Professional Documents
Culture Documents
Evolution's Eye: Susan Oyama
Evolution's Eye: Susan Oyama
Susan Oyama
Science and Cultural Theory
Susan Oyama
Designed by C. H. Westmoreland
Acknowledgments vii
Introduction
Notes
References
Index of Subjects and Names
This page intentionally left blank
Acknowledgments
The following people were of special help in writing this book. By offer-
ing comments on the pieces when they were originally being written, ad-
vice and other aid at varying points in the preparation of the book itself,
or both, they demonstrated once again the value of critical and gener-
ous colleagues. My deep thanks go to Pat Bateson, Ben Bradley, Linnda
Caporael, Richard Francis, Eli Gerson, Peter Godfrey-Smith, Russell
Gray, Paul Griffiths, Megan Gunnar, Jack Hailman, Mae-Wan Ho, Tim
Johnston, Evelyn Fox Keller, William Kessen, Philip Kitcher, Richard
Lewontin, Lenny Moss, Katherine Nelson, Jeff Ricker, Barbara Herrn-
stein Smith, Elliott Sober, Don Symons, Peter Taylor, Rasmus Winther,
and Cor van der Weele. Miriam Angress, Pam Morrison, and Reynolds
Smith of Duke University Press were responsive and supportive; what
more could any author ask?
This page intentionally left blank
Introduction
ally I have used the term developmental systems theory () to describe
this approach, and I will continue to use it here.1 It is short and snappy,
already in use, and convenient for indicating a rough grouping of alter-
native conceptualizations and critiques. I have tended to employ the term
with qualifications, however, indicating that if a theory is taken to be
only a hypothesis-generating machine, then perhaps perspective or ap-
proach would be preferable. What I have in mind is broader: a kind of
conceptual background that serves to orient more specific empirical and
theoretical endeavors. It can also allow different interpretations of exist-
ing work. (See van Gelder and Port for a discussion of this framing
or orienting function of approaches to cognition.)
The need for a term cannot be ignored, but any act of individuation in-
volves the marking of commonalities and dissimilarities in order to pick
something out from the background. The drawing of such outlines can
be a delicate matter. Reciting a creed or compiling lists of insiders risks
essentializing a more or less loose assemblage in inappropriate ways. Nor
is this the place for a history or an analysis of related approaches.2 I will,
however, sketch a number of key ideas and methods that characterize the
work presented in this book. Fuller presentations are found primarily in
the chapters of part ; to varying degrees these features also characterize
the writings of the colleagues with whom I am most closely associated.3
Although we don’t always agree on technical details or on the precise
form and limits of necessary theoretical reworkings, I think we share
a commitment to a view of development as interactive emergence over
time and to exploring the implications of taking this gradual construction
seriously. Following the ideas and methods below is a list of work that
was especially important for the development of my thinking, along with
some remarks on research and metaphor.
ters and , for instance, I show how the reasons for calling features of
organisms ‘‘genetic’’ or ‘‘environmental’’ tend to be used asymmetrically.
These analyses are developed in subsequent chapters (especially ) to
demonstrate how genes are frequently given causal priority even when
similar reasoning could be used for other causes. In such cases, what is
presented as a justification for giving the gene special status is actually
a consequence of already having done so. Undoing these conventional
asymmetries highlights, among other things, point .
. The developmental and evolutionary interdependence of organism
and environment. I replace conventional wisdom’s ‘‘interaction’’ between
separate and independently defined organisms and environments with
‘‘constructivist interaction.’’ This is not to be confused with the ‘‘inter-
action term’’ in some statistical analyses; that is why ‘‘constructivist’’
is important. In constructivist interaction organisms and their environ-
ments define the relevant aspects of, and can affect, each other. This is
called the ‘‘interpenetration’’ of organism and environment by, for ex-
ample, Lewontin (; see also Lewontin, Rose, and Kamin ). See-
ing life in these terms rather than following the traditional cleavages be-
tween nature and culture, body and mind, permits different and fruitful
ways of conceiving biology, psychology, and society, as well as different
relations among the disciplines (see Ingold ). Such a view suggests
point .
. A shift from ‘‘genes and environment’’ to a multiplicity of entities,
influences, and environments. One benefit of thinking in terms of mul-
tiple systems, interconnected and measurable on more than one scale of
time and magnitude, is that the stunningly oversimplified distinction be-
tween genes and environment resolves into a heterogeneous and equally
stunning array of processes, entities, and environments—chemical and
mechanical, micro- and macroscopic, social and geological—of the sort
outlined in chapter . Evolution’s eye widens, and so do ours. Hence my
preference in the following chapters for systemic interaction over one-
way causal arrows, for mutually influencing levels rather than the col-
lapsed levels that often go with unidirectional causation. The shift from
genes-and-environment to multilevel systems is coupled with the next
point.4
. A shift from single to multiple scales. Looking through developmen-
tal systems lenses means, among other things, looking hard and repeat-
Introduction
edly in a variety of directions and with both near and distant focuses.
Movement among scales, both of magnitude and time, is important: from
interactions of molecules inside cells to those between persons, from the
brief periods involved in the action of a hormone in the nervous sys-
tem to changing relations among conspecifics over the life span, from the
short-term dynamics of a population of organisms in a habitat to the slow
procession of generations through evolutionary time.5
Moving among scales in this way not only enables us to see more, it
also gives us a more acute sense of the many relations among these scales.
The outcome of an aggressive encounter between animals, for instance,
can lead to physiological changes, including changes in gene activity (see
Gottlieb , , , on bidirectional influences between structure
and function; Lehrman ). Similarly, rapid processes can influence,
and be influenced by, much slower ones, as might happen if gradual cli-
matic change altered the kind of food available to a creature. The animal’s
digestive processes might adapt (with or without accompanying genetic
change), and this might eventuate in a different foraging range, which
could then lead to other developmental and evolutionary changes (see
P. P. G. Bateson a; Gray ; Johnston and Gottlieb ; Lewontin
; Sterelny, Smith, and Dickison ).
There is still a place for the ‘‘provisional single-mindedness’’ men-
tioned in chapter , the tactical decision to focus on one level for a par-
ticular purpose. It is important to recognize that such an approach tempo-
rarily delimits the context for the investigation; eventually the questions
and findings must be recontextualized. There is, after all, no research
that does not limit the scope of entities and variables studied. Items
through lead ineluctably (for me, at least) to item .
. Extending heredity. I have taken two kinds of critiques, of innateness
in individuals and of the relations between evolving populations and their
niches, and connected them by means of a broader notion of heredity
than has been countenanced in contemporary science. A good part of my
writing, the present volume included, involves the working out of that
connection and its many ramifications both inside and outside biology.
The earlier work of the American comparative psychologists and En-
glish ethologists who are mentioned below in ‘‘Resources’’ is probably
the most widely known for questioning internalist conceptions of innate-
ness and instinct. Lewontin (e.g., ) wrote on these topics as well, but
Introduction
Resources
Research
The writings mentioned above are rich with examples of research on the
kinds of relationships described in my section on ideas and strategies.
Many others can be found in the works cited in note . Russell Gray
() brings together some especially useful references, although he
notes that research on what he calls ‘‘evolutionary and ecological cas-
cades’’ is not plentiful, probably because the questions have not been
posed in appropriate ways. Similarly, Cor van der Weele (), who
offers many findings on environmental factors in development, notes the
relative neglect of this topic. Eva Jablonka and Marion Lamb (:ix)
supply many citations in addition to their own findings on ‘‘epigenetic in-
heritance,’’ and observe that existing examples of such phenomena tend
to be overlooked.
Indeed, a ‘‘body of research,’’ as opposed to a mass of disconnected
and hard-to-find bits, comes into being through the synthetic conceptual
work in which these people are engaged. What an instance of empirical
work ‘‘shows’’ can change as the observations are placed within a con-
ceptual framework that helps them make sense and from which other
research can be generated.
Directions for research discussed in the present volume (especially in
Introduction
Evolution’s Eyes
Linked to the first reading of evolution’s eye, and so to the puzzle of the
place of development in evolution, is a second one: the eye of the re-
productive needle. This is sometimes referred to as the ‘‘bottleneck of
reproduction’’: the narrowing of the life cycle between successive gen-
erations of certain organisms to a single cell (Bonner ). This one-
celled stage is not universal, as anyone knows who has watched a runner
or a leaf grow into a flourishing plant. Consider, however, organisms that
do begin a new life cycle as we do, as fertilized eggs, or zygotes. Many
biological treatments downplay the rest of the richly varied and crucially
Introduction
P P P P
¢
¢
¢
¢
G ¢ G ¢ G ¢ G ¢
This phenotype develops and dies, passing only its G to its offspring.
The significance of the developing organism is thus graphically mini-
mized by showing it as a literal offshoot. At the same time, organisms be-
come mere consequences of the genes’ causal powers. The needle’s-eye
understanding of reproduction and heredity explains how we can have
an evolutionary biology that gives such short shrift to the developmen-
tal dynamics of the life cycle. Anything other than a gene has as difficult
a time getting through reproduction’s strait gate as that poor camel in
the parable, straining to squeeze through the eye of a needle.11 And yet,
as I show in the chapters that follow, there is only a bottleneck if one
accepts a particular narrow definition of heredity. The existence of all
those single-celled stages, furthermore, is possible only because they are
preceded by, and surrounded by, an enormous functional and structural
complexity.
The first two eyes, the blind one turned to development by natural selec-
tion and the narrow needle’s eye of heredity, set the themes of part ,
‘‘Looking at Development and Evolution.’’ My conviction that evolu-
tion’s eyes need to be opened, or rather that they actually take in more
than most theories do, helps to motivate the arguments in part , ‘‘Look-
ing at Ourselves.’’ There, two more meanings of evolution’s eye come into
focus. Although present in earlier chapters, these additional meanings are
rendered more salient by a shift in orthography. Chapter , on contin-
gency, is a ‘‘hinge’’ piece insofar as it joins the use of parity of reasoning
in the developmental systems approach to the principle of symmetry in
science studies, bringing to the fore the epistemological issues that are
especially important in part .
My first I is the person constituted by a double duality: the famil-
iar nature-nurture dichotomy and the body-mind dualism with which it
is closely associated. In this well-worn but still dominant framework,
the development of persons must be achieved by somehow joining the
natural to the acquired, the biological to the cultural, or, if you will,
the body to the mind. I examine aspects of these dualities in part , but
in part I pay more attention to their consequences for our everyday
lives. The widespread and rising interest in evolution has been accompa-
Introduction
the fray, to reflect on what the question actually is, whether there is only
one, and what a definitive answer could look like. The more important
the issue, the more urgent the need to frame it well, and I believe that
the perspective presented here provides a better basis than the traditional
ones for individuating precise questions and possible—or sometimes im-
possible—ways of seeking their answers.
The previous reading of part ’s title was the unintegrated I of the di-
chotomies between nature and nurture, body and mind. Knowing, think-
ing, perceiving, and remembering are all activities of living beings in
particular settings, and the other theme of ‘‘Looking at Ourselves’’ is
the epistemological aspect of organism-environment interdependence in
developmental systems. My interest in the relations between scientists’
questions and the phenomena they find, then, while evident in the first
part, is more prominent in the second. These issues are taken up in chap-
ters and . In the latter I also address the question of what is taken
to fall inside or outside legitimate theory. The first-person I is also the I
of the knowing subject, then, who may be, but is not necessarily, a sci-
entist; the emphasis here is on the role of point of view in organizing
knowledge.
Besides the familiar binaries referred to above as the double duality
of nature and nurture, body and mind, are many others that pervasively
organize our looking, whether the object of regard is evolution, develop-
ment, ourselves, or the world in general. There are insides and outsides,
autonomy and dependence, necessity and contingency, male and female.
Yet, one needn’t be a triops to see beyond them.13 There are excellent
grounds for challenging dualisms in biology and its sister sciences. Tim
Ingold () has noted that anthropologists rarely think to question the
distinction between biology and culture, but the same could be said of the
individual-society opposition in anthropology, sociology, and psychol-
ogy. Much of social constructionism, in fact, not only fails to question
these divisions but is actually grounded in them (chapters and ). In-
deed, a poignant aspect of some of this work is that it often begins with
a desire to defend the reality and significance of the social against what
seems a hypertrophied biology (or individual psychology), but then ends
Introduction
by ratifying some of the very beliefs and practices that feed that biology’s
unbalanced growth.14
By the same token, to inquire whether knowledge is made more by the
knower or by the object, even if one’s reply places greater stress on the
former, may be to rely on the same notions of reified ‘‘information’’ we
constantly encounter in biology. The connection is not only analogical;
the two discourses overlap because nature-nurture arguments brim not
only with genetic information, but also with information from or about
the world. The divided organism referred to above, in fact, is frequently
thought to be made from these two kinds of information. Textbook dis-
cussions of instinct characteristically begin with the rationalist and em-
piricist philosophers and go on to disputes about how much knowledge
is biologically given and how much must be gained through experience.
I suspect that many ideas of sensory information (and, while we are at
it, of mental representations) are not only as problematic as their genetic
counterparts in discussions of development; they are, finally, instances of
that same usage.
If information must be understood relationally, as a difference that
makes a difference (or, if you will, the information theorist’s reducer of
uncertainty), then the same constructivist interactionism that integrates
insides and outsides in ’s accounts of development should serve for
epistemological investigations as well. This kind of construction does
not depend on a distinction, say, between the constructed and the pre-
programmed, or, for cognition, between cognitive construction and accu-
rate representation.15 Just as a developmental interaction changes the or-
ganism in some way, at some scale (and is ultimately defined by such a
change), so experience changes us, for the short term or the long. An in-
formal memory check suggests that once phenotype came to refer to the
organism, it was restricted pretty much to the body (often just the appear-
ance of the body). More recently, behavior has been included. Can knowl-
edge, thought, indeed, memory itself, be excluded in any principled way?
The Book
Looking at Development
and Evolution
This page intentionally left blank
1 Transmission and Construction:
its limit and found it lacking. Having given up genetic programs, he now
speaks of internal representations and descriptions. In doing so he is like
many other workers who have been faced with the contradictions and
inadequacies of traditional notions of genetic forms and have tried to re-
solve them, not by seriously altering their concepts, but by making the
forms in the genome more abstract: not noses in the genes, but instruc-
tions for noses, or potential for noses, or symbolic descriptions of them.
This solves nothing. At most, it communicates a certain discomfort with
the preformationism implicit in traditional formulations, but it does not
alleviate the problem; it just makes the wording vaguer.
Since so many scientists accept the idea that the body is constructed by
genetically guided processes, it is not surprising that they look for par-
allels to the development of body structure when they turn to psycho-
logical and social levels. Some of our mental, even communal, life, they
reason, is phylogenetically derived, just as our bodies are, and is there-
fore attributable to the same kinds of genetically guided development.
But insofar as cruder forms of nature-nurture distinctions imply an in-
variant biological base with environmentally wrought variations, they are
incapable of supporting this complication. A consequence is that mo-
tivation, thought, perception, and behavior (and eventually culture) are
divided into pseudolevels. And so scientists fret about how to identify the
inherited aspects of the various human temperaments (Plomin and Rowe
; by ‘‘inherited,’’ by the way, they do not simply mean ‘‘showing
heritable variation in a particular population’’), how to distinguish bio-
logical from cultural universals (Harris ), and so on. Both variable
and universal features, that is, are somehow divided into those that are
(mostly) formed from within and those that are (mostly) sculpted from
without. The dichotomous has become quadrifurcate.
But what could it mean to say that behavior of any kind is ‘‘phyloge-
netically derived’’? We are told that the behavior is ‘‘programmed’’ or in
some other way ‘‘in the genes.’’ If we are unhappy with this description,
we are hastily assured that no trait, morphological or behavioral, is actu-
ally in the genes, but rather that information or rules are encoded in the
Transmission and Construction
I admit that the term construction is problematic in some ways. For one, it
often implies an acting subject. While this is perfectly appropriate when
we are speaking of knowing subjects construing their worlds and their
own actions, it is inappropriate at other levels. We find, for example, de-
scriptions of development or behavior in which genes ‘‘foresee’’ con-
tingencies, ‘‘direct’’ events, ‘‘recognize’’ stimuli, ‘‘select’’ or ‘‘program’’
outcomes. This is construction seen as revelation (Monod :), as
fulfillment of a plan, and it is precisely the construction of ontogenetic
‘‘construction’’ that gets us into so much trouble. My meaning is quite
different. Construction, in my view, does not require a subject or sub-
ject surrogate, which is often what the gene amounts to in accounts of
development (Tobach’s cryptanthroparion; Weiss’s anthropo-
morphic principle, or my own homunculoid gene), but it does require
multilevel systems constructing themselves, bringing about the condi-
tions for their own further change. I see interactive construction, then, not
as the fulfillment of a plan, and not necessarily as the activity of a know-
ing subject, but as a developmental phenomenon that is not explicable by
only one set of its constituents or by only one of its levels.
Developmental Systems
tation. Witness one old way of defining genotype and phenotype, less
frequently seen now but not yet completely superseded: Nash, in a text
on the psychobiology of development, explains that the genotype is ‘‘the
actual genetic material,’’ while the phenotype is ‘‘the appearance of the
material’’ (:). This vision fits very well, of course, with the com-
mon account of Mendelian genetics, in which recessive traits go unde-
tected in the first generation, only to reappear in the next.1 Appearances,
we learned, can be deceiving.
Phenotypes, in this view, can ‘‘match’’ (Lindzey, Hall, and Thompson
:) or ‘‘correspond with’’ (Nash :) the genotype; or they
may not. By extension, a phenotype can also match a genotype other
than its ‘‘own.’’ The phenocopy is generally defined as just this sort of
phenomenon: ‘‘an environmentally produced analogue of some genetic
property’’ (Alland :). Norman Munn (:) gives a pair of defi-
nitions not generally used these days: Similar traits with dissimilar under-
lying genes are said to be phenotypes, and similar traits based on the
same genes are genotypes. Phenocopies could be examples of the former,
though Munn seems to be thinking of the appearance of a dominant char-
acter in heterozygous and homozygous individuals. He does not give a
term for dissimilar traits based on the same genotype; as we shall see, this,
too is basic to phenocopying.
An exploration of two of the ways the concept of the phenocopy has
been employed will reveal not only some assumptions underlying its
usage, but also the contradiction inherent in definitions like Alland’s.
Why would an author pronounce the nature-nurture controversy dead in
one breath, and in the next characterize some properties as genetic and
others (possibly identical, morphologically and physiologically) as envi-
ronmentally produced? Surely some hint is to be found in the techniques
of experimental analysis, which Alland (:) mentions as useful in
separating genetic from environmental effects. As valuable as such analy-
sis can be, it frequently leads to viewing not just variation but also char-
acters, developmental sequences, abilities, and behavior as ‘‘genetic,’’ a
practice that creates conceptual problems while clarifying little.
Looking at Development and Evolution
type, under the influence of nongenetic agents, has become like the one
normally caused by a specific genotype in the absence of the nongenetic
agents.’’
Although some phenotypes are species-typical and some are not (one
sense of genetic), some are associated with major genes and some are not
(another sense), and some characters show heritable variation in a given
population and some do not (yet a third), all characters proceed from, and
are directed and limited by, a given genotype and are thus totally genetic.
Each also depends on, and is limited and directed by, a given environment
and is thus totally environmental. This is so whether observed variation
is traceable to environmental differences or to genetic ones. Each kind of
factor, in fact, is often effective only through the other. Genes influence
development by altering the immediate environment for various constitu-
ents of the organism, while the environment frequently influences the
rate and timing of gene transcription. Although it is customary to speak
of genes as actively organizing an inert environment during ontogenesis,
Yuwiler (:–) describes the active part played by enzyme prod-
ucts and substrates in ‘‘regulating the direction of metabolic flow.’’ They
are not to be seen, he says, as simply passive material used by the genes.
Take a child with phenylketonuria (), a metabolic disorder that
usually leads to mental retardation if untreated but that can often be con-
trolled by instituting a special diet early in life. The normality exhibited
by this child on a proper diet is not an environmentally produced nor-
mality that phenocopies genetic normality; it is the result of a particular
combination of unusual genome and unusual (for us) environment. The
mental retardation of a child on an unregulated diet is similarly the
result of coaction, or constructivist interaction; it is no more or less ge-
netic than it is environmental. Either genotype or diet can be seen as the
‘‘cause’’ or determinant of level of intellectual functioning, depending on
what comparison one is making and what is being varied. The point is
that every genotype has many possible developmental pathways, many
leading to the normal range (or ranges, as in the second use of phenocopy)
and some of them not, and that often a given phenotype may be reached
via many different sets of genotype-environment relationships.2
Looking at Development and Evolution
Genetic Potential
marine worm with free-swimming larvae. The larvae that eventually at-
tach themselves to mature females are masculinized by hormones pro-
duced by the female and become males. Sex changes, behavioral and re-
productive, are observed among the coral reef fish Labroides dimiatus.
