Академический Документы
Профессиональный Документы
Культура Документы
SYNTHESE LIBRARY
STUDIES IN EPISTEMOLOGY,
LOGIC, METHODOLOG Y, AND PHILOSOPHY OF SCIENCE
Managing Editor:
JAAKKO HINTIKKA, Florida State University, Tallahassee
Editors:
DON ALD DA V IDSO N, University of Californiil. Berkeley
GABRltL NUCHELMANS, University of Leyden
WESLEY C. SALMON, University ofPittsburgh
VOLUME 190
EVOLUTIONARY
EPISTEMOLOGY
A Multiparadigm Program
with a complete
Evolutionary Epistemology Bibliography
Edited by
WERNER CALLEBAUT
Limburgs Universitair Centrum, Belgium &
Rijksuniversiteit Limburg. The Netherlands
and
RIK PINXTEN
Rijksuniversiteil Gem. Belgium
TABLE OF CONTENTS
vii
PREFACE
PARTI:BACKGROUND
CALLEBAUT and RIK PINXTEN
Evolutionary
Epistemology Today: Converging Views from Philosophy, the
Natural and the Social Sciences
IL YA PRIGOGINE / The Meaning of Entropy
HENR Y C. PLOTKIN / Evolutionary Epistemology and the
Synthesis of Biological and Social Science
RENE THOM / Epistemology of Evolutionary Theories
CECILIA M. HEYES / Cognisance of Consciousness in the Study of
Animal Knowledge
WERNER
57
75
97
105
Evolutionary?
TABLE OF CONTENTS
vi
ARTHUR I. MILLER /
Biography of Ideas
CLAUDE LAMONTAGNE /
267
Sensorimotor Emergence: Proposing a
Computational "Syntax"
APOSTEL
/
Evolutionary Epistemology.
Epistemology. History and Neurology
LEO
283
Genetic
311
327
337
365
381
PART V: BIBLIOGRAPHIES
and WERNER G.
Evolutionary Epistemology Bibliography
405
GENERAL BIBLIOGRAPHY
433
INDEX
451
PREFACE
This volume has its already distant or1g1n in an international conference on Evolutionary Epistemology the editors
organized at the University of Ghent in November 1984. This
conference aimed to follow up the endeavor started at the
ERISS (Epistemologically Relevant Internalist Sociology of
Science) conference organized by Don Campbell and Alex Rosenberg at Cazenovia Lake, New York, in June 1981, whilst injecting the gist of certain current continental intellectual
developments into a debate whose focus, we thought, was in
danger of being narrowed too much, considering the still
underdeveloped state of affairs in the field.
Broadly speaking, evolutionary epistemology today consists of two interrelated, yet qualitatively distinct investigative efforts. Both are drawing on Darwinian concepts,
which may explain why many people have failed to discriminate
them. One is the study of the evolution of the cognitive
apparatus of living organisms, which is first and foremost
the province of biologists and psychologists (H.C. Plotkin,
Ed., Learning, Development, and Culture: Essays in Evolutionary Epistemology, New York, Wiley, 1984), although quite
a few philosophers - professional or vocational - have also
felt the need to express themselves on this vast subject
(F.M. Wuketits, Ed., Conce ts and Approaches in Evolutionary
Epistemology, Dordrecht Boston, Reidel, 1984). The other
approach deals with the evolution of science, and has been
dominated hitherto by (allegedly) 'naturalized' philosophers;
no book-length survey of this literature is available at
present.
Having explored the already overwhelming literature
labeled 'evolutionary epistemology' (see the comprehensive,
up-to-date bibliography by Campbell, Heyes and Callebaut at
the end of this volume), we felt that on the whole, little
had been accomplished in terms of either dependable theory or
relevant application. As Ron Giere was to put it at the Ghent
conference,
what evolutionary epistemology seems to be
lacking today i.s a really good Kuhnian exemplar, a paradigmatic problem solution. Our vaulting ambition with the conference was to lay the necessary groundwork for the transformation of evolutionary epistemology from a rather heterogeneous
collection
of
or
less
viii
rically fertile,
PREFACE
PREFACE
ix
PREFACE
nology") consists of six papers. Donald Campbell's contribution is a plea for 'selection theory' as a (psychological and
sociological, i.e. naturalistic) theory of justification for
beliefs,
"near
at
justificationism
of
mainstream Anglo-American analytic philosophy and that predominant variety of evolutionary epistemology which passes
the justificatory buck to biological evolution" (p. 139). In
addition,
he
sive ('Panglossian') adaptationism evolutionary-epistemological approaches often share with biological orthodoxy. Andrew
Clark's paper usefully complements Campbell's by pointing out
that evolutionary epistemology will never rebut the traditional skeptic (a point endorsed by Campbell) and suggesting new
questions (as opposed to those of the traditional faoundationalist enterprise) the evolutionary epistemologist might
rightfully ask and try to answer; e.g. the question whether an
corre-
biological ("evolutionary epistemology") and the philosophical ("evolutionary philosophy of science") program - in great
detail, thus preparing the way for an answer to the crucial
question: "What, i f anything, does evolutionary epistemology
contribute to philosophy of science?".
Whereas Vollmer
remains tributary, on the whole, to a rather traditional
epistemological viewpoint (grossly simplified: positivism cum
Popper cum Kuhn),
evolutionary
to
demonstrate
science
as
must be seen as a phenomenon rather unlike biological mutation" (p. 183). Linnda Caporael's paper, which rounds off
PREFACE
xi
PREFACE
develops
on
linguistic
of
The Editors.
REFERENCES:
See the Evolutionary Epistemology Bibliography and the General Bibliography at the end of this volume. For reasons of
identification,
some
~-sign;
Part I:
Background
recognized as such,
usually
understood,
viz.
primarily
descriptive inves-
genetic
epistemology
in
the
Piagetian
tradition.
biological
evolution
via
natural
selection
epistemologically
relevant",
but
a majority goes no
further and rejects,
or is non-commital on,
natural
selection analogues
in understanding
the
history of
science. Some of the implications of this distinction, to
which Vollmer's paper is wholly devoted, will occupy us
later on (see especially section 3.2.). Here we only
observe that both programs
are more or less equally
represented in this volume, and that several authors have
achieved
a
more
careful
and
more
balanced
appreciation of the dichotomy ("literal vs. metaphorical
extension of evolutionary biology") supposedly implied by
their existence than was customary until very recently.
Before delving into the heart of the matter, it will
be useful to put the current fascination for EE in proper
perspective. We contend that it cannot be dissociated from
a spreading conviction that the sciences themselves are
gradually becoming "evolutionary" in scope and method.
The structure of the paper is as follows:
In the first section, the call for an EE will be brought to
bear on (a growing awareness of) changes in the metaphysics
underlying any scientific endeavor; changes which are often,
but
not
necessarily
adequately,
summarized in the
catch-word "Mechanism is not enough". A full-fledged EE will
have to do more than coping with Quine's (philosophical)
challenge of a naturalized epistemology (Quine, 1969") and
(scientific)
bio-cognitive
program
implementing =the
suggested by Lorenz' and Campbell's identification of the
evolution of life with a knowledge process
already two
formidable tasks in themselves. In addition, we will argue,
EE must come to grips with the emergence of an "evolutionary
paradigm" in the physical sciences and in the sciences of
man.
nism?
or "Galilean",
"Philosophy
sciences in
ought
to
imitate
its methods,
so far
the
successful
as to proceed
is
usually
also
offered
as
panacea
for
the problems
of science and society
by ecologically
inspired thinkers such as Fritjof Capra (Capra, 1982) and by
other
latter-day holists.
And finally,
in a rather
intriguing but illuminating reversal of the roles that are
traditionally assigned to mechanistic vis-a-vis organicist
thinking in the various sciences,
a form
of anti-mechanism
also
pops
up
in
Margaret
Boden's
plea
for
an
anti-reductionist reading of artificial intelligence, when
she endorses a "computational approach to life and mind
( ... ) entirely compatible with notions of human freedom"
and opposes it to "the mechanization of the world-picture
brought about by the
natural sciences", in which she
includes the life sciences (Boden, 1984, p. 317) (4).
Mechanism, many scientists and philosophers agree
nowadays, is not enough (5). But, we are inclined to ask, do
we really have good reasons to cast our fate to an
organicist alternative which on closer inspection turns out
to be defined in an essentially negative way (holism vs.
reductionism,
synthetic
VS.
analytical
method, heuristics
all,
hasn't this been
biology,
psychological
interactionism,
organicist
conceptions
is
any)
of
"holism",
"organicism",
etc.
(cf.
such as "emergence",
or Itautonomy of a domain of
Methodological mechanism.
In terms of the Hegelian
concept of Aufhebung, the proper alternative to mechanism
ought to point to something transcending its mere negation.
After all, Newtonian
science can boast of quite some
achievements! Where shall we look for such an animal? Robert
Brandon, elaborating ideas previously defended by Marjorie
Grene (Grene, 1974) characterizes mechanism as a methodological (as contradistinguished from an ontological) thesis,
viz. as the search for mechanisms which in fact explain how
the phenomena under investigation are produced. Invoking one
or several mechanisms he takes to be necessary for any
science to be able to model the processes it wants to
investigate:
"To model a process is to offer some more or
less
plausible
hypothesis concerning the
mechanism underlying the process.
Thus any
process
capable
of being
modelled
is a
mechanistic process." (Brandon, 1985, p. 346.)
Mechanisms may thus consist of springs and gears, but also
of, say, "small peripheral populations and geographic isolating barriers" in the case of evolutionary biology (ibid.).
Most
importantly,
mechanistic
explanations
do
not
than the
10
chaotic
Newtonian
systems
systems studied in
equilibrium
thermodynamics,
nor,
for
that
matter,
classical equilibrium
whose
behavior,
at
in the
or nearcertain
II
mechanics,
"may
forget
perturbations",
he
is
12
"An
evolutionist
consists
in
development,
looking
as
perspective
at
progress,
history
as a
essentially
( ... )
as
succession of
13
evolutionary thinking goes deeper than the Darwinian commitment to the historical hypothesis that natural selection is
the preeminent force of evolution" (p. 153). Thus drift
could be thought of as a process of "random selection"(!):
"Sampling error
may transform a population without any of
the organisms in it changing at all. ( ... ) When drift
modifies the composition of the population in this way, it
is not because the indivi- dual organisms change but because
they vary" (ibid. ) .
Unfortunately,
apart from
this
one
additional
example, Sober's claim that (all?) "other evolutionary
forces are
conceptualized
answering
"What-for?"-
in
the
same
way"
(i.e.
in
terms of variational
explanation) remains entirely
programmatic. We insist on making this point because a case
can
be made for
distinguishing between
evolutionary
explanations of a functional or teleological
kind
questions
and
evolutionary
explanations
that do not involve function or teleology.
The theory of evolution by natural
selection clearly
belongs to the first category, the essence of which can be
rendered as follows:
"Put
cryptically,
trait A's
existence is
explained in terms of what A does. More fully,
A's existence is explained in terms of effects
of past instances of A; but not just any
effects: we cite only those effects relevant to
the adaptedness of possessors of A." (Brandon,
1981, p. 103.)
On the other hand, it is also clear that "(t)here is nothing
teleological about the theory of evolution by random drift
or theories of speciation" (ibid.). Hull, we take it, would
agree: Biologists such as Eldredge and Gould (1972) are
claiming that
species
are
static
systems, possibly
homeostatic systems.
However,
been sug-
14
is
not monolithic,
in terms
of
explanatory
to
use
such
words
as 'design',
adaptation,
produce
conscious
and
since
a well-designed
natural
intervention
selection
structure
of
can
without any
God's super-human
constitutes a
form of
some aspect
of the
between the adaptation
goal-or end-directedness tt ,
or finalistic framework
(Plotkin, 1982, pp. 4-5, and references therein). Even
those few philosophers who take the teleology of organisms
to be "intrinsic"
in the sense of irreducible to a
non-teleological, yet explanatorily adequate account, as a
reductionist such as Ernest Nagel (E. Nagel, 1977) would
rather have it - are careful enough to disentangle organic
teleology both
from
intentional
teleology (involving
15
but
undid
the
argument
to
designer",
the evolutionary
does "not
need
progress
to
a known-in-advance
goa111,
which is
what e.g. Kuhn (1970, pp.171-173) was rejecting.
Rather,
"( ... ) we can have the niche-filling model in
which - however meandering science is -, if the
nature of the physical world is involved
in one
So
far so
good.
Cyberneticians
and
systems
theorists-
16
equivalent
to
them.
17
scientific
or
other.
(The
subject
is
always
18
itself
is
continuous
"knowledge
process"
behaviors;
(ii)
have
19
1984),
But
as well as of levels of
such
worries
underdeveloped
need
not
stage of EE.
not,
selection (Brandon,
concern
There
is
an
rations
to
the
mere
in
1982).
the present,
Most important is to
see that
Waddingtonian
biologists
approach to evolution,
us
awareness
among
post-
that a
"genetic bookkeeping"
reducing all phenotypic conside-
"statistical
abstraction" of mean
be
synonymous to
evolution,
as
is
sometimes done in
On the other hand, it is true that biological evolution is "like all historical processes in not being fully
determined by law. Those laws that hold are statistical in
nature" (von Schilcher & Tennant, 1984, p.74). Thus probability (weak causation), an irreversibility (Dalla's law,
justifiable only by means of a statistical argument) that
does not preclude openness (Lewontin, 1978), and maybe even
20
means of blind
an
increasing
wel1-adaptedness
to
the
environment
is
of course,
Lorenz's solution,
Herbert Spencer,
is to reduce the status of Kant's a
prioris to that of ontogenetic a prioris, themselves the
products of phylogenetic evolution. As Wuketits captures it
in his "second postulate" of EE:
"Innate dispositions are the outcome of natural
selection;
they are the products of selective
mechanisms, which, among all 'innate products',
21
Kantian)
epistemology
and epistemological
metaphysics
against EE by interpreting the former as prescriptive and
the latter as merely descriptive (Wuketits, 1986, p.200).
Thus the fertile dialectic of an EE both presumptively
descriptive of knowers (and the world to be known) and
hypothetically normative, as called for by Campbell (e.g.
1987), would never get a chance to evolve (13).
A multiple-level model of evolution. - The work of
Plotkin and Odling-Smee seems to offer further material
perspective
(cf.
Boyd
of
a model of
organisms adapting
of
to
an
the individual
environment. The
organism
is the
here in
22
be a second
such
process.
they take to
origins:
ing
knowledge in the form of relationships."
(ibid. )
As evolutionary epistemologists, we will have to study and
integrate Piaget's model at this level in a more systematic
way than was done in the past. Leo Apostel, a long-time
student of Piaget, offers the following reasons for taking
genetic epistemology much more seriously than most advocates
of EE usually do:
"the Piagetian brand of EE has the unique
distinction: a)
of being based on an adequate
field of observation and experiment (as is also
the case
for Lorenz's EE);
b) of
being
expressed by means of algebraic tools which
attempt to capture development as such (and here
23
as
such,
as authors
such as
24
another domain,
that of evolutionary economics. According
to Kenneth Boulding,
one of the basic differences between
The
genetic information
artifacts,
but in
artifacts of great
themselves,
of human
two other
very
large
numbers of
variety
human
beings
blueprints,
libraries,
computers,
25
(1979,
p.9),
who writes that
"(i)deas, beliefs and
sentiments, from the point of view of behavioral biology,
are important only to the extent that they influence
behavior".) After all, even the most rarefied forms of
adaptation remain subject in principle to the inexorable
test of selection. That is the kernel of the very idea of a
"nested hierarchy", albeit that in most cases - and happily
so - it will be our ideas that "die in our stead" (Popper).
This issue should be disentangled from the rather
different
questions
concerning
the
realist vs.
instrumentalist (etc.) character of scientific knowledge
and the nature of the "non-ultimate" selection criteria for
scientific belief and belief change.
Contrary to what many
evolutionary epistemologists would maintain, EE does not
"prove" the truth of scientific realism as it has come to be
defined
of late (van
Fraassen,
1980;
Churchland &
Hooker, 1985; Paller & Campbell, 1987). For a defense of the
mildly agnostic view that "given the natural possibility of
alternative life-styles, needs, capacities and cognitive
structures it makes
no sense to
identify our ideal
scientific model of reality with the ultimate nature of the
world-in-itself" (Clark,
1986, p.1S8), see Clark (1983,
1986).
But let
us return to Boulding's
proposal. His
definition of a
sphere of human
tl
and
Campbell,
this
volume,
on "selective
cross-borrowing")
lends
itself
to
an interpretation
altogether different from his (in terms of "super sex") as
well. Thus Hirshleifer (1977, p. 14) was able to maintain
with equal justification that in the sphere of economics,
"no
genetic recombination
himself
offers
pretation.
More
no
cultural evolution,
could
account for
good
is
involved"
reasons to
importantly,
he fails to
the
fact
at
like
so
many
specify the
that
all. Boulding
we
seem
writers
on
force(s) that
to
observe a
intuition
of
Mendelian
and
transmission geneticists
26
energy
to the careful delineation
of genotypic and
phenotypic levels in human action and cognition, is Sidney
Winter. He reminds us that
one of the characteristics of
the "tandem mechanism of variation and selection" (Mayr) is
that
"the environment does not act directly
on genes
genotypic
gies ll ,
or,
"decision
rules",
tion. "What the environment operates on, and rewards and punishes, is not the rule but the actions evoked
from the
rule by variables in the environment itself" (ibid.).
Disregarding here the methodological problems which
the actual identification of the genotypic behavior routines
involves (they
certainly are not
identical with the
routines as formally
described by, say, an organizational
chart), a potentially very useful idea seems to emerge: One
of the ways (but certainly not the only way!) in which the
27
correctly
remarked
maintain,
it
seems
us
to
does
that
actual
so regularly
indeed
to
in
paper, Karin
reliance
practice
fact,
"as
on
some
is
rendered
why
be
the
we
would
insist
that
"naturalistically".
An
of us
This
action
m.ay deviate
of
"rule-following"
important
step
most
aware of.
in
be
this
28
superior in
beliefs can
to
avoid
"occultism"
as
well
as
"the
rational
of
science,
of science seems
It is
implausible that
obtain a
"( ... )
a matter not of biological but
rationally prefe7en~ial
elimination
historical
transm1SS10n
in
the
present
owing
to
of
of
reasoned
of purposive consideracase
considerations
oriented
towards
welfare
and affective
well-being" (Rescher, 1977, p. 133).
It shares this question-begging feature (non-naturalism)
with Toulmin's (1972) model, in which a "cunning of reason"
is invoked to avoid the tautology that "might is right:
selective survival
is success" (Lakatos). (We will return
to this issue in section 3.2.) On a more general level
(cultural evolution) than the one considered by Rescher
himself (scientific evolution), we will retain the idea
that teaching methods (instead of teaching specific ideas
or behaviors) is an important type of "replication by
selective borrowing" (Campbell).
On the other hand, if the genotype-phenotype distinction from biology is appropriately generalized (one of us
has attempted this in terms of a self-simplification of
the phenotype in the genotype, which functions essentially
as an economizing device), it becomes possible to envisage
a nested hierarchy of genotype-phenotype distinctions on
29
consequences
of
earlier
of Itfrozen accidents",
and
and
modifications,
the
persIstence
lopmental lock ll
work on the
view "innateness"
as a degree
property related to
the degree
Wimsatt's approach
therto
unexplored
genotype/phenotype
analogues as well.
the
physical
of generative entrenchment,
environment as generatively
entrenched in a
30
31
A
nested hierarchy
of
{"vicarious"} selectors
(Campbell,
1974a: ten levels) going from more or
less automatic or subconsious processes to the very
sophisticated processes of science.
(ii):
Entitativity:
obvious entities
("natural kind"
specimens)
present themselves unambiguously to the
(i):
perceiver.
His
evolutionarily
adapted
"necessary"
recognition
of
these
systems of
will make
entities
(Campbell, 1973).
(iii): Triangulation: truth and validity are dependent on
the
perspective(s) of the knower. Thus, Campbell
(1974b) pleads in favor of "multiperspectivalness" in
science,
(iv):
(v):
This reversal
more internal
32
multivariate process?
In
the latter
case,
wouldn't we
basic
ingredient
of
creativity
characteristic
of
human
considerable price;
no
33
general
has
been
classification
achieved
as
of the solutions
of
now".
What
of such equations
are
we
to
infer
namely Popper's -
has
been
an
34
35
views
on
the
self-organization of
"embedded"
for
system in
human
cognition
(which
Prigogine's sense)
may
is
also
an
in principle
Nevertheless,
biologists
and
other
hitherto,
"more
emphasis
there
adherents
has
is
of
been
sentiment
biological
placed
among
EE that
upon the
and
(i)
more
frame for
beginning our
characteristic of science:
36
37
of
original
contributions,
the punishment
of fraud and
other aspects of self-policing) in functionalist sociological terms, see Hull (1978). This is one of the very few
examples we know of an attempt to consistently extend the
functional
"logic"
of
evolutionary
explanation (cf.
section 1.2.) to the aspect of culture called science, which
is neither prone to dogmatic reductionism (despite its misleading title, referring to sociobiology) nor to facile
metaphorization.
In his (1982), Hull has explicitly pointed out that
there is a third route open to EE: building more general
theory. A crucial difference between, on the one hand, a
direct analogical or metaphorical extension of the form:
(1) biological exemplar -> model of culture (e.g. science)
and, on the other hand, building a more general model (or
theory, etc.), which is then respecified, of the form:
(2) biological exemplar
(e.g. science)
->
general model
->
model of culture
38
Dennett's:
We indeed want
Kornblith's
can,
of
course,
definition,
question
the
adequacy
of
listic;
i.e.
the methodologically
happens
is
methods which,
natural
on our
whatever exists or
susceptible
to
explanation
"through
although paradigmatically exemplified in the
sciences,
are
continuous
from
domain
to domain
( ... )"
(Arthur
Danto).
Robert Richards, discussing
Campbell's BVSR model of scientific knowledge acquisition,
fell in the same anti-naturalistic trap when he wrote:
"Ideas are selected and retained by men for a
variety of explicit and implicit reasons: power,
passion,
reason
enjoy.
inertia,
39
animals') "minimal rationality" (Cherniak, 1986) in a nonvicious way (i.e. in a way not already presupposing rationality).
Here
again
we
think
the
evolutionary
epistemologist can
learn a great deal from the way
certain economists have dealt with the problems of limited
rationality (Callebaut, 1983; for a recent review see Simon,
1983).
Towards an integral epistemology? - A few quotations
will make clear we are still far away from the "integral
epistemology"
some EE enthusiasts claim is already coming
of age:
(i) According to Campbell (e.g. 1974a, 1974b, 1977), science
is
a particular
system of
"winnowing"
and quantified prediction, designed so that outcomes quite independent of the preferences of
the investigator are possible." (ibid.)
(ii) In a remarkable comparison of Kant's and EE's problems
and notions, Vollmer (1983') tries to convey the conviction
that
evolutionary
epistemologists
have
firm
and
all similarities,
Darwinian ll ,
(iii)
the complex
regards as "not
phenomenon of
40
41
The model is
so
general as
evolutionary forces
to be capable of
complementing -
dealing with
trial-and-error adaptation"
(1982a, p.157). He
sums up a list of counterexamples and
forgotten cases, such as
hitch-hiking genes and the autopoiesis of organisms (18). He then pleads for a model that
will systematically take into account the dialectical relationship between organism and environment in the process of
evolution.
All this is still at the level of the evolution of
biological organisms and not that of EE proper. Extrapolating the investigation to EE, Lewontin remarks that the
basic structure of the trial-and-error metaphor implies the
unidirectional adaptation of organisms (e.g. scientific
groups) to their environment. Nevertheless, in all knowledge
processes
constantly
and
most certainly
reconstruct
their
in
science, organisms
environment
and
are
42
as one of
versial
interpretation
of
Waddington's
as
"collaboration"
by itself un-
meaning of
words
activity".
formation is
such
certainly
insufficient to
and "synthetic
explain organismic
43
(technological)
some
intervention.
environmental
Negative
stability,
feedback
however
requires
transient, to
should be the
voiced by evolutionary
epistemologists on
However,
as far as
empirical
EE,
biological
it uncritically
44
Since no one can seriously deny today that science is essentially a collective phenomenon (see, e.g., Knorr Cetina,
1981,
or Boon, 1983, for hard data), Hull's observation
applies to EE
as well. We will return to this point immediately.
Moreover, Hull warns us - and we think h~s point is
well taken - against overly optimistic expectatlons as to
what an EE will ever be able to accomplish:
"No strictly biological theory of evolution is
going to explain very much about the content of
human
conceptual
systems
because
these
particularities are not the sort
of thing
evolutionary theory is designed to explain.
( ... ) If it cannot make ( ... ) predictions about
the genetic makeup of biological populations,
it certainly will not be
able to explain
comparable changes in societies or conceptual
systems." (Hull, 1982, p. 275.)
Keeping in mind these limitations, we can still rightfully
ask what advocates of EE can learn from biologists, be
they neoDarwinian hardliners or challengers of the
orthodoxy. As Ho and Saunders (1984b, p. 3) remind us,
"(t)here have always been
critics of the neo-Darwinian
synthesis: independent thinkers who steadfastly refused to
lose sight of the fundamental problems of evolution which
the theory does not address". Piaget can be seen as one of
them.
Conrad Waddington and
his
Theoretical Biology
Club, the "arch antisynthesist" Richard Goldschmidt, best
known for his "hopeful monster" idea (Gould, 1983, p. 89),
D'Arcy Thompson and others have raised many disturbing
questions about the modern synthesis. Yet the precipitating
factors for the current turmoil seem to have been primarily
some recent discoveries in molecular biology (Kimura's
"neutral theory" according to which many mutations may be
neutral with regard to natural selection - Kimura, 1983)
and in paleontology (the saltationism controversy - e.g.,
Eldredge & Gould,
1972; Eldredge & Cracraft, 1980).
According to Ho and Saunders, the neutral mutation concept
"presaged the fall from dominance of the genetic theory of
natural selection -
of theories
45
nature).
(cf.
Prigogine
on
the
"over-creativity" of
a specific
amino acid
by
change,
which
is
vanishingly
small
( ... ) Now, the latter supposition contains the
false assumption that we have a universe of
pure numbers devoid of physics or chemistry. In
fact,
experiments under simulated prebiotic
conditions
consistently tell
us
that the
probability space of prebiotic proteins is much
more restricted ( ... ) The same experiments show
that the other assumption implicit in such
formulations - that function is a very rare and
special quality created only as the result of
natural
selection
is also false ( ... )"
(ibid. )
Ho and Saunders therefore vindicate Henderson's famous
thesis of the "fitness of the environment". They point to
the one- sidedness of the Darwinian notion of adaptation
(cf. Lewontin, 1978; 1982a), which misses the reciprocal in
the relationship:
"the
environment
is fit
for the
origination and evolution of organisms". Inherent in the
alternative theory of evolution which the authors represen-
46
ted in
their reader
determinism
want
to
advocate is
the
concept of
voiced against EE
as
a theory
of
scientific evolution is
progressive,
albeit in a
become
"primary
to
evolution",
hence "epistemolo-
Darwinian view)
willing to pay
47
A very popular
metaphorical unit in
EE
is a
generalized
concept of information (cf. Lorenz, 1973) or
knowledge (see Plotkin's and Heyes' papers), apparent also
in Campbell's unification of the concepts of "selector"
and "learning system" as elements of hierarchical systems.
The forces of evolution which are accepted are the same as
those in the synthetic theory (cf. section 4), but they are
now exerted on analogues of biological information at
"higher" levels of evolution, viz. culture
(Campbell,
1974a, 1974b; Boyd & Richerson, 1985, p. 3). The original
biological paradigm has been "blown up". There is no harm
in such a procedure, provided it can be justified .(cf.
section 4).
An internal justification has in fact been offered by
independent scholars. The "nested hierarchy of selectors"
presupposes the unity of the biological world, comprising,
e.g., both the amoeba's and (at the other extreme of the
spectrum) the scientist's system of adaptation (Campbell).
48
i.e.
organisms
in their interactions
with their environments. This
critique has been brought forward by Plotkin and Odling-Smee
(1979) and it applies to most of EE. It is a fundamental
critique because it focuses precisely on
the units of
selection in EE. Plotkin and Odling-Smee (1979) claim that
not the knowledge systems (and the selectors) per se are the
proper units of evolution for EE to deal with; rather it
are
the
clusters
of knowledge-cum-environment units
evolving in a
constant
dialectical relationship (cf.
section 2.1.). Their critique is again convincing, we
conjecture, but it is not clear in what way this will
affect the hierarchical organization Campbell and Lorenz
are speaking about. This is a program that needs to be
worked out in the future.
A lot of fuss has been made recently about the claims
of some sociobiologists (especially Dawkins, 1976, 1982) to
be justified to dispense with all units of selection
"higher" than the ("selfish") gene: e.g. the group, the
individual organism of traditional Darwinism, and even the
genome considered as an organized whole (Mayr, 1976, p. 49,
p. 61). According to George C. Williams, who started this
razzamatazz, it is to be conceded that "it is unrealistic to
49
coefficient
relative
to
another
at
the same
locus at any
given
point
in time. Such
coefficients are numbers that can be treated
algebraically, and conclusions
inferred from
one locus be attributed to
the effect of
selection acting independently at each locus."
(ibid. )
This is a daring claim, and a lot of energy has been spent
in recent years in order to refute it. These arguments (see
most notably Wimsatt, 1981b; Sober, 1984; and Lloyd, 1986)
are rather technical and are not directly relevant to our
theme per se. But the upshot of this controversy, which is
essentially about the possibilities and limitations of
reductionistic research strategies, is of paramount interest
uncompromising reductionism
are characterized
have
been
posited
in
more
by
a number
or
less
arbitrary manner
(Wimsatt, 1980b).
The lesson EE can learn from this is that the Scylla
of
reductionism and the Charybdis of anti-reductionism
(viz. group- selectionism as an all-pervading phenomenon a
la
Wynne-Edwards)
are
both
to
be distrusted.
Pluralism reigns once again
(cf.
Sober, 1981"), and
advocates of EE can breathe without having to worry too much
(at present!) about the weak anchoring of their concepts in
"hard" biology. Dawkins shows them the way
here,
with
his
of
of
neo-Darwinism is
Dawkins' work on
"replicators" (Dawkins,
1982) and Hull's on "interactors"
(Hull, 1982), the amended concepts, we bet, will be quite
suitable for the evolutionary
epistemologist's purposes.
Hull's own work on the historical
nature of conceptual
50
age.
ACKNOWLEDGEMENTS
Our warmest thanks go to Leo Apostel, Barbara Frankel, Jens
Robert Maier, Arthur Miller and Neil Tennant for
their oral comments on the "position paper" for the 1984
Ghent EE conference out of which this introductory paper
grew; to Jean Paul Van Bendegem for allowing us to borrow
some ideas from an unpublished manuscript on the relation
between Prigogine's and Thorn's theories; to Aderito Sanches
for remaining one hundred percent a Newtonian psychologist;
and especially to Celia Heyes, Karin Knorr and Henry
Plotkin, whose exemplary homework saved us from many a
mistake. Our very special thanks go to Don Campbell, for
everything.
H~yrup,
NOTES
(1) Since few readers will be acquainted with this unexpected and maybe unsollicited reference relevant to EE
historiography,
we
add
the German
original to our
translation of fragments 5, 6, and 8 from Brecht's "tiber
'Das Ding an sich"': (5) "Der Baum erkennt den Menschen
mindestens so weit, als er die Kohlensaure erkennt." (6)
"Zur Erkenntnis des Baums gehort flir den Menschen die Benutzung des Sauerstoffs. Der Begriff des Erkennens muss also
weiter gefasst werden." (8) "Das Leben seIber ist ein Erkenntnisprozess.
Ieh
51
rather more in
"Erkenntnistheorie
muss
be
sein ll (the
a critique of language
in the
In the lecture
delivered
when
he
received the
metaphor,
52
to follow. Cf.
Wimsatt (1986a).
(6)
Van
Parijs
(1981),
McKelvey
(1982) and
which seems to
(7)
The latter point
reflects Hull's qualified
endorsement of Jon Elster's critique of functionalism in
social-scientific explanation. According to Elster (e.g.
1983, ch. 2), genuine functional explanation requires that
the actual operation of feedback loops passing through the
relevant
social
group
(etc.)
is
demonstrated.
"Functionalist" sociologists rarely, if ever, do this. Hull
thinks that Elster's requirement is not overly restrictive,
but that "at least sometimes it can be met by social groups
in their production of conceptual systems" (Hull, 1982, p'.
314).
(8) Sober also stresses - correctly we think - that
"the idea that variation in nature is not only something
that can be explained but is also an explanatory principle
in its own right" (Sober, 1984, p. 135) is at the heart of
what Mayr (e.g. 1976, ch. 3) has called population thinking.
Mayr
opposes
this
to
typological
or
essentialist
thinking in the Platonic-Aristotelian tradition (not to be
confused with what Aristotle thought himself, at least
according to Mayr:
Mayr, 1982, p. 11). The ontology
underlying population thinking has important consequences
53
to
exclude here
all cases of
"false consciousness",
54
evolution,
(19)
The evolution of
55
science is
to characterize as "naturwiichsig",
i.e.
where wasteful
variation by definition precedes the selection of scientific
"alternatives"
(Schifer,
1983).
