NATURAL SELECTION AND THE EVOLUTIONARY STATUS OF CULTURE
Lyle Steadman
Department of Anthropology
Arizona State University
Human Behavior and Evolution
Society Meeting
Binghamton, New York
August 5, 1993
Because much of human behavior is
cultural, evolutionary explanations of human behavior will never take us far if
they cannot account for culture and traditions.
The main point of this paper is that Darwin’s discovery of natural
selection can contribute greatly to an understanding of culture. I discuss first natural selection.
Due to a focus on genes and gene
pools, evolutionists have lost sight of part of the explanatory power of
Darwin’s concept of natural selection.
Darwin was a naturalist who was attempting to explain how organisms came
to have the various TRAITS they exhibit—the particular shape of a bird
beak, for example, or why humans smile, or why males pursue females. He discovered that when certain traits that
had some chance of being inherited helped individuals leave descendants, those
traits in a similar environment tended to increase in frequency along with the
descendants. This discovery of natural
selection identified the main cause of the persistence and change in
frequencies of inheritable traits.
Genes, of course, were yet to be discovered.
Toward the end of his life in the
late 1800s, mainly because of Lord Kelvin’s estimate of the age of the earth,
now known as false, Darwin lost confidence in his theory of natural selection
being able to explain the existence of the incredible variety of traits that
had evolved. While neo-Darwinism
flourished, and [Herbert] Spencer’s phrase “survival of the fittest” remained
popular, the theory of natural selection came to be ignored. In 1900 the discovery of genes became known,
and the cause of traits, including behavior, came to be seen as basically
genetic. Eugenics [the science of the
improvement of species] became extremely popular, chairs for eugenecists were
established in universities, and writers used genetics to support their notions
of racial superiority. The latter
sparked a rebellion among social scientists, who argued that behavior was determined
solely by the environment (see Derek Freeman, Margaret Mead and
Samoa, The Making and Unmaking of an Anthropological Myth, 1983). To a great extent biology became the study of
genes and their effects. Geneticists had
eclipsed the naturalists. But in the
1930s several brilliant biologists put together what has been termed “the
modern synthesis,” bringing Darwin’s natural selection back into biology, but
applying it to genes and gene pools rather than to traits. “Population thinking” came to be
emphasized. As [Martin] Daly and [Margo]
Wilson write, “modern biology equates natural selection with the
differential reproduction of genotypes” (1983: 11). Evolution was seen as changes in a gene pool,
and natural selection was seen as having to do with the cause of that
change. However, it must be emphasized,
asexual organisms are as subject to natural selection as sexual organisms, but
have no gene pools [because all individuals of a species have exactly the
same genes]. Therefore, evolution and
natural selection should not be defined by changes in gene pools.
It was not until the 1960s that
Darwin’s discovery of natural selection was again used to explain inheritable
TRAITS. George Williams, in his book Adaptation
and Natural Selection (1966), used natural selection, not to explain gene
frequencies in gene pools, but to explain persistence and change in frequencies
of inheritable traits—inheritable phenotypes.
At least in the nascent discipline of sociobiology, despite the
genetically focused definitions of natural selection in the first chapters of
books, attention focused actually on the Darwinian adaptation of traits, not
genes. Population thinking fell out of
vogue and behavior no longer was assumed to be aimed at the maintenance of a
species’ gene pool.
Let me give an example from one of
the best books written in sociobiology.
In the first chapter of Sex, Evolution, and Behavior, Daly and
Wilson argue that natural selection implies changes in gene pools, and insist
that there has to be a change in gene frequencies if natural selection has
occurred. They write,
“natural selection is powerless to produce evolution
if phenotypic differences in survival and fertility are unrelated to genotypic
differences, for in such a circumstance no systematic changes in the gene pool
would accrue . . . Phenotypic differences must have a basis in genotypic
differences if selection is to operate” (33).
In
addition, “the natural selection that drives evolution is a process of competition
between genotypes within a population or species . . . The result is that the
species’ gene pool, the set of genes that in a sense is the species, is
altered” (35).
However, they also write, “natural
selection operates directly on phenotypes, not genotypes. It is not the genes that live or die, breed
or help their relatives, but the realized animal. The result is that both phenotypes and gene
pools evolve” (32). And arguing against
[Richard] Dawkin’s emphasis on the gene, not the individual, as the basic “unit
of selection,” they write, “what we observe are individual organisms behaving,
and it is this behavior that we wish to understand” (31). Darwin would agree wholeheartedly.