The disappearance of the male from a group is followed by increasingly
male behavior, including aggressive displays, from the dominant female.
Within four days ‘‘she’’ is able to mate with the remaining females, and
fertile sperm are produced in fourteen to eighteen days (Lerner and
Libby :). S. J. Gould (:chap. ) gives many examples of
species with alternative reproductive forms that appear in response to
changes in such variables as periods of light and dark, food supply, and
crowding, as well as amphibian metamorphosis that occurs only under
certain environmental conditions. In such multiple developmental pos-
sibilities lies a good measure of evolutionary staying power, because
they offer additional survival capabilities in the face of changing circum-
stances.
There is evidence that English-speaking psychologists, largely antibio-
logical for so long, are moving away from that stance. This, I think, is
a good thing. If it is to be a valuable development, however, the trip
back toward the biological must be more than a mere intellectual pendu-
lum swing, a shift of position along the same old conceptual dimensions
(more is written into the genetic program than we thought, etc.).
An example of misunderstanding of the biological is revealed by the
alacrity with which certain scientists have adopted the term heritability,
often to fill the vacuum left when innate and inherited fell from favor. The
connotations of fixedness that used to go with the latter two words have
frequently been transferred to heritability. The polemic on intelligence
and race still echoes, along with suggestions that we might, on the basis
of low test scores (taken as indexes of inherent potential), foreclose the
educational futures of certain children. Other areas, including sex differ-
ences, aptitude testing, and, in a more speculative vein, ‘‘human nature’’
as described by some theorists, are also characterized by assumptions
of developmental fixity and continuity. It has even been suggested that
heritability should replace instinct (Whelan ).5
In deploring such thinking I am not making a claim for infinite techno-
logical efficacy or unlimited ‘‘malleability’’; the point is not that every-
thing can be changed or improved, but rather that responsiveness to vari-
What Does the Phenocopy Copy?
If all that lies ahead is biological reductionism, then I do not greet the
new day with enthusiasm. If, on the other hand, we can extricate our-
selves from the intellectual muddle created by reifying potentials and
Looking at Development and Evolution
Howard Gruber and Isabel Sehl () studied the ways people cooper-
ate to construct knowledge that is not available to either of them alone.
Pairs of people were presented with a box in which shadows of an object
could be seen. Because each person looked at the object from a different
angle, each saw a different shape. One might see a circle, for example,
whereas the other might find a triangle. Together, the partners had to con-
struct an object that could project both of those shapes. Clearly, some
ways of working were more effective than others: Domination was not
particularly helpful in moving beyond a one-sided view, and mindless
compromise in the absence of real constructive work (the object is partly
round and partly triangular) was inadequate as well.
Like the partners at Gruber and Sehl’s task, developmentalists and
evolutionists have often had difficulty communicating with each other.
Sometimes the two groups have had divergent interests, and the lan-
guage in which they have communicated has prevented them from real-
izing it. It is possible that their difficulties are also related to the pre-
suppositions they bring to that shadow box we call science, and thus to
what they are able to construct there. Viktor Hamburger (), as noted
in chapter , has described the exclusion of embryology from the neo-
Darwinian synthesis. The lack of accord between developmentalists and
evolutionists on the evolutionary role of embryological development has
had a partial analogue in studies of behavior. Witness the exchanges in
the s, s, and s between American comparative psycholo-
gists and European ethologists (Lehrman , ; Lorenz ). To
some extent these groups talked past each other because they were inter-
ested in different matters. But I think they also had genuinely different
conceptions of development.
been guided by his explication of the ‘‘four whys’’: () the evolutionary
history of the behavior, () its survival value, () the mechanisms by
which it occurs, and () its development. Tinbergen thought that fail-
ure to distinguish among these questions had caused considerable con-
fusion. It still does. Developmental psychologists have not traditionally
thought much about the first two, and when they have, they have not
always kept their whys straight. Nor have they necessarily been helped
by their colleagues in biology. Too often, the price of admission to the
biological brotherhood has been a view of development marked by reli-
ance on assumptions that are decidedly uninformed by systems thinking,
despite a vocabulary liberally sprinkled with systems terms (e.g., Fish-
bein ; Scarr and McCartney ). The notions of genetic programs
and instinct, for instance, draw some of their authority from their associa-
tions with evolutionary thought. Programs and instinct carry all sorts of
other implications about mode of development and kinds of mechanism,
however—autonomy, internality, spontaneity, naturalness, resistance to
perturbation—neatly tying together the four whys that Tinbergen distin-
guished. (He did not claim that they were unrelated, just different. He
even discussed some ways of relating them []; see also P. P. G. Bate-
son .)
Imagine for a moment two people at a shadow box. This time they have
been given different instructions: One partner is to discern the object’s
shape, and the other must determine whether or not it is moving. Imagine
also that they must use an ambiguous vocabulary. Rooving, for instance,
means both ‘‘round’’ and ‘‘moving.’’ One person reports that the object is
round, and the other concludes that it is mobile. I am suggesting, in this
crude way, that their plight resembles that of scientists who must work
with ambiguous terms such as inherited, genetic, biological, and matura-
tional.
The ambiguity of these terms permits the blurring of Tinbergen’s whys.
If one worker discovers that a bit of behavior is present in phylogenetic
relatives, for instance, the other may conclude that it will be develop-
mentally stable. If one claims that a pattern is adaptive, the other may
deduce that it develops independently of experience. Evidence for one
why masquerades as evidence for another. If their aims and terms were
clarified, the researchers might conclude that they were pursuing sepa-
rate projects. But the goal is not separatism; it is rather the recognition of
Ontogeny and the Central Dogma
The usual way of construing the relationship between ontogeny and phy-
logeny involves several interrelated ideas. First, evolution is defined by
changes in gene frequencies. The genes are thought to produce pheno-
types by supplying information, programs, or instructions for the body
and for at least some aspects of the mind. Some genes produce better
phenotypes than others and are differentially passed on. Although in this
view the environment is necessary for proper development, its effects
on the phenotype are evolutionarily insignificant because only inherited
traits are transmitted. Causal power and information are carried in the
, and living things are created by an outward flow of causality and
form from the nucleus.
This conception of the ontogeny-phylogeny relationship seems to call
for two kinds of development: one for inherited traits and one for every-
thing else. This is true even though statements are routinely made about
the impossibility of attributing traits completely to one or the other. De-
spite their reassuring ecumenical ring, such statements either retain the
dichotomy or turn it into a continuum. Emblematic of a trendy but failed
‘‘interactionism,’’ they are responses to a multitude of developmental
observations that call traditional formulations into question; their short-
comings are examined later. (For critiques of this kind of well-inten-
tioned but conceptually misguided interactionism, see Lewontin, Rose,
and Kamin ; Oyama , ; Tobach and Greenberg .) Al-
though they scornfully dismiss ‘‘extreme views’’ that attribute behavior
entirely to the genes or entirely to the environment, the devotees of this
popular interactionism mistake compromise and relabeling for concep-
tual resolution.
Looking at Development and Evolution
We need to alter our conceptions of ontogeny and phylogeny before
we can bridge Hamburger’s () ‘‘nucleocytoplasmic gap.’’ We do not
need more conciliatory declarations that nature and nurture are both im-
portant, but rather a radical reformulation of both. All too often, as we
shall see below, people confuse the genotype with the phenotype. Those
who manage to avoid that oft-sprung trap may still become obsessed with
genes because they have been encouraged in innumerable ways to think
of them as prefiguring or making the phenotype. I propose the follow-
ing reconceptualizations, in which genes and environments are parts of
a developmental system that produces phenotypic natures:
. Nature is not transmitted but constructed. An organism’s nature—
the characteristics that define it at a given time—is not genotypic (a ge-
netic program or plan causing development) but phenotypic (a product
of development). Because phenotypes change, natures are not static but
transient, and because each genotype has a norm of reaction, it may give
rise to multiple natures.
. Nurture (developmental interactions at all levels) is as crucial to typi-
cal characters as to atypical ones, as formative of universal characters as
of variable ones, as basic to stable characters as to labile ones.
. Nature and nurture are therefore not alternative sources of form and
causal power. Rather, nature is the product of the processes that are the
developmental interactions we call nurture. At the same time, that pheno-
typic nature is a developmental resource for subsequent interactions. An
Ontogeny and the Central Dogma
and sentences, of genetic codes and grammars, is to some extent the re-
sult of notions of life as language. The genes become the repository of
true nature; molecular meanings are contained in a phenotypic vessel,
which is sometimes treated with so little regard as to render it virtually
transparent. A few paragraphs ago I mentioned confusion between geno-
type and phenotype. This may have stuck some as too bizarre to be cred-
ited. Yet consider some psychologists’ practice of effectively bypassing
the phenotype: Plomin (:, ) speaks of people responding to
children’s ‘‘genetic propensities’’ and ‘‘genetic differences’’ rather than
to the children themselves. Similarly, Scarr and McCartney (:)
refer to experiences that ‘‘the genotype would find compatible.’’ The lat-
ter authors use the term developmental system, but their insistence that
genes and environment play quite different roles in this ‘‘system’’ re-
veals the distance between their understanding of this term and my own:
In their scheme, of course, the genes play the determining role. (De-
spite the importance of both, that is, some causes are more equal than
others. For contrast, see Fogel and Thelen ; Valsiner , on sys-
tems dynamics.)
Genes appear to link evolution and development in two ways. First, they
are the material link that promises to make sense of heredity. The adop-
tion of Francis Crick’s () Central Dogma of the one-way flow of
information as the ruling metaphor for development forged a second,
conceptual link. The dogma states that information goes from genes to
proteins, never from proteins back to genes; as metaphor, it takes many
forms (programs, blueprints, instructions; see Newman ), but they
always involve the emanation of basic developmental causation from the
. An outward flow of information and power achieves the translation
of the genetic message in ontogeny.
A subtle transition is thus made from ‘‘messages’’ about molecules to
messages about bodies and minds—quite a different thing, whether we
realize it or not. The shift is from gene transmission to trait transmission.
Focus on the gene as the prime mover of ontogeny leads to all sorts of
assumptions about genetic control of development as the defining char-
Ontogeny and the Central Dogma
acteristic of certain traits; this in turn leads to the need for another kind
of process to explain everything else.
The more reductionist one is, the harder it is to appreciate the gap be-
tween the molecular and the organismic levels. The kind of reduction-
ism I am speaking of here is not the provisional single-mindedness that
allows detailed investigation of a mechanism. It is rather the desire to in-
terpret the whole world in terms of that mechanism, or at least in terms
of the level at which it was studied; it is the failure to shift levels or point
of view, whether from inability or from some conviction that to do so
would be soft-headed. The genetic program holds a fatal attraction for
such minds.
Crick’s Central Dogma has come to have the quality of an unquestion-
able truth.We forget that it is a hypothesis about specific molecular inter-
actions that is open to empirical support or refutation. But what could
challenge the ‘‘plain truth’’ that development is controlled by a genetic
program? What would count as evidence for or against such control? One
biologist told me that ‘‘the whole of molecular biology’’ demonstrated the
reality of the genetic program—hardly the language of normal scientific
inference. Genetic programs have so dominated mainstream thinking that
people have rarely been called on to defend them. Programs have func-
tioned as an ‘‘enabling concept,’’ important not only in research but in
the legitimation of a kind of reductionist explanation (Yoxen :).
Richard Dawkins (:) speaks of as programs, instructions,
and algorithms, and explicitly denies that these terms are metaphors. Of
airborne seeds he says, ‘‘it is the plain truth [that it is raining instruc-
tions]. It couldn’t be any plainer if it were raining floppy discs.’’ Organ-
isms become ‘‘natural-technical objects structured by logics of domina-
tion,’’ ‘‘biotic components in a technological communications system,’’
‘‘command-control systems’’ (Haraway –:, , ).
When system is simply shorthand for ‘‘machine governed by a pro-
gram,’’ it usually signals a preoccupation with static centralized control
rather than the sort of distributed, dynamic, contingent control under
consideration here. (For an analysis of the popular use of systems and its
links to a preoccupation with control, see Rosenthal :chap. .) We
thus have an uneasy association of systems terms with a style of expla-
nation that is, as previously noted, uninformed by the kind of systems
theory that offers the most to developmental studies.
Looking at Development and Evolution
Two Strategies
The problem with attributing some parts of the phenotype to the genes
and some to the environment is that developmental processes and prod-
ucts are simply not partitionable in this way. Nor are the various criteria
for making the distinction consistent. Although the concerns motivat-
ing any particular nature-nurture distinction may be interesting and im-
Ontogeny and the Central Dogma
of the genotype in the phenotype’’ (p. ), and the ‘‘genotype has only
those degrees of freedom that are inherent in its genes’’ (p. ). Although
she claims that it is incorrect to say that ‘‘heredity sets the limits on devel-
opment’’ (p. ), that is exactly what is entailed by the notion of genetic
degrees of freedom. This is all quite ironic because the array of pheno-
types that could be associated with a given genotype is just the array in
which all differences are environmentally determined. The environment,
after all, is seen as ‘‘selecting’’ the particular outcome. The norm of re-
action is therefore a nice demonstration of the joint determination of the
phenotype. Every organism incorporates ‘‘information’’ from genes and
environment in a complex that cannot be partitioned as variance is parti-
tioned. But such mundane truths do not seem to be the point here, which
is rather a kind of metaphysical urge to contain ontogenetic variety within
genetic boundaries.
The problem with this imperialistic version of developmental dual-
ism is that it is vacuous; a genotype has just those developmental possi-
bilities that it has (though who is to say what they are). Used this way,
the program no longer has empirical content. It is more like a symbol
of ultimate faith. Or it may be only a fancy way of saying that poten-
tial is finite. In fact, one variant of this idea is ‘‘programmed potential’’:
Mayr (:) claims that ‘‘the range of possible variation is itself in-
cluded in the specifications of the code’’ of the genetic program. But be-
cause the range of possible phenotypes is defined by the set of genotype-
environment pairings, what is the point of attributing that range to just
one member of the pair? And why insist that the range be fixed at fertil-
ization? Potential must be a developmental concept if it is to be useful. It
cannot be treated as a fixed quantity somehow inscribed in the genome
(Horowitz ; Lewontin ). As many have noted, it is just this idea
of fixity that has led people to draw conclusions about things such as
intellectual potential from heritability figures.
A given genome may certainly have several developmental possibili-
ties. But those possibilities vary with the developmental state of the or-
ganism and the context. Traditional notions of maturation, readiness, and
embryonic competence turn on the realization that possibilities must
emerge in ontogenesis. A bee larva has at one moment the potential to
become a queen or a worker but a short time later the worker-to-be may
no longer aspire to royalty. Its genes are the same, but its effective poten-
Looking at Development and Evolution
‘‘’’
ration has largely escaped scrutiny, probably because nativists and em-
piricists alike shared basic beliefs about physical development (Oyama
). Even the most dedicated behaviorist requires a body and a reliable
set of operants and reflexes to begin a conditioning story. Yet, research-
ers eventually realized that some learning is species-typical, and indeed
is crucial to many ‘‘instincts’’ (avian imprinting is the classical case, but
see also Hailman ).4 This threatened two traditional renderings: of
instinct as unlearned and of learning as arbitrarily variable. The resolu-
tion proposed by some theorists, however, involved not a reconsideration
of the notion of genetic programming, but an increase in the scope of the
program. Not only are bodies and unlearned behavior in the genes, now
some learning is in there, too.
These theorists’ attempts to replace the innate-acquired distinction
with closed and open programs or with inherited ranges of possible forms
(Lorenz , ; Mayr ), or innate, genetically determined epi-
genetic rules (Lumsden and Wilson ) merely blur the distinction be-
tween traditionally conceived nature and nurture when they should be
questioning the very basis of that distinction. Fishbein’s () descrip-
tion of ‘‘canalized’’ learning as genetically preprogrammed development
is typical of this approach, but many others have made similar attempts
to reconcile species-typical learning with the ideal of genetic control
(Freedman ; J. L. Gould ; Shatz ; and see Johnston’s
critique of innate templates in avian song learning). Perhaps these efforts
are unsurprising in light of the fact that species-typical, and more often,
species-specific, became euphemisms for instinctive and innate in the dis-
course of workers who tried to take critiques of instinct into account
but did not realize that conceptual tightening can take more than an ad-
justment of the lexicon. (An even more recent example of such hedg-
ing is the ubiquitous term constraint, discussed in chapters and .)
Such maneuvers unfortunately tend to be presented as the leading edge
of developmental theory, where an ‘‘evolutionary perspective’’ too often
means making more and more refined nature-nurture distinctions while
attributing more and more to the formative, directive powers of the genes.
These theorists claim to be eliminating the nature-nurture dichotomy, but
in reality they are simply renaming it and shifting phenomena from one
side of it to the other.
Parallel with the inclusion of some learning in the concept of instinct,
Ontogeny and the Central Dogma
some investigators (Lorenz ; Mayr a) saw that any idea of
species-typical development would have to include the possibility of
branching pathways to accommodate certain kinds of adaptive variation
(alternative morphologies or behavior patterns). External events, that is,
often intrude into the supposedly autonomous maturational sequence to
move the organism onto one or another path.
So we see that learning can be necessary for the development of be-
havior usually defined by the absence of learning (instinct), and divergent
pathways can be crucial for development usually considered unilinear
(maturation). This shows the impossibility of consistently categorizing
developmental phenomena as innate or learned. Putting them on a con-
tinuum defined by varying amounts of genetic control does not solve the
problem; it multiplies it. The hazards of switching from one definition of
innateness to another were pointed out above in the discussion of Tin-
bergen’s whys; in some of the works cited previously we see the conse-
quences of confusing innate as species-typical, innate as predetermined,
innate as conferring survival advantage, innate as unlearned, innate as
having an evolutionary history, and innate as independent of the environ-
ment. These failures of consistency could have challenged the concep-
tual framework of developmental dualism. Instead, however, the offend-
ing phenomenon in each case was simply assimilated to the old system,
giving us genetically programmed learning and genetically programmed
developmental branching. The apparent adaptiveness of these violations
of ontogenetic autonomy, as well as their selectivity and presumed evo-
lutionary histories, compelled workers to find a way to attribute them to
the genes, even though this necessitated finessing the Central Dogma of
development (that ‘‘the biological’’ is created exclusively by the outward
flow of genetic information). These efforts, however well-meaning, are
finally just superficial responses to profound conceptual problems. That
such phenomena can be treated nondualistically is evidenced by Caro and
Bateson’s () analysis of alternative tactics.
play ‘‘cognitive’’ roles that the child is not yet ready to perform. (See
also Rogoff and Wertsch , on Vygotsky’s concept of ‘‘zone of proxi-
mal development.’’) Kaye sees many universal skills as constructed in
interaction rather than revealed in maturation. Yet he is quite happy to
speak of the abilities the infant brings to early infancy as innate, maturing
‘‘according to the designs of the genetic program’’ (p. ). He explicitly
excludes from psychology the study of maturation, which is ‘‘guaranteed
by the genotype’’ (p. ). (The problem is not that he attributes the wrong
things to maturation, but that he attributes maturation and innateness to
a genetic program.) He also criticizes those who liken development to
a train ride ‘‘in which the very process of the journey is determined by
its destination . . . because it suggests that the child knows where he
is headed’’ (pp. –). Ironically, the intrinsic genetic program is an
explanation-by-destination in which it is not the child who knows where
he is headed but his genes.
Kaye (:) rejects some nativist accounts of human development,
charging that those who overestimate the functions present at birth ‘‘try
to explain away some of the mysteries that have led so many psycholo-
gists to begin looking at infants in the first place.’’ But the concept of
innateness does just that. (Kuo also made this point about the concept
of instinct, in .) Kaye’s assertion that infants ‘‘inherit certain aspects
of their social environments as much as they inherit their nervous sys-
tems’’ (p. ), although a provocative step in the right direction, reveals
the problem. In my terms, he has mixed developmental influences (what
I call ‘‘interactants’’ or ‘‘means’’) with developmental products. Genes
and social environments are inherited interactants, available to be used in
constructing a life cycle. Nervous systems and social skills, being pheno-
typic, must develop.
