ß                                                                                                                                              INTRODUCTION
trial society. Postmodern values become prevalent, bringing a variety of societal changes, from equal rights for women to democratic political institutions and the decline of state socialist regimes. The emergence of this Postmodern value syndrome is described in the following chapters.
This book demonstrates that there are powerful linkages between belief systems and political and socioeconomic variables such as democracy or economic growth rates. It also demonstrates coherent and to some extent predictable patterns of change in values and belief systems. These changes in worldviews reflect changes in the economic and political environment, but they take place with a generational time lag and have considerable autonomy and momentum of their own. Major cultural changes are occurring. They have global implications that are too important to ignore.
CHAPTER     1
Value Systems: The Subjective Aspect of Politics and Economics
Modernization and Postmodernization
Economic, cultural, and political change go together in coherent patterns that are changing the world in predictable ways.
This has been the central claim of Modernization theory, from Karl Marx to Max Weber to Daniel Bell. The claim has given rise to heated debate during the last two centuries. This book presents evidence that this claim is largely correct: though we cannot predict exactly what will happen in a given society at a given time, some major trends are predictable in broad outline. When given processes of change are set in motion, certain characteristics are likely to emerge in the long run.
The idea that social and economic change go together on coherent trajectories has been attractive but controversial ever since it was proposed by Marx. It is intellectually exciting because it not only helps explain economic, social, and political change, but may even provide a certain degree of predictability. So far, most efforts at prediction in human affairs have been exercises in hubris; it is common knowledge that many of Marx's predictions were wrong. Human behavior is so complex and influenced by such a wide range of factors, operating on so many levels, that any claim to provide precise, unqualified predictions is likely to go unfulfilled.
We do not make such promises: one cannot foretell the precise course of social change. Nevertheless, certain syndromes of economic, political, and cultural changes go together in coherent trajectories, with some trajectories being more probable than others. In the long term, across many societies, once given processes are set in motion, certain important changes are likely to happen. Industrialization, for example, tends to bring increasing urbanization, growing occupational specialization, and higher levels of formal education in any society that undertakes it (Lerner, 1958; Deutsch, 1964). These are core elements of a trajectory that is generally called "Modernization."
This trajectory also tends to bring less obvious but equally important long-term consequences, such as rising levels of mass political participation. Thus, although we cannot predict the actions of specific leaders in given countries, we can say that (at this point in history) mass input to politics is likelier to play a decisive role in Sweden or Japan than in Albania or Burma. And we can even specify, with far better than random success, what issues are likely to be most salient in the politics of the respective types of societies.
8
CHAPTER    1
The Modernization trajectory is linked with a wide range of other cultural changes. As we will see, certain cultural values are conducive to the economic accumulation and investment that make industrialization possible, and the sharply contrasting gender roles that characterize all preindustrial societies almost inevitably give way to increasingly similar gender roles in advanced industrial society.
But social change is not linear. Although a specific Modernization syndrome of changes becomes probable when societies move from an agrarian mode to an industrial mode, no trend goes on in the same direction forever. It eventually reaches a point of diminishing returns. Modernization is no exception. In the past few decades, advanced industrial societies have reached an inflection point and begun moving on a new trajectory that might be called "Postmodernization."
With Postmodernization, a new worldview is gradually replacing the outlook that has dominated industrializing societies since the Industrial Revolution. It reflects a shift in what people want out of life. It is transforming basic norms governing politics, work, religion, family, and sexual behavior. Thus, the process of economic development leads to two successive trajectories, Modernization and Postmodernization. Both of them are strongly linked with economic development, but Postmodernization represents a later stage of development that is linked with very different beliefs from those that characterize Modernization. These belief systems are not mere consequences of economic or social changes, but shape socioeconomic conditions and are shaped by them, in reciprocal fashion.
Modernization Theory: The Linkages between Culture, Economics, and Politics
The study of Modernization played a major role in social science in the late 1950s and early 1960s. Severely criticized subsequently, since the 1970s the Modernization concept has been widely considered discredited. As Pye (1990) has argued, it may be time to reexamine it. This chapter does so, presenting new empirical evidence and proposing a modified view of how Modernization works.
The central claim of Modernization theory is that industrialization is linked with specific processes of sociopolitical change that apply widely: though preindustrial societies vary immensely, one can meaningfully speak of a model of "modern" or "industrial" society toward which all societies tend to move if they commit themselves to industrialization. Economic development is linked with a syndrome of changes that includes not only industrialization, but also urbanization, mass education, occupational specialization, bureaucratization, and communications development, which in turn are linked with still broader cultural, social, and political changes.
One reason why Modernization theory aroused such great interest was its promise of predictive power: it implied that once a society entered the trajectory of industrialization, certain types of cultural and political change were
CULTURE,    POLITICS,    AND    ECONOMICS
9
likely to take place, ranging from lower birth rates to greater penetration by government, higher life expectancies, increased mass political participation, and perhaps even democracy. Some critics caricatured Modernization theory as implying that economic development would easily and automatically produce liberal democracies, and they dismissed this outlook as naive ethnocen-trism. In fact, most Modernization theorists made more qualified prognoses than this, but if we drop the gratuitous assumption that Modernization is easy and automatic, even this claim does not seem totally implausible today.
Modernization theory has been developing for over a century. A wide variety of social theorists have argued that technological and economic changes are linked with coherent and predictable patterns of cultural and political change. But there has been continuing debate over the causal linkages: does economic change cause cultural and political change, or does it work in the opposite direction?
Marx emphasized economic determinism, arguing that a society's technological level shapes its economic system, which in turn determines its cultural and political characteristics: given the technological level of the windmill, a society will be based on subsistence agriculture, with a mass of impoverished peasants dominated by a landed aristocracy; the steam engine brings an industrial society in which the bourgeoisie becomes the dominant elite, exploiting and repressing an urban proletariat.
Weber, on the other hand, emphasized the impact of culture: it was not just an epiphenomenon of the economic system, but an important causal factor in itself; culture can shape economic behavior, as well as being shaped by it. Thus, the emergence of the Protestant Ethic facilitated the rise of capitalism, which contributed to both the Industrial Revolution and the Democratic Revolution: this view held that belief systems influence economic and political life, as well as being influenced by them.
Some of Marx's successors shifted the emphasis from economic determinism (which suggests that the revolutionary Utopia will come spontaneously) toward greater emphasis on the impact of ideology and culture. Thus Lenin argued that by itself, the working class would never develop sufficient class consciousness for a successful revolution; they needed to be led by an ideologically aware vanguard of professional revolutionaries.
Mao emphasized the power of revolutionary thinking even more strongly. Breaking with Marxist orthodoxy, he held that China need not wait for the processes of urbanization and industrialization to transform it; if an ideologically committed cadre could instill sufficient enthusiasm among the Chinese masses, a communist revolution could succeed even in an agrarian society. Mao's faith in the power of ideological fervor to triumph over material obstacles seemed justified by the Chinese communist victory in 1949 over forces with vastly superior financial resources and manpower. On the other hand, the fact that ideological determinism has limits was demonstrated by the disastrous failure of the Great Leap Forward in 1959: to develop a complex society, it seems, one needs experts with specialized knowledge, as well as right-
10
CHAPTER    I
thinking masses. When building a drainage system or constructing a steel mill, there are ways that work and ways that do not work, regardless of one's ideological perspective.
While conceding an important role to cultural factors, recent Modernization theorists such as Bell (1973) viewed changes in the structure of the workforce as the leading cause of cultural change. For Bell, the crucial milestone in the coming of "Postindustrial society" is reached when a majority of the workforce is in the tertiary sector of the economy, producing neither raw materials, nor manufactured goods, but services. This leads to a massive expansion of formal education, driven by the need for an increasingly skilled and specialized workforce. Other writers such as Lerner (1958) and Inkeles and Smith (1974) emphasized the importance of formal education as the main factor shaping a "modern" worldview.
Does Modernization lead to democracy? In the late 1950s, Khrushchev's reforms gave rise to hopes that the communist bloc might be on the brink of democratizing. The emergence of scores of newly independent postcolonial nations in the 1960s intensified these hopes. But optimism collapsed after the communist elite drove Khrushchev from power in 1964, the Soviet world settled down into a seemingly permanent authoritarian regime under Brezhnev, and authoritarian regimes took over in most postcolonial nations. Rostow (1961) had argued that economic development was inherently conducive to democratization, but by the 1970s most social scientists were skeptical of the idea. Authoritarian regimes seemed to be a permanent feature of the world— even (or perhaps especially) in those communist states that had achieved impressive economic growth. Industrialization could give rise to either democracy or dictatorship.
We propose a revised view of Modernization theory. We agree with the Modernization theorists on their most central point: that economic development, cultural change, and political change are linked in coherent and even, to some extent, predictable patterns. Some trajectories of change are more probable than others because certain configurations of values and beliefs, and political and economic institutions, are mutually supportive—while others are not. Thus, if one knows one component of a society, one can predict what other components will be present with far better than random success.
But while we follow Marx, Weber, and their successors in believing that change tends to take predictable rather than random trajectories, we differ from most Modernization theorists on four essential points:
1.   Change is not linear. It does not move in one continuous direction until the end of history. Instead, it eventually reaches points of diminishing returns and has begun to move in a fundamentally new direction during the past few decades.
2.  Previous versions of Modernization theory were deterministic, with the Marxist version tending toward economic determinism and the Weberian version sometimes tending toward cultural determinism. We believe that the relationships between economics and culture and politics are mutually support-
CULTURE,    POLITICS,    AND    ECONOMICS
11
ive, as are the various systems of a biological organism. It would be senseless to ask whether the behavior of the human body is "really" determined by the muscular system, the circulatory system, the nervous system, or the respiratory system: each plays an essential role, and all activity ceases if any of them breaks down. Similarly, political systems and economic systems require a supportive cultural system—otherwise they would need to rely on naked coercion, which almost never endures for long. Conversely, a cultural system that was incompatible with its economic system would be unlikely to endure. Economic determinism, cultural determinism, and political determinism are all oversimplified: the causal linkages tend to be reciprocal. Unless these systems are mutually supportive, they are unlikely to survive.
3.  We reject the ethnocentric perspective of those who equated Modernization with "Westernization": At one point in history, Modernization was concentrated in the West; today it is evident that the process is global, and that in some ways East Asia is now leading the process of Modernization. In keeping with this outlook, we propose a modified interpretation of Weber's (1904-5) thesis concerning the role of the Protestant Ethic in economic development. Weber was correct in viewing the rise of Protestantism as a crucial event in the Modernization of Europe. However, its impact was not unique to Protestantism but was mainly due to the fact that its acquisitive rationality supplanted a set of religious norms that are common to most preindustrial societies and that inhibit economic achievement. Protestantism was uniquely Western, but acquisitive rationality is not. Although industrialization occurred first in the West, the rise of the West was only one version of Modernization.
4.  Democracy is not inherent in the Modernization phase, as some Modernization theorists suggested. There are alternative outcomes, with fascism and communism being the most prominent alternatives as Moore (1966) has pointed out. But democracy does become increasingly likely as societies move beyond the Modernization phase into Postmodernization. In the Postmodern phase, a distinctive syndrome of changes occur that make democracy increasingly likely—to the point where it eventually becomes costly to avoid.
We have stated four ways in which our view—which might be termed Post-modernization theory—differs from Modernization theory. Let us provide more detail on these points. Chapter 3 will present empirical evidence that supports the central claim underlying both Modernization theory and Post-modernization theory: that technological and economic changes tend to be linked with specific types of cultural, political, and social change. In other words, history tends to move in coherent and to some extent predictable patterns.
Socioeconomic Change Is Not Linear
The prevailing direction of development has changed in the last quarter century, and this shift is so distinctive that, rather than continuing to use the term "Modernization," we prefer to speak of "Postmodernization." The term "Post-
12
CHAPTER    1
modern" has been used with scores of different meanings, some of which are associated with a cultural relativism so extreme that it approaches cultural determinism: it asserts that culture shapes human experience almost entirely, unlimited by any external reality. Nevertheless, the term conveys an important insight, suggesting that the process known as Modernization is no longer at the cutting edge, and that social change is now moving in a fundamentally different direction. Moreover, the literature on Postmodernism suggests some of the specific attributes of this new direction: it is a move away from the emphasis on economic efficiency, bureaucratic authority, and scientific rationality that characterized Modernization, toward a more human society with more room for individual autonomy, diversity, and self-expression.
Unfortunately, the word "Postmodern" has become loaded with so many meanings that it is in danger of conveying everything and nothing. In architecture, the term has a clear meaning, designating a style of architecture that departs strikingly from the bare functionalism of "modern" architecture, which had become sterile and aesthetically repelling. The first glass box was a stunning tour de force, but by the one-hundredth box, the novelty had worn thin. Postmodern architecture reintroduced a human scale, with touches of adornment and references to the past, but incorporating new technology. In a similar vein, we suggest that Postmodern society is moving away from the standardized functionalism and the enthusiasm for science and economic growth that dominated industrial society during an era of scarcity—giving more weight to aesthetic and human considerations and incorporating elements of the past into a new context.
Neither Cultural Determinism Nor Economic Determinism
We disagree with the cultural determinism that is sometimes linked with the concept of Postmodernism. Postmodern writers are certainly correct in thinking that everyone perceives reality through some kind of cultural filter. Moreover, these cultural factors are steadily becoming a more important component of experience as we move from societies of scarcity, in which economic necessity limits one's behavior rather narrowly, to a world in which human will increasingly prevails over the external environment, allowing broader room for individual choice: this is a major reason why the Postmodern perspective has become increasingly credible.
But we reject the notion that cultural construction is the only factor shaping human experience. There is an objective reality out there too, and it applies to social relations as well as to natural science. External reality is crucial when it comes to the ultimate political resource, violence: when you shoot someone, that person dies regardless of whether he or she believes in ballistics or bullets. Similarly, though an architect has considerable scope for choice and imagination, if one forgets objective engineering principles, the building may collapse. Partly for this reason, architecture has preserved a healthy respect for reality. Similarly again, among physicists and astronomers, cultural biases play
CULTURE,    POLITICS,    AND    ECONOMICS
13
a minimal role. Despite some nonscientists' garbled references to the Heisenberg uncertainty principle, there is a worldwide consensus among natural scientists that they are studying a reality that exists independently of their preconceptions;1 a theory eventually triumphs or is rejected depending on how well it models and predicts that reality—even if it violates people's long-standing beliefs.
The fact that some Postmodern writers' grasp of the physical sciences is a bit shaky was demonstrated rather strikingly in 1996, when Alan Sokal, a physicist irked with Postmodernist claims that objective reality had dissolved in the physical sciences, submitted an article to Social Text, one of this school's leading reviews. His article, entitled "Transgressing the Boundaries: Toward a Transformative Hermeneutics of Quantam Gravity," began: "There are many natural scientists, and especially physicists, who. .. cling to the dogma. . . that there exists an external world, whose properties are independent of any individual human being. ... It has thus become increasingly apparent that physical 'reality,' no less than social 'reality,' is at bottom a social and linguistic construct; that scientific 'knowledge,' far from being objective, reflects and encodes the dominant ideologies and power relations of the culture that produced it" (Sokal, 1996: 217-18).
Though the text that followed was full of nonsense, this viewpoint was all too congenial to many Poststructuralists. Sokol went on to solemnly proclaim a long series of palpable absurdities about physical reality, including claims that the force of gravity and pi were socially constructed.
According to the New York Times account, this article was reviewed by a half dozen members of the review's editorial board, none of whom seemed to realize that the piece was a broad self-parody; they caught on shortly after the article was published, when the author himself revealed that it was a hoax.
This is not the first time that an august body has taken pi to be a social construct. In the nineteenth century, the Indiana state legislature passed a resolution officially declaring that pi would henceforth be a round 4.0, instead of the
1 The Heisenberg principle is often misread as indicating that the laws of physics do not really govern the universe, which is fundamentally disorderly and unpredictable. At the ultimate level of smallness, the universe is probabilistic, not deterministic. Thus, the behavior of individual photons is unpredictable. But large numbers of photons behave in ways that are indistinguishable from being deterministic; and since human beings normally only deal with enormous numbers of photons, the behavior of light can be predicted very accurately by deterministic physical laws.
Other laws of physics are also slight oversimplifications of reality. For example, though the laws of gases say otherwise, it is conceivable that all of the air molecules in the reader's vicinity could suddenly rush to the far end of the room and remain there until you died a horrible death. The reader need not worry. This is technically possible, but the probability is so overwhelmingly low that it would not be expected to occur even once during the entire lifetime of the universe (or even in many lifetimes of the universe). At the microlevel, the universe is probabilistic; this is a very significant fact. But it is extremely misleading to leap from this fact to the conclusion that Newton and Avogadro had it all wrong, and that the universe is disorderly and your brain could spontaneously explode at any moment. Technically, it could. But it's not likely to happen until long after the sun and the stars have all disappeared from the sky.
14
CHAPTER    1
inconvenient 3.1416; but this may be the first time that the proposition has been accepted by a panel of Ph.D.'s.
Despite this bit of entrapment, Postmodern thinkers are making a valid and profoundly important point in emphasizing that everyone's perception of reality is shaped by his or her subjective values and preconceptions. Moreover, these factors help shape even natural scientists' perceptions of reality— though not quite to the extent that some Postmodernists seem to think it does.
As Kuhn (1962) pointed out, objective tests alone do not immediately cause an entire scientific paradigm to be rejected; as inconsistent observations accumulate, the dominant paradigm may increasingly be called into question and new explanations proposed, but the new paradigm generally comes to be accepted through intergenerational replacement of scientists, more than through conversion of the older scientists. This reflects the fact that the cognitive structures of the older generation are organized around the old paradigm; it is far easier for the new generation to integrate their thinking according to the new paradigm than it is for the older generation, which would have to dismantle elaborate cognitive structures of inconsistent previous learning. At any given time, natural science reflects a cross-cultural consensus depending, ultimately, on how well given interpretations model and predict an external reality. The fine arts are at the opposite extreme. Aesthetic preferences largely are a matter of cultural predispositions.
Social phenomena fall between these extremes. Human behavior is heavily influenced by the culture in which one has been socialized. But objective factors set limits too, a recent example being the collapse and abandonment of state-run economies from Czechoslovakia to China: in running an economy, there are ways that work and ways that do not work.
Nevertheless, the term "Postmodern" is potentially useful: it implies that social change has moved beyond the instrumental rationality that was central to Modernization and is now taking a fundamentally different direction. This book does not discuss in any detail the various writers who have been labeled Postmodern: it is not about them. It deals with a set of empirical changes that are taking place among mass publics and will examine some specific ways in which the direction of social change has shifted. They include the fact that, while Modernization was not necessarily linked with democratization, Post-modernization does seem to be inherently conducive to the emergence of democratic political institutions.
Functional Analysis and Predictable Syndromes of Change
Economic, cultural, and political change go together in coherent patterns. The two most influential proponents of Modernization theory, Marx and Weber, agreed on this point. They disagreed profoundly on why economic, cultural, and political changes go together. For Marx and his disciples, they are linked because economic and technological change determines political and cultural
CULTURE,    POLITICS,    AND    ECONOMICS
15
changes. For Weber and his disciples, they are linked because culture shapes economic and political life.
Both Marx and Weber had major insights. We believe that economics shapes culture and politics—and vice versa. The causal linkages tend to be reciprocal. Political, economic, and cultural changes go together because societies without mutually supportive political, economic, and cultural systems are unlikely to survive for long: in the long run, the respective components either adapt to each other or the system flounders. And systems do indeed flounder: most of the societies that have ever existed are now extinct.
A culture is a system of attitudes, values, and knowledge that is widely shared within a society and is transmitted from generation to generation. While human nature is biologically innate and universal, culture is learned and varies from one society to another. The more central and early learned aspects of culture are resistant to change, both because it requires a massive effort to change central elements of an adult's cognitive organization, and because abandoning one's most central beliefs produces uncertainty and anxiety. In the face of enduring shifts in socioeconomic conditions, even central parts of culture may be transformed, but they are more likely to change through intergenerational population replacement than by the conversion of already socialized adults.
By culture, we refer to the subjective aspect of a society's institutions: the beliefs, values, knowledge, and skills that have been internalized by the people of a given society, complementing their external systems of coercion and exchange. This is a narrower definition of culture than is generally used in anthropology, because our purpose here is empirical analysis. We will examine the degree to which internal cultural orientations and external social institutions are linked empirically, rather than simply assume that they are. Building everything into one's definition of culture would make the concept useless for this type of analysis.
Any stable economic or political system has a compatible and supportive cultural system which legitimates that system. The people of that society have internalized a set of rules and norms. If they had not, the rulers could only get their subjects to comply with their rules by external coercion, which is costly and insecure. Moreover, to be effective in legitimating the system, cultures set limits to elite as well as mass behavior—shaping the political and economic systems, as well as being shaped by them. The process is not teleological, but it operates as if it were: societies with legitimate authority systems are more likely to survive than those without them.
Like Axelrod (1984), we find the evolutionary perspective a useful way to analyze how cultures and institutions develop: certain characteristics survive and spread because they have functional advantages in a given environment. Elster (1982) argues that functionalist interpretations of institutions are fundamentally flawed because they anthropomorphize institutions, postulating a purpose without a purposive actor—a view that has become widely accepted. But this criticism actually only applies to a crude and naive type of function-
16
CHAPTER    1
alist interpretation. Biologists today regularly use functionalist interpretations, especially when dealing with evolution. For example, plants are said to have developed bright flowers and nectar in order to attract bees so that the bees will fertilize them. Other plants are said to have developed poisonous leaves to discourage animals and insects from eating them. The newly hatched cuckoo chick pushes the other eggs out of the nest so that the parent birds will devote all their efforts to nourishing the cuckoo. And mammals living in the far North have developed white fur in order to be less visible against the snow.
Although they use this interpretation, neither biologists nor social scientists accept the crude teleological assumptions that Elster attributes to functional analysis. This mode of explanation is not used because biologists think that flowers or newly hatched cuckoos are consciously planning ahead or because they believe that evolution is guided by an anthropomorphic force. They use it because it is the most direct and parsimonious way to discuss the interaction between random mutations and natural selection that causes most mutations to die out—except for those with some functional advantage that enhances the organism's chances for survival. The mutations do not occur in order to serve some function; but they survive and spread because they do. A similar principle applies to functional interpretations of society. Dawkins (1989) argues convincingly that cultural traits or "memes" that function relatively well in a given environment replicate and spread for the same basic reason as do genes: they confer a survival advantage. Axelrod (1984) has demonstrated that certain strategies of conflict or cooperation function better than others and eventually drive out competing strategies.
Among the numerous types of societies that ever existed, the great majority have disappeared and the process is still going on. At the start of the twentieth century, absolute monarchy was the most widespread form of government. Today it has dwindled to a handful of surviving cases. Fascism spread rapidly in the 1920s and 1930s, and then all but disappeared in the 1940s, with a few loosely fascistic regimes surviving until the 1970s. The most recent case of mass extinction among societies has been the sudden collapse of communist regimes, which until recently controlled one-third of the world's population. Authoritarian state-run economies proved to be unworkable and uncompetitive in a high-technology environment. Although many of the ex-communist societies are still run by ex-communist elites, even the hard-liners among them are unlikely to return to the Stalinist model: it is a type of society that eventually proved to be dysfunctional.
Political institutions are also shaped by processes of natural selection. Some institutions survive for long periods, but most do not: three-quarters of the national constitutions now in effect were written since 1965. And even the surviving institutions undergo mutations. Thus, legislatures no.longer initiate much legislation in most societies, but they do fill a legitimating function. Legislatures themselves do not possess a conscious will to serve a legitimating function—but the fact that they fill this function is a major reason why they survive and spread. A great many new constitutions have been written in the
CULTURE,    POLITICS,    AND    ECONOMICS
17
past decade, and virtually all of them give prominent roles to legislatures. This reflects a widespread awareness that in the contemporary world those political systems that have legislatures are more likely to enjoy legitimacy and to survive and flourish than are those without them.
Is the Modernization Concept Ethnocentric?
A standard criticism of Modernization theories is that they are either ethnocentric or teleological or both. Some of the early Modernization literature did simplistically equate Modernization with becoming (1) morally superior and (2) like the West. The flaws in this perspective are pretty obvious. Few people would attribute moral superiority to Western society today, and it is evident that East Asia is now at the cutting edge of Modernization in many respects. But there is nothing ethnocentric in the concept that social change tends to take coherent, broadly predictable trajectories. In a given economic and technological environment, certain trajectories are more probable than others: it is clear that in the course of history, numerous patterns of social organization have been tried and discarded, while other patterns eventually became dominant. At the dawn of recorded history, a wide variety of hunting and gathering societies existed, but the invention of agriculture led to their almost total disappearance. They were displaced because agriculture has functional advantages over hunting and gathering. An account of the displacement of hunting-gathering societies by farming societies in precolonial Africa attributes this shift to an interaction between economic, biological, and cultural factors:
Farming and herding yield far more calories per acre than does hunting wild animals or gathering wild plants. As a result, population densities of farmers and herders are typically at least 10 times those of hunter-gatherers. That's not to say that farmers are happier, healthier or in any way superior to hunter-gatherers. They are, however, more numerous. And that alone is enough to allow them to kill or displace hunter-gatherers.
In addition, human diseases such as smallpox and measles developed from diseases plaguing domestic animals. The farmers eventually became resistant to those diseases, but hunter-gatherers do not have the opportunity. So when hunter-gatherers first come into contact with farmers, they tend to die in droves from the farmers' diseases.
Finally, only in a farming society—with its stored food surpluses and concentrated villages—do people have the chance to specialize, to become full-time metalworkers, soldiers, kings and bureaucrats. Hence the farmers, and not the hunter-gatherers, are the ones who develop swords and guns, standing armies and political organization. Add that to their sheer numbers and their germs and it is easy to see how the farmers in Africa were able to push the hunter-gatherers aside. (Diamond, 1993)
Although a few hunting and gathering societies still survive today, they comprise less than one one-thousandth of the human population. After supplanting them, agricultural societies were dominant for many centuries, until
18
CHAPTER    1
the industrial revolution finally gave rise to a fundamentally new pattern of society. The transition to industrial society is far from complete, but today almost every society on earth has at least begun to industrialize, and it seems likely that within the next century, most of humanity will live in predominantly urban industrialized societies.
This does not mean that all societies will be identical. Industrial societies have a wide variety of cultures and institutions. But their common characteristics are also striking: virtually without exception, they are characterized by high degrees of urbanization, industrialization, occupational specialization, the use of science and technology, bureaucratization, reliance on legal-rational authority, relatively high levels of social mobility and emphasis on achieved rather than ascribed social status, high levels of formal education, diminishing sex role specialization, high standards of material well-being, and much higher life expectancies than were ever achieved in agrarian or hunting and gathering societies. Hunting and agriculture will not disappear from the earth—but they will no longer be the predominant way of life. They will shape the worldview of a small minority of the population (and even the remaining hunters and farmers will have their lives transformed by the fact that they live in a predominantly urban industrial world).
It is neither ethnocentric nor teleological to assert that hunting and gathering societies gave way to agricultural societies. It is a simple historical fact. It would be ethnocentric to assert that the people living in one type of society are inherently wiser, nobler, or morally superior to those living in another—but that gratuitous claim has nothing to do with the logic of the effort to discern which type of society is most likely to survive and spread in a given economic and technological environment. The people of industrial society are not more admirable than those of agrarian society, nor does history have an anthropomorphic preference for the former; but it is clear that a majority of the world's population once shifted from hunting and gathering into the agrarian mode— and are now moving into the industrial mode. They have done so because in a given technological and economic environment, certain forms of society have functional advantages over others. Moreover, modern industrial society is not the end of history. The process of cultural evolution is still going on. This book will explore the cultural changes that go with both Modernization and Postmodernization.
For many years, it has been alleged that cultural interpretations of society are inherently conservative. This is a half-truth. The Marxist Left did indeed view emphasis on cultural factors as reactionary, but more recently the Postmodern Left has strongly emphasized the crucial role played by subjective perceptions and cultural values. From this perspective, recognizing the decisive influence of cultural factors is considered a prerequisite to social progress.
Nevertheless, there is some truth in the idea that culture itself tends to be a conservative influence. The cultural approach argues that (1) people's responses to their situation are shaped by subjective orientations that vary cross-
CULTURE,    POLITICS,    AND    ECONOMICS
19
culturally and within subcultures, and (2) these variations in subjective orientations reflect differences in one's socialization experiences, with early learning conditioning later learning, making the former more difficult to undo. Consequently, action does not simply reflect external situations. Enduring differences in cultural learning also play an essential part in shaping what people do and think.
These postulates of the cultural approach have important implications for social change. Cultural theory implies that a culture cannot be changed overnight. One may change the rulers and the laws, but to change basic aspects of the underlying culture generally takes many years. Even then, the long-run effects of revolutionary transformation are likely to diverge widely from revolutionary visions and to retain important elements of the old pattern of society. Furthermore, when basic cultural change does occur, it will take place more readily among younger groups (where it does not need to overcome the resistance of inconsistent early learning) than among older ones, resulting in intergenerational differences. An awareness of the inertia linked with cultural factors may be dismaying to those who would like to believe they have a quick fix for deep-rooted social problems. But this awareness is essential to any realistic strategy of social change, and therefore is likely to produce policies that are more effective in the long run, than a perspective which simply denies that cultural factors are important. An awareness of the fact that deep-rooted values are not easily changed is essential to any realistic and effective program for social change.
The Marxist Left saw cultural factors as opiates of the people—forms of false consciousness that could only distract the attention of the masses from the real problems, which were economic. They found it attractive to believe that the proper indoctrination could speedily wash away all previous orientations: if the right elite, guided by the one true ideology, could take power and enforce the right programs, all social problems could be quickly solved.
Unfortunately, Marxist programs designed to bring swift and massive change to entire societies overlooked the reality of cultural persistence. When these programs did not correspond to the deep-rooted values and habits of the peoples on whom they were targeted, they could be implemented only through massive coercion. The most ambitious programs of rapid social change required enormous coercion and failed nevertheless: Stalin's Forced Collectivization and Great Purges and Mao's Great Leap Forward and Great Cultural Revolution not only failed to create a New Soviet Man, or a new Chinese culture, but led to enormous human suffering and ultimately were immensely counterproductive.
The Postmodern Left tends toward the other extreme, sometimes presenting culture as virtually supreme. There are no objective limits or standards: everything is determined by one's cultural perspective—to such an extent that any reference to objective reality is viewed as almost reactionary.
Both of these extremes distort the role of culture. This book presents empirical evidence that culture is a crucial part of reality. But it is only part of it.
20
CHAPTER    1
Change Is Not Linear: Postmodernization
Another way in which early versions of Modernization theory were deficient lay in the fact that they presented a linear view of social change: the future, everywhere, would simply be more of the same. Marx's tendency to do this is particularly well known, but he had plenty of company. With the advantages of a longer time perspective, it has become evident that such linear projections are far too simple. Although industrial society has become widespread (as Marx correctly predicted), it is not the end of the road. This book presents evidence that, beyond a certain threshold, social change takes a fundamental change in direction. In the past few decades, advanced industrial societies have moved through an inflection point, from the Modernization phase into a Post-modernization phase.
This book does not examine the intellectual history of Postmodern thought and will refer to Postmodern writers only in passing. It is, instead, an empirical analysis of how a Postmodern worldview is spreading among mass publics: as it will demonstrate, a Postmodern cultural shift is taking place that manifests many of the key characteristics discussed by Postmodern thinkers. This book will not discuss how Postmodern thought developed among these writers; but we will examine the reasons why they have become widely influential. No one has fully explained why Postmodern culture has emerged: a vast amount has been written about it, but the explanation has been almost entirely at the level of the intellectual history and permeation of Postmodernism. This is an important aspect of Postmodernism, but it is not an adequate explanation of why popular culture today is strikingly different from what it was a generation or two ago. Should we assume that the masses have been profoundly influenced by the writings of Foucault and Derrida? They may have had some (largely indirect) impact. But the change is mainly due to the fact that the firsthand life experience of mass publics in recent decades has been profoundly different from that of earlier generations. Deep-rooted changes in mass world-views have taken place that enabled Postmodern ideas to find a receptive audience. This is why a Postmodern worldview that would almost certainly have been generally rejected a generation earlier has gained widespread acceptance in the last few decades.
It is not easy to give a brief account of Postmodern thought: there are several different versions of Postmodernism, and multiple readings of given authors. The literature is complex, contradictory, full of hyperbole, and sometimes reads like gibberish. Question: What is the difference between the Mafia and a deconstructionist? Answer: A deconstructionist makes you an offer you can't understand.
Ambiguity is a central component of Postmodern worldviews, and some writers seem to consider it a virtue. This is unfortunate because, underlying the ambiguous rhetoric, a real and important phenomenon is emerging. Another key tenet of Postmodernism is incredulity toward all metanarratives: all ideologies, religions, and other overarching explanations including natural sci-
CULTURE,    POLITICS,    AND    ECONOMICS
21
ence (and Postmodernism itself) cannot be believed. There is no external standard against which theories can be tested.
This perspective is carried to an extreme by Lyotard (1979) who depicts natural science as having dissolved into a relativism characterized by abrupt raptures and sudden unforeseen changes of direction. His interpretation, which has had wide influence, implies that science, like normative thought, is no longer oriented by any external reality. Baudrillard (1983) also tends toward this extreme, implying that there is no objective reality out there.2 This picture of science is one that few natural scientists would recognize. It is true, as Kuhn (1962) pointed out, that the development of knowledge is partly a social enterprise in which, when paradigm shifts occur, there is a temporary breakdown of the prevailing theoretical consensus. Kuhn's finding concerning the structure of scientific revolutions is frequently misinterpreted to mean that science itself is culture-bound. This is not the case: as we have noted, when a paradigm shift occurs, the split in acceptance is mainly along generational lines, based on different degrees of commitment to prior learning. The fact that science has a hermeneutic aspect does not mean that Indian or Chinese scientists are rejecting an interpretation that is accepted by French or German scientists. Instead, what occurs is an intergenerational culture lag.
But even these historic paradigm shifts involve much less discontinuity than Lyotard seems to imagine. Thus, Einstein's astonishing and paradigm-shifting breakthrough did not cause the previous body of scientific knowledge to be discarded. Newtonian physics continued (and continues) to function quite adequately: it simply became a special case within a broader Einsteinian framework. Many decades later, Newtonian calculations were used to take people safely to and from the moon: Einstein's limits become significant only under far more extreme conditions than are normally experienced on earth, or even in lunar voyages.
The way for Einstein's revolution was prepared by a series of findings that were inconsistent with the implications of Newtonian physics. Einstein developed a new theory that resolved these inconsistencies and generated a number of precise predictions that were then confirmed by a series of empirical tests that left little room for doubt that Einstein was right. These findings (with some delay) gave rise to a new theoretical consensus that gained acceptance from Buenos Aires to Tokyo.
Today, we seem to be on the brink of a new paradigm shift in physics—but it is unlikely to consign previous research to oblivion. Instead, the work of both Newton and Einstein will continue to apply, though within a still broader the-
2 Thus, in 1991 Baudrillard asserted with characteristic hyperbole that the Gulf War did not take place: it was all a media event (Baudrillard, cited in Lyon, 1994: 52). But whether or not the war took place was not simply a question of one's cultural perspective: thousands of corpses testified to the fact that it was a reality. From the opposite end of the ideological spectrum, German revisionist historians have argued that the Holocaust did not really take place—it is just a case of the victors writing history. In this case, millions of corpses constitute a fact that goes beyond questions of interpretation.
22
CHAPTER    I
oretical framework. The emerging Grand Unified Theory is designed to integrate all of the laws of physics into one coherent theory that will account for everything that has happened in the physical world from the birth of the universe to the present moment. Far from disintegrating into discontinuous and mutually incomprehensible islands of short-lived insights, natural science seems to be moving toward a mega-metanarrative. This is precisely the opposite of what Lyotard's followers seem to believe.
Nevertheless, stripped of its hyperbolic extremes, the literature on Post-modernity is dealing with a very real and important phenomenon: the world (or, at least, large parts of it) has moved onto a different trajectory from the one it had been following since the industrial revolution. And this new trajectory corresponds, in many respects, to what Postmodern observers claim is happening. Although there still is an external reality out there, culture does indeed have a tremendous influence on how reality is perceived. Moreover, the relative importance of culture seems to be increasing. On this new Postmodern trajectory, economic rationality determines human behavior less narrowly than before: the realm of the possible has expanded, and cultural factors are becoming more important. An empirically demonstrable cultural shift is taking place. The great religious and ideological metanarratives are losing their authority among the masses. The uniformity and hierarchy that shaped modernity are giving way to an increasing acceptance of diversity. And the increasing dominance of instrumental rationality that characterized Modernization is giving way to a greater emphasis on value rationality and quality of life concerns.
As this book will demonstrate with empirical evidence, a Postmodern shift in mass values and attitudes actually is taking place. This is why the ideas of Postmodern writers have found a receptive audience in recent decades. Although our analysis of empirical evidence cannot solve Postmodernity's normative questions, it does enable us to identify where the Postmodern shift is occurring and how fast it is moving, and it helps explain why it is taking place.
Has the entire world suddenly turned Postmodern, as some writers seem to assume? The empirical answer is No. Instead, some societies (such as Nigeria) are starting to modernize; others (such as China) are now modernizing very rapidly; still others (such as South Korea), seem to be reaching a turning point where they may be about to begin Postmodernization; and still others, such as Britain, Germany, and the United States, are well into the Postmodernization process—but even they do not lead the world in this respect. As we will see, the evidence indicates that the Nordic countries and the Netherlands are now the most Postmodern societies on earth.
This book will not merely chart the progress of Postmodernization; we will propose a theoretical explanation of why it is taking place. Before doing so, let us try to categorize Postmodernist thought: to a large extent, the changes that are occurring among mass publics correspond to these ideas. But how true this is depends on what version of Postmodernism one has in mind.
One could start by dividing Postmodern thought into three broad schools:
CULTURE,    POLITICS,    AND    ECONOMICS
23
1.   Postmodernism is the rejection of modernity: that is, of rationality, authority, technology, and science. Within this school, there is a widespread tendency to equate rationality, authority, technology, and science with Westernization. From this perspective, Postmodernism is considered to be the rejection of Westernization.
2.  Postmodernism is the revalorization of tradition. Since Modernization drastically devalued tradition, its demise opens the way for this revalorization.
3.  Postmodernism is the rise of new values and lifestyles, with greater tolerance for ethnic, cultural, and sexual diversity and individual choice concerning the kind of life one wants to lead.
These three versions of Postmodernism all capture important elements of what is taking place; though they are not incompatible, they emphasize different things.
Let us start with the rejection of modernity. Modernization offers great rewards, but imposes huge costs. It dismantles a traditional world in which the meaning of life is clear; warm, personal communal ties give way to an impersonal competitive society geared to individual achievement. Industrialization vastly increases human productivity; but (especially before labor unions and working-class political parties bring countervailing pressures to bear against capitalism) it gives rise to inhuman working conditions. Marx criticized not only the ruthless economic exploitation of early capitalism, but also the tremendous psychological costs of industrialization.
Decades later, Weber saw the rationalization of society as an inexorable aspect of Modernization; though it facilitated economic growth and public order, ever-increasing rationalization was disenchanting the world, forcing humanity into a painfully narrow iron cage of bureaucracy and mass production. What Weber deplored was the ubiquitous penetration of instrumental rationality: the rationality of immediate means was driving out the rationality of ultimate ends. Subsequently, Heidegger (1946, 1949) and Horkheimer and Adorno (1947) carried the critique of modernity farther, arguing that the instrumental rationality of industrialization had, ironically, undermined any absolute moral standards and given rise to new forms of irrationality and repression, culminating in the horrors of Hitler and Stalin. Instrumental rationality had virtually banished value rationality.
Today, this trend is beginning to reverse itself: instrumental rationality gained an exaggerated predominance during the rise of industrialization, but today, for reasons we will discuss in this book, a growing segment of society is concluding that the price is too high. Rationality, science, technology, and authority are here to stay; but their relative priority and their authority among mass publics are declining.
Within this first version of Postmodernism, there is a widespread tendency to confound rationality, authority, technology, and science with Westernization. Some of the (now outmoded) Modernization literature also equated Modernization with Westernization. If Postmodernism is the rejection of modernity, it would logically follow that Postmodernism is the rejection of
24
CHAPTER    1
Westernization. This perspective is found in the work of Lyotard and Derrida, who tend to equate modernization with Western imperialism.
Western imperialism was an important phenomenon: it was brutally imposed on the rest of the world, it deserved to be rejected, and it deserves the scorn with which Postmodern writers treat it. But equating Modernization with Westernization is not a useful way to proceed. It emphasizes superficial and accidental aspects of Modernization and ignores the core process. Wearing Western clothing was not crucial; industrialization was. Moreover, it is inaccurate to equate modern imperialism with Westernization. In the number of people it subjugated, the Japanese empire was the second largest colonial empire in history and was fully as oppressive as any Western empire.
The essential core of Modernization is a syndrome of changes closely linked with industrialization: this syndrome includes urbanization, the application of science and technology, rapidly increasing occupational specialization, rising bureaucratization, and rising educational levels. It also includes one more thing, which was the motivating force behind the whole process: industrialization was a way to get rich.
By getting rich, one could dispel hunger, acquire military strength, and obtain a number of other desirable things, including a much longer life expectancy than was possible in preindustrial society. Adopting a life strategy aimed at getting rich becomes compellingly attractive from the perspective of low-income societies, once it has been demonstrated that it can be done. Furthermore, as we will show in this chapter, economic development actually seems to be conducive to subjective well-being (though only up to a certain point in history). In short, industrialization and the Modernization syndrome that goes with it were an attractive package. It carries a high cost, and from the viewpoint of advanced industrial society these costs may seem excessive. But from the perspective of most preindustrial societies, it seemed worth the price.
This constitutes another crucial difference between Modernization and Westernization: Western imperialism was imposed on non-Western societies, which almost universally rejected it when they were free to do so. By contrast, the goal of Modernization (that is, the industrialization syndrome) has now been adopted by almost every society on earth—and non-Western societies show no sign of wishing to abandon it. Quite the contrary, it is being pursued today with far more enthusiasm in the non-Western world than in the West. The Postmodern critique of Modernization comes overwhelmingly from within Western societies.
By the 1960s, the tendency to equate Modernization with Westernization had been abandoned by most Modernization theorists. And even if one goes by obvious external indicators, this concept has been outdated since at least 1980, when Japan became the world's leading automobile producer—outdoing the United States at Fordism itself. During the ensuing decade, Japan also attained the highest GNP per capita of any major nation, leading the world in attaining the fruits, as well as the tools, of Modernization. Historically, the Industrial Revolution occurred first in the West. But there is nothing uniquely
CULTURE,    POLITICS,    AND    ECONOMICS
25
Western about technology and industrialization, or even bureaucratic rationality. Mathematics came to Europe from India and Egypt. China was the technologically most advanced society in the world for most of the past 2,000 years, losing its technological lead only in the seventeenth century (and it is not inconceivable that the nation will regain it). Similarly, another key aspect of modernity—bureaucracy—originated in China. The idea that rationality and technology are Western inventions is simply a myth. In the modern era, Westerners raised them to unprecedented levels and applied them to production to an unprecedented degree, but they are part of the human heritage, not something uniquely European. Today, East Asian and Southeast Asian societies are achieving the world's highest rates of economic growth and are at the cutting edge of Modernization in numerous other respects. Japan has become the world leader in various aspects of modernity, from consumer electronics to human life expectancy. And in recent years a growing flow of Western experts have made the pilgrimage to Japan to study the secrets of Japanese management, just as the Japanese earlier made the reverse voyage to learn industrialization from the West.
Another perspective views Postmodernism as the revalorization of tradition. This reverses one of the most prominent trends associated with Modernization. In the early modern era, the astonishing achievements of science and industry gave rise to a myth of Progress and radically discredited tradition. "New" became virtually synonymous with "good." But more recently, the instrumental rationality of modernity has lost its prestige. This has not only opened the way for tradition to regain status, but created a need for a new legitimating myth. In the Postmodern worldview, tradition once again has positive value—especially non-Western traditions. But the revalorization of tradition is sharply selective. Despite their ubiquitous presence in the traditional societies of both the Western and non-Western world, the norm that "Women's place is in the home" and the stern prohibition of extramarital sex are not among the aspects of premodern tradition that Postmodern writers admire.
The rise of new values and lifestyles is a profoundly important aspect of what is taking place today, throughout advanced industrial society. Derrida (1979, 1981) emphasizes this aspect of Postmodernity. Although Postmodern-ization does involve a downgrading of modernity and a revalorization of tradition, the emergence of a new culture is even more crucial, in our view. The best documented example of the rise of new values is the intergenerational shift from Materialist to Postmaterialist value priorities that seems to be taking place throughout advanced industrial society (Inglehart, 1971, 1977, 1990); but the rise of new values and lifestyles is taking place across many other aspects of life, from sexual orientation to religion.
Critical Theory
Apart from the Postmodern thinkers, Habermas (1984, 1987) has developed the most influential recent philosophical critique of modernity. Habermas dif-
26
CHAPTER    1
fers from the Postmodern school on a number of points. One major disagreement is that, while Postmodernism tends to depict Modernization as a basically bad choice and rejects it, Habermas argues that while it imposed high costs, it also brought major benefits. Modernization is an unfinished project; we should build on it rather than reject it. Although we think that the process of change has taken a fundamentally new Postmodern turn, we agree with Habermas on this point. Industrialization provided more than just noisy, polluting automobiles and mindless television sitcoms. It provided two things that would be considered valuable from almost any cultural perspective: (1-) greatly enhanced chances for survival, as measured by human life expectancy, and (2) higher levels of subjective well-being. Empirical evidence will be presented below in support of these assertions.
Another major disagreement centers on the fact that Postmodern thinkers conclude that there is no longer any basis by which universal moral standards could be validated: both God and Marx are dead. Habermas has not given up: he argues that moral norms may be merely social conventions, but if they are, it is imperative to develop rules for arriving at universally acceptable conventions. In a new version of the social contract, Habermas argues that a rational basis for collective life can be achieved only when social relations are organized so that the validity of every norm depends on a consensus arrived at in communication free from domination. Against the Postmodern position that moral rules are simply myths created by the ruling elite to justify the social order they control, Habermas argues that it is possible to reach a moral consensus that is not simply dominated by the ruling elites. Here again, we think he is right, and this debate raises a crucial question: Are cultural norms simply tools of the ruling elite? In order to answer this question, let us examine the relationship between authority and culture.
Authority and Culture
Marx defined ideology as false consciousness—that is, a consciousness shaped by power-holders to justify their right to rule (and to exploit), and to make it seem inevitable. The insight that culture is closely linked with power is important. It would be naive to believe that culture is neutral: in virtually every society, it legitimates the established social order—partly because the dominant elite try to shape it to help perpetuate their rule.
One of the leading themes in the literature on Postmodernism is the claim that culture is used to legitimate political authority; Foucault is a prominent advocate of this view. An extreme version of this position would hold that every reality is a politically constructed system of myths, and the key task of the social critic is to deconstruct these myths, which are simply a means to justify privilege and exploitation.
Without a doubt, culture serves to legitimate the social order. From an elite perspective, this may even be the most important thing it does. But it certainly is not the only thing it does. Culture integrates society in terms of common
CULTURE,    POLITICS,    AND    ECONOMICS
27
goals, satisfies intellectual and aesthetic needs, and finally—no insignificant point—also places some restraints on elites.
The extreme position, that mass belief systems are completely dominated by elite interests, assumes a degree of mass manipulability that is simply unrealistic. Recent historical developments illustrate this point. Thus, after 70 years of controlling the Soviet Union's educational systems, public discussion, the mass media, churches, and all other channels of communication to an historically unprecedented extent, the Soviet elite ultimately was not able to shape the worldviews of their people to conform to their goals: toward the end, not even the Soviet elite really believed the official ideology.
Western advanced industrial societies are also changing—whether their elites like it or not. A modern worldview that was once firmly established has gradually given way to Postmodern values that emphasize human autonomy and diversity instead of the hierarchy and conformity that are central to modernity. In both cases, a major factor leading to basic cultural change was the fact that the life experience of a new generation gave rise to new perceptions of reality. For the reality of one's firsthand experience ultimately intrudes. The official truth, propagated by the dominant elite, usually has a great deal of influence. But the firsthand life experience of ordinary people also counts—and ultimately may have even greater credibility than the official truth. How do established worldviews begin to crumble?
Why Is the Postmodern Shift Occurring?
The shift toward Postmodern values is not the first time that a major cultural shift has occurred. The transition from agrarian society to industrial society was facilitated by a shift from a worldview shaped by a steady-state economy. This worldview discouraged social mobility and emphasized tradition, inherited status, and communal obligations, backed up by absolute religious norms; it gave way to a worldview that encouraged economic achievement, individualism, and innovation, with increasingly secular social norms. Today, some of these trends linked with the transition from "Traditional" to "Modern" society have reached their limits in advanced industrial society, where change is taking a new direction.
This change of direction reflects the principle of diminishing marginal utility. Industrialization and Modernization required breaking the cultural constraints on accumulation that are found in any steady-state economy. In Western European history, this was achieved by the rise of the Protestant Ethic, which (though it had a long intellectual history) was like a random mutation from a functional perspective. If it had occurred two centuries earlier it might have died out. In the environment of its time, it found a niche: technological developments were making rapid economic growth possible, and the Calvinist worldview complemented these developments beautifully, forming a cultural-economic syndrome that led to the rise of capitalism and eventually to
28
CHAPTER    J
the industrial revolution. Once this had occurred, economic accumulation (for individuals) and economic growth (for societies) became the top priorities for an increasing part of the world's population; they are still the central goals for much of humanity. But eventually, diminishing returns from economic growth lead to a Postmodern shift that in some ways constitutes the decline of the Protestant Ethic.
Advanced industrial societies are now changing their sociopolitical trajectories in two fundamental respects:
1.  Value systems. Increasing emphasis on individual economic achievement was one of the crucial changes that made Modernization possible. This shift toward Materialistic priorities entailed a de-emphasis on communal obligations and an acceptance of social mobility: increasingly, social status became something that an individual could achieve, rather than something into which one was born. Economic growth came to be equated with progress and was seen as the hallmark of a successful society.
In Postmodern society this emphasis on economic achievement as the top priority is now giving way to an increasing emphasis on the quality of life. In a major part of the world, the disciplined, self-denying, and achievement-oriented norms of industrial society are giving way to an increasingly broad latitude for individual choice of lifestyles and individual self-expression. The shift from "Materialist" values, emphasizing economic and physical security, to "Postmaterialist" values, emphasizing individual self-expression and quality of life concerns, is the most amply documented aspect of this change; but it is only one component of a much broader syndrome of cultural change.
2.  Institutional structure. We are also reaching limits to the development of the hierarchical bureaucratic organizations that helped create modern society. The bureaucratic state, the disciplined, oligarchical political party, the mass-production assembly line, the old-line labor union, and the hierarchical corporation all played enormously important roles in mobilizing and organizing the energies of masses of people; they made the industrial revolution and the modern state possible. But they have come to a turning point for two reasons: first, they are reaching limits in their functional effectiveness; and second, they are reaching limits in their mass acceptability. Let us consider both factors.
Functional Limits to the Expansion of the Bureaucratic State
The rise and fall of the Soviet Union illustrates the limits of the centralized, hierarchical state. In its early decades, the USSR was remarkably efficient in mobilizing masses of relatively unskilled workers and vast quantities of raw materials to build the world's largest steel mill and the world's largest hydroelectric dam, and to attain one of the fastest rates of economic growth in the world. Although Stalin starved and murdered millions of Soviet citizens, the economic and military achievements of the Soviet state were so impressive that they convinced many people throughout the world that this type of society was the irresistible wave of the future. Soviet economic growth was re-
C V L T U R E ,    POLITICS,    AND    ECONOMICS
29
markable in the 1950s, was still impressive in the 1960s, tapered off in the 1970s, and stagnated in the 1980s. Partly, this happened because a hypertro-phied bureaucracy paralyzed adaptation and innovation. Bureaucracy is inherently deadening to innovation, and this problem became acute once the Soviet Union had moved past the stage of simply importing already proven technology from the West and was attempting to innovate in competition with the West and Japan. But the problem was not only the failure of central economic planning to cope with an increasingly complex and rapidly changing society. It also reflected a collapse of motivation and morale. Absenteeism rose to massive proportions, alcoholism became a tremendous problem, and confidence in government eroded until finally the entire economic and political system collapsed. Although the Soviet example is the most striking case, similar manifestations of the diminishing effectiveness of hierarchical, centralized bureaucratic institutions can be seen throughout industrial society. State-run economies are giving way to market forces; old-line political parties and labor unions are in decline; and bureaucratic corporations are losing ground to more loosely organized and participatory types of organization.
These organizational and motivational changes are intimately related. One reason for the decline of the classic bureaucratic institutions of industrial society is the fact that they are inherently less effective in high-technology societies with highly specialized workforces than they were in the earlier stages of industrial society. But another reason for their decline is the fact that they also became less acceptable to the publics of Postmodern society than they were earlier, because of changes in these people's values.
The mass production assembly line broke down manufacturing into simple standardized routines that were repeated endlessly. This was marvelously effective in turning out masses of relatively simple, standardized products. But a price was paid for the increased productivity that resulted: the workers became cogs in huge centrally coordinated machines. Marx, Weber, and others were concerned with the alienation and depersonalization of industrial society that made one's work uninteresting, dehumanizing, devoid of meaning. In societies of scarcity, people were willing to accept these costs, for the sake of economic gains. In affluent societies, they are less willing to do so.
Modern bureaucracy makes a similar tradeoff involving loss of individual identity and autonomy for the sake of increased productivity; this enables it to process thousands or millions of people, using standardized routines. It, too, is inherently depersonalizing: in a rational bureaucracy, individuals are reduced to interchangeable roles. Bureaucracy strips away spontaneity, personal likes and dislikes, individual self-expression and creativity. Nevertheless it was an effective tool for coordinating the efforts of hundreds or even millions of individuals, in the large organizations of modern society.