Because Darwin knew nothing of
genes, the discovery of Darwinian natural selection therefore does not imply
genes; it does imply inheritable, replicable traits. But the point I am trying to make here is
certainly not that genes are irrelevant to natural selection, but that they are
not the only elements subject to natural selection. The essence of Darwin’s discovery is that
anything inheritable and replicable is likely to influence its own
frequency in subsequent generations. Because
genes are inheritable and replicable, their frequency will be influenced by the
effect they have on their “host’s” success in leaving descendants. This is also true of traditions, which are
inheritable and replicable. Traditions
are the key to understanding the evolutionary significance of culture.
Traditions are culture that has been
transmitted from ancestor to descendant.
But what is culture? The term was
first defined, in an anthropological way, more than a century ago by the man
sometimes called the “father” of anthropology, Sir Edward Burnett Tylor. The first sentence of his massive tome, The
Origins of Culture reads: “Culture or Civilization, taken in its wide ethnographic
sense, is that complex whole which includes knowledge, belief, art, morals,
law, custom, and any other capabilities and habits acquired by man as a member
of society” (1958: 1).
Basically, this is a list of
examples rather than a precise statement of the necessary ingredients that
constitute an example of culture. Today,
more than a century later, there is still no widely accepted, explicit
definition, even though culture lies at the core of anthropology. What I propose to do now is to identify the
necessary elements which constitute any agreed upon example of culture.
All anthropologists would agree that
whatever else is implied by “culture,” it does imply something LEARNED. That is, if it is not learned it cannot be
cultural. But learning itself does not
imply culture; learning is not sufficient for culture to exist. Learned behavior is widespread among animals,
but very few scientists would identify most of that behavior as cultural.
Culture also is described as “shared
behavior.” But individuals certainly can
share and exhibit similar behavior that is not cultural. They can exhibit non-learned, genetically “programmed”
behavior, like blinking when a bright light appears, or sneezing when they
inhale pepper. Individuals can also
share learned behavior that they have acquired independently of one
another. For example, by experiencing
being burned by fire, they can learn individually to avoid it. But such learned and shared behavior is not
considered cultural. In addition, all
mammals automatically learn to identify unique individuals. The behavior resulting from this memory of
unique individuals and their behavior, of fundamental importance to kinship
cooperation everywhere, is not necessarily acquired from any other
individual. I can identify my parents
and my kids unerringly out of a million people.
And my dog can do the same. While
this is clearly learned behavior, shared and exhibited by mammals in general,
it is not cultural.
What seems essential to culture is
not just that it is learned and shared but that it is ACQUIRED from
another individual and potentially transmittable to a third. That is, the additional, and I believe final,
ingredient necessary for something to be identified indisputably as cultural is
that it be acquired, through learning, from another individual, and then
copied, or reproduced. Cultural behavior
must be copied from another individual.
That is, an individual must first experience that behavior in another,
remember it, and then reproduce it. If
it is learned, but not copied, it cannot be observed by others, and hence
cannot be transmitted to them.
Copying is crucial to the
acquisition and transmission of speech.
And because speech allows humans to experience vicariously certain
behavior, it allows them to copy such behavior.
For example, by experience a myth, a traditional story, humans can
experience certain behavior described in the myth and then copy it. But humans copy much more than speech, or
even behavior that is described by speech.
An apprentice working for a mute shoemaker can learn to make shoes by
copying his behavior. Indeed, most of
the techniques we master, even complex techniques, have been copied from
others.
However, copying implies neither cooperation nor a
group. A human can copy the song of a
bird, and use that song while hunting to attract and kill it. Likewise, he can copy the behavior of an
enemy, and use that behavior to do his enemy in, showing that the transmission
of culture need not be cooperative or social.
Culture, thus, does not imply a group (nor, in Tylor’s words, a
society).
Finally, knowledge, argued by many social scientists
to be cultural, need not be. For example,
a coyote pup following its mother exhibits cultural behavior. By following her, it copies her behavior in
going down that particular path. By
going down the same path as its mother, however, the coyote will experience the
environment of that path independently of its mother. The same can be true of a child following its
mother. Thus, even though such
experiences themselves constitute knowledge, they are not cultural.
Therefore, learning, or knowledge, which can be
exhibited in cultural behavior, obviously, is not restricted to culture. Any organism that is identified as learning
anything, whether it is behavior that is learned and copied or not, exhibits
knowledge. The knowledge necessarily
involved in cultural behavior is only the memory of the behavior
sufficient for its copying.
Thus, to sum up, the necessary conditions for
culture are no more than that it be behavior that is learned and copied from
another individual. Any behavior that
has been learned and copied, I am proposing, is an example of culture. Disproof of this hypothesis of the usage of
the term culture, would consist of an observable example of culture, acceptable
to anthropologists, that either is not based on learning or has not been
copied.
