The Domain of Politics

• Author: Gerald Garvey
• DOI: 10.1177/106591297002300107
• Published date: 01 March 1970
• Date: 01 March 1970
• Subject Matter: Articles

THE DOMAIN OF POLITICS
GERALD GARVEY
Princeton University
HIS
PAPER has three aims. First, it will briefly consider some implications
of recent trends in economic theory for political science. Second, and based
thereon, it will develop an approach to the definition, analysis and formal
measurement of &dquo;power.&dquo; Though using economic concepts, the proposed definition
will be consistent with that advanced in leading works of political theory associated
with the names of Easton, Lasswell and Dahl. It will, moreover, treat power as a
more complex phenomenon than are those represented by most variables in eco-
nomic theory (e.g., &dquo;utility&dquo;), and hence as requiring a somewhat more elaborate
theoretical conceptualization. Third, the paper will suggest some significant propo-
sitions which logically follow from the proposed definition of power. In particular,
the proposed definition’s usefulness in demarcating the &dquo;domain of politics&dquo; from
other fields of study will be considered.
I. ECONOMIC THEORY AND POLITICAL SCIENCE
A
long tradition holds &dquo;utility&dquo; to be the central concept in economics,,- and
an even longer tradition -
tracing at least to Aristotle -
settles on &dquo;power&dquo; as a
foundational concept in the systematic study of politics.2 But no boundary neatly
separates the province of utility from that of power; nor does any recognized line of
demarcation separate the intellectual jurisdiction of the economist from that of the
political scientist. As a result, an amazingly high percentage of the most influential
books in recent political theory have been by economists. Arrow’s Social Choice
and Individual Values ( 1951 ) , Downs’ Economic Theory of Democracy ( 1957),
Black’s Theory of Committees and Elections ( 1958 ) , Schelling’s Strategy of Con-
flict (1960), Buchanan and Tullock’s Calculus of Consent (1962), Boulding’s Con-
flict and Defense (1962), Olson’s The Logic of Collective Action ( 1965), and
Downs’ Inside Bureaucracy ( 967 , are a few titles in this category. Dahl’s Preface
to Democratic Theory ( 1956) and Riker’s Theory of Political Coalitions ( 1962),
though written by political scientists, draw heavily on economists’ concepts and
approaches. And the current prominence of economists in practical politics and in
important policy-making positions throughout the English-speaking world is not
in question.
For reasons to be considered below, the political scientist’s traditional notion
of power seems likely to be a casualty of the ascendancy of economic theorizing.
James March, in an essay selected as the best paper presented at the 1964 American
Political Science Association National Convention, has already observed that
&dquo;power&dquo; is increasingly regarded as an unsatisfactory explanatory concept by politi-
1
Generally, see E. Halevy, The Growth of Philosophic Radicalism (English translation, M.
Morris, 1952).
2
See Politics III, 1-2; Aristotle’s discussion of the varieties of rule is in Politics III, 7 et seq.
The "power politics" school achieved modern prominence in the so-called "Chicago
School" of Merriam in the 1930’s. See, e.g., Robert Dahl, "The Behavioral Approach
in Political Science,
"American Political Science Review, 55 (December 1961), 763-64.
120

