Propaganda as Signaling

Author(s): Haifeng Huang

Source: Comparative Politics , July 2015, Vol. 47, No. 4 (July 2015), pp. 419-437
Published by: Comparative Politics, Ph.D. Programs in Political Science, City University
of New York

Stable URL: https://www.jstor.org/stable/43664158

REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/43664158?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Comparative Politics, Ph.D. Programs in Political Science, City University of New York is
collaborating with JSTOR to digitize, preserve and extend access to Comparative Politics

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
 Propaganda as Signaling

Haifeng Huang

Why do authoritarian governments engage in propaganda when citizens often know that
their governments are propagandizing and therefore resist, ignore, or deride the mes-
sages? In China, for example, much of the government's official discourse and rhetoric
are so obsolete and far-fetched that they have been widely ridiculed1 and dubbed the
Chinese government's "own worst enemy."2 Some scholars have even argued that Chinese
citizens with more exposure to state media reports may, in fact, have less trust in the gov-
ernment, since more official messages make them see more inconsistencies between the
propaganda and the reality.3 Nevertheless, the Chinese government maintains a massive
institutional structure and devotes enormous resources on a daily basis for the production
of propaganda in the form of state media publications and programs, political education
in schools, ideological campaigns, and various kinds of rituals and ceremonies.4
Similar examples abound in other countries. Former Syrian president Hafiz al-Assad
was regularly portrayed in official Syrian political discourses and extravagant rituals as
being omnipresent and omniscient, carrying such titles as the "first teacher," "savior of
Lebanon," and even "premier pharmacist." As such, he knew "all things about all issues,"
even though such communications were not believed by ordinary Syrians or even by
members of the regime directly responsible for their dissemination.5 Similarly, the North
Korean propaganda apparatus is rarely concerned about the veracity of its various
preposterous pronouncements, such as the one claiming a "big and bright halo"
floated above Kim Jong II's alleged birth place for an hour on the occasion of his birth-
day,6 but seems to insist that the supposed virtues of the leaders and the magical
revolutionary history of the country must not be judged by their (lack of) factual
accuracy but be accepted on their own terms.7 More generally, in communist coun-
tries propaganda posters and slogans can be found everywhere even though few
people really care to read them.8 These countries also emphasize ideological and
political education in schools, but such courses are so unpopular that the Vietnamese
government, for instance, has recently resorted to offering free tuition to attract col-
lege students to study Marxism-Leninism and Ho Chi Minh Thought.9
Why do authoritarian governments engage in unpersuasive propaganda? The
common understanding of political propaganda is that it is a means to indoctrinate
419

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
Comparative Politics July 2015

the masses with pro-regime values and attitudes. However, for indoctrination to be
effective, one has to be convinced by the content of the propaganda. Pretentious
propaganda that is not persuasive is at odds with this goal. In this paper I propose that
propaganda is often not used for indoctrination, but rather to signal the government's
strength in maintaining social control and political order. More specifically, by being
able to afford significant resources to present a unified propaganda message and
impose it on citizens, a government that has a strong capacity in maintaining social
control and political order can send a credible signal about this capacity and dis-
tinguish itself from a weak government, hence implicitly intimidating the masses
who may otherwise contemplate regime change. In other words, such propaganda is
not meant to "brainwash" people with its specific content about how good the gov-
ernment is, but rather to forewarn the society about how strong it is via the act of the
propaganda itself.
An example from Chinese media will be useful to illustrate the above argument.
Media commercialization in China in the last few decades has bred market-oriented
print, broadcast, and cable media outlets, and lively coverage of social and economic
issues now thrives in the country, even though political discussions are still controlled.10
Against this sea change of the media sector, however, there exist some notable anoma-
lies: the reportage of the country's prime-time television news program, China Central
Television (CCTV)'s Xinwen Lianbo (Network News Broadcast), and the preeminent
official newspaper, People's Daily, have remained notoriously formalistic, ritualistic,
and ideological. Consider Xinwen Lianbo, which remains the most high-profile televi-
sion news program in China as government regulations have made it virtually the only
TV news program available around dinner time for most families. In contrast to the
tremendous changes in the society that the news program is supposed to cover, and even
to many other programs of the CCTV itself, the content, language, and format of
Xinwen Lianbo have remained largely unchanged. Its coverage has invariably focused
on party leaders having meetings, attending ceremonies, receiving foreign guests, or
touring local areas, often without reporting the substance of those activities. The use
of archaic and stilted prose glorifying the government has made the program a constant
target of mockery among ordinary citizens;11 media scholars in China call it "a theatrical
privately not believed by people at all but nevertheless performed with a clear and
rich tone and promoted as real."12 Despite this, the state orders almost all provincial
TV stations in the country to simulcast Xinwen Lianbo at 7 p.m. every evening.
Why does the Chinese government do this? The signaling theory suggests that
citizens may dislike and remain unpersuaded by the content of Xinwen Lianbo, but
its continual existence and the fact that the government can easily bombard the nation
with this much-ridiculed program demonstrates its strength and capacity.13 This is not
to say that Xinwen Lianbo has no other functions. Chinese citizens who view official
media as government mouthpieces will nevertheless consume their reports in order
to learn the government's policy positions.14 But the fact that the style of Xinwen Lianbo
is conspicuously different from other news programs of CCTV and has stubbornly
remained so despite numerous calls to reform the program and make it "keep pace with
420