Kaye associates evolution with innateness and so is forced to circum-
scribe his ‘‘interactionism.’’ (Innate behavior is cross-referenced to evo-
lution in his index.) It follows (:–), then, that the ‘‘genetically
determined behavioral tendencies’’ of parents are inherited biologically,
while the rest is inherited by the mechanisms of cultural evolution. Kaye’s
attempt to reconcile the role of experience in development with evolution
is a commendable one. If he had applied his constructivist thinking to
the concepts of innate behavior and maturation as well as to social devel-
opment, his account would not have required the developmental dualism
Looking at Development and Evolution
that now permeates it. This chapter contains many examples of nondual-
istic research on the developmental role of naturally occurring experi-
ences. It is true, as Kaye says, that developmental psychologists seek a
set of ‘‘givens’’ with which to begin their inquiries, but there is a differ-
ence between taking a set of abilities as given, or present, at some age
and attributing them to a particular kind of developmental process.5
opment on the organism’s own activity and its interactions with conspe-
cifics takes us even farther out. Biochemical and even social interaction
with parents or siblings can begin before birth or hatching (Gandelman,
vom Sall, and Reinisch , on the influence of fetal position on devel-
opment of mice; Gottlieb , on the effect of prehatching experience in
ducklings) and can obviously be of great importance later as well (John-
ston and Gottlieb ; Lickliter and Gottlieb ).
Of great utility here is Gottlieb’s () concept of bidirectional rela-
tionships among gene action, physiology, function, and social influences.
The principle is nicely demonstrated in Vandenbergh’s () account
of the regulation of mouse puberty by other mice. Vandenbergh’s paper
also serves as a model for relating individual and population levels. A
good presentation of many infant-parent relationships is found in Cairns
(), which shows the multiple ways investigators may fruitfully move
among different fields. That book is also full of illustrations of the so-
cial embeddedness of development, and thus of the ways that similarity
across generations may be maintained or compromised by interactions
among organisms. This embeddedness is also evident in research on the
transgenerational perpetuation of behavioral sex differences in rodents
(Moore ) and of food preferences and other behavioral patterns in
a wide range of species (Galef ). Similarly, Trevarthen (:)
shows how human infants’ ‘‘mental partnership with caretakers’’ extends
their abilities to act. He comments ironically on the idea of the ‘‘isolated
thinker’’ and declares that infants ‘‘must share to know’’ (p. ).
These are all examples of developmental research that highlights the
connectedness of the emerging organism to its surroundings, not its in-
sulation from them. Although such attention to developmental context
provides an excellent vehicle for the study of individual variation, it does
not limit researchers to such study. Indeed, it can also show the ways in
which specieswide patterns of development are maintained by stably re-
curring contexts, and how these patterns of development play a role in
maintaining that very recurrent stability.
It seems quite possible that much of this research comes from its au-
thors’ appreciation of systemic complexity and their realization that one
must move an investigation beyond the boundaries of the organism in
order to understand the organism fully. This ability to see links and inti-
mate interchanges with the surround as developmentally fundamental is
Ontogeny and the Central Dogma
Interaction in Ontogeny:
Sources of Variation, Sources of Form
tem, and thus may vary with it.We can speak, then, of genetically or envi-
ronmentally determined variation, but not of genetically or environmen-
tally determined traits. The fact that the difference between two cell types
is due to extracellular conditions does not make the cells themselves
‘‘environmentally determined’’ any more than a trait that shows heri-
table variation in a population is ‘‘genetically determined.’’ Phenotypic
‘‘messages’’ are constructed in interaction. This is true whether we study
species-characteristic development, as Piagetians do, or species-variable
ones, as students of temperament or personality do. Constructivist inter-
actionism, that is, should not be associated only with variability or mu-
tability. It is not, as previously noted, a code word for environmental
determinism; nor does it signify some overarching preoccupation with
‘‘plasticity.’’
Interactions between chemicals, between tissues, between organisms,
and between an organism and the inanimate environment are parts of
the developmental system, and the immediate context of the interaction
may be as important as the identity of the interactants. For some animals,
‘‘context’’ is not restricted to physical environment. Interpretation of the
situation is crucial. Consider the change in effective context when a sub-
ject of the television program Candid Camera realizes what is afoot. Or
think of the dilemma of the psychologist who wonders whether subjects
simulating some social process are a good model for what occurs outside
the laboratory. The question is really about what situation the subjects
are in, and inspection of the room, even through a one-way mirror, will
not necessarily give the answer.
The vocabulary of interactionism has been widely adopted, but the full
implications of a constructivist interactionism have not been accepted
nearly as readily as its terminology. Taking interactionism seriously
means rejecting the Central Dogma as a metaphor for the control of de-
velopment, even for development of the body. (Notice that this meta-
phorical notion of information flow is independent of the question of re-
verse translation in molecular biology, which is the empirical issue Crick
was addressing.) The one-way causation it implies is inconsistent with
the reciprocal, multiple causation actually observed in vital processes.
Interaction requires a two-way ‘‘exchange of information’’: Genetic and
organismic activity are informed by conditions, even as they inform those
conditions. This is nothing more than the bidirectionality so commonly
Ontogeny and the Central Dogma
What Is Inherited?
are passed on. The rest of the normal environment is also quite reliably
present (that is what it means to be a normal environment), and when it
is not present, some other environment is.8 Biologists also speak of cell
state being stably ‘‘inherited’’ (Alberts et al. :); the key here is
not change in genetic material, but transgenerational (here, generations
of cells) stability of the cell type. There are, in other words, both species-
typical aspects of developmental systems and variations in those systems,
genetic and otherwise, and an organism inherits the entire complex. The
fact that we daily acknowledge the indispensability of both genes and
surround to the development of all characteristics and yet continue to at-
tribute some of that development mostly to the genes and some to ‘‘other
factors’’ suggests that our theoretical vocabulary has not kept up with
our observations. Both evolutionary and developmental studies remain
largely genecentric (Goodwin ).
What passes from one generation to the next is an entire developmen-
tal system. Heredity is not an explanation of this process, but a statement
of that which must be explained (see my previous discussion of links
between generations). The concept of evolving developmental systems
gives a unified view of development while integrating it with evolution.
Dualism is no longer required; the inherited-acquired distinction, as long
as it is construed as a distinction between kinds of developmental processes
or sources of form, can be eliminated—not modified or turned into a mat-
ter of degree, but eliminated.9
Ontogenetic means are inherited; phenotypes are constructed. This en-
largement of the idea of inheritance seems outrageous to minds trained
to identify it with the genes, but I am only making explicit what is rou-
tinely taken for granted. No one claims that genes alone are sufficient for
development, or denies that environments, organic and inorganic, micro-
scopic and macroscopic, internal and external, change over organismic
and generational time. What is missing from most accounts is the syn-
thetic processes of ontogenetic construction.10 Inheritance is not atom-
istic but systemic and interactive. It is not limited to genes, or even to
germ cells, but also includes developmentally relevant aspects of the sur-
round—and ‘‘surround’’ may be narrowly or broadly defined, depending
on the scope of the analysis. Inheritance can be identified with ‘‘nature’’
only if it embraces all contributors to that nature, and nature does not
reside in genes or anywhere else until it emerges in the phenotype-in-
Looking at Development and Evolution
transition. Nature is thus not properly contrasted with nurture in the first
place; it is the product of a continual process of nurture.
Having redefined inheritance in this way, we must also redefine on-
togeny. This is difficult for behavior scientists, who are used to squab-
bling over degrees of biological programming of personality or behavior
but who tend to take programming of the body for granted. Although
there may be doubt, that is, about whether sex roles and aggression are in
the genes, there is usually no doubt at all that sex organs, teeth, and claws
are. (Recall the incomplete interactionism discussed above under ‘‘Deep
Problems, Superficial Solutions’’). But a unidirectional flow of genetic
information doesn’t account for a tooth or a claw, red as it may be, any
better than it accounts for the most idiosyncratic behavior. The develop-
mental system, on the other hand, accounts for the emerging phenotype
in a way the naked genome cannot.
Some fear that the concept of the developmental system requires them
to give up too much; in fact, it only eliminates a troublesome set of as-
sumptions and inferential habits. Many of the issues formerly associated
with the false opposition between nature and nurture (or biology and cul-
ture) can still be addressed, but this time clearly formulated and properly
distinguished from others. Some are largely evaluative (many concepts
of normality, for instance) and are answerable not solely by empirical in-
vestigation but by moral discourse as well. If this conceptual unpacking
is performed, we will be less likely than we have been in the past to make
ungrounded predictions, to draw illegitimate conclusions from our data,
to posit distinctions where there are none.
The developmental system is a mobile set of interacting influences and
entities. It includes all influences on development, at all levels of analysis.
Any particular investigation will obviously focus on a limited portion of
the system. For an embryologist, the scale of cells and organs defines the
investigative field, and higher-level aspects of the system can generally
be taken for granted. In some cases, though, as we have seen, it becomes
useful to pay attention to other factors as well, for the wider environment
may also intrude; witness the effects of radiation or of various chemicals
on embryogeny, or other ways in which the experience of a mother may
Ontogeny and the Central Dogma
. The genome, whose parts interact and move about in ways now being
described by molecular biologists (Dillon’s book is called The
Inconstant Gene; see also papers in Milkman ).
. Cell structure, including organelles, some of which have their own
distinctive , and seem originally to have been internal symbionts
(L. Margulis ).
. Intracellular chemicals, some of which (e.g., messenger from
previous generations) may allow considerable developmental
progress before the organism’s own genes are transcribed at all (Raff
and Kaufman ).
. Extracellular environment—mechanical, hormonal, energetic—parts
of which, like the extracellular matrix, are created by the cell itself or
by other cells.
. Parental reproductive systems, both physiological and behavioral;
prenatal effects are common, and cross-fostering experiments can
Looking at Development and Evolution
In many life cycles, a variety of factors can bring about the major
branchings of the developmental pathway discussed earlier. Other influ-
ences contribute to less dramatic variations, including variations in learn-
ing and anything else that helps define the norm of reaction. These factors
are also part of the developmental systems of these organisms.
An old maxim goes, ‘‘For want of a nail the shoe was lost, for want of a
shoe the horse was lost, for want of a horse the rider was lost . . .’’ Loss
of a particular rider might even lead to losing an entire battle. To know
whether that nail makes the difference between losing and winning the
battle, shouldn’t we know what kind of battle it is, on what terrain it is
being joined, what the command structure is, and who the opponent is?
Even if it could be shown that the loss of a battle were traceable to a lost
nail, this would not make the nail an adequate causal explanation for the
entire complex of events that constituted the battle. Indeed, it is the entire
complex that defines the nail’s role.
The nail can become a nail ‘‘for’’ losing battles only in a world that
is sufficiently stable and integrated that the entire battle—to say nothing
of the geopolitical circumstances that led to it—can be re-created with
some regularity. A gene is a gene ‘‘for’’ a given phenotypic difference
only if other aspects of the ontogenetic complex (not to mention popula-
Ontogeny and the Central Dogma
nificant political and moral repercussions far beyond the research com-
munity. It has multiple sources in our philosophical history. It influences
the classification of individuals in ways that profoundly affect their future.
It influences our view of what is possible for individuals and for the
species as a whole. Thus, it affects the manner in which we mobilize for
maintenance or for change—indeed, whether we mobilize at all.
Hamburger () says that embryology was not integrated into the
neo-Darwinian synthesis that had apparently unified so much of biology
for three principal reasons. First, evolutionists tended to focus on the out-
ward flow of influence from the nucleus, whereas embryologists focused
on the cytoplasm as crucial in determining differential gene activation.
Second, the preformationist implications of the notion of particulate in-
heritance made embryologists uneasy. Third, evolutionists emphasized
trait transmission across generations and neglected trait elaboration over
the life cycle. These factors are still barriers to effective communication
between at least some developmentalists and some evolutionists. They
raise thorny problems for those who try to be both at once; as we have
seen, nucleocentrism is problematic even for those who have no special
interest in evolution. Hamburger calls for an interactive view to bridge the
nucleocytoplasmic gap into which so much misunderstanding and acri-
mony have been spilled. I believe this need is met by the constructivist
approach described here. This approach offers a way of speaking about
complex transgenerational continuity and variability, about stability and
change in both species and individuals, while allowing us to acknowledge
the intricacy and contingency of the processes observed in ontogeny, a
way to think in evolutionary terms without being committed to a de-
velopmental dualism in which contingent nurture is pitted against geneti-
cally predetermined nature. We do not need the genetic program in order
to have an evolutionary perspective.
4 Stasis, Development, and Heredity:
Richard Lewontin (, a) has described two models of change that
have been treated as alternative explanations for development and evolu-
tion. Each combines two ideas that are not necessarily connected. In each
case, I suggest, it is possible to accept the first idea and reject the second.
In the transformational model of evolution: () change in a collection
is explained by change in its constituent entities, and () change in the
entities is predetermined and uniform.
In the variational model: () only some of the variants in a collection
are perpetuated, and () the variants themselves are static.
Change is assumed to be internally driven in the former model and ex-
ternally directed in the latter. Notice, though, that these assumptions stem
from the second component in each—predetermined change and static
Stasis, Development, and Heredity
hold on our thought and action, but what is finally important is how we
construe the concept and how we demonstrate its appropriateness. The
developmental system includes not only the organism but also the fea-
tures of the extraorganismic environment that influence development. By
positioning the organism in the ecological context in which it exists and
grows, this view overcomes the alienation of organism and environment
of which Lewontin has complained ( and elsewhere; see also Gray
) while restoring change and relation to the world we study.
It is true that one can take the snapshot attitude mentioned above, stop-
ping time to create arrays of features for comparison. Transgenerational
stability can then give the impression of immobility; photographs of simi-
lar objects, even if widely separated in time, may seem to be images of the
same object.We speak of ‘‘the shark,’’ for instance, remaining ‘‘the same’’
for millions of years, as though it were a single shark, not a succession of
life cycles. Interestingly enough, George Williams (:), seeking to
rebut Sober’s criticism of his gene-level account of evolution, makes this
very point: that phenotypes do not actually persist across generations,
but rather recur. Genes, he says, persist. They are thus the real objects of
selection. The obvious objection is that genes themselves recur by repli-
cation. This Williams counters by detaching the notion of the gene from
its physical embodiment, characterizing it instead as a ‘‘weightless pack-
age’’ of information coded in the structure of . But even leaving aside
the complexities of overlapping and discontinuous genes and the prob-
lems they pose for the concept of coding, the information-as-sequence
must be reconstituted as well. Disembodiment seems not to solve the
problem of discontinuity.Williams seems, in fact, to be saying that physi-
cal persistence is not important after all. Reliable availability in each life
cycle, however, is. Although I agree with him, I think this admission
undermines his insistence on the primacy of the genetic level.
Williams (:) writes of ‘‘natural selection of information for its
effects’’ and says the logic of selection would permit other mechanisms of
information transmission besides replication. But if the particulars
of gene replication are not necessary to the logic of natural selection, one
could argue that evolution would proceed even if the genes materialized
only a nanosecond before they were used. Although does not behave
Looking at Development and Evolution
this way, the chemical complexes for transcription and translation are as-
sembled on the spot, just as many other aspects of the developmental
system are. More to the point, if ‘‘information’’ is defined by its contribu-
tion to development, as Williams’s phrase implies, one need not look far
for alternative means of transmission; organisms have a variety of media
at their disposal. Cell structure (both membrane and cytoplasm) is part of
the developmental system that connects the generations, as are chemicals
derived from maternal and myriad extracellular and extraorganis-
mic influences. Many of these are provided by parental physiology and
behavior; others are reliable aspects of the larger environment, including
the social environment. The ‘‘informational’’ function of any influence is
determined by the role it plays in the developmental system as a whole.
Regularity of gene function is thus a result of developmental regularity
as well as a cause of it. Any biologically interesting notion of informa-
tion must be interactively defined in this way, and what is crucial is not
permanence but availability at the appropriate time. Activity, not stasis,
and relation, not autonomy, are central to this conception. Persistence is
beside the point in accounting for reliability.
It is possible for features to evolve without appearing in each gen-
eration or in all natural environments, and even the most reliable fea-
tures have an ontogeny. If developmental courses are regular, they can
for some purposes be taken for granted. (It does not follow that taking
them for granted makes them regular, and taking them for granted is not
the same as understanding them.) One can explain a bed of red flowers by
pointing out that only certain kinds of bulbs were planted, even though
bulbs have no flowers, the flowers are fleeting, and under other condi-
tions the bulbs might have produced blooms of a different shade, or none
at all.
Dynamic stability, then, is needed if natural selection is to affect a
population. It is a contingent matter whether and how the conditions
for construction of a feature are correlated with the conditions for its
use. The variants must recur, and so must the causal background against
which a given phenotypic difference makes for a difference in fitness.
But, as we have already observed, people often reverse this reasoning
and assume developmental stability from some evolutionary argument or
other. Having concluded that a feature is in some sense adaptive, they
Stasis, Development, and Heredity
clude all of them; the organism then inherits (has available to it) a highly
complex, widely ramified developmental system.
Some seek to solve the problem of the developmental insufficiency of
the genes by including in the hereditary package some cytoplasm, tran-
scripts of maternal genes, and other cellular features. This is marginally
better than focusing on naked . It tends to pass muster because it re-
mains comfortably within the boundary of the egg, the traditional vehicle
of heredity. But no egg can develop in a vacuum, and other ontogenetic
requirements must be provided in some way. Many are reliably present in
the wider environment, but many others may have to be supplied by the
organism itself, by parents and other conspecifics, or even by organisms
of other species. (Consider mutualisms, in which members of one species
feed, protect, transport, or pollinate members of another. Systems can
also be nested, as with endosymbionts.) The developmental system in-
cludes, then, not just genes, but whatever else in the living or nonliv-
ing environment contributes to or supports development. The system’s
constituents and configuration shift with its activity. In this view many
parts of the changing complex are themselves generated by it. Hormones
produced by the developing organism, for instance, serve as the chemi-
cal environment for further chemical interactions. Activity may provide
necessary self-stimulation or social contacts. Whether the interaction is
within the organism or between it and its milieu, both external and inter-
nal conditions are important, and the dynamic exchange characteristic of
biological processes finally makes the division artificial.
If, on the other hand, what is inherited is that which makes one trait
different from some other one (the ‘‘inherited differences’’ approach),
then not only the influence that makes the difference but the system in
which it makes that difference must be inherited. There are also many non-
genetic influences that can play this role of difference maker, generation
after generation (Cohen ; Gray ). Consider the ability of customs
and social structure to maintain differences among humans. Whether one
speaks of traits or differences, then, what one must ultimately reckon
with is a full developmental system.
Development is thus basic to all notions of inheritance. What I have
in mind is not the sort of unfolding of predestined fate involved in the
transformational model. That would be revelation. Forms emerge. They
are neither imposed from without nor ‘‘translated’’ from within. Nor do
Stasis, Development, and Heredity
the genes give the basic outlines while the environment provides the de-
tails. Just as unsatisfactory is a dual model (some things unfold, some do
not) or a continuum (there are degrees of programming). All develop-
ment involves contingency and ordered (but real) change. It is a process
generated by interaction of a heterogeneous mix of influences, and its
mobile unity and order derive from its embeddedness in this matrix of
causes, not from insulation against it.
enhances the likelihood that the new system will recur. If the perpetua-
tion of variations is a joint function of active organisms and surrounding
conditions, and if development is structured but not predestined, then the
variational and transformational accounts of population change can be
synthesized (Gray ).
I have touched on some ways the environment has been personified in de-
scriptions of natural selection. I have also maintained that the treatment
of the ‘‘innate’’ as produced by static internal plans is more a product of
conceptual confusion than an adequate empirical description. And I have
examined the role of the metaphor of trait transmission in joining these
views of evolution and development.
These mutually reinforcing metaphors often lead us to associate evolu-
tion with stasis. Stasis (or, better, some stable set of conditions and rela-
tions) is projected backward or forward in time. This is a paradox because
evolution is defined by change, depends on change, and makes other
kinds of change possible. Nevertheless, as I mentioned earlier, people in-
fer developmental fixity from a phylogenetic history or assume a history
of natural selection when a trait is pronounced ‘‘biological.’’ Heredity
does link development to evolution, of course, but the relation is mis-
construed when the processes are misconstrued. I have proposed a wider
definition of inheritance than is generally used, one based on process
rather than product, on the ‘‘transmission’’ of developmental resources
rather than of traits.
None of the assumptions of stasis associated with evolution by natu-
ral selection is justified. Stasis at selection is permitted by a selection
mechanism but not required by it, even though our observational prac-
tices tend to freeze time in cross sections. The assumption that organisms
under selection are passive reflects a tendency to think about these inter-
actions as choices made by an outside agent. The stability requirement,
that a selected entity must actually manifest itself in successive genera-
tions with some regularity, entails reliable ontogenetic reconstruction,
and this is a matter of process, not immobility (Ho a, b).
Stasis, Development, and Heredity
are told that we must carefully ‘‘teach our children altruism, for we can-
not expect it to be part of their biological nature,’’ a nature defined by
their selfish genes (Dawkins :).
Allied to this notion of natural state is the assumption that such ten-
dencies will be hard to deflect. Attempts to do so are likely to fail or
to incur unwanted consequences. (Kitcher :chap. speaks of pre-
cluded states and states that preclude important desiderata. See Oyama
:chap. , for other examples.) Natural tendencies will press for ex-
pression, and we may suffer if they are denied. Instincts and psycho-
logical drives fit well here. We are even told what kinds of political sys-
tems are incompatible with our biological nature (E. O.Wilson :).
