POLITICAL

POWER
A Reader in Theory and Research

Roderick Bell

David V . Edwards

R. Harrison Wagner

The Free Press, New York

Collier-Macmillan Limited, London

 Copyright © 1969 by The Free Press
A DIVISION OF THE MACMILLAN COMPANY
Printed in the United States o f America

All rights reserved. No part of this book may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopying, recording, or by any information storage
and retrieval system, without permission in writing from the Publisher.

Collier-Macmillan Canada, Ltd., Toronto, Ontario

Library of Congress Catalog Card Number: 75-84214

First Printing
Contents

page
Preface vii
I. The Study of Political Power: An Introduction to the Problems o f Theory and
Measurement 1
1. T h e C o n c e p t o fP o w e r a n d t h e S t u d y of P o l it ic s 3
R. Harrison Wagner
2. P o l it ic a l P o w e r : T h e P r o b l e m o f M e a s u r e m e n t 13
Roderick Bell
II. Political Power and Political Science 29
Introductory note 29
3. How t o S t u d y C o m m u n it y P o w e r : T h e P l u r a l is t A l t e r n a t iv e - 31
Nelson W. Polsby
4. A C r it iq u e o f t h e R u l i n g E l it e M o d e l 36
Robert A. Dahl
5. “ P o w e r E l i t e ” o r “ V e t o G r o u p s ” ? 42
William Kornhauser
6. T h e B a l a n c e o f P o w e r : P r e s c r i p t i o n , C o n c e p t , o r P r o p a g a n d a 53
Ernst B. Haas
III. The Concept of Power 67
Introductory note 67
\ 7. N o t e s o n t h e O b s e r v a t io n a n d M e a s u r e m e n t o f P o w e r 69
Herbert A. Simon
8. T h e C o n c e p t o f P o w e r * 79
Robert A. Dahl
9. Two F a c e s o f P o w e r * 94
Peter Bachrach and Morton S. Baratz
10. D e c is io n s a n d N o n d e c i s i o n s : A n A n a l y t ic a l F ram ew ork c 100
Peter Bachrach and M orton S. Baratz
11. S o m e A m b ig u it ie s i n t h e N o t io n o f P o w e r 110
William H. Riker
x Contents

IV. Theory Formation and Measurement

Introductory note
U 2. I n f l u e n c e , L e a d e r s h ip , C o n t r o l 123
Dorwin Cartwright
& 13. A n I n t r o d u c t io n t o t h e T h e o r y a n d M e a su r e m e n t o f I n f l u e n c e 166
James G. March
% 14. M ea su r em en t C o n c e pt s in t h e T h e o r y o f I n fl u e n c e 181
James G. March
o 15. Sp u r io u s C o r r e l a t io n : A C a u s a l I n t e r p r e t a t io n 194
Herbert Simon
16. T h e L i n k a g e B e t w e e n C o n s t it u e n c y A t t it u d e s a n d C o n g r e s s io n a l
V o t in g B e h a v io r : A C a u s a l M o d e l 202
Charles F. Cnudde and Donald J. McCrone
17. A M e t h o d f o r E v a l u a t i n g t h e D is t r i b u t i o n o f P o w e r in a C o m m it t e e
System 209
L. S. Shapley and Martin Shubik
18. A T e st o f t h e A d e q u a c y o f t h e P o w e r I n d e x 214
William Riker
hl9. M e a s u r e m e n t o f S o c ia l P o w e r , O p p o r t u n it y C o st s, a n d t h e T h e o r y
of T w o - P e r s o n B a r g a in in g G am es 226
John C. Harsanyi
* 2 0 . M e a s u r e m e n t o f S o c i a l P o w e r i n « - P e r s o n R e c ip r o c a l P o w e r S i t u a ­
t io n s 239
John C. Harsanyi

V. The Concept of Power and Theories of PoliticalSystems 249

Introductory note 249
*21. O n th e C o n c e p t o f P o l i t i c a l P o w e r * 251
Talcott Parsons
22, T o w a r d a P u re T h eo ry of T h r ea t S ystem s 285
Kenneth E. Boulding
2 3 . D if f e r e n t ia t io n o f P o w e r * 293
Peter Blau
«24. T he Str u c t u r e of I n flu en c e 309
Edward C. Banfield
* 2 5 . C o n c e r t in g A c t io n by I n f l u e n c e 325
Edward C. Banfield
2 6 . I n t e r n a t io n a l St r u c t u r e , N a t io n a l F orce, and the Balance of
W orld Pow er 335
Kenneth N. Waltz
2 7 . M e a s u r in g t h e C o n c e n t r a t i o n o f P o w e r i n P o l it ic a l S y s t e m s 346
Steven J. Brams