In our op,n,on, the
relatively large autonomy ("Eigendynamik") of scientific
evolution might eventually turn out
to be
the main
obstacle hindering the construction
of a full-fledged
REFERENCES
See the cumulative
of this
write:
I.PRIGOGINE
58
possible?
Here we are discussing one of the great problems
of philosophy, one which should be connected with our global
view of the world : How could science have nothing to say
about it?
Still, I believe that in a sense Schrodinger was
right.
Classical
science,
even
when
one
includes quantum
and
as long as
orientation in
time is not
there is an
problem of
Science is now in
a pos1tl0n
witnessed
great
revolutions
in
to the understanding
of the nature
59
In this sense,
we can say that our period is
rediscovering time.
But let us first ask the question: How
was time conceived of at the beginning of the century?
Interestingly enough,
opinion.
there
consensus of
a sense more
'real'
using a
'spatialized'
caricature
of
time;
to
deal with
Aristotle asks:
from?
Is
it
inscribed in
the
nature of
since Aristotle.
Is time only a human illusion or is it a
cosmic property of which man is but one of the realizations?
60
I.PRIGOGINE
in
position
to
choose
between
the
two
the idea
or time seen as a
the
destruction
of
us
the structures
with
the first
61
existence is
due
to non-equilibrium
states,
the
system
has
'choice'
between various
The "attractor"
I. PRIGOGINE
62
(a)
.*
... .
.....
(bl
Attracteur
(cl
Figure 1:
Three types of attractors for dynamical systems.
Point attractor (1a); line (limit-cycle) attractor (1b);
fractal (non-integer dimensional) attractor (1c).
63
may appear,
and
give
to the
properties.
Dissipative systems may forget perturbations. These
systems are
characterized
by
attractors.
The most
elementary at tractors
are points
or lines such as
one may
64
I.PRIGOGlNE
the appearance of
large
number
of
variables,
acting
in a
of creativity
pointed
out
irreversibility seems to
be
characteristic of human
by
David
H.
Ingvar,
inscribed in
the structure of
future";
while
65
responsible
for recalling
of
primary sensory
macroscopic
context only.
if there
But
in
the
is a constructive
s = k log P
which relates entropy S to probability P (k is a universal
constant, called Boltzmann's constant by Planck).
But let
us
then ask:
What is Probability?
Is the use of
probability
an
effect
of
our
ignorance
or
the
very
mechanics,
physicists
have
been inclined to
think
that
is
interesting
to
contrast
the
opinion
of
( ... )
in support of some false doctrine" (Eddington, 1948;
Lewis, 1930).
I.PRIGOGINE
66
~b
(a)
(b)
Figure 2.
Evolution of a system in phase space r. s (t)
maps point a on point b (2a); there is a conservation of
volume in phase space (2b).
Science is
67
often as
controversial as
modern art: we
of
known
to
present a
the
inverse characteristic
we
know
time
that
scales
involved are
no calculation done
predict?
If we have an infinite
info~ation
about initial
68
J.PRlGOGINE
For
stable dynamical systems,
the situation is
different: increasing the precision, we come closer and
closer to the prediction of the deterministic dynamics.
We
can
expect
that
probability
and, through
Boltzmann's equation, entropy, will be the expression of
dynamical instability and not, as assumed by classical
physics, the mere expression of our ignorance.
Let us describe more finely the difference in behavior
of an idealized dynamical system in which the evolution can
be depicted by trajectories, and the 'real' dynamical
evolution, corresponding to a system about which we have
finite information.
As mentioned, physicists have usually
looked at dynamical evolution in terms of trajectories in
an appropriate space, called phase space.
They have been
working on the evolution of sets of points which occupy
some volume zone in phase space.
A characteristic feature
of classical mechanics (conservative dynamical systems) is
that this volume (the 'measure',
in the mathematical
language) is conserved in time.
For unstable dynamical systems the volume in phase
s~ace
will be highly deformed or even broken into small
p1eces.
The destruction of the initial 'simple' volume
gives the appearance of an approach to equilibrium, in which
all the points would be uniformly distributed in the phase
space.
Conservation
of volume
in
phase
space and
conservation of entropy are closely related.
That is the
reason wh~, in classical dynamics, entropy is strictly
constant 1n time.
Initial conditions can be restituted.
Indeed, the fragments of the initial volume could be brought
back simply by inverting the direction of time.
But this description is an idealized one, as it is
based on the traditional idea of trajectories.
As we have
seen, the concept of a given point in phase space loses its
operational meaning for unstable dynamical systems, and we
must show that we
then go into a
quite different
description, the thermodynamical description.
This is a
description in
terms of
processes,
of
measure;
finite information,
but we
expect,
to find a description in
in the
terms
case of
of new
69
~ bw/;////gz/d
(0)
(b)
(e)
Figure 3:
The baker transformation.
General operation of
baker transformation on state (a) gives state (c) through
intermediary state (b).
tal
IIII
I'
iI
iii
A,
D
I
(bl
A,
[J ~
B,
e,
B,
e,
Figure 4:
Contrasting evolutions of ensembles of
points in the baker transformation. Contracting, vertical
fibers (4a); dilating, horizontal fibers (4b).
I
I
70
I. PRIGOGINE
where
is a 'deformation'
for unstable
of
dynamical systems.
cannot
go
into more
given.
In
'delocalized'
other
words,
we
deal
more
with
new
can be made
space
is
square; the
transformation,
direction
of
Our
dilating
physical time is
fibers.
Instead
to equilibrium
oriented in the
of
remaining
constant,
71
the extension of
the domain
occupied
by
p (its
p decreases
in time.
Contrary
see the
fundamental
contrast
objects were
with
classical
conservative
be
made
only
of
such
disappears without
science, whose
systems such
systems,
there would be no
understand
as an
We now
It
would
be
interesting
to
consider
also similar
potentiality;
'actuality'
the
and
to
wave
function
transfer
functions.
understand
is
referred
potentiality
to
into
which a wave
this
function
is
replaced by an
ensemble of wave
I. PRIGOGlNE
72
features
of
the universe,
such
as dynamical
instability.
In this new frame, when we apply it to open
systems, we can understand the appearance of differentiation
in sufficient far-from-equilibrium conditions, as well as
the evolution of such differentiated regions of space into
forms of existence which become more and more independent of
the outside world.
Science can now distinguish
between
two formulations
solution open
time-symmetry.
orientation in
time.
Therefore,
the only
73
'visions
of
the
world',
the one
the
kai usteron".
Henry C. Plotkin
University College London
1. INTRODUCTION
The chapter will bring together some ideas and approaches,
none of which is especially original, in order to try and
answer the question of what a general theory in biology that
makes contact with, and is meaningful to, the social
sciences might look like.
It will be especially concerned
with those elements that are owed to the EE that is defined
in scope by Campbell's review (Campbell, 1974a). This will
be done in two stages.
First, the kind of conceptual road
that I think must be followed if a synthesis is ever to be
achieved will be briefly explored.
The second will touch
upon some of the central issues on which a synthesis stands
or falls.
The reader who is impatient with metatheory may
want to go directly to the second part of the essay.
2. THE METAPHYSICS OF PROCESS AND CHANGE
Many biologists and social scientists, of course, dislike
and distrust theory in direct proportion to its intended
breadth. 'Premature synthesis' is a charitable judgement and
'grand theory' a term of abuse. To the ingenue raised on a
diet of the philosophy of modern physics this is a curious
judgement. What, after all, is the goal of science if it is
not
to provide
an ever smaller
set of explanatory
principles of ever wider compass? Every scientist would
agree with that, albeit grudgingly.
How then, apart from
ignorance, do we account for the continuing Balkanization
of biology and the social sciences?
It is suggested here
that one important reason lies in certain deeply held
beliefs and assumptions about how phenomena in biology are
best explained and which outweigh surface sentiment about
how science in general works.
My argument is that the
thinking of biologists is driven by two contradictory
'forces' - a background and rather hazy notion of generality
on the one hand, and more specific conceptions of the nature
of biological explanation on the other. And by evidence of
75
W. Callebaut and R. Pinxlen reds.), Evolutionary Epistemology, 75-96.
1987 by D. Reidel PubliJhing Company.
H.C,PWTK/N
76
between
functions
and
structures,
and
so
on.
The
them
some are
obvious
and/or
trivial,
others are
metaphysics,
of course,
science has
77
metaphysics in
correct or incorrect.
fruitful or unfruitful.
science is not
something that is
of synthesis between
biological sciences.
This is the metaphysics of process.
Bohm's essays in Volume 2 of the Waddington series Towards a
Theoretical Biology are an especially fine exposition of
such a view.
For Bohm
Things,
objects,
relatively
entities
constant
from
are
a
abstractions
process
all science
"'All is process.
in the universe.
of
of
what
movement
is
and
78
H.C.PWTKIN
certain
relatively
static,
or
entities (e.g.
the
of
some basic
shapes of
the clouds are the
results of
movements of molecules of air and water). Still
later, we come to the notion of a creative
process, in which there are no basic objects,
entities or substances, but in which all that is
to be observed comes into existence as a certain
order,
metaphysics
might be
achieved
interpret biology in
then
synthesis
if
it
process terms.
between
social
79
proves
If this
and
possible to
can
be done
biological science
mechanisms
are viewed
become a commonplace
most,
phenomena
processes,
recent
biology
years
that
It has
many, perhaps
development,
cognitive
processes in
complex
in
of
which
structures
elementary
that
enter
constituents
into
give
rise to
higher
order causal
to
give rise to
not
take
the
form
of
"thing",
but
is
common~sense
and
philosophical
meanings
bears
about
it
is
is
so
to
a
I
obvious that it
repeating, as follows.
I make the same assumptions
the adaptive attributes of phenotypes that most
biologists do,
viz.
H.C.PWTKlN
80
word
in
this
context
because
knowledge,
as
used
in
81
as typical Df neD-Darwinism.
They are quite right in their
criticism.
But thDUgh LewDntin (1982) includes EE in his
strictures, in this instance he is wrong.
If any arm Df
H.C.PWTKlN
82
adaptational theory
stresses
the
dynamic
nature of
adaptations, it is EE.
It is to Piaget especially that we
owe a debt in this regard.
However, this realization
of
the dynamic, mutualistic, dialectic relationship that holds
between phenotypic adaptations and the environment, and
that must include the development of these attributes and
their relationships
within
any
living
system, this
realization is at a very early stage of formulation though
it finds expression at several different
levels
of
analysis.
For example, Patten's (1982) coevolution is cast
in general systems and ecological terms; Lewontin's (1982)
'metaphor of construction' in which individual organisms
are active constructors of their environments is
set at
paving stones
are process;
where adaptations
are,
to use
across generations.
83
characteristics
different temporal
matter.
Ranged
framework.
characteristics
into
some
single
coherent
theorists
is
much
due
to
cultural
84
H.C.PWTKIN
or part of an
organism,
theory.
Lorenz
of
there".
In a world that is ceaselessly changing, in the
metaphysics of process where "creation and transformation of
order is fundamental", "what is out there" is not constant.
Hence the need for knowledge to change also. Thus not only
must knowledge be permanent in the sense of being stored in
a stable form, but it must also be updated, and these are
clearly related properties.
Furthermore it takes time to
update knowledge,
how
much
time
depending
on the
characteristics of the processes of knowledge-gain. During
that time of updating "what is out there" may continue to
change.
This is a well recognized problem in adaptational
spoke
"generational
deadtime"
and
Waddington
of the
"uncertain
futures
problem".
An
especially insightful treatment was by Sommerhoff (1950)
whose analytical biology was cast in terms
of three
adaptational processes - short, medium and long term - in
each of which a "back reference period" limits the accuracy
of an adaptation.
The significance of Sommerhoff's work
lies in his recognizing the need to partition adaptations on
the basis of the temporal characteristics of the processes
forming them.
SOCIOCUl TURAl
I
lEARNING
level 4
level3
IMMUNOLOGICAL
DEVELOPMENTAL
level 2
DEVELOPMENTAL
GENETIC
level,
I
I
knowledge
Figure
GENETIC
85
86
H. C. PLOTKlN
rigorously.
A significant part
of
the analysis
some
universal
set
of
world
rests on
relative to
knowledge-gaining
conceptual
reaches of
of how
lines would be
learning,
and
an argument in
problem
if
we
solving
so
on,
can
that
be
they
better
can be
biological evolution.
meant by the latter.
by a set
evolutionary analogies.
a device that
87
production
of
and
the
further
regenerate
variants.
More
phase
i.e. the
realistically
device
it
or algorithm
heuristic is
chance
component.
produces
some
of
a mix of
what
conservative
worked
before -
pragmatism - it
and a radical or
heuristic
does,
to be
to the past -
of selected variants.
those
novel variants
heredity itself is
the source of
88
H.C.PWTKIN
why
psychologists
have
tended
to
ignore classical
conditioning or to belittle its importance. The Skinnerian
operant is a more flexible learning form because the learner
generates a higher proportion of chance variants and the
ratio of conservative: chance is thus lower;
and some
evidence suggests that the ratio decreases significantly
with perceived changes in the environment.
Here the
learning is more obviously a process of "selection". In
general terms, the lower the ratio the greater the seeming
inventiveness and creativity of the learner, since by
creativity is meant the appearance of the novel and the
unexpected.
Thus, contrary to traditional approaches to
learning theory that emphasized learning's 'instructional'
nature and which could not cope with novel, creative
solutions in learning, the evolutionary analogy can at least
begin to account for creativity. It presents the possibility
of a really general theory of learning, that will deal with
learning forms ranging from the most dull and uninventive to
those that many still feel are quintessentially human. It is
worth pointing out that no other approach to learning comes
anywhere near such a wide explanatory range.
Again, I think that it is a reasonable assumption that
the different ratios that may characterize different forms
of learning have evolved.
And again one of our important
needs 1S to
understand
how
learners
partition the
environment in terms of change and rates of change, with the
evolution of
different
ratios
of conservative: chance
variants matching such rates of change.
A recent development in
evolutionary theorizing,
Maynard Smith's application of the theory of games (Maynard
Smith, 1982) does have an important bearing on this kind of
use of the evolutionary analogy since the r- t- r heuristic is
itself a primitive games theory approach to the analogy.
Maynard Smith accepts that the notion of evolutionarily
stable strategies or phenotypic stable strategies can be
extended to developmentally stable strategies and culturally
stable strategies.
The ESS approach does not contradict
what I am suggesting here,
rather it complements it.
It
tells us how these different ratios (strategies in Maynard
Smith's view) are to be explained in terms of competition
between individuals.
Thus not only might the ESS approach
tell us how different ratios of the analogy have evolved,
EXAMPLES OF HIERARCHIES
LIFE
INHERITANCE
89
(Simon, 1973)
COGNITION
ecosystem
Mendelian genetics
society
organIsm
chromosomes
organisations
tissue
genes
small groups
cell
DNA
individuals
I
I
I
organelle
macromolecule
I
I
I
I
I
I
I
eNS programmes
elementary information
processing units
90
H. C. PLOTKIN
However,
even
a cursory examination of
Simon's
other two
91
H. C. PLOTKIN
92
to
be
altered
appropriately.
What is the relationship between
structural and
control hierarchies? Are they entirely independent of each
other, as depicted on the left of Figure 7?
Is it simply
the case that every level of a control hierarchy is a part
of a structural hierarchy and no more than that? This is an
causes
their
uninteresting
subsequent
answer,
and
behaviour
intuitively seems to
be wrong.
Does
the
difference between
them
reflect
the old
form-function dichotomy?
This is one view advanced by
Pattee (1973).
Let us consider another approach based on
Bohm's notion of the stages of order.
Let us assume that
Simon is essentially correct in that
structural hierarchies
except
in
a historical
sense.
The nature
of the
93
a:
::>
f-
::>
--'
o
Q:
Q:
fVI
f-
------- leVel2
lL.
:r
f-
a.
/1
--I~vell
~
,:
~
structure
Quasi-eQuilibrium
creative
hierarchy of
qualities,
for
instance
the
dimension;
terms of
command
and a control
its structural
structure
in
94
H. C. PLOTKIN
95
teacher I
Science.
Does
that mean
96
H. C. PLOTKIN
A process, in other
98
R.THOM
as
pertalnlng
to
the
arbitrariness
(or
the
formalization, we
99
any history is
a word
constructed
with these
and all
a substratum-space
geometry.
This is
taxonomy.
(The
indispensable
which usually
when
one
includes a
wants
to know
in question may of
similarity,
upon
course be
which
it
is
It is
in
linguistics,
all;
according to the
100
R.THOM
It will
generally
be
observed
that axiomatic
constraints noted in the associations of forms are not
enough to determine
the process:
they allow neither
foresight from the past nor hindsight from the present. This
can be
seen even in linguistics,
where phonological
constraints of primary articulation have to be completed by
constraints
of
secondary
articulation,
that
is
to say,
nor too
in
compliance,
(there is no merit in
case
intellectual
moreover,
with
natural ethics
honesty
consists
in
and in this
avoiding
abusive
even
if,
working,
as we are, on a closed
useful
or
true,
causality
comes
in
concatenations a~pear
corresponding
aX10ms
according
to
Frege.
This
is
where
once
again;
cause
~
effect
immediately intelligible, and the
by
which
they
are
imposed
seem
101
is commutative
with a very
common situation;
as we move up
the causality
graph, we find that there are more actants involved and that
the observed facts are wearing thin. This situation could be
described as resulting from a 'diffuse causality', creator
of that fact initially observed. We have here the origin of
a notion important in historical theory, the notion of
tendency. The way the graph converges towards the fact, as
well as to a possible multiplication of facts of the same
102
R.THOM
kind,
may
'tendency',
be
the
manifestation
of
the
advent
of
one, but
to convey
~ausality
becomes
evident straightaway.
of
causal influences
described
earlier
on;
quantum
mechanics
are
also
103
to mathematical
can be
quantitative 'laws'
theories claiming to
'pregnant' concepts,
submitted to
constraints
expressed by
In sociological
whose mode of
action is often
difficult to
the theories.
In biology, people
discerned.
In this
we touch upon
options of
permanent
character (Holton's
Ithemata l
(1973, and the answer may not be clear. But in
general (fortunately for
science), these great principles
and it
is only
104
R.THOM
NOTE
1. Here we use 'pregnance' as the equivalent of the French
'pregnance' with the meaning proposed by R. Thorn in earlier
writings (1983).
COGNISANCE OF CONSCIOUSNESS
IN THE STUDY OF ANIMAL KNOWLEDGE
Cecilia M. Heyes
Cambridge University
1. INTRODUCTION
'biological evolutionary
theories of science',
epistemology' from
Vollmer contrasts
"There
are
two
interrelated but distinct
programs which go by the name 'EE'.
One is the
attempt to account for the characteristics of
cognitive mechanisms in animals and
humans by a
ideas,
culture
straightforward extension
of the biologi:al
theory of evolution to those aspects or tra1ts
of animals which are the biological substrates
of cognitive activity,
e.g., their brains,
sensory systems, motor systems etc.
The other
program attempts to account for the evolution of
scientific
theories
and
in
C.M.HEYES
106
stated:
to
substantiate
the
Darwinian
claim
that man's
are
denied
or
pronounced
irrelevant. This
107
non-existence
or
unique
inaccessibility
of
animal
consciousness
constitutes
for
many
scholars
an
insurmountable obstacle in the path of any attempt to render
coherent the notion of animal knowledge (2).
The first and
second sections of
animal
cognitive
demonstrating
any
psychology,
current
neither
utility
for
is
the
persuasive
concept
in
of
2. CONSCIOUSNESS AS AN OBJECT OF
EMPIRICAL INVESTIGATION
"What is it like to be an animal? What do
monkeys, dolphins, crows, sunfishes, bees and
ants think about?
Or do non-human animals
experience any thoughts and subjective feelings
at all?
People have always been fascinated by
the
question
of
our
speculations
thoughts
and
constructively,
feelings
of
other
about
the
species?"
C.M.HEYES
108
109
necessary to
110
C. M.HEYES
of conscious
member to understand
his companions'
that
are
then
thoughts.
Agreeing that this
describes an important
function of consciousness,
Griffin contests Humphrey's
conclusion that these capacities are confined to certain
primates, and argues that since they would also be very
useful to social insects one may expect to find them there
also.
No, Griffin's claims for consciousness in invertebrate
species are
not just a consequence
of idiosyncratic
semantics.
They result from
the
rejection
of two
traditional 'rules' for the attribution of consciousness.
The first is the 'similar brains' argument which runs
roughly as follows.
Assuming some form of a materialist
theory
of
mind
which
connects
or
identifies
neurophysiological events with conscious experience, and
we
as
humans
conscious,
mind
should be
III
In
case may
we interpret an action
as the
writings,
he homes in on
the same
C.M.HEYES
112
movement.
Similar
attributions
have
been
observed with
II3
experiments
suggest
an
association
between
the
C.M.HEYES
114
we
have
to
pay
careful
conscious
system.
When he suggests that the two properties of this
input are "the dual functions of consciousness", he makes it
clear that this is a figure of speech which combines
different levels of analysis.
Griffin recently had an
opportunity to clarify his position on the causal efficacy
of consciousness when,
in an
exchange
of
'Letters to the
Editor',
Harnad (1985)
asked him:
"As
to studying
consciousness, how is one to study an entity that (unlike
say, quarks) is causally superfluous in any theory?"
If a disclaimer such as Shallice's were available to
Griffin, surely this would have been the occasion on which
to use it.
Instead, Griffin mistook Harnad's challenge for
115
thinking?"
a
of
reason to
conscious
to ask Griffin:
'Why, under some conditions
should we flout scientific tradition and say of an animal
that it has a certain belief?', he would ultimately reply
'Because it is true'. A proposition of this kind might be
acceptable if it had either empirical force or conceptual
coherence, but in failing to demonstrate either, Griffin's
If one were
good
reading
the
foregoing
or with my
may
have become
fairly ponderous
pointing
out that,
in view
of
the
rising
popularity of
C.M.HEYES
116
Griffin's
issues
may not
be
so
familiar
to
psychologists, the
specialists
in other
fields,
and
that the
problems associated
with the
attribution of mind to other persons and to other species
are not completely co-extensive.
However, I heed the
drumming fingers and hope that their pace will slacken as I
make a few comments on Dennett's recent attempt to introduce
some conceptual discipline into the proceedings.
Dennett (1983) addresses himself to the same audience
as Griffin: "the new ethologists [who], having cast off the
straightjacket of behaviourism and kicked off its weighty
overshoes,
are looking about
somewhat insecurely for
something presentable to wear". The latter frequently cites
Dennett as the provider of a philosophical imprimatur for
his views, but (and this is reflected in the absence of
reciprocal citation) Dennett is arguing quite a different
point. The principal difference between the two authors is
that Dennett is apparently making an epistemological claim,
whereas Griffin is making an ontological claim about animal
consciousness.
Applying his theory of explanatory modes or
'stances' (elaborated in, for example, Brainstorms, 1978),
Dennett argues that it is useful under some conditions to
explain animal behaviour using the language of conscious
intention, that is, in sentences characterised by their
referential opacity (6) and apparently descriptive of mental
states
such
as
belief,
desire,
fear, recognition,
expectation etc. 'Apparently' is stressed because Dennett
emphatically denies
that if a
hypothesis couched in
intentional
terms
successfully
predicts
an animal's
behaviour, then that animal must experience the mental
states to which those terms commonly refer when they are
applied to humans. Ironically, this dissociation is so
fundamental to Dennett's argument (it is the means by which
he avoids the dualism inherent in Griffin's approach), that
it can be easily overlooked.
I would draw the attention of
those in doubt to Dennett's endorsement of the application
of the intentional stance to population genetics, embryology
and morphogenesis (1983, Footnote 17), and to one of the
passages in which he states his position explicitly:
" ... my point is that one should not confuse
the
predictive success
of the intentional
stance (in some domain) with confirmation of a
particular
representation
manipulation
hypothesis." (Dennett, 1983, p. 381, emphasis
in the original)
117
Non-intentional account
Tom (like other vervet monkeys) is prone to
three flavors of anxiety or arousal; leopard
anxiety, eagle anxiety, and snake anxiety. Each
has its characteristic symptomatic vocalization.
The effects on others of these vocalizations
have a happy trend, but it is just tropism, in
both utterer and audience.
First order intention
Tom wants to cause Sam to run into the trees
(and
he has this
noise-making trick that
produces that effect; he uses the trick to
induce acertain response in Sam).
Second order intention
Tom wants Sam to believe that (i) there is a
leopard (ii) he should run into the trees.
C. M.HEYES
118
seems to
such as these
that
"The
question
is
empirical.
The
tactic
of
aprioristic
('armchair')
investigation, but of
using the
stance to
suggest
which brute
empirical questions to put to nature. We can
test the competing hypothesis by exploiting the
rationality assumption." (Ibid. p.347.)
In the case of the hypotheses listed above, Dennett's
commitment to this assumption (that the use of intentional
idioms carries a presupposition of rationality in the
creature or system to which the intentional states are
attributed) leads him to suggest that the non-intentional
explanation could be discounted if Tom silently climbed a
tree upon seeing a leopard when no conspecifics were around;
and that the second-order account would be preferred to the
third-order one if it could be demonstrated that Tom could
make Sam climb a tree by some means other than his cry, e.g.
by imitating the sound of a leopard growling.
There are a couple of things to note in connection
with these examples of the way in which empirical data can
affect the choice of intentional explanations. The first is
that Dennett does not imagine that any of the relevant types
of hypothesis can be readily swept away by 'crucial'
experiments;
the
behaviourist and
contexts
or
societies in
which one
119
must rule
out (or
in)
such possibilities as irony,
metaphor,
story-telling,
and
illustration
('second-intention'
uses
of
words,
as
philosophers would say) that we must avail
ourselves of such high-powered interpretations."
(Ibid. p.347.)
From his instrumentalist position, Dennett could not be
advocating that we choose among hypotheses according to
their likely correspondence with events inside animals'
heads.
Rather, the process of selection, and the selection
criteria themselves
(rationality and simplicity) have
presumably been designed by Dennett to maximise the chances
of the intentional stance realising its potential as a
heuristic.
But what is its "theoretical utility" in
cognitive ethology?
Dennett is quite clear about this: its
purpose is to provide descriptions of real life animal
competences that will facilitate the tasks of the artificial
intelligence and information processing specialists. It is
incumbent upon them, and not the 'mentalist', to explain
behaviour in terms of causally active internal states (to
borrow a phrase from Patricia Churchland's commentary).
Dennett is not so clear about how intentional accounts will
assist AI and information processing theory; or rather, the
unique advantages of the intentional
stance are not
specified.
"The
intentional
profile
or
stance
characterization of an animal
or for that
matter, an inanimate system - can be viewed as
what engineers would call a set of specsspecifications for a device with a certain
overall
information-processing
competence.
An
C.M.HEYES
120
the
right
"black
box"
cognitive
interface
between
specialities:
competences observable
in the field,
121
C.M.HEYES
122
and
123
The
ambiguity
becomes apparent as he
of
Dennett's
simplicity
criterion
account for
contextual
principle,
conditioned
responses, and
variables.
This
a
satisfying
repast in
but it is not very useful in practice - the very
psychologist's
consciousness,
and
set
realist
them
down
approach
intact
to
animal
within
his
124
C.M.HEYES
instrumentalist framework.
In their original context they
are comprehensible
in relation to their
history and
connection with other premises, even if we have changed our
views enough to see them as misguided.
Their role in
Dennett's approach is mysterious.
They might have been
assimilated
thoughtlessly,
or
designed to
make the
ethologist feel 'at home', in accordance with the first of
my versions of his claim.
In summary, Dennett's contribution provides a much
needed (and very entertaining) lesson on the topic of
scientific realism for cognitive ethologists and, pending a
more explicit account of how intentional descriptions are to
be unpacked into information processing models, it may
predict cooperation between two much divided disciplines.
However, in its present form it does not achieve its purpose
of alleviating our intellectual insecurity.
Indeed, it may
aggravate the
condition.
Having
been
assured that
intentional explanations are legitimate, we are uncertain as
to why this is the case.
Is it just that they will keep us
working, or do they have a more specific function? If,
ignoring our suspicion of a snub, we start generating
intentional explanations for a given behaviour, we soon
discover that they breed like toads and are beset by a
plague.
Remembering that the search for reductionist
explanations is no longer interesting, and that lower order
intentions are to be preferred to higher order ones, we are
in a quandary about where to begin culling the beasts. As in
our confusion we reach out for the axe of the rationality
assumption, we find that it was made of rubber all along,
and only put there to give us the illusion of strength. It
is perhaps at this moment that the cognitive ethologist
decides to hang up his field glasses, become a cognitive
psychologist, and have nothing further to do with talk about
consciousness or intention.
4. COMPROMISING ON CONSCIOUSNESS
This heading is not just an example of the alliterative
indulgence common among
those
who
employ
the term
'evolutionary epistemology'. As is usual in the final
section of an essay of this kind, I shall tentatively
recommend some compromises (more respectably described as
syntheses) among the programmes reviewed, to promote a more
vigorous approach to the study of animal knowledge. But
125
first,
will point out some compromises that may have
already been made by animal cognitive psychologists, which
are
in
their work.
epistemologist's interest
attitude
toward both
who
used
the concept of
"cognitive maps", is
Hull in
to animal behaviour.
However,
he was similar to
C.M.HEYES
126
and
goes
on:
"Both
in
human and
While
sympathetic
to
Walker's
confusion
and
resulting
these
authors
suggest,
description
as,
of
an
internal
127
the
intricacies
of
which
can
be
easily
forgotten.
'Sahlins' fallacy' (Maynard Smith) exemplifies the way
in which
the
analogical role that the
products of
introspection play in the strategy can be mistaken.
"The central idea of kin selection is that a
gene A, causing an animal to be more likely to
perform an act X, may increase in frequency in a
population even if act X reduces the individual
fitness (expected number of offspring) of the
animal itself, provided that the act increases
the fitness of animals related to the actor.
Following Haldane, biologists refer to such acts
as
'altruistic'.
It
mathematical
to
other
species of
128
C.M.HEYES
129
C.M.HEYES
The uncertainty of ACP's commitment to evolutionary
thinking can be seen as the outcome of an historical
compromise.
I will
outline their
students
of
animal
behaviour;
proposes
a manner
those
relating
to
communication
or
and
of
131
apparent
human
inappropriately),
predisposition
Dennett offers
to
attribute consciousness
some
specific guidelines
the necessary
consciousness
C.M.HEYES
have
Many
authors (e.g.
provided
questions
detailed
about
human
consciousness
concerning
that
need
to
the
be
addressed.
These include examination of its ontogeny,
relationship with neural structure and function (using
electrical recording techniques and studies of brain damaged
subjects), and association with problem-solving behaviour.
The
studies
proposed
seek
a
correlation
between
neurophysiological or behavioural events and mental states,
predictive
value
as
descriptive
device.
Thus,
behaviour,
subject's
receive
own
intentional
exclusive
description
consideration?
of
And,
133
their behavior
would
not
value.
ACKNOWLEDGEMENTS
I am grateful to Werner Callebaut, Don Campbell, Dave
Oakley, and Henry Plotkin for their guidance during the
preparation of this essay.
This research has been supported
by a Harkness
Fellowship of The Commonwealth Fund of New York.
NOTES
1.
Hull's (1982) tripartite classification contrasts with
those cited in recognising that attempts "to present a
general analysis of evolution through selection processes
which applies equally to biological, social and conceptual
evolution" (p.275) may represent an alternative strategy to
those of reasoning by analogy from, or literal extension of,
neo-Darwinian theory.
This addition provides an important
clarification
of the
motives
of
those evolutionary
epistemologists whose
work is regarded by
Bradie as
anomalous in the context of his typology, and thereby
reminds us that the
explanation
of animal cognitive
processes does not always depend on literal extension of
evolutionary theory.
Whatever its analytic strategy, the
elaboration of a powerful biological EE demands conceptually
sound descriptions of the competences of a broad range of
species, and some factors affecting the supply of these raw
data are the subject of this essay.
2.
I have chosen to use the term 'animal knowledge'
for two reasons: First, surpr1s1ng as it may seem to the
philosopher, 'knowledge' is a relatively neutral term in the
context of the literature I will be discussing. Had I opted
134
C.M.HEYES
conviction.
4.
Tennant
(1984)
exemplifies a more rigorous
approach.
This paper is also pertinent in it recognition
that, owing to their lack of ecological validity and an
asymmetry
in
the
logic
of
deriving
evidence for
intentionality from behaviour, ape language experiments
cannot be used to deny second-order intentionality to the
species concerned.
5.
Refer to Nagel (1974) for the most elegant
exposition of the problem of underdetermination as it
applies to the attribution of animal mind.
6.
135
8.
Elster identifies consistency of desires and of
beliefs as a
minimally
necessary
condition for the
attribution of rationality, and argues for the partial
independence of rationality and
intentionality on the
grounds that there are comprehensible intentions that are
based on beliefs which could not all be true and be
believed.
The example he gives concerns a man who hangs a
horse shoe on his door because he believes (inconsistently)
that it will bring luck even to those who do not believe in
its magical properties. A hypothetical example involving an
animal can be drawn from Premack and Woodruff's (1978)
research on deception by the female chimpanzee Sadie. Sadie
believes (i) that a malevolent trainer requires her signal
to decide which of two boxes contains food, and (ii) that
the malevolent trainer can see through the sides of the
boxes to their contents. Sadie pointed to the 'wrong' box
because she wanted the banana and believed (inconsistently)
that the malevolent trainer required her signal to decide
which box contained the food.