Many social scientists have been quick to accept biologists’ definition of
a universal, fixed human nature and to draw conclusions about possible
social arrangements (see, e.g., Peterson and Somit ). One question
never addressed in these treatments is why only deviation from the pro-
gram is thought to entail stress. This is a peculiar omission unless one
assumes that the natural is also good and easy. This seems a small point
until one realizes that inordinate focus on the putative costs of changing
may divert attention from the costs of not changing.
The natural state model encourages a sequence of inferential leaps
among the present, the past, and the future. It assumes just the conclu-
sions we should be questioning and takes for granted just the complicated
developmental and social systems we should be investigating. Finally, it
draws attention away from the many kinds of regularity that would have
been required as preconditions of such evolution.
Many questions about evolution, adaptiveness, or the role of genes in
development are not only technical questions; they are also about the de-
gree to which we feel we ought to view aspects of our lives as necessary,
even good. When human customs and activities are at issue, the develop-
mental stability we must assume extends far beyond what is required to
produce recognizable morphology.6 It even extends beyond ‘‘behavior,’’
if by this we mean the relatively well defined patterns of movements ob-
served in many other animals. Social life is made up not of movements,
but of acts, which are largely defined by intent and meaning.7 Inten-
tions are construed, and meanings constructed, by agents interacting in a
social-physical context that itself influences the significance of the inter-
action. We need a view of development that takes more seriously both
Stasis, Development, and Heredity
the organism and its intimate interchanges with its surroundings (includ-
ing its social surroundings). How do the nested systems work? A given
child-rearing practice may have different meanings in different places or
at different times, and thus different implications for the child’s develop-
ment and its relations with others (not to mention its reproductive for-
tunes). The human mind, in all its subtlety and variety, is social from its
inception (a developmental psychology that is not social is no psychology
at all).
A Case of Meta-Recapitulation?
During the time I have been occupied with the nature-nurture opposi-
tion, I have become sensitized both to the various guises in which this di-
chotomy appears and to structurally similar ones in other fields (figure ).
One of the sources of the nature-nurture dichotomy in science is the field
of epistemology, in which a classic problem has been the origin of knowl-
edge. The disputes between rationalists, who insisted on innate ideas as
the source of knowledge, and empiricists, who credited the senses, did
much to set the framework for more recent disputes. Similar oppositions
are found in other fields as well.
Selected Dichotomies
Epistemology : innate vs. experiential sources of knowledge
History : internalist vs. externalist explanations of change
Anthropology : biology vs. culture
: culture vs. geography
Biology : history vs. ecology (ecology)
: physical necessity vs. chance (origin of life)
: mosaic vs. regulative development (embryology)
Parallels
Internal External
Of the many points that could be made here I wish to highlight just
four. (Three of these are sketched out in Oyama , and Gray gives
more extensive treatments in his a, , and .) In both debates,
internal causes are contrasted with external ones; in both, fixity is asso-
ciated with the former and malleability with the latter; and in both there
has been considerable oscillation between internalist and externalist per-
spectives. Finally, similar compromises have been proposed, as theorists
attempt to reconcile internal to external factors. These are considered in
turn.
pregiven ‘‘biological base,’’ and both are seen as externally directed. Se-
lectionist explanations are found in the literatures of operant condition-
ing, neurogenesis, the immune system, and cognitive and linguistic de-
velopment.4
If the internalist perspective defines developmental studies, the field
of evolution presents a rather different picture. Variational processes, in-
cluding genetic drift and natural selection, dominate the modern syn-
thetic theory. Populations are ‘‘molded’’ by these external forces. Ortho-
genesis, a transformational process, was resoundingly rejected by the
Modern Synthesis. The virtual disappearance of this internalist, ‘‘de-
velopmental’’ perspective from evolutionary theory is generally seen as
the triumph of scientific Darwinism over mystical Lamarckian pro-
gressivism.5 Effectively abolished from contemporary evolutionary dis-
course (but see Grehan and Ainsworth ), orthogenesis is probably
too heavily freighted a term to gain serious attention now. Its internalist
connotations persist, however, in the literature on developmental con-
straints, a literature that is gaining in volume and influence. A notable at-
tempt to integrate its diverse critical and empirical viewpoints into main-
stream neo-Darwinism is evident in the group report on a conference
convened for just that purpose (Maynard Smith et al. ). In addition
to presenting a typology of constraints, this paper repeatedly points out
the importance, and difficulty, of separating internal factors from external
ones.
Oscillation
Compromises
Not only does one find oscillation between poles in both the developmen-
tal and the evolutionary debates, but the same kinds of (largely unsat-
isfactory) compromises described in chapter are encountered as well.
Developmentalists have often ‘‘solved’’ the nature-nurture problem by
dividing the territory: by attributing some features to genetic control and
others to environmental influences. But if there are fundamental concep-
tual difficulties with the very notion of genetically or environmentally di-
rected development, it can scarcely be an improvement to combine them
in some odd organic patchwork.
Apparently more judicious is the attempt to quantify the relative
amounts of genetic and environmental influence on various features.
There have been many important critiques of these strategies of pheno-
type partitioning, as well as of the partitioning of variance associated
with such statistical techniques as the analysis of variance (); even
the seemingly more promising tack of tracing phenotypic ‘‘information’’
to the genes and the environment is problematic.8 One difficulty with the
partitioning of variance is that it frequently slides into the partitioning
Ontogeny and Phylogeny
behavioral studies. (See, e.g., Stearns ; Thomson ; for discus-
sion, see Gray .)
Sober has analyzed attempts to apportion causal responsibility to the
genes and the environment. He concludes that questions about relative
causal contributions to the phenotype are not locally answerable because
an organism cannot be affected by one and not the other. One can, how-
ever, ask how much difference one or the other made to a particular set
of outcomes (Sober ). There is a distinct possibility that the same
is true of attempts to treat constraints and natural selection as compet-
ing forces. Elsewhere, Sober () discusses disputes over the relative
strength of natural selection and correlated characters (that is, characters
whose evolutionary alterations are linked because they are developmen-
tally linked).When he asserts that the disagreement is not really about the
power of natural selection (which is the way it is often cast), but rather
about the ‘‘power of mutation,’’ he seems to be identifying the same con-
fusion between local and nonlocal analyses as exists in nature-nurture
arguments. What appear to be questions about relative causal contribu-
tions to a particular outcome are intelligible only if recast as questions
about a class of outcomes. He suggests that one could ask ‘‘how often a
trait is maintained by pleiotropy [multiple effects of a gene] even though
it is selected against, or is eliminated by pleiotropy even though there
was selection for it. This is a far cry from looking at a single trait whose
presence in a population is the joint product of selection and pleiotropy
and asking which contributed more to its evolution’’ (pp. –). Inter-
estingly enough, Stearns () asserts that selection and constraint are
involved in all evolution; the problem is to determine the relative influ-
ence of external and internal factors. Behavioral scientists justify their
continued pursuit of genetic and environmental ‘‘components’’ in pre-
cisely the same way.
Above, Sober (:) refers to the ‘‘power of mutation.’’ ‘‘Muta-
tion’’ can refer both to alterations in sequences and to the phenotypic
consequences of such alterations. Restrictions on the range of pheno-
typic results of genetic mutations surely involve not just constraints on
changes but constraints on the rest of the developmental systems in
which they occur as well. A particular change may have no effect
in some systems, and a variety of effects in others. The outcome will de-
pend both on the alteration and on the rest of the system. It might thus
Ontogeny and Phylogeny
Earlier in this chapter I pointed out that the nature-nurture and con-
straints-selection dichotomies are not only structurally similar, they are
related in substance as well. In fact, they are mutually reinforcing. Evolu-
tionary theory maintains developmental dualism by the logic described
in chapters and : Evolution is typically defined by change in gene fre-
quencies. Organisms must therefore be explicable in terms of genetic ‘‘in-
formation.’’ Genetic transmission is thus the needle’s eye through which
innate characters must pass. In Bonner’s () arresting image, the life
cycle must (often) narrow to a single cell,13 so inherited ‘‘nature’’ must
be passed on in the zygotic . Acquired characters, those formed by
environmental action, are excluded and thus rendered inconsequential to
evolutionary change. So evolutionary theory seems to require develop-
mental dualism. Developmental theory, in turn, insofar as it embraces the
transformational model of ontogeny (as predetermined, uniform, and in-
ternally generated), reinforces dualism in evolutionary studies by legiti-
mizing this vision of autonomous change.
Resolving the nature-nurture dichotomy involves, ironically, taking de-
velopment seriously. In a discussion of dichotomies in biology, Gray
(; see also P. Bateson ; Johnston ; and Lehrman , )
suggests that this resolution has been achieved in studies of behavioral
development, but I think his statement should be taken less as an accu-
rate report from the front than as a rhetorical device to prod evolutionists
into action. (What biologist, after all, would be indifferent to having psy-
chologists held up as models of intellectual sophistication?) I am less
sanguine. One of the many factors conspiring to maintain, and repeat-
edly reconstruct, the nature-nurture complex is psychologists’ increas-
ing attention to evolutionary biology. Although the distinction between
inherited and acquired traits that evolutionists want to make can be man-
aged without any specific assumptions about the nature of developmental
processes, it seldom is; in fact, it is typically treated as a statement about
developmental mechanism.
Ontogeny and Phylogeny
Synthesis
The somewhat peculiar title of this chapter is not just an allusion to items
of s popular culture. It refers as well to Steven Jay Gould’s (a)
book Wonderful Life, in which he discusses the fossils of the Burgess
Shale, a rich bed of paleontological remains in Canada. Many of these
fossils belong to extinct phyla. An extinct species or genus is hardly re-
markable, but the disappearance of groups as high in the evolutionary
hierarchy as the phylum may be, retrospectively, at least, more arresting:
Consider that the vertebrates, as varied as they are, are only a subdivision
of the phylum Chordata.
Gould enlists the bizarre Burgess creatures, which he calls ‘‘weird
wonders,’’ in the service of one of the dominant themes in both his popu-
lar and his scholarly writing: contingency in evolution. Like many other
scientists, Gould frequently argues against the widespread idea that the
course of evolution is somehow necessary: progressive, goal directed,
always moving from the less to the more complex, and culminating in
those marvels of reflective intelligence, ourselves. Again like many
others, however, Gould (e.g., :) has no difficulty attributing these
qualities of orderly goal direction to development; hence his use of the
common metaphor of the computer program to explain developmental
processes.1
The two words evolution and development have intertwined histories
that reflect changing understandings of what we would now call the evo-
lution of populations and the development of organisms. Much of this
book is about the way the relationship between these two formational
processes is conceptualized. I will not address the problems of mass ex-
tinction that are presented in Wonderful Life, but I will use Gould’s dis-
cussion of the Burgess Shale creatures as a pretext for examining the way
Looking at Development and Evolution
Gould’s argument in Wonderful Life is that the very existence of the phy-
lum Chordata, and therefore of ourselves, is an evolutionary accident.
He repeatedly uses the metaphor of the lottery and says that ‘‘this model
strongly promotes the role of contingency, viewed primarily as unpre-
dictability, in evolution’’ (a:). It might seem that his primary
concern is epistemological. This impression is strengthened when, in the
same passage, he challenges his readers to contemplate the elaborate de-
signs of these extinct creatures and to say what defect explains why they,
and not others, vanished from the earth. But simple prediction (or retro-
diction) is surely not Gould’s only point. He is also making an argument
about what kinds of causal processes are involved in evolution, ‘‘an un-
predictable sequence of antecedent states, where any major change in any
step of the sequence would have altered the final result. This final result is
therefore dependent, or contingent, upon everything that came before—
the unerasable and determining signature of history’’ (a:–).
The Accidental Chordate
Much could be said about this passage, but for now, we may note
that Gould is joining the epistemological issue of predictability with the
ontological one of the nature of evolutionary process, as chains of cer-
tain kinds of causal dependency. Gould certainly didn’t invent the asso-
ciation between these particular meanings of contingency (Cahn ;
Dray ).2 Nor is the association always pernicious. As I noted above,
however, the distinction between the predictability of processes and the
nature of those processes will be crucial to my discussion of develop-
mental dynamics.
Development involves the repeated arising and transformation of com-
plexes of interacting processes and entities. These occur not because
of preordained necessity, and not, obviously, by ‘‘mere chance,’’ if this
means an absence of regularity or causal relation. Rather, they come
about through systems of contingencies whose organization may itself
be contingent, and in any case, must be accounted for. Reliable, species-
typical life courses can be seen as contingent in a number of ways (e.g.,
not absolutely necessary, causally dependent on factors that may in turn
be uncertain). They may still be highly predictable, and thus noncontin-
gent, in another sense. Unlike evolution, development presents us with
repeating cycles. Despite many kinds of variation, these can be so simi-
lar across generations that they offer themselves for comparison with the
transmission of property by social institutions of inheritance. They thus
invite confusion between the epistemological and ontological aspects of
contingency. The reliability of these cycles tempts us to think that their
products are somehow delivered in a package, ready-made and whole.
There are many other intergenerational bridges besides the chromo-
somes, but the genes are often endowed with extraordinary causal pow-
ers, sometimes being credited with the capacity to generate the next cycle
virtually de novo. The sheer predictability of many aspects of develop-
ment has been explained by invoking special causal processes, directed
by molecular agencies that are immortal, omnipotent, omniscient (even
immaterial; see Dawkins ; and Williams ). This involves a sys-
tematic privileging of insides over outsides and active controlling agents
over passive materials: The environment is seen as supporting, modu-
lating, and constraining development while primary formative power is
reserved for the genes (Doyle ; Keller ; Oyama ).
An alternative is to explain such reliably repeated sequences of events
Looking at Development and Evolution
worm (N) which they proceed to eat. Unable to masticate it, they pull it
back and forth causing cradle to rock and put baby to sleep.’’
Figure is another example of a chain of contingencies. Natalie Angier
(:C), of the New York Times, says of recent discoveries about the ras
pathway in cell division (the ras gene, named for rat sarcoma, is believed
to be implicated in many cancers):
Molecular biologists are now gazing upon their glittering prize, a funda-
mental revelation into how the body grows. The sight is astonishing to
behold. It is epic.
It deserves a crash of cymbals, a roll of the tympanum—and a hearty
guffaw. It turns out the much-exalted signaling pathway of the cell is a
kind of molecular comedy, in which one protein hooks up to a second
protein that then jointly push a button on an enzyme that pushes a button
Looking at Development and Evolution
on another enzyme that makes this knob slide into that hole—all in all
like something Wile E. Coyote might have pieced together from one of
his Acme kits.
Looking at Ourselves
This page intentionally left blank
7 Essentialism, Women, and War:
In the s and s certain biological theorists and feminists con-
verged on ‘‘essentialist’’ accounts of war that are strangely similar. At
first glance it seems an unlikely development, given the frequency with
which feminists and antifeminists have lined up on opposite sides of the
nature-nurture rift. At second glance, though, perhaps the alliance is not
so surprising after all. We seem to be in the midst of a pendulum swing
‘‘back to nature.’’ This movement, in turn, is probably part of a more gen-
eral trend in the United States toward conservatism and a certain brand of
romanticism, although the issue is complex. Apart from the current em-
phasis on so-called traditional values, though, the convergence on ‘‘bio-
logical’’ views reflects some very pervasive beliefs that are as evident in
environmental determinist approaches as they are in biological ones.
By ‘‘essentialist,’’ I mean an assumption that human beings have an
underlying universal nature that is more fundamental than any variations
that may exist among us, and that is in some sense always present—per-
haps as a ‘‘propensity’’—even when it is not actually discernible. People
frequently define this preexisting nature in biological terms and believe
that it will tend to express itself even though it might be somewhat modi-
fied by learning, and thus partially obscured by a cultural veneer.1
When I say that social determinist and biological determinist ap-
proaches share many assumptions, I mean that even when they have dis-
agreed on the proper ‘‘balance’’ of cultural and biological explanation,
they have treated the general project of finding the correct proportions
of environmental and genetic influence as a valid one. They have also
tended to agree that the possibility of change was somehow illuminated
by their disputes.
Looking at Ourselves
The Argument
Lionel Tiger and Robin Fox (:–) say that war ‘‘is not a human
action but a male action; war is not a human problem but a male prob-
lem.’’ If nuclear weapons could be curbed for a year and women could
be put into ‘‘all the menial and mighty military posts in the world,’’ they
say, there would be no war. Tiger and Fox immediately concede that
this proposition is a totally unrealistic fantasy, because the human ‘‘bio-
grammar’’ (a term they use more or less the way others use ‘‘genetic pro-
gram’’) ensures that such a thing could never happen. Men, they say, have
evolved as hunters who band into groups and turn their aggressiveness
outward against common enemies or prey. Political structures in modern
societies are formed on this primeval hunting model, and men naturally
dominate these structures as well. Tiger speculates that women, who he
says do not bond and cooperate as effectively as men, could be given
positions in government by special mandate. He feels, however, that it
might be unwise to expect that even this effort could effect much change
(:–). Presumably, attempts to subvert natural tendencies are
not likely to succeed.
Barash (:–) accepts the idea that males bond and exclude
women from political power, although he emphasizes the grounding of
male aggression in competition for reproductive opportunities and repro-
ductively relevant resources (p. ). He argues that women are allowed
to hold political power only if they are in some sense ‘‘desexed’’ by age,
Essentialism, Women, and War
less aggressive than men, that war is caused by male aggression, and
that women are thus somehow more capable than men of bringing about
peace—or, at least, would be less destructive than men if they were in
power. It is a skeletal argument, of course, abstracted from very different
sorts of writings. The transition from individual to international conflict
in such works is not necessarily direct. For Tiger and Fox (), Barash
(), and Zanotti (), for example, war is specifically the aggres-
sion of male-bonded men in groups; and Tiger (:) distinguishes
between individual and group violence. Furthermore, the connection be-
tween aggression and peacemaking is not clear. Especially in the work
of the male scientists cited above, it is females’ lack of aggressiveness,
rather than any positive quality, that is emphasized. Even in a world in
which women have traditionally been defined by their deficits, I’m not
sure that peacemaking and peacekeeping should be seen as merely pas-
sive (to invoke another loaded dichotomy) results of low levels of aggres-
sion. Radical feminists are more likely to implicate the positive female
qualities of nurturance, sensitivity to connections, and peacefulness.3
The arguments about the sexes and aggression discussed above all re-
quire that aggression be unitary. They depend on, and encourage, certain
kinds of illegitimate ‘‘lumping.’’ One sort of lumping is definitional: All
sorts of behavior, feelings, intentions, and effects of actions are grouped
together as aggressive. Tiger’s (:) definition, for example, is so
broad that it embraces all ‘‘effective action which is part of a process of
mastery of the environment.’’ Violence is one outcome of such aggres-
sive activity, but not the only one. That women do not bond in aggressive
groups implies, then, that they are less capable of ‘‘effective action’’ in the
service of ‘‘mastery of the environment’’—a sweeping statement indeed,
and one that bodes ill for any political action on the part of women.
Another habit that allows us to treat aggression as a uniform quan-
tity is cross-species lumping, equating very different phenomena, so that
mounting and fighting in rodents, for example, territoriality in fish or
birds, and hunting, murder, and political competition in humans are all
‘‘aggressive’’ behaviors. Then there is the lumping of levels of analysis,
Looking at Ourselves
It should be clear by now how I feel about several common strategies for
dealing with biological arguments.When someone says, ‘‘It’s biological,’’
we sometimes reply, ‘‘No, it’s not, it’s cultural,’’ when instead we should
be asking why the cultural and the biological are treated as alternatives
in the first place, and just what we (and they) really mean by either expla-
nation: Not all environmental influences are cultural ones, for instance. I
call this the Protesting Too Much Syndrome, because we are often afraid
that the trait in question is ‘‘biological’’ in one or more of the mistaken
senses described above. Similarly, when someone says that women are
innately inferior, we counter, ‘‘No, we’re not, you are,’’ rather than re-
jecting the assumption of essential nature that allows pronouncements of
this sort. This second strategy, the Protesting Too Little Syndrome, en-
tails agreeing that differences are biological, but reversing the evaluative
polarity. Male nature is bad; female nature is good. While it offers the
momentary satisfaction of turning the tables, it is based on all the ideas
about nature that get us into so much difficulty. This is too great a price
to pay for Mother Nature’s favor. The solution is not to protest precisely
the correct amount, or to find the degree of biological determination that
is just right, like Goldilocks trying to find comfort in a house that is not
her own. Rather, it is to protest a whole lot about the very rules of the
discourse.
