International Journal of Public Opmion Research Vol. 7 No. 4 0954-2891/95 $3.00
THE RENAISSANCE OF A RUBRIC:
POLITICAL CULTURE AS CONCEPT
AND THEORY
William M. Reisinger
ABSTRACT
In recent years, several prominent scholars have attempted to reinvigorate politicalculture
theory and defend it from the challenge that rational-choice theory poses. Yet
the primary threat to the 'renaissance' of political-culture theory comes not from
rational-choice theory but from political culture's continued weakness as concept and
theory. Even after the theoretical and empirical defenses provided in recent years,
political culture remains poorly explicated along seven distinct dimensions: (1) how to
define the concept; (2) how to disentangle subcultures (for example an elite political
culture) from a society's overall political culture; (3) how to integrate the many
individual-level orientations of which the concept is composed; (4) how to create a
societal-level variable from individual-level components; (5) if the foregoing have been
resolved, how to measure the concept; (6) how to derive hypotheses about individual
political behavior from the subjective orientations under study; and (7) how political
culture interacts with institutions and other attributes of a polity to produce a propensity
for certain types of political outcomes. With these tasks left uncompleted, political
culture remains no more than a rubric under which different authors focus on different
individual orientations, employ different measures and different methods of aggregating
the orientations, then test different propositions about the links between those individual
orientations and politics.
In the late 1980s, the defenders of political culture's role in comparative
politics made a determined counter-attack against the advancing battalions
of rational-choice theory. The counter-attack employed both theoretical and
empirical weaponry. Wildavsky (1987) laid down ground cover by sniping
at the exogeneity of preferences in rational-choice theory. 'But where do
people's preferences come from?' was Wildavsky's question. His answer:
I would like to thank James L. Gibson, Vicki L. Hesli, Kristen Hill Maher, Arthur H. Miller, Thomas
Remington and Mitchell Seligson for their insightful critiques of draft versions of this artide. This article
was written while at the Obermann Center for Advanced Studies of the University of Iowa. I thank the
Center's staff and its director, Dr. Jay Semel, for their help.
© World Association for Public Opinion Research IQ95
THE RENAISSANCE OF A RUBRIC 329
'culture.' Eckstein (1988, p. 789) claimed that cultural theory is 'one of two
still viable general approaches to political theory and explanation' (the other
being the rational-choice perspective) and that deciding between the two is
a top item on the agenda of political science. Eckstein then provided more
detail on the theoretical assumptions underlying a 'culturalist' approach to
politics. Neither of these authors attempted to support the culturalist
approach empirically. Inglehart's (1988) argument for a 'renaissance' of
political culture, by contrast, dealt little with the theoretical underpinnings
of the approach, providing instead empirical support for its utility in
understanding differences among Western democracies. Coming close together
as they did, these three articles placed the debate about mass orientations
and politics back onto the center stage it had vacated in the 1970s (for
histories, see Barry 1970, Almond 1990, Brint 1991) and sparked a new
wave of analysis of political cultures and their impacts (e.g. Diamond 1993b).
Gabriel Almond, one of the founders of the approach, recently (1993)
referred to these developments as a 'return to culture.'
Although citizen orientations do deserve attention from those engaged in
the comparative study of politics and although survey research should
continue to be an important 'technology' for measuring and analyzing citizen
outlooks, conceptual and theoretical difficulties continue to confront the
'renaissance of or the 'return to' political culture. Those who rode out in
the late 1980s to save political-culture theory fought bravely and did score
points against simplistic rational-choice propositions. Certainly, given the
many disagreements about basic postulates that remain among rational-choice
modellers, these authors make a convincing case that cross-national differences
in mass political orientations can be ignored only at great peril. Yet in
promoting political culture as a concept with attendant theory, they failed
to provide a clear set of tools for those who would like to measure and
comparatively analyze the mass-level political cultures of different societies.
The main threat to political culture, in other words, comes not from a rival
approach but from behind its defenders' own lines. For the culturalist
approach to aspire to being a viable general approach to political theory
and explanation, its proponents must shore up both the concept of political
culture and the theory that links political culture to political processes. I
seek, therefore, to specify seven intellectual tasks, the completion of which
would shore up political culture as a concept and the cultural approach as
a means to acquire knowledge. With these tasks left uncompleted, political
culture remains no more than a rubric under which different authors focus
on different individual orientations, employ different measures and different
methods of aggregating the orientations, then test different propositions
about the links between those individual orientations and politics.
330 INTERNATIONAL JOURNAL OF PUBLIC OPINION RESEARCH
POLITICAL CULTURE'S CRITICS AND DEFENDERS
Most of political culture's recent defenders (Wildavsky 1987 is a noteworthy
exception: Thompson et al. 1990) fall within the individual-oriented, surveybased
approach that excludes behavior and institutions from the definition of
culture—the approach that grew largely from the work of Gabriel Almond.1
Throughout this article, I direct my attention to this mainstream understanding.
The research that this approach generated has been influential. Few textbooks
on politics or comparative politics lack a chapter or section discussing it. Yet
several problems with the Almondian conception of political culture and its
usage in the major early works gained prominence in the 1970s and early 1980s
(Barry 1970, Pateman 1971, Almond and Verba 1980, Patrick 1984). In one of
the best-known critiques, Barry (1970, p. 88) focused on three difficulties: in
definition, in formulation of hypotheses and in providing evidence of causality.2
Some saw the arguments made in the pathbreaking study, Almond and Verba's
The Civic Culture (1963), as implying that there was only a single path of
political development—the Anglo-American one—and as thus cloaking an
ideological argument in quasi-theoretical language (e.g. Brown 1979, p. 3).
As these criticisms emerged, the argument that each society has a 'political
culture' shaping its politics began to vie with theories of rational choice to serve
as the dominant conceptual framework for the comparative study of politics
(see, for example, Rogowski 1974, Almond and Verba 1980, Eckstein 1988).
Political culturalists claim that political science ought to benefit from psychology,
sociology and anthropology as well as (or more than) from economics. For
them, two societies sharing all the same institutional and structural conditions
could nevertheless differ in their political dynamics if their political cultures
differed significantly. For instance, the Spanish transition to democracy in the
1
It is important to note that a 'political culture,' as that term is used by its proponents within mainstream
political science and as I use it here, corresponds only imperfectly to what anthropologists would call a
culture (though they have a multiplicity of definitions of the term). 'Culture,' for anthropologists, usually
incorporates behavior patterns as well as subjective orientations. Anthropologists are less likely to separate
culture from structure or from institutions than are the political scientists employing the Almondian sense
of political culture. Though a few political scientists argue in favor of incorporating political behavior into
our understandings of political culture, most agree that using political culture to explain political behavior
is a primary goal, and thus they must not conflate the two. Also, of course, the 'culture' in political culture
does not refer to popular culture (literature, film, music) even though the latter influences politics and
political attitudes.
1
Almond (1990) includes in his list of the four major attacks on political-culture theory the claim that
political-culture theory is determinist in attributing causality from culture to behavior or institutions. He
mentions Barry and Pjtcman as theorists who have raised this issue. Almond dismisses this as a straw-man
polemic and states specifically that political culture is generally agreed by its users to both influence and be
influenced by structure and performance. 'The causal arrow goes both ways.' Almond is correct to discount
that charge as a valid critique of the political culture approach. Yet it is ironic that Almond neglects entirely
a damaging and less easily dismissed critique (the one that actually seems to have been Barry's point): merely
to present correlations between subjective orientations and a dependent variable (such as stable democracy)
says nothing al all about causality.
THE RENAISSANCE OF A RUBRIC 331
late 1970s and early 1980s is likely to have only partial applicability to the
transition in, say, Poland because of those countries' cultural differences.