If intentional explanations
are plausible even when they portray
the
animal as
irrational, then Dennett's insistence that hypotheses be
selected on this basis loses its foundation.
9. The difference between the investigation of animals
and humans within Dennett's scheme is associated with the
process of procuring a description of "what is represented",
In the human case, this description is provided by the
subject in the form of a verbal report (apparently) on their
conscious experience.
Taking this report seriously is part
of what Dennett calls 'heterophenomenology' because it does
not require the investigator to assume that the subject is,
in fact, conscious; and it consists of asking what must
really be going on inside the person's head for them to
represent it in this way.
Thus, with regard to persons, an
intentional description of a behavioural competence is, in
addition to the competence itself, an important part of the
explicans.
Judging
by the
1983
paper, Dennett is
recommending
something
rather different
to cognitive
ethologists. It is they who must generate the intentional
136
C.M.HEYES
Part II:
redundant update,
something new.
is just another
an
erroneous
emphasis
on
continuity
in
139
W Callebaut and R. Pinxten (eds.), Evolutionary r./Jisremology. 139-158.
/987 hy D. Reidel Publishing Company.
D. T. CAMPBELL
140
In
all
adaptationism
4.
must
versions
the
construction
be
of
avoided.
EE,
Panglossian
of
consensuses
among
physicists.
6. (An epicyclical solution to a problem created
by
my dogmatic adherence
to
Selection Theory):
Selection Theory
emphasizes
the
role
of "retention"
in this domain
were to
non-commital
on,
natural
selection
analogues
in
141
An important development is
rationality contribute
to
our survival
as
a species"
question,
open
to
investigation
by
natural
They
continue
advances.
to
be
indispensable,
moreover, as science
science depends
on continued
142
D. T. CAMPBELL
has
been ascertained.
ecologies
"typical ll
of
biological
143
evolution,
are
the skeptics'
induction
This does
assumed
provide
not,
in
of course,
provide any
the justification of
richer
grounds
for
better
induction.
answer
has
But
further
it
to
been
may
elaboration,
evolutionary
model,
with
lts attentlon to
past selective
beliefs
the
occasions
are misleading.
limited,
on
which
perceptually
Biological emphases
neuro-perceptual
machinery,
and
and
given
upon finite,
upon
metabolic
inspires
an
retention-and-reproduction"
abstract
algorithm
"variation-selective-
appropriate
to
all
Biological
evolution
is
only
one
such
exemplar.
Trial-and-error learning, radar, sonar, computerized problem
solving, and human thought are others. Biological evolution
has specifics which may be biologically necessary but not
part of the abstract algorithm. Many theories of scientific
evolution retain in their analog irrelevant biological
specifics.
A substantial
class
of the
critics
of
evolutionary theories of scientific progress (e.g. Cohen,
1973; Kary, 1982; Putnam, i 983) are criticizing theories of
this type in ways which do not apply to Selection Theory.
While Q'Hear's (1984) mistaken expectation that evolutionary
epistemologies must assume that scientific beliefs are
selected
accurately
scientific
by
contribution
to
human
survival
does
not
Selection Theory.
D. T. CAMPBELL
144
to
overcome their
This includes Toulmin (of 1972 and 1981, but not 1967),
Richards (1981), and Hull (1982, 1983). Along with this
goes a failure to specify how the differential prospering of
lineages could be due to the greater validity of the beliefs
transmitted (rather
than,
for
example, to political
centrality in high-prestige graduate training programs,
the acceptability of their beliefs
to extrascientific
powers, culture, etc). For me, the crucial advantage of
natural-selection analogs is the argument that the referents
of the beliefs
(e.g.,
"physical reality" or "Nature
Herself") playa role in the differential survival of the
beliefs. A plausible scenario as to how this could occur is
a
necessary aspect of
~ny
epistemologically relevant
evolutionary
personally
theory
doubt
of
that
sc~ence.
an
adequate
Given
this
theory
progress can
be achieved
without
conversions to alien theories which are
some cross-lineage
motivated, in part,
agenda,
of scientific
this
are no
of their day,
145
seems
to
an
me
an
exception
to
the
theory of
Mendelian, without
needed to defer to
changed to
become
impressive
early career
as an anti-
a vigorous
146
D. T. CAMPBELL
opportunistic in terms
contemporary scientific
of established
power
in the
establishment.
He participated,
however,
in a
social
system
employi~g
Itreplications"
as
divination rituals
1n
experimental
its mutual
persuasion
intellectual
self-serving
career
would
lack
of
if
he remained
147
of search
(introduced by
assumptions
tentatively
about
trusted
the
combination
"already
established"
beliefs)
it
will
of
purpose and
"knowledge"
not
obviate
or
the
"knowledge,"
Intelligent
one has
no
choice
but
to
or trusted belief.
More
common
incompatibility
"random"
or "blind '1
creative
the
lS
between
scientists.
assertion
the
variation,
If
one
explore blindly.
achieved "knowledge"
of
critic's
descriptive
connotations
to
pays
full
attention
to my
is
enough
model
1) A blind-variation-and-selective-retention
of the
process
of system to environment.
2)
The
many processes
which
shortcut
a more
full
blind-variation-and-selective-retention process
are in themselves inductive achievements, containing
wisdom about the environment achieved originally by
blind variation and selective retention.
3) In addition,
such shortcut processes contain in
their
own
operation
blind-variation-and
-selective-retention
process
at
some
substituting for overt locomotor exploration
life-and-death
winnowing
(Campbell, 1960, p. 380.)
of
organic
level,
or the
evolution.
148
D. T. CAMPBELL
theory,
and
avoid
Panglossian
excess
"adaptationism"
(Gould, 1982; Lewontin, 1982), or the
epistemological analog of "bean-bag genetics," in which
atomistic components have only uniform main effects (not
among each other) vis-a-vis the selecting
As an aspect of the same caution, one should
interactions
environment.
all
requirements of the
adapting organization
provide an
internal selective system,
winnowing out disorganizing
the internal
structure
of
the
roots,
their round
goodness of
of
fit between
Within evolutionary
developments
stressing
homozygosity
environments,
prevalence
loci.
in
in
species
theory
the
looseness
a number
stably
adapted
to
their
heterozygosity
at all
wild
populations of
and alleles.
of
But closest to
the spirit
of
neutral genes
my emphasis on
equilibria
(Gould,
this doctrine
stable
over
environmental
1982).
emphasizes
long
time
changes
Using
that
spans
occur,
palentological evidence,
organismic
during
and
which
contrast
substantial
this
with
149
gradualist-adaptationist
model in
which each organism
closely
tracks the
moving
target
of environmentalopportuni ty.
Piaget's (1971) distinction between assimilation and
accomodation has the appropriate character. Once a child has
achieved
a
structured
schema
for
negotiating with
environmental events, it tends to hang on to it, treating
new types of experience as though they were appropriate to
the old schema, resulting in schema stability at the expense
of poor fit to environmental opportunity. Frequent and
serious errors in expectations eventually motivate a schema
change, to a new one in which more of the environmental
regularities are "assimilated" into the schema. Kuhn's
(1962) concepts of paradigm, normal science, anomaly, and
paradigm change describe a similar sequence of cognitive
revisions.
"Normal
science fl
of the
organization,
of
variation and
selection of
alternative organized
subunits,
as in Simon's (1969)
"architecture of complexity" and "semi-decomposability."
This is equivalent to a stress on hierarchical organization
with a node of selection at each level of organization
(Campbell, 1974c). This model too predicts a fit to the
environment with imperfections due to the organizational
structure.
outside
view
inside
can see an
projected on
upside-down
the far
wall)
image
was
of the
toy for
D. T. CAMPBELL
150
leading intellectuals
from da
Vinci
through Kepler,
Descartes, and Pascal at least (Crombie, 1967). And it
provided compelling awareness that any real-world event, no
matter what its true third dimensionality, that produced a
similar two-dimensional retinal display would be perceived
the same, a preoccupation that persisted in the trompe
l'oeil artistic tradition up to the introduction of the
photograph, and is revived in the Ames distorted room' and
rotating trapezoidal window (Ittleson, 1952). But still more
upsetting to the implicit assumption of clairvoyant vision
was the realization that the fluid and mode of flow of the
animal spirits in the neural tubes was basically the same
for the tactile perception of heat, cold, pain, and touch,
and by extension, for vision and hearing: that is, that the
mechanical intermediaries
representations, that all
but what
we
do
me to have
been an essential
element
in
the
preparation
of
the
conditions
for
the
appearance of evolutionary theories based upon
adaptation.
In a way,
adaptation
through
natural selection is a providentialist notion,
and the idea that men, like other beings, are
adapted
to
their
life
conditions
is a
providentialist thesis." (Harre, 1980, p. 33.)
"Hume's providentialism emerges in his
psychological theory of belief. Clearly, we
would be best constituted for practical life
were we to come to believe in the continuance of
151
manner of
its
being
conceived'
(Treatise, I,
of
"justification"
used
by
all
modern
variations.
We know of
in the
generation,
publication,
teaching,
and
believing of
scientific truth claims that are irrelevant or inimical to
improving the competent reference of beliefs that it becomes
hard to argue for a dominant role for "Nature Herself" in
the selecting. This is in contrast to the case we can make
for Her role in the biological evolution of the eye and
brain.
A reflexive use of biolog~cal evolutionary theory
provides a complementary perspect1ve. Both Cartesian and
evolutionary providentialists could plausibly say "(God)
(Natural Selection) would not have given us untrustworthy
eyes." But even if they noted that the social system of
SC1ence requires great (albeit selective) trust of fellow
scientists,
neither
the old providentialist
nor the
evolutionary epistemologist
would be apt to
find it
plausible
to argue
that
"(God)
(biological Natural
Selection) would not have given us untrustworthy fellow
scientists." If we can in fact often validly trust fellow
scientists, this is because of culturally evolved and
fragile social systems, not because of innate honesty and
152
D. T. CAMPBELL
objectivity.
Hull (1978)
has raised the issue in a
sociobiological framework, in a pioneering essay that should
have appeared in Social Studies of Science rather than in
The Journal of Animal Behaviour.
The problem is due to the fact that for us vertebrates
there
is genetic
competition
among
the cooperators
(Campbell, 1972, 1975, 1983). For the social insects whose
cooperators are almost completely sterile, one can say:
"(God) (Natural Selection) would not have given an ant
worker untrustworthy scouts." But for us social humans, the
belief assertions or public truth claims we make have
important utilities for us other than optimally guiding our"
own (and our identical twin's or clone's) behavior. We have
selfish (including nepotistic) interests in what others
believe and in what others believe we believe, often
motivating belief assertions that differ from those that
would optimally guide our OWn behavior vis-a-vis the objects
that are nominally the referents of the truth claim. In
addition to, or instead of "valid description," we have an
often conflicting interest in influencing the decisions that
our
listeners
will make.
(This
becomes
a problem
particularly when "secondary groups" rather than "primary
groups" are involved. The problem may not have been acute
for a stage in human social evolution in which inbred tribes
of 100 or so were in intense close competition with other
similar tribes, with individuals unable to change tribes
successfully.
But
it
is
acute
in
all
secondary
organizations,
including
science.)
We
must
J01n
Merton
many
philosophers
of
science
now
recognize
the
153
too simplistic.
It seems
contradicted
by,
or at
D. T.CAMPBELL
154
practice)
of
science
as
contagious
social
movement
in
the
previous
section.
Let
me
sketch some
beliefs about
by vision and
cosmological
as gravitation
touch,
and
realms,
magnetism,
such as stars
to our enterprise.
intermediate
functions.
route
to
selective
selection
("Preselection"
is
advantage
for
term
quite
often
provide the
different
used
in
155
into
different
internal
unanimities.
This
combination
of intragroup
homogeneity
and intergroup
differences would set the stage for a "group selection" of
beliefs, were some beliefs to be accompanied by greater
group efficacy than others. Such differential propogation
would typically be by cultural diffusion (e.g., selective
borrowing
of
ideologies
from
more
successful
groups,
shows
widespread
beliefs
in
rewarding
reincarnations
and/or
occurrence
ancient civilizations
Buddhism
burdened
reincarnations and
in
its
10
afterlives.
and
(Ancient
punishing
Chinese
folk
believers
with both punishing
to 16 specialized hells.) The
of
lavishly wasteful
156
D. T. CAMPBELL
to
calculate
their
net
hedonic
payoff
with
a
time-perspective extending beyond their current bodily life,
and hence act so as to further group purposes rather than
individual purposes were these to be calculated for a
biologically limited lifetime (Campbell, 1975, 1983).
Today, after four centuries of negotiation between
science and religion, we tend to relegate such beliefs to
the metaphysical or supernatural. I want instead to join a
long-standing anthropological observation
(see Shweder,
1986, pp. 171-176, for a review) and classify these beliefs
(from the point of view of the believer) as being about
invisible physical reality, i.e. as being about forces that
are causally linked to physical observables. They are thus
in the class with
atoms, etc.
gravity,
says that in
our predicament is
to trust the
great bulk of
157
the truth.
I have
Ratio"
(1978)
epitomized
(granting
this
as
the
"99/1 Trust/Doubt
revolution" in
This, I argue, is
clairvoyant knowers).
As such, the 99/1 Trust/Doubt Ratio should hold too in
biological evolution. Indeed, it does. From a physicists'
point of view, the extreme loyalty of gene duplication in
division
the
outside
is
more remarkable
ensemble
of
blueprints.)
is
meaningless
proteins
All
must
and
be
the
the
genes
"trusted ll
in
context
that
the
are
of the
their
duplication
That
mutatlon
rate
is very low,
an essential
aspect of scientific
The
evolutionary
providentialism
that
provides
fallibilist justification to visually supported beliefs in
ordinary physical objects cannot be extended to science, for
reasons of the evolutionary biology of species in which
there is genetic competition among the cooperators. Instead,
D. T. CAMPBELL
158
as functional.
Louis Boon
Rijksuniversiteit Limburg
1. PROBLEMS OF RATIONALITY
Much of the fiercest debate in philosophy of science over
the last decades has concerned the rationality of science.
Indeed it has sometimes seemed that we have two warring
cultures, rationalists and antirationalists, in philosophy
of science.
important?
The crux is that rationality is held to be the
vehicle of progress in science.
It is through the rational
application of scientific method that
we improve our
knowledge of the world.
If the rationality of science is
attacked,
the very idea of cognitive
progress seems
endangered and the specter of relativism ralses its ugly
head. This seems to be an assumption shared by both parties
to the debate.
The strong program and the ethnomethodology
of
science,
both
highly
critical
of
scientific
rationality, proudly proclaim themselves relativist.
The
philosophical landscape consequently
has become neatly
divided between the forces of darkness and the
forces
of
light: rationalists for progress and anti-rationalists for
relativism.
Is this conjunction of progress and rationality (and
its converse) necessary?
Could we develop a theory of
science in which we would not predicate a privileged
rationality for science and its practitioners and still
remain faithful to the idea of cognitive progress? The
latter can - despite the heroic rhetoric of Paul Feyerabend
hardly be denied when we take a long term or global
perspective. Despite an occasional "Kuhn ian loss", knowledge
has grown and accumulated in terms of the number of facts,
the explanatory power of theories, the precision of our
predictions, or the accuracy and scope of description. Such
a rough concept of progress seems condoned even by authors
who have been branded irrationalist and relativist (Kuhn,
1969, p.170, p.206; Bloor, 1976, p.34).
159
W. Callebaut and R. Pinxten (eds.), Evolutionary Epistemology. 159-/77.
/987 by D. Reidel Publishing Company.
160
L.BOON
Lakatos
lists
inductivism,
conventionalism,
critical
used
these
various
methodologies
to
justify their
decisions. Skillfully capitalizing on this fact, Feyerabend
has tiredlessly pointed out: In science, anything goes
(Feyerabend, 1975).
Some philosophers have argued that we should not take
what scientists say
too seriously but
should rather
concentrate on how they act (most notably Lakatos). However,
this would seem to get rationality even deeper into trouble.
The theory that should be preferred according to methodology
A, and which is in fact preferred by a given scientist,
may, as far as the scientist is concerned, be preferred
because he adheres to methodology B.
The better theory is
chosen for the wrong reasons.
What this means can be shown
In their paper on Ptolemy's and
by looking at an example.
Copernicus's theories, Lakatos and Zahar argue that there
were 'objective' grounds for Kepler and Galilei to prefer
Copernicus,
because
even
his
rough
model
of the
Commentariolus had excess predictive power over its
Ptolemeic rival, only to retract this statement immediately
in a footnote: "Note that this statement does not say
whether
and
why Kepler and Galileo
actually became
'Copernicans'" (Lakatos and Zahar, 1978, p. 188). Of course,
this footnote is necessary, because the authors know that
other reasons determined preference in this case.
By restricting attention to the actual 'choice' of
scientists,
progress
and
rationality are effectively
separated.
Implicitly, one must now fall back upon a
'cunning of
reason'
that guides blind or irrational
161
restricted
to
scientific
activities.
Too narrow
L.BOON
162
slip,
successive
theories
show correspondence
relations.
On this level we witness the progress, or the
progressive order, that is so characteristic of science:
deeper, more precise explanations of an increasing number of
facts.
The local, short term, development of science refers
to the multitude of traditions, contexts and situational
logics in which individuals or groups grope for strands of
our
immediate
future
or
environment
and to
163
L. BOON
164
a
good opportunity
for
formulating a few
requlrements of a non-rational model of the evolution of
scientific knowledge.
Progress must not
be coupled
with rationality.
Rather, the global or overall structure is the unintended
result of individual actions.
Of course this must not be
taken to mean that individuals have no goals or purposes or
that they do not apply standards.
The idea of unintended
effects refers to the fact that 'good' intentions are not
sufficient to explain the structure of knowledge on the
global
level.
At
this
level
knowledge
acquires
characteristics that were neither planned nor intended in
local communltles.
There is a discrepancy between the
'plans' of individuals (rationality) and overall effects.
Hence an evolutionary model of science must show how the
'individual' is eliminated, how his ideas are expropriated.
It is because of the social nature
of science that
pres~nts
researchers
may
work
without
consensus
over
theories,
I will
divided
such
a model,
165
species.
Each
new
variation
arises
against a
only
overturned
few
our
start in
assumptions.
most
basic
a small area
Copernicus's
conceptions
of
or concern
theory
the
which
cosmos,
Too many
we
criticize.
much more
must
always
trust
more
than
we
can
doubt or
than
progressiveness
and
that
the
166
L.BOON
some sense
Bohr was
that
successful
variations
selection environment.
new
167
themselves
variation
influence
the
among predators,
to which variations
will be subjected.
In the early days of molecular biology
scientists could orient themselves toward the established
selection environment of biochemistry, but also to selection
mechanisms of a more physicalist nature.
The latter were
more drastic and resulted in dramatic developments.
Early hypotheses on the nature of the new area are
imported from other fields, and tend to be of a rather
general nature in any case.
However, each successful
variation will make later selection more stringent. A new
interpretation will have to take earlier successes into
account, and should preferably be (slightly) innovative,
simpler,
The
nature
of new variations
L.BOON
168
generate a long
series
of
idea that
'sports',
ever
deeper
variations and to
large
discontinuous variations,
independent
could
arise.
as
discontinuous
segregation
Addition
and
or
random
variations.
omission
assortment
of
Through
new forms
factors
were
mechanisms of evolution.
With this new heuristic
he was
able to undertake a long series of experiments
aimed
at
articulating and refining his saltationist evolutionary
program.
Failure after failure was
turned into victory.
Of course with hindsight we now know that the saltationist
program failed and that Bateson's work between 1900 and 1910
contributed primarily to genetics and less to evolutionary
theory.
However, this fact makes no difference to the
historical analysis and diagnosis.
Despite the power of
the Mendelian heuristic, Bateson's research stagnated again
after a number of years, and new traditions such as
Drosophila-genetics took over. This course of events is a
quite general phenomenon.
A strong heuristic assures a program of a great
plasticity, whereas a weak heuristic quickly leads to
rigidity.
Whatever its power however, there always comes a
moment when the heuristic loses its momentum, when it has
burnt up its suggestions.
The interplay of variation and
selection hardens the cognitive network.
All heuristics
169
L. BOON
170
geneticists
were
working
with
micro-organisms"
171
in
science can
do its
been excluded.
To
mention
in
historical
works.
Variations
Nobel
fame or
are selectively
his
research
group,
internal
172
L.BOON
the
particular
field
in
question, Donohue
173
confirmationist
bias.
Only
potentially
positive
contributions, those that fit the tradition, are worthwhile
pursuing.
Results that refute or problematize a tradition start
to become interesting at higher
levels of selection.
However, because of the interconnectedness of cognitive
elements on these levels, isolated refutations or anomalies
will pot carry much weight.
The network harbors sufficient
ploys to parry such an isolated assault.
Only when
anomalies start to heap up does the course of research start
to change, and even then only when a new candidate for
further research has presented itself. Hence, on lower
levels of selection, refutations are uninteresting, while at
higher levels refutation is a long winded, cumbersome and
complicated
process.
Perhaps this
explains why the
methodological ideologies of scientists so often have a
verificationist bias, and why they so often display an
emotional revulsion
for
falsificationism
(Mulkey and
Gilbert, 1981). On the level of the research laboratory and
research group,
the characteristic forms of selection
involve 'indexical' negotiation processes on the viability
of hypotheses or findings.
The constructivism of these
negotiations is itself a variation-selection process through
which researchers are helped to filter the expected signals
from the noise.
Once the article that communicates their
his
work is written,
however,
all traces of these
negotiations are eliminated.
The same goes
for the
negotiations that take place between the
editors and
referees of journals.
Almost all of these will have
vanished from the published version.
In part scientists orient themselves in their field
through
personal
contacts.
on the
L.BOON
174
Of the
immense
mountain of articles
that are
published, the bulk is never cited or mentioned again (de
Solla Price, 1963, p.49). Only a relatively small number of
articles is used by others in further research and is cited
accordingly.
In fast-growing fields, the life span of
research - the period during which other people refer to it
is short.
The bulk of published articles disappears
behind the horizon within a couple of years.
The immense
volume of research that thus proves its redundancy, shows
that the selection performed by editors and referees is
relatively mild.
The necessary conditions for a piece of
science to be published must be lenient.
These facts
underline what has been said before on the minor role of
negative selection in science.
Only the use that others
make of published research, in their own local problem
situations,
effects
real
selection.
In
science
articles.
overviews of
problems are
necessary to
175
A few words on
journals in
as
the institution of
the
natural science is
the journal is
allow it
lacking.
to
contributions in
remains a
these fields.
also present,
function as in
Often
structured or
balanced,
selection
coherent structure.
After passing a number of
thresholds and a form of downward causation,
on journals.
Textbooks are for socialization. Students
learn the trade from them, but when they start their own
research in a local situation, they will have to unlearn
what they know to some extent.
Of course students are not the sole public at which
textbooks are aimed. They also have relevance for scientists
who are indirectly linked with a field by the necessity to
borrow some
of
its elements.
As
the logic
of evolution
176
L.BOON
presuppose a
solid consensus in
science or -
even worse -
a counterpart that
globally
dominant
theories are
science.
in
science
has the
177
Universitat Bielefeld
Department of Sociology
1.
INTRODUCTION
have
on
theoretical
succeeded
occasion
controversies
in getting
found
with
themselves
embroiled
philosophers,
them to notice
we
in
rarely
("constructivist lt )
and
microscopic
than
K. KNORR CETINA
180
That
the formula of
mutation
and
natural selection
of theoretical vocabularies
is
hardly
deniable.
It
can
apparently accommodate
successfully
Popperian
falsificationism
and
Merton's
normative functionalism (Hull, 1978), interest models (it
can be applied to Pickering, 1980), and, if I understand
Campbell (1986) correctly, hermeneutic interpretation. Of
course, this expansionist success of evolutionary theory can
to some degree be explained by the fact that the analogy is
drawn on a very general level, and it may profit from the
fact that Darwinian concepts have long filtered into lay
discourse
and
social
science
disciplines.
It
is
disconcerting nonetheless, for it seems that the more
phenomena
one covers
by
a trim
conceptually parsimonious
to
translation into
other
vocabularies, the
doing these
lSI
concrete level,
underscores recent
and
sociology of
everyone
ultimate source
who
writes
equates
of variation in biological
mutations,
the
evolution, with
conceptual innovations.
As an example, consider first Toulmin's theory of
scientific evolution, best summarized in an article he wrote
in the American Scientist in 1967.
As most evolutionary
theorists, Toulmin distinguishes between the generation and
selection
among
these
innovations.
Conceptual
variants are
the
products
of
individual scientific
innovators; their merits are collectively judged and their
fate decided by communities of specialists.
At any given
time, intrinsic (or intellectual) and extrinsic (or social)
factors jointly affect conceptual variation. For example,
social factors may limit the occasion for scientists to
speculate freely, and intellectual factors may help them to
focus their theorizing on promising lines of thought.
Selection,
on the other hand, in Toulmin's model is
182
K. KNORR CETINA
ultimately
rational.
In the common
situation where
disciplinary
problem
solving
is
based
upon agreed
disciplinary goals and standards, the selective perpetuation
of conceptual
variants
is
straightforward~y rational
(intellectual).
When scientific work
pertalns to the
"rational frontier" (Toulmin 1972, p.241) where standards of
selection are themselves up for reappraisal, scientists may
have to take "rational bets" (1972, p.246). These may not
be arrived at by accumulations of bare facts of Nature, but
they are nonetheless "objective" in the sense that they are
backed by scientists' collective experience and that they
will be judged, in the long run, in the light of their
actual practical sequels for our understanding of Nature and
not by personal considerations (1972, p.244).
with Toulmin the published but not yet accepted
products of scientific work
constitute
the
pool of
variations.
Campbell's
well-known
(1974)
theory of
scientific evolution differs from
Toulmin's in
a number of
respects.
Most noteworthy, perhaps, is his much broader
perspective on variation and selection which prompts him to
include among conceptual innovations the novel not yet tried
"everything goes"
to advocate counter-induction
modelled
resistance
aspect
upon
than
of
theories
biological
their
of
sociocultural
evolution
specification
evolution
has
caused
of
innovation
more
as
scientific or otherwise,
have been
183
discussed in
terms
of
'coupled'
versus
evolution,
Lamarckism
versus
Darwinism
'uncoupled'
and
human
conceptual
seen
and
hence
must
be
as
K. KNORR CETINA
184
knows,
concepts are
cannot
be produced at
mutations),
mutations,
which
but
perceive
they
and
are,
like biological
unlike biological
fortunate accidents, by which I mean occurrences
scientists
also
interpret
(preselect)
as
felicitous
opportunities
for
success.
Thus for an
occurrence to enter the selection processes of experimental
trial and disciplinary IIconsensus formation ll ,
it must have
react in
terms of the
coded formulation,
and thus has real consequences in
practical action. If the thought that crosses my mind comes
through as a worry, I shall do what I can to prevent it from
of variants
selection. I have
pride themselves
to be good
185
was that "they were not dumb, they just worked on the wrong
things" (Knorr Cetina, 1981, p.74). The clues one gets from
such scientists refer to their ability to discriminate,
are
unexpected
experimental
outcomes
discrimination and
selection,
and
not
of
unjustified or
"innovative"
or
as
an
"idea"
invariably
involves a
construed
as
novelties
through
editing
practices
and applaud an
earlier
original inventor in
K. KNORR CETINA
186
shown with
the "rediscoveryll
respect to
of Mendells laws.
These strategies vary, but they would all seem to imply that
novelty is the outcome of locally and historically variable
- and reversible - imputations of novelty negotiated between
participants from within and without science.
In essence
this is what attribution theory predicts and what has been
established in at least one extensive historical case study
(Brannigan, 1982).
In accord with calling the recognition
of an occurrence as
a
potential
innovation primary
imputation, we might refer to these later designations as
secondary imputations.
into
Construal as
all aspects of
novel
scientific
or
interesting penetrates
inquiry:
scientist may
her area,
frontiers of knowledge,
to give
Thus construal as
idea
is
primarily a
matter
of "seeing-asll
think
that
is
an
an editorial
process in all
understanding
of
the
three senses. I
editing
strategies
187
Mind-creativity
is
a
phenomenon
intractable depth.
But if we avoid it
scientific
discovery as
an
of
by
editorial process,
seemingly
describing
do
we not
variations,
but they
customarily
call
do
not account
for
"discovery", or
scientific "innovation".
Why not?
Let me expound the
thesis by a brief excursion into the metaphor or analogy
theory of scientific innovation.
As has been argued (Black, 1962), much of what we
scientific
investigation.
In
evolutionary
terminology, the
188
K. KNORR CETINA
as they do in
"innovative"
of
variants
and
processes
which
explain
the
If we equate
the latter with the former, we not only disregard the larger
part of the picture discovery represents in scientific
practice,
but
we also misconstrue
a
phenomenon as
exclusively mental in which mental factors may play only a
small and in any case
a non-decisive
role.
Innovations do
paper.
matter
I shall return to
this point at the end of this
But first I want to stress that it is quite another
to explain
variation than it
is to explain
189
their
interactions
between
organisms and
the ...
emergence of functional design (or adaptation) in
nature".
Phrased differently, the term natural selection
refers to a variety of factors which in an given context are
responsible
for
causing
differential
survival
and
reproduction among (genetically) variant individuals, and
for absolute changes in numbers and diversity of different
populations over time (Corning 1983, p.30-31). Recall the
textbook example
of industrial
melanism
as recently
summarized by Corning (1983, p.31).
Before the industrial
revolution, a light colored strain of the moth (Biston
betularia) predominated in the English countryside over the
darker, "melanic" form.
Apparently, when the light moth
rested on tree trunks it was almost invisible against the
then light trees, while the dark form "stood out" and was
preyed upon more frequently by birds.
With the industrial
revolution, smog from factories blackened the tree trunks in
industrial areas, and the relative visibility of the light
and the dark form was reversed.
Now the darker moth became
190
K. KNORR CETINA
rates.
It might be
behavior
of
the moth,
its
polymorphism
process
resemblances to historical
(2)
the process
is contextual
direction
by
some
law-like
principle which
accounts for
evolutionary progression
although retrospectively, of
course, certain past trends may be discerned.
To these characteristics on which there seems to exist
a broad consensus among evolutionists (see Jacob 1977;
Corning 1983,
that
191
origin
of
these
variations
in genetic
nor
natural disasters notwithstanding
, for all
environments irrespective of their inhabitants.
Now in
cultural evolution environments are not necessarily defined
by a shared locale or by physical proximity. Instead, they
might be defined in terms of frequency of interaction and
accessibility, a social structural phenomenon dependent upon
participants' awareness and access strategies. These are
not, however, to be confused with outsiders' (say social
scientists' or philosophers') similarity classifications
than natural environments must be "lived" ("transacted") in
order to be effective.
To portray them adequately within
theory, the analyst has to adopt what Edge once called "a
radically participant-centered perspective".
want to
192
K. KNORR CETINA
educational
structures
but
have
little
to
do
with the
nothing
intrinsically
to
do
though
it
may
at times
coincide
with
scientists' official disciplinary or
professional self-identification. In fact, what constitutes
these environments is entirely an empirical question. It
cannot be
(e.g.
decided by
disciplinary)
in everyday life.
simply adhering to
the institutional
specialty
about the
communities which
have
dominated
our thinking
of science,
and not
only since
Kuhn (1962)?
The problem with
specialties, I think, is that like scientific disciplines
they
are
not
properly
speaking
"organizations",
goal-oriented
co-operative
little
systems,
controls (5).
by virtue of
systematic ("emic"!(6
characteristics such as organized
action in concert or political superstructures. Instead,
"scenic"
perspective,
specialty members'
local
dealings
appears
like the
integrated cosmos of a nature painting.
The design of the
various
parts
of
the
human
body
is
never
for
long
193
networks"
or
"research
areas"
in
terms
of
which
some
to scientific disciplines,
or in any
case to the
interpretation
of
conceptual
selection.
enterprises"
(which
conceptual selections
have
denied),
are primarily or
he
presumes
that
194
K. KNORR CETINA
But
where would
these
research decisions
take place if
not in the
laboratory, that is in the very context of scientific
discovery?
What is consensus formation if not a proces of
selective affirmation of knowledge claims in and through
further research?
Surely it is not a process of opinion
formation disengaged from actual laboratory decisions. And
even if such a process existed, it would count for little,
for it is quite clear that disengaged opinions have an
oblique, if any, relationship to action. What counts with
respect to the final survival of a knowledge claim is
plainly scientific practice, and this takes place at the
or
"weeded
out",
then
factors
which
rule out
internal
wholesale
specialty
considerations.
relegation
of. selection
environments
and
195
processes to
"disciplinary"
only
"working"
relationships of interaction
and influence
196
K. KNORR CETINA
as a process
evolution. By specifying
in which situation-specific
What
sense as natural
selection
scientific
factors
product
over
matter.
in
another
selection is contextual:
the
cannot
choice
be
of
one
concretely
possible
from
contemporary
analysis of
networks
of related
working contexts (transscientific
fields). It might also be noted that from the point of view
of scientific work, contextually contingent operations are
opportunistic in the sense that scientists do not only take
account of
to
'contextual' pressures.
participation,
create
scientists
these
environments with
opportunities
as
they
present
themselves
in
between
selection
in
biological
197
the disanalogy
between
natural
and "rational"
conditions".