Notes 363

Bibliography 395

10
Decisions and Non­
decisions:
An Analytical
Framework
Peter Bachrach
,.5 i
\ l \\r j
and Morton S. Bar atz
\ v
In this article Bachrach and Baratz am plify their earlier
criticism o f com m unity power studies, and develop their
own scheme for analyzing political decision-making.
Originally published in 57 A m erican P olitical Science
Review (1963) 632-642, it is reprinted here with the
perm ission o f that jo u rn a l and the authors.
I n recent years a rich outpouring of case
-studies on community decision-making has
been combined with a noticeable lack of
generalizations based on them. One reason
for this is a commonplace 5 we have no
general theory, no broad-gabge model in
terms of which widely different case studies
can be systematically compared and con­
trasted.
Among the obstacles to the development
of such a theory is a good deal of confusion
about the nature of power and of the things
that differentiate it from the equally im­
portant concepts of force, influence, and
authority. These terms have different mean­
ings and are of varying relevance; yet in
nearly all studies of community decision­
making published to date, power and in­
fluence are used almost interchangeably,
and force and authority are neglected.1 The
researchers thereby handicap themselves.
For they utilize concepts which are at once
too broadly and too narrowly drawn: too
broadly, because im portant distinctions
between power and influence are brushed
over; and too narrowly, because other con­
100
Notes to chapter 10 are found on pages 370 to 372.
cepts are disregarded—concepts which, had
they been brought to bear, might have
altered the findings radically.
Many investigators have also mistakenly
assumed that power and its correlatives are
activated and can be observed only in
decision-making situations. They have over­
looked the equally, if not more important
area of what might be called “ nondecision­
making,” i.e., the practice of limiting the^
scope of actual decision-making to “ safe”
issues by manipulating the dominant com­
munity values, myths, and political institu­
tions and procedures. To pass over this is to
neglect one whole “face” of power.2
Finally, the case studies are often based
upon inarticulate, perhaps unsound, pre­
mises which predetermine the findings of
“ fact.” 3 A variety of complex factors affect
decision-making—the social, cultural, ec­
onomic, and political backgrounds of the
individual participants; the values of the
decision-making body as an entity in itself;
the pressures brought to bear on the decision­
makers, individually and collectively, by
groups at interest; and so on. To say, as
some do, that these factors are equally im­
portant is as far from the mark as it is to
assume as others do, that only one is of
overriding significance.4
What is required, then, is a model in
terms of which the determinants both of
decision- and nondecision-making can be
appraised, taking full account of the distinct
concepts of power, force, influence, and
authority. In this paper we are not so
ambitious. We attempt only to lay some of
the groundwork for a model, seeking (1)i to
.clarify the attributes of what we consider
key concepts for any study of decision- and
nondecision-making and the essential differ­
ences among them, and (2) to show how
these concepts can be utilized more system­
atically and effectively in case studies.
i
It is customary to say that this or that
person or group “ has power,” the implica­
tion being that power, like wealth, is a
possession which enables its owner to
secure some apparent future G ood.5 An­
other way of expressing the same point of
view is to say that power is a “ simple
 property . . . which can belong to a person Decisions and Nondecisions: 101
or group considered in itself.” 6 An A nalytical Fram ework
Peter Bachrach and Morton S. Baratz
For at least three reasons this usage is
unacceptable. First, it fails to distinguish
clearly between power over people and ^ Let us now suppose that a second man
power over m atter; and “power in the approaches the sentry and, like the first, is
political [or economic or social] sense ordered to stop or be shot. But the second
cannot be conceived as the ability to stranger ignores the order, attempts to
produce intended effects in general, but smash through the gate, and is forthwith
only^such effects as involve other, persons.. fatally wounded. If we assume that the
. . .” 7; Second, the view that a person’s intruder’s intention was to sabotage the.
power is measured by the total number of military installation, we can have no doubtV
desires that he achieves is erroneous; one J h a t his and the sentry’s values were in
cannot have power in a vacuum, but only in direct conflict. Even so, the sentry’s,.fatal
relation to someone else. Third and fmost shot did «of constitute an exercise of power. ; ’
inportant; the common conception of the For it did not bring about compliance to •
phenomenon mistakenly implies that pos­ his order—and it did not because, appar- .
session of (what appear to be) the instru- ently, the intruder valued entry to the base ■’
rpents of power is tantam ount to possession more highly than either obedience to the
/6 f power itself. Such a notion is false sentry’s order or his own wellbeing.
because it ignores the fundamental relational Suppose, finally, that a third man ap­
attribute of power: that it cannot be pos­ proaches the sentry box, a man who wants
sessed; that, to the contrary, the successful to die but cannot bring himself to the act of
jexercise of power is dependent upon the self-destruction. He therefore deliberately / \
relative importance of conflicting values in ignores the sentry’s command and is duly j
the mind o f the recipient in the power rela- shot to death. Did someone in this situation
tionship! /; &'■' ■" • -v ■ have power and exercise it ? As we see it, the y 54
A few illustrations should clarify and “ victim” did—for it was he, cognizant of
enlarge our position. Imagine, first, an the conflict of values between himself and - ,•
armed military sentry who is approached the guard, who utilized the latter’s supposed,
by an unarmed man in uniform. The sentry sanction to achieve his own objective.9 y )
levels his gun at the intruder and calls out, We reiterate that power is relational, ás,
“ Halt or I ’ll shoot!” The order is promptly opposed to possessive or substantive. Its
obeyed. Did the sentry therefore have relational characteristics are threefold. ,
power and exercise it? So it would seem; _First, in order for a power relation to exist
but appearances could be deceiving. For ^ there must be a conflict of interests or values _
suppose that the intruder obeyed, not . between two dr more p e r s ^ s or groups, y i
because he felt compelled to do so in the ...Such a divergence is a necessary condition J
face of the threatened sanction, but because of power because, as we have suggested, if ‘
he was himself a trained soldier for whom A and B are in agreement as to ends, B will
prom pt obedience to a sentry’s order was^ freely assent to A’s preferred._ course_of
part of a system of values he fully accepted.;8/ action; J n which case the situation will
>If that was the case, there was no conflict of involve authority . rather than power.10 ;
goals or interests between the two princi­ . Second, a power relationship exists only if'! V
pals; the sentry’s threatened sanction was B actually bows to A’s wishes. A conflict of
irrelevant, and the result would have been interests is an insufficient condition, since ( fi
the same if he, and not the intruder, had A may not be able to prevail upon B jto l .^
been unarmed. Because the soldier put change his behavior. And if B does not .
' obedience to a sentry’s order at the top of comply, A’s policy will either become a, , \
his schedule of values, the threat of severe dead letter or will be effectuated through the '
deprivations had no bearing on his be­ . exercise of force rather th a n . through. ^•
havior. In such circumstances it cannot be .power.11 Third, a power relation can exist . v