Inglehart (1988, p. 1228) threw down the gauntlet colorfully, 'Both social analysis
and social policy would be much simpler if people from different societies were
interchangeable robots. But a large body of evidence indicates they are not.
The peoples of given societies tend to be characterized by durable cultural
attributes that sometimes have major political and economic consequences.'
Moreover, theorists of political culture assert that cultural differences across
societies are neither idiosyncratic nor unknowable. They argue that they have
derived theoretical expectations as to which specific aspects of any society's
political culture influence its political dynamics. Thus, certain key cultural
differences specified ahead of time can be employed to explain cross-society
differences rather than making the task of cross-society research seem futile.
In part, the sharp contrast between the rational-choice approach and the
political culture approach (a la Eckstein) has been overstated. Barry (1970,
pp. 180-2) points out in his critical comparison of the two approaches that one
can posit culturally derived values as underlying a rational actor's preference
ranking (see also Elkins and Simeon 1979, Lane 1992). Still, the differing
emphases mentioned above carry theoretical import. For example, Eckstein
argues that actors must have general predispositions to act in certain ways in
certain situations and that, without such dispositions, their actions would be
'erratic: patternless, anomic' A rational-choice theorist would likely chuckle at
such melodrama and reply that human beings are quite capable of deciding
how to act in response to the nature of the situation itself. Eckstein (1988,
pp. 790 and 792), in turn, provides a reply to the effect that decision costs
become overwhelming if people must calculate their response to every situation.
In the 1970s, political culture experienced a decline in popularity. Still, the
original rationale for the study of political cultures remained: how can scholars
satisfactorily explain cross-national differences in politics without attending to
the subjective orientations of the societies' members? By the late 1980s, a
reaction to the criticisms had set in: perhaps political culture could be defended
from its attackers and its popularity restored? As noted, significant efforts came
from Eckstein, Inglehart and Almond. Eckstein (1988) focuses on spelling out
the social-psychological assumptions on which an understanding of political
culture can rest, assumptions which can accommodate certain types of cultural
change. Inglehart (1988 and 1990) analyzes time series data for each of nine or
more Western democracies—the type of dataset not available to students of
political culture in the 1960s—and shows significant continuities over time in
countries' differing mean scores for life satisfaction (which he views as an
operationalization of 'diffuse support for democratic institutions'), interpersonal
trust and non-support for revolutionary change. He defines a democratic political
332 INTERNATIONAL JOURNAL OF PUBLIC OPINION RESEARCH
culture with these orientations. Almond (1990, pp. 144-5) raises and dismisses
four critiques of political-culture theory: the argument that it is deterministic,
the claim that class structure determines attitudes, the argument for adding
behavior to the definition, and the argument that rational behavior can explain
cross-cultural differences. He also defends the political culture approach, first,
by citing its ancient antecedents and, second, by showing the breadth of
sophisticated empirical research to which political culture arguments have given
rise in the last few decades.
WHY THE COUNTER-ATTACK FELL SHORT
These articles, and the re-thinking of political culture they present, are important
contributions to the discipline. They provide answers to several important
critiques. They are most convincing in underscoring the importance to comparative
politics of cross-national differences in the patterns of citi2en outlooks.
They nicely reiterate the theoretical rationale for measuring and analyzing
such individual outlooks as tolerance, willingness to compromise, self-efficacy,
propensity to participate and interpersonal trust.
It is a short and convenient step from there to placing such values under the
rubric of 'political culture.' A rubric provides a useful title for referring to a
class of objects without the assumption that the objects together constitute a
unified whole. For example, Inglehart's (1988, p. 1218; 1990, p. 44) use of
interpersonal trust, life satisfaction and support for revolutionary change as
three variables grouped together under the label of 'civic culture' seems to be
using culture as a rubric. The important variables are the three individual
values. Inglehart does not deal with how they might together form a single
concept called '[civic] culture.' (Neither Inglehart's 1988 article nor the slightly
revised chapter in his 1990 book clarifies how the three variables are analyzed
in the LISREL model, whether as three separate variables or as a single index
created somehow out of the three variables). Of course, Inglehart's work makes
a genuine contribution by advancing the debate about the nature and impact
of comparative patterns of individual political values. The very significance of
his findings make it harder to see the need for claims about there being a larger,
more complex 'culture' at work.
Nevertheless, few advocates of the study of political culture accept it as a
rubric that provides simplicity of wording when referring to the cross-national
study of values. The recent defenders assert that political culture can serve as
a concept in comparative empirical, usually survey-based, research. The raisons
d'etre of employing the term political culture require this. First, its purpose is
to be an attribute of an entire society: each unit in a comparative study has one
and only one political culture, however depicted. Second, this attribute of a
THE RENAISSANCE OF A RUBRIC 333
given society or social grouping—its culture—must be comparable to the
cultures of other societies or social groupings. (This could mean comparing the
culture of one society to that of another at the same time or comparing a
society's culture at one time to its culture at a different time). The results of
the comparisons—cross-society similarities and differences in cultures, whether
explicit or implicit—must lend insight into political processes or outcomes.
Defenders of the culturalist approach to comparative politics assert that political
culture already can serve these purposes. They assert, in other words, that it
is possible to make and defend statements of the following type: 'Russia's
political culture is of the X type. Because this makes it different from Spain's
political culture, which is of the Y type, the otherwise very similar institutions
and level of economic development will produce different political outcomes in
Russia than in Spain.' (Alternatively, the statement might be that 'Russia's
political culture falls at this point along the spectrum of political cultures, while
Spain's....') It is in sustaining these more ambitious claims that the recent
defenders fail. The need remains to strengthen the concept itself and the
scholarly approach of which it is the centerpiece.
What would characterize an ideal scholarly study in which political culture
was more than a rubric for the study of values? It would, I argue, meet seven
straightforward—but in practice extremely challenging—requirements. The
first five requirements listed below focus on political culture as an empirically
grounded concept. Meeting these five requirement would solve the three
problems that Sartori (1984, pp. 28-35) argues a definition must solve: the
border problem (to be settled by denotative definitions), the membership
problem (precising definitions) and the measurability problem (operational
definitions which generally hinge on the search for valid indicators). The final
two requirements I discuss deal with the concept's place in any theory of
politics.
An ideal culturalist study would, first, clearly and specifically define political
culture (using terms generalizable to many settings). Second, the author would
specify the population to which the concept applies (masses vs. activists vs.
elites, for example) and how sub-cultures relate conceptually to the overall
societal culture. Third, the author would indicate and justify how, if the concept
were defined as having multiple components, he or she had combined the
various components into a single concept. Fourth, the author would detail the
means by which he or she employed individual-level attributes, such as political
values, to construct something that applies to a society, namely its political
culture. Fifth, this exemplary piece would explain which research techniques
are appropriate for measuring a society's political culture as the author defines
it and why. Sixth, the author would make explicit the hypothesized link between
culture and individual behavior. Seventh and finally, if the author intends to
334 INTERNATIONAL JOURNAL OF PUBLIC OPINION RESEARCH
use the concept to explain some aspect of political outcomes—such as whether
democracy takes root—he or she would explicate his or her assumptions about
how a society's political culture relates to other politically relevant features of
the society, including the development and functioning of political institutions.
Ideally, this contribution, to have the postulated impact, should find its approach,
especially its definition of the term political culture, widely accepted so that
research could thereafter cumulate.