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
 Haifeng Huang

the times" shows that Xinwen Lianbo is not merely about announcing government
positions or policies.
The term "signaling" here thus refers to the indirect provision of information - in
the present case, the government's strength - through the government's act of doing the
propaganda, rather than its usual meaning in everyday language: direct provision of
information as contained in what the government is saying in the propaganda. Such
indirect provision of information is possible because propaganda is costly, particularly
at the scale implemented by authoritarian states like China, and, therefore, the willing-
ness and/or ability to undertake such costly actions constitute a credible signal of
the government's capacity and resources. An analogy can be found in the literature
on political campaigns and advertising in democracies, which can be expensive but
often contain little new information to voters. By "burning money" publicly, however,
such campaigns and advertising can signal some otherwise non-verifiable attributes of
the candidates.15
In the article I use unique survey data from China to test the above argument; a
simple game-theoretic model is contained in the appendix.16 Consistent with the theo-
retical prediction, I find that Chinese college students who are more familiar with the
government's propaganda messages embedded in their ideological and political educa-
tion courses are not more satisfied with the government, but they are more likely to
believe that it has a strong capacity in maintaining political order and are, hence, less
willing to express dissent. This is notable because those courses are not really about the
government's repressive or political control capacity, but rather the greatness and glory
of the party-state. That the government is capable of delivering the pompous and some-
times ludicrous propaganda without much overt opposition, however, has implied to the
students that the government is strong.
This signaling theory of propaganda is a complement to rather than a substitute for
the standard indoctrination theory of propaganda given that in many contexts propaganda
(such as subtle media messages or literary works) can indeed change the recipients'
opinions and imbue them with attitudes favorable to those in power. Citizens' political
awareness and ability to resist government propaganda also vary.17 The purpose of this
paper is not to replace the indoctrination theory but to suggest that indoctrination is
not the only function of propaganda. Some propaganda may not influence the masses'
political values and attitudes but can nevertheless affect their behavior and promote
regime stability by displaying the government's strength, capacity, and resources. As
will be discussed in the concluding section, propaganda that does and does not induce
belief can be called "soft propaganda" and "hard propaganda" respectively. This study
focuses on the latter.

Propaganda: Indoctrination or Signaling

Traditionally, propaganda is understood and, indeed, defined as "the attempt to trans-

mit social and political values in the hope of affecting people's thinking, emotions,
421

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
Comparative Politics July 2015

and behavior."18 This has been the premise of almost all classic and standard works on
political propaganda, including those of early theorists,19 scholars of authoritarian and
communist propaganda,20 and critics of Western media.21 Scholars of political propa-
ganda in China have similarly emphasized how propaganda shapes the values, opinions,
and attitudes of the masses, and call it "thought work" following the terminology of
the Chinese government.22 They argue that while the country's propaganda system, par-
ticularly the media system, has been significantly updated and commercialized since
the Maoist era, with the Maoist goal of transforming the nature of human beings for
socialist revolution and construction discarded and Maoist tactics such as struggle and
criticism sessions replaced by more modern and even entertaining practices, the fun-
damental goal of Chinese propaganda is still to instill the public with nationalism, con-
sumerism, and/or New Confucianism, or otherwise to "guide public opinion" and make
it conducive to regime and social stability.23 I call this standard and prevailing view the
indoctrination theory of propaganda.
While indoctrination indeed occurs when the content of propaganda persuades its
recipients and scholars such as those cited above have provided significant insights
about how it has been done in various contexts and time periods, the fact that in many
other situations propaganda does not actually induce belief shows that indoctrination
is not the only story of propaganda and sometimes not even the main story. Following
the Spence signaling game framework,24 I develop a signaling theory of propaganda,
which states that authoritarian governments engage in seemingly unproductive and
wasteful propaganda activities not to imbue the masses with pro-government attitudes,
but to demonstrate their strength in social control. In fact, for this demonstration
of strength to be well taken, propaganda may sometimes need to be dull and unpersua-
sive, so as to make sure that most citizens will know precisely that it is propaganda
when they see it and hence get the implicit message.25
The full theoretical model explicating the logic of the signaling theory is in
the appendix; here I sketch its main idea. There are two players: the government
and the citizenry. The government's capacity for maintaining social control and
political order can be strong or weak; a strong government can defeat a political
challenge of the citizens with a higher probability than a weak government. The gov-
ernment knows its capacity, but citizens do not; although, prior to the game, they
have a belief about the probability that the government is strong. Citizens' prefer-
ences and costs of challenging the government are such that they will rebel against
the government if they know it is weak, but will not rebel if they know it is strong.
Such a rebellion should be understood as a political challenge or revolution against the
regime rather than loyalist and localized protests asking the government to address some
specific grievances.
In the game, the government moves first, deciding how much propaganda to
produce. After observing the government's level of propaganda, citizens decide whether
or not to launch a rebellion. Crucially, the cost of producing the same amount of pro-
paganda is lower for a strong government than for a weak government. In other words,
a government that is strong and more capable in social control can maintain a propaganda
422

