Systematic Study of Comparative Politics
The study of comparative politics is in the midst of fundamental change.
In the past quarter century, no part of the discipline of political science
has witnessed a more thoroughgoing self-examination and reorienta-tion than
has the study of comparative politics. Indeed, its scope and direction have
shifted so dramatically that interested students have increasingly felt
trapped in a seemingly impenetrable maze of shifting theoretical emphases,
unfamiliar methodology, and abstruse terminology. The deprecation and dismantling
of traditional approaches to the study of politics have been accompanied
by sharp controversy and extended debate. This, in turn, has fostered as
much blind dogmatism as it has promoted constructive empirical and theoretical
The transitional nature of comparative political analysis is illustrated
by the adamant refusal of many to adopt new perspectives and to move in
newly charted directions, while others enthusiastically and uncritically
embrace new approaches and methods at first sight. Intense controversy continues
to swirl about the following kinds of issues: the definition of politics
and the scope of its study; the nature and roie of theory; the normative
and empirical dichotomy; the value and role of measurement and quantification;
the techniques of research in cross-cultural settings; the difference between
configurative and comparative study; the relation of interdisciplinary concerns
to the study of politics;
the need for policy oriented political analysis; and the persisting debates
for and against the various theoretical approaches to the study of poli-
tics. This list, which contains the perennial issues as well as queries
more recently developed, reHects the concern for the very core of political
analysis. Confrontation concerning such central issues in fact documents
the dynamism that infuses the contemporary study of comparative politics.
The Traditional Study of Comparative Politics
The patterns of the traditional study of comparative politics in American
political science were established most firmly during the late nineteenth
and early twentieth centuries. During^hese years, .political science developed
an identity that resulted in its recognition as a legitimate and fruitful
focus of inquiry- At the same time, the important' relationship that "political
analysis maintained with the study of history,) philosophy, and law was
appreciated and preserved. Traditional oei3m-parative political studies
reflected a significant concern for both historical perspective and the
norms of political behavior. A further contribution of the traditional
approach was its role in educating large numbers of American citizens
about their own political institutions. Finally, traditional comparative
government produced a host of studies that carefully and minutely described
political structures and institutions. These studies remain a valuable
source of data for comparative political analysis.
Despite contributions such as these, other characteristics infused the
traditional study of comparative politics, severely impeding the development
of systematic comparative inquiry and comparative political explanation.
This may be described in terms of six major characteristics that marked
the traditional approach: configurative description, formal-legalism,
parochialism, conservatism, nontheoretical emphases, and methodological
insensitivity.1 These characteristics have strongly marked the study of
politics in the past and they continue to persist today in varying form
politics emphasized detaileddescription of particular political systems
or particular aspects of these systems. Such procedure was, in fact, not
comparative politics at all, but was more accurately configurative government.
The scholar drew configurations of governmental institutions, describing
them in varying degrees of intricacy. Characteristic studies included
discussions and checklists of important names, dates, relevant historical
events, sections of legal and constitutional documents, and committee
and cabinet organization.
Most ordinarily, one scholar would present configurative descriptions
of four or five European governments within one volume, each described
according to its peculiar genius. Following World War II, a trend was
established in which different scholars would describe different governments,
and these would then be bound together in one volume and presented as
a textbook in comparative government. Occasionally, introductions, essays,
or monographs would appear, attempting to compare selected aspects of
these configurations. Thus. in the introduction of his widely read text.
Henry Russell Spencer wrote about the "far-reaching contrasts between
Westminster and Paris as to the relation of voters and ministers to elective
parliament; between the constitutional kingship that is an honorific badge
worn by Victor Emmanuel III of Italy, and that which is a tool deftly
wielded by Gustav of Sweden;
between Supreme Courts in countries which do and countries which do not
possess a strong tradition of constitutional law and custom, which continues
and grows and binds while parliamentary measures come and go."3 At
the most, therefore, the traditional study of comparative politics involved
comparative description. Most studies, however, concentrated upon configurative
descriptions of governments, one by one. Very little is comparative about
this kind of exercise-4
Formal- Legalism For years, the study of comparative politics reflected
a preoccupation with formal structures and legal strictures. Emphasis
was placed upon the organized and evident institutions of government,
and studies concentrated almost exclusively upon constitutions. cabinets,
parliaments!, courts, and bureaucracies.
The study of political science as a separate discipline-from which comparative
politics developed as a sub-field-evolved out of the study of public law
and institutional history.5 The title of one mid-nineteenth century government
textbook reveals the strength of the formal-legal scholarly tradition:
The Citizen 's Manual of Government and Law: Comprising fhe Elementary
Principles of Civil Government; A Practical View of the State Governments,
and of the Government of the United States; A Digest of Common and Statutory
Law, and of the Law of Nations; and A Summary of Parliamentary Rules for
the Practice of Deliberative Assemblies.^ Courses were frequently in the
format of comparative constitutions of the world. In the mid-1920s, the
political science department at the University of Michigan offered a course
entitled "Charter and Ordinance Drafting." During this same
period, Columbia University taught courses under the headings "Statutes"
and "Problems of the Law of Taxation."
One of the earliest comprehensive American attempts to prepare a text
in comparative politics was Woodrow Wilson's The Slate, first published
in 1889.7 This 686 page study contained a 36 column index that devoted
26 lines to "constitution," 25 lines to "Senate,"
23 lines to "law," 19 lines to "legislation," and
12 lines to "township." There was no reference to such concepts
as "power," "interest," and "group." One
line was devoted to "authority" and "competition,"
and two lines each to "revolution" and "slavery."
A 1927 Round Table Conference of the American Political Science Association
that met to discuss the field of comparative government strongly endorsed
the formal-legal perspective. As one participant pointed out, comparative
politics should be particularly concerned with "the problems of federalism,
judicial procedure, parliamentary practice, administration, and local
government." He concluded by stating that "the comparative constitutional
law of Anglo-American countries is a most satisfactory field because our
law libraries offer unusual facilities."8 This deep-seated formal-legal
tradition which stresses documents rather than political activities and
the formal institution rather
'As late as 1914. only ,38 of some 300 American colleges and universities
had distinct departments of political science. See Anna Haddow, political
Science in American Colleges and Universities. 1636-1900 (New York: D.
Appleton-Century, 1939), p. 263.
"Andrew W. Young, The Cifiien s Manual of Government and Law, rev.
ed. (New York:
H. Dayton, 1858).
Woodrow Wilson, The State: Elements of Historical and Practical Potilics
(Boston: D.C. Heath. 1889).
""Reports of Round Table Conferences: Comparative Government,"
Thf American Political Science Review 21 (May 1927):393.
than the informal processes of competition is still evident in many contemporary
Parochialism Comparative political analysis in the past focused quite
exclusively upon European government. But even within the study of comparative
European politics, the scope was narrowed to consider again and again
the same four societies: Great Britain, France, Germany, and Russia. The
tenacity of this pattern is illustrated by the fact that eight different
comparative texts devoted exclusively to this popular foursome consider
these societies one by one in the identical country by country order.9
Various reasons have been proposed for this constricted emphasis. John
Burgess wrote at the turn of the century that Great Britain, the United
States, Germany, and France were "the most important states of the
world."10 One set of authors, who also included Italy in their study,
attempted to explain their five country choices as follows:
"The omission of materials relating to the other governments of Europe
was due entirely to the lack of space and a desire to concentrate on the
countries commonly included in college and university courses.""
In a much more recent volume, the authors stress the "big four"
which they "deem most significant to Americans both as factors in
international politics and as political laboratories."12 Other reasons
for this scholarly ethnocentricity derive from the knowledge gap caused
by the inaccessibility of data and the inconvenience of research in much
of the world. Since most non-Western societies apparently lacked formal
governmental structures, many felt that little of substance was there
to study anyway. Thus, one pioneering political analyst could write in
1900 that Asian nations "have created no real states."13
Concentration upon the more "advanced," "democratic,"'
and "civilized" political systems carried a significant normative
impact as well. If comparative methods were at all introduced, American
and British political structures were elevated as the models. Other political
systems tended to be presented as deviations from these standard cases
and were ^ thus often labelled as exotic or alien. Most discussion of
African, Asian, and Middle Eastern governments was left to archaeologists/
orientalists, missionaries, diplomats, and itinerant adventurers- The
impact of this - neglect by social scientists is still seen in the absence
both of incisive studies of numerous non-Western societies as well as
of cross-societal studies that are able to draw deeply upon the Afro-Asian
experience for purposes of comparative analysis.
