After the Third Wave of democracy, which began in the 1970s and blossomed around the world in the following decades, democracy came to define people’s aspirations for the ideal political system. But people had many different understandings of the meaning of democracy. Some associated democracy with individual freedom, others with institutional limits on government power, and still others with paternalistic government that takes care of the people’s needs. Such divergent understandings shape people’s expectations about their roles as citizens and the demands that they place on governments, influence the degree to which citizens feel satisfied with the kind of government they have, and hence have the potential to influence the stability of regimes.

What factors influence citizens’ understandings of democracy? This chapter argues that while social and institutional factors are part of the answer, in China and Taiwan, individuals’ cultural norms also have important effects on citizens’ understanding of democracy.

Before probing the effects of culture, the chapter illustrates the diversity of views of democracy across Asia with data from Asian Barometer Survey I (2001–3). The second section then suggests two contrasting conceptions of democracy that contend for adherence in the Chinese cultural sphere, liberal democracy and guardianship or minben democracy. Adherents of the guardianship view of democracy are likely to identify democracy as government that produces good substantive results for the people rather than government limited by procedures that protect the rights of the people. In order to map citizens’ views according to these ideal types, the third section analyzes the answers to an open-ended question asking people in China and Taiwan to use their own words to describe what democracy means to them. A comparison of the answers from China and Taiwan reveals that the institutional differences between the two societies did not eliminate the convergence of their citizens’ views on the nature of democracy. This suggests the persistent influence of culture. The fourth section explores how cultural variables affect people’s willingness to form a view on the nature of democracy.

The fifth section offers a cultural theory to explain why people understand democracy in different ways. I argue that people in the Chinese culture area have had to decode the foreign concept of democracy with the help of their existing normative beliefs. I will suggest that persons committed to the norms of hierarchical OTA and allocentric DSI are more likely to understand democracy in terms of guardianship, whereas reciprocal OTA and idiocentric DSI predispose people toward a procedural understanding of democracy. I test this theory by empirically assessing the roles of socioeconomic factors, institutions, and culture in shaping people’s understanding of democracy. Although socioeconomic resources help explain why people develop some understanding of democracy, they do not explain which understanding of democracy people adopt; those are shaped by the cultural norms into which people were socialized. These findings hold true for people in both mainland China and Taiwan. The chapter concludes with a discussion of the implications of the findings for authoritarian resilience in China and democratic reversal in East Asia.

Diverse Views of Democracy in Asia

“We live in a democratic age,” Fareed Zakaria asserted. “Over the last century the world has been shaped by one trend above all others – the rise of democracy. In 1900 not a single country had what we would today consider a democracy: a government created by elections in which every adult citizen could vote. Today 119 do, comprising 62 percent of all countries in the world” (Zakaria Reference Zakaria2004, 13). This trend is consistent with the modernization theory of political development, as described in Chapter 4, which predicts that the universal values of liberty and democracy will gradually conquer the world as modernization proceeds.

However, certain parts of the world seem to be exceptions to this pattern. Year after year the annual Freedom in the World reports have shown countries that are deemed “not free” to be concentrated in Eurasia, the Middle East, and central Africa.¹ In Asia, too, at the time of our 2002 survey, Freedom House rated about half of the countries “free” and the rest “partly free” or “not free.”Among the “not free” countries was China, which maintained its authoritarian political system despite abandoning a planned economy and undergoing rapid economic growth. Many surveys have shown that the majority of people in China are satisfied with the authoritarian regime, despite the fact that the regime still deprives its citizens of basic rights (Chen Reference Chen2004; Shi Reference Shi2001; Tang Reference Tang2005).

In the rest of Asia as well, authoritarianism commands widespread prestige. According to Asian Barometer Survey I, in 2002 less than half the public in South Korea and Taiwan thought that democracy was the best form of government, and a majority of citizens supported a possible authoritarian alternative. The number of citizens who harbored reservations about democracy was significantly large in Thailand, the Philippines, Taiwan, and Mongolia. Even in Japan, the region’s oldest democracy, citizens showed low support for the political system (Chu et al. Reference Chu, Diamond, Nathan and Shin2008).

Citizens often gave seemingly inconsistent answers to survey questions. Average answers to the question “If 1 indicates entirely unsuitable, and 10 indicates entirely suitable, please tell us how suitable you think democracy is for your country?” ranged from 8.75 in Thailand to 6.67 in Taiwan (see Figure 7.1).

Figure 7.1. “How suitable is democracy for your country?”

The high level of demand for democracy in China suggests that the citizens there are dissatisfied with their regime. Yet when asked, “What is the level of democracy in your country now on a scale from 1 to 10,” Chinese citizens gave an average answer of 7.22 (see Figure 7.2), suggesting that their country was already nearly as democratic as they wanted it to be.

Figure 7.2. “What is the level of democracy in your country now?”

The scores from Taiwan were puzzling in the opposite way. The level of democracy desired, or demand for democracy, was the lowest among the countries surveyed, and the perceived level of democracy already achieved – the supply of democracy – was higher than the demand.

As a final illustration of the point, consider the correlation between the levels of democratic achievement reported by respondents in the Asian Barometer Survey and the same countries’ Freedom House scores, which can be considered as objective outside assessments.² The Pearson’s correlation coefficients (not shown here) between countries’ Freedom House scores and their scores on the ABS democratic perception scale are not statistically significant. This suggests that the meaning of democracy inthe minds of Asians is different from that in the minds of the experts who apply the Freedom House coding scheme to Asian countries.

What explains such puzzling responses in Asia to a range of questions about democracy? The problem arises from the fact that survey questions like these are built on the assumption that people around the world understand the meaning of democracy in the same way. The surveys assume that democracy is democracy; neither cultural, nor social-structural, nor institutional differences can influence how people in different countries understand it. But in fact, as suggested earlier, people under the influence of different cultures and different political institutions do not understand democracy in the same way.

Alternative Understandings of Democracy

Once democracy became the key word symbolizing the ideal type of government, all regimes, including authoritarian ones, began to claim that they were a kind of democratic polity. If in the past the struggle was over whether democracy was the best form of government, it is now concentrated on the very definition of democracy.³ In this process, the definitions and understandings of democracy have changed and evolved over time and in different places.

Two stylized understandings of democracy can help us classify the views of our respondents in China and Taiwan. In a view widely accepted among Western social scientists and democracy promotion organizations, democracy is a set of institutional arrangements that constrain the power of government. Open and competitive elections, in which people can choose government leaders, lie at the heart of this arrangement.⁴ In order for democratic institutions to function properly, the system must allow a free flow of information so that people can make informed decisions when choosing government leaders through elections. The democratic system encourages people to participate in politics to press political leaders to adopt the policies they want. Consistent with the norm I have labeled reciprocal OTA, democracy legitimizes opposing one’s government, that is, replacing it with a different one through established procedures. I label this concept “liberal democracy.”

According to Robert Dahl, the “perennial alternative to democracy is government by guardians” (Dahl Reference Dahl1989, 52). Historically, the idea of guardianship has appealed to many different political thinkers and leaders around the world. “If Plato provides the most familiar example, the practical ideal of Confucius, who was born more than a century before Plato, has had far more profound influence over many more people and persists to the present day” (Dahl Reference Dahl1989, 54). Since 500 BCE, the theory of minben, or the Chinese concept of government by guardians, has profoundly influenced political thinking in China.

Minben doctrine and liberal democracy have two important features in common: both assign central status to the people’s welfare as a purpose of governance, and both give political elites the responsibility for making decisions toward this goal. Unlike liberal democratic theory, however, minben doctrine invests elites with full authority to use their own judgment in policy making without influence from the people, limits the function of ordinary people’s political participation to that of communicating information about local conditions, and sets a high bar for people to demand any change of government. The traditional Chinese minben version of guardianship is similar to the guardianship doctrine of Leninism, which assigns elites absolute power to make decisions for the people. Ever since the victory of the Chinese Communist Party (CCP), the official People’s Republic of China definition of democracy has been built on the overlapping dual traditions of minben and Leninism.

There are four important differences between liberal democracy and guardianship democracy. First, they assign different methods for building the ideal government. For liberal democratic theorists, the ideal government can be achieved only by implementing procedural arrangements for people to choose their leaders through periodic elections. For guardianship democrats, only a body of “highly qualified people – guardians, if you like – can be counted on to possess both the knowledge and virtue needed to serve the good of everyone” (Dahl Reference Dahl1989, 55).⁵ The ideal government requires an institutional design that can select and train people with the knowledge and skills to make wise decisions for everyone in the society.

Second, the two theories use different standards to evaluate legitimacy. According to liberal theory, government legitimacy can be acquired only through open and fair elections. The minben doctrine, however, calls for assessing a government by the substance of its policy outputs; how political leaders acquire power is not as critical as whether the government can bring substantive benefits to its people. For example, the Tang dynasty emperor Li Shimin is considered one of the best rulers in Chinese history, even though he assassinated his brother and put his father under house arrest in order to usurp power. After his coronation, Li rapidly developed the economy of the country, accepted constructive criticism from his ministers, and successfully defended China from the Mongol invasion. He is esteemed by Chinese historians and ordinary people alike, despite the way he acquired power, because he brought his subjects tangible benefits and good governance. But this is not to suggest that minben doctrine does not hold political leaders to high standards. In fact, their claim to legitimacy could be considered even more fragile than that of democratic governments. Governments built on minben doctrine have to sustain their legitimacy through a constant flow of good policy outputs; in democratic systems, elections empower political leaders to rule for a fixed time period even if their policy performance is poor.

