On the Difference between Radical and Methodological Behaviorism

Author(s): Willard day

Source: Behaviorism, Vol. 11, No. 1 (Spring, 1983), pp. 89-102
Published by: Cambridge Center for Behavioral Studies
Stable URL: http://www.jstor.org/stable/27759016 .
Accessed: 21/10/2011 16:49
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Cambridge Center for Behavioral Studies is collaborating with JSTOR to digitize, preserve and extend access
to Behaviorism.
http://www.jstor.org
On
the Difference between
Radical
and Methodological Behaviorism
Willard
Day*
University of
Nevada,
Reno
Abstract
Radical
behaviorism,
a behaviorism associated with the views of B.F.
Skinner,
is often contrasted with
methodological
behaviorism. The nature of this contrast is
explored by examining
what Skinner himself has had to
say
about the difference. It
is
pointed
out that Skinner has
repeatedly
called
upon
the distinction
throughout
the course of his
career,
and that
essentially
the same
argument
is made each time. It
is
argued
that the two
perspectives
differ
largely
on
epistemological grounds,
with
radical behaviorism
manifesting
a
revolutionary
if
pragmatist perspective
and
methodological
behaviorism
representing
the
logical positivist-derived philosophy
of science
so common in social science
today.
Introduction
There is
a
great
deal of confusion
right
now about what behaviorism
is,
and
what behaviorism is not. Yet it is
widely accepted among
the
parties
involved that
some sort of
major
contrast exists between radical and
methodological
behavior
ism,
and that is the distinction I am
going
to discuss here.
First,
I will
report
a
revealing
anecdote.
Then,
I will make some
preliminary
comments of
my
own
by
way
of
drawing
the contrast in bold
outline,
and
finally
turn to what Skinner has
had to
say
about the difference between radical and
methodological
behaviorism.
First,
the anecdote. Several
years ago,
"The Talk of the Town" in The New
Yorker
magazine (April
4, 1977)
carried an account of a visit
by
the
psychologist
David Bakan to New York to
give
a lecture. The account in The New Yorker
begins
this
way:
Professor David Bakan blew into town like one of those blasts of fresh air
out of Canada. Some friends in Academe had informed us that
Bakan,
who
teaches
psychology
at York
University,
in
Toronto,
is
a
hard-hitting
critic of
his
fellow-psychologists,
and that he is
especially
hard on
people
like B.F.
*
An earlier draft of this
paper
was
presented
as an invited address at the
meeting
of the Midwestern
Association of Behavior
Analysis, Chicago, May
1977.
89
Willard
Day
Skinner,
who insists that the
only proper study
of
psychology
is observable
behavior. Since the behaviorist
approach
has achieved the status of ortho
doxy
in most
university psychology departments,
Bakan comes on
like an
anti-establishment
figure, (p. 28)
I am not interested here in the confusion on the
part
of The New Yorker in
identifying
Skinner with the behaviorism that is orthodox in most
university
psychology departments.
Most
psychologists
I know
regard
Skinner's outlook as
every
bit as anti-establishment as
that, say,
of David Bakan. What I want to call
attention to is the claim that in a certain sense the behaviorist
approach
"has
achieved the status of
orthodoxy
in most
university psychology departments,"
to
repeat
the
phrasing
of The New Yorker. That kind of behaviorism is
methodologi
cal behaviorism. Yet this kind of behaviorism is linked
by
The New Yorker to the
insistence "that the
only proper study
of
psychology
is observable behavior." What
they
are
talking
about is
publicly
observable behavior. Since radical behaviorism is
different from
methodological behaviorism,
we should not be
surprised
to
find,
then,
that radical behaviorism has a
special
account to
give
of
privately
observable
behavior;
and indeed Skinner's discussion of the distinction between radical and
methodological
behaviorism turns on
this issue.
However,
let me continue with the
story
from The New Yorker. As it turns
out,
Skinner himself
happened
to be in the audience
as Bakan was
speaking.
The New
Yorker interviewed Skinner
privately
after the
lecture,
and the account contains
some of the
things
Skinner had to
say.
I focus here
only
on the remark
by
Skinner
with which the
story
ends. In
noting
that his own
approach
had often been under
attack
by
his
colleagues,
Skinner said that in
general
he had
no
complaint
to make
about the treatment his work has received. "These
things
take hold
slowly,"
he said.
"We are
challenging
some
very
old ideas."
I would want to
say
that the chief
way
in which radical behaviorism differs from
methodological
behaviorism is that radical behaviorism involves a radical chal
lenge
to some
very
old
ideas,
whereas
methodological
behaviorism does not.
Skinner often
speaks
of these
very
old ideas as "the traditional view." But what is
involved is no less than the
general concept
of
mind,
as an
interacting
network of
concepts
taken to be
capable
of
explaining
human
action,
which
gradually emerged
in the evolution of Greek culture. Skinnerians
speak
of
making
use of this
concep
tual
system
to
explain
behavior
as
"mentalism,"
and mentalism can indeed be
succinctly
defined as the
practice
of
taking
mental and
psychological
states to be the
causes of behavior
(see, e.g., Skinner, 1938, p. 3; Day,
1976, p. 88f.).
