COGNITION AND HISTORY: THE EVOLUTION OF

INTELLIGENCE AND CULTURE

Henry Plotkin
University College London

The Inaugral Spinoza Lecture on Science and Society

delivered to the International Institute of Social History
(Internationaal Instituut voor Sociale Geschiedenis),
Amsterdam, in October 1996

Mailing address:
Department of Psychology
University College London
London WC1E 6BT
England

e-mail address:
henry.plotkin@ucl.ac.uk
 2

There is a wonderful complexity to the matter of human evolution, those

qualities of the human mind that we call intelligence, and how each has influenced

the other. Human intelligence has evolved, and human evolution has been

influenced by human intelligence. A crucial, perhaps defining, feature of human

intelligence is our unique capacity to enter into that thing we call culture. So we are

concerned here also with the evolution of culture and the way in which culture has

affected our evolution. These are profoundly important matters in our

understanding of what it is to be human and how we came to be so. Central to the

scientific understanding of these matters is the issue of history as cause, and so it

is there that I begin.

The Causal Force of History

It will come as no surprise to social historians that history matters, to use

S.J.Gould's phrase (Gould, 1986). However, this has not always been a universally

accepted position in science. Two examples of how causes are framed in science

will illustrate different approaches to the issue of history as having causal force.

The first is that of the Russian chemist Mendeleev who invented in the middle of
 3

the last century the Periodic Table of the elements, building on Dalton's earlier

proposal that each element has a characteristic atomic weight. The Periodic Table

was a brilliant insight, being based on Mendeleev's understanding that atoms have

an internal structure. It is this structure that causes elements to have the properties

upon which Mendeleev ordered the elements in his Table; and it is this structure

that enabled him correctly to predict the existence of elements, and their properties,

not then known. When later they were discovered, the Periodic Table and the

theory underpinning it was vindicated and Mendeleev's fame was ensured. Now the

point is this. The Periodic Table, a fabulously successful scientific insight, was

based on the assumption that the laws of chemistry are absolutely constant and

have always held. Had Mendeleev lived and worked 4.5 thousand million years

ago, when the Solar System formed, he could have used the same observations and

insights then to construct the Periodic table as he did in the 19th century. For

example, since the beginning of time on this planet, when the elements sodium and

chlorine have been mixed together they have formed a compound, table salt, which

has rather different properties from those of its constituent elements. This accords

with the principles upon which the Periodic table was constructed and as a 19th

century chemist, Mendeleev assumed not just that observations that he made of

table salt on Monday would be the same on Tuesday and Wednesday, but that he

could have made the same observations and drawn the same scientific insights

thousands of millions of years ago. He was correct in this, but the general
 4

conclusion that was drawn that there is no temporal element to the laws of

chemistry, that time in the form of historical antecedence does not have causal

force, we now have reason to doubt.

Our second example comes from Darwin, who learned from 19th century

geologists another way of thinking about causation. When Darwin visited the

Galapagos Islands he observed over a dozen different species of finch, sometimes

unique to different islands in the group and each distinctive, especially in their

beaks and feeding behaviour. He concluded that these different species were all

descended from a single mainland species, individual members of which had

sometime in the past found their way to one of the islands, the descendent species

resulting from the gradual migration of birds to other islands and the relative

isolation of the islands from one another. This explanation of Darwin's finches was

just one case, one illustration, of Darwin's wider thesis which asserts that all living

things are related because all of life forms a tree-like structure that is descended

from a common origin. Darwin's theory included an exposition of the processes

that drive evolution, and these processes have the same quality of apparent

timelessness as does atomic structure; whenever and wherever evolution occurs,

these processes must be present. But the nature of causal explanation of Darwinian

theory embraces historical antecedence as cause, and hence is very different from

that espoused by Medeleev. In order to understand why a finch has this beak shape

now, today, one has to invoke not only those universal and timeless processes of
 5

variation, selection and the transmission of selected variants, but also the form of

beak that these processes acted upon in the past, yesterday.

It is important to distinguish the "history-as-cause" explanation of

evolutionary biology from the covering law model of explanation of the logical

positivists, where an assumed general law is applied to specific instances. An

example of the latter would be the application of Newton's law of gravitation,

which is the covering law, to explain the specific instances of how far golf balls

will travel when struck with equal force here on Earth or on the moon. In

evolutionary biology, however, historical antecedence as cause asserts that a beak

form of a finch in the past is part-cause of the current form of beak; that if the past

beak form had been different, then so likely would be its current form. In

evolutionary biology, where you are coming from is a cause (not just an instance)

of where you are now. The very essence of evolutionary explanations is that history

has causal force.

