The Return of the Gene, Filozofia, Filozofia - Artykuły
[ Pobierz całość w formacie PDF ]
Journal of Philosophy, Inc.
The Return of the Gene
Author(s): Kim Sterelny and Philip Kitcher
Source: The Journal of Philosophy, Vol. 85, No. 7 (Jul., 1988), pp. 339-361
Published by: Journal of Philosophy, Inc.
Stable URL:
Accessed: 26/10/2008 14:48
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.
Journal of Philosophy, Inc.
is collaborating with JSTOR to digitize, preserve and extend access to
The Journal
of Philosophy.
__________________________-
+-
-
*
___
THE JOURNAL OF PHILOSOPHY
VOLUME LXXXV, NO. 7,JULY
1988
,
F~~~~~~~+
.
-4
THE RETURN OF THE GENE*
W
e have two images of natural selection. The orthodox
story is told in terms of individuals. More organisms of
any given kind are produced than can survive and repro-
duce to their full potential.
Although
these
organisms
are of a
kind,
they are not identical. Some of the differences among them make a
difference to their prospects for survival or reproduction, and
hence, on the average, to their actual reproduction. Some of the
differences which are relevant to survival and reproduction are (at
least partly) heritable. The result is evolution under natural selection,
a process in which, barring complications, the average fitness of the
organisms within a kind can be expected to increase with time.
There is an alternative story. Richard Dawkins' claims that
the
"unit of selection" is the
gene. By
this he means not
just
that the
result of selection is (almost always) an increase in frequency of some
gene in the gene pool. That is uncontroversial. On Dawkins's con-
ception, we should think of genes as differing with respect to proper-
ties that affect their abilities to leave copies of themselves. More
genes appear in each generation than can copy themselves up to
their full potential. Some of the differences among them make a
*
We are equally responsible for this paper which was written when we discovered
that we were writing it independently. We would like to thank those who have
offered helpful suggestions to one or both of us, particularly Patrick Bateson,
Robert Brandon, Peter Godfrey-Smith, David Hull, Richard Lewontin, Lisa Lloyd,
Philip Pettit, David Scheel, and Elliott Sober.
1
The claim is made in The Selfish Gene (New York: Oxford, 1976); and, in a
somewhat modified form, in The Extended Phenotype (San Francisco: Freeman,
1982). We shall discuss the difference between the
two
versions in the final section
of
this
paper, and
our
reconstruction
will be
primarily concerned
with
the later
version
of
Dawkins's
thesis. We shall henceforth refer to The
Selfish Gene as SG,
and to The Extended Phenotype as EP. To forestall any possible confusion, our
reconstruction of Dawkins's position does not commit us to the provocative claims
about altruism and selfishness on which many early critics of SG fastened.
0022-362X/88/8507/0339$02.30
?
1988 The Journal of Philosophy, Inc.
339
340
THE JOURNAL OF PHILOSOPHY
difference to their prospects for successful copying and hence to the
number of actual copies that appear in the next generation. Evolu-
tion under natural selection is thus a process in which, barring com-
plication, the average ability of the genes in the gene pool to leave
copies of themselves increases with time.
Dawkins's story can be formulated succinctly by introducing some
of his terminology. Genes are replicators and selection is the struggle
among active germ-line replicators. Replicators are entities that can
be copied. Active replicators are those whose properties influence
their chances of being copied. Germ-line replicators are those which
have the potential to leave infinitely many descendants. Early in the
history of life, coalitions
of
replicators began to construct vehicles
through
which
they spread copies
of themselves.
Better replicators
build better vehicles, and hence are copied more often.
Derivatively,
the vehicles associated with them become more common too. The
orthodox story focuses on the successes of prominent vehicles-in-
dividual organisms. Dawkins claims to expose an underlying struggle
among the replicators.
We believe that a lot of unnecessary dust has been kicked up in
discussing the merits of the two stories. Philosophers have suggested
that there are important connections to certain issues in the philo-
sophy of science: reductionism, views on causation and natural kinds,
the role of appeals to parsimony. We are unconvinced. Nor do we
think that a
willingness to
talk about
selection
in
Dawkinspeak brings
any commitment to the
adaptationist
claims which
Dawkins also
holds. After
all, adopting
a
particular perspective
on selection
is
logically independent from claiming that selection is omnipresent in
evolution.
In our judgment, the relative worth of the two images turns on two
theoretical claims in evolutionary biology.
1. Candidate units of selection must have
systematiccausal conse-
quences.If Xs are selectedfor, then X musthavea
systematic
effect
on its expected representationin future generations.
2.
Dawkins's
gene
selectionismoffers a more
general theory
of evolu-
tion. It canalsohandlethose
phenomena
whichare
grist
to the millof
individualselection,but there are evolutionaryphenomenawhichfit
the pictureof individualselection ill or not at all, yet whichcan be
accommodatednaturallyby the gene selectionmodel.
