<<

. 44
( 62 .)



>>

the special form of social organization found in the social insects that
became the focus of the superorganism tradition itself.14
I have already said that perhaps the most central ¬gure in the super-
organism tradition as it developed with this focus on the social insects
was William Morton Wheeler, professor of entomology at the Museum
of Comparative Zoology at Harvard University from 1908 until his retire-
ment in 1933. Wheeler™s career extended over a ¬fty-year period, and
his publications included many popular articles as well as a dozen books,
including the in¬‚uential Ants (1910), Social Life Among the Insects (1923),
and The Social Insects (1928).15
Wheeler™s 1911 essay “The Ant-Colony as an Organism” developed the
idea of thinking of an insect colony as a higher-order organism, and his hi-
larious, insightful 1920 essay “The Termitodoxa, or Biology and Society”
introduced the term “superorganism” to describe such a colony. In the
earlier essay, Wheeler had de¬ned an organism functionally as anything
that has the capacities for nutrition, reproduction, and protection, and
in the latter essay he correlates these with “castes” in termite societies:
respectively, workers, “the royal couple,” and soldiers.
Cognitive Metaphor in Biology and Social Science
276

Edward O. Wilson has implied that Wheeler™s defense of his thesis that
the ant colony is literally an organism follows readily from the breadth of
Wheeler™s functional de¬nition of an organism. In fact, Wheeler actually
places little weight on the de¬nition and concentrates instead on showing
that ant colonies share a wider range of features with paradigmatic or-
ganisms, such as individuality and regeneration. Wheeler™s deepest moti-
vation for defending the idea of colonies as superorganisms derives from
and in turn further stimulates his sustained interest in the sociality of
organisms in general and of certain species of organism in particular.
Wheeler concludes “The Ant-Colony as an Organism” with some brief
comments about what he calls “the problem of the correlation and coop-
eration of parts,” saying

If the cell is a colony of lower physiological units, or biophores, as some cytologists
believe, we must face the fact that all organisms are colonial or social and that one
of the fundamental tendencies of life is sociogenic. Every organism manifests a
strong predilection for seeking out other organisms and either assimilating them
or cooperating with them to form a more comprehensive and ef¬cient individual.

Wheeler suggests that colonies provide a more experimentally tractable
system for exploring this problem than do individual organisms. This, in
turn, presupposes that there is one and the same problem about colonial
and organismic life, and that it may have been solved in one and the same
way.16
We can express this problem in two ways, one suggested by Wheeler™s
name for it, the other by his putative solution, “egoistic altruism.” As a
problem of the parts, we might see this as the problem of why and how an
individual, whether it be a colony or an individual organism, manages
to coordinate its parts harmoniously, given their own organismic charac-
ter. Conversely, the idea of egoistic altruism “ of cooperating in order to
bene¬t one™s own self “ suggests not a problem of parts but of wholes, of
why and how it is that individual units, each satisfying Wheeler™s three-
pronged functional characterization of organisms, come to form larger
such units. So expressed, this approximates the problem of altruism, the
problem of explaining how phenotypes that differentially bene¬t indi-
viduals other than their bearer could have evolved through a process of
natural selection.
Although the superorganism tradition has maintained a focus on the
forms of sociality within the social insects, it has also shared Wheeler™s
interest in sociality in general, often with the social insects serving as a
paradigm for thought. Wheeler himself adapted his trichotomy between
Group Minds in Historical Perspective 277

nutritive, reproductive, and protective organismic functions from their
role in de¬ning the organism and in characterizing castes within termite
colonies, to serving as the basis for distinguishing three forms that “true
societies” could take, depending on which of these functions predomi-
nate in the emergent social behavior. Corals and higher vascular plants,
in which a colony is formed asexually by reiterated budding, exemplify
a nutritive society, the social insects a reproductive society, and ¬‚ocks and
herds of birds, mammals, and ¬shes, in which unrelated individuals act
together for protection, a defensive society. For Wheeler, such eusocial
groups both bene¬t the individuals that participate in them and, since
Wheeler thinks of them as types of superorganisms, they also bene¬t
themselves. Since there are two levels of adaptedness, the individual or-
ganism and the superorganism, there are two recipients of evolutionary
bene¬t.17
This concern with the form and levels of sociality within the superor-
ganism tradition has been guided by a pluralistic view of the level at which
natural selection operates, that is, the view that natural selection can op-
erate on a range of biological entities, from genes and cells to organisms
and groups. Group selection has been viewed as the most problematic
part of such a pluralistic view. Although I do not want to enter into sub-
stantive discussion of the debate over the agents of selection here, note
brie¬‚y two paths for exploring the interplay between group selection and
the conception of sociality within the superorganism tradition.18
First, consider the path from the superorganism tradition to a con-
ception of sociality via the appeal to group selection. The extension of
Darwinian natural selection to at least groups of organisms that are them-
selves organisms, superorganisms, makes perfect sense, since the natural
selection that takes place between such groups is a form of competition
between individuals. Since group selection has been traditionally invoked
principally to explain the origins and evolution of sociality, within the su-
perorganism tradition sociality itself is a natural feature of individual or-
ganisms, one that arises from the process of group selection. Here group
selection promotes sociality.
Second, consider a more direct connection between the superorgan-
ism tradition and the naturalness of individual sociality, and its implica-
tions for thinking about the “levels” at which natural selection operates.
For superorganisms there is a clear sense in which the opposition between
individual and group interest that has been taken as constitutive of the
classic problem of altruism does not exist. The competing interests of in-
dividual organisms within a superorganism have played to a draw, much
Cognitive Metaphor in Biology and Social Science
278