said that the guard exerted power. only if one of the parties can threaten to
 e 102 III T h e Concept of Power (d) The person threatened is persuaded n
that the threat against him is not idle, that
invoke sanctions: power is “ the process his antagonist would not hesitate in fine
^ of affecting policies of others with the help actually to impose sanctions. To illustrate,
of (. . . threatened) severe deprivations for if a famous general calculates that the
nonconformity with the policies intended.” 12 President lacks the will or the popular sup­
It must be stressed, however, that while the port to employ his Constitutional preroga­
availability of sanctions—that is, of any tives, he may ignore—even defy—the
promised reward or penalty by which an President’s policy instructions.16 Or, again,
actor can maintain effective control over the success of a resistance movement based
policy—is a necessary condition of power, on the principle of nonviolence rests in
it is not sufficient. It is necessary simply large measure upon the assumption that
because the threat of sanctions is what those who can invoke sanctions will refrain
,e differentiates power from influence13; it is from doing so, that value conflicts within A
insufficient f because the availability of a will prevent him from carrying out his threat
sanction Mdbws A with power over B only against B. In point are the Indians who sat f
if the following conditions are m et: on the railroad tracks in defiance of the
(a) The person threatened is aware of British and got away with it because (as the
what is expected of him. In a power situa­ Indians well knew) the British put a higher ;•
tion there must be clear communication value on human life than on obedience to i
between the person who initiates policy and their orders.17
the person who must comply.14 If our We can now draw together the several ele­
imaginary sentry challenges a man who ments of our conception of power. A power
understands no English or is perhaps deaf, relationship exists when (a) there is a conflict
the sentry has—at least at the moment he over values or course of action between A
f and B; (b) B complies with A ’s wishes; and
issues his order—no power. In other words, (c) he does so because he is fearful that A will
power has a rational attribute: for it to deprive him of a value or values which he, B, 0
exist, the person threatened must com­ regards more highly than those which would
prehend the alternatives which face him in have been achieved by noncompliance.18
vc
choosing between compliance and non- Several points must be made in reference Í
compliance. to this definition. First, in speaking of \ /•
" (b) The threatened sanction is actually power relations, one must take care not to
4 regarded as a deprivation by the person overstate the case by saying that A has
who is so threatened. A threat by the power over B merely because B, anxious to'' r V
W avoid sanctions, complies with a given
LV President to “ pijrge” a Congressman for
Cç, failure to support the Administration’s policy proclaimed by A. This could well b e ;V
tV legislative program would be to no avail if an inaccurate description of their relation-
the Congressman reckoned that his chances ship, sinceyVs power with respect to B may
•V '
for reelection woMd tíè increased rather . be extremely limited in scope, i.e., in range
than reduced by Presidential intervention. of values affected.19 Thus, the power of a
(c) The person threatened has greater traffic policeman over a citizen may be con­
esteem for the value which would be sacri­ fined to the latter’s activities as a motorist— >
ficed should he disobey than for another and no more than that. Moreover, in -
value which would be foregone should he appraising power relationships account
comply. Fear of physical injury did not must be taken of the weight of power, i.e.,
deter those Southern Negro “ sihtfers-in’’ the degree to which values are affected, and ‘
who put greaterlk o r è 'by the righítebUsness of its domain, i.e., the number of persons
of their cause. It is worth noting at this affected.20 For example, the power of the
stage that threatened deprivations are often Chairman of the House Committee on Vi
ineffectual because the policy-initiator, in Ways and Means is limited mainly to fiscal
deciding what sanction to invoke, mistaken­ affairs; but within this scope he wields
ly projects h|& own values into the minds of immense power in the determination of
his subjects^15; Federal tax and expenditure policies
(weight), which affect a vast number of
persons—up to and including at times the
President himself (domain).
Finally, account must be taken of what
Friedrich has dubbed the “rule of antici­
p ated reactions. ” 2HTh£ jp ro b le ^ posed by
this phenomenon is that an investigation
might reveal that, though B regularly
accedes to A ’s preferred courses of action,
A in fact lacks power over B because A
just as regularly tailors his demands upon
B to dimensions he thinks B wifl^accept. As
an illustration, if the President submits to the
Congress only those bills likely to be pala­
table to a majority of lawmakers, he can
hardly be said to have power over the Con­
gress simply because all his proposals are
enacted into law.
11 í '" .*• ;
In Robert Bierstedt’s opinion, “ force is
manifest power . .. Force . . . means the re­
duction or limitation or closure or even
total elimination of alternatives to the social
action of one person or group by another
?person or group. ‘Your money or your
life’ symbolizes a situation of naked force,
the reduction of alternatives to two.” 22
porce, in short, is power exercised.
We reject this view. As we see it, the
essential difference between power and
'force is simply that in a power relationship
one party obtains another’s compliance,
while in a situation involving force one’s
objectives must be achieved, if at all, in the
face of «^com pliance/23' Thus, if A ’s
demand for B’s money or his life prompts
B to surrender his wallet, A has exercised
power—he has won B’s compliance by
threat of even more severe deprivations.
But if A must kill B to get the money, A
has to resort to force—he must actually
invoke the threatened sanction—and there­
by perhaps expose himself to severer de-
privations too. By the same token, if and
when thermonuclear weapons are trans­
formed from instruments of a policy of
deterrence into activated missiles of death,
power will have given way to force.
There is another difference between the
- two concepts. A person’s scope of decision­
making is radically curtailed under the
duress of force; once the fist, the bullet, or
Decisions and Nondecisions: 103
An A nalytical Fram ework
Peter Bachrach and Morton 5. Baratz
the missile is in flight, the intended victim is
stripped of choice between compliance and
noncompliance. But where power is being
exercised, the individual retains this choice.
Put another way, in a power relationship
it is B who chooses what to do, while in a
force relationship it is A .24:)." j'-h, \ . ; Y |r :-
It follows from the foregoing that manU ^
pulation is an aspect of force, not of pow er’f" ) ;
For, once the subject is in the grip of the^ . ,
manipulator, he has no choice as to course
of action. It can be said, therefore, that !
force and manipulation (as a sub-concept
under it) are, in contrast to power, non-
rat ion al.
An additional distinguishing attribute of
force is that in some circumstances it is non-
jelational. For instance, if B is shot in thév\v.
back by an unknown robber, he and his .
assailant have only a minimal interrelation­
ship—especially when compared to a power
confrontation where B must decide whether
to accede to A’s demands. A similarly^
minimal relationship obtains in cases i n - \ ^ ' -•
volving manipulation, where compliance is .
forthcoming in the absence of recognition
on the complier’s part either of the source "
or the exact nature of the demand upon him. t
In short, force and manipulation, like i \ *'
power, involve a conflict of values; b u t ', ; /
unlike power, they are non-rational and
tend to be non-relational.
A number of implications may be drawn ^
