for it by Herbert Spencer, Karl Marx, Max Weber, and
Emile Durkheim. What has not changed as much as it might is the hos-
tility still provoked not only among sociologists but among anthropolo-
gists and historians by the extension of the concepts and methods which
have proved so successful in the explanation of the evolution of species to
the explanation of the evolution of human cultures and societies which is
directly continuous with it. Much of this hostility is based on misunder-
standing, whether wilful or not, and some of it may reflect a deep-seated
vii
viii Preface
unwillingness to acknowledge how much of human, as of natural, history
is due to chance. But if this book can help to diminish the antagonism
which neo-Darwinism still attracts, and thereby encourage sociologists
younger and better qualified than myself to carry forward the research
programme to which it points the way, its publication will have been fully
justified.
My thanks are due to John Davis, who as Warden of All Souls arranged
for me to give a series of seminars on these topics at Oxford in the aca-
demic year 2002–3; to an anonymous referee for constructive comment
on, and criticism of, the penultimate draft; to Ian Hacking for pointing
out the weaknesses of an earlier version of the Epilogue; to Richard Fisher
of Cambridge University Press for his encouragement and help in seeing
the book through to publication; to Zeb Korycinska for compiling the
index; and to Hilary Edwards for many years of indispensable secretarial
assistance.
‘T h e s i ng l e be s t i de a …’
1
When Karl Marx wrote to Ferdinand Lassalle in January 1861 that
On the Origin of Species had furnished him with a natural-scientific ground-
ing for the class struggle in history, he could hardly have guessed the extent
to which, by the end of the following century, Darwin’s theory of natural
selection would have been triumphantly vindicated but his own theory of
class struggle falsified by subsequent events. It was not as if Darwin suc-
ceeded in his lifetime in meeting the objections to the theory of natural
selection which his critics levelled against it. Nor could he, since he had no
possible way of knowing how biological inheritance actually works. The
irony on which many of his biographers and commentators have remarked
is that he could have read Mendel’s subsequently world-famous paper
about cross-pollinated garden peas. But, even if he had, it would not have
given him all the answers he needed. Decades later, self-styled Mendelians
could still be anti-Darwinians. It was only when the ‘new synthesis’, as it
came to be called, brought together evolutionary theory and population
genetics that it became possible to resolve the seeming conundrum that if
biological inheritance was blended selection couldn’t operate at all (which
in fact it can), while if it was particulate it couldn’t operate in the way
that Darwin supposed (which in fact it does). Still less could Darwin have
anticipated the discoveries of molecular biology which made it possible to
integrate evolutionary theory with an understanding of how information
affecting phenotype through the construction of proteins is transmitted
from one organism to another by being encoded in strings of DNA, or
the developments in statistics, game theory, and computer science which
practising behavioural scientists now take for granted.
1
2 The theory of cultural and social selection
It might therefore seem that to comparative sociologists of the twenty-
first century seeking to explain the evolution of distinctive patterns of
collective human behaviour, Darwin’s own writings can be of no more
than antiquarian interest. But nobody who has read The Descent of Man as
well as On the Origin of Species can continue to think so. Darwin’s presci-
ence is all the more remarkable in the light of what he didn’t and couldn’t
know not only about population genetics and molecular biology but also
about primate ethology, palaeoanthropology, archaeology, linguistics,
demography, and developmental and cognitive psychology. Sexual selec-
tion, which, except for its treatment by R. A. Fisher in one early paper
and then his classic The Genetical Theory of Natural Selection of 1930, was
almost completely ignored for a century after Darwin’s death, was to
become one of the fastest-growing areas of evolutionary biology (Miller
1998) and to bear directly on issues central to feminist theory (Campbell
1999); Darwin’s observations of animals’ capacity for deception and
manipulation were to be fully borne out in studies of the ‘Machiavellian’
intelligence which we share with other species (Byrne and Whiten 1988;
Whiten and Byrne 1997); his view of linguistic ability as ‘an instinctive
tendency to acquire an art’ directly anticipates the late twentieth-century
alliance of traditional linguistics with developmental psychology and
neuroscience (Pinker 1994: 20); his discussion of the derivation, as he put
it, of the ‘so-called moral sense’ from the ‘social instincts’ anticipates the
whole modern literature on altruism and the evolution of cooperation
between unrelated conspecifics in both animal and human populations;
and he was as aware as any present-day theorist of cultural group selection
of the possibility that selective pressure can operate not only on compet-
ing individuals but also on competing groups.
It has often been remarked that the idea of variation was no more ori-
ginal to Darwin than the idea of selection. But it was by combining them
as he did that he took the teleology out of evolutionary theory altogether.
It is because what he called ‘descent with modification’ – hereafter ‘her-
itable variation and competitive selection’ – resolves the age-old problem
of qualitative change without recourse to either special creation or a pre-
determined goal that the philosopher Daniel Dennett can call Darwin’s
‘the single best idea anyone has ever had’ (Dennett 1995: 21). The psycholo-
gist Donald T. Campbell, who was perhaps the first behavioural scientist
to appreciate the implications of natural selection being only one among
other special cases of a more general evolutionary theory, aptly credited the
Darwinian paradigm with providing what Campbell called ‘the universal
nonteleological explanation of teleological achievements’ (1974: 420). Not
The Darwinian legacy 3
only does natural selection explain more about human behaviour than the
overwhelming majority of twentieth-century sociologists were willing to
concede, but the heritable variation and competitive selection of informa-
tion which affects behaviour in the phenotype is a process which operates
also at both the cultural level, where the information is encoded in memes –
that is, items or packages of information transmitted from mind to mind
by imitation or learning – and the social level, where it is encoded in rule-
governed practices which define mutually interacting institutional roles. The
definition of ‘institution’ in the literature of sociology has been almost as
troublesome as that of ‘culture’ and ‘society’ themselves, and some sociolo-
gists talk about ‘social practices’ when they have informal, interpersonally
acquired behaviour-patterns as well as, or instead of, formal, institution-
ally imposed ones in mind. But a categorical distinction has to be drawn
between behaviour regulated by formal inducements or sanctions attaching
to extra-familial economic, ideological, or political roles designated as such
and behaviour regulated by informal habits and conventions. The mecha-
nisms of heritable variation and competitive selection are quite different
in biological, cultural, and social evolution. No less important than the
recognition that natural selection cannot by itself account for the diver-
sity of collective human behaviour-patterns is the recognition that cultural
and social selection, which have too often been assimilated (including by
Campbell) under the rubric of ‘sociocultural evolution’, are, as subsequent
chapters will abundantly demonstrate, not at all the same thing.1 There
are not two but three levels at which evolution drives human populations
down the open-ended, path-dependent trajectories which continue to gen-
erate new patterns of collective behaviour out of old.
2
The purpose of this book is not so much to defend selectionist theory
against its critics as to suggest how the agenda of comparative sociology
should be reconstructed in its terms. Yet anti-Darwinism is, to this day,
as much a part of the Darwinian legacy as neo-Darwinism is. It may
no longer be fuelled by the righteous indignation of pious Victorians
As pointed out by Brown (1991: 40), anthropologists have been particularly prone not only to
1
contrast ‘cultural’ and ‘social’ jointly with ‘biological’ but to treat ‘a culture’ and ‘a society’ as
synonymous. This is understandable where, as in much of the ethnographic record, cultural and
social boundaries coincide. But sociologists will seldom if ever find an exact fit if they map the
distribution of shared memes onto the distribution of interacting practices across a designated
institutional catchment area.
4 The theory of cultural and social selection
determined to remain on the side of the angels when confronted with
the distasteful suggestion that we are all descended from apes. But there
are still many people, sociologists included, for whom ‘neo-Darwinism’
implies either the social-Darwinian racism of the late-nineteenth century
or the reductionist sociobiology of the late-twentieth. They may agree
that the theory of natural selection is no more likely to be discarded in
favour of Archdeacon Paley’s theology than present-day physical theory
to be discarded in favour of Aristotle’s. But they are as resistant as any
pious Victorian to the application of Darwin’s fundamental insight to the
behaviour of creatures with minds like ours. They may no longer believe
that we are distinguished from our primate ancestors by our possession of
immortal souls. But they are not persuaded that the conduct of self-con-
scious human beings who actively choose between alternative courses of
purposive action can be explained within a paradigm which was inspired
by, and should be restricted to, the behaviour of creatures guided only by
perception and instinct.
To this, there are two related answers. The first is that, as Darwin well
knew, purposive decision-making is not unique to humans. The second
is that there is nothing about purposive decision-making which removes
it, and the behaviour resulting from it, from the possibility of selection-
ist explanation. Darwin himself was, in his own words, ‘very far from
wishing to deny that instinctive actions may lose their fixed and untaught
character, and be replaced by others performed with the aid of the free
will’ (1882: 66). Nowhere in this book will it be implied, let alone asserted,
that there is no such thing as choice between alternative possible courses
of action. It is a truism of sociology that social relationships are created,
interpreted, and negotiated by the interacting individuals whose relation-
ships they are. But purposive decision-making, whether by human beings
or the members of any other large-brained species, is a natural, not a
supernatural, phenomenon. We need to explain without recourse to what
Dennett calls ‘skyhooks’ how the behaviour of organisms with minds
and therewith purposes affects the course of cultural and social evolu-
tion. Behind all individual decisions there are detectable genetic, cultural,
and social influences which have guided them, and there are detectable
mechanisms through which they have played their part in the evolution
of the collective behaviour-patterns which distinguish one kind of culture
or society from another.
The implication for sociology of Darwin’s fundamental insight is not
that heritable variation in culturally or socially transmitted information
affecting phenotype is random in the same sense that genetic drift is
The Darwinian legacy 5
random. It is that the causes of variation in information affecting behav-
iour in the phenotype cannot by themselves provide the explanation of
what turn out to be its consequences. To cite a familiar topic to which I
shall return in Chapter 3, the causes of variation among different human
populations in their shared attitudes to, and beliefs about, what Darwin
called ‘unseen spiritual agencies’ are at the same time readily ascertain-
able and extremely diverse: the exemplary preacher, the hallucinated
visionary, the charismatic prophet, the vengeful moralist, the shamanis-
tic interpreter of dreams, the world-abjuring hermit, the messianic cult-
leader, and the author or reinterpreter of holy writ may be acting under
any number of different influences, giving expression to any number of
idiosyncracies of character, and seeking to realize any number of practi-
cal as well as spiritual aims. But the success or failure of their teachings
depends not on how their own mental states have come to be what they
are but on the features of their environment which do or don’t favour the
reproduction and diffusion of the memes which they transmit to their
disciples by imitation or learning and then, if things go well, propagate
through institutional roles like those of priest and schoolteacher in which
successive incumbents replace one another independently of purely per-
sonal relationships.
That is very different from claiming that innovators’ aims have no part
whatever to play in the explanation of the ongoing course of cultural
and social evolution. Innovators often act with deliberation and design
with foresight. Suppose, to remain with the same example, that a cyni-
cal evangelist has conducted careful market research before formulating a
doctrine then spread by well-incentivized acolytes among a gullible pop-
ulation of prospective converts. This may not be the way that visionar-
ies, preachers, and prophets usually proceed. But in such fields of human
endeavour as engineering or medicine or architecture – to say nothing
of advertising – innovations are launched, tested, and modified (or dis-
carded) as part of a deliberate strategy of competitive selection directed
to finding and then exploiting the fittest mutations. In that sense, creative
innovation is the antithesis of random variation, even where randomiza-
tion is itself, as it sometimes is, a deliberate strategy. But Campbell, who
in an early paper argued for what he called ‘blind variation and selective
retention’ (Campbell 1960), meant by ‘blind’ not that engineers, doctors,
or architects don’t know what they’re doing, or that there is no distinction
to be drawn between the winnowing of alternative designs by personal
choice and their winnowing by external agency, but that would-be inno-
vators cannot anticipate the consequences of their own or other people’s
6 The theory of cultural and social selection
discoveries until they have made them. On this point, neo-Darwinians
can wholeheartedly agree with a twentieth-century philosopher resolutely
hostile to the very idea of social science who quoted with approval the
remark of Humphrey Lyttelton that if he knew where jazz was going he’d
be playing it already (Winch 1958: 94). The human capacity for active
pursuit of innovation doesn’t remove the mutations and combinations
of information affecting phenotype which result from it into a creation-
ist world beyond the reach of selectionist theory. That, in turn, does not
mean that comparative sociologists can answer the questions which con-
cern them by directly applying the models which have served the theory
of natural selection so well to the very different, although in some ways
analogous, mechanisms by which cultural and social evolution are driven.
But it does mean that they can profitably put the idea of heritable vari-
ation and competitive selection to use in the service of the ‘just what is
going on here?’ approach which the economist Robert M. Solow sees as
distinguishing biological from physical science (1997: 56–7) and recom-
mends to human behavioural scientists in consequence.
3
There is, however, another view of the Darwinian legacy which is per-
haps more insidious because it is held with equal conviction by anti-
Darwinians on the one side and ultra-Darwinians on the other. It is that
to the extent that Darwinian theory can be applied to the evolution of
human cultures and societies, it can only be as the theory of natural selec-
tion itself. To both anti- and ultra-Darwinians, a Darwinian sociology is
either applied biology or it is not Darwinian. To bring the paradigmatic
conception of heritable variation and competitive selection and its associ-
ated terminology to bear on human cultural and social behaviour as such
is, on both their views, a merely metaphorical exercise.
The answer is again twofold. In the first place, much of the vocabulary of
science is metaphorical and none the worse for that: ought we to stop talk-
ing of electricity as a ‘current’ because it isn’t in fact a liquid? In the second,
the objection has force only where demonstrably metaphorical terms are
used to disguise the weaknesses of propositions which, when construed lit-
erally, can be shown not to be as well validated as the metaphor implies
that they are. Thus, to say that cultures evolve through the heritable vari-
ation and competitive selection of information affecting phenotype which
has been transmitted from mind to mind by imitation or learning would be
illegitimately metaphorical either if there were no such information affecting
The Darwinian legacy 7
phenotype or if the information transmitted was always reproduced in the
receiving mind without any possibility of mutation. But the information
is not a metaphor. It is the reality. There is no other thing for which it
is standing proxy. Nor does the transmission of information have to mir-
ror the genetic analogy directly. Both ‘mutation’ and ‘combination’ can
have as precise a meaning to comparative sociologists studying the repro-
duction and diffusion of memes and practices as to computer scientists
splicing codes for programs and crossing them over in order to see what
happens next. It is the language in which historians’ accounts of cultural
and social change are conventionally narrated which is metaphorical: new
ideas do not literally ‘march forward’ or ‘take off’ or ‘surmount barriers’, as
they do in countless history books, any more than rebellions against the
institutional status quo are literally ‘ignited’ or ‘defused’ or ‘undermined’
in countless others.2 ‘Fall of the Roman Empire’ is as much a metaphor
as ‘changing climate of opinion’, or ‘bourgeois ascendancy’, or ‘industrial
revolution’, or ‘seeds of popular discontent’, or ‘wave of protest’. The two
most famous metaphors in the literature of sociology – Marx’s ‘base and
superstructure’ and Weber’s ‘elective affinities’ – have been as misleading as
they have been influential. However difficult it may be to explain why some
items or complexes of information affecting phenotype are favoured over
time by competitive selection where others are not, it is a matter of literal
fact that in cultural evolution some mutations or combinations of informa-
tion transmitted by imitation or learning are more successfully reproduced
and diffused in adjacent or successive populations of carriers than others,
just as in social evolution are some mutations or combinations of informa-
tion encoded in practices defining institutional roles.
At this point, the alliance between the anti-Darwinians and the ultra-
Darwinians breaks down, since they have wholly incompatible opinions
about how it comes about that the collective behaviour-patterns observed
When, for example, a historian of Britain tells how in the late seventeenth century ‘high culture
2
moved out of the narrow confines of the court into diverse spaces in London. It slipped out of
palaces and into coffee houses, reading societies, debating clubs, assembly rooms, galleries and
concert halls; ceasing to be the handmaiden of royal politics, it became the partner of com-
merce’ (Brewer 1997: 3), what he is literally narrating is the diffusion of competitively selected
memes previously carried within aristocratic status-groups into environments favourable to their
probability of further reproduction. And when another (Blanning 2002: 15) says that ‘The public
sphere was a neutral vessel, carrying a diversity of social groups and ideologies. Depending on the
date of its journey, its carrier is usually labelled ‘scientific revolution’, the ‘crisis of European con-
science’, or the ‘Enlightenment’. These, together with plenty of others, do not have to be thrown
overboard, but the argument will be advanced here that room needs also to be found for freight
with a less modern or progressive appearance’, his metaphorical way of putting it cries out –
metaphorically speaking – to be cashed in the literal language of selectionist theory.
8 The theory of cultural and social selection
in different human cultures and societies are what they are observed to be.
To the ultra-Darwinians, cultural and social selection must ultimately be
reducible to natural selection: mutations in heritable information, however
transmitted, are fit and hence selected if and only if they somehow maxi-
mize the inclusive reproductive fitness of the organisms carrying them.
But to the anti-Darwinians, this presumption is even more unpalatable
than the suggestion that cultural and social selection, although not reduc-
ible to natural selection, are significantly analogous to it: for them, human
cultures and societies can only be explained in their own terms without
reference to any theory, model, or analogy drawn from biology. But on
any serious reading of the literature of comparative sociology, it is as obvi-
ous that not all cultural and social variation can be explained by natural
selection as it is that there are many aspects of human cultural and social
behaviour which natural selection does explain (however reluctant the
anti-Darwinians may be to acknowledge it). Some behavioural ecologists,
when confronted with as challenging a counterexample as a creed which
enjoins monastic and clerical celibacy, will go to extravagant lengths to
find some hypothesis about inclusive reproductive fitness which is consist-
ent with the data, just as some cultural anthropologists, when confronted
with as challenging a counterexample as the universally higher probability
of young adult males engaging in lethal violence than either older males
or coeval females, will go to extravagant lengths to find some hypothesis
which will link the data to presumptively autonomous cultural variables.
But once Campbell’s general point is taken, the question to be addressed
is not which mechanism of heritable variation and competitive selection
of information affecting phenotype is at work, but how those which are
simultaneously at work relate to each other. At the three different levels
of natural, cultural, and social selection, there are three different types of
behaviour: evoked behaviour, where the agent is responding directly and
instinctively to some feature of the environment; acquired behaviour,
where the agent is imitating or has learned from some other agent, whether
directly or indirectly; and imposed behaviour, where the agent is perform-
ing a social role underwritten by institutional inducements and sanctions.
To cite another familiar topic – warfare: a comparative sociologist who
studies it is at the same time studying the evoked behaviour of young
adult males genetically predisposed to initiate or respond to violence under
arousal or provocation, the acquired behaviour of members of cultures in
which violence on behalf of the in-group is positively valued and success-
ful warriors admired, and the imposed behaviour of recruits into their
societies’ military roles in which they are subject to formal punishments
The Darwinian legacy 9
for disobedience or desertion whatever the memes acquired by imitation
or learning which they are carrying inside their heads.
Darwin’s legacy to comparative sociology, therefore, carries an injunc-
tion to reanalyse the evidence of the archaeological, ethnographic, and
historical record in terms of heritable variation and competitive selection
of information affecting phenotype in something of the same way that
Darwin himself reanalysed evidence reported by ‘pre-Darwinian’ obser-
vers of animal species. His own occasional sociological asides are, admit-
tedly, not very felicitous. No present-day sociologist will gain much from
being told that ‘The Greeks may have retrograded from a want of coher-
ence between many small states, from the small size of their whole coun-
try, from the practice of slavery, or from extreme sensuality’ (1882: 141).3
But the neo-Darwinian sociologist’s objective is in principle no differ-
ent from that of the neo-Darwinian biologist as set out by Francis Crick
(1988: 139): ‘To produce a really good biological theory one must try to see
through the clutter produced by evolution to the basic mechanisms lying
beneath them, realizing that they are likely to be overlaid by other, sec-
ondary mechanisms.’ In comparative sociology, Crick’s ‘clutter’ is ‘noise’
in its information-theoretic sense: sociologists have somehow to identify
and trace from among all the enormous amounts of information affect-
ing behaviour in the phenotype the heritably variable and competitively
selected memes and practices without which the distinctive cultures and
societies documented in the archaeological, ethnographic, and historical
record would not have evolved into being what they are.
‘A n y a n i m a l w h at e v e r …’
4
Like Hume, to whom nothing was (in his own words) ‘more evident’ than
that ‘beasts’ are endowed with thought and reason as human beings are,
and Aristotle, who regularly calls animals phronimoi (i.e. endowed with
practical reasoning), Darwin had no hesitation in attributing to both pri-
mates and domestic animals a range of both intellectual and emotional
capacities shared with ourselves. Indeed he credits animals with jealousy,
gratitude, emulation, vengefulness, shame, curiosity, deliberation, mem-
ory, association of ideas, imagination, wonder, and even the rudiments of
A very different answer to this longstanding question is put forward in tentatively selectionist
3
5
Once, however, it is accepted by comparative sociologists that they both
can and should ‘get inside the natives’ heads’, as the saying goes, with-
out compromising their academic credentials in the process, they are con-
fronted with the long-standing question of just how far inside the natives’
heads it is necessary or possible to go. The question is no less relevant to
the study of behaviour from within the neo-Darwinian paradigm than it
is from within any other. If all behaviour is a phenotypic effect of infor-
mation internalized by its carriers, does or doesn’t a sociologist studying
12 The theory of cultural and social selection
a distinctive behaviour-pattern have to be able to say exactly how that
information is subjectively understood by ‘them’? Darwin was well aware
that empathy has its limits: ‘who’, as he asks, ‘can say what cows feel,
when they surround or stare intensely at a dying or dead companion?’
(1882: 201), thus implicitly anticipating the disdain which Evans-Pritchard
was later to express, in the context of anthropological fieldwork, for what
he called ‘if-I-were-a-horse’ arguments. But any systematic and diligent
observer of other people’s behaviour will be able not only to report what it
is they are doing but also to describe, at least to some extent, ‘how it feels’
or ‘what it is like’ to be one of ‘them’.4 Evans-Pritchard denied that it was
possible for an Oxford don to ‘work himself into’ the mind of a serf of the
time of Louis the Pious (1962: 58). But Oxford dons who have mastered
the documentary sources for the reign of King Louis do have a better
idea than other people of what was going on in the minds of the serfs
of that time, just as Evans-Pritchard had a better idea than people who
had not lived as he had done in the 1930s among the Azande and Nuer of
what was going on in theirs. Sociologists seeking to decode complexes of
culturally and socially transmitted information whose phenotypic effects
they observe, even if they make no claim to be reliving in their own
imaginations the natives’ inner mental states, can and do both ascertain
and distinguish between the natives’ intentions and their motives.
This applies, moreover, no less to chimpanzees who lack the means to
put their intentions and motives into words than it does to Oxford dons
who debate their own and other people’s intentions and motives with a
fluency and subtlety which encompass both their intentions in, and their
motives for, doing so. Under the right observational conditions, Darwin,
if taken to the Arnhem Zoo, could tell readily enough whether Luit
has come into violent physical contact with Yeroen with the intention
of inflicting physical harm and not merely by accident, and if with the
intention of inflicting physical harm, whether from a motive of revenge or
of unprovoked aggression. It is true that these terms are notoriously dif-
ficult to define precisely. But sociologists have no need to join – although
all may, and some do – the philosophical debates which surround the
notions of intentionality and agency as such. They need only to recognize
that intentions are constitutive of actions (what are the natives doing in
making the sounds and gestures observed?), whereas motives are causes
of actions (why are the natives performing the actions constituted by
‘Describe’ is here and throughout used in the sense taken by Runciman (1972) from Toulmin and
4
6
So just how unselfish are the unselfish almsgivers with their ‘so-called
moral sense’? Arguments over this longstanding conundrum have
remained peculiarly intractable from Darwin’s day to this, not least
because it has been an additional matter of dispute whether they are for
psychologists or philosophers to arbitrate. But sociologists will do well to
leave philosophy to the philosophers. Sociology’s concern is with the dif-
ferences of motive between generosity to relatives and friends, unrecog-
nized generosity in which the donor takes pride in remaining anonymous,
generosity in pursuit of a reputation for generosity and the resources to be
able to afford it (the so-called ‘handicap effect’), and generosity inspired
by the hope of future favours in return (including reward after death),
together with the second-order category of ‘strong reciprocators’ (Gintis
2000) or ‘altruistic punishers’ (Boyd et al. 2003) who are motivated by a
desire to punish those who fail to punish the selfish non-cooperators even
at personal cost to themselves. How far such behaviour is to be admired
or not is up to you. Darwin, having remarked that ‘any animal whatever,
endowed with well-marked social instincts … would inevitably acquire
a moral sense of conscience, as soon as its mental power had become
as well, or nearly as well developed, as in a man’, goes on to pose the
‘extreme case’ of human beings ‘reared under precisely the same condi-
tions as hive-bees’: in that event, he says, ‘there can hardly be any doubt
that our unmarried females would, like the worker-bees, think it a sacred
duty to kill their brothers, and mothers would strive to kill their female
daughters; and no one would think of interfering’ (1882: 98–9). Although
this invites the obvious objection that human beings couldn’t be reared
as hive-bees without having ceased to be human beings, it is an effective
rhetorical device for bringing home to readers of the Descent the point
that the altruistic dispositions on which they may pride themselves are
no less than the behaviour of animals, including the insects, a product
of the same long process of heritable variation and competitive selection.
16 The theory of cultural and social selection
The difficulty is rather that Darwin at the same time continues to credit
human beings with what he calls a ‘higher’ standard of morality and to
deny to other species ‘disinterested love for all living creatures, the noblest
attribute of man’ (1882: 126). He even at one point, when discussing recip-
rocal altruism, uses the word ‘low’ for the motive which leads one person
to help another ‘in the expectation of receiving aid in return’ (1882: 132).7
To the extent that he is pointing to the importance of the distinction
between a strategy of unconditional cooperation and the strategies of ‘tit-
for-tat’, ‘generous tit-for-tat’, or ‘win-stay, lose-shift’ in iterated plays of
Prisoner’s Dilemma, he is anticipating a whole field of subsequent game-
theoretic research. But by wording it as he does, he returns to his readers
in one hand the conventional moralizing judgements which he had earlier
taken away in the other.
Perhaps, in Darwin’s case, this was a perlocutionary move to calm
his respectable readers’ fears about his own respectability.8 Or perhaps it
reflects a descriptive interest in the subjective experience of acting, as it
seems to the agent, out of conscience, a sense of duty, and a wish to be
good. But he ought, if he was to be consistent, to have held fast to his
example of the hive-bees. Selectionist sociology is, and cannot but be, neu-
tral between the different moralities whose content it reports and whose
reproductive success (or failure) it seeks to explain. Despite Darwin’s ref-
erences to Kant, his and his neo-Darwinian successors’ ideas about the
evolution through natural selection of the ‘so-called moral sense’ stand
or fall whatever his own or anybody else’s agreement or disagreement
with the arguments expounded in Kant’s Fundamental Principles of the
Metaphysic of Ethics.
T wo fa l s e t r a i l s
7
But if Darwin’s idea was such a good one, why did his legacy not sooner
and more effectively influence sociology for the better? In the event, its
early influence was for the worse – not through any fault of Darwin’s own,
7
Which might be bound up, as for example among the virtuous female patrons of the early
Jesuits, with ‘an evident sense of satisfaction arising from an initiative allocated to them’ (Hufton
2001: 331) as well as the hope of reward in the life to come.
8
As, it has been suggested, he may have sought to make his views about the mental and emotional
capacities of animals more palatable to his readers by singling out domestic dogs as models of
fidelity and obedience rather than concentrating on monkeys and apes (Kroll 1997).
The Darwinian legacy 17
but because of the way in which the idea of evolution was deployed by
the two most influential sociologists of the nineteenth century: Herbert
Spencer and Karl Marx.
For all their differences, Spencer and Marx shared a common concep-
tion of social evolution as leading, in due course, to a state of the world in
which technological advance would ensure universal prosperity and vio-
lent conflict would give way to peaceable cooperation. For Spencer, the
driving force of social evolution was competition between individuals,
whereas for Marx it was conflict between classes. But both believed that
the final outcome of the struggle would be benign. This resolute opti-
mism, however groundless, no doubt increased the attraction of both of
their respective theories to their respective disciples at the opposite ends
of the political spectrum. But both failed equally to draw from Darwin’s
original insight what would have been the appropriate lesson for both of
them. It was not simply that they both retained a teleological notion of
evolution as leading to a pre-determined end-state, or even that they both
conflated their confident predictions of the future of mankind with their
personal values and preferences. They also, albeit for opposite reasons,
diverted the study of cultural and social evolution away from analysis of
the relation between information affecting phenotype, its carriers, and the
environment in the direction of celebration of the winners and dismissal
of the losers in the struggle between (for Spencer) more and less vigorous
and therefore deserving individuals or (for Marx) economic classes whose
triumphs derive from a dialectic of successive antitheses and their subse-
quent resolution.
This is not to say that their respective influence was wholly retrograde.
Much of Spencer’s sociology has helped to make the subject what it is
quite independently of his value-judgements about social policy, includ-
ing his analysis of the feedback between structure and function, his
perception of the part played in social evolution by the unintended con-
sequences of purposive actions, his emphasis on the influence of military
as distinct from administrative leadership, and his recognition of the sim-
ultaneity of trends towards both heterogeneity and integration in social
institutions. Nor can Spencer himself be charged with all the prejudices
of his disciples: he was, as his commentators sometimes forget, a fervent
anti-imperialist. Likewise, there is a sense in which even anti-Marxist
sociologists are Marxists too: nobody attempting a comparative analysis
of human societies and the relations of economic, ideological, and polit-
ical institutions to one another can neglect what Marx called the social
relations of production. But out of Marx’s theory of class struggle there
18 The theory of cultural and social selection
developed a whole erroneous school of thought which came to be labelled
‘Historical Materialism’, and out of Spencer’s theory of individual com-
petition there developed a whole erroneous school of thought which came
to be labelled ‘Social Darwinism’. How could this have happened? How
could both these opposite versions of evolutionary theory have so misin-
terpreted the idea of heritable variation and competitive selection?
8
Since it was Spencer who coined the phrase ‘survival of the fittest’, and
since, for all the differences between Spencer’s idea of evolution and
Darwin’s, Darwin never repudiated it, it is not surprising that Social
Darwinism should be derivable from premises both conceptually and
empirically consistent with ‘descent with modification’. There are, after
all, genetic differences between members of the same population, and
their environment does impose a selective pressure on individuals who do
compete with each other for shares of what are limited resources. But
to get from there to the doctrines of the late nineteenth-century Social
Darwinists requires a series of inferences which are in no way entailed by
the theory of natural selection. If, as it seemed to many of Darwin’s con-
temporaries, there was a clear correlation between inherited behavioural
aptitudes and observable physical differences, it by no means followed
that differential aptitudes are evidence of innate superiority rather than of
the effects of different environments on an evolved anatomy and psych-
ology common to the human species as such; and if, at the same time,
the competition for scarce resources between rival populations was visibly
intensifying, an ethic of ‘rugged individualism’ and ‘strenuous life’ was
by no means the only adaptive response to it. Darwin himself remarked
not once but twice in the Descent (1882: 65, 178) how he was struck, on
the Beagle, by how closely the Fuegians on board ‘resembled us in dispos-
ition and in most of our mental faculties’ and by ‘the many, little, traits of
character showing how similar their minds were to ours’.
The diffusion and reproduction of Social Darwinism, accordingly,
has itself to be given a selectionist explanation. Spencer’s doctrines were
particularly adaptive in the economic, ideological, and political environ-
ment of the United States in the 1880s and ’90s – an environment of rapid
expansion, unbridled competition, aggressive individualism, and ram-
pant plutocracy. When, later on, Social Darwinism came to be criticized
almost as vehemently as it had previously been welcomed, it was equally
easy to point to the changes in that environment which had made it
The Darwinian legacy 19
an altogether less plausible, as well as less appealing, guide to either the
foreign or the domestic policy-makers of a society which had, for the
time being at least, lost some of its previous certainty about its ‘mani-
fest destiny’. But that was not the end of the story. By 1937, when the
American sociologist Talcott Parsons opened his book The Structure of
Social Action by quoting from the historian Crane Brinton the rhetorical
question ‘Who now reads Spencer?’, it seemed that it was. But Spencer
was then rediscovered – with or without acknowledgement – by a new
generation of sociologists under the rubrics of ‘development’ and ‘mod-
ernization’ as the United States increased its economic, ideological, and
political power in the aftermath of the Second World War. Teleology of
a kind which all nineteenth-century sociologists would have recognized
came back into its own.
But this was not, any more than the Social Darwinism of the 1880s and
’90s had been, an authentically Darwinian understanding of heritable
variation and competitive selection. ‘Modernization theory’, of which
more in Chapter 4, extrapolated from a selective narrative of ‘Western’
history a projection of the future of the less ‘developed’ world, and in so
doing surreptitiously invoked the very presupposition which Darwin had
undermined. No doubt the neo-Spencerians were as entitled to speculate
that the other societes in the world would come to resemble the United
States as were the neo-Marxists to speculate that the United States, along
with the other societies in the world, would in due course evolve into
a single-party state and command economy on the model of the Soviet
Union. But once teleology is unsustainable, no culture or society is or
ever can be historically privileged over the rest.
9
Historical Materialism, like Social Darwinism, starts from premises con-
sistent with Darwin’s own. Competition for scarce resources does give
rise to collective as well as individual conflict, and heritable variation in
information affecting forms of social organization does generate different
modes of production. But to get from there to the doctrines of Marxism-
Leninism requires a series of inferences no more entailed by selectionist
theory than is Social Darwinism. There is no support in anything that
Darwin wrote for the idea of evolution as a cumulative series of antago-
nisms in which the conflict between successive pairs of exploiting and
exploited classes culminates in the revolutionary victory of the prole-
tariat. It was not completely illogical for Engels, in his speech at Marx’s
20 The theory of cultural and social selection
graveside, to claim that Marx had ‘discovered the law of development
of human history’ in the same way that Darwin had done for ‘organic
nature’. But nor was it illogical for Darwin, during his own lifetime, to
retain even more reservations about Marx than he did about Spencer.
Marx can no more be held responsible for everything said and done in
his name than Spencer can for everything said and done in his. But Marx
was categorical in his prediction of the overthrow of capitalism, whether
(as he originally foresaw) through the revolution of a fully industrialized
proletariat or (as he later came to think possible) by a revolutionary trans-
formation within a still part-agricultural, part-industrial economy. To that
extent, he can be read as a group-selectionist who saw economic classes as
winners in the competitive struggle where and when their members have
subordinated their individual interests to the collective cause. But neither
Marx’s original theory nor his later conjectures can be reconciled with
heritable variation and competitive selection understood in non-teleolog-
ical Darwinian terms. Marx ignores, or at best grossly underestimates,
the range of observable variation both in the form and the content of
the information affecting phenotypic behaviour and in the environmental
influences which affect its probability of reproduction and diffusion. It is
not just that modes of persuasion and coercion can influence the evolu-
tion of modes of production as well as the other way round, but that even
where similar-looking sequences of social evolution occur they are not the
outcomes of a dialectical process of the kind to which he subscribed.
In one respect, moreover, Marx’s theory of social evolution could be
said to be even more unabashedly teleological than Spencer’s, since Marx
saw the attainment of the predestined goal as being hastened by the
bourgeois converts to Marxism who would help to bring about the pro-
letarian revolution which, as Marxists, they knew to be inevitable. But
neither Spencer nor Marx conceived of competitive selection as acting on
variations in heritable information which, however they come about, are
selected only for their own likelihood of continuing reproduction and fur-
ther diffusion. It is true that Marx’s individual capitalists are not required,
any more than Spencer’s are, to be aware of the harmonious long-term
future to which their short-term self-interested actions will in due course
lead. But paradoxically, it was Max Weber who saw competition between
both individuals and groups continuing indefinitely into an unknowable
future; and he, although explicitly sceptical of the applicability to it of the
notions of adaptation (Angepasstheit) and selection (Auslese), was, as I have
argued elsewhere (Runciman 2001), something of a selectionist in spite of
himself.
The Darwinian legacy 21
‘Ou r f or m e r s e l f -r e spe c t ’
10
The quotation this time is from Nietzsche, in the Genealogy of Morals
(III.25), where he says, echoing Kant, that both the natural and what he
calls the ‘unnatural’ – that is, the human, self-referential sciences – ‘talk
human beings out of their former self-respect’. And so they do. It has
often been complained that science has undermined traditional com-
placencies about what used to be called ‘man’s place in nature’ without
offering anything similarly comforting in their stead. But what does it
mean to charge Darwin, in particular, with having done so? The dis-
tinctive properties of the human mind are no less distinctive for being
the outcome of many millennia of heritable variation and competitive
selection which cumulatively differentiated the brains of our ancestors
from those of their primate cousins. The ‘dignity of man’, to echo a
phrase of Kant’s quoted by Darwin himself, is not made less dignified
because of the length of time it has taken to become such or the nature
of the mechanism which has enabled it to do so. If we see ourselves as
the beauty of the world and paragon of animals, noble in reason, infinite
in faculty, and godlike in apprehension, does it matter what had to hap-
pen to the physiology of our species before Shakespeare could express it
in words like these?
One answer is that for some human beings, at least, it is still a ser-
iously disturbing idea that our minds should be nothing more than an
unusually complex kind of physical system. Although for T. H. Huxley’s
grandson Andrew, joint winner of the Nobel Prize for Physiology in 1963,
‘the possibilities of a physical system as complicated as the brain are so
vast that there is no need for a human being to feel that he is reduced
in stature if his thoughts and feelings are governed by physical events in
his brain cells’ (1983: 15), that is not going to satisfy those who still feel
that, as they are likely to put it, mind is more than matter and people
more than things. Behind such responses is a long history of suspicion
that scientists like the Huxleys are seeking not to explain, but to explain
away, our deep and subtle meditations and conjectures about ourselves
and our relations to the world and to each other. Even the most intelli-
gent of animals lack, as Darwin conceded, our speculative capacities and
our moral and aesthetic sense, and to be told that these are merely events
on a par with sensations of pain or pleasure shared with conscious but not
self-conscious organisms is to be told that they are something other than
22 The theory of cultural and social selection
what many people will say that they know them from their own subject-
ive experience to be.
But suppose you are standing awestruck in contemplation of the beauty
of what you regard as one of the masterpieces of your cultural tradition. Is
the significance to you of your aesthetic experience modified by whatever
explanation you are given of the evolutionary origins of artistic sensibility
as such? Suppose it is indeed the case that sexual selection explains much
more about our taste for what we call ‘beauty’ than has generally been
supposed, and that our remote female ancestors were disproportionately
attracted to potential mates who had the skills to decorate their bodies
and ornament their dwelling-places in original and striking ways. If so,
it follows that we have something closely in common with the bower-
birds whose reproductive success is, as Darwin remarked in Part II of the
Descent (and subsequent research has fully confirmed), directly linked to
both the quantity and the quality of bower-decoration by males. But that
is not going to make any difference to what you feel when you contem-
plate your chosen masterpiece,9 any more than if our aesthetic sense has,
after all, no biological function as such but is an incidental by-product of
the ongoing process of natural selection during the Pleistocene.
There is nothing controversial in the idea that our subjective experi-
ences are causally influenced by both history and environment. As J. S.
Mill put it in his essay On Liberty, the same causes which have made a
man a ‘Churchman’ in London would have made him ‘a Buddhist or
Confucian in Peking’. But whatever those causes may be, the subjective
mental experiences which result can be given a phenomenological descrip-
tion of what it feels like to have them in exactly the same way. It is true
that for people to be told why they believe what they do can sometimes
lead them to change their beliefs, or at least to realize that they need some
other reason to accept them than unthinking deference to their childhood
mentors. To be educated by Mill (or Marx or Weber or Durkheim or any-
one else) in the comparative sociology of religion may cause the London
Churchman to question the Christian doctrines in which he was reared
in a way that he would not otherwise have done. But Darwin undermined
human beings’ self-respect only where, for the human beings in ques-
tion, self-respect depended on retaining a conviction that we are the spe-
cial creations of God destined to fulfil a pre-ordained purpose; and that
The same applies to the ‘beauties of nature’. If it is true that our aesthetic responses to land-
9
scape are to be explained because they made our ancestors more likely to choose habitats which
enhanced their reproductive fitness (Orians and Heerwagen 1992), is that going to diminish your
enjoyment of your favourite view?
The Darwinian legacy 23
conviction has been repudiated both before and since by human beings
who have felt no loss of self-respect in consequence. Indeed, it has some-
times been those who reject any idea of an all-knowing Creator whose
sense of human self-respect has been the stronger on that very account.
11
The Darwinian legacy, therefore, denies neither the meaning nor the
value to you of your awareness of self and the feelings which accompany
it. What it may, on the other hand, do – which is a very different thing –
is make you more aware than you would otherwise be that your subject-
ive mental experiences are a seriously unreliable guide to the explanation
of why you behave as you do. Introspection can, pace the Behaviourists,
tell us quite a lot about ourselves. But it cannot tell us whether our intui-
tively persuasive explanations of our own behaviour, and that of the group
or community to which we belong, are valid. As to freedom of will, it
is undeniable that if any sociologist studying your behaviour is foolish
enough to offer you a prediction of how you will vote, what product you
will buy, whether you will go to work tomorrow morning, or who you
will choose to invite to a meal, you can falsify the prediction from no
other motive than to discomfit the predictor. But once you have decided
to do what you then proceed to do, your behaviour becomes part of the
large-scale collective behaviour-pattern which makes the distinctive cul-
ture and society of which you are a member the culture and society that
they are, and the sociologists who want to explain that pattern can start to
frame and test whatever population-level hypotheses they think will best
do so. If a neo-Darwinian hypothesis stands up better than its rivals, it
may thereby expose a disconcerting discrepancy between the selectionist
answer to the ‘just what is going on here?’ question and what the natives
themselves take the answer to be on the strength of what it feels like to
them to be doing what they do. But that no more invalidates the neo-
Darwinian explanation than the neo-Darwinian explanation disauthenti-
cates the phenomenological description in terms of the natives’ subjective
mental states.
The resulting disjunctions between the underlying process at work and
the experience of change (or the lack of it) as described by those who
live through it may, accordingly, undermine the self-esteem of those who
have persuaded themselves and others that events which they claim to
have predicted and worked for came about because and only because
they did. But once again, selectionist explanation of the evolution of the
24 The theory of cultural and social selection
culture and society within which they held their beliefs and attitudes
and occupied and performed their institutional roles neither underwrites
nor undermines their own accounts of how it felt to them to be doing
what they did. Darwin’s legacy leaves us still inhabiting the conceptual
world within which we describe our own and other people’s inner mental
states – the metaphors and similes, the ‘it seemed as ifs’ and ‘it was as
thoughs’, the adjectival and adverbial embellishments, the appeals to fic-
tion and poetry, the imaginative depictions of loves and hates and joys
and sorrows, and the hermeneutic exploration and discussion of all the
thoughts and feelings which between them make up what is convention-
ally meant when we talk about ‘the human condition’.
12
But Darwin did, all the same, have a lasting impact on one hitherto com-
forting assumption, as Nietzsche did too. It was an unconscious alliance in
the sense that Nietzsche’s own references to Darwin make it clear that he
didn’t understand – or, perhaps, chose not to understand – what the the-
ory of natural selection actually says. But between them they mounted a
powerful joint assault on the faith in human progress which was shared by
so many nineteenth-century Christians and non-Christians alike. Again,
it is not a matter of self-respect – or not unless your sense of self-respect
depends on a sense of playing a part in a historical drama with a guaran-
teed happy ending. But On the Origin of Species and The Descent of Man
give as little encouragement as Beyond Good and Evil and The Genealogy
of Morals to those who would like to believe that despite all the mistakes,
crimes, and sufferings of the past mankind is bound to come through in
the end to a world of universal reason, freedom, and peace. Ironically, it
was not that Darwin’s contemporaries all saw it that way. Just as Spencer
and Marx both managed to reconcile what they took Darwin to be saying
with their own triumphalist predictions of the harmonious future which
would in due course evolve out of the competitive struggle between indi-
viduals (for Spencer) or classes (for Marx), so did many Christians of dif-
ferent denominations, Prime Minister Gladstone included, manage to
reconcile an acceptance of what they saw as the unchallengeable conclu-
sions of the theory of natural selection with their belief in the providence
of God. The widely different reactions of the Protestant churches, in par-
ticular, is a topic on its own. There were probably not many who shared
the view of the Congregationalist minister Myron Adams that ‘that soci-
ety is most fit which is most Christlike’ and ‘the relatively superior, in
The Darwinian legacy 25
respect of mildness and reasonableness, has been gradually supplanting
the inferior and fiercer’ (quoted by Roberts 1988: 182) – a tendency about
which Nietzsche wholeheartedly agreed but which he equally wholeheart-
edly deplored. But once seemingly teleological outcomes have been shown
to be explicable in terms of heritable variation and competitive selection,
the damage has been done. It is always possible to argue, with Archbishop
Temple of Canterbury and others, that natural selection must itself have
been designed by the mind of God. But once God has chosen to design a
non-teleological world, those who are trying to explain what they observe
in the world have to come up with non-teleological explanations of it.
It does not follow that the idea of progress has been deprived of any
meaning whatever, or that Darwin’s contemporaries and compatriots who
saw themselves as living in an ‘Age of Improvement’ were not entitled to
do so. The natives of nineteenth-century Britain could, and did, point to
advances in transportation, technology, sanitation, diet, and real income
per head which meant that by criteria shared by Darwinians with their
opponents the world – or, at any rate, their part of it – was to that extent
getting better all the time. But they had then to recognize the force of the
argument that there might be nothing immutable about those criteria,
and nothing inevitable about the continuation of improvement as defined
by them. Evolution is by definition movement away from a previous state
of a system towards a new and different one. But that is all. For those
who cannot bear to abandon the conviction that competition and conflict
will somehow be reconciled in a peaceable and harmonious future for
the human species, it may not be enough. But as Darwin himself said in
the concluding paragraph of the Descent (1882: 619): ‘we are not here con-
cerned with hopes or fears, but only with the truth so far as our reason
permits us to discern it’.
Ch apter 1
C onc e p t s a n d pr e s u pp o s i t ions
1
For all the influence that neo-Darwinian evolutionary theory has had
over the past few decades across the human behavioural sciences all the
way from economics to archaeology, sociology is still at the stage likely,
in Thomas S. Kuhn’s original words about ‘paradigm-change’, to be char-
acterized by ‘deep debates’ about ‘methods, problems, and standards of
solution’ which ‘serve rather to define schools than produce agreement’
(1962: 48). But its agenda too is being changed in ways that the die-
hards – or ‘hold-outs’, as Kuhn called them – may continue to deplore
but can no longer reverse.
Some of Kuhn’s followers claimed – and he himself sometimes seemed
to imply – that rival paradigms are radically incommensurable. But com-
mitment to a neo-Darwinian conception of heritable variation and com-
petitive selection of information affecting behaviour in the phenotype
doesn’t make it impossible for sociologists who are so committed to talk
to others about ‘just what is going on here’ in terms readily comprehen-
sible to, and translatable salva veritate between, both sides. Confronted
with the same set of uncontested observations of a routine occurrence of
a wage negotiation in a capitalist economy, a Selectionist sees an insti-
tutional practice which has out-competed its rivals where a Marxist sees
exploitation of proletarians by a capitalist class, a Behaviourist sees a his-
tory of operant conditioning, and a Rational Choice Theorist sees a bar-
gain being struck between self-interested utility-maximizers. No amount
of more detailed reportage of the behaviour observed in the particular
case will lead them to discard their respective presuppositions. But they
can all understand each other perfectly well. They can all recognize that
their different ways of seeing the same thing apply to a wider range of
27
28 The theory of cultural and social selection
collective behaviour-patterns than the one instance of wage-bargaining
and that if their different approaches are used to frame and execute alter-
native research programmes, some will in due course generate explana-
tions which fit more closely than those of their rivals with evidence which
they can all check for themselves. There is no implication that the suc-
cesses of even the most successful of the rival programmes can be shown
to be conclusive beyond the possibility of rejection or modification, or
that a single crucial experiment or case-study can decide between them.
But the failures of the least successful do have a way of turning out to be
terminal. It is no more likely that phrenological explanations of behav-
iour culturally defined as criminal will reappear in twenty-first-century
textbooks of sociology than that twenty-first-century textbooks of biol-
ogy will come to include exposition of Bergson’s concept of élan vital.
At the same time, it is not as if the neo-Darwinian paradigm has com-
prehensively displaced its predecessors, or ever will. Even the most extrav-
agant claims for what it can do for comparative sociology will not require
that the previous literature of the subject be consigned to the flames, as
will be readily apparent from the topics chosen for discussion in Chapters
3 and 4. Reconciliation is categorically impossible only if, for example, the
rival research programme is directed to establishing which of the behav-
iour-patterns offensive in the eyes of God cause sinners to be punished in
this world rather than waiting for the next. This is not because theological
explanation of accidents, sicknesses, and misfortunes is without a history
of its own which is as intellectually coherent and copiously documented
as its rivals. When the Rev. Thomas Jackson, in the first half of the sev-
enteenth century, said that ‘Pestilence is above all other diseases catching,
and such as have been most observant of its course tell us men of covetous
minds or unseasonably greedy of gain are usually soonest caught by it,
though exposed to no greater or more apparent visible danger than others
are’ (quoted by Thomas 1971: 88), he could not be accused of advancing
a hypothesis unrelated to empirical evidence. Indeed, he might even be
construed as advancing a hypothesis grounded in a theological version of
selectionist theory. But the eventual abandonment of this kind of provi-
dential epidemiology is a textbook instance of a paradigm-change. Not
only was divine punishment theory overtaken by the findings of medical
research, but it threw up two anomalies of just the kind that bring the
presuppositions underlying a hitherto plausible theory into question: first,
the ‘greedy of gain’ were on balance less, not more, likely to die of pesti-
lence than the poor; and second, those smitten included infant children
incapable of behaviour which under any plausible definition could be
The neo-Darwinian paradigm 29
classified as ‘covetous’. Hypotheses like these are perfectly comprehensible
to rival theorists who reject them.1 It is as clear what ‘sin’ meant to Thomas
Jackson as what ‘descent with modification’ meant to Charles Darwin.
But selectionist sociology excludes providential retribution as an explana-
tory concept as categorically as it does both the dialectical ‘Aufhebung’ of
Historical Materialism and the ‘Manifest Destiny’ of Social Darwinism.
2
The neo-Darwinian sociologist’s short answer to the ‘just what is going on
here?’ question accordingly is: ‘the phenotypic behavioural outcomes at
population level of information transfer.’ Once the collective behaviour-
patterns of the members of any culture or society chosen for study are
traced back to the items or packages of information which are being acted
out in them, the agenda is to ascertain what that information is, where it
is encoded, how (and how accurately) it has been reproduced, and why it
and not another item or package has been selected. For this, there needs to
be found a causal link between the specific memes or practices being acted
out in the behaviour observed and the features of the environment which
have enhanced their relative reproductive fitness; and that in turn depends
on finding the place down the chain of phenotypic effects where selective
pressure has come to bear. Sociologists of all theoretical persuasions are
well aware that unintended consequences can make more difference to
what happens in the world of collective human behaviour than intended
ones. But the extended phenotypic effects of information affecting phe
notype can equally well be either. The populations which comparative
sociologists have to analyse if they are to account for the evolved differ-
ences between cultures and societies are populations of items or packages
of information (including, if they are deontic in logical form, instructions
or strategies), not of the individual minds or the dyads of institutional
role-incumbents that are their carriers. It is then on the carriers interacting
Providential epidemiology has a history which not only goes back, in Europe, to pre-Christian
1
convictions that plagues and other natural disasters are a punishment for offending the gods but
forward to mid-nineteenth-century Britain and the Evangelicals’ doctrine of ‘Atonement’. Hilton
(1988: 155, 113) quotes Thomas Watson’s influential textbook of 1843, Lectures on the Principles and
Practices of Physic, where he says that ‘It is ours to know in how many instances, forming indeed
a vast majority of the whole, bodily suffering and sickness are the natural fruits of evil courses’,
as well as the Rev. Charles Vansittart’s A Sermon on the Famine of 1847, where he says that the
Irish Famine is ‘a visitation – a chastisement for our national sins and crimes – for the grasping,
gambling, monopolizing, covetous spirit, which of late has engrossed the minds of all ranks of
classes and men’.
30 The theory of cultural and social selection
directly with the environment, not the information itself, that selective
pressure presses. That is why they are sometimes referred to not as ‘carri-
ers’ (or alternatively ‘vehicles’) but as ‘interactors’. The process of selection
necessarily involves both the information reproduced and the organisms,
minds, or roles that carry it, and the same entities function as both inter-
actors and replicators. But in the application of selectionist theory, as a
leading evolutionary game theorist has aptly put it, ‘it is the strategies
that come to the fore; the individuals that implement them on various
occasions recede from view’ (Skyrms 1996: 10).
3
All selectionist explanations are ‘just-so stories’ in which the link between
selective pressure and evolutionary outcome is identified only in hind-
sight. The term ‘just-so story’ is commonly used in a pejorative sense,
which is perhaps why some philosophers of science prefer to talk about
‘how possibly’ explanations. But there is nothing inherently wrong with
asking how the elephant got his trunk. Thanks to the theory of natural
selection, the just-so story of how he really did get his trunk can be cor-
rectly told. Sometimes, selectionist hypotheses cannot be conclusively
validated because the evidence necessary to test them against possible
alternatives is unobtainable. Hypotheses grounded in the theory of cul-
tural selection have to confront the inadequacy of the documentary and
archaeological record in the same way that hypotheses grounded in the
theory of natural selection have to confront the inadequacy of the fossil
record, while hypotheses grounded in the theory of social selection have
often to confront the problem of overdetermination by multiple selective
pressures all driving the evolution of a society’s modes of production, per-
suasion, and coercion in the same direction. But the difficulty of finding
the evidence which will conclusively invalidate rival explanations is not an
argument against selectionist theory. The literature of comparative soci-
ology cannot but be full of both more and less plausible just-so stories. But
a just-so story has always to be the right one, whether about how Protestant
Christianity evolved out of Catholic Christianity, or factory out of arti-
san production, or universal suffrage out of oligarchy, or Gothic out of
Romanesque architecture, or electronic out of electrical technology, or
pidgin out of standard English. Selectionist just-so stories combine theor-
etically grounded accounts of the mechanisms of transfer of information
affecting behaviour in the phenotype with narratives of path-dependent
sequences of historical events in which the relatively higher probability of
The neo-Darwinian paradigm 31
reproduction and diffusion of one rather than another item or package of
heritably variable information explains the outcome observed.
The just-so stories of selectionist sociology diverge most clearly from
conventional historical narratives at the points where the deliberate
choices and conscious objectives of individual agents are agreed by obser-
vers of all theoretical schools to have had a significant influence on the
subsequent course of cultural or social evolution. Biographically told, as
narrative historians commonly do, the story is likely to be seen by the
agents themselves, and therefore by the narrator, as a succession of ingeni-
ous attempts to achieve the agents’ chosen aims in either collaboration or
conflict with other purposive agents. But although, in selectionist the-
ory, it is explicitly recognized that successful innovations may contain, in
the phrase of Herbert A. Simon (1990: 2), ‘large components of conscious
invention’, when it comes to explaining how it is that cultures and soci-
eties come to be of the kinds that they are, innovators need to be seen as
carriers of what might as well be random inputs into the ongoing process
of heritable variation and competitive selection of information rather than
as executants of master plans which are then helped or hindered by what
might as well be random inputs from the environment. Descriptively, the
second can hardly fail to give the more authentic account of what it feels
like to reach a chosen goal despite the obstacles standing in the way: nei-
ther Julius Caesar nor Napoleon Bonaparte are likely to feel like, or think
of themselves as being, chance intruders into the ongoing course of social
evolution. But that is what they are. Although competitors for influence
or power whose ambitions have been realized may present themselves as
winners who have overcome these obstacles through their exceptional
foresight and strength of will, to the selectionist sociologist they are initia-
tors of variations in memes or practices whose extended phenotypic effects
and long-term reproductive fitness they could no more predict than they
could control. It explains nothing to say that a lasting population-level
change in cultural or social behaviour-patterns accords with an individual
decision-maker’s aims unless and until it has been shown what selective
pressures in the environment have caused those aims to be realized.
4
The psychologist John Bowlby (1969: 50) was the first person to make
explicit the distinction between evolutionary adaptedness and designed
adaptedness. But in either case, the right just-so stories will have to spec-
ify what it was in the environment which gave the critical mutations or
32 The theory of cultural and social selection
recombinations of information their greater probability of reproduction
and diffusion than their competitors. A novel strategy which turns out
to have far-reaching population-level effects may have been adopted after
careful trial-and-error, after diagnosis of the entrails of an animal sac-
rificed to the gods, after tossing a coin to decide between two available
alternatives, or after taking advice from a management consultant. But
the underlying process of heritable variation and competitive selection of
information affecting phenotype is the same.
Nor does the neo-Darwinian paradigm presuppose, as its opponents
persistently allege, that competitive selection brings about a Panglossian
best of possible worlds. By definition, it produces winners and losers. But
success is always relative, optimization is always to a local standard, and
adaptation is always a matter of trade-offs. It is quite possible that for all
the designed adaptedness of innovations introduced by medical research-
ers into the unending war against disease, both cultural and social evo
lution will one day come to an end because of the competitive advantage
which natural selection gives to bacteria over human beings. But in that
event, bacteria will have proved themselves ‘better’ than humans only in
the trivial sense that they are still there and we are not. Likewise, it can
readily be seen that some designs are ‘better’ than others, as much in the
arms races of the natural world in which both predators and prey, such
as cheetahs and antelopes, become more and more efficient at pursuing
and being pursued, as in the arms races of the human world, in which
lethal weapons become more and more ingeniously constructed and cost-
effectively manufactured. But for all the ingenuity of the solutions to
problems of design which are arrived at by natural, cultural, and social
selection alike, the natural, cultural, and social worlds are full of demon-
strable imperfections relative to the theoretical optimum. The degree of
imperfection which is compatible with local reproductive success is a
matter to be examined case by case, and the technique of reverse engi-
neering which evolutionary biologists apply to the study of organisms and
evolutionary psychologists to the study of the human brain can be no less
profitably applied by evolutionary sociologists to the study of cultures and
societies.
5
In sociology, as in biology, the resemblances generated by the evolution-
ary process can be as remarkable as the differences, and where they occur
some means has to be found of establishing whether similar collective
The neo-Darwinian paradigm 33
behaviour-patterns are the outcomes of convergent evolution, homologous
descent, or lateral diffusion. Sometimes, the resemblances are the evident
outcome of designed adaptedness and deliberate imposition: any story of
the evolution of Japanese society after the Meiji Restoration of 1868 will
include the determination of its rulers to follow the example set by the
practices, roles, and institutions of what they saw as the more successful
societies of the West. But no such story can account for the resemblances
which the Spanish conquistadors discovered – to their surprise – between
the institutions of the societies of the American continent and the institu-
tions of their own. Similarly, the resemblances between the laws promul-
gated by the kings of Anglo-Saxon England and Hammurabi’s law-code
as promulgated in Babylonia two millennia earlier can only be explained
as the convergent evolution under common selective pressures of wholly
unrelated societies which were, however, remarkably similar in their
modes of production, persuasion, and coercion alike. On the other hand,
there are many law-codes which have been formulated in direct imita-
tion of others, or even designed on request by a professional lawgiver like
Philolaus of Corinth, who was expressly summoned for the purpose to
Thebes (Aristotle, Politics 1274b). In cultural selection likewise, when
the folktales handed down within seemingly separate cultural traditions
look – as they often do – like identical reflections of universal human
concerns evoked by common experience, it may nevertheless be unwise
to rule out the possibility of lateral transmission for which the migra-
tion of a single carrier from one population to another can be enough.
But the subsequent reproduction and diffusion of the imported memes
will then depend on the receiving environment, which may favour them
either because they have a selective cultural affinity with the indigenous
tradition, or because they evoke a biologically adaptive self-protecting
emotion (Heath et al. 2001), or both.
In natural selection, lateral gene transfer and hybridization are suffi-
ciently unusual that the just-so stories of descent with modification nor-
mally start from the assumption that species are reproductively isolated.
Cultures, by contrast, are much more permeable than species, and socie-
ties are always vulnerable to invasion by mutant practices exogenously
imposed. But phylogenetic models borrowed from biology have been
applied to good effect in the construction of trees of material artefacts as
well as of languages, and in the absence of documentary records they can
make it possible to test the consistency with the data of rival hypotheses
about the vertical transmission of the information which finds expres-
sion in distinctive behaviour-patterns from carpet-weaving to bridewealth
34 The theory of cultural and social selection
(Mace et al. 2005). If, as is theoretically possible, lateral transmission
of either acquired or imposed information affecting phenotype were to
be sufficiently continuous and extensive, it would call into question the
whole idea of heritable variation and competitive selection: both cultures
and societies would lose their internal coherence, no enduring cultural
traditions or social institutions would become established, and com-
parative sociology would be reduced to the study of shifting networks
of short-term interpersonal influence or institutional power. But both
cultures and societies tend, for all their disanalogies with species, to be
both stable and resilient. Although concepts, techniques, and life-styles,
like organizational forms, codes of law, and systems of government, can
all be borrowed from elsewhere, this is nearly always within and not in
place of a pre-existing set of beliefs and attitudes and a pre-existing set
of economic, ideological, and political institutions. I shall come back to
the topic of ‘punctuated equilibrium’ in Chapter 4. But in both cultural
and social evolution, partial modification is much more often than total
extinction and comprehensive replacement the answer to ‘just what is
going on here?’
T h r e e l e v e l s of va r i at ion a n d s e l e c t ion
6
In Section 3 of the Prologue, I used the example that sociologists of war-
fare are simultaneously observing three different kinds of behaviour: bio-
logically evoked, culturally acquired, and socially imposed. The distinction
applies no less to other equally longstanding sociological topics: in the
comparative sociology of kinship and descent, for example, the differ-
ences between populations can be fully understood only if due attention
is paid to genetic relatedness and cultural construction of kinship terms
and social construction of inheritance practices. But the comparative soci-
ology of warfare brings out particularly clearly the three separate levels at
which the three different mechanisms of transmission from carrier to car-
rier of information affecting phenotype operate both in parallel and in
interaction with one another to generate collective behaviour-patterns of
distinctive kinds.
There is, as so often, a definitional problem. But it is not a serious
one. Wars need to be distinguished from skirmishes, raids, ambushes,
or feuds (to which I return in Section 9 of Chapter 2). Battles between
bands of hunters and gatherers are very different from battles between the
The neo-Darwinian paradigm 35
professional armies of large sovereign states. But in each case, even if cas-
ualties are deliberately kept low, there is organized lethal violence going
on, and its explanation does not depend on being able to specify what
does or does not amount to full-scale ‘war’. Lethal violence is unmistak-
ably discernible not only among chimpanzees2 and present-day hunter-
gatherers but in the evidence for pre-literate peoples (Keeley 1996). There
is no prelapsarian idyll in the history of the human species when it was
unknown. The archaeological record is full not only of clubs and daggers
but of trophy skulls, traces of spear-inflicted injuries, and so-called ‘parry
fractures’, and wherever face-to-face combat is studied directly, evoked
behaviour can be observed in immediate responses to direct physical
threat, in spontaneous support for endangered comrades, and in instinc-
tive retaliation even at high personal risk – to say nothing of such readily
understandable behaviour as the incorrigible edging to the right, vividly
described by Thucydides, of Greek infantry ranks whose stabbing weap-
ons were carried by their right arms while their shields were carried on
their left.
But even a cursory glance at the historical and ethnographic record
discloses the extent to which beliefs about, and attitudes to, warfare and
the resulting behaviour in which they find expression are acquired by imi-
tation and learning from parents, mentors, peer-groups, and role-models.
Populations which are not only indistinguishable genetically but also share
a common ecological environment can be significantly more or less war-
like than one another. Even when reared in an exceptionally non-violent
culture, young adult males recruited into a foreign army may turn out
to fight as ferociously as the rest: a revealing example is the recruitment
of young males of the Malaysian Semai, who are one of the most peace-
able peoples in the ethnographic record, by the British after the Second
World War in order to fight a Communist insurgency. But that could be
said to make it all the more striking that the Semai should have evolved
a culture in which the naturally selected aggressive dispositions of young
adult males are unusually effectively restrained. When the historical and
ethnographic record is surveyed overall, it reveals that in the overwhelm-
ing majority of places and at the overwhelming majority of times bravery
in battle is admired and cowardice condemned, successful combatants
Chimpanzees, that is, of the kind first closely studied (Wrangham and Peterson 1996) and most
2
closely related to ourselves. Bonobos, although less peaceable than earlier thought (Dunbar 2004:
85–6), are significantly less aggressive than either pan paniscus or homo sapiens. De Waal (2005:
221) summarizes the position of homo sapiens by saying that ‘Being both more systematically bru-
tal than chimps and more empathic than bonobos, we are by far the most bipolar ape.’
36 The theory of cultural and social selection
are held up for admiration in history, fiction, and myth,3 those who have
died in battle are honoured and their memory preserved, conquests and
annexations are both celebrated and legitimated, the killing of enemies
is sanctioned by explicit moral codes, and the active deployment of the
means of coercion against the hostile ‘other’ is taught to be both a right
and a duty. But the variation both within and between different cultural
traditions, from the Semai at one end to, say, the Aztecs at the other, can-
not be accounted for except as the product not only of evoked responses
to different environments but also of beliefs and attitudes acquired by
imitation and learning within different cultural traditions with path-
dependent histories of their own.
At the same time, however, no sociologist of warfare can be unaware
of the power of the institutional inducements and sanctions enabling
the roles through which societies control and exercise lethal violence to
function as they do. In the early stages of the transition from culture to
society, the inducements and sanctions are interpersonal only: friends or
kinsmen are recruited ad hoc,4 and cowards or deserters are punished,
whether physically or by shame and ostracism, without there being any
formal role to which disciplinary power is attached.5 But societies lacking
the roles constitutive of states cannot compete, if it does come to full-scale
war, with societies in which practices defining institutionalized military
roles enjoy an overwhelming advantage. Armies – to say nothing of navies
and in due course air forces – are groups of a kind which shared cultural
norms and purely interpersonal influence cannot sustain by themselves.
The behaviour of the fighters in institutionalized armies, as opposed to
the fighters in hostile gangs or feuding families, is ‘imposed’ not merely
in the sense that their commanders’ orders are backed by the state’s
monopoly of the means of coercion but in the sense that the roles into
which recruitment takes place are defined by practices over which their
3
The convergent evolution of a cultural tradition of heroism like that of ancient Greece among
African populations where the proud and courageous warrior is likewise admired and extolled is
documented in detail by Iliffe (2005).
4
As, for example, by Telemachus in Homer’s Odyssey. Although the example is fictional, it is symp-
tomatic that when Telemachus has to find a crew with whom to set out for news of his father, he
can only do so by calling on contemporaries who are also friends (III.363), and when Antinous
asks him how he managed it (IV:642–4), the only alternative he puts forward is house servants
of Telemachus’ own. One way of putting the difference is to say that his crew are members of an
‘association’, in contrast to, for example, the crew of Nelson’s Victory, who are members of an
‘institution’ – namely, the British navy.
5
Thus among the Bergdama (Schapera 1956: 83), a band member might be singled out for physical
punishment, but the decision would be ‘reached casually round the campfire’ and its execution
entrusted for the occasion to the younger men.
The neo-Darwinian paradigm 37
incumbents have no say. Instructions which all recruits are required to
follow now override whatever culturally transmitted beliefs and attitudes
they bring with them to the soldierly roles which they occupy and per-
form. The patriotic volunteer brought up in a family with a long-standing
cultural tradition of military service may fight more bravely than the
reluctant conscript. But both are subject to a compulsory power different
in kind from that of the gang leader or family head.
7
Once it is recognized that there are not two but three different mechanisms
by which, and levels at which, the heritable variation of information affect-
ing phenotype and the pressure of competitive selection on its extended
phenotypic effects generate the observed resemblances and differences
between human populations, it follows not only that the notion of ‘socio-
cultural’ evolution needs, as I have insisted already, to be broken down
into its two separate components, but that the notion of ‘gene-culture
co-evolution’, which has become a standard term in the vocabulary of
the neo-Darwinian paradigm, needs to be supplemented by both ‘gene-
society co-evolution’ and ‘culture-society co-evolution’. The terminology
is discretionary. Either ‘nature-culture’ or ‘gene-meme’ – or ‘meme-gene’
(Bull et al. 2001) – co-evolution would be a more logical formulation than
‘gene-culture’. But the distinction is fundamental, however worded,6 and
however frequently glossed over by sociologists analysing the workings
of large and complex societies with long ‘sociocultural’ histories behind
them. Selectionist theory has the wide-ranging implications for compara-
tive sociology that it does precisely because the idea that all behaviour is
the acting-out of information transmitted through the process of herita-
ble variation and competitive selection applies no less when the behaviour
is the acting-out of information institutionally imposed than when the
referred in the opening Section of the Prologue, is further compounded by the distinction, sig-
nificant as it is, drawn by the philosopher John Searle between ‘brute’ and ‘institutional’ facts
(1995: Ch. 2). For example: the physical properties of gold are a ‘brute’ fact, whereas the uses to
which it is put by human beings are an ‘institutional’ one; but the value attached to it as personal
adornment (‘jewellery’) is a cultural construction, whereas its function as a bullion reserve in
the vaults of a central bank is a social construction. In an influential game-theoretic analysis of
self-reinforcing equilibrium in medieval merchant guilds by the economic historian Avner Greif
(2006: Ch. 4), meme-practice co-evolution is clearly at work, but Greif ’s chosen definition of an
‘institution’ lumps rules, beliefs, norms, and organizations together as the generators of regulari
ties of collective behaviour.
38 The theory of cultural and social selection
information has been interpersonally acquired by imitation or learning
independently of the practices encoded in the agents’ social roles.