DEFINITIONS
One clear difficulty is that political culture has almost as many definitions as
audiors who employ it (see Patrick 1984, Lane 1992, p. 363). Almond and Verba
(1963, p. 13) defined a society's political culture as the set of 'attitudes toward
the political system and its various parts, and attitudes toward the role of the
self in the system' held by its members. Their book and its definition were
widely influential, yet, strangely, not even those accepting this general definitional
approach wanted to adopt precisely the same definition. This included one of
The Civic Culture's co-authors, Sydney Verba! In a later essay, Verba (1965,
p. 513) defined political culture as 'the system of empirical beliefs, expressive
symbols, and values which define the situation in which political actions takes
place.' Verba's definition illustrates one of the problems that scholars found
early on with the Almond and Verba definition: the term 'attitudes' is too
limiting for characterizing the components of a political culture. Yet those
providing definitions substituted alternative words—beliefs, values, orientations,
expectations, symbols, perceptions, knowledge, affect, norms—too freely, often
without considering the implications for the concept.3
This profusion of definitions does not worry all scholars. Certainly much of
the variety of definitions can be explained as a discussion among scholars about
how best to define the term to give it general applicability. So, some see the
lack of an agreed-upon definition as a sign of creative application. Yet the ability
to modify the meaning of the concept according to one's research parameters
poses problems as well (on this point, see also Lane 1992). It amounts to what
Sartori (1984, p. 35) has called 'collective ambiguity' about a concept, which
can reach the point of 'destroying a discipline as a cumulative fabric of
knowledge.' It also, ipso facto, reduces the accumulation of findings by making
' Despite the different ways that culturalists formulate the term, they tend to agree that it ought to refer
to something more deep-seated than public opinion. The latter typically indicates orientations toward
contemporary issues, actors and events in civic life. Political culture, by contrast, is reserved for orientations
toward the fundamental institutions and processes of the political system (Rosenbaum 1972, ch. 5). As
WHdavsky (1987) notes, this is one strength of the culturalist approach, because it makes the formation of
preferences an object of study, which rational-choice theory does not. It does, however, raise the stakes
involved in defining and measuring political culture.
THE RENAISSANCE OF A RUBRIC 335
it difficult to contrast results from different research projects. Thus, any
'renaissance' of political culture should rest on more widespread definitional
agreement than found among the concept's pioneers.
How do those seeking to rehabilitate political culture deal with definitional
issues? Inglehart (1988), intent on examining empirical evidence, provides no
explicit definition. He implies that he is following the Almond and Verba
understanding of the concept. Yet he is quite casual about it, for he actually
departs from their definition when he describes the political-culture approach
as dealing with 'certain supportive habits and attitudes among the general public'
(Inglehart 1988, p. 1204, italics added). If by 'habits,' Inglehart refers to forms
of behavior, habits would be, strictly speaking, excluded from Almond and
Verba's definition. In his 1990 book-length study (p. 18), Inglehart defines
culture as 'a system of attitudes, values and knowledge that is widely shared
within a society and transmitted from generation to generation.'
Eckstein, despite arguing that the major challenge for culturalists is conceptualization
and theory development, never actually defines political culture.
He is laudably clear on denning the components of a political culture:
Orientations are not 'attitudes': the latter are specific, the former general, dispositions.
Attitudes themselves derive from and express orientations; though attitudes may,
through their patterning, help us to find orientations. If orientations frequently occur
in collectivities they may be called 'culture themes,' ... It is conventional to regard
orientations as having diree components: cognitive elements that, so to speak, decode
experience (give it meaning); affective elements that invest cognition with feelings that
'move' actors to act; and evaluative elements that provide goals towards which actors
are moved to act (1988, pp. 790-1, italics in original).
The only evidence that Eckstein presents about how orientations or predispositions
relate to political culture comes when he mentions 'culture themes'
and when he summarizes (p. 792) by saying that 'the patterns of such predispositions
vary from society to society.' Nor is it clear what to make of such
things as 'frequently occurring orientations' or 'patterns.'
Almond (1990, pp. 143-4) s u r n s U
P w n a t
political culturalists (of this school)
accept about the term:
Political culture theory defines political culture in this four-fold way: (1) It consists of
the set of subjective orientations to politics in a national population or subset of a
national population. (2) It has cognitive, affective and evaluative components; it includes
knowledge and beliefs about political reality, feelings with respect to politics, and
commitments to political values. (3) The content of political culture is the result
of childhood socialization, education, media exposure, and adult experiences with
governmental, social and economic performance. (4) Political culture affects political
and governmental structure and performance—constrains it, but surely does not
336 INTERNATIONAL JOURNAL OF PUBLIC OPINION RESEARCH
determine it. The causal arrows between culture and structure and performance go
both ways.
Almond's definition (points 1 and 2) is extremely broad. One might therefore
imagine all culturalists of this school accepting Almond's wording as a single
working definition to improve comparability. Yet, a definition so general actually
increases the challenges of producing a workable empirical conceptualization.
It will likely fall short of Sartori's (1984, p. 56) standard that a concept's
empirical definition contain enough characteristics to identify the referents and
their boundaries and that it include no non-essential property among the
necessary, defining properties. Notice also that Almond and others use the term
'set' of orientations while Eckstein mentions the 'patterns.' The former are
denning a society's culture as all the orientations taken somehow together as a
whole. Eckstein's wording emphasizes not the whole but relationships among
the parts. Verba's (1965, p. 513) and Inglehart's (1990, p. 18) 'the system of
adds its own wrinkle, though it would seem to be closer to Eckstein's 'patterns'
than to 'sets.' These differences are not trivial; they have important ramifications
for the third challenge discussed below as well as for the conduct of research.
(The mathematics of'sets' and that of'systems' are quite distinct, for example).
So, while common adoption of Almond's general depiction might be helpful,
further specification of an empirical definition remains a task facing political
culture theorists.
WHOSE CULTURE IS BEING STUDIED?
The concept of political culture must be clear about the populations to which
its definition allows it to apply. 'Students of culture who wish to refine its
utility as an explanatory concept must develop precise means of identifying the
culture-bearing unit in different situations' (Elkins and Simeon 1979, p. 129).
For example, an important element of recent debates over processes of democratization
has been whether elites alone must have democratic values or whether
the mass public must also have (or come to have) such values (for a review, see
Diamond 1993a, pp. 2-7). Defenders of political culture are quite willing to
admit that sub-cultures, including the elite sub-culture, are important in most
countries. They point to the potential increase in theoretical leverage to be
gained by exploring how political subcultures relate to one another. The ability
to examine such subcultural clashes is an advantage of political-culture theory.
(For recent defenses of the benefits of analyzing the political culture of
subnational groups, see Wildavsky (1987) and Lane (1992). Note, however, that
both Wildavsky and Lane are explicit in rejecting the mainstream focus on
individual orientations in favor of using the group as a unit of analysis).
THE RENAISSANCE OF A RUBRIC 337
Yet acknowledging that elites differ from non-elites in attitudes, in the
sophistication of their world view, and in much else does not by itself justify
distinguishing between elite and mass political cultures. The elite must be
culturally distinct from the rest of society. And, if such is the case, then it
undercuts the existence of a political culture characterizing the larger society
of which they are part. If, for example, one argues that members of the
Communist Party of the Soviet Union shared a distinct political culture, that
suggests that there was no single Soviet political culture but at least two.
Similarly, to contrast the Ukrainian subculture and the Russian subculture
within the Ukrainian state is worthwhile, but, if one demonstrates the existence
of two different cultures, additional theoretical steps will be needed to characterize
the political culture of the entire Ukrainian society. To the extent that
the object of study is some form of a culture, it must refer to an entire social
grouping. If one argues that ethnically Lithuanian and ethnically Polish citizens
of Lithuania have distinct political cultures, one must maintain that separation
throughout the analysis. Yet for many research questions, such as democratization,
the nation-state is the proper unit of analysis. How to take the
reality of multiple political cultures within a political state into account in
cross-national studies remains an unanswered question, one which would depend
on, among other things, the definition of culture that had been adopted.
AGGREGATING ORIENTATIONS FOR EACH INDIVIDUAL
Any time one employs a concept as a single variable with multiple components
(orientations, values or whatever), one must clarify and defend the means by
which the components were merged. In many cases, this is a straightforward
methodological problem to resolve: an indexing task. The scholar would specify
and defend particular weights and additive techniques. Identifying and agreeing
upon techniques for aggregating separate orientations would facilitate cumulative
discussions of the political influence of different political cultures.