Natural
selection
"is
at
best
Hull seems to
authors in thinking
to
actions.
perhaps
globally maximizing
selections.
198
K. KNORR CETINA
will agree,
states
it
and even
is
difficult to
maintain
that societies,
as universities change
in
the process to
sununary
record.
As an
199
normally
of
selective
retention
needs
variations,
but
does
not
replication and
much more
components of technical work
interested in
varying the
in order to adjust to local
at the same
are either
time
the process in
produced is
out", whereas
Finally, if,
K. KNORR CETINA
200
as
have
argued,
understood as
process of
"discoveryll
a process
of
in
science
is
selective
editing
than
novelty generation,
the consequences
editing
of
this
In
short,
my
argument
say with
variations,
phenomenon
perception of variations.
theory needs to
makes
as a
no differences
with respect to a theory of scientific variation, whereas it
does deal a further blow,
I think, to a theory of
mutation-like scientific innovation. Variation theory might
simply include the point that designated novelties result
from various processes of
this
better
is
regard
for the
that
and explore
status and
whatever selection
to scientific 'mutations'
evolutionary theory as
and
sociological
understanding of
NOTES
1. There are exceptions of course. See for example Hesse
(1980), Nickles (1980), and Giere (1984).
2. Although it is not the first attempt to develop an
EE. Don Campbell has made me aware of the fact that Simmel,
as early as 1895, wrote a paper on the relationship between
selection theory and epistemology (Simmel, 1895).
3.
recognitionll,
use of
the
term "primary
of
a cybernetic
system, see
8.
201
Hull (1982).
1.
and many
similar remarks,
thing.
In
Lorenz
fact,
and Popper
they
have
of
seem to talk
taken
their
many
views on
or
in
various
complex
consists in establishing
systems?
And
isomorphisms,
if
understanding
203
W Cal!ehaur and R. Pmxren (eds.). EvolutIOnary epistemologv. 203-221
1987 hy D. Reidel PubfishinKCompany
204
G. VOLLMER
evolutionary
interdiscipline
cultural
that
mediates
evolution?
Why
not
between
call
biological
it
EE?
and
seminal contribution
Thus,
In
his
essay
'Evolutionary
Epistemology',
Donald T.
Campbell says" ... the natural
selection paradigm of such knowledge increments
can
be
generalized
activities,
such as
science".
contention,
but
only
agree
with this
it as one of the main
tasks of this book to draw the generalizing
comparison, proposed by Campbell, between the
different mechanisms by which different living
systems
acquire
not
to
other
epistemic
learning, thought and
regard
and
store
the
information
205
and its
evolution,
with the processes of cognition.
Popper, on the other hand, being a philosopher occasionally
drawing on biology,
is studying scientific knowledge,
abstract theories, the results of cognitive processes and
the evolution of science.
In order to signalize this
distinction
terminologically,
'evolutionary epistemology'
we
shall
use
the
term
True,
both
parties
like
to
stress
the
analogies
The
supporting
They
are
read,
cited
and
criticized as
is essentially Darwinian.
legitimate to look
206
G. VOLLMER
extremely interesting.
In a sense, the search for such
analogies has been quite successful.
There is, indeed, a
Both theories
- are theories of knowledge
- center on the growth of knowledge
information in time
- find substantial continuity between animal
and human knowledge
- stress the evolutionary character of the growth
of knowledge
refer to
knowledge
- may be subsumed under Campbell's conception of
is another matter)
1.
Common
traits
ln
Lorenz'
EE
and
Popper's
207
3.
Finding
structures
in
different
systems
is
an
There is,
minimal description.
by definition,
no redundancy in a
G.VOLLMER
208
every scientist,
teacher?
Are not scientific instruments made and
enable more and better distinctions ?
What is more,
used to
quite generally, a
measure of quality.
This is true for any instrument, be it
a conceptual one like a terminology, a theory or a research
program, or a material one like a telescope, a microscope or
abilities
traits
such as sense
organs, perceptional
and other cogn1t1ve structures.
We might even
both cases
advances.
Thus,
increases
in
resolving
identification
complementary aspects of
and
progress,
power
are
seen
discrimination
both in
as
are
science and in
evolution.
It would be quite one-sided to restrict the
concept of progress to one of them (2). Heuristically, it is
a better strategy to keep the situation symmetric. If with
respect to a special subject the differences have been
stressed long enough it might be worthwhile to look instead
for common traits. And if for a couple of objects analogies
and common traits have been widely discussed,
it may pay to
209
evolutionary
epistemology
evolutionary
philosophy of science
proponents
dealing with
evolution of cognitive
systems and abilities,
cognitive processes
evolution of knowledge
(mainly scientific),
cognitive results
"evolution" under-
organic evolution
cultural evolution
stood as part of
concept of evolution
quite specific
qui te general
correlation to
strong
weak, metaphorical,
organic evolution
relevant time scale
(essential identical)
millions of years
tentative, heuristic
tens of years
biology (genetics,
theory of evolution,
history of science
and technology
ethology J neuroscience) J
psychology J linguist ies
philosophical
discipline
(in particular)
scope
(cognitive levels)
reference
(Objects of
"selection")
regula t i ve idea
or property
approximated by
relation
ideal (f ic t it ious)
final state
... not reached
due to
epistemology
philosophy of science
("cognitology")
perceptual and experiential (mesocosmie) knowledge
all cognitive systems
(amoeba to men, Martians,
possibly even machines
("theory dynamics")
theoretical (or scientifie knowledge
hypotheses, theories
(let theories die instead
of their adherents)
fitness
truth
adaptation
"convergence" (Bavink)
"ver isimili tudell(Popper)
"partial truth ll (Bunge)
truth provides fitness
G.VOLLMER
210
never regained
non-Darwinian
(al1-or-none decisions)
(forgotten theory)
may be formulated anew
processes
unconscious
opportunistic
conscious
critical
variations
information
aimless, blind
playful
copying errors
by genetic inheritance
to own descendants
progress
is an inevitable, but
induced by
transmission of
unintended by-product
of evolutionary processes
systematic
problems
by publication
proved
innovations
rate of change
constraints on
innovative trials
character of
constraints
learning strategy
mind-body problem
saltatory,
sometimes radical
"revolutionary"
few
mainly logical
(no contradictions, e.g.)
but also epistemological:
formulizabili ty in a
finite, recursive, intersubjective, argumentative
languagej projectability
to our physical periphery
211
Let us
content
ourselves
with
called
'evolutionary
epistemology',
not one.
as we do,
had nothing in
and need
not be repeated.
Our main point is that there are such
decisive differences and that the term 'EE' being used for
both approaches, may obscure them.
What is more, the
ambiguity
of our
term has
in fact
led
to grave
misunderstandings and,
inevitably, to unnecessary attacks
directed against EE
as
a whole.
The
following sections,
some
IS IT DARWINIAN?
Is the
reveals
with.
That
there is
change
in science
This question
quite familiar
(and in history)
G.VOLLMER
212
ones?
Is evolution characterized by its (low)
pace, by
the gradualness of its steps, by its continuity? Are the
concepts of evolution and of revolution mutually exclusive,
or. do they overlap?
Perhaps, revolution is nothing but
rapid evolution, hence evolution after all?
Are there revolutions in science?
And what is normal
in science : stasis, or evolution, or even revolution? Does
Kuhn's distinction of normal science
(dominated by a
paradigm) as against revolutionary science (characterized by
a paradigm shift) really make sense? (Kuhn, 1962; Toulmin,
1970)
We shall not try to answer these questions here. The
reason is that EE and evolutionary philosophy of science
do not differ in this respect.
Whatever their differences:
both of them view the growth of science as evolutionary
(as was indicated in Table 1).
However, the concept of
evolution used here is quite general.
It is, in fact,
universal: all real systems evolve.
But by no means do
they evolve according to the same laws.
In particular,
they need not obey Darwinian principles.
And this leads
back to our main question: is the evolution of science
Darwinian?
Again,
we
are
confronted
with
an
intricate
terminological problem: when is an evolutionary process
Darwinian, and when is a theory describing it a Darwinian
theory?
Are Darwin's personal preferences relevant? Are
they decisive?
But Darwin of all people believed (herein
following Lamarck)
that acquired
characters could be
inherited! Was even Darwin not a Darwinian after all? Which
principles are constitutive for a Darwinian theory of
organic evolution,
and which are peripheral?
Is the
principle of natural selection sufficient to characterize
Darwinism, or is it just a necessary ingredient amongst
others?
How much may we add to Darwin's theory without
distorting it, and, even more exciting, how much of it may
we sacrifice without destroying it?
And again, we shall not try to answer all these
questions. In particular, we shall not give a complete
characterization
of
Darwinism.
We
shall
rather content
213
selection'),
These
science
despite
traits
are
essential
and
necessary
all
similarities,
conceptions,
to
these
transmission,
analogies
features
and
to less
variation,
and
But are
they
Darwinian?
Being
generalizations of Darwinian
principles, they cannot contradict the latter. But being so
general, they are, at the same time, much too weak to
account for organic evolution.
This has been shown by
several authors
criticizing
Darwinian
models of the
evolution of science (4).
It turns out that the common ground
between a
Darwinian-type theory on the evolution of cognitive systems
(EE in Lorenz' sense) and an evolutionary but non-Darwinian,
philosophy of science (in Popper's sense)
is
quite
limited.
In other words, the greatest common denominator
of Lorenz' and Popper's approaches is, though existing,
forbiddingly small.
Thus, the term 'evolutionary epistemology', as it is
used now,
G.VOLLMER
214
systems.
and
EE
even
Darwinian
approach
to
cognitive
an organism's 'innate
cognitive structures,
are correct, whether they are adequate not only to cope with
reality but to build a (partially isomorphic) internal
reconstruction of the world, is not to be judged from its
mere survival.
There is a much
the adequacy of
215
or adequacy
of
cognitive
structures
correctness
(by its
G.VOLLMER
216
we
trust a
theory
which
has
been severely
tested and has survived the test unscathed. And the same
kind of argument is used by EE when evolutionary success
under competition is used as an indicator for truth.
7.
217
This
characterization
of
the
relation between
knowledge and theories of knowledge is adequate under
historical
as
well
as
under
systematic
aspects.
Epistemological efforts have developed in a virtuous circle
with
science,
and epistemological
arguments
may be
reconstructed
in
like
manner.
As
Einstein
says:
mesocosmic
knowledge,
and
pre-scientific
that
is
experience,
perception,
household
will
neither
support
nor
refute
scientific
theories,
and
not
the
theories of physics
which
explicitly
contradict
those
intuitions.
Inconsistencies
between
intuition and science suggest that such roots
don't even
exist:
If
there are,
indeed, biological roots
for
G.VOLLMER
218
in
abstraction
general.
and
Memory
and
generalization,
learning,
creation
curiosity,
and
use
of
and,
constitutive
for
science.
is
still
another
respect
in
which
EE does
early
grasp of
objects,
events,
probabilities,
etc.,
than our more recent scientific
theories could ever hope or claim to (Tversky & Kahnemann,
219
NOTES
1. The minimality of a description can never be proven
definitely, as Gregory Chaitin has shown, using arguments of
Kurt Gadel and Alan Turing (cf. Chaitin, 1975).
2. Thus, if Riedl and Kaspar endow all organisms with
a ratiomorphous cognitive principle called "hypothesis of
identification" and characterized as "the expectation that
dissimilarities may be ignored in similarities, and that
similarities will prove similar even in what has not been
perceived" (see Riedl, 1984b; Kaspar, 1984), then this
assertion
must
be
supplemented
by
a corresponding
"hypothesis
of discrimination",
characterized
as the
expectation that objects known to be dissimilar will prove
dissimilar in even more aspects.
in other
contexts focussing,
of losing
with the
under the
carefully
220
G.VOLLMER
not accessible
to
us,
we shall
have
to
turn,
as
biologists
had,
to
ontogenesis." (Piaget, 1972).
From this, it is evident that Piaget would have welcomed an
evolutionary account of human cognition as aspired by EE.
But it is also evident and, besides,
well-known that
Piaget's work is ontogenetically oriented and
that his
'genetic' epistemology refers to the genesis
of knowledge
in the individual. So much so that Hans "G. Furth, one of
his disciples and interpreters, in the preface to his book
on Piaget, makes the following telling remark:
"I am unable to see how one can feel at home
with his Piaget's model of intelligence unless
one sees intelligence as a prolongation of
organic development. Without a biological basis,
Piaget's formal logical model becomes what to
many
it
unfortunately
appears:
a
cold,
artificial system of ratiocination that has no
relevance to full-blooded real life." (Furth,
1969, p. XVII, preface).
221
ways,
for
example
in
debates
concerning
innate
that
According
to
excellent contributions of
this
terminological
Louis Boon,
choice,
the
not dealing
1. INTRODUCTION
What kind of an Epistemology is an EE?
In what sense, if
any,
can such an epistemology
provide
'a dialectic
resolution to many old controversies' (Campbell, 1974a)? To
answer such questions is to begin the somewhat neglected
task of placing evolutionary epistemology in its proper
philosophical niche. The task is an important one since, as
we shall see, much recent criticism and distrust of the new
epistemology is bred by confusion
over
its intended
significance.
The naturalised approach, I shall argue,
should in no way be promoted as a means of resolving old
controversies.
For
are bracketed
questions
it poses.
are reviewed
more suitable,
non-foundationalist
fund of philosophical
the new setting.
niche and
sketches the
addressed in
A.J.CLARK
224
our experiences,
waking and sleeping alike. How, that is, can we know we are
not being deceived by some evil demon or, to adopt a more
modern formulation, how do we know we are not just 'brains
in a vat' (Putnam, 1981) in the laboratory of some alien
scientist?
Such questions go to make up what Barry Stroud
has called the philosophical problem of our knowledge of
the external world (Stroud, 1984). Standard epistemology,
as Richard Rorty has insisted at length (Rorty, 1980), thus
grew up in part as an attempt to resolve this philosophical
problem by formulating a theory of knowledge which secured
human wisdom against the sceptical challenge.
In this
environment
the rationalists
maintained
Hume's
'simple
impression'
and
we know
ourselves aright
as biological organism,
is, from
225
A.J.CLARK
226
contemporary opposition
precisely
encourage.
the
kind
of
to EE can be seen
misconception
to
which
spring from
such remarks
to the truth'.
The evolutionary epistemologist, Rorty
believes, is a1m1ng to 'de-transcendentalise epistemology
while nevertheless making it do what we had always hoped it
might' (Rorty,1980
p. 299). And what we always hoped it
might do was to justify our knowledge in some independent,
objective way. Yet the naturalised claims, as Rorty rightly
points out, are themselves as dependent on sUbjective social
3. PHILOSOPHICAL QUESTIONS
What
kind of
foundationalist
Callebaut
and
epistemological
work remains
once the
project
is
abandoned
or
bracketed?
Pinxten
claim that "EE can
offer a
227
have
some)
could
have
come
about;
i.e.,
in
Quine's
which,
if
not discontinuous
with science, is at
228
A.J.CLARK
philosophical concern.
For philosophers expect their best
arguments and conjectures to be in some sense independent of
particular experimental results or empirical considerations.
lies
in
its
providing
basis
of
scientific
229
much
of
an evolved
framework of a
is such a
(4)
The question
of
whether
the evolutionary
epistemologist, by picturing us as knowing the world only
indirectly
and
'presumptively'
by means
of evolved
strategies (see Campbell, 1974a, p. 418), must therefore
recapitulate the much-criticised Kantian divide between
Appearances
and
the
world-in-itself.
Is independent
reality, for the evolutionary theorist, necessarily an
indescribable 'something = X' hidden forever behind the veil
of perception and cognition (O'Hear, 1984)?
Taken together, such questions amount to the threat of
a new scepticism.
Far from filling in for Descartes' God
the problem is that evolutionary epistemology may unveil
Cartesian demons in nature herself.
Demons which block out
direct and unbiased contact with reality and which further
deceive
us
with
species-specific
priori
conceptual
frameworks.
The new scepticism thus uses one part of our
knowledge (viz. evolutionary theory) to question the status
of the rest.
Whether such a scepticism can be consistently
maintained, and if not, how the evolutionary arguments are
to be otherwise accommodated, provide further areas for
useful philosophical
discussion.
questions could be further expanded
230
A.]. CLARK
want
to
say,
as
a breeding
NOTES
1. Here and throughout the paper I use the term 'EE' to
refer
to
what
Donald
Campbell
calls
'biological
Evolutionary Epistemology' and Gerhard Vollmer calls EE (as
opposed to 'EE'). I therefore use it to refer
to the
account of knowledge consequent upon the identification of
human cognitive systems as, in part at least, the products
of actual biological evolution.
I do not use it to refer
to the metaphorical usage ('EE') of a natural selection
model of
theories.
the
generation
and
retention
of
scientific
231
1. INTRODUCTION
Modern technology is a pinnacle of human progress, the
perfection of reason mirrored in design.
It maintains that
bit of Enlightenment hubris captured in the term Homo
sapiens.
Why,
then, does it seem easier to design
technology than to make decisions about how to use it? That
technologies can puzzle or disappoint or seem out of
control, is explained - by their opponents - as temporary
yet resoluble oversights of rational solutions, or - by
their supporters
as the failure to demonstrate to all
concerned the reasons that the benefits outweigh the costs.
(In spite of the acronym MAD, nuclear armaments debates are
conducted in this vein).
Alternatively, some of our
technologies may puzzle, disappoint, or seem out of control
because the model of human nature is inappropriate. Despite
economic,
hedonic,
and
nonmative
justifications,
the
L. R. CAPORAEL
234
235
gene:a!ive
rules,
prototype
formation,
microbehaviors,
sets
up
her
brother's
murder,
and
murders her
236
L. R. CAPORAEL
237
mechanisms
integrating
instinctive
and
learned
behavior
mechanisms.
Ontogenetically,
then,
these
L. R. CAPORAEL
238
attachment.
I emphasize small in this discussion because
the predominant part of human evolutionary history has been
written in hunter-gatherer sized groups.
3. TECHNOLOGY IN HUMAN EVOLUTION
What are the effects of technology on human biological
evolution?
For many years, lithic technology (usually
associated with hunting behavior) was believed to be a
'prime
mover'
in
human
evolution.
This
view
has
been
239
Norms, cultural
strategies for
interacting among
groups,
are
developed
largely by
exploiting the lability of group membership and redefining
it along different dimensions (cf.
Brewer, 1981; Brewer &
Campbell, 1976; Tajfel, 1981), important functions of myth,
ritual, and moral preachings. For the better part of human
history,
however,
between-group
association has been
temporary, limited not by the cognitive capabilities for
forming larger (albeit often unstable) groups, but by the
lack of technologies to more effectively exploit the habitat
to support a higher density of individuals over extended
time periods.
Given extremely favorable habitat conditions
and the development of technology for more effectively
exploiting the ecology, the flexibility for maintaining and
redefining group membership lent itself to patching together
primary
groups
into
supraordinate
social
systems-
ultrasociali ty.
Humans did not biologically adapt to higher density
units created through technology because there was no (or
insufficient) selective pressure to do so.
Sociality and
the intelligence it gave rise to presented such a tightly
integrated biological organization (i.e., canalization) that
by the time high residential density was viable, suitable
biological mechanisms already existed for reorganizing at a
supraordinate
level.
Fundamentally,
the
'natural
environment' of humans did not change - nor has it changed
yet.
It was maintained by psychological and cultural
factors.
The sociality mechanisms, which constitute the
capacity for culture, 'parse' the stimulus information of
the social environment into small group 'grammar'. For the
practical exigencies of daily living, individuals still
organize themselves along dimensions relevant to small
groups
for example, colleagues, friends, relatives, and
strangers
with various levels of intimacy, strengths of
coalitions, and expectations for reciprocation.
Thus far I have proposed that the role of technology
in human evolution has been to facilitate the functioning of
existing morphological and social adaptations. Technology
does alter the gene pool by allowing forms to reproduce that
otherwise might not (e.g., near-sighted individuals), but
there is
no systematic selection
effect because the
distribution of any single technology across habitats is
variable. Dobzhansky (1962) discusses some cases where
240
L. R. CAPORAEL
241
hope
an example
identification is
may
prove illustrative.
Individual
species-specific characteristic, and
develop
the ability to
remember
and
recognize
faces, an
interaction between
biology and experience.
But the
evaluation of and preference for faces having 'handsome'
features is learned in the social group. With individual
experience we may incorporate the group's evaluation and
preference, reject it and influence a shift in the ideal,
remain marginal with respect to the group in our preferenc~
or even find a new group.
We
can
also
use
the
face
in technological
applications, as when Chernoff (1971) faces are generated by
computer
and used
to
present
complex
multivariate data.
L. R. CAPORAEL
242
and
increase
the
comfort
associated with
whites
of
our enemies'
eyes,
and
The
perception
of
243
technology
as out
of control or
mechanisms
for sociality -
it may continually
alter their
epistemology -,
believers
in
processes that
demonstrate the
of
boundedness
humans.
The
of rationality
intruding
on
or
disrupting
the normative
rationality by
does
not
conform
to
normative
theory;
rather,
people
however, to
The rational
methods discovered by
244
L. R. CAPORAEL
improving our
improving our
understanding of ourselves.
NOTE
1.
of a mind-extensive
technologies
do
technology.
limit
Of
course, body-extensive
mind-extensive
technologies
somewhat shape
ideational content.
But
function of mind-extensive technology is not
and
the dominant
to extend the
and mind-extensive
Partm:
The Piagetian approach
in psychology,
which is more precise than
epistemology
or
knowledge"
(Nivnik,
in:
Silverman, 1980, p.1S).
If this were the case, Piaget's genetic epistemology (GE
henceforth) could be viewed as a kind of philosophical
wisdom developed by a very special philosopher, who valued
science greatly and needed hard facts in order to help
himself philosophize.
But Piaget's system is much wider
than just the extension of a psychological theory. In order
to understand how wide-ranging his project was, we have to
go back to his 1918 novel Recherche, where he expounded for
the first time a model that will appear again and again in
later works.
of
the sciences
c. GILLIERON
248
Mathematics
Psychology
~ Biology
~YSiCS
.-------
the consequence of
a double
evolution
in
science, or
rather of an
oscillation:
"Scientific
thought ( ... )
oscillates between two poles; the mind explains physical
reality through mathematics; but physical reality explains
the mind and mathematics through biology" (Piaget, 1929,
p.147). Scientific evolution corresponds, from one point of
view, to a progressive mathematization of reality, and from
another, to self-explanation through biological laws (lato
sensu).
It must be noted that this complementarity goes
along
with an opposition between
two epistemological
attitudes:
IIBetween
the
two
poles
of
mathematics
and
in
mathematics,
as well as in
the problem
of the
249
the circle.
This is a difficult point if one wants to
escape the criticism of psychologism - which Piaget did not,
according to some epistemologists - as well as that of
logicism.
Piaget's
facts.
But
here
i.e.,
we have to
based on
Several relationships
analogically and need
are
still
a careful
the
specific
Gillieron,
in prep.).
classification
of
In addition,
implicit or treated
examination (Borel &
essential for it to
be a 'circle',
at least topologically,
the
same.
These
invariants are
A.
c.
interdisciplinary
connections
show
circular
C. GILLIERON
250
Epistemology of
psychology (often)
borrowed from biology
,.
-------
ipis-t:~orOg; of
biology -:
(ghostly), - - -,'
--
"
(see text).
A,
251
the interaction of
the organism
C. GILLIERON
252
sciences
concepts and
253
C. GILLIERON
254
of
and
even
anticipate
their
own
applicability.
If
mathematical
structures
'express'
the
object,
the
coordination which leads to mathematical structures must be
a fundamental law of the object.
If we exclude any aprioristic
explanation, this
harmony must result from the psycho-bio-physical nature of
genesis.
Logico-mathematical structures emerge out of a
real process, they are deductive constructions by concrete
subjects, i.e., organisms.
The ties are internalto the
subject:
mathematics
are
the
result
of reflective
abstractions starting from the opera tory structures of the
subject which depend upon the coordination of his actions,
which depend upon the nervous connections, which depend'upon
the organic
regression.
structures
Meanwhile,
and
lilt is
so
this
255
on,
in
a perpetual
one
must
insist
mathematization.
i.e.,
The first
evident,
'objective'
fact
unquestionable,
on
the
respect to
for
epistemological
this problem of
the seven-year-old,
C. GILLIERON
256
does
not
belong
to
the
class
transformations.
Meanwhile, it
surface area; how does one decide?
of relevant-for-substance
is
relevant
for,
say,
mathematizable.
Meanwhile,
unlike the physicist, the
logician-child does not question
this mathematization;
structures are autonomized before any physicalization. It is
only progressively that she realizes that this application
must
be
justified
theoretically.
The psychogenetic
progression from the logician to the geometrist and the
physicist mirrors the anticipatory nature of mathematics.
4. GENETIC EPISTEMOLOGY, PSYCHOLOGY AND BIOLOGY
GE,
like
any
other
'derived'
epistemology,
is
interdisciplinary. It has been shown that far from being an
appendix of psychogenetic theories, one of its central
theses comes from the image of the closing circle of
sciences, mainly due to the internal epistemology of logic
and mathematics.
of
mathematical thought
is a
major epistemic fact;
neglecting
it,
any
relevant
epistemology
would be
impossible" (1984, pp.
54-55).
Meanwhile, if the links
between logic and psychology are fundamental
for the
epistemologist, one cannot underestimate the 'organic' link
between biology and psychology.
Analyzing this link is not" as easy as it may seem.
The existence of two factions in psychology, the scientists
('serious'
psychologists)
and
the
humanists ('soft'
mentalists) is a living picture of the fundamental problem
in psychology, as summarized by Greco (1967, p.937): "It is
the misfortune of the psychologist: he never is sure that he
is 'making science'.
If he does, he is never sure that it
is psychology".
To the difficulty of delineating the
257
discipline,
whose
constructs
an
'organism'
may
serve the
not
to
relate
and
an
'environment',
but rather to point to the
unequivocal trace in the environment of the
structures that make it integrated into the
organized systems" (Meyer, 1967, pp. 785-786).
C. GILLIERON
258
living
system
organism is at
the same
time a physico-chemical
It is the
locus and
the
source of
behavior and mind.
That
is
why
"psychology is first a biology" (Piaget, 1968, p.119), but
a biology not restricted
to the study
of
the material
aspects of the
living beings;
preoccupied,
on
the
contrary,
with the organizing role of subjectivity.
This antireductionist identification of the material
domains of psychology and biology may explain the solitary
position of Piaget with respect to the classical dichotomy
of monism and dualism.
chains
at
parallelism
the
conscious
level
are parallel.
Such
is on a par with that of mathematics and
ones
of
asymmetrical
dependence,
but
of
mutual
determination.
GE,
problems -
from
double point
organization.
of
morphogenesis
to
evolution
from this
view of
physico-chemical and
For
indeed,
at every
level,
psychic
organlc
regulations refer, in their functioning, to those components
of
subjectivity
that
significations
are.
Only
significations may
direct
behaviors,
from
the most
elementary to the most evolved. Biological genesis presents
produced by the
minds become
discovered,
'being'
whose
origin
is
of structures is their
more and
more remote.
structuring"
"The
(Piaget, 1968,
p.120).
The
organism
is
the
paradigm
structure,
consequently, and "if we knew our own organism through and
through it would, on account of its double role of complex
physical object and originator of behavior, give us the key
to a
general
theory of
structures"
(ibid, p.40).
(Unhappily, biology is far from that point ... )
Coming back to the circle of sciences, which the
provisionally distinguishes the four groups of disciplines
(the Popperian worlds 1,2,3, & logic?) we may now view it
259
are
correlative.
Meanwhile,
these
terms are
H"
>".
0
H"
"
"....
TERTIUM QUID
(NEO)DARWINISM
OPERATORY
TRIAL/ERROR
CONVENTIONALISM/
PRAGMATISM
CONSTRUCTIVISM
EMPIRICISM
PHENOMENOLOGY
PLATONIC IAN
REALISM
APRIORISM
EPISTEMOGOLY
Theories of
knowledge
".,'"'
H"
.,z
rt
P-
"ro
"
"ro
.,",....
...,
.,...
structures . . . .
Construction ~.
Empirical
data
Ready-made
Interaction
Object
primacy
Subject
primacy
Both
Internal
ASSOCIATIONISM
GESTALT
EMERGENCE
Both
(NEO) LAMARCKISM
"DENKPSYCHOLOGY"
"FACULTIES"
PREFORMISM
CREATIONISM
PSYCHOLOGY
Theories of
intelligence
Internal
External
,.... External
'"<
H"
>".
>".
'"X
Factors
BIOLOGY
Theories of
adaptation
E
"',
Cl
(")
261
Lamarckism,
associationism and
empiri~ism,
c. GILLIERON
262
novel
behaviors),
Hence
nea-Darwinian
263
epistemologies
(internal
and
derived)
are
self-contradictory!
A true nativist like Fodor, who comes from
'formal' disciplines, may well object that the structures
of knowledge
are innate,
while contents depend upon
individual experience.
What is known necessarily reflects
the potential structures, since everything that is, was
possible.
However, according to Piaget, it remains to be
explained why any specific possible becomes real. The core
of the disagreement between Piaget and nativists rests upon
the notion of 'potentiality'. No structure exists without a
content,
and the
actualization of a
structure comes
from
by considering
Piaget,
evolution
hut behaviors as
internal,
hereditary
(mutations,
by
definition,
are imperfections of the
mechanism of conservation). Such a conception of the genome
seems paradoxical:
evolutionary
'strategies'
mechanisms
and
its brilliant
or reorganizing capacities are the
c. GILLIERON
264
selectionism,
adaptation
is,
in
a way,
akin to
an
immunological
process against
external perturbations ( ... ).
It is all the
more clear in the linguistic theory that Chomsky
has been professing for nigh on 30 years. The
child's learning of his or her mother tongue
appears as a progressive pruning of the inner
tree of universal language, which itself comes
from a central kernel which is innate and
existed before all experience. Faced with such a
frozen landscape, we can only wonder how the
freezing occurred ( ... )" (op. cit. p.229).
The structures without content of selectionist theories are
hypostatized and 'frozen', as is 'genetic information' when
one ignores the psychological face of epigenesis.
From the
constructivist
point of
view,
remains
role
as
experiencer' 5
structures,
i.e.,
aware
of
originator
of
the
all
of
experiential
particles"
(von
265
formalizing,
disciplines,
266
C. GILLIIJRON
"L'union de
la mathematique et
de l'experience, c'est
embarrass 'a
necessitarian progressivist
like
Piaget', we
of creative scientific
of
scientific
progress.
clue
to deeper
development
namely,
that
pre-scientific
psychologists,
II
sc ientific
thought"
thought
is
(Einstein,
1934).
of
Some
reached a similar
conclusion.
A fundamental thesis of Piaget's genetic
epistemology is that the evolving structures in scientific
thought exhibit parallelisms in structures in the formative
psychological processes in children. Inspired in large part
by Piaget's writings, I have approached these fundamental
problems by using results from case studies in the history
of science as data for theories of cognitive psychology.
Conversely, this method tests the claims of the cognitive
theories themselves concerning the construction of knowledge
and creativity (1).
This essay I shall focuses on the part
of my research in which results of historical analysis of
atomic physics during 1913-1927 are taken as data for
genetic epistemology.
We will find parallels between
the hierarchical
structures that are the varlOUS formulations of atomic
physics
during 1913-1927
and
structures
in genetic
epistemology.
And we will find that decisive scientific
progress is signaled by establishment of the following
structures that Piaget found to be essential
in the
construction of pre-scientific knowledge
namely, the
permanent object (i.e., the Bohr atom), reversibility (i.e.,
acceptance of the wave-particle duality
of light and
267
W. Cal/chalit and R. Pinxten (cds.), Hvo/Ulionary Epllfemo{ogv, 267-281
/987 hy D. Reidel PuhlishinJ.[Company.
268
A. I. MILLER
269
trying to
We
this
case
study cannot
be removed
to
a
hierarchy of equilibrated
structures that are
increasingly better approximations to physical reality.
Each level is richer than the one below.
Piaget divides the
development of representative
activity into
three
stages:
sensorimotor, egocentric
representative,
and
operational.
At the end of the
sensorimotor period there are fleeting pictures of objects
from
the
world
of
sense
perceptions.
The
egocentric
representative
stage
is
divided
into
two
parts:
preconceptual thought and intuitive thought.
The onset of
preconceptual thought
is the
construction
of object
permanency which signals the start of semiotic functions.
The dependence on imagery during pre conceptual thought
decreases during the transition to intuitive thought. In
270
A. I. MILLER
which
the system
271
is unstable.
The reason is that the revolving planetary
electrons radiate away energy and eventually spiral into the
nucleus. But most atoms are stable, else we would not be
sitting here!
In 1913 Bohr incorporated atomic stability into a
theory in which by axiom there is a lowest orbit beyond
which a planetary electron cannot fall.
Another axiom is
that only certain orbits are permitted.
While in one of
these allowed orbits, the electron does not radiate any
energy.
Bohr referred to the allowed orbits metaphorically
as "waiting places" (Bohr, 1913). The planetary electron's
downward transition between allowed orbits results in the
emission of light.