We now understand that even the Periodic table and the laws of chemistry

that it entails have not been constant in time. Shortly after the origin of the universe

no elements existed; the elements emerged over a period of millions of years as the

universe developed. Thus, cosmologists and astrophysicists tell us that history has

causal force in chemistry too, and that with the exception of the deepest laws of

quantum theory, there is no causal constancy in the universe - 19th century science

was just fooled by the slowness of change. But evolutionary change in biological
 6

systems is on an altogether different temporal scale. There are now well

documented cases, including incidently Darwin's finches, of evolutionary change

occurring over just tens or hundreds of years. This is a drop in the ocean of

geological time. Much of evolutionary change does, it is true, occur over longer

periods, but nonetheless the rate of change is such that it is measurable and hence

tangible: you can put your hand on a fossil of early humans, or a stone axe, and

actually feel the historical causes that have shaped our present existence. We can't

do that with physics or chemistry.

Well, Darwin's finches are just one very, very tiny branch on the tree of life

that extends back thousands of millions of years. Another, even smaller, branch is

that of the Hominoidea, a grouping of apes within the Primate order that includes

human beings. Best present evidence, which is molecular, points to the chimpanzee

as our nearest living relative, and indicates a common chimpanzee-human ancestor

to have lived around five to six million years ago. Of significance to the issue of

cognition and cognitive evolution, which is my theme, is what might have

happened during the period extending from the first appearance of human-like

creatures, which is usually put at about two to two and a half million years ago, to

the origins of Homo sapiens, around one hundred and fifty to two hundred and

fifty thousand years before the present. In that two million year period the human

brain doubled in absolute size and quadrupled in size relative to body mass. Over

this same period there was little or no increase in the brain size of any other species
 7

of primate - indeed, there is no known comparable increase in brain size over so

short a period of evolutionary time in any other species at any time in the history

of life on Earth.

So, whatever else human beings are, we are a relatively recently evolved

species and our evolution has involved unprecedented change in the organ system

that subserves those mental capabilities that we call intelligence. If we are really

to understand the human mind then, it seems to me, we have to take into account

the evolutionary, historical, causes that shaped our minds and made them what they

are now. There has been, to some degree there still is, a reluctance on the part of

social scientists, and other scholars within the humanities, to accept that the human

mind is a product of evolution just as is the anatomy of our hands or the way in

which we metabolize carbohydrates. This is, I think, a profoundly mistaken, and

in effect an anti-scientific, stance. A true and complete understanding of the mind,

including intelligence, has to include an evolutionary perspective. We don't yet

understand how to do this and what its implications really are, but I will try here

to sketch out some of those implications and possibilities.

The Origins of Intelligence

The psychometricians' conception of intelligence as something that people

have, that is measured by IQ tests, and which varies from individual to individual,

is not the conception of intelligence that I am considering here. There may be some
 8

quality of mind that can be measured in this way, but it is an irrelevance to any

biological approach to intelligence, and it is this more general conception of

intelligence as a biological phenomenon that needs some explanation here. All

living things are localized gradients of negative entrophy, regions of order and

organization, in a universe pervaded by positive entrophy, by increasing disorder.

The only way in which such states of negative entrophy can be developed and

maintained is by exploiting energy sources. It takes energy to maintain these

negative gradients. Unable to utilize the energy of the sun directly in the way that

plants are able to do, which trap solar energy within a complex chemical web, most

animals have to make contact with and feed off the energy of organisms that are

able to do this. Putting it rather abstractly, animals need to maintain some form of

matching relationship between themselves and these indirect solar energy sources,

and their principal way of doing this is by moving about in, and acting upon, the

world. In order to be successful in maintaining this matching relationship, the

behaviour of animals must be guided by information. If the information comes only

via genes that direct nervous system wiring such that sensory inputs lead to

relatively fixed behavioural responses, then that behaviour is called an instinct. If

some of that information comes from memory stores in the brain that have been

laid down by individual learning, then the behaviour is intelligent. The internal

mechanisms of learning, memory, thought and creativity which have evolved to

supply this guidance is what I call collectively intelligence.

 9

In broadest terms, then, it is the source of information guiding behaviour that

determines whether we think of it as instinctive or intelligent. Thus defined,

intelligence is probably quite ancient, first appearing some five hundred million or

more years ago. Obviously, it is not confined just to humans, and it takes many

forms. Since information is order, it is energetically costly, and trade-offs between

costs and benefits must have evolved. G.C.Williams' (1966) principle of the

economy of information which asserts that living things will evolve in such a way

as to maintain information in the cheapest form possible is one theoretical

expression of such a trade-off. Now, because instincts cost less than learned

behaviour both in genetic and nervous system terms, and because the nervous

system is one of the most metabolically expensive tissue in our bodies, Williams'

principle tells us that there had to be powerful and pervasive forces selecting for

the evolution of intelligence. We think we know what the source of these forces

are.

The rate at which information can be acquired by a species and stored within

the gene-pool of that species is limited by the rate at which individuals can

reproduce themselves, that is, the period that extends from individual conception

to when that creature is itself reproductively competent. In a small rodent this is a

period of a few months; in humans it is in the region of twelve to thirteen years.