A by now textbook example of ‘gene-culture’ (i.e. gene-meme)
co-evolution is adult lactose absorption (Durham 1991: Ch. 5; Holden
and Mace 1997). Contrary to what was for many years an untested
assumption of medical research, milk-drinking is far from univer-
sal among human populations. Instead, the physiological capacity to
absorb lactose has evolved to high frequencies in populations where
an ecological environment in which vitamin D and metabolic calcium
are lacking is combined with a long cultural tradition of dairying. The
gene involved in the metabolism of lactose has been identified, and its
high frequency in Northwest Europe, as among the East African Tutsis,
demonstrated. The precise relation between the cultural information
encouraging milk-drinking and the genetic information permitting lac-
tose absorption is not easy to disentangle, but the cross-cultural evi-
dence strongly suggests that the necessary genetic modifications were
driven at least in part by the continuing transmission by imitation and
learning of packages of information in which milk was positively evalu-
ated as a source of well-being – a conclusion supported by comparative
analysis of the folklore and creation myths of cultures which do and
don’t have a norm of milk-drinking. Then, however, as social selection
also comes into play, the institutional practices and roles of the dairy
industry combine with culturally constructed representations of cows
to cause consumers in market economies to buy more and more dairy
products (Wiley 2007).
Adult milk-drinking and the genetic modification that it requires are,
however, a long way away from the topics in the world of armies, banks,
temples, courts (in both senses), taxation, conscription, bureaucracy,
churches, businesses, governments, and so on with which sociologists
are accustomed to deal. With these, we are confronted by behaviour-
patterns generated by large-scale, long-lasting, and wide-ranging insti-
tutional, and nor merely intra-familial or intra-associational, inequalities
of power. Conformity is no longer a matter of cultural conventions and
personal sanctions, but of obedience to rules imposed by the incumbents
of roles to which there attaches formal entitlement to punish infringe-
ment or deviation. The transmission of information through imitation
and learning still continues, as it must. The people who are to occupy
and perform their interacting roles as generals and soldiers, priests and
parishioners, masters and slaves, parliamentarians and electors, profes-
sors and students, capitalists and wage-labourers, judges and plaintiffs,
The neo-Darwinian paradigm 39
or tax-collectors and tax-payers all have to get into their heads and retain
there the correct information about what to do in order to be what –
institutionally speaking – they are. But what they do when they are is
no longer just a matter of imitation or learning. The difference is not
that social, as opposed to cultural, rules have to be followed, since even
where rulers’ commands are backed by overwhelming physical force
recalcitrant individuals may choose incarceration or death rather than
obedience. The difference is that the information encoded in institu-
tional practices makes a society’s roles what they are independently of
how their successive incumbents have come to learn to perform them or
what their individual motives are for doing so. To sign a labour contract,
do military service, pay a poll tax, render dues to a manorial lord, stand
trial for witchcraft, manage a coal-mine or a shipyard or a factory, run
for elective office, or sue a fellow citizen for damages is to do something
different in kind from following a recipe for cake-making, playing a tune
on the violin, wearing a copper amulet as a palliative for rheumatism, or
laying flowers on your mother’s grave.
Why then is it that the difference has been ignored to the extent
that it has in the literature on ‘sociocultural’ evolution? Durham, for
example, in the book just cited, explicitly allowed that behaviour can
be ‘imposed’ in the sense that agents’ otherwise free choices can be
subject to the decision of others more powerful than they, but treated
this simply as an unexplained externality which from time to time redi-
rects what would otherwise be the course of cultural evolution. It is
not in dispute that such effects can be a matter of purely interpersonal
inequalities between the older and the younger, or the eloquent and the
inarticulate, or the skilful and the incompetent, or the physically strong
and the physically weak. But in the world of economic, ideological, and
political institutions, behaviour is imposed through the practices which
define the relationship between dyads of complementary social roles
and the systacts – classes, orders, age-sets, castes, status-groups, and so
on – which they form. The resulting differences between one society
and another are just as much the outcome of a process of heritable vari-
ation and competitive selection irreducible to cultural evolution as the
beliefs and attitudes which distinguish one culture from another are the
outcomes of a process of heritable variation and competitive selection
irreducible to biological evolution. In cultural evolution, mutant memes
emerge through reinterpretation in the minds that carry them. In social
evolution, mutant practices emerge through renegotiation between the
incumbents of interacting roles.
40 The theory of cultural and social selection
8
The distinction between cultural and social selection is both analytic and
historical. Analytically, one way to bring it out is to consider just what
is going on when Captain Bligh and his companions are cast off from
the Bounty by Fletcher Christian and the mutineers. Bligh is deprived at
a stroke of the institutional inducements and sanctions which sustained
his role as ‘Captain’: he can no more grant a favoured subordinate pro-
motion than he can order a recalcitrant one to be court-martialled. His
companions entrust him with the steering since it is on his navigational
skills that their survival will depend. But if for any reason they decide to
replace him at the tiller by another of their number he will be powerless
to prevent them. So far as food is concerned, they now offer a textbook
instance of ‘immediate-return’ (which is adaptive for hunter-gatherers), as
opposed to ‘delayed-return’ (which is adaptive for agriculturalists), behav-
iour. These are just the circumstances for which was devised the game of
fair division known in the British navy as ‘Who shall have this?’ In this
game, when a fish or sea-bird has been caught, it is divided up in the sight
of everyone into as near-equal portions as feasible. One person chosen
at random then goes into the bows of the lifeboat, looks straight ahead
and calls out the names of the complement in a random sequence, while
another in the stern holds up the pieces in what is likewise a random
sequence. It may well be Captain Bligh or someone like him who organ-
izes the game and teaches his companions its rules. But in doing so, he
is behaving no differently from the informal, ad hoc leader of a band of
hunters and foragers whose advice is followed about where to look for a
water-hole where there may be a chance of finding game. It is only when
the lifeboat has safely reached Australia that Captain Bligh once again
becomes, as he instantly does, ‘Captain Bligh’.
Historically, the transition from acquired to imposed behaviour has
taken place wherever and whenever interpersonal relations conceptual-
ized as leader/follower, senior/junior, mentor/novice, or foreigner/indigene
have been supplemented by, or converted into, formal relations between
institutional role-incumbents explicitly recognized as such in the natives’
language. It can be effected either by agreement or under compulsion. Its
first occurrence did not depend on agriculture, although agriculture and
the domestication of animals as well as plants gave it a powerful impetus.
It did not require literacy, which only emerged several millennia later.
It did require language, but language long predated it – by how much,
nobody knows. It may have been many tens of thousands of years after the
The neo-Darwinian paradigm 41
evolution of art, ritual, and myth that there evolved economic, ideological,
and political institutions of societies thereby distinguishable as such. In
the absence of written records, sociological reconstruction depends on
such inferences as can be drawn from the archaeological record. But the
evidence from the area between the Eastern Mediterranean littoral and
the Syro-Arabian desert (Bar-Yosef 2001) is more than adequate to under-
write a story of sedentary or semi-sedentary foragers who during a period
of local warming lasting from about 14,500 to 12,800 years ago began
to live in permanent or at least seasonal hamlets or villages, to domes-
ticate plants and animals, to harvest wild grains with sickles, to manu-
facture baskets, fish-hooks, decorated bone implements and kitchen
equipment, and to bury their dead with grave offerings which include
pet dogs. Sedentism can be both limited and reversible, and agriculture,
when it becomes an option, is not necessarily a more attractive way of
life than hunting and gathering. Nor are hunter-gatherers without well-
established conventions of behaviour, consistent intra- and extra-familial
relationships, division of labour, and organized collaboration for com-
mon purposes as well as ritual activities informed by jointly accepted atti-
tudes and beliefs. But once material resources begin to be accumulated
and stored and families or households to settle at designated locations
which their members regard as their own, the necessary conditions for
the transition from acquired to imposed behaviour are in place. As settle-
ment sizes become larger, face-to-face relationships more tenuous, and
problems of social control more severe, the need for institutional roles
by which repeated interactions can be made predictable independently of
purely personal characteristics becomes increasingly pressing. Discipline
of children within the face-to-face group and informal coalitions of like-
minded adults against free-riders, defectors, and self-aggrandizers are no
longer enough to keep the peace. Since we have no way of knowing just
what ‘they’ were saying to one another, we have no way of knowing what
the first such roles actually were. Perhaps the plaster-covered rounded
bench found in a small Natufian building at Ain Mallaha (Bar-Yosef 1998:
163) was sat on by one or more people in whom formal power like that of
a judge or magistrate or chief had been vested. Or perhaps those who sat
on it were not distinguished from others in any way which went beyond
some immediate short-term purpose or need. But whatever the answer, by
the time of the 10-hectare, 1,000-person settlements of the period known
as ‘Pre-pottery Neolithic B’, let alone the 20,000-strong Sumerian cit-
ies, their populations could not have cohered and reproduced themselves
without institutionalized political, economic, and ideological roles of a
42 The theory of cultural and social selection
kind unthought of by their hunting and foraging forbears. It is a matter
of continuing debate among archaeologists whether it is appropriate to
speak of a ‘Neolithic Revolution’. If the question at issue is about cogni-
tion and material culture and the relation between the two, the answer
would appear to be no: there was a gradual ongoing cultural evolution of
a classic Darwinian kind. But sociologically, the answer would appear to
be yes: the culture-to-society transition (Runciman 2001b) was a revolu-
tionary transformation of interpersonal relationships and consequential
behaviour-patterns without precedent in the earlier history of the human
species.
In the vocabulary of present-day hunters and gatherers, the presence
or absence of words for institutional as opposed to personal roles in the
vernacular terminology is a visible symptom of the difference. Thus, the
!Kung San are reported by Lee (1979: 344) as applying to themselves, as
opposed to Bantu headmen, the word for ‘chief’ only ‘in a derisory man-
ner’, and among the Batek the Malay word for ‘headman’ (penghulu) had
to be imposed on their informal leaders by the Department of Aboriginal
Affairs (Endicott 1988: 123). The terminology may be fluid and ambigu-
ous, and it is not to be supposed that the first transition from culture to
society happened overnight. But if a representative member of the popu-
lation is asked by a visiting anthropologist, ‘Why do you behave towards
so-and-so as co-operatively you do?’, the answer ‘Because he is my friend’
or ‘Because he is a courageous fighter’ is different in kind from ‘Because
he is our hereditary monarch’ or ‘Because we have just elected him to be
our commander-in-chief’.
Somebody has to have been the first incumbent of any novel institu-
tional role, just as somebody has to have been the first person to have a
novel belief or attitude inside their head. But a person can, for example,
have the hitherto unknown idea of drawing in the sand or on the wall of
a cave an object or person or animal and therefore introducing ‘art’ into
their culture without initial reference to anyone else. In social evolution,
by contrast, the behaviour is necessarily a two-sided affair. An artist can
be an artist without a public, but a capitalist employer can’t be a capital-
ist employer without a market in labour. Back in the Palaeolithic, there
may have been many occasions when a successful claimant to supernat-
ural insight sought recognition as permanent head of an ongoing follow-
ing of disciples, or a household head controlling more stored foodstuffs
than others tried to become the acknowledged collector and distribu-
tor of a community’s surplus, or a successful leader in skirmishes with
other groups or bands tried to assume formal authority over all activities
The neo-Darwinian paradigm 43
connected with warfare. Or perhaps there were communities in which
‘elders’, having at first been no more than males living beyond a certain
age, became credited with economic, ideological, and political powers
entitling them to control hunting and the distribution of game, to deploy
supernatural sanctions against deviants, and to order young men into bat-
tle without taking part in the fighting themselves. In later and more com-
plex societies, too, there are many borderline cases. Is the Greek kolax,
or ‘flatterer’, the incumbent of a social role (Millett 1989: 32), or is his
relation to his rich patron personal rather than institutional? Later still,
under the Roman Empire, the ‘sycophants’ dining at the tables of great
men might be not just personal cronies but carriers of publicly recognized
practices which functioned to channel political influence both upwards
and downwards between rulers and ruled. But only after the line has
been crossed do children start to grow up in a world not merely of par-
ents, friends, mentors, siblings, and culturally defined exemplars but of
roles whose incumbents wield economic, ideological, and political power
which attaches to them independently of occupancy by any particular
person.7 Then, but only then, does social mobility – the inter- and/or intra-
generational upward or downward movement of individuals from one
to another institutional role located in three-dimensional social space –
become a standard topic on the sociological agenda. Individual rank-
orderings are familiar from other species as well as from human groups.
But the alpha male in a chimpanzee troop is not, except metaphorically,
its ‘king’, any more than the most admired female in a rank-order of per-
sonal prestige based on physical appearance is, except metaphorically, a
‘beauty-queen’.
9
Although no direct inference can be drawn from the vocabularies of the
hunters and foragers documented in the ethnographic record to the peoples
whose culture has to be inferred from the archaeological record, there are
many well-documented peoples whose vernacular terminology illustrates
the transition from culture to society. Radcliffe-Brown, for example, noticed
long ago that in the North Andaman language men of outstanding personal
qualities were spoken of simply as ‘big’. But among the South American
Institutional roles can also exist without incumbents, as most obviously in interregna. In a strictly
7
hereditary monarchy, succession is instant: le roi est mort, vive le roi! But there was a ten-year
interregnum in the Lombard kingship during an early stage of their occupation of Italy (Paul the
Deacon, History of the Lombards III.16).
44 The theory of cultural and social selection
Achuar, the juunt, who is literally a ‘great man’, is a household head who
is also a ‘war herald’ (chicharu) whose role empowers him to organize
defence against, or negotiation with, potential enemies (Descola 1996: 290).
Some Inuit call an informal leader ‘he who thinks’, but the ‘knowledge-
able one’ (nindibuga) of the Australian Mardujarra does not merely offer
advice but trades his esoteric knowledge for a share of the young men’s meat
(Tonkinson 1988: 157). A good example of a transitional designation is the
Alaskan umialik or umealaq – a family head of some personal wealth who
is a boat-builder and captain of its crew and who is allotted a larger share of
the catch in recognition of his greater responsibility (Halstead and O’Shea
1989: 18). In twentieth-century Papua New Guinea, a successful war leader
(watenge) could control ‘vast’ amounts of wealth flowing in exchange net-
works (Feil 1987: 249), and the position became heritable. In a borderline
case, it has to be established whether or not the designation implies that
there are recognized inducements and sanctions attaching to it which are
transferred, whether automatically or after an interval, to a successor: thus,
among the Cree hunters of Northern Quebec, the uuchimaau is the leader
of an extended kin-group who controls a hunting ground and the animals
on it and will be replaced by another if he fails in the management of the
resources for which he is held responsible (Scott 1988: 39–40).
Just as, however, there are anthropologists who underestimate bio-
logical influences on cultural evolution, so are there sociologists who
underestimate cultural influences on social evolution. Consider what
happens when money and the institutions which it brings with it displace
gift, barter, and customary exchange. For Marx, in particular, money was
a destructive innovation which undermines social solidarity and substi-
tutes exploitative practices and roles for non-market transactions in kind.
Such effects do undoubtedly occur: on any theory, it is unsurprising that,
for example, among the Kinnaurese of the Western Himalayas the intro-
duction of cash and a market economy modified the centuries-old binana
system of customary work exchange between different castes (Raha 1978).
But the range of variation is much wider than Marx allowed for. It has
often been found that ‘the meanings with which money is invested are
quite as much a product of the cultural matrix into which it is incorpo-
rated as of the economic functions it performs as a medium of exchange,
unit of account, store of value, and so on’ (Parry and Bloch 1989: 21). All
over the world, from India to the Andes and from the Roman Empire8
8
In the later empire, the wealth which entered the establishment of a Christian ‘holy man’ would
be symbolically cleansed by ‘a dramatic story in which the original giver of the wealth had first
been “taught a lesson” by the saint’ (Brown 1995: 64), like the twentieth-century Lebanese lord
The neo-Darwinian paradigm 45
to twentieth-century Madagascar, detailed analysis of the ways in which
monetary practices and the roles of their carriers are interpreted and
symbolized reveals meme-practice co-evolution at work, with the conse-
quence that money as a potentially disruptive social force is culturally
purified and tamed.
The dependence of institutional practices on culturally transmitted
beliefs and attitudes is easiest to see in societies’ modes of persuasion.
A collective behaviour-pattern in which deference is paid to superiors
in social status in an ideological hierarchy of honour and respect is self-
evidently vulnerable to a memetic mutation which causes the sanctions
underpinning it to lose their force. But reproduction of the practices
defining economic roles through which material goods are exchanged
according to institutional rules is equally vulnerable to a refusal to accept
those rules as valid. So too is reproduction of the practices defining politi
cal roles to a refusal to be intimidated by the sanctions encoded in law
and enforced by soldiers or police. Practices in all three dimensions of
social space are always at the mercy of the fragility of the cultural con-
structions on which all social constructions depend. What the French
sociologist Pierre Bourdieu calls the ‘habitus’ – the set of conventions dic-
tating appropriate behavioural strategies which the members of a com-
mon culture derive from their observation of each other’s behaviour – is
usually sufficiently robust to sustain the economic, ideological, and polit-
ical institutions which they occupy and within which they perform their
roles. But it cannot be taken for granted, as prudent rulers even of long-
established and seemingly well-disciplined societies are well aware.
10
Given that there are three levels of selection, there are by definition three
levels of adaptation. The concept of adaptation has long been controver-
sial in evolutionary theory, partly because it can be used to cover any
kind of re-equilibrating systemic response to exogenous disturbance,
partly because of its mistakenly alleged implication of Panglossian opti-
mization, partly because of the need to distinguish biological adapted-
ness from cultural or social adaptedness, and partly because adaptive
‘taught a lesson’ by the sheikh whose disinterested distributions of largesse are in large part
funded by the lord (Gilsenan 2000: 102–3). Among present-day Swedish conservative Protestants,
even the most seemingly commoditized monetary transaction ‘partakes of a sacred quality in that
it contributes to the global Faith nexus of production, distribution, and associated missionary
activities’ (Coleman 2004: 431).
46 The theory of cultural and social selection
behaviour may or may not be an adaptation so designed as to be adaptive
in a particular local environment. Selectionist theory accommodates both
maladaptations and ‘exaptations’ where the genes, memes, or practices of
which they are the phenotypic effects are adaptive in their present envir-
onment for reasons other than those for which they were selected in an
earlier one. But it is always a matter of relative advantage: a mutation
or recombination of information affecting behaviour in the phenotype,
whether biologically, culturally, or socially transmitted, is adaptive to the
extent that its probability of reproduction within the relevant population
is greater than that of its competitors. The just-so stories of how cultures
and societies have come to be of the kinds that they are are stories about
how the impact of their environment on the phenotypic effects of their
constitutive memes and practices has benefited not, or not necessarily,
their carriers, but themselves. Variation can also be adaptive at one level
and not another, so that the problem becomes one of finding some way to
establish which is the stronger evolutionary force. Moreover, there is the
further complication, although a terminological rather than a substantive
one, that ‘natural selection’ can be used to refer either to genetic selec-
tion or to selection which comes about through the extraneous winnow-
ing or even extinction of the population of carriers. When, for example,
Chindasuinth succeeded to the kingship of Visigothic Spain in 462 CE
and, according to the chronicler known as ‘Fredegar’ (IV.82), put to death
700 of the Gothic aristocracy, this was selection of the same kind as when
a species is driven extinct not by maladaptive genetic variation but by a
climate change resulting in a catastrophic shortage of the foodstuffs on
which its members depend for their survival.
A vivid example of how a culturally adaptive mutation can be bio-
logically maladaptive is the red trouser (pantalon rouge) worn by French
infantry soldiers at the outset of the First World War (Tuchman 1962: 48).
It cost many of its wearers their lives in the environment in which they
went into battle. But it encoded an emotionally resonant patriotic sym-
bol (‘le pantalon rouge, c’est la France!’) whose potential encouragement of
in-group loyalty was transmitted down successive generations of recruits
as a part of their initiation into the culture of the army they had joined.
Before long – before, that is, the population of potentially reproductive
red-trousered recruits had been winnowed by their more effectively cam-
ouflaged German opponents – the higher command imposed a change.
But it was a change which in the summer of 1914 had been unthinkable.
That cultural mutations can be biologically maladaptive may be too
obvious to require illustration: to take the extreme case of the Shakers, if
The neo-Darwinian paradigm 47
a whole population is converted to the view that it is wrong to procreate
there will soon be nobody left to pass on their genes at all. But a more
widespread setback for natural selection is the persistent decline in fertil-
ity in modern human populations to levels well below what they could
be. This topic has a long history in the study of both animal and human
behaviour, and it is well recognized both that reproductive fitness is not
a matter of numbers in a single generation and that quality of offspring
is as important as quantity. But the global data on present-day human
populations do not endorse the expectation that parents who invest more
in a smaller number of healthier children will have more grandchildren.
Just what explains the difference between relatively undifferentiated agri-
cultural or horticultural societies in which parents have as many chil-
dren as they can and are admired for doing to and industrial societies
in which (with Roman Catholic and some other sub-cultural exceptions)
they prefer to restrict their number of children far below the theoret-
ical maximum is uncertain: it is not a function only of mortality rates
or differential fertility among inbred or outbred populations or levels of
disposable income. But for the present, at any rate, in the competition
between natural and cultural selection, cultural selection is winning in
more affluent and better-educated populations. In countries like the one
in which this book has been written, parents are deciding (or imitating
other parents who have so decided) to behave in a way which does little
or nothing to maximize the probability of ongoing replication of their
descendants’ genes, whatever it may do for their children’s average quality
of life or, for that matter, their own. And then there is the further pos-
sibility that the behaviour is neither evoked nor acquired but imposed
by institutional sanctions, like the ‘one-child’ policy of the rulers of
twentieth-century China, so that gene-practice co-evolution supplements
gene-meme co-evolution. Behavioural ecologists are not wrong to point
to well-researched cases where predictions derived from their models are
confirmed (e.g. Borgerhoff Mulder 2000). But comparative sociologists
are not wrong to point to demographic fluctuations of whose evolution-
ary significance they are fully aware but whose explanation may depend
on both cultural and social influences which override the innate dispos-
ition to maximize inclusive reproductive fitness, as among middle-class
families in Victorian Britain (Banks 1981).
Just as cultural selection can compete directly with natural selection,
so can social selection compete directly with cultural selection. A well-
documented example is the introduction of the practice of wage-labour
into societies whose mode of production has hitherto been constituted
48 The theory of cultural and social selection
by practices defining the roles of slaves, clients, sharecroppers, pawns,
domestic workers, or junior kin. Sometimes, the competition between
alternative practices is acted out entirely at the level of social selection:
an entrepreneur in a pre-capitalist economy who is the first to assemble
and supervise a body of wage-workers and then market a finished product
directly to a growing number of consumers thereby raises his turnover
sufficiently to be able to lower his prices and leave his rivals reliant on
putting-out and domestic production no choice but to follow suit or go
out of business. But sometimes, wage-labour fails to displace its competi-
tors for reasons which are cultural, not social. Among the Baoule of the
Ivory Coast, for example, young men regarded wage-labour as demean-
ing and therefore refused to undertake it in their own home territories,
although they were willing to migrate seasonally in order to work for
wages elsewhere (Iliffe 1983: 55). Where the employer’s role is underwritten
by sufficiently powerful institutional sanctions, the outcome may depend
solely on the employer’s calculation of self-interest and the practice will
be imposed directly. Roman landowners often supplemented slaves with
wage-labourers at certain times of year or for certain kinds of farmwork:
Varro (De Re Rustica I.xvii.3) is explicit about the advantages of being able
to chose them for their age and physical suitability and to interview them
in person about their experience with previous employers. But prospect-
ive employers as well as employees may have acquired by imitation and
learning beliefs and attitudes which dispose them against the practice of
wage-labour even on terms economically advantageous to both individual
members of the role-dyad carrying it.
There is wide scope for designed as well as evolutionary adaptedness at
all three levels. At the biological level, the breeders of animals with whom
Darwin so assiduously corresponded are the obvious example. But the
advertisers in a capitalist economy who market-test their clients’ products
before they are launched into the population of prospective consumers are
doing much the same, and rulers of newly subordinated populations can
deliberately exploit their cultural traditions for the rulers’ own social pur-
poses. In the culture of pre-Inca Peru, a norm of generalized reciprocity
was widely diffused among both the local descent-groups and the regional
lords who provided the corvée labourers drawn from the descent-groups
with food, beer, cuttings, and seed. Under the institutions imposed by
the Inca, the agricultural surplus was systematically extracted for the
benefit of rulers holding an effective monopoly of the means of coercion.
But the rituals and routines of the local descent-groups were scrupulously
respected and the food and beer continued to be distributed to the corvée
The neo-Darwinian paradigm 49
labourers just as if nothing had changed (Murra 1980). Manipulation of
this kind can extend also to evoked behaviour, as in the deliberate attempt
of the promoters of the Catholic Counter-Reformation to arouse the
desired emotions in susceptible congregations by vivid and sumptuous
sights and sounds9 in addition to formal readings or teachings or stand-
ard devotional rituals. But the effects of such manipulation will be what
the designers intend only if its environment is favourable to the mutations
in information affecting phenotype which the designers have introduced.
Attempts by the rulers of Soviet Russia to use purpose-designed material
structures to move the beliefs and attitudes of its citizens in the direction
that they wished were notably unsuccessful (Humphrey 2005).
11
It is an all too well-known impediment to the testing of rival hypotheses
in comparative sociology that controlled experimentation is impossible.
A biologist can knock out a chosen gene in an experimentally engineered
population of mice, but neither memes nor practices are susceptible to simi-
lar treatment. A sociologist can’t knock out of the Roman Catholic Church
the memes encoded in a papal encyclical in order to see what difference
that would have made to the behaviour of the faithful or knock out the
practices encoded in an amendment to the American Constitution in order
to see what difference that would have made to the behaviour of American
judges and politicians. At best, the evidence for some specific cultural or
social innovation, or some specific change in the ecological or demographic
environment, will enable its impact to be traced to the point of suggesting
a plausible answer to the kind of ‘what if?’ question which comparative
sociologists would always like to be in a position to answer. For example:
an observer of the Kachin of Upper Burma who saw how their conical
clan structure and communal ownership of land in an environment where
population growth came regularly up against the ecological barrier of the
fallow cycle generated a succession of regular transitions from egalitarian-
ism to hierarchy and back again (Friedman 1975) might well wish to engin-
eer the transfer of some segment of the Kachin population into a different
environment in order to see what difference it would make to the practices
Responses evoked by the manipulation of light have a history which goes back from the massed
9
searchlights of Nazi party rallies through the oil-lamps in the basilicas of early medieval Gaul
and the oil-lamps and pine torches ‘casting all manner of shadows and reflections inside the
gloomy Mithraea’ of the Roman Empire (Clauss 2000: 125) to the tallow lamps flickering on the
drawings on the walls of the Altamira and Lascaux caves.
50 The theory of cultural and social selection
defining the roles constitutive of their modes of production, persuasion,
and coercion. In the event, that is just what happened when some Kachin
groups moved down from the highlands into the plains of Assam: they
evolved into small, class-based states. Similary, the Maori emigrants from
the same founding population who colonized the Chatham Islands pro-
vide a ‘natural experiment’ (Diamond 1997: 55) in which the climate and
ecology of the Chatham Islands gave the Moriori, as they were called, no
option but to revert from farming to hunting and gathering.
There is also the occasional possibility of following culturally dissimilar
populations into near-identical environments. Richerson and Boyd (2005:
21) cite a study by the rural sociologist Sonya Salamon (1985) as an exam-
ple whose ‘essential elements’ make it tantamount to a ‘real experiment’.
The descendants of German-born immigrants who settled in a part of
Illinois subsequently settled also by immigrants from other parts of the
United States preserved a distinctive set of cultural traits despite there
being no difference between the adjoining areas in the quality of the soil.
The German-Americans retain what Salamon calls the ‘yeoman’ values
which lead them to pressure their children to remain on the land and to
mix arable farming with labour-intensive dairying and livestock rearing
in a way which supports a larger number of family members on a more
limited acreage. Notice that the contrast between them and their more
profit-hungry ‘Yankee’ neighbours does not merely illustrate the adapt-
edness of a pattern of acquired behaviour not modified by a shared eco-
logical environment which might have been expected to evoke a common
set of strategies. It also illustrates the capacity of an acquired behaviour-
pattern to resist the selective pressure of a capitalist mode of production
which would impose on the German-Americans the same practices of
selling and renting as those approaching fixation among the ‘Yankees’.
A further difficulty in assessing the relative evolutionary force of biolog-
ical, cultural, and social selection is the timescale required for the genetic
composition of a human population to undergo significant change. Recent
research suggests that it may be shorter than used to be thought. But there
is one well-known population where for many generations strict endogamy
has been combined with higher birth-rates among families of relatively
high status, extensive nepotism and in-group altruism, religious and lin-
guistic separatism within which proselytizing is targeted principally at the
rich and talented, high levels of parental investment and control, cultural
selection against apostasy, social selection of practices defining intellec-
tually demanding academic and mercantile roles, and a hostile external
environment which constrained participation in alternative activities
The neo-Darwinian paradigm 51
and reinforced in-group conformism: the Jews (MacDonald 1993). As
MacDonald is well aware (1993: viii), ‘Jewish history … has been to a con-
siderable extent a social construction performed by highly interested par-
ties bent on vindicating very basic moral and philosophical beliefs about
the nature of Judaism, Christianity, and gentile society generally.’ But evi-
dence for the part played by natural selection in the cultural and social his-
tory of Judaism has become stronger since then, and at the time of writing
the hypothesis of a three-way interaction, although not definitively con-
firmed, looks increasingly plausible. Ashkenazi Jews, in particular, whose
low rates of marrying out are confirmed by genetic analysis, are known
both to score above the mean level in intelligence tests and to be likelier
than others to suffer from genetic diseases such as Tay-Sachs which can,
along with other symptoms, improve the linkages between brain cells. If
this does turn out to be what has been going on for over 2,000 years, there
could hardly be a clearer illustration of how genes, memes, and practices
can co-evolve within the same population to distinctive effect.
S e l e c t ion of w h at ?
12
The question of just what the process of selection is actually selecting has
been controversial since Darwin’s day. It makes equal sense to say that it
selects phenotypes (the Galapagos finches with the longer beaks), pheno-
typic effects (the beaks themselves), or genes ‘for’ longer beaks. In the
wake of the ‘new synthesis’ of evolutionary theory and population genetics
and the subsequent discoveries of molecular biology, it has become widely
accepted that genes and their interaction with their own immediate envir-
onment are the answer which best helps to explain just what is going on.
But genes work in different ways – they can split or overlap or be nested
in one another – and other mechanisms besides replication of DNA are
involved in natural selection, including ‘epigenetic’ inheritance involving
cells with the same genotype but different phenotypes which transmit
their phenotypes to their descendants (Jablonka and Lamb 1995).
In the study of cultural selection, argument has largely revolved around
the requirement that memes, as presumptive replicators, should be dis-
crete units capable of copying themselves independently of the phenotypic
effects through which their adaptedness is decided. But sociologists have
no need to be committed to the view that ‘genuine’ replicators must cre-
ate strictly accurate copies of themselves and persist intact for a sufficient
52 The theory of cultural and social selection
length of time to be able to do so. In the first place, in both cultural and
social evolution, selection is often blended rather than particulate;10 and in
the second, both memes and practices are continually modified through-
out the lives of the individuals who carry the reinterpretable memes in
their minds and the renegotiable practices in their roles. Moreover, evolu-
tionary sociologists are free to identify different combinations of informa-
tion at different levels of complexity as objects of selection in the same way
that evolutionary linguists can identity phonemes or syllables or words
or even grammars as the objects of selection out of which languages are
formed, or evolutionary economists can identify organizational routines
or techniques of production as the objects of selection of which competing
business firms are the carriers. It might be argued that if information is
what is being selected it is by definition reducible to binary discrimina-
tions, and these are therefore ‘the’ units of mutation, transmission, and
selection alike. But to suggest that biologists, let alone sociologists, ought
to pitch their explanatory hypotheses at that level would be like suggesting
that chemists ought to pitch their hypotheses at the level of sub-atomic
particles. Identification of the critical memes and practices whose transfer
from carrier to carrier makes different cultures and societies into what they
are is not an exercise in particle-hunting. In the evolution of cultural arte-
facts, for example, the critical objects of selection are not ‘primitive’ units
of information but packages of ‘techno-memes’ (Stankiewicz 2000: 230)
assembled in a design space which generates the domain of possibilities
open to innovators searching for locally optimal solutions.11
It is possible also for the same information affecting phenotype to be
selected at either the biological or the cultural or the social level. In evo-
lutionary game theory, the object of selection is by definition a strategy.
Strategies come in many forms, and games are of many kinds. But con-
sider the classic ‘Hawk-Dove’ game analysed by Maynard Smith (1982) in
which, where animals are contesting for territory in a habitat favourable
for breeding, ‘Hawks’ escalate and continue until injured or the opponent
10
Where it is blended, the familiar image of the ever-branching evolutionary tree is no longer
accurate: it is then as if branches could grow into as well as out of one another, so that a better
image is accordingly a ‘braided stream, with different channels flowing into one another and
then splitting again’ (Shennan 2002: 84).
11
Techno-memes are sometimes, as in the early stages of industrialization in eighteenth-century
Britain, encoded not in any verbal or written instructions at all but in the ‘know-how’ of expe-
rienced artisans, with the consequence that they can be diffused only by the migration – which
employers may well seek to prevent – of the exceptionally skilled workmen who carry the know-
how inside their heads (Matthias 1979: Ch. 2). Cipolla (1972: 50–1) reports a case where some
Swedish workmen were literally kidnapped in the 1660s and transported in stages to France,
where Colbert wanted to start an iron industry on the Swedish model.
The neo-Darwinian paradigm 53
retreats while ‘Doves’ display but retreat at once if the opponent escalates.
In the real world, simple pairwise contests between animals who always
behave the same way are much less likely than contests in which there is
a wider choice of possible strategies, the contestants sometimes follow one
and sometimes another, and the payoff in fitness to the individual con-
testant depends not on the strategies of successive individual opponents
but on some average property of the population as a whole. But in the
context of comparative sociology, the interest of the strategies of ‘Hawk’
and ‘Dove’ is that they can equally well be observed in the behaviour
of human beings contesting for territory, and in such cases the resulting
behaviour-pattern can be either acquired or imposed rather than evoked.
To borrow from Avital and Jablonka (2000: 95), ‘The crucial point, there-
fore, is not the precise mechanism for acquiring a new preference or pat-
tern of behaviour, but the circumstances that allow the transmission of
this information in a way that ensures its transgenerational [to which add
lateral, as well as vertical] reproduction.’
13
The word ‘meme’ is disliked as much by the ultra-Darwinians on the one
side as by the anti-Darwinians on the other: to the ultra-Darwinians,
it implies a much closer analogy between natural and cultural selection
than can in fact be sustained, while to the anti-Darwinians it implies that
human beings are not as different from other species in their mental cap-
acities as they are. But once it has been recognized that culture can evolve
through heritable variation and competitive selection independently of
natural selection acting on genes, it is a matter of simple convenience to
have a word to stand for whatever items or packages of information make
up the messages transmitted from mind to mind through imitation or
learning by which behaviour in the phenotype is affected.12 ‘Meme’ also
has the advantage that in the years since Richard Dawkins coined it13 it
12
Richerson and Boyd (2005), who originally drafted their book using ‘meme’ throughout, then
decided to replace it with ‘cultural variant’ (although it survives on their page 244, where they
say that ‘Modern societies, by vastly enlarging the scope for nonparental transmission, have
also magnified the chance of choosing maladaptive memes’). But ‘cultural variant’ not only de-
emphasizes the critical connection with the idea of information transfer from mind to mind; it
also invites the response from anti-Darwinian cultural anthropologists that that is what they
have been studying all along.
13
The fundamental idea behind it was set out the year before by F. T. Cloak, Jr, who argued that
‘The survival value of a cultural institution is the same as its function; it is its value for the
survival/replication of itself or its replica(s)’ (1975: 72).
54 The theory of cultural and social selection
has displaced its competitors sufficiently successfully to be cited as itself
an example of an adaptive meme (Ball 1984). Provided, therefore, that
it carries no pre-emptive implication about just how cultural selection
works and acknowledges that memes are being continuously reinterpreted
and reconstructed in the course of their carriers’ development, sociolo-
gists have no good reason for refusing to accept it into their standard
vocabulary.
One plausible way of looking at memes is to see them as the software
of the hard-wired human brain. But although psychologists may one day
discover just what is going on inside the natives’ heads between the time
when information transmitted by imitation or learning is first heard or
read or inferred from observation and the time when it is acted out in
phenotypic behaviour, sociologists have for the time being to go on treat-
ing the brains of the members of the populations whose shared beliefs
and attitudes they are studying as black, or very nearly black, boxes.
The information has gone into the natives’ heads, and the consequential
behaviour is subsequently observed, but we don’t know just how infor-
mation received is turned into the acting-out of a collective behaviour-
pattern different from the one which preceded it, or just which are the
items or packages of information whose adaptedness has been critical.
There the natives are, sacrificing (let us say) to a tutelary deity to pro-
tect their community from harm. Assume that we can be sure that they
are not just unthinking conformists doing whatever their neighbours do,
or Machiavellian hypocrites in pursuit of personal advantage, and that
they are neither re-enacting an empty ritual for the benefit of tourists
or film-makers or doing what they do under compulsion from rulers or
their agents. The extended phenotypic effects of the memes of which they
are the carriers can be observed in the altars at which the sacrifices are
made, the music to which the prayers are set, the supplicatory gestures of
the officiating priest, and so forth. But we simply do not know just what
is going on as the interaction between the carriers and the environment
enhances or diminishes the adaptiveness of some relative to other memes
in the minds of the carrying population. We can only say that there is a
complex of co-adapted memes which has displaced its competitors and is
being acted out in the same sort of way that a machine acts out its soft-
ware program.
Suppose, to give a notional example, that a sociologist comes across a
culture in which the natives all carefully wash their hands before every
meal. (I say ‘notional’ not because table manners aren’t a serious topic
in comparative sociology, but because I don’t propose to enter into
The neo-Darwinian paradigm 55
serious discussion of them here. Hand-washing will reappear in passing
in Chapter 5.) It explains nothing to say that the natives carry inside their
heads a meme which enjoins the washing of hands before meals. One pos-
sible answer to the ‘just what is going on here?’ question is an instrumen-
tal, external, just-so story: the critical meme has the form of a conviction
that diseases caused by the ingestion of contaminated food14 can attach to
unwashed hands, and the natives therefore punish by ridicule, insult, and
refusal of commensalism anyone who eats without using the hand-basins
which are provided in every eating-place. But another possible answer is an
expressive, internal just-so story: the critical meme differentiates the ritu-
ally pure in-group from the stigmatized ‘other’ by symbolizing a notional
decontamination which has nothing whatever to do with physical health.
The least plausible hypothesis is individual trial-and-error learning: in
no culture do children work out for themselves that washing their hands
before meals is a good thing to do in the way that they work out how to
get their food into their mouths without spilling it all over their faces. But
some method has to be found without the benefit of controlled experimen-
tation to establish which just-so story is the right one.
This is not made any easier by the volume of information – the ‘memetic
bombardment’, as it is sometimes called – to which children in all cultures
are subjected from infancy onwards. Think how much information is
acquired by British children from parents, mentors, and peer-group mem-
bers about anything from the rules of ‘Noughts and Crosses’ to the tune of
‘Pop Goes the Weasel’. Neither of these is likely to feature prominently in
sociologists’ just-so stories of the evolution of the distinctive sub-cultures
among which the British population is distributed. But what about beliefs
about, and attitudes towards, the Russian Revolution or the life and say-
ings of Jesus of Nazareth? The problem is as familiar to historians study-
ing the influence of the printed word as it is to anthropologists studying
that of the oral traditions of pre-literate peoples or psychologists that of
rumours spread laterally across a population of contemporaries. Suppose
a historian of nineteenth-century Europe studying the evolution of a cul-
ture in which there has emerged a bourgeois norm of cleanliness discovers
that an etiquette-book which enjoins, among other things, the washing
of hands before meals has gone though innumerable editions over many
decades and has enjoyed a consistently high level of sales. What reason is
14
Food taboos have in fact been observed much more frequently in relation to meats, with their
inbuilt risk of pathogenic infection, than fruits (Fessler and Navarrete 2003).
56 The theory of cultural and social selection
there to suppose that the memes encoded in the etiquette-book explain
the conformity to the bourgeois norm of purity and not precepts being
inculcated in any case by both instruction and example down successive
generations of hand-washing bourgeois parents and like-minded school-
teachers? The historian Robert Darnton has remarked that it is impos-
sible ‘to argue from the sales patterns of books to the behaviour patterns
of human beings’ (1979: 40), and that ‘we still have only a vague sense
of how readers construed texts’ (1996: 184). Yet Darnton’s own research
has shown that successive editions of Diderot’s Encyclopédie sold well to
readers of the kind who voiced radical opinions in 1789. Nobody is going
to argue from this that without the Encyclopédie the French Revolution
(to which I return in Chapter 4) would not have taken place. But it is a
clear example of a time-lag effect where memes encoded in books find
subsequent phenotypic expression in an environment which has changed
in such a way as to enhance their adaptedness. Etiquette-books, like ser-
mons, tracts, manuals, newspaper editorials, or readings from holy writ,
may simply be repeating instructions with which those who receive them
are familiar already, and many recipients may be no more likely to fol-
low them in consequence than they were in any case. But the behaviour-
patterns which distinguish both literate and non-literate cultures from
one another would not be what they are if the memes encoded in their
written texts and oral traditions had no phenotypic effects whatever on
which selective pressure came to bear. As another historian has said of the
moralizing novelettes of Harriet Martineau, which in the estimate of her
publisher were each being read by some 140,000 people in the mid-1830s,
‘the message was so intrusive, the moral assumptions so pervasive, that
it is difficult to see how even the most casual reader could have escaped
from them’ (Himmelfarb 1984: 169). The trouble is that we have no way of
finding out in retrospect just how many readers’ behaviour was actually
modified in consequence. When Tom Paine’s Rights of Man in a sixpenny
format and Hannah Moore’s anti-revolutionary tracts priced at about a
penny were alike being bought in their hundreds of thousands, perhaps
their readers all agreed with the author already. Since we can’t experi-
ment, we just don’t know.
14
I have talked up to now about the transmission of memes from mind
to mind as if the concept of social, as opposed to individual, learning
was less problematic than it is. But as researchers into animal behaviour
The neo-Darwinian paradigm 57
well know, it is often difficult to distinguish empirically between ‘true’
imitation on the one hand and learning through some form of social influ-
ence on the other. From one point of view, imitation is a kind of learning:
the learner learns what to do by copying other people who are washing
their hands, reading encyclopaedias, and so forth. But then so can learn-
ing by instruction be said to be a kind of imitation, the difference being
that now the learner copies the behaviour by following the mentor’s verbal
or written instructions rather than mimicking the mentor’s behaviour dir-
ectly. ‘Do as I do’ is one form of instruction and ‘do as I say’ another. The
hymns sung by the church choir on Sundays may be sung either because
the singers sing what they have heard other people singing or because a
choirmaster has taught them how to read the printed score. A contradic-
tion may arise when, as often happens, instructions given by parents or
teachers are at variance with the behaviour of the parents and teachers
themselves: ‘You tell me to do what I’m told, but you often refuse to do
what you’re told yourself.’ But this is merely a special case of the compe-
tition between rival memes which is going on all the time inside every
native’s head. For the purposes of comparative sociology, the question is
why the winning memes which distinguish one culture from another at
population level are the winners, whether their reproduction and diffu-
sion came about by instruction from teachers to learners, by imitation of
role-models who are not even aware of being imitated, by a combination
of both, or by something in between.
The more important difference between imitation and learning is that
learning is notoriously costly. Individual trial-and-error learning is par-
ticularly so: that is why it has as little to offer comparative sociology as
it does, even though cultural evolution would slow to a halt without the
heritable variations which individual innovators provide. Cultural tradi-
tions will be more likely to persist unchanged among populations where
individual learning is costly, and in a population consisting entirely of
unreflective imitators and social learners there would be no escape from
cultural inertia. But social learning, even if more efficient than learning
by individual trial-and-error, still imposes a cost on both teacher and
pupil which learning by direct imitation of a mentor’s observed behaviour
does not. Evolutionary game theorists are able to model in detail condi-
tions under which ‘imitate the most successful strategy’ will itself be the
most successful strategy. There are obvious advantages to the naive indi-
vidual player in imitating perceived success rather than trying to learn
the programme that the most successful players are following, let alone
trying out a range of alternative possible strategies and seeing which one
58 The theory of cultural and social selection
does best. Under some conditions, formal instruction will despite this
be favoured over informal exhortation and simple example. But this is
likely to be in an environment of social as well as cultural selection, where
practices defining the roles of teachers remunerated by public or private
employers are transmitted down successive generations within institu-
tions where students, novices, or apprentices are being formally trained to
occupy specialized economic, ideological, or political roles.
The distinction between imitation and learning is particularly problem-
atic in relation to artefacts, given how difficult it can be to infer by simple
inspection the essential features of their design. They contain the critical
memes in the sense in which an instruction manual contains them in its
own different form, but not in the sense in which genes are contained in
organisms. The problems of reverse engineering faced by archaeologists
studying pots, pictures, monumental architecture, statuary, the lay-out
of houses and workplaces, gardens, weapons, musical instruments and so
forth may seem of only marginal relevance to comparative sociologists
concerned with large-scale differences between cultures and societies
of different kinds and long-term qualitative changes from one kind to
another. But all these are a part of what differentiates one culture or sub-
culture from another, and they all come into existence through the trans-
mission of information from mind to mind by imitation and/or learning.
So: just what is going on here?
The answer is that artefacts are extended phenotypic effects of techno-
memes which at the same time encode them. When archaeologists ‘dig-
ging for memes’ (Lake 1998) detect in the morphology of buried artefacts
a sequence of variations culminating in a locally stable design, they are
recovering from objects which are themselves a medium of transmission
information previously handed down by imitation or learning from one
to another generation of artificers. They are thereby enabled to under-
take an exercise in reverse engineering which reads back from the mater-
ial objects they have unearthed to the memes which were inside the heads
of their designers, whether the reading-back is an exercise in inferring the
assembly-instructions from study of a finished mechanical contrivance or
in inferring the allegorical symbolism from study of a finished work of art.
The important difference is between genuine imitation, whether by fol-
lowing what the mentor does or learning to do what the mentor says, and
uncomprehending copying. It is quite possible to envisage, for example, a
native artist copying pictures of the crucifixion of Jesus without the least
idea of what it represents and thereby unwittingly reproducing an image
which, when subsequently decoded by others, turns out to be adaptive in
The neo-Darwinian paradigm 59
an environment of active Christian proselytization. But this would be no
more than a special case of indirect transmission of information affect-
ing phenotype from mind to mind which in itself is unproblematic. The
printers of the bibles which the missionaries are distributing to the natives
don’t have to have marked, learned, and inwardly digested, as opposed to
merely seen, the words which they reproduced while their minds were on
other and lower things.
15
When the objects of selection are practices rather than memes, it is not to
be assumed that rules encoded in legislation about institutionalized roles
necessarily determine phenotypic behaviour on the ground. To quote the
medieval historian Chris Wickham, ‘we have to abandon the legal history
paradigm, and do it quite consciously. Instead, we have to confront the
harder talk of comparing, not rules, but the sets of local practices that
constituted whole societies’ (Wickham 1994: 207).
Since the carriers of the information affecting phenotype are now dyads
of interacting roles, it follows that the individual incumbents of both
must share not only an understanding of its meaning but an acceptance
of the inducements and sanctions which underwrite its implications for
how they are to behave towards one another. The landlords and tenants,
or priests and parishioners, or generals and soldiers have not only, as I put
it in Section 7, to know what to do in order to be what they are; they have
also to accept that the practices which make them so, renegotiable as they
may be, are not theirs to alter at will. Consider what goes on in a Catholic
country where some male adolescents grow up with the ambition of enter-
ing the priesthood. It is not enough for them to inform themselves about
Christian doctrine, study the text of the Bible, and adhere to the norms
of Christian conduct. Nor is it enough for them to learn what priests do
(how a mass is conducted, what is involved in administering extreme unc-
tion, and so forth). They have to be institutionally qualified and assigned
to a parish where their qualifications are accepted by their parishioners as
carrying the ideological authority of the Roman Church. Admittedly, the
doctrine of sanctity of clergy may lead ordinands to continue to regard
themselves as priests even if they so behave as to be defrocked: ‘Not all
the water in the rough, rude sea Can wash the balm from an anointed
king.’ But I cannot resist repeating (from Runciman 1983: 60) the Jewish
joke about the mother who is telling her friend about ‘my son the cap-
tain’: sure enough, there is the son parading around in nautical dress with
60 The theory of cultural and social selection
his talk of ‘abaft’ and ‘astern’ and ‘belay there!’; but the friend says, ‘So to
you, he’s a captain; to me, he’s a captain; but to captains, is he a captain?’
All the same, it is always possible that both cultural and social selection
are simultaneously determining the form taken by a collective behaviour-
pattern. Consider the practice of sharecropping. It is sometimes but by no
means always found in an exploitative relationship in which the landlord’s
control of the means of production imposes it on the tenant irrespective
of whether the tenant believes it to be advantageous or feels it to be fair.
In such environments, the adaptedness of the practice is a joint function
of the local ecology and the relative bargaining power attaching to the
landlord’s and tenants’ roles: in nineteenth-century France, for example,
landlords used sharecropping contracts not only to realize economies of
scale in livestock rearing but at the same time take advantage of family
labour at a time of rising wages (Carmona 2006: 237). Conversely, the bal-
ance of negotiating power can be in the sharecropper’s favour where able-
bodied young men work the land for elderly proprietors unable to work it
for themselves in exchange for not only a share of the produce but inher-
itance of the land itself on the proprietor’s death (Robertson 1980). But
cultural selection can at the same time be at work in favouring one rather
than other type of sharecropping contract. Richerson and Boyd (2005:
85), drawing now on a study by Burke and Young (2001) of Illinois farm-
ers, offer an example in which the division of the crop between landlords
and tenants converges on a small number of simple ratios which are freely
chosen through imitation or learning (or, it might be, individual learning
and rational choice). Variations of this kind are a clear illustration of how
cultural reinterpretations can influence the outcome of negotiations con-
ducted within an established social institution. Cultural selection is not
swamped by social selection simply because the transition from culture to
society generates inequalities of economic, ideological, or political power
attaching to interacting roles and systacts.
16
There remains the vexed question whether selective pressure bears only on
individuals as the carriers of information affecting phenotype, or whether
it bears also on groups as such. Discussion of this topic was for a long time
dominated by the arguments of biologists who were held to have conclu-
sively refuted the view that self-sacrificing altruism on the part of indi-
vidual members of a group, including limitation on number of offspring,
could be explained by its contribution to the greater good of the group
The neo-Darwinian paradigm 61
rather than its function in maximizing the prospects of reproduction of
the genes of the altruist (or the altruist’s kin). The revisionists did not,
however, argue that there are no conditions under which natural selection
operates on competition between groups rather than between individuals.
They only insisted that, as W. D. Hamilton put it in demonstrating for-
mally the characteristically greater power of individual selection for altru-
ism when grouping is random, or nearly random, ‘we have to consider
whether the population can get into the specified state, and, if it can,
whether its present trend will continue’ (1996 [1975]: 333). If, for example,
a species is divided into several separate populations established by dif-
ferent founders, then competition for territory and resources between
one group and another may be a stronger influence on the probability
of reproduction of their members’ genes than within-group competition
between individuals. But once there is migration between groups, indi-
vidual competition will again come to the fore. It would accordingly be
a mistake for sociologists to assume that at the biological, let alone the
cultural or social, level group selection has been shown to be incapable of
acting as a significant evolutionary force. On the contrary: there are good
reasons for thinking that intra-group cooperation was favoured by inter-
group selection throughout the entire course of hominid evolution.
The possibility of cultural group selection was explicitly envisaged by
Darwin himself in an often-quoted passage of the Descent (1882: 132). He
argued that groups with a higher proportion of members motivated by a
sense of ‘glory’ to sacrifice themselves for the good of the rest will have
a competitive advantage over groups with a lower proportion, by which
he appears to have had in mind (not that he could have put it that way)
gene-meme co-evolution in which the memes in the heads of the mem-
bers of the first group would make them more likely to reproduce their
genes successfully than the members of the second group. The time which
it will take for a group to be driven to extinction by its competitors is
difficult to calculate. Soltis et al. (1995), using data on local group extinc-
tions among the small warrior societies of Papua New Guinea prior to
pacification, estimated that without acceleration by individual choice it
might take a thousand years for the replacement of the relevant traits in
a metapopulation. But once cultural selection is operating alongside nat-
ural selection, migration between groups does not pose the same diffi-
culty for group selection, since imitation and learning enable migrants to
acquire the memes which are dominant in the group they have joined and
the cultural traditions of different groups are likely to be sufficiently sta-
ble for individual variation within them to be relatively low. Furthermore,
62 The theory of cultural and social selection
memes themselves can outreproduce their competitors through cultural
group selection whether or not they enhance the reproductive potential of
the population of the group.
At the level of social selection, competition between institutionally
structured groups such as armies or business enterprises or political parties
is so familiar to sociologists that the extinction of the losers could almost
be said to be a routine observation. It may be that the behaviour of the
winning group is no more than a statistical effect of the traits of its indi-
vidual members (the winning army wins purely because it has a larger pro-
portion of more courageous soldiers). But often it is because the outcome
depends on the successful integration of different packages of informa-
tion carried by different individuals or role-dyads which is acted out in
active collaboration consequential on communal decision-making. Many
traits predicable only of groups, but not of their individual members, can
make the memes and the practices carried by the winning group mutually
co-adaptive. Consider the account of the much-studied Nuer/Dinka rela-
tionship given by the anthropologist Raymond Kelly (1985), which Sober
and Wilson (1998: 191) cite as ‘a smoking gun of group selection in action’.
The domination of the Nuer over the Dinka, with whom they originally
shared a common ancestry and a common ecology, is attributed by Kelly
to a critical mutation in the nature and scale of bridewealth payments.
The difference in the minimum acceptable level of these was sustained by
phenotypic effects which extended to herd management, grazing require-
ments, and herd sizes. The tribes with comparatively high ideal bridewealth
payments were then those whose practices generated a form of military
organization which the Dinka were unable to counter. The outcome can-
not be explained either by demographic differences or by attributing to the
Dinka a reluctance to fight. If Kelly’s just-so story is the right one, then,
whatever scope there may be for disagreement over its details, there can be
no doubt that group selection through heritable variation and competitive
selection of both memes and practices was just what was going on.
The arguments over individual versus group selection are in some
ways reminiscent of the arguments over ‘Individualism’ versus ‘Holism’
on which much intellectual energy was expended by both sociologists
and philosophers during much of the twentieth century. In both cases,
the issues at stake were apt to be obscured by mutual accusations of bias.
Individualists could find themselves charged by Holists with condoning
the evils of capitalist property relations and encouraging the uninhibited
pursuit of self-interest, while Holists could find themselves charged by
Individualists with implicitly advocating totalitarian forms of government
The neo-Darwinian paradigm 63
and the curtailment of freedom. But ‘Individualists’ are not committed to
denying that individual behaviour has to be placed in its social context or
that groups have properties not predicable of their individual members,
and ‘Holists’ are not committed to positing metaphysical entities to which
are attributed properties of individual minds or to explaining the course of
social evolution by reference only to the properties of groups, communities,
and systacts (or even societies, empires, and civilizations). There is nothing
contentious in the observation that cooperative groups can secure for their
members advantages which the individual members could not secure for
themselves and that some groups not only exploit but alter their environ-
ments more effectively than others through what evolutionary biologists
call ‘niche construction’ (Laland et al. 2000), so that their members’ genes,
memes, and practices alike are all more likely to be reproduced – with the
continuing possibility of further mutations – as a result.
Lastly, there is no need for comparative sociologists to be concerned
over different definitions of what constitutes a ‘group’. Since institutional
practices depend on interaction between the incumbents of two com-
plementary roles, social evolution could be said to be a matter of group
selection by definition. But it makes no difference whether selection is
thought of as operating on populations of individual role-incumbents or
metapopulations of role-dyads. These are merely two different ways of
reporting the same thing. In the evolution of economically, ideologically,
and politically stratified societies out of hunting and foraging bands (or
‘band-clusters’), it is entirely plausible to suppose that ‘proto-households’
(Bergucki 1999: 152) competing with one another for resources were better
fitted to exploit their environment by risking-taking initiatives than were
individual hunters and foragers. In large and complex capitalist industrial
societies, technological innovations are subject to selective pressures from
the research community and the market which act on both the individual
traits of risk-taking technologists and on the structural properties of the
laboratories, firms, and institutes within which they work. Hitherto adap-
tive techno-memes and the practices associated with them can then be
driven to extinction either by the invasion of the team, laboratory, firm,
or institute by rival mutants or by the extinction of the laboratory, firm,
or institute through loss of funding or personnel. Whatever the right just-
so-story, it can be told either as a story about laboratories, firms and insti-
tutes or as a story about technologists. There is no need to complicate the
formulation and test of a proffered explanation by argument over whether
one way of phrasing it is to be preferred over another a priori.
Ch apter 2
Di sp o s i t ions, c a pac i t i e s, a n d s us c e p t i bi l i t i e s
1
Since we – human beings, that is – are the surviving one among many
hominid species all descended from primordial apes, it is not surprising
that the explanation of much of our behaviour can be traced as far back
as it can in our biological inheritance – with the corollary, among others,
that if we had been descended from monkeys we would have had female
kin-bonding in our evolved past of the kind which is lacking among apes.
The similarities between our behaviour and that of the chimpanzees in
the Gombe rain forest, the Arnhem zoo, and the Yerkes field station are
unmistakable: there they are, fighting, playing, imitating, seducing, griev-
ing, deceiving, collaborating, and showing off just as human beings in
all cultures and societies do. But how much does that help us to explain
the similarities and differences between cultures and societies with which
comparative sociologists are concerned?
The standard response of twentieth-century sociologists was to insist
that since our innate dispositions can be radically diverted, refined,
encouraged, modified, or suppressed in different ways in different groups,
communities, cultures, and societies, they are of no more relevance to
comparative sociology than that human beings in all cultures and societies
laugh, cry, yawn, dream, and express their emotions in facial expressions
which Darwin correctly hypothesized to be naturally rather than cultur-
ally selected (although they can, at the same time, be simulated with the
intent to deceive). But more to the point are first, the many examples where
deliberate attempts to modify behaviour – notably where the upbringing
of children is concerned – are frustrated by evoked responses which owe
nothing to either culture or society; and second, the many instances in
the ethnographic and historical record where a similar pattern of evoked
64
Natural selection and evoked behaviour 65
behaviour can be observed in widely different cultural and social envir-
onments. Cultural and social selection both drive different human
populations down one rather than another of a wide range of different
evolutionary pathways. But not even the most intransigent anti-Darwinian
can deny that there are some innate, species-wide dispositions, capacities,
and susceptibilities which impose some inescapable limits on the possible
extent of variation between one culture or society and another.
The paradox is that our biological inheritance includes both the innate
constraints which mean that not just anything is thinkable or doable,
any more than just anything is digestible, and the innate capacities which
enable us to display a far greater variety of behaviour than any other spe-
cies. As it has been aptly put by Ernest Gellner (1989: 516), ‘Question: how
can a species genetically granted by Nature such remarkable freedom and
licence, nevertheless observe such restraint, such narrowly defined lim-
its, in its actual conduct?’ The diversity is not merely between groups or
between individuals within groups but within individuals themselves, as
memorably described with characteristic insight by Montaigne in his essay
‘On the Inconstancy of our Actions’. The incumbents of even the most
tightly structured institutional roles are still the ‘irrational, disorderly,
unpredictable, and spontaneous’ persons which all of us are (Coleman
1990: 197). Yet the diversity of cultures is a diversity of stable combinations
of memes acquired by imitation and learning, and the diversity of soci-
eties is a diversity of stable sets of practices defining the roles constitutive
of the ongoing modes of production, persuasion, and coercion which are
comparative sociology’s stock-in-trade. How can this be? At population
level, evolution somehow favours relatively stable equilibria, and consist-
ency within a far from unlimited range of collective behaviour-patterns is
maintained over most of the world for most of the time.
2
Whatever surprises future research on the human mind may hold, the
distinction which folk psychology makes between attitudes and beliefs
(or, if you prefer, between passion and reason, or the heart and the head)
remains as self-evident in theory and indispensable in practice as it has
always been. Whatever the relation, under the heading of ‘attitude’,
between preferences, norms, and values, and whatever the criteria by which
the strength and consistency of beliefs are to be judged, a contrast between
the two is rightly taken for granted by sociologists of all theoretical persua-
sions. But that does not in itself make the relative stability of our collective
66 The theory of cultural and social selection
behaviour-patterns any easier to explain, since our brains have not evolved
in such a way as either to give our intellectual superegos mastery over our
emotional ids or, conversely, to subordinate reason to the point of becom-
ing, in Hume’s celebrated phrase, the ‘slave’ of the passions. It is rather a
matter of ‘the parallel operation of the emotional and the rational brain
within every human being’, with the ventromedial prefrontal cortex pro-
viding an interface between them (Massey 2002: 24). Nor has natural selec-
tion designed us to be as good as we might like to suppose at either logical
thought or consistent emotion: for all our psychological advantages over
our primate cousins, we are as careless in our reasoning – particularly our
statistical reasoning – as we are impulsive in our feelings. Yet the beliefs
(however weakly held) and attitudes (however erratically felt) which coex-
ist inside all of our heads work together in such a way that stable collective
behaviour-patterns persist over successive generations. However exactly it
came about, our ancestors’ larger brains with their proportionately larger
neo-cortices enabled them to become more skilled at interacting with one
another than their primate cousins and thereby preventing the aggregation
of unrelated strangers into larger and larger groups from leading to either
ungovernable anarchy or headlong fragmentation.
Thus despite our individual irrationality, disorderliness, unpredictabil-
ity, and spontaneity, the human species appears to have an innate predis
position to conformity. When a leading theorist of rational choice says
that the ‘main reason habitual behaviour permeates most aspects of life is
that habits have an advantage in the biological evolution of human traits’
(Becker 1996: 9), he is implicitly appealing to evolutionary psychology as
well as explicitly appealing, as he does, to Aristotle and Adam Smith. The
study of both the reasons and the passions continues to be fraught with
conceptual as well as empirical difficulties. But all human beings have them
both, and can recognize them in others. We all have a naturally selected
capacity for quick intuitive evaluation (Haidt 2001) as well as for ‘fast and
frugal’ (Gigerenzer and Goldstein 1996) ratiocination. Moreover, we are
all, whatever the culture in which we are reared, not only ‘theory-inventors’
trying to make sense of the world around us but ‘theory-retainers’ (Carey
1985: 914). Readiness to subject acquired beliefs to lengthy and careful test
is a late and limited product of cultural, not natural, selection. Although
we are all aware that other people cling with equal conviction to beliefs as
well as attitudes incompatible with our own, we are reluctant to take as
seriously as might be expected of us the possibility that they, not we, might
be right. Ask yourself how often you have ever actually won (or lost) an
argument with anyone about religion or history or politics. All beliefs are
Natural selection and evoked behaviour 67
rational to the people who hold them, however unconvincing their reasons
to the members of other cultures or other members of their own.1 Our
innate desire to be relieved from uncertainty makes us at the same time
explanatorily voracious, theoretically credulous, and obstinately loyal to
the fellow members of our local sub-culture.
Once, therefore, a package of co-adapted memes acquired, from
whatever source, by imitation or learning has made the workings of the
immediate environment more manageable, the natives in all cultures
and societies are predisposed to keep it in their heads, whether they are
Roman litigants convinced that their failure to win their case must be
due to their opponents’ recourse to a binding spell, or Shang Chinese
soothsayers heating animal bones in order to read the future from the
resulting T-shaped cracks, or Egyptian priests ‘partial to exegesis on the
worst possible etymological principles’ (Kirk 1980: 203), or chartists in a
stockbroking office claiming to have anticipated a downturn in the equity
market, or astrologers impervious to the suggestion that their occasional
successful predictions could be due to coincidence, or laboratory experi-
mentalists adept at finding reasons to ignore results which conflict with
their favourite hypotheses. As theory-retainers, we not only use reasoned
argument to justify rather than challenge convictions already held but
display ‘a prevalent, and disturbing, tendency to assimilate any new infor-
mation to existing theories’ (Kuhn 1991: 268). As is illustrated as well by
the belief in witchcraft among early modern Europeans as among Evans-
Pritchard’s Azande or the Etoro of Papua New Guinea or the popula-
tion of ‘ancient’ Rome, it is not intellectual argument which will of itself
drive to extinction the memes whose phenotypic effects the historian or
anthropologist has observed. ‘How can you behave as if you believed such
nonsense?’ is as ineffective a question addressed to worldly judges and
learned divines in a self-consciously sophisticated intellectual culture as
it is to illiterate peasants living at the margins of subsistence. The argu-
ments deployed against the persecution of witches in Britain by Francis
Hutchinson in 1718, which found ready acceptance, were much the same
as those offered by Thomas Ady in 1656 and Reginald Scot in 1584, which
had been treated as ‘part of the radical fringe’ (Bostridge 1997: 3). Not all
conformity to conventional wisdom is to be explained as directly evoked
by situational cues or motivated by an inbuilt desire to behave like other
That general statement is not invalidated by professions of ‘faith’. Credo quia impossibile is an
1
3
Some sociologists may still wish to insist that once cultural, and sub-
sequently social, selection are under way, natural selection can explain
nothing more about the differences with which comparative sociology
is concerned than the extent to which collective behaviour-patterns are
‘ultimately’, or ‘at the end of the day’, constrained by the design of the
human brain on one side and the drive to sustain inclusive reproductive
fitness on the other. They will refuse to accept that, to take an obvious
example, both a creed which enjoins holy war against all unbelievers and
a creed which prohibits the waging of war in any form are nothing more
than alternative expressions of the same innate predispositions evoked by
different local environments. True as it may be that the ‘so-called moral
sense’ is a product of natural selection and that what we call ‘conscience’
was biologically adaptive for our remote forebears, some human beings
think it a sacred duty for fathers to kill daughters who have dishonoured
their families by an illegitimate pregnancy while others think it a sacred
duty to refrain from killing not only fellow humans but animals likewise.
How can natural selection account for that?
70 The theory of cultural and social selection
The answer to that question is that no neo-Darwinian sociologist is
committed to claiming that it can. But research needs to be designed for
the purpose of finding out just how far the behaviour in question which
might, in the environment in question, be the acting-out of an innate
predisposition of a universal kind is or isn’t a product of natural rather
than cultural or social selection. If, for example, the members of a non-
literate population of hunter-horticulturalists in the Ecuadorian Amazon
perform no differently in a test of conditional reasoning from Harvard
undergraduates (Sugiyama, Tooby, and Cosmides 2002), are we to con-
clude that there is a similar cheater-detection mechanism in the natives’
heads in all the cultures and societies in the historical and ethnographic
record? A comparative study designed to test whether norms of fairness
and reciprocity override the pursuit of economic self-interest in the same
way in fifteen small-scale societies (Henrich et al. 2004) found that,
although there was much more inter-group variation than in experiments
with students as subjects, in none of them was the experimental behav-
iour consistent with the rational-choice model of the economics text-
books. When the one-shot, two-person Ultimatum Game (in which both
proposer and respondent earn nothing if the responder refuses the offer
of a proportion of a fixed sum of money) was played by the Machiguinga
of the Peruvian Amazon forest, the mean offer was 28 per cent as against
48 per cent among graduate students in Los Angeles, and Machiguinga
respondents almost always accepted offers below 20 per cent which the
Los Angeles sample almost never made in the first place. In Papua New
Guinea, by contrast, where a gift is likely to be culturally construed by
the recipient as a strategy designed to enhance the status of the donor
and place the donee under a potentially unwelcome obligation, generous
offers of more than half the money were often rejected; and where social
selection comes to bear, the offers of players in societies integrated into
the practices and roles of a market for goods and labour were likely to be
significantly higher than the offers of players who were not. The inescap-
able conclusion is that the different degrees of selfishness and unselfish-
ness displayed by the members of different populations in experiments
which model real-life strategic choices reflect the simultaneous workings
of natural and cultural and social selection.2
cooperators) among university undergraduates in sixteen countries in which it was found that
‘antisocial punishment is harsher in participant pools from societies with weak norms of civic
cooperation and a weak rule of law’ (Hermann et al. 2008: 1366).
Natural selection and evoked behaviour 71
Be h av iou r a l u n i v e r s a l s
4
Disagreements about the scope, nature, and significance of presump-
tively innate species-wide determinants of behaviour are a longstanding
feature of the literature on so-called ‘universals’ observable in any and
all cultures and societies. No comparative sociologist will dispute that
there are some behaviour-patterns which are common to all members of
the human species – food is procured, infants are nurtured, languages
are learned, and utensils are made. But what does this tell us about the
relative strength of the forces of natural, cultural, and social selection?
A universal behaviour-pattern may or may not be innate, and an innate
behaviour-pattern may or may not be universal.
Surprisingly long lists of supposed universals are compiled from time
to time only to invite the objection that unless they are made to hold
good by pre-emptive definition there are always exceptions to be found
somewhere in the ethnographic or historical record – or, alternatively,
that where they do hold good, they are of no more than trivial interest,
since what matters is the difference in the ways that different people in
different cultures and societies do these commonplace things. But the
universal most important for comparative sociology is the one which
sociologists (unlike both brain scientists and philosophers3) most often
take for granted: the universal comprehensibility of the terms in which
the members of different human populations represent themselves as
such. No anthropologist has ever returned from the field reporting the
existence of a human community about which nothing more can be said
than that its members interact with one another in an apparently non-
random yet wholly baffling way. Not only can all physiologically normal
human beings learn to speak a language, but they can all learn to talk
to each other between as well as within languages about their different
culturally transmitted beliefs and attitudes, contested as these may then
turn out to be.4
3
Wittgenstein, for all his insistence on the distinctiveness of different ‘forms of life’, is explicit that
a researcher in a strange country with a wholly alien language who wonders how to tell whether
the natives are really giving, receiving, understanding, obeying, and rebelling against orders has
to rely on mankind’s universal mode of behaviour (Handlungsweise) as the system of relevance
(Bezugssystem) for doing so (Philosophical Investigations §206).