from this reasoning. One is that the actualV 1
application of sanctions is an admission of
defeat by the would-be wielder of power.
And so it is, to the extent that the. prior i
Jhreat of sanctions failed to bring about the *
desjred behavior. A good case in point is \^ -
the action of President Harry S. Truman in fS*.
1951 when he relieved General D ouglasr '
M acArthur of his command in the Pacific
on grounds of insubordination. By con- " --
tinuing to air in public his policy differences
with the Administration, M acArthur vir­
tually compelled Truman to dismiss him.
The President’s decision to apply sanctions
was, however, an admission of defeat, an
implicit recognition that he could not. by
 104 III T h e Concept of Pow er gressmen by withholding patronage. But as
a President exchanges a job appointment
power or authority, obtain M acArthur’s for votes—that is, as he successfully utilizes
compliance to the Administration’s policy this source of power—his reserves for
of a negotiated settlement of the Korean effecting further compliance dry up. As a . /
hostilities. To be sure, policy defeats of this corollary ^repeated threats to invoke sane-'
k ind may prove to be only partial. For if tions—threats never carried out—will gra- '
the resort to force against one party effect- dually lose credibility in the minds of those v
ively deters noncompliance on the part of threatened, until at length the threats cannot %
others, now or infuture,~t’He’empToymei^ produce the desired behavior. This, in the
of sanctions becomes a fresh declaration of view of many, was the basic flaw in the im­
the existence of power. This is, of course, plementation of the stated American policy
the rationale of all who undertake punitive during the late 1950s of “ massive retaliation
actions against others: the use of force in at times and in places of our own choos­
*one situation increases the credibility of ing.” 20 The same phenomenon applies to
threats to use it in others. interpersonal relationships: a threat to
At the same time, it is im portant to recog­ withdraw one’s love for another may be
nize that resort to force can result in a loss highly potent the first time, yet prove
of power. Two cases can be distinguished. totally ineffectual if used again.
First, the invocation of sanctions often
/* I/..' .
causes a radical reordering of values within HI ! J /flr ' " ■
the coerced person (as well as in those per­ ■ One person has influence over another
sons who identify closely with him), thereby within a given scope to the extent that the-J
undermining the pre-existing power rela­ first, without resorting to either a tacit or an v>
tionship. A good illustration is provided ' overt th reat of severe deprivations, causes .
by the largely abortive attempts of the the second to change his course of action.
Nazis during World War II to pacify the ; Thus, power and influence are alike in th at .
populations of occupied countries by killing each has both rational and relational
civilian hostages. Contrary to German attributes. But they are different in that the
^expectations, this policy produced a marked exercise of power depends upon potential
stiffening'^f resistance; evidently, the num­ sanctions, while the exercise of influence
ber of “ prisoners” who put a higher value does not. And there is an important differ­
on freedom than on life itself rose sharply. ence between influence and manipulation:
Second, the deprivation may prove in in situations involving the latter, but not
retrospect far less severe than it appeared - the former, A seeks to disguise the nature
in prospect, as a result of which future non- and source of his demands upon B a n ^ if
compliance is not discouraged and may A is successful, B is totally unaware that-;
even be encouraged. For example, a child - something is being demanded of him. ~ >
whose punishment for misbehavior is the Although power and influence can and •;/
tem porary loss of a prized toy may find, ex must _be^distinguished, the line between
' post facto, that the loss is entirely bearable, them is usually difficult to draw. This is
th at the satisfactions he gained from acting especially true where B’s reasons for acting
up are greater at the margin than Jhe_ in accordance with A’s wishes are confused
jilternative foregone. In such circumstances, or multiple; in such circumstances B himself
obviously, future defiance of parental will be unable honestly to say whether his
orders is more likely than not. behavior was prompted by a fear of sane- .
Just as power may be lessened when force ..tions. or, rather, by his esteem for “ higher”
is resorted to, so also may power be lessened values (e.g., wealth, respect^rpòwèr^ wisdo
when it is successfully exercised, i.e., when "tharfthe one immediately ;a t stake. Does the
compliance is obtained by mere threat of ambitious young man who 7 submits un­
sanctions. As an illustration, Presidents of happily to the every dictate of his rich uncle
the United States have traditionally sought do so because he admires wealthy men
to exercise power over recalcitrant Con­ (influence)
i or because he feels that un- *
questioning obedience is the price of a
generous inheritance in the future (power) ?
Does the Majority Leader who unwillingly
manages an Administration b ill in the
Senate do so because he is in awe of the
Presidency and hence of the "man who
occupies the office (influence), or because
he fears the President will actually punish
him for noncompliance (power)? To say
that the decisive test in situations like these
turns on whether compliance is “ voluntary”
or “ involuntary” is, in our judgment, not
particularly helpful.26
The difficulty in distinguishing sharply
and clearly between power and influence is
further complicated by the fact that the two
are often mutually reinforcing, that is,
power frequently generates influence and
vice versa. On this score, the case of Senator
Joseph R. McCarthy of Wisconsin is
especially instructive.27 Shrewdly posing
as the principal defender of the national
security at the very moment when that
became the dominant social value vice the
inviolability of civil liberties * McCarthy
managed for a period to stifle virtually all
opposition to himself and what he stood for
(influence). And from this base he was able
to gain power, that is, to affect the making
of actual decisions (votes in the Senate, acts
of the Executive, etc.) by threats of severe
deprivations (intervention in State political
campaigns, destruction by accusation of the
careers of appointive officials, etc.). By the
same token, however, as public fears about
national security subsided and concern for
civil liberties grew, M cCarthy’s capacity to
influence others sharply waned—and so,
too, did his power.
Just because the distinction between
power and influence is often blurred does
not, however, lessen the importance of
making the distinction. Nikita Khrushchev
has little or no influence over Americans,
yet it is obvious he exercises considerable
power over us. On the other hand, the
Supreme Court of the United States has
widespread influence (and authority) over
us both individually and collectively; its
power is slight indeed.
IV ' ■’•"■'■'n
While authority is closely related to
Decisions and N ondecisions: 105
An A nalytical Fram ework
Peter Bachrach and Morton S. Baratz
power, it is not a form thereof; it is, in fact,
antithetical to it.28 In saying tffist_ we
reject both the traditional definition of
authority as “ formal power”29 and that
which conceives it as “ institutionalized
power.” 30
To regard authority as a form of power
is, in the first place, not operationally
useful* If authority is “ formal power,” then
one is at a loss' to know who has authority
at times when the agent who possesses
“ formal power” is actually powerless; to
say that Captain Queeg continued to have
authority on the USS Caine after he was
deposed of his command by the mutineers
is to create needless confusion. Further­
more, to define authority as ‘Tormal power”
is to fail to delineate the bounds of authority,
other perhaps than to say that it ends where