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
 Haif eng Huang

apparatus, carry out propaganda activities, and impose the messages on citizens more
easily and efficiently than a weak government.
This means that citizens can make inferences about the type of the government
by observing whether it is willing to produce a high level of propaganda, even if the
content of the propaganda itself is not believed by the citizens. I show that there is a
unique separating equilibrium in this game: a strong government produces a sufficiently
high level of propaganda that a weak government is unwilling or unable to produce and,
therefore, distinguishes itself from the latter. Citizens are then deterred from rebellion
when they observe a sufficiently high level of propaganda not because it induces a more
positive view of the government but because they now know that the government has a
strong capacity for defeating a rebellion and maintaining political control.
Scholars have long noted that propaganda is often not intended for (or does not
result in) persuasion. Hannah Arendt states in her analysis of totalitarianism that the
"true goal of totalitarian propaganda is not persuasion, but organization of the polity,"
and that "what convinces masses are not facts, and not even invented facts, but only
the consistency of the system of which they are presumably part."26 In Vaclav Havel's
memorable discussion of why a greengrocer in Soviet-era Eastern Europe placed politi-
cal slogans that few people would read in his store window among the fruits and veg-
etables, he notes that such slogans formed part of the panorama of everyday life, and
that "while they ignore the details, people are very aware of that panorama as a whole,"
which "reminds people where they are living and what is expected of them."27 These
insightful observations, however, were brief and not fully developed or elaborated.
The work that is most closely related to this paper is Lisa Wedeen's ethnographic
study of the cult of Hafiz al-Assad in Syria, which also seeks to understand why an
authoritarian regime would engage in preposterous propaganda and rituals that do not
convince citizens. Her arguments are multifaceted but focus on how official discourses
clutter public space and drive alternative messages underground, provide the correct
"grammar" and formula for acceptable speech,28 and thus habituate citizens to behave
"as if" they believe in official rhetoric. She also hints at signaling, for example, when
she notes that "the greater the absurdity of the required performance, the more clearly it
demonstrates that the regime can make most people obey most of the time,"29 but for her
this is part of an extensive and intertwined mechanism that compels citizens to practice
what Timur Kuran calls "preference falsification,"30 and not analyzed separately.
This article explicitly develops and formalizes the signaling mechanism of pro-
paganda and subjects it to systematic testing. It thus integrates formal theory with
public opinion survey in the study of authoritarian politics. The contemporary litera-
ture on authoritarianism, including its formal theory component, has focused on how
authoritarian rulers use political institutions for power sharing and social co-optation,31
but not the formation of public opinion. By connecting the traditionally disparate
methods of research, the article can make unique contributions to the study of authori-
tarian politics.
There is also a nascent game-theoretic literature on news media in authoritarian
countries, which focuses on the censorship of news and information.32 The only other
423

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
Comparative Politics July 2015

formal-theoretic paper to date on propaganda is by Chris Edmond, who shows through

a global game model that exaggerating the government's ability to repress can lead
citizens to coordinate on not challenging the regime.33 The difference between his study
and this paper is that in this paper information about the government's strength is not
directly provided, but indirectly inferred from the government's actions.
With regard to China studies, two other papers have also used the signaling frame-
work, although in different contexts. Victor Shih holds that junior officials in China
often use "nauseating" public praises of a senior leader to send credible signals about
their loyalty.34 When explaining the prevalence of conservative and dogmatic rhetoric
in China's reform era, I argue that the reformist central government uses such rhetoric
to conceal its objective and to control the pace of local reforms.35 This article studies
state-society relations rather than internal dynamics within the government.

Survey Evidence

Background If the signaling theory correctly describes some aspects of the reality
in an authoritarian country, there should be certain important forms of propaganda that
will make citizens more likely to believe that the government is strong in maintaining
political order and social stability, and thus decrease their willingness to oppose the
government, although they are not necessarily made more satisfied with the regime.
I already discussed China's flagship state news program Xinwen Lianbo to illus-
trate the signaling theory. Here, I more formally test the theory by examining the
effects of ideological and political education in Chinese universities, which has played
a crucial role in sustaining the Chinese Communist Party's (CCP) rule. After the 1989
student-led Tiananmen Movement that brought the Chinese government to the brink
of collapse was put down, the paramount leader Deng Xiaoping concluded that "[o]ur
biggest mistake in the last ten years was education, by which I mainly mean ideo-
logical and political education."36 Since then mandatory ideological and political edu-
cation courses have been reemphasized in college curricula.37 The current leadership,
while pursuing economic and social reform measures to correct severe economic imbal-
ances and alleviate social grievances, has continued and even intensified ideological
work, including recently engineering major counteroffensives against seven "false ideo-
logical trends and positions,"38 which have also been implemented on college campuses.
With the expansion of higher education in China, the impact of political and ideological
lecturing reaches a very large proportion of the young-adult population. Requiring
political education to this "captive audience" on college campuses can also have
ramifications beyond students' college careers.
In this study I measure the students' exposure to ideological and political education
by a set of questions from the courses they have taken in previous semesters. Although
these courses are mandatory, the attention students actually pay to them varies, and so
the classes will have different levels of effects on the students. Chinese students and
even many instructors generally view such courses as nuisances, rituals that they dislike
424

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
 Haifeng Huang

but have to observe. Students also typically regard the courses as useless for their
future careers. When asked how they treated the political education courses, only 8 per-
cent of the students surveyed in the study reported that they somewhat actively studied
for the courses, with the rest acknowledging that they listened to lectures only casu-
ally, did not listen to lectures at all, relied on cram sessions to prepare for exams, or
simply skipped some classes. Such attitudes are hardly unique to this sample.39 Stu-
dents' performances in such courses are therefore largely a function of their incentive
and ability (including cognitive ability such as memory) to achieve a high overall
academic standing, since grades in these courses constitute part of their GPA. Condi-
tional on their overall academic standings as well as some other factors that will be
controlled for, the students' familiarity with the materials and lecturing approximately
reflect random or idiosyncratic factors not systematically correlated with their politi-
cal attitudes that influence their attention and exposure to the ideological and political
education. In other words, there is some randomness in the amount of treatment for
each student. If the signaling model correctly describes the role of state propaganda in
such courses, those students with more exposure to the courses, in the sense of being
able to recollect more teachings from past courses, will be more likely to believe that
the government is strong, but not more likely to believe that the government is good.

