System and Process: Polar Concepts for Political Research

• Published date: 01 September 1961
• Date: 01 September 1961
• DOI: http://doi.org/10.1177/106591296101400311
• Subject Matter: Articles

SYSTEM AND PROCESS: POLAR CONCEPTS FOR
POLITICAL RESEARCH
RICHARD B. WILSON
University of Colorado
F
THERE IS A CORE to the intellectual process perhaps it may best be
t characterized as a search for order amidst diversity and chaos. The human
mind appears to demand some systematic ordering of the bewildering array
of objects and events which it confronts, to the end that reality may be compre-
hended in its essential aspects and that knowledge may effectively be utilized
in pursuance of social ends.
That sub-area of knowledge which we presently call political science has
clearly not been exempted from this demand. Plato’s search for the &dquo;good state,&dquo;
Aristotle’s classification of constitutional practices, Machiavelli’s recipe for power,
the Hobbsian Leviathan, the Utilitarian Calculus, even the Marxian dialectic -
differentiated though they may be in doctrinal and substantive content - all rep-
resent efforts to organize political data within a unified and integrated system of
interpretive concepts.
There seems to be a general belief, however, that until recent years Amer-
ican political analysis has been singularly unconcerned with comprehensive
systematization.’ Work in political theory is excoriated either for its excessive
&dquo;historicism&dquo; or for over-preoccupation with values; empirically oriented research
is criticized as disconnected, &dquo;hyper-factual,&dquo; pragmatic, and problem-oriented: in
neither area, runs the complaint, has there been sufficient attention to the de-
velopment of a unifying, descriptive theory.2
2
Response to these criticisms has produced in the past decade several tenta-
tive suggestions and a few positive efforts to construct such a theory. It will be
the purpose of this paper to explore in an admittedly incomplete manner some
elements common to these efforts, to assess certain of their consequences and im-
plications, and to attempt an evaluation of the results as they have emerged thus
far. It may prove helpful at the outset to summarize the central thesis which
will be developed and defended, and to indicate certain areas of orthodox con-
troversy which will be avoided.
Current conceptualizations of a unified, theoretical &dquo;system&dquo; of political
knowledge are mainly an outgrowth of our present focus on individual and social
behavior as the primary data of political analysis;3 thus, the dominant motif of
these systems is a logically integrated structure of empirically verifiable generali-
zations purporting to describe uniformities or patterns of political behavior and
to facilitate prediction of future behavior.4 From this structure further verifiable
1
See David Easton, The Political System (New York: Knopf, 1953), p. 4; Avery Leiserson,
"Problems of Methodology in Political Research," Political Science Quarterly, LXVII (Decem-
ber 1953), 558-84.
2
G. E. G. Catlin, A Study of the Principles of Politics (New York: Macmillan, 1930).
3
Heinz Eulau, Samuel Eldersveld, and Morris Janowitz (eds.), Political Behavior (Glencoe: Free
Press, 1956), p. 4.
4
Leiserson, op. cit., p. 583.
748