“ real power” begins. For those who believe
in limited or constitutional government
such a construction is unthinkable.
To argue that “ formal power” is circum­
scribed by law is also no answer. For it
assumes without warrant the legitimacy of
law. A policeman who demands obedience
in the name of a law that i*. considered
basically unjust will possess 1L v authority
in the eyes of persons steeped in the Anglo-
American legal tradition. N or is the problem
completely solved by conceiving of authority
in terms of constitutional legitimacy. Such
a conception presupposes that all members
of the community give allegiance to the
constitution and the courts which interpret
it. Do Federal courts have the authority to
issue desegregation orders to southern
school districts ? According to many South­
erners, including some learned in the law,
the answer is in the negative. ( 7^)
Friedrich’s analysis of authority s e e m s J o ^
us the_most__appropriate. He defines the .
concept as “ a quality of communication ” ^
that possesses “ the potentiality of reasoned
elaboration.” 31 Like power, a u th o ritie s
here regarded as arelational co n cep t:Jtjs
not that A possesses authority, but that B
regards A’s communication as authoritative .
Also like power, an authority jrelationshir
 106 III T h e Concept of Pow er
implies rationality—although of a different
order. That is, in a situation involving
power, B is rational in the sense that he
chooses compliance instead of defiance
because it seems the less of two evils.32 But meaningful to a majority of the people.
in a situation involving authority, B com­
plies becausd he rèeognizes that the com­
m and is reasonable"in terms of his own
values; in o th er words, B defers to A, not
because he fears severe deprivations, but
because his decision can be rationalized.33
It is not essential, however, that A’s
directive be supported by reasoning; it is
sufficient that the potentiality of such
reasoning be present and recognized.34
If B believes that A ’s communication
allows for reasoned elaboration when in
fact it does not, it is 4‘false” authority.35 game (authority). Authority, in short, can
When the source of obedience shifts from
“ genuine” to “false” authority and B _society^_it can perform the valuable function.
realizes that the communication cannot be
elaborated effectively, then a relationship
initially involving authority has been trans­ for their actions must be po^ntiallyffiÉifiecr
formed into one involving power. For
example, if a policeman demanded entrance
to your house, you would probably comply
on the implicit assumption that his demand
was potentially supportable by reason.
However, should you discover, once he was
in, that his demand was not justifiable, your
further compliance would undoubtedly
derive from his exercise of power, not
authority. The point is that the policeman’s
' badge, uniform and gun—his symbols of
“ formal power”—do not constitute his
authority. Whether he actually has that
dependsupon the authoritativeness of his
communication, and that depends to a con­
siderable degree upon the reasonableness
of his command.
If the officer’s elaboration of his demand
to enter was sound in terms of the law, did , riawTlT
he not have authority? Within the frame
of our example, the answer is both no and
yes. No, as far as you were concerned, since
the elaboration did not make sense in terms
of your own values. Yes, as far as society
and its courts are concerned—provided, of
course, that they themselves considered the
law to be authoritative. As can readily be
seen, in this kind of situation—which occurs
frequently—authority is both a source of
and a restraint upon the exercise of power;
.it both justifies and limits the use of power.
But to those who believe in democracy this
affords small comfort, unless authority
itself is grounded upon reasoning that is
As a final note, it is worth observing that’
just as authority~can T3e transformed into
power, so can ^the re\ 5rse_obtain. ‘‘Brain­
washing” after the manner of George
Orwell’s “ Big Brother” (and his real-life
counterpart in Communist China) is a
gruesome case in point; to obey Big Brother
is not enough; you must love him. A differ­
ent kind of illustration of the same point
is the parent who uses the threat of spanking
(power) to produce filial discipline which is
based on acceptance of certain rules of the
cut both ways. In a humane and heal thy-
of limiting the behavior of men, especially ;
those in official positions, to legitimate ads.;
by “reasoned elaboration” in term s, of
values of a sane society. However, if the
value frame of the society is pathological,
authority, even as we have regarded it, can
become a tool in furthering the state of
pathology.
v
Perhaps the best way to summarize our
effort to draw careful distinctions among
power and related concepts is to apply them
in ã “ reaFworld” context—say, a Southern -
community where white citizens have,
decided to" Cabide by a Federal court’s de- n
segregation õfífer. Às should be evident in^
the accompanying table, we ..assume that
different persons in the community^ had
different reasons for bowing before the
Local officials and local businessmen, for
example, were fearful of severe deprivations
—they responded to an exercise of power.
Those whites we style as “ moderates,” on
the other hand, fall into two distinct
groups: (a) those (Group I) who accepted
as legitimate and reasonable the substantive
logic underlying the Court order, and (b)
those (Group II) who rejected the sub­
stantive ground but accepted the judicial
procedure as legitimate and reasonable. Both
groups, that is, responded to authority, in
the vital senses that both perceived the
Court’s decree rationally and both con­
sidered it (even though on different grounds)
to be capable of “reasoned elaboration.”
A third body of whites—whom, following
David Riesman, we label the “ other-
directed”—complied not because they feared
severe deprivations (power) nor because
they thought the order was reasonable and
legitimate (authority) but because they felt
obliged to follow the lead of those in the
community they most respect (influence).
Stated differently, although the “ other-
directed” group regarded the Court’s ruling
as illegitimate and unreasonable both on
TABLE I. Hypothetical Behavior o f Southern Whites to a Desegregation Court Order
Concept
Power
(relational, demand rationally perceived,
conflict of values, threat of severe sanc­
tions)
A uthority
(relational, demand rationally perceived
and considered reasonable, possible
conflict of values, no severe sanctions)
Influence
(relational, demand rationally perceived,
conflict of values, no severe sanctions)
M anipulation
(non-relational, non-rational, no conflict
of values nor sanctions)
Force
(relationalto non-relational, nonrational,
application of severe sanctions)
Power, A uthority etc.
Decisions and Nondecisions: 107
An A nalytical Fram ework
Peter Bachrach and Morton S. Baratz
substantive and procedural grounds, it
“ went along with its betters.”
Like those who were other-directed, the
“ masses,” too, deferred to the newly
dominant viewpoint in the community. But,
Nunlike the former, the latter did so with
Clittle or no awareness of the issues at stake
or of the fact that they were reversing their
previous stand on the general question.
The “ masses,” in other words, did not make
a'Conscious choice) between compliance and
noncompliance with the Court order;
following the pattern of manipulation, they
simply conformed.
Subject
Groups Which Choose Compliance
State and local officials
(threat of criminal contempt)
Businessmen
(threat of economic boycott and race strife, re­
sulting in loss of profits)
Moderates I
(substantive grounds for Court’s ruling reasonable)
Moderates II
(substantive grounds unreasonable, but judicial
process legitimate and reasonable)
“ Other-Directed” Persons
(judicial ruling, substantively and procedurally
unreasonable, but apprehensive of standing in
community)
Groups Which Choose Neither Compliance N or Non-
compliance
Mass
(conform to dominant behavior in community, with