Methodological Behaviorism
First,
with
respect
to
methodological
behaviorism,
it would be a mistake to
restrict one's
conception
of
methodological
behaviorism to the commitment to a
publicly-observable
data-base alone. Most
psychological
research that I know
of,
including
radical behaviorist research however
characterized,
involves in one
way
or another the
analysis
of
publicly-observable phenomena.
The trouble comes in
getting hung-up
about
observability.
As The New Yorker
puts it,
professional
90
Difference
between Radical and
Methodological
Behaviorism
orthodoxy
in
psychology
involves
an
insistence that
psychology study only publi
cly-observable
behavior. This
insistence,
as a
kind of intraverbal
fetish, betrays
that
a broader set of
professional
beliefs and values lies at the heart of
methodological
behaviorism.
Methodological
behaviorism involves a
widely accepted professional
orientation towards how one should conduct
psychological
research in
general.
Verbalizations of this orientation amount to a kind of crude
philosophy
of science.
In an earlier
paper (1976)
I have illustrated at some
length
the kind of
things
psychologists
and
philosophers
have had to
say by way
of
explaining
what
they
take
methodological
behaviorism to be.
Basically
what is involved are views to the effect
that scientific
knowledge
is different in nature
from,
and
intrinsically superior
to,
common sense
knowledge.
It is obtained
by
controlled
experimental
research,
the
objectivity
of which is assured
by
the use of
professionally
endorsed methods of
experimental design.
The
legitimacy
of
knowledge
claims
resulting
from
psycholog
ical research is
generally
assessed
by techniques
of statistical inference. The
long
range
aim of
psychological
research is to arrive at scientific laws which are then
taken to
explain
behavior. In one
way
or another
?
but
generally involving
a
mixture of material found in
psychological journals
and
patterns
of
thought
exist
ing
in the
lay
culture
?
the researcher arrives at notions which he
regards
as
potentially interesting
as
explanations.
Modest
potential explanations
are advanced
as
hypotheses,
more elaborate ones as theories. These
hypotheses
and theories are
then
subjected
to
experimental
test. The immediate aim of research is to
try
to
verify
these
hypotheses
or theories as true. More or less
successfully
verified theories are
regarded
as
psychological knowledge.
And so on.
This
professional
outlook involves
a kind of
philosophy.
At
least,
its intellectual
coherence
can be
inspected philosophically.
It is similar in
ways
to a kind of naive
realism,
and it is at least
historically
derived from
logical positivism, operationism,
and the behaviorism of the 1940's.
Many psychologists
find it
easy
to relate this
orientation to
contemporary
and
sophisticated philosophies
of science
generally
assuming
a
positivistic
stance,
such as that of Karl
Popper.
However,
I
prefer
to
regard methodological
behaviorism
simply
as a
professional psychological
orienta
tion that is best accounted for in terms of social and cultural variables that have
functioned to
keep
the
profession roughly
all
together
as a whole.
My point
is that all these familiar values should also be included
among
the
range
of
practices regarded
as
methodological
behaviorism,
not
simply
the insis
tence that the
study
of
psychology
be restricted to the field of
publicly-observable
behavior.
However,
I want to call attention to another feature of
methodological
behaviorism. Standard
experimental design
within
general experimental psychol
ogy
is often linked to
methodological
behaviorism in the
following way.
In its
simplest
form
a
psychological experiment
is taken to involve a
comparison
between
an
experimental
group
and
a control
group.
The two
groups
share a common
dependent
variable,
which is said to involve measurement of some
property
of a
response.
The two
groups
differ in that the
experimental
group
involves
manipula
tion of an
independent
variable,
which is said to be the
manipulation
of a stimulus.
Thus
any experiment
can be said to be behavioristic in
spirit
because it
attempts
to
demonstrate
a
relationship
between stimulus and
response.
That
is, any experiment
91
Willard
Day
can
be
regarded
as a form of S-R
psychology.
Not
only this;
but when
any
experiment
is
successful,
the
manipulation
of the
independent
variable is
widely
taken to have been the cause of the measured differences on the
dependent
variable.
Thus the standard
experiment
involves not
only
S-R
psychology,
but S-R
psychol
ogy
after the model of the reflex
arc,
where the
power
to cause the
response
is vested
fully
in the
prior
stimulus. The notion of causation that is involved here is the
common one of antecedent causation. Thus one can see
why
Skinner
speaks
of
cognitive psychology
in About Behaviorism
(1974)
as a
variety
of S-R
psychology,
since the
cognitivists
are
basically
interested in
identifying cognitive
mechanisms
that are
causally
antecedent to behavior.
Methodological behaviorism,
in its
linkage
with
general experimental
method,
is also linked to a model of antecedent causation. It is of course the
general
experimental
method that is orthodox in
university psychology departments.
The
method is
employed
in the
investigation
of
hypotheses
and theories
concerning
a
virtually
limitless
variety
of
psychological processes
and states. The
processes
and
states are
themselves not
directly publicly
observable. Their
presence
is inferred
from the successful outcome of the
experiment. Methodological
behaviorism relies
on the model of antecedent causation almost
universally.
Thus in
methodological
behaviorism it is
commonly
taken that
any
hypothesized
or inferred
psychological
processes
or states are the
presumed
causes of behavior. This is mentalism. The
hypotheses involving
these
antecedently
causal
psychological processes
and states
generally
have their source in the
conceptual system commonly employed
in our
culture. Thus there is in orthodox
experimental psychology
no
resistance at all to
these
"very
old ideas" which it is the business of radical behaviorism to
challenge.
