http://www.nytimes.com/2001/07/15/books/review/15TURNEYT.html

July 15, 2001

'The Imitation Factor': Guppy Love

By Jon W. Turney

onkey see, monkey do. Does monkey have culture? Yes, according to Lee Alan
Dugatkin. And so do rats, pigeons, whales . . . oh, and guppies.
Individuals in all these species imitate some aspects of others' behavior.
So they have a way of passing on information that is separate from the DNA
they bestow on their offspring.

Guppies choosing mates are at the heart of Dugatkin's story in ''The
Imitation Factor.'' Female guppies' hard-wired preferences for the right
male -- one sporting, say, orange scales -- count for little if they see
another female mate with a more nondescript specimen. Mr. Average Guppy
suddenly is more attractive to the opposite sex. Being desired makes one
desirable.

Just that kind of slippage between animal and human behavior pervades
Dugatkin's writing. It is built into the pitch he makes for his research
program at the University of Louisville. One motive for that program is to
challenge the notion that human beings are radically different from other
creatures. The other is to devise experiments that might offer pointers
toward a general theory of culture.

In some ways, the scope for treating humans as exceptional is shrinking
fast. We have far fewer genes than most biologists thought when the Human
Genome Project began a decade ago, and share many of those with everything
from worms to fruit flies. And evolutionary psychologists, whom Dugatkin
cites approvingly, regard the mind as an evolved organ much as a
physiologist regards the liver. But however much reports of the unity of
the biosphere may appeal, enough distinctly human abilities remain to
support an argument that discontinuity with the rest of the living world is
as significant as continuity.

Those distinct abilities, especially in language and arts, are at the heart
of human culture, which is why the idea of a biological theory of culture
is so challenging. But here we have to be even more careful than when
appraising claims that genes determine human behavior -- because the
crucial terms are asymmetrical. ''Gene'' is a word taken from general
biology. When applied to humans, it means pretty much the same thing as
when applied to animals, though some of the powers popularly ascribed to it
these days are far-fetched. ''Culture,'' on the other hand, is a word with
many meanings in human life and society, heavily laden with connotations
from other disciplines: high, low and pop culture; beliefs and values; a
form of life. Take it into biology, and it may or may not mean the same
thing. This offers temptations few popular authors can resist. Can Dugatkin
avoid the kind of overinterpretation this invites? No chance.

To be fair, he tries to offer a usable definition of the kind of culture he
is talking about. It begins promisingly by invoking information passed by
imitation from one individual to another of the same species. And he cites
many elegant experiments -- mostly involving mirrors and models to control
what guppies see, or think they see -- that show imitation in action.
Unfortunately, he then extends his definition to include information passed
down by ''teaching,'' a word that raises the same problems as ''culture.''

Here the data are much thinner, and interpretation more contentious. Mainly
involved are cases where parents ''teach'' their offspring, by
demonstration, and induce behavior that is standard for the group. Hunting
in big cats is a well-known example. This contrasts with some impressive
imitation experiments, in which new behaviors can spread through a group.
It is also easier to account for in conventional evolutionary terms, if
what members of a new generation learn from parents (who share their genes)
increases their chances of thriving.

In fact, it is often possible to redescribe the imitation experiments in
terms of genetic propensities, as Dugatkin acknowledges. Mate choice by
guppies is obviously closely linked to whose genes get passed on. One of
his more ingenious experimental setups involves a female forced to choose
between a male with coloring she likes and a shabbier individual she just
saw mate with another fish. If the experimenter varies the colors, the
results can be read as comparing the attractions induced by ''genes'' and
''culture.'' But they can equally well be seen as showing the competing
influences of two genes or sets of genes. One instills a strong color
preference. The other confers a tendency to copy the last female seen
mating. If the highly colored males tend to contribute better genes of
other kinds sometimes, but not always, it is easy to imagine how changes in
the environment would keep the alternative, imitative tendency alive.

At this level, the book is a fascinating account of the difficulties of
designing laboratory experiments to dissect unambiguously something as
complex and variable as behavior -- even the behavior of a fish with a
brain the size of a pinhead. But the contrast between this careful
experimentation and the lack of caution about larger claims becomes more
painful as the book goes on. Calling the closing chapter ''Animal
Civilization'' is unhelpful. The oxymoron makes one more skeptical about
the earlier chapters. So this is what Dugatkin means when he allows that
his work on guppies stands genetic models of behavior ''on their head'' and
forces scientists to rethink the ''dramatic impact'' of culture on animal
behavior. Here is where the ''golden age of theoretical biology'' is leading.

THE truth is that Dugatkin has some solid results that may be important to
behavioral ecologists and evolutionary theorists. He does not have even the
beginnings of a biological theory of culture in its larger senses. A brief
chapter on ''memes,'' the term for cultural units of selection coined by
Richard Dawkins and recently elaborated by Susan Blackmore, simply
underlines this in being largely disconnected from the rest of the book.

It is an old point, but one worth repeating. The culture of guppies
involves observing other guppies, and mating with this fish or that. The
culture of guppy researchers includes building elaborate apparatus to
control what guppies see, recording and analyzing the resulting data,
writing scientific papers about them, defending them in conferences and
seminars, and then telling the world. It also includes the poems of T. S.
Eliot, Chopin etudes, Cubism, thermonuclear weapons, cloned sheep and
superstring theory. Using the same word for both is surely a sign of
spending too many hours in the lab.

Jon W. Turney is the senior lecturer in science communication at University
College, London.