little or no recognition of the problem nor aware­
ness of complying)
Groups Which Choose Noncompliance
Defiant official subject to contempt of Court
(incarceration reflects that values underlying de­
fiance overshadow values gained by compliance)
Extreme segregationists
 108 Hi T h e Concept of Pow er frame of reference. On the other hand, if
' B’s compliance is grounded, not on a fear
Under the heading of groups not comply­ of deprivations but on acceptance of A’s
ing with the Court order are officials who values, the resulting decision is a hybrid
are incarcerated and fined for criminal con­ case, in the im portant sense that A sought
tem pt (force) and segregationist groups that to exercise power but in fact exercised
are beyond the reach of the Court. Suffice authority. Similarly, cases can be identified
it to say that the behavior of these groups— in which A has sought to exert authority
geared as they are to a different set of values while B’s compliance was given because he
—also can be analyzed and categorized in was influenced (see diagram). The combina­
yterms of power and its related concepts. tions are many—particularly if the analysis
also takes into account situations where two
V I \ i '2 \ \ > ,i or more of the phenomena come into play
V?'■' /a " For our purposes, a decision is “ a set of : simultaneously.39 The point is, in all events,
V"/.- c ^.actions related to and including the choice off tha^a_decision cannot be said to be a result
one alternative rather than another . . . ,” 36," of power or influence or authority or force
or, more simply, “ a choice among alternative-; unless and until it is specified from whose
N" »;V, modes of action. . . .” 37 Thus, we di ffer' point of view the decision is being examined,
N! sharply from Lasswell and Kaplan, to i.e., from that of the one who seeks com-
“ ffN, -■■■' whom a decision is “ a policy involving^ ' pliance or the one who gives it.
A" f;: severe sanctions (deprivations).” 38 The basis It may be objected that this approach is
(- _for the contrast between our definition and unworkable for empirical analysis because
.. theirs is clearcut: they hold that ^decisions it necessitates mind-reading. We think not.
-r , As/are brought about solely by the exercise_of The courts of law do, and so can we, dis­
^c^i>Qwer?while we believe that power is neither tinguish between “ specific” intent and in­
. the only nor even the major factor under- tent inferred from actual behavior. We be­
\-j: lying the process of decision-making and lieve, in other words, that it is both feasible
'A
SÒ reactions thereto. We believe, in fact, that and necessary to deduce from detailed
in some situations power is not involved at observation of the situation why persons
A'
all, that in such situations the behavior of act as they do.40 To put it still another way,
decision-makers and their subjects alike there is no shortcut, no simple and mechani­
a ;-..
:can be explained partially or entirely in cal method, for gaining a full understanding
v J terms of force, influence, or authority. of the decision-making process.
A A Our position can be clarified by reference , We concede that our approach is less
A OXA to the following diagram. Tw<^4m.jportant workable than that of Lasswell and Kaplan,
points may be drawn from it. First, every i Dahl, and others of that “ school.” On the
; social, decision involves interaction be- other hand, because ours provides a broad­
r; . tween "the 'one or more persons seeking a er conceptual frame within which to analyze
given goal and the one or more persons decision-making, it makes easier the com­
whose compliance must be obtained. Thus, parative study of the factors underlying
Y if A’s attem pt to exercise power or influence
Means By Which Reason Why
or whatever over B is ignored, there is no
Compliance Is Sought Compliance Is Forthcoming
decision^
Seconds compliance can be sought through
1. Power 1. Power
the exercise of one or any combination of
the four phenomena indicatLT onT he dia­
gram . However, if compliance is forth­ 2. Influence 2. Influence
coming, it may or may not stem from the DECISION
same source. For instance, if B bows to A ’s
3. Authority' 3. Authority
wishes because A has threatened sanctions
which B wishes to avoid, the resulting de­
cision is one of “pure” power; both parti­ 4. Force 4. Force
cipants made their choices in the same Figure 1. Diagram o f impulse and response.
 different decisions on diverse circumstances.
A road is thereby opened toward the devel­
opment of a body of general theory with
respect to the decision-making process.
Moreover, because we distinguish carefully
among the forces at work in any given
situation, we minimize the risk of putting
unwarranted emphasis upon one factor to
the exclusion, wholly or partly, of others.
Stated more bluntly, we put the phenomen­
on of power in proper perspective: we
recognize that while decision-making fre-
quently does involve power relationships,
. it very often does n ot.
VII <
The other side of the coin is ^«decision-
making. When the dominant values, the
accepted rules of the game, the existing
power relations among groups, and the
instruments o_f force, singly or in combina­
tion, effectively yprevent certain-grievances
from developing into full-fledged issues]
which call for decisions, it can be said th at
a nondecision-making situation exists. This
phenomenon is clearly distinguishable from
the negative aspects of decision-making
(deciding not to act or deciding not to
decide), since the mere existence of the
“ mobilization of bias,” to use Schatt-
schneider’s phrase, is sufficient to prevent
a latent issue from becoming a question for
decision.
It might be objected that since a non-
decision, by definition, is a nonevent, it is
not observable, and, therefore, is not an
operationally-useful concept. Although it is
true that a nondecision is not visible to the
naked eye, a latent issue is discernible and
so is the mobilization of bias. Thus it^qan
be said that the nondecision-making proems
(the impact of the mobilization of bias up­
on a latent issue), in distinction to a non­
decision, is indeed subject to observation
and analysis.
In their perceptive study, Small Town in
Mass Society, Vidich and Bensman, without
Decisions and Nondecisions: 109
An A nalytical Fram ework
Peter Bachrach and Morton $. Baratz
calling it such, analyze the nondecision-:
making process in Springdale.41 For ex­
ample, they relate that the school admini­
strators in the community had basic
grievances but, cognizant of the dominant
rural values prevailing in the community,
the established tradition of deciding all
town issues by unanimous vote, and the
predominance of nonprofessionals in posts
of leadership, the schoolmen prudently kept
their grievances to themselves. In choosing
this course of action, the school officials
admittedly made a decision. But it was not
one brought about by any decision or com­
bination of decisions by others with respect
to their grievances. Quite the contrary, it
reflected the schoolmen’s realization that,
by sustaining the mobilization of bias, the'
leaders of the community—even i f J n\ '
directly and unconsciously—could, would,
and often did exercise authority, power and
influence against them.
In those instances when a latent issue of
the type which is usually kept submerged is
successfully pushed forward and emerges as
a public issue (for example, the recent
emergence of Negro demands in the South),
it is likely that the mobilization of bias will
be directly and consciously employed
against those who demand a redress of
grievances by the decision-making organ.
In such instances, the decision-making proc­
ess preempts the field previously occupied
by the nondecision-making process. And in
so doing, it necessarily jeopardizes....the
previously-established mobilization of
bias.
If the concept of mondei^qmvmakin^
proves a useful tool of analysis, it appears to
us at this juncture that it can be effectively
studied in terms of the categories suggested
in this paper for the examination of de­
cision-making.