I am not saying that we ought to throw out everything and start from
the ground up. We couldn’t do it even if we wanted to. But when there
are ample grounds for doubting the validity of a conceptual framework
or a set of issues, as is the case with the nature-nurture complex, we do
Essentialism, Women, and War
I could say more about aggression and about women, and maybe even
a little about war. Perhaps the reason I haven’t done so up to this point
is that the essentialist theories I have been discussing don’t say much
about these topics either. Instead, I have focused on the ways we think
and talk about these topics. War is about politics, diplomacy, economics.
It is about historical continuity and change in relations among people,
not about brain centers, testosterone levels, or rough-and-tumble play.
War is like a fight between individuals only by analogy, just as certain
encounters between groups of ants are wars only by analogy. Perhaps it
is significant that the book sociobiologist Barash coauthored on prevent-
ing nuclear war (Barash and Lipton ) contains nothing about differ-
ing capacities and contributions of males and females, but instead lists
very pragmatic suggestions for effective action. I would never claim that
women have no role in national and international politics, but neither can
I make sense of the notion that we ought to be somehow inserted into
public life because of some mythic direct line to life, peace, and love.
Men are not a plague, and women are not a cure.
8 The Conceptualization of Nature:
Nature as Design
The controversies I find most interesting are the ones that are notable
for the questions they raise rather than the solutions they produce. One
can argue about whether computers have minds or not, for instance, but
the more intriguing question is, What does it mean to have a mind?—
and indeed, Is a mind something one can ‘‘have’’? Or one can wonder
whether or not apes have language, but the real challenge is to say pre-
cisely what language is. Thus far we have often been concerned with the
endless debate over whether or not some trait is ‘‘in’’ the genes: We have
considered just what the phrase (or its opposite) could possibly mean,
and what it shows about the way we think about the nature of organisms
and their development. Examination of these questions reveals some un-
spoken assumptions. To begin with, they have spatial implications. Minds
and language are thought to be things we either possess or don’t; they
have a location, typically in our brains. ‘‘Genetic’’ traits (or programs for
them) are in the . Behind these assumptions is the image of isolated
individuals whose properties (possessions!) can be enumerated without
paying much attention to their activities or surroundings.
To use biology as a basis for design—to ask it to guide our activities, in
devising a more reasonable agriculture, say, or ecologically safer meth-
ods for processing waste, is to turn outward, not inward, but a similar
situation arises. On the one hand we can wonder just which designs are
biologically natural and which are not. Or we can ask whether nature has
a design ‘‘there’’ for us to imitate. To turn Gertrude Stein’s dismissive
comment about Oakland, California, into a query, is there a there there?
Does Nature have a nature, an essence, independent of us but knowable
by us?
The Conceptualization of Nature
Small-n-nature, Big-N-Nature
My reflections in this chapter will be a departure from the rest of the book
in two ways. First, my investigations of the relations between the internal
and external domains have concentrated on the concept of internal bio-
logical nature. I once called this ‘‘small-n-nature,’’ as in ‘‘human nature.’’
Here, though, I turn my attention to Nature ‘‘out there’’: Nature with a
capital N (these terms are from Oyama ). Second, instead of indicat-
ing the dangers of multiple meanings, as I did in chapter , I will celebrate
ambiguity. My emphasis on constructive interaction remains. This time,
however, it is applied not to the ontogeny of organismic ‘‘designs,’’ but
to the designs we see in the external world.
By insisting on construction, I don’t mean that we simply project our
internal designs onto a passive, chaotic Nature. This would be to ac-
cept the objectivist-subjectivist split criticized by Humberto Maturana
and Francisco Varela (), by certain feminist theorists (Harding and
O’Barr , for instance), by Lakoff and Johnson (), and by others
who are dissatisfied with what they consider to be seventeenth-century
notions of objectivity. Rather, external Nature, like internal nature, is
co-constructed over time, through intimate engagement with the world.
Nature is multiple but not arbitrary, and design is brought forth with the
world in that fusion of knowledge and action also described by Maturana
and Varela (). Just as I have denied the validity of traditional notions
of immutable biological nature inside us, written in our genes at the mo-
ment of our conception and existing independently of our developmental
interactions, I now question the idea that big-N-Nature has a unitary, eter-
nal nature that is independent of our lives in the world. In fact, the kinds
of interactions in which we participate influence the Nature we design.
(The ambiguity in the preceding sentence is intentional: I mean to refer
both to our conceptualizations of Nature and to the kinds of concrete
changes we bring about in the world.)
I am going to make some comments that may seem subversive to the
notion of ‘‘biology as a basis for design,’’ but ultimately I don’t think
they are. Along the way, I will discuss concepts of design and nature and
the virtues of multiple perspectives in elaborating these concepts. I will
touch on biology as a way of knowing, the relationship between knowers
Looking at Ourselves
and knowledge, and the necessity of taking responsibility for our contri-
butions to the knowledge (and therefore the Nature) we construct. This
may seem rather a lot, but I wish only to sketch a set of issues, not to
provide an exhaustive treatment, so my comments on each topic will be
brief.
about knowledge and activity that are now being strongly challenged
from many directions.
Two incidents prompted my own thinking about these issues. One was a
comment by Mary Catherine Bateson about the much-used metaphor of
the earth as mother. She offered some alternatives—the earth and our-
selves as co-parents, for example, or the earth as child—and made the
gentle suggestion that we needn’t insist on only one. There might be some
virtue, that is, in a kind of tentative flexibility in our thinking about this
relationship—or rather, these relationships.
The second prompt was a story in the New Yorker by Ursula Le Guin
(). Entitled ‘‘Half Past Four,’’ it is not so much a story told from sev-
eral points of view as it is a story that is itself progressively transformed.
In a series of scenes, people and events shift. Ann is first presented in
an encounter with her father, Stephen, and with Toddie, the retarded son
of her father’s new wife, Ella. In the next scene Ann is Todd’s sister,
and the two siblings discuss their absent father, who has started a family
with another woman. Then Ann is Ella’s daughter, meeting Ella’s suitor,
Stephen; and so on. Identities, histories, and even sexual orientation are
permuted and transmuted while the many-dimensioned space is bound
together by recurrent themes and images.
Reading the story for the first time, I found myself backtracking impa-
tiently in the careless reader’s attempt to recover what an initial scan had
missed. I fussed: Are Toddie and Ann siblings or half-siblings? Who is
this Ella, anyway? It was only after several of these truth-gathering for-
ays that I realized that getting the facts straight once and for all was not
the point. Rather, the concatenated scenes could be read for their richly
proliferating relations, including the relations among those relations.
These two experiences led me to reflect on the value of multiple per-
spectives—not as several lines of sight converging on the same object, in
which the goal is a single, more accurate view, but rather as paths to rich-
ness, to curiosity, to a sense of possibilities and myriad connections, di-
vergences, and discontinuities. Anthropologists Gregory and Mary Cath-
erine Bateson speak of the diversity of reference, of point of view, in
ethnographic knowing (:).2 (Perhaps it is only coincidental that
Looking at Ourselves
Can we envision another kind of biology, one less tied to a search for
the one timeless truth that will structure our lives, a biology that recog-
nizes our own part in our constructions of internal and external natures,
and appreciates particular perspectives as vehicles for empathy, investi-
gation, and change? If a metaphor is something through which we can
think (G. Bateson and Bateson :chap. ), and perhaps through which
we can create/discover ourselves, can we find different metaphors for our
world, and thus construct different ways of being in it?
We must take responsibility for the Nature (and the biology) we con-
struct.We do not, however, manufacture either our own natures or Nature
out there as detached, Godlike subjects. Our responsibility is not the
responsibility of unmoved movers, absolute originators bringing order
to chaos. Rather, the construction is mutual; it occurs through intimate
interaction. By the same token, we do not simply record facts about ex-
ternal Nature, any more than we are simply manifestations of an internal
nature encoded in some genetic text. ‘‘Information,’’ that is, is not in-
dependent of us, and because this is so, we cannot disclaim a kind of
ownership. Our cognitive and ethical responsibilities are based on our
response-ability, our capacity to know and to do, our active involvement
in knowledge and reflection.
This is a productive ambiguity of subjects and objects, of multiple per-
spectives, of ourselves as nature’s designers and as nature’s designs, as
designers of our designs of (and on) nature, of our own natures as prod-
ucts of our lives in nature. If Nature is our technical or moral teacher,
it is not as a radically separate, independent source of information-stuff,
but rather as a source of differences, a world we interrogate with particu-
lar questions and particular concerns in mind, and whose responses we
interpret in the light of our own history with it.
I respect and value much of the work on alternative biologies, but I also
think we ought to be cautious about claiming one true account of Nature’s
nature (although I have in the past come close to doing this myself ). I am
suspicious of quests for single truths; there is a strong tendency to take
what is pre-scribed (‘‘already written,’’ whether in the genes or in the
language of the earth and streams, and therefore in no way dependent on
Looking at Ourselves
us) as prescription, that is, as a received formula for action. We should re-
sist this temptation to seek ‘‘the natural’’ as an authoritative guide to our
lives. It is, after all, the temptation to deny our presence in every truth we
see. Our presence in our own knowledge, however, is not contamination,
as some may fear, but the very condition for the generation of that knowl-
edge. Biology is one activity among many, and design is the pattern that
emerges with that activity. Finally, this includes even our conceptions of
Nature’s own design.
Donna Haraway writes of science as fetish: ‘‘an object human beings
make only to forget their role in creating it, no longer responsive to
the dialectical interplay of human beings with the surrounding world’’
(:). Elsewhere, she calls on us to forsake our search for a total-
izing unity, to give up our ‘‘dreams for a perfectly true language, of per-
fectly faithful naming of experience’’ (b:). In place of the quest
for a lost innocence, an original wholeness to make our lives intelligible
and to ground our politics, she advocates a politics of coalition. Such a
coalition would be based not on a common language, but on a ‘‘powerful
infidel heteroglossia’’ (b:).
Much of what I have just said could be seen as a denial of the desir-
ability, or even the possibility, of taking biology as a basis for design. In
fact, though, once our terms are reworked in a way that fits the enterprise,
just the opposite is true. It is ‘‘biology’’ as timeless truth that I reject,
not biology as a human activity entered into with responsible awareness.
The first defines reality once and for all; the second is closely connected
to the persons and life circumstances of scientists and does not need to
claim transcendent truth. We can’t help revealing our notions of biologi-
cal design in our other creations, just as we can’t help showing ourselves
in our practice of biology. Probably the most common and intuitively ap-
pealing way of conceptualizing natural design, in fact, is by analogy with
our own activity. To acknowledge our part in constructions of Nature is
to accept interaction as the generator of ourselves and of our interrela-
tions, of knowledge, and of the world we know. Both we and the world
are expressed in this dance, even as we are created and know ourselves
in it.
If any of this makes sense, then the perspectives of the scientists who
are exploring alternative approaches are extraordinarily precious. Their
work shows us selves both responsive and response-able (and so, respon-
The Conceptualization of Nature
Certainly the sociobiologists’ assurances are meant to show that the fuss
of the s has been misguided. Bad old words such as instinct and in-
nate have largely been replaced by substitute phrases that are supposed
to be innocent of unwanted implications. Thus we hear of evolutionarily
derived or canalized behavior, adaptive central tendencies, evolved mo-
tives, and facultative adaptations (ones that take different forms under dif-
ferent conditions, thus demonstrating that adaptiveness does not always
entail developmental or behavioral rigidity). Sociobiology seems to have
Looking at Ourselves
The phrase ‘‘the body in the mind’’ comes from the title of Mark John-
son’s book. George Lakoff and Johnson () have investigated
the ‘‘objectivist’’ view of mind as disembodied and abstract, arguing for
Bodies and Minds
an alternative that does not rest on a separation of mind from body (see
chapter ). I am sympathetic with this depiction of embodied thought,
and these critiques of the opposition between subjectivism and objec-
tivism show many affinities with my own analyses. I borrowed the title
for a different reason, however: to highlight the association of innateness
with the body.
Johnson and Lakoff wish, among other things, to heal the mind-body
rift that has been so important in the Western philosophical tradition;
Johnson’s ‘‘body in the mind’’ is thus intended to signal an integration. I
use the phrase, however, to indicate a failure of integration, a continued
reliance on a set of ancient disunities, in which nature is associated with
the body, and nurture with the mind. Calling some aspect of the mind
‘‘biological’’ thus invests it with ‘‘bodiness.’’ This special status may de-
pend on any of the meanings of innateness. If a trait has counterparts in
other species, for instance, or appears universal in humans, or follows
family lines in some population (shows heritable variation), it may be
dubbed ‘‘genetic’’ or ‘‘biological.’’ Other definitions are based on appar-
ent survival or reproductive utility, the presence of identifiable morpho-
logical or biochemical correlates, apparent spontaneity or irrationality,
presence in infants, or absence of obvious learning. By attributing part
of the mind to nature, then, people place the body in the mind. Because it
is a notion of body that is rooted in a dualistic framework, however, the
result is not integration but only a more complicated dualism.
Scientists have not been completely oblivious to the weaknesses of
the nature-nurture opposition; recall the disclaimers mentioned above.
But disclaimers and adjusted terminology are not enough. Concepts can
survive considerable turnover in lexicon. E. O. Wilson () defines
sociobiology as the study of the biological bases of all social behavior.
Innumerable nonsociobiologists (physiologists, geneticists, anatomists,
etc.) also consider such ‘‘bases’’ central to their work, so the continu-
ing muddle about what a biological basis is can hardly be termed minor.
(Add those who devote time to showing that something or other is not
biological, and one has included much of the scholarly community.) 1
All behavior requires a body (and therefore involves genes and envi-
ronment, physiological correlates, brain structures, etc.), so nothing is
served by partitioning the mind into biological and nonbiological pieces.
Yet it is common for arguments about innate behavior to be buttressed by
Looking at Ourselves
After all, we learn and think by virtue of our bodies (the claims of some
artificial intelligence theorists notwithstanding). The problem of the dis-
embodied mind is not solved by embodying just the ‘‘innate’’ portions,
for this would imply that other portions must be accounted for in some
other way. However we gerrymander the outlines of evolutionary ‘‘na-
ture,’’ there will always be some residual chunk of acquired ‘‘nurture’’
hovering mysteriously in thin air, without a ‘‘biological base’’ to support
it. But if natures are the result of the continuous nurture of developmen-
tal construction, we ought to object when we are told that nature interacts
with nurture (or biology with culture, etc.), just as we do when we read
about material body-stuff interacting with immaterial mind-stuff.
ary research to insist that the behavior they describe is not inevitable or
immutable.4 It is important to keep challenging that reflex assumption.
Full assimilation of the lesson is hampered, though, by the associations
described above, and by the fact that milder versions of the fixity argu-
ment perpetuate its spirit while softening its outlines. If the innate is not
immutable, some authors tell us, it is at least hard to avoid or change.
M. S. Smith (:, ), for instance, implies that ‘‘strong cultural
pressures’’ are needed to overcome ‘‘evolved dispositions,’’ because evo-
lutionary theory tells us that ‘‘some aspects of human behavior are easier
to change than others.’’ Apparently relying on the opposition between
biological emotion and (nonbiological?) reason, Gruter (:) warns
that laws prohibiting acts that ‘‘fulfill survival functions’’ are unlikely to
be effective because strong emotion is impervious to reason. According
to Alexander (:), it will be hard to persuade men to use contra-
ception because it reduces their reproductive opportunities. It will also be
hard, he thinks (Alexander :), to achieve altruism within groups
without intergroup hostility because competition among groups has been
crucial in evolution.
As we have seen, there are other variants of the idea that the ‘‘biologi-
cal’’ is hard to change, including the notion that biology sets limits on
potential. The point is that they are variations on a theme, and are as un-
tenable as claims of strict immutability. Notice that I am not making a
case for ‘‘infinite potential,’’ whatever that might mean. Some behavior
may be impossible to change. But there is often no simple way of find-
ing out how hard something will be to influence, short of trying. We are
sometimes surprised by the outcome of our efforts. Responsiveness may
well depend on the larger context; even patterns that were stable over
evolutionary time may not be so in new environments, and vice versa.
Many theorists, in fact, have emphasized the important role of behav-
ioral innovation in bringing about evolutionary change (P. Bateson a;
Gray ).
A desire for simplicity and certainty is understandable, particularly
with respect to socially important issues such as violence, competition,
and hierarchy, or family and gender relations. It is risky, however, to con-
clude that some aspect of contemporary life is inevitable on the basis of
a phylogenetic or a functional argument. It is seldom certain that a statis-
tical phenomenon like the allegedly greater frequency of child abuse by
Bodies and Minds
fers from the Thornhills’ in emphasizing male hostility, they state: ‘‘[We]
assume the existence of a polygenic substrate that generates a closed be-
havioral program linking hostility and female vulnerability with forced
copulation. We perceive this connection as ‘‘hard wired’’ and the ge-
netic substrate encoding it as essentially fixed in the human population.
From this perspective, we view rape as a human universal and on this
level, probably recalcitrant to reprogramming via individual experience’’
(Shields and Shields :).
The model also includes an ‘‘open’’ program for ‘‘assessing female
quality and vulnerability’’ (p. ). The authors state that their model of
an innate threshold for rape implies that changes in men’s attitudes or
women’s status would do little to reduce rape. Although the authors dis-
approve of sexism, they say there is little chance of changing the rapists
themselves: ‘‘If the evolutionary hypothesis of a genetically fixed sub-
strate strongly predisposing men to rape under the appropriate proximate
conditions were true, then rehabilitation via education might be difficult,
if not impossible.’’ Instead, they advocate sure and severe punishment,
which would not only deter rape, but if ‘‘sufficiently severe . . . could act
as artificial directional selection against lower thresholds’’ (Shields and
Shields :–).
There seems to be a confusion here between (a) fixation of a gene or
genes in a population and (b) rigid developmental fixity (unresponsive-
ness to variation in developmental environment). These two are quite
distinct, of course. What links them is not empirical or logical necessity,
but rather the very system of beliefs about intractable genetic nature that
we are examining here. Concerns about fixity are not confined to evolu-
tionary thinking. The point is that the assumption of static nature is very
pervasive, and is not fully eliminated by invoking the environmental sen-
sitivity of facultative adaptations. The nature-nurture dualism threatens
to prejudge empirical questions that should be investigated on their own.
Biology is often thought to define the timeless truth that underlies shifting
appearances. Consider the spatial metaphors in the preceding sentence:
Biological reality lies below psychological appearances; it is the founda-
Bodies and Minds
tion for everything else. In this vision of a layered reality, the biological
is given special status. Related to this is the postulation of real, evolu-
tionary goals underlying everyday motives. The conceptual play here is
between evolutionary consequences and human intentions; by sleight-of-
thought, a history of natural selection is transformed into a contemporary
psychological truth. Although we frequently invoke the distinction be-
tween ‘‘ultimate causes’’ (conditions prevailing over evolutionary time)
and immediate ‘‘proximate causes,’’ we disagree on just how to keep
them apart. One problem is that the distinction tends to collapse: Ulti-
mate causes are transmuted into proximate ones, often in the guise of
genetic programs. Thus, in some human sociobiology, the evolutionary
past is present in the innate selfishness that is supposed to underlie all
behavior.
It is often said that knowledge of mechanisms (i.e., proximate causes,
including psychological processes) is unnecessary to evolutionary argu-
ments—that organisms need not ‘‘consciously’’ try to maximize fitness,
but only act as though they were doing so. This misleads even as it in-
structs. Organisms may do things that have, to some degree, been asso-
ciated with fitness over evolutionary time, but fitness maximization itself
need not be the proximate goal. The organism may well be tracking (ad-
justing its behavior to) some other variable. To say that organisms are not
maximizing fitness consciously, furthermore, is to invoke an unconscious,
which is conventionally seen as the proximate cause that really explains
behavior. For instance, Hrdy and Hausfater () say that human par-
ents calculate, consciously or not, the costs and benefits of infanticide.
(See also Shields and Shields ; and Thornhill and Thornhill , on
males’ calculation of costs and benefits of rape.) This idea of unconscious
reckoning fits very well with the idea of evolutionary causes ‘‘under-
lying’’ more superficial phenomena, perhaps in the form of genes ma-
nipulating organisms (Dawkins ), as well as with the Freudian view-
point about hidden psychic forces. And so we find declarations about
what people are really trying to do, regardless of what they think or say;
for example, M. S. Smith (:) tells us that, whatever they may
think, parents’ ‘‘real goal’’ is fitness maximization.