In the case of an individual's subjective orientations, however, indexing may
be inappropriate. One school of political psychology has been at work examining
how different types of subjective orientations inter-relate. This work is generally
referred to as the study of schemata (Conover and Feldman 1984, Hamill et al.
1985, Miller et al. 1986, Kuklinski et al. 1991, Lodge et al. 1991). The essence
of this literature is that one cannot understand the impact of individuals'
attitudes or evaluations unless one knows the cognitive processes by which the
brain relates different types of information to one another. If this school carries
weight, it certainly raises yet another hurdle for the political culture researcher:
to provide a theoretical rationale for the cognitive relations among the subjective
orientations under study. If schema are at work in individuals, then atheoretical
338 INTERNATIONAL JOURNAL OF PUBLIC OPINION RESEARCH
index building would confuse more than it would enlighten. The conceptual
movement from individual to society is unlikely to be successful if one cannot
correctly move from type of cognition to individual. Wildavsky's work (1987,
Thompson et al. 1990) takes an approach that avoids the problem of aggregating
components by arguing that 'culture' is in fact the unifier, the shaper of
cognitions. Culture is not comprised of attitudes, orientations, etc. but produces
them. This approach may have difficulty finding common acceptance, however,
since it makes secondary the analysis of survey data on levels of trust, life
satisfaction and other commonly studied orientations.
The challenge of weaving many different psychological orientations together
is heightened by the nature of the definitions of political culture. In line with
their intent to use political culture as a systemic variable, scholars have defined
it in comprehensive terms, e.g. as l
all the important ways in which a person is
oriented' (Rosenbaum 1972, p. 4, italics added). Measuring such a complex
construct for a single individual seems daunting. In practice, of course, empirical
analyses rarely incorporate more than a few aspects of any definition of
political culture and therefore can shed only partial light on this multifaceted
phenomenon.
Almond and Powell (1992, pp. 39-43) have suggested a modification of
Almond's understanding of political culture that recognizes the need to aggregate
the multiple components included in his definition of the term. They argue
that each individual's orientations fall into three categories: orientations toward
the political system, the political process and the policy outputs of the system.
Instead of discussing how one could form these into a general political culture,
they assert that each society therefore has a system culture, a process culture
and a policy culture. This tack naturally leaves the question of aggregation or
indexing open: determining how various orientations combine to produce a
single system culture, process culture or policy culture poses the same challenge
as figuring out how to measure a society's political culture.
AGGREGATING INDIVIDUALS' ORIENTATIONS ACROSS A SOCIETY
Repeatedly when attempting to determine the meaning and scope of political
culture, difficulties crop up with the fourth aspect of a satisfactory conceptualization:
constructing a societal-level or group-level variable out of individual
attributes. The emphasis that political culture is more than numerous
distributions of individual orientations was particularly important to those
developing the term. They saw political culture as a way to overcome the
excessive individualism of behavioral political analysis by Unking individual
attributes to the 'total political system' (Pye 1965, p. 9). In Pye's (1965, p. 7)
words: 'The notion of political culture assumes that the attitudes, sentiments
THE RENAISSANCE OF A RUBRIC 339
and cognitions that inform and govern political behavior in any society are not
just random congeries but represent coherent patterns which fit together and
are mutually reinforcing.' Dittmer (1983, p. 23) stresses that '[P]olitical culture
is not just a haphazard collection of beliefs and values, but a system in which
the various parts must bear an integral relationship with one another.' Almost
every standard treatment of the topic makes references to political culture as a
whole. In the words of Elkins and Simeon (1979, p. 129): 'Political culture is
the property of a collectivity—nation, region, class, ethnic community, formal
organization, party or whatever. Individuals have beliefs, values and attitudes
but they do not have cultures.'
The proponents of political culture, then, clearly intended to create a unified
concept. That is, when comparing the politics of various societies, culturalists
seek to explain similarities and differences in part by employing a single variable
entitled 'political culture,' the values of which differ across those societies.
However, in setting out to conceive and then measure such a societal variable,
political culturalists examine attributes of individual human beings. This is the
great attraction of the approach and the source of a good deal of its richness.
In Verba's (1965, p. 517) words, 'although the political behavior of individuals
and groups is of course affected by acts of government officials, wars, election
campaigns and the like, it is even more affected by the meanings that are
assigned those events by observers. This is to say no more than that people
respond to what they perceive of politics and how they interpret what they
see.' Who could disagree with such a proposition?
Yet how can we, in practice, create a concept that applies to a collectivity of
thousands or millions of people out of hypotheses about individuals? How can
we, for instance, treat a society's political culture as a 'system of control' (Verba
1965, p. 517)? Creating a societal concept out of individual orientations is
logically possible, but it presents daunting definitional and practical problems.
The challenge is to overcome the 'individualist fallacy'—the fallacy of deriving
conclusions about a higher level of aggregation from data on individuals without
a theoretical rationale that links the two levels (Simon 1962, Scheuch 1969,
Webster 1973, Mayer 1989, p. 201, Hannan 1991). This issue was cogently
presented over two decades ago by a leading light, Lucian Pye (1972), yet no
one has met the challenge. Interestingly, none of the recent defenses of a cultural
approach even cite Pye's article. Moreover, as noted above, theorists use different
formulations to characterize the link between individuals and society, including
'the set of,' 'the pattern of,' 'a syndrome of and 'the system of.'
In cross-national empirical work, analysts have too frequently been willing
to characterize an entire society with a single number without providing a
theoretical rationale linking the individual level of analysis to the societal. When
this occurs, political culture becomes hard to distinguish from the nemesis its
340 INTERNATIONAL JOURNAL OF PUBLIC OPINION RESEARCH
proponents originally sought to vanquish: national character. Culturalists (e.g.
Almond and Verba 1963, Pye 1965, Inglehart 1990, p. 17) have derided those
who spoke of a country's national character because they were simplistically
characterizing an entire, complex society with a single label. Yet reaching
cross-national conclusions based solely on a measure of central tendency veers
close to this. Dittmer (1983, p. 10) and Mayer (1989, pp. 181-2) refer to this
as creating a 'modal personality.' Hannan (1991, p. 6) put it as follows: 'unless
the addition operation corresponds to some social phenomenon, the group level
concept has no meaning other than as a kind of "aggregate" psychological
concept.'
The lack of theoretical guides for creating society-wide numbers has implications
for measurement strategies as well. Virtually anything that a researcher
would like to measure as an individual value, attitude or other type of orientation
will vary in the intensity with which individuals hold it. For instance, in seeking
to determine whether a society is characterized by 'greater' inter-personal trust
than other societies, the analyst needs to distinguish between 'more heart-felt'
trust and 'more citizens who fall into the "trusting" category.' When a survey
item can take on only two values, the respondents' intensity of preference is
ignored altogether. The resulting mean reflects a proportion, not an average.
Even when respondents can select from a larger number of categories and
thereby indicate the intensity with which they hold the value, assumptions have
to be made about the meaning of the numbers assigned to the categories in
order to employ the answers as quantitative data (Tufte 1969, Lodge and Tursky
1979, 1981). Two societies with quite different patterns of inter-personal trust
could produce the same mean. Some authors employ both the mean of a variable
and the proportion of respondents falling into one of two collapsed categories
at different times. For example, when Inglehart (1988) examines cross-national
and over-time data on responses about people's satisfaction with their life as a
whole, his first figure presents the proportion who select 'very satisfied,' while
his second figure switches to mean scores.