But its effectiveness and its acceptability are eroding. Postmodern values give a higher priority to self-expression than to economic effectiveness: people are becoming less willing to accept the human costs of bureaucracy and of rigid social norms. As this book will demonstrate, Postmodern society is char-
30
CHAPTER    1
acterized by the decline of hierarchical institutions and rigid social norms, and by the expansion of the realm of individual choice and mass participation.
Up to the middle of the twentieth century, "Modernization" was an unambiguous term. It referred to urbanization, industrialization, secularization, bureaucratization, and a culture based on bureaucratization—a culture that requires a shift from ascriptive status to achieved status, from diffuse to specific forms of authority, from personalistic obligations to impersonal roles, and from particularistic to universalistic rules. In some areas this Modernization process is still going on. But elsewhere, trends that were central to the Modernization process have undergone a fundamental change of direction.
For example, one of the most striking phenomena of the past two hundred years was the rapidly expanding scope of government. Industrial societies became increasingly centralized, hierarchical, and bureaucratized. Until recently, highly centralized state-run economies and societies like the Soviet Union seemed to be the logical end point of Modernization. One might view this trend as profoundly progressive, with the Marxists, or deplore it as threatening to human liberty, with Schumpeter (1947) and Orwell (1949)—but the growth of government seemed inexorable. At the start of the twentieth century, government spending in most societies consumed from 4 to 10 percent of gross domestic product. By 1980, it ranged from 33 to 60 percent of a much bigger output in Western societies, and 70 to 80 percent in some socialist societies. Increasing government ownership and control of the economy seemed to be the wave of the future.
It was not. During the 1980s, further expansion of the state reached a point of diminishing returns, both functionally and in terms of mass acceptance. It first ran into growing political opposition in the West and then collapsed in the Eastern bloc.
The mass production assembly line and the mass production bureaucracy were the two key organizational instruments of industrial society, and in the early phase of Modernization they had a high payoff—enabling factories to produce millions of units and governments to process millions of individuals through standardized routines. But the trend toward bureaucratization, centralization, and government ownership and control has reversed itself. Modern economies lose their effectiveness when the public sphere becomes overwhelmingly large. And public confidence in hierarchical institutions is eroding throughout advanced industrial society.
Cultural Changes Leading to Postmodernization
An equally basic change in the direction of change has been a shift in the predominant norms and motivations underlying human behavior. Virtually all agrarian societies were characterized by value systems that stigmatized social mobility. This was inevitable, given their steady-state economies. The main source of wealth was land, which is in fixed supply: the only way to become rich was by seizing someone else's land—which probably required killing the
CULTURE,    POLITICS,    AND    ECONOMICS
31
owner. Such internal violence was threatening to the survival of any society and was repressed by norms that emphasized acceptance of the status into which one was born and stigmatized the ambitious and the arriviste. At the same time, traditional societies emphasized duties of sharing and charity— which helped compensate the poor for the absence of social mobility, but further undermined the legitimacy of economic accumulation.
The rise of a Materialistic value system that not only tolerated economic accumulation but encouraged it as something laudable and heroic was a key cultural change that opened the way for capitalism and industrialization. Weber (1904-5) examined this process in The Protestant Ethic and the Spirit of Capitalism, but his work can be seen as a case study of a more general phenomenon. Today the functional equivalent of the Protestant Ethic is operating most vigorously in East Asia and is fading away in Protestant Europe, as technological development and cultural change have become global phenomena.
Precisely because they attained high levels of economic security, the populations of the first nations to industrialize have gradually come to emphasize Postmaterialist values, giving higher priority to the quality of life than to economic growth. This shift has been taking place throughout advanced industrial society during the past few decades, as we will see in chapter 4. With this has come a shift from the politics of class conflict, to political conflict based on such issues as environmental protection and the status of women and sexual minorities. Marxist ideology, based on economic determinism, was an immensely influential guide for interpreting the transition from agrarian to "modern" or industrial society. It is outmoded for the analysis of "Postmodern" society.
To clarify what we mean by this term, let us examine the specific changes that are linked with Postmodern values. Some of these trends differ radically from those of Modernization.
The Origins of Postmodern Values: Existential Security
A new worldview is gradually replacing one that has dominated Western society since the Industrial Revolution. The consequences of this transformation are still taking shape, and elements of the older culture are still widespread, but the major features of the new pattern can be discerned.
This shift in worldview and motivations springs from the fact that there is a fundamental difference between growing up with an awareness that survival is precarious, and growing up with the feeling that one's survival can be taken for granted.
The urge to survive is common to all creatures, and normally survival is precarious. This reflects a basic ecological principle: the population of any organism tends to rise to meet the available food supply; it is then held constant by starvation, disease, or predators. Throughout most of history, this principle has governed the lives of all organisms, including humanity. Until very recently, the survival of most human beings was precarious.
32
CHAPTER    I
Eventually, culture began to soften the competition for survival among humans. Although the ways in which this was done varied enormously from one society to another, virtually all traditional societies established cultural norms that limited the use of violence and repressed aspirations for social mobility. On one hand, they emphasized sharing and charity among those who were relatively well-off, stigmatizing accumulation as greed; and on the other hand, they justified acceptance of the existing social order by the poor. And cultural norms limiting reproduction softened the ruthless competition for survival that overpopulation brought.
A few centuries ago, cultural changes in Protestant Europe led to the reversal of the traditional stigma against economic accumulation, and a Materialistic worldview began to spread. Using new technology and organizational techniques, production began to outpace population growth. Nevertheless, well into the twentieth century, severe economic scarcity still prevailed widely: the Marxist view that people and history were motivated primarily by the struggle for economic goods was a fairly accurate first approximation of the driving force underlying the modernizing phase of industrial society.
The economic miracles and the welfare states that emerged after World War II gave rise to a new stage of history, and ultimately laid the way for the rise of Postmodern values. Fundamental changes in formative experiences have given rise to a distinct value system among a growing segment of those raised in advanced industrial societies during the years since World War II. The postwar birth cohorts in these societies grew up under conditions profoundly unlike those that shaped previous generations. They differed in two respects: first, the postwar economic miracles produced levels of prosperity that were literally unprecedented in human history. Real per capita income in most industrial societies rose to levels several times as high as had ever been experienced before the war, and in some cases (such as Japan) to levels 20 or 30 times higher than ever before. The economic pie became much bigger; this alone would tend to encourage a greater sense of economic security.
But the impact of unprecedented prosperity interacted with a second factor: the emergence of the modern welfare state. A sense of existential security, not absolute wealth, is the crucial variable, and the welfare state reinforced economic growth in producing a sense of security. The pie was much bigger than ever before, and it was distributed more evenly and more reliably than before. For the first time in history, a large share of the masses grew up with the feeling that survival could be taken for granted.
This led to a process of intergenerational value change that is gradually transforming the politics and cultural norms of advanced industrial societies. The best documented aspect of this process is the shift from giving top priority to economic and physical security, to giving top priority to self-expression and the quality of life. This shift from Materialist to Postmaterialist priorities has been measured annually since 1970 in surveys carried out in a number of Western societies. A massive body of evidence is now available, and it demonstrates that an intergenerational shift has been taking place in the predicted di-
CULTURE,    POLITICS,    AND    ECONOMICS
33
rection. This shift from Materialist to Postmaterialist value priorities has brought new political issues to the center of the stage and provided much of the impetus for new political movements.
More recent research indicates that the rise of Postmaterialism itself is only one aspect of a still broader process of cultural change that is reshaping the political outlook, religious orientations, gender roles, and sexual mores of advanced industrial society (Inglehart, 1990). These changes are related to a common concern: the need for a sense of security that religion and absolute cultural norms have traditionally provided. In advanced industrial societies during the decades since World War II, the emergence of unprecedentedly high levels of prosperity, together with the relatively high levels of social security provided by the welfare state, have contributed to a decline in the prevailing sense of vulnerability. For the general public, one's fate is no longer so heavily influenced by unpredictable forces as it was in agrarian and early industrial society. This has been conducive to the spread of Postmodern orientations that place less emphasis on traditional cultural norms—especially those norms that limit individual self-expression.
The Theory of Intergenerational Value Change
Let us reexamine the theory of intergenerational value change in light of recent findings. Our theory is based on two key hypotheses (Inglehart, 1977):
1.  A Scarcity Hypothesis. An individual's priorities reflect the socioeconomic environment: one places the greatest subjective value on those things that are in relatively short supply.
2.  A Socialization Hypothesis. The relationship between socioeconomic environment and value priorities is not one of immediate adjustment: a substantial time lag is involved because, to a large extent, one's basic values reflect the conditions that prevailed during one's preadult years.
The scarcity hypothesis is similar to the principle of diminishing marginal utility in economic theory. The complementary concept of a need hierarchy (Maslow, 1954) helped shape the survey items used to measure value priorities. In its simplest form, the idea of a need hierarchy would probably command almost universal assent. The fact that unmet physiological needs take priority over social, intellectual, or aesthetic needs has been demonstrated all too often in human history: starving people will go to almost any length to obtain food. The rank ordering of human needs varies as we move beyond those needs directly related to survival; Maslow's need hierarchy does not hold up in detail. But there does seem to be a basic distinction between the "material" needs for physiological sustenance and safety, and nonphysiological needs such as those for esteem, self-expression, and aesthetic satisfaction.
The recent economic history of advanced industrial societies has significant implications in light of the scarcity hypothesis. For these societies are a striking exception to the prevailing historical pattern: they still contain poor peo-
34
CHAPTER    I
pie, but most of their population does not live under conditions of hunger and economic insecurity. This has led to a gradual shift in which needs for belonging, esteem, and intellectual and aesthetic satisfaction became more prominent. Other things being equal, we would expect prolonged periods of high prosperity to encourage the spread of Postmaterialist values; economic decline would have the opposite effect.
But it is not quite that simple: there is no one-to-one relationship between economic level and the prevalence of Postmaterialist values, for these values reflect one's subjective sense of security, not one's economic level per se. While rich individuals and nationalities tend to feel more secure than poor ones, these feelings are also influenced by the cultural setting and social welfare institutions in which one is raised. Thus, the scarcity hypothesis must be interpreted in connection with the socialization hypothesis.
One of the most pervasive concepts in social science is the notion of a basic human personality structure that tends to crystallize by the time an individual reaches adulthood, with relatively little change thereafter. This concept permeates the literature from Plato through Freud and extends to the findings of contemporary survey research. Early socialization seems to carry greater weight than later socialization.
This, of course, does not imply that no change occurs during adult years. In individual cases, dramatic behavioral shifts are known to occur, and the process of human development never comes to a complete stop (Erikson, 1982; Levinson et al., 1979; Brim and Kagan, 1980). Nevertheless, human development seems to be far more rapid during the preadult years than afterward, and the great bulk of the evidence points to the conclusion that the statistical likelihood of basic personality change declines sharply after one reaches adulthood (Block, 1981; Costa and McCrae, 1980; Jennings and Niemi, 1981; Jennings and Markus, 1984).
Taken together, these two hypotheses generate a clear set of predictions concerning value change. First, while the scarcity hypothesis implies that prosperity is conducive to the spread of Postmaterialist and Postmodern values, the socialization hypothesis implies that neither an individual's values nor those of a society as a whole are likely to change overnight. Instead, fundamental value change takes place gradually; largely it occurs as a younger generation replaces an older one in the adult population of a society.
Consequently, after a period of sharply rising economic and physical security, one would expect to find substantial differences between the value priorities of older and younger groups: they would have been shaped by different experiences in their formative years. But there would be a sizable time lag between economic changes and their political effects. Ten or 15 years after an era of prosperity began, the age cohorts that had spent their formative years in prosperity would begin to enter the electorate. A decade or so might pass before these groups began to occupy positions of power and influence in their society; another decade or so would pass before they reached the level of top decision makers. But their influence would become important long before this
CULTURE,    POLITICS,    AND    ECONOMICS
35
final stage. Postmaterialists are more highly educated, more articulate, and politically more active than Materialists. Consequently, their political impact tends to outweigh that of the Materialists.
The socialization hypothesis complements the scarcity hypothesis. It helps account for apparently deviant behavior: on one hand, the miser who experienced poverty in early years and relentlessly continues piling up wealth long after attaining material security; and on the other hand, the saint who remains true to the higher-order goals instilled by his or her culture, even in the face of severe deprivation. In both instances, an explanation for the seemingly deviant behavior of such individuals lies in their early socialization.
The unprecedented economic and physical security of the postwar era has led to an intergenerational shift from Materialist to Postmaterialist values. The young emphasize Postmaterialist goals to a far greater extent than do the old, and cohort analysis indicates that this reflects generational change rather than aging effects. At the time of our first surveys, in 1970-71, Materialists held an overwhelming numerical preponderance over Postmaterialists, outnumbering them by nearly four to one. By 1990, the balance had shifted dramatically, to a point where Materialists outnumbered Postmaterialists by only four to three. Projections based on population replacement suggest that by the year 2000 Materialists and Postmaterialists will be about equally numerous in many Western countries (Abramson and Inglehart, 1992).
Postmaterialists are not non-Materialists, still less are they anti-Materialists. The term "Posr-materialist" denotes a set of goals that are emphasized after people have attained material security, and because they have attained material security. Thus, the collapse of security would lead to a gradual shift back toward Materialist priorities. The emergence of Postmaterialism does not reflect a reversal of polarities, but a change of priorities: Postmaterialists do not place a negative value on economic and physical security—they value it positively, like everyone else; but unlike Materialists, they give even higher priority to self-expression and the quality of life.
Thus, Inglehart (1977: 179-261) found that an emerging emphasis on quality of life issues was being superimposed on the older, class-based cleavages of industrial society. Although social class voting was declining, it had by no means disappeared (and was unlikely to do so). But while the old class-based polarization over ownership and control of the means of production had once dominated politics, it was increasingly sharing the stage with new Postmaterialist issues. Both industrial and preindustrial cleavages persisted, beside cross-cutting new issues.
The shift from Materialist to Postmaterialist priorities is a core element of the Postmodernization process. In early industrial society, emphasis on economic achievement rose to unprecedented levels. While traditional societies stigmatized social mobility and individual economic accumulation, modern industrial societies provided a positive evaluation of economic achievement. The Captain of Industry became a cultural hero, and the nineteenth-century U.S. Supreme Court interpreted "the pursuit of happiness" to mean "freedom
36
CHAPTER    I
to accumulate property." The core societal goal of the Modernization process was economic growth. This made a good deal of sense. Early industrializing nations had only recently acquired the technological means to cope with chronic scarcity. In such societies, where malnutrition is the main cause of death, economic achievement is an overwhelmingly important part of the pursuit of happiness. The transition from preindustrial society to advanced industrial society brings a change from a life expectancy of 35 or 40 years, to one of 75 or 80 years. This is a huge improvement.
As the possibility of starvation receded from being a major concern to an almost insignificant prospect for most people, prevailing values gradually changed. Economic security is still something that everyone wants, but it is no longer a synonym for happiness. Increasingly, the publics of advanced industrial societies have come to emphasize quality of life concerns, sometimes giving environmental protection priority over economic growth. Thus, emphasis on economic achievement rises sharply with the Modernization process, but then levels off as Postmodernization occurs. Societies in which Postmaterial-ists are most numerous have lower growth rates than those in which Materialists are overwhelmingly predominant—but the former tend to have higher levels of subjective well-being. Postmodernization brings declining emphasis not only on economic growth itself, but also on the scientific and technological developments that make it possible; emphasis shifts from coping with survival, to maximizing subjective well-being.
The Risk Society
Ironically, as survival has become unprecedentedly secure, the peoples of advanced industrial societies have become increasingly sensitive to risk. Indeed, one of the most influential critics of postmodern society characterizes it as Risk Society (Beck, 1992). According to this diagnosis, the distributional conflicts over "goods" (such as property, income, and jobs) that characterized industrial society have given way to distributional conflicts over "bads," such as the risks of nuclear technology, genetic research, and the threat to the environment. With industrialization, the religious certainties of feudal society were eroded, but they gave rise to an increasing degree of existential security; with the rise of Postmodern society, the risks of life have become incalculable and increasingly escape the control mechanisms of society. In this updated version of the doctrine of late capitalism, the ecological crisis takes over the role previously played by the legitimation crisis of late capitalism.
It is ironic that in societies where human life expectancy has risen by 20 years during the last century, concerns about risk have become central political issues. It is ironic, but logical: for it is precisely because the risk of starvation has receded almost to the vanishing point that people have been able to redirect their concerns from pervasive daily uncertainty concerning survival to more remote concerns such as the ecological crisis. The very success of the welfare states of advanced industrial society in providing an unprecedented de-
CULTURE,    POLITICS,    AND    ECONOMICS
37
gree of existential insecurity has given rise to the expectation that the state can and should ensure everyone against all uncertainties. As Samuelson has put it,
The reason for this paradox is entitlement: a postwar word and concept. By entitlement, I mean more than the catalogue of well-known government benefits (Social Security being the most prominent) or various modem "rights" (such as the "right" of those in wheelchairs to public ramps). Entitlement expresses a modern conviction, a broader sensibility, that defines Americans' attitudes toward social conditions, national institutions and even the world. Increasingly, we have come to believe that certain things are (or ought to be) guaranteed to us. We feel entitled. Among other things, we expect secure jobs, rising living standards, enlightened corporations, generous government, high-quality health care, racial harmony, a clean environment, safe cities, satisfying work, and personal fulfillment. (Samuelson, 1995: 4)
What Samuelson attributes to American society holds true of other Postmodern societies. As long as people were overwhelmingly engaged in coping with survival, more remote concerns had little salience. But the attainment of existential security does not bring Nirvana. Postmodern society has brought increasing attention to quality of life problems, and far more demanding standards for societal performance. As a net result, people probably worry as much as ever, but they worry about different things: there are profound differences in the behavior and worldviews of people who feel insecure about their personal survival and people who worry about global warming.
Stress, Coping Stategies, and Belief Systems
Far-reaching though it is, the rise of Postmaterialist values is only one aspect of a still broader process of cultural change that is reshaping orientations toward authority, religion, politics, gender roles, and sexual norms among the publics of advanced industrial society. What is driving this broad shift from survival values toward well-being values? This question is illuminated by recent research in social psychology on the relationships between stress, coping strategies, and belief systems.
People who feel that their survival is threatened react with stress; this stimulates efforts to cope with the threat. But high levels of stress can become dysfunctional and even life-threatening. One's belief system mediates the response to new or threatening situations, helping the individual deal with stress and shaping the strategy used to cope with the threat. If one has a belief system that provides a sense of predictability and control, it reduces stress to a level conducive to coping behavior (Rotter, 1966). In the absence of such a belief system, people experience a sense of helplessness, leading to withdrawal instead of coping behavior; these withdrawal responses may take the form of depression, fatalism, resignation, or alcohol or drug abuse (M. Inglehart, 1991).
Virtually all of the world's major cultures have belief systems which provide reassurance that, even though the individual alone cannot understand or
38
CHAPTER    1
predict what lies ahead, it is in the hands of a benevolent higher power. One's future may be unpredictable, but this higher power will ensure that things work out. Both religion and secular ideologies provide assurance that the universe is not random, but follows a plan which guarantees that (in this world or the next) everything will turn out well. This belief reduces stress, enabling one to shut out anxiety and focus on some immediate coping strategy. Without such a belief system, extreme stress is likely to produce withdrawal reactions.
Religion is the dominant influence on the belief systems of most preindus-trial societies. In religious world views, the higher power is an omniscient and benevolent God. Stress is reduced by a system of absolute rules that govern many aspects of life and maximize predictability. In secular societies, the state or a strong political leader fills the role of the higher power. Under conditions of great unpredictability, people have a powerful need to see authority as not only strong, but also benevolent—even in the face of evidence to the contrary.
Communist ideology provided a functional equivalent to religion, furnishing an explanation of how the universe functioned and where history was going. Although many of Marx's predictions eventually turned out to be wrong, the ideology provided a sense of predictability and reassured people that infallible leaders were in charge.
The Authoritarian Reflex
In societies undergoing an historical crisis, a phenomenon has been observed that might be called the Authoritarian Reflex. Rapid change leads to severe insecurity, giving rise to a powerful need for predictability. Under these circumstances, the Authoritarian Reflex takes two forms:
1.  Fundamentalist or nativist reactions. This phenomenon frequently occurs in preindustrial societies when they are confronted with rapid political and economic change through contact with industrialized societies; and it is often found among the more traditional and less secure strata in industrial societies, especially during times of stress. In both cases, the reaction to change takes the form of a rejection of the new, and a compulsive insistence on the infallibility of old, familiar cultural patterns.
2.  Adulation of strong secular leaders. In secularized societies, severe insecurity brings a readiness to defer to strong secular leaders, in hopes that superior men of iron will can lead their people to safety. This phenomenon frequently occurs in response to military defeat or economic or political collapse.
Thus, disintegrating societies often give rise to authoritarian and xenophobic reactions. Pogroms broke out in the declining years of Czarist Russia, and after its collapse power was seized by rulers who were even more ruthlessly authoritarian than the czars. Similarly, the Great Depression of the 1930s helped bring Hitler to power in Germany and contributed to the rise of fascis-tic dictators in a number of other countries, from Spain to Hungary to Japan.
Massive insecurity is conducive not only to a need for strong authority figures to protect one from threatening forces, but also to xenophobia (Tajfel,
CULTURE,    POLITICS,    AND    ECONOMICS
39
1978; Tajfel and Turner, 1979; Hamilton, 1981; Jackson and Inglehart, 1996). Frighteningly rapid change breeds an intolerance of cultural change, and of different ethnic groups. Thus, in the United States during the late nineteenth and early twentieth centuries, when the price of cotton went down, lynchings of Blacks went up in .the South. This was a reaction to insecurity, not a cognitive response to the belief that Blacks were manipulating the price of cotton: the lynchers were aware that Blacks had little influence on the cotton market (Beck, Massey, and Tolnay, 1989). Similarly, the Great Depression of the 1930s gave rise to the twin phenomena of Hitler and anti-Semitism—and ultimately, to the Holocaust. There was nothing inevitable in this horror story. It occurred in a society that previously had been more tolerant toward Jews than had Russia or France and had one of the most socially integrated Jewish communities in Europe. It reflected traumatic insecurity caused by military defeat and political and economic collapse, rather than anything uniquely German. In a hauntingly parallel phenomenon, the collapse of the economic and political systems of what used to be the Soviet Union and Yugoslavia has given rise to ultranationalism and "ethnic cleansing."
Postmodernism: Declining Emphasis on Political, Economic, and Scientific Authority
All societies depend on some legitimating formula for authority: unless their leaders' decisions are seen as legitimate, they rest solely on coercion. A central component of Modernization was the shift from religious authority to rational-bureaucratic authority, justified by claims that the governing institutions were conducive to the general good.
A major component of the Postmodern shift is a shift away from both religious and bureaucratic authority, bringing declining emphasis on all kinds of authority. For deference to authority has high costs: the individual's personal goals must be subordinated to those of a broader entity. But under conditions of insecurity, people are more than willing to do so. Under threat of invasion, internal disorder, or economic collapse, people eagerly seek strong authority figures who can protect them.
Conversely, conditions of prosperity and security are conducive to pluralism in general and democracy in particular. This helps explain a long-established finding: rich societies are much likelier to be democratic than poor ones. This finding was pointed out by Lipset (1960) and has been confirmed most recently by Burkhart and Lewis-Beck (1994). The reasons why this is true are complex (we will examine them in chapter 5); but one factor is that the authoritarian reflex is strongest under conditions of insecurity.
Until recently, insecurity was a central part of the human condition. Only recently have societies emerged in which most of the population did not feel insecure concerning survival. Thus, both premodern agrarian society and modern industrial society were shaped by survival values. But the Postmodern shift has brought a broad de-emphasis on all forms of authority.
40
CHAPTER    I
Changing Religious Orientations, Gender Roles, and Sexual Norms
The rise of Postmodernism is the reverse of the Authoritarian Reflex: Postmaterialist values characterize the most secure segment of advanced industrial society. Postmaterialist values developed in the environment of the historically unprecedented economic growth and the welfare states that emerged after World War II. And they are a core element of a Postmodern shift that is reshaping the political outlook, religious orientations, gender roles, and sexual norms of advanced industrial society. Two factors contribute to the decline of traditional political, religious, social, and sexual norms in advanced industrial societies.
The first is that an increasing sense of security brings a diminishing need for absolute rules. Individuals under high stress have a need for rigid, predictable rules. They need to be sure of what is going to happen because they are in danger—their margin for error is slender and they need maximum predictability. Postmaterialists embody the opposite outlook: raised under conditions of relative security, they can tolerate more ambiguity; they are less likely to need the security of absolute rigid rules that religious sanctions provide. The psychological costs of deviating from whatever norms one grew up with are harder to bear if a person is under stress than if a person feels secure. Taking one's world apart and putting it together again is extremely stressful. But Postmaterialists—people with relatively high levels of security—can more readily accept deviation from familiar patterns than can people who feel anxiety concerning their basic existential needs. Consequently, Postmaterialists accept cultural change more readily than others.
The second reason is that societal and religious norms usually have a function. Such basic norms as "Thou shalt not kill" (the Judeo-Christian version of a virtually universal social norm) serve an important societal function. Restricting violence to narrow, predictable channels is crucial to a society's viability. Without such norms, a society would tear itself apart.
Many religious norms such as "Thou shalt not commit adultery" or "Honor thy father and mother" are linked with maintaining the family unit. Various versions of these norms are also found in virtually every society on earth because they serve crucial functions. But in advanced industrial society, some of these functions have dwindled.
The role of the family has become less crucial than it once was. Although the family was once the key economic unit, in advanced industrial society one's working life overwhelmingly takes place outside the home. Similarly, education now takes place mainly outside the family. Furthermore, the welfare state has taken over responsibility for survival. Formerly, whether children lived or died depended on whether their parents provided for them, and the parents' survival depended on their children when they reached old age. Today, though the family is still important, it is no longer a life or death relationship; its role has largely been taken over by the welfare state. The new generation can survive if the family breaks up—or even if neither parent is around. One-
CULTURE,    POLITICS,    AND    ECONOMICS
41
parent families and childless old people have vastly better chances for survival under contemporary conditions than ever before. As long as it threatens the survival of children, society is apt to view divorce as absolutely wrong: it undermines the long-term viability of society itself. Today, the functional basis of this norm and other norms reinforcing the two-parent family has eroded: does that mean that society changes its values? No—at least, not immediately.
Cultural norms are usually internalized very firmly at an early age, and backed up by prerational sanctions. People's opposition to divorce does not simply reflect an individual's rational calculation that "the family is an important social unit, so I should stay married." Instead, divorce tends to be made a question of good and evil, through absolute norms. Norms that constrain people's behavior even when they strongly want to do something else are norms that have been taught as absolute rules, and inculcated so that their consciences torture them if these norms are violated. Such societal norms have a great deal of momentum. The mere fact that the function of a given cultural pattern has weakened or disappeared does not mean that the norm immediately disappears. But it opens the way for that norm to weaken gradually, especially if those norms conflict with strong impulses to the contrary.
Norms supporting the two-parent heterosexual family are weakening for a variety of reasons, ranging from the rise of the welfare state to the drastic decline of infant mortality rates, which means that a couple no longer needs to produce four or five children in order for the population to reproduce itself. Experimentation and testing of the old rules takes place; gradually, new forms of behavior emerge that deviate from traditional norms, and the groups most likely to accept these new forms of behavior are the young more than the old, and the relatively secure, more than the insecure.
The Postmodern shift involves an intergenerational change in a wide variety of basic social norms, from cultural norms linked with ensuring survival of the species, to norms linked with the pursuit of individual well-being. For example, Postmaterialists and the young are markedly more tolerant of homosexuality than are Materialists and the old. This is part of a pervasive pattern. Postmaterialists have been shaped by security during their formative years and are far more permissive than Materialists in their attitudes toward abortion, divorce, extramarital affairs, prostitution, and euthanasia. Materialists, conversely, are likely to adhere to the traditional societal norms that favored child-bearing, but only within the traditional two-parent family—and that heavily stigmatized any sexual activity outside that setting.
Traditional gender role norms from East Asia to the Islamic world to Western society discouraged women from taking jobs outside the home. Virtually all preindustrial societies emphasized childbearing and childrearing as the central goal of any woman, her most important function in life, and her greatest source of satisfaction. In recent years, this perspective has been increasingly called into question, as growing numbers of women postpone having children or forego them completely in order to devote themselves to careers outside the home.
42
CHAPTER    I
Existential Security and the Rise of Postmodern Values
Throughout advanced industrial society, there is evidence of a long-term shift away from traditional religious and cultural norms. This decline of traditional norms is closely linked with the shift from Materialist toward Postmaterialist values. In terms of face content, this is not obvious: none of the survey items used to measure Materialist/Postmaterialist values makes any reference whatever to religion or to sexual or gender norms. Nevertheless, all of these values are components of a broad cultural change linked with the transition from industrial to postindustrial society. The shift to Postmaterialism and the decline of traditional religious and sexual norms go together because they share a common cause: the unprecedented levels of existential security attained in contemporary advanced industrial society that grows out of the economic miracles (both Western and Asian) of the past several decades, and the rise of the welfare state.
In the highly uncertain world of subsistence societies, the need for absolute standards and a sense that an infallible higher power will ensure that things ultimately turn out well filled a major psychological need. One of the key functions of religion has been to provide a sense of certainty in an insecure environment. Not only economic insecurity gives rise to this need: the old saying that "there are no atheists in foxholes" reflects the fact that physical danger also leads to a need for belief in a higher power. But in the absence of war, prosperity and the welfare state have produced an unprecedented sense of security concerning one's survival. This has diminished the need for the reassurance that religion traditionally provided.
These same factors have weakened the functional basis of a pervasive set of norms linked with the fact that, throughout most of history, the traditional two-parent family was crucial to the survival of children, and thus, of society. These norms ranged from disapproval of divorce, abortion, and homosexuality, to negative attitudes toward careers outside the home for married women. As we will see, it is precisely in the most advanced welfare states that mass adherence to traditional religious and family norms has declined most rapidly. This is no coincidence. These factors are also changing another major aspect of people's worldviews: respect for authority is declining throughout advanced industrial society.
The difference between feeling secure or insecure about survival is so basic that it has led to a wide-ranging but coherent syndrome of changes, from the "survival" values that characterized agrarian and early industrial society, to the "well-being" values that characterize advanced industrial society.
The difference between whether one views survival as uncertain, or assumes that it can be taken for granted, is central in shaping people's life strategies, giving rise to very distinct worldviews. Throughout most of history, in both agrarian and early industrial society, survival has been uncertain for the great majority of the population; consequently, they have emphasized survival values. Postmodern values grow out of the unprecedented mass prosperity of ad-
- U L T V R E ,    POLITICS,    AND    ECONOMICS
43
vanced industrial societies in which, for the first time in history, large segments of the public take survival for granted. These contrasting value systems have ramifications that extend across politics, economics, sexual and family norms, and religion, as table 1.1 illustrates.
The shift from modern to Postmodern values is eroding many of the key institutions of industrial society, through the following changes:
1. In the political realm, the rise of Postmodern values brings declining respect for authority, and growing emphasis on participation and self-expression. These two trends are conducive to democratization (in authoritarian societies) and to more participatory, issue-oriented democracy (in already democratic societies). But they are making the position of governing elites more difficult.
Respect for authority is eroding. And the long-term trend toward increased mass participation is not only continuing, but has taken on a new character. In large-scale agrarian societies, political participation was limited to a narrow minority. In industrial society, the masses were mobilized by disciplined elite-led political parties. This was a major advance for democratization, and it resulted in unprecedented numbers of people taking part in politics by voting— but mass participation rarely went much beyond this level. In Postmodern society the emphasis is shifting from voting, to more active and issue-specific forms of mass participation. Mass loyalties to long-established hierarchical political parties are eroding; no longer content to be disciplined troops, the pub-
Table 1.1
Security and Insecurity: Two Contrasting Value Systems
Survival Is Seen as
Insecure
Secure
1. Politics
Need for strong leaders
Order
Xenophobia/fundamentalism
2. Economics
Priority to economic growth Achievement motivation Individual vs. state ownership
3. Sexual/Family Norms Maximize reproduction—but only in
two-parent heterosexual family
4. Religion
Emphasis on higher power
Absolute rules
Emphasis on predictability
De-emphasis on political authority Self-expression, participation Exotic/new are stimulating
Quality of life = top priority Subjective well-being Diminishing authority of both private and state ownership
Individual sexual gratification Individual self-expression
Diminishing religious authority Flexible rules, situational ethics Emphasis on meaning and purpose of life
44
CHAPTER    I
lie has become increasingly autonomous and elite-challenging. Consequently, though voter turnout is stagnant or declining, people are participating in politics in increasingly active and more issue-specific ways. Moreover, a growing segment of the population is coming to value freedom of expression and political participation as things that are good in themselves, rather than simply as a possible means to attain economic security.
But these changes have had a traumatic impact on the old-line political machines of industrial society, which are in disarray almost everywhere. Throughout the history of industrial society, the scope of state activities had been growing rapidly; it seemed to be a law of nature that government control of economy and society would continue to expand. That trend has now reached a set of natural limits—both for functional reasons and because of eroding public trust in government and a growing resistance to government intrusion. The people of each society tend to assume that this erosion of confidence is due to factors unique to their own country; in reality, it is taking place throughout advanced industrial society.
Xenophobia thrives under conditions of rapid change and insecurity. Today, this is especially evident in what used to be Yugoslavia and the Soviet Union, and ethnic hatred has not disappeared even in more secure industrial societies. But xenophobia is less widespread in secure societies than in insecure ones; and in long-term perspective, the more secure societies seem to be moving toward increasing acceptance of diversity. Finally, Postmodern politics are distinguished by a shift from the class-based political conflict that characterized industrial society, to increasing emphasis on cultural and quality of life issues.
2.  In the economic realm, existential security leads to increasing emphasis on subjective well-being and quality of life concerns; for many people, these become higher priorities than economic growth. The core goals of Modernization, economic growth, and economic achievement are still positively valued, but their relative importance is declining.
There is also a gradual shift in what motivates people to work: emphasis shifts from maximizing one's income and job security toward a growing insistence on interesting and meaningful work. Along with this comes a twofold shift in the relationship between owners and managers. On one hand, we find a growing emphasis on more collegial and participatory styles of management. But at the same time, there is a reversal of the tendency to look to government for solutions to such problems and a growing acceptance of capitalism and market principles. Both trends are linked with a growing rejection of hierarchical authority patterns and rising emphasis on individual autonomy. Ever since the era of laissez-faire capitalism, people have almost automatically turned to government to offset the power of private business. Today, there is a widespread feeling that the growth of government is becoming functionally ineffective and a threat to individual autonomy.
3.  In the realm of sexual behavior, reproduction, and the family, there is a continued trend away from the rigid norms that were a functional necessity in
CULTURE,    POLITICS,    AND    ECONOMICS
45
agrarian society. In these societies, traditional methods of contraception were unreliable, and children born outside a family with a male breadwinner were likely to starve; sexual abstinence except in marriage was a key means of population control. The development of effective birth control technology, together with prosperity and the welfare state, have eroded the functional basis of traditional norms in this area; there is a general shift toward greater flexibility for individual choice in sexual behavior, and a dramatic increase in the acceptance of homosexuality. This not only continues some of the trends associated with modernity, but breaks through to new levels. Gays and lesbians have come out of the closet, and unmarried parenthood is a normal part of prime time television.
4. In the realm of ultimate values, we also find both continuity and striking change. One of the key trends associated with Modernization was secularization. This trend has continued, where established religious institutions are concerned: the publics of most advanced industrial societies show both declining confidence in churches and falling rates of church attendance and are placing less emphasis on organized religion. This does not mean that spiritual concerns are vanishing, however: for we also find a consistent cross-national tendency for people to spend more time thinking about the meaning and purpose of life. The dominance of instrumental rationality is giving way to growing concern for ultimate ends.
These trends reflect the unprecedented security that has developed in Postmodern society. Economic accumulation for the sake of economic security was the central goal of industrial society. Ironically, their attainment set in motion a process of gradual cultural change that has made these goals less central— and is now bringing a rejection of the hierarchical institutions that helped attain them.
Predicting Cultural Change
The theory of value change generates a number of clear predictions. Table 1.1 outlines a set of qualitative shifts linked with growing existential security. This table shows what kinds of values we would expect to become more widespread as Postmodernization takes place. But the theory is not limited to qualitative predictions concerning the general direction of cultural change. It also generates a set of quantitative predictions concerning where and how fast these changes should occur. The scarcity hypothesis postulates that a sense of existential security is conducive to Postmodern values. This gives rise to the following predictions:
1.  In cross-national perspective, Postmodern values will be most widespread in the richest and most secure societies; the publics of impoverished societies will place more emphasis on survival values.
2.  Within any given society, Postmodern values will be most widespread
46
CHAPTER    1
among the more secure strata: the wealthier and better educated will be most likely to hold a whole range of security values, including Postmaterialism; the less secure strata will emphasize survival priorities.
3.   Short-term fluctuations will follow the implications of the scarcity hypothesis: prosperity will enhance the tendency to emphasize well-being values; economic downturn, civil disorder, or war will lead people to emphasize survival values.
4.   Long-term changes will also reflect the scarcity hypothesis. In societies that have experienced high levels of security for several decades, we should find a long-term shift from survival values toward well-being values. This is not a universal trend that sweeps the entire world, like the popularization of pop culture fostered by the global mass media. Instead, the shift toward well-being values is occurring mainly in those societies that have attained such a high level of prosperity and safety that a substantial share of the population takes survival for granted; it is not found in societies that have not experienced rising prosperity. On the other hand, it is not a uniquely Western phenomenon: it should appear in any society that has experienced the transition to high mass security.
The socialization hypothesis postulates that neither an individual's values nor those of a society as a whole will change overnight. In connection with the scarcity hypothesis, this generates three additional predictions:
5.  In societies that have experienced a long period of rising economic and physical security, we will find substantial differences between the value priorities of older and younger groups: the young will be much likelier to emphasize well-being values than the old. This reflects the fact that the young experienced greater security during their formative years than did the old. Fundamental value change takes place mainly as younger birth cohorts replace older ones in a given society.
6.  These intergenerational value differences should be reasonably stable over time: though immediate conditions of security or insecurity will produce short-term fluctuations, the underlying differences between younger and older birth cohorts should persist over long periods of time. The young will not take on the values of the old as they age, as would happen if the intergenerational differences reflected life-cycle effects; instead, after two or three decades have passed, the younger cohorts should still show the distinctive values that characterized them at the start of the period.
7.  In cross-national perspective, large amounts of intergenerational change will be found in those countries that have experienced relatively high rates of economic growth: if differences between the values of young and old were a normal feature of the human life cycle, they would be found everywhere. But if, as our theory implies, this process of value change is driven by historical changes in the degree of security experienced during one's preadult years, then the age differences found in a given society will reflect that society's economic history: the difference between the values of young and old will be largest in countries like Western Germany or South Korea that experienced the greatest
CULTURE,    POLITICS,    AND    ECONOMICS
47
increases in prosperity during the past 40 years; and conversely, the difference between the values of young and old will be small or nonexistent in such countries as Nigeria and India, which experienced relatively little increase in per capita income from 1950 to 1990.
Thus, high levels of prosperity should be conducive to high levels of Post-materialism and other Postmodern values; high rates of economic growth should produce relatively rapid rates of value change and relatively large intergenerational differences.
8. Finally, the theory of intergenerational value change not only yields predictions about what kinds of values should be emerging and where, but even predicts how much value change should be observed in a given period of time. Since the change is based on intergenerational population replacement, if one knows the distribution of values across birth cohorts in a given nation and the sizes of the cohorts, one can estimate how much change will be produced in a given time span, as a result of intergenerational population replacement. With the four-item Materialist/Postmaterialist values battery, for example, population replacement should produce a shift toward Postmaterialism of approximately one point per year on the Materialist-Postmaterialist percentage difference index (Abramson and Inglehart, 1992).
Authoritarianism and the Postmodern Shift
We have just described a syndrome of cultural changes through which people are shifting from one belief system to another. Under conditions of insecurity people seek strong authority; this is part of a worldview that also embraces eth-nocentrism, traditional gender roles, and traditional religious norms.
This is not the first time that such a configuration of orientations has been observed. Several decades ago, Adorno et al. (1950) demonstrated that orientations toward authority, aggression toward outgroups, and a high degree of adherence to social conventions go together in a syndrome that they called The Authoritarian Personality. This work was controversial, evoking numerous critiques on both theoretical and methodological grounds. Despite massive criticism, this thesis generated an immense body of research that has survived and evolved over the years, with particularly significant recent contributions being made by Altemeyer (1981, 1988).
From the outset of our research, the Authoritarian Personality thesis seemed potentially relevant to the rise of Materialist/Postmaterialist values that are at the core of Postmodern values. A standardized set of authoritarianism items was used in a cross-national exploration of nationalism and internationalism. The results were disappointing: dimensional analysis showed that the authoritarianism items did not cluster together as they theoretically should (Inglehart, 1970).
Subsequent pilot tests gave similar results. Authoritarianism items showed relatively weak relationships with each other; some were closely related to the Materialist/Postmaterialist dimension, but others tapped quite different di-
48
CHAPTER    I
mensions. Authoritarianism, as originally operationalized, has a poor empirical fit with Materialism/Postmaterialism.
The theoretical basis of authoritarianism is not necessarily incompatible with that of Materialism/Postmaterialism, but there are important differences in focus. The initial concept of authoritarianism emphasizes the psychody-namics of harsh discipline in early childrearing, rather than influences from the broader economic and political environment. On the other hand, Hyman and Sheatsley (1954), in their critique of the original study, argue a cognitive explanation: certain respondents, especially those from a lower socioeconomic level, may show an authoritarian-type response because this is a more or less accurate reflection of the conditions governing their adult lives; Alte mey er also endorses this interpretation. Our own interpretation of the genesis of Materi-alist/Postmaterialist values contains elements of both positions. It emphasizes the importance of early experiences, but links them with one's formative experiences as a whole, and not just parental discipline.
The original authoritarianism hypothesis does not predict either the age-group differences or the social class differences that are strikingly evident in our data. Quite the contrary, studies of authoritarianism have found that children tend to be more authoritarian than adults. It would not be impossible to reinterpret the Authoritarian Personality hypothesis in such a way as to explain the age and class differences. One might argue that childrearing practices vary according to social class and have changed over time. But if one did so, one would then need to seek an explanation of why they vary and why they have changed. Quite probably, one would eventually trace this explanation to the economic and political changes on which we rest our own interpretation.
Another important distinction between authoritarianism and Materialist/ Postmaterialist values lies in the way they are measured: authoritarianism reflects levels of support for given positions; Materialist/Postmaterialist values deal with priorities—that is, the relative rank of various goals. This distinction is crucial, and will be discussed at some length in chapter 3. Our theory implies that an intergenerational change in priorities is taking place—and notthat people no longer value economic security. Nevertheless, the two streams of research agree on one major point: orientations toward authority are related to a broad range of other orientations, forming the core of a coherent worldview.
Changing Mass Values: Testing Our Predictions
We now have a large body of empirical evidence on cultural change, from surveys carried out in more than 40 societies over the past 25 years. Using these data, this book will test these predictions. Chapter 4 focuses on the relatively detailed and abundant body of data concerning the Materialist/Postmaterialist value shift; chapters 8 and 9 examine the evidence of a much broader process of cultural change involving religious, civic, sexual, and economic norms as well as Materialist/Postmaterialist values. The following chapters examine survey data from societies containing 70
C V L T U R E ,    POLITICS,    AND    ECONOMICS
49
percent of the world's population. For 21 of these societies, we have time series data from the World Values surveys carried out in 1981 and 1990. For several societies, we also have detailed time series data on value changes from 1970 to 1994. The evidence from these surveys indicates that advanced industrial societies are moving on a common trajectory. To a striking degree, societies in Western Europe, North America, Latin America, Eastern Europe, and East Asia are undergoing similar cultural changes in politics, economics, sex and gender norms, and religion. Although they have widely varying cultural traditions and start from very different levels, they are generally moving in the same direction.
Do the values linked with secure survival actually move in the predicted direction from 1981 to 1990? As we will see below, on the whole our predictions hold up very well when tested against data from the 21 nations surveyed in both 1981 and 1990. About 40 variables were strongly correlated with existential security. These variables move in the predicted direction in most countries for which data are available. Moreover our predictions hold up best in those countries that experienced relatively prosperous circumstances; they fail to apply in those countries that experienced economic decline and political upheaval—precisely as the theory implies.
These findings suggest that social science can sometimes have predictive power: when we are dealing with relatively enduring aspects of the outlook of given birth cohorts, we can anticipate that change will tend to move in a specific direction, as intergenerational population replacement occurs. Other factors such as the rise and fall of the economic cycle or war and peace will also shape the outlook of a given society at a given time. But in the long run, across many societies, such situational factors tend to cancel each other out: the influence of intergenerational population replacement, on the other hand, tends to work in a specific direction for many decades, and its cumulative impact can be great.
This study was motivated by the belief that mass belief systems have important economic, political, and social consequences. Although it has long been believed that given cultural patterns tend to go with given economic and political systems, this belief has rested mainly on impressionistic evidence: it has been difficult to demonstrate empirically because, until recently, cross-culturally comparable measures of beliefs and values have not been available on a global scale. Empirical evidence from 43 societies demonstrates that cultural patterns are, indeed, linked with important economic and political variables— and that the cross-level linkages are astonishingly strong.
Chapter 5 examines the causal linkages between culture and democracy in greater detail; chapter 6 focuses on the linkages between culture and economic growth. In both cases, the evidence suggests that culture is not just a dependent variable, but has an important impact on both democracy and economic growth.
The evidence we will examine makes it clear that—as both Marx and Weber argued—belief systems, economics, and politics are intimately related. Their
50
CHAPTER    1
linkages seem to reflect neither a simple Marxian causality (with economics driving culture and politics) nor a simple Weberian causality (with culture driving economics and politics), but reciprocal causal relationships. Cultural, economic, and political systems tend to be mutually supportive in any society that survives for long. They help shape each other, and they are changing the world in ways that are to some extent predictable.
CHAPTER     2
Individual-Level Change and Societal-Level Change
The next several chapters examine the linkages between individual-level value change and changes at the societal level. This chapter investigates how economic development brings changes in human life strategies—and then examines the ways in which cultural changes can give rise to legal and institutional changes. Chapter 5 will analyze how belief systems influence the emergence of democratic institutions, chapter 6 examines the impact of values on economic growth, and chapter 7 examines their impact on political cleavages. In analyzing the linkages between belief systems and societal variables, the first question one is likely to ask is, Do the values and attitudes of individuals affect their behavior? If they do not, then changes in these values and attitudes could scarcely have any impact on the society as a whole. And it has often been claimed that people's attitudes have no impact on their behavior.
Do Attitudes Shape Behavior?
In the 1930s, an American social scientist reported that, in response to a written inquiry, most of the restaurant owners whom he contacted said they would not serve Chinese customers; but when he appeared at these same restaurants with a young Chinese couple, almost all of them actually did so (LaPiere, 1934). He concluded that attitudes were irrelevant to actual behavior. This finding was so counterintuitive and so interesting that it was widely cited for several decades. And as recently as the 1960s, a review of empirical studies concluded that attitudes were generally "unrelated or only slightly related to overt behaviors" (Wicker, 1969: 65).
A more recent review of 88 attitude-behavior studies comes to a very different conclusion: Kraus (1995) finds that attitudes significantly and substantially predict future behavior. Furthermore, the most important factor associated with high attitude-behavior correlations was whether the research design used the same level of specificity in the attitudinal and behavioral measures— as Fishbein and Ajzen (1975) had suggested 20 years earlier. Not surprisingly, broad global attitudes do not necessarily predict specific behaviors. For example, one's answer to the question "Are you a liberal or a conservative?" is not nearly as good a predictor of voting behavior, as is one's voting intention. And the question "Do you believe in God?" does not predict church attendance as well as the question "Do you think it's important to go to church?" Belief in God is a more global attitude than is emphasis on church attendance. On the other hand, global attitudes are relatively good at predicting global patterns of
66
CHAPTER    2
Culture is resistant to change, partly because people tend to believe whatever their society's institutions teach them. But one's worldview is also influenced by one's firsthand experience—and if the two are in conflict, one's firsthand experience may have even greater credibility than what one is told. This is one reason why political systems, even totalitarian ones, have limited ability to reshape their culture. People are sensitive to those aspects of reality that directly affect them. This was crucial in the shift toward Postmodern values. The younger birth cohorts in advanced industrial societies perceived, during their formative years, that survival was not precarious and that they could take it more or less for granted. This experience was profoundly different from the conditions that have shaped most people's lives throughout most of history. It led to pervasive changes in worldviews. For these birth cohorts, maximizing their economic gains no longer maximized their subjective well-being as it had for earlier generations.
People did not consciously set out to change their worldviews. New outlooks and new modes of behavior, like random mutations, arose for a variety of reasons—and some of them spread. Even within a given birth cohort, many continued to accept the established norms of industrial society; but others took on new orientations and transmitted them to some of their peers through social learning processes. Change has been uneven. But the new lifestyles have spread gradually—and in the last analysis, they have done so because they represent more effective ways to maximize survival and subjective well-being under new conditions. At a much earlier stage of history, new norms linked with the rise of modern society (such as the Protestant Ethic) gradually spread in somewhat similar fashion. We lack detailed information on how this took place, but it seems to have occurred more slowly than the rise of Postmodern values, which in some ways represents its reversal. In both cases, culture was gradually reshaped by changes in the socioeconomic environment; and these cultural changes eventually produced feedback that helped reshape political and economic life.
Postmodernization is a shift in survival strategies. It moves from maximizing economic growth to maximizing survival and well-being through lifestyle changes. Once industrialization had become possible, Modernization focused on rapid economic growth as the best way of maximizing survival and well-being. But no strategy is optimal for all times. Modernization was dramatically successful in raising life expectancies, but it has begun to produce diminishing returns in advanced industrial societies. Emphasizing competition, it reduces the risk of starvation, but increases psychological stress. With the transition from Modernization to Postmodernization, the trajectory of change has shifted from maximizing economic growth to maximizing the quality of life.
CHAPTER     3
Modernization and Postmodernization in 43 Societies
Introduction
As we have seen, Modernization theory falls into two main schools: (1) a Marxist version, which claims that economics, politics, and culture are closely linked because economic development determines the political and cultural characteristics of a society, and (2) a Weberian version, which claims that culture shapes economic and political life. Despite an enduring debate between the two schools, they agree on one crucial point: that socioeconomic change follows coherent and relatively predictable patterns. This means that key social, political, and economic characteristics are not randomly related; they tend to be closely linked so that by knowing one such trait, one can predict the presence of other key traits with much better than random success.
Cultural relativists, on the other hand, claimed that it would be ethnocentric not to believe that all cultures are equally conducive to economic development and democracy. And dependency theorists viewed a given society's culture as irrelevant to economic development and democracy, which were determined by the forces of global capitalism. Both of these views imply that the relationships between culture and economics and politics are more or less random.
This chapter presents a broad overview of a huge body of data from more than 40 societies. It demonstrates that, far from being randomly related, specific cultural, economic, and political variables are closely correlated. Although this chapter does not attempt to demonstrate whether causality flows in the Marxist or the Weberian direction, the linkages we find indicate that at least one version of Modernization theory was right. Subsequent chapters will probe more deeply into the causal linkages.
Although we find strong support for the central claim of Modernization theory, we disagree with it on several narrower points—above all, the notion that socioeconomic change is linear. Instead, we find evidence that a major change of direction occurs when societies reach an advanced level of industrial development. The Modernization phase involves the familiar syndrome of industrialization, occupational specialization, bureaucratization, centralization, rising educational levels, and beliefs and values that support high rates of economic growth; but among advanced industrial societies, a second syndrome of cultural and institutional changes emerges in which economic growth becomes less central, and there is rising emphasis on the quality of life and democratic political institutions.
68
CHAPTER    3
Cross-sectional Evidence of Change over Time
This chapter will undertake something that verges on heresy: we will examine hypotheses about changes over time in the light of cross-sectional evidence. This procedure has been criticized (quite appropriately) in the past: taken by itself, cross-sectional evidence is an uncertain indicator of change. There is no substitute for time series data if one hopes to draw firm conclusions about social change. In keeping with this idea, much of this book is devoted to analyzing time series data concerning sociocultural change.
Nevertheless, we are convinced that the World Values surveys can usefully supplement the available time series evidence, providing additional insight on patterns of cultural change. Global cross-national data are needed because the available survey data on this topic are largely drawn from advanced industrial societies and limited to the past few decades. The World Values survey provides a much broader range of variation than has ever before been available, bringing together data from 43 nations throughout the world, covering the full range of economic and political variation.
If we had survey data covering the entire period from the early nineteenth century to the present we could analyze the interplay between changing cultural values and economic and political Modernization over many decades. We could then determine which came first, cultural change or economic or political change. But such data are not available. The analysis of cross-sectional data offers the nearest substitute. Examining the orientations of people in poor societies gives some sense of what prevailing mass orientations in today's rich democracies may have been like when these countries were poor and predemocratic.
Conversely, comparing the worldviews of rich and poor countries provides some idea of how the outlook of the publics of poorer countries may change if their societies become industrialized and economically secure. We do not view these changes as deterministic: economic and technological changes interact with political, cultural, and other variables. The cultural heritage of a given society may facilitate or retard Modernization; and determined leaders can repress or accelerate social change. Nevertheless, as we will demonstrate, it is possible to identify a specific syndrome of cultural values and beliefs that is likely to be present, if urbanization, industrialization, higher education, and other components of Modernization become widespread.
Inkeles and Smith (1974) suggested that this should be true, but their conclusion was based on a comparison of the belief systems of those working within the "modern" and premodern sectors of six developing societies, and it did not compare societies at various levels of development. This analysis, for the first time, demonstrates the existence of fundamentally different world-views between the publics of preindustrial and industrial societies, confirming Inkeles and Smith's insight. But our analysis goes a step farther: it also analyzes the cultural differences between societies of scarcity (both preindustrial and early industrial) and economically affluent "Postmodern" societies.
MODERNIZATION   AND    POSTMODERNIZATION
69
Modernization and Postmodernization in Cross-sectional Perspective
The concepts of both Modernization and Postmodernization are based on two key assumptions:
1.  Various cultural elements tend to go together in coherent patterns. For example, do societies that place relatively strong emphasis on religion also tend to favor large families (or respect for authority or other distinctive attitudes)? If each culture goes its own way, elements such as these would be uncorrected, and one would find no consistent patterns of constraint.
2.  Coherent cultural patterns exist, and they are linked with economic and technological development. For example, industrialization was accompanied by secularization in Western history. But some observers argue that, since some Islamic countries such as Iran and Libya have grown rich without secularization, there is no linkage between economic development and secularization. This argument ignores the fact that Modernization is not just the possession of large oil deposits: it is a syndrome of cultural, economic, and technological changes closely linked with industrialization—a syndrome that Iran and Libya have not experienced, and which does tend to be linked with secularization.
Together, these two postulates imply that some patterns are more probable than others—and hence, that development is to some extent predictable. Is economic development linked with coherent cultural patterns, distinct from those found in less developed societies? If so, then cross-national surveys should reveal clear patterns, with one syndrome of orientations being found in economically developed societies, and another syndrome being found in less developed societies. If such patterns are found, the evidence would support Modernization theory. Furthermore, it would imply that sociopolitical change has an element of predictability.
Do coherent cultural patterns exist, and are they linked with levels of economic development? To answer this question, we will analyze the World Values survey data on key values and beliefs among representative national samples of publics around the world. This survey was designed to test the hypothesis that economic development leads to specific changes in mass values and belief systems—which in turn produce feedback, leading to changes in the economic and political systems of these societies. We do not assume that all elements of culture will change, leading to a uniform global culture: we see no reason to expect that the Chinese will stop using chopsticks in the foreseeable future, or that Brazilians will learn to polka. But certain cultural and political changes do seem to be logically linked with the dynamics of a core syndrome of Modernization, involving urbanization, industrialization, economic development, occupational specialization, and the spread of mass literacy.
Change is not linear in any system subject to feedback. This complicates our analysis. If the process of economic-cultural-political change moved smoothly in one continuous direction, a cross section of the world's societies would
70
C H A P T E S    3
show a simple developmental progression of cultural changes as one moved from the least developed to the economically most developed societies. Analogously, a cross section of the earth's surface sometimes reveals neatly ordered geological layers, with the oldest stratum of rock lowest and the newer strata located above the older ones. But reality is not this simple: social change produces feedback, which eventually changes the direction of change. Thus, we are likely to find patterns similar to those produced by tectonic upheavals, in which identifiable geological layers are shifted and juxtaposed with other strata. The result is not chaos, but neither is it a simple layering from oldest to newest strata.
We suggest that we will find the residue of two major waves of change (along with many lesser ones) mirrored in the World Values survey's cross section of the world's cultures: the distribution of these cultural traits reflects the processes of Modernization and Postmodernization, respectively.
The literature on Modernization focuses on the first of these two processes. It argues (correctly, we believe) that a broad syndrome of changes has been linked with modern economic development. These changes include urbanization, industrialization, occupational specialization, mass formal education, development of mass media, secularization, individuation, the rise of entrepreneurs and entrepreneurial motivations, bureaucratization, the mass production assembly line, and the emergence of the modern state. The material core of this process is industrialization; and though the industrial revolution originated in the West, this process is not inherently Western and should not be confused with Westernization. Although there are arguments about what the "real" driving force is behind this syndrome, there is widespread agreement that these changes include technological, economic, cultural, and political components.
Religion and Economic Growth
We propose a modified interpretation of Weber's thesis concerning the role of the Protestant Ethic in economic development. Weber was correct in arguing that the rise of Protestantism was a crucial event in modernizing Europe. But this was not due to factors unique to Protestantism—it has been argued that everything Weber ascribed to Puritanism might equally well be ascribed to Judaism (Sombart, 1913). European Judaism had an outlook that was in some ways modern, but it could not transform Europe because it held a marginal position there. The crucial impact of Protestantism was due to the fact that it supplanted a set of religious norms that are common to most preindustrial societies, and which inhibit economic achievement; and it replaced them with norms favorable to economic achievement.
Because they experience little or no economic growth, preindustrial economies are zero-sum systems: upward social mobility can only come at someone else's expense. In any preindustrial society that has endured for some time, the cultural system is likely to have adapted accordingly: social status is
MODERNIZATION   AND    POSTMODERNIZATION
71
hereditary rather than achieved, and the culture encourages people to accept their social position in this life, emphasizing that meek acceptance and denial of worldly aspirations will be rewarded in the next life. Aspirations toward social mobility are sternly repressed. Such value systems help to maintain social solidarity and discourage economic accumulation in a variety of ways, ranging from norms of sharing and charity, to the norms of noblesse oblige, to the potlatch and similar institutions in which one attains prestige by recklessly giving away one's worldly goods.
For Weber, the central element in the rise of modernity was the movement away from traditional religious authority to secular rational-legal authority: a shift from ascriptive status to impersonal, achievement-based roles, and a shift of power from society to state. Traditional value systems must be shattered in order for modern economic development to take place. In a society undergoing rapid economic expansion, social mobility is acceptable, even a virtue. But in hunting and gathering or agrarian societies, the main basis of production— land—is a fixed quantity, and social mobility can only occur if an individual or group seizes the lands of another. To preserve social peace, virtually all traditional cultures discourage upward social mobility and the accumulation of wealth. They help to integrate society by providing a rationale that legitimates the established social order, in which social status is hereditary; but these cultures also inculcate norms of sharing, charity, and other obligations that help mitigate the harshness of a subsistence economy.
The Confucian system was an exception in one important respect. Although (like virtually all traditional cultures) it inculcated the duty to be satisfied with one's station in life and to respect authority, it did permit social mobility based on individual achievement, through the Confucian examination system. Moreover, it did not justify meek acceptance of one's lot in this world, by stressing the infinitely greater rewards that this would bring in the next world. It was based on a secular worldview: if one were to rise, one would do so in this world or not at all.
On the whole, however, the traditional value systems of agrarian society (China included) are adapted to maintaining a stable balance in unchanging societies. Accordingly, they tend to discourage social change in general and accumulative entrepreneurial motivation in particular, which is stigmatized and relegated to pariah groups if tolerated at all. Economic accumulation is characterized as ignoble greed. To facilitate the economic accumulation needed to launch industrialization, these cultural inhibitions must be relaxed.
In Western society, the Protestant Reformation helped break the grip of the medieval Christian worldview on a significant part of Europe. It did not do this by itself. The emergence of scientific inquiry had already begun to undermine this worldview. But Weber's emphasis on the role of Protestantism captures an important part of reality. Prior to the Reformation, Southern Europe was economically more advanced than Northern Europe. During the three centuries after the Reformation, capitalism emerged—mainly in Protestant countries, and among the Protestant minorities in Catholic countries. Within this cultural
72
CHAPTER    3
context, economic accumulation was no longer despised. Quite the contrary, it was highly respected because it was taken as evidence of divine favor: those whom God had chosen, he made rich.
Protestant Europe manifested a subsequent economic dynamism that was extraordinary, moving it far ahead of Catholic Europe. Shifting trade patterns, declining food production in Southern Europe, and other factors also contributed to this shift, but the evidence suggests that cultural factors played a major role. Throughout the first 150 years of the Industrial Revolution, industrial development took place almost entirely within the Protestant regions of Europe, and the Protestant portions of the New World. This began to change only during the second half of the twentieth century, when those regions that had been most strongly influenced by the Protestant Ethic—and had become economically secure—began to deemphasize economic growth. As we will argue, they did so precisely because they had become economically secure. At the same time, an entrepreneurial outlook had emerged in Catholic Europe and (even more strikingly) in East Asia, both of which are now showing higher rates of economic growth than Protestant Europe. The concept of the Protestant Ethic is outdated if we take it to mean something that can only exist in Protestant countries. ButWeber's more general concept that culture influences economic growth is a crucial insight.
Modernization: The Shift from Religious Authority to State Authority
Secularization is inherently linked with Modernization. This holds true despite frequent assertions that a rapid growth of fundamentalist religion is taking place throughout the world. This interpretation reflects a misconception of what is happening, generalizing from two very different phenomena. The apparent rise of religious fundamentalism reflects two disparate elements:
1.  Advanced industrial societies in North America, Western Europe, and East Asia, traditional forms of religion have been, and still are, declining, as we will demonstrate. During the past 40 years, church attendance rates have been falling and adherence to traditional norms concerning divorce, abortion, suicide, single parenthood, and homosexuality have been eroding—and continue to erode. Resurgent fundamentalist activism has indeed been dramatic: gay bashing and the bombing of abortion centers have received widespread coverage in the mass media, encouraging the perception that these actions have a rapidly growing constituency. They do not. Instead, precisely because fundamentalists correctly perceive that many of their central norms are rapidly eroding, they have been galvanized into unprecedented activism. But this reflects the rearguard action of a dwindling segment of the population, not the wave of the future.
2.   Islamic fundamentalism, on the other hand, does have a growing mass constituency. But it is growing in societies that have not modernized: though
MODERNIZATION   AND    P O S T M O D E R N I Z A  T I O N                                          73
some of these societies are rich, they have not become rich by moving along the Modernization trajectory of industrialization, occupational specialization, rising educational levels, and so on, but simply by virtue of the fact that they have large oil revenues. Even without modernizing, it is possible to become rich if one possesses large petroleum reserves that can be sold to industrialized countries, enabling traditional elites to buy the external trappings of Modernization.
The possession of this wealth is important: it has enabled oil-rich fundamentalist regimes to obtain such things as automobiles, air conditioning, modern medical treatment for elites, and, above all, modern weapons: without them, the fundamentalist regimes would be perceived as militarily weak and technologically backward—and their mass appeal and prospects for survival would be far weaker.
Modernization involves more than the shift away from cultural traditions (usually based on religious norms) that emphasize ascribed status and sharing, toward placing a positive value on achievement and accumulation. For Weber, the key to Modernization was the shift from a religion-oriented worldview to a rational-legal worldview. There were two key components of Modernization.
1.  Secularization. Weber emphasized the cognitive roots of secularization. For him, the rise of the scientific worldview was the crucial factor that led to the decline of the sacred/prerational elements of religious faith. We suggest that, more recently, the rise of a sense of security among mass publics of advanced welfare states has been an equally important factor in the decline of traditional religious orientations. This difference in emphasis has important implications. The cognitive interpretation implies that secularization is inevitable: scientific knowledge can diffuse across national boundaries rapidly, and its spread is more or less irreversible. By contrast, the rise of a sense of security among mass publics takes place only after a society has successfully industrialized; and it can be reversed to some extent by rapid change or economic decline. Thus although scientific knowledge has been permeating throughout the world for many decades, religious fanaticism continues to flourish in societies that are still in the early stages of industrialization; and fundamentalist movements continue to emerge among the less secure strata of even the most advanced industrial societies, especially during times of stress.
2.  Bureaucratization. The process of secularization paved the way for another key component of Modernization, Bureaucratization, the rise of "rational" organizations, based on rules designed to move efficiently toward explicit goals, and with recruitment based on impersonal goal-oriented achievement standards. A prerequisiite for bureaucratization was the erosion of the belief systems supporting ascriptive traditional authority and zero-sum economies; and their replacement by achievement-oriented, rational, and scientifically oriented belief systems that supported the authority of large, centralized bureaucratic states geared to facilitating economic growth. The core of cultural Modernization was the shift from traditional (usually religious) authority to rational-legal authority.
74
CHAPTER    3
Along with this went a shift of prestige and socioeconomic functions away from the key institutions of traditional society—the family and the church— to the state, and a shift in economic activity from the small family enterprise to mass production that was state-regulated or even state-owned. Globally, it was a shift of prestige and power from society to state.
During the modernizing phase of history, it seemed (to Marxists and non-Marxists alike) that the direction of social evolution was toward the increasing subordination of the individual to a Leviathan state having superhuman powers. The state would become an omnipotent and benevolent entity, replacing God in a secular world. And for most of the nineteenth and twentieth centuries, the dominant trend (the wave of the future, as it was sometimes called) was a shift from societal authority toward state authority, manifested in the apparently inexorable growth of the economic, political, and social role of government. Even non-Marxist thinkers such as Schumpeter (1947) reluctantly considered the triumph of socialism to be inevitable. And until recently, even such mainstream figures as Lindblom (1977) thought that the only question was whether socialism would triumph over capitalism, or whether capitalism and socialism would continue to coexist. The possibility that socialism might give way to capitalism was not even entertained.
The Postmodern Shift
The socialist leviathan-state was the logical culmination of the Modernization process, but it did not turn out to be the wave of the future. Instead, the expansion of the bureaucratic state eventually approached a set of natural limits, and change began to move in a new direction. Figure 3.1 illustrates what happened. From the Industrial Revolution until well into the second half of the twentieth century, industrial society underwent Modernization. This process transformed political and cultural systems from traditional regimes legitimated by religious belief systems to rational-legal states legitimated by their claim to maximize the welfare of their people through scientific expertise. It was a transfer of authority from family and religious institutions to political institutions.
Within the last 25 years, a major change in the direction of change has occurred that might be called the Postmodern shift. Its origins are rooted in the economic miracles that occurred first in Western Europe and North America, and later in East Asia and now in Southeast Asia. Coupled with the safety net of the modern welfare state, this has produced unprecedentedly high levels of economic security, giving rise to a cultural feedback that is having a major impact on both the economic and political systems of advanced industrial societies. This new trajectory shifts authority away from both religion and the state to the individual, with an increasing focus on individual concerns such as friends and leisure. Postmodernization deemphasizes all kinds of authority,
MODERNIZATION   AND    POSTMODERNIZATION                                          75
RATIONAL-LEGAL
AUTHORITY Economic Growth
Achievement
Motivation
#
c .o
*
©
Religious &
Communal
Values
Steady-State Economy
TRADITIONAL
AUTHORITY
Figure 3.1. The shift from Modernization to Postmodernization: changing emphasis on key aspects of life.
whether religious or secular, allowing much wider range for individual autonomy in the pursuit of individual subjective well-being.
The core function of culture in traditional society was to maintain social cohesion and stability in a steady-state economy. Norms of sharing were crucial to survival in an environment where there was no social security bureau and no unemployment benefits: in bad times, one's survival depended on how strongly the norms of sharing were inculcated.
The importance of these norms is almost certain to be underestimated by anyone brought up in an individualistic society. In relatively traditional societies such as Nigeria, even today people feel a strong obligation to help take care of not only their immediate family, but their brothers, sisters, cousins, nieces, nephews, and old friends and neighbors. These norms are highly functional in traditional societies: they enable people to survive who would otherwise starve. In industrial societies, this sense of obligation has eroded almost to the point of extinction.
The core project of Modernization is economic growth, and the means to attain it is through industrialization—the systematic application of technology
DEEMPHASIS OF
AUTHORITY
Maximizing
Well-being
Postmaterialist Values
76
CHAPTER    3
Table 3.1
Traditional, Modern, and Postmodern Society: Societal Goals and Individual Values
Traditional
Modern
Postmodern
Core Societal    Survival in a steady-Project             state economy
Individual Value
Authority System
Maximize economic growth
Maximize subjective well-being
Traditional religious       Achievement motivation     Postmaterialist and
and communal norms Traditional authority
Rational-legal authority
Postmodern values
De-emphasis of both legal and religious authority
to maximize the output of tangible things, such as wheat, textiles, coal, steel, and tractors.
In Postmodernization, the core project is to maximize individual well-being, which is increasingly dependent on subjective factors. Human behavior shifts from being dominated by the economic imperatives of providing food, clothing, and shelter toward the pursuit of quality of life concerns.
Even economic behavior becomes less a matter of meeting the survival needs and becomes increasingly oriented toward attaining subjective well-being. Economic growth continues, but output consists less and less of tangible things that contribute directly to survival, and more and more of intangibles whose value is subjective. The Postmodernization writers are on target in emphasizing the increasingly.subjective nature of life experience in advanced industrial society.
For example, government has become an enormous sector, now employing a larger proportion of the U.S. workforce than does industrial manufacturing. Government services are intangible, and their value is highly subjective—people even disagree about whether their value is positive or negative. Computer software, education, research, entertainment, and tourism have all become major industries. Unlike food, clothing, and shelter, their products are intangible and their value is largely subjective. Computer software, microchips, and entertainment have become three of the United States' largest exports, but the value of the film or silicon or disk on which they are stored is negligible. A successful motion picture or computer program may be worth hundreds of millions of dollars; another film or program that costs just as much to produce may be virtually valueless. Ideas and innovation are the crucial component—and their value is whatever people feel it is worth. With psychotherapy and tourism, this is equally true: they have become major economic activities, and their value lies almost entirely in their contribution to subjective well-being.
Table 3.1 compares the societal goals and individual value systems underlying traditional, modern, and postmodern society. As it indicates, the core societal goal of traditional society is survival under the conditions of a steady-state economy, in which social mobility is a zero-sum game. During the
MODERNIZATION   AND    POSTMODERNIZATION
77
Modernization phase, by contrast, the core societal project is maximizing economic growth—and, in both capitalist and socialist societies, it tends to be carried out by ruthlessly extracting the necessary capital from an impoverished populace, regardless of the costs to the environment and quality of life. In Postmodern society, by contrast, the top priority shifts from maximizing economic growth to maximizing subjective well-being.
From Survival Values to Well-being Values
Individual-level value systems reflect the core societal project of the three respective types of societies. Traditional societies vary enormously, but virtually all of them emphasize individual conformity to societal norms limiting violence, sexual behavior, and economic accumulation; and encouraging acceptance of the existing economic and social order. These norms are usually codified and legitimated within a religious framework. Perhaps the most central individual-level change linked with Modernization is the rise of achievement motivation; but the broad shift toward instrumental rationality weakens all traditional norms.
During the Modernization era, there was a consensus throughout industrial society that economic growth was not only a good thing, but virtually the ultimate good: though Marxists and capitalists disagreed sharply about how the fruits of production should be distributed, both sides shared an implicit consensus that economic growth was desirable. This consensus was unquestioned because it seemed self-evident. Economic growth and scientific discoveries constituted Progress: they were good almost by definition.
During the Cold War there was a similar shared sentiment that the question of whether East or West was the better society would be decided by which one achieved the most economic growth. And during the first half of the Cold War, the Eastern bloc seemed to be winning by the test that really counted: high growth rates. In 1972 Meadows et al.'s The Limits to Growth called this consensus into question, arguing that economic growth was not desirable and should be brought to a stop before it was too late. Shortly afterward, Schumacher's (1973) Small Is Beautiful questioned another key principle of the Modernization era: the tendency to equate Biggest with Best—a tendency that was widespread, but especially strong in the socialist bloc, where bigness and centralization were elevated almost to the rank of moral virtues. Both of these critiques reflected the emergence of well-being values, a core element of Postmodernism.
From Achievement Motivation to Postmaterialist Motivation
In the Postmodern shift, values that played a key role in the emergence of industrial society—economic achievement motivation, economic growth, economic rationality—have faded in salience. At the societal level, there is a radical shift from the priorities of early industrialization, and a growing tendency
78
CHAPTER    3
for emphasis on economic growth to become subordinate to concern for its impact on the environment. At the individual level, maximizing economic gains is gradually fading from top priority: self-expression and the desire for meaningful work are becoming even more crucial for a growing segment of the population. And the motivations for work are changing, from an emphasis on maximizing income as the top priority, toward increasing emphasis on the quality of the work experience. There is even some willingness to accept ascriptive criteria rather than achievement criteria for recruitment, if it is justified by social goals.
Scarcity has prevailed throughout most of history: it follows from the ecological principle that population normally rises to meet the available food supply and is then held in check by starvation, disease, and war. The result has been chronic scarcity, with the possibility of starvation shaping the daily awareness and life strategies of most people. Both traditional and modern societies were shaped by scarcity, but industrial society developed the belief that scarcity could be alleviated by individual achievement and economic growth, a radical change in outlook.
The root cause of the Postmodern value shift has been the gradual withering away of value systems that emerged under conditions of scarcity, and the spread of security values among a growing segment of the publics of these societies. This, in turn, grows out of the unprecedentedly high levels of subjective well-being that characterize the publics of advanced industrial society, as compared with those of earlier societies. In advanced industrial societies, most people take survival for granted. Precisely because they take it for granted, they are not aware of how profoundly this supposition shapes their worldviews.
Starvation is no longer a real concern for most of the people in high-technology societies, where production has been increasing much faster than the rate of population growth. These societies have attained unprecedentedly high life expectancies and unprecedentedly high levels of subjective well-being. One consequence of this fact is the rise of Postmaterialist values, but this is only one component of a broader cultural shift. The emergence and spread of Postmaterialist values is only the tip of the iceberg—one component of a much broader syndrome of cultural changes that we term Postmodernization. There are several additional important components.
Growing Emphasis on Individual Freedom and Rejection of Bureaucratic Authority
The shift from traditional society to industrial society brought a shift from traditional authority to rational bureaucratic authority. In most societies, this simply substituted political authority for religious authority. But in Postmodern society, authority, centralization, and bigness are all under growing suspicion. They have reached a point of diminishing effectiveness; and they have reached a point of diminishing acceptability.
ODERN1ZATION   AND    POSTMODERNIZATION
79
Every stable culture is linked with a congruent authority system. But the Postmodern shift is a move away from both traditional authority and state authority. It reflects a declining emphasis on authority in general—regardless of whether it is legitimated by societal or state formulae. This leads to declining confidence in hierarchical institutions. Today, political leaders throughout the industrialized world are experiencing some of the lowest levels of support ever recorded. This is not simply because they are less competent than previous leaders. It reflects a systematic decline in mass support for established political institutions, and a shift of focus toward individual concerns.
Because Postmaterialists view self-expression and political participation as things that are valuable in themselves, the Postmodern phase of development is inherently conducive to democratization. There is nothing easy or automatic about this tendency. Determined authoritarian elites can repress it almost indefinitely, though at growing cost to the morale and cooperativeness of their subjects. Similarly, the institutional structure and cultural heritage of a given society can facilitate or retard this tendency, as can external pressures and other macropolitical factors. But as economic development takes place, mass input to the political process becomes increasingly widespread and effective. Economic development leads mass publics to place growing emphasis on participatory values.
In addition to the changes in core societal goals, individual values, and authority systems outlined in table 3.1, the Postmodern shift has two other aspects.
First, as Postmodern philosophers argue, an essential attribute of post-modernity is a diminishing faith in science, technology, and rationality. One of the core components of Modernization was a growing faith in the power of science and rational analysis to solve virtually all problems. At the elite level (especially among Postmodern writers) Postmodernization is linked with a diminishing faith in rationality and a diminishing confidence that science and technology will help solve humanity's problems. This change in worldview has advanced farthest in the economically and technologically most advanced societies. And insofar as industrial society's culture of instrumental rationality is identified with the West, Postmodernity is linked with a rej ection of the West. But for mass publics, Postmodernity has also brought a rejection of the Soviet model, which was even more hierarchical and instrumentally oriented than the Western version of industrial society.
Initially, Postmodernism focused on discontent with the dehumanizing aspects of modernity as manifested in the West. Many of the most prominent Postmodernist thinkers even considered themselves Marxists (and some still do). But it was inevitable that Postmodernization would eventually lead to the rejection of hierarchical, bureaucratic, centralized big government in the socialist world as well, where it was most extreme. This contributed to an unexpected development: the collapse of socialism. State socialism failed because (1) it no longer functioned well, in advanced industrial society—though it had functioned relatively well during the Modernization era, and (2) be-
80
CHAPTER    3
cause it was no longer acceptable. The declining effectiveness and acceptability of massive, centralized bureaucratic authority contributed to the collapse of state socialism, as did the fact that Postmodernization brings an inherent tendency toward democratization, linked with its growing emphasis on individual autonomy.
Elements of Continuity between Modernization and Postmodernization
Postmodernization continues some of the trends that were launched by Modernization, particularly the processes of specialization, secularization, and individuation. The growing complexity of advanced industrial society results in increasing specialization in all areas of life. But the processes of secularization and individuation have taken on a new character.
Secularization
Weber attributed the decline of religious belief largely to the rise of the scientific worldview, which gradually replaced the sacred/mystical prerational elements of religious faith. Although the scientific worldview has lost its glamor, secularization continues—but for a new reason: the emergence of a sense of security among the economically more advanced societies diminishes the need for the reassurance that has traditionally been provided by absolute belief systems, which purport to provide certainty and the assurance of salvation, if not in this world at least in the next.
It would be a major mistake to equate either Modernization or Postmodernization with the decline of religion. Modernization does require the dismantling of some core aspects of traditional religion—in particular, it abolishes traditional tendencies to equate the old with the good, and the rigid rejection of social mobility and individual economic achievement. But—significantly—in the Protestant Ethic thesis, Weber argued that this was accomplished by having one type of religion replace another. The Marxist route to modernity achieved this by replacing traditional religion with a secular ideology that initially inspired widespread Utopian hopes and expectations of a new sort of Judgment Day that would come with the revolution. As it lost its ability to inspire such hopes, Marxism began to crumble.
In some form or other, spiritual concerns will always be a part of the human condition. This remains true after the shift from Modernization to Postmodernization. A core element in Postmodernization is the decline of instrumental rationality (equating economic growth with the good) to value rationality, seeking human happiness itself, rather than the economic means to that end. Although Postmodernism goes with a continuing decline in traditional religious beliefs, it is linked with a growing concern for the meaning and purpose of life.
MODERNIZATION   AND    POSTMODERNIZATION                                          81
Individuation
With industrialization, the erosion of religious social controls opened up a broader space for individual autonomy, but this space was largely taken up by growing obligations to the state. The Postmodern shift away from both religious and state authority continues this long-standing shift toward individuation, but in a much stronger form. Increasingly, individual rights and entitlements take priority over any other obligation.
Globally, there is a great deal of cross-national variation in degrees of Modernization: even today, only a minority of the world's population live in industrialized societies. An even smaller proportion of humanity live in the rich and secure advanced industrial societies in which Postmodern value systems have taken root.
Consequently, we would expect to find two main dimensions of cross-cultural variation across the 43 societies we are about to analyze. During the past two centuries, the two most pervasive and important processes that have shaped them have been (1) Modernization and (2) Postmodernization. Accordingly, we would expect the world's societies to vary according to the degree to which they have been transformed by these two processes. Furthermore, a given society's position on these two dimensions should be closely linked with its level of economic and technological development: societies that are only beginning to industrialize should manifest relatively traditional belief systems; those that are now in the stage of rapid industrialization should manifest value systems keyed to maximizing economic growth; and societies that had already attained high levels of existential security some time ago should have undergone an intergenerational value shift toward Postmodern values that give priority to subjective well-being over economic growth.
Modernization and Postmodernization Dimensions: Empirical Findings
We have outlined the patterns of cross-cultural variation we expect to find, and why. Now let us examine cross-cultural variation empirically, as reflected in survey data from 43 societies. Our first question is whether the various religious, social, economic, and political components of given cultures are randomly related, or whether they go together, with certain coherent combinations being more probable than others. Figure 3.2 shows the results of a principal components factor analysis of the data from representative national surveys in the 43 societies included in the 1990-91 World Values Survey. The responses to each of the variables used here are boiled down to a mean score for each country; using the society as the unit of analysis, we can examine cross-cultural variation in a wide range of norms and values.
Figure 3.2 sums up an immense amount of information. It presents an overview of findings from the World Values surveys, showing the relationships
82
CHAPTER    3
			SECULAR-RATIONAL					
				AUTHORITY				
	-1.00		-.80         -.60         -.40	-.20             0           +.20         +.40         +.60        +.80		+1.00		
	+1.00		I       I       I       I       I       I       I	1       1       1       1       1       1       1       1       1       I       1       1 Abortion		-	^ I.UU	
	+.80			ok Achievement      @ motivation Thrift    ©   Determination ®                 ©		-	+.80	
	+.60		Discuss ® politics	® Interested                Responsibility in politics                             •      Divorce ok Politics		—	+.60	
	+.40			® Important Trust people		—	+.40	
			Not        State/employee happy     @ management	® Homosexua		—		
	+.20	—	®	imagination			+.20	C3 Z
<			Hard Woman        @ work	RATIONAL-LEGAL                          Tolerance       Affect ) t                                              w             balance		_		
>			neels   s Child         M°ney		*                                       Ecology         ®			ED
_ a. => (J)	0 -.20	-   children" ""«?*   ®   ,          SUB @    both parents -     Reject  Technology outgroups   sta(e#				rialist   o ues rtant e    I ction „ on		
				> TRADI1	>  WELL-BEING    Women.s®       #       m va movement '                                                           Friends impc IONAL                                       ® ,   •                    Li Havľ Leisure     „,isf.			-J -I Ul
			responsible Trust science	free  important     ®				
			© Jobs to own nationality	choice            .   •    ,				
				In good health				
	-.40		Respect       Good & evil •                    ®			~	-.40	
		_		Want		__		
			parents          are clear	many				
			m	® children				
	-.60						-.60	
			Respect	®                   m				
			authority	Work     ®             Family important				
				important    Obedience				
	-.80	-	i       i       i       i       i       i       i	@ National pride ® Religious faith @God is important ,       ,       p Religion important..... 1       1       1       i       i       i       i       1       1       1       1       1		:	-.80	
	1.00		1       1       1       1       1       1       1				-1.00	
	-1.00		-.80         -.60         -.40	-.20             0           +.20         +.40         +.60        +.80 TRADITIONAL		+1.00		
				AUTHORITY				
Figure 3.2. Variation in the values emphasized by different societies: traditional authority vs. rational-legal authority and scarcity values vs. Postmodern values. Source: 1990-93 World Values Survey. This figure shows the first and second principal components emerging from a factor analysis of data from representative national surveys of 43 societies, aggregated to the national level. The scales on the margins show each item's loadings on the two respective dimensions. The items in italics (e.g., "reject outgroups") are multi-item indicators.
between scores of items. Subsequent chapters will probe more deeply into the causal relationships between key variables and examine changes over time. This figure is based on the responses given by nearly 60,000 respondents in 43 societies. We do not provide the full text of each question used here. A short phrase (such as "Abortion OK") is used to convey the gist of each item on figure 3.2; for the full text, see Appendix 2. The 43 variables used here reflect a much larger number of questions: some of them are based on responses to
ODERNIZATION   AND    POSTMODERNIZATION
83
whole batteries of questions. "Affect balance," for example, sums up each respondent's answers to the 10 questions in the Bradburn Affect Balance Scale; "Postmaterialist Values" sums up the responses to a series of questions through which each respondent ranks a set of 12 basic goals; and "Achievement Motivation" sums up responses to four items concerning important values for a child to learn; "Reject outgroups" also sums up the responses to several questions.
Furthermore, these variables were chosen to reflect a considerably larger number of related items that show similar patterns. "God is important," for example taps a cluster of more than 30 items that measure the extent to which religion is, or is not, an important part of the respondent's life. Similarly, "Life satisfaction," "Affect balance," and "Not happy" reflect a larger cluster of items that tap subjective well-being. To avoid redundancy, and to limit figure 3.2 to a readable size, we have only included the most sensitive indicators of each cluster. Figure 3.2 depicts the structure underlying responses to more than 100 questions dealing with many aspects of life in 43 societies, providing a global overview of basic cultural patterns.
Figure 3.2 shows the relationships between scores of variables covering a wide variety of topics ranging from religion to politics to sexual norms to attitudes toward science. These diverse orientations tend to go together in coherent patterns. For example, certain societies place relatively heavy emphasis on religion: the people of these societies also show high levels of national pride, prefer to have relatively large families, would like to see more respect for authority, tend to rank relatively low on achievement motivation and political interest, oppose divorce, and have a number of other distinctive cultural orientations. The people of other societies consistently fall toward the opposite end of the spectrum on all of these orientations, giving rise to a vertical dimension that reflects Traditional versus Secular-Rational orientations.
Figure 3.2 greatly simplifies a complex reality—in a sense, it is a one-page summary of the entire 1990 World Values Survey. It is, of course, an oversimplification. The present author has written two books on Postmaterialist values alone, and in this analysis, these values serve as only one indicator of a much broader Survival—well-being dimension. Nevertheless—to a surprising degree—reality fits this simplified model: over half of the cross-national variance among these variables can be explained by two dimensions that reflect the Modernization and Postmodernization processes, respectively.
Our first major finding is that there is a great deal of constraint among cultural systems. The pattern found here is anything but random. The first two dimensions that emerge from the principal components factor analysis depicted in figure 3.2 account for fully 51 percent of the cross-national variation among these variables. Additional dimensions explain relatively small amounts of variance. Moreover, these two main dimensions are robust, showing little change when we drop given items, even high-loading ones. The vertical axis reflects the polarization between Traditional authority and Secular-Rational authority; the horizontal axis depicts the polarization between a cluster of items labeled Survival Values and another cluster labeled Well-being Values.
84
CHAPTER    3
The scales on the borders of figure 3.2 indicate each item's loadings on these two dimensions.
Just two dimensions account for over half of the cross-national variance among these items: this also means that about half of the variance in these values and orientations is not explained by the Modernization and Postmodern-ization dimensions. It is important to keep this in mind. Historical change cannot be entirely reduced to universal processes: to a great extent, each society works out its history in its own unique fashion, influenced by the culture, leaders, institutions, climate, geography, situation-specific events, and other unique elements that make up its own distinctive heritage. General explanatory factors can never account for everything in cross-cultural research. Just as each individual is unique, each society is unique (and each historical moment is unique). Thus, while we find the metaphor of evolution useful in describing how social change works, we do not equate evolution with determinism. Certain strategies for coping with a given environment are far more probable than others: such a strategy represents a mutually supportive combination of economic, technological, political, and cultural factors, and one that is likely to survive—while other, almost limitless, dysfunctional combinations prove abortive. But social change also involves less systematic factors that make each society unique.
Brilliant and instructive books have been written about the ways in which given societies differ from others. This book focuses on the general themes underlying the cross-national pattern, not because we are uninterested in the unique aspects of given societies—few things are more fascinating—but because the common themes are also interesting, and because any book that undertakes to deal with more than 40 societies almost inevitably must focus on what is common, rather than on what is unique. The evidence examined here indicates that common underlying themes do exist: it suggests that roughly half of the cross-national variance in these values and attitudes can be accounted for by the processes of Modernization and Postmodernization, while the remaining half of the variation reflects factors that are more or less nation-specific.
Religion plays a much more important role in some societies than in others. In Nigeria, fully 85 percent of the population said that religion is "very important" in their lives; in South Africa, the figure was 66 percent; in Turkey, 61 percent; in both Poland and the United States, 53 percent; in Italy, the figure was only 34 percent; in Great Britain, France, and Germany, the figures were 16, 14, and 13 percent, respectively; in Russia, it was 12 percent; in Denmark, 9 percent; in Japan, it was 6 percent; and in China, 1 percent.
"Do societies that place relatively strong emphasis on religion also tend to favor large families?" The answer is an unequivocal "Yes," as the proximity of "Religion important" and "Want many children" near the bottom of figure 3.2 suggests: the correlation between these two items is r = .51 (significant at the .001 level). Moreover, societies characterized by an emphasis on religion also tend to place relatively strong emphasis on work, as the proximity between "Work important" and "Religion important" suggests (r = .62, signifi-
MODERNIZATIONAND    POSTMODERNIZATION                                          85
cant at the .0000 level). The emphasis here is on having work, for the sake of survival; in economically more developed societies, people place much greater emphasis on work as a source of personal satisfaction. Relatively traditional societies also tend to stress "Obedience" as an important quality to teach a child (r = .58), and to view the family as relatively important ("Family important," r = .56). And, as one would expect, those societies in which the public considers "Religion important" also tend to be those in which the public believe that "God is important," and to say that religious faith is an important quality to teach a child ("Religious faith"): these are almost 1:1 relationships (r = .95 and .87, respectively). These last two linkages are obvious; the others, though intuitively plausible, are not. All of these items have high loadings on the vertical dimension, labeled "Traditional Authority" vs. "Secular-Rational Authority."
Societies that place relatively strong emphasis on religion are characterized by very distinctive norms concerning sexual behavior, childrearing, the role of women, and fertility rates; they have distinctive attitudes toward divorce, abortion, and homosexuality; they also place relatively strong emphasis on deference to authority; and they have distinctive norms concerning economic achievement and distinctive motivations for work.
It is not particularly surprising that societies in which religion is relatively important have distinctive norms concerning abortion, childbearing, and the role of women. But these differences also extend to areas in which the connection is far from obvious. For example, societies in which religion is important are characterized by much higher levels of national pride than those in which it is not, as figure 3.3 demonstrates. Here, the horizontal axis shows the percentage in each society who say that God plays an important role in their lives. The people of societies that rank high on this variable show much higher levels of national pride than do those that rank low. China is a deviant case, with a high level of national pride despite being overwhelmingly secular, and West Germany deviates in the opposite direction, showing a lower level of national pride than its level of religiosity would predict. But the overall linkage is remarkably strong and significant at the .0000 level (see figure 3.3).
As these findings suggest, high levels of constraint exist between various cultural attributes. For example, if we know that a society ranks high on national pride, we can pretty accurately predict its position on childrearing practices, religiosity, and a number of other important attributes. But the pattern extends even farther. Societies that emphasize the importance of religion tend to attach low importance to politics, as the locations of "Religion important" and "Politics important" (far apart from each other on the vertical dimension) suggests: the correlation between the two is —0.39. And these same societies tend to place low emphasis on "Thrift" and "Determination" as important qualities to teach a child (r = —.57 and —.59, respectively). As we will see in a more detailed analysis in chapter 5, emphasis on these values is part of an Achievement Motivation syndrome that is strongly linked with the economic growth rates of given societies.
86
CHAPTER    3
80
70
60
■S     50 -
40
30
20 -
10
		Ireland
		• •                    ---Indla             USA                        ^-""^
		Poland J^-**'^     M.
		J«T              Nigeria ^^^^®Turkey              9 Canada                      ^^         „   •.      „ ® ., @                         .^*^                    S. Africa  Brazil
	® Slovenia „ .± .              ,    ,     . Britain             Iceland	^^Mgentina    WIe|ico
	®         ©           ®	
__	Latvia                         Austria ®	/                    N. Ireland             „ ®, ^""^            Lithuania                         chlle
@	Norway                  ,,   ®                ^* ©                       Hungary ^*% Denmark &                                               ^^-"''Spain	Por'u9ai       Italy
£hina	Sweden  @                                 ^-^"^	0              ®
		® Czechoslovakia
—	Japan^^-^*"^                        Bulgaria j-"""^  @   E. Germany ^s^    Russia ^s^                  m„~,™.,      Netherlands «-"^                        Moscow               s, ®                ■        ©	
		
™	W. Germany 1                                               I	i                      i
20                              40                              60
% Giving HIGH rating to importance of God in life
80
1i
Figure 3.3. Coherent values patterns on the traditional vs. secular-rational authority dimensioi the linkage between religiosity and national pride.
Horizontal dimension shows percentage ranking importance of God in their lives as relativel high (i.e., scores of 7-10 on a 10-point scale ranging from "not at all important" [1] to "very in portant" [10]). r = .71, significant at .0000 level.
Coherent Value Patterns: The Postmodernization Dimension
In the Postmodernization phase of development, emphasis shifts from maximizing economic gains to maximizing subjective well-being. This gives rise to another major dimension of cross-cultural variation, on which a wide range of orientations are structured. So far, we have been discussing items with high loadings on the second principal component, labeled "Traditional Authority" vs. "Rational-Legal Authority." This dimension reflects the Modernization process, in which authority moves away from a traditional (usually religious) basis, toward increasing emphasis on impersonal bureaucratic authority. This is an important dimension, accounting for 21 percent of the variance among these 47 variables. But it is overshadowed by the first principal component, which accounts for 30 percent of the total variance. This dimension taps "Survival Values" versus "Well-being Values." A very sensitive indicator of this dimension is "Postmaterialist Values" (located near the right-hand pole of the horizontal axis on figure 3.2). This is a central element in a much broader cultural configuration.
Societies with large numbers of Postmaterialists tend to be characterized by a relatively strong sense of subjective well-being. Their publics tend to express high levels of satisfaction with their lives as a whole ("Postmaterialist Values" has a .68 correlation with "Life satisfaction"). Moreover, they report relatively high levels of positive affect (saying that within the past few days they felt in-
ODERN1ZATION   AND    POSTMODERNIZATION
87
terested in something, or proud, or pleased about having accomplished something) rather than negative affect (reporting that they were restless, or felt lonely, or upset because someone criticized them), which produces high scores on the Bradburn "Affect balance" scale. Furthermore, the publics of societies with high levels of Postmaterialism are likely to rate themselves as "In good health," (r = .58) and are not likely to describe themselves as "Not happy" (the correlation with "Postmaterialist Values" is —.71).
Subjective well-being is a condition, not a value, and is not correlated with Postmaterialism at the individual level. But high levels of subjective well-being are a key element in the cultural syndrome called Postmodernism. When a society attains high levels of economic security and subjective well-being, it is conducive to Postmaterialist values; but further economic development does not necessarily bring increased subjective well-being.
The linkage between Postmaterialism and subjective well-being is a cultural syndrome, not an individual-level ideology. It reflects the fact that societies with high levels of economic development not only have relatively high levels of objective need satisfaction (being relatively well-nourished, in good health, and having relatively high life expectancies); but their publics also experience relatively high levels of subjective security and well-being, which leads to an intergenerational shift toward Postmaterialist values. This cultural syndrome has gone largely unnoticed in previous Modernization literature, but manifests itself clearly when one has survey data covering a sufficiently broad range of countries.
At the individual level, however, Postmaterialists do not report relatively high levels of subjective well-being. Far from being a paradox, this is central to their nature: Postmaterialists have experienced relatively high levels of economic security throughout their formative years. They develop Postmaterialist priorities precisely because further economic gains do not produce additional subjective well-being: they take economic security for granted and go on to emphasize other (nonmaterial) goals. Moreover, they have relatively demanding standards for these other aspects of life—to such an extent that they often manifest lower levels of overall life satisfaction than do Materialists in the same society.
This leads to another finding that at first seems paradoxical. Generally, within any given society, the rich show higher levels of subjective well-being than the poor, as common sense might suggest. But Postmaterialists are an exception: they are richer (and have better education, more prestigious occupations, etc.) than most people—but they do not rank higher on subjective well-being than other people. This is significant. It reflects the fact that, as given nations become advanced industrial societies, they reach a point of diminishing marginal utility at which maximizing economic gains (for the individual) or economic growth (for the society) no longer results in higher levels of subjective well-being (we noted this phenomenon in chapter 2). From this perspective, it is perfectly rational to cease making economic efficiency and economic growth top priorities, and give increasing emphasis to quality of life concerns.
This cultural syndrome is pervasive and lies at the heart of Postmoderniza-
88
CHAPTER    3
tion. The publics of societies with high proportions of Postmaterialists do not emphasize "Hard work" as one of the most important qualities to teach a child (reflected in a loading of —0.67 on the Scarcity-Security dimension); instead, they emphasize "Tolerance" and "Imagination." Similarly, their publics do not view more emphasis on "Money" as a desirable change.
The polarization between survival values and well-being values extends to family values as well. The publics of societies with high proportions of Post-materialists tend to reject the proposition that "A woman needs children" to be fulfilled, and disagree that "A child needs both parents," in a home with both a father and a mother, to grow up happily. There is a growing emphasis on self-realization for women, linked with a shift of emphasis from the role of mother to emphasis on careers.
"Respect parents" and "Respect authority" show strong loadings on both dimensions in figure 3.2. Their loadings indicate that both the Modernization process and the Postmodernization process are linked with declining respect for authority. And "Good and Evil are clear" has a negative relationship with both the shift from traditional authority to rational-legal authority and the shift from survival values to well-being values. A growing moral relativism is linked with both Modernization and Postmodernization. In traditional societies, moral rules are absolute truths, revealed by God. At the opposite extreme, in Postmodern society, absolute standards dissolve, giving way to an increasing sense of ambiguity.
We have argued that these two dimensions reflect the Modernization process and the Postmodernization process, respectively. And the fit is generally good. For example, the rise of Achievement Motivation is strongly linked with the vertical (Modernization) dimension. Moreover, the rankings of the global domains of life fit the expected configuration: as we move up the vertical dimension we see a shift in emphasis from family and religion (as indicated by "Family important" and "Religion important") toward increasing emphasis on the state ("Politics important"). Then, as we move from left to right on the horizontal dimension, we move away from emphasis on both traditional authority and state authority, toward increasing emphasis on individual concerns: "Leisure important" and "Friends important" show loadings of .66 and .72, respectively, on the Postmodernization dimension.
An emphasis on science and technology was a core element of modernity. But the publics of societies with high proportions of Postmaterialists (at the Postmodern end of the continuum) tend to have little confidence that scientific advances will help, rather than harm, humanity ("Trust science" has a negative correlation with "Postmaterialist values" that is significant at the .001 level); similarly, they tend to doubt that more emphasis on "Technology" would be a good thing. Conversely, these same societies have relatively high levels of support for the "Ecology" movement. The fact that societies shaped by security tend to reject science and technology is a major departure from the basic thrust of Modernization—another reason why this dimension reflects change in a Post-modem direction.
MODERNIZATION   AND   POSTMODERNIZATION
89
100
I
80 -
60 -
o
e
o
s: <a > a sz o
40
20-
-		Russia	TuJey      india Lithuania                  ® @			
_	Bulgaria ®	^s.        Latvia Belarus^.      ® Moscow ^-<s	m                       ® \E?tonia           JaPan		China ®	
-			Czechoslovakia © N.		_...   Mexico Chile  @ ®	
					Portugal	N. Ireland
			Slovenia ®		Austria ©        ^ Argentina   Italy any                  W. Germany Brazil   •  Britain Spain ® e               ® Be	@ USA ©
-			E	Germ @ Franc ®		Ireland ■s.        ®    Canada \^           ® Finland«. ĹTwly         fe,a„, ® Sweden^^^^ Denmark      ^^-^
		I                                I	I		I	Netherlands ® © i               i
20
30                   40                    50                    60                   70                    80
% "High" on life satisfaction (scores of 7-10 on 10-point scale)
90
Figure 3.4. Coherent values patterns on the survival vs. well-being dimension: the linkage between life satisfaction and rejection of homosexuals as neighbors (part of the "Reject outgroups" cluster), r = —.75, significant at .0000 level.
Societies influenced by Postmodern or well-being values tend to be markedly more tolerant than are those characterized by survival values. These societies emphasize "Tolerance" as an important quality to teach a child. Similarly, their publics are less likely to "Reject outgroups," saying that they would not like to have foreigners, people with AIDS, or homosexuals as neighbors; and they are relatively likely to feel that homosexuality is acceptable ("Homosexual OK"). Both of these correlations with "Postmaterialist values" are significant at the .001 level. Moreover, societies with relatively high levels of subjective well-being rank relatively low on intolerance of outgroups, as figure 3.4 illustrates.
The outgroup dealt with here is homosexuals, but the same pattern applies to rejection of other outgroups. In Russia and Belarus, where subjective well-being was extremely low in 1990, fully 80 percent of the public said they would not like to have homosexuals as neighbors. In such societies as Denmark or the Netherlands, where overall life satisfaction was much higher, only about 10 percent of the public were unwilling to have homosexuals as neighbors. Numerous other orientations are closely related to whether a society has high or low levels of subjective well-being.
Security is conducive to tolerance and conversely, insecurity is conducive to xenophobia. The narrower one's margin for survival is, the more likely one is to fear that strangers are threatening. This is especially true if the strangers
90
CHAPTER    3
speak a foreign language or hold different values and therefore seem incomprehensible and unpredictable.
In an agrarian or hunting and gathering society in which the land supply is just sufficient to feed the existing population, the arrival of a foreign group poses a direct threat to survival: in such a situation, xenophobia is realistic and almost certain to arise. In a technologically advanced society with a growing economy, foreigners may be tolerated or even welcomed. They do not pose a threat to survival and may even enhance the standard of living. But in times of economic or political crisis, even advanced industrial societies are prone to xenophobia, as the rise of fascism during the Great Depression demonstrated, and as recent events in Western Europe and the United States continue to demonstrate. But the severity of xenophobia tends to be proportionate to the degree of insecurity; hence, ethnic conflict is far more severe in Eastern Europe, where the economic systems and political systems have collapsed, than in Western Europe: far more people have been killed in ethnic conflicts in Eastern Europe, by several orders of magnitude.
No culture is immune to xenophobia, but it tends to be most intense where insecurity is most severe. Conversely, at the individual level, Postmaterial-ists—those who have grown up under conditions of relative economic and physical security—tend to be relatively tolerant of people with different ethnicity or sexual orientations. Similarly, they are relatively supportive of the "Women's movement." The rise of security values seems conducive to increasing tolerance of diversity, an essential component of democracy.
An environment of security and subjective well-being seems to foster not only tolerance, but a whole cluster of traits that are conducive to democracy. For example, well-being values are linked with high levels of interpersonal trust (as reflected in the .66 loading of "Trust people" on this dimension). Moreover, a participant public is an essential component of democracy—and one of the defining characteristics of Postmaterialist values is the fact that they give a high priority to self-expression and participation in decision making at all levels, including the political. Postmaterialism constitutes a central component of Postmodern values. Are these values linked with stable democracy? As we will see shortly, the answer is Yes.
In addition to its emphasis on science and technology, another key characteristic of Modernization was its tendency to bureaucratize all aspects of life, with the biggest bureaucracy of all resulting from the seemingly inexorable growth of government. But Postmodern values are linked with declining support for big government: believing that the state (rather than the individual) should take more responsibility to ensure that everyone is provided for ("State responsible") is linked with survival values, and not with well-being values; the same is true of support for "State/employee management" rather than owner management. Support for big government was a central component of Modernization. It does not go with Postmodern values, which is another indication that Postmodernization reflects a fundamental change of direction.
MODERNIZATION   AND    POSTMODERNIZATION
91
The analysis presented here is not the only possible way to slice the data. If one applies varimax rotation to the factors, one gets a somewhat different solution. Similarly, one can generate a three-dimensional or six-dimensional or even a 30-dimensional solution. Doing so produces a far more complicated result that might superficially seem more scholarly. But if one is looking for the simplest possible configuration, a reasonable approach is to use principal components analysis and focus on the first dimensions. Figure 3.2 is what emerges then—a structure that sums up a surprisingly wide range of phenomena in just two dimensions that capture over half of the total cross-national variance in this array of orientations. Additional dimensions exist, but they explain relatively small amounts of additional variance. The reality is that cross-cultural variation is surprisingly orderly and can be interpreted with a relatively parsimonious model.
Another critique of this approach would be to point that it is based on the assumption that our questions have comparable meaning to people from 43 widely varying societies, who were interviewed in 31 different languages. Our questionnaire was, of course, designed with this in mind: building on extensive previous cross-national survey research and extensive pilot testing, with input from social scientists on five continents, it was designed to ask questions that do have a shared meaning across many cultures. If we had asked questions about nation-specific issues, the cross-cultural comparability would have broken down. In France, for example, a recent hot political issue (linked with Islamic immigration) was the question whether or not girls should be allowed to wear scarves over their heads in school. This question would have had totally different meanings (or would have seemed meaningless) in other societies. On the other hand, a question about whether religion is important in one's life is meaningful in virtually every society on earth, including those in which most people say it is not. The same is true of questions about respect for authority, or about how many children one would like to have, or whether or not one is satisfied with one's life as a whole.
Moreover, the cross-national placement of societies underlying this configuration of worldviews is astonishingly coherent. As we will see below, societies that show similar cultural orientations in our surveys fall into compact and theoretically meaningful clusters. Working independently and without knowledge of each other's findings, the World Values survey investigators in the five Nordic countries came up with relatively similar results. So did our colleagues in Nigeria and South Africa, and so did those in the four Latin American countries and in Eastern Europe. The evidence suggests that the World Values survey group was generally successful in framing cross-nationally meaningful questions. If these items had idiosyncratic meanings in each society, we would not have attained such a parsimonious and coherent structure. Instead, the orientations that went together in one society would be unrelated in other societies, and it would take 15 or 20 dimensions to explain half of the variance.
92
CHAPTER    3
Where Are Given Societies Located on These Dimensions? A Cultural Geography of the World
In most respects, the two dimensions in figure 3.2 show a good fit with the attributes we would expect to find if they reflected the shift from Traditional to Modern values, and from Modern to Postmodern values, respectively. But in one important way the pattern seems wrong: the growth of big government was a central aspect of Modernization. For many decades, the all-encompassing socialist state was thought to be the wave of the future: it was the logical culmination of the trend toward bureaucratization and state authority. If so, we would expect to find emphasis on "State/employee management" and "State responsibility" located near the top of the vertical dimension. Instead, we find them occupying roughly neutral positions on this dimension. Why? In order to understand the answer, let us examine the specific national cultures underlying this pattern.
Figure 3.5 shows the location of each society on the two dimensions we have been examining. To locate them in this space, dummy variables were created for each of our 43 societies; these variables were mapped onto the two dimensions shaped by the worldviews of the respective publics. Because these dummy variables are extremely skewed (each having one country coded "1" and 42 countries coded "0"), the correlations with the cultural dimensions are modest; but if we combine countries into larger groups (such as the Nordic group or the Latin American group) the correlations with the ideological space become quite strong. The societies that show similar cultural orientations in our surveys (and therefore are near each other on this figure) fall into intuitively plausible clusters.
Our broadest generalization is that the value systems of richer countries differ systematically from those of poorer countries. The poorer countries tend to be located toward the lower left on figure 3.5, with the richer ones falling into the upper right-hand quadrant. Although there are some deviant cases (the United States having much more traditional values than its GNP per capita would predict), the overall correlation between values and economic development is very strong.
But the pattern is coherent in many additional respects. For example, all four of the Latin American societies included in the 1990 World Values Survey fall into one cluster, reflecting the fact that, in global perspective they have relatively similar value systems. The two African societies fall into another cluster; and the three Confucian-influenced societies of East Asia fall into another cluster—which partly overlaps with another cluster containing the former communist societies. The historically Catholic societies of Western Europe fall into another compact cluster. Although church attendance in Western Europe has collapsed, the historically Protestant societies of Northern Europe fall into another cluster (with Eastern Germany located at the intersection of the Northern European cluster and the ex-communist cluster, as its historical experience might suggest). The United States and Canada constitute a North American
MODERNIZATION   AND    POSTMODERNIZATION
93
SECULAR-RATIONAL AUTHORITY
first principal component factor loadings
- +.20
5
UJ
m
UJ
TRADITIONAL
AUTHORITY
Figure 3.5. Where given societies fall on two key cultural dimensions. Source: 1990-93 World Values Survey. Positions are based on the mean scores of the publics of the given nation on each of the two dimensions.
cluster—that could be expanded to include the other English-speaking societies. Poland is an outlier, having more traditional values than the other ex-communist societies of Eastern Europe: it does not fit into any coherent cluster. But on the whole, the value systems of a majority of the world's people are anything but random: though shaped by a variety of factors, they manifest a remarkably coherent pattern that can be interpreted parsimoniously.
94
CHAPTER    3
Societies that are close to each other on figure 3.5 show relatively similar responses to most of the questions that were asked in the World Values survey. For example, though the peoples of the United States and Canada differ in many ways, they have relatively similar basic values in comparison with most other societies: accordingly, they are located close to each other on figure 3.5.
Just how close are they? To answer this question, let us examine the responses to the first six questions asked in the survey. The publics of each society were asked "Please say, for each of the following, how important it is in your life." They rated the relative importance of "Work," "Family," "Friends," "Leisure," "Politics," and "Religion." This list covers a broad range of human concerns, and the various publics differed a great deal in how much importance they attached to the respective aspects of life. The cross-national rankings for nearly 350 variables are shown in Basanez, Inglehart, and Moreno (1997), and they support the claim that societies that are close to each other on figure 3.5 tend to show similar values and beliefs in many other ways; here, we will limit ourselves to these six variables.
Since 43 societies are ranked here, the greatest possible distance between any two societies would occur if one of them were ranked first and the other were ranked forty-third: all 41 of the other societies would fall between them. Conversely, the smallest possible distance would occur if the two societies had consecutive ranks, with none of the other societies falling between them. Finally, if the two societies were randomly distributed, about 20 societies would fall between them.
The publics of the United States and Canada give relatively similar ratings to these six aspects of life: on the average, only 3.3 other societies fall between them. The values of Canadians and Americans are much more similar to each other than they are to those of most other peoples. The British are also relatively close to the Americans, but by no means as close as the Canadians. On the other hand, the publics of the United States and China generally give these six domains quite different ratings: on the average, they are separated by nearly 24 societies, as the following ratings show:
Cultural Distance from the United States (mean number of societies between the United States and given society)
United States-Canada         3.3
United States-Britain         9.0
United States-France        16.7
United States-Japan         21.8
United States-Russia        22.2
United States-China         23.8
This principle applies generally. One finds a similar pattern if one views the world from a Swedish perspective, for example. Sweden, Norway, and Denmark are located relatively near each other on figure 3.5, and they make rather
MODERNIZATION   AND    POST MODERNIZATION
95
similar ratings of the six aspects of life. On the average, only 3.8 societies fall between Sweden and Norway in these rankings, and only 5.8 societies fall between Sweden and Denmark. On the other hand, Sweden and France are separated by a mean of 13.6 societies, while Sweden and Russia are separated by an average of 17.8 societies, Sweden and Japan by 18.2 societies, and Sweden and China are separated by an average of more than 19 societies.
Given groups of nations take coherent positions on the two dimensions. For example, Norway, Iceland, Denmark, Finland, and Sweden—the five Nordic countries—form a compact cluster located in the upper right-hand quadrant of figure 3.5: all five have related histories and similar cultures, ranking fairly high on the cultural outlook associated with rational-legal authority, and ranking very high on Postmodern values. To some extent, these countries are geographically proximate, but the fact that they are prosperous and traditionally Protestant welfare states seems more important than their geographic proximity. Thus the Netherlands, which is not a Nordic country but was historically Protestant and is today a prosperous welfare state, falls squarely into the middle of the Nordic group. Although geographically located next door to Belgium and sharing a common language with half of Belgium, the Netherlands is culturally much closer to the Nordic countries than to Belgium. Historically, the Netherlands has been shaped by Protestantism; even the Dutch Catholics today are remarkably Calvinist. And although the churches themselves are now a fading influence in Western European society, religious traditions helped shape enduring national cultures that persist today. Thus, culturally, the Netherlands is located somewhere between Norway and Sweden.
Belgium, France, Italy, Spain, Portugal, and Austria constitute another cluster in the cultural space of figure 3.5. Although church attendance has declined drastically, all of these countries were historically Roman Catholic. Furthermore, this cluster is adjacent to a Latin American (and overwhelmingly Catholic) cluster containing Mexico, Argentina, Chile, and Brazil. These predominantly Catholic countries form a fairly coherent group. One could even expand it to include the four other historically Roman Catholic countries, Poland, Hungary, Slovenia, and Lithuania. The last four countries are outliers, probably because of their divergent histories since 1945: the rising prosperity experienced by Western European Catholic countries in recent decades had much less impact on them, and they are more permeated by survival values than are the rest of the Catholic group. On the Modernization dimension, however, their values are almost as traditional as those of other Catholic countries (and they have more traditional values than the other exsocialist countries). As Basanez (1993) demonstrates, the Protestant-Catholic differences do not simply reflect the fact that the historically Protestant countries tend to be richer than the Catholic ones: controlling for GDP/capita, the value differences between them remain significant at the .001 level.
Nevertheless, there is no question that traditional orientations are closely related to a society's level of economic development. Almost all of the economically less-developed countries fall into the lower left-hand quadrant of figure
96
CHAPTER    3
3.5, with cultures that emphasize traditional authority and survival values. But interestingly enough, all five of the English-speaking societies (Britain, Canada, the United States, Ireland, and Northern Ireland) fall into a cluster located in the lower right-hand quadrant: these countries have relatively strong security values, but much more religious-traditional values than most other countries at their economic level. This is particularly true of Ireland and Northern Ireland, which have a traditional/religious outlook that is fully as strong as that found in India, South Africa, or Brazil—with only Nigeria being markedly more traditional.
The former West German and East German regions of Germany were still independent states when these surveys were carried out and were sampled separately. Although West Germany falls into the upper right-hand quadrant with the other Western European societies, and East Germany into the upper left-hand quadrant containing most of the historically communist societies, the two societies are relatively close to each other on the two main cultural dimensions. This is significant. From 1945 to 1990, the communist regime made a massive effort to reshape East German culture to support a Marxist and atheistic authoritarian regime. Simultaneously, the Western powers launched massive efforts in West Germany to remake political culture to support a market-oriented Western liberal democracy. The evidence indicates that 45 years under radically different regimes did have an impact: by 1990 the two societies were some distance apart, especially along the Postmodernization dimension. But even more impressive is the fact that, in global perspective, the basic cultural values of the two societies were still relatively similar. This natural experiment indicates that, even when it makes a conscious and concerted effort to do so, the ability of a regime to reshape its underlying culture is limited. After 45 years under diametrically opposed political and economic institutions, in their basic values East Germany and West Germany remained as similar to each other as are the United States and Canada.
Almost all of the socialist or ex-socialist societies fall into the upper left-hand quadrant: these societies are characterized by survival values and a strong emphasis on state authority, rather than traditional authority. Poland is a striking exception, distinguished from the other socialist societies by its strong traditional-religious values. China is an outlier in the opposite direction—the least religious and most state-oriented society for which we have data. These societies' positions reflect their distinctive cultural heritages. On one hand, adherence to the Catholic church has been a mainstay of the Polish struggle for independence since 1792. The church continued to play a vital role in this struggle throughout the 1980s, revitalizing the role of religion in the national culture.
China, on the other hand, has had a relatively secular cultural system for 2,000 years; and bureaucratic authority developed within the Confucian system long before it reached the West. Thus China and the other Confucian-influenced societies of East Asia have had one major component of modern culture for a very long time. Until recently, they lacked the emphasis on science
MODERNIZATION   AND    POSTMODERNIZATION
97
and technology and the esteem for economic achievement that are its other main components; but their secular, bureaucratic heritage probably helped to facilitate their rapid economic development once these were attained. China's traditional emphasis on the state may have been reinforced by four decades of socialism. Japan, another Confucian-influenced society, and both Eastern and Western Germany are also characterized by relatively strong emphasis on rational-legal authority.
Most of the socialist and ex-socialist societies are oriented toward rational-legal, rather than traditional-religious, authority. Their people have experienced four to eight decades of socialist regimes in which religion has been systematically repressed and in which it is perfectly realistic to consider politics important because economic life, cultural life, and even one's chances of survival depend on the state. The socialist states were probably the most heavily bureaucratized, centralized, and secularized societies in history, and they held science and technology in such esteem that their elites legitimated their power by the claim that they ruled, not through the unscientific and fallible process of majority rule, but according to the principles of scientific socialism. By these standards, the socialist states represented the culmination of Modernization—and the fact that, on figure 3.5, they are located near the Modernization pole of the Traditional Authority-Rational-Legal Authority dimension seems appropriate. But figure 3.2 revealed one surprising anomaly: one would expect that such key ideological components of the socialist state as its tendency to hold the "State responsible" for providing for everyone's needs should also cluster near the Modernization pole and gain maximum support in the socialist societies. It does not. Why?
We suspect that if these surveys had been carried out a decade or two earlier, support for state management and state responsibility would have been relatively strong in the ex-socialist societies. Most of them had experienced relatively high economic growth rates from 1945 to 1975 or 1980. Up to this point, they seemed to be functioning well: they had done a good job of providing the basic necessities for nearly everyone and were able to conceal or repress criticism of their shortcomings in other aspects of life. Support for a state-run economy and society was probably a good deal higher then, in socialist countries, than it was in 1990. In this simpler, more orderly world, we would have found "State responsible" located near the Modernization pole. And the empirical picture may now again be closer to this model than it was in 1990: as the transition to market economies proved unexpectedly traumatic, reform communist elites returned to power in a number of ex-communist societies during the early 1990s.
But reality is complex. In 1990-91, when these surveys were carried out, the socialist economic and political systems were collapsing; and mass support for state-run economies had withered away in these societies. The classic model of state socialism was surviving only in North Korea and Cuba, and paradoxically mass support for socialism was no longer the wave of the future for industrial society, but had become a Third World phenomenon.
98                                                                                                                                    CHAPTER    3
SECULAR-RATIONAL AUTHORITY
first principal component factor loadings
-.30             -.20             -.10                0               +.10             +.20             +.30
+.30
+.20
+.10
SURVIVAL
second principal   o
component factor loadings
-.10
-.20
-.30
-.40
-.30             -.20             -.10                0               +.10             +.20             +.30
TRADITIONAL AUTHORITY
Figure 3.6. Where given societies fall on two key cultural dimensions: religious influences. Source: 1990-93 World Values Survey. Positions are based on the mean scores of the publics of the given nation on each of the two dimensions.
Institutional Determinism of Culture?
As we have seen, the historically Protestant countries of both Northern Europe and North America tend to cluster together to form one large group; similarly, the historically Roman Catholic countries of Western Europe, Latin America, and Eastern Europe tend to cluster together, forming another broad but reasonably cohesive cluster. Despite the enormous recent changes linked with economic and social Modernization, and despite the tremendous sociopolitical changes linked with communist domination of four traditionally Catholic societies throughout the Cold War, in global perspective the historically Catholic societies still have relatively similar cultural values—as do the historically Protestant societies. As figure 3.6 illustrates, the Catholic societies form a group characterized by more traditional values, and by greater emphasis on survival values, than holds true of most Protestant societies. At first glance, this might seem to constitute strong evidence for an institutional de-
- 0    WELL-BEING
MODERNIZATION   AND   POS T MODERNIZATION
99
terminist interpretation: the religious institutions of these societies led them to develop different cultures.
If institutional determinism is simply taken to mean that a society's institutions are among the factors that help shape its culture, it is undoubtedly correct. But institutional determinism is often pushed to a much more extreme claim than this. It is taken to mean that institutions alone determine a society's cultural values, so one need not really take cultural factors into account: if one changes the institutions, the culture automatically changes to fit it. If one examines the evidence more closely, it is clear that this position is untenable.
There are tremendous cultural differences between Protestant and Catholic societies, but for the most part they do not reflect the direct influence of the Catholic and Protestant churches today. For the direct influence of the church today is very slight in many of these countries. Although church attendance remains relatively high in Poland and Ireland (and the United States), it has fallen drastically in most of the historically Catholic countries of both Western and Eastern Europe; and it has fallen even more drastically in most historically Protestant European societies, to the point where some observers now speak of the Nordic countries as post-Christian societies: church attendance has plummeted toward zero. The societies that were traditionally Catholic still show very distinct values from those that were traditionally Protestant—even among segments of the population who have no contact with the church. But these values persist as part of the cultural heritage of given nations, and not through the direct influence of the religious institutions. This cultural heritage has been shaped by the economic, political, and social experience of the people, including the fact that the Protestant societies industrialized earlier than most of the Catholic societies—which at an even earlier stage of history may, in turn, have been linked with religious differences (as Weber suggests), but is certainly not a case of mere institutional determinism.
There is a remarkable degree of coherence to this pattern. Forty of the 43 societies fall into compact and historically meaningful clusters, such as Latin America or Eastern Europe or East Asia. There are only three outliers from this perspective: Poland plus Ireland and Northern Ireland (with the two parts of Ireland being closely linked). Both Poland and Ireland might be described as hyper-Catholic societies: they are Roman Catholic societies that for centuries were occupied and dominated by more powerful non-Catholic neighbors and that responded to pressures toward cultural assimilation by an intense reem-phasis on their Roman Catholic heritage as a means of preserving their national identity. Ironically, this may have led to a similar reaction on the part of the Irish Protestants, who constitute a small minority within Ireland as a whole and might be described as hyper-Protestants. Poland and both parts of Ireland strongly emphasize traditional cultural values concerning not only religion, but also politics, gender roles, sexual norms, and family values. They illustrate the fundamentalist reaction to threat.
In most countries, these cultural differences reflect the entire historical experience of given societies, and not the influence of the respective churches
100
CHAPTER    3
today. This point becomes vividly evident when we examine the value systems of such societies as the Netherlands and Germany—both of which were historically predominantly Protestant societies, but (as a consequence of different birth rates and different rates of religious attrition) have about as many practicing Catholics as Protestants today. Despite these changes in their religious makeup today, both the Netherlands and Germany manifest typically Protestant values. Moreover, the Catholics and Protestants within these societies do not show markedly different value systems: the Dutch Catholics today are as Calvinist as the members of the Dutch Reformed Church.
The historically Protestant and Catholic societies are not randomly distributed on our cultural map—far from it, they constitute coherent clusters. The communist ideology has been described as a secular religion, and the historically communist societies also make up a coherent cluster, which partly overlaps with the Catholic cluster: 13 of the 14 formerly communist societies fall into a compact cluster in the upper left-hand quadrant, and this Eastern European cluster could easily be expanded to include China and Eastern Germany. And as we have noted, though they are located on two different continents and span the Catholic-Protestant divide, the five English-speaking societies are also relatively near to each other on this cultural map: a common language is the unifying factor in this case. But the most pervasive influence of all seems to be economic development. If one draws a diagonal from the lower-left corner to the upper-right corner of figures 3.5 or 3.6, it traces the transition from poorer to richer societies. Both the Modernization dimension and the Post-modernization dimension are strongly correlated with a society's level of economic development: the values of richer societies differ systematically from those of poorer societies on both dimensions. Clearly, institutional determinism would be a far too simple interpretation of the evidence. Although the impact of religious institutions is evident, economic, political, geographic, linguistic, and other factors also play important roles. The worldview of a given people reflects its entire historical heritage.
To What Extent Was Modernization Theory Correct? Coherent Cultural Patterns Exist
We have found that constraint does exist among cultural patterns. To get a sense of just how true this is, let us imagine two extreme models, ranging from a world wholly without cultural constraint, to one with total constraint. In the former model, each society goes its own way: the fact that it possesses one specific cultural attribute has no influence on whether other attributes are present. Cultural components are randomly related. The other extreme model is one of total determinism: only a few cultural patterns are possible, and if one major component of a pattern is present, all the other elements are also present in every case.
As one would expect, the empirical findings do not fit either extreme model,
MODERNIZATION   AND    POSTMODERNIZATION                                       101
but they come much closer to the constrained model than to the random one. Fully half of the cross-cultural variation among this broad array of variables can be captured in just two dimensions. This picture is certainly not one of complete determinism: these two dimensions do not explain all of the variation among these 43 cultural indicators. But they do account for 51 percent of the variance—vastly more than the less than 5 percent that they would explain in a random model.
These Coherent Cultural Patterns Are Linked with a Society's Level of Economic Development
The fact that constrained cultural patterns exist does not, by itself, demonstrate that Modernization theory is correct: coherent cultural patterns might be found exclusively in given regions (such as Western Europe) as a result of given historical or religious traditions (such as Protestantism or Buddhism) without having any relationship to economic and technological change. Modernization theory, by contrast, implies that economic development is strongly linked with given cultural patterns—either because economic development produces specific types of culture, or because certain cultural patterns produce economic development. In short, Modernization theory implies that coherent cultural patterns exist, and that these patterns are linked with a given society's level of economic development. As figure 3.7 demonstrates, this clearly is the case.
We argued that the vertical dimension on figures 3.2 and 3.3 reflects the Modernization process, while the horizontal dimension reflects Postmodern-ization. The evidence presented in figure 3.7 indicates that economic development is conducive to both Modernization and Postmodernization, which are two successive stages of development: a society's per capita GNPis correlated with Modernization values at the .60 level, and at the .78 level with Postmodernization values. But other economic indicators show quite different relationships to these two respective dimensions. For example, the percentage of the labor force in the manufacturing sector is strongly correlated with the vertical (Modernization) dimension (r = .63) but much less strongly linked with the horizontal (Postmodernization) dimension (r = .22); but Postmodern values are strongly linked with the percentage of the labor force in the service sector (r = .79).
Bell argues that Postindustrial society has arrived when a majority of a society's workforce is employed in the service sector. There is a good deal of overlap between Postmodern society and Postindustrial society. But Bell's concept of Postindustrial society emphasizes changes in the structure of the workforce, while the term "Postmodern society" emphasizes cultural changes linked with economic security; we have argued that existential security is a key factor underlying Postmodernization. In keeping with this contention, we find that prosperity has as strong a relationship with Postmodern culture as does the composition of the workforce.
Similarly, the percentage of a given society's population having a secondary
102
CHAPTER    3
DC O I H 3 <
< Z
O
+1.0
+0.8 -
+0.6
+0.4
+0.2 -
D <
H-        0.0
-0.2
CC
ob <
ľ3 Ü LU W
-0.4
-0.6
-0.8
1.0 L—L
% in
Manufacuring
Sector
Growth GNP/Capita
Central Govt Expenditures as % of GNP
% with
Secondary
Education
a
Life 0    GNP/Capita
Expectancy        ®
% in Service ®        Sector % with Higher     ® Education
# of Years Democratic
High Fertility Rate
©
% Income to Top 20%
@
'''l___''''
-1.0       -0.8      -0.6      -0.4      -0.2        0.0          0.2          0.4
SURVIVAL vs. WELL-BEING
0.6
0.8
1.0
Figure 3.7. Economic and social correlates of two key dimensions of cross-cultural variation. Source: 1990-93 World Values Survey.
or higher education shows a .71 correlation with the Rational-Legal pole of the Modernization dimension—and a .63 correlation with the Well-being pole of the Postmodernization dimension. These findings support the claim by Lerner (1958), Inkeles and Smith (1974), and others that rising educational levels have contributed to major cultural changes.
The consequences of growing up in a setting in which one can take satisfaction of one's survival needs for granted, rather than in a society of scarcity, seems to have been underestimated by Bell, Inkeles and Smith, and others. The strong linkage between Postmodern values and a society's GDP/capita supports the interpretation that these are, indeed, security values: they have an extremely strong tendency to be found in relatively prosperous societies.
An even more striking contrast between the economic correlates of Modernization and Postmodernization is the fact that Modernization values have a substantial positive linkage with economic growth rates, while Postmodern values have a negative linkage with economic growth: Postmodern societies are even richer than modernizing societies, but they show lower economic growth rates.
Overall, the evidence in figure 3.7 suggests that if educational levels con-
MODERNIZATION   AND    POSTMODERNIZATION
103
tinue to rise and the workforce continues to move out of farming and manufacturing into the service and knowledge sectors, and if per capita income continues to rise, then we can expect to see a gradual modernization of preindus-trial societies and a shift toward Postmodern values, in advanced industrial societies.
Is Cultural Change Taking Place?
The fact that Postmodern values are strongly linked with economic development does not necessarily prove that as economic development takes place, these values will become more widespread. The linkage might be spurious. Long-term time series data are needed to demonstrate whether or not the predicted changes are taking place and help sort out the causal relationships. We don't have extensive time series data for most of the variables examined here, but we do have some—and virtually all of the available evidence points to the conclusion that a shift toward Postmodern culture is taking place in advanced industrial societies.
The most abundant time series data (by far) relate to Materialist/Postmaterialist values. Though this is only one component of the much larger syndrome of Postmodern values, it is a key indicator of this broader syndrome. Evidence of a shift from Materialist to Postmaterialist values will be presented in chapter 4; it shows that Postmaterialist values have increased in the past quarter century, and that they have increased at almost exactly the rate predicted by the intergenerational population replacement model (about one point on the values percentage difference index per year).
If Postmaterialist values are moving in the predicted direction, this suggests that the entire set of closely correlated Postmodern values may be moving in the same direction, since theoretically they share a common set of causes: the transition from conditions of scarcity to the relative security of Postmodern society.
This broader shift toward Postmodern values in general does seem to be taking place. In chapters 7 and 8 we test the hypothesis that all values having reasonably strong correlations with Postmaterialism are part of an intergenerational shift linked with population replacement. To do so, we first identified 40 variables that were strongly correlated with Postmaterialist values and were included in both the 1981 and 1990 World Values surveys. Our expectation was that whatever values were positively correlated with Postmaterialism would become more widespread over time, other things being equal. When we compare the 1981 responses, with responses to the same questions in the same countries in 1990, we find that in the great majority of cases these values shifted in the predicted direction. The available evidence suggests that cultural change is taking place in the predicted direction.
As hypothesized, the apparent decline of traditional values is strongly linked with economic growth: as figure 3.7 indicates, a country's growth rate from
104
CHAPTER    3
1965 to 1990 shows a .44 correlation with emphasis on Rational-Legal Authority (and a corresponding negative correlation with Traditional Authority). In chapter 6 we probe into the relevant causal linkages: the findings suggest that culture shapes economic life, as well as being shaped by it. In the early stages of industrialization, achievement motivation seems to contribute to economic growth. But insofar as this growth produces prosperity, in the long run it leads to cultural changes that tend to de-emphasize achievement motivation, leading to lower rates of economic growth. This points to still another contrast between Modernization and Postmodernization: while the Modernization process is linked with high rates of economic growth, Postmodernization is not. Quite the contrary, relatively high growth rates show a negative linkage (r = -.47) with well-being values. In part, this reflects the fact that Postma-terialists do not emphasize economic growth, and they tend to give priority to protecting the environment, if forced to choose.
Traditional values are negatively linked with economic growth, but positively linked with high fertility rates, as figure 3.7 demonstrates (r = .48). As we saw earlier, societies with Traditional values tend to emphasize the family and have relatively large numbers of children. It seems that this is not just a matter of lip service. A society's values and its actual fertility rate are closely linked, probably in a causal relationship. This tends to set up a self-reinforcing process: traditional values not only seem to inhibit a set of norms that are conducive to economic growth, they also encourage population growth rates that tend to offset the effects of whatever economic growth does occur, making it still more difficult to raise per capita income. Conversely, both Modernization and Postmodernization are linked with declining birth rates, so the pie gets divided up among fewer people—another example of how cultural and economic factors constitute mutually reinforcing syndromes.
Modernization, Postmodernization, and Democratization
Finally, the Postmodernization process has important political implications. Inkeles and Diamond (1980), Inglehart (1990), and others have argued that economic development is linked with cultural changes that are conducive to democracy, an argument that has been disputed by dependency theorists, neo-Marxists, and some rational choice theorists. As figure 3.7 indicates, there is no correlation whatever between the Modernization axis and the number of years for which a given society has been democratic. As Moore (1966) pointed out, modernization can give rise to either democratic or authoritarian regimes. But there is an amazingly strong correlation between the Postmodernization dimension and democracy: r = .88. We suggested earlier that high levels of subjective well-being, coupled with Postmodern values, including interpersonal trust, tolerance, and Postmaterialist values, should be conducive to democracy. The empirical evidence is remarkably strong: this cluster of cultural traits clearly is linked with stable democracy. One could argue that this
MODERNIZATION   AND    POSTMODERNIZATION                                       105
cultural syndrome is conducive to democracy, or that democracy somehow gives rise to a culture of trust, tolerance, subjective well-being, and Postmaterialist values, or that the cultural syndrome and the political institutions are mutually supportive. We will analyze the causal linkages more closely in chapter 5; for the moment, we will simply observe that Postmodern values and stable democracy go together very closely.
It has been known for some time that democracy is more likely to be found among relatively prosperous countries than among poorer ones (Lipset, 1960). The present body of evidence supports that conclusion. But the linkage between culture and democracy found here is even stronger than the linkage between economic development and democracy. This finding suggests that economic development by itself does not automatically produce democracy; it does so only insofar as it gives rise to a specific syndrome of cultural changes.
Putnam (1993) supports this interpretation, using aggregate time series data from 20 regions of Italy, extending from 1860 to the mid-1980s. He finds that given regions have varying degrees of a cultural syndrome termed "Civic Community" (characterized by trust, tolerance, solidarity, civic engagement, political equality, and civic associations), which is strongly correlated with the effectiveness of democratic institutions in the given regions. Economic development is also related to democratic effectiveness; but controlling for civic traditions, economic development has no impact whatever. On the other hand, a region's level of civic involvement in 1900 not only predicts subsequent civic involvement and institutional performance, but also helps explain subsequent economic development.
Why Does Change Follow Predictable Patterns?
Coherent trajectories of cultural change exist, with some cultural patterns being far more probable than others. Why?
The evidence suggests that in the long run, cultural change behaves as if it were a process of rational choice—subject to substantial cultural lags, and subject to the fact that the goals being maximized vary from one culture to another and can only be understood through empirical knowledge of the specific culture. In this loosely rational process, peoples are maximizing a variety of goods, the most basic of which is survival—and their cultures are survival strategies for a given people. Cultural variation tends to follow predictable patterns because some ways of running a society work better than others. If one is willing to concede that most people prefer survival to nonsurvival, then both Modernization and Postmodernization are linked with outcomes that almost anyone would consider "better." As figure 3.7 indicates, there is a .59 correlation between a society's female life expectancy and its level of Modernization, and a .65 correlation between life expectancy and Postmodernization. Neither process bestows moral superiority, but both Modernization and Postmodernization are linked with a markedly lower likelihood of dying prematurely from
106
CHAPTER    3
disease or malnutrition. Life expectancy today ranges from as low as 39 years in the poorest countries to almost 80 years in developed ones. This is a difference that few people would fail to appreciate, regardless of cultural orientation. It is one reason why the modernization syndrome had such pervasive appeal. Successful industrialization requires a relatively competitive, impersonal, bureaucratic, achievement-oriented form of social relations that tends to be dehumanizing and stressful; but in societies of scarcity around the globe, it came to be viewed as a worthwhile trade-off. Islamic fundamentalism remains an alternative model insofar as oil revenues make it possible to obtain many of the advantages of modernization without industrializing; but we would not expect this model's credibility and mass appeal to outlast the oil reserves.
The Modernization process brought substantial gains in life expectancy by maximizing economic growth, making it possible to sharply reduce the two leading causes of death in preindustrial societies—malnutrition and disease. But the linkage between economic development and rising life expectancy eventually reached a point of diminishing returns.
Postmodernization represents a shift in survival strategies, from maximizing economic growth to maximizing survival and well-being. Modernization focused on rapid economic growth, which provided a means to maximize survival and well-being under the conditions that emerged when rationalization and industrialization first became possible. But no strategy is optimal for all time. Modernization was dramatically successful in raising life expectancies, but it has begun to produce diminishing returns in advanced industrial societies. By emphasizing competition it reduced the risk of starvation—but it probably also increased psychological stress. As we have seen, subjective well-being levels are lower in the communist and ex-communist societies— which are, by some criteria, the most "modern" of all societies—than in the most traditional societies. These low levels of subjective well-being in the ex-socialist world are almost certainly linked with the current crisis in the former communist societies, but the 1981 World Values Survey found low levels there even before the collapse of communism. During the decade before 1990, symptoms of severe demoralization and psychological stress were evident in the former Soviet Union and were manifest in high rates of alcoholism and declining life expectancies.
Postmodernization, on the other hand, has a mildly negative linkage with economic growth, but a strong positive linkage with subjective well-being. With the transition from Modernization to Postmodernization, the trajectory of change seems to have shifted from maximizing economic growth to maximizing the quality of life.
Cross-sectional Evidence of Social Change
Cross-sectional data can be a useful supplement to time series data in understanding processes of socioeconomic change. Although time series data pro-
MODERNIZATION   AND    POSTMODERNIZATION
107
vide the only reliable measurements of changes over time, appropriate cross-sectional data can extend the scope of one's perspective in time and space: its configuration may reflect processes that occurred over many decades or even centuries.
Interpreted in connection with the available time series data, the cultural configurations found in the 43-nation World Values survey suggest that coherent, and even to some extent predictable, trajectories of political and cultural change are linked with given socioeconomic developments. These trajectories are not deterministic: the leaders and institutions, and the cultural and geographic heritage, of a given society also help shape its course. And development does not move in a simple, linear fashion: all trends eventually change direction.
But neither is socioeconomic change random and unpredictable, with each society following its own idiosyncratic course. On the contrary, change tends to follow clear configurations, in which specific clusters of cultural characteristics go together with specific types of political and economic change. The familiar Modernization syndrome of urbanization, industrialization, and mass literacy tends to have foreseeable consequences such as increasing mass mobilization. Modernization is linked with given cultural changes, such as a growing emphasis on Achievement Motivation, and a shift from Traditional to Rational-Legal Authority, which encompasses dozens of more specific changes.
Similarly, the emergence of advanced industrial society, with an increasing share of the public having higher education, being employed in the service sector, and feeling assured that their survival needs will be met, gives rise to a process in which high levels of subjective well-being and Postmodern values emerge—and in which a variety of attributes, from equal rights for women to democratic political institutions, become increasingly likely.
CHAPTER     4
Measuring Materialist and Postmaterialist Values
Materialist/postmaterialist value priorities are only one component of the much broader configuration that constitutes the Postmodernization dimension. But Materialist/Postmaterialist values are by far the best documented aspect of this broader configuration, having been measured in cross-national surveys carried out regularly since 1970. The initial research on intergenerational value change was based on this dimension, and it continues to play an important role in research on cultural change. This chapter examines the nature of Postmaterialist values and the controversy over how these values can be measured most effectively; the next chapter analyzes the intergenerational shift from Materialist to Postmaterialist priorities during the past quarter century.
Over the years since these values were first investigated, there has been considerable controversy about what they were tapping, how values should be measured, and how many dimensions one should use to analyze Materialist/Postmaterialist value priorities. This controversy continues; several important critiques of measurement techniques have appeared recently. The 1990 World Values Survey measured Materialist/Postmaterialist value priorities by asking the respondents in more than 40 societies a series of questions that began as follows: "There is a lot of talk these days about what the aims of this country should be for the next ten years. On this card are listed some of the goals which different people would give top priority. Would you please say which one of these you, yourself, consider the most important? And which one would be the next most important?"
Twelve goals were rated, but in order to reduce the task to manageable proportions, they were presented in three groups of four items; each set of four goals contained two items designed to tap Materialist priorities, and two designed to tap Postmaterialist priorities. These 12 goals are shown in abbreviated form in figure 4.1; the full text of this battery appears in Appendix 2.
The choices deal with broad societal goals rather than the immediate needs of the respondent: we wanted to tap long-term concerns, not one's response to the immediate situation. Among these 12 goals, six items were intended to emphasize survival needs: "Rising prices," "Economic growth," and "Stable economy" designed to tap emphasis on economic security, and "Maintain order," "Fight crime," and "Strong defense forces" designed to tap emphasis on physical security. These items tap two distinct types of needs, but both are "Materialist," in that they are directly related to physiological survival. We hypothesized that they would tend to go together, with only those who feel secure about the satisfaction of both needs being likely to give top priority to belonging, self-expression, and intellectual and aesthetic satisfaction—Post-
MEASURING    VALUES
1973 Surveys
Postmaterialist
More say on the job (59) Ideas count (52) Freedom of speech (44)
Less impersonal society (54) More say in government (49)-
,  ._ Fight rising prices (-34) Strong defense forces -36 J; ElľonomicU ^ (<39)' Stable economy (-41)- -Figh, agains, crime (-41) "Maintain order (-47)
Materialist
■ More beautiful cities (05)
109
1978 Surveys
Postmaterialist
More say on the job (60) Ideas count (52) Freedom of speech (43)
Less impersonal society (54) More say in government (49)
Strong defense forces (-36) Stable economy (-43)
'More beautiful cities (14)
•Fight rising prices (-32)
Economic growth (-45) Fight against crime (-47) Maintain order (-50)
Materialist
Figure 4.1. Stability over time: the Materialist/Postmaterialist dimension in nine Western nations in 1973 and 1978.
Mean factor loadings from each of the nine European Union countries surveyed in September 1973 and October-November 1978. Items with Materialist polarity are in boldface type; items with Postmaterialist polarity are in italics. Based on principal components analyses of ranking of the 12 goals. Source: Inglehart, 1990: 140.
materialist needs that the remaining six items are designed to tap. We view all of these needs as universal: every human being needs sustenance and safety— but also desires self-expression, esteem, and aesthetic satisfaction and has a sense of intellectual curiosity. Thus one finds art and music and other products of the search for beauty in all societies, and one finds magic, religion, myths, or philosophy, reflecting the desire to understand and interpret the meaning of life in even the poorest societies. Hungry people may not give top priority to self-expression and intellectual concerns, but given some respite from the struggle for survival, people will act on these needs unless circumstances stifle them. Our expectation, therefore, is that emphasis on the six Materialist items will tend to form one cluster, with the Postmaterialist items in another distinct cluster.
All cultures are multidimensional. Change can occur in sexual norms, tastes in food or music, political party preferences, trust in government, religious outlook, and numerous other aspects of life. The question is not whether the Materialist/Postmaterialist dimension is the only dimension along which change can occur, but whether such a dimension exists—and if so, whether significant change has been taking place in the predicted direction.
Our theoretical framework implies that emphasis on economic security and on physical security will tend to go together—and that those who feel insecure
110
CHAPTER    4
about these survival needs have a fundamentally different outlook and political behavior from those who feel secure about them. The latter are likely to give top priority to nonmaterial goals such as self-expression, belonging, and intellectual or aesthetic satisfaction.
Emphasis on economic and physical security were expected to go together for two reasons: (1) From a macrosocietal perspective, war tends to produce both economic and physical insecurity—both hunger and loss of life. Consequently, those generations that have experienced war are likely to feel less secure about both. (2) From a microsocietal perspective, poor individuals tend to be exposed to both economic and physical insecurity—both poverty and relatively high crime rates. The more affluent strata have resources that shield them, to some extent, from both.
Satisfaction of the survival needs, we hypothesized, leads to growing emphasis on nonphysiological or "Postmaterialist" goals. Alarge share of the public in Western societies have been socialized in an environment that provides an unprecedentedly secure prospect that one's physiological needs will be met. Consequently, Western publics' responses should tend to polarize along a Materialist/Postmaterialist dimension, with some individuals consistently emphasizing Materialist goals, while others tend to give priority to Postmaterial-ist goals.
To test this hypothesis, we performed principal components analyses of the rankings of these goals in each of the countries surveyed in 1990-93. For this analysis, each item was recoded as a separate variable with codes ranging from "1" to "3," indicating whether the given item was ranked as most important in its group of four items, as second most important, or as one of the two least important items.
Our variables are based on relative rankings, not absolute scores. This is crucial to operationalizing our hypothesis, but it means that the items are not independent, which raises problems for factor analysis. If one were rating only two items, for example, the rank of the first item would determine the rank of the second, automatically generating a -1.0 correlation between them. With three items, one would expect negative correlations of about .5. With a pool of four items, random answering would generate negative correlations of about .3 between all four items. Thus, there is a tendency for the ratings of all these items to be negatively correlated, which tends to spread any four items over more than one dimension. Nevertheless, as our empirical results show, this effect is dominated by an even stronger tendency for Materialist items to be chosen together, on one hand, and for Postmaterialist items to be chosen together, on the other.
The Materialist/Postmaterialist dimension has proven to be remarkably robust over time. As figure 4.1 demonstrates, factor analyses of survey data from the nine countries that were members of the European Community in 1973 reveal two clear clusters: at the top of the continuum, five Postmaterialist items cluster together, showing positive polarity; and at the opposite pole, all six Materialist items cluster together, showing negative polarity. This pattern is re-
MEASUR1NG    VALUES
111
markably uniform across all nine nations (detailed country-by-country results appear in Inglehart, 1977: 44-47). Moreover, this battery was used again in a subsequent survey in these same countries in 1978, and as figure 4.1 shows, the results were very similar to those from 1973. The results show a cross-national consistency that is truly remarkable. In survey after survey, five items— the same five items in every country—cluster near the positive end of the continuum. Six items—again, the same six in every country—are grouped near the negative pole. The remaining item falls near the midpoint.
The items that cluster toward the negative pole are the six Materialist items, and five of the six Postmaterialist items fall into the opposite group. A single item—the one concerning "More beautiful cities"—does not fall into either cluster. This item does not behave according to our expectations, a fact that we must explore in more detail. But the other 11 items live up to expectations to an almost uncanny degree. The consistency of responses to these items cannot be attributed to such common sources of spurious correlation as response set, which often occurs when respondents simply rate a series of items as either "good" or "bad" (or "very important" or "less important"): some respondents give similar ratings to a whole series of items. In the present case, this is impossible: one must rank each goal as being more important or less important than the others, in a format that gives no cues to the "right" answer.
Given respondents tend to be preoccupied with a consistent set of needs located toward either the Materialist or Postmaterialist side of the continuum. Eleven of the 12 items fall into two separate clusters, reflecting Materialist and Postmaterialist priorities, respectively. The item designed to tap aesthetic needs ("more beautiful cities") fits into neither cluster; with the same consistency by which the eleven other items did fit into their expected places, this one fails to show a strong positive loading in any of these countries. Why?
The answer is that this item does not simply evoke aesthetic needs, as it was intended to do. Instead, it also taps an Urban/anti-Urban dimension on which collective economic development is seen as conflicting with one's personal security. Here, this item shows a surprisingly strong relationship with the safety needs (see Inglehart, 1977:45-50). For many people, the term "cities" evokes fears of crime.
The relationship between aesthetic concerns and Postmaterialist values is clarified by another analysis utilizing the Rokeach Terminal Values Survey (see Rokeach, 1973) together with the Materialist/Postmaterialist battery, both of which were included in the American component of the Political Action study (Barnes et al., 1979). Factor analysis of the 12-item Materialist/Postmaterialist battery plus the 18-item Rokeach battery reveals an interesting structure. As we would expect, a number of dimensions are needed to capture the configuration of responses. But the Materialist/Postmaterialist dimension remains clearly recognizable, and several of Rokeach's items show substantial loadings on it. In the Postmaterialist cluster of this dimension, we find "Equality" and "Inner Harmony," plus a pair of items relating to the intellectual and aesthetic needs—"Wisdom" and "A World of Beauty." Ironically, the item we
112
CHAPTER    4
Table 4.1
The Materialist/Postmaterialist Dimension in 15 Western Nations, 1990 (Loadings on First Principal Component in Factor Analysis)
Less Impersonal Society                                              .60
More Say on Job                                                          .62
More Say in Government                                            .50
Ideas Count More than Money                                   .51
Freedom of Speech                                                      .34
More Beautiful Cities                                                  . 18
Strong Defense Forces                                           — .26
Fight Rising Prices                                                 - .28
Fight against Crime                                               —.39
Maintain Order                                                      — .55
Economic Growth                                                   — .59
Maintain Stable Economy                                    — .63
Note: Items with Materialist polarity are in boldface type; items with Postmaterialist polarity are in italics.
Source: 1990-91 World Values Survey data from France, Britain, West Germany, Italy, the Netherlands, Denmark, Belgium, Ireland, Norway, Sweden, Denmark, Finland, Iceland, Canada, and the United States.
designed to tap the aesthetic needs fails to show the expected empirical relationships—but the item developed by Rokeach does. Analysis of the responses to our own item concerning "more beautiful cities" revealed an unexpected tendency for this item to be linked with emphasis on "the fight against crime." Including the word "cities" in this context seems to evoke a concern with safety among some respondents; for them, the cities are unbeautiful not only because they are dirty but because they are dangerous. Rokeach's item makes no reference to cities and evokes aesthetic concerns in unmixed form—and consequently falls into the Postmaterialist cluster as the need hierarchy hypothesis would suggest. In short, the anomalous results obtained with this item seem to reflect imperfect formulation of our "aesthetic" alternative, rather than an indication that aesthetic concerns are not part of the Postmaterialist syndrome. When unambiguously formulated, they are.
The pattern found in 1973 and 1978 continues to manifest itself in the 1990s. The loadings on the first factor in 15 Western countries are shown in table 4.1 (the detailed results for each country are shown in Abramson and Inglehart, 1995). In this table, the goals designed to tap Materialist priorities appear in boldface; the goals intended to tap Postmaterialist priorities appear in italics. With almost incredible consistency, in nation after nation, emphasis on the six items designed to tap economic and physical security goes together, forming a coherent Materialist cluster; in every case, emphasis on the five items de-
MEASURING    VALVES                                                                                                                       JL1-J
signed to tap belonging, self-expression, and intellectual satisfaction also goes together, forming a clearly defined Postmaterialist cluster.
A "Unidimensional" Theory of Value Change?
The Materialist/Postmaterialist thesis postulates that an intergenerational value shift is taking place along a dimension defined by two poles, termed "Materialist" and "Postmaterialist" values, respectively. This space is, by definition, one-dimensional and, though human values are almost infinitely multidimensional, the theory focuses on this one dimension. In this sense only, the theory might be called "unidimensional," though this is a potentially misleading label.
Just how misleading this label can be becomes evident in a recent analysis by Buerklin, Klein, and Russ (1994). Claiming that the Materialist/Postmaterialist value change theory is "unidimensional," they interpret this term to mean that we believe all human values constitute one dimension. If this claim were true, then the analysis carried out by Buerklin et al. would be a valid test of the theory: if the Materialist/Postmaterialist value change thesis actually did hold that all human values constitute a single dimension and that everything on it is undergoing intergenerational change, then if one can find any value that is not undergoing an intergenerational shift, the thesis would be disproven. Buerklin et al. analyze a battery of questions that measures a number of different things and have no difficulty in finding various values that are not undergoing intergenerational change, so, they conclude, they have refuted the
thesis.
But do we really believe that human values are unidimensional, in the sense that Buerklin et al. claim? Demonstrably, no: though our analyses focus mainly on the Materialist/Postmaterialist dimension, we have, from the start, also analyzed various other value dimensions.
Our analysis of the Materialist/Postmaterialist shift focuses on one dimension. It does so because the theory generates clear predictions about the type and direction of change one should find along this specific dimension: un-precedentedly high levels of security are producing an intergenerational shift from giving top priority to security, to giving top priority to self-expression. The investigation focuses on this one dimension because the theory does not predict what might be happening in other value domains. The Materialist/Postmaterialist dimension is a central and important part of mass value systems, but (obviously) it does not cover everything. There are almost limitless numbers of other values, and changes might conceivably be occurring in almost any of them, ranging from communism versus capitalism, to whether people prefer Madonna to Mozart, or chocolate to vanilla, but the theory does not deal with them. Nevertheless, it is clear that we recognized from the start that other value dimensions also exist and discussed them at some length (Inglehart,
114
CHAPTER    4
1977:45-51). In subsequent publications, (Inglehart, 1978,1979) we analyzed the relationships between Materialist/Postmaterialist values and the 18 Rokeach Terminal Values which are designed to tap the full range of human concerns. And in Inglehart (1990) we explored the relationships between Materialist/Postmaterialist values and a wide variety of other values, extending from religion to politics to sexual norms. Clearly, the claim that the theory is "unidimensional" in the sense of considering the Materialist/Postmaterialist dimension to be the only value dimension that exists, or the only dimension worth analyzing, is groundless.
The Study of Value Priorities: Rankings versus Ratings
The Materialist/Postmaterialist thesis deals with priorities, not levels: this a crucial distinction. We did not think that people had never before valued freedom of expression: history makes it clear that it has been valued for centuries. Nor did we think that the new generation no longer valued economic security and order, but now preferred crime and deprivation: it is pretty obvious that most Postmaterialists want peace and prosperity, like everyone else. In other words, we did not think it likely that there had been a reversal of polarity in core societal goals. But it did seem possible that a more subtle change had occurred: though both old and young still valued both prosperity and freedom, there had been a change in the relative priority attached to such goals. While the older generation were likely to give top priority to economic and physical security over self-expression if they had to choose, we hypothesized that a growing segment of the postwar generation would give self-expression priority over economic and physical security, if it came to a crunch.
In short, the shift is not a reversal of polarities, in which most people formerly were opposed to freedom of speech or having more say on the job, but now favor them; nor is it one in which the Postmaterialists no longer like the items with Materialist priority, but have come to prefer crime and high inflation. It is, instead, a shift in priorities—something that would not necessarily be revealed in favorable/unfavorable ratings of given goals but would, nevertheless, have a crucial impact on the choices they made.
The intergenerational change is based on a shift in priorities. Consequently, we set out to determine which, among a set of almost universally desired goods, have top priority among different generations. Priorities reflect the way in which one ranks various goals (all of which may be considered desirable and important). Consequently, they are best measured by a ranking method. One can, under some circumstances, infer priorities from ratings, but ratings are generally not well suited to that task.
Building on criticisms made earlier by Herz (1979), Flanagan (1982, 1987), and Klages (1988, 1992), Buerklin et al. imply that the ranking technique is a means of forcing respondents to say things they do not really mean: "We wish to make it clear that the so-called 'Ranking' procedure violates the principle
MEASURING    VALUES
115
of Independence of Measurement. One obtains from this 'extreme forced-choice situation' (Klages, 1992: 26) only a distorted and imperfect picture of the value space. . . . The ranking procedure forces the respondent to differentiate value items along a single dimension" (Buerklin et al., 1994: 585; my translation).
Contrary to this claim, the use of ranking techniques is a commonly used and widely accepted method in survey research. It is virtually unavoidable in any well-designed study of basic values. Thus, it was used by Milton Rokeach in his classic studies of human values. Rokeach is quite explicit about why he used ranking instead of ratings. Even though rankings are a far more time-consuming method than simple ratings, he required his respondents to rank each of his 18 terminal values and 18 instrumental values because he realized that practically everyone attached positive polarity to all 36 items and would score almost all of them as "important" or "very important" if they were simply asked to rate them. He did so because he understood that one's value priorities are a genuine and crucial aspect of one's motivation—and something quite distinct from what is measured by ratings.
As we have noted, the ranking method tends to cause each item in a given pool to be negatively correlated with all of the other items in that pool. With random answering, each item in a pool of four items would have a negative correlation of about — .33 with each of the other items. This means that the first two items will tend to fall at the opposite poles of the first dimension; but the third item will be negatively correlated with both of them, which tends to force it off the first factor and onto another. The same is true of all additional items: ipsivity tends to force them off the first factor. We get a Materialist/Postmaterialist dimension not because of, but in spite of, the ranking format. The fact that we nevertheless obtain a clear Materialist-Postmaterialist dimension as the first principal component in survey after survey reflects the fact that people are not answering these items randomly: certain people tend to choose Materialist goals consistently, while others tend to emphasize Postmaterialist goals consistently. The format gives no cue concerning which are which: they cluster together because of their content, not their format.
By contrast, the rating method tends to produce a one-dimensional solution: insofar as it encourages response set, it produces a first principal component on which everything is positively correlated with everything else, regardless of content, so that all items have high positive loadings on the first principal component. As we will see below, this is exactly the pattern that appears in the data collected by both Buerklin et al. and Bean and Papadakis (1994) when they employ the rating technique (see table 4.4, first unrotated factor). These authors interpret this as simply meaning that people want all of these goals, overlooking the fact that this tendency is inflated by response set.
It is perfectly true that most people would like to eat their cake and have it too—and this poses no problem when no choice is necessary. But the crucial situations in politics (and life in general) arise when choices are necessary— when one must choose between civil liberties and maintaining order, or be-
116
CHAPTER    4
tween building highways and protecting the environment, or choosing which of two attractive people one will marry. In such situations, ratings may provide little or no guidance: one may like both alternatives. The choice is determined by one's priorities.
We certainly do not claim that the rating method is never useful; we often use it. But any battery designed to measure changes in basic value priorities will, almost inevitably, be extremely vulnerable to response set. This is true because any such battery will (and should) consist of items that have the following characteristics: (1) almost everyone considers them desirable, and (2) almost everyone considers them important. They would not be basic human values unless they had these characteristics.
It is conceivable that one might, for some reason, want to measure people's priorities among superficial and unimportant values, but it would make no sense to do so in connection with the study of intergenerational change. Such values would not be central and deep-rooted enough to show enduring intergenerational differences: they would change readily, in response to the immediate environment.
Ratings are a perfectly valid means of collecting some types of information, but they are not well suited to measuring priorities: people can and do give equally high ratings to many goals, without revealing their underlying priorities. This does not mean that one does not have any underlying priorities— they can be readily and reliably measured with properly framed questions. The question is not whether one should always use ratings or always use rankings: both techniques are useful, but it is important to be aware that they measure two different aspects of reality.
The Problem of Response Set
One of the classic methodological problems associated with having respondents rate a long series of items in identical format is the danger that the ratings will be shaped, not by the items' content (as the investigator assumes), but by the fact that they are asked in an identical format: the similarity of format cues the respondent into giving similar responses to a whole series of items, especially if they appear near each other in the questionnaire. Respondents with relatively little education are especially prone to do this, since this way of responding requires no cognitive effort or information—but, of course, it yields misleading data. This problem has been well known for decades, having been revealed as a major problem in the research underlying the classic Authoritarian Personality study; partly because the problem was already recognized so long ago, it is sometimes lost sight of in contemporary research. It remains as real as ever.
The ratings obtained from the battery used by Buerklin et al. seem to reflect the questionnaire format as much as the content of the items used. The ranking method is much less vulnerable to response set: by its very nature, one can-
MKASURING    VALUES                                                                                                                        11/
not simply run down the list, giving similar scores to a series of consecutive items: one is forced to select one item for top priority, another for second priority, and so on. This method has the disadvantages of ipsivity, but it is capable of measuring values or attitudes with much less noise than tends to be present in responses to rating scales.
It also is more useful for testing a specific theory. It targets the inquiry much more sharply than do ratings. It enables one to focus on measuring the respondents' priorities along a specific dimension. This, of course, implies that one has a specific theory in mind that one wishes to test: it is too tightly focused to be useful at the exploratory stage, or if one is not guided by some clear theory.
Rankings versus Ratings: An Empirical Comparison
As we have pointed out, this theory is about priorities, not levels of support. Consequently, the items in the Materialist/Postmaterialist battery need to be ranked, not simply rated. The ranking method has two real disadvantages: (1) it is much more time consuming than ratings, and (2) it produces ipsitive data, which violate one of the assumptions of some statistical models; this is a difficulty, though not an insuperable one (Jackson and Alwin, 1980). Ranking, however, has two major advantages that, for present purposes, far outweigh the disadvantages: first, rankings are not determined by response set: when the respondent ranks the items, it is impossible to simply race down the list, giving relatively high (or relatively low) ratings to all of them. One is forced to consider each item separately, in relation to the others. Response set seems to be such a serious problem in the ratings obtained by Buerklin et al. that it is unclear if any but the first item in their battery is rated primarily in terms of its content: after the first item in the series, all of the succeeding items get uniformly high ratings (Inglehart and Klingemann, 1996). Second—and even more important—rankings are a fundamentally different approach, which is more suitable for the measurement of priorities than are ratings: rankings enable one to measure an aspect of reality that ratings may completely fail to uncover.
Buerklin et al. (1994) have raised an important and basic methodological question, but their database does not enable them to answer it since they employed neither the format nor the items used to measure Materialist/Post-matearialist values. What would they have found if they had used items that actually did tap the Materialist/Postmaterialist dimension—and then had also taken the precaution to ask them as both ratings and rankings, to test the effects of response set? Fortunately, this question is not merely hypothetical: exactly this was done in an article published almost simultaneously with the one by Buerklin et al. The results are extremely interesting.
Bean and Papadakis (1994) also argue that the Materialist/Postmaterialist dimension can better be measured by ratings than by rankings, and they pro-
118
CHAPTER    4
Table 4.2
Factor Analysis of Materialism-Postmaterialism
Items Based on Rankings: First Principal Component
Less Impersonal Society Ideas Count More than Money More Say in Government Freedom of Speech More Say on Job More Beautiful Cities
Maintain Stable Economy Fight Rising Prices Strong Defense Forces Fight against Crime Economic Growth Maintain Order
61
58 48 45 44 21
19
22 29 35 48 55
Note: Items with Materialist polarity are in boldface type; items with Postmaterialist polarity are in italics.