Conservatism In the past, the study of comparative government tended to
stress the permanent and the unchanging. Political institutions were examined
in terms of an evolutionary development which found fulfillment in the
immediate present- But while these institutions had a past, they apparently
had no future. The study of British politics, for example, was reduced
to a description of the evolution of institutions beginning with the Magna
Carta. This historical process traced the unfolding of the unique and
enduring in British political experience. Hil! and Stoke wrote in the
preface of their volume that they placed emphasis "on the more basic
and permanent phases of the European political systems-"1"1
Scholars consciously stressed the changeless quali-1 ties of political
institutions and such phrases as "time-honored," "immutable,"
and "eternal" principles of government were sprinkled throughout
the literature. The various texts reflected a deep feeling for tradition,
precedent, order, stability, and conserving evolution.
This insensitivity to political modernization and transforming change
was in large part due to the crucial period that political science spent
in incubation with precedent-sensitive law and past-oriented history.
In a world where the processes of political development were perceived
as having "remained throughout clear and almost free from considerable
irregularities" and in which "the lines of advance are seen
to be singu-larly straight,"15 it is not surprising that there was
little concern for or study of political dynamism.
For many years. Western scholars and statesmen alike assumed that all
political systems were inexorably and inevitably evolving in the
direction of liberal democracy. In such a climate, democracy and stability
were inextricably intertwined in the minds of men. Alexis de Toque-vine
wrote, for example, that individuals living in democratic communities
"are forever varying, altering, and restoring secondary matters;
but they carefully abstain from touching what is fundamental."18
NontheoreHcal Emphases The concern for rigorous and systematic em-! pineal
theory building wasToreignto most traditional scholars ofcorn^... paratrve
politicsTVirtually no attempt was made tq_rormulate lightly o-, organized
ancTtestable generalizations relating to political processes. Although
classic political tKFnkers such as Machiavelli, Hobbes, Montesquieu/ and
Madison introduced numerous perceptive and incisive general insights into
political study, they were not primarily concerned with operationalizing
concepts and building scientifically rigorous theory. The great nineteenth
century political sociologists, including Weber, Mosca, and Pareto, were
somewhat more interested in scientific theory building. Within the early
modern political science discipline, however, there was little concern
for generating empirical theory.
The traditional emphasis was placed heavily upon normative theorizing,
a process that had deep roots in the history of American political science.
Indeed, the study of politics was first smuggled into the curriculum of
American colleges and universities under the cloak of such subjects as
moral philosophy and ethics. William Patey. whose classic study had a
profound impact upon American political science in its formative years,
wrote: "Moral Philosophy, Morality, Ethics, Casuistry, Natural Law,
mean all the same thing: namely, that science which teaches men their
duty and the reasons of if. "1T
By the turn of the twentieth century, the emphasis had come to focus on
"the good citizen." Knowledge of the duties and responsibilities
of citizenship and a deepening preoccupation with political philosophy
became the major concerns of political science. The former emphasis has
survived until contemporary times in the curriculum of many of the nation's
secondary schools, while the latter continues as an important branch of
political study in American universities and graduate schools. The concern
with normative political theory, or political philosophy, therefore, traditionally
dominated the discipline in a manner that tended to discourage the appearance
of empirical theory. Subjects such
as moral philosophy, ethics, and jurisprudence were considered and explicitly
A certain traditiona] bias has been formed against a science of politics
since it was (and still is in many circles) argued that applying rigorous
theoretical investigation in this area was impossible. The emphasis was
placed upon the unique, variable, and unpredictable nature of political
phenomena. Although nineteenth century thinkers sometimes termed politics
a science, they tended to define the latter in broad and varying ways.
Thus, Sir Frederick Pollock argued at the turn of the century that "there
is a science of politics in the same sense, and to the same, or about
the same, extent, as there is a science of morals."19
Methodological /nsensifiuify The field of comparative politics was from
the beginning limited in the methodological and research procedures that
its practitioners could utilize. Other than the classic schemes introduced
by scholars such as John Stuart Mill, Auguste Comte, and Herbert Spencer,
little was available in the character of refined and systematic comparative
methodology.20 Even these classic presentations of research logic, however,
were seldom utilized by political scientists. Techniques for selecting,
collecting, and ordering data were undeveloped and un systematic. Dominant
methods consisted of irregular observation and secondary source examination.
These kinds of procedures were buttressed and reinforced by personal intuition,
impression, and insight, the value of which varied greatly from scholar
to scholar. An early president of the American Political Science Association
once stated that "politics is an observational, and not an experimental
Sophisticated and systematic methodology was by and large not an area
of basic concern to traditional comparative government scholars. The descriptive
and formal-legal study of institutions did not demand deep behavioral
and scientific research techniques. The gathering and perusal of documents
and the intellectual interpretation of the legal intricacies of formal
government were considered relevant procedures to the tasks at hand. Not
surprisingly, therefore, as late as 1933 only one American university
offered a thorough course in the methods of politi-
cal research.22 More surprisingly, a 1952 analysis of 797 research projects
underway in 61 different graduate departments of political science indicated
that at most 15 of these studies were at all concerned with methodology
and/or the philosophy of science.23
These six characteristics have been presented as the predominant features
that have marked the traditional study of comparative politics. Their
presence and importance have varied from study to study as works occasionally
appeared that were not so parochial or descriptive or static. Indeed,
few studies exhibited all six characteristics in an unadulterated and
unequivocal manner. However, these were the primary patterns, and their
presence is stilt very much in evidence in current comparative political
analysis. The interesting question concerns how these characteristics
were able to dominate the field for so long and why they continue to persist
and to exhibit such lasting endurance.
Much of the reason resides in the manner in which alt six patterns interlock
and intertwine. An emphasis on one tends to develop and reinforce an emphasis
on another. Parochialism, for example, increases the stress placed upon
configurative description and formal-legalism. An interest in political
systems in genera! involves increased contexts of analysis, inviting more
comparison and less configuration. An acquaintance with less developed
and non-Western societies indicates very soon the futility of understanding
politics primarily in terms of laws and formal rules. Parochialism also
lends support to the static emphasis. A more serious concern with those
societies that reflect the widest gaps between "tradition" and
"modernity" calls attention to the critical problems of social
change and political development. In its limited analytical contexts,
parochialism also does little to suggest comparative theory building and
it fails to stimulate thought towards the development of fundamentally
new research techniques- Pronounced ethnocentric political analysis has
the further effect of indiscriminately portraying one's own political
system as the ideal system. In this kind of situation, normative theory
clearly assumes more importance than empirical theory.
Descriptive and formal-legal emphases in turn encourage parochialism.
The minute description of formal institutions does not lend itself to
the study of non-Western politics, nor do descriptive and formal approaches
do much to recommend the significance of the fundamental processes of
political change. Thus, these traditional characteristics sup-
port one another through
a persisting pattern of reciprocal reinforcement. This explains why leading
scholars as late as 1944 could refer to comparative government "as
a discipline in a status of suspended animation."24
The Transforming Study of Comparative Politics
The 1950 Hawley-Dexter study of political science research underway in
American universities tallied only 7 out of 797 projects that could be
clearly and definitely considered in the area of comparative politics.
The entire article devoted but four sentences to comparative politics
and concluded: "At best, emphasis on the field is slight."25
Less than a decade and a half later, Somit and Tanenhaus found the profession
designated comparative politics as the field in which the most significant
work was being done.28 The first issue of the new journal Comparative
Politics, published in 1968, described the field as an "exploding
culture" and in the same issue one theoretician referred to the "continuing
fermentation process" of the "new comparative politics."27
This changing emphasis is documented in more detail in figure 1, which
indicates the twentieth century development of comparative politics courses
in ten leading American universities. In 1925, approximately one of ten
undergraduate political science offerings was in the comparative area;
in 1945, the proportion had come to be nearly one in five; and in 1965,
the proportion was rapidly approaching one in three.28
An early reaction to the traditional study of comparative politics began
to develop in the 1920s, but not until the mid-1950s did significant and
systematic new trends began to appear. The formation and
crystallization of these trends developed as a direct reaction to the
characteristics that marked traditional comparative politics. As particular
aspects of traditional comparative study were deeply attacked, the impact
was felt throughout the network of mutually reinforcing emphases and methods.
The behavioral revolution that occurred during this period assisted in
undercutting former approaches and in unravelling the web or past procedures.
Within the social sciences in general, dissatisfaction with approaches
that have stressed only configurative description has been increasing.