A third difference between these two conceptions has to do with the role of political participation. In liberal democratic theory participation is a basic right enjoyed by everyone. Consistent with the norm of reciprocal OTA, it includes the right to demand a change of government through established procedures. By contrast, although minben doctrine requires political leaders to listen to people’s opinions in order to gather information and avoid unnecessary mistakes, it limits ordinary people’s political participation in the normal political process to communication or remonstration.⁶ According to Confucianism, harmony cannot be achieved by preventing people from expressing their views.⁷ But because ordinary people are understood to be shortsighted and to lack the information to make wise decisions, political leaders are not required to follow common people’s opinions. Instead, they are expected to make decisions based on their own expertise and judgment of society’s collective interest.

These two different views of participation were expressed during interviews that I conducted in China. When I asked a villager in Hebei province about the meaning of democracy, he told me that it means the government allows people to express their opinions.⁸ When I asked a college student in Beijing the same question, he replied, “Democracy means that ordinary people enjoy the right to participate in politics to elect their leaders and to replace their government.”⁹ Without clear theoretical guidance, both answers would be coded as “democracy means participation.” However, the political connotations of these two answers are fundamentally different. For the Hebei villager, democracy means a benevolent ruler who gives people an opportunity to express their views rather than preventing them from speaking out. For the college student, democracy means that citizens enjoy the right to exercise influence over government to try to get it to adopt the policies they want. The behavioral consequences of these alternative definitions are different. For the college student, if the government fails to comply, he is empowered with political rights to “oppose” it, that is, to demand a change of government using normal democratic procedures. For the peasant, the rulers’ failure to respond to his demands, in and of itself, does not give him the normative power to oppose his government.

Fourth, the two theories offer different understandings of freedom, a concept closely linked to democracy. Freedom in the liberal tradition is that area of individual action that the state does not limit by validly grounded laws. In The Federalist Papers, James Madison pointed out that a good government must be able to control the governed and at the same time control itself. To him, democracy meant order plus liberty (Hamilton et al. Reference Hamilton, Madison, Jay and Shapiro2009 [1787–1788]). Huntington made a similar argument in Political Order in Changing Societies (Huntington Reference Huntington1968). To balance these two goals, the freedom of individuals must be constrained to achieve order.John Stuart Mill resolved the tension between freedom and order by defining freedom as the absence of social coercion (Mill Reference Mill1993 [1859, 1863], 12–16). Isaiah Berlin differentiated positive liberty from negative liberty (Berlin Reference Berlin and Hardy2002 [1969]).¹⁰ To him, liberty in the negative sense defines the area within which people are, or should be, left to do what they want without interference by others.

Guardianship democracy values positive rather than negative freedom. According to Zhang Dongsun, “the Chinese from ancient times had nothing like the Western [liberal] concept of freedom.” The concept of zide, or “getting [something] in, by, or for oneself” might seem to resemble this idea, but it actually referred to a kind of positive freedom that can be achieved only through a long process of self-cultivation to make one change for the better over time (Tan Reference Tan2003, 167–75). Mao likewise argued that true freedom can be achieved only by a cumulative process of self-perfection. He believed that one is thoroughly free only when he or she no longer has the impetus to take negative social actions.

The two concepts of freedom were easily confused when the term “freedom” (or “liberty”) was first introduced in China. Without proper words to describe the concept, the term was originally literally translated as “unrestrained.”¹¹ Although people later adopted the term ziyou (“from oneself”) to translate the concept of freedom, without the distinction between positive and negative liberty, ziyou carries a connotation more closely connected to anarchism than to the meaning of freedom in liberal theory.

My interviews confirmed that people in China understand freedom in these two different ways. For example, one state-owned enterprise (SOE) worker told me that democracy means ziyou, and he elaborated by saying, “Democracy means I can do whatever I want to do!” “Does ziyou mean you can ignore traffic lights?” I asked. “Of course, real ziyou means I can do whatever I want to do. Otherwise, what does ziyou mean and why would I want ziyou?”¹² A member of an NGO, however, who also told me that democracy means ziyou, explained, “When people are exercising their freedom they should not infringe on the freedom of others.” “Then how do you know if you infringe on other people’s ziyou?” I asked. “The boundary is defined by law,” he answered. The meanings of freedom in the minds of the two men were quite different: to one it meant the absence of all constraints; to the other it meant the right to do whatever one wants within the boundaries defined by law.

The differences between the liberal and guardianship concepts of democracy are often obscured in general discourse. The way in which the word “democracy” was translated into Chinese – minzhu – further complicates the issue. Xiong Yuzhi traced minzhu back to its origins in the Chinese classics and found that the original meaning of the word was “master of people” (Xiong Reference Xiong, Lackner, Amelung and Kurtz2001). Chen Pengren pointed out that when minzhu was used in Shang Shu¹³ it referred to the ruler who made important decisions on behalf of his people (Chen Reference Chen1989, 17). After the Han dynasty, the term minzhu was used as an abbreviation of weimin zuozhu, which means “rule for the people,” or more specifically, decision making by government officials on behalf of the people to protect their interests. The term weimin zuozhu does not recognize the sovereignty of people or give people political rights, even though it requires government to protect their interests. In 1864, when W. A. P. Martin translated Wheaton’s Elements of International Law, the word minzhu was used for the first time to translate the idea of a democratic political system (Xiong Reference Xiong, Lackner, Amelung and Kurtz2001, 74). The particular Chinese phrase chosen by translators, I would argue, leaves a lot of room for traditional culture to influence how people understand democracy.