4
To go so far as to say that ‘any language can be used to convey any proposition, from theological
parables to military directives’ (Pinker 2002: 37) is to invite Evans-Pritchard’s rhetorical question
‘How do you render into Amerindian language “In the beginning was the Word”?’ (1965: 14). The
72 The theory of cultural and social selection
It is therefore not only an ethnographic but a biological fact that whatever
items or packages of information find phenotypic expression in the cultural
or social behaviour-patterns of any human population chosen for study,
they can in principle be decoded, whatever the mechanism or the route
by which they have been evoked, acquired, or imposed. In practice, it may
be that they can’t, whether because of aberrant genes inherited or brain-
damage suffered by the chosen native informant or for the social, rather
than (or perhaps as well as) cultural, reasons for which some ‘secret soci-
eties’ do indeed remain secret. Rare as they may be, there are cases of indi-
vidual linguistic behaviour which is incomprehensible alike to the visiting
observer and to the fellow members of the speaker’s own culture (and may
well, therefore, lead to the speaker’s behaviour being clinically explained as
‘pathological’ and pejoratively described as ‘mad’). But consistent patterns
of collective behaviour are never incomprehensible in this way. Distinctive,
culturally selected complexes of representations, attitudes, and beliefs, like
distinctive, socially selected modes of production, persuasion, and coercion,
presuppose both a shared underlying ontology and a shared underlying
acceptance of meaning without which the necessary minimum of consist-
ency and stability of behaviour wouldn’t be achievable at all (‘understand-
ing’ in the descriptive, phenomenological sense of ‘what it really feels like’
to be one of ‘them’ being, as always, a separate but not unrelated matter).
Dispositions, capacities, and susceptibilities do not, however, find
expression of their own accord in solitude. It is another universal that they
have to be activated by social contact. Just how this comes about is a con-
troversial question in cognitive science, where some psychologists assert
but others deny that ‘protoconcepts’ are innately encoded in the neural
structure of the human brain. If our brains are modular rather than gen-
eral-purpose in design, locally variable behaviour-patterns may be more
likely to be evoked than acquired. But sociologists do not need to take
sides on this contested issue. They need only to recognize that there are
some universal understandings of number, time, space, movement, and
physical bodies, some conceptual universals in interpersonal language,
and some universal intuitions about purposive artefacts. There are also
some universals in ‘fundamental motivational mechanisms’ (Heckhamen
and Schultz 1999) and ‘behavioural development’ (Bateson and Martin
answer is that of course you can’t do it in one move. But languages can be extended to incorpo-
rate parts of other languages – it happens all the time – and in due course Amerindian speakers
can be taught to understand ‘In the beginning was the Word’ as well (or badly) as English speak-
ers can be taught to understand ‘EN APXH HN O ΛΟΓΟΣ’ or, for that matter, the vocabulary
of Amerindian cosmology.
Natural selection and evoked behaviour 73
1999), and a universal capacity for ‘empathy’ (Hoffman 1981). Moreover,
just as there are some behavioural universals whose sociological relevance
is undeniable, so are there some conceivable behaviour-patterns which are
nowhere to be found in the ethnographic or historical record. There are
no human populations in which infants are all freely traded as market-
able commodities any more than there are populations where Hamilton’s
Rule is stood on its head – unconditional cooperation with outsiders but
ruthless egoism towards close genetic kin.
Anthropologists are, as a rule, more disposed than other behavioural
scientists to insist on the magnitude of the cultural differences which our
shared biological inheritance permits. Look (they will say) at how differ-
ently time itself is conceptualized by the Nuer or the Mursi or whoever by
comparison with ‘us’ and our clocks and calendars and schedules and rou-
tines geared to the exigencies of our parochial social institutions, or how
differently space is conceptualized not only by people who live in differ-
ent ecological environments (forest, jungle, savannah, desert) but by people
with different poetical and mythical traditions, or how differently humanity
itself is conceptualized in animistic or totemistic cultures from a ‘naturalis-
tic’ culture like the anthropologist’s own (Descola 2005). There have been,
and still are, very many people in the world whose ideas about, and behav-
iour towards, what ‘we’ call ‘nature’ are radically different from our own,
and to whom the distinction deployed in this book between the natural,
the cultural, and the social would be entirely alien. But anthropologists are
equally concerned to insist on the common humanity of themselves and the
peoples whose beliefs and attitudes they document. This can sometimes look
uncomfortably like having it both ways, particularly when the aim in view is
to privilege the culture being studied at the expense of the anthropologist’s
own. But there need not be any inconsistency. Reportage, let alone explan-
ation or description, of the diversity of beliefs and attitudes documented in
the ethnographic record depends on the application of terms which reflect
the universal biological inheritance which makes it possible to report them
in the first place. The more the culture which the anthropologist has been
studying is different from the anthropologist’s own, the more the anthro-
pologist’s success in elucidating the difference testifies to an innate species-
wide psychological inheritance shared between ‘us’ and ‘them’.
5
Unarguable as the evidence is for a universal human psychology, the
equally unarguable range of cultural variations which it makes possible
74 The theory of cultural and social selection
continues to pose some longstanding conceptual questions which are no
less problematic for selectionist sociologists than for any others. How far
is ‘love’ as observed in ancient Athens the same as in present-day Beijing?
How far is ‘anger’ as observed among the Hopi Indians the same as among
the English bourgeoisie? And so on. The languages of different people are
full of words for emotions which are unique to themselves: what, to go no
further afield, is the English equivalent of the German angst? But neither
the differences in the way in which what ‘we’ call ‘love’ and ‘anger’ are
conceptualized nor the conventions governing their appropriate behav-
ioural manifestations in different cultures entail the conclusion that there
is nothing which deserves to be called ‘love’ or ‘anger’ to be found inside
the heads of the natives in all cultures and societies. It is necessary only
to recognize that ‘If we want to posit universal human emotions, we must
identify them in terms of a language-independent metalanguage, not in
terms of English folk words for emotions’ (Wierzbicka 1992: 120). There
is no reason for selectionist sociologists to deny that there are local eth-
nopsychological variations in meaning (Mallon and Stich 2000) or that
love and anger are differently manifested in outward behaviour, as well
as differently classified and prioritized, in different human populations.
But however fluid the vernacular terminology, and however ambiguous
the exchange of phenomenological descriptions of subjective experience
between the members of different cultures, no anthropologist has ever
found one so peculiarly distinctive that neither love nor anger can be
attributed to the natives unless those words are given a meaning applica-
ble to nobody other than themselves. Love and anger are demonstrably
being felt by infants still too young for memes acquired by imitation and
learning, let alone institutionally imposed practices, to account for it, just
as they demonstrably make and act on presumptive connections between
things observed in the world whereby one occurrence is held to be pro-
duced by another, antecedent one. In no physiologically normal infant’s
brain is the amygdala not activated by whatever in its environment evokes
a response recognizable as fear. Nor is there anywhere in the world where
children are so reared that all of their behaviour is either solipsistically
utilitarian or else part of an indefinite series of existential actes gratuites.
Beliefs and attitudes which are susceptible to intersubjective under-
standing, intrinsically connected to one another, and acted out in distinc-
tive collective behaviour-patterns can now be detected inside the natives’
heads in any and all human populations not only through their verbal
reports which anyone who has learned their language can understand
but by the application of neurological and chemical techniques such as
Natural selection and evoked behaviour 75
positron emission tomography which locate them in the brain. Such tech-
niques will not by themselves resolve the difficulties of comparative soci-
ologists confronted with behaviour which raises the question whether, as
one anthropologist has put it, ‘a forgotten message, or a sign that betrays
some meaning, or a message that no-one understands, is an example of
communication, in any strict sense. To say that people in the society con-
tinue to do it because they feel an obscure appropriateness or satisfaction
in what they do is to speak in terms of stimulation and response rather
than in terms of communication and message’ (Lewis 1980: 37). But in all
cultures, whatever exactly is going on inside the natives’ heads, the memes
whose phenotypic effects have been acted out are not totally undetectable.
The observer of an initially baffling behaviour-pattern is still able to infer
something about the attitudes to which it gives expression, and to deci-
pher something about the beliefs which it implies – which is just what
Lewis himself proceeds to do for the Gnau of New Guinea whom he has
been studying in the field. The observer may conclude that the appro-
priate parallel is with musical rather than verbal meaning (Baranowski
1998), or that if the natives are unable to give an answer when asked for
a specific symbolic meaning it is precisely because a verbal expression of
its meaning is ‘not necessary to the communicative function of the ritual
acts’ (Arno 2003: 808–9), or that the meaning to be derived from the con-
catenation of the significata of a ritual is so abstract, complex, and at the
same time emotionally charged as to be ‘ineffable’ (Rappaport 1999: 256).
But in none of these cases are the anthropologists telling their readers that
the natives’ behaviour has no meaning whatever. Their own interpreta-
tions of the evidence they report are themselves evidence for the universal,
innate human capacity to go beyond information explicitly given – not
(pace Steven Pinker) to the point that its meaning can be extracted and
translated directly and in full into the anthropologist’s own language, but
to the point that the anthropologist can, by talking to the natives in their
own language, establish what additions to the anthropologist’s own lan-
guage that would require.
6
The innate human capacity for calculation of individual self-interest con-
tinues likewise to be problematic for comparative sociology. But that is
not because to attribute rationality to actions is of no explanatory value so
much as because some sociologists think that other sociologists think that
the appeal to ‘rational choice’ explains much more than it actually does
76 The theory of cultural and social selection
(Zafirovski 1999). Rational choice theory has even been criticized as ‘pre-
Darwinian’ in its explanatory logic (Vanberg 2002: 25), and as having an
‘ontological framework’ into which the well-attested phenomenon of fellow-
feeling cannot be fitted at all (Sugden 2002: 63). Evolutionary game theo-
rists can demonstrate formally how, for example, strategies of commitment
can escape extinction even where they fail the test of modular rationality
(Skyrms 1996: 44), and behavioural game theorists can supply convincing
experimental evidence for systematically irrational choices (Camerer 2003).
Brain scientists applying transcranial magnetic stimulation to players of
experimentally constructed games can see just what is going on inside the
players’ heads when norms of fairness override calculations of pecuniary
self-interest. But it is as plain that human beings in all cultures and socie-
ties have an innate capacity for making choices in accordance with their
assessment of the probability of more or less satisfaction to themselves from
the outcome as it is that the individual choices which they make are often
driven by motives which can be labelled ‘rationally self-interested’ only by
stretching the definitions of ‘rationality’ and ‘self-interest’ so widely that the
differences between memes of the one and the other kind have to be rein-
stated within it in other terms.5 It does not matter for the purposes of com-
parative sociology whether the competing memes are then categorized as
‘rational’ versus ‘normative’, or as rational in one sense (‘purpose-rational’)
as opposed to rational in another (‘value-rational’). Selective pressures some-
times favour the one and sometimes the other, and both result from what
natural selection has done for our brains.
It is only to be expected, therefore, that the ethnographic record will
be full of cultures where there is no ‘functional or utilitarian reason’
which can explain, for example, the plants and trees used in the mak-
ing of Wala canoes (Tilley 1999: 109).6 But it is equally to be expected
that there are cultures in which ‘people approach religion in the way that
they approach other objects of choice. They evaluate its costs and benefits
and act so as to maximize their utility’ (Iannacone 1995: 172) – a claim
supported, in the case of the United States of America, by the evidence
for hypotheses derived from rational-choice theory on denominational
mobility, conversion ages, and patterns of denominational intermarriage.
5
Yair (2007) argues that James S. Coleman, although a leading proponent of rational choice the-
ory as the paradigm for sociological explanation, relies in his own empirical studies on expressive
motivation to impel rational agents to seek to maximize their short-term utility.
6
In the archaeological record there are similar examples, as in pre-Hispanic Andean metallurgy,
where gold and silver were incorporated into the body of an object at high cost in time and labour
instead of being plated on its surface (Dobres and Hoffman 1994: 218).
Natural selection and evoked behaviour 77
No comparative sociologist seriously supposes either that the choice of
materials for building canoes is nowhere made for functional or utilitar-
ian reasons at all, or that religious affiliation is everywhere a matter of
cost-benefit calculation alone. Nor, in any case, is it always rational to
act ‘rationally’: if playing to win leads to peer-group hostility and insti-
tutional demotion, is it not rational to play (sometimes) to lose? It can be
just as mistaken for sociologists to assume that the information affecting
phenotype to which a particular environment gives selective advantage
must be of the form presupposed by rational choice theory as to assume
that it can’t be when, however indignantly the natives themselves may
deny it, that is just what it turns out to be.
7
What, then, about the innate disposition to ‘probe the social order for
weaknesses and look for openings to improve one’s standing’, as Frans
de Waal puts it (1996: 102)? Before the transition from culture to society,
the ‘social order’ is a system of interpersonal ranking only. But the motiv-
ation to rise in rank within such an order can be just as strong as in a
system of institutional roles and inter-systactic conflicts. If the sociology
of our Palaeolithic ancestors was similar to that of the ethnographically
documented hunting and foraging bands, self-aggrandizers were held in
check by joking, teasing, enforced sharing, vigilant monitoring, counter-
dominant coalitions, and occasional assassinations. But it does not follow
that would-be self-aggrandizers were not present in every successive gen-
eration (Hayden 1998: 18). Leaders were unable to turn themselves into
rulers, but that is not a reason to assume that they would not have done
so if they could. The behaviour which restrains the aggression of the self-
aggrandizers is itself aggressive behaviour.
It is not impossible that natural selection worked to reduce the repro-
ductive success of the self-aggrandizers. If the resentment which they pro-
voked led to the formation of coalitions whose members cooperated in
punishing them, and if the punishment took the form of depriving them,
whether by assassination, exile, or denial of access to nubile women, of the
opportunity to have children, then perhaps ‘the imperative for authori-
tarianism was somehow cleansed from the genes of our pre-human ances-
tors, only to be revived in relatively recent history as a cultural adaptation
to the need to domesticate plants and animals in response to popula-
tion pressures’ (Binmore 2001: 162, citing Knauft 1991). There was ample
time during the long millennia of the Pleistocene for significant changes
78 The theory of cultural and social selection
in the human genome. But how then account for the evidence of the
archaeological record which overwhelmingly supports the conclusion that
‘hereditary social status will develop everywhere the economic and social
circumstances will allow it’ (Maschner and Patton 1996: 101)? It does not
require market towns or agricultural estates or public works. Sustainable
aquatic resources are quite enough, as illustrated by the Kwakiutl of the
Northwest Pacific Coast with their chiefs and slaves and their celebrated
‘potlatches’. It is true that the transition from culture to society need
not lead to the emergence of a role to which there attaches the quasi-
despotic power of an alpha male chimpanzee. There are ‘covenants with-
out the sword’ (Ostrom et al. 1992). Cooperation can be imposed among
competitors for limited resources by a role whose authority derives, in
game-theoretic terms, from the addition to the players’ decision-free of a
parameter representing the cost of self-financed monitoring and a strategy
for negotiated agreement (Ostrom 1990: 16). Nor need females be carriers
of the same strategies as males. But that still does not make Palaeolithic
sociology a just-so story of gene-meme co-evolution in which the previ-
ously inherited disposition to ‘look for openings to improve one’s stand-
ing’ was bred out of our ancestors only to be culturally reselected many
millennia later.
Although we cannot be sure that the results of present-day psycho-
logical experiments are a reliable guide to what was going on inside our
Palaeolithic ancestors’ heads, one of the first things to look for would be
an innate desire to punish cheats who flout the terms of an implicit social
contract (Cosmides and Tooby 1992; Fehr and Gachter 2002). Among
the members of hunting and foraging bands in continuous face-to-face
contact with one another, a mutation which turned anger into righteous
anger could hardly fail to be adaptive. Furthermore, whenever our ances-
tors first began to speculate about the existence of supra-human agencies
responsible for the workings of the world in which they found themselves,
the idea of reciprocity could readily be projected from the inter-human
onto the human/supra-human level. Propitiation, compensation, and sac-
rifice are all deep in our evolutionary past (Burkert 1996), and could be as
readily evoked in reaction to what was perceived as interference in human
affairs by presumptive supra-human agencies as by the behaviour of fel-
low human beings, whether the supra-human agents are thought to be
well-disposed towards human beings or not (or, as ‘Dualists’ have not
unreasonably hypothesized, the world is a battleground in which forces
working for harmony and good are fighting it out with forces working for
chaos and evil).
Natural selection and evoked behaviour 79
The biological inheritance of the hunters and foragers of the Pleistocene
did not, therefore, make their interpersonal relationships a world of
endemic mutual hostility and physical violence any more than of continu-
ous cooperation and undisturbed harmony. But their relationships were
no more relationships of strict equality than their motives were authentic-
ally egalitarian. Enforced sharing might eliminate virtually all differences
between them in material possessions, but an individual marginally disad-
vantaged in access to food could suffer as badly as a member of an impov-
erished underclass in a stratified society. Differences in personal prestige
might be slight and evanescent, but an individual consistently less well-
regarded than others could suffer as much in disesteem as the member of a
socially stigmatized caste. Counter-dominant coalitions might prevent any
individual from dominating the rest of the band by force or the threat of
force, but a deviant singled out for punishment could be as unpleasantly
treated as one institutionally convicted of criminality in a society with law-
courts, police, and prisons. The standard measure of inequality – the Gini
coefficient – can be as high in a group in which one person alone has a
symbolically valued possession – say, a gold necklace or a jade axe – as in a
market society much of whose abundant wealth is concentrated in the own-
ership of a relatively small number of exceptionally rich families or house-
holds. Contests between cheats, defectors, free-riders, and self-aggrandizers
and the strong reciprocators determined to punish them are waged at all
levels of inequality and in all ecological and demographic environments.
E volu t ion a r y a da p t e dn e s s
8
The environment in which our Pleistocene ancestors hunted and foraged
for many millennia was anything but uniform (Foley 1996; Irons 1999).
Far from there being a stable ‘Environment of Evolutionary Adaptedness’,
there was continuous modification of the ancestral environment both
by large variations in climate and ecology and by the niche-construct-
ing behaviour of our ancestors themselves. The successive mutations in
the design of their brains gave them the capacity to react to dangerously
unpredictable environments in whatever ways might enhance their repro-
ductive fitness in face of them. Thereafter, not only do biologically inher-
ited dispositions, capacities, and susceptibilities persist under conditions
in which they may have ceased to be adaptive, but they become adaptive
for other reasons and in other ways.
80 The theory of cultural and social selection
Consider the example of the ‘sweet tooth’. It is often cited as an example
of a physiological trait which, having evolved in an environment where it
was manifestly adaptive as promoting ingestion of a nutritional source of
energy, survives into a different environment where it ceases to be so. It
has not yet been explained as authoritatively as adult lactose absorption
has been. We do not know just when, where, or why it evolved. But once
it had, its interest to comparative sociology lies in, among other things, its
connection with the institution of slavery. ‘Who says sugar, says slavery’,
from the Normans and Venetians in Sicily, Cyprus, and Crete through
the Portuguese colonization of the Atlantic Islands to the point where, in
the mid-eighteenth century, 1.4 million of slaves in the Americas – 40 per
cent of the total – are to be found working on sugar plantations (Klein
1986: 60).7 Culture was part of the story too, particularly in the fast-
spreading fashion for drinking chocolate (and also the taste for drink-
ing rum). But the prodigious demand for sugar in Europe was an evoked
response traceable to a biologically inherited trait which was then, in the
environment of much later times, reinforced by cultural selection and sat-
isfied by the social selection of the practices constitutive of the institution
of slavery (about which I shall have more to say in Chapter 4).
Another example is the universal response to rhythm. The sensations
evoked by the beating of drums, the chanting of songs in unison, the
propulsion of boats or canoes through water by coordinated rowing or
paddling, and the performance in step of dancing or marching are docu-
mented across an extremely wide range of places and times. But in the
context of warfare between the infantry armies of the European states
of the seventeenth century, ‘Military drill, as developed by Maurice of
Nassau and thousands of European drillmasters after him, tapped this
primitive reservoir of sociality directly’ (McNeill 1982: 131),8 with the
consequence that group selection favoured those armies of foot-soldiers
which parade-ground training had welded into more effective fighting
units than their competitors. As a just-so story so phrased, this one has
7
The United States is, as its historians well know, very different in this respect: ‘Even at its antebel-
lum peak, sugar was never more than a minor Southern crop that utilized less than 10 per cent
of the slave labor force’ (Fogel 1989: 29). It does not follow from ‘who says sugar says slavery’ that
‘who says slavery says sugar’. Mines are as important to the history of slavery as plantations. But
it is a plausible conjecture that if the Romans had known about sugar they would have used slaves
to produce it, and rich Romans would have consumed it in quantity.
8
McNeill also quotes from the Maréchal de Saxe: ‘I shall be told, perhaps, that many men have
no ear for music. This is false; movement to music is natural and automatic. I have often noticed,
while the drums were beating for the colours, that all the soldiers marched in cadence without
intention and without realizing it. Nature and instinct did it for them’ (1982: 133 n.12).
Natural selection and evoked behaviour 81
to be modified to allow for the objection that foot, as opposed to arms,
drill could not become institutionally established until the era of arti-
ficially levelled parade grounds and macadamized roads: seventeenth-
century armies were under much the same environmental constraint
as the proverbially militaristic Spartans had been who, although they
advanced into battle in close formation and were trained in quite com-
plex infantry manoeuvres, weren’t, and effectively couldn’t be, taught to
manoeuvre strictly in step given the nature of the terrain on which they
fought. But that is not an argument against the universal bonding effect
of close-order drill which McNeill himself experienced as an infantry
recruit in the Second World War.
The possibility of deliberately exploiting innate dispositions and sus-
ceptibilities has long been recognized by the people who do it. Orators
are often aware of their ability, as Albert Speer put it in reminiscing
about Hitler and Goebbels, ‘to unleash mass instincts’ (1970: 47), and the
success of professional teachers of rhetoric from the Greek and Roman
world through the Renaissance to the present day is testimony at once
to those instincts and to what can be learned and taught about how to
channel them into patterns of evoked behaviour.9 Those whose behav-
iour is influenced in this way may be unlikely so to explain it to them-
selves. They may even be actively resistant to the suggestion, just as many
people become angry when told that close and long-lasting friendships
between fellow humans are an evolutionary product of the reproductive
advantages under ancestral conditions of readiness to engage in recipro-
cal exchange of favours (Tooby and Cosmides 1996: 131). Soldiers writing
home from the battlefront about their willingness to die for the ‘father-
land’ or ‘motherland’ may be no more able to account for the appeal to
them of the symbolic equation of nation with family than children fright-
ened of animals or the dark are aware that they are responding instinct-
ively to what Bowlby called ‘natural clues’ (1973: 161). But that is what is
going on, whether they realize it or not. Politicians appealing to party
loyalty and corporate executives denouncing whistle-blowers are practis-
ing applied evolutionary psychology no less than are the advertisers who
use photographs of sexually attractive women to sell sports cars to pro-
spective purchasers presumed to be male.
Darwin was, once again, there already: ‘The impassioned bard, orator, or musician, when with
9
his varied tones and cadences he excites the strongest emotions in his hearers, little suspects that
he uses the same means by which his half-human ancestors long ago aroused each other’s ardent
passions, during their courtship and rivalry’ (1882: 573).
82 The theory of cultural and social selection
Evoked behaviour has, moreover, an importance in warfare which goes
beyond the immediate self-protective responses evoked in face-to-face
combat to the bonding effects of sustained companionship under condi-
tions of stress. Richerson and Boyd (1999) concluded from studying the
evidence for the comparative performance of the American and German
armies in the Second World War that German soldiers who were recruited
from the same town or district, were kept together during training, joined
the same platoons or companies, were led into battle by the same officers
who had commanded them as recruits, and when withdrawn from the
front were retrained by those same officers fought much more effectively
than American soldiers who were trained in temporary groups, were sent
to the front individually, and joined large divisions composed of strangers
within which casualties were replaced piecemeal. This conclusion would,
admittedly, have surprised neither Xenophon nor Frederick the Great.
Xenophon’s writings about Greek infantry warfare consistently empha-
size the value of comradeship as opposed to instruction, and Frederick the
Great more than once recorded his conviction that recruits drawn, as his
were, from the same cantons were more likely than others to support each
other automatically in battle and encourage each other’s willingness to put
their lives at risk. The difference in performance is not to be explained in
terms of evoked behaviour alone: as I remarked in Section 6 of Chapter 2,
cultural and social selection are both at work also in differentiating the
performance of armies and thereby influencing the subsequent reproduc-
tion and diffusion of the memes and practices of which they are the car-
riers. But McNeill’s ‘tapping’ of the ‘primitive reservoir of sociality’ can
confer a marginal advantage on one army over another which is poten-
tially as decisive as either warrior norms acquired through imitation and
learning or military discipline imposed by institutional sanctions.
9
It should not, however, be supposed that evoked behaviour is ‘instinct-
ive’ in a sense that would imply that its motivation always comes from
the emotional rather than the rational brain. Since human beings have
inherited an intellectual repertory which, limited as it may be, still makes
us potential Bayesians and potential game-theorists, these capacities too
must have been adaptive in the millennia during which they were natur-
ally selected in parallel with our repertory of emotions.
The idea that Bayes’s theorem models an inferential process biologically
inherited from our remote ancestral environment is apt to be queried for
Natural selection and evoked behaviour 83
two reasons: first, because Bayesian reasoning is excessively dependent on
unascertainable prior probabilities of the evidence given only the back-
ground knowledge; and second, because the computations which the the-
orem requires are beyond the capacity of human beings to carry them
out inside their heads when confronted with real-life choices in complex
environments. But a part of the human psychological repertory is an
ability to modify a belief in the recognition that the probability of its
being true depends on the probability of later evidence for that probabil-
ity being true. The cultural inertia too easily attributed by some sociolo-
gists to ‘peasant’ communities does not totally inhibit cultivators from
studying the yields which their neighbours obtain from their plots and
modifying their own techniques in consequence, just as seemingly unre-
flective working-class adolescents in industrial societies have been shown
to modify their educational aspirations in the light of what they observe
about their prospects of upward social mobility. Bayesian decision-takers
may be less often found, and may make the mean behaviour of the pre-
sent population less different from the mean behaviour of the previous
population when they are, than unthinking imitators of prestigious role-
models or of the majority of what they regard as their peer-group. But a
recognizably Bayesian response triggered by a changed environment does
sometimes provide the explanation of differentiated behaviour-patterns
where pre-existing memes and practices are becoming relatively less adap-
tive. To accept that natural selection has not equipped the human brain
to do probability problems at all well is not to deny that it has given it a
capacity for modification of initial beliefs in the light of observed out-
comes without an explicit calculation of probabilities having to be carried
out as such.
Similarly, collective behaviour can conform to game-theoretic models
without its being assumed that the natives are consciously applying a
pay-off matrix which they have been carrying around as such inside their
heads. Coleman, in analysing escape panics, shows how they differ from
the classic Prisoner’s Dilemma because of the possibility of coordination
conditional on the responses of others over an initial period of time fol-
lowing the response immediately evoked by obvious physical danger: the
dilemma is that to delay running for the exit as soon as the fire breaks
out in an enclosed area both increases the opportunity for coordination
which would benefit everyone and heightens the risk that if others start
to run survival will turn out to depend on having been among the first
to do so. Such panics, which pre-game-theoretic authors were apt to
attribute to collective consciousness, mob psychology, herd instinct, or
84 The theory of cultural and social selection
quasi-hypnotic suggestion, are an obvious example of evoked behaviour
set off by the universal instinct for survival, with the consequence, as
Coleman argues (1990: 214 n.7), that the emotion labelled ‘panic’ is cre-
ated by the high cost of taking time before making the potentially fatal
decision to run or not to run for the exit. In examples like this, rational
choice theory and evolutionary psychology combine to produce some of
the most plausible just-so stories of all.
10
One of the standard topics in comparative sociology which invites ana-
lysis along these lines is the feud.
Feuding is abundantly documented in the ethnographic and histori-
cal record across a range of times and places from pre-Islamic Arabia to
medieval France to the Southern United States, the Peloponnese, Corsica,
Andalusia, Algeria, East Africa, Montenegro, Albania, and others besides.
The melodramatic tales culturally transmitted in song and story, whether
in the gestes of the minstrels, the Icelandic sagas, the Spanish cantars, or
simply the folk memory of kin-groups nurturing their culturally inher-
ited hatreds of one another, reflect behaviour-patterns with which their
audiences were all familiar. The ‘highly structured poetic genre’ exem-
plified in particular in the Icelandic sagas (Fentress and Wickham 1992:
167; Dunbar et al. 1995) reflects both the structure of the feuds and the
relations of kinship among killers and killed. It is striking how often the
stories mirror Hamilton’s Rule directly: the intensity of the desire for
vengeance is proportionate to the genetic proximity of avenger to victim,
compensation is accepted in settlement only when the scale of it corre-
sponds to the closeness of the relationship, and importance is particu-
larly attached to the chastity of nubile female relatives which, if violated,
requires young adult males to seek out and kill the violator. To borrow
McNeill’s metaphor about the bonding effect of close-order infantry drill,
the feud ‘taps a primitive reservoir’ of kin-based solidarity and masculine
aggression. In principle, moreover, feuds can continue for as long as the
memory of injuries not fully avenged. A feud is ‘not a finite series of hos-
tile acts which can be said to have both a beginning and an end’ (Black-
Michaud 1975: 37). Even where settled by arbitration or negotiation of
some kind, or seeming to peter out over time or across physical distance,
feuds can be evoked all over again by a chance face-to-face encounter,
or an accidentally reawakened territorial dispute. But the feud is a dis-
tinctive behaviour-pattern which persists only under a specific range of
Natural selection and evoked behaviour 85
environmental conditions in which evoked and acquired and imposed
behaviour all play a part.
One of the more frequently observed of these conditions is the absence
of a strong central state – that is, of practices defining roles to which
there attaches the power to impose a monopoly of the means of coercion
on the feuding parties. But no lawlike generalization to that effect can
be framed – feuding persists in some strong states, and is absent in some
weak ones. Similarly, there is a positive correlation between pastoral econ-
omies and the feud, but some village-dwelling agriculturalists engage in
feuding. The feud is a textbook example of convergent evolution where
different cultures and societies in which it is found have just-so stories
which are unique in some respects to themselves but at the same time
point to similar selective pressures at work. The biologically inherited
readiness of young men to go out and kill the killers of their kin is sup-
ported by memes enjoining defence of familial honour handed down by
imitation and learning in an environment of institutional practices and
roles which are always permissive of, and sometimes positively favour-
able to, it. Feuds cannot be explained in the same way as wars – they
involve the killing of fellow members of the same society, not of others, in
ways that conform to certain specific rules. As has been said of the society
depicted in the sagas, ‘The social and governmental order that the advo-
cacy system reinforced made Icelandic feud possible’ (Byock 1982: 38). Or
as Evans-Pritchard said of the Nuer, ‘Blood feuds are a tribal institution,
for they can only occur where a breach of law is recognized, for they are
the way in which reparation is obtained’ (1940: 150). Variations in this
common behaviour-pattern are explicable by reference to local selective
pressures which derive, for example, from the dispersal of fraternal inter-
est-groups through non-patrilocal residence (Kang 1979), or the relation
of compensation in blood-money to the institutions of a market economy
(Otterbein 2000). As it is put in a study of the feud in nineteenth-century
Corsica, where court records provide an unusually reliable source for the
precise number and nature of the killings, the participants ‘were highly
strategic even as they sacrificed individual interests for collective action.
They reliably reproduced a feuding system not because of a commitment
to a society-wide system of values or out of purely individual self-interest,
but out of devotion to their own family units – a devotion on which they
acted in a highly strategic way’ (Gould 2000: 700).
A thoroughly researched just-so story about interpersonal vio-
lence which rests on a co-evolutionary explanation at all three lev-
els is the analysis of the ‘culture of honor’ in the American South by
86 The theory of cultural and social selection
Richard E. Nisbett and Dev Cohen, who in their Preface explicitly
endorse the principles underlying evolutionary psychology as ‘an emer-
ging metatheoretical base for the social sciences that is particularly rele-
vant to cultural studies’ (1996: xvii). The readiness of white, non-Hispanic
Southern males to respond with lethal violence to personal insults or
threats to their families and property is amply evidenced in the homicide
rates disaggregated by Nisbett and Cohen from US Department of Justice
records. Particularly striking are the correlations with rural residence in
herding as opposed to farming regions. Furthermore, the regional dif-
ferences between the South and Southwest on the one hand and the
non-South on the other are significantly more marked in the case of
‘argument-related’ than of ‘felony-related’ homicide. The ecological cor-
relations are underpinned by both survey data and experimental results,
and alternative hypotheses convincingly disposed of. The descendants
of the Scotch-Irish immigrants who settled on the frontier herding ter-
ritories have inherited a culture enjoining violence in the protection of
property, physical retaliation against personal insult, distrust of public
authority, and unquestioning loyalty to family and kin. In this they differ
from non-white Southerners as much as they do from Northern whites.
But their attitudes are coherently formulated and consistently expressed,
and they pursue their families’ interests as strategically as anyone else.
The explanation is not to be sought in behaviour genetics: there is no
reason to suppose that they are innately predisposed to violence to a sig-
nificantly greater degree than the generality of young adult males. But a
biologically inherited potential has come to be evoked to an extent that it
is not elsewhere because of the selective pressures which have favoured a
co-evolution of memes and practices through which feuding persists.
H u m a n s o c iol o g y a s pr i m at e s o c iol o g y
11
The science writer Matt Ridley has described (1996: 159) the sense of
uncanny familiarity with which he responded to an account of the Wars
of the Roses which he happened to read shortly after reading de Waal’s
account in Chimpanzee Politics of the goings-on in the Arnhem zoo: here
were Margaret of Anjou, Edward the Fourth, and Warwick the Kingmaker
behaving just like Luit, Nikki, and Yeroen in their struggles for place and
power. Parallels of this extended kind between human and animal behav-
iour have a long history in literary and philosophical discussion of human
Natural selection and evoked behaviour 87
nature. But the combination of advances in neo-Darwinian theory with
the findings of primatologists working in the field has given them greater
precision as well as point. We now have book titles like not merely The
Naked Ape and The Third Chimpanzee but The Arbitrary Ape, The Thinking
Ape, The Ape That Spoke, The Protean Primate, The Promising Primate, and
even Primates in the Classroom: An Evolutionary Perspective on Children’s
Education. There is thus a sense in which comparative sociology might be
called one among other sub-disciplines of comparative primatology. On
what grounds is ‘chimpanzee politics’ a less legitimate term of behavioural
science than ‘vigilant monitoring among hunter-gatherers’ or ‘kingship in
Sub-Saharan Africa’ or ‘the capitalist state’?
Sociologists may protest that this example is too easy. Struggles for per-
sonal dominance among a small number of contenders within a close-knit
transitive rank-order are a different matter from the varied and continu-
ous movements of persons between roles, and of roles themselves in three-
dimensional social space, of the kind familiar from human societies with
complex systactic structures and long histories of evolving modes of pro-
duction, persuasion, and coercion. But I recently had an experience very
similar to Ridley’s on rereading a classic study of English politics in the
period from 1675 to 1725 (Plumb 1967). The players in that game were
contending for power in an institutional environment dominated by a
royal court and a bicameral parliament in one part of which replacement
of representatives of one party or faction by another in periodic elections
could be decisive for the maintenance or loss of status. It is a world yet
further away from the Arnhem zoo than the Wars of the Roses. But the
favours given and received among the politicians and their supporters still
correspond to the mutual grooming of the chimpanzees; the monetary
emoluments – or bribes, if you will – correspond to the chimpanzees’
gifts or exchanges of food; the votes cast at the hustings or in parliament
correspond to the support which chimpanzees give to one another in cir-
cumstances of actual or potential conflict; the higher-ranking males in
both cases enjoy pickings and perquisites from which the lower-ranking
are excluded; coalitions form and reform among both with equal ease
and speed; rituals of submission are performed at court or in parliament
as they are in the Arnhem zoo; in both cases, ties of kinship, reciprocal
altruism, and ‘Machiavellian’ social manipulation are equally visible; and
in both cases, there can be discerned differences in cultural traditions
shared by sub-groups within the population as a whole.
No less striking are the parallels in the details of evoked face-to-face
behaviour. Stereotypical gestures, facial expressions, the keeping of physical
88 The theory of cultural and social selection
distance, and actions such as bending and bowing (or alternatively rising
upright) are readily observable in humans and chimpanzees alike. So too
is the emotional instability manifested in abrupt reversals in conduct.
Humans and chimpanzees alike switch from defiance to submission. Both
make sudden displays of sympathy or affection, both intervene unexpect-
edly as either participants or mediators in conflicts previously watched with
apparent detachment, and both are prone to outbursts of collective anger.
The vocalizations of chimpanzees (laughter included) carry meaning,10
and their lack of vocabulary, grammar, and syntax does not prevent them
from having some sense of their own identity and attributing some degree
of knowledge and intention to others (including humans). They not only
retaliate immediately against their enemies but take delayed revenge. They
recognize allies as allies and rivals as rivals over extended periods of time.
They jointly patrol the boundaries of territories which they regard as their
own. They often quarrel, but often become reconciled with those with
whom they have quarreled. They tend each other when sick or injured. If
two chimpanzees consistently groom each other, help each other, play with
each other, pay attention to each other, huddle up with each other, and
share with each other, what more do they have to do before the primatolo-
gist studying them is allowed to talk of them as ‘friends’ who, as Aristotle
stipulates (Nicomachean Ethics 1156a 3–4), want to do good to each other
and are conscious that the other does the same?
12
But chimpanzees are still … well, chimpanzees. Luit, Nikki, and Yeroen
can’t sit down round a conference table to renegotiate their roles in the
way that Margaret of Anjou, Edward the Fourth, and Warwick the
Kingmaker can. If a primatologist came back from the field having found
a chimpanzee population with monumental architecture, an ancestral
mythology, and a council of elders, it wouldn’t be a chimpanzee popu-
lation. It would be a population of other primates, however exactly to
be classified, with a critical difference, whatever exactly it might be, in
their genetic inheritance and the design of their brains. This would, as in
our own case, have taken many generations since the last common ances-
tor. But it would be no less a difference of kind. As Darwin allowed, ‘no
animal is self-conscious, if by this term it is implied that he reflects on
10
To an extent that those of vervets, for example, do not (Cheney and Seyfarth 1990); by compari-
son with chimpanzees, vervets can almost be labelled solipsists.
Natural selection and evoked behaviour 89
such points as whence he comes or whither he will go, or what is life and
death, and so forth’ (1882: 83). But that is just what human beings can’t
help doing. We are not just a loving and hating and playing and fighting
and cooperating and dissembling and grieving and rejoicing primate but
a narrating and speculating and sacralizing one. However important it is
that sociologists should recognize both that evoked behaviour is perva-
sive in all human societies and that acquired behaviour is not unique to
the human species, none of the research which has vindicated Darwin’s
alleged anthropomorphism has gone beyond the reservations which he
himself explicitly made.
From the perspective of comparative sociology, therefore, the way to
put it is to say that somewhere down the evolutionary sequence which
leads to ‘us’ primatologists have to turn into anthropologists. We do not
know, and may never know, just when our ancestors started to talk to
each other. If, as the anatomical evidence suggests, they had the phys-
ical capacity to do so by about 300,000 years ago, they may have used it
to shout, growl, laugh, sing, and cheer long before they used it to debate
what to do tomorrow or reminisce about what happened last week. But
at some unascertainable time, they started to exchange moralizing judge-
ments and metaphysical speculations as well as sharing practical infor-
mation, agreeing plans for the future, and comparing recollections of the
past. They began to ask each other such questions as ‘How can you excuse
such bad behaviour?’ and ‘Do you think the spirits of our ancestors are
watching us?’ The anthropologist who tracked them down in the field
could infer their beliefs and attitudes not only by observing their outward
actions, but also by learning their language, hearing what they thought
about their actions, and discussing with them what they meant in talking
about them as they did.
Thereafter, there is no escape from the perennially troublesome dis-
junction between the natives’ and the observer’s conceptualizations of
their behaviour. No primatologist at the Arnhem zoo or the Yerkes field
station or in the Gombe rain forest is concerned with a conflict between
‘their’ account of their behaviour and ‘ours’. But every anthropolo-
gist is concerned with it from the first day of fieldwork, whether ‘they’
are speakers of a language which the anthropologist has to learn from
scratch or of a language which is also the anthropologist’s own. The pro-
cedural maxim which follows from the distinction drawn in Section 5 of
the Prologue between intentions and motives remains as relevant as ever:
the natives’ reports of what it is that they are doing are privileged over
the observer’s at the same time as the observer’s explanation of why they
90 The theory of cultural and social selection
are doing it is privileged over the natives’. But once the natives are willing
to answer questions about their own subjective experiences, the anthro-
pologist observing them has both the benefits and the attendant risks of
access to evidence about what they are doing and why which not even the
most intelligent chimpanzee, dolphin, or elephant can provide.
Primatologists whose experiences in the field have enabled them to
detect the differences in temperament and personality between one chim-
panzee and another may, like Darwin himself, refuse to accept that their
subjective mental states are indistinguishable, and still less that they have
none. Might there not after all be a chimpanzee phenomenology (to say
nothing of a dolphin or elephant phenomenology)? Why shouldn’t the sub-
jective experiences of Luit, Nikki, and Yeroen lend themselves to incorp-
oration into a novel as persuasive as William Golding’s The Inheritors, in
which the encounter between Neanderthals and humans is narrated from
the perspective of a hypothetical Neanderthal protagonist whose ‘cloudy,
static, non-abstract awareness of life’ is a world ‘where neither action nor
its corollary, speech, contains any subordinate clauses’ (Green 1963: 46)?
Authentic accounts of what other people’s experience is like for them
don’t require the account to be given in words which are there inside
their heads. Vivid and illuminating descriptions of mental states can be
phrased in language which couldn’t possibly have been used by ‘them’,
whether a child (or a chimpanzee) in a state of boiling rage who doesn’t
know what ‘boiling’ is, or a Machiavellian ruler (or the alpha male at the
head of a chimpanzee troop) who ruled long before Machiavelli was born,
or a Palaeolithic Neanderthal (or chimpanzee) Othello consumed by sex-
ual jealousy who has never seen a performance of Shakespeare. But the
critical difference remains that humans can, but chimpanzees can’t, have
put to them by the observer the suggestion that the proffered description
of what is going on inside their heads does authentically fit their case.
There is no evidence, after decades of field and laboratory research, that
chimpanzees have the capacity for meta-representations. ‘Machiavellian’
chimpanzees who, like Darwin’s decoy elephants, ‘well know what they
are about’ still cannot to be taken to be conducting an interior mono-
logue about what they are about of the kind that human beings do.
Chimpanzees are able to exchange information with one another to a
greater degree than used to be believed. But no primatologist has found a
chimpanzee who has, to quote Darwin again, developed a power of rea-
soning to the point of speculating, even ‘vaguely’, on ‘his own existence’
(1882: 94). When Frans de Waal observes two unruly female teenage chim-
panzees being punished by the rest of the colony in the Arnhem zoo for
Natural selection and evoked behaviour 91
refusing to enter their sleeping quarters and thereby delaying everybody
else’s evening meal (1996: 89), it is difficult to resist the inference that
something very like ‘serve ‘em right!’ is going on inside the punishers’
heads. But that still doesn’t license a further inference to Darwin’s ‘so-
called moral sense’. Subsequent experiments by de Waal and others have
shown that not only chimpanzees but capuchins (Brosnan and de Waal
2003) register awareness of a difference between equitable and unequitable
exchanges. But evoked responses to ‘unfair’ rewards or ‘grabbing’ instead
of sharing of food still fall short of self-consciously virtuous restraint of
the impulse to grab, assignation of value to a sense of fairness as such,
assessment of the relative strength of a claim of need as against a claim
of desert, and judgement on the behaviour of both self and others as a
violation of a specifically moral norm. Among chimpanzees, tolerance for
inequity may increase with social closeness (Brosnan et al. 2005), but they
do not exhibit the other-regarding preferences and concern for the welfare
of unrelated group members found in humans (Silk et al. 2005).
Some readers, including particularly dog-owners, may respond by say-
ing that they have seen animals display unmistakable signs of guilt at
having disobeyed a command. But that is to confuse guilt with shame.
Guilt presupposes the violation of a norm for which the violator is aware
of deserving to be punished because the norm carries an inherent moral
obligation to conform (keeping a promise when it is personally disad-
vantageous to do so) as opposed to a norm which does not (speaking the
vernacular language correctly). There is, moreover, a parallel between
the so-called moral and the so-called aesthetic sense. Just as chimpan-
zees can be observed responding with what might be thought to approach
righteous anger at the behaviour of unruly teenagers who delay the even-
ing meal, so can they be observed responding to the sight of a fire or a
waterfall with what might be thought to approach aesthetic appreciation.
But to borrow from Kant (despite my remark about him at the end of
Section 6 of the Prologue), to say that something is beautiful is to say that
everyone else ought to approve of it too. The chimpanzees standing awe-
struck in front of a fire or a waterfall do not have the same thing going
on inside their heads as Henry Adams standing awe-struck in front of
Chartres Cathedral (and then deciding to tell the rest of us how deserving
of admiration it is). Only among primates who can talk to each other can
cultural selection begin to generate out of a naturally selected moral sense
the diversity of both moral and aesthetic attitudes which, like the diver-
sity of causal beliefs about the workings of the world, is for comparative
sociologists, not cultural primatologists, to explain.
Ch apter 3
Be l i e f s a n d at t i t u de s
1
How then should comparative sociologists proceed when contrasting
one human culture or sub-culture with another? What kind of just-so
story will best explain why these and not those memes are the ones being
acquired and acted out?
However the process of heritable variation and competitive selection of
memes is modelled, it will have to be estimated as accurately as possible
just how widely the distinctive beliefs and attitudes and the representa-
tions underlying them are distributed among the population of carriers.
Obvious as this may be, there are more instances than there should be in
the historical and ethnographic literature where an author implies without
sufficient qualification that the members of the culture chosen for study
were all carrying much the same bundles of memes inside their heads. It
might turn out that some representations, at least, are shared with only
insignificant differences between the information in one native’s head and
another’s. Perhaps every Hellene shares a broadly similar representation of
Zeus, every Navajo a broadly similar representation of hozho (harmony,
beauty), and every Polynesian a broadly similar representation of mana.
But the ideal type of a traditional culture, in which every native holds
exactly the same beliefs and shares exactly the same attitudes, has never
been more than that. Not only does there have to be an explanation of how
an allegedly unquestioned set of inherited memes came to be selected in
the first place, but the capacity of the human mind actively to reinterpret
or reconstruct any inherited meme means that any complex of memes by
which the natives’ collective behaviour is guided is likely to be challenged
by at least some of the members of each successive generation. This is so
even if the behaviour-pattern is the product of meme-practice co-evolution
92
Cultural selection and acquired behaviour 93
and an imposed ideology is underwritten by institutional sanctions attach-
ing to the roles of elders or priests or literati or schoolteachers or party
cadres. It is true that there are (or at any rate, have been) cultures whose
members are unaware of any other cultures different from their own. But
that does not mean that the culture in question did not evolve out of an
earlier one in the past or will never evolve into another one in the future,
even if it remains shielded from invasion by the memes of any other.
Nor should it be assumed that beliefs, any more than attitudes, are
carried inside the natives’ heads without vacillation or ambivalence,1 even
where a widely shared belief is of a self-evidently practical rather than
metaphysical kind. Quesalid, Franz Boas’s Kwakiutl shaman, is proud of
his successes, but he is well aware how much trickery he and his fellow-
sorcerers administer (Lévi-Strauss 1963: 175–8), just as Zande witch-doctors
‘know that their extraction of objects from the bodies of their patients is
a fake, but they believe that they cure them by the medicine they admin-
ister’ (Evans-Pritchard 1937: 255). In matters of health and sickness, it is
only to be expected that uncertainty will be felt, if not always admitted,
both by shamans, sorcerers, oneiromancers, faith-healers, astrologers, and
doctors on the one side and by their patients and clients on the other. It is
not surprising in any culture to come across the advice given by Samuel
Johnson in a letter to his friend Bennett Langton dated 17 April 1775:
‘My opinion of alternative medicine is not high, but quid tentasse noce-
bit?’ – what harm will it do to have given it a try? The anthropologist
Mary Douglas, having reported the !Kung Bushmen as laughing out of
court the suggestion that they believe that their rain rituals cause rain,
also reports of the Dinka that ‘Of course they hope that rain rituals will
cause rain’ (1966: 58, 68), while Evans-Pritchard reports of the Nuer that
‘they are little interested in ritual for bringing rain and even think it pre-
sumptuous to think of asking God for rain before sowing’ (1956: 200).
The elder Pliny, in his rambling, encyclopaedic Natural History, at one
point denounces ‘magic’ as ‘the most fraudulent of arts’ (XXX.1) but at
another (XXVIII.4) concedes that it works. There is no human popula-
tion whose members have no beliefs whatever inside their heads. But not
‘Half-belief ’ in the sense of uncertainty needs to be distinguished from ‘half-belief ’ in the sense
1
that the natives follow a rule such as ‘never walk under a ladder’ despite disavowing the ‘supersti-
tion’ behind it (Campbell 1996). Does the congregation ‘really’ believe in purgatory? Do the vot-
ers ‘really’ believe that the performance of the national economy will be significantly improved by
a change of government? Perhaps not. But what matters to comparative sociology is whether, in
the culture or society under study, they behave as if they do.
94 The theory of cultural and social selection
even the most vehement protestations of belief are a guarantee that the
protester is immune to the possibility of doubt.
Since, however, comparative sociology is concerned only with the
reproduction and diffusion of information which does demonstrably
affect population-level behaviour-patterns, this means, in cultural evo-
lution, the displacement of one set of memes by another which reaches
the point that the later one has phenotypic effects sufficiently consistent,
widespread, and distinctive for sociologists of all persuasions to agree that
a new culture (or sub-culture) has emerged. There will always be scoff-
ers and sceptics like Diagoras of Melos, who responded, according to
Diogenes Laertius (VI.59), to the sight of the votive dedications in a temple
of the sea-god Poseidon with the remark that there would be many more
of them if those who had been drowned had been able to do the same.
But the temple and the dedications were there for Diagoras to be sceptical
about, and they were part of the complex of memes and their extended
phenotypic effects covered by the conventional designation ‘Greek reli-
gion’. It is also possible for competing memes to co-exist side by side in
the same heads, with the result that, for example, an Anglo-Saxon king
keeps a pagan altar alongside a Christian one (Bede, Ecclesiastical History
II.15). But once a large-scale qualitative change in a previously stable cul-
tural behaviour-pattern is reported as having occurred, its explanation
depends on finding out where, down the chain of extended phenotypic
effects, selective pressure has come to bear in such a way as to favour what
have turned out to be the more reproductively successful mutations and
combinations of the particular memes to which the behaviour-pattern is
giving phenotypic expression.
It is not to be expected that the natives will all agree in their beliefs
about, let alone their attitudes to, the past any more than the present. The
anthropologist Robert Borofsky, doing fieldwork on the Polynesian atoll
of Pukapuka from 1977 to 1981, was astonished to find in operation a trad-
itional form of social organization which had been deliberately revived
shortly before but on the evidence of previous anthropologists, as well as
of missionaries and government officials, appeared never to have existed
at all. Extensive interviews with elderly informants revealed that there
had indeed existed more than once in the past a two-village, as opposed
to three-village, form of social organization, but previous outside obser-
vers had failed to ask the questions which would have revealed it, and
the informants’ views of just what went on, let alone just when and why,
differed significantly among themselves (1987: Ch. 1). But that is itself just
what goes on when, for example, the recollections of strikes and protests
Cultural selection and acquired behaviour 95
among the members of working-class communities in Europe and the
United States are studied in the same way (Fentress and Wickham 1992:
Ch. 3). All cultural traditions are inventions in the sense that somebody
has to have been the first person to put together the package of memes
which is reproduced in recognizably similar form down subsequent gener-
ations of readers or listeners. The inherited histories or folktales or myths
by which the natives’ behaviour is affected do not need to have been
selected for the degree of their correspondence with observations which
have been tested for their accuracy. It is true that in any culture, there
are numerous beliefs about the past which are no more disputed than
the beliefs transmitted by formal learning down successive generations
of British schoolchildren about the dates of the battles of Hastings and
Trafalgar. But ‘traditional’ representations of the past become such wher-
ever, whenever, and in whatever way the environment in which they are
handed down has enhanced the adaptiveness of their constituent memes,
whether or not they have a demonstrated correspondence with ‘the facts’.
2
Although the naturally selected capacity for fast and frugal reasoning
and quick intuitive evaluation does not exclude second-order reflection
about the possible merits of other competing theories, such reflection is,
as I suggested in Section 2 of Chapter 2, rarely found outside of, and not
always within, sub-cultures where norms of disinterested enquiry and test
of received opinion are among the memes routinely acquired from teach-
ers by their pupils. The Fijian chief who said to the Protestant mission-
ary ‘muskets and gunpowder are true, and your religion must be true’
(Sahlins 1983: 519) may have been speaking with a touch of irony. But as
Donald T. Campbell remarked in one of his articles, ‘Priests can be as
ignorant as the superstitious populations they dominate of the true adap-
tive functions of the belief-systems they perpetuate’ (1975: 1107).
In general, memes which find expression in stable collective
behaviour-patterns are more likely to have been inherited by imitation
than by learning, and if by learning less by individual trial-and-error
than by initially passive social learning from parents and mentors. Once
a change is under way, frequency-dependence is sometimes a sufficient
explanation by itself. But time and again, there is an exemplary carrier
of a mutation who is imitated by others on the basis of personal prestige,
not only in cases of religious (or, for that matter, political) conversion
but across the whole spectrum of phenotypic effects from burial customs
96 The theory of cultural and social selection
to natural-scientific research. In France, it was the precedent set by a
pious daughter of Pepin the First which diffused the replacement of the
Merovingian custom of burying the dead with their accoutrements by
the Christian custom of unfurnished graves a whole two centuries after
the Franks had become Christian (Fletcher 1997: 259). In England, the
success of the memes constitutive of the program of experimental science
was significantly advanced by the invasion of the sub-culture of natural
philosophy by the ‘conventions, codes, and values of gentlemanly con-
versation’ (Shapin 1994: xvii) of which Robert Boyle was an exemplary
carrier. And sociologists studying the formation of ‘public opinion’ in the
mid-twentieth-century United States found that the relevant information
was transmitted in a ‘two-step flow’ from leaders to followers within their
local communities (Katz and Lazarsfeld 1955).
Such deference to prestige is no guarantee that the memes transmitted
will be advantageous to their individual carriers, if acted on, in the way
that the imitators themselves may suppose. It is likely enough that once
prestige no longer attaches to the conspicuous burial of valuable material
goods alongside the body of the deceased the heirs will gain materially by
retaining them, that once experimental methods become demonstrably
effective natural philosophers will help to further their careers if they
adopt them, and that the implicit presumption on the part of followers of
opinion that the articulate leaders have access to information which they
do not is at least as likely to benefit them as the assumption, in a textbook
example, of diners who, when they see queues for places in two externally
indistinguishable restaurants, conclude that those in the longer queue are
in it because it is for the one that has been found by experience to be the
better of the two. Once, however, social as well as cultural selection is
involved, the distribution of institutional power may be more important,
evolutionarily speaking, than that of beliefs and attitudes, and memes
which would otherwise have been better adapted than their competi-
tors become less so. It is true that the incumbents of roles to which there
attaches institutional power cannot force other people to believe things
they regard as untrue or approve of things which they regard as repug-
nant, even with muskets and gunpowder: as Kant said, it is the body, not
the mind, that bows to a superior, and willingness to die for strongly held
convictions is as well documented for the sectarians of the Eternal Mother
cult in Qing China as for the early Christian martyrs. Nor can conversion
be achieved simply by attacking, as would-be reformers so often do, the
extended phenotypic effects of the offending memes. Although books can
be burned, statues smashed, altars desecrated, and churches or temples
Cultural selection and acquired behaviour 97
pulled down, this may entrench the offending memes instead of extirpat-
ing them. Sociologists of religion have often observed that small, strict
sectarian communities can be stronger and longer-lasting than larger and
less restrictive ones. But in a society whose practices include institutional
sanctions deployed by the incumbents of the topmost roles against any-
one suspected of adherence to a prohibited creed, doubters, opportunists,
and lukewarm conformists are not only more likely to conceal their con-
victions and refrain from behaviour which might attract the attention of
the authorities, but also less likely to pass the offending memes down to
their children. As in natural selection, the criterion of adaptedness is not
the size of the population of carriers.2 But social selection is likely to be a
stronger evolutionary force than cultural selection in populations where
commitment is limited to what A. D. Nock, in his classic study of con-
version, called ‘mere adhesion’ (1933: 7). The changed institutional envir-
onment then has the same effect in winnowing the strong from the weak
as does a changed ecological environment where biological rather than
cultural survival is at stake.
To the extent that the distribution of memes is demonstrably correlated
with the economic, ideological, or political power attaching to their carri-
ers’ roles, something therefore needs to be retained within neo-Darwinian
sociology from what used to be called, somewhat misleadingly, the ‘soci-
ology of knowledge’. The causal impact of the carrier’s role may be either
positive, in the sense that a systactic interest is directly served, or nega-
tive, as when, to borrow an example from Karl Mannheim (1936: 175), the
memes constitutive of the Christian idea of brotherly love run up against
the practices constitutive of the institution of serfdom: the medieval
knight can be a committed adherent of the Church, loyal to his superiors,
courteous to his equals, and generous to his inferiors, but he cannot act
out a strategy of unconditional cooperation with his fellow human beings
without abandoning his role. The appeal of different doctrines of salva-
tion to the members of different systacts has been a commonplace among
sociologists of religion since Weber, and the answer to the ‘just what is
going on here?’ question has to include the extent to which memetic muta-
tions or combinations which modify an existing norm of conduct are on
average more likely to be reproduced and diffused among carriers whose
roles are of one rather than another kind. Just as Methodism in nine-
teenth-century England appealed to tradesmen and artisans more than to
As George C. Williams rhetorically asks, ‘Is a fox population less successful than the more
2
3
Cross-cultural comparisons in shared attitudes and beliefs are made no
easier by the lack of agreement among both the participants and the
researchers who report them about the terminology for doing so. There
is not only the familiar question touched on in Section 5 of Chapter 2
whether it is legitimate to apply to the memes in other people’s heads
terms to which there are no corresponding terms in their own vernacu-
lar language. There is also, even within the population whose vocabu-
lary is the same as the observer’s own, the problem of whether to assign
to a familiar conventional category the same set of beliefs and attitudes
that the natives do. In the culture within which this book is written, it
is easy to point to collective behaviour-patterns which both the academic
observer and the natives can agree to be ‘religious’: there are the worship-
pers singing hymns on Sunday, the theologians expounding the Old and
New Testaments, the mourners at the burial service voicing their faith
in an afterlife, and so on. But what about civil weddings, the national
anthem, taboos on the eating of domestic pets, fortune-telling, faith-heal-
ing, Christmas presents, initiation rituals, ‘New Age’ spirituality, rever-
ence for Nature, and commemoration of the war dead on Armistice Day?
The selectionist response to that question is not to argue over the defi-
nition of ‘religion’ but to identify the co-adaptive memes being acted out
in the behaviour-patterns of the members of the culture or sub-culture
Cultural selection and acquired behaviour 99
under study which are phenotypic effects of either beliefs in supra-human
agencies on the one hand or sacralizing attitudes on the other. As soon
as that is done, some of what is conventionally labelled ‘religion’ begins
to look increasingly like some of what is conventionally labelled ‘science’,
and some of what is conventionally excluded from ‘religion’ begins to
look increasingly like a part of it. A causal belief, however hesitantly or
ambivalently accepted as a guide to action, is no less a causal belief if
the agency held to operate on what goes on in the world is an invisible
quasi-personal being or an invisible impersonal force. When the emperor
Marcus Aurelius (Meditations XII.38) said that we know that gods exist
because we experience their power, he was voicing what countless human
beings over many millennia have accepted as a truism: somehow or other
the world works in many ways independent of human choice or interven-
tion, and the fact that it does work, however mysteriously, makes it plausi-
ble to suppose that somebody or something is making it work as it does.3
In many cultures, the local ontology allows for impersonal and quasi-
personal forces in, so to speak, the same breath: when, to take a Roman
example, Livy (III.vii.1–2) reports the abatement of a plague, he is frankly
open-minded about whether the cause was divine intervention or an
improvement in the weather. To Newton’s successor as Lucasian Professor
at Cambridge, the existence of ‘demons’ was ‘no more to be denied,
because we cannot, at present, give a direct solution of them, than are Mr
Boyle’s experiments about the elasticity of the air’ (Clark 1997: 306). This
is just as ‘scientific’ a belief as when the Fang people of Cameroon tell a
visiting anthropologist that those of them who are particularly successful
in oratory, business, horticulture, or witchcraft have been born with an
internal organ called evur which cannot be detected directly (Boyer 2001:
76). It may turn out in further discussion that the Fang have different
standards of evidence and proof from ‘us’. But their beliefs about evur are
no less beliefs about how the world works than are our beliefs about genes
and chromosomes and Newton’s successor’s beliefs about demons.
Where the memes acted out in the behaviour-pattern observed are
traced to the emotional rather than the rational brain, it is equally
A belief that the workings of the world are under the control of one or more quasi-personal agen-
3
cies benevolently disposed to human beings is, on the other hand, anything but a truism. To the
contrary, when one of Aristophanes’ comic slaves (Knights, line 32) asks another why he believes in
gods and is told ‘because they’ve got it in for me, don’t you see?’, the questioner finds the answer
entirely convincing. Cf. Cohn (1993: 50) on the culture of Mesopotamia: ‘As servants of the gods
the Mesopotamians knew better than to trust in the goodwill of their masters. That gods could
act ruthlessly, could even be wantonly destructive and cruel – that was obvious already in the
100 The theory of cultural and social selection
misleading to assume that the adaptedness of ‘religious’ attitudes is to be
differently explained from the adaptedness of attitudes unconnected with
beliefs about quasi-personal superhuman agents. Durkheim’s distinction
between the sacred and the profane, although deployed by him in sup-
port of his much criticized conception of religion as the worship of soci-
ety by itself, captures the universality of the urge to attach intrinsic value
to culturally constructed kinds independently of any instrumental value
as a means to some practical end. Sacralizing memes which are acted out
in rituals of veneration are easily discernible in ‘irreligious’ sub-cultures
in such fields of human activity as art, sport, and politics.4 It may be that
their probability of continuing reproduction is enhanced where they are
combined with memes constitutive of a belief that the behaviour they
dictate is approved by a god, or a saint, or the spirits of the ancestral dead.
But sacralizing memes can make an atheistic soldier on the field of battle
as ready to die to keep a regimental flag out of the hands of the enemy as
a Christian martyr thrown to the lions to die rather than abjure.
Where the memes located in the rational brain and those located in the
emotional brain are mutually adaptive, it makes no difference whether the
design of the combination is deliberate or unconscious. It doesn’t matter
whether or not Joseph Smith, the founder of Mormonism, mixed a cock-
tail of beliefs drawn from Protestantism, Freemasonry, and Hermeticism
and then cunningly laced it with attitudes of approval of higher educa-
tion and polygamy because he decided, after careful reflection, that that
was how he would have the best chance of winning the largest number of
prospective converts. The right just-so story has to show, whatever Joseph
Smith thought he was doing, what it was about that combination which
accounts for Mormonism’s rate of diffusion of 40 per cent per decade for
a hundred years. Marxism offers a similar example, despite – or, perhaps,
because of – Marx’s explicit repudiation of conventional ‘religious’ doc-
trine. Many commentators – Joseph Schumpeter, Karl Popper, and Leszek
Kolakowski included – have remarked on the traits which Marxism,
or Marxism-Leninism, shares with the Christian culture to which it is
ostensibly opposed – a combination of hypotheses about why the world
is as it is with eschatological prophecies, charismatic leaders, sacred texts,
ritual initiations, ceremonial processions, annual festivals, and not least
days of the Sumerian city-states, and it remained obvious so long as Mesopotamian civilisation
existed.’
4
Their adaptedness may well have been deliberately designed, as for example in the appropriation
of Roman symbols in the myths and rituals of Fascist Italy (Gentile 1996). The similar appropria-
tion of the Roman past during the French Revolution is briefly discussed in Chapter 4.
Cultural selection and acquired behaviour 101
Lenin’s mausoleum as ‘the symbolic centre, the “holy shrine” ’ (Lane 1981:
80). The extended phenotypic effects of the memes which the rulers of the
Soviet Union wished to diffuse among the Russian and other subordinate
populations may not be a reliable guide to what was in fact going on in
the natives’ heads: if the public expression of dissent is sufficiently effec-
tively repressed, both rulers and dissenters may mistakenly believe that the
memes prescribed by the ruling ideology are more widely diffused than in
fact they are (Kuran 1995). But belief in the inevitability of Communism
and disapproval of the ethos of capitalism are likely to enhance the rate of
diffusion of each other. The temptation to be resisted, here as elsewhere,
is the Fallacy of Elective Affinity – that is, the assumption that a con-
ceptual affinity is a causal one when it isn’t. However plausible the con-
nection in the observer’s mind, it may not be there in the natives’ heads
at all. A fatal flaw in Weber’s initial formulation of his celebrated thesis
about the ‘Protestant Ethic’ was his attribution to the Calvinist idea of
predestination the motivation which drove ‘inner-worldly ascetic’ entre-
preneurs to reinvest their hard-won profits in their businesses rather than
spend them on self-indulgent private consumption or ostentatious public
display. So stated, the thesis fails because neither Weber nor anyone else
has produced the evidence to show that Protestant entrepreneurs were in
fact motivated by beliefs about predestination to behave in a way that
Catholic entrepreneurs did not and they themselves would not otherwise
have done. That does not rule out the possibility of a co-adaptedness of
other Protestant – or, more accurately, Puritan – memes and certain of
the practices which Weber regarded as critical to the evolution of ‘mod-
ern’ capitalism. But that is a very different just-so story from the one that
he advanced in 1904, and I shall return to it at the end of this chapter.
4
The range of beliefs about how the world works, wide as it is, is far from
unlimited, and the resemblances between them down the ages and across
the globe are too marked to be attributable solely to chance. But the reverse
engineer who is looking for the memes whose combination forms the
locally optimal design will time and again find that its probability of con-
tinuing retention in the natives’ heads is highest where a belief which is not
straightforwardly obvious is combined with an attitude which is strongly
felt. The appeal of what are defined by the local culture as ‘miracles’ is
that they at the same time invite assent to some violation of conventional
assumptions about causes and effects and generate abnormally heightened
102 The theory of cultural and social selection
feelings of veneration or repugnance. When an anthropologist comment-
ing on Brazilian millenarism says that the ability of a charismatic leader
to ‘create a performance that showcases his divine power and teachings’,
when combined with ‘the development of miracle stories by his devotees’,
is critical to the process of sacralization (Pessar 2004: 46), the same could
equally be said not only of the charismatic instigators of secular political
uprisings and charismatic healers who defy the predictions of orthodox
practitioners of medicine but of charismatic arbiters of fashion. The young
men-about-town who see ‘Beau’ Brummell walking down St James’s Street
with his cravat untied are at the same time led to believe that their status
will be enhanced rather than diminished by violating accepted sartorial
convention and encouraged in their disdain for those whom they regard as
too timid or too obstinate to follow Brummell’s example.
The search for the critical memes being acted out in the emergence of
a novel culture or sub-culture sometimes leads to the formulation of a
just-so story of immediate plausibility. The novel beliefs so clearly approxi-
mate to what psychologists define as a ‘cognitive optimum’, and the atti-
tudes with which they have a selective (as opposed to elective) affinity so
obviously enhance the self-esteem of the believers, that their reproductive
success is virtually guaranteed. But sometimes, the disciples and converts
are attracted to beliefs which are not only extravagantly counter-intuitive
but difficult to comprehend and to attitudes which involve the sacral-
ization of rituals and life-styles which are not only eccentric but pain-
ful. This seeming paradox is addressed by the anthropologist Harvey
Whitehouse in terms of the distinction which he draws between ‘doctri-
nal’ and ‘imagistic’ modes. He himself restricts it to ‘religion’, defined for
his purpose in terms of ‘any set of shared beliefs and actions appealing to
supernatural agency’ (2004: 2), and he avoids the word ‘meme’ because
he wishes to adopt a position intermediate between the memeticists’
emphasis on population-level frequencies of discrete, culturally selected
units on the one hand and the cognitivists’ emphasis on innate, gen-
etically selected biases on the other. But he is explicit that ‘The theory
advanced here operates on principles of selection’ (2004: 74), and there
is no reason not to extend it to cover ‘religions’ which have no ‘gods’.
‘Doctrinal’ memes depend for their reproductive success on semantic
memory, routinization, and standardization of authoritative teaching,
where ‘imagistic’ memes depend on episodic memory, infrequent per-
formance, and high arousal in groups without centralized leadership. The
two are not mutually exclusive.5 The search, as always, is for the memes
5
Historically it may be that in Palaeolithic Europe, at any rate, an imagistic mode preceded the
doctrinal mode, and the emergence of a doctrinal mode required political and priestly roles and
Cultural selection and acquired behaviour 103
which combine in a co-adaptive complex, and the right just-so story has,
as always, to estimate the trade-offs. In the ‘doctrinal’ mode, routiniza-
tion confers a selective advantage on orthodox over heretical variants of
the crititical memes, and the trade-off is that routinization succeeds at the
cost of diminished motivation. In the ‘imagistic’ mode, the trade-off is
between the intense cohesion generated by rare climactic rituals and the
reduced probability of diffusion to other groups without radical mutation
of the critical memes in the course of transmission. But many success-
ful movements, parties, denominations, and sects are successful precisely
because their distinctive behaviour-patterns are a phenotypic effect of a
combination of co-adaptive doctrinal and imagistic memes.
Behind the selective reproduction and diffusion of the most successful
memetic combinations there can always be discerned the innate dispos-
ition to form teams and cliques differentiated by cultural markers. As I
remarked in Section 2 of Chapter 2, there is abundant experimental as
well as ethnographic and historical evidence for the ease with which the
commitment of recruits to such deliberately constructed groups can be
engineered. But there is also much experimental as well as ethnographic
and historical evidence for the degree to which this commitment can
be reinforced by rituals of initiation.6 In the culture within which this
book is written, rites of passage extend all the way from elaborate public
ceremonials like the coronation of kings or queens through secret rites
like those of the Freemasons to the initiation by symbolic humiliation
of new members of schools or street gangs or army regiments. It is some-
times possible for the same result to be achieved simply by the shared
experience which is a function of the roles occupied and performed by
the members of the group: the ‘baptism of fire’ undergone by soldiers who
have not previously been ordered into battle is enough by itself to bind
them to their companions-in-arms in the way discussed in Section 8 of
Chapter 2. But the reverse engineer will look for designs which combine
‘doctrinal’ memes which are easily understood and frequently repeated
with ‘imagistic’ memes whose esoteric and internally generated meaning
is encoded in an infrequent but highly emotive performance difficult to
perhaps literacy. But on this point, ‘Whitehouse’s theory is weakened by the fact that the pre
literate societies for which an imagistic mode of relation is proposed encompass an immense
degree of economic and political variability, which, in Western Asia for instance, ranged from
small-scale, egalitarian, highly mobile hunter-gatherers, to hierarchically organized, sedentary
hunter-gatherers, to farmers living in permanent villages with two-storey mud-brick architecture’
(Mithen 2004: 20).