Scholars have become preoccupied with "how" and "why"
questions and are less prone to rest content with the "what"
inquiry. Methods and modes of explanation have received growing attention
in each of the major fields generally included under the term political
science.29 Although describing political configurations is an essential
part of scientific political analysis, it represents only one important
step in the process of theory building. It is increasingly recognized
that description must be utilized to promote explanation and that configurative
exercises must become an integrated part of comparative analysis. Much
of the reason for the demise of the purely descriptive emphasis resides
in the close relationship that existed between this focus and the prevalence
The roots of reaction to formal-legalism in American political science
can be traced back to the 1920s and 1930s and to the pioneering writings
of scholars such as George E. G. Catlin, Charles E. Merriam, and Harold
Lasswell.3' These political scientists made similar pleas in the sense
that they all cautioned against formal-legal-institutional study and called
instead for the comparative analysis of power and control relationships.
Not until the 1950s and 1960s, however, was the traditional emphasis upon
formal-legalism replaced by a series of systematic new approaches/ all
of which purported to confront and explain the fundamental patterns of
political behavior. A major reason for the steady decline of formal-legal
study resided in the continuing discovery and increasing knowledge of
political systems where formal-legal structures were obviously only of
The mid-twentieth century witnessed the development in the United
States of a deep and real interest in African, Asian. Middle Eastern,
and Latin American political processes. The era of colonialism had been
shattered; the United Stales had established firm political and economic
commitments throughout the world; sharp competition developed between
rival ideologies in the struggle over the non-Western areas of the world;
international and interregional organizations of varying import and impact
were born; and spectacular technological advances in the fields of transportation
and communications increased contact between and among all societies.
At the same time, area study programs began to appear within American
universities, and students increasingly began to study the history, language,
culture, and politics of societies that had been formerly considered exotic.32
Private and public scholarship funds were made available to students and
scholars interested in the sociopolitical processes of Afro-Asian and
Latin American societies. Supported by a background in social science
and methodology and armed with linguistic tools such as Arabic, Hindi,
Swahili. Persian and Japanese, these scholars have rapidly increased the
contexts necessary to successful comparative study. Table 1 reveals the
tremendous growth of non-Western and Latin-American oriented courses in
the political science curricula of ten major universities.
In 1925, only one course in non-Western politics was offered in all of
the ten institutions of higher learning studied. This was Political Science
135 taught at the University of California and entitled "Political
Development of China." Four decades later, these universities accounted
for 65 different courses in non-Western and Latin-American political systems.
Today, approximately every other course offered in the field of comparative
politics concerns Third World systems. Many decades ago, George Catlin
warned that "political studies will not be advanced by a political
science of the arm-chair."33 What Karl Loewen-stein pointed out a
bit prematurely in 1944 can now be reiterated more appropriately: "Supercilious
aloofness from other peoples' political institutions-another facet of
intellectual isolationism-has definitely come to an end."34
In the 1950s, political
scholars began to focus seriously and systematically upon the problems
of political development and social change. Words such as "modernization,"
"change," "development," "revolution," "reform,"
"transformation," and "process" began to appear everywhere
in the literature. Leading comparative specialists immediately confronted
these issues and their analyses remain among the most incisive contributions
to the study of political change. Scholars such as Samuel P. Huntington.
Gabriel Almond, Fred W. Riggs, David E. Apter, Manfred Halpern, Lucian
W. Pye, Dankwart A. Rustow and Leonard Binder continue to move the study
of comparative politics in a direction that will enable it to address
successfully the challenge of modernization- Not accidentally, most of
these theoreticians are fundamentally concerned with non-Western societies-
A deep knowledge of such societies directly reflected the existence of
rapid change, great development gaps, and concomitant tension, frustration,
conflict, and violence. This, in turn, occasioned a reexamination of Western
political patterns in general and American politics in particular.35 The
issue of political development is presented as holding universal import.
Finally, contemporary comparative politics is marked by a deepening concern
for theoretical and methodological considerations. Here the thorny issue
of behavioralism becomes acutely relevant. Voices for and against a changing
political science have used this term to buttress their respective stances
and to vindicate their obvious dogmatism. The "be-havioralists"
too often have considered themselves scientific sages because of their
facility in the arts of statistics and neology, while the "anti-behavioralists"
have tended to sneer defensively at anything and everything that suggests
scientific method and theory building. Few have bothered to define "behavioralism"
cautiously, concisely, and clearly.
, Behavioralism can be defined as the systematic search for political
patterns through the formulation of empirical theory and the technical
analysis and verification thereof. Behavioralism involves two basic emphases:
"the formulation of concepts, hypotheses, and explanations in systematic
terms" and "empirical methods of research."36 The essence
of a behavioral science is that "all hypotheses be experimentally
confirmed by reference to publicly observable changes in behavior."37
According to David Easton, the behavioral movement within the discipline
of political science has involved special concern for the following eight
characteristics: regularities, verification, techniques, quantification/
values/ systematization, pure science, and integration.38 There are regularities
in political behavior that can be stated and tested. The discovery and
validation of such uniformities can be furthered and ascertained by rigorous
techniques involving the selection, collection, and ordering of data.
This task can in turn be facilitated by the introduction of precision
through measurement and quantification. Behavioral-ism does not discard
values, but rather stresses the need to distinguish j carefully and analytically
between fact and value.39 Systematization ' involves a recognition of
the need to relate empirical research to theory building and to stress
the symbiotic relationship between theory and data. The emphasis called
"pure science" refers to the logical precedence of theory and
explanation to the specific activities of political engineering. Finally,
behavioralism often calls for the utilization of cross-disciplinary approaches,
methods and findings.
Various scholars have stressed one or another of these characteristics
of behavioralism- Indeed, intra behavioral divisions have increasingly
resulted in allegations and accusations reflected in terms such as "non-behavioral,"
"prebehavioral/' "pseudo behavioral/' and "postbehav-ioral."
Perhaps the major fissure in past dialogue among behavioralists has been
between'those seeking to confine the meaning of behavioralism to the sophisticated
employment of quantitative techniques and i precision of measurement and
those identifying behavioralism more 1 with the scientific methodology
of conceptual rigor, hypothesis formu- i lation, and theory building.
Although controversies such as these have ', often generated a narrow
academic dogmatism, they have also con- '
tributed to a continuing evaluation and reassessment of current trends
in political analysis.
The behavioral approach as equated with technique has been challenged
fundamentally by what is increasingly referred to as the "post-behavioral
revolution." David Easton's "Credo of Behavioralism" is
being replaced by what he terms a "Credo of Relevance."40 The
post-behavioral revolution does not displace behavioralism in its broader
dimension but sensitizes the political scientist to the role of values
and the importance of policy considerations. This trend will play an important
role in charting the future directions into which the study of comparative
politics will move.
The major dissatisfaction with behavioral emphasis has been the preoccupation
with technique rather than substance, contemplative theorizing rather
than policy relevant theory, and neutral academic conservatism rather
than progressive social transformation. The marked tendency to theorize
about theory (metatheory) and to amass and study only that data conducive
to exact measurement has occurred in a world increasingly convulsed by
division, discrimination, poverty, and violence. In his 1969 presidential
address to the American Political Science Association, David Easton stated
that "the search for an answer as to how we as political scientists
have proved so disappointingly ineffectual in anticipating the world of
the 1960s has contributed significantly to the birth of the postbehavioral
revolution." He pointed out that "mankind today is working under
the pressure of time. Time is no longer on our side."41 Sheldon Wolin
argues that in a world that "shows increasing signs of coming apart
. . . official political science exudes a complacency which almost beggars
description."42 Hans J. Morgenthau has spoken about the tendency
of contemporary political scientists "to retreat from the burning
political problems of the day into a kind or methodological or factual
irrelevance." He argues that the "innocuous" political
scientists of the past "created a respectable methodological wall
between themselves and the political problems of the day and were exactly
those who were irrelevant then and who are forgotten today."43 August
Heckscher has written that "research which disavows any responsibility
except that of being objective and non-utilitarian may well qualify as
'pure.' But it is the kind of purity which a society-particularly a society
in an age of change-can overvalue."44
In a world caught in the midst of fundamental change, problem selection
and problem solution assume increasing importance- There is an embryonic
but marked trend away from the speculative/contemplative and a movement
towards the meaningful application of social and political knowledge.
The methodology of behavioral science is increasingly being related directly
to the sociopolitical issues of the day. Thus, David Easton foresees a
"politicization of the profession," while Harold
Lasswell calls for a "problem-solving attack on political science."'15
Yet/ despite the fact that the priorities and perspectives of the discipline
are being increasingly reassessed as technique is measured against substantive
contribution, there is little doubt that the behavioral concern for methodological
rigor will long remain. The question is one of emphasis and purpose, and
the movement is one that stresses the confluence of scientific methodology
and problem solution. In the mainstream of this postbehavioral movement
are a number of related tendencies that promise to characterize the future
study of comparative politics. There are four broad categories of trends.