Data and Measurement In the following section, I report the results of a unique survey
conducted in a mid-sized and mid-upper-ranked university in eastern China in the spring
semester of 201 1. Although not top-ranked, the university is one of the key national uni-
versities under the direct supervision of the Ministry of Education, offering degrees in
engineering, social sciences and management, natural sciences, and humanities, with a
slight emphasis on engineering as is typical in China (a legacy of socialist planning).
In addition, because the university is mid-upper ranked, survey participants in the project
were also potentially more representative of Chinese college students in general than those
from top elite universities, who have usually been the target of academic surveys. Since
instructors from different universities often teach the political education courses dif-
ferently, for the purpose of this paper it is important to restrict the testing of the students'
ideological and political "learning" within the same university so that students have had
the same lectures and instructors.40
The survey was implemented in a university-wide required second-year course as a
class activity to assess teaching effectiveness and student opinions. One out of every
two sections of the course in the university's main campus was selected for the survey,
which thereby covered about half of its sophomore population and all but some small
majors. The formal survey, reported below, had 1,250 respondents altogether; however,
not all respondents answered every question on the questionnaire, thus the variation in
the number of observations reported in the tables below. The survey was anonymous
and conducted in group settings to ensure that the students knew their answers would
not be individually identified.
To avoid any potential priming effects, the survey first asked the students about
their political attitudes and opinions before measuring their exposure to ideological
425

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
Comparative Politics July 2015

and political education. The opinion questions include their satisfactions with China's
overall situation, performance of the central government, performance of the respon-
dent's local government, the Chinese government's competence in governance, and
China's political system, on a five-point Likert scale.41
To measure the students' appraisal of the government's strength in maintaining
political order, the survey asked them about the capacity of the government to "main-
tain social stability." The phrase "social stability" rather than "regime stability" was
used in order to reduce the political sensitivity of the question, but the term "maintain-
ing social stability" (weiwen) is broadly understood in China as a code word for main-
taining the stability of the existing regime. The issue of political sensitiveness also
made it impractical to measure this variable with multiple questions. In addition, the
survey asked about their willingness to protest and dissent, in other words, to partici-
pate in assemblies/demonstrations and in student strikes, which were the two primary
forms of rebellion in China's 1989 student movement. For obvious political concerns,
the students could not be directly and explicitly asked about "rebelling against the gov-
ernment" or "challenging the national regime,"42 but the wording used in the survey
was sufficient to tap into the students' inclination for political dissent.
Following the above questions on the dependent variables, the students were tested
by fifteen multiple-choice questions based on the two ideological and political education
courses they had taken in previous semesters. This represented their exposure to state
propaganda, the study's independent variable.43 The two courses were "Principles of
Marxism" and "Modern Chinese History," which focused on justifying and glorifying
the roots of China's current political system and its revolutionary path. The course on
modern Chinese history was not a standard history class but one with a significant ideo-
logical component, and the survey questions on that course were all about ideological
elements of the course rather than historical facts. Respondents were instructed to
answer these questions according to what they had learned in classes. All in all, the test
questions were similar to political knowledge questions in public opinion studies, except
that the "correct" answers were based on the official political discourse in China.
Two examples will help elucidate the nature of these test questions. One of the
questions asked: "What is the essence of elections in capitalist countries?" The choices
were (A) "division of power between the capitalist class and the proletarian class,"
(B) "expression of citizens' wishes and demands through electoral competition and
political participation," (C) "an important measure to mediate the interests and con-
flicts within the ruling class," and (D) "the principle that people are the masters of
their own country." The "correct" answer according to the standard official political
discourse in China is (C), with the implication that elections in the West are just
façades and, therefore, the lack of elections in China is not a bad thing. Another
question asked: "Among all the complicated contradictions and conflicts in modern
China, what was the most important one?" The choices were (A) "the conflict between
the proletarian class and the capitalist class," (B) "the conflict between feudalism and
the masses," (C) "the conflict between the peasant class and the landlord class," and
(D) "the conflict between imperialism and the Chinese nation." The "correct" answer
426

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
 Haifeng Huang

to this question is (D), which (implicitly) highlights the communist party's role in
achieving China's national independence.
A respondent's propaganda score, which reflected his or her exposure to and
familiarity with the ideological and political lecturing, was simply the number of ques-
tions answered correctly. The control variables included the students' general aca-
demic standings as discussed above, political efficacy, gender, family income, and CCP
membership.44 Age and education were not included because the respondents were all
college sophomores.
Table 1 shows the summary statistics of the survey. One thing that stands out is
that the respondents' mean political satisfactions were only around the midpoint of
the five-point scale, but their average willingness for political dissent was very low,
suggesting that the state has made them reluctant to participate in such activities even
though they were not really satisfied with the regime. With regards to the propaganda
scores, no one attained the perfect or almost perfect score (15 or 14) but some got zero,
which, again, indicates the students' general lack of interest in the subjects. The mean
and the mode of the distribution (7) were just below half of the total number of questions.

Table 1 Summary Statistics of the Survey

Wiable Mean Std. Dev. Min Max

Satisfaction with China's overall situation 2.97 0.93 1 5

Satisfaction with central government performance 3.36 0.90 1 5

Satisfaction with local government performance 2.81 1.02 1 5
Satisfaction with government competence 3.30 1.02 1 5
Satisfaction with political system 3.42 1.05 1 5
Evaluation of government capacity for social stability 3.59 1.05 1 5
Willingness to join assemblies and demonstrations 0.62 0.75 0 3

Willingness to join student strikes 0.61 0.77 0 3

Propaganda score 7.12 2.63 0 13

Academic standing 5.79 1.97 0 10

External efficacy 2.03 1.05 1 5

Internal efficacy 2.78 1.10 1 5

Female 0.42 0.49 0 1

Family income 4.56 1.59 0 10

CCP member 0.19 0.39 0 1

427

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
Comparative Politics July 2015

Results Given the ordinal nature of the dependent variables, I analyze the data with
ordered logit regressions (I have also run OLS regressions, with results consistent with
the theory). Table 2 shows the results on the respondents' satisfaction with China's
overall situation, performance of the central and local governments, government com-
petence, and the political system. As the table clearly shows, one's exposure to ideo-
logical and political education ("propaganda score") had no significant relationship