Let me turn
now,
by way
of
contrast,
to some
preliminary
remarks about radical
behaviorism. In the Introduction to About Behaviorism Skinner makes the
point
that behaviorism is not the science of human
behavior;
it is the
philosophy
of that
science. Most of the discussion in the Introduction deals with matters relevant to the
history
of behaviorism as a
general movement,
so
it is neutral with
respect
to the
distinction between radical and
methodological
behaviorism.
However,
the Intro
duction ends with the statement that the most common
criticisms of behaviorism
are most
effectively
answered
by
a
"special discipline"
within the broader field of
behavioral science. This
special discipline,
Skinner
says,
has come to be called the
experimental analysis
of behavior. He
goes
on to
say
that the behaviorism he
presents
in the book as a whole
?
that
is,
radical behaviorism
?
is the
philosophy
of this
special
version of a
science of behavior.
However,
in
my opinion
it is too
simple
to
regard
radical behaviorism as
nothing
more than the
philosophy
of the
experimental analysis
of behavior. The
experimen
tal
analysis
of behavior is a
sub-culture within the broader field of
experimental
psychology.
Yet cultures evolve. Current methods within the
experimental analysis
of behavior are
clearly
related at least
historically
to Skinner's own research
practices.
I am
thinking
here
principally
of his work with Ferster
published
in
Schedules
of Reinforcement (1957).
Yet the
range
of Skinner's
empirical
research
interests has been much broader than
that,
as one can see in Cumulative Record
(1959).
92
Difference
between Radical and
Methodological
Behaviorism
Research that has come to be called the
experimental analysis
of behavior is
closely
identified with work
published
in the Journal
of
the
Experimental Analysis
of
Behavior. Yet the thrust of the research
reported
in that
journal
is
becoming
increasingly
difficult to
separate
out from research in the
general experimental
psychology
of
learning, although
it is true that the
journal
tries to hold the line to
research
revealing properties
of the behavior of individual
organisms.
Fifteen
years
ago
one
might
have wanted to
identify
the
experimental analysis
of behavior with
strategies
of research as
discussed
by
Sidman in Tactics
of Scientific
Research
(1960), yet
there has been
pressure
to make
increasing
accommodation to research
designs
and aims more
commonly
associated with
methodological
behaviorism.
The
thing
to look at is Zeiler's Editorial in the
January (1977)
issue of the Journal
of
the
Experimental Analysis of
Behavior,
where he raises the issue of what
subject
matter
properly belongs
in the
journal.
He
says,
"An overview of
contemporary
experimental psychology
indicates that
previously sharp
distinctions between tradi
tions and
approaches
have become blurred"
(p. 1).
He adds
that,
"Actually
the
Journal never
legislated
a
particular
theoretical or
metatheoretical stance"
(p. 1).
He
goes
on to
say,
"In
my view,
breadth is beneficial to the Journal and to
experimental psychology
as a
whole,
so I will do what I can to further JEAB as a
forum
representing
both the mainstream and the frontier of
psychological
research
and
theory" (p. 1).
Please do not misunderstand me: I am not
objecting
to the
experimental
analysis
of behavior. It is
true,
I do
object
to
restricting
interest in radical behavior
ism to an interest in that
type
of research.
However,
I am
merely calling
attention
here to the fact that research
practices
are
subject
to
contingencies
of social
evolution, very
much in the sense discussed in Skinner's books
Beyond
Freedom
and
Dignity (1971)
and About Behaviorism
(1974).
Zeiler is
particularly poignant
in
his Editorial when he
points
out that the editors of
journals
are
essentially helpless
in
stopping
this
professional
cultural evolution. Be that as it
may,
I trust it is clear
that radical behaviorism is not in
any
sense the
philosophy
of the
general experi
mental
psychology
of
learning.
If radical behaviorism is the
philosophy
of a
special
science of
behavior,
then
there are certain
things
that can be said about the nature of this
philosophy.
The
things
to be said of the
philosophy
in radical behaviorism cannot be said of the
philosophical perspectives generally
associated with
methodological
behaviorism.
Radical behaviorism bears
striking
similarities to the functionalism of William
James. A
relevant,
if
angry,
reference is the article
by
John Malone in the Fall 1975
issue of the
journal
Behaviorism. The historical ties of radical behaviorism to the
shaping
verbal
community
of American
philosophers
in the 1920's
are
very strong.
Skinner's
approach
to the
conceptualization
of ethical
problems,
so
puzzling
to the
general
intellectual
community
of
today,
is
a
development
of the ethical naturalism
of
Ralph
Barton
Perry
at Harvard and of the
pragmatist
John
Dewey,
who carried
forward the
legacy
of
pragmatism
he inherited from William James.
However,
it is
not in ethics but in
epistemology
where the influence of
pragmatism upon
Skinner's
thought
is most
strongly
to be seen. The
pragmatist
character of Skinner's
episte
mology
has been
emphasized by
Zuriff
(1980)
and also
by Day (1980). Today,
few
93
Willard
Day
academic
philosophers
can be
regarded
as
having
been socialized within the
prag
matist tradition. Yet Skinner's
conceptualization
of
knowledge
and truth as seen in
Verbal Behavior and About Behaviorism,
which must
inevitably
bear
significantly
upon
how the aims of scientific research for radical behaviorism
are to be
concep
tualized,
is
conspicuously pragmatist
in
spirit.