Source: National survey of Australians' Attitudes toward Welfare, 1988 (N= 1,814), inBeanandPapadakis, 1994: 275.
ceed to carry out an appropriate and well-designed test of this claim. To do so, they first administered a national survey in Australia that measured Material-ist/Postmaterialist values both by rankings (as advocated by Inglehart) and by ratings. They find that when the respondent uses the ranking method, one obtains a first principal component on which all six of the items designed to tap Materialist priorities form a cluster having negative polarity, and all six items designed to tap Postmaterialist priorities show positive polarity. In short, the authors find a clear Materialist-Postmaterialist dimension, with the Materialist items near one pole and the Postmaterialist items near the other pole. Table 4.2 shows their results. As this table indicates, the fit with theoretical expectations is virtually perfect. Although two of the items have relatively weak loadings on their respective factors, all 12 items show the polarity predicted by the theory. And although these results come from Australia, they show a pattern that is almost identical to results obtained by other investigators using the same methodology, in dozens of other countries.
Using a rating format, however, the authors get a completely different set of results: the first principal component is one on which all 12 items are given relatively high (or relatively low) ratings. When subjected to varimax rotation, this factor breaks down into a Materialist factor and a Postmaterialist factor (which are uncorrelated since varimax rotation was used). Using this approach, there seems to be no such thing as a Materialist/Postmaterialist dimension. Table 4.3 shows the results obtained from ratings.
The results seem contradictory. The authors argue that the rating approach reflects reality better than the ranking approach because most respondents are really in favor of all 12 goals: they want freedom of speech and law and
MEASURING    VALUES
119
Table 4.3
Factor Analysis of Materialism-Postmaterialism Items Based on Ratings:
First Principal Component Analysis with Varimax Rotation
	First Unrotated Factor	Rotated Factors	
		Postmaterialism	Materialism
Materialist Items			
Maintain Order	.47	.06	.67
Fight Rising Prices	.56	.39	.41
Economic Growth	.46	.04	.69
Strong Defense	.42	.04	.63
Stable Economy	.56	.24	.59
Fight Crime	.51	.09	.70
Postmaterialist Items			
More Say in Government	.53	.52	.19
Protect Free Speech	.49	.51	.14
More Say in Jobs	.61	.70	.11
Beautify Cities and Countryside	.54	.65	.05
More Humane Society	.64	.75	.08
Society Where Ideas Count	.56	.74	-.04
Source: National survey of Australians' Attitudes toward Welfare, 1988 (N = 1,814), in Bean and Papadakis, 1994: 278.
order—and economic growth and environmental protection and low inflation and more say in government and strong defense forces. The rating approach is more "flexible" than the ranking approach because it enables respondents to give high ratings to all 12 goals. Most respondents give priority to the Materialist goals over the Postmaterialist goals (or vice versa) only when they are constrained to do so by a ranking format.
Which approach captures reality? It depends on which aspect of reality you want to capture. Ultimately, it comes down to the question, Do you want to measure priorities or response levels?
It is perfectly true, as Bean and Papadakis point out (together with Klages and Buerklin et al.), that nearly everyone would like to attain all of these goals. All 12 goals were designed to tap things that nearly everyone considers desirable: this is inherent in any battery designed to measure changing priorities among basic values. Thus, in an unconstrained format, people can give positive ratings to virtually everything. This tendency is strengthened by the fact that a rating approach puts a dozen positive goals in an identical format, creating a series where the easiest thing to do is run down the list, giving similar ratings to item after item. This is a perfect setting for response set to occur. The rating list is a convenient way to quickly elicit a long stream of responses. It seems extremely quick and efficient. The problem is that it can yield virtually meaningless responses. This format systematically tends to create or inflate
120
CHAPTER    4
positive correlations between the items in a series. Furthermore, it provides a systematic bias: less-educated people are particularly likely to be influenced by response set. It is cognitively undemanding: one need not differentiate between the items—one can simply give a series of high (or low) ratings.
The less educated are particularly prone to response set, and in keeping with this fact it is very significant that Bean and Papadakis find (when, and only when they use ratings) that the less educated constitute the high scorers on both their Materialist dimension and their Postmaterialist dimension (Bean and Papadakis, 1994: 280). In striking contrast to this finding, in virtually every survey ever carried out, the Postmaterialists identified by the ranking method consist of those with relatively high educational levels—to such an extent that some analysts have even suggested that Postmaterialist values simply reflect high educational levels (Duch and Taylor, 1993). This is a remarkably clear illustration of the fact that the rating method fails to identify those with Post-materialist priorities: the "Postmaterialists" identified by ratings have a mean educational level below that of the sample as a whole. Needless to say, the "Postmaterialists" identified in this manner do not show the other demographic characteristics of true Postmaterialists: most strikingly, they are not concentrated among the young.
Ratings elicit information quickly—but they are not well suited to measuring priorities. Rankings, on the other hand, are much more time-consuming and difficult than ratings. With this format, the respondent cannot simply race down the list giving similar (usually high) ratings to every item. He or she must painstakingly decide which, among a series of desirable goals, is most important, which ranks second, and which are least important. This is a more demanding method. But for analyzing politics, this provides much more useful information than simply knowing that given individuals rate almost everything as "extremely important."
This is true because politics is, above all, a question of choices. True, environmental protection does not always conflict with economic growth; nor does maximizing individual freedom necessarily conflict with maintaining order. But these are the easy situations, when the solution is obvious and everyone can be satisfied. The crunch comes when two or more highly important and desirable goals do conflict with each other: it is precisely then that important political issues arise. Sometimes you cannot eat your cake and have it too. Under these circumstances, one is forced to make choices. And it is then that priorities become crucial.
The ratings approach is quite appropriate for many purposes. But it is not an appropriate way to measure value priorities—which is the topic on which the Materialist/Postmaterialist value change thesis focuses. The theory does not hypothesize that some people are opposed to economic and physical security, while others like it. It assumes that these are goals sought by almost everyone. The shift from Materialist to Postmaterialists priorities occurs when a given segment of the population has been raised with a sufficiently high level of economic and physical security that further emphasis on them has diminishing
MEASURING    VALUES
121
marginal utility. It is precisely because Postmaterialists have security (and not because they reject it) that they give top priority to other goals.
Thus, Bean and Papadakis (1994) and Buerklin et al. (1994) are perfectly correct in saying that most people (Materialists and Postmaterialists alike) would like to have all'of these goals—and that, if they were unconstrained, they would give positive ratings to nearly all of them. The ratings results that they obtain from Australia and Germany, respectively, demonstrate this point. This is, indeed, one aspect of reality. Hence, the ratings approach produces a first principal component on which all 12 goals have positive polarity.
But this does not mean that there is no such thing as Materialist-Postmaterialist priorities. Quite the contrary, the data collected by Bean and Papadakis (1994) demonstrate clearly that these very same respondents are perfectly capable of differentiating between Materialist and Postmaterialist priorities, and they do so when the question is posed in a way that measures priorities. Priorities tap a different aspect of reality from that measured by ratings; and priorities are important because reality often constrains individuals to make choices.
Priorities are not an artifact of survey research. They exist inside people's minds and can be measured reliably—and they emerge even when (as with Bean and Papadakis) the investigators have a skeptical attitude toward them. Indeed, their data replicate the pattern that was predicted by the theory of Materialist/Postmaterialist value change with a remarkable fidelity. This theory was developed 25 years ago, by an American working in Western Europe. Over the past two decades, literally scores of surveys in dozens of countries have discovered an almost identical configuration, in which one segment of the population (which tends to be the older and the economically less secure strata) gives top priority to the items that were designed to tap Materialist priorities; while another segment (the younger, more prosperous and better educated) give priority to a set of items that were designed to tap Postmaterialist priorities. These findings, from 43 societies not including Australia, are presented below. This latest set of findings, from a survey in Australia, indicates that a virtually identical configuration of societal priorities emerges in yet another country on another continent.
The study by Bean and Papadakis (1994) adds a well-executed methodological study to the picture, and it provides convincing evidence that the format in which the questions are posed is crucial. The authors conclude that the rating method is a more appropriate way to measure Materialist/Postmaterialist values than the ranking method, since it "allows for a more flexible and realistic account of the choices made by most social actors." The authors have demonstrated that, free from any constraint, people will give high ratings to both Materialist and Postmaterialist goals. They interpret this to mean that the rating format taps reality more faithfully than the ranking format.
But it depends on which aspect of reality one wishes to understand. If one is asking the question, Do people want to attain all of these goals, or only some of them? then the rating format is appropriate, and the answer is clear: most
122
CHAPTER   4
people want all of them. But if one is interested in the question, What priorities do people act on, when forced to choose between two desirable goals? then the ranking method is appropriate.
Comments on Bean and Papadakis's study by both Inglehart (1994) and Hellevik (1994) appeared with the original article. Both comments noted, independently, that the results from the ratings approach gave strong indications of response set. Hellevik commented,
Bean and Papadakis see the correlated indexes of Materialism and Postmaterialism as a substantively interesting result of allowing a choice of "both-and" instead of forcing the respondent to make priorities. In my view they overlook methodological problems of the rating method, problems which actually may explain their results. Through my work with national and comparative analyses of cultural indicators, using mainly scales, we have reached exactly the opposite conclusion: ranking is preferable to rating. The reasons for this are both substantive and methodological. The two measurement methods give information on separate aspects of value preferences which are both of interest. Rating indicates absolute levels of support, while ranking indicates the priorities among values—even if they have similar levels of support. (Hellevik, 1994: 293)
In their reply to these critiques, Bean and Papadakis (1994) reported that they had tried to correct for response set by using a standard method—anchoring the responses around each respondent's mean score across the set of items and analyzing the extent to which a given response was higher or lower than this mean. When they did so, they found, to their surprise, that their result mirrors the result based on rankings, with the Materialist items at one pole, and the Postmaterialist items at the opposite pole (Bean and Papadakis, 1994: 297). They took this to mean that their correction for response set had failed. There is no reason to believe that it had: a more straightforward interpretation is that response set is, indeed, responsible for the pattern they get using ratings—and that when it is corrected, one obtains the Materialist/Postmaterial-ist dimension that these items were designed to tap.
What Is the Right Level of Analysis: One Dimension? Or Two, Four, or N Dimensions?
The Materialist/Postmaterialist concept is unidimensional by definition and operates at a relatively high level of generality. This level of analysis has proven productive. It is perfectly possible, however, either to ascend to higher levels of generality—examining the degree to which Materialist/Postmaterialist values are themselves part of a still broader configuration of values (as we did in chapter 3)—or to descend to lower levels of generality, examining its component subclusters, as Herz (1979), Flanagan (1982, 1987), and Van Deth (1984) have done. Herz (1979) used multidimensional scaling to obtain a two-dimensional
E A S U R I N G    VALUES
123
structure from the Materialist-Postmaterialist rankings. But this two-dimensional array revealed a clear watershed, with all of the Materialist items on one side and all of the Postmaterialist items on the other side (see Inglehart, 1990:143).
Van Deth (1986) went even farther: he used factor analysis, smallest space analysis, multidimensional scaling, and unfolding techniques to explore the structure of the standard 12-item battery Materialist/Postmaterialist values battery based on rankings. He found that, although the detailed placement of individual items varied, these methods all produced basically the same result: here, too, the most prominent feature was a clear watershed, with all of the Materialist items at one side of the graph and all of the Postmaterialist items at the other side. The results indicated that the use of an additive index, based on the distinction between Materialist and Postmaterialist items, was justified.
As all of these analyses demonstrate, the Postmaterialist dimension can, indeed, be broken down into its constituent subclusters. But, at the same time, they confirm the robustness of the broader Materialist-Postmaterialist dimension: if one measures people's priorities, then no matter how you slice the data, one comes up with an overall pattern on which the Materialist items are at one side, and the Postmaterialist items are on the opposite side.
Klages (1988, 1992) prefers to deal with two dimensions that he describes as "Duty/Acceptance" and "Self Realization." His analysis does not deal with a change in priorities, but with high or low levels of support for two more or less unrelated dimensions. He is free to do so, and he may well come up with interesting findings concerning changes in levels of support for these goals: it is not an either/or choice.
Varimax rotation will usually do what it was designed to do—break a dimension down into its components; but it is absurd to interpret this fact as meaning that the dimension in question "really" has two or more subcomponents, and that, consequently, one is obliged to operate at the lower level of generality. The reality is not only that the battery used to measure Materialist/ Postmaterialist values has subclusters that were designed to tap economic and physical security, respectively, but that each of these subclusters can also be broken down further, until one reaches the individual items that make them up. It is perfectly legitimate to operate at the lowest possible level of generality, examining attitudes toward "The fight against crime," for example, or any other individual item that interests one. Similarly, it is perfectly possible to operate at Flanagan's slightly higher level of generalization, or at the level of the Materialist/Postmaterialist dimension, or (since Materialist/Postmaterialist priorities prove to be correlated with a broad range of other values) to examine whether these values themselves fit into a still broader configuration such as "Survival Values" versus "Well-being Values." It depends on the theoretical purpose one has in mind.
To date, dimensional analysis of data from more than 40 countries around the world has demonstrated that, underlying the value systems of many different peoples, a Materialist/Postmaterialist dimension does exist. Further-
124
CHAPTER    4
more, as the Materialist/Postmaterialist value change theory implies, and as we will see in the next chapter, the values measured by this dimension are linked with economic security, both at the microlevel—with individuals from relatively prosperous families being more likely to be Postmaterialists—and at the macrolevel—with Postmaterialist values being far more widespread in rich countries than in poor ones. In addition, as the theory implies, these values show evidence of intergenerational change, with Postmaterialist values being far more widespread among the young than among the old.
A number of critics implicitly assume that one must either focus on levels of support for given values, as measured by ratings; or value priorities, which are best measured by rankings. They overlook the fact that value priorities and levels of support reflect two distinct aspects of reality. One is free to measure either. A good deal of empirical evidence indicates that the most important intergenerational value change now taking place involves a change in priorities: this does not rule out the possibility that important changes may also be occurring in levels of support for given goals. For value priorities and relative levels of support for given goals are two distinct aspects of reality. Both are worth studying. It is not an either/or choice.
The Postmaterialist Dimension in Global Perspective
The 1990 World Values Survey measured Materialist/Postmaterialist priorities in 43 societies, using the ranking method. As we saw above in figure 4.1, the results from 16 Western democracies show remarkable cross-national similarity. In every case, the same five items clustered near one pole of the first principal component. These items deal with such goals as freedom of expression and having more say on the job or in political life. All five of these items had been designed to tap Postmaterialist priorities. Six items—again the same six in every country—clustered near the opposite pole. These items emphasize economic and physical security. All six were designed to tap Materialist priorities. The remaining item fell near the midpoint.
Exactly the same pattern holds true in 16 Western European and North American societies—including several that had not previously been surveyed. People respond to the five Postmaterialist items (shown in italics) as if they tapped a similar underlying value, placing them in one cluster. People also respond to the six Materialist items (shown in boldface type) as if they tapped something very different. The one remaining item is ambivalent. The detailed results for each of the 43 societies included in the 1990 World Values Survey are reported in Abramson and Inglehart (1995).
There is strong evidence that the Materialist/Postmaterialist battery taps a cross-nationally comparable phenomenon throughout the Western world. Is it merely a Western phenomenon, however, or does it, as the theory implies, emerge in any society that is undergoing an intergenerational shift toward higher levels of prosperity and security?
MEASURING    VALUES
125
Table 4.4
The Materialist/Postmaterialist Dimension in Four Latin American Countries (Loadings on First Principal Component in Factor Analysis)
More Say in Government                                       .67
More Say on Job                                                   .60
Less Impersonal Society                                        .52
Freedom of Speech                                                .36
Ideas Count More than Money                               .34
Economic Growth                                             —.15
More Beautiful Cities                                         -.27
Strong Defense Forces                                      —.27
Maintain Stable Economy                                —.31
Fight Rising Prices                                           —.37
Fight against Crime                                          —.49
Maintain Order                                                —.56
Note: Items with Materialist polarity are in boldface type; items with Postmaterialist polarity are in italics.
Source: 1990-91 World Values Survey data from Mexico, Argentina, Brazil, and Chile.
Do non-Western publics see things from a completely different perspective? Apparently not. Four Latin American societies were included in the 1990-91 World Values Survey, and the same basic structure applies (see table 4.4). There are some variations from the pattern found in advanced industrial societies: already known to be a deviant item, "More beautiful cities" shows negative polarity here, and "Economic growth" is only very weakly linked with the Materialist cluster. But all five of the Postmaterialist items show positive polarity and fall into the predicted cluster.
More strikingly still, the pattern that emerges in Japan and South Korea is practically identical to the pattern we find in Western Europe and Latin America. Although these two East Asian societies started with profoundly different cultural traditions from those of the West, they have both become advanced industrial societies—and their publics respond in a fashion that is almost indistinguishable from that of Western respondents. As table 4.5 demonstrates, the five Postmaterialist items cluster at one pole, and the six Materialist items cluster at the opposite pole. Moreover, the item concerning "More beautiful cities and countryside," which gave ambiguous results in the West, shows a clear Postmaterialist polarity in both Japan and South Korea. The two East Asian countries not only conform to theoretical expectations, they actually show a slightly better fit than do most Western countries. This is a very interesting finding. It indicates that the emergence of a polarization between Materialist goals and Postmaterialist goals is not a uniquely Western phenomenon. It is a phenomenon of advanced industrial society that emerges with high levels of
126
CHAPTER    4
Table 4.5
The Materialist/Postmaterialist Dimension in East Asia (Loadings on First Principal Component in Factor Analysis)
Less Impersonal Society                                              .70
More Say on Job                                                          .60
More Say in Government                                            .50
More Beautiful Cities                                                  .38
Freedom of Speech                                                      .45
Ideas Count More than Money                                   .24
Fight against Crime                                               —.30
Fight Rising Prices                                                 — .31
Strong Defense Forces                                           — .45
Maintain Order                                                      — .50
Maintain Stable Economy                                    —.61
Economic Growth                                                   —.62
Note: Items with Materialist polarity are in boldface type; items with Postmaterialist polarity are in italics.
Source: Pooled 1990-91 World Values Survey data from Japan and South Korea.
economic development—even among societies that started with very different cultural heritages.
There are societies in which the polarization between Materialist and Post-materialist values is less distinct, but the division is not between Western and non-Western cultures. The countries that deviate most from the Materialisť Postmaterialist configuration are the ex-socialist countries, even those that have historically had close ties with the West.
The 1990-91 World Values Survey includes samples from 11 societies of the former Soviet bloc, as well as China (see table 4.6). In 11 of these 12 countries the five Postmaterialist items cluster together in one group. Poland is the sole exception, and even there, four of the five Postmaterialist items cluster together. Thus, even in the communist and ex-communist societies, we find a clear and consistent Postmaterialist cluster. But there is a tendency for emphasis on "economic growth" to fall into the Postmaterialist cluster. This is a striking deviation from what we find everywhere else, and it would be surprising if we found it in societies with market economies. But within the societies that have had state-run economies, it is understandable. "Economic growth" (like everything connected with the economy) has quite different connotations in a state-socialist society from what it has in a market economy.
In 1990 the socialist societies had authoritarian political systems and state-run economies that had become stagnant to the point of paralysis. The existing system of state controls had widely come to be seen as incompatible with economic growth. Even in China, which continued to experience rapid growth,
measuring  values                                                                                          127
Table 4.6
The Materialist/Postmaterialist Dimension in 11 Ex-socialist or Socialist Countries (Loadings on First Principal Component in Factor Analysis)
More Say in Government                                            .73
Less Impersonal Society                                             .46
More Say on Job                                                          .44
Freedom of Speech                                                      .27
Economic Growth                                                     .18
Ideas Count More than Money                                  Al
Maintain Stable Economy                                    -.02
Strong Defense Forces                                           —.37
More Beautiful Cities                                              -.38
Fight Rising Prices                                                 -.43
Maintain Order                                                      — .50
Fight against Crime                                               — .57
Note: Items with Materialist polarity in Western countries and East Asia are in boldface type; items with Postmaterialist polarity in Western countries and East Asia are in italics.
Source: Pooled 1990-91 World Values Survey data from Belarus, Bulgaria, China, Czechoslovakia, East Germany, Estonia, Hungary, Latvia, Lithuania, Poland, Russia, and greater Moscow.
this growth was taking place entirely in the private sector. State-run industries were stagnant. In this context, as of 1990, economic growth had come to be seen as something that could only be attained by breaking free from a massive and sclerotic state bureaucracy and turning the economy over to individual initiative. In contrast with the West, where an emphasis on economic growth was linked with loyalty to the established order, in the state-socialist world many viewed economic growth as a goal that could only come through radical socioeconomic change. Moreover, these changes were closely linked with the Postmaterialist emphasis on individual autonomy: they required the liberation of the individual from state authority. Hence we find the apparent paradox that, in many of the former state-socialist societies and in China, respondents who gave high priority to economic growth were the same people who emphasized "giving the people more say in government," "more say on the job," "freedom of speech," and "a less impersonal, more humane society."
Further analyses support the conclusion that in these societies Postmaterialist values are linked with support for reduced government involvement in the economy. In the 1990-91 World Values Survey, respondents were asked to place themselves on a 10-point scale measuring their attitudes toward governmental versus private ownership of business and industry. In the non-socialist countries, people with Postmaterialist values tend to be slightly more favorable to the idea of a state-run economy than are those with Materialist values.
128
CHAPTER    4
This is not a strong relationship, but it reflects the traditional ideological heritage of the Left.
The situation is dramatically different in Eastern Europe, the societies of the former Soviet Union, and the People's Republic of China. In all 12 of these societies, people with Postmaterialist values are less favorable to a state-run economy than are the Materialists. The polarity of "progressive" values has reversed. Throughout the ex-socialist societies, people who place themselves on the Left and those with Postmaterialist values reject the idea of a state-run economy.
There are also cross-national differences in the polarity of the goal of "trying to make our cities and countryside more beautiful." This item was designed to tap aesthetic values. But analyses of Western societies showed that it had a surprisingly strong linkage with safety needs.
The "beautiful cities" goal has differing associations in different societies in the World Values survey. The variations follow a consistent pattern. When the 12-item measure was first developed in 1973, the "beautiful cities" item was more or less neutral, with a slight Postmaterialist polarity (Inglehart, 1977). In the 1990 surveys, this item has a slightly stronger Postmaterialist polarity than it had previously, in most Western societies. And, as we noted, the goal has an even clearer Postmaterialist polarity in Japan and South Korea. In the four Latin American countries, on the other hand, the goal of "more beautiful cities" clearly emerges as a Materialist value; and it is even more clearly a Materialist goal in the societies of the former Soviet bloc.
In the West this goal, originally designed to tap aesthetic values, now appears to be performing its original mission, though still only weakly. The fact that this item falls into the Materialist cluster in Eastern Europe seems to reflect severe environmental deterioration in these societies (see Inglehart and Abramson, 1992b). In these societies, environmental pollution has become a massive and life-threatening problem, one that is far more severe than in the West. In the former state socialist countries, pollution is not perceived primarily as an aesthetic problem, but as one that is directly life-threatening. This perception is far from groundless. The death rates in some regions of Russia, the former East Germany, Poland, and the former Czechoslovakia are shockingly high, for reasons attributable to massive industrial pollution. In the West the goal of "more beautiful cities" has an ambivalent meaning, tapping aesthetic needs to some extent, but also having secondary connotations of urban disorder and crime. In Japan and Korea, where urban crime rates are much lower than in most Western countries, this item unequivocally taps Postmaterialist concerns. In the former Soviet bloc countries, with life-threatening levels of pollution, it tends to tap basic Materialist concerns for the effects of industrial pollution on human survival.
In the former communist world, economic and physical security cannot be taken for granted nearly as much as in the rich industrial societies of the West and East Asia. Hence, as one might expect, the Materialist-Postmaterialist dimension is less consistent, and less clearly crystallized in the ex-socialist so-
MEASURING    VALUES
Table 4.7
The Materialist/Postmaterialist Dimension in India (Loadings on Second* Principal Component in Factor Analysis)
More Say in Government                                            .61
Less Impersonal Society                                             .58
More Say on Job                                                          .43
Ideas Count More than Money                                   .38
Freedom of Speech                                                      .06
Strong Defense Forces                                           — .01
Economic Growth                                                  — .08
Maintain Order                                                      —.19
Maintain Stable Economy                                    —.34
More Beautiful Cities                                              - .35
Fight Rising Prices                                                 — .46
Fight against Crime                                               — .49
Note: Items with Materialist polarity are in boldface type; items with Postmaterialist polarity are in italics.
Source: Pooled 1990-91 World Values Survey data from India.
*In India, the first principal component reflects a polarization between emphasis on economic priorities and emphasis on maintaining military and domestic order; all of the Post-materialist items are neutral on this dimension.
cieties than it is in the West and East Asia. Although a recognizable Materialist/Postmaterialist dimension does emerge in the former Soviet bloc, it accounts for less variance than it does in richer advanced industrial societies. In three of these societies (Lithuania, Czechoslovakia, and Hungary) this dimension emerges only as the second principal component in the factor analysis.
The Postmaterialist phenomenon is only marginally present in such low-income societies as India and Nigeria, both of which had per capita incomes of approximately $300 in 1990. As the theory implies, there are very few Post-materialists in these countries, and mass values do not crystallize along this dimension as strongly as they do in richer societies (see Abramson and Inglehart, 1995, for country-by-country results). A recognizable Materialist/Postmaterialist dimension emerges in India, but it accounts for less variance than in almost any other country and is the second principal component instead of the first (see table 4.7).
Although interesting cross-national differences in the structure of these values exist, the evidence suggests that the core meaning of Materialism/Post-materialism is similar across this wide range of societies. The quest for economic security is much more politicized in societies with state-run economies than in societies with market economies, which gives a distinctive meaning to items that refer to the economy. Apart from this, however, the Materialist goals show consistent results, and the five Postmaterialist items behave in a similar
130
CHAPTER    4
way in all types of societies, East and West, as well as North and South. The results indicate that an additive index, based on summing up the five Postmaterialist items (producing an index having scores ranging from 0 to 5) can be used in virtually any of the 43 societies included in the World Values survey. Consequently, we can construct a cross-nationally comparable index of Mate-rialist/Postmaterialist values, based on the nine items that do have globally consistent polarity (assigning neutral polarity to the two items referring to the economy, and the item referring to "more beautiful cities")- This index is used throughout this book, except when only the original four-item index is available.
On the whole, the cross-national similarities underlying mass responses to the Materialist/Postmaterialist values items are far more striking than the cross-national differences.
CHAPTER     5
The Shift toward Postmaterialist Values, 1970-1994
Introduction
The basic values of publics throughout advanced industrial society have been undergoing a gradual intergenerational shift during the past several decades. Although different countries have shifted at different rates, economic and technological changes have had a broadly similar impact across these societies. As Postmodernization theory implies, the process applies to advanced industrial society in general.
In 1970 we hypothesized that the value priorities of Western publics were shifting from Materialist values toward Postmaterialist values—from giving top priority to physical sustenance and safety, toward heavier emphasis on belonging, self-expression, and the quality of life (Inglehart, 1971). The predicted intergenerational value shift could not be demonstrated until many years had passed; and whether or not it was occurring has been hotly disputed (Böltken and Jagodzinski, 1985; Thomassen and Van Deth, 1989; Trump, 1991; Clarke and Dutt, 1991). Only in recent years has a sufficiently long time series become available to test the prediction reliably. This chapter examines cross-national survey data over a 24-year period. The results show a clear and statistically significant trend toward Postmaterialist values in almost all of the societies for which we now have detailed time series measurements over this period. These values also show short-term fluctuations linked with changing rates of inflation and unemployment, as the value change thesis implies; but the long-term trend seems to result mainly from intergenerational replacement.
Evidence from the 43 societies surveyed in the World Values surveys enables us to test the value change hypothesis on a broader basis than has ever before been possible. The thesis implies that we should observe two quite different findings, both of which are important: societies with high levels of economic development should have relatively high levels of Postmaterialist values, and societies that have experienced relatively high rates of economic growth should show relatively large differences between the values of younger and older generations. As we will see, the evidence supports both hypotheses. It appears that this value shift occurs in any society that has experienced sufficient economic development in recent decades so that the preadult experiences of younger birth cohorts have been significantly more secure than those of older cohorts. Large intergenerational differences are found in societies that have experienced rapid growth in GNP per capita and are negligible in societies that have had little or no growth. And these value differences are enduring characteristics of given birth cohorts. Accordingly, as intergenerational
CHAPTER     6
Economic Development, Political Culture, and Democracy: Bringing the People Back In
A generation ago, Lipset (1959), Rostow (1961), Dahl (1971), and others argued that economic development leads to democracy. This claim was disputed by dependency school writers, who argued that development was more likely to lead to bureaucratic authoritarianism than to democracy; and more recently, Arat (1988) and Gonick and Rosh (1988) claimed to have disproven Lipset's thesis on the basis of empirical analyses. Nevertheless, the evidence indicates that development is indeed conducive to democracy (Bollen, 1979,1980,1993; Bollen and Jackman, 1985; Brunk, Caldeira, and Lewis-Beck, 1987).
Figure 6.1 shows one piece of the evidence. It reveals that as of 1987, out of the 42 countries with per capita incomes under $500 only one (India) was democratic. Among countries with incomes from $500 to $1,000, only four out of 15 were democracies. But among countries with incomes over $6,000, 20 of the 26 were democratic (the exceptions being then East Germany, Czechoslovakia, Kuwait, Saudi Arabia, United Arab Emirates, and Singapore). Although there is no one-to-one relationship between economic development and democratization, rich countries are much likelier to have democratic institutions than are poor countries.
This relationship between democracy and economic development is not merely cross-sectional—it helps predict which countries are most likely to become democratic. Thus, during the avalanche of democratization that took place during the four years after 1987, most of the countries that began the transition to democracy were drawn from the upper middle income group, with per capita incomes from $1,000 to $6,000. This includes such countries as Chile, Nicaragua, Turkey, South Korea, and most of Eastern Europe (including both Czechoslovakia and East Germany). By 1992, democracy remained very rare in low-income countries, but a majority of the countries in the upper middle income group had governments that came to power through free elections.
Burkhart and Lewis-Beck (1994) have provided the latest and most conclusive demonstration of the fact that economic development is conducive to democracy, using more reliable time series data and more rigorous methodology than that of Arat (1988) or Gonick and Rosh (1988), and finding (1) that economic development is conducive to democracy, and (2) but democracy is not conducive to economic development. On the latter point they confirm Helii well's (1994) finding that, despite the spectacular success of the authoritarian model of development in East Asia, economic development is about as likely to take place in democratic as in authoritarian regimes.
These are important findings, but they leave a major question unanswered:
BRINGING    THE   PEOPLE    BACK   IN
161
Under-$500        $500-$1000       $1000-$6000 Annual GNP/Capita in 1987
Over $6000
Figure 6.1. Economic development and the transition to democracy. Source: GNP per capita from World Bank, World Development Report, 1989, pp. 164-65. "Democracies" consist of countries coded as "free" in Raymond D. Gastil (ed.), Freedom in the World, 1986-87 (New York: Freedom House, 1987): 30-34. This figure includes only those countries for which data are available for both variables.
Why does economic development lead to democracy? Is the linkage between development and democracy due to wealth per se? Apparently not: if democracy automatically resulted from simply becoming wealthy, then Kuwait and Libya would be model democracies. It is important to bear this point in mind: wealth alone does not automatically produce democracy. It seems clear that additional steps are involved. This chapter argues that economic development is conducive to democracy provided that it brings certain changes in culture and social structure.
Lerner (1958), Deutsch (1964), and others had argued that modern economic development brings social mobilization, facilitating mass participation in politics, which helps prepare the way for democracy. We believe that they were right, but this is only part of the story. Building on the work of Almond and Verba (1963), Inglehart (1988, 1990), Diamond, Linz, and Lipset (1990), and Putnam (1993), we will demonstrate that economic development is conducive to democracy not only because it mobilizes mass publics, but also because it tends to give rise to supportive cultural orientations. Our analysis utilizes empirical measures of culture from the 1990-93 World Values surveys carried out in 41 societies around the world representing 70 percent of the world's population, to demonstrate that both social structure and political culture play important roles in the emergence and survival of democracy.
The idea that political culture was linked with democracy had great impact following the publication of The Civic Culture (Almond and Verba, 1963), but went out of fashion during the 1970s for a variety of reasons. The political cul-
162
CHAPTER    6
ture approach raised an important empirical question: whether given societies had political cultures that were relatively conducive to democracy. Some critics alleged that this approach was "elitist" in finding that some cultures were more conducive to democracy than others: any right-minded theory should hold that all societies are equally likely to be democratic. The problem is that by tailoring one's theory to fit a given ideology, one may come up with a theory that does not fit reality, in which case, one's predictions will eventually go wrong, and the theory will provide misleading guidance to those who are trying to cope with democratization in the real world.
By the 1980s, though the concept of political culture was still unfashionable in American academic circles, observers in other countries, from Latin America to Eastern Europe to East Asia, were coming to the conclusion that cultural factors played an important role in the problems they were encountering with democratization. Thus Mikhail Gorbachev observed, "We are now, as it were, going through the school of democracy afresh. We are learning. Our political culture is still inadequate. Our standard of debate is inadequate; our ability to respect the point of view of even our friends and comrades—even that is inadequate" (Gorbachev, cited in Brzezinski, 1989: 44). Even in Latin America, where the dependency perspective had been extremely influential, cultural factors are now being accorded a key role in democratization. Thus in 1990, a conference of leading politicians and intellectuals from throughout Latin America concluded
Democracy and sustainable development will depend in large measure on the ability of individual societies to modernize from within. . . . Changes in the practical exercise of power and the reorganization of systems of production give rise to changes not only in political, social and economic institutions, but also in culture and in the behavior of individuals molded by that culture. The other vital dimension of the challenge facing societies in the early stages of democratization is the forging of a democratic political culture. (Declaration of Montevideo, 1990, cited in Albala-Bertrand, 1992: 156-57)
Cultural factors have been omitted from most empirical analyses of democracy—partly because, until now, we have not had reliable measures of them from more than a handful of countries. When cultural factors are taken into account, as in Inglehart's (1990) and Putnam's (1993) work, they seem to play an important role.
We will briefly describe which factors are important to democracy and why, and discuss the nature of our dependent variable, democracy, in presenting the theory underlying this analysis. We claim that economic development leads to two types of changes that are conducive to democracy: it gives rise to social structural changes that mobilize mass participation; and cultural changes that help stabilize democracy.
Structural changes. Industrialization tends to transform a society's social structure, bringing urbanization, mass education, occupational specialization, growing organizational networks, greater income equality, and a variety of as-
BRINGING    THE    PEOPLE    BACK   IN
163
sociated developments that mobilize mass participation in politics. Two aspects of this "Modernization" syndrome are particularly relevant to democracy:
1.  Rising educational levels, which produce a more articulate public that is better equipped to organize and communicate, and
2.  Rising occupational specialization, which first shifts the workforce into the secondary sector and then into the tertiary sector. These changes produce a more autonomous workforce, accustomed to thinking for themselves on the job and having specialized skills that enhance their bargaining power against elites.
Cultural changes. Economic development is also conducive to cultural changes that help stabilize democracy. We find two particularly central factors:
1.  A culture of trust. In authoritarian regimes, the usual way to handle opposition is to imprison or execute its leaders. A crucial element in the rise of democracy is the emergence of the norm of the "loyal opposition": instead of being viewed as traitors who are conspiring to overthrow the government, the opposition is trusted to play by the rules of the democratic game. This means that if the opposition wins an election, the governing elite will turn power over to them, trusting that they will not be imprisoned or dispossessed, and that (after a given time) the new elite will hold elections in which they can freely compete for power.
2.  Mass legitimacy. Legitimacy, or diffuse mass support, can help sustain democratic institutions through difficult times. It is an asset to any regime, but it is crucial to democracies. Democratic institutions can be imposed by elites or even by foreign conquest—but whether they survive depends on whether they take root among the public. With democratization, the public becomes a crucial political factor.
Positive outputs from a political system can generate mass support for the political incumbents. In the short term, this support is based on calculations concerning "What have you done for me lately?" But if a given regime's outputs are seen as positive over a long time, the regime may develop "diffuse support" (Easton, 1963)—the generalized perception that the political system is inherently good, quite apart from its current outputs. This type of support can endure even through difficult times. As we will demonstrate, a sense of subjective well-being among the public of a given society is an excellent indicator of whether or not that regime possesses legitimacy—indeed, it is a better indicator than responses to direct questions about how strongly one supports democratic political institutions.
Three Aspects of Democracy: Stability, Level at a Given Time, and Short-Term Shifts
Democracy is a multidimensional phenomenon. But to date, most empirical analyses have focused on a single aspect of democracy as the dependent variable. Thus, Inglehart (1990) analyzed the stability of democracy—operationally, the number of years that democratic institutions had functioned con-
164
CHAPTER    6
tinuously in a given society. But this is just one aspect of democracy; many other analyses have focused on levels of democracy at a given time. This chapter examines the linkages between economic development, sociocultural change, and three different aspects of democracy:
1.  The long-term stability of democracy,
2.  The level of democracy at given points in time, and
3.  Short-term changes in levels of democracy: for example, the causes of the sudden surge of democracy that followed the collapse of socialist regimes in Eastern Europe and the former Soviet Union in 1989-91.
Careful specification of the aspect of democracy to be examined is crucial, since different causal factors are important with each aspect of democratization. These factors may even reverse their polarity in connection with different aspects of democratization. Thus, though there is strong evidence that prosperity is conducive to democracy, the fall of authoritarian regimes may be precipitated by economic collapse. Political culture tends to be most important in consolidating democracy and enabling it to endure through difficult times. Situation-specific factors (such as the death of Franco) are often the immediate cause of the transition to democracy. But once democratic institutions are in place, their long-term survival depends on the presence or absence of supportive orientations among the citizens. The growing importance of mass preferences is inherent in the very nature of democracy. If democratic institutions do not attain enough deep-rooted mass support to weather difficult times, the citizens can simply vote democracy out of existence: they did so in Weimar; they would have done so in Algeria if democracy had not been suspended there; and they may do so in some of the Soviet successor states.
Long-Term Stability of Democracy
Mass political culture's most crucial role concerns the long-term stability of democracy: political culture stabilizes democracy by providing an enduring base of mass support.
Democratic institutions can be implanted by a handful of elites or even imposed by foreign conquest, as they were in Germany and Austria at the end of World War I, and in Germany, Japan, and elsewhere at the end of World War II. Democracy can be imposed from above or from outside, but whether or not it survives through good times and bad depends on whether its institutions have built up deep-rooted cultural attachments among the citizens. Various writers have stressed the importance of this factor. Weber emphasized the importance of legitimacy; Easton spoke of "diffuse support"; Almond and Verba discussed the "Civic Culture"; Putnam (1993) showed how "Civic Orientations" contributed to the effectiveness of democracy in Italy; and in an analysis based on data from 24 societies surveyed in the 1981 World Values Survey, Inglehart (1990) demonstrated that interpersonal trust and subjective well-being were closely linked with the long-term survival of democratic institutions.
BRINGING    THE    PEOPLE    BACK   IN
165
The appropriateness of analyzing the stability of democracy has been debated. Lipset (1959), Müller (1988), and others measured democracy in ways that included stability along with measures of political rights and civil liberties. This practice was criticized by Jackman (1973), Bollen (1980), and Bollen and Jackman (1985). "The fusion of stability and democracy measures," Bollen and Jackman argue, "makes it impossible to interpret observed associations of 'democratic stability' with other variables, because it is never clear whether degree of stability or degree of democracy is the operative factor at work" (Bollen and Jackman, 1985: 612). For example, in answering the question "Does democracy lead to greater income equality?" this procedure can be confusing—possibly leading one to conclude that democracy leads to income equality, when it is actually due to stability. Bollen and Jackman's writings make a strong case for not confusing the stability of democracy with the extent of democracy—which, we emphatically agree, are two distinct things. But they do not constitute a blanket injunction against studying the stability of democratic institutions—which is an important variable in itself.
Our analysis does not confuse these two aspects of democracy: we will carry out analyses in which the dependent variable is the extent or level of democracy. But we will also analyze the stability of democracy. This analysis does not use democratic stability as a proxy for degree of democracy; instead, it is explicitly designed to focus on the factors that enable democratic institutions to survive over time, addressing the question "What factors enable given societies to remain above the threshold at which the top political leaders are chosen by free and competitive elections?"1
Levels of democracy at a given point in time and the stability of democracy in given societies are both significant. Whether or not democratic institutions survive through good times and bad depends on whether they have built up deep-rooted cultural attachments among the citizens. Weimar Germany had a constitution that was, on paper, as democratic as that of any society on earth; the level was high. But democratic norms did not take root, and these institutions proved unstable. In 1933 the German people turned power over to Adolf Hitler. Similarly, although Bulgaria and Slovenia today are coded by Freedom House as having virtually the same levels of democracy as Britain and Swe-
1 Our measure of the stability of democratic institutions is the number of continuous years from 1920 to 1990 during which top leadership was chosen by free and competitive elections and ranges from 0 to 70. Accordingly, it reflects a wide range of degrees of stability (without attempting to simultaneously capture the relative extent of democracy across the various societies). Coding it requires that one determine whether top leadership was selected by free and competitive elections. This is a judgment call, which may be uncertain in marginal cases, but in most instances there is almost universal agreement on whether or not given elections were free and competitive. But any system for measuring either the extent or the stability of democracy requires human judgments about whether or not given societies fall above given thresholds. Competent coders usually can decide whether a given society did, or did not, hold genuinely competitive elections—and probably can do so with even greater reliability than they would attain in judging whether a given society falls at level "3" or level "4" on the Freedom House ratings of political rights or political liberties—which many analysts (ourselves included) use.
166
CHAPTER    6
den, democratic institutions have functioned much longer in Britain and Sweden than in Bulgaria or Slovenia—a fact that no prudent analyst would ignore in assessing the prospects for survival of democratic institutions in the respective societies. Stability and levels of democracy are two different things, and both are important.
Levels of democracy have risen repeatedly during the nineteenth and twentieth centuries. When modern democracy first emerged in Great Britain and the United States, suffrage was limited to the middle class. In subsequent waves, it was extended to the lower middle class, the working class and to exslaves; in the 1920s, it was extended to women; and in the 1970s, to 18-20,-year-olds. In subsequent years, mass political participation has continued to become more active and more issue-specific, as increasingly educated electorates have extended their repertory of techniques designed to influence elite decision making.
Jackman and Miller (1996) claim that this expansion of mass participation makes it impossible to measure democratic stability: one cannot start counting the number of years during which a given society has been "democratic" until it has become completely democratic by today's standards. In fact, this constitutes a problem only if one adheres to a unidimensional concept of democracy. Dahl (1971) distinguishes between two key aspects of democracy, contestation and inclusion; and he argues that democracy is more likely to survive over time if contestation precedes broad mass inclusion. Thus, in British history, elite contestation began with the Magna Charta, which forced the king to share power with the nobility, and which constitutional authorities accord an important role in introducing pluralist norms into the British political culture. Mass democracy began to emerge in the nineteenth century, and democratic norms had become widespread and generally accepted long before the latest major extension of the franchise, to 18-year-olds in the 1970s. It would be absurd to claim that British democracy began only in the 1970s (or even in the 1920s, when women obtained suffrage). These were indeed important stages in the extension of mass inclusion, but genuine contestation (in the form of freely contested elections) existed well before that time. To insist that democracy does not exist until the process of mass inclusion has been completed would be to define democracy as an empty cell: the process is probably not yet complete even now, because levels of democratic participation will almost certainly continue to rise.
Shifts to (and from) Democracy
In contrast with its role in sustaining democracy over the long run, political culture has a very different relationship to short-term changes to and from democracy. Indeed, the same cultural factors that stabilize and sustain democracy can also help stabilize authoritarian regimes. Thus, though high levels of legitimacy and trust are crucial to the survival of democracy, they would not explain short-term shifts toward democracy. Instead, one would expect low levels of legitimacy and trust to be linked with the collapse of authoritarian
BRINGING    THE    PEOPLE    BACK   IN
167
regimes, possibly opening the way for a transition to democracy. Thus, the short-term consequences of cultural factors are very different from their long-term functions. Gradual cultural changes can give rise to conditions that become increasingly favorable to the rise of democratic institutions, but the immediate precipitating factor is likely to be some macroevent such as defeat in war or an intergenerational transfer of power from hard-line leaders to reformist leaders. Accordingly, the literature on transitions to democracy tends to focus on elite-level events rather than on underlying changes in culture or social structure (e.g., O'Donnell, Schmitter, and Whitehead, 1986).
Levels of Democracy at a Given Time Point
Most empirical analyses of the factors conducive to democracy have used the level of democracy at a given point in time as the dependent variable. Although the most crucial function of cultural factors is their role in sustaining democratic institutions over time, they are also linked with the level of democracy found at given points in time. But clearly, the strength of this relationship will vary from one time point to another.
In 1790, there were three democracies in the world. In 1900, there were about a dozen. In 1919, there were about two dozen, and by 1991 there were more than 60. After each wave of democratization, many of the new democracies failed to survive. Thus, the number of democracies declined sharply during the period between the two World Wars, and again in the 1960s and 1970s; and some of today's new democracies will probably not survive.
Each major wave of democratization weakens the correlation between cultural factors and democracy, because a massive surge of democratization tends to bring into the "democratic" category a large new group of societies that rank lower on prodemocratic culture than the long-established democracies: the latter are societies in which democracy has survived for a long time partly because they have high levels of these cultural characteristics. If, by some happy stroke, every nation in the world were suddenly to adopt democratic constitutions, the correlation between culture and democracy would automatically drop to zero—but this would probably be a temporary situation. Unless the new democracies developed such cultural attributes as interpersonal trust and legitimacy, their democratic institutions would be unlikely to survive major economic or political crises; and the processes of cultural change and attrition would eventually bring back a correlation between civic culture and democracy. Thus (as we will demonstrate) the strength of the relationship between polit-ical culture and democracy differs sharply before and immediately after a major wave of democratization. This means that when one uses levels of democracy as a dependent variable, the time point one chooses is crucial.
The number of democracies in the world has been increasing and, in the long run, we think the trend will continue. It will do so because economic development tends to bring changes in social structure and culture that are favorable to democracy. Let us examine these processes in more detail.
168
CHAPTER    g
Changes in Social Structure: Cognitive Mobilization and the Rise of Citizen Intervention
The literature on social mobilization has chronicled how industrialization and urbanization led to mass literacy, the rise of organized labor, mass political parties, and the emergence of universal suffrage (Lerner, 1958; Deutsch, 1964; Inkeles and Smith, 1974). These were profoundly important developments that brought previously parochial masses into political relevance. These processes increased mass political participation, but they did not necessarily bring about democracy. Instead, depending on the social and economic context of the given society, they could either give rise to mass democracy, or to fascism or communism. All three forms of government emphasized mass participation; indeed, both fascism and communism regularly attained higher levels of mass attendance at political rallies and higher voting turnout than liberal democracies ever did. But with fascism and communism, it was almost entirely elite-led participation, designed to mobilize mass support for policies already chosen by the elites—and not participation through which the masses chose between competing elites and alternative elite policies.
Democratic theory emphasizes two central elements: elite competition and mass participation. In the first half of the twentieth century, democracy (unlike fascism and communism) was based on genuine elite competition; but mass participation was still largely orchestrated by elites even in the democracies. Democracy continues to evolve. In advanced industrial society, the process of cognitive mobilization gives rise to more active and more demanding types of mass participation. This makes it increasingly difficult for democracies to limit mass publics to an elite-directed role, and increasingly difficult for authoritarian systems to survive: they face rising mass pressures for liberalization.
The coming of advanced industrial society leads to a syndrome of intergen-erational changes that bring significant further increases in citizen intervention in politics. A long-term rise in educational levels and in mass political skills has characterized all industrial societies. An extension of social mobilization beyond the transformations brought by urbanization and early industrialization, this process has been termed "cognitive mobilization" (Inglehart, 1977). While social mobilization manifested itself in visible changes of location and occupation, cognitive mobilization is based on invisible changes that upgrade individual skills. These changes have momentous political consequences.
Cognitive Mobilization reflects rising levels of education and changes in the nature of work, from simple routine operations to tasks requiring specialized knowledge and autonomous judgment. The publics of advanced industrial societies become accustomed to thinking for themselves in their everyday jobs; at the same time, they become more articulate and skilled at organizing people. The skills they learn through higher education and in their work life make them increasingly skillful political participants.
Democracies existed before the industrial era. But in polities that are too large for face-to-face interaction, political participation was limited to a mi-
RINGING    THE    PEOPLE    BACK    IN
169
nority of the population. Ancient Athens was a democracy by the standards of its time, but one that excluded a large slave population, a large foreign population, and all women. Even as recently as the eighteenth century, democracy in the United States was limited to a minority of the population which excluded Blacks, women, and, in some states, those who fell below certain property-ownership thresholds. By contemporary standards, neither classical Athens nor the early United States would qualify as democracies. Mass mobilization is a prerequisite for the contemporary version of democracy.
Mass political participation develops in two major stages, one based on an older mode of elite-led political participation, and the other on a newer mode linked with cognitive mobilization. The institutions that mobilized mass political participation in the late nineteenth and early twentieth century—labor union, church, and mass political party—were hierarchical organizations in which a small number of leaders or bosses led masses of disciplined troops. These institutions were effective in bringing large numbers of newly enfranchised citizens to the polls in an era when universal compulsory education had just taken root and the average citizen had a low level of political skills. But while these elite-directed organizations could mobilize large numbers, they produced only a relatively low level of participation, rarely going beyond mere voting.
By itself, voting is not necessarily an effective way for citizens to exert their control over national decisions. It can be, and sometimes is, manipulated by elites. The extreme example is the communist people's democracies which regularly attained far higher levels of electoral participation than any liberal democracy—but did so in an institutional framework that kept real decision making entirely in the hands of the elites. Voting can be an effective step toward empowering the citizens, but it is not a very discriminating one. In one-party states, it is nothing more than a way for the ruling elite to elicit mass endorsement. And even when competing parties are present, it may only mean that the citizens get to choose one set of elites or another, and then let them make the actual decisions for the next several years.
A newer elite-challenging mode of participation is emerging that expresses the individual's preferences with far greater precision than the old. It is issue-oriented and based on ad hoc groups rather than on established bureaucratic organizations. It seeks specific policy changes, rather than simply giving a blank check to the elites of a given party. This mode of participation requires relatively high skill levels.
The most readily available indicator of political skills is one's level of formal education. In part, participation levels reflect skill levels, and sheer literacy seems sufficient to produce voting. The citizens of most Western democracies reached this threshold generations ago. But while mere literacy may be sufficient to produce high rates of voting, taking the initiative to seek specific policy changes at the national level seems to require higher education. This is particularly true of the more elite-challenging types of political behavior: as Barnes et al. (1979) demonstrate, high educational levels are closely associated with participation in elite-challenging forms of political action. But the
170
CHAPTER    6
Barnes et al. study goes a step farther and develops measures of political skills; they prove to be an even stronger predictor of unconventional political behavior than is education—and far stronger than social class.
Educational statistics give a good indication of the progress of cognitive mobilization over time, since governments have kept records of the numbers of students enrolled at various levels for many decades. These statistics tell a dramatic story. Early industrial society introduced universal primary education, bringing widespread literacy, and as industrial societies have developed knowledge-based economies, enrollment in higher education has increased enormously.
As a result of the explosive expansion of higher education during the past 50 years, younger cohorts have much higher educational levels than older ones, throughout advanced industrial society. In the United States, for example, only about a third of the cohort born during the decade before 1925 received any secondary or higher education. Among the cohort born from 1966 to 1972, over 90 percent have done so. In Russia, the rise is even steeper, moving from about 10 percent among the oldest cohort, to almost 90 percent among the youngest.
Because its educational attainment is a relatively stable attribute of a given birth cohort, intergenerational population replacement has foreseeable consequences. One can project the educational level of a given population 10 or 20 years into the future with considerable accuracy. And the consequences are significant.
The rise of postindustrial society or information society (Bell, 1973, 1976) leads to a growing potential for citizen participation in politics. Increasingly, not only one's formal education but also one's job experience helps develop politically relevant skills. The assembly-line worker produced material objects, working in a hierarchical system that required (and allowed) very little autonomous judgment. Workers in the service and information sectors deal with people and concepts; operating in an environment where innovation is crucial, they need autonomy for the use of individual judgment. It becomes inherently ineffective to attempt to prescribe innovation from above, in hierarchical fashion. Accustomed to working in less hierarchical decision-structures in their job life, people in the tertiary, or information and service, sectors are relatively likely to have both the skills and the inclination to take part in decision-making in the political realm as well.
Inglehart (1990) presents evidence of a long-term rise in mass skills in coping with politics that is transforming the mass basis of politics in Western industrial societies. Throughout advanced industrial society, publics are becoming more apt to want democratic institutions, and more adept at applying pressures to get them. These changes in mass skills and values are not the only factors that matter. A determined elite can repress public demands for democratization for a long time. But as an industrial society matures, the costs of repression rise: it stultifies initiative, bringing a demoralized, inefficient economy, and a technology that falls behind world standards.
RINGING    THE    PEOPLE    B A C K   IN
111
The new mode of political participation is far more issue-specific than voting is, and more likely to function at the higher thresholds of participation. It is new in the sense that only recently has a large percentage of the population possessed the skills required for this form of participation. And it is new in that it makes the public less dependent on permanent, oligarchic organizations.
Thus, as cognitive mobilization proceeds, the established organizations become progressively less effective. Possessing a wide range of alternative channels of information and input, people rely less and less on permanent organizational networks such as labor unions, churches, and urban political machines. Both union membership rates and church attendance have been falling in most Western countries, and traditional political party ties have also been weakening. This tends to depress voter turnout, which is heavily dependent on elite-directed mobilization, and may require little or no cognitive response to current issues. High rates of voter turnout are a good thing, to be sure. But we should bear in mind that the one-party communist regimes regularly reported voting rates of 98 or 99 percent. Electoral turnout is desirable, but it is not a reliable indicator of citizen input. Although electoral turnout has stagnated, elite-directing types of participation, aimed at influencing specific policy decisions, are becoming more widespread.
