BEHAVIORAL AND BEAIN SCIENCES (1990) 13, 63-108
Printed in the United States of America
Jonathan Schyil
Department of Psychology, Haverford College, Haverford, PA 19041
Electronic mail: ischull@hvrford.hitnet
Abstracts Plant and animal species are information-processing entities of such complexity, integration, and adaptive competence that
it may be scientifically fruitful to consider them intelligent. The possibility arises from the analogy between learning (in organisms)
and evolution (in species), and from recent developments in evolutionary science, psychology and cognitive science. Species are now
described as spatiotemporally localized individuals in an expanded hierarchy of biological entities. Intentional and cognitive abilities
are now ascribed to animal, human, and artificial intelligence systems that process information adaptively, and that manifest
problem-solving abilities. The structural and functional similarities between such systems and species are extensive, although they
"are usually obscured by population-genetic metaphors that have nonetheless contributed much to our understanding of evolution.
In this target article, I use Sewall Wright's notion of the "adaptive landscape" to compare the performances of evolving species to
those of intelligent organisms. With regard to their adaptive achievements and the kinds of processes by which they are achieved,
biological species compare very favorably to intelligent animals by virtue of interactions between populations and their environ-
ments, between ontogeny and phylogeny, and between natural, interdemic, organic, and species selection. Addressing the question
of whether species are intelligent could help to refine our ideas about species, evolution, and intelligence, and could open new lines of
empirical and theoretical inquiry in many disciplines.
Keywords: animal behavior; artificial intelligence; cognitive science; evolution; insight; intelligence; interdemic selection; learning;
natural selection; organic selection; parallel distributed processing; punctuated equilibrium; species; species selection
1. introduction
In this target article I argue that plant and animal species
are information-processing entities of such complexity,
integration, and adaptive competence that it may be
scientifically fruitful to consider them intelligent. Al-
though I do not mean to suggest that species are sentient,
or conscious, even the idea that they might be intelligent
is admittedly unorthodox. Nonetheless, I hope to show
that the possibility follows from recent developments in
biology, psychology, philosophy, and cognitive science,
and that workers in these and other disciplines will have
to address a variety of novel issues before the idea can be
evaluated properly. I also hope to show that the effort
would be worthwhile,forit could lead to the clarification
and elaboration of our ideas of intelligence, species, and
evolution, and it could open new lines of empirical and
theoretical inquiry.
The analogy between learning (in individuals) and
evolution (of species) is at least as old as the idea of
evolution (e.g., Lamarck 1809). There are also deep
similarities underlying contemporary theories of learning
and of evolution,forat the heart of most such theories is a
variation on Darwin's theory of random variation and
natural selection. Indeed, it has been argued that learn-
ing, intelligence, and the acquisition of new knowledge
can only arise from selected variation, and that any theory
of elaborated intelligence or higher cognition will have to
acknowledge the operation of multiple, nested levels of
variation and selection (see, e.g., Campbell 1974; Csanyi
1989; Dennett 1975). [See also Skinner: "Selection By
Consequences" BBS 7(4) 1984.]
My thesis is that plant and animal species process
information via multiple nested levels of variation and
selection in a manner that is surprisingly similar to what
must go on in intelligent animals. As biological entities,
and as processors of information, plant and animal species
are no less complicated than, say, monkeys. Their adap-
tive achievements (the brilliant design and exquisite
production of biological organisms) are no less impres-
sive, and certainlyrivalthose of the animal and electronic
systems to which the term "intelligence
5
' is routinely (and
perhaps validly) applied today. So perhaps we should
recognize them as intelligent systems of considerable
scope and study them for the light they can throw on the
nature of both evolution and intelligence.
Regardless of how we ultimately answer the question,
"Are species intelligent?" the attempt to answer it could
make it easier to perceive and investigate patterns of
populational change and adaptation, and could throw new
light on interactions of ontogeny and phylogeny, evolu-
tion and intelligence, natural, organic, interdemic, and
species selection, and artificial and natural intelligence.
Thus, I have two major goals: first, to construct a con-
ceptual "macroscope" for viewing aspects of biological
evolution that are usually obscured; and second, to pro-
voke a reappraisal of the concept of intelligence and of the
entities to which we apply it. To these ends, I develop the
concept of species as meta-organisms, and compare their
behavioral competence in evolutionary time to the com-
petencies of intelligent animals behaving in "real" time.
© 1990 Cambridge University Press 0140-525X190 $5.00+.00 63
https://doi.org/10.1017/S0140525X00077542
Downloaded from https://www.cambridge.org/core. Eastman School of Music, on 17 Apr 2019 at 20:07:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms.