Matched-Dependent Behavioralism: the Cargo Cult in Political Science

• Author: Francis D. Wormuth
• DOI: 10.1177/106591296702000401
• Published date: 01 December 1967
• Date: 01 December 1967
• Subject Matter: Articles

MATCHED-DEPENDENT BEHAVIORALISM:
THE CARGO CULT IN POLITICAL SCIENCE
FRANCIS D. WORMUTH
University of Utah
E. MILLER AND JOHN DOLLARD, in their
EAL
study of imitation,’
identity a type which they call &dquo;matched-dependent behavior.&dquo; They illus-
jL ~ trate it with a story. The father told Jim, aged six, and Bobby, aged three,
that he had hidden a piece of candy for each of them in the living room. Jim looked
in the fireplace: no candy. Bobby looked in the fireplace: no candy. Jim looked in-
side the piano bench: no candy. Bobby looked inside the piano bench: no candy.
Jim looked under a cushion on the sofa, found a piece of candy, and ate it. Bobby
looked under the cushion: no candy. &dquo;Bobby was now helpless&dquo; ; he could think of no
other place to look. Finally his father gave him the candy. &dquo;On a succeeding trial of
the same game, exactly the same thing happened. The younger child would look only
in the places already examined by his older brother. He would not respond to place
cues by looking for himself.&dquo;
More than twenty years ago Gordon W. Allport, in an address as Divisional
President to the Division of Personality and Social Psychology of the American Psy-
chological Association, complained that &dquo;many of us seem so stupefied by admiration
of physical science that we believe psychology in order to succeed need only imitate
the models, postulates, and methods of physical science.&dquo;2 The same stupefaction has
gripped those political scientists who call themselves behavioralists. We may liken
the physical scientist to Jim; the behavioralist, who emulates his models, postulates,
and methods, is Bobby. Miller and Dollard called Bobby’s behavior maladaptive, but
they were wrong, for he always got his candy. Similarly the behavioralists have never
found a piece of candy; but they are given their candy by the Ford Foundation.
Gordon Allport was not complaining of the undertaking of psychology, but of
the diversion of that undertaking into imitative and unrewarding channels. There is
no intention here to complain of the undertaking in political science, as old as Aris-
totle, to deal dispassionately and systematically with the data. In particular there is
no intention to condemn
the use of psychology -
or at any rate psychologies like All-
port’s -
in the explanation of politics. But the use of psychology is not new; it goes
back to Plato.
Harold D. Lasswell has said that the &dquo;behavioral upswing&dquo; dates from the work
of Charles E. Merriam at the University of Chicago in the 1920’s and early 1930’s.3
But in fact the method and the assumptions of Merriam were traditional. Behavior-
alism is novel; Robert A. Dahl has called it a successful revolution.~
4
1

Social Learning and Imitation (New Haven: Yale U. Press, 1941), excerpted in Melvin H.
Marx (ed.), Psychological Theory: Contemporary Readings (New York: Macmillan,
1951), pp. 543-49, at p. 545.
2
"Scientific Models and Human Morals," in Marx, op. cit., pp. 156-70, at p. 157.
3

The Future of Political Science (New York: Atherton Press, 1963), p. 37.
4
"The Behavioral Approach in Political Science: Epitaph for a Monument to a Successful Pro-
test," American Political Science Review, 55 (1965), 763-72, at p. 766.
809