The Iron Law of Oligarchy is being weakened. Advanced industrial society brings an increasingly educated and occupationally specialized public. As the workforce shifts from doing routine tasks, toward becoming specialists, doing tasks that require individual judgment and autonomy, they become less amenable to centralized hierarchical control.
Cultural Changes Conducive to Democracy
The spread of democracy reflects not only changes in social structure, but also cultural changes. The study of political culture grew out of the tragic events that led up to World War II. In the aftermath of World War I, democratic regimes were set up in Germany, Italy, Poland, Spain, and many other formerly authoritarian societies. On paper, some of them looked like ideal democracies. But when they encountered the severe economic difficulties of the 1920s and 1930s, democracy failed to survive in many cases. Why did this happen? Great Britain, the United States, and the Nordic countries also experienced severe economic distress during the Great Depression, but democracy survived there; in contrast, democracy gave way to fascist regimes in Germany, Italy, Japan, Spain, Hungary, and elsewhere, preparing the way for the greatest bloodbath the world had ever known.
The classic Civic Culture study (Almond and Verba, 1963) addressed the question "Why did democratic institutions survive in some countries but not in others?" Manifestly, it was not just a question of constitutional engineering. The laws and constitution of the Weimar Republic were as democratic as those of any nation in the world—but they did not take root. An authoritarian out-
172
CHAPTER    6
look remained widespread throughout German society, and when distress and insecurity became severe, the Germans voted Hitler into power in free elections. Facing comparable problems, the British, Americans, and various other peoples were relatively steadfast in their support for democracy. Democracy, apparently, is not just a matter of elite-level arrangements; the basic cultural orientations of the citizens also play a crucial role in its survival. Almond and Verba set out to measure the relevant orientations empirically, to determine whether there really were underlying differences in the political cultures of stable democracies, as compared with those of unstable democracies.
Ideally, to explain the role of cultural factors in the survival of democracy, Almond and Verba would have used data on cultural conditions from the period before the rise of fascism: a cause must precede its effect. But survey research techniques had not yet been developed in that period, and they would have needed a time machine in order to go back and collect such data. However, culture is by definition a relatively stable aspect of a society. If so, one would expect to find significant elements of the cultural differences that contributed to the survival of democracy in Britain and the United States, and to its failure in Germany and Italy, that were still visible in the orientations of the respective mass publics in 1959, when their fieldwork was earned out. Almond and Verba set out to determine whether such differences existed.
Democratic institutions had recently been transplanted to Germany, Italy, and Japan. Would they take root this time, or would they fail again?
This basic question is of far more than academic interest again today, when democratic institutions have recently been installed in scores of formerly authoritarian societies, from Argentina to Russia—and where their fate remains uncertain. Authoritarian forces seem to be making a comeback in a number of the Soviet successor states; and a protofascist party won more votes than any other party in the 1993 Russian parliamentary elections, which gives rise to the chilling question: Will Russia's fate be like that of the Weimar Republic?
The Importance of Societal Trust
Partly, the answer depends on the development of a culture of trust. Interpersonal trust plays a crucial role in democracy. Democratic institutions depend on trust that the opposition will accept the rules of the democratic process. One must view one's political opponents as a loyal opposition who will not imprison or execute you if you surrender political power to them, but can be relied on to govern within the laws, and to surrender power if your side wins the next election.
Banfield (1958) found that Southern Italian society had much lower levels of trust than Northern Italy; this severely hindered the large-scale cooperation between strangers that is essential to both economic development and successful democratic institutions. Almond and Verba (1963) also argued that a sense of interpersonal trust is a prerequisite for effective democracy. They found that the
BRINGING    THE    PEOPLE   BACK   IN
173
publics of Italy and West Germany were characterized by lower levels of interpersonal trust, readiness to participate, and other attitudes conducive to democracy than were the British and American publics. The relative weakness of the "Civic Culture" in Germany and Italy presumably contributed to the failure of democracy in those societies in the period before World War II.
Testing these ideas in a broader cross-national context, Inglehart (1990) found that interpersonal trust and related cultural orientations were strongly linked with both economic development and with stable democracy. He emphasized, however, that culture is a variable, not a constant: though cultural characteristics tend to change slowly, they can and do change. Thus, while Southern Italians were still markedly less trusting than Northern Italians in 1990, and the Italian public still had lower levels of interpersonal trust than the British or Americans, levels of trust had gradually risen in Italy. Starting from an almost incredibly low level in the 1959 Almond and Verba survey, only 8 percent of the Italian public had agreed that "most people can be trusted"; but this figure rose to 27 percent in 1981 and to 30 percent in 1986. Even in 1990 Italy ranked lower than the United States or Britain, but it showed a gradual upward trend.
As the classic literature on political culture implied (but could not demonstrate directly), trust is linked with the survival of democratic institutions. The 1990 World Values data reveal a strong positive correlation between interpersonal trust and the functioning of democratic institutions throughout the world, as figure 6.2 demonstrates. The vertical axis reflects the number of years during which democratic institutions have functioned continuously in a given country. This measure ranges from long-established stable democracies to authoritarian states and societies in which democratic institutions have just been established—and may or may not survive.
The overall pattern in figure 6.2 confirms theoretical expectations that have never before been tested against a global database. Levels of interpersonal trust among mass publics are closely linked with the number of years for which democratic institutions have functioned continuously in those societies, showing a highly significant .72 correlation globally. In most stable democracies, at least 35 percent of the public express the opinion that "most people can be trusted"; in almost all of the nondemocratic societies, or those that have only recently started to democratize, interpersonal trust is below this level.2
It seems likely that democratic institutions are conducive to interpersonal
2 Because the 1990 World Values Survey finds a surprisingly high level of interpersonal trust among the Chinese public (higher than in any other nondemocratic society), we checked these results against another national survey carried out in China, using the same question. An October 1993 survey of urban China carried out for Ichiro Miyake and Kazufumi Manabe by the Institute for Public Opinion Research of the People's University of China, with an N of 1,920, also shows higher levels of interpersonal trust in China than in any other nondemocratic or newly democratic society (for further details, see Manabe, 1995). The level of interpersonal trust shown for China in figure 6.2 reflects the combined results from the 1990 World Values Survey and the 1993 survey by Manabe and Miyake.
174
CHAPTER    6
O) ü
C
'in
•>. o
(S
u o
E o ■a
3 O 3 C
"43 C O
o
Before 1920 1925 1930	-					w^%		•    ®®
1935	-							
1940								
1945 1950	-				Italy @             © Austria           # West Germany	•		
1955				France ®				
1960								
1965	-							
1970 1975 1980 1990 After	Brazil • © Turkey I		Portugal    ,.           ^    í?   •*■      „				China f»                              I	i
10
20
30
40
50
60
70
Percentage saying "most people can be trusted"
Figure 6.2. Stable democracy and interpersonal trust.
r = .72, N = 43, significant at .0000 level. Number of years for which democratic institutions have functioned continuously in given country, by percentage saying that "most people can be trusted." Source: 1990-93 World Values Survey.
trust, as well as trust being conducive to democracy. We do not have the long time series database that would be needed to sort out the causal linkages between culture and institutions. But there is no reason to simply assume that institutions determine culture, rather than the other way around. As Putnam (1993) has demonstrated in the Italian case, cultural patterns already present in the nineteenth century seem to have helped shape the economic and political developments of given regions in the twentieth century. We suspect that culture and social structure tend to have a mutually supportive relationship in any stable social system. The available evidence cannot determine the causal direction, but it does indicate that culture and political institutions have a strong tendency to go together—with trust and stable democracy being closely linked, as the political culture literature has long claimed.
Stable Democracy and Legitimacy
In recent years, formerly authoritarian regimes from East Asia to Central Europe and the former Soviet Union have held their first free elections. But it is one thing to adopt formal democracy and another to attain stable democracy.
RINGING    THE   PEOPLE    BACK   IN
175
Immediately after World War I, a number of new democracies were established, many of which did not survive the stresses of the interwar era. The most tragic and fateful case was that of Germany. Democratic institutions were seen by many Germans as a foreign element that had been forced on them by defeat in World War I. Authoritarian elites still held influential positions, and the underlying mass political culture was not congruent with democratic institutions (Eckstein, 1961, 1988). Democracy failed to develop the deep-rooted allegiance among the mass public that might have enabled it to weather difficult times. Formal democracy can be established by elites—but once politics is decided by free elections, the orientations of the masses become crucial. In Weimar Germany, Hitler became chancellor through free elections.
The Weimar Republic collapsed in the face of economic difficulties because it lacked legitimacy and because an authoritarian political culture persisted. But culture is a variable, not a constant. It can change gradually, as the history of Germany after World War II demonstrates. Democracy slowly established roots among the West German people after 1945 (Boynton and Lowenberg, 1973; Baker, Dalton and Hildebrandt, 1981). By the 1980s, West Germany had become a stable democracy.
Weimar Germany never had a chance to develop this kind of legitimacy. Associated with defeat from its start, it soon faced the hyperinflation of the 1920s; it was unable to maintain internal order; and it finally collapsed under the impact of the Great Depression in the 1930s. Several decades later, the Bonn regime did develop legitimacy, but it did so gradually. Throughout the first decade of its existence, a large proportion of the German public continued to agree with the statement that "the Nazi regime was a good idea, badly carried out." As recently as 1956, a plurality of the West German public still rated Hitler as one of Germany's greatest statesmen; 1967 was the first year in which an absolute majority of respondents rejected that claim (Conradt, 1993: 51-52).
Democratic institutions gradually won acceptance. At first this acceptance was based on the postwar economic miracle; by the late 1950s, the Bonn republic had achieved remarkable economic success. The 1959 Civic Culture survey showed that while many British and American citizens expressed pride in their political institutions, few Germans did. But the West Germans did take pride in their economic success (Almond and Verba, 1963). Mass support for the democratic regime in Bonn continued to grow with continued economic achievement, though economic success was not the only reason for its growing legitimacy. The institutions of the Federal Republic (unlike those of Weimar) maintained domestic order and provided for a peaceful transfer of political power from a hegemonic party to the opposition in the 1960s. By the late 1970s, the West German public was more apt to express satisfaction with the way their political system was functioning than were most other Western European peoples, including the British. Democracy had finally developed roots in West German society.
176
CHAPTER    6
Subjective Well-being and Legitimacy
Political economy research deals with similar processes leading to the development of mass support, but it normally has a short-term focus. If the economic cycle has been going well, support for the incumbents increases; if the economy has done poorly, support for the incumbents declines. In the short run, the response is to "throw the rascals out" (Kramer, 1971; Lewis-Beck, 1986; Markus, 1988). Support for a democratic regime has similar dynamics but is based on deeper long-term processes. Recent economic success may enhance support for the individuals in office. But if, in the long run, people feel that life has been good under a given regime, it enhances feelings of diffuse support for that regime. Thus, feelings of overall subjective well-being play a key role in the growth of legitimacy. Legitimacy is, of course, helpful to any regime, but authoritarian systems can survive through coercion; democratic regimes must be legitimate in the eyes of their citizens, or, like the Weimar Republic, they are likely to collapse.
In preindustrial society, chronic poverty was taken for granted as a normal part of life. But in industrial society, mass publics have come to expect their governments to provide for their well-being. Thus, in industrial society, reasonably high levels of subjective well-being have become a necessary though not sufficient condition for stable democracy: societies with high levels of subjective well-being can function as democracies, though they do not necessarily become democratic unless they also have high levels of trust and other preconditions; societies with low levels of subjective well-being are likely to have coercive governments or to collapse in the face of mass demands for radical change.
Satisfaction with one's life as a whole is one of the best available indicators of subjective well-being, and it has been surveyed regularly in the Euro-Barometer studies. A society's prevailing level of subjective well-being is a reasonably stable cultural attribute—and one that has important political consequences. If a society has a high level of subjective well-being, its citizens feel that their entire way of life is fundamentally good. Their political institutions gain legitimacy by association.
Surprising as it may seem at first glance, satisfaction with one's life as a whole is far more conducive to political legitimacy than is a favorable opinion of the political system itself. Mass satisfaction with the way the political system is currently functioning has only a modest linkage with stable democracy; but satisfaction with one's life as a whole is a strong predictor of stable democracy (Inglehart, 1990). On reflection, it makes sense that satisfaction with one's life as a whole is a stronger predictor of stable democracy than is satisfaction with the political system. For politics is a peripheral aspect of most people's lives; and satisfaction with this specific domain can rise or fall over night. But if one feels that one's life as a whole has been going well under democratic institutions, it gives rise to a relatively deep, diffuse, and enduring basis of support for those institutions. Such a regime has built up a capital of mass support that can help the regime weather bad times. Precisely because overall life satisfaction is deeply rooted and diffuse, it provides a more stable basis of sup-
RINGING    THE   PEOPLE    BACK   IN
177
Before 1918		_							ytíVľffW	f®
	1920	-							>^2Jjmn	h
C	1927	-							<= 'C °>.£      ô) £ ô -S = SI ~	I
"E o									,<; m ä £    ilJtcaj o    m         -z   . vvtnä	
o ,.	1934									
C Q)										
3 O U.  C	1941	—								
O >>									Japan           Italy	
	1948	-							•                «       „West Germany	
S3  V)										
3  3									Austria	
K O	1955	_							0	
(fl   ? C-									France ®	
o 5	1962	-								
										
o	1969	-								
o										
1           1976 Q 1983 1990 After 1990		m	í G	1	t	«J ndia     S @         o ^ *  o (S   rtr ,c CDU  7J(b 5? &   «*'C ®    ® 1     1		í	(B Portugal £     Spain ®        c          ® i      &      ± '-   ü »-    ®  /&&   £• ® '5 3 ®^*o& ,£* ^® o    ©ť" ......	
0   7  14  21  28  35  42  49  56  63  70  77  84  91
Subjective Well-being Index
(Mean of: % Happy Minus % Unhappy, and % Satisfied Minus % Dissatisfied)
Figure 6.3. Stable democracy and subjective well-being, r = .82, N = 42, significant at .0000 level. Number of years for which democratic institutions have functioned continuously in given country, by percentage ranking "high" on subjective well-being index. Source: 1990-93 World Values Survey.
port for a given regime than does political satisfaction. The latter is a narrower orientation that taps support for specific incumbents at least as much as support for the regime. Accordingly, political satisfaction may fluctuate rapidly over time, with adherents of the Left expressing higher levels when their party is in office, and supporters of the Right showing higher levels when their party is in power—but overall life satisfaction is relatively stable. Political satisfaction mainly taps support for the current incumbents; life satisfaction taps support for the type of political system, or regime.
Figure 6.3 shows levels of subjective well-being in more than 40 societies, based on combined responses to questions about life satisfaction and personal happiness. It examines a broader range of societies than ever before, including a number of authoritarian societies and new democracies. As this figure shows, societies characterized by a relatively strong sense of subjective well-being are far likelier to be stable democracies than societies characterized by a low sense of well-being, confirming earlier findings (Inglehart, 1990). The correlation (r = .82) is remarkably strong. Our interpretation is that, because
178
CHAPTER    6
a sense of subjective well-being is diffuse and deep-rooted, it provides a relatively stable basis of support for a given type of regime.
When people are dissatisfied with politics, they may change the parties in office. When the people of a given society become dissatisfied with their lives, they may reject the regime—or even the political community, as in the case of the Soviet Union and Yugoslavia. Only rarely does mass dissatisfaction reach this level.
Research on subjective well-being in many countries has virtually always found that far more people describe themselves as "happy" than as "unhappy" and far more people describe themselves as satisfied with their lives as a whole than as dissatisfied (see, for example, Andrews, 1986). The data from the 1990 World Values Survey reveal the lowest levels of subjective well-being ever recorded in research on this subject. In the surveys carried out in Russia, Belarus, and Bulgaria, as many people described themselves as "unhappy" as "happy"; and as many said they were "dissatisfied with their lives as a whole" as said they were "satisfied." This is an alarming finding. Normally, people tend to describe themselves as at least fairly satisfied with their lives as a whole, even in very poor societies. But in 1990, these three societies ranked far below even the poorest countries such as India, Nigeria, or China. Subjective well-being had fallen to unheard-of levels. It seems significant that in all three societies, the system of government collapsed during the year following these surveys—and in the Soviet case, the political community itself also collapsed, breaking up into successor states.
Postmaterialist Values, People Power, and Democracy: The Interaction between Mass Publics and Elites
Democratization is not something that automatically occurs when a society's people attain given skill levels and a given threshold of value change. The process can be blocked or triggered by societal events. For Eastern Europe, Gorbachev's accession to power was important: he made it clear that the Red Army would no longer intervene to stop liberalization in these countries. This, together with economic failure, was a triggering event that explains why liberalization suddenly took place throughout the region in 1989-91, rather than a decade earlier or later. But this catalyst would not have worked if underlying societal preconditions had not developed. These preconditions were not present earlier: with the possible exception of Czechoslovakia (the most developed society), none of the Eastern European countries were stable democracies before World War II. Ironically, an unintended consequence of the relative security and rising educational levels provided by four decades of communist rale was to make Eastern European publics less willing to accept authoritarian rule and increasingly adept at resisting it. Such cultural changes can be repressed by domestic elites or by external military force. But by the late 1980s, such countries as Poland, Czechoslovakia, Hungary, and East Germany were ripe for democratization.
BRINGING    THE    PEOPLE    BACK   IN
179
Once it became clear that the threat of Soviet military intervention was no longer present, mass pressures for democracy surfaced almost overnight.
These forces interact with the elites in control of a given society. The generational transition that brought Gorbachev to power could, conceivably, have brought some other less flexible leader to the top. This might have delayed the process of reform for some years, but it would not have held back the clock forever.
The impact of changing values on mass potential for unconventional political action is not limited to Western societies. East Asian societies show the same phenomenon; indeed, rising mass participation began to manifest itself in South Korea before the recent surge of democratization in Eastern Europe. In 1987 an unprecedented wave of demonstrations swept South Korea, demanding direct election of the president. The government yielded, and the ensuing elections in December 1987 were the fairest in South Korean history, with the opposition actually winning a clear majority of the vote. Only the fact that the two main opposition candidates split their vote almost evenly enabled the governing party's candidate to win. In the early 1990s, Taiwan, facing similar pressures from an increasingly educated and articulate populace, also adopted freely contested elections.
China went through a somewhat similar crisis in 1989, but it ended with bloody repression of the dissidents. This illustrates an important point: democratization is never automatic. It reflects the interaction of underlying social changes and specific historical events and leaders. A resolute authoritarian elite can respond to demands for reform by slaughtering the citizens involved. But in choosing this course, one pays a price: the loss of legitimacy and citizen cooperation. In part, the Chinese leadership's choice of this option was feasible because China was still at a considerably less advanced level of development than the other nations we have discussed. Its per capita income was only a fraction of that in South Korea, Taiwan, or most of Eastern Europe. China's prodemocracy movement, in 1989, was mainly based on the younger and better educated strata in the urban centers. Its repression brought little repercussion among China's vast rural masses, which still comprise the great majority of the population.
Subjective well-being levels seem to have been falling throughout the socialist world during the 1980s. The most reliable evidence comes from Hungary, the only ex-socialist nation in which the World Values survey was carried out in 1981 as well as 1990. Both happiness and life satisfaction fell by about 20 points from 1981 to 1990: in the former year, Hungary ranked at about where Turkey and Mexico are on figure 6.3; but by 1990, it had fallen to the level of India. A local survey was also carried out in one region (Tambov oblast) of the Russian republic in 1981, using the World Values survey questionnaire. Comparing these results with those from the 1990 survey of Russia indicates that subjective well-being fell even more markedly in Russia than in Hungary.
A large decline in the subjective well-being of a given public is unusual and may portend major changes in the society. The decline in subjective well-being in Hungary and Russia probably was linked with the deepening economic and
180
CHAPTER    s
political crises of the socialist world in the 1980s. In the Soviet case, it is clear that the decline of subjective well-being was not simply a mass reaction to elite-level events, for our findings of unprecedentedly low subjective well-being among the Russian people were registered before the economic and political system broke down in August 1991. The decline of subjective well-being among mass publics preceded the collapse of communism and the breakup of the Soviet Union.
We suspect that under the Weimar Republic, the German public also manifested low levels of subjective well-being. It is too soon to say whether the former Soviet Union will follow the path of Weimar or that of Bonn. The Russian economy is beginning to recover. But it is clear that in 1990-91 diffuse support was at alarmingly low levels; it would be rash to assume that democracy is safely installed in the former Soviet Union.
Although dependency theory itself has largely been abandoned, the heritage of its efforts to discredit political culture still lingers. Recent interpretations of democratization tend to focus on elite bargaining or on economic factors outside the individual, de-emphasizing the role of mass publics. This is one-sided. It is also ironic, because democracy is, by definition, a system in which mass preferences determine what happens. Mass political culture is certainly not the only factor; but, we argue, it plays a crucial role—particularly in consolidating democracy and enabling it to survive over the long term. It is time to reevaluate the role of political culture. We are in a better position to do so than ever before, because we now have a database that makes it possible to examine the linkages between mass belief systems and political institutions in global perspective.
Empirical Analyses: Three Aspects of Democracy
Let us summarize our key theoretical points. Our central claim is that economic development is linked with democracy because it tends to bring social and cultural changes that help democracy emerge and flourish. The assertion that cultural factors play an important role in sustaining democracy is the most controversial part of this claim, but we believe that social change is also important. These two types of change play quite different roles in relation to different aspects of democracy. Economic development may encourage democracy, but democracy does not emerge automatically. It emerges and flourishes insofar as economic growth produces the social and cultural changes we have just discussed. These factors impact differently on three different aspects of democracy: (1) the amount of change toward democracy in a given period, (2) the level (or extent) of democracy, and (3) the persistence of democracy over time.
Table 6.1 examines the impact of cultural factors on each of these three aspects of democracy, using multiple regression analysis. We have already seen (in figures 6.1 and 6.2) that well-being and trust are closely linked with the stability of democratic institutions. Table 6.1 demonstrates that (controlling for each other's effects) they both have powerful linkages with stable democracy,
B r 1 N G I N G    THE   PEOPLE    BACK   IN                                                                     181
Table 6.1
Cultural Values and Democracy: Multiple Regression Model
Independent Variable	Stability of Democracy 1920-95	Level of Democracy 1990	Level of Democracy 1995	Change in Level 1990-95
Culture Well-being	0.74** (6.34)	0.14** (7.77)	0.05** (2.95)	-0.09** (-4.08)
Trust	82.91** (4.00)	-1.17 (-0.35)	-0.07 (-0.02)	1.10 (0.28)
Intercept	-37.13	3.09	8.82	5.73
Adjusted R2	.76	.66	.20	.33
Number of Cases	41	43	43	43
Notes: Entry is unstandardized OLS coefficient. Coefficient divided by standard error is in parentheses.
*Variables significant at .05 level **Variables significant at .01 level
giving preliminary support to Inglehart's (1990) findings from the narrower range of countries in the 1981 World Values Survey, that trust and well-being are conducive to stable democracy. But well-being and trust have quite different relationships with each of our three dependent variables. They explain a very large proportion (76 percent) of the variance in stability of democracy, and a large proportion (66 percent) of the variance in levels of democracy in 1990; but their linkage with levels of democracy in 1995 is much weaker (explaining only 20 percent of the variance). The relatively weak linkage between culture and levels of democracy in 1995 reflects the fact that a major historical change took place from 1989 to 1995: an avalanche of new democracies emerged, partly through the collapse of communism in the former Soviet Union and Eastern Europe, but also through a major wave of democratization in other societies from South Korea to South Africa. Among the 41 independent polities in the 1990-93 World Values Survey, more than one-third began a transition to democracy during this period.
Virtually all of these new democracies had much lower levels of well-being and trust than the already established democracies, which greatly weakened the relationship between political culture and democratic institutions in 1995. But whether or not democratic institutions survive in these new democracies will depend, in large part, on the extent to which their publics develop a sense of well-being and interpersonal trust.
The change in the relationship between these cultural variables and changes in level of democracy from 1990 to 1995 is even more dramatic. Subjective well-being shows a strong negative linkage with this variable: the societies that were most likely to shift toward democracy were those in which the public
182
CHAPTER    6
Table 6.2
Social Structure and Democracy: Multiple Regression Model
=----------------------	Stability of	Level of	Level of	Change
Independent Variable	Democracy 1920-95	Democracy 1990	Democracy 1995	in Level 1990-95
Social Structure Percent Service Sector	1.50** (3.54)	0.33** (6.51)	0.12** (2.96)	-0.21** (-3.24)
Percent Higher Education	0.82** (2.95)	0.03 (1.02)	0.08'** (2.85)	0.04 (0.97)
Intercept	-62.50	-6.84	3.86	10.70
Adjusted R2	.55	.65	.48	.19
Number of Cases	41	42	42	42
Notes: Entry is unstandardized OLS coefficient. Coefficient divided by standard error is in parentheses.
♦Variables significant at .05 level »»Variables significant at .01 level
showed the lowest levels of subjective well-being. Thus subjective well-being shows strong relationships with all four dependent variables, but reverses its role in connection with short-term changes. While high levels of well-being are linked with stable democracy and high levels of democracy, low levels of well-being are linked with short-term shifts away from authoritarian institutions. This finding supports our interpretation that subjective well-being is crucial to the legitimacy of political institutions: when it is absent, neither democratic nor authoritarian institutions are likely to endure.
Table 6.2 shows the linkages between democracy and our two indicators of cognitive mobilization, occupational structure, and educational level. As hypothesized, both variables have strong positive linkages with the stability of democratic institutions: societies with a large service sector and societies in which a relatively large proportion of the given age cohort receives "tertiary" education (as defined by the World Bank) are much likelier to be stable democracies than are other societies. Both variables are also linked with levels of democracy in both 1990 and 1995 (though the linkage with education falls below significance in the former year). But these two variables explain relatively little of the variance in the changes that took place from 1990 to 1995—and here again, we find a reversal of polarity: the proportion of the economy in the service sector shows a rather strong but negative relationship with change.
Let us now undertake a more comprehensive analysis of how culture and social structure relate to economic development, and to each of the three aspects of democracy. Table 6.3 shows the results of OLS regression analyses measuring the impact of culture, social structure, and economic development on democratic stability. Model 3.1 includes all three types of independent variables,
bringing   the  people  back in
Table 6.3
Stability of Democracy: Multiple Regression Model
183
Independent				
Variable	. Model 3.1	Model 3.2	Model 3.3	Model 3.4
Culture				
Well-being	0.25	—	0.36**	0.44**
	(1.90)		(3.09)	(3.03)
Trust	57.07**	—	47.51**	82.43**
	(3.08)		(2.74)	(4.02)
Social Structure				
Percent Service	0.51	0.53	__	0.78*
Sector	(1.59)	(1.67)		(2.00)
Percent Higher	0.05	0.12	_	0.30
Education	(0.30)	(0.56)		(1.47)
Economic				
GNP/capita, 1990	0.15**	0.25**	0.18**	__
($100s)	(3.91)	(6.87)	(5.23)	
Intercept	-44.40	-24.96	-24.02	-66.19
Adjusted R2	.86	.80	.86	.81
Number of Cases	41	41	41	41
Notes: Dependent variable is the number of years for which democratic institutions functioned continuously in the given society from 1920 to 1995. Entry is unstandardized OLS coefficient. Coefficient divided by standard error is in parentheses.
*Variables significant at .05 level
**Variables significant at .01 level
and it explains fully 86 percent of the variance in the number of years for which democratic institutions have functioned consecutively in these 41 societies. Taking the other variables into account, interpersonal trust and GDP/capita emerge as the key factors, both being significant at well above the .01 level. Subjective well-being also seems important, being significant at very near the .05 level. Neither occupational structure nor educational level shows significant effects.
When we drop the two cultural variables (in model 3.2), the proportion of explained variance drops to .80 and the impact of economic development rises markedly, taking up most of the slack. But when we drop the two social structural variables (in model 3.3), the proportion of explained variance remains unchanged, at .86; the cultural and economic variables take up all of the slack, with subjective well-being and interpersonal trust both being significant at above the .01 level. Finally, when we drop GNP/capita from the regression, the proportion of explained variance falls to. 81; the two cultural variables take up most of the slack, though the percentage in the service sector also rises to the .05 level of significance.
184
CHAPTER    6
Our model is robust and indicates that the impact of economic development on stable democracy seems to work mainly through its tendency to bring cultural and (to a lesser degree) social changes. Dropping GNP/capita from the model reduces the explained variance by only five percentage points; though the linkage between development and democratic stability is very strong, most of its impact seems to pass through the cultural variables (and excluding them reduces the explained variance even more than does excluding GNP/capita).
Burkhart and Lewis-Beck (1994) have argued convincingly that economic development leads to democracy, and not the other way around. Building on their analysis, we would conclude that the most plausible interpretation of these results is that economic development leads to stable democracy mainly (though not entirely) insofar as it brings changes in political culture and social structure. This model could be depicted as follows:
SOCIAL STRUCTURE
ECONOMIC   X                                       ^     STABLE
DEVELOPMENT v                                      _ DEMOCRACY
CULTURE
Can We Use 1990 Measures of Trust and Well-being as Indicators of Their Levels at an Earlier Time? The Stability of Cultural Variables
Before we go any farther, let us take up a basic problem involved in any endeavor to measure the impact of political culture on long-term democratic stability. Empirical measures of political culture from most of the world's societies have not been available until quite recently; consequently, any analysis of culture's impact on long-term stability must necessarily use recent measures to help explain events that took place in earlier years. Thus, the analysis in table 6.3 uses cultural measures carried out in 1990 to explain democratic stability from 1920 to 1995: obviously, we would prefer to have cultural measures from 1920 or earlier for this analysis, but such data are not available.
Using a 1990 measure of culture to explain the stability of democracy from 1920 to 1995 depends on the assumption that the cultural variables are relatively stable. But this assumption (though fortified by countless anecdotes about the stability of the cultural characteristics of given nationalities through the ages) has never been proven empirically. As we will show, there is strong empirical evidence that the cultural characteristics dealt with here actually are relatively stable. We have already seen one piece of this evidence: the fact that, from the Almond and Verba study in 1959 to the present survey in 1990, South-
RINGING    THE    PEOPLE    BACK   IN
185
65 60 55 50 45 40 35 30 25 20 15 10 5
J___I    I    I___L_J___LJ___.......I    I    I    I    I    I    I
75    76    77    78    79
'80    '81    '82    '83
Year
'84    '85    '86    '87
'89    '90    '91     '92    '93    '94   '95
Figure 6.4. Cultural differences are relatively enduring but not immutable: cross-national differences in satisfaction with one's life as a whole, 1973-95. Source: Euro-Barometer surveys carried out in each respective year.
era Italy has been characterized by much lower levels of interpersonal trust than Northern Italy—a finding that is entirely consistent with Putnam's (1993) evidence that the contemporary differences in political performance that he found between Northern and Southern Italy can be traced back to cultural differences that already existed more than a century earlier.
Although we have only fragmentary evidence concerning the long-term persistence of interpersonal trust, we have much more detailed evidence concerning another of our key variables. Overall life satisfaction has been measured in the Euro-Barometer surveys carried out in the member countries of the European Union every spring and fall from 1973 to the present. As figure 6.4 demonstrates, overall life satisfaction shows an impressive degree of cross-national stability in the European Union countries, from 1973 to 1995. Although a society's level of subjective well-being can and does change gradually over time, high or low levels are a relatively stable attribute of given societies. The correlation between a given country's level of life satisfaction at the first time point for which data are available, and its level in 1995 (the latest time point for which we have data), is .81: for most societies, this covers a 22-year time span, and it represents a truly impressive level of stability. Furthermore, as inspection of figure 6.4 demonstrates, this stability maintains itself throughout the period from 1973 to 1995, and not just at the two endpoints: in every year for which we have data, the Dutch and the Danes always rank near the top, while the Italians, French, and Portuguese always rank near the bottom.
To provide a yardstick by which to evaluate the stability of this basic cultural orientation, let us ask: How does it compare with the stability of the most
186
CHAPTER    6
frequently used of all economic indicators, per capita GNP? Relative levels of wealth are generally considered to be very stable. This assumption is well founded: with few exceptions, the relatively rich nations of 1900 were also the relatively rich nations of 1995; and most of the societies that were relatively poor in 1900 were still relatively poor in 1995. Accordingly, during the 20-year period from 1970 to 1990, GNP/capita was relatively stable, showing a correlation of .73 among the societies in the 1990 World Values Survey. But—surprising as it may seem to those who view economic data as "hard" and cultural data as "soft"—our cultural indicator shows even greater stability over time than does the economic indicator!
The data from the 1981 and 1990 World Values surveys enable us to test the stability of key cultural characteristics on a broader scale, using the data from the 24 societies on five continents included in both of these surveys. The results are impressive. Our index of subjective well-being (based on overall life satisfaction and reported happiness) shows a correlation of .86 between the levels measured in 1981 and the levels measured in 1990: this is even higher than the .81 correlation shown in figure 6.4. Moreover, interpersonal trust (as measured in the 1981 surveys) shows an amazingly high correlation of .91 with interpersonal trust in 1990. By comparison, the per capita GNP of these same countries in 1980 shows a correlation of .88 with their per capita GNP in 1990: a stability level about as high as that of our two cultural indicators. All of these figures are high. When one speaks of "rich countries" versus "poor countries," one is indeed dealing with a relatively stable attribute of most societies. But this is equally true of our two political culture variables. Relative levels of interpersonal trust and subjective well-being seem to be as stable attributes of given societies as are their economic levels.
Cultural variables are often thought of as vague and ethereal simply because we usually have only vague, impressionistic measures of them. When measured quantitatively, basic orientations such as these display impressive stability. This is an important finding, which supports the claim that cultural variables have an autonomy and momentum of their own.3 Moreover, it suggests that our measures of political culture carried out in the 1980s and 1990s may be reasonably good indicators of how these societies ranked in earlier decades: though we cannot go back in time and measure the orientations of these publics in the 1920s, we need not abandon the effort to understand how political culture contributes to the long-term survival of democratic institutions.
Let us examine this problem from another perspective. The question is "How well can data from 1990 be used in multivariate analysis, to stand in for data measuring the same variable at an earlier point in time?" In table 6.3 we
3 Culture has a significant degree of autonomy from economic factors. Though affected by economic events, culture is not simply a consequence of economic change: (1) it is shaped by many other factors besides economic ones, including wars, great leaders, major diseases, and other historical events; and (2) even insofar as they are shaped by economic factors, cultural changes have significant time lags. Thus culture has a momentum of its own and can influence economic factors as well as being influenced by them.
bringing  the  people  back in                                                            187
Table 6.4
Stability of Democracy, Using GNP/Capita in 1990, 1980, 1970, and 1957
as Independent Variables: Multiple Regression Model
Independent				
Variable	Model 4.1	Model 4.2	Model 4.3	Model 4.4
Culture				
Well-being	0.25	0.34**	0.45**	0.46**
	(1.90)	(2.59)	(3.42)	(3.35)
Trust	57.07**	63.75**	66.86**	73.33**
	(3.08)	(3.43)	(3.51)	(3.77)
Social Structure				
Service Sector	0.51	0.57	0.68*	0.67
	(1.59)	(1.79)	(2.06)	(1.95)
Higher Education	0.05	0.07	-0.06	-0.09
	(0.30)	(0.35)	(-0.26)	(-0.37)
Economic				
GNP/capita, 1990	0.15**	—	—	—
(SlOOs)	(3.91)			
GNP/capita, 1980	—	0.25**	—	—
(SlOOs)		(3.56)		
GNP/capita, 1970	—	—	1.20**	—
(SlOOs)			(3.11)	
GNP/capita, 1957	—	—	—	0.02*
(SlOOs)				(2.53)
Intercept	-44.40	-54.96	~ 62.05	-60.31
Adjusted R2	.86	.85	.84	.83
Number of Cases	41	41	41	41
Notes: Dependent variable is the number of years for which democratic institutions functioned continuously in the given society from 1920 to 1995. Entry is unstandardized OLS coefficient. Coefficient divided by standard error is in parentheses.
*Variables significant at .05 level
"Variables significant at .01 level
used 1990 GNP/capita to measure the impact of economic development on cultural and social change, and on democracy. We used the 1990 data to be on the same footing with the cultural variables, which were also measured in 1990. But this means that the economic data have the same problem of chronology as the cultural data: causes precede effects, and we are using economic data from 1990 to explain democratic stability from 1920 to 1995.
In fact, we get essentially the same results when we employ economic indicators from much earlier time points: GNP/capita is a relatively stable attribute
188
CHAPTER    6
of given societies and, although absolute levels of income vary from one year to the next, the relative positions of given societies are reasonably stable. Thus, our regression model yields virtually identical results when we use GNP/capita in 1990, as when we use GNP/capita from earlier times.
Table 6.4 shows the results of multiple regression analyses of democratic stability, using GNP/capita in 1990, 1980, 1970, and 1957, respectively, as our indicators of economic development levels. Although the coefficients vary slightly, the same basic model emerges: GNP/capita, subjective well-being, and interpersonal trust are the key variables in every case. Moreover, our various models all explain approximately the same amounts of variance, ranging from a low of 83 percent (using GNP/capita in 1957) to a high of 86 percent (using GNP/capita in 1990).
As we have seen, our two cultural variables are fully as stable over time as is GNP/capita. Although we do not have measures of well-being and trust from earlier decades, we suspect that if we could obtain them our analysis would produce similar results, with the same basic model emerging. This, at any rate, is what happens when we use economic indicators from earlier points in time. Our model proves to be robust, suggesting that relatively stable cultural indicators from 1990 can serve as surrogates for cultural indicators from earlier points in time.
Testing Additional Variables
Let us now examine the impact of a variety of additional variables that the literature suggests may play important roles in democracy. After preliminary discussion, each of these variables will be tested in multivariate analyses.
The Importance of Organizational Networks
Alexis de Tocqueville stressed the importance of networks of voluntary associations, arguing that democracy had emerged and flourished in America because its people participated in numerous and extensive networks of voluntary associations. This fostered cooperation and trust, which were essential to the successful functioning of democratic institutions. Putnam (1993) also emphasized this factor, arguing that Social Capital plays a crucial role in both political and economic cooperation. Social Capital consists of a culture of trust and tolerance, in which extensive networks of voluntary associations emerge. These networks provide contacts and information flows that are, in turn, supportive of a culture of trust and cooperation: economics does not unilaterally determine culture nor does culture determine economics. The two are intimately intertwined and mutually supportive in any society that flourishes for any length of time.
Putnam's work makes an important contribution to sorting out the causal linkages between economic and cultural factors, facilitated by his development
RINGING    THE    PEOPLE    BACK   IN
189
of an exceptionally long time series of economic and cultural indicators. Analyzing Italian regional-level data from the nineteenth century to the 1980s, Putnam found that certain regions had relatively high levels of social capital, while others had much lower levels. These levels were fairly stable attributes of given regions; and they were strongly linked with the economic development level of those regions. But Putnam's analysis dispels any assumption that these regional cultural differences are simply a consequence of their respective levels of economic development. Putnam found that levels of civic involvement around 1900 predicted civic involvement levels 60 or 70 years later far better than did economic factors. More strikingly still, he also found that levels of civic involvement around 1900 predicted subsequent levels of economic development even better than did economic variables. Putnam's analysis indicates that cultural factors help shape economic life, as well as being shaped by it.
The World Values surveys provide information about organizational memberships. The respective publics were shown or read a list of 16 types of voluntary associations and asked, "To which, if any, of these organizations do you belong?" The surveys cover the following types of organizations: labor unions, religious organizations, sports/recreation organizations, educational/cultural organizations, political parties, professional associations, social welfare organizations, youth groups, environmental organizations, health volunteer groups, community action groups, women's organizations, Third World development groups, animal rights groups, and peace movements.
Rates of organizational memberships vary greatly across societies. The lowest level of membership was recorded in Argentina, where only a cumulative 23 percent belonged to any of the 16 types of organizations: the average rate of membership was slightly over 2 percent. The society with the highest rate of organizational membership was the Netherlands, where membership in these organizations averages 16 percent.
These data underrepresent the low-income societies. This battery was not asked in India or Turkey because many of these organizations scarcely existed there; it was asked in Nigeria, but was framed to imply "Do you sympathize with these organizations?" Consequently, we do not have comparable data from these cases. These questions were asked in a number of relatively low-income countries, however, with particularly interesting results from China. Although China shows a lower rate of organizational memberships than most advanced industrial societies, it has a high rate for a largely rural society.
These data enable us to examine the relationship between organizational membership and stable democracy in an unprecedentedly broad perspective. Figure 6.5 shows the overall relationship between rates of membership in these 16 types of organizations and the number of years for which democratic institutions have functioned in the given society.
Our findings support the Tocqueville-Putnam hypothesis: membership in voluntary associations is strongly linked with stable democracy. The overall regression coefficient is .65, significant at the .0001 level. Societies with high
190
CHAPTER    6
eíore 1920 1925	—		\	\ \	WW	•            8 Iceland Nether lands
1930						
1935	-					
1940	-					
1945 1950	-	© Japan	Austria Italy                                        Ge™any			
1955	-		e		-	
1960	-					
1965						
1970	-					
1975	-	Spain ®	(i4&/			
1980 1985 1990 After	Argentina © -       Romania • i			South >      -J? &       Korea ^ ^            • ■v it/	®	I
50                             100                               150                            200
Cumulative percentage belonging to 16 types of organizations
250
Figure 6.5. Democracy and voluntary associations.
r = .65, N = 35, significant at .0001 level. Number of years for which democratic institutions have functioned continuously in given nation since 1920, by cumulative percentage belonging to 16 types of voluntary associations (e.g., the Dutch public report a cumulative 259 percent belonging to the 16 types of organizations, for a mean 16 percent belonging to each type). Source 1990-93 World Values Survey.
rates of membership are far more likely to be stable democracies than those with low rates of membership. Subsequent analyses will examine whether this holds up in multivariate analysis.
Support for Revolutionary Change and Support for Gradual Reform
Inglehart (1990) used a political culture index composed of interpersonal trust, subjective well-being, and the percentage supporting revolutionary change to explain democratic stability, finding that societies characterized by relatively high levels of support for revolutionary change are less likely to be stable democracies than other societies. Müller and Seligson (1994), in a reanalysis of Inglehart's data (plus six Central American countries), used support for gradual reform in their analysis instead of support for revolutionary change, arguing that the former provides a stronger explanation of shifts toward democracy than does the latter. Both of these variables show reasonably high levels of stability over time—though not as high as that found with subjective well-being and interpersonal trust. Among the 24 societies included in both the
BRINGING    THE    PEOPLE   BACK   IN
191
1981 and 1990 World Values surveys, the correlations between levels in 1981 and levels in 1990 for each of these four variables were:
Interpersonal trust                           .91
Subjective well-being                      .86
Support for gradual reform              .80
Support for revolutionary change     .74
Thus, although there is an enduring tendency for certain societies to be characterized by relatively high levels of support for revolutionary change, this is a less stable variable than interpersonal trust or subjective well-being. These variables will also be examined in multivariate analyses.
Income Inequality
Much of the literature on democracy has emphasized income equality as an important factor in connection with stable democracy. This literature points out that very high levels of income inequality lead to extremist politics in which the dispossessed have nothing to lose and a great deal to gain by radical change—and the privileged elite has an enormous stake in maintaining the status quo at almost any cost. This is a recipe for extremist politics. Conversely, a reasonable degree of income equality is conducive to the spirit of compromise and moderation that is crucial to democratic politics. Furthermore, a diverse economy with many attractive jobs in the tertiary sector makes the elite more willing to accept rotation out of office: in such a setting, government is not the only route to prosperity and power; one may even have greater opportunities to earn a high income out of office than in office. Cross-sectional evidence suggests that economic development tends to produce greater income equality—which could be one reason why economic development is linked with democracy. We would expect income equality to be positively correlated with democracy.
Ethnolinguistic F rationalization
Müller and Seligson (1994) used an Index of Ethnolinguistic Fractionalization from Taylor and Jodice (1983) in their analysis, finding that ethnic diversity makes democratization less likely. We will examine whether this finding holds up when tested in the context of the much broader database provided by the 1990 World Values Survey.
EMPIRICAL RESULTS
As table 6.5 demonstrates, none of these additional factors has a significant impact on stable democracy; and in no case does adding them to the regression analysis increase the percentage of variance explained. The basic model shown in table 6.3 explains 86 percent of the variance in democratic stability; none
192
CHAPTER    6
Table 6.5
Stability of Democracy: Multiple Regression Models Testing Impact of Support for Revolutionary Change, Support for Gradual Reform, Income Equality, and Organizational Memberships
			Model 5.3	Model 5.4
Independent	Model 5.1	Model 5.2	(Income	(Organizational
Variable	(Revolution)	(Reform)	Equality)	Membership)
Culture				
Well-being	0.27	0.32	0.32*	0.25
	(1.90)	(2.05)	(1.96)	(1.46)
Trust	59.90**	56.49**	54.79*	49.44*
	(3.01)	(2.88)	(2.27)	(2.24)
For Revolutionary	0.16	—	—	—
Change	(0.51)			
For Gradual	—	-0.22	—	—
Reform		(-0.90)		
Social Structure				
Percent Service	0.59	0.56	0.51	0.62
Sector	(1.71)	(1.71)	(1.25)	(1.62)
Percent Higher	0.02	0.04	0.05	0.02
Education	(0.09)	(0.21)	(0.23)	(0.01)
Percent Income to	—	—	-0.36	—
Top 20%			(-0.95)	
Membership, 16 Types	—	—	—	0.05
of Organizations				(1.05)
Economic				
GNP/capita, 1990	0.16**	0.16**	0.14**	0.14**
(SlOOs)	(3.47)	(3.42)	(3.18)	(3.31)
Intercept	-52.44	-35.01	-30.87	-51.21
Adjusted R2	.85	.85	.86	.85
Number of Cases	39	39	32	34
Notes: Dependent variable is the number of years for which democratic institutions functioned continuously in the given society from 1920 to 1995. Entry is unstandardized OLS coefficient. Coefficient divided by standard error is in parentheses.
*Variables significant at .05 level
"Variables significant at .01 level
RINGING    THE    PEOPLE    BACK   IN
193
of the models shown in table 6.5 explain more than this proportion of the variance. Thus, neither support for revolutionary change nor support for gradual reform has a significant independent impact on stable democracy.
It seems likely that income equality is conducive to democracy. To test this hypothesis, the percentage of a country's income going to the top 20 percent of the population was included as an explanatory factor in this analysis.4 The findings were interesting. There is a clear tendency for advanced industrial democracies to have higher levels of income equality than preindustrial or newly industrializing societies, most of which are not stable democracies. But the highest levels of income equality were found among the ex-socialist countries—and when they are included in the analysis, the zero-order correlation between income inequality and stable democracy dwindles almost to zero (r = — .06). Democracies tend to have reasonably high levels of income equality, and we believe that this is conducive to democracy (though it may also be a consequence of democracy to some extent, since democracy shifts political power to the public, enabling them to press for more egalitarian social policies). But income equality does not seem to be the main cause of democracy.
Moreover, the relationship between income equality and democracy is not linear. Very low levels of income inequality lead to economic inefficiency and political instability. But, apparently, extremely high levels of income equality can only be attained by coercive governments.
We also tested the impact of membership in voluntary associations, our indicator of Putnam's concept of Social Capital. Although it is strongly correlated with stable democracy, this variable did not show a statistically significant impact when we control for the effects of other variables. This does not prove that it plays no role; it simply indicates that it is not among the two or three variables most strongly linked with stable democracy. Putnam (1993) views organizational membership as contributing to democracy largely because it is conducive to interpersonal trust and cooperation: consequently, we would expect organizational membership to be highly correlated with interpersonal trust, and it is. With only 41 observations and a good deal of overlapping variance (as is the case here), only a few variables are likely to have a statistically significant impact, and in this analysis organizational membership shows a positive but not statistically significant linkage with stable democracy. It also shows positive but not statistically significant linkages with levels of democracy in 1990 and 1995; but, as we will see below, organizational membership does show a statistically significant linkage with changes in levels of democracy from 1990 to 1995.
Although the results are not shown in table 6.5, we also examined the impact of ethnolinguistic fractionalization, finding that it does not show a signif-
4 Data on income equality are from World Bank (1993). These data were not available for the Soviet successor states, which were assigned the mean score observed across the Eastern European ex-communist societies. This is only roughly accurate but should be in the right ballpark: income equality was certainly higher in these societies than in non-communist societies.
194
CHAPTER    6
icant effect on stable democracy, and that adding it to the model does not increase the percentage of variance explained. Although one can point to horror stories about the difficulties that ethnic diversity may pose for democratic governance, from Nigeria in the 1960s to Bosnia in the 1990s, diversity does not seem to rank among the most crucial factors. In global perspective, multiethnic societies are only slightly less likely to be stable democracies than are more homogeneous societies.
Interpersonal trust, subjective well-being, reasonable levels of income equality, low levels of extremism, relatively high levels of political participation and organizational membership, and Postmaterialist values are all part of a highly intercorrelated syndrome that might be called a "prodemocratic culture." And all of these variables are closely correlated with stable democracy. But interpersonal trust and subjective well-being have the highest correlations with stable democracy. In this regression analysis, with a huge amount of shared variance and only 40 cases, only these two variables show statistically significant relationships with stable democracy. We think it highly unlikely that trust and well-being are the only relevant parts of this syndrome. Social reality is usually more complex than that. Quite possibly, all or many of the other elements of this closely related cluster of variables help sustain democratic institutions. The present analysis indicates that cultural variables play an important role in the survival of democracy over time, with trust and well-being constituting the two most prominent cultural variables. It seems unlikely that these two variables alone shape the outcome: more probably, they serve as indicators of a broader cultural configuration that is conducive to democracy.
Levels of Democracy: 1990 and 1995
Table 6.6 analyzes the factors linked with level of democracy in 1990, using the five variables in our basic model. There are some interesting changes from the results in table 6.3. Subjective well-being has a highly significant linkage with levels of democracy in 1990, as it does with democratic stability. But social structure plays a more important role, and economic development a less important one, than it did with stable democracy. The percentage of the economy in the service sector is the most important single factor shaping levels of democracy in 1990.
Interpersonal trust does not have a significant impact on levels in 1990—indeed, it shows a weakly negative linkage with it. The same occurs with the percentage receiving higher education: though it has a strong positive zero-order relationship with democracy in 1990, it shows a weakly negative relationship in this regression analysis. The education finding seems to reflect a familiar phenomenon: education and the size of the service sector are highly correlated, and under certain conditions, including both of them in the regression causes a reversal of the sign on the variable with the larger measurement error—in this case, education (Achen, 1985).
bringing   the  people  b
Table 6.6
Level of Democracy in 1990: Multipl
Independent Variable.
Model 6.]
Model 6.2
Model 6.3
Model 6.'
Well-being	0.07**	—	0.11**	0.08**
	(2.93)		(3.98)	(3.49)
Trust	-1.30	—	-4.75	0.29
	(-0.39)		(-1.40)	(0.09)
Social Structure				
Percent Service	0.18**	0.27**	—	0.20**
Sector	(3.18)	(5.31)		(3.44)
Percent Higher	-0.01	-0.02	—	0.01
Education	(-0.18)	(-0.51)		(0.42)
Economic				
GNP/capita, 1990	0.01	0.02**	0.02**	—
($100s)	(1.60)	(3.08)	(2.64)	
Intercept	-2.79	-4.40	4.10	-4.12
Adjusted R2	.76	.71	.70	.75
Number of Cases	42	42	42	42
Notes: Dependent variable is the level of democracy in 1990 as measured by the Freedom House ratings for Political Rights and Civil Liberties, combined into an additive index. Entry is unstan-dardized OLS coefficient. Coefficient divided by standard error is in parentheses.
♦Variables significant at .05 level
**Variables significant at .01 level
The importance of economic level diminishes markedly here, by comparison with its linkage to democratic stability. When both culture and social structure are included in the model, GNP per capita does not show a significant impact on the 1990 level of democracy (though it does show a powerful linkage when either of these factors is dropped). Moreover, eliminating GNP/capita from the model reduces the percentage of explained variance by only one percentage point, from 76 to 75: here again, we find indications that, although economic development is strongly linked with democracy, its effects work mainly through the changes it brings in culture and social structure.
We also examined the impact of several additional variables, including support for revolutionary change, support for gradual reform, levels of income inequality, organizational memberships, and ethnolinguistic fractionalization. None of these variables shows a statistically significant linkage with level of democracy in 1990, and none of them greatly increases the proportion of variance explained.
196
CHAPTER    6
Table 6.7
Level of Democracy in 1995: Multiple Regression Model
Independent				
Variable	Model 7.1	Model 7.2	Model 7.3	Model 7.4
Culture				
Well-being	-0.01	—	0.01	-0.01
	(-0.64)		(0.55)	(-0.02)
Trust	-3.20	—	-4.34	-1.23
	(-1.03)		(-1.36)    -	(-0.40)
Social Structure				
Percent Service	0.10	0.10*	—	0.12*
Sector	(1.90)	(2.16)		(2.20)
Percent Higher	0.06	0.05	—	0.08**
Education	(1.89)	(1.79)		(2.72)
Economic				
GNP/capita, 1990	0.01*	0.08	0.02**	—
($100s)	(2.11)	(1.52)	(3.74)	
Intercept	5.76	4.90	10.94	4.11
Adjusted R2	.50	.50	.39	.46
Number of Cases	42	42	42	42
Notes: Dependent variable is the level of democracy in 1995 as measured by the Freedom House ratings for Political Rights and Civil Liberties, combined into an additive index. Entry is unstan-dardized OLS coefficient. Coefficient divided by standard error is in parentheses.