6
In an intriguing early example, Aronson and Miles (1959) showed by administering an ‘embar-
rassment test’ to American college women who had volunteered for the experiment that sever-
ity of initiation was positively related to group preference after holding constant motivation for
admission.
104 The theory of cultural and social selection
forget. Conformity can then be reinforced by meme-practice co-evolution
through the institutional sanctions which attach to the roles of the elders,
priests, or officials who control and deploy the means of persuasion in the
society concerned.
It requires, accordingly, a just-so story of selection at all three levels to
account for how it is that, as Macaulay put it, ‘The experience of many
ages proves that men may be ready to fight to the death, and to perse-
cute without pity, for a religion whose creed they do not understand, and
whose precepts they habitually disobey.’ Consider the massacre of its pris-
oners, women and children included, by the Covenanting army after the
Battle of Philiphaugh in 1645 (Smout 1969: 64) to the cry of ‘Jesus and no
quarter!’ What can possibly explain the butchery of these innocents, who
posed no conceivable threat to their killers, in the name of the preacher
of the Sermon on the Mount? It is not enough to appeal to evolutionary
psychology for the proposition that just about anything will serve as a
rallying-cry by which to evoke collective behaviour, even if, for example,
soldiers in English armies who responded automatically to the invocation
of ‘St George for England!’ had no idea who St George (if he ever existed)
actually was. Nor is it enough to cite the Monophysites who walked about
the streets of fifth-century Alexandria shouting, ‘Cry the name of Jesus!’
to startled bystanders, or the ‘Honk if you love Jesus!’ stickers displayed
by car-drivers in the North American ‘Bible Belt’. The naturally selected
traits which predispose adult males to lethal violence against out-groups
were combined, in this instance, with culturally selected memes drawn
from an Old Testament conception of a God who enjoined his followers
to smite the enemy in His name, and socially selected practices impos-
ing soldierly obedience to orders from above. The rank-and-file of the
Covenanting army may have been less motivated than the Covenanting
ministers by the prospect of purging the Scottish church of prelacy in the
name of ‘King Jesus’. But they obeyed them readily enough. It is a clear
example of an exaptation where a mutant meme initially diffused among
the disciples of a charismatic founder come to be acted out in the subse-
quent course of cultural and social evolution in a phenotypic behaviour-
pattern directly contrary to what the charismatic founder had in mind.
5
Memes acted out in patterns of acquired behaviour can also be adaptive
when they take the form of beliefs which make the believers’ environment,
as they see it, more manageable for them even though their behaviour is
Cultural selection and acquired behaviour 105
stigmatized by the fellow-members of their own culture and punished by
the rulers of their own society. This takes us to the category of behaviour
conventionally labeled ‘magic’. As with ‘religion’, no useful purpose is
served by arguing over its definition. But it is another of the words – there
are more to come in Chapter 4 – which it seems that sociologists cannot
do without, even though they may be as much of a hindrance as a help
in the identification of the critical memes or practices whose adaptedness
explains the behaviour observed.
Any attempt at comparison across different cultures has to reckon not
only with the scepticism of an anthropologist like Edmund Leach, who
claimed that he had ‘almost reached the conclusion that the word has
no meaning whatever’ (1982: 133), but with that of a historian for whom
‘magic, as a definable and consistent category of human experience, sim-
ply does not exist’ (Gager 1992: 24). But it can be asked about any culture
or sub-culture whether its members draw a distinction between know-
ledge and related skills regarded as open, legitimate, and accessible, and
knowledge and related skills regarded as secret, illegitimate, and access-
ible only to initiates. Forms of knowledge and skill accepted as legitimate
may also be believed to be accessible only to disciples who have under-
gone prolonged training or are endowed with exceptional talent, like
Freudian psychoanalysts or Ndembu ‘doctor-adepts’,7 or apprentices who
have mastered the ‘mysteries’ of their craft. But when a Greek agyrtēs and
a Visigothic sorcerer both lay claim to a mysterious ability to control the
weather, or African and Roman farmers alike complain that their neigh-
bours’ crop-yields are due to their surreptitious invocation of a supra-
human power, or ‘witches’ (whatever the vernacular term) are believed
to be inflicting otherwise inexplicable misfortunes on their neighbours
by supernatural means in Europe, the Americas, sub-Saharan Africa,
India, Madagascar, and Papua New Guinea, these different cultures all
have in common a distinction between ‘good’ knowledge and ‘bad’. The
practitioners may continue to be disapproved of even when the know-
ledge becomes generally accessible. But they then cease to be ‘magicians’.8
7
The parallel extends to the requirement that the female Ndembu doctors ‘must at some time
have been patients (ayeji) themselves, and may not become practitioners until they are generally
regarded as having been cured after undergoing the ritual’ (Turner 1968: 57).
8
In some cases where the word is used, it is questionable whether the ‘magician’s’ esoteric knowl-
edge or skill is culturally stigmatized at all: ‘magic’ was ‘by no means illegal or socially deviant
in Egypt’ (Rigner 1995: 53). In early colonial Ghana, by contrast, a pejorative general ethnonym
was applied to outsiders stigmatized as ‘uncouth primitives’ who were nevertheless believed to
‘possess access to an array of the most potent ritual powers, together with the esoteric knowledge
necessary to mobilize them in defence of the cultural order’ (Parker 2004: 401).
106 The theory of cultural and social selection
Thus, poisoning is explicitly associated with magic in Cicero’s (as before
him in Plato’s) writings and in the Lex Cornelia de sicariis et veneficis of
81 BCE, and the rumours surrounding the death of Germanicus (Tacitus,
Annals II.69) are echoed by those surrounding the affaire des poisons at
the court of Louis XIV (Somerset 2003). But in a cultural environment
which includes an accepted pharmacology, poisoning loses its occult sta-
tus (even if it is still not as easy to detect as police and prosecutors would
like). Conversely, technological expertise can continue to be suspected as
‘magic’, as it often was in Renaissance Europe, until its efficacy, however
difficult for the uninitiated to explain, comes to be taken for granted as
part of everyday routine.
Once the reproduction and diffusion of ‘magical’ memes is approached
in this way, the confusion which bedevilled much of the earlier literature
between behaviour which is the phenotypic effect of ‘instrumental’ and
‘expressive’ memes is dissolved. Nineteenth- and early t wentieth-century
anthropologists and sociologists were often too ready to designate as
‘magic’ behaviour-patterns which are the phenotypic effects of memes
involving no belief one way or the other about the effectiveness of the
behaviour in enlisting or manipulating occult agencies. Sometimes,
admittedly, it is difficult to get far enough inside the natives’ heads to tell.
When Malinowski (1935: 101) reports the Trobriand talala (‘we make to
flower’) rite as ‘meant to impart fertility to the soil’, on what evidence can
he be sure whether the Trobrianders are celebrating a process which they
know will succeed as a matter of course, seeking to persuade an unseen
quasi-personal agency to ensure that it does, or acting out a strongly felt
wish that the soil should be as fertile as it can be made? But where wax is
melted in an ‘execration ceremony’ (Faraone 1993: 63), the wish thereby
symbolized, although it may have the perlocutionary effect of frightening
those against whom it is directed, is no more the acting-out of a package
of memes which includes a causal belief than is, for example, the carv-
ing of an inscription in fourteenth-century Andhara which threatens that
‘He who steals endowed land … will be reborn as a maggot in excre-
ment for 60,000 years’ (Talbot 2001: 12).9 There are also performative, as
opposed to either instrumental or expressive, rituals. When the Roman
The surviving Anglo-Saxon charters are similarly full of threats to the effect that, for example,
9
‘if anyone is so presumptuous towards God as to desire to alter this, God Almighty shall remove
him from the bliss of heaven to the abyss of hell, unless before his death he make amends as fully
as possible’ (Robertson 1939: LXXXV). Did the bishops and earls who witnessed the document
really believe that? Selectionist theory suggests that what is going on here is not an affirmation of
belief but the signalling of a presumptively costly commitment to cooperation (Sosis 2003; Sosis
and Alcorta 2003).
Cultural selection and acquired behaviour 107
fetial priests threw a spear into enemy territory – or, if the border was a
long way away, into a piece of land near the temple of Bellona which by a
deliberate legal fiction was designated as such (Beard et al. 1998: 132–3) –
they thereby transformed a state of peace into a state of war in exactly the
same way that the Maring, by hanging ‘fighting stones’ from the central
poles of their houses, transform opponents into enemies to be engaged in
full-scale battle with lethal weapons (Rappaport 1999: 75). But in these
cases, the memes in the natives’ heads are not to be equated with the
memes encoded in charms, spells, or incantations which are believed to
cure a sickness, excite the passion of a lover, safeguard a valued posses-
sion, frustrate a competitor, or appease the ghosts of the unburied dead. If
the critical memes acted out in the observed behaviour have not been cor-
rectly identified at the outset, the selectionist just-so story of their repro-
duction and diffusion will be in danger of generating an explanation of
something which was not going on at all.
In practice, the reverse engineer trying to infer from the ethnographic
or historical evidence the essential features of an adaptive memetic design
will often find that its successful reproduction depends on a combination
of beliefs and attitudes of which some are locally defined as magical at
the same time that others are not. ‘Magical’ memes can often be detected
in among both the routine, down-to-earth precepts of doctors, farmers,
or technicians on the one side and the abstruse, high-flown speculations
of preachers, theologians, or mystics on the other. The competition for
adherents is often a competition between rival bundles of memes in which
the adaptedness of claims to acquaintance with questionably acceptable
occult agencies accessible only to initiates depends on their combin-
ation with straightforwardly legitimate claims of practical efficacy. When
Bishop Hincmar of Rheims recommended medicinae ecclesiasticae as a cure
for sexual impotence (Flint 1991: 294) he was linking a claim to reputable
authority to what might otherwise be thought a suspect claim to disrep-
utable expertise, just as the self-styled ‘occultists’ of nineteenth-century
England linked the prestige of Victorian science to what might otherwise
be thought a suspect claim to have discovered a hidden, supra-sensible
world (Owen 2004) – a claim which some eminent contemporary practi-
tioners of ‘science’ were willing, at least provisionally, to entertain (Lamont
2004). There is no reason to suppose that either Bishop Hincmar or Sir
William Crookes (President of the Royal Society) were cynical confection-
ers of whatever combinations of memes they thought likeliest to appeal to
the largest number of people. But whether they were or not, they might as
well have been – as were, for example, the well-dressed, smooth-talking,
108 The theory of cultural and social selection
self-appointed Bemba witch-finders observed by the anthropologist Audrey
Richards in what was then Northern Rhodesia, whose ritual for the detec-
tion and cure of sorcery was the phenotypic effect of a judiciously con-
cocted combination of traditional and Christian memes (Richards 1935).
One of the more successful examples on record which has time and
again emerged at different times and places is the so-called ‘evil eye’. Like
the feud, it is a product of natural as well as both cultural and social selec-
tion. The anxiety aroused by a fixed and prolonged stare can be traced far
back in our evolutionary past (Burkert 1996: 43), and ‘the attribution of
suspicion to eye-to-eye contact, a straight look, is so widespread – and not
restricted to human populations – that it does not require exaggeration to
say that a psychological universal may explain why the concept is specific-
ally the evil eye’ (Spooner 1976: 284). In the Norse sagas, when a magician
is apprehended a sack is at once put over his head for fear of it (Kieckhefer
1989: 50). But the evil eye is not found worldwide. It is not unknown in
Islam, but it is not found in China. It is less common in Anglophone
than in Hispanic America (mal ojo), and less common in India than in
Mediterranean Europe. It was generally recognized in the worlds not only
of ‘Ancient’ Greece and Rome but of Byzantium. The remedies prescribed
for it typically involve expressive or performative effects as well as, or as
part of, instrumental techniques (Herzfeld 1986). In the Graeco-Roman
world, there is a lack of archaeological evidence for it to match the exten-
sive literary evidence, but this can plausibly be explained by reliance either
on words and gestures to counter it or on amulets or charms made of per-
ishable substances (Dickie 2001: 97). After the advent of Christianity, the
Church might, as in Byzantium, seek to replace the ‘pagan’ conception
of it with the idea of saintly intervention in favour of the all-seeing eye
of God (Maguire 1994), or, as in modern Greece, denounce folk remed-
ies as sinful and stipulate the use of holy water in their place (du Boulay
1974: 66). As with the feud, no lawlike cross-cultural generalization can
be framed from which its incidence could be predicted in advance. But
the environments most favourable to the probability of its continuing
reproduction are relatively small communities relatively little differenti-
ated in social distance between roles where feelings of envy are readily
evoked and relative good or ill fortune held to call for special explanation
(Schoek 1969: 261; Limberis 1991: 171; Ogden 2001: 107; Boyer 2001: 222).
Each just-so story has distinctive features of its own. But the appearance
of the evil eye in widely separate cultures under similar environmental
pressures is as clear-cut an example of convergent evolution as the conver-
gent evolution in different species of the eye itself.
Cultural selection and acquired behaviour 109
E nc ode d w h e r e ? T r a ns m i t t e d how ?
6
There is a bewildering diversity of places where memes can be encoded
and routes by which they can be transmitted from mind to mind. This in
itself is one obvious reason why the transmission of information from one
carrier to another by imitation or learning is so different from the bio-
logical transmission from one carrier to another of information passively
copied in the receiving organism with only the possibility of occasional
and limited mutation. But from the perspective of comparative sociology,
the coherence and stability of widely different cultures and sub-cultures is
evidence for the remarkable capacity of functionally equivalent memetic
combinations to reproduce those of their component units of information
which affect behaviour in the phenotype despite the information loss and
intentional or unintentional distortion and reconstruction which, except
in such cases of stock formulation as the Christian ‘Lord’s Prayer’ or the
hymns of the Indian Rig Veda, is always more likely than perfect copy-
ing. Even when a norm of conduct is transmitted from parent to child,
or mentor to pupil in strict deontic form (‘cremate the dead’, ‘blow the
trumpets at the new moon’, ‘abstain from pork and alcohol’, ‘give your
mother a present on her birthday’), there is scope for alternative interpret-
ations; and even when the distinctive behaviour-pattern observed is acted
out in the same way by every member of the population, the package of
memes inside one native’s head may, as I have emphasized already, be
quite different from that inside another’s. But the right just-so story will
always be a story of how certain critical memes have succeeded in sur-
viving translation from one to another language or medium or mode of
transmission with sufficiently little modification that they find expression
in phenotypic behaviour whose constitutive intentions are recognizably
the same. The individual differences between carrier and carrier recede as
always from view to disclose the bundles of information which have been
favoured over their competitors at population level.
The relative reproductive fitness of a complex of co-adapted memes
can be influenced both by where it is encoded and by how it is transmit-
ted. Memes encoded verbally are not necessarily more adaptive on that
account. Nor are written or spoken words, however widely diffused, nec-
essarily the repositories of the memes of which the behaviour observed in
the phenotype is the expression. There is no doubt that the printing press,
and after it radio, television, tape-recording, and email, have enhanced
110 The theory of cultural and social selection
the range as well as the speed of memetic transmission. But they have also
enhanced the volume of redundancy. The Protestant Reformation is a
familiar example where the printed word is agreed among its historians to
have had a significant influence in replacing one set of ‘religious’ memes
with another in the heads of the population of much of Northern Europe.
But the voluminous output of anti-papal pamphlets from Calvin’s Geneva
was less effective in transmitting the Protestant message than the metrical
psalms, which Calvin borrowed from the Huguenots of Strasburg and
became, in the words of the historian Diarmid MacCulloch (2003: 307–8),
‘the secret weapon of the Reformation not merely in France but wherever
the Reformed brought new vitality to the Protestant cause’. They could
be easily memorized, they could be sung by women and men alike in
streets or market-places as well as churches, and they were accessible to
those whose level of literacy fell well short of what was needed to master
Calvin’s Institutes (which, in any case, not many of them could afford to
buy). What is more, singing the metrical psalms in unison could evoke
in a Protestant crowd what MacCulloch calls ‘ecstatic companionship’ of
the kind known, thanks to recent research, to be a psycho-pharmacolog-
ical effect of the endorphins produced in the brain by singing in unison.
Not any words sung to the same tunes would have had the same effect,
any more than any rallying-cry will do as well as any other. Indeed, one
of the features of the psalms which a reverse engineer would not fail to
notice is that they gave the singers a choice which included, for example,
those of Psalm 115 which MacCulloch cites as ‘the perfect accompaniment
for smashing up church interiors’. But the words would not have had the
same effect without the tunes.
On the other hand, the printing of newspapers could promote the dif-
fusion of memes with a significant effect on phenotypic behaviour even
where the greater part of their contents was information of a different
kind included for a different purpose. An example of this is the creation
by that means of ‘imagined communities’ (Anderson 1983: Ch. 4) among
the population of Hispanic America in the years between the restriction
of the printing press to Mexico and Lima, which lasted until the end
of the seventeenth century, and the achievement of independence from
Spain. Much of the information transmitted will have entered the read-
ers’ heads as little more than noise – reports of changing commodity
prices, of the movement of ships, of marriages or social events or political
appointments which would be rapidly forgotten. But some of what stayed
behind inside their heads generated a sense of shared cultural identity
between people who, although aware of one another’s existence, had no
Cultural selection and acquired behaviour 111
personal knowledge of, or contact with, them. The ‘imagined community’
was the sub-culture of locally born creoles whose beliefs and attitudes and
consequential behaviour came to differentiate them increasingly from the
Spanish-born peninsulares who might be indistinguishable from them in
appearance, language, and religious denomination alike. As so often, the
longer-term consequences were not perceived at the time. But when the
institutional environment changed, creoles in separate regions of Hispanic
America responded to the changes in ways which they would not other-
wise have done, even to the point of putting their property and sometimes
their lives at risk in what they now perceived as a common cause.
Before information could be transmitted from mind to mind by the
technical means which are now taken for granted, the diffusion of memes
from one population to another was normally effected by the physical
movement of their carriers. Traders, soldiers, preachers, performers, crafts-
men, colonists, exiles, pilgrims, and explorers could all bring with them into
the territories into which they moved memes from their culture of origin
which could be copied, with varying degrees of distortion or reinterpret-
ation, and then reproduced down successive generations of the receiving
population (or, usually, some sub-culture within it). Buddhist memes ori-
ginating in the Ganges Valley were carried from one newly-founded hill-
top monastery to another, Dualist memes originating in the Balkans were
carried along the trade routes to North Italy and from there across the
Alps; and Christian memes were carried from Europe to China by Father
Ricci and Father von Bell, who brought with them as ‘salesmen of a special
kind’ (Landes 1998: 337) their mechanical clocks and astronomical obser-
vations. But later, too, affiliation to the nascent Social Democratic Party in
Sweden was diffused along ‘mesolevel networks’ which emerged from the
routes travelled by itinerant agitators (Hedström et al. 2000). In the course
of these peregrinations there is always scope for distortion of the carriers’
messages along the way. But the distortion observed is much less than the
scope for ongoing reinterpretation of memes might suggest. Some mem-
bers of the receiving population reject the message outright. But those who
accept it tend to accept the package as an integrated whole. Or at any rate,
they do so sufficiently for its phenotypic effects to be easily visible: the
converts conform to a new behaviour-pattern which immediately distin-
guishes them, irrespective of personal differences, from the hold-outs who
continue as before. It is Gellner’s Paradox from Section 1 of Chapter 2:
distinctive cultures and sub-cultures are much more coherent and durable
than the mutability of memes and the volatility of their individual carriers
would appear to allow for.
112 The theory of cultural and social selection
7
The transmission of memes differs from the transmission not only of genes
but of practices in the speed with which it can take place. Natural selec-
tion, even at its fastest, works too slowly in human populations to explain
cultural and social changes which can, after all, take place within a single
generation. But social selection, when contrasted with cultural selection, is
as a rule relatively lethargic: in the metaphor of Fernand Braudel (1966: II,
62), societies ‘seldom march with the steps of giants’. Information affect-
ing phenotype transmitted by imitation and learning can spread like the
proverbial wildfire. Rumours and fashions can be transmitted not only
within but across whole populations in a matter of days. In social selec-
tion, by contrast, even if a radical change is imposed by the incumbents
of roles to which there attaches a monopoly of the means of production,
persuasion, and coercion, it may take years before the novel pattern of
interacting roles is firmly in place and the practices defining them consist-
ently reproduced in the relevant dimensions of social space. Rome wasn’t
built in a day, either physically or institutionally. But Wesley’s preaching
could (we are told) convert as large an audience as his voice could reach in
the course of an afternoon.
This aspect of acquired behaviour serves also to re-emphasize the rela-
tive unimportance of formal instruction in much of cultural selection.
Almost before (as the saying goes) you can turn round, all the kids on
the block have turned their baseball caps back to front and switched
allegiances from one rock group to another. This may not be the stuff
of which major cultural changes are made. But time and again, when
a population-level cultural change is examined in detail, it turns out to
involve almost immediate imitation of behaviour observed and almost
immediate acceptance of information verbally transmitted from a source
that the recipients are willing to trust. In this, horizontal transmission of
memes is very different from vertical, where the socialization of children
involves much more transmission of memes by formal instruction over a
longer period. In many of the smaller and simpler societies documented
in the ethnographic record, the transmission of memes, and particularly
techno-memes, is more likely to be vertical or oblique than horizontal,
and to be relatively slower on that account. But in large and complex
societies, memes inherited from parents account for less of the behav-
iour observed in the adult phenotype relative to parental genes on the
one hand and horizontal transmission from the non-shared environment
on the other than used to be widely supposed (Harris 1998). Moreover,
Cultural selection and acquired behaviour 113
children are more immediately receptive to novel memes than their elders
are. This raises the theoretical possibility, which I touched on in Section 5
of Chapter 1, that among populations where the choice of life-styles is
sufficiently wide and novelty as such is sufficiently highly valued, the
rate of memetic mutation will rise to the point that cumulative selection
becomes virtually impossible. But as long as the rate of variation does not
rise to the point that it swamps the rate of reproduction, the extinction
rate of emergent sub-cultures will not be so rapid that the ever-branching
Darwinian tree grows no more solid branches but only (to remain with
the arboreal metaphor) short-lived cultural twigs.
Meanwhile, in any case, there are always the hold-outs. No matter how
fast a mutant meme may be galloping up the slope of the S-shaped logis-
tic curve, there will be some natives into whose heads it fails to penetrate.
The hold-outs’ motives may well be of different kinds, and behaviour
genetics may well have a contribution to make to the explanation of why,
within a given population, some are more resistant than others. But where
the cultural and social environment strongly sustains the continuing
reproduction of long-established ‘traditional’ memes, neither frequency-
dependence nor biased transmission in favour of prestigious role-models
can accelerate the rate of conversion. For all the dedication and skill of
the Jesuits who set up their missions along the route which Father Ricci
had followed from Macao to Beijing, the Chinese literati were unlikely to
conclude from what the Jesuits had to tell them about European astron-
omy, cartography, and music that they should take seriously a view of
the cosmic order and the place of humanity within it which seemed to
many of them ‘only a bastard form of Buddhism sometimes mixed with
borrowings from Islam’ (Gernet 1982: 455). From a selectionist viewpoint,
however, the story of the Jesuits in the Far East has a twofold interest.
On the one hand, it illustrates the obstacles in the way of any attempt to
displace a set of established beliefs and attitudes by imitation and learn-
ing alone without imposition of the political, ideological, and economic
practices which have supported Christian missionary endeavours in other
parts of the world. But on the other hand, the continuing reproduction
of the novel memes within the few Christian communities where they
achieved fixation illustrates the strength of the pressure to conform which
is brought to bear within close-knit communities of adherents to teachings
culturally defined by the majority population as unorthodox and institu-
tionally persecuted as heretical. When the Jesuits were readmitted into
Japan after their protracted expulsion under the Tokugawa regime, those
who arrived in Nagasaki were astonished to be greeted by descendants
114 The theory of cultural and social selection
of the original converts who came down from the hills clutching in their
hands the crucifixes symbolic of their prohibited faith.
8
Taken together, the speed with which memes can invade a hitherto
unaffected population and the resistance by which they can nevertheless
be met within it constitute one of the reasons for which it can be claimed
that memetic transmission is ‘very similar to the transmission of infec-
tious diseases’ (Hewlett and Cavalli-Sforza 1986: 23). But this analogy,
like the gene-meme analogy itself, is not to be taken too far. The impli-
cation of pathology may be appropriate to outbreaks of mob violence or
medieval dancing manias. It may also be of rhetorical value to polemicists
who regard ‘religious’ beliefs and attitudes as not only erroneous in con-
ception but pernicious in effect, or to heresy-hunters, xenophobes, and
counter-revolutionary propagandists eager to stigmatize their cultural
enemies as carriers of plague. To think of a meme as if it were a ‘virus
of the mind’ which copies itself wherever the opportunity offers (Brodie
1996) may help to bring out both that memes’ reproductive success can
be at the expense of their carriers and that some prospective carriers will
be resistant to it. But an infection requires only a single carrier, whereas
the displacement of one meme by another may require confirmation or
reinforcement from multiple sources (Centola and Macy 2007). Whatever
exactly memes are, they aren’t bacilli, and the ‘changes of mind’, as we
describe them to ourselves, which alter cultural behaviour-patterns do not
leave us, like diseases, either dead or crippled or cured.
One possible rejoinder is that the notion of ‘thought contagion’ (Lynch
1996) need not exclude by definition the possibility that, whatever the
information being passed from one person’s mind to another’s, it can be
benign rather than pathogenic.10 There is a whole psychological literature
on ‘disinhibiting’, ‘echo’, and ‘hysterical’ contagion as distinguishable
forms of interpersonal influence (Levy and Nail 1993), and such influ-
ence can be – according to your point of view – either for good or for
ill. But the routes of memetic transmission which sociologists find them-
selves tracing in their comparisons between one culture or sub-culture
10
There is an intriguing precedent in a letter to Rousseau of 6 June 1761 written by a Genevan
correspondent in the aftermath of a civic festival, which is more often cited for its early invoca-
tion of the triad of égalité, liberté, and fraternité: ‘si le mot Contagion pouvoit être pris en bonne
part, je vous dirois: la Contagion de l’amitié publique, avoit gagné tous les individus de la Société’
(Correspondance complète, ed. R. A. Leigh, IX: 12).
Cultural selection and acquired behaviour 115
and another are far more diverse than the direct interpersonal transfers
to which a literal reading of the term ‘contagion’ is appropriate. Memes
can be encoded in so many different places, and pass through so many
successive mutations before they find outward expression in phenotypic
behaviour, that the search for the right just-so story of their local adapt-
edness may be as much hindered as helped by the parallels which the
term ‘contagion’ suggests even if the implication of pathology has been
disavowed.
Two examples drawn from the sociology of twentieth-century Britain
illustrate this in different ways.
The first, which has the advantage of well-attested quantitative evi-
dence, is the extent to which ‘ “social contagion” effects’ (Ermisch 2005:
23) may explain the rapid rise in child-bearing outside marriage in Britain
in the last quarter of the twentieth century. The percentage rose from
9 per cent in 1975 to 40 per cent in 2000. There is no evidence for a
decline in marital fertility rates during that time, and the statistics for
aborted conceptions show no increase. The researchers did not get far
enough inside the natives’ heads to assess the reciprocal adaptedness of
their beliefs and attitudes about cohabitation, contraception, marriage,
and childbearing. But it is apparent that what was going on was imita-
tion of a chosen reference group, and that the probability of cohabitation
was a function of the expected probability of cohabitation in the group
being imitated. A mathematical model of the relation between actual and
expected proportions cohabiting can be constructed which is consistent
with the data and shows two stable equilibria between which tempor-
ary changes in the environment that alter parenting behaviour can cause
a dramatic move. The increase in childbearing outside marriage fits the
familiar S-shaped logistic curve. But it does not follow that anything is
gained by reporting it in epidemiological terms. It is a straightforward
just-so story of frequency-dependence. After a slow start, the transmission
of memes which encode attitudes and beliefs about cohabitation outside
of marriage generates a rapidly increasing rate of imitation which then
starts to decline towards a new equilibrium where the hold-outs are no
longer influenced by the expected probability of cohabitation in the ref-
erence group. It is a story more of symbolic detoxification than of qua-
si-epidemic contagion. Imitation of a reference group is not the same as
catching an infectious disease.
The second example is one which furnishes another of those quasi-
experimental contrasts of which comparative sociologists are always in
quest. A mid-twentieth-century observer studying British industrial
116 The theory of cultural and social selection
relations during the inter-war depression could well have wished to be able
to engineer the insertion, as if by introducing a bacillus, of the beliefs and
attitudes of militant trade-unionism into the working-class population of
a region where unemployment was relatively low and the workforce rela-
tively unorganized in order to see whether militancy proved to be conta-
gious or not (Whiting 1983). On a ‘contagion’ hypothesis, the expectation
would be that the militant memes would start to be transmitted from the
newcomers to those of the local workforce with whom they came into
contact and that these would in turn infect their fellow locals until the
point was reached where individual attributes of the hold-outs were such
as to keep them immune. But that is not what happened when the experi-
ment was, in effect, realized by mineworkers with a strong cultural trad-
ition of militancy who migrated from regions of high unemployment to
where the expanding motor-vehicle industry was recruiting workers for
its assembly-lines. The locals did not, in the event, ‘catch’ the newcomers’
memes. Instead, they adopted a strategy of free-riding on the increasingly
intransigent negotiations with the employers which the newcomers con-
ducted in their institutional roles as trade-union convenors empowered to
organize collective withdrawals of labour.
9
I remarked in Section 3 of Chapter 1 that sociologists are often confronted
with a problem of overdetermination. There are so many selective pres-
sures favouring the reproduction and diffusion of a practice such as wage-
labour, or formal education, or military conscription which has invaded a
hitherto unexposed population that the increase in the number of wage-
workers (and their employers), students (and their professors), and soldiers
(and the officers commanding them) invites more explanations than it
needs. But there are cultures, too, in which so many memes which appear
to be altering phenotypic behaviour are encoded in so many media of
transmission and transmitted along so many pathways that the reverse
engineer, who cannot conduct the experiments by which the problem
could in principle be resolved, is at a loss to distinguish the critical muta-
tions from the noise.
A familiar topic on which this difficulty is particularly acute is the com-
plex of memes and practices subsumed under the concept of ‘nationalism’ –
another word which sociologists, anthropologists, and historians are quite
unable to do without but which makes it more difficult, not less, to identify
the particular memes which in different local contexts find expression in
Cultural selection and acquired behaviour 117
‘nationalistic’ behaviour.11 Is it ‘modern’, or also ‘ancient’, ‘secular’ or ‘reli-
gious’, ‘progressive’ or ‘reactionary’? Does it apply equally to appeals by
‘medieval’ monarchs to their subjects to think of themselves as ‘English’ or
‘French’, to the Anglo-Saxon representation of Engla Lond, and to the senti-
ments which Thucydides attributes to Pericles in his commemoration of the
Athenian war-dead? There is obviously a difference between the memes in
the heads of Pericles’ audience and those in the heads of nineteenth-century
European or twentieth-century African and Asian nationalists aware of
living in an environment of industrialization, secularization, and the con-
trol of the means of persuasion by the incumbents of increasingly powerful
institutional roles. But from the perspective of selectionist sociology, the
explanandum is the emergence, whether by convergent evolution, lateral
diffusion, or homologous descent, of cultures in which the universal, innate
disposition to stigmatize out-groups defined as ‘other’ takes a local form in
which self-identification as members of a distinctive in-group involves more
than language, religion, territory, and kinship by themselves. Not any trait
can, as one sociologist has put it, be ‘packaged into a nation’ (Bruce 1998:
89), any more than any words could have been set to the metrical psalms
which were the ‘secret weapon’ of the Protestant Reformation or any battle-
cry could be substituted for ‘St George for England!’ But the problem is to
account for the collective behaviour-patterns of populations whose members
would not act as they do if they did not think of themselves as belonging
to a ‘nation’ (or a word in their vernacular language which corresponds to
it, which may simply be a word designating the ‘such-and-such’ people). It
is, moreover, an exercise not only in meme-tracing but in practice-tracing.
The reverse engineer who has identified the memes whose combination and
reproduction has enabled them to outcompete memes encoding loyalty to
clan, or locality, or systact, or creed has also to identify the practices which
enhance the probability of continuing reproduction of those memes by
attaching institutional sanctions to the roles of rulers and their agents who
by controlling the means of persuasion can impose on the population a for-
mally nationalist ideology.
This, however, can sometimes be done. Consider the just-so story told by
the historian Linda Colley of the emergence in eighteenth-century Britain
of a distinctive form of nationalism which can be traced to an identifiable
combination of critical memes. The phenotypic effect to which Colley
This is brought out in the revealing admission by an author who has written extensively on ‘nation-
11
alism’ that ‘we can really only hope to demonstrate an ‘elective affinity’, in terms of subjective
meaning between certain features of national identity and ideology, and particular beliefs and
practices of one or more ethnic communities and kingdoms over a period of time’ (Smith 2003: 5).
118 The theory of cultural and social selection
draws particular attention is the readiness of young and even middle-
aged males to volunteer for military service during the Napoleonic wars
and the ‘speed and unanimity with which militia and volunteer corps in
different parts of Great Britain unfailingly assembled during the various
false alarms of a French landing’ (1992: 310). The 100,000 militiamen who
had volunteered for service by the time that the Supplementary Militia
Act of 1796 was in full effect will have done so from a variety of individ-
ual motives: fear of unemployment, deference to landlords or clergy, sus-
ceptibility to persuasion by family or friends, or simply the prospect of a
legitimate outlet for innate pugnacity. Nor does Colley neglect to mention
the evoked behaviour observable when recruiting parties arrived in small
villages and exploited to the full the impact of their wind instruments,
drums, and cymbals. But the turnout of such a significant proportion of
the age-cohort at population level reflects the reproduction and diffusion
of distinctively Protestant memes when they were combined with stereo-
typical representations of French ‘Papists’ and encoded in tracts, homilies,
newspapers, almanacs, Foxe’s Book of Martyrs, sermons, folktales, songs,
plays, ceramics (Brewer 1982: 328), and bonfire nights. Inevitably, there
are no means of reconstructing the evidence which would make possible
a conclusive test of Colley’s hypothesis. But if the data could cover in full
the provenance, upbringing, and peer-group relationships of both volun-
teers and non-volunteers, the standard techniques of multivariate analysis
would generate the quasi-experimental contrasts from which it could be
inferred not only how much a greater or lesser exposure to separately iden-
tifiable memes increased or diminished the likelihood of volunteering but
by how much more a combination of them did so. If Colley is right, the
probability of volunteering would be found to increase exponentially to
the extent that patriotic and Protestant memes were identified side by side
inside the heads of British males socialized into the distinctive culture of
‘Britishness’ which evolved in the aftermath of the ‘Glorious Revolution’
of 1688 and the Act of Settlement of 1701.
C ol l e c t i v e m i n d s a n d w h at Mov e s t h e m
10
Although selectionist theory admits no such entity as a group mind
or Zeitgeist or conscience collective, and unanimity is not to be assumed
even in the most ‘traditional’ cultures, that does not require all talk of
the ‘Greek’, or ‘Protestant’, or ‘Victorian’, or even ‘Asian’ as opposed
Cultural selection and acquired behaviour 119
to ‘Western’ mind (Nisbett 2003) to be expunged from the literature
of comparative sociology,12 any more than talk of the ‘Amerindian’ as
opposed to the ‘Cartesian’ conception of the relation between nature and
culture (Viveiros de Castro 1998) or Hindu, Buddhist, and Confucian
as opposed to Christian conceptions of ‘evil’ (Parkin 1985). Nor does it
imply any disagreement with the cultural anthropologists’ insistence that
meaning is always a function of social context and individual behaviour
inseparable from social relationships. What it does, however, require is
for such talk to be validated by accurate reports of the distribution and
ongoing reproduction of identifiable memes which significantly influ-
ence the course of cultural and social evolution. Herbert Spencer’s bulky
two-volume Descriptive Sociology, or Groups of Sociological Facts, with its
peculiar list of such headings as ‘watchmen on every public building’ and
‘smoke tobacco till senseless’, still serves as an awful warning of what to
avoid in this respect. Memes which have significant influence on changes
in large-scale cultural behaviour-patterns have to be sifted out from the
noise, and they have to be shown to reproduce with sufficient fidelity,
despite the conscious reinterpretations and unconscious distortions, for
assertions of the form of ‘the Greeks now believed …’ or ‘the Victorians’
attitude became …’ to stand up to detailed scrutiny.
It should need no re-emphasis that such assertions must be advanced
with care. Think how foolhardy it would be to generalize about, say, the
‘Christian mind’ or ‘the Buddhist mind’. Any such attempt would imme-
diately disclose the wide diversity of memes carried in the heads of self-
proclaimed Christians or Buddhists, even if all of them can be traced back
by homologous descent to their origins in doctrines attributed respectively
to Jesus, son of a carpenter of Nazareth, and Siddartha Gautama, son of
a Kshatria of the Sakya clan. Only in local sub-cultures such as (per-
haps) Anglican Christianity in mid-sixteenth-century Southern England,
or twentieth-century Theravada Buddhism in rural Thailand, might it
be possible to identify packages of memes with consistent phenotypic
effects which approached fixation at population level. In even the most
restricted environment, the search is likely to lead to the discovery not
of a core tradition but only of family resemblances between related clus-
ters of memes. The problem is the same as that familiar to sociologists
That is not to say that Nisbett’s conclusions about the different memes which he claims to find
12
in ‘Western’ as opposed to ‘Asian’ heads are correct or that the criteria which he applies in dis-
tinguishing ‘Western’ from ‘Asian’ are not open to criticism: Lloyd (2007: 169) points out the
irony that a characteristic insight of Aristotle’s ‘looks distinctly “Eastern” and conflicts with
what Nisbett takes to be characteristically “Western” ’.
120 The theory of cultural and social selection
studying the systactic structures of capitalist industrial societies who
find that survey respondents mean very different things when they say in
interviews that they regard themselves as belonging to the ‘middle class’
or the ‘working class’, even if their self-attributions turn out to correlate
significantly at population level with their life-styles and voting habits.
Conversely, similarities between different local environments can explain
the convergent evolution of, for example, representations of intermediate
supernatural beings to whom there is attributed the power to influence
human affairs – the saints and devils of Christianity on the one side and
the devas and Nat spirits of Buddhism on the other.
Construction of the right just-so stories of cultural evolution is often
made more difficult still by the fact that just as the same individuals
can occupy roles within several different institutions, so can they carry
in their heads the memes of several different sub-cultures, like Bede’s
Anglo-Saxon king cited in Section 1. To go back to the elusive concept of
‘nationalism’: what did ‘Englishness’ mean to the population of England,
as opposed to ‘Britain’, at the beginning of the twentieth century? That
it did have a meaning shared by the overwhelming majority of the popu-
lation of England can hardly be doubted. Both domestic and foreign
observers were well aware of an English sense not only of nationhood
but of national superiority – which Engels characteristically denounced
as a sense of ‘imaginary’ national superiority – extending across all sys-
tacts and milieux. But bellicose patriotism co-existed both with a marked
sub-cultural differentiation of working-class from middle-class attitudes
and beliefs13 and with what one historian calls the ‘culture of local xeno-
phobia’ which on occasion gave loyalty to ‘parish, township, hamlet, or
estate’ (Snell 2003: 4) precedence over loyalty to either nation or class.
The volunteers from all sections of their local communities who took part
in the pageants celebrating the Alfred Millenary in 1901 (Readman 2005:
171) were giving expression to an unforced interest in commemorating an
explicitly ‘English’ past. But that could be quite consistent with a repudi-
ation of the beliefs and attitudes of the English elite and a refusal to accord
approval to what many saw as England’s exploitative mode of production,
inegalitarian mode of persuasion, and undemocratic mode of coercion.
If there were some memes affecting phenotypic behaviour which virtu-
ally all English men and women had acquired by imitation or learning
from parents or teachers during the course of their upbringing, there were
Gender and ethnicity can be relevant also, as illustrated by Byrne (2007) in a study based on
13
11
The best prospect for identifying the critical memes is where there is a
diversity of sub-cultures within a larger common culture and the environ-
ment changes in such a way as to reveal those whose mutation or, as it may
be, extinction has had a detectable effect on phenotypic behaviour across
the population of the common culture. Of this, there is an instructive
example in one well-documented part of the culture of ‘ancient’ Greece.
The ‘Greek mind’ included a unanimous awareness of a shared cultural
identity which distinguished the ‘Hellenes’, as they saw themselves, from
the Egyptians to the south, the Persians to the east, and the Thracians
and other ‘barbarians’ to the north (Macedon being a socially as well
as culturally troublesome borderline case). But the hundreds of separate
poleis (and the tribal ethnē which survived alongside them) transmitted
down successive generations their own distinctive cults, coins, calendars,
dialects, and artefacts whose accurate reproduction was sustained by
strict in-group conformity and sustained out-group hostility. Their sense
of shared Hellenism did not inhibit them in the least from fighting one
another. On the contrary: peace was, as Plato said (Laws 626a), merely
a state of undeclared war between poleis; ‘neighbourtown’ (astygeitones)
wars, as Aristotle called them (Politics 1330a), were regarded as a normal
part of the experience of the adult male citizen; and Herodotus (I.30.4–5)
has Solon say that an otherwise unknown Athenian called Tellus was the
happiest of men because he died in a victorious battle in precisely such
a war and was accorded a public burial at the spot where he fell. The
Hellenic culture, in other words, of which all Greeks saw themselves as
members, was among other things a warrior culture which not merely
sanctioned but encouraged organized lethal violence against fellow
Greeks whose sub-cultural differences it thereby helped to preserve and
even exaggerate. But this warrior culture was being radically modified
under an unanticipated selective pressure at the very time that Plato and
Aristotle were taking it for granted.
At that time, intra-Hellenic wars were being fought by citizen
armies of heavy-armed infantry drawn up in close order (and instinct-
ively edging, as I mentioned in Section 6 of Chapter 1, to the right)
122 The theory of cultural and social selection
who engaged in hand-to-hand fighting with stabbing weapons which
was brutal, strenuous, and frightening to a degree unparalleled among
other societies of peasant agriculturalists (Dawson 1996: 50). It can-
not be explained in terms of either a behavioural-ecological hypothesis
about maximization of inclusive reproductive fitness or an evolutionary-
psychological hypothesis about a supposedly universal ‘territorial
imperative’. Although sexual exploitation of foreign or servile women
was a matter of course (Dover 1974: 210) and concubines were appor-
tioned among victorious Greeks after the defeat of the Persians at Plataea
(Herodotus IX.81.1–2), women were not normally part of the spoils of
victory. The enslavement of the women and children of a defeated polis,
where it occurred, was motivated by the wish to inflict exemplary pun-
ishment, not to augment the victor’s population. Nor was the objective
the annexation of territory. The enemy’s crops might be destroyed and
vines deliberately damaged, but without occupation and subsequent
exploitation of their productive land. The citizen-soldiers of the hop-
lite phalanx fought to hold their own against potential threats to their
independence, or to take or retain control of a strategic location that
might be used against them, or to support an ally who called for help,
or to exact revenge, or to prevent a threatening accretion in the power
of a rival polis. The true significance of hoplite warfare, in the words of
one historian, is to be found ‘in its close links to almost every major
feature of the culture’ (Connor 2004: 35), or in the words of another, in
its being ‘more than a tactical formation. It represented a way of life, a
code of manliness and morality’ (Ferrill 1985: 145).
Young Greek males acquired from parents, mentors, and role-models
a criterion of prestige whereby it was accorded not to heroic warriors of
the type of Hector and Achilles but to those who could be relied on not
to break ranks. Commanders fought alongside their men and shared with
them a local social life in which they knew each other by name and ate
(and more importantly drank) together in clubs or messes (Murray 1991).
Cowardice was not merely stigmatized individually but ideologically
sanctioned by strong reciprocity. Victories were celebrated, but never by
Roman-style triumphs. Battles were fought in accordance with a sequence
of prescribed rituals in which invocation of chosen gods was combined
with symbolic reaffirmation of civic solidarity: omens, oaths, prayers,
libations, divinations, sacrifices, trophies, dedications, funerals, and com-
memorations constantly reminded performers and their audiences of the
collaboration incumbent on them as fellow Athenians, fellow Spartans,
and the rest. Whatever the proportion in the various armies of reluctant
Cultural selection and acquired behaviour 123
conscripts and cowardly hoplites (Christ 2006: Chs. 2 and 3), the solidarity
reaffirmed and strengthened by wars between poleis was both a collect-
ive self-differentiation against the ‘other’ and a cultural celebration of the
social role of the self-governing, arms-bearing, adult male citizen.
The obligation to fight when called on to do so was supplemented in
the package of mutually adaptive memes by specific injunctions both to
commemorate the fallen and to dedicate the spoils of war to the appro-
priate god. The tradition of paying homage to the fallen pervades not
only the literary sources, including the speeches put by the historians into
the mouths of generals or political leaders, but also the many inscrip-
tions recorded by Pausanias in the first century of the Christian era and
confirmed by subsequent archaeology. At Athens, an empty bier was gar-
landed and carried in procession for any whose bodies had not been recov-
ered (Thucydides II.34.3–4), and Plato has Socrates emphasize (Menexenus
234a) that the poorest citizen who dies in battle will have no less splendid
a funeral than the richest one. At the same time, the helmets and shields
of the enemy dead were used to ‘crown the pure dwelling-places of the
gods’ (Aeschylus, Seven Against Thebes, line 278) on a remarkable scale,
both at local temples and at pan-Hellenic sanctuary-sites such as Delphi
and Olympia. It has been calculated by applying a multiplier taken from
the surviving Panathenaic amphorae that an average of 1,000 helmets
per year were dedicated by Olympia alone between 700 and 500 BCE
(Snodgrass 1980: 131). At this distance, and without the benefit of evidence
from participant-observation or questionnaires, it is difficult to tell how
far the guiding motive was conspicuous display of superior status rather
than expression of gratitude to a quasi-personal supra-human agency
presumed to have favoured the winning side. But bronze was expensive,
and the hoplite panoply required many hours of skilled labour. Economic
rationality had nothing to do with it.
The subsequent demilitarization of the collective Greek mind is a classic
just-so story of meme-practice co-evolution. The change which brought it
about was the rapid expansion in and after the Peloponnesian war in the
number of mercenaries who fought in intra-Hellenic wars on Hellenic ter-
ritory. The role of mercenary warriors is attested from the earliest period
of hoplite equipment and tactics, and surplus males, particularly from the
more mountainous and infertile regions of Greece, were always available
for hire in the service of the rulers of Egypt or Lydia. But as campaigns
became longer, tactics more diversified, sieges and ambushes increasingly
supplemented pitched battles between opposing phalanzes, the ville-foyer
(Garlan 1974: 277) from which citizen-soldiers sallied out to defend their
124 The theory of cultural and social selection
farmsteads became a ville-bastion protected by elaborate fortifications,
and commanders became managerial strategists rather than fighting tac-
ticians. Richer societies recruited increasing numbers of mercenaries from
poorer ones and the ratio of mercenaries to citizen soldiers steadily rose.
Mercenaries presented themselves at known centres of recruitment, and
would appear ‘even unbidden’ (McKechnie 1989: 87) at places from which
rumours had spread about forthcoming military operations. The mercen-
aries formed cooperative groups with their own internal sub-cultures. The
memes which had found phenotypic expression in the campaigns fought
by citizen-hoplites were steadily driven to extinction. Young Greek males
no longer assumed, as their fathers and grandfathers had done, that they
would at some time in their lives have to be ready to prove themselves
in combat. Rituals commemorating the fallen were no longer performed
when these were ‘strangers’ (xenoi), as Xenophon and Demosthenes called
them. Panoplies taken from the enemy dead were no longer dedicated in
local temples or at pan-Hellenic sanctuary-sites. And the stigma attaching
to cowardice steadily weakened as the miles gloriosus became a stock fig-
ure on the comic stage (Humphreys 1978: 174, citing Webster 1953: 39).
12
So far, it might be said, so good. But selectionist theory has also to account
for what doesn’t go on – that is, for the absence of change where it might
be expected as well as its occurrence where it might not. In cultural evo-
lution, where both frequent mutation and rapid diffusion of memes are
commonplace, there are at the same time cultures where it needs to be
explained why the collective mind remains unmoved even though the
population is open to invasion by a variety of competing beliefs and atti-
tudes expounded by active and eloquent innovators. Of this, the culture
of ‘ancient’ Rome is a particularly revealing case because of its unique
combination of ‘religious’, ‘scientific’, and ‘magical’ memes transmitted
down successive generations of the Roman population from the early
monarchy until the adoption of Christianity as the official religion of
the Roman state. Despite widespread and sometimes drastic changes in
its social institutions, the culture of Rome passed down from generation
to generation with no effective challenge whatever to ancestral tradition.
Why not?
Nowhere in Roman culture did there ever emerge the idea that the wis-
dom of the ancestors (maiores) ought to be discarded and replaced, or that
a ‘Mandate of Heaven’ might be withdrawn from an unjust ruler, or that
Cultural selection and acquired behaviour 125
entitlement to occupancy of a political role might depend on conformity
to a standard of morality laid down from on high, or that rulers might be
under a binding obligation to meet the needs or further the interests of
the ruled. There is no reason to doubt the authenticity of Cicero’s disap-
proval of the abuse of power by a provincial governor like Verres, or his
admiration for Brutus and Cassius as tyrannicides, or his willingness to
regard Roman laws of inheritance as framed for the benefit of men and
therefore unjust to women (de Republica III.x). But it is symptomatic that
when, in his forced retirement from politics, he set himself to write two
dialogues based on Plato’s Republic and Laws, his ideal republic turns out
to be the Roman state and his ideal laws those of the state already agreed
to be best, i.e. Rome. Neither the pontifices nor any other of the priest-
hood saw themselves as carriers of a transcendental ideal by which the
holders of economic, ideological, or political power might be judged and
found wanting. Nor, for all the interest shown by literate Romans in the
doctrines of the Stoics, did that interest extend beyond acknowledgement
of the ideals of moderation, clemency, detachment, and self-discipline in
the exercise of power. There is nothing in the writings of Posidonius (so far
as they can be reconstructed), who appears to have been the philosopher
most admired by members of the Roman elite, which suggests other than
that the triumphs of Roman imperialism were comfortably in accordance
with divine providence. However many Romans dabbled in philosophical
speculation, or combined together in the name of a chosen divinity for
ceremonial purposes, this never amounted to a self-identification with a
creed and associated injunctions about the right governance of the good
society. At most, it was felt that, as one authority puts it, ‘the gods could
not be asked positively to maintain the state, while the Romans were
behaving in ways that must undermine it’ (Liebeschutz 1979: 100).
Yet this lack of what Weber called a ‘salvation ethic’ has to be recon-
ciled with the abundant evidence that Romans at all systactic levels, and
not merely those whom their social superiors regarded as ‘stupid’ (Ovid,
Fasti II.531), shared with Marcus Aurelius, as quoted in Section 3, a belief
in the capacity of unseen quasi-personal agents to influence the course of
both natural events and human affairs. It is unlikely that any of them ser-
iously expected to see gods appear on earth in seemingly human form.14
14
Pausanias says he believed that gods at one time used to appear among men but then ceased
to do so (VII.2.4), as does Catullus in a poem (LXIV.386) where the gods are alleged to have
appeared in the time ‘when piety was not yet held in contempt (nondum spreta pietate)’. Compare
Sir Thomas Browne in Religio Medici: ‘that Miracles are ceased I can neither prove nor absolutely
deny’.
126 The theory of cultural and social selection
But not only prayers and sacrifices but the reading of omens, the cast-
ing of spells, mystery cults, haruspications, incantations, dedications, and
incubations were facts of everyday Roman life. There were the predictable
sceptics, scoffers, and cynics, and the politicians who manipulated cus-
tomary rituals for their own self-serving purposes. But the consultation
of the Sibylline Books was not merely play-acting, and the despatch of
Fabius Pictor to consult the Delphic Oracle after the Carthaginian vic-
tory at Cannae was not merely a hypocritical exercise in public relations.
The Romans regularly celebrated seasonal festivals, displayed votive ter-
racottas at sanctuary sites, erected statues of guardian spirits in houses or
at crossroads, credited practitioners of the occult with suspect ‘magical’
powers, and treated certain places, spaces, and features of the natural
world as sacred. Due dates were studiously adhered to, traditional formu-
lae repeated, and gods correctly addressed.15 Conformity to established
procedures and formulae did not rule out reinterpretation of inherited
memes: an example for which there is good epigraphic as well as literary
evidence is the cult of the ‘Arval Brethren’, whose rituals, after being ini-
tially concerned with ensuring the fertility of the soil, were modified to
include vows and sacrifices directed to the well-being of successive emper-
ors. Foreign deities were admitted into the Roman pantheon, and foreign
cults tolerated as long as they posed no threat to the Roman state. But
nowhere in all this can there be found any serious challenge to ancestral
tradition.
The answer to the conundrum is to be found not in Roman religion
(however ‘religion’ is defined) but in the memes and practices constitutive
of Roman law. From its shadowy beginnings in and before the Twelve
Tables of the mid-fifth-century BCE, historians and sociologists of all
theoretical persuasions have acknowledged its uniqueness. The early leges
regiae seem to have been as much ius divinum as ius civile. But even if
some trace of supra-human authority survives in the Twelve Tables, it dis-
appears in the simplified legis actio (Jolowicz and Nicholas 1972: 79). The
selectionist just-so story can be summarized by saying that the politiciza-
tion of Roman religion evolved in co-adaptive relation to the seculariza-
tion of Roman law. It is a story of meme-practice co-evolution in which
the practices defining the roles of the Roman priesthood clearly reveal
just what was going on. The pontifices were for a long time responsible
Not that this was peculiar to paganism: Christian reverentia towards the saints involved a
15
homologous ‘etiquette towards the supernatural, whose every gesture was carefully delineated’
(Brown 1981: 119).
Cultural selection and acquired behaviour 127
for control of the Roman calendar as well as for the rules which ordi-
nary citizens were required to follow in such matters as the burial of the
dead. But they never claimed theocratic legitimacy, and they never chal-
lenged the presuppositions on which the ius civile rested. The conflicts
which periodically broke out among the incumbents of priestly roles were
social conflicts over who was to occupy the roles to which there attached
control of the priestly colleges, not cultural conflicts over doctrine.
Roman priests were no more disposed than Roman jurists to question
the entitlement of the rulers to rule, and the jurists were explicit about
the impropriety of challenging the rationes for things ancestrally estab-
lished (a maioribus constituta). Although the jurists recognized an implicit
distinction between the spirit and the letter of the law – aequitas versus
ius – they concerned themselves in their opinions (responsa) only with the
application of traditional legal concepts and procedures to the particular
needs and circumstances of those to whom their advice was being given:
nowhere did any of them expound a doctrine of rights or a philosophy
of law which could be construed as seriously critical of the system with
whose interpretation they were charged. Anachronistic laws were not
abolished but simply allowed to lapse. In the absence of an earlier law
which could be taken as a precedent, the appropriate guide was custom
(consuetudo). When the eminent Gaius Cassius Longinus, as reported
by Tacitus (Annals XIV.42–3), said that he had never doubted that what
had been laid down in former times was always right, and change always
for the worse, he was voicing conventional senatorial opinion at its most
uncompromisingly reactionary. But he was not arguing against would-be
reformers for whom respectful allegiance to the wisdom of the ancestral
maiores as inherited by the senatorial patres was inherently misconceived.
There were no heterdox reformist memes against whose carriers argument
needed to be made.
In the Roman mode of persuasion, the practices which defined the role
of priest made its incumbents agents of the Roman state no less than did
those which defined the roles of the magistrates in the mode of coercion.
Not every magistrate was a priest. But priests and magistrates shared a
common systactic origin as well as inheriting the same traditional set of
memes. (They also, in many cases, shared their genes, although high mor-
tality rates among the senatorial elite entailed frequent recourse to adop-
tion.) Rome was not a ‘traditional’ society in the sense that its political
and ideological institutions were embedded in each other to the degree
that no distinction could be drawn between them. But it evolved in such
a way that the law became disembedded from religion without religion
128 The theory of cultural and social selection
becoming disembedded from politics. Ambitious individuals had no
scope for upward social mobility within a Roman church institutionally
distinct from the Roman state. When Augustus as emperor arrogated to
himself the role of pontifex maximus, his behaviour was consistent with a
trend going back two centuries earlier when a pontifex maximus had been
at the same time a consul. Not until after the conversion of Constantine
was there an area of social space in which mutant practices could form
roles whose incumbents might have been carriers of memes expressive
of a tension between some transcendental moral standard and the per-
formance of their traditional economic, ideological, and political roles by
the acknowledged members of the Roman elite. Cultural inertia, in this
instance, was due neither to memetic isolation nor to the costs of imita-
tion or learning, but to the effect of the practices defining the roles con-
stitutive of Rome’s ideological and political institutions on the ongoing
course of meme-practice co-evolution.
W i n n i ng or t hod ox i e s a n d l o s i ng h e r e s i e s
13
In cultural evolution, heresies are losers by definition, since if they succeed
they become orthodoxies in their turn. But the outcome of the competition
between orthodoxy and heresy is, for the reasons outlined in Chapter 1,
seldom decided by dispassionate assessment of intellectual merit: even in
the history of mathematics, there are innovators who to their successors
are pillars of orthodoxy but were initially stigmatized as heretics. We have
seen that where the carriers of memes culturally defined as unorthodox
are subjected to social sanctions as heretical and therefore subversive, the
‘mere adherents’ can be expected to fall away at the same time as the
zealots are strengthened in their commitment. But the prospect of impos-
ition of institutional sanctions only arises after the heretical memes have
been sufficiently widely diffused by imitation and learning for the need
to counter them to be perceived by the incumbents of the roles to which
there attaches the power to impose the sanctions.
In any culture, the growing child inherits an initial set of memes from
parents or other guardians and mentors. Where both parents share the
same memes with institutional mentors such as schoolteachers and priests,
the probability of their ongoing reproduction in relatively unmodified
form is high. As they grow towards adulthood, some will be more recep-
tive to the alternatives than others for reasons which behaviour genetics
Cultural selection and acquired behaviour 129
may be able to explain.16 But in comparative sociology, the problem for
the reverse engineer is to identify what it is in the memetic design of a
heresy which enables it to mount a successful challenge to the established
orthodoxy at population level. Designed adaptedness may be involved,
but it may be involved on both sides in a cultural arms race. The memetic
engineers of the Roman Catholic Church who transferred the doctrine
of purgatory from the fourth book of Gregory’s Dialogues across into the
appendix to the Second Lateran Council’s constitution cum sacrosancta
(Le Goff 1984: 285) could fairly congratulate themselves on having pro-
moted a heightened consciousness of sin among the faithful at the same
time as encouraging bequests for masses to be said on behalf of the dead.
Conversely, sectarian competition at a time of religious revival can sig-
nificantly enhance the adaptedness of rival memes (Froese 2001). Chance
can be expected to play its usual part. The great Byzantine controversy
between Iconophiles and Iconoclasts which the Iconophiles finally won
in the year 843 was only settled after a succession of wholly unpredictable
twists and turns in dynastic inheritance, ecclesiastical policy-making, and
the fortunes of war. The leading protagonists may have had strong and
well-argued views about whether images are symbols through which wor-
shippers can be brought closer to the unseen reality which exists beyond
the range of human vision or whether to treat them as objects of worship
is a blasphemous repudiation of God’s own commandments to the people
of Israel. But how many of the worshippers put it to themselves in those
terms? Reproduction of Iconophile memes was favoured by the appeal of
their extended phenotypic effects among carriers for whom pictures of
Christ, his mother, his apostles, and his saints in churches and private
houses answered a felt need for visible representations of unseen powers
thought capable of giving or withholding comfort and protection in times
of misfortune and danger. But the existence of such a need is no guarantee
that it will be met. The right just-so story has to identify both the critical
memes whose probability of diffusion and reproduction was enhanced
as the environment changed and the critical events, which might as well
have been random, which changed the environment as they did.
Not all orthodox memes are expelled from a cultural population
through direct confrontation with their heretical competitors. They may
simply cease to be transmitted down successive generations as they pre-
viously were. According to the so-called ‘secularization thesis’, which
was itself something of an orthodoxy among late twentieth-century
16
Relevant also may be association of rebelliousness with birth order (Sulloway 1996).
130 The theory of cultural and social selection
sociologists of religion, this is just what was going on in the many cultures
whose members no longer shared their parents’ or grandparents’ beliefs in
gods, demons, witches, ancestral spirits, miracles, portents, prophecies,
and ghosts. That in the course of cultural evolution many such beliefs have
gone extinct in populations among which they had once seemed close to
fixation is not in doubt. But a decline in the rate of reproduction of ‘reli-
gious’ practices and a narrowing of the scope of ‘religious’ institutions are
perfectly compatible with an increasing intensity of individual ‘religious’
experiences (Gorski 2000: 162). The ‘secularization thesis’ is as mistaken
if it is interpreted as positing an extinction of beliefs in quasi-personal
supra-human agents as if it is interpreted as positing an extinction of sac-
ralizing attitudes. Adaptive mutations within long-established traditions
such as Judaism, Christianity, and Islam have taken the form not only
of heretical variants of the orthodox core but of explicit reaffirmations
of orthodoxy of the kind conventionally labelled ‘fundamentalism’. The
problem for selectionist sociology, here as elsewhere, is to separate cases
of meme-practice co-evolution from cases where the universal, innate
psychological receptivity to memes constitutive of metaphysical beliefs
and sacralizing attitudes generates a new ‘religion’ by cultural selection
alone. The selective pressures in the institutional environment are as obvi-
ously relevant to the probability of reproduction and diffusion of the
memes which originated in the heads of Confucius or Marx as to that
of the memes which originated in the heads of Muhammad or Luther.
Commitment to any set of shared metaphysical beliefs and sacralizing
attitudes will be biologically adaptive to the extent that it enhances the
probability of within-group cooperation and therefore, as Darwin envis-
aged, a higher probability of reproduction of the genes of the members of
the group relative to the groups with which they are competing. But what
will explain the cultural adaptedness of the particular memes which have
this effect, and where will the reverse engineer find cultural group selec-
tion at work in its purest form?
Of all the heresies which have become orthodoxies, early Christianity
affords the most promising example on record for addressing that ques-
tion. This is not because Christianity’s eventual success would have
been as astonishing as it assuredly would to any contemporary obser-
ver. The same could be said of Islam and the extraordinary sequence of
chance events which guided its early history. It is because, unlike Islam’s,
Chistianity’s just-so story comes in two distinct parts, first of cultural
and then of social selection. Without Constantine’s victory at the Milvian
Bridge and his attribution to it of the support of the Christian God,
Cultural selection and acquired behaviour 131
Christianity would have remained one among many rival creeds compet-
ing for adherents within the Roman world. Only after the institutional
innovations which followed, including not least the immunities granted
by Constantine to the clergy ‘in the first ardour of his conversion’ (Jones
1964: 92), did the economic, ideological, and political sanctions attaching
to the roles of the emperors and their subordinate officials guarantee its
success. But Constantine’s choice would not have arisen at all had not
such a small group become such a big problem (Brown 1971: 65) – so
big that the emperor Diocletian was prodded by his Caesar Galerius into
publishing the edict of February 203 which initiated what the Christians
called the ‘Great Persecution’. The early Christians had been members of
a small, obscure, and studiously apolitical Jewish sect. As it grew, the roles
of elder or preacher evolved into the role of bishop, and bishops became
local grandees with whom Roman officials dealt as such. But the diffusion
of Christian memes prior to Constantine’s conversion is a just-so story of
cultural selection to which social selection is almost totally irrelevant.
14
During that early period, there was none of the organized mission-
ary activity of the kind later undertaken across Europe and beyond.
Christian memes were acquired by informal transmission at workplaces
or on street-corners or among friends or by the personal influences of an
estate-owner or household head whose dependents followed his (or some-
times her) lead. The younger Pliny’s celebrated testimony, in a letter to
the Emperor Trajan from Bithynia, to the universality of Christianity’s
appeal has to be qualified by its rarity in the Roman army, among the
senatorial elite, or in the barracks of the rural gang-slaves. But it was nei-
ther (as Nietzsche maintained) a ‘slave uprising’ of the oppressed nor (as
Weber maintained) a religion of urban artisans who had bought, or were
buying, their freedom. Christians were, as Pliny told Trajan, drawn from
all ages, all ranks of society (‘omnis ordinis’), and women as well as men.
Yet it is not as if Christianity was unique in holding out to its adherents
the hope of a life after death or the comforts of ceremonial, mystery, and
companionship in this one. Nor was it unique in its monotheism, since
even apart from Judaism an increasing number of philosophers and intel-
lectuals were ready to postulate a single, supreme deity. Nor was it unique
in its offer of supernatural help in sickness: numerous competitors offered
the same, with the same attestations by eye-witnesses to support them
(including, on one occasion, a cure allegedly affected by the Emperor
132 The theory of cultural and social selection
Vespasian). Nor is it as if the traditional religion of Rome had lost its erst-
while support, as evidenced not least by the time it took for ‘paganism’, so
called, to be driven to the cultural margin which the carriers of Christian
memes had once occupied and non-Christians to become, as one histor-
ian puts it, ‘outlaws at last’ (MacMullan 1984: 101). The explanation of
Christianity’s diffusion has to be located somewhere in the sub-culture of
the local communities which were ‘exclusive and totalistic in a way that
no club nor even any pagan cultic association was’ (Meeks 1993: 78).
Christians were initiated through the ritual of baptism (which they
might be permitted to defer, if they so chose, until nearer to death) into
a worldwide community whose members were thereby symbolically
cleansed of sin and offered the prospect of salvation. The overwhelming
majority of them remained in their same social roles after their conver-
sion, and doctrinal disputes, acrimonious as they might be, were of as
little concern to the generality of ordinary men and women in the streets
of Rome or Antioch or Ephesus or Edessa as to those landowning gen-
try and local politicians who chose adherence to Christianity for prag-
matic reasons of their own. But they met with fellow Christians at regular
intervals to reaffirm their commitment, participated in common meals,
assisted one another during life, and cared for one another at death. They
were not expected to take literally the injunctions of the Sermon on the
Mount. They were expected only to be willing to offer some part, at least,
of their time, attention, and money to their churches. Rival Christian
communities often quarrelled with one another, and with their bishops,
thereby giving obvious pleasure to pagan commentators like Ammianus
Marcellinus. But among the injunctions which they were expected to take
seriously was a duty of benevolence towards out-group members as well as
fellow Christians.
Almsgiving as such was not a phenotypic effect of uniquely Christian
memes. But in the Jewish cultural tradition from which Christian
almsgiving descended, distributions from the synagogues were much
more closely restricted to fellow members than were distributions from
the churches. There was in the Roman cultural tradition also the idea
of disinterested benevolentia. But benefactions were made ob honorem to
enhance the prestige of the benefactor on the model of the relationship
between patron and client, not out of Pauline agapē: recall from Section 5
of the Prologue the disdain of the urbis moderator Lampadius for the
beneficiaries of his largesse. In Stoic philosophy, it is proper to offer help
to the suffering, but out of mercy (clementia), not pity (misericordia).
Distributions of goods or money were periodically made by the state,
Cultural selection and acquired behaviour 133
but by the criterion of rank, not need. ‘Alimentary’ schemes like Trajan’s,
which provided for the maintenance of children up to the age of fifteen or
sixteen, ‘did not constitute the one exception to the rule that the Roman
world had little compassion for the poor and that no ancient government
ever regarded them as their responsibility’ (Woolf 1990: 227). No other
agency provided orphanages or almshouses or homes for widows or hos-
pitals in the way that the Christian churches did.17 Convincing evidence
for the uniqueness of Christian charity comes not only from Christians
like Lactantius (in Book 6 of his Divine Institutes) but from Christianity’s
opponents, most notably from the letters of the ‘apostate’ emperor Julian
in which he points to the need for the priests and high priests whom he
created to match their Christian counterparts in benevolence to those in
need, including prisoners as well as the poor.
Whoever the recipients, however, Christian almsgiving takes us dir-
ectly back to Section 6 of the Prologue and Darwin’s contrast between
the ‘low’ motives of those who expect something in return and the
motives of the genuine altruists who exemplify the ‘noblest attribute of
man’. Christians who were generous towards non-Christians, although
they did not look to them for anything in return, might still have been
motivated by the anticipation of reward in the world to come. If they ser-
iously believed that by giving away their earthly possessions they would
secure posthumous bliss, rational self-interest would presumably dictate
that they did so. This is the cost-benefit calculation explicitly presented
by Clement of Alexandria in The Rich Man’s Salvation, where he offers as
a ‘beautiful deal’ (ō kalēs emporias) the purchase of eternal incorruption
with perishable cash. But there is no evidence that this is what was in fact
going on inside the converts’ heads. The expanding Christian commu-
nities were guided neither by the primitive communism depicted in the
Acts of the Apostles nor by the asceticism of the later monastic movement.
Renunciation on the model of the advice of Jesus to the rich young man
in the gospel story to give up all he had and ‘follow me’ was as rare as
self-chosen martyrdom. Cyprian of Carthage appears to have been one
of the few who gave away his personal fortune, but there are hints in the
sources that he either kept some of it back or later acquired other property
of his own. In any case, to buy favour from God on the model of a mar-
ket purchase or of the do ut des of pagan sacrifice would be the antithesis
17
There were infirmaries (valetudinaria). But they were not for the poor. Initially, they were for
soldiers, and then by extension for members of the imperial court or the households of the rich
(Hands 1968: 141).
134 The theory of cultural and social selection
of Christian agapē. The ‘prime almsgivers’ were the rich (Countryman
1980: 114). But regular donations on a scale compatible with the normal
conduct of daily life were the staple of Christian almsgiving. Much of
what each local community received was applied to the support of the
churches themselves, the maintenance of their own officials, and help to
Christians in need including the ransoming of captives. In Rome, accord-
ing to Eusebius (Ecclesiastical History 6.43.11), the Christian community
looked after 1,500 needy persons in addition to its own officials. But the
difference was that they combined the sanction of disapproval which sus-
tained conformity to the injunction to give with a norm of benevolence to
people to whom they had no other ties of any kind.