Those
skeptical
of
Dawkins's picture-in particular, Elliott Sober,
Richard
Lewontin,
and
Stephen Jay Gould-doubt whether genes
can meet the condition demanded in
(1).
In
their view, the phenom-
ena of epigenesis and the extreme sensitivity of the phenotype to
THE RETURN OF
THE GENE
341
gene combinations
and environmental effects undercut genic selec-
tionism. Although we believe that these critics have offered valuable
insights into the character of sophisticated evolutionary modeling,
we shall try to show that these insights do not conflict with Dawkins's
story of the workings of natural selection. We shall endeavor to free
the thesis of genic selectionism from some of the troublesome ex-
cresences which have attached themselves to an interesting story.
I. GENE SELECTION AND BEAN-BAG GENETICS
Sober
and Lewontin2
argue against
the thesis that all selection is
genic
selection by contending that many instances of selection do not
involve selection for properties of individual alleles. Stated rather
loosely, the claim is that, in some populations, properties of individ-
ual alleles are not positive causal factors in the survival and repro-
ductive success of the relevant organisms. Instead of simply resting
this claim on an appeal to our intuitive ideas about causality, Sober
has recently provided an account of causal discourse which is in-
tended to yield the conclusion he favors, thus rebutting the pro-
posals of those (like Dawkins) who think that properties
of
individual
alleles can be causally efficacious.3
The general problem arises
because replicators (genes) combine to
build vehicles (organisms)
and the effect of a gene is critically de-
pendent on
the
company
it
keeps.
However, recognizing the general
problem,
Dawkins seeks to disentangle the various contributions of
the
members of the coalition of replicators (the genome). To this
end,
he offers an analogy with a process of competition among
rowers for seats in a boat. The coach may scrutinize the relative times
of different teams but the competition can be analyzed by investi-
gating the contributions of individual rowers in different contexts
(SG 40/1 91/2, EP 239).
Sober's Case.
At the
general level,
we are left trading general
intuitions
and
persuasive
analogies. But Sober (and, earlier, Sober
and Lewontin)
attempted to clarify the case through a particular
example.
Sober
argues
that heterozygote superiority is a phenome-
non that cannot be understood from Dawkins's standpoint. We shall
discuss Sober's example in detail; our strategy is as follows. We first
set out Sober's case: heterozygote superiority cannot be understood
as a gene-level phenomenon, because only pairs of genes can be,
or
fail to be, heterozygous. Yet being heterozygous
can be
causally
2
"Artifact, Cause and Genic Selection," Philosophy of Science,
XLIX
(1982):
157-180.
'See Sober,
The Nature
of Selection (Cambridge: MIT, 1984),
chs.
7-9, espe-
cially 302-314.
We shall henceforth refer to this book as NS.
342
THE JOURNAL OF PHILOSOPHY
salient in the selective process. Against Sober, we first offer an anal-
ogy to show that there must be something wrong with his line of
thought: from the gene's eye view, heterozygote superiority is
an
instance of a standard selective phenomenon, namely frequency-
dependent selection. The advantage (or disadvantage) of a trait
can
depend on the frequency of that trait in other members of
the
relevant population.
Having claimed that there is something wrong
with Sober's argu-
ment, we then try to say what is wrong. We identify
two
principles
on
which the reasoning depends.
First is a general claim about causal
uniformity. Sober thinks
that there can be selection for a property
only
if
that property
has a positive uniform effect on reproductive
success. Second,
and more specifically, in cases where the heterozy-
gote
is fitter, the individuals have no uniform causal effect. We shall
try to undermine both principles, but the bulk of our criticism will be
directed against
the first.
Heterozygote superiority occurs when a heterozygote (with geno-
type Aa, say) is fitter than either homozygote (AA or aa). The classic
example is human sickle-cell anemia: homozygotes for the
normal
allele in African populations produce functional hemoglobin
but are
vulnerable to malaria, homozygotes for the mutant ("sickling")
allele
suffer anemia (usually fatal), and heterozygotes
avoid anemia while
also having resistance to malaria.
The effect of each allele varies with
context, and the contexts
across which variation occurs are causally
relevant. Sober writes:
In thiscase, the a alleledoes not havea uniquecausalrole. Whetherthe
gene a will be a positiveor a negativecausalfactor in the survivaland
reproductivesuccessof an organismdependson the geneticcontext.
If
it is placednext to a copy of A, a willmeanan increase
in
fitness.
If it is
placednext to a copy of itself, the gene willmeana decrement
in
fitness
(NS 303).
The argument
against Dawkins expressed here seems to come in
two parts. Sober
relies on the
principle
(A) There is selection for propertyP only if in all causally
relevant
backgroundconditionsP has a positiveeffect on survival
andreproduc-
tion.
He also adduces
a claim about the particular case of heterozygote
superiority.
(B)Althoughwe can understandthe situationby notingthatthe hetero-
zygotehas a uniformeffect on survivaland reproduction,the property
[ Pobierz całość w formacie PDF ]