The hiatus between the
philosophical
connections of radical and
methodologi
cal behaviorism was further widened in the
1940's,
as Skinner makes the crucial
epistemological
moves in his
paper
for the 1945
Symposium
on
Operationism.
In
this
paper
Skinner turns
logical positivism upside
down,
while
methodological
behaviorism continues
on its
own,
particular logical-positivist
way.
At the same
time,
analogous developments
were
taking place, apparently independently,
within
analytical philosophy
in
England.
Thus there are
tight
connections between radical
behaviorism and the later
philosophy
of
Ludwig Wittgenstein (Day, 1969).
The
issue centers around the
analysis
of
meaning.
The immediate
sequel
to
Wittgenstein
was the
philosophical
orientation called
"ordinary language analysis."
Yet it is
important
to realize that for Skinner the
sequel
was a
proposal
for the
empirical
analysis
of verbal behavior.
Almost
every
psychologist
I know
conceptualizes operationism
after the fashion
of
logical positivism
and
methodological
behaviorism. Yet it is in his 1945
paper
on
operationism
that Skinner first draws the distinction between radical and methodo
logical
behaviorism. This
paper
cannot be understood unless it is
clearly
seen that
by operational
definition Skinner means
very
explicitly
the functional
analysis
of
verbal behavior
involving
the term to be defined. There is much concern on the
part
of critics these
days
about what sort of intellectual move is involved in the so-called
"translations" of common mentalistic
expressions
that Skinner
constantly
makes.
Yet Skinner discusses the "translation" of terms in this 1945
paper,
and there these
translations can be seen to involve
steps
towards the functional
analysis
of verbal
behavior. It is
certainly
clear that in 1945 Skinner
conceptualized operationism
as
the functional
analysis
of verbal behavior.
Also,
at that
time,
he
appeared
to
regard
attempts
to
carry
out such an
enterprise
in
practice
as a
relatively straightforward
matter.
Yet as
interesting
as
the
philosophical
affiliations with Skinner's
thought may
be,
and as different as
they happen
to be from those associated with
methodological
behaviorism,
I think it is a mistake to
regard
Skinner
as
primarily
a
philosopher.
I
say
this in an effort to
put
in
perspective
the fact that most of Skinner's
published
works consist of material other than the
reports
of
findings
of
empirical
research. In
all this
writing
Skinner's verbal behavior is
clearly largely
under the audience
control of the
professional psychological,
rather than
philosophical,
verbal com
munity. Why
not
regard
Skinner in this
respect simply
as the
spokesman
for his own
particular
orientation within
psychology?
That orientation is of course radical
behaviorism. Skinner's
writing
tends to
engage philosophy
because it constitutes
a
serious
challenge
to traditional values that are
widespread
both within
philosophy
and the
lay
culture at
large.
The force of Skinner's
challenge
to
epistemology
is that
the
concept
of
knowing anything
is
given
an
unremittingly psychological
treatment
within the framework of a behavioral
analysis.
The force of his
challenge
to
94
Difference
between Radical and
Methodological
Behaviorism
traditional
philosophy
of science is his
attempt
to
give
a
strictly
behavioral account
of the realities which
actually
constitute scientific
practice.
This is
psychology.
If
one were to
regard
Skinner's written work as
essentially philosophy,
then it would
seem to me that its critical assessment should
principally
be the business of
philos
ophers,
and his books should be
taught primarily by philosophers
in their own
departments. If,
on the other
hand,
all this
writing
is
largely
a contribution to
psychology,
as I
suggest,
then the relation between the
great
breadth of
subject
matter treated in the written material and the
practices
which in fact constitute the
experimental analysis
of behavior needs
very
much more
specifically
to be
spelled
out.
For
many years
it was fashionable
among philosophers
to take
potshots
at
Skinner, largely,
in
my opinion,
in a
misguided
effort to attack
methodological
behaviorism. However there are indications
now that
philosophers
are
coming
to
have a more accurate
understanding
of radical behaviorism. A case in
point
is the
recent article
by
Jon
Ringen (1976). Ringen's
chief task in the article is to
give
a
technical
philosophical analysis
of issues
concerning intentionality
in
operant
behaviorism,
in
light
of Charles
Taylor's (1964) highly
influential attack
upon
behaviorism
along
these lines.
Ringen's accomplishment
in the article is to show
that
Taylor's objections
can in no
way
be
brought
to bear
upon
Skinner's
conceptualization.
However,
what I want to call to
your
attention
are some of the
things
that
Ringen
has to
say
in the concusion of his article.
Ringen
notes that "the fundamen
tal contribution of
operant
behaviorism
may
well be to have uncovered and
precisely
described functional relations between behavior and
controlling
variables
in the environment"
(p. 245).
Yet he
goes
on as follows:
Judgments
about the
plausibility
of the
operant program
must
currently
be
made under circumstances where
no
operant psychologist
has
provided
a
careful,
precise, specification
of
contingencies
of reinforcement under
which the
complex
behavioral
repertoires
which interest
psycholinguists
and social
psychologists
are
acquired
and maintained.
Furthermore,
there
are
incredible,
practical,
ethical,
and intellectual obstacles
to
carrying
out
such an
analysis.