These are finite periods of time that sometimes occupy significant proportions of

the average life-span of an animal. In humans it approaches one fifth of our lives.
 10

Any changes in the world that occur at rates higher than this rate of generational

time cannot be detected by, and hence information about such changed conditions

cannot be incorporated into, the gene pool. If these changes are important to

survival, and many, like the spatial position of vital resources or the identity of

social allies and enemies, are, then such animals will have evolved additional,

individual information-gaining processes that are capable of tracking such rapid

changes in their world. In most general terms, this temporal sampling limitation of

genetic information-gaining processes is what has caused the evolution of

individual intelligence in animals, and maintains it in place despite its high energy

costs.

Some Implications

The previous section is a highly condensed form of a rather lengthy

argument that derives from a number of sources, which have recently been brought

together between single covers (Plotkin, 1995). This thesis on the origins of

intelligence as rooted in the need for some animals to track high rates of change

has a number of important implications, just two of which are presented here. One

is that the tabula rasa view of mind and intelligence is untenable; the other is that

genetic reductionist accounts of the evolution of any species whose individuals are

intelligent are unworkable.

 11

The British Empiricist philosopher, John Locke, argued that the mind is at

birth like a blank slate upon which experience writes. Hence he denied the

existence of innate or a priori knowledge. The notion of a blank slate became an

important part of empiricist thinking and has also had a strong following in the

scientific psychology of this century. It has been at the centre of most of

associationist learning theory and of behaviourism, and it has been applied as much

to animals as humans. It has also usually had an appeal to the liberal-minded non-

scientist, suggesting as it does that humans as cognitive generalists can learn and

think anything; any failure to do so is because of impoverished experience or

inadequate teaching. In the context of the age-old debate about the relative

contribution of nature and nurture to human disposition and behaviour, the tabula

rasa position is a vote for nurture and against nature. The evidence, however, tells

us that this is wrong. I will give just two examples. One comes from the study of

animals the significance of which is that what we know has been achieved with

rigorous experimentation.

We know with the certainty that experiments give that songbirds have to

hear the song of their fellow males at a certain age if they are to acquire normal,

species-typical song. What is extraordinary is that if they are exposed to the song

of another species during this song-sensitive period, they acquire neither their own

species' song nor that of their "mentor". They have to hear their own species' song

for normal song to be acquired. In other words, these are animals that come into the
 12

world knowing what it is that they have to learn. This wonderful contradiction,

this paradox for any follower of Locke, is resolved by the realization that cognitive

abilities are the products of evolution. All of the processes that we tend to think of

as being in the service of nurture themselves have nature, that is, they have

evolved, and so cognitive modules are primed or predisposed by information

coming from the genes. There are other examples from other kinds of animals,

ranging from insects like honey bees to mammals like voles, of similarly

constrained learning. As a general rule, animals learn what is good for them,

Konrad Lorenz's (1965) great insight, and that can only mean that learning is a

product of evolution.

A similar account can be given of human learning and intelligence, but in

our case, because of the limitations placed on experimentation, the evidence is less

direct. The famous example is language. At birth infants have a pronounced

sensitivity to human speech sounds; this rapidly narrows to the speech sounds

specific to the linguistic environment in which they are being raised; the pattern

and rate of language learning is remarkably uniform across languages, irrespective

of which of the world's five and a half thousand documented languages is being

acquired; the rate at which new words are learned is breathtaking given that it

occurs without formal tuition from caregivers or others; the same extraordinary

effortlessness marks the acquisition of syntax and grammar, which again is almost

always entirely untutored - this must be so because few adults are aware of the
 13

rules governing their own language. All of this, apart from the sensitivity to speech

sound, applies as much to deaf children who are raised within a linguistic

environment of handsigning as they do to children with normal hearing who are

raised to speak Dutch or Zulu. The fact that language is a human-specific learned

trait, and that its acquisition has never been explained using generalist learning

principles like those of associative learning, adds further weight to the Chomskian

view (see Pinker, 1994 for a recent review) that language is the product of an

innate organ of mind, and that learning a language is the result of a specialized,

predisposed, cognitive module. In the same sense that I applied the phrase to

songbirds, we humans come into this world knowing that language is one of the

things that we have to learn.