Although the
drastically altered
classical mechanics could still be used to calculate allowed
orbits, it could not trace the electron's transition.
Consequently, in transit the electron is unvisualizable; it
disappears and appears like the smile of the Cheshire cat.
We may describe these developments
with genetic
epistemology as follows (see levels 1 and 2 in the figure).
Rutherford assimilated the scheme of classical mechanics to
his 1911 data with no attempt at adjustment. In 1913 Bohr
adjusted or accommodated the disequilibrated version of
atomic physics with its unstable atom to a new equilibrated
structure, that is, the 1913 Bohr theory. Central to the
1913 Bohr theory is the permanence of the solar system
atom.
Consequently,
at this
juncture
Bohr's theory
displays representative act1v1ty analogous to egocentric
representation.
Rutherford's 1911 conception of the atom
(the product of assimilation) is the signified. Bohr's
1913 atom (the product of accommodation) is the signifier.
The signifier or image represents the atom, but neither is
it the atom itself nor is the signifier the image of the
scheme
in
Bohr's atomic theory.
Because, as Bohr
emphasized in 1913, the images from classical mechanics were
imposed on the atomic theory through the perception-laden
interpretation of the mathematical symbols of ordinary
mechanics.
272
A.!, MILLER
exercises
the
of
the macrOCQsmos in
terrestrial
world
great
magic
is
on
rooted
the small
obviously
mankind's
mind;
in the superstition
stars".
among others,
referred
to
as
our
'customary intuition',
to
be avoided
start in atomic
are
atomic
transitions.
In
this
way
the
characteristics of the
scattered
light
wave can be
correlated properly with the incident one.
Although we may
visualize an atomic oscillator to be a billiard-ball like
electron attached to a spring, there are so many possible
transitions that a constituent electron is neither localized
273
representation
"intuitive thinking".
scenario.
is
what
Piaget
Historical data
refers
to
as
by radiation.
Bohr included probability with its inherent
discontinuity in
place of light quanta.
A stunning
prediction was the non-conservation of energy and momentum
in the individual atomic processes.
To most everyone's
relief, in early 1925 this prediction was empirically
disconfirmed.
there was
a sophisticated
oscillators that
probability (see
A.I.MILLER
274
*
*
r
r
THOUGHT EXPERIMENTS
r
r
r
*
BOHR (1924)
~
BOHR (-1923)
~
BOHR (1913)
~
RUTHERFORD I S DATA
CLASSICAL MECHANICS
275
mechanics.
At
of spontaneity" (1962).
That
scheme to data for which the
in a state
A.I.MILLER
276
direct
response
to
the
interplay
between
atoms
and
radiation to
just this problem.
Among the scheme's
ingredients was the virtual oscillator metaphor of the
1913-1925.
of
structures
in
the
genetic
Bohr
theory from
epistemology often
Some
flexibility
in
structure
'reflective
abstraction',
process where reorganization
formation
is
offered
by
notion of
an
disorganized
For example,
invariant or
conserved
field of knowledge
quantity
entering a
to precipitate creativity.
data,
manipulations
neither
the
meaning
of
intermediate
mysterious properties
This situation was
277
of matter,
i.e.,
wave or particle,
with no reversibility.
was
evoked.
Heisenberg's
a wave or
correspondence
and
matter
exclusively
as
waves
that
mechanics,
with its accompanying
imagery.
resisted reversibility concerning the physical
is,
wave
Heisenberg
schemes of
particle and wave; in his opinion the wave mechanics did not
quantum
mechanics
to
give
restrictions
on
A. I. MILLER
278
to
or
our measuring
perceive only
space-time
pictures
of
customary
intuition
must be
reversibility
and
conservation.
Systematization
was
constructed
because the 1927
structure possesses the
mathematical formulation and physical interpretation that we
use today for non-relativistic quantum
in
1927,
Heisenberg
achieved
mechanics.
only
However,
reversibility
of
electric
charge
emerged
from
the
electron's
wave
description.
This problem was solved in 1927-1928 through
the invention of a far-reaching mathematical prescription of
transforming the wave
and
particle
states
of
matter and
279
inherent
in
Bohr's
complementarity
principle,
and
thus
conservation too.
This leads us to suggest that in 1928
Heisenberg assimilated the 1927 structure of the quantum
mechanics as he understood it to second quantization (see
levels 7 and 8 in the figure). "The resulting accommodation
was to the
1928
structure within which he achieved
conservation.
280
A. I. MILLER
was
required
from
But this
either the
structures or theories
rich ones.
NOTES
1. This essay further explores issues from my recent book,
Imagery in Scientific Thought: Creating 20th~Century Physics
(1984) .
281
SENSORIMOTOR EMERGENCE:
PROPOSING A COMPUTATIONAL "SYNTAX"
Claude Lamontagne
School of Psychology
University of Ottawa
1. INTRODUCTION
Within the general context of the Conference's theme as
sketched by Callebaut and Pinxten in their position Paper,
the present paper belongs to this set of papers which can
be said to provide elements for an answer to the question of
the relevance of the Piagetian perspective as a means of
tackling some of the central issues raised by Evolutionary
Epistemology.
In this line of thought, the
paper's
contribution is 'seen' as being twofold. Firstly, the
paper stresses a key limitation in
Piaget's proposed
solution to the problem of 'genesis', or 'development', a
limitation which, although fully acknowledged by Piaget
himself and by a number of 'classical'
as
well as
'neo'-Piagetians (see e.g. Pinard, 1981), still seems to
have gone totally unnoticed by many, namely the failure to
define convincingly the process upon which 'development'
itself
critically rests:
the process of 'abstraction
reflechissante' (from now on referred to as 'reflective
abstraction').
And secondly, the paper proceeds to present
the
first
steps
of an
attempt
to
alleviate the
above-mentioned limitation, an attempt to bring to focus the
with
as a matter of
context-setting,
C. LAMONTAGNE
284
or "cognitive state of
equilibrium",
as
such.
Piaget has, no doubt, given quite a bit of attention
to these former, "diachronic", aspects of the problem:
"It is evident that a structural analysis of
this kind needs a complement: a model that
accounts for change. Piaget does not limit
himself to the framework of the structural
analysis
of equilibrium
states;
his main
interest is in the transition from one state of
equilibrium to the next, that is to say in
mechanisms of transition between
the older
structures and the
new ones.
It is this
qualitative
jump that is
central
in his
preoccupations ... " (Inhelder et aI., 1977, p.7)
However, a major problem arises from the fact that it seems
extremely difficult to achieve
clear understanding of
exactly how Piaget followed up on
this
concern for
'mechanisms' through which more global cognitive structures
emerge from more local ones: explicit theoretical constructs
are rather scarce.
One key notion nevertheless stands out
in
the
conceptual
'constellation'
inhabiting
this
problem-space, that of 'reflective abstraction':
"Already in 1950 the author insisted ( ... ) on
the
necessity
to
distinguish
between an
abtraction
referring
to
objects
and
a
'reflective abstraction'. The latter originates
in the actions or operations of the subject and
transfers to a superior level what is gathered
in
an inferior level
of activity: hence,
differentiations generate at the superior level
new and generalizing compositions." (Piaget et
aI., 1977, P.5)
285
SENSORIMOTOR EMERGENCE
2. SENSORIMOTOR EMERGENCE
From the very onset of our fascination for the Piagetian
endeavour, it had appeared that the 'evolutionary' principle
underlying the primary
concern for "how more global
cognitive structures emerge from more local level ones" had
not been convincingly brought to bear on the most primitive
(hence most fundamental) cognitive structures of all: the
perceptual and motor
ones. Consequently, we had set out to
tackle this task, and some ten years were devoted to
seeking a satisfactory account of the 'nature' of perceptual
and motor systems, with a clear emphasis, throughout the
first
seven years,
on
perceptual
piagetian
research
on
two
extra
counts:
the form
of "neural network
design",
in
most
C. LAMONTAGNE
286
concern
for
neurophysiological
and tied
plausibility.
bear on
SENSORIMOTOR EMERGENCE
287
and
frequencies in
and various
for perception,
and
288
C. LAMONTAGNE
(ii)
discretize
luminous
temporality
into
'moments of light
occurrences',
and (iii) discretize
luminous quantity into 'amplitudes of light occurrences'. Of
course, these dimensions of the environmental energetic
medium do not need to be specified, in the receptor (or
nervous)
domain,
through
an
identical
physical
dimensionalization: if positions of light in space, in the
SENSORIMOTOR EMERGENCE
case
289
of visual perception,
do translate
into positions of
dimension
of
medium concerned
dimensionalized
neural
events,
along
whatever particular
{p, m, f}
C. LAMONTAGNE
290
is any combination
from the following
. , m k }
sSi'
.. "
55}
SENSORIMOTOR EMERGENCE
291
human one.
however,
11M's
must
differ in
a minimum of
as different,
only
one
any two
aspect
of
allows
for,
the
actual
variety
of
those
292
C.LAMONTAGNE
any 'unique'
combination of
Position
value, a
SENSORIMOTOR EMERGENCE
293
define the
"setting
of
O:EE's
in
critical computational
instance,
amount
",
same moment /
same
same moment /
some "residual"
local" EE's in
294
C. LAMONTAGNE
Positions
characterizing O:EE's).
This
leads to the
recognition of the necessity of generalizing our terminology
about O:EE's to EE's of any level, and of generalizing our
terminology about "setting O:EE's in critical computational
relationship with 1:EE's" to "setting n:EE's in critical
computational relationship with n+1:EE's".
In general, then, we shall say that "more local" EEls
which establishes a
SENSORIMOTOR EMERGENCE
295
introduced
as
consequence
dimensions
by
the
of
'type
the identity
of subset'
296
C.LAMONTAGNE
SENSORIMOTOR EMERGENCE
297
structural filiation
from Informational
Interface
Modalities
(11M's)
onwards,
an
obvious
genetic filiation
pattern 'emerges': neurophysiological
'evolution' of perceptual and motor neural networks has no
choice whatsoever but to start with
11M's and their
associated 'knowledge units', Level-O Epistemic Entities
(O:EE's), and to proceed step by step, over the highly
ordered succession of EE-Ievels achieved by setting current
level EE's (or 'more local' EE's) in critical relationship
with next-level-up EE's (or 'immediately more global' EE's)
through the
processes
of Essential
and Differential
Characterization.
In the beginning ... then ... there are Informational
Interface
Modalities,
and
their
associated
Level-O
Epistemic
Entities
(O:EE's).
Progressively, organisms
appear which access Level-1, and then Level-2, Level-3,
Level
298
C. LAMONTAGNE
"identity seekers",
Essential
Characterization
is
but
the
conjectured
irrelevance of some dissimilarity between n:EE's, and that
it seems better to keep the extent of the dissimilarity as
limited as possible; and the second type of considerations
points to the fundamental belief that systems tend to
conform themselves to certain rules of "processing economy",
exploiting for instance, in the case of neurophysiological
in that,
feature values.
In order
to allow
for efficient visualization
of this latter
argument, it might be worth introducing at this point the
"uniformization principle" which we applied in defining our
working representation
of
11M's neural
domains
and their
SENSORIMOTOR EMERGENCE
299
continuums
as
those
tapped
by
their environmental
counterparts, and are therefore free to exploit whatever
continuums seem to offer some other type of advantage.
Concretely,
our
spatial uniformization principle
amounts to treating
11M
neural
domains
(and their
local-global ramifications) as multi-dimensional spatial
cell
arrays.
Let us
consider,
for
instance, the
intra-modality diversity implied by an 11M whose {p, m, f}
reads:
{Input, Photonic, (Positions, Moments, Amounts)}
where
"Positions"
stands
for
some
two-dimensional
orthogonal set
of
8x8
discrete
(and quantitatively
equivalent) "slabs" of space, where "Moments" stands for a
set of 2 discrete (and quantitatively equivalent) "slabs" of
time, namely moment(O) (present) and moment(-l) (immediate
past), and where "Amounts" stands for a set of 2 discrete
"slabs"
of luminance, namely amount (0) (darkness) and
amount(l) (light) (See figure 10).
In this overall context, then, "greater similarity"
amongst EE's translates into "greater proximity" of EE's,
restrict
emergence
paths
(i.e.
local-global
ties)
exclusively to those falling within the boundaries set by
"contiguity"
relationships,
as
extreme
form
of EE
similarity, and we embodied this constraint in a formal
neural network which, expressed according to the "spatial
uniformization" principle, takes the form illustrated in
Figure 11, where Co stands for any single n:EE supporting
cell, and where (C"
C2 , C 3 , ... , Cn ) stand for Co's
contiguous n:EE supporting cells, n varying according to the
particular geometries of n:EE supporting cells.
The different thresholds introduced in the network
(the 2's and the n) in fact imply local-global ties which,
quite apart from being based exclusively on contiguity
relationships between single n:EE's and other n:EE's, are
also
based on
whatever
single
n:EE)
whatever
chosen
possible
to ignore,
of
all
dimension(s).
the
contiguous
occurrence of
n:EE's
along
possible
single
relationship: it
in the definition of
general
made it
the "contiguity"
various geometries
300
C. LAMONTAGNE
positions
(SxS)
moments
(x2)
=~~~ii~I-;~%~~~~iI~~~l11:Q
amounts;
(x2) ~
1__
I
I
-1 L~
Figure 11.
SENSORIMOTOR EMERGENCE
It
301
was decided to
"contiguity
primitive"
start "testing"
the
in the context of a
above~defined
perceptual and
where "Positions"
discrete
(and
quantitatively
equivalent) hexagonal
discrete (and
geometry"
of
the
neural
domain
associated
with
such a
perceptual IIM.
Most critical for eventual "contiguity
relationships" is of course the hexagonal geometry of the
"positional space": for anyone non-boundary cell, it allows
for exactly six (6) contiguous cells. Two convergent types
of considerations in fact led to the choice of this
particular geometry for the "positional space": the first
type has to do with the particular ease with which the
hexagonal geometry lends itself to establishing uniform
contiguity relationships, which stands in sharp contrast
with the difficulty which other geometries,
like the
"dominant" orthogonal one, present vis--vis this same issue;
and the second type of considerations has to do with the
extremely widespread occurrence of
the
hexagonal (or
quasi-hexagonal) geometry in "natural constructs" in general
11M,
"Positions"
stands for
set
of
discrete (and
quantitatively equivalent)
portions
of
space, namely
position(1) (agonist muscle) and position(2) (antagonist
muscle) ,
where "Moments" stands for a set of
discrete (and
quantitatively
equivalent)
portions
of
time, namely
moment(O) (present) and moment(~1) (immediate past) ,
302
C. LAMONTAGNE
Figure 12.
63
~
1
Momen ts
Amounts
Positions
Figure 13
SENSORIMOTOR EMERGENCE
303
Perceptual 11M:
Motor 11M:
or
or
{I,P, (P,M,A))
~
yielding a potential intramodality EE-diverslty
of:
~
p ri'I ali
p ri'I Ii
ri'I ii
ri'I ii
p ri'I Ii
p ri'I ii
p ri'I ii
(a)
P
P
P
P
P
P
P
ri'I ii
ri'I ii
ri'I ii
ri'I ii
ri'I ii
ri'I ii
ri'I ii
(b)
single layer
portion of
Level-O EE's
(O:EE's)
single layer
portion of
Level-1 EE's
(1:EE's)
Figure 15.
C. LAMONTAGNE
304
and
where "Amounts"
quantitatively
equivalent)
portions
force" continuum.
Now these two 11M's open
well
as
intra-modality
up
64
of
onto
EE diversity.
discrete (and
the
"contraction
inter-modality as
We chose to address
variations within
feature-sets.
Each
11M
was
therefore
of the respective
attributed
an
EE
Level-Q
was to be tested.
Figures 15, 16, and 17 show the
prototypical formal "wiring" patterns obtained by applying
the
"contiguity primitive"
diversity combinations
of
to the first
the perceptual
three (3)
11M
II
p ,m,a''
(i.e. those
primitive
to
the
"Moments"
feature space
(Figure 16),
that the 1:EE supporting
neural domain
(right-hand side portions of 7a and 7b) does not follow the
dimensionalization strategy adopted from the start, where
the vertical dimension was firstly split in "blocks" of time
(i.e. moments), each of which was then split in "blocks" of
quantity
reversed,
(i.e.
the
amounts):
vertical
dimension being
firstly
split in
SENSORIMOTOR EMERGENCE
portion of
O:EE's
portion of
l:EE's
305
portion of
O:EE's
portion of
l:EE's
(b)
(a)
Figure 16,
portion of
O:EE's
portion of
l:EE's
(a)
Figure 17,
portion of
l:EE's
portion of
O:EE's
(b)
306
C. LAMONTAGNE
order to mark the fact that in this ~e:y case it does not
present a "genuine"
diversity, pa1r1ng moment(O) with
mom:n~(-l) appearing as strictly equivalent (given identical
pos1t10n and amount), in terms of outcome, to pairing
moment(-l) with moment(O).
It should also be noted, this time regarding the
application of the contiguity pr1m1t1ve to the "Amounts"
feature-space (Figure 17), that if the particular. wiring
implied represents a formal
possibility, it does not
represent
an actual
p,m,~ or
p,Gt,a)
is applied repeatedly
characterization"
being
implemented
through
the
very
,i.e. recursion,
within
obviously
basis!!
become
incapable of
Introducing
doing it
recursion
in
the
on
the
a stand-alone
"differential
characterization"
process
simply meant
repeating the
repetitive application of the contiguity primitive (onto the
newly created sets of EE supporting cells) until no cells
SENSORIMOTOR EMERGENCE
307
be "necessary",
if one is to
lIinhibition tl
appears
necessary in
hope
reaching neural structures
presenting functional
similarity with the visual cortex of higher mammals, it does
not seem to be
"sufficient"
for
this
purpose: an
order to rid
arising
from
primitive
the
the
also
to
be
308
C. LAMONTAGNE
3. CONCLUSION
But what does all this have to do
with "reflective
abstraction"? Are we not stretching the notion far beyond
its "breaking point"? How serious is it to argue that the
combinatorics
of
"EE
local-global
contiguity-based
relationships" prefigures the whole Piagetian "equilibration
process"
in general, including the key-aspect of the
acqulsltlon of conservations,
abstraction" in particular?
the
If,
and
Piagetian
II
re flective
of
model"
context of
in
our research
endeavour
that it
aims at
"conservation",
and
"reflective
perceptual
and
motor
emergence
process,
where
sets
of
309
SENSORIMOTOR EMERGENCE
point
of
fact.
In
our
particular
case,
"empirical
abstraction"
is carried out
through
the
actual emergence
of
our endeavour
to
the
concept
of
"conservation".
As
can
be
"conservation", which
C. LAMONTAGNE
310
that we
of our
ACKNOWLEDGEMENTS
The financial
support
of the
National
Science and
Engeneering Research Council of Canada over the past few
years of research,
and
the
powerful, fruitful, and
passionate
conceptual resistance offered
by Jean-Roch
Beausoleil,
Celine Cote, Denis Belisle, Jean Bernabe,
Jean-Pierre Delage, Alain Desrochers, and Pierre Tremblay,
over the past ten years of research,
are gratefully
acknowledged
as determinant factors in
obtaining the
above-reported research results.
Leo Apostel
Rijksuniversiteit Gent
1. INTRODUCTION
The papers I will comment upon
(those by Gillieron,
Lamontagne, and Miller) were brought together because all
three of them are concerned with 'genetic epistemology'. As
the editors of this book have remarked, an EE looks for a
model of the history of science in phylogenesis, while
Piaget himself has mainly examined the analogies between
ontogenesis and the development of concepts, in his genetic
psychology in general and in his main treatise on GE (3
vols.) in particular. Piaget & Garcia (1983) offer the only
fully worked out parallel between the history of science and
genetic epistemology.
One might thus claim that GE does
not belong to EE.
If this were true, the models offered in
the papers I am going to discuss would fall outside the
scope of EE.
They
would simply be
examples of a
naturalistic epistemology according to which the study of
the evaluation and history of science must
itself be a
science. The sciences taken as paradigms are different
however:
deductive neurology in
As far as
I can
see,
her
conclusion
is
not
On each
312
L. APOSTEL
is nea-Darwinism
with something
added
proliferation of vicarious selection levels.
every
by a
to
As
proliferation,
non-random,
random one
it:
long
the
the
as
EE
teleonomic
on another
level.
In fact, during the conference this book is the
result of, Donald Campbell emphatically insisted that his
work has the nature of a challenge: he calls on others to
discover or refute the existence of his levels and he is
fully aware that the EE he is presenting now is purely
speculative. To a lesser degree Jean Piaget
also deviates
from classical nee-Darwinism.
His deviations
are more
severely sanctioned in biological
circles because his
phenocopy hypothesis seems incompatible
with molecular
biology and comes close to Lamarckism.
The idea that a
species which has, during one or n generations, adapted to
external influences, and then reconstructs from the genome
remains:
What
is
the most
theories
and
neo-Darwinism.
of
everyday
skills
with
an
enriched
313
L.APOSTEL
314
with
strong
enough
forces,
the
intuitions of
a 'reductionism'?
if
we know
biology
and
biology
explains
psychology.
The relations
were
may
be
related, magnetism
means of physics,
that
physics
will
be
derivable
from
considers, in 1987,
pure mathematics).
Sometimes the arrows mean 'are used in' and sometimes they
mean 'are explained by';
(iii) The vicinity between
315
science;
(iv) The relation between psychophysiology and
logic 1S similar to no other one: causal chains on the
neurological level correspond to implications on the level
of consciousness. So GE claims (as any theory of entailment
and of causality can show)
that these two relations
(causality and entailment) do not have the structural
properties to solve, by some kind of structural dualism, the
mind-body problem which is lurking at the core of EE. The
relations between EE and the Piagetian circle of the
sciences are
unclear.
To be sure,
if a biological
explanation
for
the
history
of knowledge
and its
self-imposed norms were available, this would allow to close
the circle at one point (the relation human biologyinformation acquisition in psychology and history). But
the circle needs much more than that.
Moreover, if mutual
assimilations and accommodation are typical of the circle,
then biology will have to be changed as much by epistemology
as inversely. Yet,
neither in history nor in behavior can
strong arguments in
favor of the circle
be found.
Inversely, even if EE were more than a program and already a
reality,
nothing would compel us to claim either
a
historical or a suprahistorical truth in favor of the circle
of the sciences.
The fact that EE does not entail the
reality of the circle is not identical to the fact that EE
is
incompatible with it.
For those who fear that EE
amounts to a
reduction of higher to
lower levels of
complexity, the possibility (not the necessity) of such an
organization
is an important argument alleviating their
doubts.
Arthur Miller's contribution is an attempt to prove,
by means of a concrete historical analysis, that the
development of one science during a certain period (quantum
physics from 1913 till 1927) can be explained by mechanisms
due to GE.
If the reader is convinced, after my discussion
of Gillieron, that after all GE is an example of EE, then
this case study becomes also a strong argument in favor of
EE; it would
enable, in its Piagetian version, to throw
light on the history of the most advanced sciences. History
- all naturalistic epistemologists agree - is the laboratory
of epistemology.
The paper refers to Miller's (1984) book.
Before entering the discussion, I want to make a preliminary
remark.
Max
Wertheimer's
Productive
Thinking
contains a
316
L.APOSTEL
view,
come
to
a decision
on
this
point.
However, Miller
dispositions."
If
the
invariance,
concepts
of
accommodation,
317
assimilation,
Miller will give them new meanings that are adequate in the
new context of history of science, This has bearing on the
relevance of his programs as to
their status in EE.
Obviously, a generalized schemata dynamics will have to be
derived from
Campbell's n-level random variation, if the
facts of Miller are to be considered as relevant to EE.
Here I would like to make some remarks about the
application of Piagetian concepts to facts of the history of
science, and address the question "Would another type of EE,
now extant, have done better with the material at our
disposal?"
In 1913, Niels Bohr,
in order to explain the
stability of an atom conceived as a minor solar system
surrounded by electrons, proposed that only certaln paths
can be followed by electrons, which emit energy in finite,
discrete amounts when they jump from one path to another;
they cannot be 'observed' during their jumps (there, no path
exists) and they can never get closer to the nucleus once
they have reached a basic trajectory.
Miller describes this as follows: The Rutherford atom
The Bohr atom (quantized) creates
was 'in disequilibrium'.
an
equilibrium,
by
'adjusting'
or
'accomodating'
Rutherford's atomic model to the facts of 'stability'
(explained by the quantization). However, this Bohr atom is
still far from stability because, keeping the initial
representation of Rutherford (the atom as a small solar
system), it 'displays representative activity analogous to
egocentric representation'.
I attract the
attention of the reader
to the
psychological terms used.
a) Presumably the Rutherford atom is 'in disequilibrium' because its stability is in contradiction with
classical mechanics. So far so good. If all contradictions
in science entail a disequilibrium, then Miller is correct.
History of science shows, however, that not all patent
contradictions create the immediate impulse to change the
theory. Therefore, I would only accept to a certain extent
the equivalence 'contradiction = disequilibrium'.
b) However, the Bohr atom is, for the very same reason
as the later history shows clearly, also in disequilibrium,
because it
certainly also contradicts classical mechanics:
electrons changing orbits do not have paths but should have
318
L.APOSTEL
them.
To call this an 'equilibration' is already difficult
to defend.
One simply jumps from one 'disequilibrium' to
another.
c) I am utterly unable to understand why the Bohr
'theory' displays 'egocentric representation'. Egocentrism
in psychology is the identification of an object with the
perspective on that object as presented to a particular
subject from a specific vantage point. Nothing in the Bohr
atom shows such 'egocentric representation'.
d) If we follow Miller's story, we repeatedly face the
mixture of a creative attempt to explain the history of
science by schemata dynamics and of the observation that
schemata dynamics cannot explain this same history.
For instance, reaching the pure formal stage is
certainly not the result of the Schrodinger-Heisenberg
stage, since both wanted to continue using visual imagery in
their work (and so with excellent results). Moreover,
Piaget's claim that accommodation and assimilation work with
equal velocity and pregnancy is clearly negated when Miller
tells us that during 1913-1923 "the thinking of Bohr and
other
physicists
emphasized
accommodation over
most
assimilation".
'preconceptual
thinking',
because
both
accommodation and
social in
because
high-class
results
of
physics
are
319
is what
to 'reversibility'
320
L.APOSTEL
Disagreements notwithstanding - the main one being the
321
ordered
neuron systems
L. APOSTEL
322
'reflexive
abstraction'
combinations
organism" .
of
internal
and
external
actions
of
the
Can
the simulations of
biological
evolution, poor as
By
remarks:
way of
general conclusion,
1.
My first recommendation would be, given the current
multiplicity of theories of evolution, to work on the
epistemological consequences of all
these explanations.
No useful purpose is served, for non-biologists, by choosing
in favor of one of the competitors;
the attempt to tie EE
strictly
to one explanatory schema has until now had the
consequence (see Donald Campbell,
Jean Piaget, Konrad
Lorenz) that either the basis was transformed in order to
make the explanandum more understandable, or that the
explanandum was
transformed in order to make the basis
suff icient.
2.
My second recommendation is to follow Miller's example,
apply various theories of the evolution of knowledge to the
history of science, while remaining fully aware that the
transposition
from one
domain to
another introduces
arbitrary choices which must
be noted first, be defended
next, and finally be overcome by using more than one
evolutionary scheme.
3.
My third recommendation is to focus by all means on the
evolution of the brain-in-environment (ecology and ethology
combined with brain science). As far as I am concerned, AI
simulations will have to be selected at their highest
levels
and
their
evolutionary
theory
should
be
reconstructed.
323
PartlV:
Extensions and applications
order to
for evolution,
of horizontal transfer
the basics
of molecular
biology.
In molecular biology life is said to be the
reaction between three sorts of elements.
One element is
the coded information, which takes the chemical form of a
linear molecule and is composed of a number of repeated
signals: the code.
The second element is the cell, or a
number of cells together: the organism, a system in which
327
328
Proteins cannot
be changed
assembly,
is changed.
the fittestf
is,
to
some extent,
329
which is
the
classical
basis
for
understanding species,
The following is
330
J.SCHELLANDD. DE WAELE
applied,
every
bacterium that was sensitive
to the
antibiotic dies and only the mutant bacteria which have
become resistant to the antibiotic through mutation survive.
This is Darwinism at it simplest - a clear proof that
Darwinism is basically correct. . However, if instead of
applying the selective pressure to the culture in isolation,
we first put its component cells in contact with other
living systems (other bacteria, or plants or whatever there
is), then we do not find that mutation has occurred, but
rather that bacteria have received a whole set of new genes
from the other living organisms.
What has happened? We know that bacteria have evolved
very dynamic genetic systems with a variety of mechanisms
for the transfer of DNA. We also know that the most
efficient mechanism for bringing new DNA into a given cell
operates between cells of almost the same type. We have
demonstrated that several types of bacteria dramatically
acquired e.g.
the ability to detoxify specific antibiotics
in this way, even when we did not intend to treat them with
antibiotics. This indicates that bacteria do not evolve
through mutation, but rather by opening themselves up to
genetic information that originates and has become adapted
in other environments. In the usual way, a genetic program,
thus acquired, that provides a better chance of survival for
the cell, will be maintained.
types
of
cells are
taxonomically
related,
331
although some
information
in
(e.g.
for
several
proteins or
if only one of
these functions or
properties has a selective advantage, nevertheless the
complete set of properties will be transferred. One change
functions.
the
Even
environment
the
selective
pressure
of an
antibiotic)
can thus lead to the acquisition in the
population of a whole set of new genes.
When such a new
gene - for which there is no selection - has significance
for the cellular system, this can produce an abrupt change
in the evolution of the species concerned.
The phenomenon of gene transfer via plasmids is very
important for understanding the evolution of microorganisms.
It indicates that their evolution comes about primarily
through horizontal transfer of information, and not through
the gradual accumulation of mutations.
7. CONSERVATION OF INFORMATION VERSUS
INNOVATION OF INFORMATION
evolution we see
a balance between
two opposing
tendencies: (i) a tendency to keep genetic information
constant within a species; (ii) a tendency toward the
dynamic transfer of genetic information from one living
system to another.
One of the systems which both allows the transfer of
genetic information and restricts the transfer to fairly
similar organisms, is sex - one of the pleasures but also
often one of the miseries of our life. Linking the exchange
of genetic information with reproduction
is extremely
efficient.
A new generation is made out of the coming
together of the genetic information
of two different
individuals who have but slightly different genetic programs
(variation on alleles).
Since what is transmitted from one
generation to another is only the information carried on by
the germ lines, there is a tremendous limitation to the
transfer of foreign information.
Only information present
in, or able to enter, the germ lines will be propagated to
the next generation.
In
332
Bacteria,
however, do not
between reproduction and exchange
information
completely
independently of
genetic
same
protein.
code
is
If the code
were not
exchange of
information
could
have
no evolutionary
significance: the two sets of information concerned would
not be compatible because the codes would be different.
Genetic engineering is based on this universality of
the code. Geneticists are looking for those signals which
make the DNA function and are making new combinations of
different DNA
pieces.
The
combinations that became
popularized
first
were those
directed by industrial
interest.
For example, the case of insulin for patients
suffering from diabetes: the genetic information for the
synthesis of insulin is taken from human cells and the
signals are modified so that the DNA can become active in
bacterial cells.
Since bacterial cells duplicate every 30
minutes the synthesis of human insulin
cells can produce large
amounts of
out of bacterial
insulin.
A new
333
different species.
This natural system of gene transfer is
studied in the Laboratorium voor Genetica (Gent, Belgium).
In this case, highly evolved organisms, namely plants,
biological significance.
334
335
The
environment was
supposed
to
pick out
conjecture and to
But
as
the 'suspect'
so
often
interesting heuristic,
conclusions.
For there
lemmas.
happens,
first
sight
is
an
method as well:
(a) the method works only if a proof or refutation has been
found;
(b)
(that is,
polyhedra) (1).
Concerning (a), it can be remarked that a large part
of
mathematics
consists of unproven
statements; and
concerning (b), that it is not always the case that
counterexamples are 'accessible'.
This means that in all
cases not covered by
and refutations,
338
J. P. VAN BENDEGEM
yn-1
> nx n - 1
>
n.
QED.
339
340
J.P. VANBENDEGEM
341
that a solution exists, say (x, y., z). Then prove that
there must be a second solution, (x', y', z') such that [x]
> [x']'[y] > [y']'[z] > [z']. The argument can now be
repeated with the solution (x', y', z') infinitely often
(hence, infinite descent). But as all these solutions must
be whole numbers, this is impossible. Values cannot become
arbitrarily small within the integers.
But in order to
construct such a proof, it was necessary to introduce
certain algebraic identities, specific for the case in
question.
And exactly that prevented generalization to a
complete proof for all cases.
Note: Only cases in which n is a prime number need to
be considered, as all mathematicians involved with FLT knew.
For suppose that n = m.p, where p is a prime. And suppose
you have proved that x P + yP = zP has no solutions. But
then it follows that x" + y" = z" can have no solutions
either. Suppose it did, then we can rewrite x" + y" = z" as
(xm)P + (ym)p = (zm)p, but then x m, ym, zm is a solution of
x P + Y P = z P, which is impossible.
Therefore in what
follows, only prime numbers will be discussed as exponents
and always indicated by p (or p').