Pointing to the unconscious implies that humans are self-deluding
rather than hypocritical. Alexander’s characterization of the conscience,
however, at least suggests the latter. It is the ‘‘still small voice that tells
Looking at Ourselves
Evolutionary Tough-mindedness
People have tended to associate biology with the body, the passions, the
irrational. This venerable network of meanings has helped generate and
maintain notions of innateness as an animal (often beastly) nature within
us humans (Klama ), at war with reason and capable of overwhelm-
ing it. This opposition between reason and instinctive passion is evident
in some of the works cited above. With the growth of sociobiological
theory, however, a fascinating, if subtle, change has occurred: Long con-
sidered our best defense against the beast within us, rational deliberation
is being appropriated to do that animal nature’s work. As we have seen,
certain theorists suppose us to be forever doing the bidding of our self-
ish, fitness-hungry genes. Their machinations conveniently unconscious,
these calculating prodigies are said to produce, and explain, our behav-
ior. Regardless of what we think we are doing, we are really serving our
genes’ interests. It seems that the biological beast has commandeered the
very rationality that used to keep it in check.
Bodies and Minds
Descartes’ Ghosts
Scientists sometimes discover the Cartesian subject not only in the hu-
man head, but also in the cells that make up that head (as well as the rest
of the body). Why should there be such striking parallels between the
problem of the subject in developmental psychology and the problem of
conceptualizing biological, especially genetic, processes? In both cases,
a single source of organizing (information-processing) power seems nec-
How Shall I Name Thee?
essary, so that the cognitivist subject and the homunculoid gene play
similar roles in the dramas of cognitive and ontogenetic formation.4 The
seventeenth-century scheme of pure individual knowledge so important
to many present-day views of mentality and the split between immaterial
mind and material body inhabit contemporary constructions of ourselves.
When psychologists drive the homunculus-in-the-head from one part of
their theories, it tends to crop up in others (Henriques et al. :;
Shotter :). Similarly, developmental biologists take pride in hav-
ing eliminated the curled-up preformed manikin from the germinal cell,
yet have no problem believing in microscopic intelligences in the nucleus
that plan and control the creation of the organism. The much-criticized
presocial, rational subject not only resembles the macromolecular sub-
ject, it is thought to be the product of the activity of that subject. Thus
we are haunted by Descartes’ ghost, not only when we directly theorize
humans, but also, as we have seen, when we describe their development.
If matter is inert, then organized processes can be explained only by ref-
erence to a structuring intelligence. But if it is interactive (think of any
chemical reaction) under changing, interdependent constraints, such out-
side direction is not needed.
ones, they need to be exorcised—not only from our heads, but from their
other places of refuge as well.
ments cannot be definitive—not only are ‘‘all the data’’ never in on any
particular question, but the human significance of the data is not at all
clear.
Just as the child’s contribution tends to get lost even in many ‘‘inter-
actionist’’ (traditional, developmentally dualistic interactionisms, not the
constructivist, systemic interactionism espoused here) descriptions of
development, so does the organism tend to disappear from tales of gene-
environment interaction. Recall also Shweder’s point about disciplinary
competition eliminating the person from ethnography. It is perhaps not
so startling that what begins as a progressive impulse to rescue people
from biological determinism often ends up subjecting them to an alterna-
tive determinism, leaving them as passive as ever (Henriques et al. ;
N. Rose ).
is a fixed quality inside people, and that there are only two such qualities:
heterosexual and homosexual. A move in a particular game of defining
beginnings in order to construct an attitude toward the present, it does
not question the coherence of the game itself. Finally, it would seem to
make one’s claim to be a person subject to empirical verification—and, of
course, to potential disconfirmation—so that one’s very selfhood would
seem perpetually poised on the moving edge of ‘‘the body of data.’’ John
Money declares (in Gelman et al. :), ‘‘Of course it [homosexu-
ality] is in the brain. . . . The real question is, when did it get there?’’ Was
it there from the beginning or put there by some contingency of experi-
ence? Shotter, on the other hand, advises, ‘‘Ask not what goes on ‘inside’
people, but what people go on inside of’’ (:).
One need not choose between speaking of insides and outsides, some-
thing that Shotter (:) implies later in the same passage (although
he sometimes uses the ‘‘outsides are more important than insides’’ style
of argument as well). If insides are interdefined with outsides, we must
ask what kinds of exchanges result in personhood. The invocation of
essential nature, then, is a complex matter whose psychological, politi-
cal, and cultural meanings must be worked out in the uncertain arenas of
everyday life.
The legitimacy of one’s self would seem to be more the point than
the precise etiology given for it: Notions of biological feminine nature
exist quite comfortably with discrimination against women. In fact, eti-
ology becomes an issue when legitimacy is in question. (There seems
no need to hinge heterosexual rights on a developmental narrative.) Pre-
cisely because such outcomes are unforeseeable, multiple, and conflict-
ing, I would not presume to advise against the moves in question. Indeed,
if a broader swing back to biological explanations is in progress, the dis-
empowered might be unable to halt it if they tried. More discussion might
even encourage scrutiny of the whole issue, which could be a good thing.
My comments, then, are meant to express my anxiety about these mat-
ters and my preference for questioning prevailing beliefs about biology,
nature, and development rather than trying to turn a dangerously loaded
dichotomy to homosexuals’ advantage.
Looking at Ourselves
in which it and the organism select and construct each other (Giddens
:; Morris ; Oyama ; Venn ).We are ecologically em-
bedded, social beings, although we are not always well served by our
situations. It is entirely natural that our constructions of ourselves and of
the world should be marked by our particular historical, cultural, social
situations (although they will not necessarily ‘‘mirror’’ them), as well as
whatever experiences may be common to those situations.
Shotter’s (:) descriptions of joint action, in which, without
being aware of it, people alter their own and others’ consciousnesses as
they construct meanings together, may help us understand some of these
developmental phenomena. In such joint action, whether antagonistic or
cooperative, outcomes are often unpredictable, cannot be traced to the
intentions of any single participant, and thus may seem to be external
to them (p. ). They can appear ‘‘natural’’ or ‘‘spontaneous,’’ and it
is easy to miss one’s own (or others’) contribution to them. This is a
view of development as formation in a system composed of the person
and the developmental environment, changing together. It is a view that
does not permit nature-nurture distinctions, and that is what the natural-
constructed distinction is.
The vision of mutually constraining influences that I present here does
not, it should be evident by now, necessarily imply harmonious relations;
indeed, I wish to avoid the implications of automatic stability, healthy
function, or homeostatic regulation that are sometimes attached to the
concept of system. One might find a vicious circle of interactions in a
family, for example, that benefits no one very much but that is stable in
the face of a variety of attempts to change it, thus fitting the usual idea of
self-regulating systems. Similarly, ecological embeddedness does not en-
tail benefit to all the organisms involved. Think of the costs of pregnancy
to a mother, or the tight integration of the ‘‘traditional family’’ into exist-
ing political and economic structures (or the prey into any predator-prey
system!).
but in which causes and effects are not ultimately distinguishable; and in
which organization need not be imposed on inert matter, but rather arises
from matter in interaction (though not all organization is living organi-
zation: consider a crystal or a whirlpool), it may be possible to conceive
responsibility in a different way. Rather than contrasting autonomously
acting persons with passive objects, perhaps we can consider these to
be two stances toward certain human interactions. One is oriented more
toward considerations and consequences as seen from the agent’s point(s)
of view and occurring in a social context in which that agent is able to
communicate acceptable, or at least intelligible, reasons for acting (re-
sponsible action, according to Shotter :, being a matter of shared
interests, not of individuals). The other takes the point of view of some
(third-person) observer. Moral agency does not require freedom from
causes (what could this mean?) or even from biological causes. Rather, it
requires, precisely, embeddedness in a causal world (see Dennett ).
Only there can one be subject to the joys, pains, desires, and perplexities
that give rise to action; only there can one affect the world; only there can
one be engaged by the exchanges that constitute human life; only there
can one be moved to encourage some outcomes and prevent others; and
only there can one be positioned among others who regard one as respon-
sible. Such positionings are not foregone, however. The earlier discus-
sion of homosexuality shows how some people are attempting a strategic
repositioning while others oppose it. In Shotter’s ‘‘political economy of
selfhood’’ (:, ), people enhance and limit each other’s oppor-
tunities for development.
Immersion in the causal network, then, is not only consistent with
moral action, it is the very condition for such action. An actor will fre-
quently construe relevant antecedents and probable consequences rather
differently from any given onlooker (including a psychologist). An im-
portant form of moral disputation, in fact, concerns the proper place-
ment of action (its meaning) in life’s colliding activities and understand-
ings, influences and resultants. Blame and credit are also redistributed as
people’s understandings of their own actions change. It is possible, by
emphasizing some connections over others, to allocate responsibility in a
variety of ways, in what Kenneth Gergen (:) calls the exercise of
‘‘practical rhetoric.’’ Although rhetoric tends to have a degraded meaning
today, persuasion and influence are hardly trivial matters. In disagreeing
Looking at Ourselves
persons that comes ready-made. Rather than restricting the proper scope
of biology, as many critics have wished to do, I broaden it to encom-
pass the entire life cycle. Biology so construed cannot be used to par-
cel out developmental (or moral) responsibility to internal and external
factors. Nor can it define an invariant core of human nature. Difficult
issues of human capabilities—how to define the ‘‘could’’ in ‘‘could have
done otherwise,’’ for example—are not thereby resolved. I suspect that
such issues, like accountability itself, are matters of complicated social
negotiation, in which scientific understandings of possibility may play a
part, but which are not definitively answerable by means of those under-
standings.
The developmental systems perspective is no less constructed than any
other. Indeed, this book is part of its ongoing construction. This does not
mean that it is arbitrary or indefensible, nor does it prevent me from be-
lieving passionately in it, or from hoping that if I explain adequately what
considerations (scientific and otherwise) led me and others to formulate
it, readers will come to see things as we do.
’
Turing, who was instrumental in breaking the Germans’ secret codes dur-
ing World War II, was a homosexual who underwent a course of chemi-
cal therapy after the war as an alternative to incarceration for ‘‘gross in-
decency.’’ 9 He died not long afterward, apparently by suicide (Hodges
). The Turing test owes its name to this man. As it is understood
today, the test states that any entity that can pass for human when com-
municating by means that conceal its physical makeup (perhaps by words
displayed on a computer screen) is considered to have a mind, although
this is not precisely the way Turing presented it. Turing () himself
called it the ‘‘imitation game,’’ and its first players included a man, a
woman, and an interrogator whose task it was to distinguish between
them.
Turing’s biographer comments, ‘‘Like any homosexual man, he was
living an imitation game, not in the sense of conscious play-acting, but
by being accepted as a person that he was not’’ (Hodges :). The
author of the original imitation game did not himself play it very well,
partly because he refused to accept the goal of being mistaken for a
How Shall I Name Thee?
I am aware that the words politics and boundary in my chapter title will
raise some red flags. Part of my project is to inquire into the provocative
power of such boundary-marking emblems. They tend to imply a particu-
Evolutionary and Developmental Formation
lar field of oppositions, and they inevitably gloss over some differences
and commonalities while proclaiming others, making it hard to move be-
yond global categorizations. This situation helps explain why there may
be a need for something like a developmental systems perspective, to
rework many of the basic oppositions in the biological and social sci-
ences. It may also help us understand why developmental systems theory
seems to be resisted for being simultaneously inside and outside existing
theory: inside (and therefore redundant) because it stresses matters that
are already acknowledged at least part of the time, and outside because it
explicitly breaches certain sacred boundaries. A not uncommon reaction
to is, ‘‘That’s completely crazy, and besides, I already knew it.’’
Parts of this chapter will serve as a review of some of the basic argu-
ments of the book. I start with the developmental systems case for broad-
ening the concepts of inheritance and evolution, which means including
within the hereditary package many factors that are traditionally consid-
ered to lie outside it. I also review some common defenses for the tradi-
tional positions. Next come some recent border disputes in evolutionary
theory. In mentioning these disagreements I try to show that ‘‘science
as usual’’ frequently involves disputes over what should be treated as
internal and external to a given phenomenon. While the notion of a de-
velopmental system admittedly tampers with some categories that mod-
ern thinkers treat as sacrosanct, then, I’m suggesting that the kind of con-
ceptual change it proposes is not different in kind from the everyday work
of theorists, and that, indeed, boundary shifting is a favorite, although
sometimes unacknowledged, technique in scientific dispute. This does
not make it right, but it may make some of my own moves look less out-
landish. Although is perhaps better known for questioning distinc-
tions than for drawing them, repositioning a boundary can not only group
together things that were formerly kept strictly apart, like genetic and
other influences on development, it can also highlight new differences
and bring the peripheral into focus. Trends in the direction of something
like developmental systems are visible even in some standard theorizing.
At various points I touch on the selfish replicators that have figured
in debates about ‘‘biological bases’’ of human behavior, and raise doubts
about the status of ‘‘informational’’ units such as genes and ‘‘memes’’
as privileged currencies and explanatory tools. More inclusive concep-
tions of development and evolution open up space to address issues that
Looking at Ourselves
have suffered relative neglect. It may even be, as I speculate in the pen-
ultimate section, that they will allow the exploration of the complexities
of cooperation and competition without foreclosing some possibilities
beforehand. I close with some comments on the position of with re-
spect to more mainstream theories, suggesting that some of the resistance
it has met stems from the fact that even in science, boundaries can be
treated as defenses against heresy rather than as lines to be drawn and
redrawn in the pursuit of various kinds of understanding. By speaking of
the politics of biology in this way, I do not intend to expose the field to
the world, unclothed and unlovely. The emperor has many clothes, more
than he usually wears, and certain ensembles, not yet tried, might serve
him better.
Redrawing Boundaries
tural change. The models use either genes or learning as the mode of
heredity, and either ‘‘having babies’’ or ‘‘having students’’ as the mea-
sure of fitness. We see here the centrality of information transmission
as a way of accounting for continuity. Also evident is the intimate con-
nection of these models with what I have been calling ‘‘developmental
dualism,’’ the doctrine that some developmental processes are (mostly)
guided from within, and some are (mostly) guided from the outside. It
becomes a simple matter to see how this conceptual geography encour-
ages genetically reductive views of human affairs (if only gene replica-
tion counts), developmental dualism (especially if ‘‘memes,’’ or cultural
units, replicate too), or both. Neither is very pleasing.
The developmental systems approach allows us to redraw the overly
restrictive boundary around the genes to include other developmentally
important influences. The point is not to open a second channel, one to
carry culture, but to trade discrete channels for interacting systems whose
processes give rise to successive generations. Regularities in these sys-
tems’ functioning support the sorts of predictable associations that make
the language of transmission possible.
come out when we talk about cooperation and competition below.) Thus
Dawkins (:) can blithely write of memes (his units of cultural
evolution, analogous to genes) that ‘‘propagate themselves in the meme
pool by leaping from brain to brain’’ precisely because, as Daniel Den-
nett (:) acknowledges, evolutionists and cognitive scientists can
‘‘finesse’’ their ‘‘ignorance of the gory mechanical details of how the in-
formation got from A to B.’’ One could suspect that more than finessing is
involved here, given the evident glee with which these authors displace
agency from humans to selfish replicators. Thus Dennett (:, )
speaks of ‘‘invasion of human brains by culture,’’ and even of interacting
‘‘meme-infested brains’’ (not people!). True, in a more benign mood he
says that brains are ‘‘furnished by’’ memes (:), but his general
thrust is conveyed by a description of selves as ‘‘created out of the inter-
play of memes exploiting and redirecting the machinery Mother Nature
has given us’’ (:). Actually, he defines the notion of a person by
such memic invasion and manipulation.
These are especially vivid examples of the ways in which develop-
mental issues and all sorts of other questions about social phenomena
can be kept at bay (or worse, ‘‘explained’’) by evolutionary theorists. If
evolutionists are content to assume the processes allowing reliable re-
appearance across generations, and wish only to document net outcomes,
then one might reasonably ask what purpose is served by the agentic lan-
guage, especially when it causes such dismay in some and lends itself to
such misuse by others. Turning genes and memes into diminutive master-
minds conflates process and consequence precisely in the way Patrick
Bateson (b) warns against. Yet this sector of the literature continues
to be populated by self-interested replicators tirelessly maneuvering for
the reproductive edge. They are prototypes of competitive selfishness,
and they design us to be their instruments, even, as discussed in chapter ,
providing our ‘‘real’’ motives and goals (see critiques in P. P. G. Bateson
; Kitcher ). Small wonder that this style of evolutionary writ-
ing has given rise to worries about the possibility, even the coherence, of
cooperation, as well as about the more general implications of evolution
for human lives.
Evolutionary and Developmental Formation
Theorists are annoyed when they are told what they have ‘‘always
known.’’ Yet there is a difference between knowing in a parenthetical,
‘‘of course it’s important’’ way about the intimacy and reciprocality of
Evolutionary and Developmental Formation
Once one has accepted a gene-based view of biological evolution one can
then ask: If genes evolve in particular ways, why not other things? If tiny
particles of matter are central to one story, atoms of thought or culture
could populate another (note once again the body-mind duality). Many
approaches to learning, society, and culture have been inspired by evolu-
tionary theory. Sometimes social arrangements are explained by biology.
Or the relation between biological evolution and cultural evolution may
be analogical: They may ‘‘interact’’ or compete, or run in parallel. We
might then envision gene-meme (or culture) coevolution. Just as genes
could manipulate a bird into making a nest to ensure their own propaga-
Evolutionary and Developmental Formation
tion (Dawkins ), an idea or tune could arrange for its own replication
by imitation (hence the term replicator). There is some parity of reason-
ing going on here, and scheming memes can be as entertaining as selfish
genes. But as Patrick Bateson () suggests in a teasing but pointed re-
joinder to Dawkins, one could also speak of nests using genes to make
more nests.
‘‘The radicals’’ should not be cast as exclusive press agents for coopera-
tion. As I noted above, evolutionists have studied it intensively, and Grif-
fiths and Gray () stress the importance of competition (as depen-
dence on limited shared resources) in . The idea of competition is
considerably altered, though, when the focus shifts from organisms or
agentlike replicators to systems of interrelated processes (see Griffiths
and Gray ). At issue is the evolutionary role given to the complex
interdependencies and integration emphasized in . Below are some
tentative comments about these ideas in what might loosely be called
a developmental systems style: one that moves with a certain fluidity
among scales and measures, taking a pragmatic stance toward research
decisions and a somewhat skeptical one toward much received wisdom.
This approach emphasizes emergent pattern from shifting, heterogene-
ous sets of interactants, and changing, multiple control (see also Taylor
). Such phenomena, which some call ‘‘cooperative,’’ are frequently
described by ecologists, developmentalists, and social scientists. They
tend to appear in evolutionary theory only as constraints on, or condi-
tions for, selection—as messy details to be bracketed out, or as strategies
to propagate genes. Although present in the theory, then, they are ren-
dered conceptually secondary. It is possible to avoid that kind of prior
privileging. These aspects of biological processes can be treated in a way
that does not always measure them against the same bottom line. One
can, for instance, concentrate on consequences of variant developmental
systems’ differing rates of self-perpetuation or on the interrelated causes
involved in the systems themselves.7 In the latter case, the ‘‘bottom line’’
is the continuity of a system’s functioning, whether it is reliably repeated
or not.
In chapter I spoke of the importance of understanding how a set of
processes keeps going partly by re-creating its own constituents (N.B.: not
how some subset of constituents re-creates itself, making and using the
rest of the system to further its own proliferative interests). I was speak-
ing of autocatalytic chemical reactions in vital processes, but the signifi-
cance of interacting, self-perpetuating, and self-preparing complexes of
resources and processes is more general. Larger-scale interactions are
Looking at Ourselves
implicated in reproduction and ontogeny as well, and they are not nec-
essarily contained within a skin (this is one of the points of Dawkins’s
Extended Phenotype, and what made that exploration so interesting
to me). The continuity of the germ line is supposed to explain the re-
peated cycling of life courses, but this leaves out the rest of the develop-
mental complex, downgrading important phenomena and encouraging
problematic conceptions of all-knowing designer genes. Disclaimers of
complete genetic efficacy are becoming more common, but they tend to
leave the basic message of powerful and clever macromolecules largely
untouched. If there is indeed a practical difference between merely ac-
knowledging phenomena and giving them full-fledged theoretical status,
then it could be worthwhile to look at evolution through different lenses.
If we ask how systems keep going, how they change or remain stable,
how changes at one level are, or are not, reflected at others, there will be
room for tracking particular units from generation to generation. They
will not have unique status a priori, so analysts may be called on to jus-
tify their decisions. A benefit of refocusing from tightly bounded self-
replicators to loosely bounded repeating systems, however, might be to
allow a more discriminating and generous view of collective activity
(Gordon in press). Just as development has never been fully integrated
into the synthetic theory, the question of long-term stability of multi-
species communities has not been adequately dealt with either (Taylor
; see also Gray ). Peter Taylor stresses the historicity of eco-
logical complexes, which must be developed or reconstructed, not just
‘‘dispersed.’’