Table 2 Overall and Political Satisfactions

China Center Local Government Political

Overall Performance Performance Competence System

Propaganda Score -0.003 0.037 0.004 0.005 0.029

(0.024) (0.025) (0.025) (0.024) (0.024)
Academic Standing -0.045 0.024 -0.071** -0.018 -0.047
(0.031) (0.032) (0.032) (0.031) (0.031)
External Efficacy 0.205*** 0.282*** 0.293*** 0.273*** 0.198***
(0.057) (0.060) (0.061) (0.057) (0.057)
Internal Efficacy -0.234*** -0.130** -0.111** -0.249*** -0.160***
(0.054) (0.054) (0.056) (0.054) (0.054)
Female -0.015 -0.156 -0.074 -0.140 0.149

(0.114) (0.117) (0.119) (0.115) (0.114)

Family Income 0.091** -0.021 0.212*** 0.066* 0.033
(0.038) (0.039) (0.040) (0.038) (0.037)
CCP Member 0.442*** 0.481*** 0.027 0.244* 0.419***

(0.145) (0.150) (0.151) (0.144) (0.146)

Intercept 1 -2.683*** -2.526*** -1.174*** -2.880*** -2.643***
(0.324) (0.343) (0.323) (0.329) (0.327)
Intercept 2 -0.995*** -1.495*** 0.089 -1.515*** -1.644***
(0.308) (0.326) (0.319) (0.311) (0.312)
Intercept 3 0.826*** 0.594* 2.092*** 0.411 0.077
(0.307) (0.321) (0.326) (0.307) (0.307)
Intercept 4 3.670*** 3.261*** 4.074*** 2.102*** 2.062***
(0.356) (0.343) (0.357) (0.315) (0.315)
Observations 1089 1079 1014 1072 1089

Standard errors in parentheses. * p < 0.1, ** p < 0.05,

428

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
 Haifeng Huang

with any of the dependent variables that measure the respondents' overall and political
satisfactions. In other words, if the purpose of ideological and political education is to
make the students more pro-government (i.e., indoctrination), it has clearly failed.
Other variables worked as expected. External efficacy, one's feeling about their
ability to influence government decision making, was correlated with higher satisfac-
tions with the country and government. Internal efficacy, one's self-assessment of their
ability to understand political affairs, was correlated with lower satisfactions, which was
expected given China's authoritarian system. Naturally, family income and member-
ship in the communist party were often correlated with more positive opinions of the
country and government.
While propaganda has not elevated the students' satisfaction with the govern-
ment system, Table 3 shows that it has succeeded in signaling the regime's strength.
Respondents with higher propaganda scores had a higher belief that the government's
capacity in maintaining social stability was strong. The effects of this enhanced belief
are also apparent in Table 3. Higher propaganda scores were associated with a lower
willingness to dissent, particularly with regards to participating in student strikes. The
coefficient on participating in assemblies and demonstrations is not significant, but it
nevertheless has a negative sign (the OLS regression with the same variables would
return a statistically significant coefficient at p = 0.09). Overall, while the statistical
significance levels of the main coefficients in Table 3 are usually at p < 0.1, they con-
trast sharply with the results on political satisfactions in Table 2, where the p-values of
the coefficients on propaganda are often around 0.8 and 0.9. Given that the general
level of willingness to dissent was already low (see Table 1), it is remarkable that a
higher level of exposure to ideological and political lecturing could further dampen
any such inclination.
In results reported in the appendix, I also found that there was no correlation
between exposure to propaganda and willingness to vote in state-sanctioned local elec-
tions (e.g., village, local people's congress, and neighborhood elections) or elections
on campus. Participation in local elections, which is encouraged by the Chinese gov-
ernment, has been shown to be associated with identification with the regime and
affective attachments to the political authority.45 That propaganda had no effect on
regime-sanctioned forms of political participation but reduced people's willingness
to dissent is revealing.
The general results are thus consistent with the signaling theory of propaganda,
but not with the indoctrination theory, since political education made the respondents
believe that the government was strong and reduced their willingness to dissent, but did
not imbue them with a higher level of pro-regime sentiments. To use the game-theoretic
terminology, political and ideological propaganda in Chinese colleges does not change
the students' political "tastes" or "preferences," but does influence their belief about the
"state of the world." Given that the ideological and political education courses focus on
justifying and glorifying the rule of the communist party, rather than touting the state's
social control or repressive capacity, this outcome would be otherwise surprising and
confusing, but is natural within the signaling framework.
429

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
Comparative Politics July 2015

Table 3 Evaluation of the Government's Strength and Willingness to Dissent

Government Capacity Assembly and Student

for Social Stability Demonstration Strike
Propaganda Score 0.043* -0.035 -0.046*
(0.025) (0.025) (0.025)
Academic Standing 0.008 0.071** 0.040
(0.033) (0.033) (0.033)
External Efficacy 0.148** 0.205*** 0.219***
(0.061) (0.059) (0.060)
Internal Efficacy -0.157*** 0.070 -0.019
(0.058) (0.056) (0.056)
Female -0.431*** -0.339*** -0.355***

(0.122) (0.121) (0.122)

Family Income 0.010 -0.017 -0.003
(0.041) (0.040) (0.040)
CCP Member 0.334** -0.275* -0.446***

(0.156) (0.155) (0.159)

Intercept 1 -2.910*** 0.524 0.151
(0.354) (0.324) (0.320)
Intercept 2 -1.844*** 2.430*** 1.959***
(0.333) (0.334) (0.327)
Intercept 3 -0.098 4.671*** 3.964***
(0.327) (0.405) (0.382)
Intercept 4 1.531***
(0.331)
Observations 937 1083 1084

Standard errors in parentheses. * p <

Concerns and Alternative Ex

potential reverse causality: the
the strength of the state to a h
than the other way around. Fo
are interested in a career in th
grades from the political educa
the most important political ex
430