I raise this point about the difference between a mean and a proportion not
to argue that multiple categories must always be used so that differences in the
distribution of intensities can be determined. It might well be appropriate to
look at how many are or are not trustful, satisfied with their life, intolerant,
efficacious or whatever. The point is that one needs a theoretical rationale for
why a particular way of summarizing a large number of individual responses is
appropriate. Political culture dieorists rarely spell out such a rationale. Inglehart's
strategy of displaying how many fall into the 'very satisfied' category is fine if
one hypothesizes that life satisfaction contributes to stable democracy in a
particular fashion: when individuals feel themselves to be very satisfied, their
politically relevant thoughts and actions are qualitatively different from others
THE RENAISSANCE OF A RUBRIC 341
in the same society who are only partially satisfied or are dissatisfied and,
furthermore, the more such very satisfied people there are in a society, the
more support for democracy will be forthcoming (in a linear fashion). One
might, by contrast, prefer to assume: (a) that life satisfaction has a range of
intensities, perhaps adequately captured by five categories, perhaps requiring
more categories or some form of magnitude scaling, and therefore that the
differences between a partially satisfied citizen and a highly dissatisfied citizen
are as important as the difference between a very satisfied citizen and a partially
satisfied citizen; (b) that the relationship between life satisfaction and stable
democracy is not purely linear but rather that (i) stable democracy only requires
that a majority be somewhat or more satisfied; (ii) stable democracy will be
threatened when more than a small minority feel themselves to be very
dissatisfied even if many others are very satisfied; and (iii) that the number of
satisfied citizens must be weighted by the political clout of different citizens
(education, wealth, efficacy, etc.). A researcher who wants to assume any of
these would not want to measure life satisfaction as Inglehart does.
Theory could provide a variety of ways out of the levels-of-analysis difficulty.
Verba (1965, p. 525) mentions the anthropological approach of defining a culture
as the orientations that are (100 percent) shared by and (100 percent) distinctive
to the members of a given group. He rightly points out how limiting such a
definition would be. An alternative—and, for political-culture theory, preferable—route
would be to spell out the theoretical links between the complex
patterns of subjective orientations that characterize the members of a society
and some construct characterizing the entire society. These links have been
called 'laws of composition' (Simon 1962) or 'auxiliary theories' (Elkins and
Simeon 1979). The more precise and measurable these links, the better.
One approach that implicitly provides a link between individuals and society
stresses the importance of the degree to which the members of a society share
certain values, even if there is not 100 percent unanimity (Verba 1965, pp. 525-6,
Elkins and Simeon 1979). Again, if the value or orientation under study can
take on more than two values, measurement issues arise. Although it is not
clear what statistic measures 'sharedness,' or consensus, when a value has
different intensities, devising such a statistic is not the major hurdle. One
possibility might be a variant of the indices of 'concentration of power' that
have been developed for the study of multi-polar systems in international
politics (Ray and Singer 1973). Employing such a measure, however, means
that the 'sharedness,' or degree of consensus, of the subjective orientation—not
the orientation itself—becomes the explanatory element.
A second possible way to overcome the levels of analysis problem is to
redefine political culture such that it incorporates only societal-level attributes.
A good example is to define political culture as a given society's symbols (e.g.
342 INTERNATIONAL JOURNAL OF PUBLIC OPINION RESEARCH
Elkins and Simeon 1979, Dittmer 1983, Welch 1987, p. 498). Symbols, by
definition, are shared among many people, indeed, among most or all of those
constituting the social group in which the symbols exist. Verba's (1965, p. 513)
definition includes the term symbols as a component of political culture, though
it also incorporates individual-level beliefs. A focus on rituals is a related
approach (Lane 1991). Similarly, some authors discuss culture as norms that
set standards of behavior (and attendant punishments for violations) common
to all members of a society. Laitin (Laitin and Wildavsky 1988) argues that
culture instills 'points of concern to be debated,' not values.
Focusing only on such societal concepts ensures that political culture will be
treated as a societal attribute, but it siphons out of the concept what many saw
as its richness and greatest potential: the ability to model individual human
psychology interacting with its environment. Also, identifying symbols, rituals,
norms or other non-individualistic phenomena is more difficult to study through
survey research (at least of the standard type) (Elkins and Simeon 1979, Welch
1987, Laitin and Wildavsky 1988).
MEASUREMENT
Almond has noted that the development of sophisticated survey-research
techniques was central to the blossoming of political-culture theory in recent
decades. Yet as Laitin (Laitin and Wildavsky 1988, p. 592) points out, a
respondent is unlikely to be fully aware of his or her deep-seated orientations
(cf. Almond 1990). Even honest respondents might give answers that are
misleading about very basic orientations. Of course, survey researchers now
have a good deal of experience in resolving such problems. They must know
what they are looking for, though. The problem for political-culture theory is
the lack of specification of (or agreement about) what exactly is involved. If a
researcher has satisfactorily dealt with the issues of definition discussed above,
then justifying the use of particular survey techniques, by themselves or together
with additional techniques, will be easy.
The five challenges discussed above focus on conceptual issues. Meeting
these challenges requires scholars to define and measure political culture within
the context of a theory of comparative politics. Not only does the larger theory
provide the framework for conceptualizing political culture, the political culture
approach seeks to go beyond description and explain political outcomes. Although
political culture theories are generally better developed than the concept itself,
more explicit attention to the place of political culture within the political
system must characterize those conducting research. Two links in particular
need to be clarified: the hypothesized link between individual orientations and
THE RENAISSANCE OF A RUBRIC 343
individual political behavior, and the link between individual behavior and the
political outcomes of interest.
EMPLOYING POLITICAL CULTURE TO EXPLAIN INDIVIDUAL BEHAVIOR
What does an individual with a particular type of outlook do (or avoid doing)
that an individual with a different outlook would not (or would) do? For a few
salient orientations, the behavior that is expected to result has been posited
clearly. A good example is interpersonal trust. A citizen who trusts other
members of society is hypothesized to be more disposed to form and join
secondary associations and to accept political losses without challenging the
democratic order (Inglehart 1988, p. 1204). For most other orientations, the
links to behavior have received little if any attention. Naturally, linking a political
culture to behavior is, given the challenges discussed above, even more complex
since one must link the balance among numerous orientations to behavior.
Moreover, empirical efforts to connect orientations and behavior require more
than just posing and testing bivariate relationships between certain orientations
and a propensity to some form of action or inaction. The presence or absence
of individual behaviors might well reflect individual outlooks less than they
reflect that polity's incentives for and restrictions on political behavior. A good
deal of the work that examined political culture in Communist systems dealt
with the issue of the link between orientations and behavior (e.g. Tucker 1973,
1987). It was an extremely difficult question. Mass protest against the regimes
in Eastern Europe and the USSR did occur, but they were sporadic, flaring up
and (prior to 1989) dying away in a different manner each time. Culturalist
approaches that stressed the antipathy of, say, Polish culture toward Soviet-style
rule could explain why large-scale protest was a potential outcome but not why
it happened when it did. Those who saw in Russian political culture certain
elements that produced relatively high support for Communist rule could not
explain the protests of the late 1980s without saying that changes had been
going on for some time. If so, why did the protests not begin earlier? Through
one or another comparative technique, an analysis must attempt to control for
participatory channels and other aspects of a political system when testing a
hypothesized link between culture and individual behavior.
POLITICAL CULTURE AND THE POLITICAL SYSTEM
This leads into the seventh and final challenge facing those who would employ
political culture: spelling out what they see as the connections between political
culture and other aspects of a political system (or to undertake to discover such
connections). It is not unfair to set this requirement before political-culture
344 INTERNATIONAL JOURNAL OF PUBLIC OPINION RESEARCH
theory. From the beginning, a society's political culture has been seen as a key
component of the entire political system. Also, the recent defenses of political
culture have explicitly sought to revive it not as mid-range theory but as grand
theory. (Recall Eckstein's 'one of two still viable general approaches to political
theory and explanation.').