Nevertheless it is a
task to which serious
operant
behavior
ists are
professionally
committed insofar
as
they
are behavioral scientists
interested in scientific
explanation
of
complex
human behavior and
not
simply
armchair
psychologists
or
behavior
technologists
satisfied with
interpretations
of behavior which bear some tenuous relation
to the
princi
ples employed
in the
operant analysis, prediction,
and control of the
goal
directed behavior of
pigeons
and rats.
(p. 247)
Personally,
I could not
agree
more
strongely
with these remarks
by Ringen.
However,
even more relevant to
my purpose
here is the bold
position
in which
Ringen places
Skinner's
psychology
at the
very
end of his article. He calls attention
to a
recent,
but
vigorous, development
within the
philosophy
of
psychology.
This
orientation,
often associated with the
leadership
of the late Theodore
Mischel,
is
95
Willard
Day
often
spoken
of as action
theory,
and it focuses on a
conceptualization
of man
centering
around his
capacities
as an
Agent.
In
focusing
on human
agency,
action
theorists confront
directly
the
crucial, anti-agency
thrust that is central to Skinner's
position.
However,
the
really interesting thing
is that
Ringen
sees
operant
behavior
ism and action
theory
as
genuine
alternatives,
as the two
major options facing
psychology today. Ringen
then
points
out the need for considerable
conceptual
clarification in connection with both
positions.
He concludes as follows:
Radical behaviorism
appears
to be the
only
serious
existing
alternative
to
common sense
mentalism,
and serious
conceptual analysis
of its technical
terms will contribute to our
understanding
what the alternatives
are.
If current assessments of the
revolutionary
character of
operant
behav
iorism
are
correct,
such a
clarification will be no small task. It will
require
something
of at least the
magnitude
of Galileo's critical discussion of
Aristotelian
physics
and
cosmology.
...
(p. 250)
Now we are
talking
about a
very
serious
challenge
within
psychology
to some
"very
old
ideas,"
as Skinner
puts
it for The New Yorker. Now we are
talking
about radical
behaviorism,
and that is
why
I find radical behaviorism
interesting.
Skinner on Radical and Methodological Behaviorism
I turn now to a review of what Skinner has had to
say
on the distinction between
radical and
methodological
behaviorism. There are five
places
to look:
(1)
the
paper
on "The
Operational Analysis
of
Psychological
Terms
(1945)"; (2) Chapter
XVII on
"Private Events in a Natural Science" in Science and Human Behavior
(1953); (3)
the section
on "Verbal Behavior under the Control of Private Stimuli" in Verbal
Behavior
(1957); (4)
the
paper
"Behaviorism at
Fifty" (1964);
and
(5) Chapters
1 and
2 of About Behaviorism
(1974).
In what follows I will refer to this material
collectively
as "the five sources." The
important thing
to see about these five sources
is that
they
all
say very
much the same
thing;
the
underlying
structure of the
argument
remains
basically
the same as it is
repeated by
Skinner.
However,
there is additional relevant material towards the end of About Behav
iorism. A statement in
Chapter
13 identifies the issue we will be concerned with: "A
science of behavior must consider the
place
of
private
stimuli as
physical things,
and
in
doing
so it
provides
an
alternative account of mental life. The
question,
then is
this: What is inside the
skin,
and how do we know about it? The answer
is,
I
believe,
the heart of radical behaviorism"
(pp. 211-212). Then,
the first three
pages
in
Chapter
14 contain a
concise
summary
of the
position
on consciousness that
Skinner attributes to radical behaviorism. The
summary begins
in this
way,
which
involves
a contrast between
methodological
and radical behaviorism:
Methodological
behaviorism and certain versions of
logical positivism
could be said to
ignore consciousness,
feelings
and states of
mind,
but
radical behaviorism does not thus "behead the
organism";
it does not
"sweep
the
problem
of
subjectivity
under the
rug";
it does not "maintain
a
strictly
behavioristic
methodology by treating reports
of
introspection
96
Difference
between Radical and
Methodological
Behaviorism
merely
as verbal
behavior";
and it was not
designed
to
"permit
conscious
ness to
atrophy." (p. 219)
From this it should be clear that Skinner believes that radical behaviorism has a
robust account of consciousness to
give.
This account consists of the
argument
I will
review from the five sources. But first I want to call attention to how Skinner
conceptualizes methodological
behaviorism in this material.
The
important thing
to be aware
of is that Skinner takes a much narrower
view
of
methodological
behaviorism than the
general professional orthodoxy
I have
discussed above. Skinner looks at
methodological
behaviorism as
something
that
happened
in the
history
of behaviorism. For
him,
it was an outlook often associated
with behaviorism
roughly
in the
1940's,
and it is linked
by
him with the
logical
positivism
and
operationism
of that
time,
which he
speaks
of as
involving
a
philosophy
of truth
by agreement.
It was a view which at times
appeared
to Skinner
to want to substitute the verbal
report
of
experience
for the
experience itself,
or on
the other hand to want to rule
experience
out of bounds for science.
There is a
point
to take from all this. Skinner's
conception
of
methodological
behaviorism is so narrow
that for him
simply
to make a
distinction between
methodological
and radical behaviorism is for him not to
engage
at all the
complex
set of
professional practices
and beliefs that are now orthodox in most
psychology
departments.
There is an
advantage
in this.
Simply
in the distinction between
radical and
methodological
behaviorism
nothing
is said about
strategies
of
research. No
objection
is raised at all to the notion of
experimental
science itself as it
is
generally
understood in the
profession.