2.2. BREAKTHROUGH PHASE:
VARIATIONS ON EXISTING METHODS
(a) Gauss was the first mathematician to attack the problem
in a different way.
But not altogether that different.
His attention was restricted to the case n = 3, so, in that
sense,
But
what
to
342
J. P. VAN BENDEGEM
open new
possibilities.
As we
will see, this pattern will return over and over again.
(b) Simultaneously with Gauss, a second line
there is a solution,
certain n, xn
solution have.
i.e.
of
yn
zn
solution
exists
and tries
to
list
properties
such a
343
interest.
In short, what I want to say is that, given
certain proof searching methods or counter-example searching
methods, there are in some sense optimal decompositions
decompositions
mentioned
nevertheless excluded.
above,
FLT is
some
decompositions
are
of integers
344
J. P. VAN BENDEGEM
or not.
345
v (s) = E
Remarkably
n=l
enough,
there
is
an
(4).
>
interesting
connection
J. P. VAN BENDEGEM
346
valuable
information
counterexamples.
The
concerning
the
result of Grunert,
size
of
possible
e.g., belongs in
Kummer
introduced,
approach of
347
348
J. P. VAN BENDEGEM
work.
some
from Furtwangler's
zn.
to
find.
Falting's
result increases
this feature's
intensity: counterexamples are of impressive size and rare
as well. Apparently you cannot have both at the same time
(at least not in the case of FLT).
(c) The importance of FLT can be judged by the fact
that in domains totally different from number theory (where
FLT originally was formulated), attention is paid to it and,
interestingly enough, sometimes with a completely different
motivation. Let me list here just two examples:
349
xn
"yn
have
the following
zn
~(X,y'Z)I~ M'I~(X,y'Z)I~
M,I~(X,y'Z)I~
M
a yn
b zn
350
J.P.VANBENDEGEM
principle.
If there would be a proof (for S or 'S) then a
contradiction would follow.
There is a beautiful example
that illustrates the problem (11). Consider all finite games
(chess, checkers, ... ). Consider now metagame the rules of
which are very simple: the first player makes a first move
by choosing a finite game. The other player then makes the
first move in the finite game, and the game proceeds until
it ends. Consider the proposition 'Metagame is finite' and
its negation 'metagame is infinite'. Surprisingly, metagame
is neither. Suppose metagame is finite. Then the first
player may select metagame. The other player then selects a
finite game, say, metagame and the first player ... So
clearly metagame is infinite. But if it is, then the first
player cannot choose metagame, but must select a finite game
and thus metagame always ends. Hence it is finite. Thus, we
must conclude that metagame is neither finite nor infinite
in principle. The first statements that were shown to be
undecidable were highly technical and far removed from daily
mathematical practice, but, later on, it was demonstrated
that problems of a more concretenature (whatever that means
for mathematicians) were equivalent to these undecidable
statements.
Hence they were undecidable too.
One such
problem was the search for a general solution to the
Diophantine problem. Roughly speaking, Diophantine problems
involve polynomials with integer
coefficients and the
question is to determine whether solutions in integers exist
or not.
Now, the general Diophantine problem has already
been shown to be undecidable (12). FLT is a special case of
a Diophantine equation. Could it be that FLT is undecidable
too?
This is precisely the question logicians are working
on. If this attempt would prove successful, it would imply,
cynically enough, that all the previous time spent in
searching for a general solution was wasted.
Note that if
it were to turn out that FLT is undecidable, this does not
exclude that special cases can be settled, as we have seen.
This kind of research or
approach
would
have been
impossible, say, fifty years ago (Godel published his famous
paper in 1931) but, today it is part of the spectrum of
mathematical activities that, either you try to prove
something, or to prove that it is wrong, or you try to prove
that it is undecidable.
Tracing back this undecidability
approach, we must return to Godel, whose work was based on
the work of Hilbert.
His project was to
show that
mathematics is complete and consistent. He wanted to do that
because there
were problems
in
the
foundations of
351
352
J.P.VANBENDEGEM
mathematical universe.
353
::::t:o:i:a:i~l:
:1 and y =1:
:1 / 1: :1)
354
J. P. VAN BENDEGEM
355
a complex is
between the
mathematical universe.
In that sense, the concept of a
make-up carries explanatory power.
Finally, the model explains why the classical view of
mathematics does not hold, viz. once a problem is solved,
that is it.
If you have a proof or a refutation of your
theorem, you can start to attack the next problem. But,
strangely enough, mathematicians spend a great deal of their
time, rewriting existing proofs, polishing, cleaning up,
trying to find shorter proofs, etc.
Why are they doing
this?
The answer in terms of the model is straightforward:
it guarantees remalnlng in existence.
Take e.g., FLT. A
great deal of the work of Furtwangler was a rewriting of the
proofs of Wieferich, Krasner and Miromanoff. What does the
model say? Wieferich establishes a S-link between a certain
P-complex and a certain method. If a S-link can now be
established between the same P-complex and another method,
then this increases
the continued
existence of that
P-complex because it now has two S-links.
In short, the
more different proofs, the better! I consider this a clear
356
1. P. VAN BENDEGEM
linear,
part of
phrase
'remaining
in
existence'
replaces
'increasing its
survival value'
in
the first draft
mentioned. It is perhaps somewhat odd to associate the
survival of a problem to its being solved or not. However,
it is plausible to assume that an isolated problem remaining
unsolved for a long time, is not likely to hold the
attention of the mathematicians.
C-links are important precisely to prevent problems to
disappear if they resist being solved. Furthermore, if a
problem is solved, this does not imply that the problem
disappears. A solved problem indicates a S-link with some
M-complex, which through some C-link might prove to be
important for an as yet unsolved problem.
Notwithstanding the numerous similarities,
it is
obvious that the model presented here is not well enough
articulated to justify
the
choice
of
a particular
evolutionary model (16). Nothing has been said about the
carriers of the mathematical ideas, viz. the mathematicians
themselves. The story presented here takes place in the
357
Hence it is in accordance
358
J. P. VAN BENDEGEM
in the link.
Try to find a P-complex such that the S-link
no
longer
holds
for
the
whole
P-complex (global
counterexample) or try to find a M-complex such that the
S-link no longer holds between that M-complex and the
original problem (local counterexample).
The first part is rather obvious, but the second part
requires some explanation.
If the original problem is
reshuffled and the S-link no longer applies, this can only
be because there is an element in the P-complex that resists
the S-link.
This is precisely what Lakatos calls a global
counterexample.
If on the other hand, the S-link is lost
because the method is reshuffled this can only be because
some element of the M-complex resists the S-link and this is
a local counterexample.
The specific element of the
M-complex that is the cause of the breakdown of the S-link
is what Lakatos calls the suspect lemma. The reshuffling is
precisely the way of making clear what the hidden lemmas
are.
Corollary.
The establishment of a S-link between a
P-complex and a M-complex does not imply that the complexes
involved are no longer subject to reshuffling.
4.
It is obvious that
CONCLUSIONS
is
to look
for other examples and see if these too fit the model.
After all, that is precisely the procedure followed in this
paper.
359
established some years ago, but with the strange effect that
some mathematicians accepted it as a S-link and some did
not,
The problem was that the proof involved the use of a
computer (some two hundred thousand quite complicated maps
had
to be colored
effectively)
and that
for some
mathematicians such a proof is not a proof since it is not
open to control by individual mathematicians,
In terms of
the approach outlined here, this means that the nature of a
S-link needs specifying in the very same way that the ways
in which the reshuffling takes place needs to be specified.
NOTES
satisfy V removed:
E +
+2,
particular
I
I
polyhedron
.,...----'"
,LL--l :
I':
I
I
'I 'I
I
: i,/}---~-7
I"
I
I'
I'
I'
I'
I'
that
~------"
1-------
does not
inner cube
J. P. VAN BENDEGEM
360
2.
n=O
->00
an infinite series,
theorem, which
n=O
of both g(x)
361
see Note 4.
x .
+ ~ C.xm. (2k)n- m
m=O
Eliminate xn from both sides and observe that on the
right-hand side each term of the remaining sum has a factor
k in it, hence it is divisible by k. But then the left-hand
side must be divisible by k as well.
Thus (x+k)n is
J. P. VAN BENDEGEM
362
2b-1,
reads:
where
b is
a certain number.
<h. ~f0110W;",,"C1""
363
h->O
11.
12.
See,
Steen (1978).
13.
e.g.,
(1983), p.
See
the
position
paper
in
this
volume
for
17.
Rijksuniversiteit Gent
1. INTRODUCTION
This
paper
epistemology
touches
on
certain
central
issues of
in relation to development and evolution. We
bring up
certain suggestions
for
the
development and
as (i)
a modifier of the
object
specification of
noun epistemology,
the noun.
or as (ii) an
In the latter
case
we
development etc.? In
EVANDAMME
366
theory
about knowledge.
In this
evolution of epistemologies~
in its turn, can be considered
as a special case of the development of knowledge. But even
does
per se.
The
same
can be
argued here.
and selection)
and
biological
systems (variation
respect.
The
Kitchener
work
~
of
Ekeland,
Thorn,
Prigogine,
Popper,
of knowledge.
We think the epistemological problems
deterministic
example,
character of
laws
are
more
related
to the
peculiar. For
The
point we
want
to make is
367
is that "knowledge",
at
least
In
Its
social and
psychological dimensions, is always changing according to a
certain context,
or
subdomain
epistemology then is
epistemology.
of
epistemology.
viewed as a standpoint
An
evolutionary
within dynamic
of
18th century
view
is that,
EVANDAMME
368
These
neither repeat
nevertheless a number of
as
is
witnessed
by
interesting phenomena,
thermodynamics.
as the fourth
Time
dimension of
geometric image,
reflection
of an irreversible
369
a biological
regularities?
3. What about freedom: how is it possible to produce the
impression of freedom or at least partial independence from
the past by the present, when taking a future action?
Is
this only an epistemological indetermination, does some
kind
of
ontological
indetermination
occur
or
is
EVANDAMME
370
dimension
is
concerned,
but
is
plainly
point of
stage
in,
at
erroneous when
least
in
the
view, we introduce
organism
is
e.g.,
Adolphe is
able
task
to perform
and
not
underlying
shows how
a certain
others.
This
cognitive stage
the person
is
kind
is
of cognitive
competence
law.
For
example,
Piaget's
cognitive
developmental
stage
law
of
individual
development
(sensorimotor
intelligence
concrete operational
intelligence
formal
operational intelligence) is "explained" by a
developmental
principle
of
increasing
equilibrium-equilibration,
and
evolutionary
stage laws are "explained" by reference to
Spencer's
principles
of
increasing
differentiation and integration, by Williston's
law
of
increased
specialization
of function,
very
central, as
371
history
or
or
generalizations
development
of
concerning
the
particular
truly universal
or nomological." (Kitchener
1983, p.793)
Kitchener himself defines a stage law as:
"(x)
{Kx (S1 (x,tw)
(S2 (x,ty) . S3(x,tz},
i.e., for all x's of a certain kind K, if x is
in a certain stage Sl at time tw then x will be
in stage S2 at time ty (where ty is later than
tw) and x will be in stage S3 at time tz (where
tz is later than ty)." (Kitchener 1983, p.792)
So in
a nutshell,
in stage laws
properties
organisms that
EVANDAMME
372
seems
to
throw
its
scientific
373
"In
traditional
(prestructural) linguistics
people were mostly occupied
with comparing
genetic related languages and tried to render
account of features of the one, while refering
to certain features of an other, instead of
trying to determine how and why a given language
evolves through the centuries. In other words,
researchers had a tendency of signalizing the
correspondences rather than to explain them."
(Martinet 1962, p.162)
They have looked for the universal, i.e. the common
core between different languages. In a certain sense it can
be argued that one looked for the invariable in the
variable.
374
F.VANDAMME
sense.
it,
in the
on
strategies
for
observation.
Strategies
in
Martinet focuses on
Nowadays,
information is
Nevertheless,
interesting
notion of
language.
elaborations
and
The syntactical,
the semantical,
pragmatical
Lamarck/Darwin controversies.
375
new impetus.
For
F.VANDAMME
376
in
the
manner
that
it
is
taken
into
account
in
its
form of symbolisation).
When looking at language change in this way, one
could be tempted to argue that we have a kind of synthesis
between Lamarck and Darwin in the Martinet approach of the
change in symbolic system. The Lamarck aspect is present in
the principle that the change is determined by its use. The
Darwinian aspect is present in the selection by
means of
functionality. But we must be aware that it is social
selection, and a social functionality with an ultimately
social survival!
Language and knowledge are primarily
social, with individual aspects. For language as well as
knowledge is meaningless and purposeless, without this
social dimension.
We are led to a series of interesting problems.: Can
this synthesis been generalised to encompass all living
systems and their changes? Do we have to speak about
language change, or may we speak about language development
and language evolution? These last terms imply some more
specific structures of change. The same certainly is valid
for knowledge. Do we simply have knowledge change or do we
have true knowledge development and knowledge evolution?
5. LANGUAGE AND DEVELOPMENT
versus
the ontogenetic
normal case,
that it
is
it
not
is
nevertheless
"default"
not
feature
automatic (meaning
of
system change).
377
and
integration,
of
in
view
the group of
of
extending
users of the
large.
In
technique
one uses,
differentiate
between
several stages
But we
change is
sets of analogous
concerned,
as far
symbolic systems
as the ontogenetic
development
(inside
the
overall
social
stage
of a
the target
ontogentic
system
in
phylogenetic process).
Defining progress at an ontogenetic level is therefore
mainly done by reference to an external target, which is
the level of development of the phylogenetic symbolic
system. It is true that in the endeavor to reach this level
a certain ontogenetic symbolic system may deviate from or
transgress
this external target. Such transgressions are
important tools in and for the further development of the
phylogenetic dimension of the symbolic system. But the
phylogenetic evolution
always remains a prerequisite for
EVANDAMME
378
more
important,
others
are
more
peripheral
or,
give
organising and
pragmatical:
defined in
terms
of
some explorative,
predictive or therapeutic perspective, for instance. The
actual status of the phylogenetic structure of the symbolic
systems plays an heuristically important
role in the
construction
of the stages,
we believe.
379
deviations
default case is
be focused on,
in
its
beyond deviation.
characterisation. The
If not,
deviation would
on
the
norm-stage
or default
6. CONCLUSION
In Vlew of this analysis of the development of symbolic
systems, in their ontogenetic as well as phylogenetic
dimension, and considering language and knowledge system as
particular cases of symbolic systems,
we propose the
following hypothesis: knowledge systems phylogenetically
evolve
(or progress in the Spencer interpretation) along
internal phylogenetic criteria towards a higher level of
adaptational
and
accommodational
efficiency.
380
F.VANDAMME
essential one.
In
this
interpretation
dynamic
epistemology,
including EE as a special case,
is the epistemology of the
future. Indeed, dynamic epistemology is the only one which
can be a base for a true applied epistemology. The latter is
so much needed in Artificial Intelligence, in Intelligent
Information Technology,
but also in politics: General
Politics as well as Science Policy
and in cognitive
education strategies as well as cognitive therapies.
NOTE
(1)
realistic
ontological
point
of
view.
In
take a
strict
incorporates
much more
ACKNOWLEDGMENT
The informal discussions and interactions of D. Campbell and
G. Vollmer were very stimulating for the elaboration of this
paper.
entities
to
be
which
explained
these
through
features
the
selection
characterize;
381
W Callebaut and R. Pinxten reds.), Evolutionary Epistemology, 381-401.
/987 by D. Reidel Publishing Company.
of the
P. VANPARIJS
382
mechanisms. I
(a)
383
transmitted)
biological
characteristics
are (genetically
or
(socially
organisms or groups.
The four
cases obtained by
crossing these two
distinctions
correspond roughly
to
the neo-Darwinian
synthesis
(biological/individual),
to so-called "group
selectio~'
(biological/collective), to so-called "cultural
ethology"
(cultural/individual)
and
to
"system
functionalism"
(cultural/collective) (Van Parij s, 1981,
section 22). Is it in principle possible for natural
selection, thus described, to operate on what we usually
call "beliefs" (or "ideologies", lIideas", etc.)?
One reason why this might not be possible is that
beliefs, unlike practices, do not directly produce external
effects on which natural selection could have a grip. A
belief
can only
conjunction with
the belief
that
having
term in this
lots of children
guarantees one's
unless those who hold that belief also value salvation. What
matters to natural selection, however, is differential
consequences, whether joint or not, whether direct or not.
It is enough, for natural selection to operate, that
It is not required by
beliefs,
the
practices if it
differentiate between
on that belief.
Since this
384
P. VAN PARUS
natural-selection
models in industrial
economics, however,
courts) can be
In the case of a
with reproductive
genetic feature,
success.
But in
of a cultural
385
process in
its pure
form
is an evolutionary
mechanism as
moreover,
either,
that
it
is
not
mechanism
of reinforcement
386
P. VANPARIJS
by
social
scientists
in
terms of inertia
and contact: a
This
kind
of
explanation
tends
to
become
between contrary
sanctions, a change
communication",
as
reviewed,
for
example,
by
Simons,
387
is true,
appealed to.
It is a remarkable feature of most social-scientific
theories of belief systems that they tend to define their
specific objects in two complementary ways - or to oscillate
between two such definitions. The beliefs they are concerned
with
are,
on
characterized as "distorted",
function"
388
P. VANPARIJS
Qvergeneralization,
as
having
no
more
than
which,
in
the absence of
interference by other
eyes,
when
belief
these states -
point
of
i.e.
systems
(systematically)
deviate from
epistemic definitions
of "ideology", "illusion",
a) because
389
The
"ideology",
advantage
of
the
functional
definitions
of
it covers all those cases and that they point straight away
to the kind of explanation which is being proposed. Of
course, the same kind of explanation could also be expressed
without making use of a functional definition, for example
by stating that some ideologies, illusions, prejudices, etc.
(more neutrally defined) perform certain functions which
account for their existence. But functional definitions make
were at work.
absence
of
"evolutionary
attraction",
beliefs
will
constitutes
a significant difference
between
the
case of
of some
390
P. VANPARIJS
Its
these
two
perspectives,
there
are obvious
391
wishes,
goods or
in
an
impending
exchange
of roles
practices,
evolutionary
explanations
can
be
392
P. VANPARIJS
the
implicit
criterion
of
optimality
is
if
one
is
to
make
sense
of
the
various
functional
organisms
conditioning.
as
generalization
of
operant
of "chances of satisfaction",
instead of "chances of
reproduction",
as the relevant maximand. Why not use
introduced earlier,
whereas reinforcement assumes that
beliefs converge on cognitive attractors, it seems essential
coincide
entertained because
said to be
satisfying, not
393
operations
of
"displacement",
"condensation",
"imbalance",
etc.
394
P. VANPARIJS
395
beliefs
by
the
interests
of
its ruling
positions
of their
Moreover,
just
396
P. VANPARIJS
transmission", i.e.
either
that
through
the selective
selective reproduction,
natural extension of the
realm is no longer an
397
of accommodation
by
which
we
end
up entertaining
"comfortable" beliefs (I discuss this sort of example as a
threat to the exhaustiveness
of the natural-selection
/reinforcement classification in Van Parijs, 1982). But can
we just subsume it under the second category, along with
accommodation, operant conditioning and adaptive preference?
(3) What is the inner structure of the second category
of evolutionary
mechanisms?
Is the
fundamental difference
398
P. VANPARIJS
however,
that cooking habits,
for example, could be
reinforced by their immediate consequences for the pleasure
people take in eating, as well as by tbeir more mediate
impact on their health.
(4)
Last but not
least,
what do we make of
"evolutionary
epistemology"
in the Popperian
vein?
Is an
for useful
comments
on earlier
NOTES
(1) This may suggest the existence of an analogy between
the pairs beliefs/practices and genotypes/phenotypes.
But
such an analogy would be misleading.
What corresponds to
the genotype/phenotype distinction in the realm of cultural
evolution is rather the distinction between what linguists
call "competence" (an internalized set of rules) and what
they call "performance" (the application of those rules in
actual speech)
or,
more
generally,
between
norms
and
level,
399
and only
in
so
far
as
this
level
A notable
exception
is
the
so-called "strong
(6)
reference
labelled
400
P. VANPARIJS
of ideology as any
social or individual
against
empirical
belief
observations
because
is
quite
it generates
(9)
At least providing it can be assumed that
maintaining them involves a cost. If this is the case, this
cost could be incorporated into the evaluation of the
consequences and the absence of the practice could then be
viewed as the evolutionary attractor. Even so, however,
beliefs would still remain different from practices in so
far as there is nothing like "cognitive attraction" applying
to the latter.
(10) See Runciman (1969, p. 173-174) for an explicit
formulation of this presumption in the context of the
explanation of religious beliefs.
(11) From an (NS-)evolutionary perspective, the very
existence of a mechanism pulling beliefs away from cognitive
attractors
is pretty puzzling.
For
whereas one can
understand that natural selection may have "wanted" to
transform weak cognitive attractors into strong evolutionary
attractors (to use an expression aptly suggested to me by
Peter Richerson), it is harder to see why and how a
mechanism could have come about and maintained itself which
generates a systematic discrepancy between what one would be
right in believing and what one takes pleasure in believing.
The answer is bound to be, roughly, that it is not always in
one's (true) interest to know the truth - or to know that
there is no truth. What I want to do in this paper, however,
401
to
substitute
consonance
for
dissonance,
see
etc.).
The
Part V:
Bibliographies
the
most
complete
previous
bibliography
(Campbell,
D.T., 1974,
Evolutionary Epistemology. In P.A. Schilpp (Ed. L The philosophy of Karl
R. Popper. The library of living
philosophers.
LaSalle, IL: Open
Court Publlshlng
Company.
Volume 14 1,
413-463.),
D. T. CAMPBELL ET AL.
406
Baldwin.
J.M. (1909). Darwin and the humanities. Baltimore:
Publishing Co. Reissued, New York: AMS Press, 1980, 118 pp.
Review
*Barr, H.J. (1964). The epistemology of causality from the point of view of
Bartley, W.W., III. (1976). The phl.losophy of Karl Popper, I: Biology and
evolutionary epistemology. Philosophia, 6(3/4), 463-494.
Bartley, W. W. , III. (1982). Philosophy of biology versus
philosopby of
407
Bjerring, A.K. & Hooker. C.A. (1979). Progress and process: The nature
of systematic inquiry.
In J.
Biirmark (Ed.). Perspectives in
metascience. Goteborg: Kungl Vetenskaps-och vitterhets-Samhallet . J
57 74.
Blachowitz, J. A. (1971). Systems theory snd evolutionary models of the
development of science. Philosophy of Science J 38 J 178-199.
Blazek, B. (1978). On scope and l~ml.ts of analog1es between evolution aDd
cognition. Proe. Syrnp. Natur. Select, Prague, 543-558.
Blazek, B. (1979). Can epl.stemology as a philosophical discipline develop
into a science? Dialectica, 33, 87-108.
Bloor,
Boden,
D.
(1983).
Macmillan, 175
Wl.ttgenstel.n:
&
A social theory of
knowledge. London:
408
D. T. CAMPBELL ET AL.
Burkhardt,
R.W..
Jr.
(1982).
The
evolutionary
development
of
an
Burks,
Urbana, IL 61801.
A.W. ( 1977). Chance, cause, reason: An inquiry into the nature of
Callebaut,
kritische
en
(!
409
Campbell, D. T.
(1986a). Science's social system of validity-enhancing
collective belief change and the problems of the social sciences. In
D.W. Fiske & R.A. Shweder (Eds.), Metatheory in the social sciences:
Pluralisms and subjectivities. Chicago: Unl.versl.ty of Chl.cago Press,
108 135.
Campbell, D.T.
(1986b). Science policy from a naturalistic sociological
epistemology. In P.O. Asquith & P. Kitcher (Eds.),
PSA 1984,
2. East Lansing, MI: Philosophy of Science Association.
Campbell, D.T. (1987). Neurological embodiments of belief and the gap in
the fit of phenomena and noumena. In A. Shimony, D. Nails, & R.S.
Cohen (Eds.), Naturalistic epistemology: A symposium of two decades.
Dordrecht & Bas ton: Re.1del.
Campbell, D. T. (1 987b). Selection theory and the sociology of scientific
validity. (This vOlume.)
Caporael, L.R. (1987). Homo sapiens, homo faber, homo socians: Technology
and the social animal. (This volume.)
Carrier, L. S. (1 9aO). Perception and animal belief. Philosophy, 55,
193~209 .
Cellerier,
G. (1984). Genetic epistemology Human Development, Sept.
(Special double issue on ontogenesis and
phylogenes.1s organized
by Archives Jean Piaget.)
Changeux, J.-P., Heidmann, T., & Patte, P. (1984). Learning by selection.
In P. Marler & H.S. Terrace (Eds.), The biology of learning. Berlin &
New York: Springer, 115-133.
Cherniak, C. (1986a). Limits for knowledge. Philosophical Studies, 49,
1-18.
Cherniak, C. (1986b). Minimal Rationality. Cambridge, MA: MIT Press
(Bradford Books).
*Child, A. (1946). On the theory of the categories. Philosophy and
Phenomenological Research, 7, 316-335.
Chomsky, N. (1966). Cartesian linguistics: A chapter in the history of
rationalistic thought. New York & London: Harper & Row.
Chomsky, N. (1968). Language and mind. New York: Harcourt, Brace & World.
Churchland, P.S. (1986). Neuroph11osoph : Toward a unified science of the
mind-brain. Cambridge, MA: MIT Press Bradford Books .
Churchland, P. S. & Church land , P.M. (1983). Stalking the wild epistemic
engine. Nalls, n( 1), 5-18.
Churchland, P~ (1979). Scientific realism and the plasticity of mind.
Cambridge: Cambridge Univers.1 ty Press.
Churchland,
P.M.
(1984).
Matter and consciousness: A contemporary
introduction to the philosophy of mind. Cambridge, MA: MIT Press.
Churchland, P.M. ( 1985). The ontolog~cal status of observables: In praise
of the superempirical virtues.
In P.M.
Churchland & C.A. Hooker
(Eds.), Images of science.
Chicago: University of Chicago Press,
35-47 (esp. 45).
Church land , P.M. & Hooker, C.A. (Eds.). (1985). Images of science: Essays
on realism and empiricism. Chicago: Chicago Univers~ty Press.
Clark, A. J. ( 1982). Evolut~onary epistemology and the dialectics of
meaning. In P. Weingartner & J. Czermack (Eds.), Epistemology and
philosophy of science. Wien: Holder-Pichler-Tempsky.
Clark, A. J. ( 1984). Evolutionary epistemology and ontological realism.
Philosophical Quarterly, 34, 482-490.
Clark, A.J. (1983). Meaning and evolutionary epistemology. Theoria, 49,
23-31.
--Clark,
A.J.
(1985). Natural anti-realism. Unpublished Ph.D. thesis,
University of Stirhng, Scotland.
Clark, A.J. (1986). Evolutionary epistemology and the
scientific method.
Philosophica, 37, 151-162.
Clark, A.J. (1987). The philosophical significance of an evolutionary
epistemology. (This vOlume.)
410
D. T. CAMPBELL ET AL.
Cohen,
J.L.
(1973).
Is the
progress
of
science
evolutionary? British
41-61 .
---
J.L.
1974. Professor Hu 1 and the evolution of science.
British Journal for the Philosophy of Science. 25(4). 334-336.
Constant,
E.W.
II.
(1978). On the d~versity and co-evolution of
Cohen,
technological
multiples:
Steam turbines
and
Corning,
P.A.
('983).
of
the
A theory of progressive
e~~F~em&lfo~li:~~a~~
(~~~~).e~~~dnUng
tir.r.:
durch Fluktuation": Hine
evolutionare Epistemo!ogie flir menschliche Systeme. Familiendynamik,
6(2), 104-122.
De Hey. M. (1972). Kennis en evolutie: Vergelijkende psychologie van
cognitieve processen. In L. Apostel et a1. (Eds.). Eenheid van de
kultuur. Meppel: Boom.
De Me~ (1982). Genetic epistemology and integrated science of
science. Revue Internationale de Philoso hie. 36(142-3), 636-679.
Dennett, D.C.
1971. IntentJ.onal systems. Journal of Philosophy. 68(4),
87-106. Dennett, D.C. (1978). Brainstorms: Ph~losoph~cal essays
on mind and psychology. Montgomery, VT: Bradford Books, MIT Press.
253 pp.
Dennett, D.C. (1981a). Three kinds of intentional psychology. In R. Healey
(Ed. L Reduction, time and reality. Cambridge: Cambridge University
Press, 37 61.
Dennett, D.C. (1981b). True believers: The intentional strategy and why it
works. In A.F. Heath (Ed.), Scientific explanation. Oxford: Oxford
University Press.
Dennett, D.C. (1983). Intentional systems in cognitive ethology:
The
"Panglossian paradigm" defended. Behavioral and Brain Sciences, 6,
343-390.
Dennett, D.C. (1984a). Elbow room: The varieties of free will worth
wanting. Cambridge, MA: Bradford Books: MIT Press, 183 pp.
Dennet~.
(1984b). Cognitive wheels: The frame problem in artificial
intelligence. In C. Hookway (Ed.), Minds, machines and evolution.
Cambridge: Cambridge University Press, 129-151.
Dennett, D.C. (1986a). Evolution, error and intentionality. Unpublished
manuscript CCM-86-1. Center for Cognitive Studies, Tufts University.
Medford, MA 02155.
Dennett, D.C. (1986b). The myth of original intentionality. To appear in
R. Viale (Ed.), Proceedings of the Conference, Human Mind-Artificial
Mind, Turin, Italy, Apn.l, 1985. Center for Cogn~t~ve Studies, Tufts
unIVersity, Medford, MA 02155.
411
412
D. T CAMPBELL ET AL
~82).
Pour
Jean
Piaget
et 1a
409 434.
Feyerabend, P.K. (1974), Popper. Objective Knowledge Inquiry, 17, 475-507.
Flohr,
Florkin,
M.
(1974).
of
Social
of molecular
Fox, R~ The matter of mind. In R. Fox, The red lamp of incest. New
York: Dutton, Chapter 7, 166-192 & 252-257.
Freeman-Moil", D. J. (1982). Evolutionary epistemology. In J. M.
Broughton
&
D. J. Freeman~Moir (Eds.), The cognitive-developmental psychology
of James Mark Baldwin:
Current theory and research ln genetic
epistemology. Norwood, NJ: Ablex, 98 121.
Frehland, E. (Ed.). (1984). Synergetics - from-microscopic
to macroscopic
order. Berlin: Springer.
Frey, -G-,- (1980).
Moglichkeit
und
Bedeutung
einer evolutionaren
Erkenntnistheorie. Zei tschrift fur philosophische Forschung, 34( 4),
1-17 ,
Furth, H.G. (1981). Piaget and knowledge: Theoretical
foundations (2nd
ed.). Chicago: Universlty of Chlcago Press (1rst ed.: 1969).
Gamble, T. J. (1983). The natural selection model of knowledge
generation:
Campbell's dictum and its critics. Cognition and Brain Sciences, 6,
353-363.
Gellner, E. (1979). Pragmatism and the importance of being earnest. In I.C.
Jarvie & J. Agassi (Eds.), Spectacles
and predicaments. Cambridge,
MA: Cambridge University Press.
Georgescu-Roegen, N. (1971). The entropy law and the economic process.
Cambridge, MA! Harvard University Press. (Chapter I: SClenCe! A
brief evolutionary analysis).
Gerhardt, G. (1983). Wider die unbelebrbaren Empiriker. Die Ar~umentation
gegen empirische VerSl0nen der Transzendentalphl1osophle bel H.
Cohen und A. Rl.ehl. Wiirzburg: Konlgshausen & Neumann.
*Ghisehn, M. T. (1969). The triumph of the Darwinian
method. Berkeley:
University of California Press.
*Ghiselin, M.T. (1973). Darwin and evolutionary psychology. Science, 179,
964-968,
--Giere, R.N. (1984). Toward a unified theory of science. In J.T. Cushing,
C.F. Delaney, & G. Gutting (Eds.), Science and reality, Notre Dame:
Universi ty of Notre Dame Press, 5-31.
Giere, R.N. (1985). Philosophy of science naturalized. Philosophy of
Science, 52, 331-356.
Gillieron, C. (1987). Is Piaget I s "genetic epistemologyl! eVOlutionary?
(This vOlume.)
Gillispie, C. C. (1968). Remarks on social selection as a factor in the
progressivism of science. American Scientist, 56, 439-450.
Glassman, R.B. (1977). How can so little braln hold so much knowledge?
Applicability of the principle of natural
selection
to mental
processes. Psychological Record, 2, 393-415.
Goldman, A.I. (1975). Innate knowledge. In S.P. Stich (Ed.),
Innate
Ideas. Berkeley: University of California Press.
Goldma~I. (1979). Varieties of cognitive appraisal. Nous, 13, 23-38.
Goldman, A.I. (1985). The relation between epistemology-and psychology.
Synthese, 64,29-68.
Goldma~
(1986). Epistemology and cognition. Cambridge, MA: Harvard
Universi ty Press.
413
Goldman,
277-305.
S.
(1975).
Hahn,
"1"7"2=182.
L.E. &
The
relevance of
psychology to epistemology.
Metaphilosophy, 6(2), 161-176.
Habermas,
J.
&
Luhmann, N.