One aspect of this task is to decide whether non-elites matter and, if so, how.
If some people's behavior counts more than others', a measure of political
culture based upon the entire mass public of a society will not suffice. Numerous
studies show that elites hold quite different beliefs and orientations than
non-elites in the same society. Moreover, equally strong evidence indicates that
elites act upon their beliefs more often and more effectively than non-elites;
this is, indeed, a definition of a member of the elite (Verba et al. 1978). In his
recent defense of political-culture theory, Diamond (1993a, pp. 2-7) makes a
strong case for the importance of elite values in the successful 'crafting' of
democratic institutions. Having done this, however, Diamond does not discuss
whether elite values might not be the same as those comprising the mass
political culture and what impact, if any, the mass public's political culture is
likely to have.
This lack of attention to the link between mass values and political outcomes
has contributed to a popular argument that political-culture theory is of little
relevance to recent democratizations. Not everyone believes that 'It is obvious
... that the emergence and persistence of a democratic government among a
group of people depends in some way on their beliefs' (Dahl 1989, p. 30). A
prominent group of scholars, sometimes dubbed the 'elitist school' of democracy,
has been contesting this view for some time (Schumpeter 1942, Schattschneider
i960, Lijphart 1968, Rustow 1970, O'Donnell and Schmitter 1986, Przeworski
1986, di Palma 1990). They propose that democracy depends to a much greater
degree upon the attitudes and behaviors of elites than upon those of mass
publics. Of course, extreme elitist perspectives are rarely tenable. To focus only
on elites and to ignore the societal bases of their power would be as misleading
as to look only from the bottom up. What the elitist school's arguments do
highlight is the lack ofconvincing arguments from political culture theorists about
the conceptual links between orientations and behavior, first, and subsequently
between individual behavior and the larger political system. Elitists have
consequently been more attracted to rational-choice models than to cultural
explanations of elite behavior (Przeworski 1986, 1991), though Karl (1991)
argues that elite and mass approaches should be reconciled.
In addition to the issue of how the mass public and the elite interact is the
need to relate mass political culture to political institutions, which must mediate
between political culture and political outcomes. The original argument in
The Civic Culture was that congruence or incongruence between the political
THE RENAISSANCE OF A RUBRIC 345
institutions and the political culture helps explain democratic stability (see
Verba 1965, p. 513, Patrick 1984, pp. 302-4, Almond and Powell 1992). Almond
and Verba argued that neither analyses of political institutions (e.g. presidential
vs. parliamentary forms) nor measures of the distribution of individual orientations
could by themselves explain democratic stability. Eckstein (1969)
developed the argument further by linking democratic stability to 'congruence'
between culture and structure. The problem has been that the power of
survey research methodology placed the emphasis squarely on measuring value
distributions. For example, after Pye (1965, pp. 9-10 and 16) stressed the
importance of treating the historical development of the entire system—'the
evolution of the institutions and value patterns'—he goes on to describe the
sole means of studying mass political culture as 'advanced techniques of survey
research' without offering any ideas about how to use survey data in tandem
with information on institutional evolution. Warwick (1990, p. 174) made the
point in the following way: 'The lack of precision in defining the independent
variable [political culture] carries through to the hypothesis: terms such as
congruence and conduciveness often hide an uncertainty over the nature or even
the direction of the causal connection between culture and system structure or
performance, an uncertainty that cross-sectional survey findings can scarcely
hope to address.' In the practice of empirical researchers, then, the tendency
has been to shift away from the original focus on congruence to a focus on the
existence and levels of certain values.
Some theorists, adopting a sufficiently long time perspective, can place
institutions as an endogenous variable—the form that institutions take in a
particular country is a function of the political culture of that country. Verba
(1965, p. 519), for example, has argued that fundamental political beliefs 'play
a major role in guiding the ways in which institutions develop and change.' He
does not explain whether he means the fundamental political beliefs of the
relatively small number of powerful political actors in a society at a given time
or the beliefs of the entire society, aggregated in some form or other. His
discussion suggests, though, that he has the latter in mind. It would not,
however, always be desirable to play down institutional rules in this manner. If
one is interested in shorter-term political outcomes, the exogenous impact of
institutions and the interplay between culture and institutions will need direct
examination. Moreover, a number of recent works highlight the centrality
of institutional practices for regime change and other longer-term political
phenomena (Elster 1988, March and Olsen 1989, Shepsle 1989, Easton 1990,
Warwick 1990, Thelen and Steinmo 1992, Stepan and Skach 1993, Weaver and
Rockman 1993). By the same token, recent efforts to rehabilitate political
culture convincingly criticize those instutionalists who treat political culture as
endogenous, largely shaped by institutions (Diamond 1993a, p. 7). As the
346 INTERNATIONAL JOURNAL OF PUBLIC OPINION RESEARCH
majority of both culturalists and institutionalists stress, the critical matter is
the interplay between structure and culture (see, e.g., Elkins and Simeon 1979,
p. 143, Inglehart 1990, p. 14).
For this reason, the process of placing a study of political culture in a
satisfactory theoretical framework must involve providing assumptions about
the ways in which cross-national differences in culture, when taking into account
cross-national institutional differences, will shape outcomes. The cultural approach
will not draw additional adherents by ignoring this. Still, the recent
defenders of the political culture approach remain vague in specifying the
relationship between culture and institutions. Eckstein (1988) sticks to clarifying
the link between orientations and individual behavior. Inglehart (1990) hypotheses
an interactive and changing relationship between values and democratic
institutions (and between values and economic growth) but only discusses this
in a long-term evolutionary sense. Given his interest in understanding the
existence of stable democracies in some societies but not others, he does not
discuss how culture relates to institutions at a particular time nor how one
could compare this relationship cross-nationally. Almond (1990) pays the most
attention to the issue, first criticizing those who attribute cultural determination
to his and others' views of political culture. He then discusses the perspective
that he developed with Powell (Almond and Powell 1992). They see political
culture as the medium through which the inputs and outputs of the political
system travel: policies emerge from structures and processes and are interpreted
through the lens of political culture, as are society's responses. With more
concrete specification of their theoretical expectations (not possible in a textbook),
the Almond and Powell perspective could be useful, though other scholars
might reject Almond and Powell's systems perspective.
CONCLUSIONS
Recent defenders of political-culture theory would have us believe that anyone
acknowledging the crucial political role of individual values must also accept
that the political culture approach is valid. Clearly, to discount the impact of
individuals and their political orientations is to fly in the face of a good deal of
evidence. Yet making such a case is not enough. Even rational-choice theorists
acknowledge the importance of popular beliefs, norms, values, etc. Rationalchoice
theorists claim instead that a rational-choice approach is of greater
theoretical utility because of the clarity of its postulates and the relative
parsimony of its models. To offer up political culture as one of the two dominant
approaches in the field and a potential rival to rational-choice theory requires
that political culture be defended as a useful and usable concept embedded in
a workable theory.
THE RENAISSANCE OF A RUBRIC 347
Seven distinct challenges face a scholar who would conduct a comparative
empirical study relating political culture in the mainstream (Almondian) sense
to patterns of political outcomes. The challenges are to define the term, to
disentangle subcultures from a society's overall political culture, to integrate
the many individual-level orientations of which the concept is composed, to
create a societal-level variable from individual-level components, to develop
techniques of measuring the resulting concept, to derive hypotheses about
individual political behavior from the subjective orientations under study and
to theorize how political culture interacts with institutions and other attributes
of a polity to produce political outcomes. No single study in this tradition has
dealt with all seven, and for none of the seven has a generally accepted resolution
been found. Thinking about the development of political-culture theory in
terms of these seven challenges makes clear why, in Diamond's (1993a, p. 15)
view, the following questions
... loom large in the literature: How broadly must these beliefs and values be shared
across various groups and strata of a nation's citizenry? How soon after the inauguration
of democracy must they develop? How important are they to democracy, relative to
other types of conditions? Which elements of political culture are most important?