This, therefore, produced a textbook example of the free-rider problem.
Whatever the individual motives of more and less generous Christian
almsgivers, the Christian communities were bound to be vulnerable, as
the satirist Lucian remarked at the time (The Passing of Peregrinus 13), to
any clever charlatan (Lucian’s word is goēs) who took money off gullible
Christians and then moved on. Internally, the Christian communities
were so constituted that selective pressure favoured strong reciprocity. The
congregations of the house-churches were small enough for their mem-
bers to know one another personally, to monitor one another’s behav-
iour directly, to preserve recollections of non-cooperation, and to pool
information about the trustworthiness of different individuals. No doubt
they were exploited from time to time in the way that Lucian suggests.
But it was, as always, the trade-off that counted. If, for every stranger
who turned out to be a cheat, there was on average more than one who
joined the community of unconditionally benevolent almsgivers and
abided by its norms, the strategy would be, however slowly, diffused; and
if Christian parents brought up their children to be unconditionally ben-
evolent likewise, it would be reproduced in proportion to their fertility
(which some historians believe to have been consistently higher on aver-
age than that of pagans).
There is, moreover, an aspect to the unconditional benevolence of the
Christians to which the American sociologist of religion Rodney Stark has
drawn particular attention (1996: Ch. 4) and which must to some degree
have increased the number of converts: their willingness to nurse non-
Christians as well as Christians during the plagues by which the popula-
tion of the Roman Empire was periodically afflicted. Stark draws on the
estimate of present-day epidemiologists that in epidemics of the kind that,
so far as we can tell, these were, rudimentary care – water, a little food,
a blanket, and an attentive nurse – can reduce what would otherwise be
Cultural selection and acquired behaviour 135
the mortality rate by as much as two-thirds. The rhetorical account given
by Eusebius (Ecclesiastical History 7.22.7, 10) of pagans abandoning their
nearest and dearest to die in the streets while Christians nursed the sick
without regard to the risk to themselves cannot be taken at face value.
But it needs only to have been the case that Christians were more likely
than non-Christians to nurse the sick, and that they nursed at least some
non-Christians even if they gave preference to Christians, for it to fol-
low that significant numbers of non-Christians will have seen for them-
selves that a significant number of non-Christians nursed by Christians
were cured. The epidemic of the year 165 was succeeded by another in
189, which according to Dio Cassius (LXII.14.3–4) killed 2,000 people
in Rome in a single day, and then by another in 250. Successive genera-
tions of pagans will therefore have included a proportion, however small,
of survivors nursed by Christians of whom some may have remained
attached to the traditional religion in which they had been brought up
but some will have been converted to Christianity by the experience and
transmitted their accounts of that experience and their conversion to their
subsequent families and friends.
Given the irremediable lack of quantitive evidence, this is yet another
topic where the mathematical models available to present-day theorists of
cultural selection cannot be put to effective use. Twenty-first-century his-
torians of Christianity are no better placed than was Gibbon himself to
revise either upwards or downwards his guess that 5 per cent of the popu-
lation of the Roman Empire were Christians by the time of Constantine’s
conversion. But the just-so story outlined above does fit the evidence that
we have. At any level approaching Gibbon’s 5 per cent, the rate of growth
of a diminutive Jewish sect which barely survived the execution of its
founder can only, in the absence of imposed conversions, be explained by
the adaptedness of some distinctive package of mutant or recombinant
memes of its own; and of these, the most plausible candidate is the pecu-
liar norm of unconditional benevolence which was extended to strangers
in a way that no competing cult or sect attempted to do.
15
The subsequent success of Christian memes in an increasing number of
mutant forms is still, in part, a story of cultural transmission by phylo-
genetic descent and lateral diffusion which could in principle be mod-
elled in the same way as that of languages or artefact types. But it turns
also into a social-evolutionary story of geopolitical differentiation, forcible
136 The theory of cultural and social selection
conversion, and colonial or imperial imposition in which institutional
sanctions are as important as interpersonal imitation and learning. By the
time that Protestantism was approaching fixation among the populations
of Northwest Europe and North America, the beliefs and attitudes passed
down from parents and mentors to children in its name were, as in the
example cited in Section 9, as much to do with nationalism as with the-
ology, and the relation of conduct in this world to the hope of salvation
in the next was such as to lead Weber to link the teachings of Luther
and Calvin to the emergence of a distinctive behaviour-pattern among
Protestant capitalist entrepreneurs.
The ‘Weber thesis’ is, as I already remarked in Section 3, untenable
in its original formulation. But if we turn from Weber’s depiction of the
Protestant capitalists of seventeenth-century Europe to his depiction of
those he encountered in person on his visit to the United States in the
early twentieth century, there emerges a selectionist just-so story in which
there is a striking similarity between the memes in the heads of the latter
and those in the heads of the members of the early Christian communi-
ties. The selectionist hypothesis is that in small-town Protestant American
communities where memes constitutive of the norms of fair-dealing with
outsiders and restrained personal consumption had been transmitted
both vertically and laterally to successive incumbents of entrepreneur-
ial roles, the commercial practices of those entrepreneurs who adhered
to them enjoyed a competitive advantage within their local economic
environment. There is no suggestion in this that Protestants were in gen-
eral more likely to be capitalists in consequence of their religious beliefs
or that in other environments, in North America as elsewhere, these
norms might not be disadvantageous in competition with the ‘Pierpoint
Morgans, Rockefellers, Jay Goulds, etc.’ whom Weber characteristically
described in Nietzschean terms as ‘beyond Good and Evil’ (1922: 214).
The just-so story is one of meme-practice co-evolution in which a locally
optimal equilibrium was uninvadable by alternative strategies for as long
as certain distinctive environmental conditions obtained.
This story begins in the Puritan communities of seventeenth-century
New England. Their founder populations were very obviously not a rep-
resentative sample of the population from which they were drawn. As
MacCulloch puts it (2003: 538), ‘not all who had crowded the Atlantic
migration boats were pure in heart or sought godliness’. But the adult
males were predominantly small manufacturers, farmers, artisans, mer-
chants, and traders who knew each other personally, engaged in repeated
transactions with one another, and remembered each other’s strategies in
Cultural selection and acquired behaviour 137
past interactions. Like the early Christian communities in the towns of
the Roman Empire, they were vulnerable to free-riders and cheats who
could exploit them in one-off encounters and then move on. But the costs
imposed by the defectors could be off-set by the gains perceived by new
entrants who, if they conformed to the community’s norms, would receive
payoffs underwritten by strong reciprocators ready to deploy the sanctions
of ridicule, censure, and ostracism against fake-signalling hypocrites who
exploited the gullible while paying lip-service to the principle of ‘Honesty
is the best policy’ as conduct pleasing to God.
At the same time, vigilant monitoring by neighbours and fellow
churchgoers reinforced the norms of frugality, diligence, and unosten-
tatious personal consumption. In the words of the historian on whom
Weber principally relied, the New England Puritan, having ‘lost the cap-
acity for luxurious expenditure’, must, when he prospers, ‘keep his money
in a strongbox or else employ it in trade’ (Doyle 1887: 35). In this, the
New England capitalists were, as Weber was well aware, different not only
from the slave-owning gentry of the Southern States but from the big
Anglican merchants of New Hampshire and Maine. But their attitudes
and life-styles are recognizable to this day in those parts of the United
States where small-town Protestant businessmen are regular churchgoers,
are totally committed to the businesses they manage, have comfortable
but not ostentatious life-styles, and practice both honesty and thrift. It
is not that ascetic lifestyles are unique to Protestant Christianity. But,
as Weber emphasized, it tends elsewhere to take the form of withdrawal
from, and repudiation of, the economic institutions of the society within
which ascetic sub-cultures are formed and sustained. Whatever reserva-
tions are necessary about Weber’s thesis in its original formulation, he was
not mistaken in identifying what he called ‘inner-worldly asceticism’ as a
distinctive variant of what he called ‘religious rejections of the world’.
Unconditional cooperation was not, and never can be, an adaptive
strategy in large and fluid populations where fair-dealers ‘lost in a sea of
anonymous others’ (Macy 1991: 2, quoting James S. Coleman) will never
be able to build up networks of mutual trust and the payoffs to free-riders
and cheats are multiplied by the opportunities available to them in one-
off encounters and the difficulty for strong reciprocators to form stable
coalitions. Over large parts of North America during the nineteenth
century, as the frontier moved westwards, the economy expanded at an
accelerating rate, and the population was steadily augmented by succes-
sive waves of incomers, the ‘anonymous others’ came to include not only
the Pierpoint Morgans, Rockefellers, and Jay Goulds but the whole army
138 The theory of cultural and social selection
of fast-talking, get-rich-quick promoters, landboomers, carpet-baggers,
pedlars, quacks, guns-for-hire, cardsharpers, and showmen who feature
in the pages of Melville’s novel The Confidence Man (Tanner 2000: Ch. 5).
But what Weber observed at first hand during his visit to the United
States in 1904 was the survival and ongoing reproduction, in the com-
munities which he visited, of memes and practices still recognizable from
Puritan New England. The small and medium-sized towns where much
of the growth in domestic output was generated during the nineteenth
century were not replicas of those of the seventeenth century. But they
were sufficiently small, and their populations sufficiently stable, for local
manufacturers, traders, and bankers to integrate new entrants into their
existing networks of churches and other voluntary associations, to moni-
tor their commercial conduct, and to bring the sanction of public opin-
ion to bear on them if they failed to conform. The cost of punishment to
the punishers was cheap, but the cost to a fake-signalling manufacturer,
trader, or financier offering a reputation for trustworthiness18 was severe.
No doubt the occasional deviants cheated their customers, reneged on
their agreements, and absconded with other people’s money. But as long
as the community remained small and stable, such episodes would be
more likely to reinforce the vigilant monitoring of new entrants than to
encourage a runaway effect of frequency-dependent defection.
How far Puritan capitalists were more successful than their competi-
tors, and how far the practices of which they were the carriers were more
adaptive in consequence, cannot be measured with anything approach-
ing the requisite accuracy. But in a fast-growing market, a readiness to
deal fairly with outsiders even without the guarantee of reciprocity, if
coupled with retention of investible surpluses which had not been dis-
sipated in personal consumption, created opportunities for increasing
market share less readily available to competitors of whom neither was
true. Some such competitors no doubt prospered, and some honest and
frugal entrepreneurs failed. But on average, over time, and until econ-
omies of scale gave decisive advantage to the large corporations with their
multiple shareholders, salaried bureaucracies, and impersonal social rela-
tionships, the gains to the carriers of still recognizably Puritan memes
outweighed the losses. There is, moreover, a similar just-so story which
has been told for Scotland. Calvinist Protestantism did nothing initially
In a revealing anecdotal example, Weber witnessed in North Carolina the baptism of a new-
18
comer who wanted (as a relation of Weber’s explained to him) to open a bank in a nearby town
(1922: 210).
Cultural selection and acquired behaviour 139
to accelerate the diffusion and reproduction of capitalist practices. But
later, when the opportunities became available in a different economic
environment, the carriers of Puritan memes were those best fitted to
exploit them (Marshall 1981).
The example is an apposite one with which to conclude this Chapter
for three reasons. First, it illustrates the continuity which persists between
the insights of the founding fathers of twentieth-century sociology and
their reformulation within selectionist theory. Second, it illustrates how
homologous descent and convergent evolution can both be involved,
as well as lateral diffusion, in just-so stories of cultural selection which
extend across large distances in both time and space. But third, it illus-
trates the benefit to be gained from the reformulation in terms of meme-
practice co-evolution of the perennial question of the relation between
‘ideal’ and ‘material’ interests as the drivers of change. In Weber’s well-
known metaphor, ideas function like switchmen directing the course of
institutional change down one rather than another railway-line. But the
metaphor is misleading twice over. There are no tracks laid down for evo-
lution to follow, and there are no termini at the end of the different lines.
The interaction between mutant or recombinant memes and mutant or
recombinant practices is continuous and open-ended, and its outcomes
are explicable only by reference to selective pressures whose effects can-
not be identified except in hindsight. In any human population where
the transition from culture to society has been made, ‘meme-practice co-
evolution’ is the right answer to the question ‘just what is going on here?’
as it bears on the fundamental questions of comparative sociology as for-
mulated in other terms by Spencer, Marx, Weber, and Durkheim.
Ch apter 4
Rol e s, s y s tac t s, s o c i e t i e s, e m pi r e s
1
It should by now be abundantly clear just how different are the mechanisms
of heritable variation and competitive selection of information affecting
phenotypic behaviour when social rather than natural or cultural selection
is the driving evolutionary force. If the archetypal just-so story of cultural
selection is the visionary preacher from whom an expanding group of dis-
ciples acquires by imitation and learning a novel set of memes and a conse-
quential life-style whose ongoing reproduction the environment turns out
to favour, the archetypal just-so story of social selection is the innovative
entrepreneur who in the environment of a putting-out economy – with or
without the memes of a ‘Protestant Ethic’ inside his head – imposes the
novel but highly adaptive practice of wage-labour on an expanding work-
force and thereby forces his competitors either to exit the market or to
employ wage-workers themselves. There is of course no single practice ‘for’
a distinctive mode of production (or persuasion, or coercion) any more
than there is a single meme ‘for’ a distinctive cultural behaviour-pattern or
a single gene ‘for’ a distinctive personality trait. But at the social, as at the
cultural and biological, level there are somewhere in all the noise and clut-
ter the critical mutations and combinations in the information affecting
behaviour at population level which have to be identified and traced.
Like cultural selection, social selection is neutral between intentional
and unintentional design. The most unequivocal examples of designed
adaptedness are to be found at times of constitutional choice when the
makers of the institutions into whose constitutive roles future incumbents
are to enter lay down the practices which will define them: elections to
political office on a specified franchise, trials for offences defined as crim-
inal before a judge and jury, compulsory recruitment of adult males into
140
Social selection and imposed behaviour 141
military service in time of war, and so on. But even there, the constant
renegotiation of practices, like the constant reinterpretation of memes in
the ongoing process of cultural selection, introduces an element of evo-
lutionary adaptedness and of scope for the diffusion and reproduction of
mutations of which the original designers had neither knowledge nor fore-
sight. In the primordial transition from culture to society, random drift
and individual trial-and-error may on occasion have been as significant
as social learning. But once the practices are there defining the natives’
institutional roles, the interacting role-dyads have formed, and the terms
for the complementary roles have entered the natives’ vocabulary, it is the
selective pressure of the environment on their phenotypic effects which
determines which practices win and which lose at population level down
the subsequent generations of carriers.
It is easiest to see just what is going on when a role is deliberately
designed by a person to whose own role there attaches the power to
impose directly a mutation of existing practices, as in Peter the Great’s
‘Table of Ranks’ of 1722, which laid down that merit rather than birth
was to be the basis of appointment to Russian governmental offices, or
Solon’s division of the citizens of Athens into four property classes in 594
BCE. But proclamations, as I have remarked already, are seldom to be
taken at face value, and rules are often flouted or ignored. There are also
many borderline cases, not only during periods of transition but when the
economic, ideological, or political power attaching to a recognized role
is so tenuous as to be hardly discernible at all. The Christian bishops of
Section 15 of Chapter 3 are as clear an example of an evolutionary transi-
tion from interpersonal influence to institutional power as are the emer-
gent roles cited from the ethnographic record in Section 8 of Chapter 1.
But sometimes the transition is impossible to make, even where the
incumbent of the role is a person of exceptional ability and determination.
Weber’s classic analysis of ‘charismatic’ legitimacy and its inherent imper-
manence shows how the leader whose command of a following rests on
personal prowess which fuses the source of authority with its agency gen-
erates the expectation of a continuous run of quasi-miraculous successes:
‘routinization’ (Veralltäglichung) can be achieved only if the ‘gift of grace’
is somehow transmitted from the person to a role which others can sub-
sequently occupy (Amtscharisma). An almost ideal-typical example from
the ethnographic record of a role which, without a change in the envir-
onment, cannot become fully institutionalized is that of a sheikh among
the Rwala Beduin: since he can bring the means of coercion to bear only
by the deployment of his personal wealth and his personal standing as
142 The theory of cultural and social selection
a ‘good man’ (rajul tayyib), it can be said that this is a society in which
‘political power does not exist’ (Lankaster 1981: 96). Indeed, something of
the same is true where, as over much of pre-colonial Africa, land is not a
scarce commodity and leaders have to ‘attract as well as restrain’ (Goody
1971: 30). The difference is that African chiefs, although their roles did
not (except in Ethiopia, which had the plough) evolve into lordship over
a peasantry tied to the soil, could nevertheless keep close coercive control
over their subordinate soldiers, women, and slaves and were maintained
in their roles by the render of food and other material goods produced
by the labour of subordinates outside of their own families and personal
followings.
The textbook example of a borderline role is that of the ‘big-men’ who
briefly appeared in Section 9 of Chapter 1. Big-men are particularly well
documented for Western Melanesia, where the standard comparison is
with the Polynesian chiefs at the head of their conical clans who com-
mand permanent bodies of professional soldiers, ceremonial attendants,
and supervisors of stored resources. But analogous practices defining simi-
lar roles are found not only among nomadic pastoralists but, for example,
in Iceland where the gođi (who is literally a ‘priest’) depends for his pos-
ition on retaining personal popularity among followers willing to fight
for him (Foote and Wilson 1970: 135), or in early medieval Brittany where
the power of local ‘machtierns’ to tax, hold courts, and take fines depends
on the goodwill of their local plebs (Davies 1987: 139). There are numerous
examples in the ethnographic record of leaders of factions whose posi-
tions derive from their personal qualities of intelligence, fighting ability,
generosity, craft skill, and knowledge of local customs. The essence of all
such roles is that, as the anthropologist Marshall Sahlins has put it, their
incumbents must ‘personally construct their power over others’ in con-
trast to roles whose incumbents ‘come to power’ (1974: 139).1 Although
Melanesian big-men are quite often succeeded by their sons, and are occa-
sionally able to designate a preferred successor, the successors have then to
prove themselves in their turn – which is made still more difficult in areas
of the New Guinea Highlands where there is a high rate of migration.
(In Brittany, by contrast, there were machtiern families extending across
several plebes.)
An archetypal just-so story of how a self-aggrandizing big-man can
bring off the transition is the story which Herodotus (I.96–100) tells of
A revealing symptom of the difference is that whereas a ‘big-man’ feasts his followers, an Anglo-
1
2
Institutional roles are, however, not only parts to be acted out – of which
the same individual can play several – but vectors within a bounded
three-dimensional social space. The inherent inequality of power within
such familiar dyads as capitalist and wage-worker, bishop and parish-
ioner, or warlord and conscript joins the dominant and subordinate roles
respectively into systacts which rise and fall in relation to one another at
the same time that individuals move up or down between the roles them-
selves. Just-so stories of, for example, the ‘decline of the aristocracy’ and
‘rise of the bourgeoisie’ are group-selectionist stories to the extent that
the diminished cohesion of the one systact and the enhanced cohesion
of the other affects the reproductive success of the associated mutations
in the practices defining ‘aristocratic’ and ‘bourgeois’ roles. In the case of
English society in its ‘early modern’ period, what was going on was a dis-
placement of lord/retainer role-dyads by capitalist/tenant (and therewith
144 The theory of cultural and social selection
tenant/wage-worker) role-dyads – an evolution to which Adam Smith
attached particular importance in the contrast which he drew between
England and Scotland. And later, when wage-workers combined to form
trade unions, the most successful were those which contained the highest
proportion of strong reciprocators ready to punish not only free-riders but
fellow workers unwilling to punish them. (The employers on their side
faced the same problem when members of trade associations reneged on
an agreement to hold out against a strike called by a trade union.)
But inter-systactic struggles resulting from institutional inequalities of
power are not always drivers of social evolution. A society can be rid-
dled with endemic economic, ideological, and political conflict without
its modes of production, persuasion, or coercion being changed on that
account. Mamluk Egypt is one example, and nineteenth-century Haiti
another. The prerequisite for change is that mutant practices must appear,
no matter from where, which either modify existing roles or create new
ones. Thereafter, their reproduction and diffusion can be a matter of either
individual or group selection. If trade unions competing for members at
the same time as they negotiate with their employers for better terms are
a textbook example of group selection, a textbook example of individual
selection is given by Marc Bloch when in his discussion of vassalage he
speaks of ‘the process of evolution which transformed what was formerly
a personal grant into a patrimonial property and an article of commerce’
(1961: 212). The mutation of a practice which had originally bound a sin-
gle vassal to a single lord into one which permitted a plurality of poten-
tially conflicting ties created by homage to different lords in exchange for
different fiefs was not, as Bloch pointed out, logically necessary. In Japan,
although vassal fiefs were both hereditary and alienable, multiple fealty
was known only as an occasional abuse. But the politically fragmented
environment of post-Carolingian Europe favoured its spread from indi-
vidual to individual to the point that the jurists were soon dealing as a
matter of course with the dilemmas and contradictions which resulted
from it.
When there does emerge a novel role defined by mutant or recombin-
ant practices, it can sometimes amount to a modal change by itself, as
when Augustus turned Rome from an oligarchic republic into an auto-
cratic monarchy. But more often, the practices defining the novel role
have to be diffused across the relevant population: a single wage-paying
entrepreneur doesn’t create a capitalist mode of production any more than
a single consecrated priest creates an ecclesiastical mode of persuasion or
a single elected office-holder a democratic mode of coercion. Sometimes,
Social selection and imposed behaviour 145
the practices are laterally transmitted by imitation or learning: a land-
lord switches from demesne cultivation to money rent because of seeing
that other successful landlords have done so, or a congregation decides to
institutionalize the hierarchy of a rival church, or a parliamentary govern-
ment models its electoral procedures on those of another. In thirteenth-
century Italy, the diffusion of the role of professional podestà is a textbook
example of ‘institutional imitation’ (Epstein 1999: 18). But sometimes, it
is not so much imitation as convergent evolution in response to inescap-
able selective pressure: if outdated military practices are not replaced in
time, the army will go down to defeat and the country be conquered.
New systacts can also be formed simply by a sufficient increase in the rate
of individual social mobility. Although a handful of manumissions leaves
the structure of the society unchanged, a steady flow out of slavery cre-
ates a systact of freedmen and freedwomen, who, even if they never act
jointly in pursuit of a perceived common interest, continue to be collect-
ively ranked in social space below those born free.
3
After the transition from culture to society, societies of increasingly diver-
gent kinds evolve out of the heritable variation and competitive selection
of practices in the same way that after the transition from nature to cul-
ture, cultures of increasingly divergent kinds evolve out of the heritable
variation and competitive selection of memes. But at the same time, these
societies – unless, like Tasmania, they have for many centuries no contacts
of any kind with any other – are likely to be in competition with other
societies for territory and economic, ideological, and political power. The
imposition of the practices of a stronger society on the population of a
weaker one does not have to be by force, although it often is. But once
the rulers of a central, metropolitan society are able to exercise effect-
ive domination over one or more peripheral societies, we are in a world
not only of societies but of empires. Empires are at the same time more
than big societies and less than leagues of allied societies of which one is
more powerful than the rest. Sometimes, forcible annexation results in
the absorption of the members of the weaker, peripheral society as addi-
tions to the central population of the stronger. The English absorption of
the Scots enabled an ambitious Scot to occupy one of the topmost roles
in what was by then the ‘British’ state, just as the Roman absorption of
the Latins enabled an ambitious inquilinus to become a Roman consul.
But in the comparative sociology of empire, the interesting cases are those
146 The theory of cultural and social selection
where the initial annexation is neither so complete that the distinction
between big society and empire is eliminated nor so short-lived, as in the
cases of Lugal-zagesi or Charlemagne or Asoka,2 that there is no time for
what might have been adaptive mutations of imperial memes and prac-
tices to be reproduced. Empires call for explanation as such where the
central society dominates the peripheral societies for an extended period
without either wholly absorbing them or wholly disengaging from them.
The peripheries may be as remote as the Philippines from Spain or as
close as Tibet to China. The centre may be represented in the periph-
eries by a handful of soldiers, traders, and missionaries, or by perman-
ent garrisons, large commercial enterprises, and an implanted network of
temples or churches or schools. The peripheries may be colonies or vassals
or tributaries or satellites or clients of the centre. But the most apt single
word is ‘protectorate’ as defined by Lord Halsbury in 1890 (Burroughs
1999: 194): ‘a convenient state between annexation and mere alliance’. The
question then is: can the relationship between the central and peripheral
roles reach a local fitness peak where it is uninvadable by practices which
will make it impossible to sustain?
The answer is no. Societies can survive revolutionary class conflict,
subversion of traditional hierarchies of status, and military defeat or civil
war. But empires can’t. No empire lasts forever, or anything like it. The
difficulties which they face are well recognized by their rulers: the costs
are so high, the frontiers so exposed, the revenues so hard to collect, the
resentments so entrenched, the administrators so inefficient, the monopoly
of military technology so impermanent, the trading routes so riddled with
extortion, bribery,3 contraband, and piracy, the rebellions so frequent, and
2
There is, however, a distinction to be drawn between the collapse of an empire on the death of
its founder and the defeat of the founder in war by a rival imperialist: in Lugal-zagesi’s case,
no sooner has he emerged out of the Sumerian transition from culture to society and the world
of assemblies and temples and landowners and kings to claim rule ‘from the Lower Sea along
the Tigris and Euphrates rivers to the Upper Sea’ (Oates 1979: 28) than he finds himself put on
display in a neck-stock at the gates of Ekur by Sargon of Agade. In Asoka’s case, although ‘frag-
mentation, local reassertion of independence, and interregional rivalries and invasions swiftly
followed the demise of the God-King-Father’, ‘the first great dynasty of Indian history continued
to rule over Magadha at least until 184 B.C.’ (Wolpert 1977: 69).
3
Some imperial rulers, and some historians, are tempted to blame ‘corruption’ for the difficulty of
maintaining central domination over the peripheries. But practices and roles whereby favours are
exchanged in apparent contravention of official rules can be stable in the precise sense in which
stability is defined in evolutionary game theory, and the pay-offs are such that the ‘corrupt’ sys-
tem is uninvadable by any mutant strategy. As Elliott says of Spanish America (2006: 229), ‘In
practice, the spread of systematized corruption endowed the imperial structure with a flexibility
that its rigid framework appeared to belie.’ Cf. Veyne (1981) on corruption ‘au service de l’état’ in
the later Roman Empire.
Social selection and imposed behaviour 147
so on. In view of them, it may seem remarkable that any empire should
manage to last as long as some of them do.4 The imperial engineer who is
the notional colleague of the religious engineer of Chapter 3 can be asked
to design the roles whose defining practices will have the best prospect of
continuing reproduction, but they will be adaptive only in an environment
where there is little or no selective pressure driving the centre and the per
ipheries in the direction of either absorption or disengagement; and such
pressure is always there. In the heyday of the most famous empires, the
prospect of disengagement may have seemed as remote as the prospect of
absorption. A contemporary observer of the British Empire at the time of
Victoria’s Jubilee in 1895 would have been as unlikely as an observer of the
Roman Empire at the time of Diocletian’s abdication in 305 to question its
success. But the history of the British Empire under Victoria is a history of
crises and improvisations which followed one after another and repeatedly
failed to produce a lasting solution to the problems which had given rise to
them, just as Diocletian’s reforms were a response to a period of breakdown
and disorder during which the Roman Empire had virtually ceased to exist.
After Victoria, it turned out to be a short step from Halsbury’s ‘convenient
state’ to colonial self-government and then to imperial disengagement, and
after Diocletian, it turned out to be a short step from the defeat and death
of the emperor Valens at Adrianople to the settlement of barbarians within
the frontiers, the sackings of Rome, the seizure of Carthage, and the by
then irrelevant deposition of Romulus Augustulus in 476. For a time, the
domination of the centre over the peripheries may look not only politi-
cally but economically and ideologically secure: the practices defining the
roles of viceroys, proconsuls, corregidors, harmosts, and commissioners
are co-adaptive both with those defining the roles of contractors, traders,
entrepreneurs, financiers, and planters and with those defining the roles of
missionaries, schoolteachers, propagandists, and priests. But they do not
do away with the selective pressures driving the system in the direction of
either absorption or disengagement.
Chance, as always, plays its part, including what might be called ‘mineral luck’. How could the
4
Athenian Empire have been sustained if the silver mines of Attica had not, by geological accident,
been spared the need for advanced pumping technology to combat the risk of flooding (Finley
1965: 30), or the British Empire without abundant cheap coal located near consumer demand and
artisan skill (Pomeranz 2000: 66)? It is true that the ‘Industrial Revolution’, which was not, soci-
ologically speaking, a revolution and only partly industrial, depended on the technological inno-
vations which allowed Britain’s mineral wealth to be exploited; but escape from the Malthusian
constraint imposed, as Ricardo believed, by the laws of nature (Wrigley 2003: 166) was only pos-
sible because of the quantity and accessibility of a mineral-based, as opposed to organic, energy
source.
148 The theory of cultural and social selection
Every empire, from those of the Egyptians and Hittites onwards, has
its own evolutionary trajectory. But if there is one common problem
which confronts all imperialists unable or unwilling either to absorb or
to disengage from the populations of the peripheral societies subordi-
nated to them, it is the problem posed by the practices which define
intermediary roles. Whether the intermediaries are agents of and from
the imperial government or local elites exercising power over the periph-
eries on terms which the imperial government has laid down, the impe-
rial government is faced with the same dilemma: the less the power that
attaches to the intermediaries’ roles the greater the risk that the peripher-
ies will escape from the centre’s control, while the greater the power that
attaches to the intermediaries’ roles the greater the risk that they will
themselves escape from the centre’s control and take the population of
the peripheral society with them. The intermediate roles may constitute a
distinctive systact located in one or more of the three dimensions of social
space. Tax-farmers are a straightforward example of an economic class,
but scribal gentries can equally constitute a distinctive systact within the
mode of persuasion and local warlords a distinctive systact within the
mode of coercion. Sometimes, as in the Ottoman and Mughal empires,
landholding by dispensation from the centre was combined with del-
egated local control of the means of coercion. Sometimes, as in Spanish
America, churches or monastic orders were substantial landowners, or
lenders to local landowners who might find it as difficult to clear those
debts as subordinate peons to clear their debts to the landowners. But
in no combination of practices and the roles defined by them was the
problem of the intermediaries ever permanently solved. Time and again,
practices which were adaptive in the short term turned out to be mala-
daptive in the long. In the mode of production, tax-farming can seem a
promising device for extracting resources from the periphery only for it
to aggravate both the excesses of the collectors and the resistance of the
payers. In the mode of persuasion, the founding of mission-stations and
schools can seem a promising device for educating the children of the
population of the periphery as the imperial government requires only for
the children to use their education to foment ideological opposition to
the centre. In the mode of coercion, recruitment of young adult males
from the periphery to serve in the imperial army can seem a promis-
ing device for reducing the likelihood of local rebellions and enhancing
the capacity to put them down only for the native troops to mutiny.
Although no lawlike generalization can be framed about such outcomes,
it is more than a truism to say that empires are harder to keep under
Social selection and imposed behaviour 149
long-term control without absorption or disengagement than they were
to acquire in the first place.
Cultural selection of beliefs and attitudes can sometimes enhance both
the acquiescence of the populations of the peripheries in their subordin-
ation to the rulers of the centre and the effort made by rulers and their
advisers to design for the purpose monumental architecture and the rit-
uals and ceremonies to be enacted at or within it. But the cultural selec-
tion of memes which cause their carriers to behave as they do through
genuine imitation or learning has as always to be clearly distinguished
from the social selection of practices which impose on the incumbents
of subordinate roles, whatever the memes inside their heads, behaviour
which conforms to the imperial rulers’ ideology. The impression made
on the emissaries from the peripheries of the Assyrian empire who were
led through the courtyards and colonnades of Assurbanipal’s enormous
palace into his presence may have been as effective as that made on the
dignitaries ushered into the presence of Louis XIV at Versailles. But
rulers of empires who are conscious of the institutional prestige which
attaches to their roles are sometimes too ready to assume that there are
therefore memes in the heads of the natives at the periphery which may
not be there at all.5 Among the peripheral elites, the attraction of the cul-
ture of the imperial centre is as evident among the belted and bejewelled
chiefs of the barbarian tribes along the Rhine and Danube in the time
of Diocletian as it is among the billiards- and polo-playing rulers of the
Indian princely states in the time of Victoria. But the resentments of the
liturgists of Roman Egypt which are eloquently documented in the sur-
viving papyri are as readily explicable as those of the representatives from
every province in British India who attended the first meeting of the
Indian National Congress in Bombay in December of 1885. How can a
discourse be engineered which will implant in the heads of people whose
labour has been ‘connected, subjugated, and made tributary’ (Bayly 2004:
64) memes which will permanently reconcile them to their subjected and
tributary roles? Some forms of acquired behaviour may be diffused from
the centre by lateral transmission throughout the entire population of the
5
The parallels between the rhetoric of imperial Britain and imperial Rome are almost too easy
to draw: ‘land of hope and glory’ matches ‘rerum pulcherrima Roma’; ‘the empire on which the
sun never sets’ matches ‘quae caret ora cruore nostro?’; the proconsular paternalism of Cromer in
Egypt matches that of Cicero in Cilicia; and the effusions of the late-Victorian propagandists for
pax Britannica are, if anything, outdone by those of Aelius Aristides for the pax Romana. But the
victims of the British soldiery’s dum-dum bullets could endorse as readily as the victims of the
Roman soldiery’s broadswords the much-quoted dictum that ‘where the Romans create a desert
(solitudinem) they call it peace’ (Tacitus, Agricola 39).
150 The theory of cultural and social selection
peripheries. Indigenous traditions can be displaced through frequency
dependence and imitation of prestigious role-models to the point that,
for example, the children of poor Jamaicans take up the game of cricket
and their parents give them ‘Christian’ names borrowed from those of
the British aristocracy. But that is no solution to the dilemma which the
imperial rulers face when they seek to persuade the population of the
peripheries of their benevolent paternalism. Paternalism implies that
the docile child will grow up to be the equal of the putative parent. But in
that event, the children become entitled to be treated as adults, whether
as fellow members of a common culture or as independents who learn or
imitate from the imperial culture as much or as little as they choose. No
more in cultural relationships where behaviour is acquired rather than
imposed than in social relationships of economic, ideological, and polit-
ical domination and subordination is it possible to preserve a stable inter-
mediate state between absorption and disengagement for more than a
limited period.
There is, moreover, a further reason for the inherent impermanence
of empires. Natural, as opposed to either cultural or social, selection
may seem relevant to the study of empires only to the extent that it has
endowed the human species with the innate disposition for in-group col-
laboration and out-group hostility which gives rise to them in the first
place. Recall, however, the general problem posed by populations larger
than can be held together by the combination of Hamilton’s Rule with
reciprocal altruism. Cultural selection helps to resolve it by altruistic pun-
ishment, strong reciprocity, and information networks through which
a reputation for trustworthiness can be monitored and assessed. Social
selection helps to resolve it by economic, ideological, and political prac-
tices which extend the reach of the power attaching to dominant roles:
regulated markets link progressively larger numbers of buyers and sellers
in more extensive economic relationships; churches and schools draw pro-
gressively larger groups of disciples into relationships of ideological sub-
ordination; and the negative allometry of elites monopolizing the means
of coercion enables 10,000 soldiers to control ten million people more
easily than one soldier can control a hundred. But they will not hold the
members of a subordinated peripheral society under the indefinite control
of a dominant central one unless absorption, whether forced or volun-
tary, turns the empire into a big society; and even big societies, like both
Russia and China at certain times in their respective histories and the
United States of America in the mid-nineteenth century, may be at risk of
fragmentation in much the same way that empires are.
Social selection and imposed behaviour 151
Mode s of produc t ion, pe r s ua s ion, a n d c oe rc ion
4
One consequence of adopting a selectionist approach to the comparison
of different societies and their constitutive economic, ideological, and
political roles is that the inadequacy of the standard sociological vocabu-
lary – feudalism, capitalism, absolutism, totalitarianism, socialism, dem-
ocracy, oligarchy, theocracy, plutocracy, and so on – becomes very quickly
apparent. (Worst of all is the entrenched metaphorical dichotomy of
‘Left’ and ‘Right’, which has done as much to retard sociological explan-
ation as to encourage tendentious political rhetoric.) Consider ‘feudalism’.
Historians have long inveighed against the ‘tyranny’ (Brown 1974) of the
concept and the contradictions and confusions which surround it. Yet
they have been consistently reluctant to discard it. Although they can,
as Bloch himself did (1961: 446), put together a list of ‘fundamental fea-
tures’ common to European and also Japanese ’feudalism’, it is equally
easy to put together a list of ‘fundamental’ differences, including for a
start the difference that in Japan the practice of vassalage defined a chain
of relationships between dominant and subordinate roles which culmi-
nated in the role of the shogun, not the emperor (tennō). If ‘feudal’ is
defined in terms not of a fusion of benefice and vassalage or conditional
military tenure or the tying of peasants to the land but simply of decen-
tralization of the economic, ideological, and political power attaching to
the society’s dominant roles, then societies as far apart as Japan, China,
Iran, Ethiopia, Poland, and Mexico at certain periods of their histories
can all be classified as ‘feudal’. But this explains nothing about how their
decentralized economic, ideological, and political institutions evolved as
they did until the different practices defining the different decentralized
roles have been identified and their adaptedness linked to the critical fea-
tures of their local environments. In the mode of production, do the local
power-holders extract economic surpluses through rent, tribute, labour
services, or taxation? In the mode of persuasion, does their status derive
from birth, life-style, or access to sacred symbols and rituals? In the mode
of coercion, are their soldiers and administrators conscripts, or clients,
or salary-earners, or arms-bearing landholders? Does a society cease to
be ‘feudal’ if the local power-holders are integrated into an international
mercantile and monetary system, or if they form a closed endogamous
caste, or if they are liable to serve by royal command in a national army?
Are societies of nomadic pastoralists ‘feudal’?
152 The theory of cultural and social selection
These rhetorical questions do not lead directly to the right just-so story
with which to explain the part-similar, part-different evolutionary trajec-
tories of the ‘feudal’ societies of post-Roman Europe. But notice what that
much-told story is not a just-so story of. The researches of the historians
of the post-Roman world who have exploited to the full the documentary,
archaeological, numismatic, epigraphic, and papyrological evidence have
left standing little or nothing of either Marx’s just-so story of a progres-
sion into a feudal out of a slave mode of production or Weber’s just-so
story of a regression from a nascent capitalism stifled by bureaucracy to
a decommercialized feudal economy.6 It is a story to which the practices
defining control of land and the relations between its controllers and the
incumbents of the subordinate roles who worked it is critical, as it was
in theirs. But it is only in part a story of the evolution of ‘serfdom’ out of
slavery. It is more a story of a diverse renegotiation in shifting local envir-
onments of practices defining relative degrees of institutional inequality
in the three dimensions of social space. Slavery, moreover, continues to
be part of it for many centuries longer, and the concept of slavery, unlike
the concept of ‘feudalism’, remains central to it as the limiting case of
total subordination. It provides both the historical and the analytical con-
trast by reference to which the powerlessness of the coloni, villani, col-
liberti, prebendarii, aldi, and originarii of the documentary sources for
‘feudal’ Europe (or their counterparts elsewhere) can be measured, and
the difference made by the mutations in the practices defining their roles
assessed.
5
There is always scope for argument about how exactly ‘slavery’ is to be
distinguished from all other forms of what anthropologists sometimes
call ‘rights-in-persons’ (Kopytoff and Miers 1977: 7) such as pawnship or
purchase of adoptive kin. But where there is a set of practices defining
a role to which there attaches no institutionally sanctioned countervail-
ing power in any of the three dimensions of social space, any word for
it in the vernacular terminology can be translated salva veritate into the
English word ‘slave’. Comparative sociology then reveals a complex of
6
A striking example is the way in which the papyrological evidence has transformed historians’
understanding of the sociology of Roman and Byzantine Egypt by revealing the diversity and
flexibility of the practices which defined the relationships between landowners on the one side
and slaves, tied tenants, sharecroppers, casual wage-labourers, lessees and sub-lessees, and full-
time resident employees on the other (Banaji 2007).
Social selection and imposed behaviour 153
variable gradations between the total subordination of the slaves and the
practices which define the roles of those above them. The relatively freer
are not necessarily better off in their own terms than some slaves. But
slaves cannot negotiate the practices defining their role except in purely
personal terms (foot-dragging, insubordination, malingering, or alterna-
tively ingratiation, manipulation, and seemingly willing compliance). If
the institutional subordination is total, it does not become less so if it
is represented in a language of fictive kinship or lawful punishment, or
if slaves are as kindly treated as Cicero treated his Tiro, or if they are
granted privileges or perquisites by their masters,7 or even if they exercise
delegated powers as extensive as those which astonished Western observers
of the Ottoman empire. Slavery is found in the societies of pre-colonial
sub-Saharan Africa no less than in Pharaonic Egypt, Rome, Anglo-Saxon
England, Muscovy, China, Korea, Zanzibar, Hispanic and Portuguese
America, the Northwest Pacific Coast, India, the Caribbean islands, the
Southern United States, Nazi Germany, and the Soviet Union.8 So wide-
spread is it in the historical and ethnographic record that its emergence,
or re-emergence, might be said to pose less of a problem for the theory
of social selection than its extinction. Its convergent evolution is – or
should be – no more surprising to comparative sociologists than that of
the feud.
Slavery is not by any means a social universal, and where found it always
exists alongside other economic, ideological, and political relationships of
domination and subordination. But there are many different environments
in which reverse engineering can show it to have been a locally optimal
design. The origin of the initial mutation is, as always, of limited relevance
to the explanation of its adaptedness. Somebody, somewhere, must have
7
Including slaves of their own: in one inscription well known to historians of Rome (Corpus
Inscriptionum Latinorum 6.5197), the tomb of a servile cashier in a provincial treasury in the reign
of Tiberius carries a dedication by no less than sixteen under-slaves who were with him when
he died. But slaves are no less slaves when they have slaves of their own than tenants are tenants
when they have sub-tenants and wage-earners are wage-earners when they also pay other wage-
earners for their labour.
8
In the Soviet Union, the inmates of the gulag incarcerated under indefinitely renewable sentences
were not traded as commodities, since in a socialist mode of production they were state not pri-
vate property, and they were freed by discharge rather than manumission. But as well as being
deprived of any share in the product of their labour and subject to physical coercion at the hands
of their guards they were deprived of their ideological status as Soviet citizens even if they were
members of the Communist Party, and their families could be deliberately broken up (Sawyer
1986: 199). Similarly, the forcibly indentured peasant labourers of Kampuchea under the regime
of ‘Pol Pot’ (Kiernan 1996: Chs. 5 and 6) were not only under the total coercive control of the
Communist Party ‘centre’ but ideologically stigmatized by a mode of persuasion which categori-
cally excluded any member of the population defined as ‘not true Khmer’.
154 The theory of cultural and social selection
been the first person to enslave enemies captured in battle rather than either
killing or ransoming them; somebody, somewhere, must have been the first
person to sell a child into slavery to avoid starvation in a time of famine;9
somebody, somewhere, must have been the first person to organize a raid-
ing party for the express purpose of kidnapping prospective slaves from
among the members of a neighbouring population; somebody, somewhere,
must have been the first person to fall into slavery by losing a wager staked
against their own person – an unusual route of downward social mobil-
ity, but one which is documented for Thailand (Feeny 1993: 97) as well as
the Kwakiutl (Macleod 1925). But where slaves are a desired commodity,
whether for their productive capacity or because of their value as items of
conspicuous consumption, and where in addition they are drawn from an
ideologically disprivileged status-group and held under close coercive con-
trol, the combination of practices by which the role is defined has a high
probability of continuing reproduction for as long as the servile population
can be replenished by purchase, capture, or breeding. Nor, given replen-
ishment, is that probability diminished in societies where the possibility of
manumission is as far from neglible as in Rome or, later, Brazil.10
At the same time – and again, as always – chance has much to do
with whether these conditions are met. In the evolution of large-scale
plantation slavery in the Caribbean and the Southern United States, the
cultivation of not only sugar but cotton and tobacco might seem over-
whelmingly favourable to the exploitation of the labour of slaves working
Not that children might not be sold into slavery for other motives than allowing their ‘heads’,
9
as an entry in the Durham Liber Vitae puts it, to be ‘taken for food in the evil days’ (Whitelock
1979: 610). In the Mamluk system, where ‘there is ample evidence for the cooperation of the
rulers and the chiefs of the nomads in the slave trade and also for the readiness of the nomads
themselves to sell their relatives’ (Ayalon 1975: 56 n.3), parents could see opportunities of a better
life for their children who might rise to positions of secondary power even though remaining at
the ruler’s disposal and at risk of summary punishment or death.
10
Despite the occasional availability of numerical evidence (e.g. Libby and Paria 2000, where the
rate in one Brazilian parish is around 1% per annum), manumission rates are notoriously difficult
to calculate. Some reported grounds for it can only have applied to an insignificant number
of cases. How often was a Visigothic slave ever freed for informing on a counterfeiter (Claude
1980: 164), or a Roman ancilla freed because she was sold with a condition against prostitution
which the buyer subsequently broke (Buckland 1908: 550)? But there are obvious advantages to
owners of slaves with marketable skills or attributes (including suitability for prostitution) in the
practice of allowing them to buy their freedom in instalments, particularly if, as ‘almost univer-
sally’ (Patterson 1982: 240), a condition of manumission is continued performance of services
to the former owner (or his or her heirs). Cicero, in one of his speeches (Philippics VIII.x.32),
implies in passing that hard-working ( frugi et diligentes) slaves could anticipate freedom within
six years. But they had to be in a position to show that they were worth freeing: contrast the fate
of Seneca’s doorkeeper left behind in the country (Epistles 12) with that of Lucilius Voltacilius
Pilutus, a city doorkeeper freed for his display of intelligence and interest in literature (Suetonius,
Grammarians 27).
Social selection and imposed behaviour 155
in gangs under the coercive supervision of overseers. But the right just-so
story is a different one. Far from being the first stage in an evolution out
of a domestic mode of production in the direction of a tributary and then
a capitalist mode, Caribbean and North American slavery was an unfore-
seen consequence of the maladaptiveness of a practice which had initially
appeared to be spreading towards fixation: indenture. Indenture, more-
over, was a mutation not, as might be supposed, of the practice of appren-
ticeship, but of the practice of service in husbandry (Galenson 1981: 6),
the difference being that the indenture was negotiable and the creditor
free to sell at any time before the servant’s debt had been extinguished
by the amount of work performed. Able-bodied young men dissatisfied
with their prospects in their society of origin could negotiate their pas-
sage to a society which offered more attractive opportunities by mortga-
ging their labour for a limited period, and employers chronically in need
of labour could hire it from a distance with the option of onward sale if
the worker failed to perform. Initially, both black and white labourers on
the tobacco plantations had laboured according to ‘customary English
practices’ (Berlin 1998: 32) which descended from the Elizabethan Statute
of Artificers. But indenture was a maladaptive practice to the extent that
employers had a strong incentive to exploit the indentured servants and
the indentured servants a strong incentive to break free in an environ-
ment where land was plentiful and labour scarce. It then so happened
that the supply of recruits ran low at a time when the supply of Africans
seized inland by local rulers for sale to slave-traders on the coast was
enough to keep down their price, and an increase in the ratio of slave to
indentured labour did not increase the marginal cost of labour in the way
that attracting more indentured servants would do. When, in due course,
slaves were no longer as readily available for purchase by the owners of
the means of production, the practice of indenture became adaptive once
again where the demand for labour in expanding economies could be met
from other parts of the world where population growth had outstripped
the capacity of the local economy to absorb it. Meanwhile, in the mid-
seventeenth century, an apposite quasi-experiment is to hand in the his-
tory of the short-lived Puritan colony established on Providence Island
in 1630, where, when imported black slaves turned out to be significantly
cheaper than white indentured servants, ‘considerations of godliness
therefore lost out to harsh financial realities’ (Elliott 2006: 103).
Such is the moral repugnance for slavery felt by the overwhelming
majority of those who write about it, and such is the appeal of the tele-
ology implicit in both Marxist and Whiggish stories of emancipation,
156 The theory of cultural and social selection
that it is tempting to underemphasize just how mutually adaptive slav-
ery’s defining practices have often been under local environmental con-
ditions. In the case of the United States, it was an institution which
had been explicitly accepted by the signatories to the Declaration of
Independence, even though repudiated in the different environment of
the North. The abolition of the Atlantic slave trade had no effect on
a society in which the slave population had become self-sustaining.11
Fugitives were distinguishable on sight from the free population, since
black/white interaction was categorically forbidden and the manumis-
sion rate was virtually nil. Slaves had no realistic prospect of amassing a
peculium sufficient to enable them to purchase their freedom, since the
local petty economy was the preserve of the poorer whites. Nor did they
have the opportunity of escaping to long-established and well-defended
maroon communities as opposed to short-lived bands of runaway out-
laws surviving as best they could on robbery and brigandage. Large-scale
rebellion was impossible to organize, and small-scale uprisings easy to
suppress. Before the war, as one historian has put it (Hahn 1990: 78), the
Southern planters ‘stood as one of the most – if not the most – imposing
of landed elites in the Western world’. The outbreak of the war was not
inevitable. Chance played its usual part. The government was initially
willing to accept the continuation of slavery if the rebels would rejoin
the Union. Even after hostilities had broken out, there might have been
a negotiated peace. Even if the trade-off was a slower rate of growth in
gross domestic product per head than that of the North, that did not
preclude the continuing reproduction of slavery.
Although slaves have often been physically indistinguishable – to the
annoyance of, among others, Athenian gentlemen in the time of Plato
who found themselves jostled in the street – and fellow members of the
same long-established society can nevertheless be condemned to out-
caste status by birth,12 the association of slavery with skin colour in the
11
Slave-breeding was not unknown in Rome (Appian, Bellum Civile 1.1.7): slave overseers, in
particular, were permitted wives and children, and there was a vernacular term for the roles
of home-bred domestic slaves (vernae). But there is no evidence, whether literary, epigraphic,
or archaeological, which would license an inference to demographic reproduction of the slave
population on anything approaching the American scale. Vernae ‘seem to have had a privileged
position and not infrequently to have become their masters’ heirs’ (Rawson 1989: 18).
12
As, notably, in the case of Japan’s ‘invisible race’ (De Vos and Wagatsuma 1966) – the Burakumin
descended from the members of residential communities attached to regional lords who were
assigned the performance of occupational tasks which the mode of persuasion stigmatized as
polluting. The American and Japanese anthropologists who studied them in the aftermath of
the Second World War found that ordinary Japanese believed them to be dirty, dangerous, and
diseased although (or precisely because) they never associated with them (Donoghue 1957).
Social selection and imposed behaviour 157
American South was an additionally adaptive aspect of its design.13 So
too was the combination of a few deliberate concessions on the masters’
terms with the twin sanctions of unrestrained physical coercion14 and the
ever-present threat of being ‘sold down river’. Asked to design a set of
practices which would give the institution of slavery the highest probabil-
ity of ongoing reproduction in a favourable ecological and demographic
environment, a reverse engineer would be likely to suggest that the slaves
should be allowed to rear their own families, but not necessarily to keep
them together; be given plots ostensibly of their own to tend for occa-
sional days or hours, but remain dependent on their masters for rations;
be conceded a ‘freedom of the soul’ (Morgan 1988: 688) which could find
expression in forms of worship, speech, and dance, but be denied any
opportunity of formal education; and be rewarded by modest privileges,
including oversight of other slaves, for loyal service at the same time as
being severely punished for disobedience. Northern denunciations of
the brutality of Southern slaveowners impervious to injunctions to ‘use
their people well’ were no more going to drive slavery to extinction than
Southern denunciations of the rapacity of Northern employers exploit-
ing their wage-labourers to the limit were going to drive to extinction
the practices defining the roles constitutive of industrial capitalism. The
practices which denied the slaves any countervailing power against their
masters were mutually co-adaptive in all three dimensions of social space.
Viewed comparatively, it was not quite the ‘peculiar institution’ which
contemporaries frequently called it; and in the event, it was driven extinct
only when the society whose peculiar institution it was had been defeated
in war.
6
Yet if slavery was as adaptive as it was in the environment of the Southern
United States, how did it come to be driven extinct in the Caribbean
without a war despite similar selective pressures in its favour? This story
has been told many times and from many different viewpoints. There are
13
Contrast the situation in Peru, where freed non-whites passed where they could as dark-skinned
Spaniards (Bowser 1974: 333).
14
Masters were not free to kill their slaves at will, but the courts seldom returned a verdict of
unlawful killing: the constraint was only the sanction of public opinion in local communities
conscious of the risk of ‘cheapening the deterrent itself ’ (Wyatt-Brown 1982: 372). As the ex-slave
Frederick Douglass deliberately understated it in his autobiography (1855: 53), public opinion
is ‘not likely to be very efficient in protecting the slave from cruelty’ in a rural community of
slaveholding neighbours.
158 The theory of cultural and social selection
stories of inexorable progress towards emancipation, stories of rival inter-
est-groups in pursuit of perceived short-term advantage, stories of saints
and heroes routing the forces of evil, stories of an inescapable contradic-
tion between slavery and capitalism, stories of emancipation ‘from above’,
and stories of emancipation ‘from below’. But when viewed from within
selectionist theory, the sequence of unpredicted events through which
it came about offers as striking a contrast to the story of the decline
of slavery in post-Roman Europe as it does to the story of its extinc-
tion in the Southern United States. Chance once again played its usual
part. Abolition was favoured by unforeseen events which ranged from
the American and French Revolutions and the slave uprisings in Haiti
and Jamaica to the changed distribution of seats in the British House
of Commons which followed the passage of the First Reform Bill by a
single vote. But whatever the disagreements between rival historians over
the might-have-beens, the abolitionists’ success at the culmination of a
protracted series of arguments and compromises depended among other
things on a diffusion and reproduction of two independent cultural muta-
tions which had happened to coincide.
No observer present at the Peace of Paris would have been other than
surprised to be told that slavery would be condemned by the Congress of
Vienna. Until the middle of the eighteenth century, abolitionist memes
barely existed outside the heads of a handful of jurists or philosophers.
The occasional criticisms voiced by Christian priests were no more suc-
cessfully diffused and reproduced than those of their Islamic counterparts
(Clarence-Smith 2006). But then there occurred a memetic mutation
centred in one particular Christian sub-culture: that of the Quakers. As
one historian reports two of the significant incidents in the story, ‘In 1758,
Philadelphia Yearly Meeting cautiously resolved to exclude from business
meetings any members who subsequently bought or sold black slaves; the
meeting also authorized a committee to visit Quaker slaveholders, begin-
ning a quiet, charitable, and ultimately effective campaign to persuade
all Friends to manumit their slaves. In 1761 London Yearly Meeting ruled
that henceforth Quaker slave dealers should be disowned’ (Davis 1984:
107). The speed with which this attitude was diffused among the mem-
bers of other Christian denominations during the following decades is
remarkable not only because it is an instance of an unexpected runaway
effect but because the Quakers were not, sociologically speaking, located
any higher within their society’s mode of persuasion than their denomina-
tional competitors. They might be admired for their piety and respected
for their quietism, but they did not command the ideological resources
Social selection and imposed behaviour 159
of either the Methodists or the established Anglican church. There was
no obvious reason to expect their opposition to slavery to be any more
widely endorsed by other Christians than their opposition to war. But
the abolitionist memes in Quaker heads were so successfully diffused
among other dissenting and evangelical congregations that their pheno-
typic effects extended to ‘recurrent bursts of petitioning on an ever larger
scale’ to Parliament, culminating in over 5,000 at the climax of the cam-
paign in 1833 (Turley 1991: 65). At the same time, those same memes were
carried to Jamaica, in particular by Baptist and Wesleyan missionaries
from whom they spread first by direct transmission from missionary to
convert and then, after a further mutation, to the expanding population
of ‘Native Baptists’ – ‘a generic term for a proliferation of sects in which
the slaves developed religious forms, more or less Christian in content,
that reflected their needs more closely than the orthodox churches, black
or white’ (Turner 1982: 58).15 When the missionaries sent back to the read-
ers of the Baptist and Wesleyan church magazines and the Anti-Slavery
Monthly Reporter reports of the ill-treatment of slaves, they combined
to adaptive effect both with Protestant doctrines of atonement and with
anti-Catholic beliefs and attitudes of the kind whose phenotypic effects
in an earlier context were discussed in Section 11 of Chapter 3.
At the same time, but for quite other reasons, slavery was coming under
increasing attack from political economists in whose eyes it was short-
sighted, retrograde, and wasteful. It is true that Adam Smith’s objection
to it was moral as well as political: he thought that it corrupted the slave
owners by gratifying their ‘love to domineer’. But he and others like him
believed that slaves, because they were denied any opportunity of better-
ing their own condition by their exertions, were bound to be less produc-
tive than wage-workers recruited in an open competitive labour market.
The puzzle in this, as Smith’s commentators have frequently pointed out,
is that his claim in Book III of The Wealth of Nations that ‘the experience
of all ages and nations’ has shown slaves to be less productive than wage-
workers is supported by nothing more than three references to Pliny,
Columella, and Aristotle respectively. In any case, once the slaves were
freed they preferred the role of smallholder to the role of wage-worker. In
Antigua and Barbados, where there was much less land available on which
they could settle, the selective pressure in favour of proletarianization was
Here, the instructive contrast is with the ‘Folk Catholicism’ of Spanish America (Palmer 1976:
15
152), which was similarly diffused among the slave population but without any equivalent con-
demnation of slavery as such.
160 The theory of cultural and social selection
for that reason stronger. But in Jamaica, the production of sugar was
halved once the former slaves were fully free and not indentured (as they
were for an initial four-year period following emancipation). The planters’
arguments against the economists turned out to be better-founded than
the abolitionists had been prepared to concede. But from a selectionist
viewpoint, it is a just-so story of meme-practice co-evolution driven by
indirect bias. The economists were believed not because their arguments
had been tested, but because they were the people who were putting them
forward and the evangelizing abolitionists were the beneficiaries of their
prestige.
The comparison which the extinction of slavery in the British
Caribbean then invites is its abolition in Brazil, where abolitionists were
increasingly successful in the years up to the formal legislation enacted
in 1880 in encouraging local slaveowners to manumit. The memes in the
Brazilian abolitionists’ heads were, however, not at all the same as those
in the heads of their British and North American predecessors. After
the Brazilian government’s belated decision, under British pressure, to
suppress the Atlantic slave trade on which replacement of Brazil’s 2½
million slaves depended, members of the Brazilian elite ‘imbibed [in the
authors’ chosen metaphor for memetic transmission] liberal values with
undiluted enthusiasm’ (Cain and Hopkins 1993: 300). But their motive
was to preserve their standing and therewith their creditworthiness in
the eyes of the British government and the City of London and to be
perceived as ‘modernizers’. The Brazilian planters had no choice but to
acquiesce in emancipation when faced not only with the cutting-off of
the supply of slaves but with ‘decentralized direct action’ (Drescher 1988:
45) by slaves migrating en masse from the plantations to ‘free zones’.
Compared to the planters of the American South, they were at a double
disadvantage: the much more widespread practice of manumission had
narrowed the ideological distance between dark-skinned slaves and free
whites, and the planters lacked the means of coercion available to their
counterparts in the American South (or Rome in the time of Spartacus)
which would have enabled them to hunt down and kill or re-enslave
the fugitives. The evolution of the Brazilian mode of production out of
near-total reliance on slavery in regions where it was based on sugar or
coffee to one based on wage-labour (including immigrants from Europe
as well as native-born Brazilians) was, when it occurred, as rapid as its
occurrence had been unforeseen by the members of the first post-colo-
nial governments. There had not, as in Mexico and Peru, been a history
of extensive recruitment of the surviving indigenous Indian population
Social selection and imposed behaviour 161
into wage-labour through the practices and roles which created the insti-
tutions of repartimiento and mita. The practices by which Brazilian slav-
ery was defined were doomed to extinction in a way that had applied
neither to the Caribbean nor to the Southern United States. The select-
ive pressures which doomed them are easily discernible in hindsight: the
story I have briefly summarized is not, so far as I am aware, contro-
versial among present-day historians of Brazil. But it demonstrates once
again just how differently the common underlying process of heritable
variation and competitive selection of information affecting phenotype
can work to produce the same evolutionary outcome in different local
environments.
7
It is, accordingly, because of the diversity of practices which fall between
slavery and other forms of domination, not because slavery is a pre-
determined stage in a teleological progression out of ‘ancient’ through
‘feudal’ to ‘modern’ society, that the just-so story of the evolution of the
modes of production, persuasion, and coercion of the societies of post-
Roman Europe is as instructive for comparative sociology as it is. The
starting-point is an environment in which plantation-type chattel slavery
like that of the Caribbean or the Southern United States was confined to
a relatively few areas in certain parts of Italy, Southern Gaul, and Sicily: as
one historian has put it, the system expounded by Cato, Columella, and
Varro, rather than driving to extinction the practices of smallholding and
tenancy, ‘filled a gap’ in Italian agriculture (Rathbone 1983: 162). It is not
in dispute among historians of Rome that there was a large-scale impor-
tation of slaves in the period following the Punic Wars. But the slaves
were distributed across the whole civilian labour force. They included the
unproductive retinues of domestic servants in the households of the rich,
the possessors of marketable skills buying their freedom in instalments,
the servile government functionaries of the emperors’ households, and
the incumbents of quasi-professional roles such as supervisor, secretary,
architect, and doctor. Numbers of successful escapees16 are even less pos-
sible to calculate than manumissions. But there were always some of both.
Weber’s arresting simile of the Roman economy devouring human beings
as a blast furnace devours coal is misleading twice over: slaves were not
16
Including those who made collusive bargains with a slave-catcher by using either their peculium
or stolen goods (Daube 1952).
162 The theory of cultural and social selection
to that degree the fuel on which its per capita output depended, and they
were not to that degree systematically worked to death.17 But the propor-
tion of slaves in the population as a whole may well have been comparable
to that in the Caribbean and the Southern United States, and its decline
has to be explained as a part, although only a part, of the just-so story of
the evolution of the modes of production, persuasion, and coercion of the
societies of post-Roman Europe.
After the fall (if that metaphor is the right one) of the Roman Empire
in the West, there was nowhere a Quaker-like cultural mutation which
substituted abolitionist memes for the long-held acceptance that one
person could legitimately be held at the total disposal of another. Men
and women still designated as servi, ancillae, or mancipia were still
bought and sold, listed alongside livestock and tools in testamentary
dispositions, denied freedom of movement, forbidden to marry whom
they pleased, and subjected to punishments of a degrading kind. The
change was that the practices defining their roles became increasingly
negotiable as the roles constitutive of the institutions of the Roman
state lost the control of the means of coercion which had previously
attached to them. The ‘slaves’ began to have goods whose use they fully
controlled, homes which they could call their own, and wives who were
recognized as such.18 Their entitlement to days off work began to be
publicly recognized. They began to receive perquisites which acquired
the sanction of local custom. Their roles were still located close to the
apex of the inverted pyramid of economic, ideological, and political
power. But a ‘slave’ who is institutionally accorded a customary right is
no longer wholly unfree.
When the different evolutionary trajectories of the different societies
which emerged in the territories of Italy, France, Germany, England,
and Spain are contrasted with one another, the underlying theme which
emerges is the simultaneous decline in the numbers of the wholly unfree
and increase in the rate of mutation of the practices defining the roles
of the relatively freer. This emphasis may result in part from the nature
17
Ironically, Weber’s simile could more aptly be applied in his own country a generation after his
death to the roles of the slave labourers in the underground ‘Central Works’ of Nazi Germany
during World War II (Speer 1970: 500). Death rates may, however, have been as high (that is,
in the German case, up to 4 per cent per month) in the slave-worked mines of Roman Spain
(Diodorus Siculus V.38), as also in the gold and silver mines of Hispanic America and Brazil.
18
This last was true also – or appeared to be true – of Brazilian slavery, where the Constituiçôes
primeiras published in Lisbon in 1719 required slaveowners to allow slaves to marry ‘without hin-
drance or threat’ (Schwartz 1985: 385). But slave unions could be broken at any time by separate
sale at the owner’s whim.
Social selection and imposed behaviour 163
of the documentary sources, particularly the records of disputes between
landholders and their tenants preserved by the landholders for self-inter-
ested reasons of their own. But the general impression which they leave is
of societies in which roles are being renegotiated all the time in all three
dimensions of social space, with the courts being ‘used within strategies
of social negotiation’ (Fouracre 2005: 49; cf. e.g. Goetz 1993: 42) which are
themselves competitively selected as the environment changes. The prac-
tices defining the roles of the minimally free, the half free, the tacitly free,
and the formally freed are subject to constant mutation despite the social
distance between lords and their inferiors in economic, ideological, and
political power alike. In the mode of production, not only the rent-payers
but the dependents rendering services on the demesnes of bipartite estates
are doing their day’s work according to customary formulae: corvée labour
is compulsory, but corvée labourers are not slaves whose obligations are
limited only by the amount of work that can be extracted from them
by force. In the mode of persuasion, even mancipia ‘subject to total eco-
nomic and legal subjection’, are, as a historian of the middle Rhine valley
has put it, ‘also seen as moral actors in their own right’ (Innes 2000: 79).
In the mode of coercion, although rebellion would be met outright with
military repression, concilia or coniurationes might secure concessions by
which open resistance would be averted. The ‘speaking tools’ (instrumenta
vocalia) of the Roman jurists, for whom the law was nothing other than
their master’s will, can now act in defence of their status as persons, how-
ever limited the countervailing power attaching to their modified roles
remains.19
The mutations of practices which generated the new institutional
relationships between the members of the dominant systacts and their
vassals, clients, soldiers, tenants, retainers, administrators, bailiffs, ten-
urial serfs, personal serfs, landless labourers, domestic workers, and
the town-dwelling merchants and artisans with whom they also dealt
were little influenced by the memetic mutations which were taking
place alongside them. There were many of these which can be as read-
ily identified and traced as mutations of practices. Beliefs and attitudes
about the creation and application of wealth, about the criteria of social
prestige, and about the use and misuse of the means of coercion were
being actively reinterpreted all the time. The ongoing story of European
19
A critical difference was that ‘being sold along with landed property, however the conveyancing
was formulated, had nothing like the personal consequences of being sold and transported. The
one kept family and humanity; the other did not’ (W. Davies 1996: 246).
164 The theory of cultural and social selection
Christianity is full of passionate doctrinal controversies. But it was not
the Church which designed the role-maps of the societies within which
its own institutional structures evolved. On the contrary, the Church
reflected them. Acceptance of equality before God and expectation of a
life to come were quite consistent with acquiescence in economic, ideo-
logical, and political inequality in this one. Slaves might be manumit-
ted as acts of personal piety, but slavery as such was never condemned
as unacceptable to God and ecclesiastical foundations were as jealous
as any lay lord of the powers which they sought to retain over their
dependants both servile and free. For the abbots and bishops, the trade-
off was between total control of the labour of the servile and extraction
of tithes and scot from the freed. Monasteries and nunneries mirrored
in their internal organization the hierarchies of the outside world. In
contrast to the just-so story told in Section 5, the selective pressures
which determined the relations between topmost and lowermost roles
in post-Roman Europe were overwhelmingly social, not cultural. In
this environment, the drivers of meme-practice co-evolution were the
practices, not the memes.
Sociological comparisons between post-Roman Italy, France,
Germany, England, and Spain (Wickham 2005) are complicated by the
difficulty that the best-documented regional just-so stories are often
constructed from within national historiographical traditions which
presuppose that one must be the model of which the others are vari-
ants. Thus, the controversies surrounding the paradigm of seigneurie
banale and mauvaises coutumes imposed by castellated lords after the
mutation de l’an mil – if such there really was (Barthélemy 1997) – are
in part a function of a longstanding disagreement between French and
other historians about whether what happened in the territory ruled,
or at any rate claimed, by Charles the Bold can be generalized as a
pan-European evolution. But the sociology of the Ile de France is not
at all the same as that of the West or South, to say nothing of the soci-
ology of Catalonia or Latium. The institutional vacuum left by the col-
lapse of Roman power offers an exceptionally favourable opportunity
to observe the process of social selection in operation in different ways
as different populations moved into the social as well as geographical
spaces now open to them. One society, in particular, offers what Bloch
himself (1961: 181) called ‘the most precious of examples – that of a
society of Germanic structure which, till the end of the eleventh cen-
tury, preserved an almost completely spontaneous course of evolution’:
Anglo-Saxon England.
Social selection and imposed behaviour 165
8
The social space that the Romans left behind them after their evacuation of
Britain was more nearly empty than anywhere except Iceland, where people
similar to those who raided and sometimes settled in England moved into a
totally unpopulated territory and created a feud-riven society of independ-
ent slave-owning farmers loosely united under the ideological leadership of
their priests. Although the archaeological record reveals some continuity of
occupation of Roman villa-sites, it is a long time before there is evidence
of elaborate artefacts, large buildings, or extensive trade; and although the
indigenous British population was neither exterminated nor expelled by
the incomers, institutional survivals other than a few churches and burial-
grounds are hard to find. The roles and institutions of ‘late’ Anglo-Saxon
society evolved out of centuries of near-continuous raids, invasions, con-
quests, and reconquests within a mode of production in which ‘the plough
is king’ (J. H. Round, quoted by Lennard 1959: 350), a mode of persuasion
controlled (despite some obstinate pagan survivals) by a Roman rather than
a Celtic Church, and a mode of coercion in which armies of high-ranking
leaders and their subordinate foot-soldiers, supported by unarmed auxil-
iaries, fought each other with swords, axes, lances, daggers, javelins, and
spears. England was not impervious to the lateral transfer of both memes
and practices from the societies of continental Europe: the imitation of
Frankish styles in female dress and jewelry is as well attested as the adop-
tion of Frankish-style consecration by anointment of a royal heir, and in
areas of Danish occupation and settlement manorial practices are much less
in evidence than further West. But social selection can be observed at work
across half a millennium in a society whose just-so story, for all the resem-
blances between its constituent roles and those of the other post-Roman
European societies, was very much its own.