(1971). Theorie der Gesellschaft oder
Sozialtechnologie - Was leistet dl.e Systemforschung? Frankfurt:
Suhrkamp. (Esp. 364 369)
Hahlweg,
K.
(1983). The evolution of science: A systems approach.
University Microfilm. Ph.D. 1983, The University of Western Ontario.
Canada, 44/10-A, 3086.
Hahlweg, K.
(1986). Popper versus Lorenz: An exploration into the nature
of evolutionary epistemology. In: A. Fine & P. Machamer (Eds.), PSA
1986, 1. East Lansing, MI: Philosophy of Science Association,
Schilpp, P.A. (Eds.) (1986). The philosophy of W. V. Quine.
LaSalle: Open Court.
H. (1984). Some introductory remarks on synergetics. In
E.
Frehland (Ed.), Synergetics: From microscopic to macroscopic order.
Berlin: Springer, 1 3.
Hamlyn, D. W.
(1978). Experience and the growth of understanding. London:
Rou tledge & Kegan Paul.
Harre,
R.
(1979).
Social be in : A theory for social
sychology.
Oxford:
Blackwell
Chapter 15:
Social
change:
Theor~es
and
assumpt ions) .
Harre, R.
(1980) KnOWledge. In G.S. Rousseau & R. Porter (Eds.), The
ferment
of
knowledge:
Studies
in
the
historiography OI
e~ghteenth century
science.
New York
&
Cambridge: Camhrl.dge
Unl.vers~ty Press, 11 54.
Harre, R. (1981a). The evolutionary analogy in social explanation. In U.J.
Jensen & R. Harre (Eds.), The philosophy of evolution. Brighton,
Sussex: Harvester Press, 161~175.
Harre, R.
(1981 b). The conditions for applying an evolutionary model to
the development of science: Commentaries on Mittelstrasse
[sic],
Laudan and Newton-Smith.
In U.J.
Jensen &
R. Harre (Eds.). The
philosophy of evolution. Brighton, Sussex: Harvester Press, 290-29~
Hattiangad1, J.N.
(1973). Mind and the origin of language. Philosophy
Forum, 14, 81-98.
Hattiangacrr,
J.N.
(1979).
Meaning,
reference
and
subjunctive
conditionals. American Philosophical Quarterly, 16(3), 197-205.
Haken,
414
D. T. CAMPBELL ET AL.
Hattiangadi,
J.N.
(1982a).
Knowing
that,
Essence
versus evolution
Duplicated
manuscript.
Dept.
Downsville J Ontario M3J 1P3.
of
and
how (a
Philosophy,
new empiricism).
York
University,
in language. Word,
philosophy
Its scientific
Herber~ (1981).
Marx,
International Philosophical
Plaget
415
D.L.
(1978). Altruism in science: A sociobiological
model of
cooperative
behaviour among scientists.
Animal Behaviour, 26,
685-697 .
Hull, D.L. (1980). Sociobiology: Another new synthesis. In G.W. Barlow &
J. Silverberg (Eds.), Sociobiology: Beyond nature/nurture? AAAS
Selected Symposium 35. Boulder, CO: Westv~ew Press.
Hull, D.L. (1982). The naked meme. In H.C. Plotkin (Ed.), Learning,
development and culture: Essays in
evolutionary epistemo~w
York: Wiley, 273 327.
Hull, D.L.
(1983). Conceptual evolution and the eye of the octopus.
Proceedings of the 7th International Congress of Logic, Methodology,
and Ph~losophy of Sc~ence. July 1 16, 1983, Salzburg, Austn.a.
Hull l D.L. (1984a). Cladistic theory: Hypotheses that blur and grow. In T.
Duncan &
T. Stuessy (Eds.), Cladistic
perspectives
on
the
reconstruction of evolutionary history. New York: Columbl.a University
Press.
Hull, D.L. (1984b). Darwinism as a historical entity. In D. Kahn (Ed.)'
The Darwinian heritage. Wellington, New Zealand: Nova Pacifica.
Hull, D.L. (1984c). Hl.storical entities and historical narratives. In C.
Hookway (Ed.), Minds, machines, and evolution. Cambridge: Cambridge
University Press.
,
Hull, D.L. (1984d). Lamarck among the Anglos. Introduction to reprinted
edition of: J.B. Lamarck's Zoological Philosophy:
An Exposition
with Regard to the Natural
Hl.story of Anl.mals. Chlcago: Chlcago
University Press.
Hull, D.L. (1985). Interactors versus vehicles. Paper presented at ICSEB,
Brighton, England, July 1985.
Husserl, E. (1970). Philosophie der Arithemetik, (Ed. L.Eley; orig. 1890).
(Esp. 359-371.)
Irrgang, B. (1986). Biologie als erste Philosophie? Uberlegungen zur
Voraussetzungproblematik und zum Theoriestatus einer Evolutionaren
Erkenntnistheorie. Philosophische Rundschau, 33(1/2), 103-121.
Isard, W. & Kaniss, P. (1978). Structure, control and language hierarchies
and world organization. Journal of Peace Science, 3(1), 63-91.
Isard, W. & Liossatos, P. (1979). Spatial dynamics and optimal space-time
development. New York: North Holland.
Izuzquiza, I. ( in press). Epistomologia y filosofias de la ciencia
postpopperianas I:
La ep~stemologl.a evoluc~on~sta.
Ov~edo:
El
Basilisco.
Jantsch, E. (Ed.). (1981). The evolutionary vision. Boulder, CO: Westview.
Jensen, U. & Harre, R. (Eds.) (1981). The Philosophy of evolution.
Brighton: Harvester Press.
Johnson,
D.M.
(1983).
Memory
and
knowledge: The epistemological
significance of biology.
American Philosophical Quarterly, 20,
375-382.
Johnston,
T .D. (1982). Learning and the evolution of developmental
systems. In H.C. Plotkin (Ed.), Learning, development, and culture:
Essays in evolutionary epistemology. New York: Wl.ley, 411 442.
Jones, W.T. & Chiaraviglio, L. (1979). Is science an adaptive system?
Behavioral Science, 24, 325-333.
Kaniss, P. ( 1978). Evolutionary change in hierarchical systems:
A
general
theory.
Regional Science Dissertation
and
Monograph
Series, 9. Ithaca, NY: Program
on Urban and Regional
Stud~es,
Cornell University.
Kanitscheider, B. (1981). Wissenschaftstheorie der Naturwissenschaften.
Berlin & New York: De Gruyter.
Kantorovich, A. (1983). The collective a priori in science. Nature and
System,S, 77-96.
Kantorovich, A. (1984). The collective a priori: A genotype/phenotype model
for the
growth of knowledge.
Institute
for
History
&
Philosophy,
Tel Aviv University.
Paper
presented
at
the
416
D. T. CAMPBELL ET AL.
417
418
D. T. CAMPBELL ET AL.
Bou er J CO:
Westv1e~
Press.
Lorenz.
K.
(1982).
contemporary
Learning,
e l.stemo 0
&
Kant's doctrine
biology.
of
Reprinted
the a priori
in
H.C.
in
the
Plotkin
light
of
(Ed.).
develo ment,
and
culture:
Essays
in evolutionar
ew Yar: Wl ey.
j a so 10
. . vans
. an
Intervl.ew r , Konrad Lorenz: The man and his ideas. New York:
Harcourt Brace JovanOvl.Ch, 1975, lS1-217.
Lorenz, K. & Wuketits, F.M. (Eds.) (1983). Die Evolution des Denkens.
Munchen: Piper. 394 pp.
Lorenz, K. (1987). Das Apriori-Problem.
In R. Riedl & F.M. Wuketits
(Eds.), Die evolutionare Erkenntnistheorie. Berlin: Parey.
Law. R. (1984) The metaphysl.cal ll.Dll.ts of evolutionary epistemology. In
F.M.
Wuketits (Ed.),
Concepts and approaches in evolutionary
epistemology. Dordrecht & Boston: Rel.del, 209-230.
Law, R. &. Spaemann, R. (Hg.). (1983). Evolution und Erkenntnis-Tragweite
und Grenzen der evolutionaren Erkenntnistheorie in philosophischer
Absicht. In K.
Lorenz &
F.M.
Wuketits (Hg.). Die Evolution
des Denkens. Munchen: Piper, 330-331.
Lowenhard,
P.
(1981)."
Consciousness:
A biological view. Goteborg
Psychological Reports. 11 (10), 1-88.
Lowenhard, P. (1982). Knowledge, belief and human behaviour. GOteborg
Psychological Reports, 12( 11), 1-71.
Lowenhard,
P.
(1986a) . The mind-body problem: Some neurobiological
reflections. In P. Hoyningen-Huene &
F.M.
Wuketits
(Eds.),
Molecules and or anisms: Problems in reductionism and s stems theor
in 101ogy. Dordrec t & Boston: Re1de
10 preparat10n .
Lowenhard, P. (1986b). Evolutionary epistemology: A commentary. In G.
Radnitzky 5: W.W. Bartley, III (Eds.). Evolutionary epistemology,
theory of rationality and sociology of knowledge. LaSalle, IL: Open
Court.
Luhmann, N. (1975). Soziologische Aufklarung, 2: Aufsitze zur Theorie der
Gesellschaft. Opladen: Westdeutscher Verlag. (Esp. 199-200).
Luhmann, N. (1984). The differentiation of advances in knowledge: The
genesis of science. In N. Stehr & V.
Meja (Eds.)' Society and
knowledge. New Brunswick, NJ: Transaction Books.
Lutsky~5). Relativism and evolutionary objectivity in the
development of scientific knowledge. Dept. of Psychology, Carleton
College, Northfield, MN 55057. Duplicated manuscript presented to
the Fifth Annual Symposium of the Jean Piaget Society, June 13, 1975.
Lilt terfelds, W. (1982). Kants Kausalkategorie - ein stammesgeschichtliches
Aposteriori? Philosophia Naturalis, 19, 104-124.
Liltterfelds,
W.
(Ed. ).
( 1986).
Transzendentale
oder evolutionare
Erkenntnistheorie? Darmstadt: Wissenschaftl1che Buchgeselischaft.
Lycan, w. (1985a). Epl.stemic value. Synthese, 64, 137-164.
Lycan, W. (1985b). Conservatism andtEe'""Qita base. In N. Rescher (Ed.),
Reason and rationalit in natural science.
MacCormac, E. .
. Hume s embo 1e l.mpressions. Southern Journal of
PhilOSO~hY, 18(4), 447-462.
MacCormac, E. . (1983). Religious metaphors: Mediators between
biological
and cultural evolution that generate transcendent meaning. Zygon,
18(1), 45-65.
-MacCormac, E.R. (1985). Metaphor as a knowledge process. Chapter 5 in E.R.
MacCormac, A cognitive theory of metaphor. Cambridge, MA: MIT Press.
E.
(1976).
Knowledge and
error:
Sketches
419
on the psychology of
Erkenntnls und Irrtum,
Clarendon.
Maturana,
In
G.A.
of
reality.
Maxwell~al~~att'968j. thSc~!~~~icBO:!~~~d~~~:;I~d
1977 .
D. T. CAMPBELL ET AL.
420
Mohr.
Mohr,
H.
(1983d).
Forschun s rogramm.
sltzungsberlchte Heldelberg:
Wl.ssenscha ten Mathemat. naturwiss. Kl.), 223-232.
AkademlB
ein
der
Honad,
J.
(1970).- L8 hasard
et
18
n~cessite.
Paris: Seui!.
D.
&
Concepts,
421
In
H.
Feigh,
M.
Scriven,
of
science as a
and G. Maxwell
working
(Eds.),
Ott,
J.A., Wagner, G.P. & Wuketits, F.M. (Eds.) (1985). Evolution, Ordnung
und Erkenntnis. Berlin-Hamburg: Parey.
Petrie, H.G. (1981). The dilemma of enquiry and learning. Chicago & London:
University of Chl.cago Press.
*Piaget, J. (1950). Introduction a I' epistemo!ogie genetique, 3 volumes.
Paris: Presses Unlversltalre de France.
Piaget, J. (1980) Adaptation and intelligence: Organic selection and
pbenocopy. Chicago: Unl.versl.ty of Chl.Cago Press.
Piaget~971).
Biology and knowledge. Chicago: University of Chicago
Press.
Piaget~ (1972). The principles of genetic epistemology. New York: Basic
Books.
Piattelli-Palmarini, M. (Ed.). (1980). Language and learning: The debate
between Jean Piaget & Noam Chomsky. Cambrl.dge: Harvard Onl.versl.ty
Press.
Piattelli-Palmarini,
M.
(1981).
Against
evolutionary arguments in
epistemology. Florence Center for the History
and the Philosophy
of Science, Florence, Italy. Paper delivered at the Symposium "Karl
Popper et la Science d'Aujourd'hui", Cerisy, July 1-11,1981, (1984
revision) .
Plotkin, H.C. (1981). The ~volution of closed and open programmes of
development. In D.R. Garrod & J.D. Feldman (Eds.), Development in
the nervous system. Cambridge: Cambridge University Press, 369 389.
Plotkin, H.C. (Ed.) (1982a). Learning, development and culture: Essays in
evolutionary epistemology. New York: Wl.ley.
Plotkin, H. C. ( 1982b). Evolutionary epistemology and evolutionary theory.
In H. C. Plotkin (Ed.), Learning,
development
and
culture:
Essays in evolutionar epistemolo . New York: Wl.ley, 3 13.
approach to the
evolution of intelligence. To appear in the Journal of Human
Evolution (a special issue on the evolution of intel1l.genceJ, 16(1).
422
D. T. CAMPBELL ET AL.
Plotkin,
H.C.
(1987d).
Game
theory,
hierarchy
POlJ,ack~J.s~~mot1ek3) ~ewF;~~k~e~~o~~io:ic~l7;~escriPtion
to socia-historical
description:
The changing metascientific
discourse. Fundamenta
Scientiae, 4(1), 1-27.
*Poppe~
(1959).
The logic of scientific discovery. London:
Hutchinson.
*Popper, K.R. (1963). Conjectures and refutations. London: Routledge &
Kegan Paul.
*Popper, K.R. (1972). Objective knowledge: An evolutionary approach.
Oxford: Clarendon Press.
Popper, K.R. (1977). Die beiden Grundprobleme der Erkenntnistheorie.
Tiibingen: J.C.B. Mohr (Pau! S1ebeck).
Popper,
K.R. (1978). Natural selection and the emergence of mind.
Dialectica, 32(3/4), 339-355.
Popper, K.R. (1981). The rationality of scientific revolutions. In I.
Hacking (Ed.)' Scientific revolutions. OXford: OXford University
Press.
Popper, K.R. (1986). Evolutionary epistemology. In G. Radnitzky & W.W.
Bartley III (Eds.), Evolutionary e istemolo , theor of rationality,
nd
evolutionaren
Erkenntnistheorie.
In R.
Riedl & F.M. Wuketits (Eds.), Die
evolutionare Erkenntnistheorie. Berlin: Parey.
Pribram, K.R. (1984). EvolutJ.on and the mind/body problem. Paper presented
at the
annual meeting
of the American
Association for the
Advancement of Science, May 27, 1984, New York, NY.
Prigogine, I. (1987). The meaning of entropy. (This volume.)
*Pringle, W.S. (1951). On the parallel between learning and evolution.
Behaviour, 3, 175-215.
Putnam~1979). The place of facts in a world of values. In D. Huff &
O. Prewett (Eds.), The nature of the physical universe. New York:
Wiley.
Putnam, H.W. (1983). Why reason can't be naturalized. Synthese. 52, 3-23.
Reprinted in H.W. Putnam, Realism and reason. CambrJ.dge: Cambridge
University Press, 229-247.
*Quine, W.V. (1969). Ontological relativity. New York: Columbia University
Press, 126-128.
Quine, W.V. (1973). The roots of reference. LaSalle, IL: Open Court.
Quine, w.v. (1975). The nature of natural knowledge. In S. Guttenplan
(Ed.), Mind and language. OXford: Clarendon.
Radnitzky, G. (1982). Popper as a turning point in the phi1oso~hy of
science: Beyond foundationalism and relativism. In P. Levinson (Ed.),
In pursuit of truth: Essays in honour of Karl Popper's 80th ~irthday.
AtlantJ.c HJ.ghlands, NJ: HumanJ.tl.es Press & BrJ.ghton: - arvester
Press.
Popper
423
culture.
BioScience. 34(7). 430-434.
Riedl,
R.
(i 978).
Uber
die
Biologie des
Ursachen-DBnkens: Ein
evolutionistiscber, systemtheoretischer Versucb. Mannbeimer Forum,
9-70.
Riedl. R. (1980). Biolo ie der Erkenntnis: Die stammes eschichtlichen
Grundlagen der Brnun t.
WJ.t
t e co a oratJ.on 0
. aspar.
Berlin: Parey.
Riedl,
R.
143-154.
Riedl.
R.
(1984a).
Evolution and evolutionary
knowledge: On the
correspondence between cognitive order and nature.
In
F .M.
Wuketits (Ed.), Concepts and approaches in evolutionary e istemology.
Dordrecht & Boston: ReJ. e , 5- .
Riedl. R. (1984b). Biolo
of knowled e: The evolutionar basis of reason.
(With the col a oratJ.on 0 R. Kaspar; translate
rom t e r German
Edition by Paul Foulkes.) New York: Wiley.
Riedl, R. (1985). Die Spaltung des Weltbildes. Berlin: Parey.
Riedl, R. (1987). AdaptJ.erungsmangel der menschlichen Vernunft. In R.
Riedl & F.M. Wuketits (Eds. L Die evolutionare Erkenntnistheorie.
Berlin: Parey.
Riedl. R. & Kreuzer. F. (Eds.) (1983). Evolution und Menschenbild.
Hamburg: Hoffmann und Campe.
Rorty, R. (1979). The unnaturalness of epistemology. In D.F. Gustafson &
B.L. Tapscott (Eds.)' Body. mind, and method. Dordrecht & Boston:
Reidel, 77-92.
Rorty. R. (1980). Philosophy and the mirror of nature. Oxford: Blackwell.
D. T. CAMPBELL ET AL.
424
of science.
Conference,
June 1981 J
Rosenberg,
J.F.
evolution
Dept. of
psychology:
Ruse,
2W1.
M.
(1985). Evolutionary epistemology: Can sociobiology help?
Synthese.
Ruse, ~85/86). Evolutionary epistemology: Let's be Darwinian not
Spencerian. Newsletter: For those interested in the philosophy of
Karl Popper, 2(3/4), 6-8.
Ruse, M. (1986). Takin Darwin seriousl: A naturalistic a roach to
philosophy.
Ox or :
ac e .
sp.
c
vo utl.onary
Epl.stemology", and ch 5: "Darwinian epistemology").
Russell, C. & Russell, W.M.S. (1962). Raw materials for a definition of
mind. In J.M. Scher (Ed.), Theories of the mind. New York: Free
Press, 535-571.
Russell, J. (1978). The acquisition of knowledge. New York: St. Martin's
Press.
Schafer, w. (1978). Normative Finalisierung. Eine Perspektive. In G. Behrne
et aI., Starnberger Studien, 1: Die gesellschaftliche Orientierung
des wissenscbaftll.cben Fortscbrl. tts. Frankfurt: SUhrkamp, 377-415.
Translated as Normatl.ve
hnall.zation,
in
W.
SchAfer (Ed.),
Finalization in science: The social orientation of
scientific
progress. Dordrecbt & Boston: Re~del, 1983, 207-231.
Schell, J. & De Wae!e, D. (1987). The exchange of genetic informatio,n
between organisms of distinct origin. (This volume.)
Seitelberger, F. (1983). Die evolution zur Erkenntnis: Leistungspotenzen
In P.
W~ingartner &
425
*Sellars, W.S. (1963). Science, perception and reality. London: Routledge &
(Eds.),
Scientific inquiry
Jossey-Bass, 98-123.
Shrader,
426
D. T. CAMPBELL ET AL.
Steele,
E.J.
(1981).
evolution:
On the
Stegmiiller,
W. (1984). Evolutionare Erkenntnistheorie, RealislIlus und
Wissenschaftstheorie. In R. Spaemann, P. Koslowski, & R. Low (Hrsg.),
Evolutionstheorie und
menschliches
Selbstverstandnis. Weinheim:
Verlag Cheml.e, 5-34.
*Stemmer, N. (1971). Three problems in induction. Synthese, 23(2/3),
287-308.
Stemmer,
N.
(1975a).
---
relative
notion
of
natural
generalization.
Stemmer,
N. (1978b). A partial solution to the Goodman
paradox.
Philosophical Studies, 34, 177-185.
Stemmer, N. (1919). Pro]ectlble predicates. Synthese, 41, 375-395.
Stemmer,
N.
(1981a).
Generalization
classes
as alternatives for
similarities and some other concepts. Erkenntnis, 16. 73-102.
Stemmer,
N.
(1981b).
The
objective
confl.rmation
of hypotheses.
Canadian Journal of Philosophy, 11, 395-404.
Stemme~39_~53. (1982). A solutl.on to the lottery paradox. Synthese. 51.
Stemmer.
N.
(1983a).
Blackwell.
is:istemology
Palo
Alto,
CA:
Stent.
Press, 249~267.
Stroud, B. (1980). The si nificance of naturalised epistemolo ., Midwest
Studies in P l. asap y,
.
epr1nted l.n
. Korn 1t
(Ed.),
Naturalizing epistemology. Cambridge, MA: MIT Press, 71-89.
427
Stroud,
B.
(1984).
Naturalized
epistemology.
The
significance
of
Tennant,
---
N.
Holder-Pichler-Tempsky, 168-17 .
Tennessen, H. (1973). Knowledge versus survival. Inquiry, 16, 407-414.
i'
D. T. CAMPBELL ET AL.
428
Vollmer J
edition, 1983).
Vollmer, G. (1983). On supposed circularities in an empirically oriented
epistemology.
In
International
Cultural
Foundation
(Ed.),
Vollmer I
by
C198tlt,
WachtershaUser,
G.
(1986).
On
429
In G.
Radnitzky &
W.W.
Bartley,
III
CEds. L
Evolutionary
epistemology,
theory
of
rationality, and the sociology of knowled e. LaSalle, IL: Open Court.
Wagner, G.P.
.
e ogl.ca baS1S 0 eva utionary epistemology. In
F .M. Wuketits (Ed.), Concepts and
a;roaches in evolutionary
Wagner
der evolutionaren
Erkenntnistheorie. In K. Lorenz & F.M. Wuketits (Eds.)' Die Evolution
der Denkens. MUnchen; Piper, 199-214.
Wagner,
G.P.
(1987).
Die
logische
Begriindung
der evolutioniren
Erkenntnistheorie.
In R.
Riedl &
F.M. Wuketits CEds.), Die
evolutionare Erkenntnistbeorie. Berlin: Parey.
*Wartofsky, M.W. (1967). Metaphysl.cs as heuristic for science. In R. S.
Cohen & M. W. Wartofsky (Eds.) In memory of Norwood Haneson.
Dordrecht: Reidel, 164-170.
Wartofsky, M. W. (1971). From praxis to logos: Genetic epistemology and
physics.
In
T.
Mischel
(Ed.),
COjnitive
development
and
epistemology. New York & London: Academic ress, 129 147.
Wartofsky,
M. W.
(1979).
Models:
Representation
and
scientific
understanding. Dordrecht & Boston: Rel.de1.
Wartofsky, M.W. (1982). Piaget's genetic epistemology and the
Marxist
theory
of knowledge.
Revue
Internationale
de Philosophie,
36(142/3), 470-507.
*Watanabe, S. (1965). Une explication mathematique du elassement d I objets.
In S. Dockx & P. Bernays (Eds.), Information and prediction in
science. New York: Academic Press. 39-45.
Watt. w.c:--( 1979). Against evolution. Linguistics and Philosophy, 3.
121-137.
Weeder. P. & Kester, D. (1982). Variation and selection,
the construction
of an industrial product: The case of Tenax. Dynamics of Science, 2,
282-310.
Weingartner. P. & Czermak, J. (Eds.). (1983). Epistemology and
Philosophy
of Science. Erkenntnis - und Wissenschaftstheorl.e. Wl.en: HolderPl.chler-Tempsky. Sectl.on 1. 5 (Evolutlonary Epl.Stemology).
430
D. T. CAMPBELL ET AL.
Wimsatt,
w.e. (1980b). Reductionistic research strategies and their
biases in the units of selection controversy. In T. Nickles (Ed.),
Scientific discovery: Case studies. Boston & Dordrecht: Reidel,
213-259.
Wimsatt, W.C. f1986a). Heuristics and the study of human behavior. In D.W.
Fiske
&
R.A.
Shweder (Eds.), Metatheory in social science:
Pluralisms and subjectivities. Chicago: Unl.vers1ty of Ch1cago Press,
293-314.
431
Wuketits.
and
Wuketits.
GENERAL BIBLIOGRAPHY
Ackerman~
N. W. & Jahoda, H. (1950). Antisemitism and emotional disorder.
New York: Harper.
Adorno, T.W., Frenkel-Brunswick, Eo, Levinson, D.J. & Sanford, R.N. (1950) .
.!h!. authoritarian personality. New York: Norton, 1969.
Allen, P.M.. Engelen, G. & Sanglier, H. (1984). New methods for policy
exploration in complex systems, communication to a UNU conference at
Hontpellier. France, under the theme Praxis and Management of Complexity. See also special issue of Environment and Planning,
BI2
(1985), 1, 1-138.
Allport, G.W. (1954). Th! ~ .2! prejudice. New York: Doubleday.
Allport, G.W. & Postman, L. (1947) .TI:!! psychology.2f!:!::!!!!2!.' New York:
Holt.
Alston, W.P. (1976). Two types of foundationalism. Journal.Q! Philosophy,
73, 165-185.
Althusser, L. (1965). Pour Marx. Paris: Kasp~roJ 1968.
Althusser, L. (1969). ~n;-;t' la philosophie. Paris: Haspero.
Althusser, L. (1970). IdeologI; et appareils ideologiques d'Etat. In:
Positions. Paris: Editions Sociales, 1976.
A1thus~ Balibar, E. (1968). !4.!:!!! capital, !, Paris: Haspero.
Apostel, L. (1961). Towards the formal study of models in the non-formal
sciences. In: H. Freudenthal (Ed.), The role of the model in mathematics, natural and social sciences. Dordrecht:Reidel-.-- - - - AposteI:"L. (1977). Discussion~. Inhelder, R. Garcia & J. Voneche
(:~ds.), Epistemologie genetique !!! eqUilibration. Neuchitel & Paris:
Delachaux & Niestle, 61-63.
Aristotle. PhYSiCS. Ed. Ross. Oxford: Clarendon.
Armstrong, D.H. (1981). The !!!E!!:! 2!. !!!!!!! and other essays.
Ithaca:
Cornell University Press.
Arnheim, R. (1971). Visual thinking. Berkeley: University of California
Press.
Arthur, W. (1984). Mechanisms .2! morphological evolution: ~ combined genetic, developmental and ecolosical approach. New York: Wiley.
Babloyantz, A., Salazar, J.H. & Nicolis, C. (1985). Evidence of chaotic
dynamics of brain activity during the sleep cycle. Working paper,
Departement de Chimie Physique II, CP 231, Universite
Libre de
Bruxelles.
Barash, D.P. (1977). Sociobiology and ~. New York: Elsevier.
Barnes. .B. (1974). Scientific knowledge .!!!!! sociological theory. London:
Routledge & Kegan Paul.
Barnes, B. & Edge, D. O!:ds.) (1982). Science.!!! ~ Readings !!! the
sociology 2!.!..!!!!!. Milton Keynes: Open University Press.
Barthes, R. (1957). Mythologies. Paris: Seui!.
Battali. J. (1983). Computional introspection. !! Memo 701, HIT, February.
Bechtel * W. (Ed.) (1986). Integrating scientific diSCiplines. The Hague:
Nijhoff.
Berger, P.L. & Luclemann, T. (1966). The .!.2!!.! construction 2! reality. New
York: Doubleday, 1967.
Bergmann,
G. (1957). Philosophy .2!. !..!!!!!. Madison: University of
Wisconsin.
Beth, E. & Piaget, J. (1961). Epistemologie mathematigue !! psychologie.
Paris: Presses Universitaires de France.
Black, H. (1962). Models!!!!! metaphors. Ithaca, NY: Cornell University
Press.
Black, H.
(1985). Ambiguities of rationality. Paper presented at conference on IINaturalism and Rationality", State University of New York,
Buffalo, March, 1985.
433
GENERAL BIBLIOGRAPHY
434
Bloor, D. (1976). Knowledge and social imagery. London: Routledge & Kegan
Paul.
Boden, M.A. (1979). Piaget. Glasgow: Fontana.
Boden, M.A. (1984). Animal perception from an artifical intelligence viewpoint.
In: C.
Hookway (Ed.), Minds, machines, and evolution.
Cambridge: Cambridge University Pre;;:------Boden, M.A. (1984). Artificial intelligence and biological reductionism. In
M.~W.
Ho & P. Saunders (Eds.), Beyond nea-Darwinism. New York: Academic Press, 317-329.
Bohm, D. (1969). Some remarks on the notion of order. In: C.H. Waddington
(Ed.). Towards ~ theoretical biology, 2. Edinburgh: Edinburgh University Press, 18-60.
Bohr, N. (1924). On the spectrum of hydrogen. Lecture delivered to the
Physical Society of Copenhagen on 20 December 1913. Reprinted in
Bohr. N.. The theory !. spectra ~ atomic constitution. Cambridge.
U.K.: Cambridge University Press, 1-19.
Borel, M.J. & Gillieron, C. La classification des sciences selon Jean
Piaget. (In preparation).
Born, M. (1923). Quantentheorie und St5rungrechnung. Die Naturwissenschaften, 27, 537-550.
Boudon, R. (1984). Le phenomeme ideologique: En marge d'une lecture de
Pareto. L'Annee Sociologique, 34.
Boulding, K.E. (1981). Evolutionary ~. Londen: SAGE.
Bourdieu, P. (1966). Le desenchantement du mende. Paris: Centre de Socielogie Europeenne.--Bourdieu. P. (1971). Genese et structure du champ religieux. Revue Francaise ~ Sociologie, 12, 295-334.
Bourdieu, P., Passeron, J.C. & Chamboredon, J.C. (1968). ~ metier ~
sociologue. Paris: Mouton & Bordas.
Brandon. R.N. (1981). Biological teleology: Questions and explanations.
Studies .!!! the History !ill! Philosophy i. Science. 12. 91-105.
Brandon. R. N. ( 1982) . The levels of selection. In: PSA 1982. 1. East
Lansing: Philosophy of Science Association, 315-32~-Brandon, R.N. (1985). Grene on mechanism and reductionism: More than just a
side issue. In: PSA 1984, 2. 345-353.
Brandon. R.N. & Buria;;:-R~(Eds.) (1984). Genes. organisms. populations:
Controversies ~ the units 2! selection. Cambridge, MA: MIT Press.
Brannigan, G. (1979). The reification of Mendel. Social Studies ! ~,
9, 423-454.
Brannigan.
G.
(1982). The social basis f scientific discoveriel!'
Cambridge: Cambridge University Press.
Brecht. B. (1929-1941/1967). Notizen zur Philosophie. In: Gesammelte ~.
20, Schriften zur Politik und Gesellschaft. Frankfurt: Suhrkamp. 125-
178.---- -
--- -
GENERAL BIBLIOGRAPHY
435
GENERAL BIBLIOGRAPHY
436
Carloye,
2!
and
mathematical
proof.
Scientific
Chomsky,
sity Press.
Columbia Univer-
Church, R. (1984). Popper's "World 3" and the problem of the printed line.
Australasian Journal .2! Philosophy, 62, 378-391.
Churchland, P.S. (1983). Dennett's instrumentalism: a frog at the bottom of
a mug. Behavioral and Brain Sciences, 6, 358.
Cellerier, G. (1980).
The construction of a theory: A prescriptive endeavour, Cahiers 4!!! londation ~:!!!!!. Plaget, 1, 177-199.
Clark, A.J. (1987). The evolutionary constraints on cognitive models (paper
draft).
Clement, J. (1982). Student's preconceptions in introductory mechanics.
~ Journal .2! PhYSiCS, 50,
66-71.
Cohen, G.A. (1978). ~ Marx's theory .2! history: A defence. Oxford:
Oxford University Press.
CODDDons, H.L., Richards, l.A. & Armon, C. (Eda.) (1984). Beyond formal
operations: Late adolescent development and adult cognitive development. New York: Praeger.
Craik, K. (1943). The!!!E!!:! 2!. explanation. Cambridge: Cambridge University Press.
Crombie, A.C. (1967). The mechanistic hypothesis and the scientific study
of vision: Some optical ideas as a background to the invention of the
microscope. In: S. Bradbury & G.L.E. Turner (Eds.), Historical !!.:.
peets of microscopy. Cambridge: Heffer, 3-112.
Dawkins, R. 0976). The selfish &!!!!!.. Oxford: Oxford University Press.
Dawkins, R. (1976). Hierarchical organization: a candidate principle for
ethology. In: P.P.G. Bateson and R.A. Hinde (Eds.), Growing points !!!.
ethology. Cambridge: Cambridge University Press, 7-54.
Dawkins, R. (1982). The extended phenotype. Oxford: Oxford University
Press.
Denbigh, K. (1981). How subjective is entropy?, Chemistry.!!!!.!:ili.i!!., 17,
168-185.