What determines the evolution and transformation of political culture over time? What
specific consequences for democracy can we attribute to various elements of political
culture?
Despite noting the absence of answers, Diamond chooses to stress the promise
of the political culture perspective. I place greater emphasis on the dangers of
neglecting further conceptual and theoretical development.
For example, as researchers investigate political values and attitudes within
societies seeking to democratize, they must be wary of replicating unquestioningly
the methods and assumptions of political-culture theory. From the beginning,
the study of political culture has most frequently been undertaken to illuminate
the prospects for democracy in one or more societies. Inglehart's (1988; 1990,
ch. 1) attempt to resuscitate political-culture theory by using better data rests
in part on relating attitudes to the number of years a country has experienced
democratic governance. Furthermore, much of the rekindled interest in political
culture in recent years stems from the collapse of authoritarian regimes, in
particular the formerly Communist regimes of Eastern Europe and the USSR.
As Barry (1970, p. 52) has noted, a natural test of political culture's impact is
to compare values before and after a regime change. A series of studies has
examined mass values and attitudes in societies formerly under authoritarian
rule in order to advance hypotheses about how these countries' 'political cultures'
influence their prospects to consolidate democracy (e.g. Brown 1989, Diamond
et al. 1989, Catterberg 1991, Gibson et al. 1992, Gibson and Duch 1993).
348 INTERNATIONAL JOURNAL OF PUBLIC OPINION RESEARCH
These analyses of political attitudes and values in democratizing societies are
enlightening and a boon to those studying the government and politics of these
societies. Yet none seriously questions the political culture approach they adopt.
Moreover, most of these researchers are actually investigating a few select
orientations. Their use of political-culture theory comes in general discussions
before and after the analyses. Those who seriously intend to analyze the
connections between one or more societies' political cultures and their political
changes—and thereby pose an alternative to rational choice models of regime
change—must attend to all seven conceptual and theoretical poblems discussed
above.
By identifying the remaining tasks standing before the 'renaissance' of political
culture, I have sought to facilitate the theoretical reinvigoration of the general
approach. Yet in the short run, we need to accept political culture as a rubric
not a concept. Admitting that researchers focus almost exclusively on a few
significant orientations—whose relationships to a society's culture becomes a
separate issue—could actually facilitate scholarly analyses of cross-national value
differences by eliminating the daunting task of reifying those orientations. Much
clarity would be gained by admitting that, primarily, we want and need to know
more about such matters as popular support for the political system, the
willingness of people to engage in different types of political behavior, and
evaluations of governmental performance. Why confuse the issue by claiming
to depict some grandiose phenomenon which is so defined that no one can
really measure it? Moreover, taking this approach would release the scholar
from the obligation to spell out a model of the entire political system. While
one cannot study political culture without attending to its place in the political
system, investigations of mass orientations are of interest for their own sake.
REFERENCES
Almond, Gabriel A. (1990): 'The Study of Political Culture.' In Gabriel A. Almond
(ed.) A Discipline Divided: Schools and Sects in Political Science, Beverly Hills, CA,
Sage.
Almond, Gabriel A. (1993): 'Forward: The Return to Political Culture.' In Larry
Diamond (ed.) Political Culture and Democracy in Developing Countries, Boulder, CO,
Lynne Reinner.
Almond, Gabriel A. and Powell, G. Bingham (eds.) (1992): Comparative Politics Today:
A World View, 5th ed, New York, Harper Collins.
Almond, Gabriel A. and Verba, Sydney (1963): The Civic Culture: Political Attitudes
and Democracy in Five Nations, Princeton, Princeton University Press.
Almond, Gabriel A. and Verba, Sydney (eds.) (1980): The Civic Culture Revisited: An
Analytic Study, Boston, Little, Brown.
THE RENAISSANCE OF A RUBRIC 349
Barry, Brian (1970): Sociologists, Economists and Democracy, Chicago, University of
Chicago Press.
Brint, Michael (1991): A Genealogy of Political Culture, Boulder, CO, Westview Press.
Brown, Archie (1979): 'Introduction.' In Archie Brown and Jack Gray (eds.) Political
Culture and Political Change in Communist States, 2nd ed, New York, Holmes &
Meier.
Brown, Archie (1989): 'Ideology and Political Culture.' In Seweryn Bialer (ed.) Inside
Gorbachevs Russia, Boulder, CO, Westview.
Catterberg, Edgardo (1991): Argentina Confronts Politics: Political Culture and Public
Opinion in the Argentine Transition to Democracy, Boulder, CO, Lynne Rienner.
Conover, Pamela J. and Feldman, Stanley (1984): 'How People Organize the Political
World: A Schematic Model,' American Journal of Political Science, 28, 95-125.
Dahl, Robert H. (1989): Democracy and Its Critics, New Haven, Yale University Press.
Diamond, Larry (1993a): 'Introduction: Political Culture and Democracy.' In Larry
Diamond (ed.) Political Culture and Democracy in Developing Countries, Boulder, CO,
Lynne Rienner.
Diamond, Larry (ed.) (1993^): Political Culture and Democracy in Developing Countries,
Boulder, CO, Lynne Rienner.
Diamond, Larry, Linz, Juan J. and Lipset, Seymour Martin (eds.) (1989): Democracy
in Developing Countries, Vols. 1—4, Boulder, CO, Lynne Rienner.
di Palma, Guiseppe (1990): To Craft Democracies: An Essay on Democratic Transitions,
Berkeley, University of California Press.
Dittmer, Lowell (1983): 'Comparative Communist Political Culture,' Studies in Comparative
Communism, 16, 0-24.
Easton, David (1990): The Analysis of Political Structure, New York, Routledge.
Eckstein, Harry (1969): 'Authority Relations and Government Performance: A Theoretical
Framework,' Comparative Political Studies, 2, 269-325.
Eckstein, Harry (1988): 'A Culturalist Theory of Political Change,' American Political
Science Review, 82, 789—804.
Elkins, David J. and Simeon, Richard E. B. (1979): 'A Cause in Search of Its Effect,
or What Does Political Culture Explain?' Comparative Politics, 11, 127-45.
Elster, Jon (1988): 'Consequences of Constitutional Choice: Reflections On Tocqueville.'
In Jon Elster and Rune Slagstad (eds.) Constitutionalism and Democracy, New York,
Cambridge University Press.
Gibson, James L. and Duch, Raymond M. (1993): 'Emerging Democratic Values in
Soviet Political Culture.' In Arthur H. Miller, William M. Reisinger, and Vicki L.
Hesli (eds.) Public Opinion and Regime Change: The New Politics of Post-Soviet
Societies, Boulder, CO, Westview.
Gibson, James L., Duch, Raymond M. and Tedin, Kent L. (1992): Democratic Values
and the Transformation of the Soviet Union,' Journal of Politics, 54, 329—71.
Hamill, Ruth, Lodge, Milton G. and Blake, Frederick (1985): 'The Breadth, Depth
and Utility of Class, Partisan and Ideological Schemata,' American Journal of Political
Science, 29, 850-70.
Hannan, Michael T. (1991): Aggregation and Disaggregation in the Social Sciences, rev.
ed., Lexington, MA, Lexington Books.
350 INTERNATIONAL JOURNAL OF PUBLIC OPINION RESEARCH
Inglehart, Ronald (1988): 'The Renaissance of Political Culture,' American Political
Science Review, 82, 1203-30.
Inglehart, Ronald (1990): Culture Shift in Advanced Industrial Society, Princeton,
Princeton University Press.
Karl, Terry Lynn (1991): 'Dilemmas of Democratization in Latin America.' In Dankwart
A. Rustow and Kenneth Paul Erickson (eds.) Comparative Political Dynamics: Global
Research Perspectives, New York, Harper Collins.