Interdisciplinary Cooperation and Communication
With the drive for more problem-centered inquiry, the often artificial
and arbitrary lines which divide the social sciences become increasingly
irrelevant and indefensible. Many of the important strides already made
in the recent resurgence in the comparative study of politics owe a great
deal to methods and approaches borrowed from sociology. Examples include
structural-functional analysis, as well as refined group, class, and elite
analysis. Future developments indicate an intellectual convergence of
political science with economics, anthropology, and social psychology.
In the search for truly comparative tools for the analysis of social and
political behavior, for example, a growing number of political scientists
have turned to economic rationality models and to supply and demand laws
applied to collective goods. Others are moving in the direction of political
anthropology as they seek to ferret out and compare the informal patterns
of interpersonal power and authority relationships. The findings of social
psychology are increasingly drawn upon by scholars seeking to apply learning
theory to the study of political socialization. These represent only a
few examples of an accelerating drive that is beginning gradually to break
down the barriers that divide the social sciences. Departments, divisions,
and institutes of social science are increasingly becoming realities.
The methodologies, theoretical frameworks, and conceptual apparatus pertinent
to comparative analysis are neither monopolized by nor concentrated within
the discipline of political science. In this spirit leading scholars such
as David Singer and David Easton have already called for the establishment
of a "Federation of Social Scientists."46
These interdisciplinary trends, however, may carry far beyond inter-social
scientific inquiry. The study of comparative political patterns, for
"Easton. "The New Revolution," p. 1060; and Harold Lasswell,
'The Future of the Comparative Method," Comparative Politics 1 (October
1968):11. "Easton, "The New Revolution," pp. 1060-61.
example, may gain profoundly from advances made in the biological sciences.
The developing field of ethology or social biology, which concerns the
analysis of animal behavior and its patterns of evolution, has already
suggested numerous provocative propositions that may go far to explain
the behavior of the political animal. The crucial comparative problems
of conflict and cooperation closely parallel the ethologists' concern
for the processes of aggression and sociability. Scholars such as Konrad
Lorenz and John N. Bleibtreu provide particularly stimulating studies
of animal behavioral patterns that are often explicitly related to human
political behavior. Comparative political analysts are only now beginning
to open their lines of communication in these less orthodox directions.47
Intersociefal Cooperation and Comprehension Comparative political study
has been marked by a continually widening universe of analysis that began
with the narrow plethora of studies focusing almost exclusively upon Euro-American
societies. This was followed by country monographs and area studies. Political
analyses of particular societies that can be utilized as case studies
for purposes of comparative analysis and theory building are in many cases
nonexistent. Country monographs that are conceptually sound, theoretically
rigorous, and problem oriented are extremely rare. Area studies which
further intraregional, cross-societal investigation are only in the infancy
of their development. By providing multicontextual laboratories, such
studies often encourage the comparative analysis necessary to the generation
of meaningful hypotheses. Future problem-oriented comparative study will
draw upon the experiences of societies and cultures anywhere and everywhere
in the world. Subcultures and intrasocietal structures and institutions
will also serve as units for comparative study. Country and area studies
will no doubt continue to develop and increase along with what Dankwart
Rustow calls "cross-regional comparison."48
The tremendous task of confronting the multitude of varied laborato-
ries of comparative political experience will increasingly be shared by
social scientists representing nations throughout the world. African,
Asian, and Latin American scholars, for example, have begun to team with
Europeans and Americans for the study of political processes in the societies
under investigation. Scholars native to the particular society o under
study carry social and cultural advantages that enable them to ) analyze
the system from within. The outside scholars bring a differing;
and more detached perspective to bear upon that same political system.
| Cooperative efforts by such multinational research teams should im-1
prove immeasurably our understanding of comparative political processes.
Scattered but serious efforts in this direction have already begun to
Special note might be made of the growing proclivity to treat the study
of American politics as an integral part of the study of comparative politics.
Surely, many of the intradisciplinary divisions are as artifi-. cial and
sterile as some of the interdisciplinary distinctions referred to above.
Problems such as political development, poverty, violence, and discrimination
are as much a challenge to the United States as to any other society.
The experiences and patterns that mark the most elaborately studied society
in the world are increasingly becoming an essential part of comparative
political investigation. There is no reason why an "American specialist"
should not be a "comparative specialist" and vice-versa. The
political and personal patterns that mark American subcultures, for example,
may very well more closely approximate those of certain Afro-Asian societies
than those of white middle class America.49 This kind of comparative analysis
promises to generate provocative and policy relevant hypotheses.
Academic Cooperation and Communication
Partially because of the growing demand upon the scholar of comparative
politics to be a theoretician, historian, statistician, mathematician,
sociologist, psychologist, linguist, and country/area specialist, and
partially because of the deep need for more imaginative analyses and prognoses,
increasing communication and continuing scholarly dialogue must inevitably
develop. The establishment of institutes, seminars, workshops, and conferences
on a more regular basis can, and in some cases has already begun to, mean
a cooperative cross-fertilization of ideas. Harold Lasswell, for example,
has recently called for "a nationwide and worldwide network of counterpart
seminars-"50 Such semi-
nars will contribute to the presentation, sharing, probing, and generation
of ideas. A system of in-depth dialogue and confrontation may someday
absorb students, researchers, and faculty of varied disciplines and backgrounds
in a systematic effort to improve our understanding of comparative political
Inlerprocedural Experimentation and Adaptation
The drive to refine the experimental, statistical, and mathematical techniques
which were emphasized and developed during the behavioral period will
continue. New quantitative procedures and measurement devices will be
constantly generated to seek precision whenever and wherever possible.
Social science research and experimental specialists will increasingly
serve as indispensable technicians for the translation, measurement, and
testing of hypotheses. The continuing advance of technology will build
deeper sophistication into methods of gathering, cataloguing, storing,
retrieving, and comparing data. In the search for more accurate data,
methodology and techniques are beginning to be borrowed from sciences
such as biology and physiology.
At the heart of the behavioral movement and the continuing transformation
of the study of comparative politics has been a growing concern for theory.
Much of the controversy that swirls about behavioralism and postbehavioralism
concerns the role of problems, approaches, tax-onomies, and data in the
theory building process. The trend to theory means many things to many
people, and because of this it demands explicit examination and clarification-
The Theoretical Mold
The spirit of generalization should dominate a university .... During
the school period the student has been mentally bending over a desk;
at the university he should stand up and look around.... The function
of the university is to enable you to shed details in favor of principles.
Alfred North Whitehead51
During a plane trip to three societies outside of the United States, an
American traveler who is interested in the question of whether there is
a relationship between a society's legislative system, literacy rate,
and form of government first visits country x which he knows is a Middle
Eastern monarchy. During his stay there, the visitor notes an absence
of any legislature and the presence of a high illiteracy rate. He then
travels to country y which is a Southeast Asian monarchy where he again
observes the absence of legislative politics and the existence of a
high rate of illiteracy.
He next travels to country z in Africa about which he has no political
knowledge. Here, the traveler notes the absence of any legislature along
with high illiteracy- He concludes that this country is a monarchy.
In this situation, the traveler engaged in both inductive and deductive
mental processes. In his first two experiences, the absence of a legislature
and the existence of a high rate of illiteracy went together with a monarchical
form of government. This was the case in the first country and it was
confirmed in the second society. Although this is an obviously weak basis
for conclusion, it is still enough to make a crude 9
induction, i.e., to generalize from particular facts. The observer expected
to find a monarchy if illiteracy prevailed and legislatures were nonexistent.