Deneubourg, J.L.. Pasteels, J.H. & Verhaeghe, J.C. (1985). Probabilistic
behaviour in ants: A strategy of errors? Journal of Theoretical
Biology, lOS, 259-271.
--- Dennett, D.
(1982). How to study consciousness empirically: Or, nothing
comes to mind. Synthese, 53, 159-180.
Depew, D.J. (1986). Nonequilibrium thermodynamics and evolution: A philosophical perspective. Philosophica, 37, 27-57.
Descartes, R. (1948). Discours de la ~thode. Paris: Bader-Dufour. (Orig.
1637. )
Descartes, R. (1955). Meditations on first philosophy. In: E.S. Haldane
(Ed.), Th! philosophical .!2.!:!.!.2!.~. New York: Ross.
Dobzhansky, T. (1962). Mankind evolving: The evolution .2! the human .!.2!.:.
eies. New Haven: Yale University Press.
Donchi;;:--X.
(1981). P300 and classification.
In: R. Galambos & S.A.
Hillyard (Eds.), Electrophysiologieal approaches to human cognitive
processing. Neuroscience Research Program Bulletin, 20, 157-161.
437
GENERAL BIBLIOGRAPHY
algebraic
438
GENERAL BIBLIOGRAPHY
GENERAL BIBUOGRAPHY
439
Gruber, H.E. (1979). The changing shape of Pisget's 'circle of the sciences I . Unpublished paper presented at the International School of
History of Science, Ettore Majorana Centre for Scientific Culture,
Ham.lyn, D.W. (1970). The theory .2!. knowledge. London: MacMillan Press.
Hamad, s.
(1985). Bugs, slugs, computors and consciousness. American
Hansell,
M.J.
Lepidostoma
(1972).
hirtum.
Case
study
!.
psycholo-
Journal.2f the
Zoological Society
larva
(London),
167, 179-192.
Rebb, D.O. (1949). The organization .2! behavior: ! neuropsychological theo!I.. New York: Wiley.
Beidegger, H. (1975).!ll.!. Grundprobleme der Phiinomenologie. Engl. tr. A.
Hofstadter. The !!!!! Problems 2! Phenomenology. Bloomington, IN:
University Press, 1982.
Heider, R. &
Simme!, H. (1944). An experimental study of apparent behaviour. ~ Journal .2! Psychology, 57, 243-259.
Heisenberg, W. (1926). Quantenmechanik. lli:!. Naturwissenschaften, 14, 889-
894.
Heisenberg, W. (1975). Discussion with Heisenberg. In: O. Gingerich (Ed.),
The nature of scientific discovery: A symposium commemorating !h!.
500th anniversary of the birth ! ~ Copernicus. Washington:
Smithsonian Press. 556-573.
Henriques. G. (1984). Le preformisme: Critique de ses fondements et de sa
valeur explicative. Archives!!! Psychologie, 52. 53-68.
Hesse. M. (1970). ~.!.!!2 analogies .!!!~. Notre Dame, IN: University of Notre Dame Press.
Hesse, M. (1974). The ~ 2! scientific inference. Berkeley: University of California Press.
Hesse, M. (1980). Revolutions and reconstructions.!!!!h! philosophY .2!.
science. Brighton, Sussex: Harvester Press.
Heyes, C.M. (in preparation). Psychological epistemology. To be submitted
to !mer ican PsycholoB ist.
Hirshleifer, J. (1977). Economics from a biological viewpoint. Journal.Qf
Law and Economics, 20, 1-52.
Ho.
H':'"='W:-& Saunders. P. (Eds.) (1984). Beyond neo-Darwinism. New York:
Academic Press.
Holldobler, B. & Michener, C.D. (1980). Mechanisms of identification and
discrimination in social hymenoptera. In H. Harkl (Ed.), Evolution of
social behavior: Hypotheses and.!!.!ll. Deerfield Beach,~i;i
Chemie, 35-57.
Holmes, W.G. & Sherman. P.W. (1983). Kin recognition in animals. American
Scientist, 71. 46-55.
--Hube1,~963). The visual cortex of the brain. Scientific American,
November.
Hull. D.L. (1974). Philosophy.Q.f biological ~. Englewood Cliffs, NJ:
Prentice-Hall.
Hull, D.L. (1976). Are species really individuals? Systematic Zoology, 25,
174-191Hull. D.L. (1980). Individuality and selection. Annual Reviews of Ecolon and SystematiCS. 11, 311-323.
- - --- --Hull. D.L. (1981). The herd as means. In: ~ 1980, 2. East Lansing. HI,
Philosophy of Science Association. 73-92.
Hull. D.L. (1983). Exemplars and scientific change. In: PSA 1982, 2. East
Lansing, HI: Philosophy of Science Association, 479-503-.Hull, D.L. (1985). Darwinism as a historical entity: A historiographic
proposal. In: D. Kolm (Ed.), The Darwinian heritage. Princeton. NJ:
Princeton University Press, 773-812.
440
GENERAL BIBUOGRAPHY
Duxbury.
Ittelson, W.R. (1952). The ~ demonstrations !!! perception. Princeton:
Princeton University Press.
Jacob, F. (1977). Evolution and tinkering, Science, 196, 1161-1166.
Jacobs, J. (1986). Teleology and reducti~n biology. Biology and
Philosophy, 1, 389-399.
1975.
S.R. &
17-29.
Lakatos, I. (1970). Falsification and the methodology of scientific research programmes. In: J. Worrall and G. Currie (Eds.), The method.2.!s:l 2i scientific research programmes. Philosophical Papers, 1,
Cambridge.
441
GENERAL BIBLIOGRAPHY
Lakatos, 1. (1976). Proofs and refutations:
!!!! logic
diSCOV1ry. Cambridge: Cambridge University Press.
.!!!.
mathematical
442
GENERAL BIBLIOGRAPHY
443
GENERAL BIBLIOGRAPHY
Mesarovic. M.D. (1964). Foundations for a General Systems Theory. In: H.D.
Mesarovic. (Ed.), Y!!!!! 2!l General Systems Theory. New York: Wiley J
1-24.
mard, 781-82l.
studies in creativity.
--
Miller-,-D-.-(1972). Ideology and the problem of false consciousness, Political Studies, 20, 432-447.
-Horgan:-c.L~g4). ~ introduction !Q comparative psychology. London:
Walter Scott.
Horin, E. (1969). La rumeur d'Orleans. Paris: Seu!l.
1l.
Nagel, E.
2!
1979.
Nagel, T.
Philosophical~, 83,
435-45l.
Nelson, K. (1973). Does the holistic study of behaviour have a future? In:
P.P.G. Bateson and P.H. Klopfer (Eds.) Perspectives .!!! ethology, l.
New York: Plenum, 281-328.
Nelson, R.J. (1984). Naturalizing intentions. Synthese, 61, 173-203.
Nickles, T. (1980). Scientific problems: Three empiricist models. In: PSA
1980. East Lansing, HI: PSA, 3-19.
Nickle;:--T. (1980b). Introductory essay: Scientific discovery and the
future of philosophy of science. In: T. Nickles (Ed.), Scientific
discovery. logic !!U! rationality. Boston-Dordrecht: Reidel.
Nicolls, C. and G. (1984). Is there a climatic attractor? Nature, 311, 529532.
-Nicolis, C. & Prigogine, 1. (1977). Self-organization !!!. nonequilibrium
systems. New York: Wiley.
Oakley, D.A.
(1985). Animal awareness, consciousness and self-image. In:
D.A. Oakley (ed.), Brain and mind. London: Methuen.
Oakley, D.A. &
Gormes, 'L.C':"'" m8~The plurality of consciousness. In:
D.A. Oakley (Ed.), Brain and mind. London: Methuen.
Olding, J. (1978). A defenCe'Ot evolillOnary laws. ~ Journal !2!!!!
Philosophy ~~, 29. 131-143.
Osgood, C.E. & Tannenbaum, P.B. (1955). The principle of congruity in the
prediction of attitude change. PsYchological Review, 62, 42-55.
Pareto, V. (1916). Trattato .Q! sociologia generale. Milano; Comunite, 1964.
Parsons. T. (1959). An approach to the sociology of knowledge. In: !!.!!!!:.
~.2f the Fourth World Congress .2f Sociology. Louvain. 25-49.
Pattee, H.H.
(1973). The physical basis and origin of hierarchical control. In: R.H. Pattee (Ed.). Hierarchy theory. New York: Brazi1ler.
71-108.
Patten, B.C.
(1982). Environs: Relativistic environmental particles for
ecology. American Naturalist. 119, 179-219.
Patterson, X.E. & Baddeley, A.D. (1977). When face recognition fails.
~ .2f Experimental Psychology:
~ Learning!!l!!
Memory. 3.
406-417.
444
GENERAL BIBLIOGRAPHY
445
GENERAL BIBLIOGRAPHY
145-150.
Learnins,
adaptation.
development!!!!!~.
In:
H.C.
GENERAL BIBLIOGRAPHY
446
Richel1e, M.N.
(1986). Variation and selection: The evolutionary analogy
in Skinner's theory.
In: S.
Hodgil and C.
Hodgil (Rds.) !:.!.:..
~ ~
Radinson,
H.
(1968).
64, 70-104.
and controversy.
Roitblat, H.L.
(1982). The meaning of representation in animal memory.
Behavioral and Brain Sciences, 5, 353406.
Roitblat, H.L., Bever:--T~ Terrace, B.S.
(Eds.) (1984). ~
Cognition. Hillsdale, N.J: Lawrence Erlbaum.
Roll-Hansen, N. (1983). The death of spontaneous generation and the birth
of the gene: Two case studies of relativism. 2!!! Studies 2f i:.
ence, 13, 481-519.
Romane;:--G.J. (1883). Mental evolution in animals. London: Kegan Paul
Trench.
- - ---- - --Rosenberg, A. (1980). Obstacles to the nomological connection of reasons
and actions. Philosophy.2! Social ~, 10, 79-91.
Rosenberg, A. (1985). The ~ .2! biological ~. Cambridge. U.K.:
Cambridge University Press.
Rosenberg, H. J .A. (1960). An analysis of affective-cognitive consistency.
In: C.L Hovland & H.J. Rosenberg (Eds.), Attitude organization and
change. New Haven: Yale University Press, 15-64.
Rouquette, H.L. (1975). 1!.!!!!!!!!!!!. Paris: P.U.F.
RWlC iman , W.G. (1969). The sociological explanation of 'religious beliefs'.
~ Europeennes ~ Sociologie, 10, 149-191.
Sachsse,
H.
(1979).
KausaliUlt-Gesetzlichkeit-Wahrscheinlichkeit.
Darmstadt.
Sahlins, M. (1976). The ~ and ~ 2! biology. London: Tavistock.
Samuelson, P.A. (19m. Maximum principles in analytical economics. ~
these, 31. 323-344.
Schife;:-1i. (1983). Normative finalization. In: W. Schlfer (Ed.), Finalization !!! ~ .TI!!.!2.i!! orientation .2! scientific progress.
Dordrecht & Boston: Reidel, 207-231.
Schaff, A. (1967). La dlHinition fonctionnelle de l'ideologie. L'Homme ~
la societe.
SchilPP':' ~(Ed.) (1949). Albert ~ philosopher-scientist. La
Salle. IL, Open Court. (third edition, 1969),683-4.
Schilpp, P.A. (Ed.) (1974). Th! philosophy 2!!!!:! Popper. La Salle, IL.
Open Court.
Schrodinger, E. (1961). Meine Weltansicht.
Hamburg-Wien: Paul Zsolnay.
English translation: !:!v !!2ili Y!!!!. Cambridge: Cambridge University
Press. Foreword to the 1960 edition.
Schon, D.A. (1963). Displacement .2! concepts. London: Tavistock.
Seyfarth, R., Cheney. D.L. & Harler, P. (1980). Honkey responses to
three different alarm calls: Evidence of predator classification and
semantic communication. Science, 210, 801-803.
Shallice. T.
(1972). Dual fWlctions of consciousness. Psychological ~
view, 79, 383-393.
Shaper~D. (1984). ~ and the !!!h !.2!: knowledge: Investigations i!2
the philosophY 2!.~. Dordrecht: Reidel.
Shapere, D. (1987). Hethod in the philosophy of science and epistemology:
How to inquire about inquiry and knowledge. In: N. Nersessian (Ed.),
.!!!! processes 91.~. The Hague: Nijhoff (to appear).
Sheldrake. R. (1982). A new science of life. London; Blond and Briggs.
Shweder, R.A. (1986)-:- Divergent r;t'ionalities. In: O. W. Fiske and R.A.
Shweder (Eds.), Hetatheory i!2 social science: P1uralisms and subiec~. Chicago: University of Chicago Press, 163-196.
Silverman. H.J. (Ed.) (1980). Piaget. philosophy and the human~.
Atlantic Highlands, N.J.: Humanities Press.
Silverman, P.S.
(1983). Attributing mind to animals: the role of intuition. Journal g! ~ !!!.2. Biological Structures. 6. 231-247.
GENERAL BIBLIOGRAPHY
447
W.--
Sober, E.
(1984). The nature !. selection: Evolutionary theory i!! philosophical focus. Cambridge, MA: MIT Press.
Sober, E. (Ed.) (1984). Conceptual issues 1!! evolutionary biology. Cambridge, MA: MIT Press.
Sommerhoff. G. (1950). Analytical biology. Oxford: Oxford University Press.
Sommerhoff, G. (1969). The abstract characteristics of living systems. In:
F .E. Emery (Ed.). Systems Thinking. Baltimore: Penguin.
Sperry, R.W. (1969). A modified concept of consciousness. Psychological
Review, 76, 532-536.
Spiro.~ (1966). Religion: problems of definition and explanation. In:
M. Banton (Ed.). Anthropological approaches 12 the study f religion.
London: Tavistock~ 85-126.
Staddon, J.E.R. (1983). Adaptive ~ and learning. Cambridge: Cambridge University Press.
Steen, L.A.
(Ed.) (1978). Mathematics today: Twelve informal essays. Heidelberg: Springer.
Stroud. B.
(1984). The significance ! philosophical scepticism. Oxford:
Oxford University Press.
Symons, D. (1980). Precis of ~ evolution .2! human sexuality (Author's
response). Behavioral and Brain Sciences, 3, 171-214.
Tajfel, H. (1981). Human groups and social categories: ~ i!! ~
psychology. New York: Cambridge University Press.
Tennant. N. (1984). Intentionality and the evolution of language. In: C.
Hookway (Ed.). Minds, machines, and evolution. Cambridge: Cambridge
University Press.
Terrace, H.S.
(1984). Animal cognition. In: H.L. Roitblat, T.G. Bever &
H.S. Terrace (Eds.). Animal cognition. Hillsdale, N.J.: Lawrence
Erlbaum.
Thoday, J .M.
(1975). Neo-Darwinian "evolution" and biological progress.
Nature, 255, 675-677.
Thorn, ~972). Stabilite structurelle II morphogenese. Paris: Intereditions.
Thom, R.
(1974). Modedes mathematigues Q! ~ morphogenese. Paris: U.G.E.
(lD/1S) .
(1975). Dlun modele de la science une science des modeles. ~
Thom, R.
these, 31, 359-374.
Thom, ~1983a). Paraboles ~ catastrophes. Paris: Flammarion.
(Ital.
orig. 1980).
Thorn. R. (1983b). Animal psychism vs human psychism. In: E. de Grolier
(Ed.), Glossogenetics. Paris: Harwood.
Thom. R. (1983c). Semiotique et theorie des catastrophes, Structures cycliques en semiotique. Groupe de recherches semio-linguistiques de
l'EHESS, 10 rue Monsieur Ie Prince, Paris 75006, V47-48. 38-58.
Thompson. M. (1981). Things in themselves. Proceedings.2! the American
Philosophical Association, 57(1).
448
GENERAL BIBLIOGRAPHY
Thompson, P. (1986). The interaction of theories and the semantic conception of evolutionary theory. Philosophica, 36, 73-86.
Tolman, E.C. (1926). A behavioristic theory of ideas. Psychological ~,
33, 352-369.
Tolman, B.C. (1951). Collected papers .!!! psychology. Berkeley: University
of California Press.
Tombaugh, J. (1984). Evaluation of an international scientific computerbased conference. Journal of Social Issues, 40, 129-144.
Toulmin, S.E. (1970). Do~dt;t~n between normal and revolutionary
science hold water? In: 1. Lakatos & A. Musgrave (Eds. >. Criticism
and the growth 2! knowledge. Cambridge: Cambridge University Press,
39-47.
Toulmin, S.E. & Goodfield, J. (1965). Th! discovery.2! l!!!!!. New York:
Harper & Row.
Trivers, R.L. (1971). The evolution of reciprocal altruism. Quarterly
Review 2i Biology, 46, 35-57.
Tversky, A. & lCahnemann, D. (1981). The framing of decisions and the psychology of choice. ~, 211, 453-8.
Ullman. S. (1979). TI!! interpretation 2i Y.!.!!!!! !!2.1!.2!l. New York.
Vandenbrande, R. (1979). La critique litt'raire: Wl espace interme.diaire.
In: W. Callebaut et al. (Rds.). Theory 2i knowledge ~!.i!!!!. policy.
Ghent: Communication & Cognition, 183-189.
van Fraassen, B.C. (1980). The scientific image. Oxford: Clarendon.
Van Parijs, P. (1978). Theorie des catastrophes et mate.rialisme historique,
Revue Francaise 4! Sociologie, 19, 195-220.
Van Parijs, P. (1982). FWlctionalist Marxism rehabilitated. A comment on
In: P. Weingartner
H51der-Pichler-Tempsky, 185-197.
Voneche, J. (1984). Introduction. In: Fondation Archives Jean Piaget (Ed.),
Phylogeny and ontogeny. Human Development, 27 J 227-232.
von Glasersfeld, E. (1974). Piaget and the radical constructivist epistemology. Paper presented at the Third Southeastern Conference of the
Society for Research on Child Development. Chapel Hill, N.C.
Vrba, B.S. & Eldredge, N. (1984). Individuals, hierarchies, and processes:
Towards a more complete evolutionary theory. Paleobiology, 10, 146171.
Waddington, C.H. (1961). The human evolutionary system. In: M. Banton
(Ed.), Darwinism and the study B!. SOCiety. Chicago: Quadrangle Books,
63-82.
Waddington, C.H. (1969). The practical consequences of metaphysical beliefs
on a biologist's work. In: C.H. Waddington (Ed.). ~! ~
.!!!! Biology. 2. Edinburgh: Edinburgh University Press, 72-81.
Walker, s. (1983). Animal thought. London: Routledge & Kegan Paul.
Watson, J. (1966). Crowing up in the Phage group. In: Cairns et a1. (Eds.).
Watson, J. (1968). The double helix. New York.
Whitley, R. (1978)":"'""' Types of science, organizational strategies and patterns of work in research laboratories in different scientific
fields. Social Science Information, 17, 427-447.
Weber, R. (19~Dialogues with scientists ~ sages: The.!.!!!.h for
unity. London: Routledge & Kegan Paul.
Wertheimer, M. (1982). Productive thinking. 2nd ed. Chicago: Chicago Universi ty Press.
Wicken, J.S. (1984). On the increase in complexity in evolution. In: M.W.
Ho & P. Saunders (Eds.), Beyond neo-Darwinism. New York: Academic
Press, 89-112.
Wicken, J.5. (1985). An organismic critique of molecular Darwinism. Journal
2! Theoretical Biology, 117, 545-561.
Williams, C.C. (1966). Adaptation .!m! !l!E!r!!~. New Jersey:
Princeton University Press.
GENERAL BIBLIOGRAPHY
449
Williams, M. (1973). Falsifiable predictions of evolutionary theory. Philosophy .2! Science~ 40, 518-537.
Williams, H. "(""i"9ii') The importance of prediction testing in evolutionary
biology. Erkanntnis, 17, 291-306.
Wilson. H.O. (975). Sociobiology: The!!!!! synthesis. Cambridge, MA;
Harvard University Press.
Wimsatt, W.C. (1981). Robustness, reliability, and multiple-determination
in science. In: M. Brewer & B. Collins (Eds.), Knowing !!!!! validating
Index of Names
Ackermann 388
Adorno 388
Allen 72
Allport 388,394,401
Althusser 387,388,395
Alston 142
Apostel 3,9,22,23,40,265,319
Aristotle 52,59
Armon 281
Arms trong 109
Arnheim 277
Arthur 19
Ashby 146,157
134
Beth 249
Black 123,187
Bloor 159,399
Boden 7,16,129,266
Bohm 77,78,79,91,92,93,95
Bohr 267,268,270,271,272,273,
274,275,276,277,278,279,
317,318,319
Boltzmann 65,207
Boon 18,24,34,37,44,221
Borel 259
Born 65,271,276
Boudon 388
Boulding 24,25,26
Bourdieu 388,391,395,401
Boyd 21,25,30,32,41,47,144,
153,154,384,385,399
Bradie 3,4,8,17,23,46,54,105
Brandon 8,9,13,14,18,19,47,52,
234
Branningan 185,186
Brecht 5,6,50
Brewer 31,155,239
Brooks 11,33
Brown 378
Biichel 15
Bunge 11,19
Burghardt 131,132,134
Burian 18,29,47,234
Burks 141
Buscaglia 257,262
Berry
Babloyantz 64
Bacon 160,207
Baddeley 241
Baldwin 146,157,265,314
Balibar 388
Banach 363
Barash 235
Barlow 342
Barnes 54,145,399
Baron 234,401
Barthes 395
Bateson 145,168
Battali 322
Beausoleil 286
Bechtel 19,39
Beckermann 235,238
Benard 61
Berger 399
Bergmann 10,371
Bergson 59,99
Berkeley 150,224
Berkowitz 386
Bernouilli 344,345,347,360,
361
Bertalanffy 7
Cairns 161,170
Callebaut 8,11,16,19,23,26,29
34,37,39,40,46,51
53,54,55,139,226,227
283,363
451
INDEX OF NAMES
452
1,4,5,9,15,17,18,20
21,22,24,25,27,30,31
32,34,35,36,37,38,39
40,41,43,47,48,53,55
75,81,82,85,86,94,
105,106,113,139,143
144,145,146,147,148
149,152,153,155,156
165,179,180,182,183
189,193,200,204,205
206,223,225,229,230
234,235,236,237,239
240,265,311,312,313
316,317,322,365
Cantor 339,351
Caplan 234
Caporael 24,235,236
Capra 7
Carloye 116,131
Carrier 30
Cartwright 14
CelIerier 264
Chait in 219
Chamboredon 388
Changeux 87
Chardin,de 220
Cheney 117
Cherniak 39
Chernoff 241
Childe 144
Chisholm 142
Chomsky 126,262
Church 34,251
Churchland 25,110,119,121,122,
140
Clark 25,35,51
Clausius 60
Clement 219,378
Cohen 143,221,395
Collins 31
Campbell
Commons 281
Copernicus 73,160,165
Corning 182,189,190,191,200
Cracraft 44
Craik 15
Crick 170,172
Crombie 150
Culbertson 236
Dallo 11,19
Danto 38
Darbishire 145
D'Arcy Thompson 44
Darwin 12,15,73,111,189,212,
374,375,376
Davis 363
Dawkins 19,48,49,53
DelbrUck 161,169
Denbigh 65
Deneubourg 72
Dennett 38,107,115,116,118,119
120,121,123,124,126,
128,129,130,132,135,
140,227
Depew 6,11,33
Descartes 6,17,51,160,223,224,
229
De Waele 26,144
Dirichlet 340
Dobzhansky 239
Donchin 113
Donini 390
Donohue 172
Dretske 30,80,128
Duhem 156
Dyson 266
Eames 132
Eddington 65,278
Edge 54
INDEX OF NAMES
Edwards 339
Eigen 7
Einstein 58,59,99,203,208,
267,272,315,366
Ekeland 366,367,368,369
Eldredge 13,44,79,90,191,
238
Ellis 141
Elster 10,36,52,120,135,197,
387,397,401
Engelen 72
Engels,E. 30
Engels,F. 390
Euclid 51
Euler 337,357,359
Faltings 348,351
Felson 112
Fermat 337-363
Festinger 393
Feyerabend 159,160,180,182
Feymnann 101
Fisher 53
Fiske 113
Flavell 253
Flewer 34
Fodor 263
Foucault 180
Fouille 146
Fox 51
Freeman-Mayer 265
Frege 100
Freud 12,73,366,388,390,393
Furth 220
Furtwangler 348,355
Galambos 113
Galileo 51,100
Gamble 147
Garcia 130
Gauss 341,342,344,353
453
Geertz 179
Georgescu-Roegen 30
Germain 342,343,344,354
Ghiselin 35,78,83
Gibbs 68
Giere 3,200,313,363
Gieryn 152
Gilbert 173,193
Gillieron 23,256,311,313,314
315
Ginsberg 144
Glasersfeld,von 264
Gmelch 112
Godel 219,251,349,350
Golbach 363
Goldmann 142,143,387
Goldschmidt 44
Goodfield 10
Goodwin 47,265,266,313
Gould 13,14,44,148,191,236
237,314
Gray 7
Greco 256
Greimas 102
Grene 8,52,78,90
Griesemer
Griffin
17
107,108,109,110,111
112,113,114,115,116
117,123,128,130,131
Grimes 378
Gruber 247,249
GrUnert 238,340
Haldane 83,224
Hamlyn 224,228
Hansell 108
Harnad 114
Haroutunian 23,42
Harre 11,27,150,151
Hebb 320,321
Heidegger 59
454
INDEX OF NAMES
Heider 112,122
Heisenberg 268,270,274,275,
276,277,278,281,
318
Hempel 13
Henderson
45
Henriques 265
Hesse 187,200
Heyes 16,30,47,55,139
Hilbert 350
Hillyard 113
Hishleifer 25
Ho 16,30,44,45,46,48
Holldobler 235
Holmes 235
Holton 103
Hooker 25
Howe 286
Hubel 307
Hull 10,13,15,16,19,37,43,
44,49,50,52,54,82,133
144,152,154,180,182,
183,187,201,371
Hume 99,150,151,224
Humphrey 110,112
lngvar 64,65
Inhelder 284
Irrgang 17
It telson 150
190
Jacobs 15
Jahoda 388
Jakobson 375
Jantsch 6
Jarvie 390,391
Jaynes
112
Jenning
146
Jacob
Jensen
11
Kahnemann 218,219,243
Kant 17,20,21,39,54,224,228
Kary 143
Kaspar 20,219
Kaufman 8
Kellert 134
Kepler 150,160,165,368,369
Kimura 44
Kitchener 366,370,371,372,373
376,378,379
Kitcher 11
Kleene 251
Knorr-Cetina 3,27,34,37,40,44,
185,188,194,195,
221
Koelling 130
Kolmogorov 67
Kornblith 38
Kramers 275,319
Krasner 347,348,355
Kreuzer
203
Kroodsma 130
Kuhn 15,27,148,159,165,169,
180,192,197,212
Kummer 344,345,346,347,348,
355
Kurland 235,238
Ladriere 252
Lakatos 28,160,166,169,337,
338,351,357,358,363
Lamarck 12,264,374,375,376
Lame 340
Lamontagne 286,307,311,320-322
Landau 238
Lang 387,388
Laudan 154,161
Lavoisier
207
Le Doux 132
Legendre 343
455
INDEX OF NAMES
Leibniz 17,224
Lelas 54
Levine 10,11,12
Levins 41
Levinson 35
Lewin 144
Lewis 65,112
Lewontin 12,13,18,19,40,41,
42,43,45,53,81,82,
148
Lindenberg 8
Lloyd 18,49
Locke 150
Lorenz 4,20,21,22,40,47,48,
54,80,84,94,110,140,
203,204,205,206,211,
213,221,229,262,322
Losee 221
Lovejoy 45
Luckman 399
Lukaszweski 239
Lyapunov 67
MacKenzie 145
Malinowski 390
Mandelbrot 63
Mannheim 387
Marler 117
Marr 120
Martinet 373-376
Martinez 70
Marx 390
Maturana
54
Maxwell 207
Maynard-Smith 50,88,90,127
Mayr 11,19,26,46,48,52,191
McArthur 234
McClelland 51
McCloskey 219
McCulloch 308
McDougall III
McKelvey 52
McMullin 28
Mendel 162,168,170,186
Mersenne 347
Merton 152,180
Mesarovic 15
Meyer 257
Michener 235
Miller 130,281,311,315,316
317,318,319,320,322
387
Mirimanoff 347,348,355
Monod 45,262
Mordell 348
Morgan,T.H. 145,170
Morgan,W. 111,123
Morin 388
Moyer 207,386
Mulkay 187,193
Mulkey 173
Milller 170
Nagel 5,14,134,189
Nelson,R.R. 27,38,91,384
Neurath 51
Newton 53,58,208,361
Nickles 53,194,200,353,363
Nicolis 32,64
Nivnik 247
Oakley 109,132
Odling-Smee 21,22,24,32,48,85,
86,94
Oeser 36,37,38
O'Hear 143,229
Olding 371
Oppenheim 8,13
Orwell 240
Osgood 393
Paller
25
456
Palmarini 262
Papert 308
Pareto 388,401
Parsons 198,387
Pascal 150
Passeron 388,401
Pasteels 72
Pattee 91,92
Patten 82
Patterson 241
Pauli 277
Peano 353
Pearson 145
Peirce 6,7
Piaget 5,6,12,20,22-24,
42,44,47,80,82,
142,149,219,220,
247-267,269,270,
273,275,279,280,
283-285,309,312,
313,314,316,318,
319,320,322,369,
370
Piatelli 262
Pickering 180
Pike 200
Pinard 283,309
Pinxten 53,54,226,227,283
Planck 58
Plotkin 3,9,14,17,19,20,21,
22,23,24,30,32,47,
48,54,79,80,85,94,
140
Poincare 67,147,367
Polanyi 27
Pollock 110
Pomian 98
Popper 5,25,33,38,39,40,97,
98,114,163,203,204,
205,206,213,366,371
Postman 401
INDEX OF NAMES
Premack 135
Prigogine 5,6,10,11,17,32,33,
35,43,45,46,58,70,
73,366
Provine 145,168
Ptolemy 160
Putman 38,134,224,226
Quine
4,21,112,136,140,141,156
225,227,230
Rescher
27,28,29,38,140,143,
197,229
Ribenboim 360
Richards,F.D. 281
Richards,R. 34,38,46,144,146
Richardson 34
Richelle 87
Richerson 21,25,30,32,41,47
144,153,154,384,
385,399,400
Riedl 20,48,219
Riemann 345
Rizzo 7
Roitblat 125,128
Roll-Hansen 145
Romanes 108,111
Rorty 224,226,227
Rosenberg 16,121,122,146,393
Rouquette 388,394,401
Rumelhart 51
Runciman 400
Ruse 45
Rutherford 270,271,274,317
Sachsse 15
Sahlins 127
Salazar 64
Salthe 79,90
Samuelson 15,52
Sanglier 72
INDEX OF NAMES
Saunders
5,16,30,44,45,46,
48
Schafer 55
Schaff 41,387,399
Schell 26,144
Schilcher,von 19,35,36
Schilpp 205,217
Schon 187
Schrodinger 57,58,72,318
Schuster 7
Seyfarth 117
Shapere 28
Shallice 114
Shannon 374
Sheldrake 47
Sherman 235
Shimony 17,140
Shweder 156
Siegel 146
Silverman 131,132-247
Simmel 112,123,200
Simon 10,29,34,39,41,42,52
90,93,149,231
Simons 386
Slovic 243
Smelser 388
Smullyan 363
Sober 10,11,12,13,18,19,21
47,49,52,140,141,234
Solla Price,de 174
Sommerhoff 84,120
Sosa 141
Spencer 20,376
Spinoza 51
Staddon 87
Steen 363
Stengers 6,11,32,43,45,
58,73
Stroud 224,230
Symons 235,242
457
Tannenbaum 343
Tarski 364
Tattersall 238
Taylor 360
Tennant 19,35,36,134,229
Terrace 126
Thagard 19,43,46,146,221
Thoday 46
Thorn 6,11,14,17,20,32,33,35,
102,104,366,367,368,369
Thompson 11,225
Thorndike 111
Tirapegui 70
Tolman 111,125
Tombaugh 242
Toulmin 10,11,23,27,28,38,144
181,182,183,192,193,
200,204,205,221
Trivers 234
Turing 219,349
Tversky 218,219,243
Ullman
129
22,23,42,44,47,
77,78,144,263
INDEX OF NAMES
458
Wagstaff 361
Walker 126,134
Watson 112,170,172
Weaver
10
Weber,B.H. 6
Weber,M. 197
Weber,R. 6
Weldon 145
Wertheimer 315
Wheeler 72
Whitehead 103
Whitley 200
Wicken 7,10,33
Wieferich 346,347,348,355,
361
Wiener 374
Wiley 11,33
Williams 11,19,48,51,83
Williston 370
Wilson 235
Wimsatt 7,10,14,18,19,23,29
30,33,49,52,79,113
314,356,363
Winter 26,27,384
Wittgenstein 228
Woodfield 15
Woodruff 135
Woodward 370
Woolgar 193
Worsley 391
Wright 190,195
Wuketits 20,21,140
Wynne-Edwards 49
Yilmar 140
Yuley 378
Zahar
160
Zamiatin
240
Zeeman 400
Ziff 374
Zuckerman
152