as one might think that they have in the case of individual cells within an
organism. Thus, the very existence of sociality is deemed unproblematic
or natural. Superorganisms are just those groups of organisms that have
overcome Wheeler™s “problem of the coordination and cooperation of
parts,” where the parts are individual organisms. Individual organisms
that are part of superorganisms have, in effect, eliminated or minimized
within-group selection. This suggests that in superorganisms the force
of natural selection between groups should be greater than that within
groups. Thus, if the sociality of individuals within superorganisms is to be
expected, so too is the strength of group selection, which is just selection
between groups. Here sociality promotes group selection.
The group mind hypothesis goes beyond a defense of the superorgan-
ism in that it postulates the emergence not simply of a novel individual but
one with a mind. Yet so far nothing I have said about the superorganism
tradition bears on the attribution of mental properties to superorganisms,
and thus on the concept of a group mind. Indeed, explicit endorsement
of the group mind hypothesis has been less pronounced in, and the hy-
pothesis itself has been more peripheral to, the superorganism tradition
than to the collective psychology tradition. This has been so for at least
three related reasons.
In the ¬rst place, there has been a greater concentration on defend-
ing the integrity of the concept of the superorganism itself, introduced
as a novel scienti¬c concept for understanding ecological and social or-
ganization, particularly that found in eusocial insect colonies. The collec-
tive psychologists, by contrast, devoted little attention to defending “the
crowd” as a legitimate, scienti¬c construal of human groups “ perhaps,
as I have implied above, less attention than they should have. Moreover,
for those like Clements who adapted the superorganism concept to make
sense of plant communities and biomes, even if populations were under-
stood as having something like an ontogeny and a physiology, there was
little motivation to ascribe distinctly psychological functions to their ob-
jects of study. Even for someone like Wheeler, the superorganism concept
primarily allowed one to treat insect colonies via a form of social phys-
iology, not psychology. Thus, with attention focused on whether insect
colonies were organisms at all, and an emphasis on adapting a physiolog-
ical rather than a psychological methodology, the question of whether
they were organisms with minds became secondary.
Second, the predication of mental or psychological characteristics to
individual animals has generally been more circumspect than it has been
to individual humans. This has constrained the readiness with which
Group Minds in Historical Perspective 279

those characteristics could be extended to groups of animals. Particularly
for those setting out to go beyond folklore about the social insects and
engage in systematic scienti¬c study of them, C. Lloyd Morgan™s canon “
never invoke a higher-level mechanism when a lower-level mechanism will
do “ has loomed large. Both the behaviorist tradition within psychology
and the developing ethological tradition associated with Konrad Lorenz
and Nikko Tinbergen offered accounts of animal behavior that, in their
different ways, minimized appeals to the mind (as opposed simply to
internal brain physiology). These traditions thus created a climate in
which psychological attribution to animals was circumspect. When one™s
focus was animals whose individual members might be thought to have
minds in only some trivial sense, these tendencies of cautious mental
attribution were ampli¬ed. If insect societies were to have a group mind,
then not only would they constitute a new entity, the superorganism,
but this entity would itself have emergent properties of a kind that its
constituents lacked. In terms that I introduced at the end of section 2,
minds within the superorganism tradition are group-only, not multilevel
traits, and this fact about them contributed to making endorsement of the
group mind hypothesis circumspect within the superorganism tradition.
Third, as we have seen, collective psychologists readily helped them-
selves to the developing notion of unconscious mental processes in in-
dividuals to explain the behavior of crowds and how they produced
something like group personality traits: They were ¬ckle, foolhardy,
emotional, irrational. For collective psychologists, the group mind was
the seamy and steamy emotional underbelly of human nature, reducing
complex and rich psychological processes and traits to simpler and more
impoverished ones. But no such processes or traits could be invoked
within the superorganism tradition without a charge of anthropomor-
phism being laid. This left those in the superorganism tradition with more
limited resources for explaining the mechanisms through which a group
mind could have evolved or operated, and truncated the list of human
mental characteristics that could be literally or even metaphorically as-
cribed to animal societies.
Given the evolutionary and functional motivations for positing super-
organisms, it is no surprise to ¬nd that the sorts of psychological proper-
ties that were postulated within the superorganism tradition were those
that served some identi¬able biological function in the relevant species.
These include the perceptual and communicative abilities involved in
gathering information about food sources and the motoric capacities to
utilize resources and avoid predators and dangers in the world. Some
Cognitive Metaphor in Biology and Social Science
280

of these abilities, such as the ability of the bee colony to locate distant
sources of nectar and regulate the relative number of foragers and hive
workers in accord with the richness of the source, or the ability of the ter-
mite colony to rapidly repair damage to its nest, manifest both some level
of intentionality and a degree of concern over the integrity of the colony.
In short, the behaviors of at least some groups is such that it seems di-
rected at self-preservation, where the self here is a colony, and the means
of achieving that goal is cognitive.
I have already intimated that both the collective psychology and su-
perorganism traditions embody some confusions. It is to one of these
that I turn next, the confusion between the group mind hypothesis and
quite a distinct thesis about the nature of individual minds, the social
manifestation thesis.


5 group minds and the social manifestation thesis
As a way of introducing this confusion, let me begin with William
McDougall™s critical dissent from a reductionist claim made by Herbert
Spencer. In defending a reductionist view of social groups, Spencer had
introduced an analogy between how emergent properties of groups of
physical entities, such as stable geometrical forms and crystallization, and
those of social groups could be understood. To Spencer™s claim that the
“structure and properties of a society are determined by the properties
of the units, the individual human beings, of which it is composed,”
McDougall says,
the aggregate which is a society has, in virtue of its past history, positive qualities
which it does not derive from the units which compose it at any one time; and in
virtue of these qualities it acts upon its units in a manner very different from that
in which the units as such interact with one another. Further, each unit, when it
becomes a member of a group, displays properties or modes of reaction which
it does not display, which remain latent or potential only, so long as it remains
outside that group. It is possible, therefore, to discover the potentiality of the
units only by studying them as elements in the life of the whole.

McDougall™s continuation of this passage reexpresses his view in terms
that make its location within the collective psychology tradition more
explicit, as follows:
That is to say, the aggregate which is a society has a certain individuality, is a true
whole which in great measure determines the nature and the modes of activity
of its parts; it is an organic whole. The society has a mental life which is not
the mere sum of the mental lives of its units existing as independent units; and
Group Minds in Historical Perspective 281

a complete knowledge of the units, if and in so far as they could be known as
isolated units, would not enable us to deduce the nature of the life of the whole,
in the way that is implied by Spencer™s analogies [to stable geometrical forms and
crystallization].19

Melded together here, in both of the halves of this long passage, are two
quite distinct ways in which one might develop the general nonreduc-
tionist view of the relationship between society and the individual that
we have seen was characteristic of adherents to the collective psychology
tradition.
First, there is the idea that groups have properties, including mental
properties, which their individual members do not have, and which are
not reducible to the properties of those members. This emergentist view
of group properties, together with the further assumption that some of
these properties are psychological, entails a version of the group mind
hypothesis that postulates group psychological traits. These traits, while
in a strict sense multilevel traits (such as being angry or irrational), are not
actually possessed by the individuals in the corresponding social group
prior to or simultaneous with their forming that group. In this sense, these
group psychological properties are something other (or more) than the
properties of those individuals. Thus, they are more like group-only than
multilevel traits.
Second, there is the idea that individuals have properties, including
psychological properties, that are manifest only when those individuals
form part of a group of a certain type. This is what I shall call the social
manifestation thesis. Precisely how this sits with the group mind hypothesis
is far less clear, since it makes a claim about the character of individual
minds, and it would seem that any group properties relevant to this claim
could be and indeed would likely be nonpsychological in character.
For example, suppose that individual people become angry or aggres-
sive in certain ways only when they form a certain type of group (for
example, a crowd). Then, unless they do so only because the crowd it-
self has a speci¬c psychological pro¬le, there is no need to posit group
psychological properties, and so no role for the group mind hypothesis.
Moreover, insofar as the social manifestation thesis does lead to the group

<<

. 44
( 62 .)



>>