From this, the generalization was formulated that all countries lacking
legislatures and possessing high rates of illiteracy are monarchies- Upon
visiting another highly illiterate and "legislatureless" society,
he concluded: All "legislatureless" and highly illiterate societies
are monarchies; this society is "legislatureless" and highly
illiterate; therefore, this society is a monarchy. This is a logical case
of deduction, i.e., formulating a specific conclusion on the basis of
a general assumption. In this described reasoning process, a generalization
was first established by induction and upon this generalization was founded
The above example represents a step-by-step replication of the method
of scientific investigation presented by the great nineteenth century
scientist, Thomas H. Huxley.52 Huxley once defined science as "trained
and organized common sense" and told his readers that theory building
is something you engage in "every day and every hour of your lives."53
Albert Einstein has written that "the whole of science is nothing
more than a refinement of everyday thinking."54 The focus upon theory
and the scientific method represents the heart of the behav- | ioral and
postbehavioral movements in contemporary political science, j Nevertheless,
the concept theory remains surrounded by tremendous . confusion.. and
many students of comparative politics still recoil men- :
tally when confronted with the subject. There are, of course, many reasons
for this apprehensive attitude-
The concept theory has been treated in an offhand and rather cavalier
manner even by leading scholars of political science. The concept has
been left completely undefined;55 it has been defined indirectly;56 it
has been defined ambiguously;57
and it has been equated with other equally confusing terms such as model
and method.58 The situation has ""7 been muddied considerably
by the necessary consideration of several / additional ill-defined and
ambiguously-presented concepts, all of which ' are crucially related to
theory and the process of theory building. These concepts include approach,
conceptual framework, definitional system, taxonomy, classificatory scheme,
typology, model, generalization, hypothesis, law, and paradigm. Philosophers
of science themselves have been unclear and inconsistent in their usage
of such terms. Certainly no common, agreed-upon system of definitions
exists that even begins to distinguish and relate a comprehensive list
of these important concepts. The political scientists' usage of these
terms is considerably more confused since their views are shaped by sporadic
secondhand reliance upon different philosophers of science.59 A further
cause of this confusion is the fact that a terminology generally agreed
upon is as much the outcome of scientific progress as the precondition
The difficulties have been compounded and continue to increase as scholars
feel compelled to make perfunctory bows in the direction of these key
concepts. Therefore, the literature is sprinkled with loosely presented
terms that happen to be most crucial to the theorizing process. There
are, for example, theoretical and conceptual approaches, theoretical and
conceptual frameworks, theoretical and conceptual systems, theoretical
and conceptual models, and theoretical and conceptual typologies. Approaches,
taxonomies, typologies, and models are constantly presented and portrayed
as theories. Taxonomies and typologies are sometimes treated as synonyms;
other times they are sharply distinguished. This same kind of inconsistency
marks the situation of model-paradigm and generalization-theory. Leading
and theory conscious political scientists the caliber of David Easton,
for example, have equated the term "conceptual framework" with
"a body of theory." "a theoretical model," -and "a
system of working hypotheses."60
The following discussion represents an attempt to define, distinguish,
and relate explicitly those-concepts central to theory building in comparative
political analysis. The uncertainty that surrounds these concepts is well
recognized, but it is also relevant to note that serious
dispute exists about even more fundamental matters such as the very methods
presented in philosophy of science.<" The "scientific method"
itself is by no means something agreed upon and accepted.ez The reader
should not be surprised, therefoie, if the definitions and distinctions
presented herein do not always coincide with his own conceptions or those
of a particular school of methodology. The definitions in the following
list have been formulated in view of five major considerations: (1) general
views on these concepts as expressed in the philosophy of science literature;
(2) meanings of these terms as used in the natural and physical sciences;
(3) importance of simplicity and conciseness; (4) need to relate these
concepts one to another and then to the overall process of theory building;
and (5) concern for the subject matter and problems of comparative politics.
Approach: A predisposition to adopt a particular conceptual framework
and to explore certain types of hypotheses towards the generation of theory.
Conceptual Framework: A schema of explicitly defined and differentiated
concepts necessary to the study of a particular subject.
Taxonomy: A subject divided into classes distinct from one another.
A taxonomy is also referred to as a classificalory scheme or classification
Typology: That kind of taxonomy in which dassificatory distinctions are
graded or ordered.
Model: A theoretical and simplified representation of the real world.
Generalization: Statement of uniformities in the relations between two
or more variables of well-defined classes.
Hypothesis: A generalization presented in tentative and conjectural terms.
Theory: A set of systematically related generalizations suggesting new
observations for empirical testing.
Law: A hypothesis of universal form that has withstood intensive experimentation.
Paradigm: The worldview which legitimates the scientific community's consensus
on what constitutes exemplary scientific research.
Approach and Theory One
of the most important concepts to the systematic investigation of comparative
politics is reflected in the term approach. Basically a very simple concept,
the idea of approach has in many respects become one of the most often
misunderstood and distorted. The major reason for the difficulty inheres
in a persisting proclivity among scholars to label and present certain
approaches as theories. Such terms as "group theory/' "functional
theory," and ^systems theory," are prime examples of this confusion.
These three, for example/ are approaches, and it is difficult to see how
they can be termed theories since they are insufficiently articulated
in terms of the systematic relationship of hypotheses.
An approach is a predisposition to adopt a particular conceptual framework
and to explore certain types of hypotheses towards the generation of theory.
At bottom/ an approach is the particular orientation that one adopts when
addressing a subject or issue. It is the line of advance that a scholar
takes in initiating his investigation. This orientating framework is of
crucial import to the theorizing process since it determines what sets
of concepts, questions/ perspectives, and procedures the researcher will
adopt in pursuing his inquiry. The approach that one selects will decidedly
shape the hypotheses that are generated and ultimately the theory that
is formulated. An approach may be implicit or explicit; it may be crudely
developed or highly refined; and it may be utilized either to describe
This volume is pre-eminently a critical survey of various theoretical
approaches to the study of comparative politics. It is not a survey of
theories. Rigorous, systematic, and explicit theories of politics are
still in relatively embryonic stages of development. Approaches, however/
provide the frameworks within which theories are constructed. This aspect
and function of an approach will be strongly emphasized in the following
pages. The major concern herein is for developing approaches to theory.
The relation of an approach to a theory is of vital importance to this
study. Thus, we use the term theoretical approach.
Conceptual Frameworks A conceptual framework is here considered more the
equivalent of a definitional schema than of some broader system of classification
or theory.63 More accurately, it is a schema of explicitly defined and
differentiated concepts that are necessary to the study of a selected
subject. A conceptual framework provides a systematic arrangement of those
conceptual tools that accompany a particular theoretical approach. It
must not under any circumstances be considered
theory since it neither
generalizes nor explains. A well-formulated and rigorously constructed
conceptual framework, however, is an essential element in the theorizing
process. If the basic concepts of an approach are ambiguous, then the
resultant theory will also be ambiguous- A confusing/ unclear/ or contradictory
conceptual schema signals confusion for any theoretical edifice that might
rise around such a framework. For this reason, Arthur Stinchcombe writes
that "social theorists should prefer to be wrong rather than misunderstood."8"
The process of conceptualization is a complex one since it is directly
entwined with such theory relevant exercises as classification and generalization.
Concepts themselves can be classified according to the level of abstraction
or stage of generality at which they are located. Karl Popper, for example,
distinguishes between universal and individual concepts. According to
Popper/ terms such as "dictator," "planet," and "HzO"
refer to universal concepts while "Napoleon," "the earth,"
and "the Atlantic" are corresponding individual concepts.85
A number of notable attempts have recently been made to confront this
issue of generality with specific regard to comparative political and
sociological study. One scholar writes about universal, general/ and configurative
conceptualizations which in turn correspond to high, medium/ and low level
Another pair of theorists analyze the situation in terms of concepts dichotomously
viewed as observables and constructs.67 Observable concepts are susceptible
to direct sensory observation and exist at a relatively low level of generalization.
Constructs are general abstractions of persons, places, or events. The
process by which one moves from the observable level of conceptualization
to the more abstract constructual level has been termed abduction.88 Observable
or low level concepts are grounded more in the "real" world
and tend to preserve the details and intricacies of the situation under
investigation. The more abstract species of concept lends itself readily
to generalization and "can transcend the limits of individual instances
precisely because its terms are not completely dependent upon any one
of these cases or
a sum of them for their
meaning."69 Despite all this, both levels of conceptualization are
crucial to theory construction. Also, a scholar's awareness of these kinds
of conceptual distinctions is critical to successful theorizing.
One of the first requirements of scientific theory building is that the
basic concepts be presented explicitly and clearly. Implicit conceptual
frameworks invite imprecision and misunderstanding. This has been particularly
so in comparative political analysis where the most vital concepts carry
an extraordinary amount of definitional haze and ambiguity, Examples include
terms such as "group/" "class/" "power,"
"authority," "structure," and "system."
Despite this, scholars of comparative politics continue to leave crucial
concepts undefined, and when they do indulge in definition they do so
sporadically and unsystemati-cally- The general trend seems to be that
only graduate students are expected to provide rigorous conceptual schemata
which must necessarily preface their term papers and theses. The point
is not to support exercises in verbal hair splitting and definition bickering,
but rather to stress the important theoretical role of clear and consistent
conceptual presentation. Carl G. Hempel argues that the first fundamental
requirement for scientific theory building is "a clear specification
of the basic concepts. .. ."70 One Nobel prize winner argues that
all science "depends on its concepts. These are the ideas which receive
names. They determine the questions one asks and so the answers one can
get. They are more fundamental than the theories, which are stated in
terms of them."71
Taxonomies, Typologies, and Models Taxonomies, typologies, and models
contribute significantly to the theorizing process. Very often they are
presented as theory, but like conceptual frameworks they are neither generalizing
nor explanatory instruments. Richard Rudner summarizes this situation
perfectly when he describes the activity of constructing taxonomies, typologies,
and models as "theoretical work" but not "formulations
A taxonomy (also referred to as a classificatory scheme or classification
system) is a subject divided into classes distinct from one another.