There is a very similar story to be told for facial recognition, the

understanding of cause-effect relations in the physical world, and the attribution of

intentional mental states to others (see Johnson and Morton, 1991; Hirschfeld and

Gelman, 1994; Sperber, Premack and Premack, 1995 for reviews). Any one of

these examples destroys the notion of the mind as a tabula rasa. But there is yet

another reason to doubt the blank slate concept. When we come into the world, it

is initially an unlabelled place (Edelman, 1987) which can be partitioned into a

virtually infinitely large number of ways. That is, the number of things that we

could learn from our quite limited sensory input combined with our ability to move,

is huge. We live within this vast space of things we could learn, and if our
 14

cognition were not predisposed, if it did not point us into the right parts of this

massive search space, the chances of our learning the things that it is important to

know are very small indeed. This is because the tabula rasa principle says that if

you can learn anything, if the slate is really blank, why then given enough time,

you will learn everything. But you would have to be given enough time. Learning

the important things then becomes a ponderous and chance-ridden process. Yet the

very reason why it evolved was because of the need to track rapid change. Slow

and haphazard is just what cognitive processes are not because if they were they

would have little adaptive value and would not have evolved at all. We know that

in those species that have evolved intelligence, cognitive skills are indeed rapid and

economical mechanisms that fit with the general life-history style of the learner;

intelligence and instinct are closely interwoven to the point that it is not inaccurate

or fanciful to think of learning as an extension of instinct. To repeat the point, but

the point is so important it bears repeating many times, resolution of the old nature-

nurture argument comes with the understanding that nurture itself has nature. In

technical jargon, learning is a consequence of the operation of a semi-autonomous

information-gaining device that is nested within the larger, more encompassing

information-gaining processes of evolution whose products are stored in genes.

In summary, most evolutionists and evolutionary psychologists believe that

the correct way to think of intelligence is, first, that it is a characteristic that is

highly constrained by a species' evolutionary history; second, that the word is best
 15

confined to a generic use; third, that contemporary cognitive psychology is correct

in thinking in terms of intelligences which are relatively discrete, discontinuous

cognitive skills; and fourth, that neither humans, nor the members of any other

species have the capacity to learn and think about anything and everything. One

way of saying all this is that the ideal of a rational, free-thinking, human being is

a fiction - or so rare as to be, in effect, a kind of freak of nature. Shakespeare, of

course, existed but was probably not very good with mathematics, or might have

been less than skilled with his hands. Newton was good with sums but couldn't

write sonnets. And most of us are entirely ordinary in all these respects. We

humans are not general-purpose thinkers and problem-solvers who can turn our

minds to anything. We are very restricted in what it is we can learn and how we

think, yet within these limitations we are economical and effective cognitive

agents.

The second implication concerns the spectre of a rampant and powerful

biological science taking over the social sciences with unstoppable reductionist

arguments and explanations, which has been with us for over a century. However,

an evolutionary approach to intelligence, which relates closely to its non-tabula

rasa nature, tells us that such reductionist arguments could never succeed. The

counter-argument goes like this. Nature is rather clever in the way that it equips

learners with knowledge about what it is they have to know if they are to survive.

It does this by feeding information gained a posteriori, through history, at the

 16

species level to the individual as a learning or reasoning predisposition where it

appears, for the individual, as a priori knowledge. The empiricist denial of innate

knowledge is wrong; the age-old rationalist view is closer to the truth and Kant's

notion of the a prioris is much closer to the way modern science views these

matters. (It is worth noting in passing that science can now adjudicate in such

philosophical disputes.) However, whilst Nature is clever, it is not prescient.

Indeed, it evolves these semi-autonomous information-gaining devices precisely

because it is not prescient. And in doing so, it in effect passes some of the causal

responsibility for behaviour onto these semi-autonomous devices. Intelligent

creatures construct in their own minds and brains the detailed picture of the world

that their genes only hint at, crucial as those hints might be in navigating through

to the right place in knowledge space. In so doing, they construct for themselves

the causes of their own behaviour. Were this not the case, if the slate were

completely written on by genes, then we would be dealing with instinct and the

causes would reside in genes and in the developmental processes by which they are

translated into action. The fact that we do make life and death decisions concerning

who we co-operate with, which resources we exploit, where we live, how we move

through space, and with whom we choose to reproduce ourselves, all on the basis

of information that comes to us through these semi-autonomous information-

gaining devices, means that any proper causal analysis of the evolution of species

whose individuals are intelligent has to locate some of those causes within the
 17

brains and minds of those individuals. Towards the end of this lecture I will give

some specific examples. The general point being made here is that genes and

development are not enough. Social scientists and scholars of the humanities quite

simply have nothing to fear from a powerful, reductionist biology. What has

happened in molecular biology since the genetic code was unscrambled by Watson

and Crick over forty years ago has been magnificent. And I don't doubt that it will

help in developing the human sciences. But it is never, on its own, going to do the

whole job.

The image I would like to leave you with is this: the slate is not blank. Both

evidence and argument tell us it is not and cannot be so. But neither is it wholly

written upon. The temporal sampling limitation of genes requires that some part of

the slate is not filled in at birth. However, because of the operation of Williams'

principle, what space is left open for experience to fill in is the least amount Nature

will allow, and it will vary with the general life-history and life-style of a species.