The differences between democracy and minben lie in the means they designed to achieve good governance, the standards they use to evaluate governmental legitimacy, the roles they assign to people’s participatory acts, and the ways they define the concept of freedom. The similarity between the two theories causes some people in Asia to understand democracy according to minben, whereas others understand it in ways that are consistent with the liberal tradition. Depending on which view a person holds, the meaning of “democracy” in the minds of Asian people can be fundamentally different from the understanding of democracy in the minds of Western experts.

How Do People in Mainland China and Taiwan Understand Democracy?

Although the political institutions in Taiwan reflect the principles of liberal democracy and those in mainland China reflect the principles of guardianship democracy, the Asian Barometer Survey found that institutional differences had only a limited impact on the way citizens in the two political systems understand democracy.

The Asian Barometer Survey examined this subject by asking an open-ended question: “What does democracy mean to you?” Interviewers were instructed to probe respondents twice after they gave the first answer. Ideally, the answers from all societies in the survey would be analyzed (they are summarized in Chu et al. Reference Chu, Diamond, Nathan and Shin2008, 12). However, since I lack the language skills to pick up subtle differences in answers given in languages other than Chinese, I concentrated my analyses on data collected from mainland China and Taiwan. The data are presented in Table 7.1.

Table 7.1. Understanding of Democracy in Mainland China and Taiwan

^a Percent entries are percent of the total sample. Total exceeds 100 percent because multiple answers are allowed for each respondent.

Source: 2002 China and Taiwan surveys, Asian Barometer Survey I (China N = 3,183; Taiwan N = 1,415).

In both societies, people understand democracy in various ways. Many people linked democracy to freedom or liberty. When coding the data, I made a distinction between freedom and liberty. The understanding of respondents who only mentioned the word ziyou in their answer was coded as “freedom.” If the respondent told our interviewers that democracy meant freedom within the boundaries defined by law, or that people enjoy the freedom to do what they want to do without infringing on other people’s rights, I coded the answer as “liberty.” It turned out that in both societies, among respondents defining democracy in terms of freedom or liberty, only about one-quarter added the proviso about not infringing on other people’s rights.

Respondents who told our interviewers that democracy means elections and other institutional arrangements to constrain political power were put in another category. In the mainland, 10.2 percent of respondents perceived democracy as elections, checks, and balances, or the division of power among different branches of government. Despite Taiwan’s fifteen years of democracy, a lower percentage of people – only 9.3 percent – defined democracy this way. In mainland China, 12.1 percent of respondents saw democracy as giving citizens the right to participate in politics, and another 11.7 percent believed it obliges government to listen to people’s opinions. In Taiwan, the figures were 8.7 percent and 9.3 percent, respectively.

At the same time, the analyses show that a substantial number of people understand democracy according to minben doctrine. For 8.7 percent of the people in mainland China, democracy means that a government takes people’s interests into consideration when making decisions or brings tangible benefits (shihui) to its people. Another 11.7 percent felt it means that government officials allow people to express their opinions.¹⁴ Similarly, 4.3 percent of people in Taiwan considered a regime democratic as long as its policy making takes people’s interests into consideration; 4.8 percent identified freedom of expression as the determining factor of a democracy. In all, 20.4 percent of people in mainland China and 9.1 percent of people in Taiwan understood democracy according to the minben tradition.

Beyond the answers outlined above, respondents also gave some answers that did not fit into the liberal-versus-minben classification. Some did not fit because they had elements of both theories – that is, that a good government is one that solicits and listens to people’s opinions when making decisions, pursues joint decision making, aspires to ensure equality in society, and provides people with a good life. Others did not fit because they were not conceptual definitions but references to specific institutions. For example, for a long time, the CCP defined democracy as “democratic centralism” (the party listens to the people, then makes decisions, and the people obey). Although the regime stopped using this term more than two decades ago, we still find people who define democracy in this way. In Taiwan, some defined democracy in a negative way, referring to their disappointment with such phenomena as ethnic mobilization by the Democratic Progressive Party (DPP), fistfights in the Legislative Yuan, and widespread corruption.¹⁵ They referred to these phenomena as “chaos” or some equivalent term or criticized them by saying that a proper democracy would demonstrate “self-control” or an equivalent term. Since these institutionally oriented responses do not reveal people’s exact definitions of democracy, I put them in the “other” category.

