IV. SUMMARY AND EVALUATION
It has been argued throughout this article that be-
haviorism, as an historical phenomenon, has been a
loosely knit
collection of doctrines and theories woven
round a central prescription
about the proper method
to use in developing a science of psychology. Behav-
iorism, first of all, has often been
associated with the
metaphysical doctrine of materialism. Nowadays there
is less reluctance to discuss metaphysical problems than
there was in the
heyday of behaviorism and of “the
revolution in
philosophy” (Ryle, 1956). Materialism has
been revived, though
it has few adherents amongst
philosophers (Smart, 1963; Armstrong, 1968).
One of
its problems has been to state coherently what could
be meant
by saying that mental and bodily processes
are identical (Hamlyn, 1964).
It would be difficult to maintain that, in the sphere
of scientific theory, behaviorism has
advanced the un-
derstanding of behavior
in any major respect. The
theory which was most widely employed was that
of
associationism which was as old as Hobbes, Hume, and
Hartley. The
behaviorists merely transferred this the-
ory
from the realm of ideas to the realm of movements.
What occupied them most
was disputes amongst
themselves within this type of theoretical
framework.
There were two major issues which divided them. The
first
related to the importance of reward or rein-
forcement in setting up S-R connections. The second
issue
concerned the relative importance of central as
distinct from peripheral
processes. It was not really
surprising that the behaviorists in fact
contributed little
in the way of theory to the understanding of
behavior;
for basically most of them were not interested in ex-
plaining behavior or even learning for that
matter.
They were interested in conditioning. Even at the
animal level
it is extremely doubtful whether rats, dogs,
cats, and monkeys in fact
learn much by conditioning
in a normal environment. Ethological studies
certainly
cast grave doubt on the omnipresence of this type of
learning. It is probable that this type of learning is
an artifact of the
situations in which animals have been
confined. The extrapolation of this
type of learning to
the human level, where the pattern of life is deter-
mined largely by social rules and
purposes, was largely
programmatic. However, behaviorists showed that as-
sociationist principles might well
apply to the learning
of simple reactions and motor habits. Little more
could
in fact be claimed for their contribution to psycho-
logical theory.
Many of the defects in behavioristic theorizing, es-
pecially their programmatic extrapolations to the
human level,
derived from their lack of clarity about
such concepts as stimulus, reinforcement, and response.
Underlying these
particular confusions were funda-
mental
confusions about the concept of behavior itself,
due to their aversion to
assuming the existence of
consciousness. Guthrie was most sensitive to this
diffi-
culty. He made the distinction
between acts and
movements and tried to arrange an experiment which
dealt only with movements; for he realized only too
well that descriptions
of behavior at the molar level
are in terms of acts and not in terms of
mere move-
ments. And we identify their acts
by reference to what
human beings have in mind when they make certain
movements. For example, an act involving the same
movements of the arm is
identified as either signalling
to a friend or fanning the face (Hamlyn,
1953).
Skinner, in his distinction between operants and
respondents, actually hit
upon a distinction which is
crucial for getting clearer about the concept
of action.
Respondent reactions like salivations and eye-blinks,
which can be dealt with reasonably well by classical
conditioning theory, are indeed reactions which can
be correlated with
stimuli. But they are not, strictly
speaking, actions; they are events that
happen to us.
When, however, we pass to Skinner's operants, to
things
done as instrumental to an end, we are entering
the sphere of action
proper. Such actions, at the human
level at any rate, cannot either be
described or ex-
plained as mere movements
exhibited at the reflex
level. For an action is not simply a series of
bodily
movements; such movements as are necessary to it are
done for
the sake of something, as Aristotle pointed
out in his criticism of the
mechanists of the ancient
world. They are classed as belonging to an
action
because of their assumed relevance to an end (
telos).
Similarly, on the perceptual as distinct from the
motor side of behavior,
the importance of consciousness
is inescapable. Human beings, and probably
animals
as well, do not often simply react to stimuli in terms
of
their purely physical properties, as the Gestalt psy-
chologists pointed out in their distinction between
the
psychological and physical or geographical properties
of the
environment. They see things as meaning some-
thing; they respond to features of situations which are
interpreted
in terms of their understanding of them.