Nature is forced into this delicate balance in the matter of how much must be

written on the slate, but one thing is clear. At birth, something is always written

there. We are all possessors of innate knowledge upon which we then build.

The Evolution of Culture

The evolution of individual intelligence as an adaptation for tracking rapid

change and adjusting to it, powerful and effective as it is, nonetheless is like any
 18

other kind of adaptation in that the consequences of individual intelligence, the

memories and the products of reasoning and thought, remained confined within

each individual organism for hundreds of millions of years in all but a single

species, or perhaps the species of the single genus Homo. Any population-level

effects came about only by the adaptive consequences of having the capacity for

individual intelligence feeding back into changes in gene-pool constitution by the

usual route of overall individual fitness gains and reproductive success. The precise

form or content of the adaptations formed by intelligence - this food is safe, that

face is not to be associated with - could not be transmitted to others and remained

locked within each intelligent creature that had gained the information. In the

context of major evolutionary events (Szathmary and Maynard Smith, 1995), the

appearance of individual intelligence was not a major transition in the history of

life on Earth, on a par with, say, the evolution of self-replicating molecules or

multicellularity. Individual intelligence was, however, a necessary precursor to

what was an evolutionary watershed, a profoundly important event in the history

of life the consequences of which we are seeing unfold before us - this was the

evolution of culture which, among extant species, is unique to Homo sapiens.

Literally hundreds of definitions of culture have been offered over the last

150 years, which can be classified in a large number of ways (Kroeber and

Kluckholm, 1952; Keesing, 1974). Many revolve around the products of culture,

be they artefacts, ideas or behaviour. A better approach, I believe, is to define

 19

culture in ways that make for an analysis of the phenomenon in terms of processes

and mechanisms. For that reason I define culture simply as shared knowledge. If

you can share what you know with others, and are able to acquire knowledge from

others, then you are a creature capable of entering into culture. There are

documented cases of a few other species, like bees, songbirds and chimpanzees,

being able to acquire information from one another. There are, though, several

features of the human capacity for learning from others that make human culture

different from anything that birds and bees can do (Tomasello et al, 1993).

Primarily, these differences concern the way in which, pathology apart, literally

every member of the social group shares in knowledge, and that knowledge is of

many different things - linguistic, dietary, dress, and abstract beliefs amongst many

other forms of knowledge. By contrast, in chimpanzees for example, only a limited

number of individuals will share a small number of isolated behaviours; and in

bees, while most members of a hive are able to share information about the

location of resources, that is all that is shared. The sheer astonishing richness of the

range of what humans share with one another marks culture out as different from

anything seen in any other species. A second unique feature of human culture is

incessant, cumulative, modification of knowledge and practice over generations.

Tomasello calls this the "ratchet effect". The conservation and elaboration of stone

tools over hundreds of thousands of years is one example; the growth of the form

and use of computers in the last four decades is an example of a spectacular ratchet
 20

effect. Furthermore, human culture is marked by an extraordinary specificity and

detail of shared knowledge which no other animal exhibits. What this all means is

that the richness, the precision, the range, and the durability of human cultural

knowledge is such that if one is looking for insights and understanding of culture

from biology, comparisons with other living species will not help.

Szathmary and Maynard Smith consider language to be a major evolutionary

event, and the reason for their choice is that with the appearance of language came,

for the first time in evolutionary history, the means of transmitting large quantities

of information between individuals which do not involve genetic transmission. This

truly is a major evolutionary event, but language cannot be the only crucial

psychological process that was necessary for culture to appear in humans, and it

does seem to me that it is culture, not just language, that is the major event, with

language being one of the necessary component processes of culture. The other

processes necessary for culture can, perhaps, be gleaned by considering one of the

most important characteristics of human culture, which is the existence of social

constructions like money, marriage and justice. All social constructions are based

on agreement. It is only because we agree that a fifty Guilder note has sufficient

value that it can be exchanged for enough goods to make a meal, or for the

purchase of a book, that it has that value. In itself, the piece of paper has miniscule

intrinsic worth. And it is only because we agree in societies like our own that

fairness should underpin our notion of justice that our beliefs and practices with
 21

regard to justice have the characteristics that they do. The notion of fairness as

central to justice is not some immutable, inexorable consequence of human

biochemistry and present in all humans. Some cultures have quite different social

constructions of justice based on social status, revenge or religious precepts. Justice

is what it is in Holland or England because we agree that it should be so, and for

no other reason. Put simply, some things exist only because we all agree to think

that they exist. Parenthetically, some biologists are sceptical about the existence

of social constructions and suggest that they are part of the mythology of the social

sciences. This is a curious stance to take because it is manifestly the case that

social constructions have real causal force. People live in the manner that they do,

and die as they do, often in large numbers in wars, because of them.