*Variables significant at .05 level
"Variables significant at .01 level
Table 6.7 analyzes the factors linked with level of democracy in 1995. When we compare the percentage of variance explained by these variables with the percentage explained by the democracy level in 1990, the proportion drops from .76 to .50. Although this analysis is based on levels of democracy only five years later than the previous analysis, it reflects the state of the world after an avalanche of changes that brought democratization to fully one-third of the nations in our sample. Consequently, structural factors explain much less than they did in the analysis of levels of democracy in 1990: in the new democracies that had emerged by 1995, democratic institutions were less firmly anchored in social structure and culture, and much more contingent on situation-specific factors (such as historical events, elite maneuvering, and the role of specific leaders) than they were in the longer-established democracies. The only statistically significant influence on levels of democracy in 1995 is level of economic development: richer societies were likelier to be democracies than were poorer ones. But when GNP per capita is dropped from the regression, social structural variables take up most of the slack: societies with a relatively
BRINGING    THE    PEOPLE    BACK   IN
197
large service sector and (even more important) societies with a relatively well-educated population were likelier to show high levels of democracy than were those with lower levels of these variables.
Our indicators of political culture are virtually unrelated to levels of democracy in 1995. The massive number of new democracies washes out the linkage between culture and democracy: as we have argued, democratic institutions can be adopted in virtually any setting. But our interpretation implies that democracy is most likely to survive and flourish in societies that rank high on subjective well-being and interpersonal trust. A culture of trust and well-being will probably develop in some of the new democracies; and democracy may fail to survive in others that rank low on trust and well-being. In the long run, both processes tend to reinstate the linkage between political culture and democracy.
As additional analyses indicate, none of the additional variables examined in table 6.5 (support for revolution, support for gradual reform, income inequality, organizational membership, and ethnolinguistic fractionalization) has a statistically significant impact on levels of democracy in 1995, although income inequality comes close to the .05 level of significance: societies with relatively low levels of income inequality were likelier to have higher levels of democracy in 1995 than those with greater inequality.
Analyzing Recent Changes in Levels of Democracy
Inglehart (1988, 1990) argued that political culture plays a crucial role in sustaining democratic political institutions: economic development is linked with democracy, in large part, because it leads to changes in social structure and political culture that are conducive to democracy.
Müller and Seligson (1994) argue that Inglehart's political culture data were collected in 1981 and therefore cannot be used to explain the persistence of stable democracy before that time (unless, of course, they tap stable cultural differences that were present even earlier). Consequently, they drop Inglehart's dependent variable and analyze changes in levels of democracy that occurred after 1981. They claim to be testing Inglehart's thesis, but their analysis is based on a model in which the dependent variable is democracy at time 2 (the 1980s), controlling for democracy at time 1 (the 1970s). This means that they are not analyzing either the extent of democracy or the stability of democracy among the societies in their sample: they are analyzing recent changes (from the 1970s to the 1980s). The authors do not attempt to conceal this fact: they refer to their dependent variable as "change in level of democracy." But their choice of recent change as a test of whether political culture is conducive to democracy has important implications that they seem to have overlooked. They use this dependent variable to address the question "Is political culture conducive to democracy?" But their analysis actually addresses the question "Is political culture conducive to the shifts in levels of democracy observed
198
CHAPTER    6
Table 6.8
Shifts in Levels of Democracy, Muller-Seligson Model: Impact of Income Inequality,
Support for Gradual Reform, and Ethnolinguistic Fractionalization
Independent Variable	Model 8.1 (1970s-1980s shifts) (Muller-Seligson, 1994: 642)	Model 8.2 (1990-95 shifts)
Culture For Gradual Reform	.62** (2.82)	-.05 (-.75)
Social Structure Ethnolinguistic Fractionalization	-.17* (2.13)	-.01 (-.59)
Income to Top 20%	-1.60** (4.10)	-.09 (-1.23)
Level of Democracy, 1980s (1990)	.32* (2.46)	.61** (3.11)
Intercept	93.1	13.1
Adjusted R2	.87	.37
Notes: Dependent variable is the level of democracy in 1980s (1995), controlling for level in 1970s (1990). Entry is unstandardized OLS coefficient. Coefficient divided by standard error is in parentheses.
*Variables significant at .05 level
"Variables significant at .01 level
from one decade to another?" It does not and cannot determine whether a given political culture is conducive to stable democracy or to high levels of democracy during a given period.
Table 6.7 replicates the Muller-Seligson analysis, using their model to analyze shifts in levels of democracy from 1990 to 1995. Model 8.1 in table 6.8 shows the results they obtained, using the data from the 1981 World Values Survey (plus six Central American societies) to analyze shifts from the 1970s to the 1980s. In their analysis, they found that income inequality was the most important influence on democratization from the 1970s to the 1980s. Support for gradual reform, ethnolinguistic fractionalization, and level of democracy at the earlier time point also had statistically significant effects. Their model explained fully 87 percent of the variance in democratization during this time period.
But their model is completely time bound. The factors governing shifts in levels of democracy from one decade to the next are largely situation-specific (in the period they analyzed, such events as the death of Franco and Argentina's defeat in the Falklands War triggered democratization). Accordingly, their model does not hold up when used to analyze the shifts toward democracy in other time periods. As model 8.2 demonstrates, using the same variables with
BRINGING    THE    PEOPLE    BACK   IN
199
the 1990 World Values Survey data, we get completely different results when we apply their model to analyze shifts toward democracy during the period from 1990 to 1995. Neither income inequality nor support for gradual reform nor ethnolinguistic fractionalization has a statistically significant impact on the shifts that took place in this broader sample of countries from 1990 to 1995. Not surprisingly, level of democracy at time 1 does have a significant linkage with level of democracy at time 2, but this is the only element of their model that survives. And the proportion of variance explained by their model drops precipitously—falling from 87 to 37 percent. A completely different group of countries shifted toward democracy in the 1980s, from those that shifted toward democracy during the 1990s—and the two groups of countries were very different in social structure and culture.
Is the breakdown of the Muller-Seligson model due to the fact that we use a broader set of nations in the analysis in model 8.2? No, it is not. Table 6.9 replicates their analysis, using the same set of nations that they examined. The results demonstrate that their findings are indeed time bound. When we focus on the same time span that they examined, we get similar results (see model 9.1). But when we analyze shifts toward democracy among these same societies during other time periods, we get quite different results. Income equality, support for gradual reform, and level of democracy at time 1 were the main influences on the shifts toward democracy that took place from the 1970s to the 1980s, in Müller and Seligson's analysis (with ethnolinguistic fractionalization approaching the 0.05 level of significance). None of these variables consistently shows a significant impact on democratization during the other time periods. Level of democracy at time 1 has a significant effect in two of the four other time periods; income equality has a significant effect in one of the four other time periods; and neither support for gradual reform nor ethnolinguistic fractionalization has a significant effect in any of the four other time periods. The factors that explain shifts toward democracy from the 1970s to the 1980s are not the same as those that explain shifts toward democracy in those same countries from the 1980s to the 1990s, or from the 1970s to the 1980s, or from the 1970s to 1995, or from 1990 to 1995. Situation-specific factors dominate structural factors, in explaining short-term change.
To understand why we get such volatile results, let us look more closely at the changes on which the Müller and Seligson analysis focuses. Figure 6.6 shows which societies changed the most during the period Müller and Seligson analyzed, based on their own data. As figure 6.6 makes clear, their approach does not analyze which nations are most democratic, or which nations have the most stable democratic institutions. Instead, their analysis focuses on the difference between two distinct sets of countries: one group consisting of Spain, Argentina, Honduras, Portugal, Greece, and Panama, which had experienced large recent changes; and another group of 21 societies that showed little or no change in level of democracy from the 1970s to the 1980s—and which lumps together the stable democracies and the stable authoritarian states and any marginally democratic societies that did not undergo major changes
Table 6.9
Effects of Civic Culture Attitudes and Macrosocietal Variables
		Equations Explaining Level			of Democracy	
Independent Variable		Model 9.1 1970s-1980s (Muller-Seligson Analysis)			Model 9.2 1980s to 1990	
Level, 1970s		0.32	(2.23)*			—
Level, 1980s		—			0.96	(6.36)**
Level, 1990		_			' —	
Gradual Reform		0.62	(2.49)*		0.16	(0.78)
Interpersonal Trust		0.04	(0.20)		-0.19	(-1.44)
GDP/capita		-0.01	(-0.05)		-0.07	(-0.72)
Income Inequality		-1.61	(-3.21)	!*	-0.20	(-0.44)
Ethnolinguistic Fractionalization        —1.62			(-1.78)		0.65	(0.91)
Intercept		93.13			14.10	
Adjusted R2		.85			.92	
Number of Cases		25			25	
		Equations Explaining		Level of D	emocracy	
Independent Variable	Model 9.3 1970-95		Model 9.4 1980S-95		Model 9.5 1990-95	
Level, 1970s	0.01	(0.07)	—		—	
Level, 1980s	—		0.53	(2.83)**	—	
Level, 1990	—-		—		0.15	(0.76)
Gradual Reform	-0.13	(-0.47)	0.50	(-1.93)	-0.25	(-0.80)
Interpersonal Trust	0.12	(0.57)	0.06	(0.35)	0.13	(+0.66)
GDP/capita	0.07	(0.46)	0.03	(0.23)	0.07	(+0.50)
Income Inequality	-1.20	(-2.20)*	-0.08	(-0.14)	-0.89	(-1.38)
Ethnolinguistic Fractionalization	-0.39	(-0.39)	0.08	(0.88)	-0.19	(-0.19)
Intercept	144.47		76.57		125.45	
Adjusted R2	.61		.73		.62	
Number of Cases	25		25		25	
Notes: Entry is unstandardized OLS coefficient. Coefficient divided by standard error is in parentheses. Estimates are based on 25 cases. Greece and Luxembourg are excluded because data on income inequality is unavailable.
*Significant at .05 level
**Significant at .01 level
BRINGING    THE   PEOPLE   BACK   IN
40-1
_ Spain - Argentina
201
©
CO
V)
©
N
<3>

30
E
8     20
>. o
(Q
a» U
o »
Q
10-
ô)
(B
5   °
-10 -i
Honduras Portugal
Greece Panama
,   Large r Change
' Itaiy ' Japan
Ireland, Luxembourg _     U.S., Britain, Norway, Denmark, Netherlands,      L Little or
Canada, Sweden, Australia, Costa Rica, Belgium ■ France, Mexico
, Nicaragua 1 W. Germany 1 Guatemala ' El Salvador South Africa
No Change
Figure 6.6. Changes in level of democracy from 1970s to 1980s in 27 nations. Source: calculated from Müller and Seligson, 1994: 648 (appendix A).
in this period. Most of the variance on which Müller and Seligson's analysis is based reflects the contrast between six societies that, for various reasons, showed major changes in the 1980s—and a heterogeneous group of societies that did not. Their analysis ranks Spain and Argentina at the top of the scale, far above the Nordic countries or the English-speaking democracies, even though the latter countries rank much higher on both level of democracy and stability of democracy: for Spain and Argentina happen to be the societies that showed the most dramatic changes during this period. Their analysis focuses on recent fluctuations that cultural differences would be unlikely to explain since culture is, by definition, relatively stable.
Müller and Seligson's dependent variable is so volatile that any cultural variables that did explain the pattern they find in the 1980s could not very well explain the pattern found in the 1970s or 1990s: the societies that make a break-
2j)2                                                                                                                          CHAPTER    6
through to democracy in one decade are not likely to be the same ones that make it a decade earlier or a decade later. For example, Spain falls at the high end of the scale in Müller and Seligson's analysis, since it adopted democratic institutions during this period. South Africa falls at the opposite end of the scale. Nevertheless, shortly after 1990 South Africa began a transition to democracy: today it would rank near the high end of their scale, having shown large recent changes. Conversely, Spain and Argentina would now drop toward the low end of the scale, since they did not show large shifts toward democracy in the last few years but declined slightly in the Freedom House ratings. When applied to a different time period, their dependent variable becomes radically different from the one they examined in the 1980s.
Müller and Seligson present a thoughtful analysis that addresses a very real problem: causes precede effects, which means that any attempt to analyze the contributions of political culture to democratic stability before 1960 (when political culture began to be measured empirically) faces difficult measurement problems. But their analysis has two flaws, either of which would be fatal to a test of political culture theory.
First, it is based on a dependent variable that does not address the question of whether cultural factors are responsible for the long-term survival or failure of democracy—instead, it focuses on the fluctuations from the 1970s to the 1980s. By their very nature, cultural differences are relatively stable aspects of given societies and hence unlikely to explain fluctuations in a given society's level of democracy from one decade to the next. Müller and Seligson's analysis controls for the long-term component of democracy and analyzes only the recent fluctuations. This gives them a dependent variable that was measured after the 1981 surveys were carried out, but it is clearly the wrong dependent variable to test the role of culture.
Their analysis is further distorted by an artifact of the data they use. The Freedom House codings of levels of democracy use the stable democracies to define the top level of their scales: from the start, they have been assigned the maximum possible score. This means that they literally cannot rise any farther—and since Müller and Seligson's analysis is based on change, this means that they cannot attain high scores on their dependent variable. This explains the bizarre pattern that is visible in figure 6.5: in virtually every case, the stable democracies get scores near (or exactly at) zero, reflecting no change. Müller and Seligson nevertheless argue that the cultural characteristics associated with stable democracy should be linked with large amounts of change on this measure. Their model specification virtually guarantees that they will not find it, and they don't.
It is perfectly legitimate and useful to analyze short-term changes in levels of democracy, as Müller and Seligson have done; but when one does so, it is important to be aware that this is a very different dependent variable from either levels of democracy or stability of democracy—and one that is mainly shaped by situation-specific factors. With this in mind, let us examine the shifts in levels of democracy that took
bringing  the people  back in                                                            203
Table 6.10
Shifts in Levels of Democracy from 1990 to 1995: Multiple Regression Model
Independent				
Variable	Model 10.1	Model 10.2	Model 10.3	Model 10.4
Culture				
Well-being	-0.09*	—	-0.10**	-0.08*
	(-2.49)		(-3.68)	(-2.56)
Trust	-1.90	—	0.41	-1.53
	(-0.41)		(0.09)	(-0.35)
Social Structure				
Service Sector	-0.08	-0.18*	—	-0.08
	(-0.99)	(-2.50)		(-0.97)
Higher Education	0.06	0.07	—	0.07
	(1.38)	(1.46)		(1.60)
Economic				
GNP/capita, 1990	0.01	-0.01	0.01	—
($100s)	(0.26)	(-1.26)	(0.67)	
Intercept	8.55	9.30	6.19	8.23
Adjusted R2	.32	.20	.32	.34
Number of Cases	42	42	42	42
Notes: Dependent variable is the shift from 1990 to 1995 on the Freedom House combined index of Political Rights and Civil Liberties. Entry is unstandardized OLS coefficient. Coefficient divided by standard error is in parentheses.
♦Variables significant at .05 level
"Variables significant at .01 level
place from 1990 to 1995, using change scores instead of level at time 2, controlling for level at time 1, as Müller and Seligson do: change scores provide a more straightforward measure of change, and one that is comparable to the other dependent variables used in this analysis.
Table 6.10 shows the results of a regression analysis of changes in level of democracy from 1990 to 1995, using the same cultural and social structural explanatory variables as in our previous analyses. One striking contrast with these previous analyses is the relatively small proportion of variance that these variables explain: while our analyses of levels of democracy and stability of democracy explained from 50 to 85 percent of the variance, the same variables explain only 32 percent of the variance in the shifts from 1990 to 1995. This reflects the fact that these recent shifts are not firmly rooted in the social or economic structure of the societies. Controlling for other factors, rich societies were as likely to shift as were poor ones, and highly educated publics were as likely to shift as were poorly educated ones. The most important influence on change is subjective well-being, but it has a reversed polarity in comparison with the previous analyses: the societies that underwent regime changes were
204
CHAPTER    6
Table 6.11
Shifts from 1990 to 1995 (Change Scores): Multiple Regression Models Testing Impact of Support for Revolutionary Change, Support for Gradual Reform, Income Inequality, and Organizational Memberships
Independent Variable	Model 11.1 (Revolution)	Model 11.2 (Reform)	Model 11.3 (Income Inequality)	Model 11.4 (Organizational Membership)
Culture Well-being	-.07 (-1.91)	-.08* (-1.99)	-.09* (-2.33)	-.08* (-2.60)
Trust	-.69 (-.01)	-1.48 (-.29)	.54 (.10)	-11.6* (-2.68)
For Revolutionary Change	.13 (-1.75)	—	—	—
For Gradual Reform	—	-.02 (-.29)	—	—
Social Structure Service Sector	-.03 (-.35)	-.07 (-.83)	.04 (.45)	-.11 (-1.60)
Higher Education	.06 (1.23)	.08 (1.44)	.02 (.41)	.02 (.46)
Income to Top 20%	—	—	-.13 (-1.45)	—
Membership, 16 Types of Organizations	—	—	—	.03* (2.74)
Economic GNP/capita, 1990 ($100s)	.004 (.32)	.001 (.09)	.001 (-.09)	.005 (.65)
Intercept	3.02	8.99	8.32	11.93
Adjusted R2	.34	.28	.36	.62
Notes: Dependent variable is the shift in level of democracy from 1990 to 1995. Entry is un-standardized OLS coefficient. Coefficient divided by standard error is in parentheses. *Significant at .05 level "Significant at .01 level				
societies in which the mass publics had low levels of subjective well-being. When well-being and trust are dropped from the analysis, the percentage of variance drops to only 20 percent (see model 10.2). When this is done, the role of the occupational structure becomes significant: societies with a relatively low proportion of the economy in the tertiary sector are most likely to have shifted. But when the social structure variables are dropped and well-being and
BRINGING    THE    PEOPLE   BACK   IN
205
trust are restored, the percentage of variance explained recovers fully, returning to 32 percent. Dropping GNP/capita from the analysis does not reduce the explanatory power of the model (it actually rises slightly). Far from confirming Müller and Seligson's finding that subjective well-being is irrelevant to democracy, this analysis suggests that the authoritarian regimes fell, in part, because they had lost mass legitimacy.
We also tested the impact of a number of other variables, with the results shown in table 6.11. When we add "support for revolutionary change" to the regression (in model 11.1), this new variable almost reaches the .05 level of significance, while subjective well-being drops just below that threshold: both of these variables tap regime legitimacy and they share a good deal of variance, reducing each others' explanatory power.
When we drop "support for revolutionary change" and add "support for gradual reform," however, the latter variable has very little explanatory power (and subjective well-being rises above the .05 significance level). Contrary to the findings of Müller and Seligson (1994), support for revolutionary change seems to have a stronger impact on short-term changes than does support for gradual reform, though neither of them increases the proportion of explained variance significantly.
Model 11.3 shows what happens when we add income inequality to the regression model. Countries with less inequality are somewhat more likely to change than those with greater inequality, but the effect is not significant and this variable does not add much to the percentage of variance explained.
Model 11.4 adds our indicator of Social Capital, organizational memberships, to the regression. Doing so increases the percentage of variance explained markedly, almost doubling the adjusted R2. In this model, interpersonal trust, subjective well-being, and organizational memberships all have significant effects: societies that are low on trust and well-being, but high on social capital, are the ones that are most likely to have shifted toward democracy from 1990 to 1995. Social Capital had shown the predicted polarity but had not attained a significant level in our previous analyses. Its inclusion here brings a dramatic change, producing what seems to be a model that is much stronger than those that lack it. This analysis is based on a significantly smaller number of cases than the other models, because we have data on organizational memberships for eight fewer societies than those included in the other analyses in table 6.11. Nevertheless, these findings tend to support the hypothesis that social capital plays a significant role in democratization.
Which Comes First: Democratic Political Culture or Democracy?
We have found strong evidence that political culture and the level and stability of democracy are closely linked. But determining the causal direction of relationships in social science is always difficult. One way to explain away the linkages between culture and democracy would be to argue that they reflect a
206
CHAPTER    6
spurious effect of economic determinism, in which economic development gives rise to a specific type of culture, and to democratic institutions, without culture being an intervening variable. This interpretation does not hold up in the light of the analyses we have just carried out: economic development is indeed important, but its effect seems to work mainly through changes that it brings in culture and social structure.
Another interpretation that might explain away the relationships we have observed between culture and democracy could be termed Institutional Determinism. This interpretation would argue that the linkages between culture and democracy exist because democratic institutions determine the underlying culture.
This model contains a grain of truth: institutions do help shape their society's culture—along with many other factors. But the plausibility of the interpretation that institutional determinism is the major explanation is severely undermined by the findings of Burkhart and Lewis-Beck (1994) that economic development leads to democracy, but democracy does not bring economic development. The causal process seems to run from economic factors to institutions, rather than the other way around—and, as we have seen, the economic factors work mainly through changes in culture and social structure.
Institutions do influence politics and economics. But they do not explain them by themselves, and the importance of their role varies greatly according to the kind of behavior in question. For example, institutions have a major impact on relatively narrow and highly formalized behavior such as voting turnout. Voting is an activity that engages the average citizen briefly once every four or five years and is highly amenable to institutional control. By simply changing the laws, for example, one can expand the electorate overnight to include women or 18-20-year-olds. Or by applying severe sanctions against nonvoting, one can produce extremely high rates of turnout. Thus, the ex-communist societies regularly reported voter turnout rates of 98 or 99 percent, and Albania may hold the world's record for electoral participation, with a reported turnout of 99.99 percent in one of Enver Hoxha's last elections. Although they were voting for a one-party slate, the regime got almost every living citizen to the polls. The point is that voting turnout is relatively easily manipulated by elites: it does not necessarily reflect any real choice or deep-rooted preferences on the part of the masses.
Stable democracy, by contrast, depends on a deeply rooted sense of legitimacy among the public. Simply making it illegal not to trust people or legally requiring everyone to be satisfied with their lives would not produce governmental legitimacy or a society of trust.
Trust and legitimacy are much more diffuse characteristics than voter turnout, and much less amenable to institutional manipulation. They reflect the entire historical heritage of the given society, with the political institutions being only one of many relevant factors. Similarly, economic growth does not seem to result from simply getting the right institutions: societies with a wide variety of institutions have failed to attain it. And conversely, high rates of eco-
BRINGING    THE   PEOPLE   BACK   IN
207
nomic growth have been achieved by societies with institutions ranging from democratic to authoritarian ones, with market economies or state-run economies, and with small-scale enterprises or huge industrial conglomerates.
The same is true of stable democracy: if it were simply a question of getting the right institutions, the world would be much nicer. One could simply xerox the U.S. Constitution and mail it out to all the governments of the world. Unfortunately, reality is not that simple: the fact that each society has a distinctive economic and social structure and cultural heritage can have a decisive impact on whether or not democracy survives in that society.
Thus, the former Soviet Union had one of the most democratic constitutions in the world (on paper), guaranteeing high levels of civil rights and political freedom, together with referenda, recall of judges, and other enlightened features. Great Britain, on the other hand, has no written constitution: the basic rules of democracy exist only as unwritten norms. But in the Soviet Union, the constitutional guarantees had no real effect, while in Britain, they were generally observed—with results that were as different as day and night. The current debate between advocates of an institutional approach and the advocates of a behavioral approach wrongly assumes that the two are separable. They are not. Formal institutions and political culture have a symbiotic relationship, with institutions becoming a behavioral reality only insofar as they become a part of the political culture.
An Institutional Determinist interpretation of these findings would argue that a society's level of interpersonal trust is determined by how long it has lived under democratic institutions. Our position, by contrast, is that interpersonal trust reflects a society's entire historical heritage, with its political institutions being merely one contributing factor. Although we lack sufficient time series data for a conclusive test, the institutional determinist model fails to hold up in those cases where we do have a substantial time series. For example, the peoples of Northern and Southern Italy have lived together under the same political institutions since unification 125 years ago. Nevertheless (as we have seen), Northern Italian society continues to show much higher levels of interpersonal trust than Southern Italy: clearly, these differences in trust levels reflect something other than the presence or absence of democratic institutions. The United States furnishes an even stronger refutation of the institutional determinist thesis. It is one of the oldest democracies in the world and shows relatively high levels of interpersonal trust (though by no means the highest in the world). But are these high levels of trust due to its democratic institutions? Apparently not—for trust in government has shown a sharp decline among the American public during the past few decades: in 1958, only 24 percent of the American public expressed distrust in the national government; in 1992, fully 80 percent expressed distrust. But this collapse of trust in government was not mirrored in a similar collapse of interpersonal trust which was relatively stable, declining only slightly. Interpersonal trust apparently moved on a different trajectory, which suggests that interpersonal trust was not determined by the American people's experience in the political sphere. Moreover, it is per-
208
CHAPTER    6
fectly clear that stable democracy does not necessarily produce high levels of trust. In the United States, it has actually been declining.
In addition to these empirical findings, there are theoretical grounds to doubt the institutional determinism model. To illustrate this point, let us turn to our other main variable, subjective well-being. There are clear reasons why democratic regimes cannot survive unless they are supported by the masses: if they are not, the public can simply vote them out of existence. The classic example was that of Weimar Germany, when Hitler came to power in 1933; this happened again most recently in Algeria, where the military took over to prevent a democratically elected Islamic Fundamentalist party from taking power; and it came close to happening in Russia in 1996 and could still conceivably happen there. But when we try to reverse the causal arrow, there is no obvious reason why democratic regimes would necessarily be more successful than authoritarian regimes in producing high levels of subjective well-being for their citizens. History indicates that they sometimes do and sometimes do not. In Germany, the Weimar regime apparently did not produce high levels of subjective well-being, but the Bonn regime did. Subjective well-being was higher under the authoritarian Soviet regime than it is under the current, more democratic, Russian regime. The World Values survey data show that in 1990 (the year before the collapse of the Soviet dictatorship), 33 percent of the Russian public were dissatisfied with their lives as a whole (scores of 1-4 on a 10-point scale). A 1995 survey of the Russian public replicated this question; it found that 51 percent of the Russian public—an absolute majority—were dissatisfied with their lives as a whole.5 Far from automatically producing subjective well-being, the experience of the Russian people with democracy so far has been linked with a decline in overall life satisfaction.
The Euro-Barometer surveys provide a less dramatic but more broadly based demonstration of the fact that living under democratic institutions does not automatically produce rising life satisfaction. As we saw in figure 6.4, subjective well-being levels were remarkably stable among established Western European democracies throughout the period from 1973 to 1995. More than 20 years of being democracies did not significantly raise their levels of subjective well-being. The evidence suggests that high levels of subjective well-being are a prerequisite for stable democratic institutions, rather than an automatic consequence of them.
Even apart from this strong empirical evidence, it seems highly unlikely that the extremely strong correlation (r = .82) that we find between subjective well-being and democracy is simply a consequence of having democratic institutions: to accept this interpretation, one would need to believe (1) that whether the masses are experiencing desperate misery or high levels of well-being has no impact on the survival of democratic institutions, but (2) that democratic institutions have almost magical powers to make people happy.
5 Results from a representative national survey of the adult population of the Russian Republic (N = 2,040), carried out by the Russian Institute of Public Opinion (ROMIR) in November 1995.
BRINGING    THE    PEOPLE    BACK   IN
209
The latter proposition simply is not plausible. To assume that a society's subjective well-being is determined by its political institutions is to assume that the tail is wagging the dog. Andrews and Withey (1976) have demonstrated that among the American public, subjective well-being is determined mainly by one's level of satisfaction with one's family life, one's marriage, job, home, friends, and leisure time—with politics making only a relatively minor contribution to overall subjective well-being. This accords with a large body of evidence that politics plays only a peripheral role in most people's lives.
Findings from the 1990 World Values Survey demonstrate that this holds true not only in America but in the world as a whole: politics is only of relatively minor subjective importance to most people. When asked how important various things were in their lives, the following percentages of the publics of our 43 societies rated the six following domains "very important":
1. 2.	Family Work	83% 59%
3.	Friends	38%
4.	Leisure	33%
5. 6.	Religion Politics	28% 13%
Politics ranked dead last, with only one person in eight considering it very important. Six times as many people emphasized the family, as emphasized politics. This may be dismaying to political scientists, but it seems to be a global reality. Politics was rated least important in almost every country.
Unlike totalitarian systems, democracies make only modest efforts to reshape their underlying cultures: the very essence of democracies is that they reflect the preferences of their citizens, rather than attempting to dictate them. It seems highly unlikely that the powerful correlation that we have found between culture and democracy exists because democratic institutions somehow create a new culture. Democratic institutions probably encourage feelings of interpersonal trust to some extent and may have some tendency to enhance subjective well-being, but the process seems to work mainly in the opposite direction: mass well-being and trust are crucial to the viability of democratic institutions.
Postmaterialist Values and the Future of Advanced Industrial Society
In advanced industrial society, prolonged prosperity and the welfare state contribute to an increasingly widespread sense that survival can be taken for granted, giving rise to another cultural factor conducive to democracy: the spread of Postmaterialist values. This is a relatively recent development, and it may be a major reason why, although early industrial societies were almost equally likely to mobilize their publics into democratic, fascist, or communist
210
CHAPTER    6
forms of political participation, advanced industrial society gives rise to democracy almost exclusively, rather than to either of the two other forms of modern political regimes.
Both theory and empirical evidence indicate that Postmaterialist values did not become a significant factor in politics until about two decades after World War II; clearly, Postmaterialist values cannot explain the stability of democracy throughout the period since 1920. Moreover, even today these values are widespread only in advanced industrial societies; they could not very well be the main factor explaining the levels of democracy found throughout the world. Nevertheless, there is reason to believe that they are-conducive to democracy in advanced industrial societies, contributing to a growing demand for higher levels of mass participation in politics.
Our theory implies that a shift toward Postmaterialist values should occur in any nation that develops high levels of economic security. As we have seen, this process seems to be at work not only in the West but also in East Asia (parts of which have now attained Western levels of prosperity) and even to some extent in Eastern Europe.
Postmaterialist values are conducive to democracy for three reasons: (1) They entail an emphasis on self-expression and participation that is inherently conducive to political participation, and, as we will see, Postmaterialists are relatively likely to act to attain democracy. (2) Postmaterialists view democracy as something that is intrinsically desirable—and not just as a possible means to become wealthy and successful. Thus, their support for democracy is more secure than that of Materialists, many of whom were initially attracted to democracy simply because it was associated with being rich. (3) In addition to their emphasis on participation and free speech, Postmaterialists tend to hold a wide range of basic democratic norms, as recent research demonstrates. Thus, Rohrschneider (1993) finds that Postmaterialist values are an important factor accounting for the presence or absence of democratic attitudes among the elites of both the former German Democratic Republic and the Federal Republic of Germany.
Research by Gibson and Duch in the former Soviet Union also demonstrates this point. Gibson and Duch (1994) developed a broad based scale of support for democratic values. This scale integrates seven subscales measuring (1) valuation of liberty, (2) support for democratic norms, (3) rights consciousness, (4) support for dissent and opposition, (5) support for independent mass media, (6) support for competitive elections, and (7) political tolerance. In their 1990 survey of the European Soviet Union, Gibson and Duch find that
Those who hold Postmaterialist values are markedly more likely to support these democratic values. In the European USSR data the percentage high in support for democratic values ranges from 14 percent for the Materialists to 80 percent for the Post-materialists, a truly remarkable difference. This is quite strong support for Inglehart's theory: Postmaterialists are much more likely to support core democratic values such as tolerance, competitive elections, etc. To what degree, though, are these findings
BRINGING    THE    PEOPLE    BACK   IN
211
spurious—that is, are they a function of other factors that contribute to the development of both democratic and Postmaterialist values? We can consider the effect of Postmaterialism on democratic values controlling for a number of attributes of the respondents that might well account for both sets of values. ... We find that Post-materialism has a substantial impact on democratic values beyond the effect of these demographic attributes. Those who are more highly educated, and who are younger are more likely to support democratic processes and institutions, and these variables alone can account for a respectable amount of variance in the dependent variable. Yet when the Postmaterialism indicator is added to the equation, an additional 10 percent of the variance can be explained. Even while controlling for age, education, social class, etc., Postmaterialists are considerably more supportive of democratic values than are Materialists. (Gibson and Duch, 1994: 20-21)
Since Postmaterialists emphasize individual freedom and self-expression, it is not surprising that Postmaterialist values correlate with democratic values, but Gibson and Duch's finding is far from tautologous. By demonstrating the linkages between democratic values and Postmaterialist values, Gibson and Duch help integrate democratic theory and the theory of value change. Their findings imply that we should find an intergenerational trend toward increasing support for basic democratic values in societies that experience economic growth and attain higher levels of mass security.
The Growing Role of Mass Political Action
Value change has important behavioral implications. In any setting, Postmaterialists are relatively ready to act to attain their political goals—but in authoritarian systems, they are the ones most likely to act in order to attain democracy.
Representative samples of the publics interviewed in the World Values survey were asked about their readiness to take part in four forms of political action: (1) joining in boycotts, (2) attending lawful demonstrations, (3) joining unofficial strikes, (4) occupying buildings or factories. Using similar questions in eight Western democracies, Barnes et al. (1979) demonstrated that Materi-alist/Postmaterialist values are strongly related to one's willingness to participate in unconventional political activities such as these, in order to press for some political goal.
The spread of citizen activism is not merely a Western phenomenon. Post-materialist values show the same linkages with unconventional political protest potential in Eastern Europe and East Asia as they do in Western countries. The data from the World Values surveys reveal that in country after country, Postmaterialists are two to four times as likely to engage in unconventional political action, as are Materialists. Figure 6.7 shows the percentage of the Russian public who said (in 1990-91) that they "have done" or "might do" all four of these activities.
J^Slá
CHAPTER    6
Materialist        Mixed    Postmaterialist
Figure 6.7. Readiness to undertake unconventional political action, by value type: Russia, 1990-91.
Percentage saying they "have done" or "might do" all four of the following: (1) join in boycotts, (2) attend lawful demonstrations, (3) join unofficial strikes, and (4) occupy buildings or factories. Source: World Values Survey carried out in Russia, December 1990-January 1991.
The spread of Postmaterialist values seems to be increasing the degree to which mass publics engage in elite-challenging political action. From 1981 to
1990, public readiness to engage in these activities became more widespread in the great majority of societies for which we have data, as we will see in chapter 10. In nondemocratic regimes, unconventional political action may play an even more important role than it does in the West: it may serve as the proximate cause by which the public obtains democratization. From Seoul to Warsaw to Budapest to East Berlin, mass participation in strikes, demonstrations, and boycotts—precisely the activities examined here—played a crucial part in the transitions to democracy launched throughout Eastern Europe in 1989, and in recent moves toward democratization in East Asia and Latin America.
People power has become an unprecedentedly important factor in politics. It proved its effectiveness again six months after these surveys, in August
1991,  when hard-liners in the Soviet Union attempted to seize power, arresting Gorbachev and rolling tanks into Moscow. But to widespread surprise, this time the Russian people did not resign themselves to authoritarian rule. Instead, citizens poured into the streets, defying the reactionary coup's leaders and building barricades around the Russian Parliament building where Yeltsin had organized resistance. Crowds of citizens brought armored columns to a halt. Miners went on strike. And entire units of tanks and paratroops went over to the resistance.
Both economic and noneconomic motives played a part in motivating mass resistance to communism. Its economic failures contributed to its downfall.
	25
0	20
Ol	
«	
	
c	15
Ü	
^,	
0	
0.	10
	5 n
BRINGING    THE    PEOPLE    BACK   IN
213
But it is equally important that the desire for freedom of speech and self-determination have become high-priority goals, for more people, than ever before in history. Postmaterialists are far likelier than Materialists to have taken part in the strikes, demonstrations, and other unconventional protest actions which brought down the communist regimes (or helped maintain the reform regime, in the Soviet case).
As the younger, better-educated, and more Postmaterialistic birth cohorts replace the older, less-educated ones in the adult population, we would expect elite-challenging political action to increase. Does it? In virtually all societies for which we have time series data, the answer is yes. In almost every country included in both waves of the World Values surveys, we find the predicted shift. The proportion of people who have actually done elite-challenging political actions during the past five years rose substantially from 1981 to 1990, as we will see below. One frequently reads journalistic accounts that mass publics have become politically apathetic, citing evidence that voter turnout has stagnated or declined. These accounts are accurate about voting, but miss the point that people display a rising potential for elite-challenging action. Voting turnout statistics convey a misleading impression of political apathy. Mass publics are becoming less likely to vote, which is a relatively elite-controlled form of participation; but throughout industrial society they are becoming more likely to engage in elite-challenging behavior.
The Impact of Mass Values on Democratization, Political Culture, and Stable Democracy
Do individual-level values have an impact on the societies in which people live? The evidence we have just examined suggests that democracy should be more likely to emerge (and survive) in societies with relatively large numbers of Postmaterialists than elsewhere. Is this the case?
The horizontal axis on figure 6.8 reflects the balance between Materialists and Postmaterialists in each country. The pattern is clear. Nations with relatively high proportions of Postmaterialists are much more likely to have had continuously functioning democratic institutions than other societies. Those with heavily Materialist publics tend to be not democratic, or to be recent (and possibly unstable) democracies. All but one of the countries that were not yet democratic in 1990 show scores below +10 on the Materialist/Postmaterialist values index on figure 6.8 (the sole exception being Mexico). All but one of the democracies had scores above that level (the sole exception being India).
One consequence of this cultural transformation is rising mass pressure for more democratic and participatory institutions. Although mass preferences alone do not determine when democratization takes place, there is a remarkably strong correlation between the ratio of Postmaterialists to Materialists and the existence of stable democracy (r = .71). Correlation is not causation. But the evidence of a causal link between Postmaterialist values and stable democ-
214
CHAPTER    6
Before 1918	
	1924
0)	
u	
£	
"35	1930
>.	
U) 3	1936
O	
3	
C	
*3	1942
C	
o	
ü	1948
U)	
c	
"E o	1954
'*s	
u	
3	1960
U.	
U)	
c	
o	1966
'■3	
3	
^	
'■t-» (A	1972
C	
O "43	1978
(S	
1».	
o	
o E	1984
Q	1990
After
		e    a«t    s s    ■•••	ese
		*  *«c  -g« oqoiic	*    <5   *.
-		\w\\\\\\	w
-		U|pan          lt|ly	
		"í"         ľ£rmany	
		France	
—		s	
\	_.		
		7/-7	
	.      /.,		
t		®*    ®^             e	
i!	nit 11	' v/ 1          1          1          1	
-a a	oaas    mase,		
-    s I	>    >		1        1
-20
+ 20
+ 40
+ 60
% "High" Minus % "Low" on Postmaterialist Index
Figure 6.8. Stable democracy and Materialist/Postmaterialist values. N = 43, r = .71, significant at .0000 level. Source: 1990-91 World Values Survey. Note: Respondents are classified as "high" on the 12-item Materialist/Postmaterialist values index used here if they gave high priority to at least three of the five Postmaterialist goals (ranking them among the two most important in each group of four goals). They are classified as "low" if they gave high priority to none of the five Postmaterialist goals.
racy goes well beyond cross-sectional evidence. It is reinforced by individual-level findings that suggest why countries with relatively Postmaterialist publics should be likelier to become stable democracies: (1) their publics give relatively high priority to individual freedom and to democratic values, and (2) their publics are relatively likely to engage in direct political action that can help bring a shift from authoritarian to democratic regimes.
When we entered Postmaterialist values into the regression analyses above, they showed positive but not significant linkages with stable democracy and high levels of democracy. These values are part of a highly intercorrelated syn-
BRINGING    THE    PEOPLE    BACK   IN
215
drome consisting of interpersonal trust, subjective well-being, a large tertiary sector, a highly educated population, and well-developed organizational networks. This syndrome is strongly linked with democracy; but Postmaterialism is not among the two or three independent variables that show significant effects. Thus far, it seems to have played only a secondary role in the emergence and spread of democracy. But if advanced industrial societies continue to produce an increasingly widespread feeling that survival can be taken for granted among a growing proportion of their populations, the political implications of Postmaterialist values may be far-reaching.
Conclusion
The evolution of industrial society makes democracy more likely. It brings gradual cultural changes that make mass publics increasingly likely to want democratic institutions and more supportive of them once they are in place. This transformation does not come easily or automatically. Determined elites, in control of the army and police, can resist pressures for democratization. But the emergence of prosperous welfare states leads to gradual long-term changes in which mass publics give an increasingly high priority to autonomy and self-expression in all spheres of life including politics. And as they mature, industrial societies develop increasingly specialized and educated labor forces, which become increasingly adept at exerting political pressure. It becomes more difficult and costly to repress demands for political liberalization. Moreover, economic development is also linked with relatively high levels of subjective well-being and interpersonal trust, which also seem to play a crucial role in democracy. With rising levels of economic development, cultural patterns emerge that are increasingly supportive of democracy, making mass publics more likely to want democracy, and more skillful at getting it.
Although rich societies are much likelier to be democratic than poor ones, wealth alone does not automatically bring democracy. But the process of industrialization does have an inherent tendency to produce changes that are conducive to democracy. In the long run, the only way to avoid the growth of increasingly articulate and effective mass demands for democratization would be to reject industrialization. Very few ruling elites in the contemporary world are willing to do so. Those societies that do move onto the trajectory of industrial society will eventually face increasingly powerful pressures for democratization.
Our findings suggest that political culture plays a much more crucial role in democracy than the literature of the past two decades would indicate. Although it does not seem to be the immediate cause of the transition to democracy, political culture does seem to be a central factor in the survival of democracy. In the long run, democracy is not attained simply by making institutional changes or through clever elite-level maneuvering. Its survival also depends on what ordinary people think and feel.
236
CHAPTER    7
25 cases, only this variable shows a statistically significant relationship, but it seems unlikely that it is the only relevant cultural variable. Postmaterialist values may also play an important role in shifting emphasis away from thrift and saving, though their effect tends to be confounded with the fact that they are found in relatively wealthy countries, which tend to have low growth rates for a variety of reasons. Also potentially important, though they do not show statistically significant results in this analysis, are a society's associational networks—which seem to play contrasting roles in developed and developing societies.
Since we do not yet have time series data on these values, we cannot reach definitive conclusions about the causal relations between culture and economic growth. An alternative interpretation to our own would be that high rates of economic growth somehow give rise to a culture that emphasizes thrift and determination. This is conceivable. But there is an obvious logic to the hypothesis that cultures that emphasize thrift and determination should tend to show high growth rates: thrift makes high investment rates possible. If we try to turn the causal arrow around, however, there is no obvious reason why rapid growth would bring increasing emphasis on thrift: quite the contrary, one would expect it to give rise to higher rates of spending. Until we have cultural time series data, we cannot regard the matter as settled, but the evidence strongly suggests that certain cultural values play an important role in economic growth. The question "Is economic growth due to cultural factors or to economic factors?" misses the point. Cultural factors are intimately linked with economic factors; and they provide a strong explanation of why, over the long term, some societies have shown much higher rates of economic growth than others. Both the encompassing tests and the structural equation models demonstrate that a model that includes both cultural factors and economic factors has a significantly better fit and explains more of the variance than does a model that relies on economic variables alone.
Economic theory has already begun to incorporate social norms and cultural factors into its models (Cole, Malaith, and Postlewaite, 1992; Fershtman and Weiss, 1993). The logical next step is to determine how cultural and motivational factors can be used to augment existing economic models in order to gain a better understanding of economic growth.
The collapse of the Soviet economy illustrates how costly it can be in the long run to refuse to consider the importance of individual incentives and motivations. Our results indicate that both cultural and economic factors are crucial to economic growth. Neither supplants the other. Future research will be best served by treating the two types of explanation as complementary.
CHAPTER     8
The Rise of New Issues and New Parties
The goals of both individuals and of societies are changing as a result of the diminishing marginal utility of economic growth. This is changing the political agenda of advanced industrial societies, giving rise to new issues, new political movements, and new political parties. This chapter examines how this is happening at both the individual and societal levels.
Changing Values and a Changing Political Agenda
The shift toward Postmodern values has brought a shift in the political agenda throughout advanced industrial society, moving it away from an emphasis on economic growth at any price, toward increasing concern for its environmental costs. It has also brought a shift from political cleavages based on social class conflict toward cleavages based on cultural issues and quality of life concerns. Huntington (1994) has gone so far as to argue that the main basis of global political conflict from now on will no longer be economic or ideological issues, but cultural issues: world politics will revolve around a "Clash of Civilizations." While this projection may be overdrawn, there is no question that ethnic and cultural issues are becoming more prominent. Economic conflicts are likely to remain important. But, while in the past they dominated the scene to such a degree that many influential thinkers accepted the Marxist view that economics was virtually the whole story, today this seems less plausible. Economic conflicts are increasingly sharing the stage with new issues that were almost invisible a generation ago: environmental protection, abortion, ethnic conflicts, women's issues, and gay and lesbian emancipation are heated issues today—while the central element of the Marxist prescription, nationalization of industry, is almost a forgotten cause.
As a result, a new dimension of political conflict has become increasingly salient. It reflects a polarization between modern and postmodern issue preferences. This new dimension is distinct from the traditional Left-Right conflict over ownership of the means of production and distribution of income. Its growing salience is transforming the meaning of Left and Right and changing the social bases of Left and Right. Historically, the Left was based on the working class and the Right on the middle and upper classes. Today, increasingly, support for the Left comes from middle-class Postmaterialists, while a new Right draws support from less secure segments of the working class. A new Postmodern political cleavage pits culturally conservative, often xenophobic, parties, disproportionately supported by Materialists, against change-oriented
238
CHAPTER   s
parties, often emphasizing environmental protection, and disproportionately supported by Postmaterialists.
Throughout most of the twentieth century, it was generally agreed that support for more state intervention in the economy was the crucial distinction between Left and Right. From a Marxist perspective, private ownership was the root problem, and nationalization of industry and state control of the economy constituted the core solution to all social problems. Abolishing private ownership of the means of production, it was thought, would eradicate exploitation, oppression, alienation, crime, and war.
Although they called themselves "liberals," the American Left also tended to view more state regulation and control of the economy as inherently good: liberals were those who supported a growing role for the state; conservatives were those who opposed it. Well into the 1970s, Western political elites continued to define the meanings of "Left" and "Right" in terms of state intervention in the economy and society (see Aberbach et al., 1981, 115-69).
This consensus has dissolved. It no longer seems self-evident that more state authority constitutes progress, even to those on the Left. One of the key developments of recent years has been a growing skepticism about the desirability and effectiveness of state planning and control, a growing concern for individual autonomy, and a growing respect for market forces. In recent years this outlook has been endorsed not only by conservatives but also by growing segments of the Left. As early as the 1960s, New Left groups emerged in the West that were highly critical of big government, viewed bureaucracy as dehumanizing, and called for devolution of decision-making power to local communities and to those directly affected by the decisions.
In an even more dramatic change, post-socialist Eastern European governments have been drastically reducing the role of the state. And in China, the pragmatists who came to power after the death of Mao, though still nominally communists, have been allowing more and more scope for individual enterprise and an increased role for market forces, though continuing to repress political pluralism. The last attempt to apply the classic policies of the Left in a major Western nation occurred in 1981, when a socialist-communist coalition won office in France. After two years of unrewarding experience with nationalization of industry and other traditional policies of the Left, the socialists abandoned the classic Marxist approach and shifted to market-oriented policies. Similar shifts toward market economies have been occurring in Asia, Africa, and Latin America, even in states led by elites who were shaped by Marxism. Today, almost no one views nationalization of industry as a panacea. While the economic Left-Right dimension still exists, its meaning has changed radically.
The transition from a state-run economy to a market economy often entails traumatic costs. In a number of ex-communist societies, this has brought former communist elites back into power under new labels. But even in these cases, there has not been a return to a Soviet-style state-run economy; instead, the policies of former communist elites have generally been limited to slow-
fi B W   ISSUES   AND    NEW   PARTIES
239
ing down the pace of change and attempting to soften its shock. Where change is occurring in this area today, it is predominantly a movement toward privatization of former state functions. The Right consists of those who are pushing for faster or more widespread privatization; the Left consists of those who resist privatization or urge that it be done more slowly.
Russia is an exception: here, in a reversal of meanings, the "Left" label now designates the reformers who are seeking a market economy with private ownership. Russia is torn between reformers on the Left, seeking a greater role for individual initiative and individual self-expression, and the still-entrenched Nomenklatura on the Right, clinging to power and privilege based on their control of the economy. Well into the 1990s, Russia was trapped between two eras, with its state-run economy collapsing but with its market economy not yet fully established. Halfway measures to move toward a more open but more competitive and less predictable society brought suffering and insecurity to a large part of the Russian people. In the 1993 elections to the Russian parliament, a majority of seats were won by a coalition of former communist hardliners and xenophobic protofascists. The future of the reform movement, and of democracy itself, is uncertain there, but the old lines of confrontation have changed irrevocably.
This change in orientations toward state authority can be analyzed on two levels. At the individual level these changes reflect the Postmodern shift in basic values; and at the societal level, they reflect the fact that the expansion of the state has reached a point of diminishing returns. The two developments are mutually supportive.
In the ex-socialist countries, overexpansion of the state eventually paralyzed innovation and economic growth, bringing their economies to the point of collapse. In the West, the problem is more limited; economic growth continues, but the welfare state is in crisis. Paradoxically, this crisis does not reflect the failure of the welfare state so much as the fact that it has succeeded in alleviating those problems it can most readily solve—and thereby helped pave the way for new types of problems to become central. The expansion of the welfare state tempered the ruthless exploitation of laissez-faire capitalism, helping it evolve into a stabler and more viable form of society. Today, in contrast with previous history, the masses do not starve even in times of severe economic decline; their standard of living has been stabilized at a modest level of economic security, reducing social class tensions. This helps explain why—in contrast to the widespread political extremism that arose during the Great Depression of the 1930s—Western nations' politics remained on a relatively even keel during the recent recessions, even though unemployment in some countries exceeded the levels experienced during the Great Depression.
But the growth of the welfare state has begun to reach its limits. When government expenditures exceed 55 percent of gross national product, as is now the case in many Western societies, there is little room for further expansion; taxation becomes massive, and the majority of the public feels the burden.
240
CHAPTER    8
Individual-Level Changes: The Postmodern Shift and the Rise of the Postmodern Politics Cleavage
The goals of both individuals and of societies are changing as a result of the diminishing marginal utility of economic growth. In this respect, cultural change behaves as if it were a rational response to the changing physical and socioeconomic environment. But culture exists in the minds and feelings of given peoples. Accordingly, it changes only insofar as what people learn and experience reshapes prevailing beliefs and values. It can be analyzed at both the individual and societal levels, which are simply two sides of the same coin. The rise of a new axis of political cleavage started with changes in the values of individuals—which then brought new issues such as abortion, environmental protection, and women's issues to a central place in the political arena. Only gradually and a good deal later did these changes reshape political cleavages and lead to the emergence of new political parties. Long-established institutions have considerable inertia and are slow to change.
For most of the twentieth century, the dominant axis of political cleavage was the Left-Right polarization based on economic issues, with the working class supporting the Left and the middle class supporting the Right. In his 1960 classic Political Man, Lipset correctly described this polarization as the most important single fact about political cleavages throughout the industrial world. In a predominantly materialistic world, conflict over income and ownership of the means of production was the central issue.
But significant numbers of Postmaterialists moved into political relevance, as the postwar generation began to reach adulthood. Postmaterialists first became visible as student protesters, during the 1960s, bringing a variety of new issues into the political arena. At that point in time their values differed sharply from those of the dominant establishment; they were outsiders and invented a whole repertory of (then) unconventional protest tactics to bring their goals to national notice. But as they reached maturity and began to occupy positions of power, Postmaterialists adopted new strategies. By the 1980s, they were becoming powerful within established political parties, or were founding successful political institutions of their own. As Postmaterialist elites took over established institutions, political extremism became less and less associated with the Left, and increasingly a tactic used by the Right.
For most of the past three decades, Postmaterialists have dominated the political agenda in most Western democracies: overwhelmingly, the new issues that were introduced during the 1960s and 1970s reflected Postmodern priorities. It was only recently that the Right staged a counterattack, often utilizing the same political techniques that the Postmaterialists had introduced during the 1960s, when they were a relatively powerless minority.
The Modern/Postmodern dimension described in chapter 3 reflects the wide array of new issues that have become prominent with the rise of Postmodern politics: these issues range from abortion to cultural change and ethnic diversity as figure 3.2 demonstrated. Although a variety of issues became salient
N E W	ISSUES   AND    NEW   P A t				TIES								
	70 60												
													
													
	>_   E												
	O   O												
	*■   *3												
	t; ° ^n												
	Q. O Q, *-3 Q- 40												
		_											
	w ä .    a> 30 O) g					rwvwrwm							
	—    O												
	X ±  20												
	=    >												
	o" LU												
	10 n												
241
0            1             2            3            4            5
Number of Posímaíerialist Goals Given High Priority
Figure 8.1. Support for environmental protection, by Materialist/Post-materialist values in five advanced industrial societies. Source: 1990-93 World Values Survey data from United States, Britain, France, West Germany, and Sweden (N = 7,473). Note: Respondents are classified as "high" on the environmental protection index if they (1) agreed with both of the following: (a) "I would give part of my income if I were certain that the money would be used to prevent environmental pollution," and (b) "I would agree to an increase in taxes if the extra money were used to prevent environmental pollution," and (2) they also disagreed with both of the following: (a) "The government should reduce environmental pollution but it should not cost me any money," and (b) "Protecting the environment and fighting pollution is less urgent than often suggested."
with the emergence of Postmodern culture, the central issue initially was the Peace Movement. But as the war in Vietnam receded into the past, environmental causes became the flagship issue. Throughout advanced industrial society (though not necessarily in developing societies), Postmaterialists are far more favorable to environmental protection than are Materialists, as figure 8.1 demonstrates.
Postmaterialist Values and Environmental Attitudes
The rise of Postmaterialist values helps account for the spectacular rise in the salience of environmental issues which has taken place during the past two decades. Postmaterialism became a significant political force during the past 25 years, as the postwar generation emerged into political relevance. Shortly afterward, environmental concerns took on an unprecedented salience throughout advanced industrial society.