Table 7.1tells us something about the persistence of culture. If cultural norms adapt rapidly to new institutions, democratization in Taiwan should have caused people to abandon the more traditional minben understanding of democracy and adopt a liberal understanding. The persistence of minben-based conceptions in Taiwan shows that institutions have a limited impact on how people define democracy. To be sure, the percentage of Taiwanese (9.1 percent) who defined democracy according to the minben doctrine was lower than that of mainlanders (20.4 percent). But this does not necessarily show that the experience of living in a democratic regime in Taiwan has brought a strong switch to a liberal conception of democracy. Indeed, the percentage of people who adopted procedural understandings of democracy in Taiwan is still lower than that of mainland China.

Another indicator of the persistence of minben culture is the sizeable number of people in Taiwan who attach negative meanings to democracy. When culture is congruent with institutions, it normalizes the acceptance of certain negative consequences associated with the institutional design and characterizes them as necessary evils. Liberal democratic theory assumes that along with the rights to participate in politics and elect leaders comes the possibility that voters may choose a bad leader; this is defined by a democratic culture as a necessary risk. If incompetent or malicious political leaders are elected, actors who have internalized a democratic culture would not question their political system. Instead, they would either start the recall process or wait for the next election to select a new leader. People who have internalized minben doctrine would not be as forgiving because viewing a bad leader as legitimate is not culturally acceptable. They might begin to challenge the validity of the electoral institutions that produced such a leader. Many people in China, for example, when facing problems generated by rural elections, began to doubt the validity of the electoral system. This seems also to have been the reaction of some people in Taiwan when the political class was perceived as behaving badly.

Ideological transformation can be a complicated and prolonged process. Unless a person under the influence of minben thinking is fully converted to liberal democratic ideology, he or she may accept the goals, the means, and other principles defined by liberal theory but may not view the possible negative consequences as acceptable costs of the new system. Thus, a new convert may evaluate the performance of the newly democratized political system according to the old standards defined by minben doctrine. After discovering that elections had produced a corrupt leader, the half-converted person would become dissatisfied with the new political system. Thus, the negative evaluation of democratic performance in Taiwan may not indicate that the new democracy does not work but that democracy in operation departs from people’s expectations. If this hypothesis is correct, then the findings in Taiwan suggest that people were only partially converted to liberal democratic ideology. This may also help to explain the low evaluation of democracy given by people elsewhere in East Asia.

Cultural Norms and “Cognitive Diffidence” about Democracy

Table 7.1 revealed that many respondents in both mainland China and Taiwan gave a “Don’t know” answer to the question “What does democracy mean to you?” “Don’t know” answers can be interpreted as reflecting a lack of interest in or attention to the subject being asked about, a characteristic I label “cognitive diffidence.” The opposite of cognitive diffidence might be termed “cognitive engagement.” Cognitive diffidence may have different sources for different topics, such as a topic’s complexity or sensitivity or its lack of salience to the respondent. As students of culture, our interest is in whether cultural factors play a role in preventing respondents from forming an understanding of what democracy means.

To explore the role of culture in shaping cognitive diffidence toward the meaning of democracy, I constructed a model with variables that will be familiar to the reader from previous chapters. Education, a primary sociological resource, increases people’s cognitive capacity, equipping them with the ability to understand politics and governmental affairs, and should diminish cognitive diffidence. Income and urban residence should contribute to cognitive engagement, since economic development is accompanied by an expansion of government activities that brings more people in contact with political life. Males are more likely than females to be cognitively engaged with politics. Age is included in the model as a control.

Psychological resources, as discussed in Chapter 6, play a critical role in cognitive mobilization, leading some people to develop concern about their political system, which should lead them to acquire some knowledge of democracy. Three variables used in previous chapters are used here again to measure psychological resources: political interest, internal efficacy, and external efficacy. Institutions are also expected to influence people’s cognitive engagement with the political system. If media control is successful, we should expect that access to official media would make people more likely to develop a certain understanding of democracy. People with regular access to grapevine rumors should also be more likely to become interested in the political system and develop an understanding of democracy. I included political fear in the model as a control variable.

Furthermore, people’s interest in the characteristics of regime types might be influenced by the performance of government. If people feel that their current government is doing a good job managing public affairs, there is no need for them to rock the boat. Alternatively, if people believe their government is not doing its job, they may seek alternative political solutions. Thus, the likelihood that people will take an interest in the characteristics of different types of regimes might be negatively correlated with the perceived performance of government. I used two questions to measure perceived government performance. The first question read, “In your opinion, is corruption a major problem here in your local area?” and the second, “In your opinion, how serious is corruption in the central government?”

I also hypothesize an important role for cultural norms. Hierarchical OTA requires people to defer political decisions to a group of qualified elites and follow their guidance, so they may not need to form an understanding of different regime types; in contrast, reciprocal OTA assigns people the responsibility to pay close attention to politics and government affairs. This suggests:

Similarly, allocentric DSI discourages people from evaluating government policies purely on the basis of their impacts on their private interests, which gives government greater room to maneuver. Idiocentric DSI legitimizes evaluating government performance purely on the basis of its impact on one’s private interests and should lead people to be more critical of public authority and more likely to think about alternatives to their current political system. Therefore:

The sources of cognitive diffidence may not be the same for all objects of cognition. I therefore compare the role of culture in encouraging “Don’t know” answers to the meaning of democracy with “Don’t knows” to another question that has a different referent: “How suitable is democracy for your country?” (Figure 7.1).¹⁶ The referent of the meaning of democracy question is democracy in the abstract, as a regime type. The referent of the suitability of democracy is the respondent’s country and its situation and political needs. For each dependent variable, I coded people giving any substantive answer as 1, and all “Don’t know” and “Not applicable” answers as 0.