11
Some Ambiguities in
the Notion of Power*
William H. Riker
This article by W illiam Riker, Professor o f P olitical
Science at the University o f Rochester, provides a
useful abstract summary o f a number o f different
attempts to explicate the concept o f power, and calls
attention to the dependence o f such efforts on different
conceptions o f causality. Originally published in 58
A m erican P olitical Science Review (1964) 341-349,
it is reprinted here with the permission o f that jo u rn a l
and the author.
T of power is often said to be cen­
h e n o t io n
tral to the analysis of politics. But while that
analysis is a very ancient activity, the con­
ceptual clarification of the notion of power
has been undertaken only in the past genera­
tion. The reason for this discrepancy I leave
to the historians of political ideas. In this
introduction I merely observe that the
clarification has not proceeded as far as is
needed, so that we are still not at all sure of
what we are talking about when we use the
term. Nevertheless there is light ahead,
owing especially to some formal definitions
that have been offered in recent years by
Shapley and Shubik, March, Dahl, Cart­
wright, and Karlsson. By reason of the
formality of these definitions the issues of
meaning have been more sharply delineated
than was previously possible. Hence we
have reached the point, I believe, where we
may confront definitions with each other
and specify precisely how they differ. In so
doing we may be able to resolve some of the
ambiguities remaining in the concept
of power. In that hope this essay is
written.
* I thank P rofessors R obert Dahl, W illiam Flanagan,
C arl H em pel, and Dennis Sullivan fo r criticism s help­
fu l in improving the argument o f this p aper. An
earlier version was delivered a t the Annual M eeting
o f the Am erican P olitical Science A ssociation, N ew
Y ork C ity , S eptem ber 1963.
110
Notes to chapter 11 are found on pages 372 to 373.
But first a personal rem ark: most con­
temporary criticism of political theory is
directed, unfortunately, at the so-called
giants of the past. In such an enterprise, it
is not personally embarrassing—indeed it is
academically fashionable and intellectually
trivial—to explain where Plato went wrong
or what Rousseau meant. W hat political
theory needs, however, is criticism of con­
temporary theory, for this is the theory that
is important in guiding political research.
But such criticism may be personally em­
barrassing, especially when, as in this in­
stance, it is directed at the work of men
whom I regard as at the very forefront of
the social sciences. I want to make it clear,
therefore, that (a) I regard the theories I dis­
cuss as a great advance, one which I have
in the past struggled to make and failed and
(b) I utter criticism not captiously but in the
spirit of contributing to the dialectic of
understanding.
I. FIVE FORMAL DEFINITIONS OF POW ER”
I start with a simple statement of the basic
elements of each of the five definitions, ig­
noring most of the subtleties of each writer’s
interpretations, and usually using the sym­
bols preferred by the authors. I have also
offered verbal translations of the formal
definitions, translations which exhibit, I
suppose, all the characteristic pitfalls of
translations generally.
Shapley, a mathematician who developed
his notions originally to discuss the value of
«-person games, was aided in applying it to
social world by an economist, Shubik.1
Their definition relates only to the power
resulting from the right to vote in a system
where voting, and only voting, determines
outcomes:
m ( i)
Pi = -------
n\
where P is the power to determine outcomes
in a voting body for a participant, /, in a set
of participants: (1,2, . . ., n} where m{i) is
the number of times i is in the pivotal posi­
tion and where pivotal position is defined
thus: when the rules define q votes as
winning,
n + 1 n
— < q < n or — h i < q < n
2 2
370 Notes fo r pages 98 to 102
Notables is to oppose tax increases; this leads
them to oppose expenditures for anything more
than minimal traditional city services. In this
effort their two most effective weapons ordin­
arily are the mayor and the Board of Finance.
The policies of the Notables are most easily
achieved under a strong mayor if his policies
coincide with theirs or under a weak mayor if
they have the support of the Board of Finance
. . . . New Haven mayors have continued to find
it expedient to create confidence in their
financial policies among businessmen by ap­
pointing them to the Board.” (pp. 81-2).
25. Dahl does discuss in general terms (pp.
79-84) changes in the level of tax rates and
assessments in past years, but not actual de­
cisions of the Board of Finance or their effects
on the public school system.
26. Ibid., p. 124.
27. Ibid., “ A rough test of a person’s overt
or covert influence,” Dahl states in the first
section of the book, “ is the frequency with
which he successfully initiates an im portant
policy over the opposition of others, or vetoes
policies initiated by others, or initiates a policy
where no opposition appears.” {Ibid., p. 66).
28. Ibid., p. 131.
29. Dahl is, of course, aware of the “ law of
anticipated reactions.” In the case of the
mayor’s relationship with the CAC, Dahl notes
that Lee was “ particularly skillful in estimating
what the CAC could be expected to support or
reject.” (p. 137). However, Dahl was not inter­
ested in analyzing or appraising to whatjgxtent
) the CAC limited Lee’s_ freedom of action.
Because of his restricted concept of power, Dahl
dfdTnot consider that the CAC might in this
respect have exercised power. That the CAC
did not initiate or veto actual proposals by the
mayor was to Dahl evidence enough that the
CAC was virtually powerless; it might as
plausibly be evidence that the CAC was (in it­
self or in what it represented) so powerful that
Lee ventured nothing it would find worth
quarreling with.
30. The fact that the initiator of decisions
also refrains—because he anticipates adverse
reactions—from initiating other proposals does
not obviously lessen the power of the agent who
limited his initiative powers. Dahl missed this
point: “ It is,” he writes, “ all the more im­
probable, then, that a secret cabal of Notables
dominates the public life of New Haven through
means so clandestine that not one of the fifty
prominent citizens interviewed in the course of
this study—citizens who had participated ex­
tensively in various decisions—hinted at the
existence of such a cabal. . . ” (p. 185).
I In conceiving of elite domination exclusively
tin the form of a conscious cabal exercising the
power of decision-making and vetoing, he over-
Ilooks a more subtle form of domination; one
in which those who actually dominate are not
conscious of it themselves, simply because their
position of dominance has never seriously been
challenged.
31. Sayre and Kaufman, op. cit., p. 640. F or
perceptive study of the “ mobilization of bias”
in a rural American community, see A rthur
Vidich and Joseph Bensman, Small Town in
Mass Society (Princeton, 1958).
10. DECISIONS A N D NONDECISIONS: AN
A N A L Y T IC A L FRAMEWORK
Peter Bachrach and M orton S. Baratz
1. See, e.g., Floyd Hunter, Community Power
Structure (Chapel Hill, 1953); and Robert A.
Dahl, Who Governs? (New Haven, 1961).
2. Peter Bachrach and M orton S. Baratz,
“ Two Faces of Power,” American Political
Science Review, Vol. 56, (December 1962), pp.
947-52. A somewhat similar view, arrived at
independently, may be found in Thomas J.
Anton, “ Power, Pluralism, and Local Politics,”
Administrative Science Quarterly, Vol. 7 (M arch
1963), p. 453.
3. See Bachrach and Baratz, op. cit., pp. 947,
952.
4. Cf. Peter Rossi, “ Community Decision-
M aking,” in Roland Young (ed.), Approaches
to the Study o f Politics (Evanston, 111., 1958),
p. 359.
5. Thomas Hobbes, as paraphrased by C. J.
Friedrich, Constitutional Government and Poli­
tics (New York, 1937), p. 12.
6. Harold D. Lasswell and Abraham K aplan,
Power and Society (New Haven, 1950), p. 75,
draw this implication from the definition of
power, i.e., “ the production of intended effects,”
in Bertrand Russell, Power: A New Social
Analysis (New York, 1938), p. 35.
7. Lasswell and Kaplan, loc. cit.
/v&\ Agreement based upon reason represents
another kind of [interpersonal relationship^l-
authority—which ísVliscussed below. ^
9. It might be argued that the “ victim” did
not actually exercise power in this instance, be­
cause he had no sanctions with which to threaten
the sentry. This objection misses the obvious
point: the “ victim” threatened the guard with
severe deprivations (dishonor, imprisonment)
if the guard did not perform his soldierly duty
by complying with the “ victim’s” command
that he (the “ victim”) be killed.
10. See part IV below.
11. See part II below.
12. Lasswell and Kaplan, op. cit., p. 76. We
have deleted “ actual or” from the parenthetical
expression because actual deprivation for non­
conformity is a property of force, rather than
power. This point is discussed further below.
The Lasswell-Kaplan definition is open to
another criticism. They observe (p. 77) that “ to
have power is to be taken into account in
others’ acts (policies).” Strictly construed, this
must mean that any and every person or group
involved—in whatever degree—in decision­
making must have power. For is not the farmer
who markets .001 percent of the total supply of
wheat “ taken into account” by other buyers
and sellers in just the same sense—though not,
of course, in the same degree—as in the General
Motors Corporation in the determination of
automobile prices? Or, to change the illustra­
tion, is it not the case that, in the literal inter­
pretation of the word, nonvoters as well as
voters “ participate,” and therefore have power,
in deciding close elections ? We should think so.