---

749
hypotheses may be deductively derived. The resulting conceptual model is
represented as a sufficient and exclusive framework for the guidance of research.
It will be argued that when regarded as the paramount orientation for political
analysis these systems are untenable and largely sterile; that the basic concepts
which compose them often lack meaning, significance, and theoretical justifica-
tion ; that fruitful generalizations in our field accrue from experience with unique
ideas, institutions, and situations rather than from correlations of those few as-
pects of behavior which are reducible to quantifiable units or to other easy meas-
ures of uniformity; and, finally, that political scientists can best serve the practical
requirements of society as well as the intellectual and theoretical demands of the
discipline by utilizing the concepts, values, and descriptive theory which are im-
plicit in an already existing &dquo;system&dquo; of political knowledge -
the &dquo;constitu-
tional&dquo; system in its broadest sense.
Development of this thesis would not seem to require excursions along cer-
tain well-traveled paths of dispute. First, there is neither need nor intention to
question here the foundations of the behavioral movement in American political
thought. While emerging from and intimately linked with that movement, &dquo;sys-
tem-building&dquo; is not a logically necessary correlate of it; nor do the majority of
behaviorally oriented researchers accept the feasibility or possibility of a compre-
hensive theoretical system in the near future .5 Second, there is neither need nor
intention to challenge unified descriptive theory on grounds that it is not value
theory. The vast majority of American political scientists, regardless of analytical
orientation, admit the vital and legitimate role of values in political thought; such
systematically inclined theorists as Easton, Snyder, and Lasswellg most certainly
do. The real dispute centers on a proper relationship between fact and value,
rather than on claims that either should be ignored. Finally, there would seem
to be no need to re-examine the claim that political analysis cannot be &dquo;scientific&dquo;
because the relevant variables are almost infinite in number, the observer is
&dquo;biased&dquo; by involvement with the data, and all conceptual frameworks are by
definition &dquo;culture-bound.&dquo; The term &dquo;science&dquo; is a sufficiently ambiguous rubric
to allow each of us, regardless of method, to claim its honorific protection. More-
over, and again regardless of method, we are all data-involved and culture-
bound ; only the test of history is likely to determine which of the various analyti-
cal orientations are most adversely affected by these strictures.
If the initial task of critical analysis is definitional clarity, then we must begin
by asking what is meant by a comprehensive systematization of political data?
What common concepts and notions are to be found in the thinking of such
theorists as David Easton, Richard Snyder, David Truman, G. E. G. Catlin, Har-
old Lasswell, Earl Latham, Morton Kaplan, etc.?
As we have indicated, the behavioral movement as a whole might be viewed
as an initial step toward a more systematic ordering of political data. But, against
5
Oliver Garceau, "Research in the Political Process," American Political Science Review, XLV
(March 1951), p. 85.
6
Easton, op. cit., chap. ix; Richard Snyder, "A Decision-Making Approach to the Study of Politi-
cal Phenomena" in R. Young (ed.), Approaches to the Study of Politics (Evanston: North-
western University Press, 1958), pp. 3-38; Harold Lasswell and Abraham Kaplan, Power
and Society (New Haven: Yale University Press, 1950), pp. xi-xiv.

---

750
the many characteristics which are admittedly common to the work of most be-
haviorists must be balanced the incommensurate concepts and categories and the
varied emphases implicit in such approaches as decision-making, small and large
group analysis, communications theory, voting behavior theory, and leadership
and personality studies -
to name but a few. Concepts which are crucial for the
development of one approach do not fit easily into the framework of the others.
Indeed, some behaviorists view this lack of conceptual unity as inevitable. As
Oliver Garceau puts the matter: &dquo;A program of research in the political process
must remain eclectic. No
single strategy or model recommends itself from a sur-
vey of the tools available or the hypotheses adumbrated.&dquo; 7 Of course, to the
extent that any one of these models is represented as a comprehensive guide for
political analysis (as appears to be the case with Truman’s or Homan’s group
theory or Snyder’s decision-making analysis) we may legitimately designate it as
an attempt at system-building.
What characteristics, then, are common to the work of those thinkers who
stress unity rather than diversity? First, and perhaps of transcendant importance,
is a rejection of the traditional ordering concepts of political science - sover-
eignty, equality, liberty, separation of powers, rights, duties, obligations, etc.
Such notions are alleged to be ambiguous and value-laden. &dquo;If, for empirical
research,&dquo; remarks Easton, &dquo;we define a good concept as one that refers to an
identifiable set of facts and that can be explained in terms of the operations
needed to discover these facts, then a good part of the terminology used in
political science falls far short of this standard.&dquo; 8 Second, dissatisfaction with
old ordering concepts prompts the search for new ones which are free of these
alleged defects and which, above all, are universally relevant to any area and type
of political behavior. Such concepts may be more accurately characterized as
&dquo;significant variables,&dquo; cast in the form of empirical generalizations and logically
related to yet higher order concepts which summarize or synthesize the patterns
of relationship discovered to exist among those variables. Thus, such notions as
power, class, status, role, group, role-oriented behavior, structured and unstruc-
tured situations, interaction pattern, feedback, etc., become, with &dquo;precise&dquo; de-
finition, the basic building blocks of a system. Third, the system-building enter-
prise culminates in the postulation of highest order concepts such as &dquo;equili-
brium,&dquo; the &dquo;authoritative...