Since the dependent variables are binary, I chose the probit model. The results are presented in Table 7.2. As expected, most sociological and psychological resources increased the likelihood that people would provide substantive answers to both questions. Exposure to official channels of communication and grapevine rumors encouraged people to develop some ideas about democracy, but only the official channels and not grapevine rumors affected the likelihood of giving an opinion on the suitability of democracy. Perceived corruption in local, but not central, government was a factor that counteracted cognitive diffidence in both cases.

Table 7.2. “Don’t Know” Answers in Mainland China

*** p < .01; ** p < .05; * p < .1

Source: 2002 China Survey, Asian Barometer Survey I (N = 3,183).

Turning to culture, the analyses show that OTA and DSI have statistically significant and strong impacts on the likelihood that people will or will not develop both an understanding of democracy and a view on the suitability of democracy. As hypothesized, hierarchical OTA and allocentric DSI encourage respondents to express cognitive diffidence about both the meaning of democracy and its suitability for China.

By conducting a parallel analysis with the data from Taiwan, we can see whether this cultural effect is the same in a different institutional setting. The Taiwan model is presented in Table 7.3. The variables of political fear and grapevine rumors were removed from the model since they are not relevant to the situation in Taiwan. I added DPP membership as a control variable. Because the party was a major force behind democratic transition, membership in the DPP may have increased people’s concern for the type of regime in Taiwan.

Table 7.3. “Don’t Know” Answers in Taiwan

*** p < .01; ** p < .05; * p < .1

Sources: 2002 Taiwan Survey, Asian Barometer Survey I (N = 1,415).

We find that sociological and psychological resources played a similar role in Taiwan in encouraging people to develop an understanding of democracy and a view on democratic suitability. In contrast to mainland China, however, urban residents in Taiwan were no more likely than rural residents to understand the meaning of democracy. This reflects the fact that economic development has largely eliminated the rural/urban differences there. In Taiwan, it was the perceived corruption of the central rather than the local government that influenced people’s cognitive attention. This discrepancy reflects the fact that the central government has a more direct influence on the life of the average citizen in Taiwan than is the case in the much larger political system of mainland China.

The most important finding for this study is that cultural norms in Taiwan play important roles in the same way as they do in mainland China in shaping people’s willingness to form views on political questions. After controlling for structural and institutional effects and government performance, hierarchical OTA and allocentric DSI in two societies with fundamentally different political institutions made people less likely to understand the meaning of democracy.

The findings in this section show that both modernization and culture influence the likelihood that people will develop views on some political issues relating to regime type. But we have not yet explored how culture may influence what views people adopt. Different understandings of democracy may have different consequences for political behavior and for political development in a society. Even if people want a regime change, those influenced by the minben tradition may not want a change to liberal democracy. Any conclusion about the impacts of economic development on political development, and particularly on regime change, would be misleading without an understanding of the impacts of structure, institutions, and culture on how people understand democracy.

How Cultural Norms Shape People’s Understandings of Democracy

The fact that democracy has become a universally accepted value does not mean that people everywhere accept the liberal conception of democracy prevalent in the West. Some people in mainland China and Taiwan do, whereas others understand democracy in terms of the minben doctrine or in other ways. Since the way in which people define democracy can have important consequences for individual behavior and the evolution of the political system, it is important to understand what factors influence the ways in which people understand democracy.

When people are exposed to a new idea, they need to decode the concept to understand its meaning and implications. Since the minben tradition is so influential in East Asia, it may influence the way in which the idea of democracy has been decoded as it has come into Asia from outside. In particular, OTA and DSI might influence people’s cognitive decoding in two ways: (1) by shaping their expectations of the ideal type of government and (2) by providing standards for people to evaluate public authority.

I hypothesize that under reciprocal OTA, government’s legitimacy depends on whether rulers are selected by ordinary people through established procedures. Under hierarchical OTA, people evaluate government’s legitimacy on the substance of its policy outputs. This suggests:

Reciprocal OTA obliges government to take people’s views into account in policy making or else face opposition and replacement, whereas hierarchical OTA allows government to disregard the people’s opinions when necessary for the greater interests of the majority. Hierarchical OTA encourages elites to listen to people’s opinions, but it confines this function to information and fact-finding.