The starting-point is an environment of small tribal kingdoms –
‘kingdoms’, since they are headed by acknowledged leaders who, as I
remarked in Section 1, are not ‘big-men’ but rulers who, in Sahlins’ phrase,
have come to power; and ‘tribal’, since relations of kindred are critical both
to the acknowledgement of common social identity and to the mainte-
nance of order. Out of these there evolved through a combination of aggre-
gation and aggrandizement (Bassett 1989) a society which, as I remarked
in Section 5 of Chapter 1, was in many ways similar, despite the obvious
differences in ecology and climate, to Hammurabi’s Babylonia. It had in
broad terms the kind of four-tiered systactic structure which convergent
evolution has brought into being at countless times and places. At the top
166 The theory of cultural and social selection
was an elite of kings, nobles, and both secular and ecclesiastical landhold-
ers to whose roles there attached a near-monopoly of economic, ideological,
and political power. At the bottom were the mendicants, outlaws, prison-
ers, and slaves: an estimated ten to twelve per cent of the population of
England were slaves at the time of the Norman Conquest (Moore 1998),
not in the houses of the rich or the administrative offices of the king, but
on the rural properties of non-noble ceorls as well as of bishops, abbots,
and thegns. Above these were the dependent agricultural tenants, the small-
holders, the craftsmen, the servants, the men-at-arms, the shepherds and
swineherds, the fowlers, the bakers, the salters, the blacksmiths, the mill-
ers, and the occasional wage-workers (hyrmen or hyringmanna). But above
these, and below the elite, there evolved out of the roles of the warrior com-
panions of the early tribal kings the roles of the adjutant administrators and
officials on whom the incumbents of the topmost roles came increasingly
to depend. Alongside the burgesses, merchants, lesser clergy, and country
gentlemen (Gillingham 1995) are the ‘riding-men’, the drengs, geneats, and
radcnihts whose roles involve the performance of services for their lords
which are not, in the ideological dimension of power, degrading. The role
of reeve (gerefa) becomes increasingly ubiquitous: there are king’s reeves
and high-reeves and village-reeves and port-reeves and toll-reeves and wap-
entake-reeves and reeves charged with the collection of tithes and scot and
reeves paying hyrmen their wages, and reeves responsible for the execution
of convicted criminals. It is a story in which the careers of the leading pro-
tagonists, dramatic as they often are, are incidental to the selective pressures
which, as the society’s total resources continued to expand, brought into
being a growing systact of roles empowered to oversee their use. But there
are, at the same time, three things which it is not a just-so story of.
To a comparative sociologist who comes to the historiography of Anglo-
Saxon England from outside, the long debate between the partisans of a
‘manorial’ model on one side and a ‘peasant proprietor’ model on the other
is puzzling. A story of rapid and wide-ranging exploitation by seigneurial
lords of unarmed and impecunious villeins is as implausible as a story of a
rapid and wide-ranging entrenchment of democratically organized bands
of free warriors holding their acreages (or ‘hides’) under no superior other
than their elected kings. The ceorls of the sources for the early kingdoms
were clearly men of some substance with rights in ploughland, meadows,
and woods and dependents of their own.20 But what went on thereafter
The laws of Ine, from whose text generations of historians have sought to squeeze every possible
20
semi- or quasi-deductive inference, are unequivocal both about the possession by lords of estates
on which they have tenants who owe them rent or labour (in marked contrast to contemporary
Social selection and imposed behaviour 167
varied in accordance with the different selective pressures which came to
bear in the different environments of different regions. By the time of the
Norman Conquest, there are numerous manors of the kind depicted in the
memorandum on estate management known as Rectitudines Singularum
Personarum. But that document comes from Bath (Harvey 1993) – a by
no means typical area where there had long been opportunities for build-
ing up large estates divided between a directly exploited ‘inland’, which
‘was recognized as privileged, in the sense of being exempt from a wide
range of public service and eventually from the payment of geld’, and
an outer ‘warland’ liable to geld from whose inhabitants public rather
than private obligations were due (Faith 1997: 16).21 Manorial inlands are
characteristic particularly of ecclesiastical estates worked by substantial
numbers of monks or ‘semi-monastic personnel’ (Blair 2005: 255) whose
roles are not easily distinguishable from those of other unfree tenants
whose labour generates the surpluses on which the minster depends for
the support of its unproductive members and its expenditure on new land
and new buildings, and the acquisition of prestige goods. There is little
doubt that extraction of surpluses and imposition of services intensified
under both lay and ecclesiastical ownership. But there are large areas of
England, both within and outside of the Danelaw, where manorial prac-
tices had hardly penetrated by the time of the Conquest – remote uplands
with scattered farmsteads and hamlets, settlements newly carved out of
woodland and waste, regions of partible inheritance and jointly held fam-
ily farms, regions where rent-paying censarii perform minimal labour ser-
vices, villages subordinated to no single lord, regions where agrarii milites
are needed for the purpose of defence, and regions where sheepfarming
by prosperous sokemen supplements the produce of ploughlands. Behind
the variegated pattern of landholding in different local environments and
the histories (to the extent that they can be reconstructed) of different
individual estates, there is a general trend away from the tributary mode
of production of the early tribal kingdoms in the direction of rent (gafol)
and tax (geld). But it is a trend within which some large estates were built
up while others fragmented, some ceorls acquired enough land to rank
Ireland, where the adaptive practice was clientship rather than tenancy) and about the relations
with one another of independent ceorls with their own households, cattle, ploughs, and shares in
common woods and meadows. In the laws of Aethelbert, King of Kent, it is explicitly envisaged
that a ceorl may have his own ‘loaf-eater’ (hlafaeta), and the payment due to him if someone kills
his loaf-eater is the same as that due to him from the first of several intruders who force their way
into his homestead (Liebermann 1903: 4).
21
Roffe (1990: 317) draws attention to the ‘elaborate calligraphic conventions’ adopted to distin-
guish tributary from demesne land in .
168 The theory of cultural and social selection
as thegns while some lesser thegns were left with manors barely deserving
of the name, and some dependent cultivators were confined to smaller
and smaller holdings while others were able to purchase their freedom
of movement in land or cash. Land was constantly changing hands not
only by inheritance, usurpation, or bequest but in an active sale and leas-
ing market (in which the surviving wills show women as well as men
participating directly), and new territories were being settled or resettled
at the same time that established holdings were being more systematically
organized and exploited to better effect.
Surprising also to a comparative sociologist is the popularity among
Anglo-Saxon historians of a third just-so story: that of the progressive
evolution of a strong central state. ‘State’, like ‘institution’, has generated
unhelpfully many definitional disagreements in the literature of sociol-
ogy, and there is always scope under any definition for alternative views
about how the strength of a state is to be assessed. In selectionist theory,
a society is a ‘state’ if and only if institutional control of the means of
coercion attaches to specialized extra-familial roles in which rulers and
their associated deputies and officials succeed one another on a regular
basis: a state is not a state, any more than an empire is an empire, if it
does not outlast the death of its founder and the charismatic legitimacy
attaching to the founder’s person. The practices which defined the roles of
the late Anglo-Saxon kings made them significantly more powerful than
their itinerant predecessors dependent, along with their retinues, on the
food-rents drawn from the different parts of their shires. They had ‘the
first regular and permanent land-tax known to the West in the Middle
Ages’ (Loyn 1962: 305) which they could on occasion use their troops to
collect (Campbell 2000: 227); they controlled a coinage which was both
sophisticated and flexible and enabled them to benefit directly from new
issues (Lawson 1993: 197); they could call up fighting-men for expeditiones
regum on something approaching a national scale; and their administra-
tion of justice extended down through the holders of sake and soke to
the hundreds and wapentakes. But they failed the standard sociological
test: they did not sufficiently control the means of coercion across the
territory occupied by those who acknowledged their titular roles. Theirs
was, in Weberian terminology, a ‘patrimonial’ state in which the assess-
ment and collection of tax, the recruitment of soldiers and auxiliaries,
and the apprehension and punishment of malefactors were in the hands
of ealdormen, abbots, bishops, and thegns whose roles were no more those
of a service nobility than they were those of a salaried bureaucracy. The
kings are careful to consult them as seniores or sapientes whenever royal
Social selection and imposed behaviour 169
initiatives are in prospect. Although the justice which they administer
is the king’s justice, they can draw profit from doing so directly. A king
can be deposed, and the witan can choose one aspirant over another. The
kings have no corps of informants and spies reporting to them. They can-
not be certain that their subordinates will fulfil their obligations to keep
bridges and fortifications in repair or discipline the lawless men of their
districts as they are supposed to do. Their law-codes are statements of
aspiration as much as of fact.22 They have not abolished the feud and they
cannot enforce payment of compensation in place of killings in revenge.
There are no royal burghs north of York. There are villani too rich to be
punished.23 The kings are acknowledged as rulers of all England, but the
unity of the kingdom is more ideological than either economic or politi-
cal. Maitland was right to remark as he did that it was their splendour
which had grown more than their power.
There is no way of knowing how English society would have evolved
without the Norman Conquest, any more than there is of knowing how
the American South would have evolved without its defeat by the North.
But to a selectionist sociologist trying to detect just what was going on up
until 1066, the two distinctive features of Anglo-Saxon England’s modes
of production, persuasion, and coercion are, first, the diversity of the roles
constitutive of them, and second, the extent of both inter- and intra-gen-
erational social mobility between those roles. Mobility rates do not in
themselves constitute a modal difference between one society and another:
individuals can rise and fall in all three dimensions of social space with-
out their movement having any effect on the practices which define the
roles between which they move. But institutional sanctions against any
possibility of mobility do constitute a modal difference: if in the mode
of production sons are compelled to follow their father’s occupations, or
in the mode of persuasion members of a stigmatized status-group can
escape from their origins only by ‘passing’, or in the mode of coercion
participation in political decision-making is forbidden to any member of
the population who is not a citizen by birth or formal decree, their rela-
tionship with the incumbents of the other roles by which their society
is constituted will be qualitatively different as a result. In Anglo-Saxon
22
Of them all, the one whose sanctions are least convincing is King Edgar’s, promulgated some
time between 959 and 963 (Liebermann 1903: 194–207). How often are we to believe that a late
payer of ‘St Peter’s pence’ took it to Rome in person and paid 120 shillings to the king on his
return, or a judge who made an unjust judgement paid the king 120 shillings, or a willing buyer
and willing seller of wool at other than the prescribed price paid the king 60 shillings each?
23
As explicitly admitted in a letter from the bishops and ‘wise men’ of Kent to King Athelstan
(Liebermann 1903: 170).
170 The theory of cultural and social selection
England, the social distance between economic, ideological, and polit-
ical roles steadily widened as the ongoing mutation and renegotiation of
practices brought into being the systactic structure outlined above. But
the gradations between roles were close and the possibility of movement
from one to another was always open. However it is to be classified, it was
emphatically not a society of castes.
Sociologists calculating mobility rates for present-day industrial soci-
eties from survey and census data are familiar with the combination of
an increase in the proportion of high-ranking roles occupied by entrants
from below with the absence of a corresponding increase in the chances
of individuals born to lower-ranking parents rising into higher-ranking
roles. ‘Structural’ mobility increases because of the expanding number
of higher-ranking roles needing to be filled, but there is little ‘exchange’
mobility in which downwardly mobile individuals born to higher-ranking
parents are directly replaced from below. But in Anglo-Saxon England
there is plentiful evidence for both in all three dimensions of social space.
Not only did large landowners need skilled assistants, but an expanding
Church needed literate priests and local lords needed experienced warriors
who would – however unreliably – pledge fealty (Abels 1988: 153). At the
same time, downwardly mobile thegns were being replaced by upwardly
mobile sokemen, ‘unhappy ex-clerks’ (John 1977: 180) were being
replaced by better-educated (gelaeredum) novices, and slaves (according
to Archbishop Wulfstan) were running away to the Vikings and enslaving
the thegns who had been their masters. If, at the time of the Conquest,
some three-quarters of the English population were in the roles of villani,
cottarii, or bordarii (with almost all women taking ‘secondary rank’, in
the technical sociological sense, from men), only a very small proportion
of them could hope to rise above their systact of birth. But the roles above
theirs were not all being filled by individuals born at that level. Moreover,
of those who remained villani some were becoming significantly richer
than others in land or plough-teams, some enjoyed higher prestige than
others as witnesses and compurgators, and some were better able than
others to defend themselves against predators or, as it might be, the agents
of royal justice. These differences evolved out of the different practices
negotiated by subordinates with their superiors24 in an environment of
endemic violence, intense competition for control of land, and a decline
24
The way in which gesith, as the vernacular term for a companion rather than servant, ‘fights a
losing battle against thegn’ from the beginning of the tenth century (Loyn 1955: 530) is directly
symptomatic of the mutation of practices defining the relationship of nobles to kings.
Social selection and imposed behaviour 171
in wergild as the criterion of status in favour of the nature of services per-
formed and the life-style that went with them. To say that it was a society
in constant flux is to restate the obvious. What is remarkable about what
was going on is that it was a society in which the roles constitutive of its
changing economic, ideological, and political institutions could be occu-
pied by individuals moving into them from other systactic origins with-
out the constraints which might have been expected to keep them within
their systact of origin – gesithcund children of nobles as nobles, ceorls as
children of ceorls, and children of slaves as slaves.
Once again, therefore, as so often in just-so stories of social selection,
it is what wasn’t going on which is as instructive as what was. To look
behind the rhetoric of the law-codes, the questionable reliability of the
narrative sources (Bede’s Ecclesiastical History included), the terminology
of the haphazardly surviving wills, charters, letters, and writs, the hints
embedded in contemporary literature (Aelfric’s Colloquy or the homilies
of Wulfstan no less than Beowulf or The Battle of Maldon), and the tan-
talizing ambiguities and inconsistencies of Domesday Book to the prac-
tices which defined the society’s roles is to be struck by the absence of
inter-systactic conflict, whether between economic classes, ideological
status-groups, or political factions. There are no peasant revolts or col-
lective withdrawals of labour, no heretical churches or millenarian sects,
no revolutionary uprisings or breakaway states. England, as its historians
have often observed, was by 1066 an old society, and by then a very dif-
ferent one from what it had been half a millennium before. But its evo-
lution had been gradual, for all the relentless competition within it for
economic, ideological, and political power. What, therefore, is the wider
implication to be drawn for comparative sociology? Does that make it an
exception, or a rule?
I nc r e m e n ta l m u tat ions
a n d pu nc t uat e d e qu i l i br i a
9
Biologists have disagreed ever since Darwin about whether natural selec-
tion proceeds more, as Darwin thought, by a protracted sequence of grad-
ual changes or by long stretches of stability punctuated by occasional
abrupt leaps. To comparative sociologists for whom examples of both
gradual and sudden change come easily to mind, it may be surprising that
the disagreement among biologists should be as vehement as it has been.
172 The theory of cultural and social selection
Don’t natural, cultural, and social selection all proceed at different speeds
under different environmental conditions? But there is a larger question
at issue. It may be that there can be listed specific antecedents of crises,
breakdowns, and upheavals which make major transformations of social
institutions more likely. But practices have a capacity to reproduce them-
selves for longer periods, and with fewer and smaller mutations, than
contemporary observers expect. Contemporary observers, however well
informed, are as capable of convincing themselves either that nothing has
changed, institutionally speaking, when in fact it has as that a revolution-
ary social transformation has taken place when in fact it hasn’t. As always,
it is the practices, not the people, on which the attention of sociologists as
opposed to historians needs to be fixed.
‘Revolution’ is yet another word which neither sociologists nor histori-
ans are able to do without but which, like ‘nationalism’ and ‘feudalism’,
means too many different things to different people. Defined as the sud-
den replacement of one set of practices and therefore roles by another
which is sufficiently pervasive and sufficiently lasting for all observers to
agree that the society has evolved out of one mode of production, per-
suasion, or coercion into another, ‘revolutionary’ change can be driven
either from below or from above, and imposed either from without or
from within. But it is more than a phase in a recurring cycle of practices,
and more than the replacement of one set of incumbents of existing roles
by another. The mutant practices have to be identified, their transmission
traced, and their adaptedness accounted for. That the makers of revolu-
tions can set in train events whose consequences are not merely differ-
ent from, but contrary to, their stated aims is a historical commonplace.
But the phrase ‘punctuated equilibrium’ captures the distinction between
qualitative change which results from a steady, ongoing renegotiation of
practices and qualitative change which results from near-instantaneous
jumps within a three-dimensional institutional design-space. Some muta-
tions in heritable information affecting phenotype are more revolution-
ary than others, and selectionist sociologists need to find out just what is
going on when they are.
Of all the revolutions in the historical record, the one which at first
sight looks closest to the ideal type of a rapid, comprehensive, and pro-
spectively lasting institutional transformation is the Bolivian Revolution
of April 1952. After three days of fighting, a textbook alliance of middle-
class radicals, mineworkers, and peasants under the leadership of an avow-
edly revolutionary party defeated the government in power and the army
supporting it. Thereafter, control of the means of coercion was handed
Social selection and imposed behaviour 173
over to civilian militias; the Indian population was accorded the ideo-
logical equality of status it had long been denied; and hacienda land was
handed over to peasant syndicates at the same time that control of the
mines was handed over to the government-controlled Corporacíon Minera
de Bolivia. But there were no mutations or recombinations of practices
sufficient to transform the roles constitutive of the modes of production,
persuasion, or coercion of Bolivian society. The army was reconstituted.
The peasants were concerned above all with holding title to the land they
had taken, and the supply of food to the cities was so far reduced that
the government became dependent on large-scale imports funded by the
United States. The hacendados were replaced by syndicate bosses, but if
they remained (as many did) they continued to perform the functions
necessary for commercial agricultural production and marketing. The
mineworkers continued to sell their labour for wages under similar condi-
tions of employment. The members of the urban middle class suffered a
drastic reduction in the value of their property as a consequence of devalu-
ation of the currency, but they were not dispossessed. The ex-hacendados
no longer enjoyed the personal services which the practice of pongueaje
had imposed on the Indians on their estates,25 but patronage roles, and
associated expectations of services from clients and their families, passed
to the bosses of the local sindicados. To sociologists whose interest is in
the roles rather than their incumbents, the interesting question is why the
practices which revolutionaries are determined to drive extinct are, as in
this case, so often capable of survival and ongoing reproduction neverthe-
less. And that applies not least to what is, to many historians, ‘the’ revolu-
tion – the one which in 1789 dismantled the institutions of what thereby
became, almost literally overnight, France’s ‘ancien régime’. As a topic in
selectionist sociology, the French Revolution offers an almost ideal-typical
case-study of a punctuated equilibrium whose evolutionary outcome was
as paradoxical as its occurrence had been unforeseen.
10
Not even those historians of France who believe that the ancien régime
was doomed to inevitable extinction will dispute the conclusion of
Georges Lefebvre that the Revolution required a ‘truly extraordinary
25
Pongueaje had, ironically, been formally abolished in 1945 by Gualberto Villaroel, who was to be
hanged by the mob from a lamp post outside the presidential palace in the following year. But
the decree was never put into force (Klein 1982: 219).
174 The theory of cultural and social selection
and unforeseeable (vraiment extraordinaire et imprévisible) combina-
tion of immediate causes’ (1954: 247). The many might-have-beens in
the sequence of events which culminated in the measures passed by the
Constituent Assembly on the night of 4 August 1789 extend all the way
from the decisions which led to the defeat of the French by the Prussian
army at Rossbach on 5 November 1757 to the decisions which led to the
withdrawal of the troops from the streets of Paris on the evening of 12
July 1789.26 All historians of the period agree that the conduct of Louis
XVI introduced a random disturbance into a sequence which would have
been very different without him. Lefebvre was as ready as any other to
concede that matters would have turned out otherwise if the throne had
been occupied by ‘un Henri IV, ou même un Louis XIV ’ (1939: 29). Had
Louis XVI been willing in 1787 to allow a commission of auditors to
oversee the government’s management of its finances, or had he allowed a
vote on the registration of new loans at the royal session of the Parlement
of Paris held on 19 November of that year, or had he agreed to announce
at the royal session of the Estates-General on 23 June 1789 that the three
orders could deliberate and vote in common, or had he not timed his dis-
missal of Necker so that the news of it broke on a Sunday (which meant
that, although the Assembly wasn’t sitting, the working population was
free to riot), there would have been no fall of the Bastille on the 14th.27
But the king would not have been faced with the choices which it fell to
him to take if there had not been the chance conjunction of the worst
harvest in living memory with the collapse of the government’s ability to
borrow money. It is true that bad harvests, and a consequential rise in the
price of bread to the point of provoking public disorder, were a perennial
risk which every eighteenth-century government faced. It is also true that
the parlous state of the government’s finances had been known since at
least the summer of 1786. But on no theory can it plausibly be hypoth-
esized that there was a predetermined connection between the two. Both
were the outcome of sequences of cause and effect which historians of the
period have analysed in detail. But to a sociologist, their conjunction was
26
We shall never know what might have happened if, at a council of war held by the King on the
morning of 16 July, Marshal de Broglie had not repeated his view of the previous day that the
attitude of the troops precluded any hope of military operations against Paris. But we do know
that at Rennes on the following day, the troops of the Artois and Lorraine Infantry and Orléans
Dragoons refused to obey the order of the military commander of the city to fire on the crowd
(Scott 1978: 60, 70).
27
Nor, in any case, need the Bastille have fallen then. Even in the orthodox Marxist account given
by Godechot (1965: 271), it is conceded that ‘a determined commander (un chef résolu)’ could
have held out against the attackers.
Social selection and imposed behaviour 175
as much a random intrusion into the course of social evolution as is, to
a biologist, a sudden climate change which alters what would otherwise
have been the evolution of an animal species.
Once, accordingly, the equilibrium of the ancien régime had been punc-
tuated, the three-dimensional institutional design space open to invasion
by mutant practices expanded on an accelerating scale. But the scope
which it offered for liberty, equality, and fraternity was much less than
that slogan implies. The royal exchequer was bankrupt and tax-collection
had been brought to a standstill, but the economy continued to func-
tion and to favour the propertied over the propertyless. The ideological
domination of the monarchy had been fatally undermined, but ascrip-
tive distinctions of social status survived. The means of coercion were no
longer in the monopoly control of the king and his ministers, but they
were still deployed by the incumbents of roles to which there attached
effective command over bureaucrats, soldiers, and police.28 Although
there was, despite violent opposition in some parts of France, no success-
ful counter-revolution, it turned out that for all the ambition of its leaders
and the fervour of their followers, the Revolution changed far less in the
economic, ideological, or political institutions of French society than its
supporters had expected and its opponents had feared.
If a comparison is drawn in terms of the institutional differences still
visible a decade after the equilibrium was punctuated, France in 1799 was
more different from France in 1789 than Bolivia in 1962 from Bolivia in
1952, although less different than either Russia in 1927 from Russia in 1917
or China in 1959 from China in 1949. Detailed regional studies have shown
that what went on in those years and thereafter was far from the same
across the whole of France. Indeed, by the time that the electoral geog-
raphy of twentieth-century France could be plotted by political sociolo-
gists, the pattern of voting was found to reflect many of the same regional
differences that had been discernible in and after 1789. But no historian
goes so far as to claim that the systactic location of the nobility as it had
been in 1789 was ever again the same. Seigneurial rights and revenues
were never restored; the same degree of deference was never accorded to
status derived solely from birth; and political careers remained open to
aspirants who laid no claim to nobility. The change was not merely cyc-
lical, and it did not merely involve the upward and downward mobility
It is symptomatic that, for example, the Convention should have re-established coercive control
28
over the itinerant book-selling colporteurs as soon as they were suspected of peddling counter-
revolutionary material (Hemmings 1987: 97).
176 The theory of cultural and social selection
of individual incumbents between roles. There is no need for either his-
torians or sociologists to start disputing among themselves whether the
French Revolution was ‘really’ a revolution. But what makes it of particu-
lar interest for selectionist sociology is that many of the mutant practices
which emerged from it turned out to be maladaptive, and many of those
which turned out to be adaptive would have been favoured by changes in
their environment which were happening in any event.
In the mode of production, although substantial amounts of previously
noble- or church-owned land changed hands, it did not pass to the mem-
bers of what would thereby have constituted a new class of independent
peasant proprietors. Many nobles recovered at least some of the hectares
they had lost, even if they might have to borrow extensively in order to
do so. The practices defining the familiar roles of cultivators, proprietors,
tenants, sharecroppers, smallholders, and landless agricultural labourers
and servants continued to be reproduced in the countryside as before. So,
in the towns and cities, did those defining the roles of the industrial and
commercial employers, the financiers, the independent professionals, and
the wage-earning employees (including the numerous domestic servants).
Industrial workers were still required to carry their pass-books (livrets),29
and their attempts at collective organization in pursuit of improved wages
and conditions were still denied legitimacy. Taxes continued to be levied
in much the same way as under the ancien régime: in the words of one his-
torian, ‘the Revolution had done little more than give them new names.
The taille became the impôt foncier, the capitation became the personnelle-
mobilière, the droits de jurande were transformed into the patente’ (Zeldin
1973: I, 710). Successive generations of young men and women continued
to succeed each other as before in roles constitutive of a formally free
market in commodities and labour.
In the mode of persuasion, the immediate impact of the Revolution
was observable in the disappearance of deferential forms of address, the
ostentatious wearing of revolutionary clothes, the decapitation of statues,
the pulling-down of churches or their conversion to stables or workshops,
and the pillaging of libraries and archives. The phenotypic effects of the
new ideology included marriages of priests and nuns and abjuration of
clerical offices, as well as the imposed substitution of Revolutionary for
Catholic rituals, reform of the calendar, and the replacement of the priest
by the schoolmaster as the moral preceptor of the young. Yet the extent
Although employers were frequently ready, as might be expected, to take on workers sans papiers
29
As is the ‘rapidité incroyable’ of the runaway diffusion of dechristianizing memes, and there-
30
with desacralizing festivals and masquerades, all over France within the same fortnight in 1792
(Bianchi 1987: 231).
178 The theory of cultural and social selection
family connections remained as useful as they always had been. The fitful
extension of the suffrage, which at first included only a miniscule propor-
tion of the total adult male population of France, did not transfer polit-
ical power to le peuple, as opposed to giving rhetorical legitimacy to those
who acted in its name. Despite the importance of women in the events of
the Revolution, they were not enfranchised until 1945.
None of this is, so far as I am aware, controversial. But the conclu-
sion to which it points is that from a sociological, as opposed to a his-
torical, perspective the Revolution was in a double sense unnecessary. It
would not have occurred at all without an extraordinary conjunction of
chance events and of decisions, or failures to take decisions, on the part
of individual agents whose characters and temperaments would in other
circumstances have had no impact on the evolution of France’s modes
of production, persuasion, or coercion. At the same time, the changes in
these which did take place over the course of the nineteenth century were
the outcome of mutations and combinations of practices which would
have been selected in any case. It is not, therefore, a question of choosing
between a just-so story of gradual change and a just-so story of sudden
transformation. The story of the evolution of French society after 1789 is
a story of both.
But there is more to it than that. The events of 1789 created not only a
vacant social design-space enabling mutant practices to be diffused across
the modes of production, persuasion, and coercion but a vacant cultural
design-space enabling mutant memes to be transmitted from mind to
mind across as well as within systactic boundaries. The outcome was that
revolutionary memes turned out to have a much higher probability of
ongoing reproduction than did revolutionary practices.
11
Beliefs and attitudes hostile to both the monarchy and the Church had
been widespread during the 1770s and 1780s, and all historians of the
period agree that the conduct of Louis XVI and Marie Antoinette did
nothing to moderate them.31 But on no theory is it plausible to sup-
pose that the memes in the heads of the philosophes and Jansenists and
That is not to say that the attacks on the hapless Marie Antoinette were all founded on fact. She
31
cannot fairly be held responsible for the affair of the diamond necklace in 1783 and the flood of
libelles which it provoked if ‘the whole scheme was a swindle organized by a gang of confidence
tricksters’ (Blanning 2002: 413). But the libelles are in themselves symptomatic of the disesteem
in which she was held.
Social selection and imposed behaviour 179
Freemasons and parlementaires and scurrilous journalists and disaffected
army officers were going to overturn the institution of monarchy. Neither
the philosophes nor the Freemasons were preachers of revolution or regi-
cide, and the grievances articulated in the cahiers of the Third Estate
cannot be construed as anticipating the program of the Constituent
Assembly. Only after the initiative for devising a solution to the problems
confronting France had passed to the Estates-General, and the power
to legislate had then passed to the National Assembly, was there a run-
away selection of memes in the heads of the deputies and their public
which turned reformist beliefs and attitudes into revolutionary ones. In
the months which followed, the phenotypic effects of novel symbols and
discourses were everywhere visible, and whatever was culturally defined
as ‘privilege’ was denounced in all its forms. Although the impervious
die-hards were not confined to the First and Second Estates, and not all
nobles and clergy were among them, the scale and speed of transmission
of information by imitation and learning was such that memes which
had been nowhere inside the heads of even the most vehement of the gov-
ernment’s critics in 1788 had reached the populations of all regions and
systacts by 1790.
From a selectionist viewpoint, the most intriguing aspect of what went
on between then and Napoleon’s coup d’ état of 1799 is the fate of the
deliberately designed and sedulously propagated rituals through which
the ideal of republican virtue was intended to inspire and uplift the entire
population of France. Some of the behaviour to be observed was as much
evoked as acquired or imposed. Clothes, caps, cockades,32 and even plate-
ware and playing-cards became objects the sight of which could evoke
not merely verbal insult but physical assault.33 Many of the tens of thou-
sands of ‘liberty trees’34 planted across the country by the Revolution’s
supporters were dug up again with equal enthusiasm by its opponents
and then replanted by the municipal authorities. But the ‘elaborate
32
As with the symbolic markers which differentiated one human group from another in the very
earliest stages of cultural evolution, the origin of the symbolism is, or might as well be, random.
The tricolor itself was a substitute, in the contemporary observation of the Duke of Dorset, for
an initial choice of green for the patriotic cockades (Hunt 1984: 57 n.12) – green being associated
with the followers of the radical Duc d’Orléans.
33
The bonnet rouge, which in 1792 was still being worn by Parisian men (but not women) as a
marker of revolutionary enthusiasm, was by 1795 exposing those still wearing it to being attacked
in the street by ‘young dandies’ (Harris 1981: 310).
34
The trees have been interpreted as a recycling of the ‘archetypal image of the tree’ which ‘drew
from the remote symbolism of genealogical trees, as if the liberty tree made visible the will to
root out the ancient hierarchical power structures’ (Malaussena 2004: 162). But perhaps they
should be decoded simply as ‘symbols of a revolution taking root’ (Hébert 1995: 141).
180 The theory of cultural and social selection
ceremonies which centred on oaths of brotherhood and loyalty’, like that
enacted by 10,000 national Guards from Eastern France on the Plaine
de l’Etoile in November 1789 (Maza 2003: 92), were intended to influ-
ence the subsequent behaviour of those who took part at the same time
as enabling them to act out sentiments of patriotic allegiance. They were
explicitly religious,35 and they deployed a complex iconography of their
own. But again, for the purpose of comparative sociology the sources of
the symbols deployed are irrelevant. Many of them vanished as quickly as
they had arisen, while others – notably the tricolor and the commemora-
tion of the fall of the Bastille – did not. But whichever were or weren’t
reproduced over the longer term, there had been set in train a sequence
of heritable variation and competitive selection of memes of which some
continued to be reproduced long after the extinction of the practices with
which they had at first appeared to have a selective, and not merely an
elective, affinity.
The history of recognizably socialist – and alongside them of anarchist
and Communist – doctrines and programmes in the culture of nineteenth-
and twentieth-century France can be told as a sequence of sectarian
disputes, personal rivalries, shifting alliances, short-lived polemical pub-
lications, and evanescent groupings which failed to outlast their charis-
matic founders. But there is clearly discernible within it all a reproduction
and diffusion of both beliefs and attitudes in which opposition to mon-
archy and nobility was selectively combined with opposition to large-
scale ownership of private property. As so often, the combinations were
as important as the mutations. The critical memes were those by which
there was linked hostility to the inheritors of aristocratic privilege and
hostility to the ‘bourgeois’ possessors, as Louis Blanc defined them, of the
‘instruments of labour’. It was before, and independently of, Marx that
the bourgeoisie came to be defined in France not as a classe moyenne but
as a quasi-noble systact at once separate from, and hostile to, the peuple.
The same opprobrium was never (so far as I can discover) directed by
republican critics at the incumbents of either military or administrative
roles. The bourgeoisie were, in their critics’ eyes, those Frenchmen who
had profited from the Revolution at the expense of those in whose name
it had been achieved and by whose exertions it had been saved from its
enemies. It was, therefore, the extinction of the mutant practices that the
Revolutionaries had sought to introduce into the society of France which
‘Religious’, that is, in the sense of what the leading historian of the Revolutionary festivals calls a
35
12
Evolution never goes backwards. Successive generations of the carriers of
the units of selection are never going to come increasingly to resemble
their ancestors, biologically, culturally, or socially. When, in Section 13
of Chapter 3, I said that the so-called ‘secularization thesis’ is mistaken
twice over, I didn’t mean that European Christianity could ever become
once again what it was in the days of Aquinas. But that doesn’t rule out
what biologists call ‘atavisms’: stick-insects, for example, have shown
36
Differences over its meaning are reflected equally in the interpretations of each other’s writ-
ings by rival historians: cf. the ‘misunderstanding of French political sensibility’ attributed to
William Doyle, in a review of the French translation of Doyle’s Origins of the French Revolution
by François Gresle (1989: 646), on the grounds that Doyle’s treatment of Lefebvre confounds
Jacobinism and Marxism.
Social selection and imposed behaviour 183
that they can re-evolve wings long after the genes of which wings are the
phenotypic effect had been thought extinct, just as there can re-emerge
phenotypic effects of memes long thought to have been forgotten and of
practices long thought to have been abolished. Conversely, traits which
were adaptive in an earlier environment can turn out to be maladaptive in
a later one to the point that they are eliminated altogether from the popu-
lation. Selectionist theory has to take account not only of sub-optimal
design but of the frequent failure of locally optimal designs to sustain the
ongoing probability of reproduction of the genes, memes, and practices
from which they were initially put together. Maladaptions are as famil-
iar in sociology (Friedman 1982; Tainter 1988; Edgerton 1992) as in biol-
ogy. But in the study of social selection, a distinction needs to be drawn
between the decline of a society and its collapse. A society can suffer a
large-scale fall in its population, its productivity, its influence over others,
and even its capacity to defend itself without its modes of production,
persuasion, or coercion being modified. But in the sudden collapse of the
Central Maya in the period conventionally designated ‘Terminal Classic’,
a drastic fall in population was accompanied by the cessation of both
monumental and residential building, the disappearance of other than
essential manufactured goods, the near-total loss of writing, and the total
extinction of the practices defining the roles of the hitherto dominant
elite. The relative weight to be assigned to the several contingent causes
of the collapse is the subject of continuing archaeological debate. But
once a tipping-point is reached, as when the rulers of the Western Roman
Empire could no longer raise sufficient taxes to pay for an army and bur-
eaucracy adequate to defend the frontiers and ensure that the taxes were
paid, long-established practices can be driven to extinction as quickly as
they had once ascended the S-shaped logistic curve.
An equally well-known but smaller-scale example is what has been
called ‘the closest approximation we have to an ecological disaster unfold-
ing in complete isolation’ (Diamond 2005: 82) – Easter Island. Diamond’s
own just-so story has been challenged by Hunt (2007), who attributes
the halted regeneration of the forest to predation by rats. But the enigma
of the giant stone statues whose ruins have astonished visitors ever since
the Dutch explorer Jacob Roggereen first sighted them on Easter Day
1722 has been solved by the combined researches of archaeologists, ecolo-
gists, petrologists, dendrologists, and climatologists. Thor Heyerdahl of
Kon-tiki fame was wrong. The island was settled (although Hunt disa-
grees with Diamond also about the date) by Polynesians from the West
who brought with them an existing cultural tradition of which statues
184 The theory of cultural and social selection
representing high-ranking ancestors were an extended phenotypic effect.
For several generations, food surpluses were adequate to support a grow-
ing population and the workforce required for the quarrying of the stone
and construction, transport, and erection of statues which could be as tall
as 30 feet and as heavy as 75 tons. Then, whatever exactly the proximate
causes may have been, accelerating deforestation and the over-exploita-
tion or extinction of edible species of birds and fish resulted in starva-
tion, demographic collapse, and internal war. Sociologically, the story is a
story of rival clans of dominant chiefs and subordinate commoners who
collaborated sufficiently to make it possible for all of them to erect their
separate statues and compete with one another in their size. The result
was a runaway effect reminiscent of the islanders of Ponapae, who con-
tinued to compete with one another in growing larger and larger yams
long beyond the point of nutritional efficiency (Bascom 1948). In a closed
environment, the memes and practices whose co-adaptation had initially
enabled the statues to be constructed and erected became increasingly
maladaptive for the natives’ genes. The islanders encountered in 1774 by
Captain Cook (as quoted by Diamond 2005: 109) struck him as ‘small,
lean, timid, and miserable’. After the collapse, there evolved a new mode
of production based on much reduced crop yields and extensive chicken-
rearing, a new mode of persuasion based on the cult of a single creator
God, and a new mode of coercion based on dominance by military lead-
ers whose role was designated by a new vernacular term (matatoa). But
no inconsistency with selectionist theory is implied in characterizing the
story which preceded those changes of mode as one of a decline which
turned into a collapse.
13
A further possibility is that social evolution is prevented from follow-
ing the path-dependent trajectory that it would otherwise have taken
by a parasitical practice which reproduces itself at others’ expense. The
example I have chosen because it is directly relevant to the topics dis-
cussed in Section 11 is the sale of offices in France between 1467, when
offices were made tenable for life subject only to royal suppression, and
1789, when venality was abolished (subject to compensation) by the
National Assembly. Sale of offices is one of those practices whose short-
term adaptedness may all too easily conceal its maladaptive longer-term
consequences. In Section 3, I cited the farming of taxes as a familiar
example where the attraction to rulers seeking to extract resources from
Social selection and imposed behaviour 185
their imperial peripheries can lead to over-exploitation by the farmers
and consequential default by the taxpayers. But in France, the range of
offices sold, many of which had been created simply in order to be sold,
extended so widely that it becomes the exceptions which call for com-
ment – the navy, the law, and the collection of direct taxes by commis-
sioners appointed specifically for the purpose. Nobody suggested at the
time that the incumbents of venal roles were being individually selected
for their ability to perform them well. On the contrary, one of Louis
XIV’s comptrollers-general is alleged to have quipped that no sooner did
the king create offices than God created fools to buy them. Nor did any-
one suggest that the practice furthered the country’s trade or its level of
productive investment or the efficiency of either its central or its local
administration. Yet by 1789, there were some 70,000 venal roles occu-
pied by what amounted to 1 per cent of the total adult male population
of France. What was going on?
The answer is to be found in the society’s mode of persuasion. The
motive of the purchasers was to acquire social prestige as institution-
ally defined by an ideology of inherited rank, and therewith the defer-
ence imposed by it, which descended from the king through the ducs et
pairs, the nobles d’ épée, and the nobles de robe to the well-born rustics
whose poverty was such that they had to have swords lent to them when
they came forward as potential delegates to the Estates-General in 1789
(Cobban 1964: 29). But the nobility was not, although sometimes depicted
as such by its critics, a caste. Of the 70,000 venal offices of 1789, over
4,000 ennobled either their incumbents or their incumbents’ heirs, and
this was a privilege for which a steady supply of bourgeois were happy to
pay and, if necessary, to borrow from intermediaries in order to be able to
do so. There were some pecuniary returns to be had from investment in
offices, even if the topmost financial roles giving access to the revenue of
the crown were the only ones whose purchasers would become seriously
rich. Some offices carried entitlement to levy fees, some entitled their
holders to interest on the capital sum they had paid, and many exempted
them from taxes to which they would otherwise have been liable. But
privileges ‘were a far more important attribute of most advertised offices
than any financial return’, and ‘privilege was as much about prestige as
profit, and not merely prestige as obvious as ennoblement’ (Doyle 1996:
155). Social standing in a local community, an occupation with duties dis-
proportionately small in relation to the ostensible role, a red robe, a fine-
sounding title, a seat on a fleur-de-lys-covered bench in a sovereign court,
and association on equal terms with others similarly privileged – these
186 The theory of cultural and social selection
were the institutional rewards which venality placed on offer and which
enabled the purchasers to rise above their systactic origins in the ideo-
logical dimension of social space.
Yet the practice of venality was constantly being criticized, and the crit-
ics always had the best of the argument – which may in part account for
the runaway willingness to abolish it in August 1789 on the part of dep-
uties, many of whom held venal offices themselves. Its few defenders saw
it more as a necessary evil than as a positive good. The Estates-General of
1614 protested against it, but in vain. Colbert, Maupeou, and Necker in
turn attempted to suppress it, but despite their efforts the market became
more active than ever and tradeable offices continued to be created or
reinstated. In Doyle’s chosen metaphor, the system ‘spawned’ lesser offices
(1996: 10) for which bidders could always be found. It was in the interests
of successive governments, whatever the personal opinions of individual
ministers, to maintain the system in being, not only because they could
not afford to buy out the existing holders but because the demand was
such that the market could bear periodic manipulation of the terms in
favour of the crown. The idea that social prestige could be bought and
sold was as offensive to those for whom patrilineal descent was the sole
legitimate criterion of noble status as it was to those for whom careers in
the service of the state should be open to talented aspirants drawn from
all systacts of origin. But the buyers got what they were paying for and
the sellers ensured that they did. The population of venal roles was unin-
vadable by mutant practices defining roles of the alternative kinds that
serviced the needs of the state in England or Germany or Russia.
The diversion of resources from what might otherwise have been pro-
ductive investment no more condemned venality to extinction than slav-
ery in the Southern United States was condemned to extinction by what
might have been the greater productivity of investment in manufacturing
industry and the employment of formally free wage-labour. But the para-
sitical nature of the practices defining venal roles becomes apparent as
soon as the ‘just what is going on here?’ question is focused on that part of
the system where the pecuniary rewards to the purchasers of offices were
highest. France had no central bank, and the amounts which could be
raised on the government’s behalf by the municipality of Paris or the pro-
vincial estates were limited. So was the scope for a reduction in expend-
iture sufficient to eliminate annual deficits. Subsequent historians have
been no more able than contemporaries were to calculate just how serious
was the state of the royal finances throughout the 1780s. But whatever
the sums involved, the government could not solve its financial problems
Social selection and imposed behaviour 187
without fiscal reform but could not afford fiscal reform until its financial
problems had been solved. Whoever they were, the incumbents of min-
isterial roles had no choice but to go on borrowing on terms sufficiently
attractive to compensate investors for the risk of default, with the conse-
quence that existing borrowings could be serviced only from cash raised
through new ones. Control of the government’s finances was in the hands
of the Farmers-General and the purchasers of the lesser offices of treasurer
or receiver who lent the king’s own money back to him at punitive rates.
There was in theory the option of deliberate default to which French and
other rulers had had recourse in the past. But once default was declared,
or even suspected, the government risked finding itself unable to bor-
row at all. No historian claims that the venality of financial offices made
the French Revolution inevitable. But venality is a textbook example of a
parasitical practice which continued to live off its hosts.
Even the vote of the National Assembly on the night of 3 August 1789
did not drive the practice altogether extinct. In 1866, when Napoleon III
allowed 628 commodity brokers to be dispossessed and compensated at
the average of seven years’ sales paid by the other, freely operating, bro-
kers, ‘The success of the operation sent a wave of panic through the 9,824
notaries, 3,419 avoués, 7,850 process servers, 412 auctioneers, 3,457 regis-
trars, and 60 advocates at the court of cassation, who made up the venal
professions’ (Doyle 1996: 315). The functions performed by the incum-
bents of these roles were, admittedly, very different from those of the
Farmers-General, whose vested interests had retarded what might other-
wise have been the reform of the system of government finance under the
ancien régime. Nor were they made available for purchase in the same way
as those advertised and marketed before the Revolution. The payments
to the government were now made as forced loans defined as ‘caution
money’, and a right of presentation of a chosen successor was substituted
for the traditional practice of resignation in favorem. But the survival of
these offices is a striking illustration of the capacity of parasitical practices
for continuing reproduction even in a much more unfavourable environ-
ment than they had previously enjoyed.
14
There are, however, many sociologists who, while accepting that the future
direction of social evolution is not predictable, still see it as having moved
in the past, and continuing to move in the present, in a general direction
which, despite the occasional reversal or retardation, is plain to see. It
188 The theory of cultural and social selection
was, after all, just such an awareness that inspired much of the agenda of
the nineteenth-century sociologists whose influence lingers to this day.
Whatever their hopes or fears, or their Eurocentric bias (to which I return
in Chapter 5), they all saw themselves as living through an unprecedented
series of fundamental changes about whose general trend there could be
little scope for disagreement. They could hardly fail to observe in the
world around them an evolution out of agricultural towards industrial
modes of production, out of theocratic towards secular modes of persua-
sion, and out of monarchical and oligarchic towards democratic modes
of coercion. The pace of change might be different in some parts of the
world from others. But it was a change from whose impact no society
could much longer escape entirely.
For it, the term ‘modernization’ came to be increasingly widely used
by sociologists writing in the second half of the twentieth century. But
that word has no explanatory value in itself. In the ongoing course of
social evolution, all societies and their modes of production, persuasion,
and coercion are modern at first and ancient in due course. The impli-
cation of an underlying teleology, whether intended or not, is difficult
to avoid: as many commentators pointed out at the time, the writings
of twentieth-century American theorists of ‘modernization’ for whom
their own society was the model which others were destined to follow
mirrored the writings of their Marxist opponents for whom the model
which others were destined to follow was the Soviet Union. If, on the
other hand, no predetermined sequence of any kind is in the author’s
mind, to characterize an institutional change as a transition out of what
is by definition ‘pre-modernity’ is merely to signal a personal criterion by
which the change so characterized is accorded exceptional significance.
To claim, for example, that ‘By general consent, modernization was first
achieved in Great Britain’ (N. Davies 1996: 764) is to say little if anything
more than that British society evolved in unprecedented ways associated
with the so-called ‘Industrial Revolution’; and to claim that the economy
of the Netherlands was the ‘first modern economy’ (de Vries and van der
Woude 1997: 257) is to say little if anything more than that no other soci-
ety had as little of its labour force employed in agriculture as early as the
Netherlands did.
But it does not follow that the concept of modernization has no place
whatever in selectionist sociology. This is because, as the historian C. A.
Bayly puts it in writing about what he calls the ‘birth of the modern world’
between 1870 and 1914, ‘an essential part of being modern is thinking you
are modern’ (2004: 10). Being ‘modern’ is not a state of society. It is a
Social selection and imposed behaviour 189
state of mind, and it has its contribution to make to the explanation of
what is going on whenever the right just-so story is one of meme-practice
co-evolution in which the phenotypic effects of that state of mind play a
part. Institutional innovators are modernizers to the extent that they are
motivated by the desire to replace existing practices and roles with others
which will be perceived as a ‘modernization’, whether they are imitating
or learning from the modernizers of another society or creating the prece-
dent themselves. (Recall from Section 6 the Brazilian abolitionists.) At the
same time, nothing prevents them from carrying both modernizing and
archaizing memes in their heads. In the case of the French Revolution,
its historians are all aware of the revolutionaries’ ‘obsessive yet mysterious
recourse to antiquity’ (Ozouf 1988: 271) whereby the imagined virtues of
Republican Rome were invoked side by side with a millenarian vision of a
future society purged of the injustice and corruption of the past.
‘Ancient’ modernity is not, therefore, a contradiction in terms within
selectionist theory. The outstanding example is ‘ancient’ Athens, whose
articulate members were not only aware, but proud, of its differences both
from other contemporary societies and from what it had itself been in its
recent past. It makes no difference that there is no word in Greek which
corresponds to the English ‘modern’. No reader of Thucydides can doubt
that he thinks of himself as a modern historian and the citizen of a mod-
ern state which has been fighting a modern war of which he himself was
a veteran.37 Equally, no reader of Euripides’ Supplicants can doubt that
he thinks of himself and his audience as members of an ‘equal-voting’
(isopsēphos) polity in which they take pride not – despite invocation of the
legendary Theseus – because it is a revival of a glorious past but because it
has made the holders of political office accountable in an unprecedented
way to the citizens from among whom they have been chosen by election
or lot. The ‘modernity’ of the Athenian mode of production has long been
a topic of scholarly debate: if a ‘modern’ economy requires factory produc-
tion, a stock exchange, double-entry book-keeping, and textbooks of eco-
nomic theory, then the Athenian economy is irredeemably ‘ancient’. But
the Athenian citizens and metics (and on occasion slaves) who lent, bor-
rowed, traded, taxed, mortgaged, discounted, foreclosed, insured, budg-
eted, price-fixed, dowered, embezzled, audited, and litigated as they did
(Cohen 1992; Loomis 1998) cannot have been unaware that in so doing
37
If, moreover, ‘functional differentiation’ is the integrating concept of ‘all modernization theo-
ries irrespective of their different schools of thought and disciplines’ (Schelkle 2000: 92), then
Thucydides’ view that ‘less complicated and less organized = early’ (Hornblower 1991: 10) makes
him a ‘modernization theorist’ who sees his own society as exemplifying precisely that change.
190 The theory of cultural and social selection
they were ‘modernizing’, however they may have put it to themselves, the
practices by which the roles constitutive of their society’s mode of pro-
duction was defined.
Even, however, where self-consciously ‘modernizing’ memes can be
clearly identified in the heads of the carriers of mutant or recombinant
practices, there is the need to establish how far those memes are the driv-
ing force of economic, ideological, or political evolution. When Bayly
says of what he calls the ‘central’ Islamic world and Northern and Eastern
Africa that ‘the aspiration to create modern states spread with bacillus-like
speed in the early nineteenth century’ (2004: 280), he is telling a just-so
story of cultural selection in which memes constitutive of the representa-
tion of the institutions of a ‘modern’ state were being rapidly transmitted
by imitation and learning from the mind of one ruler to another. But it is
sometimes difficult to detect how far cultural rather than social selection
is just what is going on. A test case is a society not referred to by Bayly:
nineteenth-century Afghanistan. Although the soldiers in the army of
Amir Sher Ali wore European-style uniforms ‘as a symbol of modern-
ity’ (Kakar 1979: 95), the practices which gave nineteenth-century Afghan
rulers significantly tighter control over the mode of coercion were not an
expression of the ruler’s self-conscious wish to appear ‘modern’. The two
most effective innovations were the attachment to the roles of generals
rather than provincial governors of control of the means of coercion and
the employment of a network of paid informers reporting directly to the
amir (for which the model was not Britain or Russia but the neighbour-
ing kingdom of Bukhara). Amir ‘Abd al-Rahman Khan further tightened
his hold over his insubordinate and quarrelsome subjects by evicting from
their roles those mullahs who refused to interpret the Quran in such a
way as to bolster his authority. But many rulers at different times and
places have put pressure on the controllers of the means of persuasion to
further their political aims. What looks like imitation or learning from
the example of others in the cause of ‘being modern’ may turn out to be
another just-so story of convergent social evolution, and some means has
to be found for discovering which it is.
15
If there is any unilinear trend which social evolution can be shown to have
followed across the globe in the millennia since the transition from culture
to society, it is ‘globalization’ itself. But this means no more than that the
geographical diffusion of economic, ideological, and political practices
Social selection and imposed behaviour 191
from place to place has reached the point that autonomous social evolu-
tion has become a thing of the past. For more than ten thousand years,
there have been societies evolving outside the reach of any possible inva-
sion by exogenous practices. Only in the middle of the twentieth century
CE did two Australian prospectors stumble, to mutual astonishment, on
the hitherto unknown societies of the New Guinea Highlands. By now,
it has ceased to be possible for there to be cases of convergent evolution of
the kind which produced the resemblances between the modes of produc-
tion, persuasion, and coercion of Hammurabi’s Babylonia and those of
Anglo-Saxon England, or between those of the Aztecs and those of their
Spanish conquerors. Even so, it is possible to envisage a future in which
natural and man-made catastrophes combine to isolate the few surviv-
ing human societies from one another and to extinguish the techniques
which would bring them back into contact. But for the time being both
memes and practices can travel anywhere in the world. We have arrived
at the end of a long and fascinating story of increasing mutual awareness
among the hitherto separate human populations of the globe. Whether,
in hindsight, that story can be divided into evolutionary stages or peri-
ods which are anything more than extrapolations by elective affinity from
presuppositions of the narrator’s own is a question to be briefly addressed
in the chapter which follows.
Ch apter 5
S t or i e s of s t or i e s
1
To insist, in the same breath as proclaiming the merits of a neo-Darwinian
approach to comparative sociology, that there is no master narrative of
human history is to invite the rejoinder that selectionist theory itself is
an attempt to impose a master narrative of its own (Landau 1991). But
although selectionist theory does bring with it a multitude of narratives,
they are narratives of a kind which leave room for many others compatible
with them. Since the agenda of comparative sociology is set by the observed
resemblances and differences between human cultures and societies, every
explanation of how a culture or society has evolved into being of the kind
that it is will be a story about winning and losing memes and practices. But
the just-so stories which succeed in so doing do not have authorial ‘plots’ in
the way that ‘good’ stories, whether factual or fictional, are required to do
(Welleman 2003). They have no conclusive endings, no contrived denoue-
ments, no privileged vantage-points, no preferential fields of thought and
action, no paeans to virtue, and no denunciations of wickedness. They are
stories about what has happened to go on in one or another part of the ever-
branching evolutionary tree (or ever-flowing braided stream) out of which
there continue to emerge new and distinctive cultural and social behaviour-
patterns which are neither inexplicable nor pre-ordained.
That does not mean that selectionist stories may not be, in their per-
locutionary effects, as exciting or dismaying or encouraging as if they
had been deliberately constructed with those effects in view. Although,
from a selectionist viewpoint, the French Revolution can be seen to have
been in a double sense unnecessary, that does not make any less compel-
lingly readable the sequences of dramatic events narrated by Carlyle or
Michelet or their many imitators and successors. Nor are stories about the
192
Selectionist theory as narrative history 193
extinction of memes and practices by which seemingly enduring cultures
and societies were previously being sustained any less dramatic than stor-
ies about the lives of their carriers. The felling of the last palm tree left on
Easter Island was an event which, however dispassionately told, can leave
few readers wholly unmoved, even though nothing is known about the
person who felled it. Selectionist stories can be as effective as any con-
ventional tale of triumphs and disasters, unexpected reversals of fortune,
astonishing coincidences, and individual or collective joys and sorrows in
generating an emotional response among even those readers who are most
firmly committed to the proposition that history-writing is, or should be,
more a science than an art. Since, indeed, chance plays as large a part as it
does in cultural and social evolution alike, selectionist just-so stories have
a capacity to elicit responses of just the kind which both tragic and comic
dramatists and novelists actively exploit in the plots which they concoct
for that purpose.
But a right just-so story, if such it is, is right whatever reactions it pro-
vokes. Since selectionist stories are always stories of winners and losers,
it is inevitable that they will be differently told by narrators who side
with the one or the other. One temptation to which this can lead is the
unblushing complacency nicely caricatured by the cultural historian
Ralph Samuel in his remarks about the schoolbook histories of British
imperialism (1998: 86): ‘Incidents double in the character of parables,
pointing up lessons in manners and morals. Intrepid explorers push into
the unknown; martyr-missionaries, like Bishop Harrington, brave the
wrath of cannibals; besieged soldiers, surrounded by howling dervishes,
display tremendous pluck.’ Or in the rather different phrasing of an emi-
nent military historian, ‘The empires were won; and they were won, in the
overwhelming majority of cases, by small groups of men exceptional in
their energy, intelligence, endurance, and physical courage, even if they
had all the power of industrial civilization behind them’ (Howard 1984:
34). Inevitably, storytelling like this provokes anti-stories in which the
heroes become villains and the villains heroes and counterfactual hypoth-
eses are advanced for the overt purpose of persuading the reader that
the winners did not deserve their success and ought not to be admired
for achieving it. But the difference made by British imperialism to what
would otherwise have been the evolution of the cultures and societies
which were for a short time constituent parts of the British Empire is the
same whatever different moral judgements different readers may be dis-
posed to pass on the carriers of the memes and practices which made the
difference that they did.
194 The theory of cultural and social selection
The charge of Eurocentrism frequently levelled against the best-known
sociologists of the nineteenth and twentieth centuries is difficult to rebut.
They did underestimate the extent of autonomous evolution in non-
European cultures and societies and the extent to which the evolution
of European cultures and societies had been influenced by innovations
diffused from the Eastern to the Western parts of the Eurasian landmass.
But they were not mistaken in their awareness that there was something
that needed to be explained in the inter-societal imbalances of economic,
ideological, and political power which they saw in the world around them.
It was not as if they saw African colonies being founded on the coasts of
North America, or Chinese armies fighting for control of the territory
of France, or Islamic mosques replacing the cathedrals and churches of
Italy, or the products of Polynesian factories being imported into Britain.
The so-called ‘Industrial Revolution’, however defined by reference to
new energy sources, new technology, or new methods of production, had
taken place in Britain and not, however nearly it might have done, in
China. But then the human diaspora began in Africa, the domestication
of plants and animals began in Southwest Asia, and primary state forma-
tion began in Mesopotamia. Those are the places where those things hap-
pened first to happen and where the just-so stories of how they did have
therefore to be set. Nor is it as if Eurocentrics were under the illusion that
the hegemony of the West was unchallengeable. It was, after all, Kipling
who wrote the poem ‘Recessional’ and Macaulay who envisaged a New
Zealander contemplating the ruins of St Paul’s. Tocqueville turned out
to be remarkably prescient about a future in which America and Russia
would one day divide the world between them. But he did not suggest
that at that point history would come to an end.
2
Just as the long quarrel between ‘Individualists’ and ‘Holists’ to which I
referred in Section 15 of Chapter 1 can (for sociologists, if not for philoso-
phers) be laid to rest within selectionist theory, so can that between the
partisans of ‘Idiographic’ and ‘Nomothetic’ explanation. The supposed
contradiction between ‘history’ and ‘science’ has its ‘evolutionary solution’
(Blute 1997): while every sequence of heritable variation and competitive
selection which creates novel biological, cultural, or social forms is histori-
cally unique, it would not be a sequence of causes and effects if there were
no lawlike regularities in the workings of the world at an underlying level.
Often enough, these can be taken for granted, just as they can in the just-so
Selectionist theory as narrative history 195
stories of natural science: a geologist who solves the puzzle of a hitherto
unexplained plateau by demonstrating that it is the result of sub-aerial
denudation down to a sea-level below that of the present day has no need
to invoke a physical law, any more than a sociologist needs to invoke a psy-
chological one when explaining an upsurge in social mobility by pointing
to the changes in the economy which have led to an increase in the ratio
of non-manual to manual occupational roles. But sometimes, it may be
by direct appeal to a lawlike regularity that the puzzle is solved: if it turns
out that the members of a group or community whose behaviour is excep-
tionally altruistic are (or believe themselves to be) genetically related then
Hamilton’s Rule will do very well. Explanations of resemblances between
cultures or societies in terms of convergent evolution imply an underlying
generalization of the form ‘similar environment, similar response’, whereas
explanations in terms of homologous descent imply that environmental
similarity is less important than the details of the local sequence of vertical
transmission by cultural or social inheritance. But the relative evolutionary
significance of analogy and homology in one or another case is not some-
thing which can usefully be debated a priori.
Once, however, a selectionist hypothesis has been shown to fit the
relevant evidence as well or better than its competitors, the just-so story
accepted as the right one may become a candidate for inclusion among the
‘sooner-or-later’ stories, which can be as persuasive, but also as problem-
atic, for evolutionary biologists as they are for evolutionary sociologists.
Was the human species bound sooner or later to emerge on the planet
where it did, or did it do so only because of a wholly extraordinary series
of mutations of which any one might never have occurred and human
beings never have evolved at all? Is the story of the evolution of species
a story of ‘a myriad of possible evolutionary pathways, all dogged by the
twists and turns of historical circumstances’ which will therefore ‘end up
with wildly different alternative worlds’, or is it a story in which ‘the con-
straints we see on evolution suggest that underlying the apparent riot of
forms there is an interesting predictability’ (Conway Morris 1998: 139)? In
cultural and social evolution, although the ‘riot of forms’ is – to date –
less diverse than that of the billion or more biological species which have
evolved since life on earth began, it is quite diverse enough to invalidate
any would-be lawlike generalizations put forward to cover them all. But
not only is the range of possible forms far from unlimited; so is the range
of possible sequences along which one succeeds another.
Conway Morris gives as an example the evolution of the whales (1998:
202). From the perspective of the Cambrian, they are no more likely
196 The theory of cultural and social selection
than hundreds of other possibilities, but ‘the evolution of some sort of
fast, ocean-going animal that sieves sea-water for food is probably very
likely and perhaps almost inevitable’. Given sufficient knowledge of
the previous history and the ongoing constraints, it becomes increas-
ingly plausible to suppose that sooner or later certain cultural and social
traits will appear in one or another culture or society, just as certain
biological traits will sooner or later appear in one or another species.
The stories of the domestication of plants and animals, or of urbaniza-
tion, or of state formation, or of industrialization, can, like the story of
the whales, be said to be about something which after a certain point
was ‘waiting to happen’ or ‘just around the corner’ or ‘only a matter
of time’ without any surreptitious reintroduction of illegitimate teleo-
logical presuppositions.
This claim – or, if you will, concession – may seem to sit uncomfort-
ably with the insistence that sociology is not and never will be a predictive
science, any more than biology will. But outcomes which had once been
impossible to imagine, let alone to predict, can come in due course to be
so increasingly probable that by the end of the sequence the result does
come to be seen by contemporary observers to be inevitable. By the time
that mineral resources, new technology, and available human labour had
come together in the British environment, a ‘factory system’ of industrial
production was as foreseeable by those who deplored it as by those who
welcomed it. But it had not been foreseeable when Adam Smith published
The Wealth of Nations in 1776, any more than when Alfred Marshall pub-
lished Industry and Trade in 1919, with its story of ‘the business point of
view’ as what he called a ‘chief feature of economic evolution’, he or any-
one else could have foreseen the mode of production which was by then
evolving in the Soviet Union.
3
The history of history-writing can itself be told as a selectionist just-so
story, as chronicles, annals, and folktales are displaced by narratives
based on eyewitness testimony or documentary records from which
reported outcomes can convincingly be linked to antecedent causes. But
they are not mutually exclusive. Selectionist just-so stories about human
behaviour-patterns do not put all the chroniclers, annalists, and folk-
lorists out of business or deny their ability to construct stories capable
of being shown to be true. Nor are selectionist sociologists in competi-
tion with other sociologists whose purpose is something other than the
Selectionist theory as narrative history 197
explanation of how the collective behaviour-patterns of the populations
of different cultures and societies have come to be what they are.
Sociologists may want only to use reports of the behaviour-patterns of
people very different from themselves to convey to their readers what it
feels like to be one of ‘them’. (I return yet again to the relation between
description and explanation later in this chapter.) Others may want only
to collect data from which to estimate frequencies, or calculate trends.
Others may want only to offer general comment on some aspect of the
human condition, or to construct concepts with which to reformulate
empirical generalizations about collective human behaviour in more
abstract terms. And yet others may want above all to put before their
readers observations of behaviour which will cause their readers to share
the emotions of approval or disapproval which those observations have
evoked in themselves.
There are, however, two kinds of story-telling, in addition to the
mythical and the providential, which selectionist theory does categori-
cally disavow. The first is the story-telling of self-styled ‘no-nonsense’
historians for whom ‘the record speaks for itself ’ and explanation is a
matter only of ascertaining the motives of the particular individuals
whose decisions influenced the course of subsequent events and the
consequences which followed from those decisions. This type of nar-
rative is well illustrated in, for example, the writings of Geoffrey Elton
(Skinner 2002: Ch. 2). To it, the selectionist response is that narra-
tives that are limited in this way, well-validated though they may be,
ignore the questions which they themselves raise about what makes the
decision-makers’ decisions effective in altering the course of cultural
and social evolution from what it would otherwise have been. The sec-
ond is story-telling of the kind in which the narrator illustrates a pre-
conceived theme about the course of cultural and social history without
reference to the relevant comparisons. To it, the selectionist response is
that just as synchronic correlation is no proof of underlying causality, so
is diachronic sequence no proof of underlying process. This type of nar-
rative is well illustrated by Norbert Elias’s The Civilizing Process. Elias
tells a well-documented and convincing story of sequential changes
in European manners and mores over a chosen period. But he ignores
similar changes in non-European cultures, including the courtly sub-
culture of Japan; and as has been pointed out ever since its original
publication in Germany in 1939, the assumption implicit in his chosen
title was being falsified in the very society in which it was being written
during the very years in which he was writing it.
198 The theory of cultural and social selection
Ag e s a n d s tag e s
4
As I lamented in Section 8 of Chapter 4, sociology has as yet no taxonomy
of either cultures or societies with anything approaching an adequate
Linnaean-cum-Darwinian rationale. But no sociologist will deny that
there is a demonstrable sequence of modes of production, persuasion, and
coercion of different kinds, just as there are demonstrable institutional
resemblances across societies widely separated in both time and place.
Symphony orchestras could no more have preceded chanting or singing
in cultural evolution, or parliamentary democracy big-men or lineage
heads in social evolution, than elephants could have preceded bacteria in
biological evolution. Imagine what would happen if an archaeologist dis-
covered a series of marks on the wall of one of the Lascaux caves alongside
the depictions of animals and people which turned out to be the notation
of a piece of music which could have been written by Mozart! The whole
evolutionary paradigm would fall apart.
On the other hand, the appearance of similar institutions at differ-
ent times and places does face selectionist sociologists with the need to
establish which similarities between roles and practices are adaptations
whose common form is explained by their common function as opposed
to either exaptations which have a reproductive advantage for a different
reason from that for which they were originally selected or by-products
with no effect at all on the subsequent course of the society’s evolution.
Which resemblances matter, and which don’t? To base a taxonomy on
a resemblance which has no evolutionary rationale is to invite the fam-
ously disobliging comment made by Edmund Leach (1961: 4) when he
likened ill-conceived classifications by unthinking anthropologists to the
idea that identifying a class of ‘blue butterflies’ could contribute to the
understanding of the anatomical structure of lepidoptera.
Consider the sociological category ‘city-state’. A large number are doc-
umented in detail in widely separate parts of the world, from the Near
East to Greece, Italy, Germany, Russia, Asia (including Southeast Asia),
Mesoamerica, and Africa both north and south of the Sahara. They can
all be compared with one another by reference to an ideal-typical list
of family resemblances (Griffeth and Thomas 1981; Hansen 2000). But
they have in common only that, by definition, their rulers all exercise
control over a surrounding territory and/or associated trading network
from a single urban centre. What is gained in the understanding of social
Selectionist theory as narrative history 199
evolution by putting Sparta (which in any case lacked an urban centre)
together with Florence, Amalfi, the German Reichstädte, Novgorod, Kano,
Tenochitlan, and Monaco? Their family resemblances and differences in
modes of production, persuasion, and coercion apply no less to societies
which fall outside the definition, including among others nomadic states,
oasis states, and the Swiss cantons which swore mutual allegiance in the
Rütli meadow in 1293. If, on the other hand, the story being told is about
the consequences for a society’s evolutionary trajectory of the dominance
of one disproportionately influential urban centre, then the dominance of
London over England between 1650 and 1750 (Wrigley 1987: Ch. 6) is as
relevant a case as Sparta’s over Lakonia or Florence’s over Tuscany. Or if
the story is about the trade-offs negotiated by the rulers of small states in
an environment dominated by large ones, then the independent existence
of Switzerland is as relevant as that of Singapore.
This does not make it a mistake to speak, in a familiar example, of
both an ‘Age of the Polis’ in ‘ancient’ Greece and an ‘Age of the Statto-
Città’ in ‘medieval’ Italy. The resemblances in their combination of
hostility between, and conflict within, a multitude of separate societies
seeking to extend their control from a single centre over their surround-
ing territories include similar inter-systactic relationships which affected
their modes of production, persuasion, and coercion alike. In both their
modes of production, landed interests competed with mercantile; in both
their modes of persuasion, status-groups defined by birth competed with
status-groups defined by personal achievement; and in both their modes
of coercion, would-be monopolists of the use of force competed with
assemblies, committees, and councils. But there are differences as strik-
ing as the resemblances. There is no equivalent in Italy to the ‘hoplite
revolution’ in Greece; there is no parallel in the age of the poleis to the
roles of bishops and clergy in medieval Italy; and although the economic
practices and roles of Athens, in particular, were less dissimilar to those of
medieval Italy than some historians of Greece have maintained, there are
no roles in any Greek polis equivalent to those of the banking dynasties
like the Bardi and Peruzzi. Nor, in any case, was the ‘city-state’ universal
in either ‘ancient’ Greece or ‘medieval’ Italy: the poleis co-existed with
powerful tribal ethnē like Thessaly, just as the statti-città coexisted with
municipal communes which neither achieved, nor in some cases wished
for, independence from the principalities embracing them.
From a selectionist viewpoint, the interesting parallel is not the con-
junction of urban centre with rural periphery, but the re-emergence
among the Italian city-states of the determination to reduce, so far as
200 The theory of cultural and social selection
possible, the likelihood of an evolution out of oligarchy into tyranny sim-
ilar to that documented among the poleis. The limits imposed by citizen
bodies on the tenure of office by a podestà or incumbent of some other
quasi-consular and therefore potentially monarchical role is at once remi-
niscent of similar limitations imposed by the citizens of some of the poleis
on their magistrates (kosmoi, archontes, or prytaneis).1 In neither case did
there evolve institutions which could be labelled ‘democratic’ as that term
is now understood. A Greek dēmos was no more all-inclusive, let alone
egalitarian, than an Italian popolo. The differences were between prac-
tices constitutive of more and less restricted forms of an oligarchic mode
of coercion. Sometimes a larger and sometimes a smaller proportion of
citizens – always male – were entitled to participate in political decision-
making, and leading families had sometimes more and sometimes less
influence over the selection of political office-holders and the decisions
that they made. The peasantry of the Italian contadi had no more say in
those decisions than the Spartan Helots (or Syracusan Killyrioi or Argive
Gymnētes or Locrian Woikiatai or Cretan Apetairoi). But this makes it all
the easier to see that what was going on was the selection under similar
environmental pressures of similar memes and practices. It was not the
result of convergent evolution alone. Not only did the Italian constitu-
tion-makers learn from and imitate one another, just as the nomotheteis of
Greece had done, but they were explicitly conscious of the Roman past,
in particular, as a precedent.2 But by the time that, in the early tenth
century, the institutions of the Italian state had been weakened beyond
the point of recovery, the cities were, to use that way of speaking, ‘wait-
ing’ to become – sooner rather than later – the centres in which local
elites, including resident bishops, would be the carriers of the practices
defining political roles. In Greece, it took much longer after the fall of the
Mycenaean palaces before the world of scattered villages, depleted popu-
lations, small-scale trading, patriarchal retinues, guest-friendships, and
local or tribal chieftains and warbands evolved into the world of the poleis
with their citizen armies and their formal diplomatic relations with one
another. But there is no covert teleological implication in pointing out the
resemblances which there were between the evolution of the oligarchic
1
The similarities are sometimes so close that, for example, the three ‘heads’ drawn from a commit-
tee of forty drawn from a ‘Great Council’ of 300 or 400 in thirteenth-century Venice (Lane 1973:
100) match the equivalent three drawn from a committee of fifteen drawn from a council of 600
in Massilia in the sixth century BCE (Strabo 179).