Faced with a complex system made of processes at scales from the
molecular to the biogeographical, it is necessary to ask: Cooperation for
what? Competition for what? There need not be just one bottom line
(B. H. Smith :chap. ), and for any indicator to track anything the
system must be kept going. Processes and entities at diverse scales are
often at least partially nested, and there may be different consequences
of an interaction at different levels (Wimsatt ). Hence my earlier
allusion to competing theorists ‘‘cooperating’’ to sustain larger games,
which may themselves be in higher-level competition. At any level, the
entities or processes must be scrupulously specified. The same is true of
the relevant outcomes. If an analyst invokes competition when compar-
ing outcomes (recall Keller’s point), the counterpart could be to speak of
Evolutionary and Developmental Formation
ket for reasons other than absolute superiority. The essays in Keywords
in Evolutionary Biology (Keller and Lloyd ) are instructive, not only
for the contemporaneous and historical variety of definitions, methods,
and assumptions surrounding important concepts, but for the stories that
can be read between the lines and between the entries. These are stories
about the dynamics of a sometimes loosely associated set of developing
research areas, impinging on each other, requiring coordinating concepts,
flourishing or not (Callon ; Taylor ). A historical perspective
can also help us understand how such topics as cooperation and environ-
mental influences in evolution (other than selective ones) gain the power
to contaminate, in these cases by association with group selection and
Lamarck (Shanahan ).
When I speak of a politics of theoretical boundaries, I do not mean
that people are always maneuvering to push some views and suppress
others, although they often are. But unintended consequences are ubiq-
uitous. There is a whole shadow domain of events and connections that
are hard to relate to anybody’s intent (Shotter ). Convenient mea-
surements tend to turn into definitions, and not only in science, despite
individuals’ explicit desires that this not happen. (Think of , or grad-
ing in school. Psychologists are taught that Alfred Binet wanted to avoid
precisely the use to which intelligence tests, some of which still bear his
name, were eventually put.) Specialized measuring technologies arise,
squeezing out other perspectives as intellectual and professional interests
become linked in complex networks. Certain kinds of analyses and expla-
nations are readily available, while others are hard even to think about.
A colleague confides, ‘‘We don’t have the language’’ to describe social
behavior.
Developmental systems treatments have sometimes been criticized for
being too radical. Yet when faced with ’s specific points, people often
reply that the matter in question has always been acknowledged within
conventional theory. I would not want the potential usefulness of this
approach to be missed because topics such as cooperation (not assessed
in an exclusively competitive frame) and environmental influences (not
treated exclusively from the point of view of natural selection) have been
stigmatized. Classification as ‘‘radical’’ or ‘‘moderate’’ implies a single
point from which distance can be measured, but I don’t think that con-
temporary evolutionary theory has such a center. Perhaps natural selec-
Evolutionary and Developmental Formation
tion comes close. Nor would I want what I believe are valid criticisms
to be hastily brushed off with a reflex, ‘‘We knew that,’’ when what is
needed is something more. Some of what is said about developmental
systems is indeed acknowledged by most evolutionists, but one doesn’t
have to be a developmental systems theorist to see the difference be-
tween some familiar-looking bits and a novel but consistent configura-
tion. Clarifying the relationships between this perspective and others (I
consider several in Oyama ; see also van der Weele ) may help
bypass some reflex objections and answer others, but the judgment about
whether work on developmental systems lies inside or outside the syn-
thetic theory can probably be made only retrospectively. Even then there
is no guarantee that the judgment will be either unanimous or enduring.
This page intentionally left blank
Notes
Introduction
tive science. For a different use of dynamical systems that nevertheless ar-
ticulates in interesting ways with some of the ideas explored in this book,
see Fontana and Buss (). Caporael () gives an account of human
social psychological evolution that takes issue with certain of the trends in
contemporary evolutionary psychology, and Andresen () and Studdert-
Kennedy () have written on language in ways that are conceptually
related to the views being presented here.
4 There is no antipathy to genes here, although one does need to say just what
sense of gene one intends. There is, however, a dissatisfaction with certain
ways of dealing with them. The diversity of meanings of gene contributes
to its rhetorical power. See Keller and Lloyd () for some discussions.
Griffiths and Neumann-Held () provide alternative understandings of
the ‘‘molecular gene’’ and the ‘‘evolutionary gene.’’ They argue that the
evolutionists’ difference makers need not correspond to specific stretches
of , and show some of the implications of this idea.
5 Slow, that is, if we take our own quarter-century chunks as paradigmatic.
Other organisms cycle more quickly or slowly, which can make for inter-
esting and sometimes consequential dynamics. Consider the ways in which
our attempts to deal with disease or insect pests can encourage their quite
rapid evolution into ever more formidable threats, in what evolutionists
sometimes call ‘‘arms races.’’ Diseases and pests, in turn, are important
influences on human evolution.
The conscious changes of scale mentioned here, incidentally, are to be
distinguished from the unacknowledged ones that sometimes result in the
bizarre confusions of genotypes with phenotypes that crop up in the litera-
ture.
6 This is so whether the subject is the ontogeny of organisms or the evolu-
tionary relationship between development and natural selection. I have in
mind here people like Alberch (a); Gould & Lewontin (); Ho and
Saunders (); Maynard Smith, Burian, Kauffman, Alberch, Campbell,
Goodwin, Lande, Raup, and Wolpert (); and Webster and Goodwin
(). See discussions in Amundson (); K. C. Smith (); Oyama,
, ; and van der Weele ().
S. J. Gould (, pp. –) observes that ‘‘nothing in the strict Dar-
winian paradigm suggests hostility to developmental issues, but the theory
offers precious little space for the major internalist and structuralist themes
of embryology.’’ This quote illustrates the view of development as struc-
turalist inside to natural selection’s outside. At the same time, it points out
the difference between a de jure theoretical entailment and a de facto prac-
tice. That is, consideration of development is not actually ruled out of evo-
Notes to Pages –
This was based on a talk given at the Schneirla Conference. It ap-
pears in slightly different form in G. Greenberg and E. Tobach (Eds.) (),
Levels of social behavior: Evolutionary and genetic aspects (pp. –).
Wichita, KA: T. C. Schneirla Research Fund.
1 By using the word wish I do not mean to suggest that people seek such a
basis only for what they value; they may also look for ‘‘biological’’ expla-
nations for negative phenomena, perhaps to explain their sense of power-
lessness before widespread or apparently intractable aspects of life.
5 Though Whelan does not seem to make these errors in interpreting herita-
bility, this substitution does not seem likely to improve matters.
This chapter is based on a talk given at the nd Annual Minnesota Sym-
posium on Child Psychology in , and appeared in slightly different
form in M. Gunnar & E. Thelen (Eds.) () Systems and development:
Minnesota symposia on child psychology, Vol. (pp. –). Hillsdale, NJ:
Erlbaum. Further discussion of Tinbergen’s () four whys can be found
in Klama (). In the light of more recent work (Griesemer, in press;
Winther, in press), my mention in note of Weismann’s separation of devel-
opment from transmission, historically an enormously important boundary,
seems too stark. It is the reading of Weismann that has become dominant,
however, and my point turns on just that dominant reading.
1 In the past I have not always distinguished clearly between the kind of inter-
actionism I criticize as being inadequate and the kind I think overcomes
those inadequacies. The former I often call ‘‘traditional’’ or ‘‘conventional’’
interactionism, the latter, real interactionism. But sometimes no modifier is
used, and I have ungenerously expected the reader to know which meaning
I intend. Others have avoided this problem by using another label (Lewon-
tin, Rose & Kamin, ; Tobach & Greenberg, ). Although I shrink
at coining yet another term, I offer constructivist interactionism as a name
for my approach.
The notion of construction has its own problems, one of which is that it
is often associated with environmental or social determinism, and so can
imply arbitrary and unlimited variation. As should be clear from the Intro-
duction and chapter , that is not what is intended here; what I wish to
emphasize is the emergence of the phenotype, as well as its own active role
in that emergence.
2 At least at the level of generality at which they are considered universal,
so that there may be heritable variation in nose shape but not much in the
presence of a nose.
3 The notion of programmatic control probably reflects the technocratic zeal
of early systems thinkers (Haraway, –; Taylor, ; Yoxen, )
better than it does the reciprocal, distributed control that characterizes a
dynamic system. For descriptions of the latter, see M. C. Bateson ()
and Fogel and Thelen ().
4 Instinct belongs to a whole set of concepts that mix evolutionary issues (the
Notes to Pages –
organs or tissues, but this would involve attributing the higher-level devel-
opment to a heterogeneous assemblage of subsystems, and tissues exert a
great deal of influence on each other. The organization among the subsys-
tems would then present additional problems for a straightforward transfor-
mational account.
3 By this I mean only statistical typicality; no essential species nature is re-
quired for such probabilistic generalizations to be made.
4 See Skinner (). On neural selection, see Edelman and Mountcastle
(); on the immune system, see Jerne (); on language, see Wexler
and Culicover (); and for a sweeping view of selectionist explanation,
as well as many references, see Piattelli-Palmarini (). It should be noted
that there is a difference between selection from an actual array of ob-
jects or responses (organisms, neurons, operants) and selection as parame-
ter setting, a difference that tends to be ignored when selectional models
from neurobiology or immunology are mustered as support for the nativist
project in the cognitive sciences.
5 This has the qualities of a good origin myth, but the story is more complex.
For more appreciative views of Lamarck, see Ho and Saunders () and
Taylor (). For more historical detail, see Gruber () and Jordanova
().
6 Or ‘‘ahistorical universals’’ (Kauffman, , p. ). Similarly, Ho and
Saunders (, p. ) declare that ‘‘a scientific study should consist in
the delimitation of the necessities which underlie the process of evolution,
without recourse to contingencies.’’ See also Goodwin (a).
7 The relation between artist and materials is being rethought, even in the
popular press. Of an exhibit of Japanese sculpture a newsmagazine reports:
‘‘The exhibition rightly contends that its artists (or any artists, if you think
about it) don’t transform their materials so much as redirect them. They
don’t make everlasting objects out of inert and characterless stuff. . . . In-
stead, they highlight a few inherent qualities of their materials for a rela-
tively brief moment in time’’ (Plagens, , p. ).
8 For classic critiques of nature-nurture dichotomizing, see Lehrman (,
). The distinction between genetic and environmental information is
associated with Lorenz (). For discussions of this partitioning of infor-
mation see Oyama () and Johnston ().
9 Gray (, p. ) maintains that constraints are considered primary. Al-
though he and I read the literature differently, neither of us likes the insis-
tence on designating one cause as dominant. The genes cannot, of course,
define an array of possible outcomes independently of an array of develop-
mental environments (each of which is actually an indefinitely long se-
quence of environments at various scales). As I mentioned earlier, further-
Notes to Pages –
the extension of the concept of ontogeny, not the use of the traditional con-
cept to curb the power of selection. Similarly, the definition of heredity
presented here is significantly broader than the ‘‘hereditary apparatus’’ of
Ho (). Consisting of the nucleus and cytoplasm, her ‘‘apparatus’’ is
more restricted than the formulations in her later papers, as in her (a),
in which she appears to argue for the inheritance of something like my
developmental system.
Johnston and Turvey (, p. ) capture the same interactive com-
plex when they speak of the ‘‘co-implicative’’ relationship between organ-
isms and their surrounds, as does Gray (b) with his ‘‘reciprocally con-
strained construction.’’ Patten’s () ‘‘environs’’ may also refer to the
same sort of inclusive complex. Whether these authors focus on develop-
ment or on function (as in this last group of works), they have in common
an interest in reducing the conceptual distance between organisms and their
surroundings. See also Taylor (); Levins and Lewontin (); Lewon-
tin, Rose, and Kamin ().
This chapter is based on a talk given at the meeting of the International
Society for the History, Philosophy, and Social Studies of Biology, at Bran-
deis University, in Waltham, Massachusetts. It appears in slightly different
form in () South Atlantic Quarterly, (), –. That paper is also
reprinted in B. H. Smith and A. Plotnitsky (Eds.) (), Mathematics, sci-
ence, and postclassical theory (pp. –), Durham, NC: Duke University
Press.
1 For discussion of the program metaphor, which is a commonplace in bi-
ology and in other fields, see Goodwin (); Nijhout (); and Oyama
().
Although the issue of progress or directionality in evolution is not closed,
it tends to be discussed by evolutionists and philosophers in rather more
rarified terms; see Nitecki (). The emphasis on (nonadaptive) chance
in Wonderful life is part of a much larger argument about hierarchy in evo-
lution and the adequacy of natural selection as an explanation for all levels
of evolutionary change. For an outline of this argument, see Gould ().
2 Both ontogeny and phylogeny, of course, are historical phenomena, and it
is precisely the conceptualization of history that is at issue here.
3 Compare this with the relations between object and context in Callon (),
and between person and setting, in Lave ().
Notes to Pages –
This chapter is based on a talk given at the Genes & Gender confer-
ence in New York City. It appears in slightly different form in A. E. Hunter
(Ed.) (), Genes & gender VI, On peace, war, and gender (pp. –),
New York: Feminist Press. It is also reprinted in M. M. Gergen & S. N.
Davis (Eds.), (), Toward a new psychology of gender: A reader (pp. –
), London: Routledge Press; and in D. L. Hull & M. Ruse (Eds.) (),
The philosophy of biology (pp. –), Oxford: Oxford University Press.
I refer in this chapter to a kind of feminist essentialism as ‘‘radical femi-
nism.’’ Alice Echols () considers this instead to be part of the cul-
tural feminism that she says supplanted the more politically oriented radical
variety by the mid-s, although she says that radical feminism also con-
tinues to be used in the way I use it here. She notes too that an emphasis on
the ‘‘eternal and unchanging’’ easily survives a switch from biological de-
Notes to Pages –
2 See chapter . To the extent that this kind of mobile multiplicity fosters
the appreciation of both particularity and commonality, it can, under the
right circumstances, even serve to encourage empathy. Trying to see the
consequences of one’s actions through another’s eyes, after all (perhaps
even including the ‘‘eyes’’ of Nature), has long been an exercise for increas-
ing moral and emotional responsiveness. As G. Bateson and M. C. Bateson
write, ‘‘empathy is a discipline’’ (, p. ). Such sensitivity can even
give rise to the sense of extended responsibility so nicely embodied in the
work on sustainable agriculture, energy-efficient design, and ecologically
sound waste treatment (by people such as John and Nancy Todd, Wes Jack-
son, Sim Van der Ryn, and Amory and Hunter Lovins) that inspired this
chapter.
3 On the relationship between the two, see Harding (). This is a complex
and difficult problem; Harding believes it is possible, and necessary, to take
a critical stance toward both.
4 On the importance in agriculture of nonscientists’ knowledge, and on alter-
native research methods that use existing conditions in underdeveloped
areas rather than requiring expensive and inaccessible equipment and tech-
niques, see Levins ().
mental ones as they are usually understood, however; at some point change
will seem marked enough that an observer will say that the person is now
different.
3 Although this view is typically dated from the seventeenth century, I do
not mean to imply total discontinuity from earlier periods. The separation
between reason and passion, for instance, or between form and matter, two
other troublesome polarities, predates Descartes. Shweder (, p. ) says
that the Platonic assumptions of cognitivism indicate a desire ‘‘for some-
thing abstract and bleached and really real.’’
4 There is a startling disanalogy: While the subject in cognitive science is the
subject of knowledge (or, better, ‘‘information,’’ because knowledge im-
plies content and meaning), not action, the genetic ‘‘subject’’ both knows
and acts, and so seems the more nearly complete subject. Yet, computer-
talk dominates in both cases, and abstract representations are more promi-
nent than material interactions. Once properly contextualized in its cellu-
lar milieu, however, the gene interacts by virtue of its three-dimensional
structure, without benefit of representations.
5 See chapter for comments on Williams’s dematerialization of the gene
into a ‘‘weightless package’’ of information.
6 I suspect we are more apt to use causal explanations when we do not under-
stand, or cannot accept, another’s reasons. This may help to explain why
reasons and causes have so often been treated as mutually exclusive.
7 My use of ‘‘person’’ to indicate both subjectivity (awareness, including self-
awareness) and accountable agency (capacity for intentional action in a
moral order) is somewhat similar to Dennett’s () personal stance, which
combines the intentional and moral attitudes, and to Shotter’s (, p. )
social accountability stance. The latter’s distinction between acting as one
requires and as one’s circumstances require (pp. –) is perhaps not as
useful. I am thus less skeptical of some notion of agency than are Henriques
et al. (), for instance; in fact, I intend by this term to include just the
sorts of attempts, assertions, and situated acts they discuss.
8 That this is necessary is evident from the way Scarr and McCartney (,
pp. , ) first speak of active children, then go on to attribute real
agency to the genes that control development. Genes are even said to find
some experiences more ‘‘compatible’’ than others (see chapter ).
9 The choice captures England’s movement from a moral view of homosexu-
ality to a more medical one. Hodges interprets Turing’s choice of estrogen
(to suppress libido) as one in favor of mind (the possibility of continuing
his intellectual work) over body (, p. ), but the apparent suicide
suggests they were indivisible.
Notes to Pages –
6 See also Depew and Weber (); and see Weber and Depew () for
an attempt to rework selection.
7 These are not mutually exclusive classes, especially when one thinks in
terms of systems in which consequences are causally important. Think of
the idea of feedback. This does not change the point about investigative
focus.
This page intentionally left blank
References
Barkow, J., Cosmides, L., & Tooby, J. (Eds). (). The adapted mind. Oxford:
Oxford University Press.
Bateson, G. (). Steps to an ecology of mind. New York: Ballantine Books.
Bateson, G. (). Mind and nature: A necessary unity. Toronto: Bantam
Books.
Bateson, G., & Bateson, M. C. (). Angels fear: Towards an epistemology of
the sacred. New York: Macmillan.
Bateson, M. C. (). Our own metaphor. New York: Knopf.
Bateson, M. C. (, November ). Talk presented at the Fifteenth Anniversary
Meeting of the Lindisfarne Association, New York City.
Bateson, P. [P. G.] (). [Review of the book The selfish gene]. Animal Behav-
iour, , –.
Bateson, P. P. G. (). How do sensitive periods arise and what are they for?
Animal Behaviour, , –.
Bateson, P. [P. G.] (). Genes, environment and the development of behav-
iour. In T. R. Halliday & P. J. B. Slater (Eds.), Animal behaviour: Vol. .
Genes, development, and learning (pp. –). Oxford: Blackwell.
Bateson, P. P. G. (). Genes, evolution, and learning. In P. Marler & H. S.
Terrace (Eds.), The biology of learning (pp. –). Berlin: Springer-Verlag.
Bateson, P. P. G. (). Problems and possibilities in fusing developmental and
evolutionary thought. In G. Butterworth, J. Rutowska & M. Scaife (Eds.),
Evolution and developmental psychology (pp. –). Sussex: Harvester Press.
Bateson, P. P. G. (). Sociobiology and human politics. In S. Rose & L. Ap-
pignanesi (Eds.), Science and beyond (pp. –). Oxford: Blackwell.
Bateson, P. P. G. (). Biological approaches to the study of behavioural de-
velopment. International Journal of Behavioral Development, , –.
Bateson, P. [P. G.] (a). The active role of behaviour in evolution. In M.-W.
Ho & S. W. Fox (Eds.), Evolutionary processes and metaphors (pp. –).
London: Wiley.
Bateson, P. [P. G.] (b). The biological evolution of cooperation and trust.
In D. Gambetta (Ed.), Trust: Making and breaking cooperative relations (pp.
–). Oxford: Blackwell.
Bateson, P. [P. G.] (). Are there principles of behavioural development? In
P. Bateson (Ed.), The development and integration of behaviour (pp. –).
Cambridge: Cambridge University Press.
Bateson, P. [P. G.], and Martin, P. (). Design for a life. London: Cape.
Beach, F. A. (). The descent of instinct. Psychological Review. , –.
Beckstrom, J. H. (). Sociobiology and the law. Urbana: University of Illinois
Press.
Bellah, R. N., Masden, R., Sullivan, W. M., Swidler, A., & Tipton, S. M. ().
Habits of the heart. Berkeley: University of California Press.
References
Caporael, L. R. (). The evolution of truly social cognition: The core con-
figuration model. Personality and Social Psychology Review, , –.
Caporael, L. R., Dawes, R. M., Orbell, J. M., & van de Kragt, A. J. C. ().
Selfishness examined: Cooperation in the absence of egoistic incentives. Be-
havioral and Brain Sciences, , –.
Caro, T. M., & Bateson, P. [P. G.] (). Organization and ontogeny of alter-
native tactics. Animal Behaviour, , –.
Churchill, F. B. (). William Johannsen and the genotype concept. Journal
of the History of Biology, (), –.
Clarke, A. D. B. (). Predicting human development: Problems, evidence,
implications. Bulletin of the British Psychological Society, , –.
Cohen, J. (). Maternal constraints on development. In D. R. Newth &
M. Balls (Eds.), Maternal effects in development (pp. –). Cambridge: Cam-
bridge University Press.