Thus:

Definition of self-interest can also influence how people understand democracy. For people who believe in allocentric DSI, collective interests are ranked above the interests of individuals, giving government the normative power to make decisions on behalf of society. As with hierarchical OTA, governments are judged on the basis of the substance of their policies. Those who believe in idiocentric DSI, on the other hand, believe that government should provide a fair opportunity for each person to pursue his or her interests. Thus:

Since idiocentric DSI defines the self as a legitimate unit of analysis in interest calculation, participation is seen as a right enjoyed by citizens in democratic institutions. Because allocentric DSI embeds private interests in the interests of collectives, government enjoys the right to ignore the interests of individuals in order to serve the interests of larger collectives or the whole society. Although allocentric DSI does allow people to express their opinions to authority, and encourages government officials to listen, as with hierarchical OTA, the norm limits the goal of these acts to information-gathering and communication. Thus:

To test these hypotheses, I classified responses to “What does democracy mean to you?” into five types (“Don’t know” answers were removed from the sample):

1. Liberty. People who defined democracy as freedom under rule of law or as freedom without interfering with the rights of other people, were coded as 1, others as 0.

2. Rights-based participation. People who understood democracy as a political system in which people enjoy rights to participate in politics, the right to express their opinions, and other human rights, were coded as 1, others as 0.

3. Elections/checks and balances. People who perceived democracy as a political system in which people enjoy the right to elect political leaders and/or as a political system with checks and balances among different branches of government, were coded as 1, others as 0.

4. Allows people to speak. People who defined democracy as a government that allows people to speak or as a government that solicits people’s opinions when making decisions, were coded as 1, others as 0.

5. Government for the people. People who described democracy as a political system in which the government takes people’s interests into consideration when making decisions or as a government that is concerned about its people, were coded as 1, others as 0.

Scores of 1 on any of the first three variables indicated that the respondent understood democracy according to the liberal concept, and scores of 1 on either of the last two variables revealed that respondents understood democracy in terms of the minben doctrine or government by guardianship.

The same independent variables used in previous analyses were used to predict how people understand the meaning of democracy. The goal was to try and tease out the impacts of structural, cultural, and institutional variables. As cultural norms, according to the argument I proposed in Chapter 1, shape people’s choices through both individual-level and community-level channels, I chose hierarchical linear modeling (HLM) for the analysis. Because the dependent variables are binary, I chose probit within HLM for the analyses. Each type of understanding of democracy is subjected to a separate regression. (Since respondents were allowed to give up to three answers to the question, a given respondent would appear in more than one of the regressions if he or she gave answers that fell into more than one category.) The results are presented in Table 7.4.

Table 7.4. Cultural Impacts on People’s Understanding of Democracy in Mainland China

*** p < .01; ** p < .05; * p < .1

^a Entries are restricted maximum likelihood unstandardized coefficients with HLM 6.06.

^b Entries are robust standard errors.

Source: 2002 China Survey, Asian Barometer Survey I (N = 3,183).

Although we found in Tables 7.2 and 7.3 that socioeconomic resources associated with modernization, psychological resources, and institutional effects all had some impact on people’s propensity to form views on political issues, Table 7.4 shows that all these variables had only weak effects in moving people away from a guardianship conception of democracy toward a liberal understanding. Education caused Chinese to understand democracy in terms of liberty and the right to participate but did not reduce the likelihood that people would define democracy as a government that takes people’s interests into consideration and/or one that allows people to express their opinions as part of the decision-making process. Urban residence, rather than fostering a Western liberal understanding of democracy, encouraged people to understand democracy in terms of the minben tradition. Youth tended to be more liberal and elders more devoted to a minben conception. Our standard measures of psychological resources showed no effect on the conception of democracy. The fact that access to the official media increased the likelihood that people would define democracy in terms of elections is initially puzzling but probably represents the success of regime propaganda in promoting the importance of village elections. Likewise, perceived government performance had a limited impact on how people understood democracy. Those who saw local government as corrupt were less likely to define democracy as a government that allows people to speak, whereas those who saw the central government as corrupt were more likely to define it in that way, which suggests that local government corruption did more to destroy the faith in guardianship democracy than did central government corruption.

Cultural norms played the largest roles in shaping how people in China understood democracy. The analyses confirm H3, H4, and H6. After controlling for social-structural, psychological, institutional, and performance variables, people with reciprocal OTA tended to understanddemocracy in terms of procedural arrangements to constrain political power, the guarantee of liberty, and the right to participate in politics. People holding hierarchical OTA were more likely to understand democracy according to the minben tradition. The effect is seen at both the individual level and the county level. People living in a reciprocal environment are more likely to define democracy according to the liberal tradition than people living in a hierarchical social environment. Likewise, idiocentric DSI at the individual level led people to understand democracy as a political system guaranteeing people liberty and the right to participate in politics. Allocentric DSI, on the other hand, led to the notion that a democratic government allows people to express their opinions when it makes decisions.

However, no empirical evidence can be found to support H5. Definition of self-interest had no impact on how people define the democratic character of a regime. Taken together, the analyses confirm that hierarchical OTA and allocentric DSI encourage people to understand democracy in terms of minben doctrine. Reciprocal OTA, on the other hand, fosters a traditional liberal understanding.