In this kind of classification
exercise, an object either falls into a defined class or it does not,
depending upon whether or not it possesses the characteristics essential
to inclusion in that class. Because of the shaded, graded, and changing
nature of much of the subject matter central to sociopolitical affairs,
however, classification is difficult. It is certainly often impossible
to classify with any kind of precision. A typology carries more flexibility
than a taxonomy since it permits ordered classifi-catory distinctions.
Typologies introduce a more subtle and precise method of classification
which is rendered especially so by the possibility for the introduction
of quantitative techniques. Formulations that concentrate upon continua,
axes, polarities, and serials are typologies, not taxonomies.73
A model is a theoretical and simplified representation of the real world.71
It is an isomorphic construction of reality or anticipated reality. A
model, by itself, is not an explanatory device, but it does play an important
and directly suggestive role in the formulation of theory. By its very
nature it suggests relationships. Whereas taxonomies and typologies divide
and order, models reconstruct. The jump from a model to a theory is often
made so quickly that the model is in fact believed to be the theory. A
model is disguised as a theory more often than any other concept. This
confusion has been complicated by the recent popularity of the term parad'xgm-a
concept often confused with both model and theory.
In ei general sense, the concept of paradigm is closely related to the
idea of model since it is derived from the Greek term paradeigma meaning
to set up as an example. For purposes of comparative political analysis/
however, paradigm will be defined in a sense relatively akin to the meaning
introduced by Thomas Kuhn, whose work has had a decided impact upon theories
of scientific development.75 The concept of paradigm refers to the fact
that (1) the particular scientific community holds basic assumptions about
what it is investigating; and (2) examples of recognized exemplary scientific
research exist that are accounted for in terms of these assumptions. An
individual who accepts this basic world view as well as the research that
develops from these
assumptions is regarded as a bona fide member of the scientific community.
A paradigm, then, is the worldview which legitimates the scientific community's
consensus on what constitutes exemplary scientific research.76
Taxonomies, typologies, and models perform several positive functions
in the theorizing process. They describe, order, clarify, stimulate, and
compare. Typing and classifying, for example, are first methods of description.
The systematic and ordered nature of such description makes it an especially
valuable tool to theory building. Indeed, the ordering of data serves
as a catalyst in the theorizing process because it clarifies what is usually
very complex subject matter, and serves to stimulate hypotheses and generate
theory. A tidy taxonomy or a refined model often reveals patterns and
relations that were formerly concealed by the haphazard and jumbled arrangement
of data. The very exercises of classifying, typing, and modeling also
often suggest new connections and relations that can be framed as tentative
generalizations. Finally, whether it is classes or types on the one hand
or isomorphisms on the other, an element of implicit or explicit comparison
is always involved. One class or type is, at least in some respects, compared
to another class or type, and continual and reciprocal comparison is involved
between models and the objects they seek to represent.
One of the major difficulties inherent in the intensive construction of
taxonomies, typologies, and models involves a loss of scientific perspective.
The development and refinement of these kinds of mental constructs becomes
an end in itself. Taxonomies and models are developed, defended, discarded,
redeveloped, and redefended in a ceaseless intellectual exercise. This
kind of process can be quite fruitful if the theoreti-cian remembers exactly
what he is doing and realizes exactly what are the intellectual contributions
of the exercise- If, however, the theorist presents these exercises as
theory, he may in fact be impeding the theorizing process. Not only does
he twist and distort the meaning of theory, but he also invests a great
deal of energy in projects not related to theory-building. This "taxonomizing"
or "modelmania" results, therefore, in the proliferation of
sterile constructs that rest outside the process of theory building.
Despite dangers such as these, taxonomies, typologies, and models represent
important steps on the road to theory. "Scientific explanation requires
the systematic ordering and classification of empirical data.""
The Essence of Theory
The analysis of comparative political theory requires that an early distinction
be drawn between two types of theory, normative theory and empirical theory.
Traditionally, the field of political theory within the discipline of
political science concerned itself with the study of normative theory.
The emphasis was placed upon norms, i.e., rules and standards, of political
behavior. Investigation centered upon how man should or ought to act politically,
and the history of the development of political values and goals dominated
the study of political theory. Discussions concerning such fundamental
values as justice, equality, freedom, and democracy have long existed
at the core of normative political theory. This important area of analysis
continues to contribute to political science and is now referred to as
political philosophy as well as political theory. Empirical theory refers
to a reliance upon experience, observation, and experiment. The basic
concern is with actual and observable behavior rather than with proper
or desirable behavior. Empirical theory involves the observation, generalization,
and explanation of actual (empirical) behavior. In this study, theory
always refers to empirical theory unless otherwise noted.78
A theory is a set of systematically related generalizations suggesting
new observations for empirical testing. Thus defined, the concept "theory"
contains three major elements: (1) theory always involves generalization,
i.e., it includes statements that highlight uniformities between two or
more variables; (2) theory suggests new observations, i.e., it draws relations
that carry different explanations of empirical reality; and (3) theory
is testable. The formulated generalizations must possess the capacity
to be tested, i.e., to be falsified on the basis of evidence drawn from
empirical examples. A theory is a general statement that systematically
calls attention to regularities and patterns. Carl G. Hempel writes that
besides a clear specification of the basic concepts, a scientific theory
requires; (1) a set of general assumptions; (2) a connection between the
theoretical statement and observable phenomena; and (3) "testability-in-principle"
of the theory, i.e., the presence or absence of observable phenomena measured
against the theory will provide confirming or discontinuing evidence concerning
Various types of generalizations fall within our definition of theory.
These range all the way from generalizations of probable applicability
to generalizations of invariable applicability. One of the most lucid
classifications of generalizations has been provided by Eugene Meehan
who distinguishes universal generalizations, probabilistic generalizations,
and tendency statements. According to Meehan, a universal generalization
is phrased in terms of "All A is B" or "If A then B."
An example relevant to comparative political analysis and one that at
the same time preserves some modicum of credibility would be: "All
military coups that result in a change in political elites are planned
and engineered by middle-ranking army officers."80 The probabilistic
generalization is limited to those statements in which a particular percentage
is an integral part of the generalization. An example would be:
"Eighty-five percent of military coups that result in a change in
political elites are planned and engineered by middle-ranking army officers."
Finally, an example of a tendency statement is as follows: "Military
coups that result in a change in political elites tend to be planned and
engineered by middle-ranking army officers." All of these classes
of statements are theoretical but only universal and probabilistic generalizations
can be termed strictly theory.81
One of the key components of a theory is that it must carry the capacity
to be falsified. If empirical evidence exists that contradicts the theoretical
statement, then the latter may be refuted. Usually, it is necessary that
the contradicting evidence be reproducible or recurring in order to disconfirm
the theory under question.82 It is impossible to confirm or prove a theory.
This is why theory construction is a continu-ing and open-ended process.
The distinguished physicist Richard Feyn-^man summarizes this important
point concerning the essence of theory in the following terms:
There is always the possibility of proving any definite theory wrong;
but notice that we can never prove it right. Suppose that you invent a
good guess, calculate the consequences, and discover every time that the
consequences you have calculated agree with experiment. The theory is
then right? No, it is simply not proved wrong. In the future you could
compute a wider range of consequences, there could be a wider range of
experiments, and you nught then discover that the theory is wrong. That
is why laws like Newton's laws for the motion of planets last such a long
time. He guessed the law of gravitation.
calculated all kinds of
consequences for the system and so on, compared them with experiment-and
it look several hundred years before the slight error of the motion of
Mercury was observed. During all that time the theory had not been proved
wrong, and could be taken temporarily to be right. But it could never
be proved right, because tomorrow's experiment might succeed in proving
wrong what you thought was right. We never are definitely right, we can
only be sure we are wrong."