But if this is what is meant by power, how can
we avoid concluding that no matter where we
look, we shall always find that power is broadly
diffused? To rephrase, if (a) we analyze the dis­
tribution of power solely in terms of decision­
making and (b) we ascribe power to all who
participate in whatever measure or with what­
ever “ weight” (“ The weight of power is the
degree of participation in the making of deci­
sions . . . ” [Ibid.], then (c) do we not necessarily to the United States. Thus the United States,
prejudge that power in real-world situations
will be widely dispersed? For further discussion
of this general question, see Bachrach and
Baratz, op. cit.
13. See part III below.
14. See Richard E. Neustadt, Presidential
Power (New York, 1960), p. 21. Compare
Thomas C. Schelling, The Strategy o f Conflict
(Cambridge, Mass., 1960), pp. 38-9.
15. This error, compounded by that of re­
garding power as something which is possessed,
may well have underlain the policy of the
United States toward Chiang Kai-Shek during
the period (1944-49) of the Chinese civil war. It
is entirely possible, that is to say, that in pro­
viding substantial amounts of armament to the
Kuomintang regime, wc mistook the instru­
ments of power for power itself; and, in addi­
tion, by interpreting the Kuomintang-Com-
munist struggle in terms of our own values, we
utterly misread the temper of the great majority
of the Chinese people.
The abortive invasion of Cuba in April 1961
is perhaps another example of the inherent
dangers in projecting our values onto a popu­
lace holding a different collection of interests.
Looking at the great bpdy^of Cuban nationals
who were apparentIy^bereft\both of individual
freedom and personal-dignity, we concluded
that we need only provide the opportunity, the
spark, which would ignite nationwide uprisings
against the Castro regime. But hindsight has
indicated how badly we misread popular feeling —-p (5^. ‘‘An Analysis of Social Power,” American
in Cuba. See Stewart Alsop, “ Lessons of the
Cuban Disaster,” Saturday Evening Post, 24
June 1961, pp. 26-27.
16. Neusiadt, op. cit., pp. 12-13. On the
general point, see also Schelling, op. cit., p. 6.
17. The point is also well illustrated by
Franco-American policy differences in the early
Notes fo r pages 102 to 103 371
1960s. Committed both to the defense
Western Europe and to strict limitation on the
number of nations with independent nuclear
forces, the United States was caught in a
dilemma in its dealings with General de Gaulle.
In the words of a contemporary observer, “ De
Gaulle . . . has played a judo trick on the
United States . . . [He] means to fashion his
‘European construction,’ based on the force de
frappe and the Franco-German axis and ex­
cluding the British and Americans. And he
means to do this under the umbrella o f the
American nuclear deterrent. .. There is precious
little the Kennedy Administration can do about
de Gaulle’s judo trick—short of removing its
nuclear protection. And this has not even been
seriously considered . . . . ‘We’re a bit like that
little Dutch boy with his finger in the dike,’ says
one Kennedy adviser. Removè the American
commitment to defend Europe, and the result
is unmitigated disaster, not only to Europe but
like the little Dutch boy, is immobilized. The
strongest power in the Western alliance has
amazingly little bargaining power in the
alliance.” Stuart Alsop, “ Should We Pull Out of
Europe?” Saturday Evening Post, 13 April
1963, p. 80. Emphasis in original.
The main point is made more pithily by
“ President H udson” in Allen D rury’s novel, A
Shade o f Difference (New York, 1962), p. 82:
“ The more real power you have, the less you
can afford to exercise it, and the less real power
you have, the more you can throw it around.”
For further discussion of the relationship
between power and commitment, see E.
Abramson et ah, “ Social Power and Commit­
ment Theory,” American Sociological Review,
Ypks23 (February 1958), pp. 15-22.
(■ 18j With Las^well and K aplan, op. cit., p. 16,
we -define a(valueyas “ a desired-evenW a goal
event. That X values Y means that X acts so as
to bring about the consummation of Y .” L * ¥ .
19. Ibid., p. 76.
20. Ibid., p. 77.
21. Op. cit., pp. 17-18. A corollary proposi­
tion could be called the “ rule of /«/^anticipated
reactions.” We refer to a situation in which one
person grudgingly conforms to what he thinks
another wants, but finds after the fact either
that he misread the other’s preferences or that
the latter never intended to invoke sanctions
for behavior contrary to his preferences.
Sociological Review, Vol. 15 (December 1950),
p. 733.
23. A major defect of Lord Russell’s con­
ception of power (see above, note 6) is that it
utterly ignores this distinction. One can prod­
uce an “ intended effect” through the exercise of
either power or force.
 372 Notes fo r pages 103 to 112
24. It is often true, when force is operative,
that A gives B the option to comply with his
demands between blows. But in such circum­
stances, should B bend to A’s wishes, he does
so out of fear of further sanctions, in which case
force is transformed into power.
25. One of the more penetrating critiques
along these lines may be found in General
Maxwell D. Taylor, The Uncertain Trumpet
(New York, 1959).
26. According to Bierstedt, op. cit., p. 731,
“ .. . influence is persuasive while power is coer­
cive. We submit voluntarily to influence while
power requires submission.” In^our view,JfJB
submits voluntarily, .power is opeM iver~butlf
he submitSL„under duress, f o r c e Is' operative.
' It is worth noting that under our definition it
would be incorrect to say that Marx “ in­
fluenced” Lenin, or that Haydn “ influenced”
M ozart, or that Jesus Christ “ influenced” the
Conquistadores. In each of these cases the
second shared the values of the first, i.e., the
relationship involved neither power nor in­
fluence, but authority. See part IV below.
27. See Richard H. Rovere, Senator Joe M c­
Carthy (New York, 1959).
'^ i \2 8 . C. J. Friedrich, “ Authority, Reason and
Discretion,” in C. J. Friedrich (ed.), Authority
(Cambridge, Mass., 1958), p. 37.
29. Lasswell and Kaplan, op. cit., p. 133.
30. Bierstedt, op. cit., p. 733.
31. Authority, pp. 35, 36.
32. As is perhaps obvious, if B chooses to
defy A, the relationship no longer will involve
power. This notion of rationality of choice is
analogous to Thomas Hobbes’s treatment of
the relationship between fear and liberty.
“ Feare, and Liberty,” he wrote, “ are con­
sistent; as when a man throweth his goods into
the Sea for feare the ship should sink, he doth
it neverthelesse very willingly, and may refuse
to doe it if he w ill: It is therefore the action, of
one that was free.” Leviathan, Everyman Edi­
tion, p. 110.
33. Friedrich, Authority, p. 36. Reasoning
also underlies the difference between authority
and influence. Thus, if B complies with A ’s
demand neither because he fears deprivations
nor because his compliance is based upon
reasoning, B has been influenced. This dis­
tinction will be further elaborated below.
34. Ibid., p. 38.
35. Ibid., p. 47.
36. Robert A. Dahl, “ The Analysis of In ­
fluence in Local Communities,” in Charles
Adrian (ed.), Social Science and Community
Action (East Lansing, Mich., 1960), p. 26.
37. Peter Rossi, “ Community Decision-
Making,” in Roland Young (ed.), Approaches
to the Study o f Politics (Evanston, 111., 1958),
p. 364.
38. Op. cit., p. 74.
39. For example, A may employ both author­
ity and power to gain B’s agreement, and B’s
response may have a similarity dual basis. An
apparent case in point is the relationship be­
tween Adolf Hitler and some of his military
chiefs during World War II. On this, consult
William L. Shirer, The Rise and Fall o f the
Third Reich (New York, 1960), pp. 366 if. and
passim.
40. The approach we have in mind is exem­
plified by the untutored, but nevertheless
penetrating, study of “ Springdale” by Joseph
Vidich and Arthur Bensman, Small Town in
Mass Society (Princeton, N .J., 1958). For fur­
ther discussion of this point, see following
section.
41. Ibid.
11. SOME AMBIGUITIES IN THE NOTION
OF POWER
W illiam H. Riker
1. L. S. Shapley and M artin Shubik, “ A
Method for Evaluating the Distribution of
Power in a Committee System,” this R e v i e w ,
Vol. 48 (1954), pp. 787-92; L. S. Shapley,
“ A Value for A-Person Games,” Annals o f
Mathematics Study No. 28 (Princeton, 1953)
pp. 307-17 and “ Simple Games,” Behavioral
Science, Vol. 7 (1962), pp. 59-66.
2. James G. March, “ Measurement Con­
cepts in the Theory of Influence,” Journal o f
Politics, Vol. 19 (1957), pp. 202-226; see also
his “ An Introduction to the Theory and
Measurement of Influence,” this R e v i e w , Vol.
49 (1955), pp. 431-51.
3. Robert A. Dahl, “ The Concept of Power,”
Behavioral Science, Vol. 2 (1957), pp. 201-15, at
pp. 202-03. Note that Harsanyi has modified
D ahl’s definition (and also Shapley’s) by adding
opportunity costs. Since these modifications do
not affect the basic theory, I have not discussed
them here. John C. Harsanyi, “ Measurement of
Social Power, Opportunity Costs, and the
Theory of Two-Person Bargaining Games,” and
“ Measurement of Social Power in A-Person
Reciprocal Power Situations,” Behavioral
Science, Vol. 7 (1962), pp. 67-80, 81-92.
4. Dorwin Cartwright, “ A Field Theoretical
Conception of Power,” pp. 183-220, in Dorwin
Cartwright, ed., Studies in Social Power (Ann
Arbor, 1959).
5. Georg Karlsson, “ Some Aspects of Power
in Small Groups,” in Joan H. Criswell, Herbert
Solomon, and Patrick Suppes (eds.), Mathemat­
ical Methods in Small Group Processes (Stan­
ford, 1962), pp. 193-202.
6. If Cartwright’s force to comply, / ab, and
force to resist, . f e could be translated into con­
ditional probabilities of compliance and re­
sistance, then Cartwright’s definition would be
exactly the same as D ahl’s formula, pi — p 2.