I am
thinking here,
for
example,
of such
things
as the
concept
of the controlled
experiment
and its aims.
The structure of the
argument
remains the same in each of the five
sources,
although
the account
given
in the
paper
"Behaviorism at
Fifty"
is more
condensed
than that of the others. The
argument
always begins
in this
way:
there is a frank
acknowledgment
at the outset that
private
events
exist,
that
they play
a role in the
control of
behavior,
that we can
speak intelligibly
of our
private experience
to
others,
that
knowledge
of what is
occurring privately
within us is at times useful and
important
to the verbal
community,
and that a functional
analysis
of behavior is
intrinsically incomplete
without an account of the role
played by private
events in
the control of behavior.
Then the
important,
and
innovative,
philosophical
move is made. The tradi
tional
problem
of
privacy
is
reconceptualized
in this
way.
There is no
question
that
private
events exist. The central
philosophical problem
is not whether or not
private
events
exist,
but how it is that we are able to come to have
knowledge
of them. This
philosophical question
is then
psychologized.
In the
psychological
treatment of this
problem
the issue is reduced
centrally
to the
problem
of how the verbal
community,
which has access
only
to
publicly
observable
stimulation,
is able to teach us to talk
effectively
under the control of stimulation
occurring privately
within us. In answer
ing
this
question
the
problem
of
privacy
is
regarded
as
having
been solved. Concen
tration on the
functioning
of the verbal
community
is the most central feature of the
discussion
throughout
the entire
complex argument.
97
Willard
Day
Skinner
characteristically
identifies four
ways
"in which a verbal
community
which has no access to a
private
stimulus
may generate
verbal behavior in
response
to it"
(1959, p. 276).
With the
exception
of the abbreviated account in "Behaviorism
at
Fifty"
these four
ways
are
simply repeated
as
the
argument
is restated in the
different sources over the course of the
thirty-year period.
First,
the verbal com
munity may
make use in its
practices
of different reinforcement of
any public
stimuli which
accompany
the
private
stimulus with reasonable
regularity.
Second,
the verbal
community may respond
on the basis of
publicly
observable collateral
responses
to the
private
stimulus.
Third,
the verbal
community "may
reinforce
a
response
in connection with a
public stimulus, only
to have the
response
transferred
to a
private
event
by
virtue of common
properties,
as in
metaphorical
and met
onymical
extension"
(1957, p. 132). Fourth, responses descriptive
of the
speaker's
own behavior
may
be based
upon externally
observable
behavior,
which then
may
be reduced in
magnitude
so that "the
response
is
eventually
made to a
private
stimulus which is similar
except
in
magnitude
to
private
stimuli otherwise accom
panied by public
manifestations useful to the
community" (1957, p. 133).
What has been said so far is the central feature of the
argument
used
by
Skinner
in the solution of the
problem
of
privacy. However,
in the five sources this material
is
consistently accompanied by
discussion of the
following
other matters. Private
events are
regarded predominantly
as
stimulation
arising
from the
interoceptive
and
proprioceptive
nervous
systems,
although
the role of the
exteroceptive
nervous
system
is also mentioned
as
playing
a
part
in the observation of our own
body.
Again
the
functioning
of the verbal
community
is the focal
point
of the
argument.
The verbal
community
is held to
play
an
important part
in
enabling
us to make the
particular discriminations,
both
public
and
private,
that we
do,
although
the
relevance of the non-social environment to the establishment of these discrimina
tions is also
pointed
out. The case of our
knowing
that we are
behaving
discrimina
tive^,
as
opposed simply
to
discriminatively responding,
is discussed
separately,
and it is held that this is
necessarily
the achievement of the verbal
community.
The
conception
of
knowledge
that is involved here is one which
emphasizes
our
capaci
ties for discriminative
responding.
This
stage
of the
argument
often draws to an end
by pointing
out how it is
largely through
the
practices
of the
differentially
reinforc
ing
verbal
community
that we can be said to know ourselves
interestingly
in
any way
at all.
The
argument
as
presented
in the five sources also
invariably
calls attention to
the fact that the indirect
public
resources
available to the verbal
community
for
establishing
verbal behavior under the control of
private
stimuli entails the conse
quence
that
private
events exercise a
less
precise
control over
verbal
responding
than does that of the
publicly
observable
environment,
where
contingencies
of
differential
reinforcement
can
be made
quite specific.
Thus an account is
given
of
why
we tend to
mistrust
reports
of what is
going
on
within others.
Lastly,
the
account of how we are
able to come to talk under the control of
private
events is
linked
directly
to an
additional account of how we are
enabled
by
the verbal
community
to describe
any aspect
our
behavior in the
way
that we do.
Separate
accounts are
given
of how the verbal
community
becomes involved in
enabling
us to
98
Difference
between Radical and
Methodological
Behaviorism
report
our current
behavior,
our
probable behavior,
our
past behavior,
our covert
behavior,
and our
future behavior. This discussion on
Skinner's
part
of the verbal
contingencies
involved in
enabling
us to describe our own behavior
may
be
pre
sented so
starkly
that it can be
quite puzzling
to the
reader,
as in About Behavior
ism. It is
helpful
to
keep
in mind that this discussion is linked
historically
to the
treatment of the
problem
of
privacy
as a
consistent and unified
practice
in the five
sources. Much of this entire
complex argument
is
succinctly
summarized at the
beginning
of
Chapter
14 in About Behaviorism. I have
already quoted
earlier the
beginning
of this
summary statement,
which is framed as a contrast between
methodological
and radical behaviorism.