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
 Haifeng Huang

CCP membership, which has been controlled for in the statistical analysis. Grades
from the ideological and political education courses themselves do not really matter
aside from affecting the GPA. That is why Chinese students, including party members,
generally regard ideological and political education as "useless" and consider related
coursework as the least important of their college career. In fact, student party mem-
bers, who had a higher incentive for government and party jobs, did not score higher
in the propaganda test than non-party members. The party members' average score
was 7.38 (standard deviation = 2.404), while the non-party members' average score
was 7.33 (standard deviation = 2.399).46 In addition, the important annual civil service
exams, which applicants for government jobs must take, primarily test intellectual
aptitude along the lines of that measured by the United States' GRE and LS AT rather
than conformity to state ideology; they cover Chinese language, mathematics, logic,
general knowledge, basic data analysis, and writing, but very little, if any, material
from college political education courses.
A related alternative explanation is that belief in and fear of the state's social control
capacity will induce more attention to ideological and political education. This concern
is also not warranted. As I have discussed earlier, Chinese students' incentive in political
education courses is to meet the degree requirement and attain a high overall academic
standing. Low grades in these courses will not result in any negative outcome other than
a low GPA. Satisfaction with and loyalty to the government, on the other hand, may
indeed provide a student with more incentive to hear what the state has to say, but
as Table 2 has clearly shown, there is little relationship between the students' exposure
to political lecturing and their satisfaction with the regime. Any potential effect of
one's fear of the government's strength on willingness to receive state propaganda will
therefore be even weaker.
Another type of concern is due to the observational rather than experimental nature
of the data. Ideally, to examine the effects of the students' exposure to propaganda, one
should run an experiment in which some students are randomly assigned to a year
of political education, while others do not receive the treatment, and then compare
the two groups. Clearly this is not feasible. As I have discussed above, however, even
though these political education classes are mandatory, students' actual attention
to them over the course of several semesters varied considerably, which reflects their
differing incentives (and abilities) to achieve good overall academic standing, as well
as other random and idiosyncratic reasons. In other words, even though they have all
received the propaganda treatment, the amount of treatment varies from one student
to another. Controlling for their overall academic standing as well as demographic
and political factors such as CCP membership, the students' ability to recall the
lecturing reflects the varying amount of treatment they have received due to these
random and exogenous factors, which can then be used to estimate the effects of
exposure to propaganda.
One may also wonder if the propaganda scores actually measured the extent to
which the students were convinced by the propaganda, rather than their exposure to
propaganda. But if that were the case, students with higher propaganda scores should
431

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
Comparative Politics July 2015

be more satisfied with the government, unlike the results in Table 2. For another exam-
ple, the survey shows that most of the students preferred Western political systems
over the Chinese system, a finding that is echoed in other surveys.47 In a question
(before the propaganda test) asking whether they agreed or disagreed with the state-
ment that "Western political systems are very appropriate for our country," a total of
73.2 percent of the students answered "agree" or "somewhat agree," while only 7.3 per-
cent answered "disagree" or "somewhat disagree," with the remaining choosing "neither
agree nor disagree." It is unlikely, therefore, that the 57.4 percent of students who "cor-
rectly" answered the political education test question about the essence of elections in the
West truly believed that democratic elections were merely a "measure to mediate the
interests and conflicts within the ruling class." Rather, they were completing that section
of the survey according to the section instruction, which was to select answers according
to what they had been taught regardless of their personal opinion.
Another concern is that the propaganda scores may simply reflect the students'
memorization ability. This concern will not affect the validity of the results. First of
all, much of the cognitive ability, including memory, is already reflected in the students'
academic standing, given that Chinese education has a well-known emphasis on memo-
rization. Secondly, even if there are aspects of one's memorization ability not captured
by academic standing, having remembered more of the state's propaganda is func-
tionally equivalent to having had more exposure to the propaganda. In either case,
the lecturing has similar cognitive effects on the recipients.
Still another concern is that even though political education does not improve the
students' satisfaction with the government and the political system, it might have effects
on other types of political attitudes, for example, fostering higher levels of attachment to
the state for nationalist reasons. It should be noted that I am not arguing that political
and ideological education does not have any conceivable indoctrination effect; the
survey did not exhaust all possible political attitude questions. The evidence does show,
however, that political education does not improve the students' views of the govern-
ment and the regime in a general sense. Regarding the specific question of nationalist
or pro-China sentiments, one of the survey questions asked the students the extent to
which they agreed with the following statement: "We should strive to maintain our
own institutions, culture, and way of life, rather than becoming more and more like
other countries." Ordered logit analysis with the same control variables as in the above
statistical tables shows that there was little relationship between propaganda scores
and pro-China sentiments (see the appendix).

Conclusion

This paper has shown through the use of a signaling model that a sufficient amount o
propaganda can serve to demonstrate a regime's strength in maintaining social control
and political order, thus deterring citizens from challenging the government, even if th
content of the propaganda itself does not induce pro-government attitudes or values
432