2
Venice, which might have been expected to be among the first, was in fact ‘late in developing an
interest in Roman republicanism as a political model’ (Brown 1991: 102). But Venice had no role
equivalent to the Roman ‘tribune of the people’.
Selectionist theory as narrative history 201
practices and roles of the age of the polis and, a millennium later, the
e volution of those of the age of the Italian stati-città.
These ‘ages’, however, were not stages in either a cycle from monarchy
to oligarchy to democracy and back to monarchy again or a linear pro-
gression from tribes to states to empires. Nor was there anything about
the roles constitutive of the institutions of the city-states which destined
them either to abandon an oligarchic for a monarchic mode of coercion
or to surrender their sovereignty to some larger territorial entity. Venice
did not merely retain its independence but acquired – for a time – an
empire, as Rome had done. It is true that twice during the fourteenth cen-
tury Venice only narrowly escaped signoria. It is also true that the failure
of the Roman oligarchy to attach the legionaries’ loyalty to the Senatus
Populusque Romanus, whose acronym they carried on their standards,
rather than to their individual commanders, led to a protracted period
of civil war which was resolved only by an evolution out of oligarchy
into monarchy. But in neither case can the outcome be explained by the
concentration of power in a single urban centre. Conversely, despite the
inability of the Greek poleis to unite against Macedon as they had done
(with some exceptions) against Persia, the Athenians might have won the
Peleponnesian War and might then have assembled a pan-Hellenic army
to launch an invasion of Persia, as Isocrates, for one, explicitly recom-
mended. The selectionist just-so story is no more (but also no less) than
a story of a similar sequence of meme-practice co-evolution which hap-
pened at two separate times and places and happened to involve a number
of relatively small urban centres controlling their adjoining territories and
fighting regularly but inconclusively with one another.
5
Yet there are some ‘ages’ which were stages of a kind. In the evolution
of both technology and art there are some which, like for example the
‘Bronze Age’ or the ‘Age of the Baroque’, can uncontentiously be dis-
tinguished in terms of what can be seen with hindsight to have been a
coherent transitional sequence of variation and selection between what
preceded and what followed them. For the purposes of comparative soci-
ology, however, to report them in those terms contributes to the explan-
ation of the ongoing course of cultural and social evolution only to the
extent that they specify qualitative population-level changes in shared
beliefs or attitudes in the collective mind in the one case, and in roles
constitutive of different modes of production, persuasion, or coercion in
202 The theory of cultural and social selection
the other. The boundaries can seldom be marked by a single date, even
where the change has come about through a punctuated equilibrium
rather than an incremental mutation of memes or practices; and even if
they can, whenever an age is claimed to be a stage, the Fallacy of Pre-
emptive Periodization needs to be no less carefully guarded against than
the Fallacy of Elective Affinity.
The Fallacy of Pre-emptive Periodization is clearly exemplified by the
conventional division of human history into ‘ancient’, ‘medieval’, and
‘modern’ which is owed to an otherwise obscure German scholar of the
late seventeenth century CE (Burrow 2007: 416). Historians seem as una-
ble to discard the ‘middle ages’ from their standard vocabulary as they
are the ‘feudalism’ which is regularly associated with them, and many are
as happy to apply it to the histories of the Near and Far East as to that
of Europe. But in relation to the story of human evolution to date, it is
as parochial as it is misleading (Smail 2008). In cultural evolution, the
big transitions are first, the emergence of language and symbolism, and
second, the emergence of literacy. In social evolution, the big transitions
are first, the emergence of the practices and roles constitutive of states, and
second, the institutional outcome of the break-out – for as long as it lasts
among the populations to which it applies – from the Malthusian con-
straint of demographic growth on food supply (Clark 2007). Odd though
it would sound to call the period between the first and second of these
transitions ‘medieval’, it would have as much of a rationale as calling the
period between the fall of the Roman Empire and the so-called European
Renaissance ‘medieval’. I have already, in Section 14 of Chapter 4,
remarked that ‘ancient’ Athens was, for Thucydides and others of his con-
temporaries, ‘modern’ in relation to what preceded it; and although, in
previous chapters, I have referred in the conventional fashion to the ‘early
modern’ period of English society, the term is a twentieth-century coin-
age which has, for comparative sociology, no explanatory value whatever.
The Fallacy of Elective Affinity is clearly exemplified in the concept
of the so-called ‘Axial Age’ in the story of human cultural evolution
(Eisenstadt 1982; Bellah 2005). For Karl Jaspers, who coined the phrase,
it was in the same pre-Christian centuries that there emerged in China,
India, Iran, Israel, and Greece the first self-conscious exercises in criti-
cal reflection on the human condition by reference to a transcendental
standard arrived at by theoretical speculation. The originality of the doc-
trines expounded by such remarkable innovators as Zoroaster, Confucius,
Buddha, the Hebrew prophets, or the Greek philosophers is not in doubt.
But to bring them all together, as Jaspers does, as belonging to a single
Selectionist theory as narrative history 203
distinctive ‘age’ in which, as he put it, ‘Man, as we know him today, came
into being’ does nothing to explain the very different evolutionary tra-
jectories which account for their varying degrees of reproductive success.
There is, no doubt, something in common between the environments in
which there emerge, then and since, ‘new models of reality, either mystic
ally or prophetically or rationally apprehended’ which are ‘propounded as
a criticism of, and an alternative to, the prevailing models’ (Momigliano
1975: 9). But there is no just-so story of convergent evolution like that
of the evil eye, or of homologous descent like that of the Austronesian
languages as they spread out into the Pacific, or of lateral diffusion like
that of the techno-memes carried by itinerant craftsmen from one society
to another. There are resemblances, but also very large differences, not
only in the memes constitutive of their teachings and the social environ-
ments into which they introduced them but in their extended phenotypic
effects on the behaviour of the members of the different populations. The
beliefs of the Greek philosophers are not at all the same as those of the
Hebrew prophets with their uniquely disorienting experience of desecra-
tion and exile; the Analects of Confucius are as different from the Platonic
dialogues or the Zoroastrian Avesta as they are from the Books of Hosea
or Amos or Ezekiel or deutero-Isaiah; the Buddhist attitude to political
power and those who wield it is no more that of Confucius than that of
either the Greek philosophers or the Hebrew prophets; the varna model
of the four orders is unique to Vedic India; and so on. And what about
the culture of ‘ancient’ Rome in the ‘Axial Age’? As I argued already in
Section 12 of Chapter 3, no tension between worldly power and a tran-
scendental standard by which its possession and use was to be criticized
was ever conceptualized in Roman culture. Nor, for all the long-lasting
influence of Plato’s Republic (which took the continuance of both slavery
and warfare for granted) did there evolve anywhere in Hellenic culture
any conception of human rights possessed independently of any given
social order. Perhaps the environments of the poleis of Greece and the
‘warring states’ of China were similarly favourable to critical reflection
on politics; perhaps the worship of Allah in Islam can be traced back by
homologous descent to the worship of Yahweh in Israel; perhaps there
was some diffusion of theoretical speculation across societal frontiers in
which Achaemenid Persia played a part. But Jaspers’ ‘Axial Age’ is a cat-
egory derived from neo-Hegelian philosophical anthropology, not neo-
Darwinian comparative sociology. It is defined by elective affinities which
are the product more of Jaspers’ own existential philosophy than of the
selective affinities in the world whose distinctive cultures and societies
204 The theory of cultural and social selection
evolved as they did. Well-documented stories can always be narrated in
such terms as ‘the growth of human self-understanding’ or ‘the cultural
crystallization of conceptual change’ or ‘the triumph of reason’. But they
are not stories which explain why the mutant memes referred to in these
terms out-competed their rivals where and when they did. Nor are they
stories in which memes constitutive of critical reflection about the human
condition drive their mythical competitors to extinction: Zoroaster’s most
influential cultural legacy was the eschatological myth of the ultimate
victory of the good God (Ahura Mazda) over the wicked Satan (Angra
Mainyu).
6
Just as, however, you are ‘modern’ if you think of yourself as modern, so
are you living in a distinctive ‘age’ if that is what, descriptively speaking,
it feels like to you. As I remarked in Section 12 of the Prologue, mid-
Victorians who saw themselves as living in an ‘Age of Improvement’ were
living in one to the extent that, by their personal standards, that was what
it was. Likewise, the Christians who saw themselves as living in a medium
aevum between the First and Second Comings of Christ were living in a
‘middle age’ to the extent that they believed that the Second Coming was
an event waiting – sooner rather than later – to occur. The difference is
that you can’t see yourself as living in what Karl Jaspers will subsequently
christen an ‘Axial’ any more than in a ‘pre-Christian’, or ‘pre-Renaissance’,
or ‘pre-Industrial’ age. You can, perhaps, see yourself as living in a
‘golden’ age if you are convinced that one day in the future your descend-
ants will look back nostalgically on the period of your lifetime as ‘golden’.
But there remains the perennial disjunction between explanation and
description from Section 5 of the Prologue and Section 12 of Chapter 2.
Yes, you are privileged over the academic observer in your sense of what
your experience feels like for you. Introspection, as I said in Section 11 of
the Prologue, can tell you quite a lot about yourself, however vehemently
the Behaviourists may deny it. But the academic observer is privileged
over you in explaining why your experience is what it is. Your own explan-
ation may be as good as, or better than, the academic observer’s. But if it
is, it is because you are yourself looking at it from an academic obser-
ver’s viewpoint. Your subjective experience is relevant to the explanation
of your cultural and social behaviour only where it is because of the terms
in which you describe it to yourself that your behaviour is different from
what it would otherwise be. You may be reflecting as deeply as you know
Selectionist theory as narrative history 205
how about your sense of living in an Age of Improvement, or Barbarism,
or the Common Man, or Warring States, or Chivalry, or Imperialism, or
in a Romantic Age, or a Dark Age, or an Age of Faith. But it doesn’t fol-
low that you know how and why your and your contemporaries’ behav-
iour is being influenced by that feeling, let alone that you have traced the
selection of the memes and practices which distinguish one such age from
another and identified the environmental pressures which have caused
them to be selected instead of their competitors.
T e l l i ng i t l i k e i t f e lt l i k e
7
It is not only because human informants can be as unashamedly devious
and wilfully untruthful as they often are that sociologists may find that
they have misled their readers. It is also because even the most candid and
self-aware informants can be as seriously mistaken as they often are about
what is going on inside their heads. If, to take a present-day example,
you live in a ‘democratic’ society and take part in its periodic elections
to governmental roles, you may think that you know better than anyone
else, and be ready to tell the sociologist who is questioning you on your
doorstep, that in a recent election you voted for candidate A because of
your personal antipathy to candidate B despite your upbringing in a fam-
ily which supported the party of candidate A. You also know better than
anyone else what it feels like to feel the antipathy that you do. But are you
sure that you know why you do, and why you do to that degree? In any
case, do you know why you and others of the electorate bother to vote at
all when the chance that your vote will decide the outcome is so small as
to be negligible? Are you aware that you’re more likely to vote if at the last
election you voted for the winning candidate or didn’t turn out to vote
but supported the losing candidate (Kanazawa 2000)? A well-informed
sociologist is much better placed than you are to explain voter turnout
at population level and the extent to which it is a function of expres-
sive motivation, or tactical calculation, or a sense of civic duty, or ‘weak
a ltruism’ – that is, utility derived from support for a candidate who will
help to implement programs that differentially help others with whom
you, the voter, feel sympathy (Jankowski 2007).
Some empirically minded readers may want to argue at this point that
individual agents’ descriptions of their own experiences are wholly irrele-
vant to the agenda of selectionist sociology. Interesting as it may be to
206 The theory of cultural and social selection
be told what it felt like to be one of ‘them’, what difference does it make
to the correct identification of the selective pressures acting at popula-
tion level on the phenotypic effects of the genes, memes, and practices
which are the protagonists in just-so stories of biological, cultural, and
social evolution? There will always be the very different way of answer-
ing the ‘just what is going on here?’ question which generates the stories
of immediate personal experience of the kind for which the sociolo-
gist Erving Goffman is rightly celebrated – stories about ‘impression
management’, ‘breaking frame’, and the whole repertory of behavioural
devices by which we all, whether consciously or unconsciously, deploy
the manoeuvres and subterfuges which are an inescapable part of human
interaction at the day-to-day, face-to-face level. But however revealing
about other people’s inner lives such stories may be, and however enter-
taining they may be to read, what do they contribute to the explanation
of how the cultures and societies within which they are going on have
evolved into being what they are?
The answer is that to exclude altogether from the literature of com-
parative sociology the subjective experiences of the carriers of the bundles
of heritably variable and competitively selected information which make
cultures and societies into what they are would be to treat those carri-
ers as if they were not, after all, psychologically different from Frans de
Waal’s chimpanzees. To say that a human culture or society is of a cer-
tain kind is to say among other things that representative members of its
constituent sub-cultures and systacts are likely to have certain subjective
experiences which are likely to influence their behaviour in certain ways.
Description of these experiences calls for skills in the choice of words and
phrases, and the construction of metaphors and similes, which are obvi-
ously different from the skills taught in courses on mathematical model-
ling or statistical methods. But both depend equally on the evidence: the
academic observer is no more entitled to tell you that you feel things that
you don’t (however little you can explain them) than you are to tell the
academic observer that your culture or society has come to be what it is
through a sequence of causes and effects that it hasn’t (however convinced
you are that it has).
8
For all the demonstrable ability of some historians, anthropologists, and
sociologists (and many novelists) to describe just what ‘their’ experi-
ences felt like to ‘them’, there are too many alternative perspectives from
Selectionist theory as narrative history 207
which cultures and societies are viewed by their members, and too many
relevant details out of which too many stories could be presented to the
reader, for sociological descriptions by different observers to be ranked
above or below one another on a single ordinal scale of authenticity and
representativeness. But authentic and representative descriptions can still
be distinguished from inauthentic and unrepresentative ones. The practi-
cal difficulty is that historians cannot interrogate the dead, and it is rela-
tively seldom that the anthropologists on whose testimony comparative
sociologists have to rely have conducted extended conversations with the
natives designed to test as rigorously as can be done the authenticity of
the descriptions to be presented to the readers back home. Some, like for
example the Colombian anthropologists Gudeman and Rivera (1990: 4, 8),
are at pains to make clear that they conduct their fieldwork as ‘a per-
petual discussion’, and to assure their readers that ‘Almost everything we
describe was put to the “test” of conversation by the rural folk.’ But notice
the quotation-marks round the word ‘test’. The danger is that such con-
versation will distort the natives’ subjective experience through the very
act of exploring them. Far from helping the anthropologists to arrive at
a better explanation of why the natives behave as they do, it may sim-
ply mislead them. What is needed is for the anthropologists to have con-
ducted extended conversations with knowledgeable native informants who
are also thoroughly familiar with the anthropologist’s own culture and
can therefore arrive at a shared understanding of the difference between
authenticating the description and validating the explanation of the dis-
tinctive behaviour-patterns which the anthropologists have observed.
An example which illustrates this more clearly than any other I have
been able to find is the anthropologist Caroline Humphrey’s account
(Humphrey 1996) of the lengthy exchanges which she had about Daur
Mongolian Shamanism with her informant and friend Urgunge Onon,
whose upbringing in a remote Mongolian village was followed by cap-
ture by bandits, schooling in Qiqihar, service in the Mongolian freedom
movement, graduation from a Japanese university, flight to the Chinese
Nationalists, and emigration to the United States at the age of twenty-eight
in 1948. Caroline talks of Urgunge as wanting to ‘explain’ Shamanism, but
it is clear from her text that he is concerned neither with the historical
origins of Mongolian Shamanism, nor with the reasons for the inherit-
ance and selection of its constitutive memes in competition with its rivals,
but with getting Caroline to understand what Daur Shamanism means
to him so that she can use her anthropological expertise to describe it to
us. The exercise is conducted on the assumption that four presuppositions
208 The theory of cultural and social selection
are shared by Urgunge, Caroline, and Caroline’s prospective readers.
First, there is a clear distinction between the literal and metaphorical: in
Caroline’s example, we, she, and Urgunge all know that when an English
newspaper carries the headline ‘Six Cheat Death in Freak Accident’, it
would be a mistake to infer that English people think of Death as a per-
son who can be literally cheated. Second, there is a range of basic human
responses shared between his, her, and all other cultures: Caroline has
no difficulty in recognizing in shamans’ songs an intuitive understand-
ing of the psychology of human relationships. Third, belief is not a clear-
cut, hard-and-fast matter: Urgunge remarks on the ambiguity with which
some Daur beliefs about shamans are held and expresses uncertainty about
what his ancestors ‘really’ believed. Fourth, the attitudes and beliefs of dif-
ferent members of the same culture are affected by their personal traits:
Caroline sees, for example, that Urgunge has, as a man, failed to recol-
lect the part that, to her knowledge, women sometimes played in Daur
Shamanism. Neither Urgunge nor Caroline expect Caroline’s readers to
be persuaded by her account to come to share Urgunge’s world-view. They
only want us to see it as intelligible and coherent and to understand how
Shamanism can seem to Urgunge, although not to Caroline or to us, to be
‘the best orientation for living in our times’.
The way in which they set about it is by suggesting parallels between
‘their’ beliefs and attitudes and ‘ours’, as when Caroline sees shamans’
dress as sharing the historical resonances of a priest’s vestments as well
as objectivizing shamans’ power. Her readers, indeed, are likely to find
that such parallels occur to them almost automatically. When Urgunge
says that he thinks of tengger (the sky) as both an ancient sage and a time-
less metaphysical entity, that is immediately reminiscent of Christian
representations of God as both an all-wise father-figure and an eternally
immanent non-corporeal presence,3 and when he tells Caroline that he
thinks of the Usun-Khan (water-lord) of a large river as an old man with
a long beard, we can remind ourselves of the statues of the old men with
long beards that depict the mythical river-gods ornamenting parks and
statues in some of our cities. Urgunge’s claim that Daur go to shamans
for help in illness, and particularly mental illness, which they cannot cure
by themselves in the same way that people like Caroline go to their GPs
Caroline confesses to having difficulty with Urgunge’s insistence that tenggur is both a person
3
and a place, like Irmu Khan, the lord of the netherworld (1996: 109). She half agrees with his
laughing rejoinder that when students say, ‘I went to Cambridge,’ they mean both the place and
the people, but they then settle by mutual agreement on ‘upstairs’ for tengger and ‘downstairs’ for
Irmu Khan.
Selectionist theory as narrative history 209
is likewise not difficult for Caroline to accept. Shamans, admittedly, have
acquired their skills through the experience of an ‘inner metamorphosis’
which activates ‘special means of engaging with unseen forces’ rather than
the kind of professional training undergone by medical students in our
culture. But then we need only remind ourselves of the Ndembu ‘doc-
tor adepts’ whom I cited as a parallel to ‘our’ Freudian psychoanalysts in
Section 5 of Chapter 3.
It comes as no surprise that Caroline is left uncertain about whether
Urgunge is, perhaps, making an anthropological joke at her expense when
he tells her that shamans and wild animals (which shamans are excluded
from hunting) are ‘cross-cousins’. But he leaves her in no doubt that Daur
live in a world which is peopled by spirits intermediate between earth
and heaven and that shamans, as well as divining illnesses and explain-
ing dreams, invoke and bargain with spirits, conduct rituals of sacrificial
payment to spirits, expel or calm spirits when they attack people, exorcize
spirits through substitute objects, and retrieve human souls which have
been stolen by them. Caroline sees the shamanic construction of caus-
ality, in contrast to that of the knowledgeable old men who preside over
the Daur hunting-camps and make the nightly meat-offerings necessary
to ensure the success of the hunt, as ‘founded on a particular, flagrantly
non-realistic, premise’ (1996: 39). But she accepts that the Daur’s ‘visions
of externality’ were ‘real, in the sense that they were the understandings
of real psychological processes, and it was the shaman’s task to catch them
and control them, to re-create a unity and balance’ (1996: 226–7). Her
willingness to see the world, with Urgunge’s help, as the Daur see it does
not extend to seeing spirits ‘slugging it out’, as she puts it, ‘against other
powers in the world’ (possible affinities with Zoroastrian and Christian
angels and demons notwithstanding). But she feels both that she under-
stands how Daur see the world as well as it is possible to do without actu-
ally doing so herself, and that Urgunge is satisfied that she has done so
and that he has used his familiarity with her culture as best he can to help
her. Her conversations with him give her readers the sense that between
them they have succeeded in getting into the collective Daur mind to
the point of being able to give the most nearly authentic description pos-
sible of what it feels like to see the world as representative members of the
Daur Mongol population do.
The exercise so carefully conducted by Caroline and Urgunge is still
not without dangers familiar to anyone who has ever conducted extended
conversations with native informants about their attitudes and beliefs,
including sociologists who have conducted ‘depth’ interviews, so called,
210 The theory of cultural and social selection
with fellow members of their own culture in which respondents are asked
questions of a kind which they have never had occasion to put to them-
selves. First, the quest for the vivid and illuminating detail may lead the
questioner to construct a description which loses in representativeness what
it gains in authenticity. Second, the respondent’s willingness to help the
questioner may lead the questioner to construct what is in effect a rede-
scription manufactured by the respondent’s incorporation of presupposi-
tions which are the questioner’s rather than the respondent’s own. Third,
both questioner and respondent may be equally unaware of the degree to
which the questioner’s questions reflect presuppositions unique to the ques-
tioner’s own culture. Mary Douglas, in discussing the difference between
the French and British schools of ethnography, asks the disturbing rhetori-
cal question ‘What would we know of the Nuer if they had been in the
French Sudan – and of the Dogon if they had been on the banks of the
White Nile? It is hard to imagine, because the Dogon now seem so unmis-
takably French, so urbane, so articulate, with such philosophical insight’
(1975: 124), and Evans-Pritchard has sometimes been suspected of project-
ing onto the Nuer presuppositions about the nature of subjective religious
experience which are in part a reflection of his own. But the difficulties in
the way of presenting to the reader an authentic and representative descrip-
tion of a culture – whether extravagantly remote from the reader’s or the
reader’s very own – is not a reason to abandon the attempt, any more than
the difficulties of finding evidence which provides a quasi-experimental
test of a causal hypothesis about the evolution of a distinctive pattern of
collective behaviour are a reason to abandon the attempt to explain it.
There is nothing in the neo-Darwinian paradigm which prevents the right
just-so stories of how cultures and societies have come to be what they are
from including descriptions of the subjective experiences of the men and
women who were the carriers in their minds and roles of the memes and
practices constitutive of them. Nor are descriptive stories any more arbi-
trarily conjectural than explanatory ones. They are, rather, discretionary in
a way that explanatory stories are not to the extent that the narrator is free
to select and emphasize some parts of the evidence in preference to others
as best conveying ‘their’ experience to ‘us’. And although description is a
speech-act categorically different from explanation, it is perfectly possible
that it may on occasion bring out from inside the natives’ heads something
of what is going on there which helps to explain what has made the memes
constitutive of their culture adaptive in their local environment. (Recall
from Section 8 of Chapter 1 the historian William H. McNeill’s subjective
experience of the bonding effects of close-order infantry drill.)
Selectionist theory as narrative history 211
9
Constructive discussion of the relation between the description and the
explanation of distinctive behaviour-patterns has been further complicated
in the literature of sociology by the notion of ‘thick description’ which
was borrowed by the anthropologist Clifford Geertz from the philosopher
Gilbert Ryle. In the example from Ryle with which Geertz opened his
influential exposition of the difference between ‘thin’ and ‘thick’ descrip-
tion (1973: 7), he contrasted a ‘thin’ description of what someone is doing
(‘rapidly contracting his right eyelid’) with a ‘thick’ description (‘prac-
tising a burlesque of a friend faking a wink in order to deceive an inno-
cent into thinking a conspiracy is in motion’). But this is no more than
an elaborately concocted illustration of the observer’s need to understand
what a person is doing in performing a physical action whose defining
intention cannot be inferred from the action itself. Geertz, however, had
more than this in mind. The actions whose defining intentions the eth-
nographer must get right have also to be placed within a context of others
which provide what Geertz called the ‘background information’ necessary
to enable the ethnographer to elucidate the ‘structures of significance’ of
focal rituals or traditional customs or dramatic events in the way that a
literary critic elucidates the latent assumptions and implicit nuances of a
text. Where this is done – and many anthropologists and historians do it
as a matter of course without feeling any need for methodological justi-
fication – it can be presented, and often is, as an explanatory exercise in
the sense that answers are being offered to ‘why?’ questions. Why are the
natives dressed like that? Because these are military uniforms and this is
a victory parade. Why are they reciting these words and phrases in uni-
son? Because they are drawn from a holy book and this is a funeral ser-
vice. And so on. But how has this society come to be of the kind where
military parades of this kind are held? How has this culture come to be
of the kind where recitations of this kind are thought to be appropriate
for commemorating the dead? Why, in other words, are these the evolved
behaviour-patterns which the ethnographer is reporting and describing to
prospective readers? To that question, ‘thick descriptions’ do not, and do
not pretend to, offer answers.
It is because of this confusion of purposes that Geertz can be singled
out as he has been by the evolutionary psychologists John Tooby and
Leda Cosmides as exemplarily guilty of the ill-conceived invocation of
culture as ‘the universal glue and explanatory variable that held social sci-
ence explanations together’ (1992: 41). The trouble, in their view, is not
212 The theory of cultural and social selection
so much the different meanings which different authors give to ‘culture’
as its use as an all-purpose and therefore unfalsifiable explanation of the
typical form of ‘women want to look younger because youthful appear-
ance is valued in their culture’. As in many such academic disputes, there
is a temptation to both sides to exacerbate their disagreement by attrib-
uting to each other opinions which they do not in fact hold. But for the
purposes of this book, and of this chapter within it, the mistake to be
avoided is that of assuming too easily that cultural anthropology as prac-
tised by Geertz and evolutionary psychology as practised by Tooby and
Cosmides are in irreconcilable contradiction with one another. They are
not. Comparative sociology needs both of them.
Geertz’s critics are right that he makes too much of ‘hunting the exotic’
(his own phrase), just as Tooby’s and Cosmides’ critics are right that they
are too ready to assume a uniform ancestral environment and therefore a
uniform human mind designed in response to it. But there is no contra-
diction between an authentic description in Geertz’s terms of what is
going on inside the natives’ heads during a Balinese cockfight and a valid
explanation of the status rivalry involved in it as an example of a psycho-
logical disposition latent in all human beings which this cultural form is
particularly well able to evoke. The authenticity of the description and the
validity of the explanation have then to be established in the usual way by
pitting them against whatever alternative descriptions and explanations
rival observers may put forward – not in a contest between description
and explanation, but in a contest between rival descriptions on the one
hand and rival explanations on the other. Comparative sociology is quite
difficult enough without sociologists who have innocuously different
purposes and interests being goaded into dispute over yet another false
dichotomy.
E pi l o gu e
T wo di s e nc h a n t m e n t s
1
To any present-day sociologist, the word ‘disenchantment’ at once brings
to mind Weber’s ‘Entzauberung der Welt’ – the displacement, in a ‘demag-
ified’ world, of the supernatural, the numinous, and the occult by direct
observation, controlled experiment, and logical argument. It was not – or
not to the extent that he supposed – an evolution unique to the West.
Nor was it either as irreversible or as pervasive as he and many of his
contemporaries and successors have been disposed to assume. But in the
aftermath of the so-called ‘Enlightenment’, the collective European mind
of the so-called ‘Middle Ages’ could be viewed with the complacency of
hindsight as a culture of mysticism, bigotry, and superstition in which
angels and demons fought for possession of human souls, the workings
of nature were apprehended through signs, portents, and miracles, and
Heaven was, so to speak, as close above our heads as Hell beneath our
feet. By the time that T. H. Huxley had debated with Bishop Wilberforce,
and Marx had published Volume I of Capital, theology had lost the hold
it once had over the study of the workings of the world. It had not been
comprehensively replaced by ‘natural’ or ‘mechanical’ philosophy. But it
had been forced to surrender its erstwhile authority in the steadily widen-
ing area which Wissenschaft now claimed as its own.
There was, however, a different disenchantment still to come. What
eighteenth- and nineteenth-century European intellectuals preserved
is as notable as what they rejected. It is not just that the beliefs and
attitudes inside the heads of bourgeois rationalists conversing in
drawing-rooms and coffee-houses were still very different from those
inside the heads of traditionally minded nobles, clerics, and an indif-
ferent and uneducated populace. Even the most radical of the secular
213
214 The theory of cultural and social selection
commentators on the human condition were less disenchanted than
they thought themselves to be if they are judged by the standards of
their twenty-first-century counterparts. In two respects, in particular,
they were not so much disenchanted as optimistically, not to say cheer-
fully, re-enchanted by the prospect that the discredited dogmas of the
priest-ridden past would be replaced by certainties of a new and better-
founded kind. The demise of teleology was, if not inconceivable, at any
rate not conceived; and none of them anticipated the debates that were
to exercise their successors on the topics of chance on the one hand and
morality on the other.
Older conceptions of chance in the European intellectual tradition had
been a shifting and sometimes contradictory amalgam of the ideas of the
capriciousness of the unseen powers by which human lives are influenced
on the one hand and the inexorability of the pre-ordained on the other.
But the underlying concern was the distressing inability of human beings
to foresee, and therefore to guard against, whatever unpleasantness Fate
(or ‘the Fates’), Providence, Luck, Destiny, God (or a god), or Fortune and
the ‘Wheel of Fortune’ (Murray 1978: 98–101) might hold in store. The
Homeric hero is at the same time at the mercy of sudden and uncontrol-
lable impulses which are visited on his thymos from without and driven
relentlessly towards an inescapable death whose manner, time, and place
cannot be altered even by a god or goddess who is by way of being on
his side. The Christian preux chevalier, however strong his faith and sin-
cere his prayers, is prey to supernatural terrors, hideous temptations, and
the risk of dying unshriven with his salvation itself at stake. But by the
eighteenth century, if not before, the mysteriousness of what lay in the
impenetrable future had begun to be not merely demystified but tamed.
The distinction between the ‘natural’ and the ‘preternatural’ could now
be resolved by simply absorbing within the first whatever of the second
retained its credibility. Actuarially calculated life-tables, fire-insurance
premiums, public lotteries, and political arithmetic (Davenant’s ‘art of
reasoning by figures upon things related to government’) made the uncer-
tainties of human life both less frightening and more manageable. Yet it
was an intellectual world as far from that of Darwin or Nietzsche as it
was from that of Aquinas or Dante – a world in which the laws of nature
to which mankind was subject along with the rest of God’s creation,
although they might not yield predictive knowledge of individual futures
any more than of next year’s market price for a bushel of wheat, were
taken to be uncompromisingly true. As apt an example as any is Joseph
Massie’s remark of 1760, in the context of the study of commerce, that ‘it
Sociology in a post-Darwinian world 215
would be possible, with good statistics, to discover the laws governing the
mechanism’ (Hoppit 1996: 521).
As with the laws of nature, so with the laws of conduct. Sin,
Redemption, and the Day of Judgement were replaced not by nihilism
but by ‘moral science’ – psychology in the service of virtue. The utili-
tarianism which was to dominate British thought on these topics, and
thereby to incur Nietzsche’s contempt, was implicit in the ‘moral sense’
theory of eighteenth-century authors who, as Elie Halévy pointed out
(1928: 13), had already inferred from the psychological principle of sym-
pathy a fusion of our personal interests with the happiness of our neigh-
bours. The dilemmas which Mill and Sidgwick were to find so hard to
reconcile, and which were to lead Weber to invoke ‘old Mill’, as he called
him (meaning John Stuart, not his father James), in the name of ‘poly-
theism’ of values, were not yet enunciated or even perceived. To be sure,
there was Hobbes in the background with his bonum and malum as mere
nomina signifying appetite or aversion. But then Hobbes was as much an
advocate of the duty, as well as the prudence, of political obedience as
Adam Smith, despite his much-quoted insistence on the self-interested
motives of professionals and tradesmen, was an advocate of the ‘moral
sentiments’ of sympathy, benevolence, and fellow-feeling. Nobody sug-
gested that virtue might not be something desirable in itself or truth not
necessarily a good.
2
One of the several ironies in the second disenchantment is that through-
out the history of twentieth-century sociology scientific method in gen-
eral, and statistical techniques in particular, appeared to make human
behaviour easier not only to explain but also to control. Their applica-
tion came to extend beyond demography, criminology, and medicine to
the pre-testing of consumer tastes in food and drink by producers and
advertisers and the interviewing of samples drawn from now enfranchised
populations by opinion pollsters in such a way as to permit not only more
accurate forecasting of the outcome of parliamentary and presidential
elections but the modification of political programmes in the light of the
preferences and prejudices thereby revealed. Condorcet’s conviction that
‘Only calculation, practised faithfully and mindfully, could create jus-
tified certainty, as opposed to the spurious certainty generated by mere
reiteration of impressions’ (Daston 2007: 128) seemed to be increasingly,
not to say triumphantly, justified.
216 The theory of cultural and social selection
The disenchantments still to come were partly technical. But the
appearance of these was accompanied by a growing awareness of problems
of a more fundamental kind. It is only to be expected that in the practice
of Wissenschaft the solution of one problem will lead to the emergence of
another. If, for example, the successful use of tests of statistical significance
raised hitherto unnoticed puzzles in the notion of randomness itself, that
did not deprive the tests of their usefulness. Estimation of the likelihood
that an apparently significant correlation might be due purely to chance
doesn’t lose its point because of a concern about whether disturbingly long
strings of the same digits may not be part of a genuinely random sequence
which cannot be derived from an algorithm shorter than itself. But what
about the uncertainties introduced by the very act of measurement? Or
about the unmanageably large effects which can be produced by extremely
small disturbances even within strictly deterministic systems? Might it be
that twenty-first-century sociologists have to face up to the implications of
living in a universe where chance, in the words of C. S. Peirce, ‘pours in
at every avenue of sense’ (Hacking 1990: Ch. 23)? What would Spencer, or
Marx, or Weber, or Durkheim have said in response to Jacques Monod’s
remark that ‘chance alone is at the source of every innovation, of all cre-
ation, in the biosphere’ (1972: 110)? In a post-Darwinian world, under-
standing of the underlying process which has made species, cultures, and
societies into what they are comes at the price of surrendering any vestigial
hope of predicting what they are going to become.
3
No twenty-first-century sociologist talks any longer about ‘moral science’
any more than about ‘laws of commerce’. The welfare economists and
social choice theorists who are the intellectual descendants of Bentham
and Mill have abandoned any claim to be able to prove to policy-makers
what transcendental ideal of the good society they should have in mind in
their attempts to base their decisions on arithmetical calculation and logi-
cal inference. To Spencer or Marx or Weber or Durkheim, however, the
Wittgensteinian notion that morality cannot be put in propositional form,
but merely demonstrated by being lived, would have seemed eccentric if
not perverse. They would have been as nonplussed to hear the literary critic
Lionel Trilling, in his opening remarks in a lecture series on ‘Sincerity and
Authenticity’, say to a Harvard audience in 1970 that ‘Nowadays, of course
[the italics are mine], we are all of us trained to believe that the moral
life is in ceaseless flux’ (1972: 1), as to find a prominent neuroscientist
Sociology in a post-Darwinian world 217
saying at the end of the twentieth century, ‘There cannot be any social
order or moral order that is objectively desirable’ (Varela 1999: 64). Any
twenty-first-century sociologist who looks back to the time when the four
of them were composing the works by which the agenda of twentieth-
century sociology was largely set is likely to be struck not merely by their
confidence in the progress of social alongside natural science but by their
readiness to see themselves as not only researchers but preceptors whose
researches entitled them to pronounce on how the members of both their
own and other societies should regulate their lives. This is as obviously so,
albeit for different reasons, in the case of Durkheim as in that of Marx or
Spencer. Marx’s injunctions to proletarians and right-minded bourgeois
to hasten the impending overthrow of capitalism, like Spencer’s injunc-
tions to governments not to intervene in the elimination of the unfit,
are too familiar to need comment. Likewise, Durkheim’s explicit moral
didacticism is as familiar as his hypotheses about the division of labour
or suicide rates or the primordial origins of religion. But Weber too, for
all his insistence that sociology, although it cannot but be ‘value-relevant’,
must at the same time be ‘value-free’, leaves his readers in no doubt about
the moral standards to which he thinks the people whose behaviour he
has analysed should conform. The student audiences who attended his
celebrated lectures on the vocations of Wissenschaft and Politik cannot
have left the lecture-hall without recognizing that he regarded intellectual
honesty as a moral duty incumbent on university teachers and politicians
guided by an ethic of responsibility as more to be admired than those
guided by an ethic of absolute ends.
The same holds for the British sociologists of the early twentieth cen-
tury, including those who came to it by way of statistics no less than those
who came to it by way of philosophy. Beatrice Webb, although she saw
herself as engaged in creating a ‘science of society’, had no doubt about
the ethical standard by which the institutions of a scientifically ordered
society would be an improvement on what they replaced. Indeed, she and
her contemporaries were heirs to a tradition which ‘from its very founda-
tion was premised on the use of social knowledge in the service of reform’
(Goldman 2007: 434). However much they might differ among them-
selves in their views about the powers and functions of the state, they
shared a common commitment to the goal of improving the condition
of the people. Loss of Christian faith, whether or not under the influ-
ence of either Darwin or Nietzsche, did not undermine their inherited
convictions about right and wrong in matters of both personal conduct
218 The theory of cultural and social selection
and public policy. None of them would have been willing to accept the
characterization of the moral life as ‘in ceaseless flux’.
It is, therefore, another of the ironies of twentieth-century intellectual
history that the cultural, and therewith moral, relativism which was to
accompany the abandonment of teleological theories of progress derived
in part from the same sociological evidence that had once been invoked
in support of it. The early advances in knowledge of the ideas and behav-
iour of peoples from outside the literate civilizations of Europe and Asia
had seemed to confirm a just-so story of the evolution of mankind in
which phylogeny replicated ontogeny and the ‘people without history’,
as the anthropologist Eric Wolf was to call them (Wolf 1982), would in
due course catch up with their betters. By the end of the twentieth cen-
tury, such presuppositions had been so far discredited that many of the
pioneering observers who had first reported on the behaviour of such peo-
ples were being pilloried by their successors as insensitive, patronizing,
and wedded alike to colonialist prejudices about supposedly backward
cultures and to capitalist prejudices about the virtues of the supposedly
rational pursuit of individual self-interest. To Spencer, Marx, Weber, or
Durkheim, the idea that their own moral standards should be placed on
a par with those whom Frazer had no compunction about calling ‘sav-
ages’ would have seemed unwarranted if not positively offensive. But any
comparative sociologists of the twenty-first century who venture to priv-
ilege their own parochial standards of behaviour over those of others can
expect to be told that the claim is indefensible as well as condescending.
In a post-Darwinian world, every code of morality is as much a locally
adaptive variant of a shared cultural inheritance as any other. The diffe-
rence between then and now is neatly captured in the contrast between
L. T. Hobhouse’s Morals in Evolution of 1906, with its story of the pro-
gress of humanity in the direction of Hobhouse’s own liberal ethical
standards, and Richard Joyce’s The Evolution of Morality of 2006, with its
story of the human capacity for making moral judgements as a product of
natural selection and the enhancement of the biological fitness of those of
our ancestors who made and acted on them.
D ou b t e r s a n d di e -h a r d s
4
The resistance of twentieth-century sociologists to neo-Darwinian the-
ory is itself an intriguing topic in the sociology of sociology. When John
Sociology in a post-Darwinian world 219
Maynard Smith, not yet known as the father of evolutionary game theory,
addressed a meeting of the Scottish branch of the British Sociological
Association on ‘Evolution and History’ in 1959 (1961: 83–94), he was as
aware as anyone in his audience of the misuses of Darwinism in sup-
port of racist ideologies and policies directed against sub-populations
culturally stigmatized as unfit. The recollection of those misuses is one
of the reasons for the die-hards’ continuing resistance. But to resist the
application of Darwinian concepts and methods to sociology because
neo-Darwinian sociology might be put to uses which the resisters would
deplore is not an argument against their validity so much as a reflection
of a fear that they will lend themselves all the more effectively to such
uses if neo-Darwinian theory turns out to be right. The irony here is
that the decoding of the human genome raises the prospect that genetic
engineering could be imposed by twenty-first-century rulers on their sub-
ordinate populations in ways which can no longer be dismissed, as could
t wentieth-century ‘eugenics’, on scientific as well as ethical grounds.
Resistance sometimes finds expression in imputations of motive
whereby neo-Darwinians are accused by the resisters of wanting to justify
the outcomes of the evolutionary process rather than explain them. This is
on occasion extended to Darwin himself, who is charged with condoning
the evils of capitalism, endorsing the doctrine of laissez-faire, and ‘siding
with the bosses and his fellow investors’ against the shop-floor (Desmond
and Moore 1991: 421). But even supposing that Darwin’s reading of
Malthus led him to see the ‘struggle for existence’ as no less inescapable
in the world of political economy than in the world of plant and animal
species, that bears neither one way nor the other on the capacity of the
theory of natural selection to withstand attempted disconfirmation. As
it happens, Wallace and Darwin held widely divergent political views,
and ‘the theory of natural selection as applied to the animal world is the
only thing on which they were in agreement’ (Winch 2001: 428). Among
Darwin’s most influential twentieth-century British successors, J. B. S.
Haldane was as far to the Left in politics as R. A. Fisher was to the Right,
and Maynard Smith was for many years a member of the Communist
Party. Maynard Smith, towards the end of his life, wondered whether his
becoming more reductionist as he grew older might have something to do
with his progressive disenchantment with Marxism, and acknowledged
that there appears to be an association between holistic views about devel-
opment and left-wing political opinions. But that did not lead him to
compromise his endorsement of the view that ‘scientific issues should be
decided by the evidence, not by ideological bias’ (1998: 43). Sociologists,
220 The theory of cultural and social selection
even more than biologists, may hope that their findings will encourage
others to share their views about how the world might be changed for
what they would personally regard as the better. But the response of many
sociologists (myself included) will be to say: if only that were so!
There is a particular irony in the vehemence of the hostility directed
by late twentieth-century Marxists against late twentieth-century
Darwinians1 in the light of Marx’s letter to Lassalle with which this
book begins and Marx’s respectful presentation to Darwin of a copy
of the second edition of Das Kapital. Like the late nineteenth-century
Christians outraged by what they saw as the threat posed to their cher-
ished convictions about the part played by God in humanity’s past, some
late twentieth-century Marxists were outraged by what they saw as the
threat posed to their cherished convictions about the part to be played by
secular revolutionaries in humanity’s future. But they might equally have
seen in the underpinning of evolutionary theory and population genetics
by molecular biology a vindication of Marx’s insistence on the material
base on which the whole of human history rests. To an avowedly Marxist
historian who recognizes that ‘developments in the natural sciences have
put an evolutionary history of humanity back on the agenda’, it explicitly
follows that ‘the DNA revolution calls for a specific, historical method for
studying the evolution of the human species’ and that history is ‘the con-
tinuance of the biological evolution of homo sapiens, but by other means’
(Hobsbawm 2007: 185–6). Although no reconciliation is possible between
neo-Darwinian selectionist theory and the teleological presuppositions
of Marxist theory in its original form, Hobsbawm’s characterization of
human history as the continuance of biological evolution by other means
can quite well stand as an apposite summary of what this book is about.
5
A further influence on late twentieth-century sociologists’ resistance to
neo-Darwinian theory was the intellectual movement known as ‘post-
modernism’. To empirical sociologists concerned only about the accuracy
of their data, the consistency of their definitions, and the reliability of
their techniques, the doubts voiced in the name of post-modernism are
of scarcely more interest than Zeno’s paradox to a ballistics expert. But
behind the post-modernists stands Nietzsche, whose part in undermining
For some vehement examples, see Pinker (2002: Ch. 7), and for some temperate comments, see
1
6
It may be the lack of agreement among selectionists about what exactly
is being selected which has encouraged some of the doubters to persist in
their doubts. But enough has (I hope) been said in this book to show that
neither the disanalogies between genes and memes or practices nor the
complexities of multi-level selection are reasons to deny that first cultural
and then social selection are continuous with, and analogous but not redu-
cible to, natural selection. Admittedly, mere assertion of what some readers
may regard as by now hardly more than a truism does not by itself generate
the just-so stories which have to be formulated and tested before any claim
to have explained how a reported cultural or social change in the state of
a specified population over a specified period came about. But that is not
a reason for denying that the truism is true. A rhetorical device sometimes
deployed by the die-hards is the claim that selectionist explanations are
merely ‘old wine in new bottles’ (Benton 2001: 216). But the accusation of
restating in other terms what is already familiar sits oddly with the attempt
to show that selectionist theory is fundamentally flawed (and selectionist
explanations therefore inherently misconceived). Doubters are always enti-
tled to raise their doubts. But they are not entitled to have it both ways.
The die-hards need to be listened to not because selectionist theory has
been, or is likely to be, overturned by their objections to it but because
Sociology in a post-Darwinian world 223
selectionists have sometimes made larger claims for it than it can sustain.
By now, the quantity and quality of findings about human behaviour
from within the neo-Darwinian paradigm is such that they are making
their way onto syllabuses that will be taught by the die-hards’ successors
to students who will wonder, like present-day geologists reading about
the nineteenth-century arguments between ‘Neptunists’ and ‘Vulcanists’,
what the fuss was about. Admittedly, not all these findings will. But
that, far from undermining the paradigm, is inherent in the evolutionary
nature of science itself.2 The perennial risk is that the advocates of any
novel paradigm will overstate their case to the point that they provoke
a reaction which goes too far in its turn. Not only are there questions
about the evolution of collective behaviour-patterns which selectionists
have answered wrongly, but there are questions which they will never be
able to answer at all. No amount of additional evidence of any conceiv-
able kind will yield a definitive account of the difference made by the
First World War to Germany’s evolution into a society of the kind that
it became under what its rulers called “National Socialism”, or the diffe-
rence made to the evolution of the economic, ideological, and political
institutions of the Soviet Union by the death of Lenin. Neo-Darwinian
sociology, however far it extends our understanding of the mechanisms
at work in cultural and social selection, will never be able to tell us more
than a limited part of what we would like to know. But it will generate
some much better-validated just-so stories than we presently have about
what has made the cultures and societies in the archaeological, ethno-
graphic, and historical record into what they have been and are.
C onc l us ion
7
I said in the Prologue that the purpose of this book is not so much to
defend selectionist theory against its critics as to suggest how the trad-
itional agenda of comparative sociology might usefully be reconstructed
in its terms. I have, in the end, done little more than sketch the outline
of such a reconstruction and put forward some examples from a range of
On which it may be worth quoting from Kuhn’s own concluding chapter (1962: 171): ‘The analogy
2
that relates the evolution of organisms to the evolution of scientific ideas can easily be pushed too
far. But with respect to the issues of this closing section it is very nearly perfect. The process
described in Section III as the resolution of revolutions is the selection by conflict within the sci-
entific community of the fittest way to practice future science.’
224 The theory of cultural and social selection
different times and places to show how the understanding of just what
is going on in distinctive cultural and social behaviour-patterns can be
advanced by it. But the selectionist agenda for comparative sociology can
be summarized, with one proviso, in a simple diagram:
gene-meme meme-practice
co-evolution co-evolution
gene-practice
co-evolution
The proviso is that the straight arrows fail to make it clear that while
they depict a historical sequence at the same time as an analytical dis-
tinction, social selection has been at work for only a very small fraction of
the time since natural selection was first supplemented by cultural selec-
tion. In the course of these few millennia, however, the rates of mutation
and extinction of both memes and practices have accelerated exponen-
tially. The same evolutionary forces which have caused hitherto separate
cultures and societies to come to resemble each other under pressures
from an increasingly global environment have at the same time generated
a multitude of new and unprecedented cultural and social forms which
have reached their own locally optimal fitness peaks. The only certainty
is that the future course of cultural and social, as of biological, evolution
will continue to be as unforeseeable as ever. In a hundred years’ time, the
cultures and societies of the twenty-second century of the Christian era
will be different from anything envisaged by any sociologist of today, and
so will twenty-second-century sociology itself.
References
225
226 References
Bassett, S. R. 1989. ‘In search of the origins of Anglo-Saxon kingdoms’, in S. R.
Bassett (ed.), The Origins of Anglo-Saxon Kingdoms. Leicester
Bateson, P. and Martin, P. 1999. Design for a Life: How Behaviour Develops.
London
Bayly, C. A. 2004. The Birth of the Modern World 1780–1914. Oxford
Beard, M., North, J., and Price, S. 1998. Religions of Rome. Cambridge
Becker, G. S. 1996. Accounting for Tastes. Cambridge, Mass.
Bede. Ecclesiastical History
Bellah, R. N. 2005. ‘What is axial about the axial age?’, Archives Européennes de
Sociologie 46: 69–89
Benton, T. 2001. ‘Social causes and natural relations’, in H. Rose and R. Rose
(eds.), Alas Poor Darwin: Arguments against Evolutionary Psychology. London
Bergucki, P. 1999. The Origins of Human Society. Oxford
Berlin, I. 1998. Many Thousands Gone: The First Two Centuries of Slavery in North
America. Cambridge, Mass.
Bianchi, S. 1987. Révolution culturelle de l’an II: Elites et peuples, 1789–1799. Paris
Binmore, K. 2001. ‘How and why did fairness norms evolve?’, Proceedings of the
British Academy 110: 149–70
Black-Michaud, J. 1975. Cohesive Force: Feud in the Mediterranean and the Middle
East. Oxford
Blair, J. 2005. The Church in Anglo-Saxon Society. Oxford
Blanning, T. C. W. 2002. The Culture of Power and the Power of Culture: Old
Regime Europe, 1660–1789. Oxford
Bloch, M. 1961. Feudal Society. London
Blute, M. 1997. ‘History versus science: the evolutionary solution’, Canadian
Journal of Sociology 22: 345–64
Borgerhoff Mulder, M. 2000. ‘Optimizing offspring: the quantity–quality trade-
off in agropastoral Kipsigis’, Evolution and Human Behavior 21: 391–410
Borofsky, R. 1987. Making History: Pukapukan and Anthropological Constructions
of Knowledge. Cambridge
Bostridge, I. 1997. Witchcraft and its Transformaions c.1650–c.1750. Oxford
Bowlby, J. 1969. Attachment and Loss. I: Attachment. London
1973. Attachment and Loss. II: Separation. London
Bowser, F. P. 1974. The African Slave in Colonial Peru, 1524–1650. Stanford, Calif.
Boyd, R., Gintis, H., Bowles, S., and Richerson, P. J. 2003. ‘The evolution of
altruistic punishment’, Proceedings of the National Academy of Sciences
100: 3,531–5
Boyer, P. 2001. Religion Explained: The Human Instincts that Fashion Gods, Spirits
and Ancestors. London
Braudel, F. 1966. La Méditerranée et le monde méditerranéen à l’ époque de Philippe
II. Paris
Brewer, J. 1982. ‘Commercialization and politics’, in N. McKendrick, J.
Brewer, and J. H. Plumb (eds.), The Birth of a Consumer Society: The
Commericalization of Eighteenth-century England. London
1997. The Pleasures of the Imagination: English Culture in the Eighteenth
Century. London
References 227
Brodie, R. 1996. Virus of the Mind: The New Science of the Meme. Seattle
Brosnan, S. F. and de Waal, E. B. M. 2003. ‘Monkeys reject unequal pay’, Nature
425: 297–9
Schiff, H. C., and de Waal, E. B. M. 2005. ‘Tolerance of inequity may increase
with social awareness in chimpanzees’, Proceedings of the Royal Society of
London Series B 272: 253–8
Brown, A. 1991. ‘City and citizen: Changing perceptions in the fifteenth and six-
teenth centuries’, in A. Molho, K. Raaflaub, and J. Emlen (eds.), City-States
in Classical Antiquity and Medieval Italy. Stuttgart
Brown, E. A. R. 1974. ‘The tyranny of a construct: Feudalism and historians of
medieval Europe’, American Historical Review 79: 1,063–88
Brown, P. 1971. The World of Late Antiquity: From Marcus Aurelius to Muhammad.
London
1981. The Cult of the Saints: Its Rise and Function in Latin Christianity.
Chicago
1995. Authority and the Sacred: Aspects of the Christianization of the Roman
World. Cambridge
Bruce, S. 1998. Conservative Protestant Politics. Oxford
Buckland, W. W. 1908. The Roman Law of Slavery: The Condition of the Slave in
Roman Law from Augustus to Justinian. Cambridge
Bull, L., Holland, O., and Blackmore, S. 2001. ‘On meme-gene co-evolution’,
Artificial Life 6: 227–35
Burke, M. A. and Young, P. 2001. ‘Competition and custom in economic con-
tracts: A case study of Illinois agriculture’, American Economic Review
19: 559–73
Burkert, W. 1996. Creation of the Sacred: Tracks of Biology in Early Religions.
Cambridge, Mass.
Burroughs, P. 1999. ‘Imperial institutions and the government of empire’,
in The Oxford History of the British Empire. III: The Nineteenth Century.
Oxford
Burrow, J. 2007. A History of Histories: Epics, Chronicles, Romances and Inquiries
from Herodotus and Thucydides to the Twentieth Century. London
Byock, J. L. 1982. Feud in the Icelandic Saga. Berkeley, Calif.
Byrne, B. 2007. ‘England – whose England? Narratives of nostalgia, empti-
ness, and evasion in imaginations of national identity’, Sociological Review
53: 509–30
Byrne, R. W. and Whiten, A. 1988. Machiavellian Intelligence: Social
Expertise and the Evolution of Intellect in Monkeys, Apes, and Humans.
Oxford
Cain, P. J. and Hopkins, A. G. 1993. British Imperialism: Innovation and
Expansion 1688–1914. London
Camerer, C. 2003. Behavioral Game Theory: Experiments in Strategic Interaction.
Princeton, N.J.
Campbell, A. 1999. ‘Staying alive: Evolution, culture, and women’s intrasexual
aggression’, Behavioral and Brain Sciences 22: 203–52
228 References
Campbell, C. 1996. ‘Half-belief and the paradox of ritual instrumental activism: A
theory of modern superstition’, British Journal of Sociology 47: 151–66
Campbell, D. T. 1960. ‘Blind variation and selective retention in creative thought
as in other knowledge processes’, Psychological Review 67: 380–400
1974. ‘Evolutionary epistemology’, in P. A. Schilpp (ed.), The Philosophy of Karl
Popper. La Salle, Ill.
1975. ‘On the conflicts between biological and social evolution and between
psychology and moral tradition’, American Psychologist 30: 1,103–26
Campbell, J. 2000. ‘Some agents and agencies of the Late Anglo-Saxon State’, in
David A. E. Pelteret, (ed.), Anglo-Saxon History: Basic Readings. London
Carey, S. 1985. Conceptual Change in Childhood. Cambridge, Mass.
Carmona, J. 2006. ‘Sharecropping and livestock specialization in France,
1830–1930’, Continuity and Change 21: 235–59
Catullus. Carmina
Centola, D. and Macy, M. 2007. ‘Complex contagions and the weakness of long
ties’, American Journal of Sociology 113: 702–34
Cheney, D. L. and Seyfarth, R. M. 1990. How Monkeys See the World: Inside the
Mind of Another Species. Chicago
Christ, M. R. 2006. The Bad Citizen in Classical Athens. Cambridge
Church, C. H. 1981. Revolution and Red Tape: The French Ministerial Bureaucracy
1770–1850. Oxford
Cicero. Republic
Philippics
Cipolla, C. M. 1972. ‘The diffusion of innovations in Early Modern Europe’,
Comparative Studies in Society and History 14: 46–52
Clarence-Smith, W. G. 2006. Islam and the Abolition of Slavery. Oxford
Clark, G. 2007. A Farewell to Alms: A Brief Economic History of the World.
Princeton, N.J.
Clark, S. 1997. Thinking with Demons: The Idea of Witchcraft in Early Modern
Europe. Oxford
Claude, D. 1980. ‘Freedmen in the Visigothic Kingdom’, in E. James (ed.),
Visigothic Spain: New Approaches. Oxford
Clauss, M. 2000. The Roman Cult of Mithras: The God and his Mysteries.
Edinburgh
Cloak, Jr, F. T. 1975. ‘Is a cultural ethology possible?’, Human Ecology 3: 161–82
Cobban, A. 1964. The Social Interpretatin of the French Revolution. Cambridge
Cohen, E. E. 1992. Athenian Economy and Society: A Banking Perspective.
Princeton, N.J.
Cohn, N. 1993. Cosmos, Chaos and the World to Come: The Ancient Roots of
Apocalyptic Faith. New Haven, Conn.
Coleman, J. S. 1990. Foundations of Social Theory. Cambridge, Mass.
Coleman, S. 2004. ‘The charismatic gift’, Journal of the Royal Anthropological
Institute n.s. 10: 421–42
Colley, L. 1992. Britons: Forging the Nation 1707–1837. London
References 229
Connor, W. R. 2004. ‘Early Greek land warfare as symbolic expression’, Past &
Present 119: 3–29
Conway Morris, S. 1998. The Crucible of Creation: The Burgess Shale and the Rise
of Animals. Oxford
Cosmides, L. and Tooby, J. 1992. ‘Cognitive adaptations for social exchange’,
in J. H. Barkow, L. Cosmides, and J. Tooby (eds.), The Adapted
Mind: Evolutionary Psychology and the Generation of Culture. Oxford
Crick, F. 1988. What Mad Pursuit: A Personal View of Scientific Discovery.
London
Countryman, L. W. 1980. The Rich Christian in the Church of the Early Empire.
New York
Darnton, R. 1979. The Business of Enlightenment: A Publicity History of the
‘Encyclopédie’ 1775–1800. Cambridge, Mass.
1996. The Forbidden Best-Sellers of Pre-Revolutionary France. London
Darwin, C. 1882. The Descent of Man, and Selection in Relation to Sex. 2nd,
revised edition. London
Daston, L. 2007. ‘Condorcet and the meaning of enlightenment’, Proceedings of
the British Academy 151: 113–34
Daube, D. 1952. ‘Slave-catching’, Judicial Review 64: 12–28
Davies, N. 1996. Europe: A History. London
Davies, W. 1987. Small Worlds: The Village Community in Early Medieval Brittany.
London
1996. ‘On servile status in the early Middle Ages’, in M. L. Bush (ed.), Serfdom
and Slavery: Studies in Legal Bondage. London
Davis, D. B. 1984. Slavery and Human Progress. Oxford
Dawson, D. 1996. The Origins of Western Warfare: Militarism and Morality in the
Ancient World. Boulder, Col.
De Vos, G. and Wagatsuma, H. 1966. Japan’s Invisible Race: Caste in Culture and
Personality. Berkeley, Calif.
De Vries, J. and van der Woude, A. M. 1997. The First Modern Economy: Success,
Failure, and Perseverance of the Dutch Economy, 1500–1815. Cambridge
De Waal, F. 1996. Good Natured: The Origins of Right and Wrong in Humans and
Other Animals. Cambridge, Mass.
2005. Our Inner Ape: The Best and Worst of Human Nature. London
Dennett, D. 1995. Darwin’s Dangerous Idea: Evolution and the Meanings of Life.
London
Descola, P. 1996. The Spears of Twilight: Life and Death in the Amazon Jungle.
London
2005. Par-delà nature et culture. Paris
Desmond, A. and Moore, J. 1991. Darwin. London
Diamond, J. 1997. Guns, Germs and Steel: The Fates of Human Societies. London
2005. Collapse: How Societies Choose or Fail to Survive. London
Dickie, M. W. 2001. Magic and Magicians in the Greco-Roman World. London
Dio Cassius. Roman History
230 References
Diodorus Siculus. History
Diogenes Laertius. Lives of the Philosophers
Dobres, M. A. and Hoffman, C. R. 1994. ‘Social agency and the dynamics of
prehistoric technology’, Journal of Archaeological Method and Theory 1: 211–58
Donoghue, J. D. 1957. ‘An Eta community in Japan: The social persistence of
outcaste groups’, American Anthropologist 59: 1,000–17
Douglas, M. 1966. Purity and Danger: An Analysis of Concepts of Pollution and
Taboo. London
1975. Implicit Meaning: Essays in Anthropology. London
Douglass, F. 1855. My Bondage and My Freedom. New York
Dover, K. J. 1974. Greek Popular Morality in the Time of Plato and Aristotle.
Oxford
Doyle, J. A. 1887. The English in America. II: The Puritan Colonies. London
Doyle, W. 1980. Origins of the French Revolution. Oxford
1996. Venality: The Sale of Offices in Eighteenth-century France. Oxford
Drescher, S. 1988. ‘Brazilian abolition in historical perspective’, Hispanic
American Historical Review 68: 249–60
Du Boulay, J. 1974. Portrait of a Greek Mountain Village. Oxford
Dunbar, R. 2004. The Human Story: A New History of Mankind’s Evolution.
London
Clark, S. and Hurst, N. L. 1995. ‘Conflict and cooperation among the
Vikings: Contingent behavioural decisions’, Ethology and Sociobiology
16: 233–46
Durham, W. H. 1991. Coevolution: Genes, Culture, and Human Diversity.
Stanford, Calif.
Edgerton, R. B. 1992. Sick Societies: Challenging the Myth of Primitive Harmony.
New York
Eisenstadt, S. N. 1982. ‘The axial age: The emergence of transcendental visions
and the rise of clerics’, Archives Européennes de Sociologie 23: 294–314
Elliott, J. H. 2006. Empires of the Atlantic World: Britain and Spain in America
1492–1830. New Haven, Conn.
Endicott, K. 1988. ‘Property, power, and conflict among the Batek of Malaysia’,
in T. Ingold, D. Riches, and J. Woodburn (eds.), Hunters and Gatherers.
2: Property, Power, and Ideology. Oxford
Epstein, S. R. 1999. ‘The rise and decline of Italian City States’, Working Paper
51/99. London School of Economics
Ermisch, J. 2005. ‘The puzzling rise in childbearing outside marriage’, in
A. F. Heath, J. Ermisch, and D. Gallie (eds.), Understanding Social Change.
Oxford
Eusebius. Ecclesiastical History
Evans-Pritchard, E. E. 1937. Witchcraft, Oracles, and Magic among the Azande.
Oxford
1940. The Nuer: A Description of the Modes of Livelihood and Political Institutions
of a Nilotic People. Oxford
1956. Nuer Religion. Oxford
References 231
1962. Essays in Social Anthropology. London
1965. Theories of Primitive Religion. Oxford
Faith, R. 1997. The English Peasantry and the Growth of Lordship. London
Faraone, C. A. 1993. ‘Molten wax, spilt wine and mutilated animals: Sympa
thetic magic in Near Eastern and Early Greek oath ceremonies’, Journal of
Hellenic Studies 13: 60–80
Feeny, D. 1993. ‘The demise of corvée and slavery in Thailand, 1782–1913’, in M.
A. Klein (ed.), Breaking of the Chains. Madison, Wis.
Fehr, E. and Gachter, S. 2002. ‘Altruistic punishment in humans’, Nature
415: 137–40
Feil, D. K. 1987. The Evolution of Highland Papua New Guinea Societies.
Cambridge
Fentress, J. and Wickham, C. 1992. Social Memory. Oxford
Ferrill, A. 1985. The Origins of War: From the Stone Age to Alexander the Great.
London
Fessler, D. M. T. and Navarette, C. D. 2003. ‘Meat is good to taboo: Dietary
proscriptions as a product of interaction of psychological mechanisms and
social processes’, Journal of Cognition and Culture 3: 1–40
Finley, M. I. 1959. ‘Was Greek civilization based on slave labour?’, Historia
8: 145–64
1965. ‘Technological innovation and economic progress in the ancient world’,
Economic History Review 18: 29–45.
Fletcher, R. 1997. The Conversion of Europe: From Paganism to Christianity
371–1386 AD. London
Flint, V. I. J. 1991. The Rise of Magic in Early Medieval Europe. Princeton, N.J.
Fogel, R. W. 1989. Without Consent or Contract: The Rise and Fall of American
Slavery. New York
Foley, R. 1996. ‘The adaptive legacy of human evolution: A search for the envir-
onment of evolutionary adaptedness’, Evolutionary Anthropology 4: 194–203
Foote, R. G. and Wilson, D. M. 1970. The Viking Achievement. London
Fouracre, P. 2005. ‘Marmoutier and its serfs in the eleventh century’, Transactions
of the Royal Historical Society 15: 29–49
Fredegar. Chronicle
Friedman, J. 1975. ‘Tribes, states, and transformations’, in M. Bloch (ed.), Marxist
Analyses and Social Anthropology. London
1982. ‘Catastrophe and continuity in social evolution’, in C. Renfrew, M. J.
Rowland, and B. Segraves (eds.), Theory and Explanation in Archaeology.
London
Froese, P. 2001. ‘Hungry for religion: A supply-side interpretation of the
Hungarian religious revival’, Journal for the Scientific Study of Religion
40: 257–68
Gager, J. G. 1992. Curse Tablets and Binding Spells from the Ancient World.
Princeton, N.J.
Galenson, D. W. 1981. White Servitude in Colonial America: An Economic Analysis.
Cambridge
232 References
Garlan, Y. 1974. Recherches de poliorcétique grecque. Paris
Geertz, C. 1973. The Interpretation of Cultures. New York
Gellner, E. 1989. ‘Culture, constraint and community: Semantic and coer-
cive compensations for the genetic under-determination of Homo sapi-
ens sapiens’, in P. Mellars and C. Stringer (eds.), The Human Revolution.
Edinburgh
Gentile, E. 1996. The Sacralization of Politics in Fascist Italy. Cambridge, Mass.
Gernet, L. 1982. A History of Chinese Civilization. Cambridge
Gigerenzer, G. and Goldstein, D. G. 1996. ‘Reasoning the fast and frugal
way: Models of bounded rationality’, Psychological Review 103: 650–69
Gil-White, F. J. 2001. ‘Are ethnic groups biological “species” to the human
brain? Essentialism in our cognition of some social categories’, Current
Anthropology 42: 515–54
Gillingham, J. 1995. ‘Thegns and knights in eleventh-century England: Who
was then the gentleman?’, Transactions of the Royal Historical Society 6th
series 5: 129–53
Gilsenan, M. 2000. Recognizing Islam: Religion and Society in the Modern Middle
East. London
Gintis, H. 2000. ‘Strong reciprocators and human sociality’, Journal of Theoretical
Biology 206: 169–79
Godechot, J. 1965. La Prise de la Bastille, 14 Juillet 1789. Paris
Goetz, H. W. 1993. ‘Serfdom and the beginnings of a “seigneurial system” in the
Carolingian period’, Early Medieval Europe 2: 29–51
Goldman, L. 2007. ‘Foundations of British sociology 1880–1930: Contexts and
biographies’, Sociological Review 55: 431–40
Goody, J. 1971. Technology, Tradition, and the State in Africa. London
Gordon, H. S. 1954. ‘The economic theory of a common-property resource: The
fishery’, Journal of Political Economy 62: 124–42
Gorski, P. S. 2000. ‘Historicizing the secularization debate: Church, state, and
society in late medieval and early modern Europe’, American Sociological
Review 65: 138–67
Gould, R. V. 2000. ‘Revenge as sanction and solidarity display: An analysis
of vendettas in nineteenth-century Corsica’, American Sociological Review
65: 685–704
Green, P. 1963. ‘The world of William Golding’, Essays by Divers Hands n.s.
32: 37–57
Greif, A. 2006. Institutions and the Path to the Modern Economy: Lessons from
Medieval Trade. Cambridge
Gresle, F. 1989. ‘La Révolution française: Notes et commentaries sur des ouvrages
récents’, Revue Française de Sociologie 30: 639–49
Griffeth, R. and Thomas, C. F. (eds.) 1981. The City-State in Five Cultures. Santa
Barbara, Calif.