Now, language alone cannot explain social constructions. I agree with Searle

that "what is special about culture is the manifestation of collective intentionality

and, in particular, the collective assignment of functions to phenomena where the

function cannot be performed solely in virtue of the sheer physical features of the

phenomenon" (Searle, 1995 pp 228). The power of money does not lie in the coin,

nor that of justice in the buildings that house our courts or the people that operate

the system. "Collective intentionality" is where the power lies, and this is a very

difficult thing to explain. In the last 18 years or so, centre stage in developmental

psychology has come what is known as "theory of mind", which is the

understanding that other people have minds as well as ourselves. We now know
 22

something of the way in which children come to understand that intentional mental

states like wanting and knowing exist in the minds of others; and we have also

come to understand the catastrophic consequences for normal social function when

the ability to attribute mental states to others fails to develop. We now believe that

the development of a theory of mind is one of the most important features of

human cognition, because it lies at the heart of social causation - we understand

people as causal agents by way of the attribution of intentional mental states.

Although the phrase theory of mind was first used in a classic paper with regard to

chimpanzees (Premack and Woodruff, 1978), there is no evidence whatsoever that

any other species is able to attribute intentional mental states to other indiviuals.

It is, like language, a human-specific characteristic, and it must have a crucial role

to play in establishing agreed or "collective" intentional states. It must be the case

that the capacity for attributing mental states to others is a necessary prerequisite

for the capacity to recognize that one's own mental states match those of others.

The kind of agreement that social constructions require can only come into being

between individuals who possess a theory of mind.

Another important aspect of human psychology that is necessary for culture

has been known to social psychologists for many years. Social forces such as

conformity, obedience and cohesiveness are known to operate with real causal

effect to establish or shift intentional states, and the direction of shift is usually

towards some social norm, towards an agreed or "collective" intentional state. The
 23

classic studies are those of Muzafer Sherif, which were carried out in the 1930s.

I am suggesting that, at a minimum, the extragenetic transmission of

information, a theory of mind module, and the evolution of a responsiveness to

social force, were the elements necessary for the appearance of human culture.

Whether through co-operation, coercion or education, we establish agreed mental

states and values on matters that relate in large part to the existence and continuing

function of the social group. Culture is truly a population-level or group-level

phenomenon. In my view it is impossible to understand it in any other way, and

this is another reason for believing that biologists armed with reductionist strategies

cannot succeed in explaining the social sciences. It must be added for those who

are knowledgeable about the problems raised in the past for the notion of group

selection by evolutionary theory that there is now an increasing acceptance that the

concept of group selection can be defended (Wilson and Sober, 1994)

It should also be noted that I have been referring to extragenetic transmission

of information at times rather than language. This is a caution born of our not

knowing, and perhaps we will never know, the order in which these essential

components of culture that I am advocating here appeared in human evolution.

There are some experts on human evolution, for example Tobias (1995), who

advocate the view that a form of language may even have existed in our

australopithecine ancestors, the ape-human creatures that mark the beginnings of

hominid evolution some four to five million years ago, and certainly existed in
 24

Homo habilis, the earliest known species of the genus to which modern humans

belong; and others, for example Lieberman (1984) who believe that language first

appeared when Homo sapiens evolved, in the region of two hundred thousand

years before the present. Yet even if Lieberman is correct, two hundred thousand

years is a long time. Nobody suggests that human language appeared suddenly and

in one single step. A rapid evolutionary event for something as complex as

language is probably to be measured in thousands, perhaps tens of thousands, of

years. It almost certainly was preceded by the evolution of protolinguistic and

paralinguistic signalling such as gestures, grunts and cries of varying intonation

and intensity. It must have evolved alongside imitation and mimesis and would

have formed, across a not insignificant period of time and many generations, an

increasingly complex and effective collective system of communication involving

hands, ears and eyes. Entangled somewhere within this complexly evolving set of

communication skills is the cognitive module that allows the attribution of mental

states to others. It is likely that developmental studies will tell us more about the

dependencies of language, imitation and theory of mind, and give us some glimpse

into the evolution of the human capacity for culture, but I would not hold my breath

on it. Lacking any evidence from other existing species, I suspect that these are

going to remain issues about which we can only conjecture.

The Causes of Human Evolution

 25

I want here to flesh out the argument presented earlier that genetic-

developmentalist reductionist accounts of human intelligence and evolution are,

literally, untenable. The argument, remember, is that the evolution of the cognitive

modules that make up the semi-autonomous information-gaining device that is

intelligence resulted in a diffusion of the causes of intelligent behaviour into those

modules; and that if intelligence has had a causal role in the evolution of our, and

others, species, then the argument has to be extended. Reductionist accounts of the

evolution of intelligent species are no more tenable than are reductionist accounts

of the behaviour of intelligent individuals. This is because if a choosing

intelligence affects survival and reproduction, then that intelligence is one of the

causes of the changing gene frequencies in that species, which means that

intelligence becomes one of the causes of the evolution of that species.