In the social sciences there is often a deceptive confidence in low level
theoretical statements since they appear to resist falsification. This
confidence is certainly premature for two basic reasons: (1) no theory
is ever confirmed and all such statements require time to be tested; and
(2) many theoretical statements are presented in forms that make falsification
impossible. As such, they cannot be referred to as theory. The political
science literature is laden with statements that are purported to be theory
but are in fact so broad, indefinite, flexible, and ambiguous that they
can indefinitely resist falsification. Feynman correctly points out that
"you cannot prove a vague theory wrong-"84 And as a leading
scholar of applied mathematics writes: "... a theory that is not
rigid enough to be disproved is just a flabby bit of talk. A theory is
scientific only if it can be disproved. But the moment you try to cover
absolutely everything the chances are that you cover nothing."85
Tendency statements fall into this intellectual trap. It is extraordinarily
difficult, if not impossible, to falsify such statements. For this reason,
tendency statements cannot be accurately considered theory. They can,
however, serve an important function in the theory building process since
they can act as general orientation guides out of which theory can be
fashioned. Tendency statements can be highly suggestive and intellectually
provocative and, if refined, can often form the basis of theory. It is
true that "a good tendency statement may be far more useful in the
explanation of political phenomena than a precise statistical generalization
that deals with trivia."86 And it is also true that a continual exercise
in building, connecting, and refining these kinds of generalizations leads
slowly in the direction of probabilistic and universal theory. The process
The subject matter of comparative politics is such that theoreticians
have not moved much beyond the generation of tendency statements and probabilistic
generalizations. There has been an orderly retreat away from the drive
for universal generalizations or high level theory and various scholars
have coined various terms for the kind of theory they feel social scientists
should develop. Thus, there has been a call for "singular generalizations,"
"narrow gauge theory," "middle range theory/' "partial
theories," "piecemeal theoretical insights," and "modest
general propositions.""7 The changing, uncertain nature of political
processes has been one major reason for the difficulty in building high
level theory. (It was Albert Einstein who once pointed out that politics
is more difficult than physics.) Another problem is the difficulty of
gathering relevant data and the further dilemma of measuring that data
or the relations that link that data.
Thus, even probabilistic theory is difficult to develop in comparative
politics. Only in cases of less complex empirical issues where it is easy
to collect relevant data can one build low level probabilistic theory.
It has been clearly and correctly pointed out that:
Probability can only be stated in numerical terms when the phenomena to
which it relates can be measured additively. In political science, that
class of measurable phenomena is quite limited and it is obvious that
many significant events cannot be so measured, though they are worth explaining.
We can produce precise statements about voting behavior, or population
movements, or various phases of economic activity, but the range of phenomena
is narrow and the boundaries of the phenomena are not easily defined.BB
Comparative political analysis, therefore, continues to oscillate between
tendency statements and probabilistic generalizations. The drive to construct
theory is confined largely to the area of relatively low level empiricalgeneralizations.
The Process of Theory Building
The actual example of theory building presented at the beginning of
this section represents a crude exercise reflecting the core of the overall
theorizing process sometimes referred to as the scientific method. Scientific
theory building involves the interrelated processes of problemation,
"The first two terms belong to David Easton. The others have been
used by Robert Merlon, Charles P. Kindleberger. Arthur L Burns, and V.
0. Key respectively. "Meehan, The Theory and Mefhoa of Political
Analysis, p. 114.
observation, generalization/ confirmation, and application.89 (See figure
2.) These overlapping steps or stages are herein analytically distinguished
one from the other for purposes of more detailed analysis. Despite the
constant movement back and forth between and among these various processes,
they are presented in the order in which they most often and logically
Problem selection is one of the most important, if least discussed, exercises
in theory building. This involves the determination of issues to be theorized
about and the construction of patterns of questions to be raised. It is,
of course, necessary to theorize about something, to draw relations between
certain variables, to explain certain phenomena. The importance of a theory
about hair styles of world political leaders is highly questionable, no
matter how rigorous and refined this theory may be. Similarly, a precise
and well-formulated theory about electoral patterns in the Afro-Asian
world may be of little value if the votes cast in these settings reflect
nothing about the actual power and authority relations in this region.
The business of problem selection is that part of the theoretical process
often referred to as the "context of discovery." Norwood R.
Hanson stresses the crucial nature of this context when he argues that
"the ingenuity, tenacity, imagination and conceptual boldness which
has marked physics since Galileo shows itself more clearly in hypothesis-catching
than in the deductive elaboration of caught hypotheses.90 Barrington Moore
writes that "techniques alone cannot define what is scientifically
The process of successful theory construction relies heavily upon such
considerations as intuition, imagination, insight, guesses, and hunches.
Although the social and political sciences tend to play down this aspect
of scientific inquiry, the more exact sciences cannot seem to stress these
elements enough. The important role of chance and the unconscious is also
an oft discussed topic in the literature on scientific method and theory
construction in the physical and biological sciences. ^ James B. Conant
has written, for example, that "the great
working hypotheses in the past have often originated in the minds of the
pioneers as a result of mental processes which can be best described bv
such words as 'inspired guess/ 'intuitive hunch/ or 'brilliant Hash of
imagination/"93 Sharp intuitive capacities are invaluable in the
selection of key problem areas and in the generation of theory.
In social and political analysis no less than in physical inquiry the
ability to make the right guesses about the right issues is a crucial
consideration- At any point in the theorizing process intuition can serve
as the triggering mechanism that uncovers important and previously hidden
patterns. New hypotheses are often generated on the basis of sensitizing
hunches. One of the important keys in developing comparative political
analysis is for the rigorous logicians and methodological technicists
to make room for this more artistic component of science. At the same
time, intuitive insights acquire more meaning when they spring up within
a well organized context of theory construction. Insights, no matter how
profoupd, must be harnessed and ordered if they are to yield fruitful
Even more dramatically illustrative of this point is the role that chance
plays in scientific investigation. Chance is a significant variable and
deserves recognition but its positive effects occur most readily in the
correct setting. One scientist writes that "probably the majority
of discoveries in biology and medicine have been come upon unexpectedly,
or at least had an element of chance in them, especially the most important
and revolutionary ones."94 Two other scientists point out, however,
that "chance favors the prepared mind. The great chance discoveries
in science were made by people who distinguished them-Selves by other
work as well. It is only the master of his subject who can turn to his
advantage the irrational and the unsuspected."95 The theoretical
processes of problemation and discovery, in short, require a climate of
Theory building requires a special kind of imagination, sensitivity, and
creativity. In a limited sense, this can be stimulated by such theoretical
exercises as classifying, typing, and modeling- It can also be partially
developed by a broad knowledge of and deep insight into world history.
"James Conant, Science and Common Sense (New Haven: Yale University
Press. 1951), P. 48.
'"Beveridge, The Art of Scientific Investigation, p. 31. Perhaps
no better example is provided of the role that chance and intuition play
in scientific discovery than that of the process by which the genetic
code was broken. This is vividly described in intensely personal terms
in James D. Watson, The Double Helix: A Personal Account of the Discovery
offhe Structure of DNA (New York: Atheneum, 1968).
"Gerald Hollon and Duane H. D. Roller, Foundations of Modem Physical
Science (Reading, Mass.: Addison-Wesley. 1958), p. 249.
And finally, the kind of imagination required to sensitize one to fundamental
problems can be encouraged by an ability to think in terms of continual
comparison. In the end, however, there is as yet no definite information
explaining why some scholars operate well and why others are uncomfortable
and unproductive in the context of discovery.
Systematic observation is the next step of the theoretician. Included
within this procedure are the development of conceptual frameworks, taxonomies,
typologies, and models. Data is collected and classified in accord with
"the scientific activity of description." 9a As there can be
no explanation without description, this procedure assumes great importance.
Even areas in the physical sciences including certain branches of biology
are still primarily involved in the observation and classification stages
of theory building. Careful and patient observation provides the basis
for generalization which in turn yields explanation and understanding.
When observation suggests certain uniformities and regularities, one may
formulate an hypothesis or tentative generalization. This generalization
is then subjected to continual testing by recourse to renewed and expanded
observational procedures. No theory is ever finally and irrevocably confirmed,
but the longer one survives rigorous testing the more powerful it becomes
as an explanatory device.
Embedded within this brief discussion is implicit reference to a reciprocal
inductive-deductive interaction. The contention that theory is essentially
a deductive operation simplifies and distorts the process considerably.
The idea that generalizations are somehow spun deduc-tively and then tested
misrepresents a situation which "is much more fluid than this and
undoubtedly always involves an inductive effort, r7 Theories are formulated
on the basis of observation, experience, and evidence. Often generalizations
thus formed become the basis for deduced conclusions. There is a constant
movement back and forth between research and hypothesis, observation and
generalization, fact and theory. Hempel points out that the two processes
are "inseparably linked," and Eulau stresses their "mutual
interdependence."88 Theories must be continually revised and modified
in accord with newly observed facts and empirical evidence. At the same
time, facts acquire
special meaning when examined
in light of generalizations suggesting new relations and interpretations.