Gudeman, S. and Rivera, A. 1990. Conversations in Colombia: The Domestic
Economy in Life and Text. Cambridge
Hacking, I. 1990. The Taming of Chance. Cambridge
References 233
Hahn, S. 1990. ‘Class and state in postemancipation societies: Southern planters
in comparative perspective’, American Historical Review 95: 75–98
Haidt, J. 2001. ‘The emotional dog and its rational tail: A social intuititionist
approach to moral judgement’, Psychological Review 108: 814–34
Halévy, E. 1928. The Growth of Philosophic Radicalism. London
Halstead, P. and O’Shea, J. 1989. Bad Year Economics. Cambridge
Hamilton, W. D. 1996 [1975]. ‘Innate social aptitudes of man: An approach from
evolutionary genetics’, in Narrow Roads of Gene Land: The Collected Papers
of W.D. Hamilton. I: Evolution of Social Behaviour. Oxford
Hands, A. R. 1968. Charities and Social Aid in Greece and Rome. London
Hansen, M. H. (ed.) 2000. A Comparative Study of Thirty City-State Cultures: An
Investigation Conducted by the Copenhagen Polis Centre. Copenhagen
Harris, J. 1981. ‘The red cap of liberty: A study of dress worn by French
Revolutionary partisans 1789–94’, Eighteenth-Century Studies 14: 283–312
Harris, J. R. 1998. The Nurture Assumption: Why Children Turn Out the Way
They Do. New York
Harrison, C. F. 1999. The Bourgeois Citizen in Nineteenth-Century France. Oxford
Harvey, P. D. A. 1993. ‘Rectitudines singularum personarum and Gerefa’, English
Historical Review 426: 1–22
Hayden, B. 1998. ‘Practical and prestige technologies: The evolution of material
systems’, Journal of Archaeological Method and Theory 5: 1–55
Heath, C., Bell, C. and Sternburg, E. 2001. ‘Emotional selection in memes’,
Journal of Personal and Social Psychology 2: 563–9
Hébert, F. 1995. ‘Une dramaturgie en Révolution: La Fête révolutionnaire à
Rouen (1789–An III)’, in M. Vovelle (ed.), 1789–1799: Nouveaux chantiers
d’ histoire révolutionnaire: Les institutions et les hommes. Paris
Heckhamen, J. and Schultz, R. 1999. ‘The primacy of primary control is a
human universal: Reply to Gould’s (1999) critique of the life-span theory of
control’, Psychological Review 106: 605–9
Hedström, P., Sandell, R., and Stern, C. 2000. ‘Mesolevel networks and the
diffusion of social movements: The case of the Swedish Social Democratic
Party’, American Journal of Sociology 106: 145–72
Hemmings, F. W. J. 1987. Culture and Society in France 1789–1848. Leicester
Henrich, J. and McElreath, R. 2003. ‘The evolution of cultural evolution’,
Evolutionary Anthropology 12: 123–35
Boyd, R., Bowles, S., Camerer, C., and Gintis, H. (eds.). 2004. Foundations
of Human Sociality: Economic Experiments and Ethnographic Evidence from
Fifteen Small-scale Societies. Oxford
Herman, B., Thöni, C., and Gachter, S. 2008. ‘Antisocial punishment across
societies’, Science 319: 1,362–7
Herodotus. Histories
Herzfeld, M. 1986. ‘Closure as cure: Tropes in the exploration of bodily and
social disorder’, Current Anthropology 27: 107–20
Hewlett, B. S. and Cavalli-Sforza, L. L. 1986. ‘Cultural transmission among Aka
Pygmies’, American Anthropologist 88: 922–34
234 References
Hilton, B. 1988. The Age of Atonement: The Influence of Evangelicalism on Social
and Economic Thought, 1795–1865. Oxford
Himmelfarb, G. 1984. The Idea of Poverty: England in the Early Industrial Age.
London
Hobsbawm, E. 2007. ‘Marxist historiography today’, in C. Wickham (ed.),
Marxist History-writing for the Twenty-first Century. Oxford
Hoffman, M. L. 1981. ‘Is altruism part of human nature?’, Journal of Personality
and Social Psychology 40: 121–37
Holden, C. J. and Mace, R. 1997. ‘Phylogenetic analysis of the evolution of lac-
tose digestion in adults’, Human Biology 69: 605–28
Homer. Odyssey
Hoppit, J. 1996. ‘Political arithmetic in eighteenth-century England’, Economic
History Review 49: 516–40
Hornblower, S. 1991. A Commentary on Thucydides. I: Books I–III. Oxford
Howard, M. 1984. ‘The military factor in European expansion’, in H. Bull and
A. Watson (eds.), The Expansion of International Society. Oxford
Hufton, O. 2001. ‘Altruism and reciprocity: The early Jesuits and their female
patrons’, Renaissance Studies 15: 328–53
Humphrey, C. 1996. Shamans and Elders: Experience, Knowledge, and Power
among the Dauer Mongols. Cambridge
2005. ‘Ideology in infrastructure: Architecture and Soviet imagination’,
Journal of the Royal Anthropological Institute 11: 39–58
Humphreys, S. C. 1978. Anthropology and the Greeks. London
Hunt, L. 1984. Politics, Culture, and Class in the French Revolution. Berkeley, Calif.
Hunt, T. L. 2007. ‘Rethinking Easter Island’s ecological catastrophe’, Journal of
Archaeological Science 34: 485–502
Huxley A. F. 1983. ‘How far will Darwin take us?’, in D. S. Bendall (ed.),
Evolution from Molecules to Man. Cambridge
Iannacone, L. R. 1995. ‘Household production, human capital, and the econom-
ics of religion’, in M. Tommasi and K. Ierulli (eds.), The New Economics of
Human Behaviour. Cambridge
Iliffe, J. 1983. The Emergence of African Capitalism. London
2005. Honour in African History. Cambridge
Innes, M. 2000. State and Society in the Early Middle Ages: The Middle Rhine
Valley 400–1000. Cambridge
Irons, W. 1999. ‘Adaptively relevant environment versus the environment of evo-
lutionary adaptedness’, Evolutionary Anthropology 6: 194–204
Jablonka, E. and Lamb, M. J. 1995. Epigenetic Inheritance and Evolution: The
Larmarckian Dimension. Oxford
Jankowski, R. 2007. ‘Altruism and the decision to vote: Explaining and testing
high voter turnout’, Rationality and Society 19: 5–34
John, E. 1977. ‘War and society in the tenth century: The Maldon campaign’,
Transactions of the Royal Historical Society 5th series: 173–95
Jolowicz, H. J. and Nicholas, R. 1972. A Historical Introduction to Roman Law.
Cambridge
References 235
Jones, A. H. M. 1964. The Later Roman Empire 284–602: A Social, Economic, and
Administrative Survey. Oxford
Kakar, H. K. 1979. Government and Society in Afghanistan: The Reign of Amir
‘Abd al- Rahman Khan. Austin, Tex.
Kanazawa, S. 2000. ‘A new solution to the collective action problem: The para-
dox of voter turnout’, American Sociological Review 65: 433–42
Kang, G. E. 1979. ‘Exogamy and peace relations of social units: A cross-cultural
test’, Ethnology 18: 85–99
Kaplan, S. 1979. ‘Réflexions sur la police du monde de travail, 1700–1815’, Revue
Historique 261: 17–77
Katz, E. and Lazarsfeld, P. F. 1955. Personal Influence: The Part Played by People
in the Flow of Mass Communications. Glencoe, Ill.
Keeley, L. 1996. War Before Civilization. Oxford
Kelly, R. C. 1985. The Nuer Conquest: The Structure and Expansion of an
Expansionist System. Ann Arbor
Kelly, R. L. 1993. The Foraging Spectrum: Diversity in Hunter-Gatherer Lifeways.
Washington, D.C.
Kieckhefer, R. 1989. Magic in the Middle Ages. Cambridge
Kiernan, B. 1996. The Pol Pot Regime: Race, Power, and Genocide in Cambodia
under the Khmer Rouge, 1975–79. New Haven, Conn.
Kirk, G. S. 1980. Myth: Its Meaning and Function in Ancient and Other Cultures.
Cambridge
Klein, H. S. 1982. Bolivia: The Evolution of a Multi-Ethnic Society. Oxford
1986. African Slavery in Latin America and the Caribbean. Oxford
Knauft, B. 1991. ‘Violence and sociality in human evolution’, Current
Anthropology 32: 223–45
Kopytoff, I. and Miers, S. 1977. ‘Introduction’, in S. Miers and I. Kopytoff (eds.),
Slavery in Africa: Historical and Anthropological Perspectives. Madison, Wis.
Kroll, E. 1997. ‘Dogs, Darwinism, and English sensibilities’, in R. W. Mitchell,
N. S. Thompson, and H. Lyn Miles (eds.), Anthropomorphism, Anecdotes,
and Animals. Albany, N.Y.
Kuhn, D. 1991. The Skills of Argument. Cambridge
Kuhn, T. S. 1962. The Structure of Scientific Revolutions. Chicago
Kuran, T. 1995. Private Truths, Public Lies. Cambridge, Mass.
Kurzban, R. and Leary, M. R. 2001. ‘Evolutionary origins of stigmatization: The
function of social exclusion’, Psychological Bulletin 127: 187–208
Lactantius. Divine Institutes
Lake, M. 1998. ‘Digging for memes: The role of material objects in cultural
evolution’, in C. Renfrew and C. Scarre (eds.), Cognition and Material
Culture: The Archaeology of Symbolic Storage. Cambridge
Laland, K. N., Odling-Smee, J., and Feldman, M. W. 2000. ‘Niche construc-
tion, biological evolution and cultural change’, Behavioural and Brain
Sciences 23: 131–75
and Brown, G. R. 2002. Sense and Nonsense: Evolutionary Perspectives on
Human Behaviour. Oxford
236 References
Lamont, P. 2004. ‘Spiritualism and a mid-Victorian crisis of evidence’, Historical
Journal 47: 897–920
Landau, M. 1991. Narratives of Evolution. New Haven, Conn.
Landes, D. S. 1998. The Wealth and Poverty of Nations: Why Some Are So Rich and
Some So Poor. New York
Lane, C. 1981. The Rites of Rulers: Ritual in Industrial Society – The Soviet Case.
Cambridge
Lane, F. C. 1973. Venice: A Maritime Republic. Baltimore
Lankaster, W. 1981. The Rwala Bedouin Today. Cambridge
Lawson, M. K. 1993. Cnut: The Danes in England in the Early Eleventh Century.
London
Le Goff, J. 1984. The Birth of Purgatory. London
Leach, E. R. 1961. Rethinking Anthropology. London
1982. Culture and Communication: The Logic by which Symbols Are
Connected: An Introduction to the Use of Structuralist Analysis in Social
Anthropology. Cambridge
Lee, B. B. 1979. The !Kung San: Men, Women, and Work in a Foraging Society.
Cambridge
Lefebvre, G. 1939. Quatre-Vingt-Neuf. Paris
1954. Études sur la Revolution Française. Paris
Lennard, R. 1959. Rural England 1086–1135. Oxford
Lévi-Strauss, C. 1963. Structural Anthropology. New York
Levy, D. A. and Nail, P. R. 1993. ‘Contagion: A theoretical and empirical
review and reconceptualization’, Genetic, Social, and General Psychology
Monographs 119: 235–84
Lewis, G. 1980. Day of Shining Red: An Essay on Understanding Ritual.
Cambridge
Libby, D. C. and Paria, C. A. 2000. ‘Manumission practices in a late eight-
eenth-century Brazilian slave parish: Sâo José d’El Rey in 1795’, Slavery and
Abolition 21: 96–127
Liebermann, F. (ed.) 1903. Die Gesetze der Angelsachsen. Halle
Liebeschutz, J. H. W. G. 1979. Continuity and Change in Roman Religion.
Oxford
Limberis, V. 1991. ‘The eyes infected by evil: Basil of Caesarea’s homily, On
Envy’, Harvard Theological Review 84: 163–84
Livy. History of Rome
Lloyd, G. E. R. 2007. Cognitive Variations: Reflections on the Unity and Diversity
of the Human Mind. Oxford
Loomis, W. T. 1998. Wages, Welfare Costs, and Inflation in Classical Athens. Ann
Arbor
Loyn, H. R. 1955. ‘Gesiths and thegns in Anglo-Saxon England from the seventh
to the tenth century’, English Historical Review 277: 529–49
1962. Anglo-Saxon England and the Norman Conquest. London
Lucian. The Passing of Peregrinus
References 237
Lynch, A. 1996. Thought Contagion: How Belief Spreads Through Society.
New York
MacCulloch, D. 2003. Reformation: Europe’s House Divided 1490–1700. London
MacDonald, K. B. 1993. A People That Shall Dwell Apart: Judaism as a Group
Evolutionary Strategy. Westport, Conn.
Mace, R., Holden, C. J., and Shennan, S. (eds.) 2005. The Evolution of Cultural
Diversity: A Phylogenetic Approach. London
Macleod, W. C. 1925. ‘Debtor and chattel slavery in aboriginal North America’,
American Anthopologist 27: 370–80
MacMullen, R. 1984. Christianizing the Roman Empire (A.D. 100–400). New
Haven, Conn.
Macy, M. W. 1991. ‘Learning to cooperate: Structure and tacit collusion in social
exchange’, American Journal of Sociology 97: 808–43
Maguire, H. 1994. ‘From the evil eye to the eye of justice: The saints, art, and
justice in Byzantium’, in A. E. Laiou and D. Simon (eds.), Law and Society
in Byzantium: Ninth-Twelfth Centuries. Washington, D.C.
Malaussena, K. 2004. ‘The birth of modern commemoration in France: The tree
and the text’, French History 18: 154–72
Malinowski, B. 1935. Coral Gardens and their Magic: A Study of the Methods of
Tilling the Soil and of Agricultural Rites in the Trobriand Islands. London
Mallon, R. and Stich, S. P. 2000. ‘The odd couple: The compatibility of
social construction and evolutionary psychology’, Philosophy of Science
67: 133–54
Mannheim, K. 1936. Ideology and Utopia. London
Marcus Aurelius. Meditations
Marshall, G. 1981. Presbyteries and Profits: Calvinism and the Development of
Capitalism in Scotland 1560–1707. Oxford
Maschner, H. D. G. and Patton, J. Q. 1996. ‘Kin selection and the origins of
hereditary social inequality: A case study from the Northern Northwest
Coast’ in H. D. G. Maschner (ed.), Darwinian Archaeologies. New York
Massey, D. S. 2002. ‘A brief history of human society: The origin and role of
emotions in social life’, American Sociological Review 67: 1–29
Matthias, P. 1979. The Transformation of England: Essays in the Social and
Economic History of England in the Eighteenth Century. London
Maynard Smith, J. 1961. ‘Evolution and history’, in M. Banton (ed.), Darwinism
and the Study of Society. London
1982. Evolution and the Theory of Games. Cambridge
1998. Shaping Life: Genes, Embryos, and Evolution. London
Maza, S. 2003. The Myth of the French Bourgeoisie: An Essay on the Social
Imaginary 1750–1850. Cambridge, Mass.
McGrew, W. 2004. The Cultured Chimpanzee: Reflections on Cultural Primatology.
Cambridge
McKechnie, P. 1989. Outsiders in Greek Cities in the Fourth Century B.C.
London
238 References
McNeill, W. H. 1982. The Pursuit of Power: Technology, Armed Force, and Society
since A.D. 1000. Oxford
Meeks, W. A. 1993. The Origins of Christian Morality. New Haven, Conn.
Miller, G. F. 1998. ‘How mate choice shaped human nature: A review of sexual
selection and human evolution’, in C. Cranford and D. L. Krebs (eds.),
Handbook of Evolutionary Psychology: Ideas, Issues, Applications. Mahwah,
N.J.
Millett, P. 1989. ‘Patronage and its avoidance in classical Athens’, in A. Wallace-
Hadrill (ed.), Patronage in Ancient Society. London
Mithen, S. 2004. ‘From Ohalo to Çatalhöyük: The development of religios-
ity during the early prehistory of Western Asia, 20,000 to 7000 BCE’, in
H. Whitehouse and L. H. Martin (eds.), Theorizing Religions Past:
Archaeology, History, and Cognition. Walnut Creek, Calif.
Momigliano, A. 1975. Alien Wisdom: The Limits of Hellenization. Cambridge
Monod, J. 1972. Chance and Necessity: An Essay on the Natural Philosophy of
Modern Biology. London
Moore, J. S. 1998. ‘Domesday slavery’, Anglo-Norman Studies 11: 190–220
Morgan, P. D. 1988. Slave Counterpoint: Black Culture in the Eighteenth-Century
Chesapeake and Lowcountry. Chapel Hill, N.C.
Murra, J. V. 1980. The Economic Organization of the Inka State. Greenwich,
Conn.
Murray, A. 1978. Reason and Society in the Middle Ages. Oxford
Murray, O. 1991. ‘War and the Symposium’, in W. J. Slater (ed.), Dining in a
Classical Context. Ann Arbor, Mich.
Nietzsche, F. The Antichrist
The Will to Power
Genealogy of Morals
Nisbett, R. E. 2003. The Geography of Thought: How Asians and Westerners Think
Differently … and Why. New York
and Cohen, D. 1996. Culture of Honor: The Psychology of Violence in the
American South. Boulder, Col.
Nock, A. D. 1933. Conversion: The Old and the New in Religion from Alexander
the Great to Augustine of Hippo. Oxford
Oates, J. 1979. Babylon. London
Ogden, D. 2001. Greek and Roman Necromancy. Princeton, N.J.
Orians, G. H. and Heerwagen, J. H. 1992. ‘Evolved responses to landscapes’, in
J. H. Barkow, L. Cosmides, and J. Tooby (eds.), Evolutionary Psychology
and the Generation of Culture. Oxford
Ostrom, E. 1990. Governing the Commons: The Evolution of Institutions for
Collective Action. Cambridge
Walker, J., and Gardner, R. 1992. ‘Covenants with and without the sword: Self-
governance is possible’, American Political Science Review 86: 404–17
Otterbein, K. 2000. ‘Five feuds: An anlysis of homicides in Eastern Kentucky in
the late nineteenth century’, American Anthropologist 102: 231–43
Ovid. Fasti
References 239
Owen, A. 2004. The Place of Enchantment: British Occultism and the Rise of the
Modern. Chicago
Ozouf, M. 1988. Festivals and the French Revolution. Cambridge, Mass.
Palmer, C. A. 1976. Slaves of the White God: Blacks in Mexico, 1570–1650.
Cambridge, Mass.
Parker, J. 2004. ‘Witchcraft, anti-witchcraft and trans-regional innovation in
early colonial Ghana’, Journal of African History 45: 393–420
Parkin, D. 1985. (ed.), The Anthropology of Evil. Oxford
Parry, J. and Bloch, M. 1989. ‘Introduction: Money and the morality of
exchange’, in J. Parry and M. Bloch (eds.), Money and the Morality of
Exchange. Cambridge
Patterson, O. 1982. Slavery and Social Death. Cambridge, Mass.
Paul the Deacon. History of the Lombards
Pausanias. Description of Greece
Pessar, P. R. 2004. From Fanatics to Folk: Brazilian Millenarism and Popular
Culture. Durham, N.C.
Pinker, S. 1994. The Language Instinct. New York
2002. The Blank Slate: The Modern Denial of Human Nature. London
Plato. Laws
Menexenus
Pliny. Natural History
Plumb, J. H. 1967. The Growth of Political Stability in England 1675–1725.
London
Pomeranz, K. 2000. The Great Divergence: China, Europe, and the Making of the
Modern World Economy. Princeton, N.J.
Raha, M. K. 1978. ‘Stratification and religion in a Himalayan society’, in J.
S. Fisher (ed.), Himalayan Anthropology: The Indo-Tibetan Interface. The
Hague
Rappaport, R. A. 1999. Ritual and Religion in the Making of Humanity.
Cambridge
Rathbone, D. W. 1983. ‘The slave mode of production in Italy’, Journal of Roman
Studies 73: 160–8
Rawson, B. 1989. ‘Spurii and the Roman view of illegitimacy’, Antichthon
23: 10–41
Readman, P. 2005. ‘The place of the past in English culture, c. 1890–1914’, Past
& Present 186: 147–99
Richards, A. 1935. ‘A modern movement of witch finders’, Africa 8: 448–60
Richerson, P. J. and Boyd, R. 1999. ‘Complex societies: The evolutionary origins
of a crude superorganism’, Human Nature 10: 253–89
2005. Not by Genes Alone: How Culture Transformed Human Evolution.
Chicago
Ridley, M. 1996. The Origins of Virtue. London
Ritner, R. K. 1995. ‘The religious, social, and legal parameters of traditional
Egyptian magic’, in M. Meyers and P. Mirecki (eds.), Ancient Magic and
Ritual Power. Leiden
240 References
Roberts, J. H. 1988. Darwinism and the Divine in America: Protestant Intellectuals
and Organic Evolution 1859–1900. Madison, Wis.
Robertson, A. F. 1980. ‘On sharecropping’, Man 15: 411–29
Robertson, A. J. (ed.) 1939. Anglo-Saxon Charters. Cambridge
Roffe, D. 1990. ‘Domesday Book and Northern Society: A reassessment’, English
Historical Review 105: 310–36
Runciman, W. G. 1972. ‘Describing’, Mind 81: 372–88
1983. A Treatise on Social Theory. I: The Methodology of Social Theory. Cambridge
1990. ‘Doomed to extinction: The polis as an evolutionary dead-end’, in
O. Murray and S. Price (eds.), The Greek City: From Homer to Alexander.
Oxford
1991. ‘Are there any irrational beliefs?’, Archives Européennes de Sociologie
32: 215–28
2001a. ‘Was Max Weber a selectionist in spite of himself?’, Journal of Classical
Sociology 1: 13–32
2001b. ‘From Nature to Culture, from Culture to Society’, Proceedings of the
British Academy 110: 235–54
Sahlins, M. 1974. Stone Age Economics. London
1983. ‘Other times, other cultures: The anthropology of history’, American
Anthropologist 85: 517–44
Samuel, R. 1998. Island Stories: Unravelling Britain: Theatres of Memory II.
London
Sawyer, R. 1986. Slavery in the Twentieth Century. London
Schapera, I. 1956. Government and Politics in Tribal Societies. London
Schelkle, W. 2000. ‘Summary: Modernization’, in W. Schelkle, W. -H. Krauth,
M. Kohli, and G. Elwert (eds.), Paradigms of Social Change: Modernization,
Development, Transformation, Evolution. Frankfurt
Schoeck, H. 1969. Envy: A Theory of Social Behaviour. London
Schwartz, S. B. 1985. Sugar Plantations in the Formation of Brazilian Society: Bahia,
1550–1835. Cambridge
Scott, C. 1988. ‘Property, practice, and aboriginal rights among Quebec Cree
hunters’, in T. Ingold, D. Riches, and J. Woodburn (eds.), Hunters and
Gatherers 2: Property, Power, and Ideology. Oxford
Scott, S. F. 1978. The Response of the Royal Army to the French Revolution: The Rise
and Development of the Line Army 1787–93. Oxford
Searle, J. R. 1995. The Construction of Social Reality. London
Seneca. Epistles
Sewell, W. H. 1981. ‘La confraternité des proletaires: Conscience de classe sous la
monarchie de juillet’, Annales 36: 650–71
Shapin, S. 1994. A Social History of Truth: Civility and Science in Seventeenth-
Century England. Chicago
Shennan, S. 2002. Genes, Memes, and Human History. London
Silk, J. B., Brosnan, S. F., Vonk, J., Henrich J., Povinelli, D. J., Richardson,
A. S., Lambeth, S. P., Mascaro, J., and Schapiro, S. J. 2005. ‘Chimpanzees are
indifferent to the welfare of unrelated group members’, Nature 437: 1,357–9
References 241
Simon, H. A. 1990. ‘Invariants of human behavior’, Annual Review of Psychology
41: 1–19
Skinner, Q. 2002. Visions of Politics. I: Regarding Method. Cambridge
Skyrms, B. 1996. Evolution of the Social Contract. Cambridge
Smail, D. L. 2008. On Deep History and the Brain. Berkeley, Calif.
Smith, A. D. 2003. Chosen Peoples. Oxford
Smout, T. C. 1969. A History of the Scottish People 1560–1830. London
Snell, K. D. M. 2003. ‘The culture of local xenophobia’, Social History 28: 1–30
Snodgrass, A. 1980. Archaic Greece: The Age of Experiment. London
Sober, E. and Wilson, D. S. 1998. Unto Others: The Evolution and Psychology of
Unselfish Behavior. Cambridge, Mass.
Solow, R. M. 1997. ‘How did economics get that way and what way did it get?’,
Daedalus 126: 39–58
Soltis, J., Boyd, R., and Richerson, P. J. 1995. ‘Can group-functional behaviors
evolve by cultural group selection? An empirical test’, Current Anthropology
36: 437–94
Somerset, A. 2003. The Affair of the Poisons: Murder, Infanticide, and Satanism at
the Court of Louis XIV. London
Sosis, R. 2003. ‘Why aren’t we all Hutterites? Costly signaling theory and reli-
gious behavior’, Human Nature 14: 91–127
and Alcorta, C. 2003. ‘Signaling, solidarity, and the sacred: The evolution of
religious behavior’, Evolutionary Anthropology 12: 264–74
Speer, A. 1970. Inside the Third Reich. London
Spooner, B. 1976. ‘Anthropology and the Evil Eye’, in C. Maloney (ed.), The Evil
Eye. New York
Stankiewicz, R. 2000. ‘The concept of design space’, in J. Ziman (ed.),
Technological Innovation and the Evolutionary Process. Cambridge
Stark, R. 1996. The Rise of Christianity: How the Obscure, Marginal Jesus
Movement Became the Dominant Religious Force in the Western World in a
Few Centuries. Princeton, N.J.
Stedman Jones, G. 2004. An End to Poverty? A Historical Debate. London
Strabo. Geography
Suetonius. Grammarians
Sugden, R. 2002. ‘Beyond sympathy and empathy: Adam Smith’s concept of
fellow-feeing’, Economics and Philosophy 18: 63–67
Sugiyama, L. S., Tooby, J., and Cosmides, L. 2002. ‘Cross-cultural evidence of
cognitive adaptations for social exchange among the Shiwiar of Ecuadorian
Amazonia’, Proceedings of the National Academy of Sciences 99: 11,536–11,542
Sulloway, F. J. 1996. Born to Rebel: Birth Order, Family Dynamics, and Creative
Lives. New York
Tacitus. Agricola
Annals
Tainter, J. A. 1988. The Collapse of Complex Societies. Cambridge
Talbot, C. 2001. Precolonial India in Practice: Society, Religion, and Identity in
Medieval Andhara. Oxford
242 References
Tanner, T. 2000. The American Mystery. Cambridge
Thomas, K. 1971. Religion and the Decline of Magic: Studies in Popular Beliefs in
Sixteenth and Seventeenth Century England. London
Thucydides. History
Tilley, C. 1999. Metaphor and Material Culture. Oxford
Tonkinson, R. 1988. ‘“Ideology and domination” in Aboriginal Australia: A
Western Desert test case’, in T. Ingold, D. Riches, and J. Woodburn (eds.),
Hunters and Gatherers. 2: Property, Power, and Ideology. Oxford
Tooby, J. and Cosmides, L. 1992. ‘The psychological foundations of cul-
ture’, in J. H. Barkow, L. Cosmides, and J. Tooby (eds.), The Adapted
Mind: Evolutionary Psychology and the Generation of Culture. Oxford
1996. ‘Friendship and the banker’s paradox: Other pathways to the evolution
of adaptations for altruism’, Proceedings of the British Academy 88: 119–143
Toulmin, S. and Baier, K. 1952. ‘On describing’, Mind 61: 13–38
Trilling, L. 1972. Sincerity and Authenticity. London
Tuchman, B. 1962. August 1914. London
Turley, D. 1991. The Culture of English Antislavery, 1780–1860. London
Turner, M. 1982. Slaves and Missionaries: The Disintegration of Jamaican Slave
Society, 1787–1834. Urbana, Ill.
Turner, V. W. 1968. The Drums of Affliction: A Study of Religious Processes among
the Ndembu of Zambia. Oxford
Van den Berghe, P. L. 1997. ‘Rehabilitating stereotypes’, Ethnic and Racial
Studies 20: 1–16
Vanberg, V. J. 2002. ‘Rational choice versus program-based behav-
iour: Alternative theoretical approaches and their relevance for the study of
institutions’, Rationality and Society 14: 7–54
Varela, F. J. 1999. Ethnical Know-How: Action, Wisdom, and Cognition. Stanford,
Calif.
Varro. De Re Rustica
Veyne, P. 1981. ‘Clientèle et corruption au service de l’état: La vénalité des offices
sous le bas-empire romain’, Annales 36: 339–60
Viveiros de Castro, E. 1998. ‘Cosmological deixis and Amerindian perspectiv-
ism’, Journal of the Royal Anthropological Institute 4: 469–88
Weber, M. 1922. Gesammelte Aufsätze zur Religionssoziologie. Tübingen
Webster, T. B. L. 1953. Studies in Later Greek Comedy. Manchester
Welleman, J. D. 2003. ‘Narrative explanation’, Philosophical Review 112: 1–25
Whitehouse, H. 2004. Modes of Religiosity: A Cognitive Theory of Religious
Transmission. Oxford
Whitelock, D. 1979. (ed.), English Historical Documents I. Cambridge
Whiten, A., and Byrne, R. W. (eds.) 1997. Machiavellian Intelligence. II: Exten
sions and Evaluations. Cambridge
Horner, V., and Marshall-Pescini, A. 2003. ‘Cultural panthropology’,
Evolutionary Anthropology 12: 92–105
Whiting, R. C. 1983. The View from Cowley: The Impact of Industrialization upon
Oxford 1918–1939. Oxford
References 243
Wickham, C. 1994. Land and Power: Studies in Italian and European Social
History, 400–1200. London
2005. Framing the Middle Ages: Europe and the Mediterranean 400–800.
Oxford
Wierzbicka, A. 1992. Semantics, Culture, and Cognition: Universal Human
Concepts in Culture-Specific Configurations. Oxford
Wiley, A. S. 2007. ‘Transforming milk in a global economy’, American
Anthropologist 109: 666–77
Williams, B. 2002. Truth and Truthfulness: An Essay in Genealogy. Princeton,
N.J.
Williams, G. C. 1966. Adaptation and Natural Selection: A Critique of Some
Current Evolutionary Thought. Princeton, N.J.
Wilson, E. O. 1978. On Human Nature. Cambridge, Mass.
Winch, D. 2001. ‘Darwin fallen among political economists’, Proceedings of the
American Philosophical Society 145: 415–37
Winch, P. 1958. The Idea of a Social Science. London
Wittgenstein, L. 1953. Philosophical Investigations. Oxford
Wolpert, S. 1977. A New History of India. Oxford
Wolf, E. R. 1982. Europe and the People without History. Berkeley, Calif.
Woolf, G. 1990. ‘Food, poverty, and patronage: The significance of the epig-
raphy of the Roman alimentary schemes in early imperial Italy’, Papers of
the British School in Rome 58: 197–228
Wrangham, R. and Peterson, D. 1996. Demonic Males: Apes and the Origins of
Human Violence. London
Wrigley, E. A. 1987. People, Cities and Wealth: The Transformation of Traditional
Society. Oxford
2003. ‘The Quest for the Industrial Revolution’, Proceedings of the British
Academy 121: 147–70
Wyatt-Brown, B. 1982. Southern Honor: Ethics and Behavior in the Old South.
Oxford
Yair, G. 2007. ‘Existential uncertainty and the will to conform: The expressive
basis of Coleman’s rational choice paradigm’, Sociology 41: 681–98
Zafirovski, M. 1999. ‘What is really rational choice? Beyond the utilitarian con-
ception of rationality’, Current Sociology 47: 47–113
Zeldin, T. 1973. France 1848–1945. I: Ambition, Love and Politics. Oxford
Index
244
Index 245
Angra Mainyu, 204 Bayly, C. A., 188, 190
animal behaviour studies, 2, 9–10, 13, 56–7, Becker, G. S., 66
86–7, 91 Bede, the Venerable, 94, 120, 171
see also primatology behaviour, types of, 8–9
anthropology, 12, 14, 73, 89–90 see also acquired behaviour; evoked
anti-Darwinism, 1, 3–4, 7–8, 10, 53, 65 behaviour; imposed behaviour
Antigua, 159–60 behavioural development, 72–3
Anti-Slavery Monthly Reporter, 159 behavioural ecologists, 47, 122
archaeological record, 9, 35, 41, 43, 58–9, 76n6, behavioural sciences, 2–3, 27
78 behaviourist psychology, 10–11, 23, 27
aristocracy, 143–4 behaviour-patterns
Aristophanes, 99–100n3 co-adaptive memes, 98–9
Aristotle literate/non-literate cultures, 56
animals’ reasoning, 9 neo-Darwinism, 23
Nicomachean Ethics, 88 phenotypes, 11–12, 27, 120–1
and Nisbett, 119n12 transmission, 33–4
physical theory, 4 universal, 71–3
Politics, 33, 121 see also collective behaviour-patterns
reason/passion, 66 beliefs
artefacts, 58–9, 72 ambivalence, 93–4
Artificers, Statute of, 155 and attitudes, 65–6, 74–5, 149
L’Artisan, 181 causal, 99
Arval Brethren, 126 converts to, 102
Ashkenazi Jews, 51 distribution of, 92
Asoka, 146, 146n2 explanations of, 222
Assurbanipal, 149 half-belief, 93n1
Assyrian empire, 149 hunter-gatherers, 68
atavisms, 182–3 memes, 67–8
Athelstan, King, 169n23 rational, 66–7
Athens, 141, 147n4, 156–7, 189–90, 199, 202 retention, 101
attitudes shared, 98, 201
adaptedness, 100 transmission of, 95
and beliefs, 65–6, 74–5, 149 belief-systems, priests, 95
distribution of, 92 Bemba, 108
hunter-gatherers, 68 benevolence, unconditional, 134–5
retention, 101 benevolentia, 132
sacralizing, 99 Bentham, Jeremy, 216
shared, 98, 201 Bergdama, 36n5
Augustus, 128, 144 Bergson, Henri, 28
Avesta, 203 big-men, 43, 142–3, 165, 198
Axial Age, 202–4 binana system, 44
Azande, 12, 67, 93 biological inheritance: see inheritance,
Aztecs, 36, 191 biological
birth order, 129n16
Babeuf, ‘Gracchus’, 182 bishops, 141
Babylonia, 33, 165, 191 Blanc, Louis, 180
bacteria, 32 Bligh, Captain, 40
Balinese cockfight, 212 Bloch, Marc, 144, 151, 164
Bantu, 42 blood-money, 85
Baoule, 48 Boas, Franz, 93
baptism, 132 Bolivian Revolution, 172–3, 175
Barbados, 159–60 Bolsheviks, 182
Bastille, fall of, 174, 174n27, 180 bonobos, 35n2
Batek, 42 bordarii, 170
Bayes’s theorem, 82–3 Borofsky, Robert, 94
246 Index
Bounty mutiny, 40 selectionist theory, 193
Bourdieu, Pierre, 45 slavery, 154–5
bourgeoisie, 143, 180, 213, 217 change
bower-birds, 11, 22 absence of, 124
Bowlby, John, 31, 81 drivers of, 139
Boyd, R., 50, 53n12, 60, 82 gradual/sudden, 171–2
Boyle, Robert, 96, 99 mutant practices, 144
Braudel, Fernand, 112 punctuated equilibrium, 172
Brazil qualitative, 2–3
abolition of slavery, 160–1 charisma, 98, 100, 102, 104, 141, 168
millenarism, 102 Charlemagne, 146
mining in, 162n17 Charles the Bold, 164
slavery, 154, 154n9, 162n18 Chartres Cathedral, 91
bridewealth payments, 62 Chatham Islands, 50
Brinton, Crane, 19 cheating, 70, 137
Britain, 47, 52n11, 55, 67, 145, 165, 188, 194 child rearing, 64, 112–13, 128–9
see also England; Scotland child-bearing
British Empire, 147, 147n4, 149n5, 193 frequency, 47
Britishness, 118 outside marriage, 115
Brittany, 142 children
Broglie, Marshal de, 174n26 memetic bombardment, 55
Browne, Thomas, 125n14 sold into slavery, 154, 154n9
Brummel, ‘Beau’, 102 chimpanzees, 10
Brutus, Marcus, 125 alpha male, 43, 78, 90
Buddhism, 111, 113, 119, 203 evoked behaviour, 91
Bukhara, 190 friendships, 88
Burakumin, 156n12 and humans, 64, 86–8
bureaucracy, 152 individuality, 88, 90
burial customs, 96 meta-representations, 90–1
Byzantium, 108, 129 violence, lethal, 35, 35n2
vocalizations, 88
Calvin, Jean, 110, 136 see also de Waal
Calvinism, 101, 110, 138–9 China, 47, 96, 111, 113, 151, 175, 202–3
Campbell, Donald T., 2, 3, 5–6, 95 Chindasuinth, 46
Cannae, Battle of, 126 chocolate, 80
canoe-building, 76, 77 Christian, Fletcher, 40
cantars, 84 Christianity
capitalism, 50, 101, 120, 136, 152, 181–2 almsgiving, 132–4
capuchins, 91 benevolence, unconditional, 134–5
Caribbean, 155, 157–8, 160 burial customs, 96
Carlyle, Thomas, 192 Constantine, 130–1
Cartesian concepts, 119 cultural transmission, 135–6
Carthage, 126, 147 cultural/social selection, 130–1
cash economy, 44 Europe, 163–4
Cassius, 125 evil eye, 108
caste, 39, 79, 170, 185 forcible conversion, 135–6
Cato, 161 God, 208
Catullus, 125n14 Great Persecution, 131
causality, 99, 209, 221 heresy/orthodoxy, 130–1
centre/periphery, 146, 146n3, 147, 148 holy man, 44–5n8
ceorls, 166–8, 171 imperialism, 136
chance and Judaism, 51
innovation, 216 love, 97, 132, 134
intellectual tradition, 214 loss of faith in, 217
mineral deposits, 147n4 martyrdom, 96, 100
Index 247
and Marxism, 100–1 practice-tracing, 29, 117
memes, 108 rallying cries, 104
Paley, 4 Colley, Linda, 117–18
Protestantism/Catholicism, 30 Columella, 159, 161
reciprocity, 134 commensalism, 55, 177
Rome, ancient, 134 Communism, 101, 180
Second Coming, 204 comparative sociology
slavery, 163–4 collective behaviour-patterns, 69
social evolution, 135–6 cultural/social differences, 29–30
supernatural, 126n15, 131–2 Darwin’s influence, 9
churches as landowners, 148 evolutionary theory, vii, 6
Cicero, 106, 125, 149n5, 153, 154n9 mutation/combination, 7
Cipolla, C. M., 52n11 neo-Darwinian approach, 28
city-states, 198–9 selectionist theory, 3–4, 16, 37, 223–4
class differences, 97–8, 120 slavery, 152–3
class struggle, 1, 17–18, 24 stability of culture, 109
Clement of Alexandria, 133 warfare, 8–9
clientship, 167n20 competition, 17, 19, 24, 31, 78
Cloak, F. T., Jr., 53n13 Condorcet, Marie Jean, 215
cockfighting, 212 conduct, norms of, 109, 215
coercion, modes of conformity, 66, 104, 126
elites, 150 Confucianism, 119, 130, 203
empires, 148 conscience, 15, 69
France, post-Revolution, 177–8 conscience collective, 118
personal wealth, 141–2 Conspirary of Equals, 182
recruitment to army, 140–1, 148 Constantine, emperor, 128, 130–1, 135
slavery, 157 contagion imagery, 114–15, 116
warlords, 148 convergent evolution, 85, 117, 139, 153, 165, 190,
co-evolution, 37, 38 195, 200, 203
co-evolution Conway Morris, S., 195–6
behaviour-patterns, 92–3 Cook, James, 184
cash economy, 45 cooperation, 2, 15, 16, 73, 78, 97, 130
conformity, 104 Corpus Inscriptionum Latinorum, 153n7
culture-to-society transition, 139 Corsica, 85
demilitarization, 123–4 corvée labour, 48–9, 163
feuding, 86 Cosmides, Leda, 211–12
gene-meme, 51, 61, 78, 224 cottarii, 170
meme-practice, 181, 224 Counter-Reformation, 49
phenotypic effects, 189 Covenanters, 104
practices as drivers, 164 cowardice, 122–3, 124
similar sequences, 201 cree, 44
and sociology of knowledge, 98 creoles, 111
cognitive optimum, 102 Crick, Francis, 9
cohabitation, 115 Cromer, Lord, 149n5
Cohen, Dev, 86 Crookes, William, 107
Colbert, Jean-Baptiste, 186 cultural anthropology, 10–11
Coleman, James S., 76n5, 83–4, 137 cultural evolution
collective behaviour-patterns biological influences, 44
comparative sociology, 69 group formation, 68
deference, 45 mutant memes, 39
explanations, 72, 197 population differences, 50–1
game-theoretic models, 83–4 selectionist theory, 23–4
meme-practice co-evolution, 92–3 subcultures, 120
memes, 95 transitions, 202
natural selection, 69 variation, 6–7
248 Index
cultural inertia, 83 Descartes, René, 119
cultural relativism, 218 descent
cultural selection homologous, 33, 117, 139, 195, 203
archetypal story of, 140 with modification, 2, 10, 18, 29
beliefs and attitudes, 149 descriptions
Christianity, 130–1 authenticity, 207, 212
as driver, 65 thick/thin, 211
empires, 150 Diagoras of Melos, 94
lateral transmission, 33 Diamond, J., 183
memes, 51–2, 62, 149 Diderot, Denis, 56
and natural selection, 47, 61 diffusion, lateral, 33, 117, 139
and social selection, 40, 60, 97 Dinka, 62, 93
culture Dio Cassius, 135
acquired behaviour, 34 Diocletian, emperor, 131, 147, 149
diversity, 65 Diogenes Laertius, 94
evolving, 53 divine providence, 29n1, 125
phylogenetic models, 33 DNA, 1, 51, 220
resemblances, 195, 198–9 Dogon, 210
rules, 39 Domesday Book, 167n21, 171
traditions, 95, 124–5 domestication, 194, 196
transmission, 135–6 Douglas, Mary, 93, 210
culture-to-society transition, 139, 145 Douglass, Frederick, 157n14
Cyprian of Carthage, 133 Dover, Kenneth, 122
Doyle, William, 182n36, 186, 187
Danelaw, 167 Dualists, 78, 111
Darnton, Robert, 56 Durham, W. H., 38, 39
Darwin, Charles, 23 Durkheim, Émile, vii, 100, 139, 216–18
altruism, 133
animals, 9–10, 13 Easter Island, 183–4, 193
and comparative sociology, 9 Eddy, Mary Baker, 98
cooperation, 130 Edgar, King, 169n22
deception, 13 Egypt, 67, 105n8, 123, 144, 148, 152n6
The Descent of Man, 2, 18, 24, 25, 61 elders, 43
empathy, 12 Elective Affinity, Fallacy of, 101, 202
facial expressions, 64 see also affinities, elective/selective
gift-giving, 133 Elias, Norbert, 197
gradual change, 171 elites, 150, 200
morality, 16 Elliott, J. H., 146n3
music, 81n9 Elton, Geoffrey, 197
The Origin of Species, 1, 2, 24 embarrassment test, 103n6
political views, 219 emotion, 74
self-consciousness, 21–2, 88–9 empathy, 12, 73
Daur, 207–9 empires, 145–6, 201
Davenant, Charles, 214 absorption of weaker, 147–8, 150
Dawkins, Richard, 53–4 centre/periphery, 146, 146n3, 147, 148
de Waal, Frans, 10, 35n2, 77, 86, 90–1, 206 disengagement, 147–8, 150
deception, 10, 13, 64 natural selection, 150
decontamination, 55 endogamy, 177
deference, 45, 96 Engels, Friedrich, 19–20, 120
Deioces, 143 England, 186
delayed-return economy, 40 Norman Conquest, 166, 167, 169
demilitarization, 123–4 role dyads, 143–4
democratic society, 200, 205 and Scotland, compared, 144
Demosthenes, 124 slaves, 166
Dennett, Daniel, 2, 4 see also Anglo-Saxon England
Index 249
Englishness, 120–1 facial expressions, 64, 87–8
Enlightenment, 213 fair division game, 40
envy, 108 faith, 67n1, 217
epidemiology/providence, 29n1 Fang, 99
epigenetic inheritance, 51 Fate, 214
equilibrium, punctuated, 34, 172, 173, feminist theory, 2
175, 182 fertility ritual, 106, 126
escape panics, 83–4 feudalism, 151–2, 172, 202
Eternal Mother cult, 96 feuding, 84–6, 165
Ethiopia, 142 Fisher, R. A., 1, 219
ethnographic record, 64–5, 107 food taboos, 55n14
etiquette-books, 56 Fourier, Charles, 181
Etoro, 67 Foxe, John, 118
eugenics, 219 France
Euripides: Supplicants, 189 anarchism, 180
Eurocentrism, 188, 194 ancien régime, 173, 175, 181, 187
Europe bourgeoisie, 143, 180, 213
Christianity, 163–4 burial customs, 96
feudalism, 151–2 Communism, 180
post-Roman, 152, 158 dechristianization, 177
see also specific countries fiscal reform, 186–7
Eusebius, 134, 135 infantry uniform, 46
Evans-Pritchard, E. E., 12, 67, 71–2n4, 85, 93, Marx, 181
210 post-Revolution, 175–6, 179–80, 181–2
evil, 78, 119 sale of offices, 184–5
evil eye, 108 share cropping, 60
evoked behaviour, 8–9 socialism, 180, 181
archaeological record, 35 suffrage, 178
child rearing, 64 working class, 181
chimpanzees, 91 see also French Revolution
Counter-Reformation, 49 Frazer, James George, 218
ethnographic record, 64–5 Fredegar, 46
face-to-face, 87–8 Frederick the Great, 82
feuding, 85 free will, 4, 23
motivational mechanisms, 82–3 freedom, degrees of, 162–3
and ‘other’, 68 Free masons, 103, 179
pervasiveness, 89 free-riding, 41, 79, 116, 134, 137, 144
rhetoric, 81 French Revolution, 173–8, 187, 189, 192
warfare, 34, 82, 118 frequency-dependence, 95, 115, 150
evolution of species, 195 friendships, 88
see also convergent evolution; cultural fundamentalism, religious, 130
evolution; social evolution;
‘sociocultural’ evolution Galapagos finches, 51
evolutionary biology, 2 game theory
evolutionary game theory, 57–8, 76, 219 collective behaviour, 83–4
evolutionary psychology, 66, 81, 84, 122, contract enforcement, 143
211–12 cooperation, 78
evolutionary theory, vii, 1, 27, 32–3 evolutionary, 52, 57–8, 76, 146n3, 219
evur, 99 merchant guilts, 37n6
exaptations, 46, 104, 198 Prisoner’s Dilemma, 16, 83
exchange, 81, 87, 91 Ultimatum Game, 70
exchange mobility, 170 Geertz, Clifford, 211, 212
expeditiones regum, 168 geld payments, 167
exploitation, 81, 122, 137 Gellner’s Paradox, 65, 111
extinction, accelerating, 224 generosity, 15–16
250 Index
genetics, 1, 50–1, 113, 128, 219, 220 Hobhouse, L. T., 218
German-Americans, 50 Hobsbawm, Eric, 220
Germany, 162n17, 186, 223 ‘hold-outs’, 27, 113–14
Ghana, 105n8 Holists, 62–3, 194
Gibbon, Edward, 135 holy man, 44n8
gift-giving, 14–15, 70, 133 honour, culture of, 85–6
Gladstone, William Ewart, 24 hoplites, 122, 123, 199
globalization, 190–1 hozho, 92
Gnau, 75 Huguenots, 110
goði, 142 human genome, 78, 219
gods, 99–100n3 Hume, David, 9, 66
Goebbels, Joseph, 81 humiliation, symbolic, 103
Goffman, Erving, 206 Humphrey, Caroline, 207–9
Golding, William, 90 Hunt, T. L., 183
Great Persecution, 131 hunter-gatherers, 35, 40, 41, 42
Greece, ancient hunter-horticulturalists, 70
hoplites, 122, 123 hunting and foraging, 63, 68, 78, 79
infantry, 121–2 Hutchinson, Francis, 67
kolax, 43 Huxley, Andrew F., 21
oligarchy, 200–1 Huxley, T. H., 21, 213
polis, 199–201, 203
see also Athens; Hellenism Iceland, 84, 142, 165
Gregory the Great, 129 Iconophiles/Iconoclasts, 129
Greif, Avner, 37n6 idiographic explanation, 194
group selection, 2, 20, 60–3, 80–1, 143, 144 Illinois, 50
groups, 63, 68 imagined communities, 110, 111
see also in-groups; out-groups imagistic memes, 102–4
Gudeman, S., 207 imitation, 57–9, 61–2, 95–6, 109, 112
guilt, 91 immediate-return economy, 40
imperialism, 125, 136
habitus, 45 see also empires
hacendados, 173 imposed behaviour
Haiti, 144, 158 conformity, 38–9
Haldane, J. B. S., 219 Durham on, 39
Halévy, Elie, 215 mutation of practices, 141–50, 171–3
half-belief, 93n1 slavery, 153–5
Halsbury, Lord, 146, 147 transition from acquired, 40–1
Hamilton, W. D., 61 in warfare, 8–9, 34
Hamilton’s Rule, 73, 84, 150, 195 Inca, 48
Hammurabi, 33, 165, 191 indentured labour, 155, 160
hand-washing, 54–6 India, 44, 105, 149
Hawk-Dove game, 52–3 Individualists, 62–3, 194
Hellenism, 92, 121–4, 203 industrial relations, 115–16, 144
heresies and orthodoxies, 128–31 Industrial Revolution, 147n4, 188, 194
heritable variation, 2–3, 4–5, 4–6, 32, 34, Ine, laws of, 166n20
37, 39 infectious disease analogy, 114
Herodotus, 121, 122, 142–3 information
heroism, 35–36, 36n3 acting-out, 37–8
Heyerdahl, Thor, 183 phenotype, 13, 59, 72
Hincmar, Bishop, 107 transmission, 30–1, 38–9, 109–10, 112
Hinduism, 119 in-groups, 51, 68–9, 73, 150
Historical Materialism, 18, 19–20, 29 inheritance, biological, 1, 15, 69, 88
history-writing, 196–7 behaviour, 64
Hitler, Adolf, 81 and cultural differences, 73
Hobbes, Thomas, 215 epigenetic, 51
Index 251
Gellner’s paradox, 65, 111 land-tax, 168
hunting and foraging, 79 Lassalle, Ferdinand, 1, 220
initiation rituals, 103 Lateran Council, Second, 129
innate disposition, 64, 65, 77, 81 law-codes, 33, 169
innovation, 5, 6, 32, 128, 216 Leach, Edmund, 105, 198
instinct, 4, 82, 84 leadership, 102, 141
institutions, 93, 96, 169 learning, 57–8, 61–2, 95, 109, 112, 141
definitions of, 3 Lebanese, 44–5n8
practices and, 39, 45, 63, 141 Lee, B. B., 42
roles and, 8, 42, 43n7, 141, 143–4 Lefebvre, Georges, 173–4, 182n36
intentionality, 12–15, 89 Lenin’s mausoleum, 101
intermediary roles, 148 Lévi-Strauss, Claude, 93
interregna, 43 Lewis, G., 75
introspection, 23, 204 liberty trees, 179, 179n34
Inuit, 44 light, manipulation of, 49n9
Ireland, 167n20 literacy, 35, 56
Islam, 103 livrets, 176
Isocrates, 201 Longinus, Gaius Cassius, 127
Italy, 161, 199–201 Louis XIV, 106, 149, 185
Louis XVI, 174, 178–9
Jackson, Thomas, 28–9 Louis-Philippe, 181
Jacobinism, 182 love, 74, 97, 132, 134
Jamaica, 150, 158, 159, 160 Lucian, 134
Jansenists, 178–9 Lugal-zagesi, 146, 146n2
Japan, 33, 113–14, 144, 151, 156n12 Luther, Martin, 130, 136
Jaspers, Karl, 202–4 Lyttelton, Humphrey, 6
Jesuits, 50n7, 113–14
Jesus Christ, 58, 119 Macaulay, Thomas Babington, 104, 194
Jews, 51, 59–60 MacCulloch, Diarmid, 110, 136
Johnson, Samuel, 93 MacDonald, K. B., 51
Joyce, Richard, 218 Macedon, 121, 201
Judaism, 51, 131, 203 Machiavellian intelligence, 2, 87, 90
Julian, emperor, 133 Machiguinga, 70
Juunt, 44 McNeill, W. H., 80n8, 81, 84, 210
Madagascar, 45, 105
Kachin, 49–50 magic, 93–4, 105, 106, 107, 126
Kampuchea, 153n8 Maitland, F. M., 169
Kant, Immanuel, 16, 21, 91, 96 maladaptations, 46, 183, 184
Kelly, Raymond, 62 Mamluk, 144, 154n9
kingship, 143, 168–9 Malinowski, B., 106
Kinnaurese, 44 mana, 92
kinship, 34, 84 ‘manifest destiny’, 19, 29
Kipling, Rudyard, 194 Mannheim, Karl, 97
knowledge, 97, 98, 105 manumission, 145, 154, 156, 160, 161, 164
Kolakowski, Leszek, 100 Maori emigrants, 50
kolax (flatterer), 43 Marcus Aurelius, 99, 125
Kuhn, Thomas S., 27, 223n2 Mardujarra, 44
!Kung San, 42, 93 Marie Antoinette, 178, 178n31
Kwakiutl, 78, 93, 154 Maring, 107
maroon communities, 156
labour migration, seasonal, 48 marriage, interdenominational, 76
Lactantius, Lucius Caelius, 133 Marshall, Alfred, 196
lactose absorption, 38, 80 Martineau, Harriet, 56
Lampadius, 14n6, 132 Marx, Karl, vii, 130, 139
land ownership/control, 152, 166–7, 170–1 base/superstructure metaphor, 7
252 Index
Marx, Karl (cont.) transmission, 56–7, 85, 109–11, 112–15, 129,
class struggle, 1, 17–18, 24 131–9, 159
on Darwin, 1 winning and losing, 192
France, 181 see also co-evolution, meme-practice
Das Kapital, 213, 220 Mendel, Gregor, 1
to Lassalle, 1, 220 mercenary warriors, 123–4
on money, 44 Mesopotamia, 194
morality, 218 metallurgy, 76n6
production, modes of, 152 metaphor in science, 6–7
proletariat, 216, 217 meta-representations, 90–1
revolution, 20 Methodism, 97–8, 159
social evolution, 17 metrical psalms, 110
teleology, 20 Michelet, Jules, 192
Marxism, 17, 100–1, 155–6, 220 middle class, 120
Massie, Joseph, 214–15 migrants, 61–2
Massilia, 200n1 military drill, 80–1, 84, 210
mathematics, history of, 128 milk drinking: see lactose absorption
Maupeou, René Nicolas de, 186 Mill, John Stuart, 22, 215, 216
Maurice of Nassau, 80 millenarism, 102
Maya, Central, 183 Milvian Bridge, Battle of, 130–1
Maynard Smith, John, 52–3, 218–19 mineral deposits, 147n4
Meiji Restoration, 33 miracles, 101–2
Melanesia, 142 mita, 161
Melville, Herman: The Confidence Man, 138 Mithraea, 49n9
Memes, 3, 29, 33, 53–4, 75 modernity, concept of, 189–90
adaptation, 54, 104–5, 123 modernization theory, 19, 188
archaeology, 58–9 Monod, Jacques, 216
beliefs, 67–8 Monophysites, 104
in books, 56 Montaigne, Michel Eyquem de, 65
coadaptive, 67, 98–9, 101, 109 monumental architecture, 149
collective behaviour-patterns, 95 Moore, Hannah, 56
competition, 31 moral relativism, 218, 221
cultural diversity, 65 moral sense, 16, 69, 91
cultural selection, 51–2, 62, 149 morality, 16, 216, 217, 218
diffusion of, 5, 97, 111, 129 Moriori, 50
displacement/replacement, 94 Mormonism, 100
doctrinal, 102–4 motivational mechanisms, 72–3, 82–3
emotional brain, 99–100 Mughal empire, 148
encoding of, 109–11 Muhammad, 130
heretical, 128 Mursi, 73
imagistic, 102–4 music, 75, 80, 81n9
imitation/learning, 9, 95 mutation
modernizing, 190 accelerating, 224
mutation, 39, 45, 113, 135, 139, 178, 181, 204 adaptive, 46, 130
phenotypic effect, 31, 54 biologically maladaptive, 46, 176
practices and, 31, 52 bridewealth payments, 62
recombinant, 135, 139 comparative sociology, 7
reinterpreted, 39, 92, 141, 181 critical, 140
related clusters, 119–20 imitation of, 95–6
reproduction of, 129 improvisation, 6
Roman Catholic Church, 49, 129 memes, 39, 45, 113, 135, 139, 178, 181, 204
Rome, 124, 127 practices, 139, 144, 163–4
sacralizing, 100
software analogy, 54 Nagasaki, 113
techno-memes, 52, 63, 112, 203 Napoleon III, 187
Index 253
Napoleonic wars, 118 paganism, 132, 133–4, 135
nationalism, 116–18, 120, 172 Paine, Thomas, 56
native informants, 89–90, 207, 209–10 Palaeolithic, 42–3, 77, 78, 102–3n5
natural selection, 46, 69, 171, 219, 222, 224 Paley, William, 4
Bayesian reasoning, 82–3 panic, 83–4
behavioural science, 2–3 panoplies, 123, 124
Campbell, 2 Papua New Guinea, 44, 61, 67, 70,
collective behaviour-patterns, 69 142, 191
conscience, 69 paradigm-change, 16, 27, 216, 217, 218
and cultural selection, 47, 61 parenting behaviour, 115
elephant’s trunk, 30 Paris, Peace of, 158
empires, 150 Parsons, Talcott, 19
evolutionary adaptedness, 79–82, 97 particulate selection, 1, 52, 52n10
gene transfer/hybridization, 33 passion/reason, 65–6
God, 25 pastoralists, 142, 151
intentionality, 13 paternalism, 150
and judaism, 51 patrilineal descent, 186
and moral judgements, 218 patrimony, 144, 168
neo-Darwinism, 4, 69–70 patriotism, 118, 120
probability theory, 83 patronage, 173, 177–8
Protestantism, 24–5 Pausanias, 123, 125n14
self-aggrandizers, 77–8 Peirce, C. S., 216
speed of, 112 Peleponnesian War, 200–1
natural/preternatural, 214 Peninsvlares, 111
nature-to-culture transition, 119, 145 Pericles, 117
Nazi Germany, 49n9, 162n17, 223 periodization, 202
Ndembu doctors, 105n7, 209 persuasion, modes of, 140
Neanderthals, 90 France, 176–7
Necker, Jacques, 174, 186 missions/schools, 148
Neolithic Revolution, 42 prestige, 185
Netherlands, 188 Quakers, 158–9
New England, 136–8 Roman, 127–8
newspapers, 110–11 Roman Catholic Church, 177
niche construction, 63, 79 scribal gentries, 148
Nietzsche, Friedrich Wilhelm, 21, 24, 25, 131, Peru, 48, 157n13
215, 221 Peter the Great, 141
Nisbett, R. E., 86, 119n12 peuple/bourgeoisie, 180
Nock, A. D., 97 pharmacology, 106
nomotheteis, 200 phenotypic effect(s), 11, 29, 37, 54, 94, 159, 189,
nomothetic explanation, 194 206
Norman Conquest, 166, 167, 169 armsgiving as, 132
Norse sagas, 108 artefacts as, 58
Nuer, 12, 62, 73, 85, 93, 210 of memes, 75, 96, 103, 108, 111, 117, 129, 183,
203
oligarchy, 200–1 of practices, 141, 183
Onon, Urgunge, 207–9 of Revolutionary ideology, 176
optimization, 32, 45 Philiphaugh, Battle of, 104
orators, 81 Philolaus of Corinth, 33
Orléans, duc d’, 179n32 philosophes, 178–9
orthodoxies, 128–31 phrenology, 28
otherness, 55, 68, 69, 73, 123 phylogenetic models, 33
Ottoman empire, 148, 153 Pictor, Fabius, 126
out-groups, 117, 150 Pinker, Steven, 75
overdetermination, 30, 116 plantation slavery, 154–5
Ovid, 125 Plataea, Battle of, 122
254 Index
Plato slavery, 155
Euthydemus, 221 tax-farming, 148
Laws, 121, 125 proletarianization, 159–60
Menexenus, 123 Protestant Ethic, 101, 136
poisoning, 106 Protestant Reformation, 110, 117
Republic, 125, 203 Protestantism, 24–5, 30, 45n8, 118, 136
Pleistocene, 22, 77–8, 79 protoconcepts, 72
Pliny, the elder, 93, 159 proto-households, 63
Pliny, the younger, 131 Proudhon, Pierre-Joseph, 181
podestà, 145, 200 providence, 29n1
poisoning, 106 Providence Island, 155
polis, 199–201, 203 psalm-singing, 110
politicians, 81, 87 public opinion, 96
Polynesians, 92, 142, 183–4 public sphere, 7n2
Ponapae Island, 184 Pukapuka atoll, 94
pongueaje, 173, 173n25 Punic Wars, 161
pontifex maximus, 128 punishment, 70n2, 150
pontifices, 125, 126 purgatory, 93n1, 129
Popper, Karl, 100 Puritan communities, 101, 136–8
Posidonius, 125
positron emission tomography, 75 Quakers, 158–9
possibilities, evolutionary, 195–6 Quesalid, 93
post-modernism, 220–1
power, 37, 96, 141, 143, 144 Radcliffe-Brown, A. R., 43
practices, 3, 29, 43, 48, 117 rain rituals, 93
adaptive/maladaptive, 148 randomization, 5
capitalist, 182 rational choice, 27, 66, 70, 75–6, 77, 84
memes and, 31, 52 reason/passion, 65–6
mutation, 139, 144, 163–4 rebelliousness, 129n16
recombinant, 139 reciprocity, 48, 81, 115, 134
renegotiation of, 39, 141, 181 Rectitudines Singularum Personarum, 167
reproduction of, 45 reeves, 166
social diversity, 65 Reform Bill, First, 158
social selection, 149 repartimiento, 161
winning and losing, 192 reproductive fitness, 79, 109, 222
predestination, 101 resemblances, 195, 198–9
Preemptive Periodization Fallacy, 202 reverse engineering, 32, 58, 101, 103, 107, 110,
pre-literate peoples, 35, 56 116, 129, 134
Pre-pottery Neolithic B settlements, 41 revolution, 20, 172–6, 182
prestige, 96, 185 rhetoric, 81
see also social status rhythm, 80–1
presuppositions, 210 Ricardo, David, 147n4
priesthood, 59, 95, 107, 125, 126–7, 142 Ricci, Father Matteo, 111, 113
primatology, 10, 14, 87, 89–90 Richards, Audrey, 108
see also de Waal, Frans Richerson, P. J., 50, 53n12, 60, 82
printed word, 110 Ridley, Matt, 86–7
Prisoner’s Dilemma, 16, 83 ritual
probability theory, 82–3 baptism, 132
production, modes of, 140 fertility, 106, 126
Athens, 189–90 ineffableness, 75
capitalism, 50 initiation, 103
France, post-Revolution, 176 for rain, 93
industrial/agricultural, 188 revolutionary France, 179–80
Marx, 152 Roman, 126
Index 255
Rivera, A., 207 sedentism, 41
Roggereen, Jacob, 183 self-aggrandizers, 77–8, 143, 165
role distance, 13 self-consciousness, 21–2, 88–9
role-dyads, 39, 59, 62–3, 143–4 self-interest, 27, 75–6
roles, institutional, 3, 5, 42, 43n7, 141, 143–5 self-respect, 21–3
Roman Catholic Church, 47, 49, 59, 129, 165, self-sacrifice, 60–1
177 Semai, 35–6
Roman Empire serfdom, 97, 152, 163
absorption of Latins, 145 Settlement, Act of, 118
and British Empire, 149n5 settlement sizes, 41
Christian holy man, 44–5n8 Sewell, W. H., 181
collapse of, 164, 183 sexual selection, 2, 22
and corruption, 146n3 Shakers, 46–7
reform, 147 shamanism, 93, 207–9
sycophants, 43 shame, 91
Roman law, 126–8 sharecropping, 60, 152n6, 176
Rome, ancient, 67 Sher Ali, 190
benevolentia, 132 Sicily, 161
Christianity, 134 Sidgwick, Henry, 215
conformity, 126 silver mines, 147n4
cultural traditions, 124–5 Simon, Herbert A., 31
divine providence, 125 skin colour, 156–7
oligarchy, 201 Skinner, Quentin, 13n5
persuasion, 127–8 skyhooks, 4
priesthood, 107, 125, 126–7 slave-breeding, 156n11
ritual, 126 slavery
slaves, 48, 156n11 abolition of, 156, 158–61
Weber, 161–2 Athens, 156–7
Romulus Augustulus, 147 Brazil, 160–1, 162n18
Rossbach, Battle of, 174 Caribbean, 157–8, 159–60, 161
Rousseau, Jean-Jacques, 114n10 chance, 154–5
routinization, 141 Christianity, 163–4
Russia, 175, 186, 194, 198 comparative sociology, 152–3
Rwala Beduin, 141 distribution of, 153–4
Ryle, Gilbert, 211 England, 165, 166
Europe, 161–4
sacralization, 99, 100 exploitation, 122
Sahlins, Marshall, 95, 142, 165 fugitives, 156
Saint-Simon, Claude, 181 as locally optimal design, 153–4
Salamon, Sonya, 50 manumission, 145, 154
sale of offices, 184–5 Quakers, 158–9
salvation ethic, 97, 125 rights, 162
Samuel, Ralph, 193 Roman, 48, 156n37, 161–3
Saxe, Maréchal de, 80n8 skin colour, 156–7
Schumpeter, Joseph, 100 sugar, 80
Scot, Reginald, 67 Weber, 131, 162n17
Scotch-Irish immigrants, 86 Smith, Adam, 66, 144, 159, 196, 215
Scotland, 138–9, 144 Smith, Joseph, 100
see also Covenanters Sober, E., 62
scribal gentries, 148 social contagion, 115
Searle, John, 37n6 social contracts, 78, 143
sectarian communities, 96–7 social control, 36, 41
sects, 97–8 Social Darwinism, 4, 18–19, 29
secularization, 126–7, 129–30 Social Democratic Party, Sweden, 111
256 Index
social mobility, 43, 83, 128, 145, 154, Stoics, 125, 132
169–70, 195 strangers, 68
social selection see also xenophobia
in Anglo-Saxon England, 171 strikes, 94–5
archetypal story of, 140 strong reciprocity, 15, 79, 122, 134, 137,
Christianity, 130–1 144, 150
and cultural selection, 40, 47–8, 50, 60, 97 structural mobility, 170
as driver, 65 subjective experiences, 205–6, 207
empires, 150 subordination, 150, 153
neutral between designs, 140 suffrage, 178
practices, 149 sugar, 80, 80n7
social status, 77, 78, 79 Sulloway, F. J., 129n16
see also prestige Sumeria, 41–2, 99n3, 146n2
socialism, 180, 181 superego/id, 66
socialization, 112–13, 128–9 supernatural, 93, 105, 120, 126n15, 131–2
society Supplementary Militia Act, 118
in decline/collapse of, 183 survival instinct, 84
diversity, 65 survival of the fittest, 18
Durkheim, 100 Sweden, 45n8, 52n11, 111
imposed behaviour, 34 sweet tooth, 80
and state, 168 Switzerland, 143, 199
‘sociocultural’ evolution, 3, 37, 39 Sybilline Books, 126
sociology sycophants, 43
evolutionary process, 32–3 systactic structures, 87, 120, 125, 143, 148,
human/primates, 86–8 165–6, 199
native informants, 209–10 systacts, 39, 60, 77, 97, 117, 127, 144, 145, 163,
Spencer, 17 170–1, 178–9, 180, 186
taxonomy lacking, 198–201
warfare, 34–5, 36 Table of Ranks, 141
see also comparative sociology Tacitus, 106, 127
sociology of knowledge, 97 talala, 106
sociology of religion, 97, 129–30 Tasmania, 145
sociology of sociology, 218–19 tax-farmers, 148, 184–5
Socrates, 123 Tay-Sachs disease, 51
Solon, 121, 141 technological advances, 109–10
Solow, Robert M., 6 techno-memes, 52, 63, 112, 203
Soltis, J., 61 Telemachus, 36n4
sorcery, cure for, 108 teleology
Soviet Union, 101, 153n8, 188 demise of, 214
Spanish conquistadors, 33 Marx, 20
Sparta, 199, 200 modernization theory, 188
Spartacus, 160 theories of progress, 218
Speer, Albert, 81 Spencer, 19
Spencer, Herbert, vii and variation, 2
competition, 17, 24 Thailand, 154
Descriptive Sociology, 119 thegns, 166, 168, 170
morality, 216, 217, 218 theory-invention/retention, 66–7
sociology, 17 Theseus, 189
teleology, 19 Thessaly, 199
Stark, Rodney, 134–5 thought contagion, 114
Statto-Città, 199–201 Thucydides, 35, 117, 189, 189n37, 202
stereotyping, 68 time, concept of, 73
stick-insects, 182–3 Tocqueville, Alexis de, 194
stigmatization, 68, 79, 105, 117, 122–3, Tooby, John, 211–12
124, 169 trade-unionism, 116, 144
Index 257
trading routes, 146–7 wage-labour, 47–8, 144, 157, 159, 166, 173, 186
traditions, 92–3, 95, 124–5, 150 Wala, 76
Trajan, emperor, 131, 133 Wallace, Alfred Russel, 219
transcranial magnetic stimulation, 76 warfare, 8–9, 34–5, 36, 82, 107
tribalism, 85, 165 warlords, 148
Trilling, Lionel, 216 warrior culture, 121–3, 124
Trobriand Islanders, 106 Watson, Thomas, 29n1
Twelve Tables, 126–7 wealth, 141–2
Webb, Beatrice, 217
Ultimatum Game, 70 Weber, Max, vii
umialik, 44 adaptation, 20
ultra-Darwinism, 7–8, 53 bureaucracy, 152
United States of America, 194 charisma, 141
as big society, 150 disenchantment, 213–14
Constitution, 49 elective affinities, 7
Declaration of Independence, 156 ideas as switchmen, 139
homicide, 86 morality, 216, 217, 218
manumission, 156 patrimonialism, 168
Protestant capitalism, 136–8 Protestant Ethic, 101, 136
rational choice of religion, 76 on Roman economy, 161–2
Scotch-Irish immigrants, 86 salvation, 97, 125
slavery, 80n7, 153, 154–5, 156, 161 slaves, 131, 162n17
wage-labour, 157 wergild, 171
universals, 71–4, 84, 131–2 Wesley, John, 112
utilitarianism, 215 whales, 195–6
utility-maximizers, 27 Whitehouse, H., 103n5
Wickham, Chris, 59
Valens, emperor, 147 Wilberforce, William, 213
valetudinaria, 133n17 Williams, Bernard, 222
Vansittart, Charles, 29n1 Wilson, D. S., 62
Varro, 48, 161 Wilson, E. O., 69
vassalage, 144, 151 Wissenschaft, 213, 216, 217, 221
venality, 184–5, 186, 187 witchcraft, 67, 105, 108
Venice, 200n1, 200n2, 201 witch-doctors, 93
vernae, 156n11 Wittgenstein, Ludwig, 71n3, 216
Verres, 125 Wolf, Eric, 218
vervets, 88n10 working class, 94–5, 116, 120, 181
Vespasian, emperor, 132 World War II, 82
Victoria, Queen, 147 Wulfstan, Archbishop, 170, 171
Victorians, 204
Vienna, Congress of, 158 xenophobia, 120
villani, 170 Xenophon, 82, 124
Villaroel, Gualberto, 173n25
violence, 8, 35, 36, 85–6 Yair, G., 76n5
Visigoths, 46
volunteers, 118 Zande witch-doctors, 93
voter turnout, 205 Zoroaster, 203, 204