The consequence of viewing the role of intelligence in this way is to make

evolution a much more dynamic and complex process. The history of this kind of

approach is to be found in Richards (1987) and Plotkin (1988a), and the details of

evidence and theory in Laland (1992) and Plotkin (1988b). Here I will offer just a

few examples. Imprinting is a phenomenon first described in birds by Lorenz, and

now understood to be present in many sexually reproducing species (Bateson,

1983). In broad terms, imprinting refers to the way in which exposure to particular

individuals affects later choice of mating partner. Imprinting is not an absolute and

irreversible outcome of development. In many species of bird and mammal

 26

imprinting looks like other forms of learning in occurring repeatedly and hence in

being reversible. Imprinting is a form of learning. Now, no behaviour more directly

affects what genes are propagated and in what combinations than does mating

behaviour. If choice of mate is an outcome of learning, then learning is entering in

a direct causal fashion into the determination of the genetic constitution of sexually

reproducing species that imprint. Here is a form of learning that does indeed have

causal force in evolution. The same argument can be run for birdsong, which we

know is learned and which has been shown to affect subsequent mating behaviour

and mate choice.

The conclusion is stark and simple. In some species there is clear evidence

that intelligence is one of the causes of evolution. Does this apply to our own

species? There is an enormous body of anthropological data about the way in

which human mate choice is determined by cultural injunctions in many different

kinds of societies. As argued in the previous section, culture is a human-specific

trait that derives from certain human-specific cognitive characteristics. The causal

force of culture in Homo sapiens is massive and pervasive. No-one is surprised at

the notion that cultural injunction constrains human mate choice; yet few social

scientists are comfortable with following through on the implications of this - that

since culture can only be explained as an outcome of human intelligence, then

intelligence is directly implicated in human mating patterns, including assortative

mating, which have been causal forces in human evolution. In this sense, it is the
 27

social scientists who should and must intrude into the biological sciences.

There are other examples apart from reproductive behaviour. The case of

lactose tolerance and intolerance is becoming famous for the way it demonstrates

the interweaving of culture and biology (Durham, 1991). Approximately two thirds

of all people in the world have varying degrees of difficulty in digesting lactose,

a sugar found in mammalian milk. Prior to weaning, the enzymes that allow lactose

absorption are present in the alimentary tracts of all people. Around the time of

weaning these levels decline in most people, and the illness that results from

drinking milk in those who are lactose intolerant can be severe - in the nutritionally

stressed, or those suffering from other illnesses, it can be fatal. Most people from

Africa and south and east Asia are lactose intolerant, whereas 95% of

Scandanavians are lactose tolerant. Across Europe there is an especially marked

gradient in the ability to absorb lactose, being high in the north and declining to the

south and east. Correlating with this gradient are customs of milk and milk product

preparation such that in northern Europe there is consumption largely of

unprocessed milk and its products like cream which are lactose rich, whereas as

one moves towards the Levant, processed milk products like yogurt and kefir,

which have greatly reduced lactose levels, become common. Accompanying the

dietary and food preparation practices are culturally propagated myths about milk

consumption, such as it being the food of the gods. It is now widely accepted that

lactose tolerance is caused by a mutant gene which became fixated at high levels
 28

in populations that combined high nutritional stress with vitamin D deficiency, the

latter being prevalent amongst people who live in regions of low sunlight levels

(unless they have other components of their diet that make good this deficiency,

but this is a detail of a much more complicated story than there is space for here).

The peoples of northern Europe benefitted in significant ways, life-saving ways,

through their ability to consume milk without it making them ill, and so the mutant

gene was strongly selected for in certain populations; however, the important point

is that the engine driving this evolutionary event was not the mutant gene but the

invention and propagation of animal husbandry and dairying practices, part of the

agricultural revolution. The agricultural revolution, as the word demonstrates, was

one of the most significant achievements of human culture in all of human

evolution.

The final example of how human intelligence and culture might have

affected human evolution is rather more tentative, but no less fascinating for that.

Earlier I mentioned the doubling (or quadrupling in relative terms) of brain size

during human evolution. In fact, the increase has not occurred gradually and

smoothly. There have been periods of relatively rapid expansion in brain size and

longer periods of stasis. One of the periods of most rapid increase occurred in the

middle Pleistocene epoch, around six to seven hundred thousand years ago, and it

coincides with evidence for the first use of fire in food preparation. The

significance of this according to Aiello and Wheeler (1996) is that cooking is

 29

essentially an externalization of the digestive process. As the controlled use of fire

spread and became commonplace amongst archaic humans, and it could only have

done so by cultural innovation and transmission, so the need for the large digestive

systems that characterize the great apes was reduced. Earlier in human evolution,

the switch to high energy foods, notably meat consumption in quantity, coincided

with another period of accelerated increase in brain size. What links these two

organ systems, brain and gut, is that they are the metabolically most expensive

organ systems in our bodies. Raise dietary quality or invent cooking and gut size

can be reduced allowing brain size to increase. The controlled use of fire was

a landmark event in human evolution. It is most likely that the practise spread

culturally, and if Aiello and Wheeler are correct, then this is an exquisite example

of human evolution having been significantly affected by human intelligence.