* -me gap that divides those immersed in empirical research from those
who spin high level formal schemes is closely related to the inductive-deductive
question. In one imaginative effort, two scholars attempt to bridge the
gap by calling for "grounded theory." They argue that theory
is best generated by constructing it out of the data of social research."
Another important study addresses the issue by attempting to distinguish
analytically between "general theory" and "auxiliary theory."'00
Implicit in all these attempts is the understanding that theory construction
is a highly integrated process in which the various intellectual operations
overlap and interlock with one another (see figure 2). Constant movement
back and forth also occurs as one refines and reconsiders. Basic conceptualization,
for example, is subject to change as the procedures of classification,
generalization, and testing are engaged in. At any point in the process,
new insights and intuitive flashes can suddenly emerge and fundamentally
alter the investigation underway. The process of theory construction involves
a pendulumlike movement between the specific and the general, the empirical
and the formal, the climate of discovery and the exercise of testing.
The final step in the theory building process concerns the application
of theory. By discovering patterns and explaining processes, theory assists
in prediction. This, in turn, permits control. Theory contributes general
guidelines upon which action and policy can be based. In the social sciences,
the preoccupation with the construction of theories only for the sake
of elegance is a luxury that we cannot afford. In a world where man limps
through the darkness of uncertainty lit only by the flames of crisis,
general beacons of enlightenment that serve to guide his steps are necessary.
In the field of comparative politics, policy relevant theories of authority
and change, for example, are among the most pressing concerns of scholars.
The Role of Theory o
The advantages that theory and the theorizing process offer to the study
of comparative politics are many. Eight major contributions flow from
the exercise of theory building. First, theory contributes greatly in
FIGURE 2 SCIENTIFIC THEORY BUILDING
selection/ collection, ordering, and storing of data. It exists as a thAad
that runs through masses of data drawing only the relevant facts to its
path. Theory provides an important instrument of discrimination for selecting
and collecting the relevant from the irrelevant in the sea of sociopolitical
experience. This prevents the never ending process of compulsive fact
accumulation that results in impressively huge but depressingly meaningless
filing silos of data. Refined and rigorously developed theories also stand
as repositories of ordered fact. They have been repeatedly measured against
relevant data and these data then stand in at least this systematized
arrangement. In this role, theory functions preeminently as a filtering
Second, theory building necessitates conceptual and methodological clarity.
The theoretician is forced to rethink many of his basic concepts, assumplions.and
hypotheses. The process of reworking and recasting verbal theories is
essential for infusing rigor and precision into the theorizing process.
This enterprise involves "clarifying concepts, eliminating or consolidating
variables, translating existing verbal theories
into common languages,
searching the literature for propositions, and looking for implicit assumptions
connecting the major propositions in important theoretical works."101
This process of reexamination and clarification not only contributes to
sounder theory, but also heightens communication and understanding between
various theorists working with similar concepts and addressing similar
The third advantage of theory is its relational stress which marks its
special significance to comparative analysis. Theory always draws relations,
builds connections, and states linkages. Facts and events are intertwined
and interrelated. Comparative politics is fundamentally a study that relates
patterns and processes that occur within two or more political contexts.
The theorizing process is the most profound and systematic way of proceeding
in this kind of study. Generalizations are essentially multicontextual,
and their construction is what has come to transform configurative -descriptive
political study into comparative political study.
Closely related to this third function of theory is the ability of theory
to transcend particular time and spatial limitations. Analysis is not
confined to describing and explaining one event in and of itself as it
has occurred at one time and in one place. Instead, this event is viewed
primarily from a very broad perspective that treats the particular happening
as only one manifestation of something much larger. This question of perspective
is important both for the confirmation of existing hypotheses and for
the development of new hypotheses. By presenting a broader picture of
relations and linkages, theory can utilize breadth to enhance depth. Theorizing
is one way of transcending particularism and narrowness which stultify
vision and creativity.
A fifth advantage of theory is that it is a form of explanation. In indicating
that particular cases fall within certain general principles and in linking
and ordering events, the theorizing process is confronting and answering
key "why" questions. Whereas description is primarily concerned
with the "what," the "when," and the "where,"
explanation concentrates upon the "why." Theory is an explanatory
tool and as such it is a crucial aid towards building and furthering understanding.
Following from these last three functions is an advantage that enables
man to attempt to foresee and forecast. Theory leads to prediction and
enhances the ability to see ahead. By relating generalizations, theories
discover and highlight patterns and trends, and it is on the basis of
that man predicts. Accurate
prediction in social and political life is very difficult but it can be
furthered by placing more reliance upon sound theory building and often
tested theoretical principles. Although prediction frequently rests on
unanalyzed trends, such prediction is inevitably risky and is rarely reliable
for more than very short periods of time and closely adjacent cases. Prediction
over any longer term must rest on some kind of theory, for only theory
provides an explanation of why we can expect the future to be as we predicted.
The seventh general contribution of theory is best described by Kurt Lewin,
who once said that nothing is so practical as a good theory. Theory assists
in policy-making and decision-making processes at all levels and in alt
systems. As a simplifying and unifying intellectual tool, it restricts
the range of choices and suggests new alternatives for action. This function
has been referred to above as one important form of theory application.
Finally, theory and the entire process of theory building act as a provocative
force for the selection of new problems and the stimulation of new research.
Discovering, clarifying, classifying, relating, explaining, and testing
breed an intellectual dynamism and curiosity that continually encourage
new explorations. In the field of comparative politics, for example, the
leading theoreticians have been those who are constantly settling upon
new problems and generating new hypotheses. This is one less recognized
reason why names like Almond, Huntington, Riggs, Eckstein, and Apter have
so dominated the recent study of comparative politics.
If it performs any or all of these functions, then theory has been to
some degree successful. Any serious attempt to engage systematically in
the theorizing process is bound to be a positive experience. It intro-
t duces a sensitivity to the importance of scientific procedure and rigor
in the process of comparative analysis and explanation. The future of
comparative political study rests upon effective theory building and the
continuing development of a body of comparative political theory -
The Goals of This Study
A proliferation of varied approaches to the study of comparative politics
has resulted in numerous and differing orientations to the construction
of theory. These approaches have been traditional and modern, highly refined
and hardly developed, explicit and implicit, more popular and less popular.
In some cases, approaches have overlapped with one another; in other instances,
they have been quite exclusive and separate. Often, concepts basic to
one approach are utilized in another approach,
sometimes with the same
meanings, sometimes with quite different meanings. Various approaches
have offered differing strengths to comparative political analysis and
theory building. In terms of relevance, rieor, and researchability, for
example, there are marked differences in the kinds of contributions that
flow from each theoretical approach.
Perhaps the greatest difficulty in analyzing the role of a particular
approach in the study of comparative politics is the absence of any clear
understanding of the nature of any one approach in relation to any other
approach. This intellectual fuzziness has had an impeding and retarding
effect upon the study of comparative politics for several reasons. First,
it has had a negative impact upon theory building. A confusing jumble
of approaches prohibits one from discerning which approach contributes
most to which stage in the theorizing process. The lack of clear analytic
distinctions here also fosters conceptual and procedural confusion which
is destructive to rigorous theory building. Second, an unclear view of
existing approaches tends to blind a scholar to his own particular theoretical
orientations. He lacks a clear awareness of what approach or approaches
he utilizes under what circumstances and in what form. Nor is he able
to distinguish lucidly the general orientations of other observers and
scholars. This kind of confusion is deepened by the fact that many scholars
adopt a changing and eclectic hodgepodge of approaches that defy identification
and clarification. The lack of awareness of basic orientating perspectives
dulls the tools of comparative political analysis from the very beginning.
Finally, a clouded view of theoretical approaches prevents one from making
discriminating evaluations of research results in comparative politics-
Unless the projects themselves present an explicit discussion of the approaches
and procedures that led to the relevant conclusions, one is often unable
to judge adequately the results put forth.
Chapters 3 through 7 in this study attempt to clarify, distinguish, relate,
and evaluate five major theoretical approaches to the study of comparative
politics: the political culture approach, the group approach, the elite
approach, the class approach, the.functional and systems approaches. The
presentation of these approaches will not rely upon the works of any particular
practitioner, but rather will represent a distillation of relevant work
through time. The major conceptual equipment of each approach will be
carefully investigated and judged in terms of the tasks at hand. All approaches
will be evaluated in terms of their capacity to confront, elucidate, and
explain certain problems. The fundamental issue against which the approaches
will be measured is the subject of the next chapter.