Thus we have Skinner's account of the heart of radical behaviorism.
Clearly
this
account remains neutral with
respect
to
strategies
of research.
However,
the
account is not at all neutral in the
way
it rests
fundamentally upon
a
viable
conception
of the functional
analysis
of verbal behavior. I have
argued
that
Skinner's discussion of these matters is not an
exercise in
philosophy,
but in
psychology.
If
so,
then it must bear some relation to
empirically
observable behav
ioral
phenomena.
What this means to me is that "the heart of radical behaviorism"
boils down to a commitment to the central relevance of a functional
analysis
of
verbal behavior. If Skinner's efforts to
distinguish
radical from
methodological
behaviorism
are to make
any professional
sense at
all,
I
argue
it is
absolutely
essential that
they
be viewed in the
light
of
on-going empirical
research directed
towards that end
?
namely,
the functional
analysis
of verbal behavior.
Now,
where I do
personally
stand on all this? What interests me about radical
behaviorism is
something
different from what I have been
talking
about. I find the
argument
about
private
events
interesting enough,
but I want to follow
up
on it. Yet
it is not
private
events as such that I want to follow
up
on at all. The
interesting thing
about the
argument concerning private
events that Skinner
gives
is that it
hinges
on
the
intelligibility, meaningfulness,
and relevance of a functional
analysis
of verbal
behavior. What interests me
is the
possibility
of a
functional
analysis
of verbal
behavior that rests
upon
a solid
empirical foundation,
and
beyond
that a similar
functional
analysis
of
on-going
human behavior as it is there to be observed in
everyday
life. I think that it is
only through
this sort of work that a serious
challenge
to
"very
old ideas" can be made. One sees another
thing
in
looking closely
at the
argument
in the five sources. In the earliest
paper (1945),
Skinner
appears
to have
great
confidence that a science of the functional
analysis
of verbal behavior is
essentially
at hand. The dimensions of such a science for him at that time seemed
clear: the science would concentrate on the
study
of how verbal behavior in an
individual
organism
is
acquired
and maintained. However
as we move
through
the
five sources the call for an
empirical analysis
of verbal behavior is
progressively
attenuated.
Although
the fundamental
dependence
of the
argument
for its coher
ence
upon
the notion of the functional
analysis
of verbal behavior does not
change,
concern
with the
subject
as a matter of scientific
or
empirical inquiry
seems to
get
pushed upstairs.
I have been
sorry
to see
this.
When it comes to the business of
verbalizing
the discrimination that can be
made between radical and
methodological behaviorism,
I
prefer
to take a behav
99
Willard
Day
ioral
approach
to the matter. I
conceptualize methodological
behaviorism as a
particular pattern
of
controlling contingencies:
it is the
pattern
of
stimulating
circumstances,
now orthodox in most
university psychology departments,
which
acts to
shape
the
professionally respectable
skills and
practices
at controlled
exper
imental research with which we are all familiar. These
stimulating
circumstances
also set
up complex,
but not
conventional,
verbal
repertoires concerning
the nature
and aims of
science,
the
appropriate interrelationship among psychological
research
findings,
and how to
approach
the
composition
of research
reports.
Radical
behaviorism,
on the other
hand,
I
conceptualize behaviorally
as the effect
of Skinner's
thought upon
the behavior of
people. Particularly
relevant is the
corpus
of Skinner's written work when it acts as a stimulus to enter into the control
of
people's
behavior.
Especially interesting things happen
when this written work of
Skinner's functions to
change people's behavior,
so that
genuinely
different
reper
toires of
responding
are set
up, repertoires
that are
genuinely
new for the
person
so
engaged.
A
particularly
desirable
consequence
of this control occurs when
people
are
equipped by
it with new behavior connected with
practices
of
research,
yielding
new research
practices
which can then be
brought
to bear
upon
an
analysis
of the
environmental control of behavior. Whenever this
happens
we find radical behav
iorism
functioning effectively
as a serious
challenge
to
"very
old ideas." To be
sure,
this new behavior is
only
under
partial
control of Skinner's written
work,
since it
also reflects the
particular history
of the reader and the realities of the current
professional
situation at
any particular period
of time.
I now summarize what I have had to
say
in this
paper
about the difference
between radical and
methodological
behavior.
First,
I take it to be in line with the
most common
professional usage
to
regard methodological
behaviorism as a name
for the broad set of
orienting
values, beliefs,
and
practices
which characterize
general experimental psychology. Mahoney's
account in his 1974 book is as
good
as
any,
so let me
simply
list the six
general
characteristics he
gives
as
frequently
representative
of
methodological
behaviorism:
1. an
assumption
of
macroscopic determinism;
. . .
2. an
emphasis
on
observability:
when two event classes are
hypothetically
mediated
by
one or more
intervening
elements
(e.g., physiological
or entero
ceptive stimuli),
the first and final events must be
observable;
3.
a
pragmatic adoption
of
operationism
in which
independent
and
dependent
variables
are each
clearly
and
objectively specified according
to
the
procedures (operations)
entailed in their measurement;.