Cohen, J. (). The privileged ape. Park Ridge, NJ: Parthenon.
Connor, J. M., Schackman, M., & Serbin, L. A. (). Sex-related differences
in response to practice on a visual-spatial test and generalization to a related
test. Child Development, , –.
Crawford, C. B. (). Sociobiology: Of what value to psychology? In C. Craw-
ford, M. Smith, & D. Krebs (Eds.), Sociobiology and psychology (pp. –).
Hillsdale, NJ: Erlbaum.
Crawford, C. B., Smith, M. S., & Krebs, D. (Eds.). (). Sociobiology and
psychology. Hillsdale, NJ: Erlbaum.
Crick, F. (). On protein synthesis. Symposium of the Society of Experimental
Biology, , –.
Cushman, P. (). Ideology obscured. American Psychologist, , –.
Daly, M., & Wilson, M. (). Homicide. New York: Aldine de Gruyter.
Dawkins, R. (). The selfish gene. Oxford: Oxford University Press.
Dawkins, R. (). The extended phenotype: The gene as the unit of selection.
Oxford: Oxford University Press.
Dawkins, R. (). The blind watchmaker. New York: Norton.
de Catanzaro, D. (). Evolutionary pressures and limitations to self-preser-
vation. In C. Crawford, M. Smith, & D. Krebs (Eds.), Sociobiology and psy-
chology (pp. –). Hillsdale, NJ: Erlbaum.
Denenberg, V. H., & Rosenberg, K. M. (). Nongenetic transmission of in-
formation. Nature, , –.
Dennett, D. C. (). Mechanism and responsibility. In D. C. Dennett, Brain-
storms: Philosophical essays on mind and psychology (pp. –). Cam-
bridge: MIT Press/Bradford.
Dennett, D. C. (). Elbow room: Varieties of free will worth wanting. Cam-
bridge: MIT Press/Bradford.
References
Dennett, D. C. (). Darwin’s dangerous idea. New York: Simon & Schuster.
Depew, D. J., & Weber, B. H. (). Darwinism evolving: Systems dynamics
and the genealogy of natural selection. Cambridge: MIT Press/Bradford.
Dillon, L. I. (). The inconstant gene. New York: Plenum Press.
Doyle, R. M. (/). On beyond living: Rhetorical transformations of the
life sciences. Stanford: Stanford University Press.
Dray, W. H. (). Determinism in history. In P. Edwards (Ed.), Encyclopedia
of philosophy (Vol. ) (pp. –). New York: Macmillan and Free Press.
Durham, W. H. (). Toward a coevolutionary theory of human biology and
culture. In N. A. Chagnon & W. Irons (Eds.), Evolutionary biology and human
social behavior: An anthropological perspective (pp. –). North Scituate,
MA: Duxbury Press.
Dwyer, P. D. (). Functionalism and structuralism: Two programmes for
evolutionary biologists. American Naturalist, , –.
Dyke, C., & Depew, D. (). Should natural selection be an explanation of
last resort? Well, maybe not the last resort, but . . . Rivista de Biologia/Biology
Forum, , –.
Echols, A. (). Daring to be bad: Radical feminism in America –.
Minneapolis: University of Minnesota Press.
Edelman, G. M., & Mountcastle, V. B. (). The mindful brain. Cambridge:
MIT Press.
Ehrenreich, B. (). Blood rites: Origins and history of the passions of war.
New York: Holt.
Ehrenreich, B., & McIntosh, J. (, June ). The new creationism: Biology
under attack. Nation, pp. , .
Elster, J. (). Nuts and bolts for the social sciences. Cambridge: Cambridge
University Press.
Erlenmeyer-Kimling, L. (). Commentary : Nature-nurture redefined. In
K. W. Schaie, V. E. Anderson, G. E. McClearn & J. Money (Eds.), Develop-
mental human behavior genetics (pp. –). Lexington, MA: Heath.
Fausto-Sterling, A. (). Myths of gender: Biological theories about women
and men. New York: Basic Books.
Fausto-Sterling, A. (in press). Sexing the body: Gender politics and the con-
struction of human sexuality. New York: Basic Books.
Fishbein, H. D. (). Evolution, development, and children’s learning. Santa
Monica, CA: Goodyear.
Fogel, A. (). Developing through relationships. Chicago: University of Chi-
cago Press.
Fogel, A., & Thelen, E. (). Development of early expressive and commu-
nicative action: Reinterpreting the evidence from a dynamic systems per-
spective. Developmental Psychology, , –.
References
Fontana, W., & Buss, L. W. (). The barrier of objects: From dynamical sys-
tems to bounded organizations. In J. Casti & A. Karlqvist (Eds.), Boundaries
and barriers (pp. –). Reading, MA: Addison-Wesley.
Ford, D. H., & Lerner, R. M. (). Developmental systems theory: An integra-
tive approach. Newbury Park, CA: Sage.
Ford, E. B. (). Genetic polymorphism. Cambridge: MIT Press.
Freedman, D. G. (). Human sociobiology: A holistic approach. New York:
Free Press.
Frodi, A. M., & Lamb, M. E. (). Sex differences in responsiveness to
infants: A developmental study of psychophysiological and behavioral re-
sponses. Child Development, , –.
Fujimura, J. H. (). Crafting science: Standardized packages, boundary ob-
jects, and ‘‘translation.’’ In A. Pickering (Ed.), Science as practice and culture
(pp. –). Chicago: University of Chicago Press.
Fuller, J. L. (). What can genes do? In C. Crawford, M. Smith, & D. Krebs
(Eds.), Sociobiology and psychology (pp. –). Hillsdale, NJ: Erlbaum.
Furth, H. G. (). Two aspects of experience in ontogeny: Development and
learning. Advances in Child Development and Behavior, , –.
Galef, B. G., Jr. (). Social transmission of acquired behavior: A discus-
sion of tradition and social learning in vertebrates. In J. S. Rosenblatt, R. A.
Hinde, E. Shaw, & C. Beer (Eds.), Advances in the study of behavior, (pp.
–). New York: Academic Press.
Galef, B. G., Jr. (). Imitation in animals: History, definition, and interpre-
tation of data from the psychological laboratory. In T. Zentall and B. G.
Galef, Jr. (Eds.), Comparative social learning (pp. –). Hillsdale, NJ: Erl-
baum.
Gandelman, R., vom Sall, F. S., & Reinisch, J. M. (). Contiguity to male
fetuses affects morphology and behavior in female mice. Nature, , –
.
Gelman, D., Foote, D., Barrett, T., & Talbot, M. (, February ). Born or
bred? Newsweek, pp. –.
Gergen, K. J. (). The social constructionist movement in modern psychol-
ogy. American Psychologist, , –.
Gergen, K. J. (). Social understanding and the inscription of self. In J. W.
Stigler, R. A. Shweder & G. Herdt (Eds.), Cultural psychology: Essays on
comparative human development (pp. –). Cambridge: Cambridge
University Press.
Gergen, M. M., & Davis, S. N. (Eds.). (). Toward a new psychology of
gender: A reader. London: Routledge.
Giddens, A. (). Social theory and modern sociology. Stanford: Stanford
University Press.
References
Haraway, D. [J.] (a). In the beginning was the word: The genesis of bio-
logical theory. Signs , –.
Haraway, D. [J.] (b). A manifesto for cyborgs: Science, technology, and
socialist feminism in the s. Socialist Review, , –.
Haraway, D. [J.] (). Animal sociology and a natural economy of the body
politic, Part : A political physiology of dominance. In S. Harding & J. F.
O’Barr (Eds.), Sex and scientific inquiry (pp. –). Chicago: University
of Chicago Press. (Reprinted from Signs, , .)
Haraway, D. J. (). Primate visions: Gender, race, and nature in the world of
modern science. New York: Routledge.
Haraway, D. J. (). Simians, cyborgs, and women: The reinvention of nature.
New York: Routledge.
Harding, S. (). The science question in feminism. Ithaca: Cornell University
Press.
Harding, S., & O’Barr, J. F. (Eds.). (). Sex and scientific inquiry. Chicago:
University of Chicago Press.
Harré, R. (). The discursive production of selves. Theory and Psychology,
, –.
Harris, M. (, July/August). Margaret and the giant-killer. Sciences, (),
pp. –.
Hendriks-Jansen, H. (). Catching ourselves in the act. Cambridge: MIT
Press/Bradford.
Henriques, J., Hollway, W., Urwin, C., Venn, C., & Walkerdine, V. ().
Changing the subject. London: Methuen.
Herrnstein, R. J. (). Darwinism and behaviourism: Parallels and intersec-
tions. In A. Grafen (Ed.), Evolution and its influence (pp. –). London:
Clarendon.
Hinde, R. A. (). Dichotomies in the study of development. In J. M. Thoday
& A. S. Parkes (Eds.), Genetic and environmental influences on behaviour
(pp. –). New York: Plenum.
Hinde, R. A. (). Individuals, relationships, and culture. Cambridge: Cam-
bridge University Press.
Hinde, R. A., & Bateson, [P. G.]. (). Discontinuities versus continuities in
behavioural development and the neglect of process. International Journal
of Behavioral Development, , –.
Hinde, R. A., & Stevenson-Hinde, J. (). Constraints on learning: Limita-
tions and predispositions. New York: Academic Press.
Hirsch, J. (/). Behavior-genetic analysis and its biosocial conse-
quences. In N. J. Block & G. Dworkin (Eds.). The IQ controversy (pp. –
). New York: Pantheon. (Reprinted from Seminars in Psychiatry, , .)
Ho, M.-W. (). Environment and heredity in development and evolution. In
References
Ingold, T. (). Life beyond the edge of nature? or, the mirage of society. In
J. D. Greenwood (Ed.), The mark of the social (pp. –). Latham, MD:
Rowman and Littlefield.
Jablonka, E., & Lamb, M. J. (). Epigenetic inheritance and evolution: The
Lamarckian dimension. Oxford: Oxford University Press.
Jackson, W. (). Altars of unhewn stone. San Francisco: North Point Press.
Jacobson, J. L., Boersma, D. C., Fields, R. B., & Olson, K. L. (). Paralin-
guistic features of adult speech to infants and small children. Child Develop-
ment, , –.
Jaffe, L. F., & Stern, C. D. (). Strong electrical currents leave the primitive
streak of chick embryos. Science, , –.
Jaggar, A. M. (). Feminist politics and human nature. Totowa, NJ: Rowman
and Allanheld.
Jerne, N. K. (). Antibodies and learning: Selection versus instruction. In
G. C. Quarton, T. Melnechuk & F. O. Schmitt (Eds.), The neurosciences: A
study program (pp. –). New York: Rockefeller University Press.
Johnson, M. (). The body in the mind: The bodily basis of meaning, imagi-
nation, and reason. Chicago: University of Chicago Press.
Johnston, T. D. (). Learning and the evolution of developmental systems.
In H. C. Plotkin (Ed.), Learning, development, and culture (pp. –).
New York: Wiley.
Johnston, T. D. (). The persistence of dichotomies in the study of behav-
ioral development. Developmental Review, , –.
Johnston, T. D. (). Developmental explanation and the ontogeny of bird-
song: Nature-nurture redux. Behavioral and Brain Sciences, , –.
Johnston, T. D. (). The influence of Weismann’s germ-plasm theory on the
distinction between learned and innate behavior. Journal of the History of the
Behavioral Sciences, , –.
Johnston, T. D., & Gottlieb, G. (). Effects of social experience on visually
imprinted maternal preferences in Peking ducklings. Developmental Psycho-
biology, , –.
Johnston, T. D., & Gottlieb, G. (). Neophenogenesis: A developmental
theory of phenotypic evolution. Journal of Theoretical Biology, , –.
Johnston, T. D., & Turvey, M. T. (). A sketch of an ecological metatheory
for theories of learning. In G. H. Bower (Ed.), The psychology of learning
and motivation (Vol. ) (pp. –). New York: Academic Press.
Jordanova, L. J. (). Lamarck. Oxford: Oxford University Press.
Kagan, J. (). The nature of the child. New York: Basic Books.
Kauffman, S. A. (). Self organization, selective adaptation, and its limits:
A new pattern of inference in evolution and development. In D. J. Depew
& B. H. Weber (Eds.), Evolutionary theory at the crossroads: The new bi-
References
ology and the new philosophy of science (pp. –). Cambridge: MIT
Press/Bradford.
Kaye, K. (). The mental and social life of babies. Chicago: University of
Chicago Press.
Keller, E. F. (). Reflections on gender and science. New Haven: Yale Uni-
versity Press.
Keller, E. F. (). Reproduction and the central project of evolutionary
theory. Biology and Philosophy, , –.
Keller, E. F. (). Secrets of life, secrets of death. New York: Routledge.
Keller, E. F., & Lloyd, E. A. (Eds.) (). Keywords in evolutionary biology.
Cambridge: Harvard University Press.
Kessen,W. (). The American child and other cultural inventions. American
Psychologist, , –.
Kimura, M. (). Neutralism. In E. F. Keller & E. A. Lloyd, Keywords in
Evolutionary Biology (pp. –). Cambridge: Harvard University Press.
Kitcher, P. (). Vaulting ambition: Sociobiology and the quest for human
nature. Cambridge: MIT Press.
Kitzinger, C. (). The social construction of lesbianism. Newbury Park, CA:
Sage.
Klama, J. (a pseudonym for Durant, J., Klopfer, P. H., and Oyama, S., edi-
tors of a book cooperatively authored by Durant, J., Honore, E., Klopfer, L.,
Klopfer, M., Klopfer, P. H., Kohn, T., Lessley, B., Nur, N., and Oyama, S.).
(). The myth of the beast within: Aggression revisited. London: Longman.
(Published in the United States by Wiley as Aggression: The myth of the beast
within.)
Klopfer, P. [H.] (). Instincts and chromosomes: What is an ‘‘innate’’ act?
American Naturalist, , –.
Klopfer, P. H. (). On behavior: Instinct is a Cheshire cat. Philadelphia:
Lippincott.
Kohn, A. (). The brighter side of human nature. New York: Basic Books.
Konner, M. (). The tangled wing: Biological constraints on the human spirit.
New York: Holt, Rinehart and Winston.
Kuo, Z.-Y. (). How are our instincts acquired? Psychological Review, ,
–.
Kuo, Z.-Y. (/). Dynamics of behavior development. (Enlarged ed.).
New York: Plenum.
Lakoff, G., & Johnson, M. (). Metaphors we live by. Chicago: University of
Chicago Press.
Lamb, M. E., Pleck, J. H., Charnov, E. L., & Levine, J. A. (). Paternal
behavior in humans. American Zoologist, , –.
References
epistemology, , , –, , , Fox, S. W., , , , unnumb.
, unnumb. note. See also note, n
knowledge Freedman, D. G., ,
Erlenmeyer-Kimling, L., freedom, , , ,
essence, , , , , Frodi, A. M.,
and appearance, , Fujimura, J. H.,
as cause, Fuller, J. L.,
essentialism, , , , , function, , ,
ethics, , , , , , Furth, H. G.,
evolution, , , , ,
as change in developmental sys- Galef, B. G., Jr., , , n
tems, , , , , , , game theory,
Gandelman, R.,
cultural, , , , , , Gelman, D., ,
as genetic change, , , , , genecentrism, , , , , ,
, , , , , , n
human, genes
Lamarckian. See inheritance: as agents, , , , , –
Lamarckian , , , , n,
evolutionary theory, , , , , n, n, n
, , , , –, as ghosts in the machine, –,
evolutionary tough-mindedness,
experience, , –, , , , homunculoid, , ,
, , , , , , , as link between evolution and
n, n development,
reliably occurring, selfish, , , n. See also
selfishness
Fausto-Sterling, A., , n genetic code, ,
feedback, , , n, n genetic imperialism, , –, ,
feminism, , , , n
Fields, R. B., genetic programs, , , –, ,
Fishbein, H. D., , , , , , , , , , ,
fitness, , , , ,
Fogel, A., , n, n open and closed, , ,
Fontana, W., n genetic transcription, , , , ,
Ford, D. H., n
Ford, E. B., n genetics, , , n, n, n
form, , , , , , , , , genotype, , n
, , , , , , and phenotype, confusion between,
n , , n
in genes, , , Gergen, K., , ,
phenotypic, , , , Gergen, M. M., – unnumb.
species, note
Index of Subjects and Names
nature-nurture opposition (continued ) , , , , , , , ,
resolution of, , , , , , , , , ,
structural analogues of, –, , n, n, n,
– unnumb. note, n, unnumb.
necessity, , , , , , , , note n n, n, n
, –, , , ,
Nelkin, D., n parity of reasoning, , , , , ,
neo-Darwinism, , , , n , . See also causal privilege;
neoteny, causal symmetry
Netting, R. M., passion, , n
Neumann-Held, E. M., , , passivity, –, , , , ,
n, n, unnumb. note , , –, ,
Newman, S. A., Patten, B. C., , n
niche, , , , , , , n Pawson, A.,
evolution of, persons, , , , , , , ,
ontogenetic, , –, –, , ,
Nijhout, F. H., , n, n , n
Nitecki, M. H., , n -situations debate,
Noonan, K. M., Peterson, S. A.,
norm of reaction, , , , , phenocopy, , ,
normality, , , phenotypes, , n
nurture, as developmental inter- developmental construction of,
actions, , , , extended,
partitioning of, , , ,
O’Barr, J. F., as secondary to genetic change, ,
objectivity, , , , ,
Olson, K. L., species-typical,
ontogeny, , , , , , , , philosophy, , , , , n
, phylogeny, , ,
Opitz, J. M., n Piaget, J., , unnumb. note,
Oppenheim, R. W., n
organism-environment interdepen- Piattelli-Palmarini, M., n
dence, , , –. See also Pinker, S., unnumb. note
interpenetration of organism and Plagens, P., n
environment Plomin, R., , , ,
organisms Plotnitsky, A., n, unnumb.
as active. See activity note
as evolutionary dead ends, point of view
as passive. See passivity in ethnography,
as static. See stasis and ‘‘information’’,
orthogenesis, , and responsibility, ,
Oyama, S., , , , , , , in science, , , ,
Index of Subjects and Names
politics. See knowledge: politics of; reproductive success, , , , ,
nature-nurture opposition: political , ,
aspects of research, , , –, ,
Port, R. F., , n, unnumb. responsibility, , –, , –
note , , –, –, ,
potential n
developmental, –, Richerson, P. J.,
evolving, Riley, D., ,
genetic, , , , , , , , Rodgers, D. A.,
, Rogoff, B.,
predispositions, genetic, , , , romanticism,
, , Rosch, E., unnumb. note
preformationism, , , , , , Rose, N., , , n
, , , , , n Rose, S., , , , , unnumb.
process structuralism, n, n note, n, n, n
programs, genetic. See genetic pro- Rosenberg, K. M., , n
grams Rosenthal, P.,
psychology, , , , , , , Rowe, D. C.,
–, n Ruse, M., unnumb. note
evolutionary, n
Salamone, C.,
Raff, R. A., , , , n Salthe, S.,
rape, – Sampson, E., ,
Rappoport, L., Sapp, J., , , , , , n,
rationalism, , n
reason, , –, , n Saunders, P. T., , , n, n,
reasons, , , –, – n
Reddy, M., Sayers, J., n, n
reductionism, , , , , , , scale, , , , , n, n.
, n See also levels: in developmental
as provisional single-mindedness, systems
, Scarr, S., , , , , n,
Reinisch, J. M., n
replication, Schackman, M.,
replicators, , , –, – Schaffner, K. F.,
, n, n Schneirla, T. C., , , , ,
representations unnumb. note (chap. )
genetic, , , , n Schwartz, B.,
mental, , , n Searle, J. R.,
reproduction, – n Sehl, I. A.,
reproductive bottleneck, , , , selection
, group, ,
Index of Subjects and Names
Wexler, K., n Wimsatt, W. C., , , , n,
Whelan, R. E., , n n
will, , , , , , , , Winther, R., , , unnumb. note
, Wolters, G., , ,
Williams, G. C., , , , n
Wilson, D. S., unnumb. note, Young, R. M., n
n Yoxen, E., , n
Wilson, E. O., , , , , , , Yuwiler, A.,
,
Wilson, M., , , , Zanotti, B., , ,
This page intentionally left blank
Susan Oyama is Professor of Psychology at John Jay
College, CUNY. She is the author of The Ontogeny of
Information: Developmental Systems and Evolution
(Duke University Press, ).
Oyama, Susan.
Evolution’s eye : A systems view of the biology-culture
divide / Susan Oyama.
p. cm. — (Science and cultural theory)
Includes bibliographical references (p. ) and index.
--- (cloth : alk. paper). —
--- (paper : alk. paper)
. Developmental psychology. . Genetic psychology.
. System theory. I. Title. II. Series.
.
.—dc -