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
 Haifeng Huang

This can explain why authoritarian governments are willing to spend an enormous
amount of resources on propaganda activities, the content of which often does not
persuade the intended recipients. A unique survey dataset from China lends support
to the signaling theory of propaganda. Students with more exposure to state propaganda
in the form of ideological and political education are not more satisfied with the govern-
ment and state, but they are more likely to believe that the state is strong in maintaining
political order and social stability, and are less willing to engage in political dissent. In
other words, the ideological and political lecturing they receive does not persuade them
about the greatness of the state but does succeed in warning them about the capacity of
the regime and hence the likely futileness of challenging it.
One may wonder why the Chinese government would signal its strength through a
socially wasteful activity, rather than through investing in useful public projects and
infrastructure building, or improving citizens' livelihood. Political propaganda is cer-
tainly not the only mechanism through which a regime sustains its rule, and the Chinese
government has indeed been doing all of the above. However, that does not mean that
the government can do without political education in schools. In fact, the inevitable
slowing down of economic growth in China means that the Chinese government cannot
simply rely on performance legitimacy to sustain its rule.48 Given the central role that
young people, especially students, play in political crises, signaling the state's social
control capacity to them may actually become more important.
Aside from the above survey evidence and the existence of seemingly outdated
state news programs such as Xinwen Lianbo, the pattern of political challenges to the
Chinese regime during the last three decades is also largely consistent with the theory
here. During the 1980s, China's first decade of reform, a notable relaxation of propa-
ganda work and ideological control accompanied the deterioration of state capacity and
a series of pro-democracy protests aimed at the entire government system, culminating
in the 1989 student-led Tiananmen movement that almost toppled the regime. This
development prompted Deng Xiaoping 's comment about the state's biggest failure of
the decade being ideological and political education. Since the 1990s, after the state
reasserted its propaganda work, there have been no large-scale political challenges to
the regime aside from some ethnic and religious-based conflicts.49 The vast majority
of social protests in China have been localized loyalist ones calling for the government
to deal with certain specific issues or punish corrupt local officials.50 To be sure, pro-
paganda is just one of many social factors and mechanisms that affect political dissent
in a country, and so the above pattern is only suggestive. Nevertheless, the association
between the intensity of propaganda work and the absence of political revolt against
the regime since the 1990s, when the country's economy has become increasingly
integrated with the global market, information and ideas flowing around the society
are increasingly pluralistic, and social grievances are on the rise, is telling.
The signaling theory of propaganda should be regarded as a complement to rather
than a substitute for the traditional indoctrination theory of propaganda. The goal of
the article is not to replace the indoctrination theory, but to point out that the purpose
of propaganda is often not limited to influencing the recipients' opinion; signaling the
433

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
Comparative Politics July 2015

government's strength, capacity, and resources and therefore intimidating the masses
is sometimes a more important goal.
This finding sheds important light on the issue of whether and how propaganda can
be effective. I will continue to focus on the Chinese case for concreteness. In recent
years there has been a debate among scholars about whether the Chinese government's
propaganda has achieved its intended goal or has been ineffective or even counter-
productive. On the one hand, Chen and Shi argue that propaganda has alienated Chinese
citizens and, using a 1993-94 national survey (not long after the Tiananmen move-
ment), show that those with more exposure to state media reports actually trust the
government and the Chinese political system less.51 Tong's case study of the publish-
ing industry during the anti-Falun Gong campaign shows that the Chinese government
has an impressive capacity in quickly producing a large amount of propaganda mate-
rials in the face of challenges, but the public appeal of such publications is meager.52
Lynch and Latham argue that the in-flow of a multitude of alternative information
from global and other non-official sources has significantly increased fatigue and skep-
ticism toward state-sponsored communications.53 Chan and Rosen focus on political
education in schools rather than propaganda in the media and publishing industry
and show with a series of reports from Chinese newspapers and journals that the
renewed emphasis on political lecturing following the 1989 Tiananmen movement
has little effect on the students' belief systems; in fact, Chinese students' admiration
and acceptance of the American political system often far exceeds that for the Chinese
party-state model.54
On the other hand, Stockmann and Gallagher find from surveys conducted in
four Chinese cities in 2005 that Chinese media's propaganda about ordinary people's
positive experiences in the legal system contributes to regime legitimacy and encour-
ages citizens to participate in the legal system.55 Kennedy, using the 2000 World
Value Survey, shows that media exposure increases Chinese citizens' satisfaction with
the national leadership. With regard to education, he finds that primary and junior
high school education has positive effects on rural residents, although those with
higher levels of education can resist state propaganda and display lower satisfactions.56
Tang, using a 1999 six-city survey, also shows that media exposure increases the
respondents' nationalism and support for China's political system.57 Yang and Tang
use a 2004 national survey and similarly find that media exposure increases Chinese
citizens' trust in the country's political institutions.58
Note that the negative quantitative evidence discussed above is either about
ideological and political education, or about media effect in the wake of major anti-
government movements, when the regime is more focused on deterrence and preven-
tion than persuasion. The more positive results, on the other hand, are from surveys
about media effects during more recent and "normal" times. With the continued
reform in the media sector, Chinese media have become much more commercialized
and diversified than those of earlier times. As a result, Chinese media reports are now
significantly more lively and interesting than before, and less preposterous or distorted
than political lecturing in schools (except for important anomalies such as Xinwen
434

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
 Haifeng Huang

Lianbo). Even though the Chinese state still ensures that media serve political purposes
by synchronizing media messages and preventing conflicting communication from being
widely circulated, the official messages conveyed in Chinese media are now much more
subtle, sophisticated, and credible.59 We can call these more subtle and persuasive mes-
sages "soft propaganda," while those pretentious and dogmatic propaganda in ideo-
logical and political courses as well as media programs and publications following
major anti-government movements can be termed "hard propaganda." The aforemen-
tioned studies have demonstrated that soft propaganda can indeed influence people's
political and social opinions, while hard propaganda will not and may even backfire.
The current study shows, however, that even though hard propaganda does not
influence the masses' opinions and attitudes toward the government, it can neverthe-
less be effective in influencing their behavior and promoting regime stability. By sur-
rounding the society with pompous and resource-consuming propaganda messages and
activities, the regime signals its strength in social control and capacity to meet poten-
tial challenges. Both soft and hard propaganda, therefore, can be effective tools for
an authoritarian government, with the former changing social attitudes and the latter
deterring political opposition. This article represents an endeavor to theoretically char-
acterize hard propagànda with support from empirical evidence, but much additional
work is needed. For example, how do the indoctrination goal and the deterrence goal
of propaganda work together in the authoritarian setting? And what is the relationship
between propaganda, both hard and soft, and other forms of social control? These and
other questions are left for future research.