242
CHAPTER    S
Postmaterialist goals are not the only factor motivating concern for the quality of the environment. In advanced industrial society, environmental protection is primarily a Postmaterialist concern; but in many developing countries, from China to Mexico, air pollution and water pollution levels are far worse than in advanced industrial societies, posing immediate problems to health. In such settings, environmental protection is not a quality of life issue, but a matter of survival; it is as likely to be supported by Materialists as by Postmaterialists. The highest levels of support for environmental protection, however, arc found in the Nordic countries and the Netherlands—which have the most Postmaterialist publics in the world (Inglehart, 1995).
Figure 8.1 shows the relationship between Materialist/Postmaterialist values and support for environmental protection in advanced industrial societies. This figure uses a 12-item battery in which the following five items tap Post-materialist priorities across virtually all 43 societies included in the World Val -ues surveys (see Abramson and Inglehart, 1995):
Protecting freedom of speech
Giving people more say in important government decisions
A less impersonal, more humane society
Giving people more say on the job and in their communities
A society in which ideas count more than money
A given individual may choose anywhere from zero to all five of these items among his or her high-priority goals.
In advanced industrial societies, these values are strongly related to support for environmental protection: as figure 8.1 demonstrates, among those who give high priority to none of the Postmaterialist goals, only 29 percent rank high on support for environmental protection; among those who give high priority to all five Postmaterialist goals, fully 68 percent rank high on support for environmental protection. This relationship has impressive strength across advanced industrial societies, especially considering the fact that none of the five Postmaterialist items makes any direct reference to environmental concerns.
It is relatively easy to give lip service to environmental protection, and many people do so. Do these attitudes have behavioral consequences? The relatively favorable attitude of Postmaterialists toward environmental causes is not just a matter of lip service: their behavior reflects their distinctive values to an even greater extent than do their attitudes. Although Postmaterialists are only about twice as likely as Materialists to favor environmental protection, they are four to 10 times as likely to be active members of environmental protection groups. And Postmaterialists are four to six times as likely to vote for environmentalist parties (in countries that have them) as are Materialists. Figure 8.2 shows the evidence from four Western societies.
The Materialist/Postmaterialist dimension has become the basis of a major new axis of political polarization in Western Europe, leading to the rise of the Green Party in West Germany, and to a realignment of party systems in a number of other countries (Inglehart, 1977, 1990; Dalton, Flanagan, and Beck,
NEW   ISSUES   AND    NEW   PARTIES
243
c & a
(3
30
25 h

20
c o.2
>  tí 2 —
ľS LU  ne
o *- lo
>    (U
>z
®   e-H U) C
■>
(S
10h
_	
-	•    West ; Germany e @
—	a e /   s Sweden i / ijy* France
-	/        <ŕ                  ^»
	\^=^~  i              i              i              i              i
0               1               2              3              4              5
Number of Postmaterialist Goals Given High Priority
Figure 8.2. Intent to vote for environmentalist political parties, by Materialist/ Postmaterialist values in four countries having such parties. Source: 1990-93 World Values Survey.
1984). During the 1980s, environmentalist parties emerged in West Germany, the Netherlands, Belgium, Austria, and Switzerland. In the 1990s they made breakthroughs in Sweden and France and are beginning to show significant levels of support in Great Britain. In every case, support for these parties comes from a disproportionately Postmaterialist constituency. As figure 8.2 demonstrates, as we move from the Materialist to the Postmaterialist end of the continuum, the percentage intending to vote for the environmentalist party in their country rises steeply: from 0 to 8 percent in Britain, from 2 to 17 percent in France, from 4 to 20 percent in Sweden, and from 2 to 27 percent in Western Germany. Pure Postmaterialists are five to 12 times as likely to vote for environmentalist parties as are pure Materialists. Environmentalist parties are not yet strong enough to govern independently and may never be so; but they have successfully advocated environmental protection policies in each of these countries and have forced the established parties to adopt stronger environmental protection policies in order to compete for their voters.
A New Axis of Political Cleavage: Postmodern versus Fundamentalist Values
Although it is more difficult to change long-established institutions than individuals' attitudes, environmentalist parties have begun to emerge in many so-
244
CHAPTER    8
cieties in which the electoral system does not tend to strangle new parties. Why? The environmentalist cause is only one of many Postmodern issues favored by Postmaterialists. This electorate is distinctive in its entire worldview: they are relatively favorable to women's rights, disabled groups, gay and lesbian emancipation, ethnic minorities, and a number of other causes. But the environmental cause has emerged as the symbolic center of this broad cultural emancipation movement: while many of the other Postmodern causes tend to be divisive, practically everyone likes clean air and green trees. Although these parties reflect an entire worldview, environmental symbols captures the issue on which they have the widest potential appeal.
Nevertheless, the rise of Postmaterialist causes has given rise to negative reactions from the very start. The French student protest movement was able to paralyze the entire country in May 1968; but it led to a massive shift of working-class voters, who rallied behind De Gaulle as the guarantor of law and order, giving the Gaullists a landslide victory in the June 1968 elections. In the same year, student protesters in the United States were able to bring down Lyndon Johnson, but they alienated much of the traditional Democratic Party electorate—many of whom threw their support to a reactionary candidate, George Wallace, enabling Richard Nixon to win the presidency. The 1972 elections were something of a replay, except that this time normally Democratic voters who were repelled by the seeming radicalism of the McGovern campaign supported Nixon: for the first time in history, white working-class voters were about as likely to vote for the Republican as for the Democratic candidate. The aftermath of these events transformed the two parties, but the United States still has a two-party system, with the same party labels as before: superficially, the system seems unchanged.
Although Postmaterialist-led parties emerged in both the Netherlands and Belgium during the 1970s, West Germany was the scene of the first breakthrough by an environmentalist party in a major industrial nation. Postmaterialist protest had manifested itself as dramatically in Germany as in the United States or France, but it was only in 1983 that the Greens were sufficiently strong and well organized to surmount West Germany's 5 percent hurdle and enter the West German parliament—bringing a significant structural change to West German politics. But more recently, the Greens have been countered by a Republikaner party characterized by cultural conservatism and xenophobia. In the 1994 national elections, the Greens won 7 percent of the vote. The Republikaner, on the other hand, were stigmatized as the heirs of the Nazis and won only 2 percent of the vote, too little to win parliamentary representation. Nevertheless, xenophobic forces have already had a substantial impact on German politics, motivating the established parties to shift their policy positions in order to co-opt the Republikaner electorate. These efforts even included an amendment to the German constitution: to cut down the influx of foreigners, the clause guaranteeing free right of political asylum was revised in 1993, in a decision supported by a two-thirds majority of the German parliament. The rise of the Green Party in Germany has also had a major impact even
NEW   ISSUES   AND    NEW   PARTIES
GREENS
245
PARTY OF
DEMOCRATIC -
SOCIALISM
SOCIAL
DEMOCRATS
FREE DEMOCRATS
. CHRISTIAN DEMOCRATS
REPUBLIKANER
Figure 8.3. The social class-based Left-Right dimension and the Postmodern politics dimension in Germany.
though only a small portion of the electorate votes for it. For the Greens, like other New Left parties and movements, reflect an entire worldview which differs fundamentally from that of the traditional Left. Despite their name, the Greens are much more than an ecological party. They seek to build a basically different kind of society from the prevailing industrial model. During the Cold War, their most massive demonstrations were directed against nuclear weapons and NATO. They have actively supported a wide range of Postmodern causes, from unilateral disarmament to women's' emancipation, gay and lesbian rights, rights for the physically disabled, Palestinian liberation, and citizenship rights for non-German immigrants. But their greatest impact on German politics has been in forcing the established parties, from the Christian Democrats to the Social Democrats, to adopt pro-environmentalist positions in order to compete for the Greens' voters.
The Greens and the Republikaner are located at opposite poles of a new political dimension, as figure 8.3 suggests. If we simply judged by their labels, this might not seem to be the case: the Republikaner do not call themselves the Anti-Environment Party; nor do the Greens call themselves the Pro-Immigrant Party. But, in fact, their constituencies are disproportionately Materialist and Postmaterialist, respectively; and these parties adopt opposite policies on the relevant issues. The older parties are arrayed on the traditional Left-Right axis, established in an era when political cleavages were dominated by social class conflict. On this axis (the horizontal dimension of figure 8.3) both elites and mass electorates place the Party of Democratic Socialism (the Eastern German
ex-communists) on the extreme Left, followed by the Social Democrats and the Free Democrats, with the Christian Democrats at the Right of the spectrum. This figure is schematic. As Kitschelt (1995) has demonstrated, the new politics dimension is not perpendicular to the long-established Left-Right dimension. Instead, the Greens are closer to the old Left on key issues, while the Republikaner are closer to the Right. But, although both elites and masses tend to think of the Greens as located on the Left, they represent a fundamentally new Left. Traditionally, the Left parties have been based on a working-class constituency and advocated a program that called for nationalization of industry and redistribution of income. In striking contrast, the Postmaterialist Left appeals primarily to a middle-class constituency and is only faintly interested in the classic program of the Left. For example, Postmaterialists are not necessarily more favorable to state ownership than are Materialists, as figure 8.12 indicates. But Postmaterialists are intensely favorable to the Left position on Postmodern issues—which frequently repel the traditional working-class constituency of the Left.
The vertical axis on figure 8.3 reflects the polarization between Postmodern and Fundamentalist values, reflecting differences in people's subjective sense of security. At one end, we find a Postmodern openness to ethnic diversity and changing gender roles; at the opposite pole we find an emphasis on familiar values (often rooted in traditional religion) in the face of insecurity. This cleavage tends to pit the Postmaterialists against those with traditional religious values. Although the classic interpretation of secularization attributed it to the cognitive spread of a scientific worldview, we have argued that the rise of a sense of security among mass publics of advanced welfare states is an equally important factor in the decline of traditional religious orientations. The cognitive interpretation implies that secularization is inevitable and more or less irreversible. By contrast, the rise of a sense of security among mass publics is far from inevitable and can be undermined by economic decline or rapid change. Fundamentalist movements continue to emerge among the less secure strata of even the most advanced industrial societies, with people reemphasiz-ing traditional values in times of stress.
As figure 8.4 demonstrates, across five advanced industrial societies, 70 percent of the pure Materialists support a policy of reverse affirmative action— that is, the position that "When jobs are scarce, employers should give priority to [one's own nationality] over immigrants." Among the pure Postmaterialist type, only 25 percent are in favor of giving preference to native-born citizens.
Figure 8.5 presents a similar comparison, based on the proportion saying that they would not like to have immigrants or foreign workers as neighbors: 19 percent of the pure Materialists take the xenophobic position, as compared with only 3 percent of the pure Postmaterialists.
On this issue, value priorities have even more impact than they do on the environmental protection issue: Materialists are almost three times as likely as the Postmaterialists to favor employment discrimination favoring the native-
5^ 4^1	70
	
■™    ^	
o- £	60
tu  U)	
.ä f	
o F	50
TS —	
	
5 «	
9. >	40
■*= o tfí ľ.	
..   >	
(A   +■<	
	30
>» c	
O   O	
	
	20
P'i	10
.£ £	
> K-	
«5        Q C/5  S	0
-	:			...	I----------i	
S?                              0                12                3                4                5
Number of Postmaterialist Goals Given High Priority
Figure 8.4. Support for giving preference to one's own nationality over immigrants when jobs are scarce, in the United States, Britain, France, West Germany, and Sweden. Source: 1990-93 World Values Survey. Note: The question was "Do you agree or disagree with the following statement: When jobs are scarce, employers should give priority to [one's nationality] over immigrants."
30
CD
ňí
m  25
1*.
o °
Z s.
2-SP 20
o2 t>   15
»a
f c
r- (S
o>ro 10
'5* £
(B C
u> —   5
ď-
0            12            3            4            5
Number of Postmateria!ist Goals Given High Priority
Figure 8.5. Rejection of immigrants as neighbors, in United States, Britain, France, West Germany, and Sweden. Source: 1990-93 World Values Survey. Note: The question was: "On this list are various groups of people. Could you please sort out any that you would not like to have as neighbors? Just call off the letters, please." The list included 15 groups, such as "heavy drinkers" or "homosexuals," with "immigrants/foreign workers" as one item.
248
CHAPTER    8
born over foreigners, and six times as likely to say they would not want to have foreigners as neighbors.1
Like Materialist/Postmaterialist values (with which it is linked), our measure of xenophobia shows strong differences across age groups, with the younger birth cohorts being more tolerant of foreigners, homosexuals, and other outgroups. But this holds true only in societies that have experienced rising security: in societies that have not experienced economic growth, the young are not more tolerant than the old. In low-income societies that have not experienced economic growth, the young are actually less tolerant of outgroups than are the old.
A Postmodern Politics axis has also taken shape in other countries, such as France, where an Ecologist Party has recently emerged at the Postmaterialist pole, and the xenophobic National Front at the other. In contrast to Germany, where the Republikaner are unlikely to surmount the 5 percent hurdle, in France's 1993 parliamentary elections, the National Front won 12 percent of the vote. Reflecting a pervasive decline of the traditional Left in the early 1990s, the French Socialist Party won only 18 percent of the vote, and the communists won only 9 percent. Meanwhile, the Ecologists got 8 percent of the vote, the strongest performance they had ever made in elections to the National Assembly. Throughout the postwar era, the communists had been the strongest party in France. In 1993 they were outpolled by the National Front and came in only slightly ahead of the Ecologists. Figure 8.6 depicts the alignment of French parties on the two respective dimensions of political cleavage.
The once-dominant Left-Right dimension based on social class and religion is increasingly sharing the stage with a Postmodern politics dimension. Although support for environmentalist parties has grown in many Western societies, there has also been a right-authoritarian reaction at the opposite pole of the Postmodern Politics dimension. Right-wing extremist parties, such as Le Pen's National Front, have been gaining votes by appealing to antiforeign sentiments. This appeal has been particularly effective among blue-collar workers who formerly voted for parties of the Left.
The social base of such parties consists disproportionately of economically and psychologically marginal segments of society, manifesting a reaction of the insecure in the face of change. Parties of cultural autonomy, on the other hand, are not necessarily xenophobic and sometimes have a very cosmopolitan outlook: thus, though they emphasize a specific cultural identity, the Flemish and Catalan "nationalists" are actually more favorable to European integration than are most of their compatriots; and the Quebecois are more
1 The "reject neighbors battery" referred to here was included in both waves of the World Values Survey, but the data are not reliable for cross-time comparisons. This battery used a scaleless format, which is very sensitive to context and interviewer effects. Moreover in 1981 it had an interviewer instruction to "code all mentions," while in 1990 each item had codes for "mentioned" or "not mentioned," with the interviewer instruction "Check a response for each item"; this attracted more choices for all items. In addition, "People with AIDS" was added to the battery in 1990.
NEW   ISSUES    AND    NEW   PARTIES
ECOLOGISTS
249
COMMUNISTS ■
SOCIALISTS •
•UDF-
■RPR
NATIONAL FRONT
Figure 8.6. The social class-based Left-Right dimension and the Postmodern politics dimension in France.
supportive of North American free trade than are Canadians in general. These parties are motivated not by xenophobia so much as by a concern for cultural identity and for autonomy in decision making, and their social base consists disproportionately of the young, the well-educated, and Postmaterialists.
In the Netherlands—one of the most Postmaterialist societies in the world— Postmodern parties have been making growing inroads for more than two decades. In the 1994 Dutch parliamentary elections, two heavily Postmaterialist parties—Democrats '66 and the Green Left—won nearly 20 percent of the vote; while at the opposite pole of the Postmodern politics spectrum, several small fundamentalist religious parties won over 5 percent of the vote.
The ex-communist resurgence in former socialist countries reflects the fact that these parties are associated with the Good Old Days of relative stability and security in the minds of their voters. But parties tied to the classic program of the Left have been faring poorly in Western countries. This is partly due to the Postmaterialist shift, and partly due to the loss of legitimacy of the economic philosophy of socialism that accompanied the collapse of Marxism in Eastern Europe and the former Soviet Union. It also reflects the fact that the electoral appeal of the long-established Western parties was based on class-based economic issues, which have a diminishing ability to mobilize voters today.
For most of the postwar era, the Italian Communist Party won 30 to 35 percent of the vote in Italian elections, but starting in the late 1970s, it went into a steady decline. As Italy's 1994 elections approached, the situation seemed to offer a golden opportunity for the communists to stage a comeback. The Christian Democratic coalition, which had dominated Italian politics throughout the
250
CHAPTER    8
period since World War II, had finally self-destructed. Most of the top leaders of both the Christian Democrats and their socialist allies were either in jail or under indictment for flagrant corruption—a classic opportunity for the opposition to come to power. In the 1994 elections, the Christian Democratic vote plummeted to 11 percent and that of their socialist allies fell to 2 percent. But the communists were unable to capitalize on this opportunity. The Democratic Party of the Left won only 20 percent of the vote, and the hard-line communists won only 6 percent. The vacuum created by the collapse of the governing coalition was filled by new parties, some of which appeared almost overnight. The traditional Left-Right axis sunk to an unprecedented low point. The leading party was the newly established Forza Italia, with 21 percent of the vote; a regional party, the Northern League, won 8 percent; and the neo-fascist vote rose to an appalling 14 percent, while at the other pole of the Postmodern politics dimension, the Greens won 3 percent. In Italy's 1996 elections, a reform Communist Party (renamed as the Democratic Party of the Left) emerged as the largest party in parliament, but it was a profoundly transformed party, dedicated to a market economy and to Italian membership in NATO.
In Britain's 1992 general elections, the Left was in an ideal position to win. The Conservatives had been in power for three consecutive terms, and the British economy was in the throes of a deep recession; moreover, the Conservative Party had an unappealing leader who ran a dull campaign. Nevertheless, Labour lost for a fourth consecutive time. The party was widely seen as still committed to old-line policies of the Left, such as state ownership of business and industry—which it was still officially endorsing even after it had been abandoned within the ex-socialist bloc. A succession of Labour Party leaders tried to drag the party back to the mainstream, but it was only in 1995, under the leadership of Tony Blair, that the party finally officially abandoned the goal of state ownership of business and industry. With its return to the mainstream, the prospect of a Labour electoral victory was finally within reach.
The rise of Postmodern values has not led to the emergence of new parties in societies like the United States, where the absence of proportional representation makes it difficult for new parties to survive. Nevertheless, it has forced the existing parties to reposition themselves. Both major parties now claim to be pro-environmentalist, and both parties are trying to find just the right balance between cultural permissiveness and traditional family values. The success of Clinton in 1992 owed much to a skillful balancing act within the future First Family. Clinton himself, a Southern WASP male, took positions on social and economic issues that were almost indistinguishable from those of the Republicans, promising a middle-class tax cut and a balanced budget, while his wife made a subtly differentiated appeal to the Postmodern constituency. Although Postmodern political parties have not emerged in the United States, it is clear that the issues that launch such parties are as powerful here as anywhere. Concern for environmental protection has a large and active constituency; but opposition to illegal immigration also has broad and in-
N E W   ISSUES   AND    NEW   PARTIES
251
creasingly articulate support. In the United States, Postmodern politics plays itself out within the two long-established dominant parties.
In France, as we have seen, the National Front became prominent by appealing to nativist sentiment among working-class voters who formerly supported the parties of the Left. Its nearest parallels on the American scene are the Christian Coalition and the anti-immigration movements. These movements represent reactions against rapid cultural change which has been occurring throughout advanced industrial society. During the past 25 years, divorce rates rose by as much as 300 percent in Western societies, while during the same period, fertility rates fell to well below the population replacement rate. Similarly, a generation ago homosexuality was something that was only whispered about. Today, gay and lesbian groups are officially organized under government and university sponsorship and are beginning to obtain legal protection of the right to follow their own sexual orientations. This change is part of a broad intergenerational cultural shift. As we have seen, younger groups are far more permissive toward divorce, homosexuality, and abortion than older groups and place much less emphasis on having children.
In addition to these cultural changes, massive immigration flows, especially those from Third World countries, have changed the ethnic makeup of most advanced industrial societies. The newcomers speak different languages and have different religions and lifestyles from those of the native population— further compounding the impression that the culture one grew up in is being swept away. The rise of militant religious fundamentalism in the United States, and of xenophobic movements in Western Europe, represents a reaction against rapid cultural changes that seem to be eroding some of the most basic values and customs of the more traditional and less secure groups in these countries. The emergence of highly visible New Right groups has led some observers to conclude that they reflect the mainstream trend. They are important phenomena—but they do not represent the wave of the future. On the contrary, New Right groups are a reaction against broader trends that are moving faster than these societies can assimilate them. This reaction against cultural change has reinforced the Postmodern politics cleavage, pitting predominantly Materialist-oriented parties against Postmaterialist parties and giving rise to a Postmodern versus Fundamentalist cleavage dimension.
The foregoing interpretation is supported by empirical analyses by some leading scholars of comparative politics. Thus, Knutsen (1989, 1995) has demonstrated that in most Western European countries, a new dimension of political cleavage has emerged, which he describes as a Materialist/Postmaterialist values cleavage; tapping a number of issues such as environmentalism and nuclear power, the core variable in this cluster is Materialist/Postmateri-alist values. This new dimension cuts across the traditional Left-Right dimension and has become an increasingly important influence on party choice in many societies—and has become the most important variable shaping political cleavages in some countries.
252
CHAPTER    8
Similarly, in an insightful analysis of the rise of Postmodern parties in Western democracies, Kitschelt refers to parties based on Postmaterialist constituencies as "Left Libertarian parties," and those at the opposite pole, with disproportionately Materialist constituencies, as "Right Authoritarian parties" (Kitschelt, 1994, 1995). Although we use different labels, our interpretations converge with those of Knutsen and Kitschelt on the key points.
The Rise of Postmaterialist Issues and the Decline of Social Class Voting
Most of the major political parties in Western countries were established in an era dominated by social class conflict, and to a considerable extent the main established political parties are still aligned along a social class-based axis. But support for new political movements and new political parties largely reflects the tension between Materialist and Postmaterialist goals. Accordingly, social class-based voting has been declining, and there has been a growing tendency for Western electorates to polarize according to Materialist versus Postmaterialist values. This development imposes a difficult balancing act on party leaders, especially those of Left parties. If they adapt to this new polarization too slowly, they lose their young Postmaterialist activists; but if they move too fast in this direction, they risk losing their traditional working-class constituency.
The rise of a new axis of politics, based on polarization between Postmodern values and traditional cultural values and the decline of class-based polarization, has left Western political systems in a schizophrenic situation. Most of the major political parties have been aligned along the class-based axis of polarization for decades, and established party loyalties and group ties still hold much of the electorate to this alignment. But the most heated political issues today are mainly Postmodern issues, on which support for change comes mainly from a Postmaterialist, middle-class base. This creates a stress that can be resolved in two ways: by repositioning the established parties or by creating new parties. Both have been taking place.
Klingemann, Hofferbert, and Budge (1994) find that there has been a gradual repositioning of party positions along the Postmodern politics axis: in an analysis of party programs in Western democracies during the last several decades, the percentage of references to social class conflict steadily declined, and the percentage of references to Postmaterialist issues increased sharply.
In the 1940s and 1950s, socialist policies were a major theme in the political party programs of Western democracies. As figure 8.7 demonstrates, socialist economic policies were mentioned in party manifestos about 15 times as often as were environmentalist policies: during 1944-59, the average party program referred to socialist economic policy about five times (socialist parties, of course, mentioned them more often than conservative parties), while the average party platform mentioned environmental matters .3 times: two-thirds of the party platforms did not mention environmental policy at all. Since then, a radical shift in emphasis has taken place. By the 1980s, environmental policy had overtaken socialist economic policy as a campaign issue (receiving
NEW   ISSUES   AND   NEW   PARTIES
253
£	12
CO	
en	11
o	
s-	
a	10
>	
C «s	9
o.	
	8
a.	
(0	7
c	
_o	6
c	
<D S	5
<*-	
O	4
L.	
a>	
£1	3
E	
3 C	2
C	
CO	1
CD	
S	
Environmental / policy       /
Socialist
economic
policy
»•ď
1944-59
(360)
1960-69
(227)
1970-79
(303)
1980-89    1990-93
(264)          (54)
Figure 8.7. Emphasis on socialist economic policy vs. environmental policy in political party programs: 1944-93. Source: Comparative Manifestos Project. Based on content analysis of 1,208 party programs issued during this period by political parties in the following countries: France, Italy, (West) Germany, Austria, Switzerland, Great Britain, Ireland, the United States, Canada, Sweden, Norway, Denmark, Finland, Iceland, Belgium, the Netherlands, Luxembourg, Australia, and New Zealand. For details, see Klingemann et al., Parties, Policies and Democracy (Boulder: West-view Press,1<
almost twice as much emphasis). By the 1990s, environmental policy dominated socialist economic policy as an electoral theme: the average party program mentioned environmental policy eleven times; socialist economic policy was mentioned only 2.5 times (with much of the mention being negative).
This seems to reflect political influences moving from the microlevel to the macrolevel: though we cannot directly demonstrate that the changes at the mass level preceded the shifts at the party program level, it seems implausible that this was a case of mass values following elite cues. For, as we saw in chapter 5, the changes at the mass level reflect a deep-rooted intergenerational change that can be traced back to the postwar economic miracles and was set in motion long before the party programs began to shift. At the elite level, the changes manifest themselves only in the 1960s and 1970s, when the postwar generation became an increasingly important segment of the electorate—and of the political activists. According to Carkoglu and Blinn (1994), by 1989 the Materialist-Postmaterialist issue dimension had become the first factor in the party programs of Western democracies, explaining more of the variance in party programs than the traditional Left-Right dimension based on the classic Marxist social class polarization over ownership of the means of production and redistribution of income. Although the parties were still perceived as po-
254
CHAPTER    8
sitioned along a Left-Right dimension, the dominant issue polarization had shifted from social class issues to Postmodern issues (Huber and Inglehart, 1995).
The rise of the Postmodern politics dimension tends to bring a reversal of social class positions: on the old Left-Right dimension, the upper income strata supported the Right or conservative position: they were the Haves and acted to preserve their economically privileged position against the Have-nots. But the Postmodern politics dimension is based not on ownership of property, but on one's subjective sense of security. It pits a Modern/Materialist worldview against a Postmodern/Postmaterialist worldview. On this dimension, those with higher levels of income, education, and occupational status are relatively secure, and increasingly, they tend to support the Left position.
Postmaterialists come from middle-class backgrounds, but they support change (Inglehart, 1977). This is conducive to a decline of social class voting, as middle-class Postmaterialists move left—and working-class Materialists move to the right.
For decades, one of the basic axioms of political sociology was the fact that working-class voters tend to support parties of the Left, and middle-class voters those of the Right (Alford, 1963; Lipset, 1960). This was an accurate description of reality a generation ago, but the tendency has been getting steadily weaker. As figure 8.8 illustrates, social class-based voting has declined markedly during the past 40 years. If 75 percent of the working class voted for the Left and only 25 percent of the middle-class voters did so, one would obtain an Alford class voting index of 50 (the difference between the two figures). As figure 8.8 shows, this is about where the Swedish electorate fell in 1948— but by 1990 the index had fallen to 26. The Scandinavian countries have traditionally shown the world's highest levels of social class voting, but it has declined sharply in all of them. In the United States, Britain, France, and West Germany, during the late 1940s and early 1950s, working-class voters were more apt to support the Left than were middle-class voters by margins that ranged from 30 to 45 points. By the 1990s, this spread had shrunken to the range from 1 to 25 points. In the 1992 U.S. presidential elections, social class voting had virtually disappeared. There were short-term fluctuations in given countries: the 1980s produced a partial resurgence of social class voting in Great Britain, for example. But all five of the countries for which we have data over this long time period show pronounced long-term declines in class voting. Overall, class voting indices in the 1990s were about half as large as they were in the postwar era. By the 1990s, the country with the highest class-voting index (Sweden) showed weaker class polarization than did the country with the lowest level in the 1940s (France).
The class-conflict model of politics is not just a straw man: a few decades ago it provided a fairly accurate description of reality. But that reality has changed, gradually but pervasively, and partly through intergenerational population replacement processes. Throughout Western Europe, social class vot-
N E W   ISSUES   AND    NEW   PARTIES
255
60 55
U)  50
c
| 45 «J 40 5 35
if«.
© 30
x
0) pc
■o co c
- 20 &~
O  iL
<
10
-	Sweden       ,.,,/a XW-"              v Great	\				
-	n Britain		o.	.n		
-	\ ö       -a    %_ •                                                 S,	1	t-v	A,		
c	/\      ° «», *'•     p/westX 'Ss uGermany\v		*y	\ \		"'O
-	b     c    D..... \    France	■°'.'Qv				^-°
	usa\ .a	•'	\ £>....	1**		«BBtí^Q
~	• o"		\	/	b...?^-<	
.......I	.....i......i	Mil	i i i i i	'Ú    I    1	*•• 1   1   1  1   1  1   1   1  M   1	i i i Vro
5 -
48 50 52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92
Year
Figure 8.8. The trend in social class voting in five Western democracies, 1947-92. Source: Adapted from Lipset (1981): 505. Updated by author with results from France and from recent elections. American data based on whites only, for comparability over time, cited in Abramson et al., (1985, 1994). The 1990 figures for European countries are from the 1990 World Values Survey.
ing indices are about half as large among the postwar birth cohorts as they are among older groups.
There has been extensive recent debate over whether social class voting has really been declining. If one focuses on selected periods for selected countries, it is easy enough to demonstrate that there has been no decline. For example, if one focuses on Great Britain and uses the 1969 low point (shown in figure 8.8) as one's starting point, one can conclude that there has been no downward trend in that country. But if one examines the entire time series since World War II, one finds a statistically significant decline that brought class voting down to half its former size.
Another approach has been to argue that Alford's manual/nonmanual dichotomy is too simple: using various more complicated ways of measuring social class, or more complex statistical procedures than comparing the Alford index over time, various analysts have argued that they find no downward trend. The great advantage of Alford's index is precisely the fact that it is so simple and straightforward: the distinction between manual and nonmanual workers is theoretically clear and easy to operationalize. It indicates a clear and obvious cutting point in any industrial society. Hout, Brooks, and Manza
256
CHAPTER    8
(1993) increase the number of measures of occupation and use logistic regression to test whether class voting has declined in the United States. Increasing the number of occupational categories does indeed increase the possibility that some combination of them may affect party choice. But their interpretation of their logistic regression results is questionable. They find, for example, that professionals voted predominantly for the Republicans in the 1950s, but for the Democrats in the 1990s. Ignoring this reversal of sign, they describe this as part of a pattern of "trendless fluctuation" because their various occupational groupings still explain about as much of the variance as ever (see Clark, 1995). Analyzing a massive database from 16 societies across four decades, and using more appropriate methodology, Nieuwbeerta and De Graaf (forthcoming) find a clear overall decline in class voting.
As social class-based voting has declined, the importance of the Postmodern political cleavage has increased. In the 1970s, Lijphart (1979) found "New Politics" parties (parties with a Postmaterialist constituency) in only three countries. In the 1989 elections to the European Parliament, New Politics parties won at least 10 percent of the vote in eight of the 12 European Community countries (Dalton, 1991b).
The Societal Level: Diminishing Marginal Returns from Economic Development
Let us turn now from analyzing cultural change at the individual level, to view it at the societal level.
Although Karl Marx died in 1883, his analysis of political conflict continued to fascinate social scientists for most of the following century. His emphasis on politics as the struggle to own the means of production captured an important part of reality in the early phases of industrial society. But with the evolution of advanced industrial society, new conflicts and new worldviews have emerged, making the economic conflicts Marx emphasized less central to political life.
This development reflects the diminishing marginal utility of economic determinism: economic factors play a decisive role under conditions of economic scarcity; but as scarcity diminishes, other factors shape society to an increasing degree. We have examined this phenomenon from an individual-level perspective; now let us examine it at the aggregate cross-national level. Forces operating at both levels converge, bringing a diminishing degree of economic determinism and class-based political conflict, as advanced industrial society emerges.
As we saw in chapter 2, human life expectancy is closely linked to a nation's level of economic development. In poor societies, life expectancies are less than 40 years, but they rise steeply with relatively modest increases in wealth, until one reaches a threshold of about $3,000 per capita income; then the life expectancy curve levels off. Economic factors become less decisive, and
NEW   ISSUES    AND    NEW   PARTIES
257
lifestyle factors more so. Similar patterns of diminishing returns from economic development are found with numerous other social indicators. Caloric intake, literacy rates, and other indicators rise steeply at the low end of the scale but level off among advanced industrial societies.
The pattern of diminishing marginal returns from economic development is not limited to objective aspects of life: it extends to subjective well-being as well. As we saw earlier, subjective well-being rises markedly with rising levels of economic development, and then levels off. Above a threshold of about $6,000 per capita, there is virtually no cross-national relationship between wealth and subjective well-being.
A rational strategy would dictate that at low levels of development, the individual should give top priority to maximizing one's income, and the society should give top priority to economic growth. But more of the same indefinitely is not a rational strategy. Beyond a certain threshold, there is a change in survival strategies, as Postmodern politics begins. Gradual cultural changes are feeding back into the political process of advanced industrial societies, leading to a change in their political agenda.
The Diminishing Political Base of the Traditional Left
Political life is also responding to a curve of diminishing marginal returns— in this case, the diminishing marginal utility of the classic program of the Left.
Equality of income distribution shows a curve of diminishing returns similar to those we saw for life expectancies and subjective well-being in chapter 2. Income equality increases sharply with economic development, up to a level of about $3,500 per capita in 1978 dollars (see Inglehart, 1990: 251); above that threshold, the curve levels off. In the overwhelming majority of countries with a GNP per capita below $3,500 (as of 1978), the top tenth of the population got more than one-third of the total income (in some cases as much as 57 percent). In none of the nations with a GNP per capita above $3,500 did the top tenth of the population get more than one-third of the total income; their share ranged from as low as 17 percent, in communist countries, to a high of 33 percent, in Finland.
Does this cross-sectional pattern reflect a longitudinal trend? The point has been debated. The most reliable longitudinal data come from economically advanced countries, most of which have shown only modest increases in income equality during the past 30 years. But if the shift is based on a curve of diminishing returns rather than a linear trend, this is exactly what we would expect. It is only in the earlier stages of economic development that we would observe large amounts of change. The United States, for example, moved toward substantially greater income equality from 1890 to 1950, but has shown some reconcentration since then. Absolute levels of income continued to rise, but relative shares changed only slightly. Most OECD countries moved toward greater income equality during the 1960s and 1970s, but the trend seems to have leveled off in the 1980s (Cusack, 1991). Conversely, Taiwan, South
258
CHAPTER   8
Korea, Singapore, and Hong Kong all have made dramatic leaps from poverty to prosperity only recently—and all have shown substantial increases in income equality (Chen 1979).
Why do we find a curvilinear relationship between economic development and income equality? In the early phase, we believe, it reflects a process of social mobilization, engendered by economic development. Industrialization leads to urbanization and mass literacy, which facilitate the organization of labor unions and mass political parties and the enfranchisement of the working class. Economic development does not automatically bring equality, but it does tend to transform the masses from isolated and illiterate peasants into organized citizens with the power to bargain for a more equal share of the pie.
But why does the curve level off among mature industrial societies? There are two main reasons. First, as a society approaches perfect equality, it necessarily reaches a point of diminishing returns. At the point where the top tenth had only 10 percent of the income, any further transfer of income would be a move away from equality. None of these societies has actually reached this point, but some were getting close. In East Germany, for example, the top tenth got only 17 percent of the total income. Norway, Sweden, and Denmark have greater income equality than the United States, West Germany, or France, which suggests that the latter countries could move further toward equality without necessarily having ineffective economies or coercive societies. But the Scandinavian countries seem to be approaching the limit of what is possible in a democratic political system. By the 1980s they were already experiencing a sharp public reaction against any further expansion of the welfare state, and began to cut it back.
Why this is so reflects a second basic principle: political support for increased income equality reaches a point of diminishing returns at a level well short of perfect equality.
As a society moves closer to an equal income distribution, the political base of support for further redistribution becomes narrower. In a poor society where the top 10 percent get 80 percent of the total income, the vast majority would benefit from redistribution. In a society in which the top 50 percent get 80 percent of the total income, far fewer people will benefit from further redistribution, and they will benefit proportionately less; one eventually reaches the point at which a majority of the voters stand to lose more than they would gain by additional redistribution. This does not constitute a moral justification for not moving further toward equality, but it does constitute a major political barrier in democratic societies. Under these conditions, the political base for further development of the welfare state is simply not there—at least not insofar as the citizens are motivated solely by economic self-interest. Ironically, further progress toward equality would come not from an emphasis on materialistic class conflict, but through an appeal to the public's sense of justice, social solidarity, and other nonmaterial motivations. Thus in the long ran, economic development makes a sense of economic deprivation both less widespread among mass publics, and a less powerful cause of political conflict.
NEW   ISSUES   AND    NEW   PARTIES
259
50
40 —
30 —
20 —
10 —
1890 1900 1910 1920 1930 1938 1950 1960 1970 1980 1988
Figure 8.9. The growth of government in OECD countries, 1890-1988.
Average government spending rates over time from 10 OECD countries: Austria, Canada, Denmark, France, West Germany, Italy, Japan, Norway, United Kingdom, and the United States. Source: Cusack, 1991.
Stated in this way, this conclusion may seem self-evident. But it is not; it has been hotly debated and is not generally accepted even now. A quarter century ago, the "end of ideology" school concluded that growing prosperity was giving rise to the "politics of consensus in an age of affluence" (Lane 1965); the subsequent explosion of protest in the late 1960s led many to conclude that this school had been completely wrong. In fact, the "end of ideology" school's analysis of what had been happening was partly correct; like Marx, they simply failed to anticipate new developments. While economic cleavages did become less intense with rising levels of economic development, they gradually gave way to other types of conflict.
By 1988 government spending had risen to nearly 50 percent of GDP in OECD countries, as figure 8.9 illustrates. This is another point of diminishing returns. Obviously, this trend cannot continue much longer: it is impossible to go above 100 percent except by running large deficits, and the costs of debt service eventually eliminate even that option. But another limit begins to take effect long before a society reaches the 100 percent level: psychologically, the 50 percent level seems to be a significant threshold. As taxation moves above this level, people begin to realize that an hour spent on tax avoidance can be more remunerative than an hour spent on one's job. And the ratio shifts very rapidly as the curve rises. When government spending (and taxation) rises above 66 percent of GNP, it may become economically rational to devote two-thirds of one's time to lobbying or tax avoidance, and a third of one's time to working. One's job eventually becomes a sideline; dealing with the govern-
o
m	rŕ.
	ra
CJ>	m
	>
n	<
c	■>,
	
a O)	|
	
(U	O
f-	c
C	|S
(II	
>	
O	
O	
m	
	
a	
260
CHAPTER   s
mental bureaucracy becomes the major focus. For some time, people may continue to work diligently, from force of habit, but a mentality of "They pretend to pay us, and we pretend to work" increasingly permeates the society.
Consequently no economy even approaches the theoretical 100 percent limit; functional requirements call a halt well short of this point. Thus, even the USSR, though ideologically committed to the position that private enterprise was morally wrong, probably never went above the 75 percent level. The rulers were forced to tolerate a sizable private sector because private agriculture and the unofficial economy were essential in staving off economic collapse. Thus, for functional as well as political reasons, by the 1980s the growth of the state was reaching natural limits. An awareness that the Marxist model was no longer working began to permeate mass consciousness.
Diminishing Returns from the Traditional Program of the Left
We have seen indications that economic development leads to a diminishing impact of economic influences on such objective characteristics as life expectancy and economic equality. But do such changes actually reshape the subjective political preferences of mass publics? The evidence suggests that they do; at high levels of economic development, public support for the classic economic policies of the Left tends to diminish.
Everyone knows that Denmark is a leading welfare state, with advanced social legislation, progressive taxation, a high level of income equality, and well over half its GNP going to the public sector. Obviously, the Danish public must be relatively favorable to these traditional policies of the Left. Conversely, everyone knows that Ireland is a relatively rural nation, with a modest public sector and no significant communist or socialist movements. Clearly, Ireland must be a bastion of conservatism on the classic Left-Right issues.
In fact, the conventional stereotypes are wrong on both counts. These stereotypes reflect patterns that were true in the past, but precisely because Denmark has attained high levels of social security—and very high levels of taxation— the Danish public has little desire for further extension of these policies. Support for the classic economic policies of the Left tends to diminish as economic development rises.
As figure 8.10 demonstrates, Greece is by far the poorest country among 11 European Community societies surveyed in 1979-8 3, and the Greek public has by far the highest level of support for nationalization of industry, more government management of the economy, and reducing income inequality. Ireland is the second-poorest country, and overall Ireland ranks second in support for these policies. At the opposite end of the spectrum, Denmark is the richest country—and has the lowest level of support for these policies. Western Germany ranks next to Denmark in economic level—and also in support for the classic Left policies.
The principle of diminishing marginal utility applies at the societal level, as well as the individual level. Greece is an economically underdeveloped coun-
N E W   ISSUES   AND   NEW   PARTIES
• Greece
• Ireland
• France • N. Ireland •Italy Belgium •• Britain
Luxembourg.^   ^eersmany
Netherlands
• Denmark
4      5      6     7     8     9     10   11    12
Gross Domestic Product/Capita in
ECU, 1992 (thousands)
Figure 8.10. Support for the classic economic policies of the Left in 11 countries, by level of economic development.
Based on responses to three questions asked in each of three Euro-Barometer surveys in 1979,1981, and 1983. Source: Inglehart, 1990: 256.
try, with many living in poverty and a small affluent elite. In such a context, the balance between rich and poor can be redressed only by strong government intervention. Denmark is a rich country that has long had some of the world's most advanced social welfare policies—and one of the world's highest rates of taxation. About 60 percent of Denmark's GNP is spent by the government; it is reaching the point at which it becomes impossible to move much further in this direction. In Denmark, further redistribution by the government seems much less urgent than in Greece—and the costs of government intervention impinge on a much larger share of the population. The incentives to press farther with the traditional economic policies of the Left become relatively weak, and public resistance becomes relatively strong.
There may be still another factor behind the decline of the traditional program of the Left. Tanzi and Schuknecht (1995) argue that the human returns on public spending are also subject to diminishing marginal returns. They analyze the increase in public spending in industrial economies over the past 125 years, assessing its social and economic benefits. They find that up to 1960, higher government spending was linked with considerable improvements in infant mortality rates, life expectancy, income equality, and educational lev-
0)	au
'5	
"5	
a.	
O	
1	80
o	
c	
o	
u uu	"5" 0)
+J	ľ!
•4— _1	(B .!» 70
o	(U
'5i	
V)	.E
ra	*j
O	O
c	c 60
*3	<8
L_	<D
O Q.	E,
Q.	
3	
tfl	
<D	50
U)	
<B	
*j	
C	
0)	
u	
\-	
	dn
262
CHAPTER   s
els. But since 1960, further increases in public spending have gone with only modest social gains—and those countries in which spending has risen most have not performed any better than those in which spending rose least. Indeed, on some indicators such as unemployment rates, the low-spending countries have done significantly better.
These findings are sure to be controversial, and they do not demonstrate a causal link: it is possible, for example, that most of the improvement in social standards up to 1960 was mainly due to rising incomes rather than public spending, and that such gains diminish once a certain income threshold is reached. But regardless of whether increased public spending brings diminishing social returns beyond a certain level, it seems clear that it does eventually bring diminishing public support.
The evidence in figure 8.10 suggests that higher levels of economic development are linked with diminishing mass support for state intervention. This figure is based on data from Western Europe. Does this pattern hold up more broadly? Data from the 1990-93 World Values Survey suggest that the phenomenon is global. As figure 8.10 indicates, throughout the industrial world public sentiment today is in favor of less, not more, government ownership. But the degree of support is closely linked to a society's level of economic development. Support for more government ownership is highest in China and Nigeria and lowest in the United States, Canada, and Western Germany. The midpoint of this scale is 5.5, about where India is located on figure 8.11. The only countries in which a majority favor more (rather than less) government ownership are China, Nigeria, Turkey, Chile, and Belarus; and the publics of Russia and India are about evenly divided. The publics of the other 35 societies favor moving away from government ownership. Marxism has become a phenomenon for which the mass appeal is found mainly in developing nations.
As figure 8.12 shows, even in Western countries, Postmaterialists are not particularly interested in government ownership of industry: though relatively enthusiastic supporters of the Left position on practically all of the Postmodern issues, the Postmaterialist constituency is not significantly more favorable to government ownership of industry than are Materialists. And in the ex-socialist societies, the situation is dramatically different: Postmaterialists are far more favorable than other groups to moving away from state ownership of business and industry. They see privatization as a step away from the rigid authoritarianism of the past, toward a society in which individual autonomy is enhanced.
In many ex-socialist societies such as Lithuania, Poland, and Hungary, the trauma of the transition to a market economy has brought reform communists back to power in free elections—but even the former communists are now disillusioned with government ownership. As one leader of the newly elected reform communist regime in Poland commented, "If one of our leaders proposed going back to a state-run economy, we would have to drop him from the party—not because he was a communist, but because he was an idiot."
HEW   ISSUES   AND   NEW   PARTIES
263
ES a S2
a °
O cz
V
China			
®			
Nigeria			
®	Belarus		
	t   ■          Va Chile Turkey      ®® ® Russia ®		
»india		Spain	
	Brazil      Moscow	©	
	®              ®		©
—	South Lith-           Korea		Japan
	uania             ©	Northern	
Romania         ®      Mexico		Ireland	Britai
©	©	©	®
	Poland		Italy
	®          Hungary		®    Norway /Sweden
	Bulgaria/ Czech.		
	•     //Latvia     Portugal		Nether- @@©x ® France
	South   ©®      i           ® Africa®              ® ©Estonia	Ireland	lands
		®	Belgium
—	® Slovenia		©
	® Argentina		Iceland ® Denmark West Finland     ®            Germany ®                        @ Austria ®   Canada ©
	East		©
	Germany		U.S.
1      '	......• i	1      I      I	1      1      1      1      1      1      1      1
$1,000
$5,000                   $10,000                  $15,000
PPC of GDP/Capita, 1990
$20,000
Figure 8.11. Economic development goes with diminishing support for state ownership.
Mean level of support for private vs. government ownership of business and industry, on a scale where 1 = "Private ownership of business and industry should be increased" and 10 = "Government ownership of business and industry should be increased." N = 43, r = -.54, p < .0001. Source: 1990-93 World Values Survey. Purchasing power estimates of GNP/capita from World Bank, World Development Report, 1993.
The policies that dominated the agenda of the Left throughout most of this century are running out of steam. Increased state intervention was desperately needed to alleviate starvation and social upheaval in the 1930s, was essential to the emergence of the welfare state in the postwar era, and still makes sense in some areas. But in others, it has passed a point of diminishing returns. The renewed respect for market forces that has emerged throughout most of the industrial world reflects this reality.
The neoconservative claim that the classic welfare state policies have failed is false, however. On the contrary, in such countries as Denmark these policies have largely solved the problems they are capable of solving—and have thereby reduced the demand for more of the same. Insofar as the succeed, they reach a point of diminishing returns, and begin to cede top priority to problems that have not been solved.
264
CHAPTER
80
U)
S   70
c
g
O
aj
| 60
'k.
a.
a
k* o
S   50
>
(S
Publics of 12 ex-socialist ^ societies /
40
30
Publics of 28
non-socialist
societies
_L
_L
Materialist    Mixed
Post-materialist
Value Type
Figure 8.12. Percentage in favor of privatization of business and industry by value type, among publics of 12 ex-socialist societies vs. publics of 28 non-socialist societies.
Percentages are those placing themselves at points 1-4 on a 10-point scale on which 1 = "private ownership of business and industry should be increased" and 10 = "government ownership of business and industry should be increased." Source: 1990-93 World Values Survey.
An attempt to turn back the clock to the savage laissez-faire policies of the early twentieth century would be self-defeating, ultimately leading to a resurgence of class conflict in all its former harshness. But the fundamentalists of the Left are equally self-defeating in their rigid adherence to a traditional program based on class conflict and state ownership and control of the means of production.
This does not mean that economic factors are no longer politically important. On the contrary, some very significant research has demonstrated strong linkages between fluctuations in the economies of Western nations and support for the incumbent political party (Kramer 1971; Tufte 1978; Hibbs, Rivers, and Vasilatos 1982). But this research has also produced a surprising finding: while support for the incumbents does reflect the performance of the national economy, it does not seem motivated by individual economic self-interest. The electorates of advanced industrial societies do not seem to be vot-
/V E W   ISSUES   AND    NEW   PARTIES
265
ing their pocketbooks, but instead seem primarily motivated by "sociotropic" concerns; rather than asking "What have you done for me lately?" they ask "What have you done for the nation lately?" (Kinder and Kiewiet 1979).
In short, economic factors remain an important influence on electoral behavior—but increasingly, they reflect sociotropic motivations rather than class conflict. The politics of advanced industrial societies no longer polarize primarily on the basis of working class versus middle class; and the old issues, centering on ownership of the means of production and government control of the economy, no longer lie at the heart of political polarization.
Conclusion
Marx set the agenda underlying modern political cleavages, which were based on ownership of the means of production and the distribution of income, and where support for the Left had a working-class base. With the emergence of advanced industrial society, the impact of economic factors reaches a point of diminishing returns. Postmodern issues take an increasingly important place on the national agenda, giving rise to a new axis based on the polarization between Postmodern and Fundamentalist worldviews; and support for sociopolitical change increasingly comes from a Postmaterialist, largely middle-class base.
The major established political parties in Western countries emerged in an era dominated by social class conflict, and to a considerable extent they are still aligned along a class-based axis. But support for new political movements and new political parties largely reflects the tension between traditional and Postmodern goals. Accordingly, social class-based voting has been declining, and there has been a growing tendency for Western electorates to polarize on a new axis, based on Postmodern issues. The established parties of the Left are trying to co-opt the Postmodern constituency, but if they move too far in this direction, they risk losing their traditional constituency, which reacts negatively to rapid change in sexual norms, gender roles, and massive immigration.
The Marxist model has lost its appeal in the industrialized world. Its emphasis on economic factors as the driving force of history provides a good first approximation of reality in the early stages of industrialization, but is of diminishing value as scarcity diminishes and new problems emerge. Similarly, the policies that are needed to counter the ruthless exploitation of capitalism in its laissez-faire stage reach a point of diminishing returns in advanced welfare states. Where government spending is already 40 to 60 percent of GNP, there is little potential to move further in this direction, and the massive power of big government itself becomes an increasingly serious problem. The old assumption of the Left that more government was automatically better has lost its credibility. But to elevate government nonintervention into a quasi-theological principle is equally untenable.
The meaning of "Left" and "Right" has been transformed. The key Marxist
266
CHAPTER   8
goal—nationalization of industry—has been abandoned by the publics of advanced industrial societies, though it remains attractive in less developed societies. Nationalization is not the panacea it once appeared to be. And insofar as it diverts attention from increasingly pressing problems concerning the quality of the physical and social environment, it can be downright counterproductive—for it provides no solution to these problems. Indeed, insofar as nationalization merges the political regulators with the military-industrial complex into one cozy elite, it may even make things worse. The nationalized factories of the Soviet bloc polluted even more than the private ones in the West. East Germany was the most severely polluted nation in Europe, with air and water pollutant levels two to three times as high as those in West Germany. And it is no coincidence that the only nuclear power plant accidents that have cost human lives occurred in the former Soviet Union, where environmentalist pressures for safety measures could not be freely organized. Although their environmental problems were even more severe, and their arms expenditures proportionately even higher than in Western societies, the political systems of the Eastern European countries made the emergence of independent environmentalist movements or peace movements far more difficult than in the West. It was relatively easy for the ruling elite to simply ignore such issues: officially, the problems underlying them existed only in capitalist countries.
The goals of individuals and the challenges facing society are different from those of a generation ago. For the past generation, Postmaterialists have controlled the agenda in advanced industrial societies, inserting a series of new causes into the political arena, and giving birth to new political movements and parties. After a generation of rapid social change, the opposition to Postmodern politics has become increasingly well organized and articulate, giving rise to movements and parties that systematically oppose the Postmodern agenda. The long-established political party institutions still reflect their origins in social class conflict, but the issues actually being debated today mainly concern support and opposition to Postmodernization.
CHAPTER     9
The Shift toward Postmodern Values: Predicted and Observed Changes, 1981-1990
A Broader Shift toward Postmodern Values
Not just postmaterialist values, but a whole range of Postmodern values are shifting in a predictable direction. This chapter and the next one examine this shift. Although we do not have year-by-year evidence of these changes (as we do for the shift from Materialist to Postmaterialist values), the 1981 and 1990 World Values surveys provide at least some evidence concerning how this broader shift toward a Postmodern worldview is moving.
The 1981 World Values Surveys showed that a wide variety of values and attitudes concerning politics, work, religion, sexual norms, and childrearing values were linked with Materialist/Postmaterialist values. They also showed significant differences between the preferences of young and old, with the old having attitudes similar to those of the Materialists, and the young having attitudes similar to those of the Postmaterialists.
These findings were based on the 1981 surveys alone: time series evidence was not yet available to show whether these age differences reflected an inter-generational shift. But we argued, on theoretical grounds, that these related attitudes were influenced by the same factors as those that motivated the Post-materialist shift and were moving on a similar trajectory:
Far-reaching though it is, the rise of Postmaterialist values is only one aspect of a still broader process of cultural change that is reshaping the political outlook, religious orientations, gender roles, and sexual mores of advanced industrial society. These changes are related to a common concern: the need for a sense of security, which religion and absolute cultural norms have traditionally provided. In the decades since World War II, the emergence of unprecedentedly high levels of prosperity, together with the relatively high levels of social security provided by the welfare state, have contributed to a decline in the prevailing sense of vulnerability. (Inglehart, 1990: 177)
Diminishing insecurity, it was claimed, is giving rise to a broad cultural shift, embracing not only Materialist/Postmaterialist values, but an entire syndrome of related attitudes. Emphasizing this prediction, the book that presented this thesis was entitled Culture Shift in Advanced Industrial Society.
We now have the cross-time data needed us to test this thesis. Is a broad cultural shift occurring?
We would not expect to find that all values are changing, of course: only those that are influenced by an underlying sense of security or insecurity. This