---

121
cal scientists who are critical in temper and who demand predictive power in their
theoretical tools. March noted that &dquo;power is a disappointing concept. It gives us
surprisingly little purchase in reasonable models of complex systems of social
choice.&dquo;
3
Wherein lies the solution in the simultaneous quest among political scientists
for increased rigor and the trend toward reduced confidence in this, one of the
discipline’s traditionally basic tools and concepts? In 1950, Duncan Black (whose
pioneering work is mentioned above) suggested one answer to this question: &dquo;A
’satisfactory’ Political Science will need a set of formal or mathematical proposi-
tions&dquo; -
the type of propositions which economics, alone among today’s social
sciences, can lend to the study of politics. Black continued that the &dquo;adequacy of
a Political Science will depend, among other things, on how far economic language
can first
...
capture and then make explicit, thoughts which relate to a field other
than Economics, viz., to Politics.&dquo; It
is no part of this paper’s purpose to repudi-
ate the use of economics’ more rigorous tools in the study of political phenomena.
But this does not mean that economics should &dquo;capture&dquo; political science in quite
the wholesale manner that Black seems to advocate.
The peculiar interest of political scientists will remain somewhat different from
that of economists, even if political science as a body of systematic knowledge yields
to the methodological imperialism of economics (and no such capitulation is yet
on the horizon!). Therefore, it will be necessary to insure that any analytical tools
borrowed from economics are appropriate to the treatment of political problems.
The danger is that &dquo;economic language,&dquo; unless invoked with an understanding of
the particular intellectual interests which gave rise to that language, will lose just
those nuances and insights that are of prime interest.
It would be helpful, therefore, briefly to review the quality of economic theory
-
or, more particularly, of utility theory -
which carries the greatest potential of
missing these complications and nuances: its elegance. From Adam Smith right
down to the present day, economic theory has manifested a rigorous fealty to the
principle of Occam’s Razor. Unnecessary assumptions, unneeded metaphysical ele-
ments, and cumbersome psychological underpinnings have been eliminated. As a
result, market-behavior theory today is founded on a handful of formally elegant,
but empirically contentless, tautologies.5 The principal ideas in this elegant body of
theory are as follows:
The act of seeking any product is (tautologically) defined as pursuit of satis-
faction. Any product with the ability to satisfy is a Good - even if this Good is
&dquo;bad,&dquo; like dope or poison. If the Good is scarce, then it is a &dquo;good&dquo; in the strict
economic sense. Scarcity means that certain goods must be sacrificed in order to
gain others: the consumer must trade potatoes for bread; the nation must trade
guns for butter; and so forth. From this fundamental interest in situations where
3
James March, "The Power of Power," in David Easton (ed.), Varieties of Political Theory
( Englewood Cliffs : Prentice-Hall, 1966).
4
Duncan Black, "The Unity of Political and Economic Science," in Martin Shulick (ed.),
Readings in Game Theory and Political Behavior (Garden City: Doubleday, 1954),
pp. 62-64.
5
See in this connection Charles E. Lindblom, "In Praise of Political Science," World Politics,
9 (January 1957), 241-42.

---

122
acquisition of goods is constrained by scarcity follows economics’ fundamental prob-
lem -
viz, to minimize the sacrifice involved in giving up some goods in order to
have others; or obversely, to maximize the satisfaction of getting whatever goods
can be acquired under existing limited resource constraints.
This problem ostensibly requires some ability to measure the sacrifice or the
satisfaction associated with alternative choices of goods. Down through Alfred
Marshall, the principle of measurement employed by virtually all economists was
cardinality, which assumed that quantities of utility can be measured (at least in
principles), just as physical entities like length and time can be measured. The idea
that an,individual’s &dquo;utility bundles&dquo; could be compared in terms of a common unit
of measure, called a utile, might have prevailed much longer but for the entry of
that Swiss polymath, Vilfredo Pareto.
Pareto, more or less in spite of himself, showed what was wrong with cardinal
utility. The cardinal utility principle made certain assumptions about a consumer’s
preferences which were unnecessary to account for observable market behavior.
Because they were unneccessary -
for purposes of economic analysis -
they were
therefore to be done away with.
Pareto reasoned that as a consumer gave up a given quantity of good Y, it
would be possible to compensate him with more of good X so as to leave him with
a combination of Y
and X such that he would take equal satisfaction from, and
hence be indifferent between, the second bundle and the bundle which he had
before the intial deduction of Y. If repeated over and over again, this procedure
would map out a series of indifference curves, each of them connecting a con-
sumer’s points of equal preference for different combinations of Y and X given by
different points on the same curve, as in Figure 1. This technique could be made
to explain behavior just as well as theories which presumed cardinal utility. But
the indifference curve technique required no assumptions about the measurability
of utility. Indeed, if numbers were used at all, they need be only ordinal numbers
(e.g., &dquo;first,&dquo; &dquo;second,&dquo; &dquo;third&dquo;) rather than cardinal numbers like &dquo;one utile,&dquo;
&dquo;two utiles,&dquo; etc. For quantitative utility, a wholly qualitative theory could be
substituted.6
Given an individuals indifference map, based strictly on his relative preferences
for different bundles of goods Y and X, the problem of determining how much Y
and how much X he should purchase, subject to his income constraints, can be
solved as shown in...