If this proposition is true, important consequences
follow. To the extent culture is not
exhibited, it cannot be copied.
“Beliefs” are often specified as part of the definition of culture
(recall Tylor’s definition). And yet, to
the extent that beliefs are internal and cannot be identified, they cannot be
copied. What is inside our skulls is
inaccessible to others, and therefore cannot be cultural. For something to be cultural it must be
exhibited. Obviously, people can say
things they do, or do not believe, and such claims can be copied. Thus, claims about beliefs can indeed
be cultural, regardless of the underlying beliefs. To the extent religion is cultural, it must
be copied.
We have come some distance from Tylor’s
definition. Culture does not have to be
acquired as a member of society, nor does it necessarily include belief or
knowledge, other than that required for its reproduction. Furthermore, as learned, copied behavior it
is as subject to natural selection as any learned behavior. Our great readiness to copy the behavior of
other individuals is based, presumably, on distinctive genes, and hence is
subject to natural selection. Like
learned behavior in general, the particular behavior copied can be crucial, for
some cultural behavior can be very unsuccessful in leaving descendants. For example, individuals who witness
behavior, even behavior they themselves condemn, nevertheless are more likely
to exhibit that particular behavior than if they had never witnessed it (Baron
and Byrne, 1977: 324-5). Individuals,
especially teenagers, who witness suicide either directly or vicariously, are
more likely to commit suicide, for suicide is now a possibility. When we witness violence we are more likely
to be violent. At the same time, when we
witness altruism, we are more likely to be altruistic.
All learning organisms that exhibit parental care
can benefit their own offspring (and hence their fitness) by what they themselves
have learned during their lifetime. But
learned behavior that is transmittable to descendants, that is, when it can be
reproduced by descendants, like genes, can influence very distant
descendants. While culture can be
acquired from anyone, even an individual of a different species, culture that
is acquired from ancestors is distinctive, and we have a name for such
behavior—tradition. Culture acquired
from ancestors is copied from individuals who have been successful in leaving
descendants. Transmitting such
successful behavior to descendants not only tends to increase the number of
one’s descendants, but also the frequency of that behavior. Because traditions are both inheritable and
replicable, like genes, they should be subject directly to natural
selection. In contrast to a gene,
however, a tradition can be acquired by all offspring—indeed, all
descendants—not just a few of them, and therefore can respond very rapidly to
natural selection. While gold and land
may be inheritable, they are not, unfortunately, replicable, and therefore are
not subject to natural selection. So,
although they may help one leave descendants, they will not increase in
frequency.
I’m sure many of you would prefer not to call what
I’m describing natural selection. But
the question comes down to this: Traditions, because they are inheritable and
replicable, can influence their own frequency in subsequent generations. That seems indisputable. What should that influence be called? Darwin, it seems to me, would accept it being
called natural selection. It satisfies
all requirements of the definition of natural selection except that it does not
require a change in gene frequencies.
But Darwin didn’t either.
I give two examples: the first is the Shakers. Mother Ann Lee, the prophet of the Shakers,
identified the “Root of evil in the act of sexual intercourse.” She encouraged her followers to leave England
and come to the U.S. [British American colonies] in 1774. There they lived together in “families” of
celibates, sharing all property in common.
They recruited young members mainly through orphanages, until
communities began to resist. In their
years of greatest vigor, in the early 1800s, the Shakers numbered over
5,000. They loved one another like
brothers and sisters—not sexually—and consequently left no descendants. Perhaps only a few converts survive today
(Albanese 1981: 155-6). Distinctive
Shaker traits were replicable, but not inheritable. The significant evolutionary effect of Shaker
traditions was extinction.
My second example: the Hutterites. When the Hutterites migrated from Europe to
South Dakota in 1872 they numbered less than 800 (Hostetler and Huntington,
1980: 3). By 1965, less than a century
later, the Hutterite “population numbered about 16, 500” (Hostetler and Huntington,
1980: 1). This population growth was
almost entirely due to an amazing degree of descendant leaving success
resulting from a “median family size of
. . . 10.4 children” (Hostetler and Huntington, 1980: 57). Hence, Hutterite traditions—including
communal living, strict obedience to church elders, prohibitions on
birth-control, and great altruism toward one another—traditions that were
inherited and replicated in each new generation, also increased dramatically
over the period from 1872 to 1965. This
increase in Hutterite traditions was due largely to the positive effect of
these traditions on the descendant leaving success of Hutterites. Hence, Hutterite traditions have increased
due to a process that is not just “somewhat analogous” to natural selection,
but a process that IS natural selection as originally formulated by
Charles Darwin.