Kuklinski, James H., Luskin, Robert C. and Bolland, John (1991): 'Where Is the
Schema? Going Beyond the U
S" Word in Political Psychology,' American Political
Science Review, 85, 1341-56.
Laitin, David and Wildavsky, Aaron (1988): 'Controversy: Political Culture and Political
Preferences,' American Political Science Review, 82, 589—96.
Lane, Christel (1991): The Rites of Rulers: Ritual in Industrial Society, New York,
Cambridge University Press.
Lane, Ruth (1992): 'Political Culture: Residual Category or General Theory?' Comparative
Political Studies, 25, 362—87.
Lijphart, Arend (1968): The Politics ofAccommodation, Berkeley, University of California
Press.
Lodge, Milton and Tursky, Bernard (1979): 'Comparison Between Category and
Magnitude Scaling of Political Opinion Employing SRC/CPS Items,' American
Political Science Review, 73, 50-66.
Lodge, Milton and Tursky, Bernard (1981): 'On the Magnitude Scaling of Political
Opinion in Survey Research,' American Journal of Political Science, 25, 376-419.
Lodge, Milton, McGraw, Kathleen M., Conover, Pamela J., Feldman, Stanley and
Miller, Arthur H. (1991): 'Where Is the Schema? Critiques,' American Political
Science Review, 85, 1357—80.
March, James G. and Olsen, Johan P. (1989): Rediscovering Institutions: The Organizational
Basis of Politics, New York, Free Press.
Mayer, Lawrence C. (1989): Redefining Comparative Politics: Promise Versus Performance,
Beverly Hills, CA, Sage.
Miller, Arthur H., Wattenberg, Martin P. and Malanchuk, Oksana (1986): 'Schematic
Assessments of Presidential Candidates,' American Political Science Review, 80,
521-40.
O'Donnell, Guillermo and Schmitter, Phillipe C. (1986): 'Tentative Conclusions About
Uncertain Democracies.' In Guillermo O'Donnell, Phillipe C. Schmitter, and Laurence
Whitehead (eds.) Transitions From Authoritarian Rule: Prospects for Democracy,
Part IV, Baltimore, Johns Hopkins University Press.
Pateman, Carol (1971): 'Political Culture, Political Structure and Change,' British
Journal of Political Science, 1, 291-305.
Patrick, Glenda M. (1984): 'Political Culture.' In Giovanni Sartori (ed.) Social Science
Concepts: A Systematic Analysis, Beverly Hills, CA, Sage.
Przeworski, Adam (1986): 'Some Problems in the Study of the Transition to Democracy.'
In Guillermo O'Donnell, Phillipe C. Schmitter, and Laurence Whitehead (eds.)
Transitions from Authoritarian Rule: Prospects for Democracy, Part III, Baltimore,
Johns Hopkins University Press.
THE RENAISSANCE OF A RUBRIC 351
Przeworski, Adam (1991): Democracy and the Market: Political and Economic Reforms
in Eastern Europe and Latin America, New York, Cambridge University Press.
Pye, Lucian W. (1965): 'Introduction: Political Culture and Political Development.' In
Lucian W. Pye and Sydney Verba (eds.) Political Culture and Political Development,
Princeton, Princeton University Press.
Pye, Lucian W. (1972): 'Culture and Political Science: Problems in the Evaluation of
the Concept of Political Culture,' Social Science Quarterly, 53, 285-96.
Ray, James Lee and Singer, J. David (1973): 'Measuring the Concentration of Power
in the International System,' Sociological Methods and Research, 1, 403-36.
Rogowski, Ronald (1974): Rational Legitimacy, Princeton, Princeton University Press.
Rosenbaum, Walter (1972): Political Culture. New York, Praeger.
Rustow, Dankwart A. (1970): 'Transitions to Democracy: Toward a Dynamic Model,'
Comparative Politics, 2, 337—63.
Sartori, Giovanni (1984): 'Guidelines for Concept Analysis.' In Giovanni Sartori (ed.)
Social Science Concepts: A Systematic Analysis, Beverly Hills, CA, Sage.
Schattschneider, E. E. (i960): The Semi-Sovereign People, New York, Holt, Rinehart &
Winston.
Scheuch, Erwin K. (1969): 'Social Context and Individual Behavior.' In Mattei Dogan
and Stein Rokkan (eds.) Quantitative Ecological Analysis in the Social Sciences,
Cambridge, MA, MIT Press.
Schumpeter, Joseph (1942): Capitalism, Socialism and Democracy, 2nd ed., New York,
Harper and Bros.
Shepsle, Kenneth A. (1989): 'Studying Institutions: Some Lessons from the Rational
Choice Approach,' Journal of Theoretical Politics, 1, 131—47.
Simon, Hebert A. (1962): 'The Architecture of Complexity,' Proceedings of the American
Philosophical Society, 106, 67—82.
Stepan, Alfred and Skach, Cindy (1993): 'Constitutional Frameworks and Democratic
Consolidation: Parliamentarism Versus Presidentiah'sm,' World Politics, 46, 1—22.
Thelen, Kathleen and Steinmo, Sven (1992): 'Historical Institutionalism in Comparative
Perspective.' In Sven Steinmo, Kathleen Thelen and Frank Longstreth (eds.)
Structuring Politics: Historical Institionalism in Comparative Analysis. New York,
Cambridge University Press.
Thompson, Michael, Ellis, Richard and Wildavsky, Aaron (1990): Cultural Theory,
Boulder, CO, Westview.
Tucker, Robert C. (1973): 'Culture, Political Culture and Communist Society,' Political
Science Quarterly, 88, 173-90.
Tucker, Robert C. (1987): Political Culture and Leadership in Soviet Russia, New York,
Norton.
Tufte, Edward R. (1969): 'Improving Data Analysis in Political Science,' World Politics,
21, 641-54.
Verba, Sydney (1965): 'Conclusion: Comparative Political Culture.' In Lucian W. Pye
and Sydney Verba (eds.) Political Culture and Political Development, Princeton,
Princeton University Press.
Verba, Sydney, Nie, Norman H. and Kim, Jae-On (1978): Participation and Political
352 INTERNATIONAL JOURNAL OF PUBLIC OPINION RESEARCH
Equality: A Seven-Nation Comparison, New York, Cambridge University Press.
Warwick, Paul V. (1990): Culture, Structure or Choice? Essays in the Interpretation ofthe
British Experience, New York, Agathon Press.
Weaver, R. Kent and Rockman, Bert A. (eds.) (1993): Do Institutions Matters? Government
Capabilities in the United States and Abroad, Washington DC, The Brookings In-
stitution.
Webster, Murray, Jr. (1973): 'Psychological Reductionism, Methodological Individualism
and Large-Scale Problems,' American Sociological Review, 38, 258-73.
Welch, Stephen (1987): 'Issues in the Study of Political Culture: The Example of
Communist Party States,' British Journal of Political Science, 17, 470-500.
Wildavsky, Aaron (1987): 'Choosing Preferences By Constructing Institutions: A Cultural
Theory of Preference Formation,' American Political Science Review, 81, 3—21.
BIOGRAPHICAL NOTE
William M. Reisinger is Associate Professor of Political Science at the University
of Iowa, Iowa City, Iowa, USA. His research focuses on public opinion, foreign
policy and democratization in the Soviet Union and in the Soviet successor
states. His publications include Energy and the Soviet Bloc: Alliance Politics
After Stalin, Cornell University Press, 1992), Public Opinion and Regime Change:
The New Politics of Post-Soviet Societies (Westview Press, 1993; co-edited with
Arthur H. Miller and Vicki L. Hesli), as well as articles in the American
Political Science Review, British Journal of Political Science, Europe—Asia Studies,
International Organization, International Studies Quarterly, Journal of Politics,
and Slavic Review.