. .
the
objectiv
ity
of
operational
definitions is often assessed via measurements of
reliability;
4. a
strong emphasis
on
falsiflability (or testability)
as the cardinal feature
of
meaningful
scientific
hypotheses
and
legitimate empirical
research;
. . .
the
methodological
behaviorist must be able to
specify
what data would
have
bearing
on the truth value of his
hypothesis;
100
Difference
between Radical and
Methodological
Behaviorism
5. an
emphasis
on
controlled
experimentation
as the ultimate
means
for
accumulating
and
refining knowledge
about
behavior;
6. a
positive
valuation of
independent replication
and
generalizability.
when a
specified relationship
has been observed
by
one or more
indepen
dent
researchers,
confidence in its
legitimacy
is
increased;
when a relation
ship
has been shown to be
applicable (generalizable)
to other
subject
populations
or similar
phenomena,
the breadth of its relevance is enhanced.
(1974, pp. 16-17)
Clearly,
it is this set of
orienting
attitudes that has become orthodox in most
university psychology departments.
It is
important,
however,
in
reading
Skinner to
be aware of the fact that his
conception
of
methodological
behaviorism is much
narrower than this. When Skinner
distinguishes
radical behaviorism from metho
dological
behaviorism,
he refers
by methodological
behaviorism to what he has
called "the
operationism
of
Boring
and
Stevens,"
a
conception
in which
private
events were either identified with verbal
reports
about
them,
or held to be
incapable
to
playing
any
role in a scientific account of behavior.
When it comes to a statement of what radical behaviorism
is,
the most
straight
forward
thing
to
say
is that it is the
attempt
to account for behavior
solely
in terms
of natural
contingencies:
either
contingencies
of
survival, contingencies
of rein
forcement,
or
contingencies
of social evolution.
However,
Skinner has
recently
chosen to
speak
of radical behaviorism as the
philosophy
of a
special discipline
within
psychological
research
spoken
of as the
experimental analysis
of behavior.
Yet it should be
kept
in mind that when Skinner
attempts
to characterize this
philosophy,
the account
consistently
involves
a
complex argument centering
on
practices
of the verbal
community
in
teaching
us to talk about
our
behavior,
both
private
and
public.
I
myself prefer
to take a
somewhat different
approach
to
conceptualizing
radical
behaviorism.
For
me,
radical behaviorism is the effect that Skinner's
thought
happens
to have
on the behavior of
people.
To the extent that this effect involves
establishing
new
repertoires
of
responding
that
seriously challenge
the
very
old
ideas we have in our culture about how behavior is to be
explained,
I believe that
radical behaviorism is both
interesting
and
important.
REFERENCES
Day,
W.F. On certain similarities between the
Philosophical investigations
of
Ludwig Wittgenstein
and the
operationism
of B.F. Skinner. Journal
of
the
Experimental Analysis of
Behavior, 1969,
12,
489-506.
Day,
W.F.
Contemporary
behaviorism and the
concept
of intention. In J.K. Cole & W.J. Arnold
(Eds.),
Nebraska
Symposium
on
Motivation,
1975. Lincoln:
University
of Nebraska
Press, 1976,
67-131.
Day,
W.F. The historical antecedents of
contemporary
behaviorism. In R.W. Rieber & K.
Salzinger
(Eds.), Psychology:
Theoretical-historical
perspectives.
New York: Academic
Press,
1980.
Ferster, C.B.,
& Skinner,
B.F. Schedules
of reinforcement.
New York:
Appleton-Century-Crofts,
1957.
101
Willard
Day
Mahoney,
M.J.
Cognition
and behavior
modification. Cambridge: Ballinger,
1974.
Malone, J.C., Jr., James, William,
&
Skinner,
B.F.
Behaviorism, reinforcement,
and interest. Behav
iorism, 1975, 3,
140-151.
Ringen,
J.
Explanation, teleology
and
operant
behaviorism.
Philosophy of
Science, 1976,43,
223-253.
Sidman,
M. Tactics
of scientific
research. New York: Basic
Books,
1960.
Skinner,
B.F. The behavior
of organisms.
New York:
Appleton-Century-Crofts,
1938.
Skinner,
B.F The
operational analysis
of
psychological
terms.
Psychological
Review, 1945, 52,
270-277,
291-294.
Skinner,
B.F. Science and human behavior. New York:
Macmillan,
1953.
Skinner,
B.F. Verbal behavior. New York:
Appleton-Century-Crofts,
1957.
Skinner,
B.F. Cumulative record. New York:
Appleton-Centrury-Crofts,
1959.
Skinner,
B.F. Behaviorism at
fifty.
In T. W. Wann
(Ed.),
Behaviorism and
phenomenology. Chicago:
University
of
Chicago Press,
1964.
Skinner,
B.F.
Beyond freedom
and
dignity.
New York:
Knopf,
1971.
Skinner,
B.F. About behaviorism. New York:
Knopf,
1974.
Taylor,
C. The
explanation of
behaviour. New York: Humanities
Press,
1964.
The unmanifest. The New
Yorker,
April 4, 1977, pp.
28-29.
Zeiler,
M.D. Editorial. Journal
of
the
Experimental Analysis of
Behavior, 1977, 27,
1-2.
Zuriff,
G.E. Radical behaviorist
epistemology. Psychological
Bulletin, 1980, 87,
337-350.
102