Many other examples, drawn from warfare, genocide, science and medicine, would

all tell the same story.

I began this lecture by pointing to the wonderful complexity in the

relationship between intelligence and evolution: intelligence evolves, then in turn

drives on the process of evolution. It brings to mind the image of a complex causal

dance in time in which effects become the causes of effects which themselves

become causes. To social historians who study the causes and consequences of

successions of cultural selection filters in the process of cultural evolution, this

 30

kind of image is, I should imagine, a familiar one. You people have always known

that history matters. I hope that I have shown you that an extension of that view to

human intelligence, culture and evolution has important consequences for

understanding ourselves, and demonstrates how links can be forged between the

biological and the social sciences.

A footnote

I am honoured by the invitation to present the inaugral Science and Society

Lecture. I am grateful to Dr. Kloosterman and his colleagues at the International

Institute of Social History who have given me the opportunity to present this

material in this way.

September 1996

References

Aiello, L.C. and Wheeler, P. (1996) "The expensive-tissue hypothesis: the brain
and the digestive system in human and primate evolution." Current Anthropology
36, 199-221.

Bateson, P.P.G. ed. (1983) Mate Choice. Cambridge, Cambridge University Press.
 31

Durham, W.H. (1991) Coevolution: Genes, Culture and Human Diversity.

Stanford, Stanford University Press.

Edelman, G.M. (1987) Neural Darwinism: the Theory of Neuronal Group

Selection. New York, Basic Books.

Gould, S.J. (1986) "Evolution and the triumph of homology, or why history
matters." American Scientist 74, 60-69.

Hirschfeld, L.A. and Gelman, S.A. eds. (1994) Mapping the Mind: Domain
Specificity in Cognition and Culture. Cambridge, Cambridge University Press.

Johnson, M.H. and Morton, J. (1991) Biology and Cognitive Development: The
Case of Face Recognition. Oxford, Blackwell.

Keesing, R.M. (1974) "Theories of Culture." Annual Reviews of Anthropology

3, 73-97.

Kroeber, A.L. and Kluckholm, C. (1952) "Culture: a critical review of the concepts
and definitions." Papers of the Peabody Museum of American Archaeology and
Ethnology 47, 1-223.

Laland, K.N. (1992) "Learning and evolutionary rates: a critical review."

Evolution and Cognition 2, 63-78.

Lieberman, P. (1984) The Biology and Evolution of Language. Cambridge

Mass., Harvard University Press.

Lorenz, K. (1965) The Evolution and Modification of Behaviour. Chicago,

Chicago University Press.

Pinker, S. (1994) The Language Instinct. London, Allen Lane.

Plotkin, H.C. (1988a) "Behaviour and Evolution." In H.C.Plotkin (ed) The Role
of Behaviour in Evolution. Cambridge Mass., MIT Press. pp 1-17.

Plotkin, H.C. (1988b) "Learning and Evolution." In H.C.Plotkin (ed) The Role of
Behaviour in Evolution. Cambridge Mass., MIT Press. pp 131-164.

Plotkin, H.C. (1995) Darwin Machines and the Nature of Knowledge. London,
Penguin Books.
 32

Premack, D. and Woodruff, G. (1978) "Does the chimpanzee have a theory of

mind?" The Behavioural and Brain Sciences 1, 515-526.

Richards, R.J. (1987) Darwin and the Emergence of Evolutionary Theories of

Mind and Behaviour. Chicago, University of Chicago Press.

Searle, J.R. (1995) The Construction of Social Reality. London, Allen Lane.

Sperber, D., Premack, D. and Premack, A.J. (1995) Causal Cognition: a

Multidisciplinary Debate. Oxford, The Clarendon Press.

Szathmary, E. and Maynard Smith, J. (1995) "The major evolutionary transitions".

Nature 374, 227-232.

Tobias, Ph.V. (1995) "The communication of the dead: earliest vestiges of the
origin of articulate language." Seventiende Kroon-Voordracht. Amsterdam,
Nedelands Museum voor Anthropologie en Praehistorie.

Tomasello, M., Kruger, A.L. and Ratner, H.H. (1993) "Cultural Learning." The
Behavioural and Brain Sciences 16, 495-511.

Williams, G.C. (1966) Adaptation and Natural Selection: A Critique of Some

Current Evolutionary Thought. Princeton, Princeton University Press.

Wilson, D.S. and Sober, E. (1994) "Reintroducing group selection to the human
behavioural sciences." The Behavioural and Brain Sciences 17, 585-608.