Skinner, for instance, was grossly
misleading when he
claimed that what we call emotions are names for
classifying behavior with respect to various circum-
stances which affect the probability of the
behavior's
occurrence; for the circumstances are those which are
interpreted by the subject in a certain light, e.g.,
as
dangerous in the case of fear, as involving somebody
else having
something which we want in the case of
envy. The relationship between
circumstances and the
subject is not one of purely physical causality
(Peters,
1965).
In brief even what the behaviorists called “behavior”
includes a range of phenomena between which there
are very important
distinctions, let alone other purely
mental phenomena such as remembering
and dream-
ing, which may have no overt
expressions and which
may lead to no overt actions. Many more
distinctions
than these can be drawn which would complicate the
picture even further. But this would not affect the two
cardinal points
that need to be emphasized: first, that
it is impossible to make such
distinctions without ref-
erence to
consciousness and, second, that behaviorists
tend to think that the form of
description and explana-
tion applicable at
the lowest level of reflex behavior
can be extrapolated to explain the much
more complex
phenomena at higher levels.
About the methodological doctrine that was the
kernel
of behaviorism—that psychology should base
itself as a science
on the type of publicly observable
data that biologists use when theorizing about ani-
mals—the first point to make is that it is an
example
of the long-standing delusion that success in science
depends
upon following a particular method. A study
of the history of science gives
no support for the belief
that science has in fact been advanced by
following
any particular method, if this is interpreted as meaning
following a particular procedure for making discoveries
or arriving at
laws. It is impossible to formulate any
method for arriving at hypotheses;
all that can be done
is to lay down general rules about testing them.
Is there then anything to be said for the behaviorists'
prescription as a
procedural rule relating to the testing
of hypotheses rather than to their
formulation? If they
had been concerned solely with animal behavior
their
prescription would have been unexceptionable but
otiose; for
there is no possibility of obtaining intro-
spective reports from animals. Insofar, however, as they
studied
animals partly with the intention of making
extrapolations to human
behavior, their prescription
seems to be very much a self-denying
ordinance, for
in science it is advisable to obtain all the evidence
available. Also the sort of observations which are ap-
propriate depends upon what is being studied. If it is
reactions such as salivation, knee-jerks, and simple
motor skills, which
were the main field of interest
amongst behaviorists, introspective reports
may not be
of great significance. If, however, hypotheses about
dreams, perception, delusions, remembering, emotional
phenomena, or moral
development have to be tested,
it is very difficult to see how much
relevant evidence
could be accumulated without recourse to reports by
the subject. And it simply will not do to say that the
experimenter is then
relying on another form of be-
havior, namely
verbal behavior. Furthermore, this
move by behaviorists is a form of conceptual
behav-
iorism. The methodological doctrine,
which is distinc-
tive of behaviorism, would
evaporate if a subject's
reports were re-admitted as evidence because they
too
were regarded as forms of behavior.
Historically, therefore, behaviorism was a salutary
corrective that was
pushed to inordinate extremes. At
a time when psychology was largely
preoccupied with
examining the minutiae of a subject's introspections
there was some point in drawing attention to what
could be publicly
observed. But this injunction unfor-
tunately was not accompanied by any suggestion of
new hypotheses
that might be tested. It functioned
mainly as a new recipe for continuing
the old associa-
tionist program. The
widespread implementation of
this recipe, however, had very important
consequences
for psychology generally. It enhanced the status of
psychology as a science amongst the scientific commu-
nity. Psychologists could now wear lab-coats like biol-
ogists and be admitted to the Faculty of Science.
Although
behaviorism was basically a philosophical
movement psychologists were now
able to part com-
pany with philosophers and set
up on their own.
Whether this separation has been beneficial in ad-
vancing our understanding of human behavior is an-
other question; for the basic problem in the central
spheres of action, motivation and emotion, perception,
learning,
remembering, etc., is to decide what is a
psychological question. In the
sphere of learning, for
instance, in which behaviorists evinced most
interest,
how much depends upon the conceptual and logical
relationships involved in what has to be learnt and how
much depends on
general empirical conditions about
which psychologists might reasonably
test hypotheses?
The work of theorists such as Jerome Bruner and Jean
Piaget, who have been concerned with human learning
and development in a
concrete rather than a program-
matic way,
raises such problems in an acute form. But
it is difficult to see how much
progress can be made
until issues of this sort are squarely faced. But to
face
them would involve a revolution in psychology as
radical as the
methodological movement which Watson
himself initiated.