---

810
Although the adjective &dquo;behavioral&dquo; was used occasionally in the 1930’s, the
term &dquo;behavioral science&dquo; appears to have been introduced by the Committee on Be-
havioral Sciences, an interdisciplinary group organized at the University of Chicago
in 1951 under the leadership of a psychologist, James G. Miller, with the encourage-
ment of Dean Ralph W. Tyler. Miller has recorded the &dquo;working assumptions&dquo; of
the Chicago &dquo;theory group&dquo; :
First, we agreed to accept as confirmation of theorems only ob jective phenomena available
to public inspection by more than one observer, excluding private experience. Second, we tried
when possible to state hypotheses quantitatively, so that they might be precisely testable and
could subsequently be corrected. Third, we attempted to make statements capable of being dis-
proved as well as proved, by crucial experiments. Finally, as will be explained below, insofar as
possible we employed dimensions of natural science related to the centimeter-gram-second sys-
tem.5
The first assumption is negative rather than affirmative : it is the rejection, com-
mon to behavioralism and psychological behaviorism, of both subjective data and
subjective method. The second and the fourth make a commitment to the mathe-
matical method of natural science. The third appears to be a bow in the direction of
Karl Popper.
But these were not in fact the controlling assumptions of the Chicago group.
Miller says that &dquo;we have found most profit in general behavior systems theory,&dquo; and
he goes on to describe this. The term &dquo;system theory&dquo; was introduced in 1945 by
Ludwig von Bertalanffy when he offered an alternative to the analytical method of
natural science.
In the world view called mechanistic, born of classical physics of the 19th century, the aimless
play of the atoms, governed by the inexorable laws of mechanical causality, produced all phe-
nomena in the world, inanimate, living, and mental. No room was left for any directiveness,
order, or telos. The world of organisms appeared a mere product of chance, accumulated by a
senseless play of mutation at random and selection; the mental world as a curious and rather
inconsequential epiphenomenon of material events.’
This &dquo;scheme of isolable units acting in one-way causality has proved to be in-
sufficient. Hence the appearance, in all fields of science, of notions like wholeness,
holistic, organismic, gestalt and so forth, which all signify that in the last resort, we
must think in terms of systems of elements in mutual interaction.&dquo;’ In short, Bertal-
anffy rejected &dquo;the inexorable laws of mechanical causality&dquo; in favor of an Aristote-
lian teleology. Lurking behind this teleology is a vitalism like that of Driesch. It is
not an open-ended vitalism like Bergson’s; it aims at a given &dquo;final state,&dquo; an Ari-
stotelian final cause.
Mathematical behavioralism, with its commitment to analytical method and to
efficient cause, is at the opposite pole from general systems theory, built as it is on an
Aristotelian or Thomistic foundation. Yet some behavioralists believe they can com-
bine the two. David Easton, in a book entitled A Systems Analysis of Political Life,8
5
"Toward a General Theory for the Behavioral Sciences," American Psychologist, 10 (1955),
513-31, at p. 514.
6
Ludwig von Bertalanffy, "General System Theory," in General Systems, 1 (1956), 1-10, at
p. 6.
7

Loc. cit.
8
New York: Wiley, 1965, p. 5.

---

811
says that his goal is &dquo;descriptive, empirically-oriented, behavioral, operational or
causal theory.&dquo;
In 1954 the Ford Foundation established the Center for Advanced Study in the
Behavioral Sciences at Menlo Park, California, with Ralph Tyler from the Univer-
sity of Chicago as Director. The Center became the first training place for behavior-
al political scientists. It still functions, but now many graduate schools offer the same
training. At Menlo Park, however, an atomistic and reductionist approach by way
of mathematics very largely displaced the holistic approach of Miller’s general sys-
tems theory, and the latter now has its strongholds at the University of Chicago and
the Mental Health Research Institute at the University of Michigan, to which Miller
migrated.
Behavioral projects, principally mathematical, are carried on by the RAND
Corporation and the Pentagon; others are financed by various organs of the national
government; the Ford Foundation has been generous with financial assistance. We
may offer a behavioral hypothesis that the number of pages of behavioral literature
published varies directly with the cube root of the number of dollars spent.
Harold Lasswell has said of the Center for Advanced Study:
The chief intellectual stress was methodological.... The methodological concern was threefold
-
( 1 ) to encourage mathematical thinking or comparably strict methods; (2) to stimulate think-
ing about social processes in human history by giving consideration to general biological theory;
(3) to provide a setting appropriate to the planning of research on problems of importance that
are neglected or pursued by inferior methods unless recognized and undertaken by interdiscipli-
nary teams.
BI
The third topic can be dismissed. There have been collaborations of behavioral
political scientists with economists, but this has not pooled methods familiar to the
two disciplines. Rather, these teams have invariably had recourse to the tools of the
mathematicians. Apparently Lasswell’s second topic is an understatement of the
ambition of general systems theory. Many persons call themselves systems theorists
- in addition to Bertalanffy and Miller, Talcott Parsons, David Easton, Morton
Kaplan, Coleman and Almond, Bertram Gross, and others. Although there are
vague resemblances between the formulations of these authors, there are greater in-
consistencies, and it is not feasible to review this literature here.
The chief impact of behavioralism has been Lasswell’s first topic, the adoption
of new mathematical tools. This is often unaccompanied by any degree of mathe-
matical sophistication; and frequently it is entirely unaccompanied by political so-
phistication.
Since Petty, at least,...