Because sociobiology fundamentally is the study of
inheritable, replicable phenotypes, or traits, traditions are a proper study of
sociobiology. What I have argues in this
paper is that the selection for traditions accounts for the widespread
existence of cultural behavior, including behavior that is ultimately
maladaptive.
(One reason humans are difficult to study is because
the explanations they give of their behavior not only may be inaccurate, but
even aimed at influencing the behavior of their audience. We must never assume that what we are told
reveals the true understanding of the speaker.
To overcome this great problem, we must make hypotheses whose accuracy
is checkable, and therefore subject to disproof.)
(John Tooby and Leda Cosmides, 1989, come close to
the definition of culture proposed here.
Also, Richerson and Boyd’s “vertical transmission” of culture is similar
to my definition of tradition.)
Appendix:
Some people argue that language, the
symbolic use of sound, is the sine qua non [something indispensable] of
culture. However, individuals can learn
the meaning of symbols (e.g., words), without exhibiting cultural behavior. For example, dogs not only can learn their
name and the names of others, but can learn to respond appropriately to such
symbols as “come,” “sit,” “fetch,” and “stay.”
That is, based on the dog’s behavior we can say it has learned the
meaning of those words, in contrast to, for example, a dog that has not. Those symbols and their referents are fixed
in the dog’s memory, even though the dog cannot reproduce them. Because it cannot reproduce the words,
however, they are not transmittable to others, and therefore are not
cultural. (The dog can exhibit behavior
appropriate to those words in the same way it can adjust its behavior to the
pattern run by a fleeing rabbit. The dog
is merely exhibiting learned, but not cultural, behavior.) Humans too, especially when young, can show
they know the meaning of certain words without being able to copy them. Indeed, a human who cannot speak, can
demonstrate that he had learned the meaning of words, even though he cannot
transmit that speech to another. Thus,
for him, the words are not cultural. So,
learning the meaning of words does not imply culture, but the utterance of the
words themselves does. (The symbols we
call words are acquired—copied—by humans from other humans.) Speech, therefore, is cultural; but the
meaning of a word (i.e., knowledge of the referent of a symbol) need not
be. Thus a cultural symbol is not merely
copied behavior, for it implies an identified and remembered association with
its referent. It is the memory of this
association with its referent that is crucial to the meaning of a symbol;
indeed, the association with its referent constitutes the meaning of a
symbol. As a symbol, the association
with its referent must be remembered; as cultural behavior, the association
with its referent must be copied. Until
it is copied, it is not cultural. In
addition, language allows humans to “witness” or “experience” vicariously some
behavior indirectly and thereby be able to copy it. Elsewhere, I have argued that myths serve
this purpose: to transmit behavior vicariously.
Parrots, when they mimic segments of
human speech, exhibit cultural, but not symbolic behavior. Apparently, they simply remember and copy the
words as sounds (the same as they do with a doorbell or jackhammer). Such behavior, therefore, would be cultural,
but not symbolic. Parrots’ readiness and
ability to copy sounds is presumably an adaptation, perhaps having something to
do with being identified as close kin, or as a member of their flock.
(Domesticated animals and plants,
from the point of view of humans may be both inheritable and replicable and
because they are used to help humans leave descendants, they have indeed
increased in frequency.)
BIBLIOGRAPHY
Albanese,
Catherine L.
1981. American Religions and Religion. Wadsworth: Belmont, CA.
Baron,
Robert A. and Donn Byrne
1977. Social Psychology. Allyn and Bacon, Inc.: Boston.
Daly,
Martin and Margo Wilson
1983. Sex, Evolution, and Behavior. Willard Grant: Boston.
Freeman,
Derek
1983. Margaret Mead and Samoa, The
Making and Unmaking of an Anthropological Myth. Harvard University Press: Cambridge, MA.
Hostetler, John A. and Gertrude Enders Huntington
1980. The Hutterites in North
America. Holt, Rinehart, and Winston, Inc.: Fort
Worth, TX.
Tooby, John and Leda Cosmides
1989. “Evolutionary Psychology and
the Generation of Culture, Part 1: Theoretical Considerations.” Ethnology and Sociobiology 10 (1-3):
29-51.
Tylor,
Edward Burnett
1958 (first published in 1871). The Origins of Culture. Harper & Brothers: New York.
Williams,
George C.
1966. Adaptation and Natural Selection. Princeton University: Princeton, NJ.