To see whether the effects of culture on understanding of democracy are independent of the institutional setting, I looked again to the data collected from Taiwan. Table 7.5 shows the impacts of structure, institutions, and culture on people’s understandings of democracy in Taiwan. Similar to mainland China, sociological and psychological resources had limited impacts on how people in Taiwan understood democracy. Despite the fact that democratic values have become a major subject of civic education in Taiwanese public schools, people with higher levels of education were not more likely to define democracy in liberal terms.

Table 7.5. Cultural Impacts on People’s Understanding of Democracy in Taiwan

*** p < .01; ** p < .05; * p < .1

Source: 2002 Taiwan Survey, Asian Barometer Survey I (N = 1,415).

The most important finding for the purpose of this study is that cultural norms in Taiwan, a society with fundamentally different institutional settings, work the same way as they do in mainland China. While reciprocal OTA and idiocentric DSI made people in Taiwan more likely to understand the meaning of democracy according to liberal theory, hierarchical OTA and allocentric DSI made them more likely to understand the meaning of democracy in terms of guardianship.

The findings in Tables 7.4 and 7.5 are likely to be weakened by the fact that each of the five concepts of democracy is compared to a mixed bag of all other concepts. A clearer view of the impact of culture on conceptions of democracy can be obtained by classifying all respondents into three exclusive categories: those who gave all of their up-to-three responses to the question “What does democracy mean to you?” in the category ofliberal-democratic answers, those who gave all of their responses in the category of minben answers, and all other persons (including those who gave mixed responses and those who gave “Don’t know” responses). The numbers in each category are shown in Table 7.6.

Table 7.6. Different Understandings of Democracy (2002)

Source: 2002 China and Taiwan surveys, Asian Barometer Survey I (China N = 3,183; Taiwan N = 1,415). Data are weighted. Adapted from Tianjian Shi and Jie Lu, “The Shadow of Confucianism.” Journal of Democracy 21, 4, October 2010, p. 128.

The technique of multinomial regression allows us to compare the believers in liberal and minben concepts of democracy to all other respondents. The results are presented in Table 7.7.

Table 7.7. Multinomial Analysis of Cultural Impacts on People’s Understanding of Democracy (2002)

*** p < .01; ** p < .05; * p < .1

Source: 2002 China and Taiwan surveys, Asian Barometer Survey I (China N = 3,183; Taiwan N = 1,415).

The findings again highlight cultural norms as the single most consistent determinant of people’s conceptions of democracy. Although structural, psychological, and institutional variables increase the likelihood that people will pay attention to the political system and develop an understanding of democracy, they have very limited impacts on how people understand democracy. Cultural norms alone have this effect among the variables we have examined.

Conclusion

The key findings of this chapter can be summarized as follows: most people in mainland China and Taiwan claim that they want democracy, but democracy means different things to different people. While sociological and psychological resources do cause people to develop some understanding of democracy, as suggested by social mobilization theory, the ways in which they understand democracy are shaped by the cultural norms they were socialized into in their early lives. Traditional culture in Chinese society does not prevent people from accepting democracy as ideal form of government, but it does help to define the particular meaning of democracy for political actors. The ideal type of governmentdefined by traditional culture significantly influences the way people in these two societies understand democracy. Studying people’s aspiration toward democracy without carefully examining what democracy means to them would cause researchers to reach inaccurate conclusions about the relationship between people’s support for democracy, regime change, and democratic consolidation.

The analyses also reveal that the institutional changes in Taiwan, rather than altering the relationship between traditional cultural norms and the way people there understand democracy, reinforced cultural impacts. Together, these findings provide a possible explanation for why Asians under the influence of Confucian culture appear to be outliers to the social mobilization theory that predicts that democratization will follow modernization.

The findings may also help explain popular support for the authoritarian system in mainland China and popular suspicion of the democratic system in Taiwan. Since many people in both societies define democracy according to minben doctrine, the standards they use to evaluate government legitimacy and performance must be different from those of liberal democrats. Because traditional minben culture bases government legitimacy on policy outcomes, depriving people of the right to elect public officials or ignoring requests made by certain people in the society does not jeopardize regime legitimacy. Even if people are not satisfied with a policy, their understanding of democracy does not include the right to try to replace the government. Policy conflicts are to be resolved by informing the existing leaders of the realities of people’s lives. Thus, despite the survey results showing that an absolute majority of people in China claim they want democracy, the way Chinese people understand democracy leaves their government great space to maneuver. This space, I would argue, constitutes the micro foundation of authoritarian resilience in China.

Cultural variables may also help explain the decline of popular support for democracy and the presence of democratic reversals in Asia in recent years. Unless people in this region fully convert to liberal values, they may accept the goals defined by liberal democratic theory and even the means defined by liberal theory to achieve these goals, but they may not be able to accept the negative consequences associated with the new institutional design. Newly converted democrats, such as the people of Taiwan, may evaluate the performance of the democratized political system according to the standards defined by minben theory.