NOTES

I am very grateful to Li Zhi for help with the data collection. I would also like to thank John J
Daniela Stockmann, Kharis Ali Templeman, and two anonymous reviewers for helpful com
financial support was provided by the Pacific Rim Research Program of the University of Calif

1. Perry Link, Evening Chats in Beijing: Probing China's Predicament (New York and L
Norton & Company, 1992).
2. James Fallow, "Their Own Worst Enemy," The Atlantic Monthly (November 2008).
3. Xueyi Chen and Tianjin Shi, "Media Effects on Political Confidence and Trust in the Peopl
of China in the Post-Tiananmen Period," East Asia : An International Quarterly, 19 (Fall 200
4. Anne-Marie Brady, Marketing Dictatorship: Propaganda and Thought Wbrk in Contem
(Lanham, MD: Rowman & Littlefield, 2008); Heike Holbig and Bruce Gilley, "Reclaiming Le
China," Politics & Policy ; 38 (June 2010), 395-422; Stanley Rosen, "The Effect of Post-4 Jun
Campaigns on Chinese Students," China Quarterly ; 134 (June 1993), 310-34; David Shambaugh,
paganda System: Institutions, Processes and Efficacy," The China Journal , 57 (January 2007),
5. Lisa Wedeen, Ambiguities of Domination: Politics, Rhetoric, and Symbols in Contem
(Chicago: University of Chicago Press, 1999).
6. Reuters, "North Korea Reports 'Bright Halo' over Leader's Birth-place," Feb. 17, 2011
http://www.reuters.com/article/20 1 1/02/1 7/us-korea-north-halo-idUSTRE7 1 G 1 YR20 1 1 02 1 7.
7. B. R. Myers, The Cleanest Race: How North Koreans See Themselves and Why It Matter
NY: Melville House, 2010).
8. '&clav Havel, The Power of the Powerless: Citizens against the State in Central-Eas
(Armonk M.E. Sharpe, 1985).
435

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
Comparative Politics July 2015

9. Associated Press, "Vietnam Seeks to Lure Students to Study Marxism with Free Tuition," New York
Times , Aug. 15, 2013.
10. Susan L. Shirk, ed., Changing Media, Changing China (New York: Oxford University Press, 2010);
Daniela Stockmann, Media Commercialization and Authoritarian Rule in China (New York: Cambridge
University Press 2013); 'liezhi Zhao, Media, Market, and Democracy in China: Between the Party Line
and the Bottom Line (Urbana: University of Illinois Press, 1998); Zhou He, "Chinese Communist Party
Press in a Tug-of-War: A Political-Economy Analysis of the Shenzhen Special Zone Daily," in Chin-Chuan
Lee, ed., Power; Money, and Media: Communication Patterns and Bureaucratic Control in Cultural China
(Evanston: Northwestern University Press, 2000).
11. For example, a widely circulated joke notes that in Xinwen Lianbo's reports "no opening (of
meetings) is not solemn, and no closing is not victorious; no speech is not important, and no applause
is not enthusiastic; no leader is inattentive, and no visit is not cordial..." See http://www.wangxiaofeng.net/
?p=1731 (last accessed April 6, 2013). A Google search for " Xinwen Lianbo Xiaohua " in Chinese (jokes
about Xinwen Lianbo) returned about 1,510,000 results, while a search in Baidu, China's dominant Internet
search engine, returned about 3,150,000 results.
12. Jiang Zhan, "Xinwen Lianbo Yinggai Zenme Gai" (How Should Xinwen Lianbo Be Changed), China
Newsweek , Sept. 28, 2011, available at http://viewpoint.inewsweek.cn/columns/columns_detail.php7id =481.
13. It can also give many citizens a sense of social stability amid China's rapid social transformations.
14. Stockmann, 2013.
15. For a review, see Andrea Prat, "Rational 'bters and Political Advertising," in Barry R. Weingast
and Donald A. Wittman, eds., The Oxford Handbook of Political Economy (New York: Oxford University
Press, 2006).
16. Due to space constraints, the Appendix is not in the print version of this article. It can be viewed in the
online version, at www.ingentaconnect.com/cuny/cp.
1 7. Barbara Geddes and John Zaller, "Sources of Popular Support for Authoritarian Regimes," American
Journal of Political Science , 33 (May 1989), 319-47.
18. Peter Kenez, The Birth of the Propaganda State: Soviet Methods of Mass Mobilization, 1917-1929
(Cambridge: Cambridge University Press, 1985), 4.
19. Walter Lippmann, Public Opinion (New York: Macmillan, 1922); Harold D. Lasswell, "The Theory of
Political Propaganda," American Political Science Review ; 21 (August, 1927), 627-31.
20. Geddes and Zaller, 1989; Kenez, 1985; Fred S. Siebert, Theodore Peterson and Wilbur Schramm, Four
Theories of the Press (Urbana: University of Illinois Press, 1956).
21. Edward S. Herman and Noam Chomsky, Manufacturing Consent: The Political Economy of the Mass
Media (New York: Pantheon Books, 1988).
22. Brady, 2008; Daniel C. Lynch, After the Propaganda State: Media, Politics, and " Thought M brk" in
Reformed China (Stanford: Stanford University Press, 1999).
23. Brady, 2008; Holbig and Gilley, 2010; Stockmann, 2013.
24. Michael Spence, "Job Market Signaling," Quarterly Journal of Economics, 87 (August, 1973), 355-74.
25. I thank an anonymous reviewer for urging me to make this point explicit.
26. Hannah Arendt, The Origins of Totalitarianism (New 'brk: Harcourt, Brace, Jovanovich, 1973), 361, 351.
27. Havel, 1985, 35.
28. For a similar point on language formalization in Chinese official speeches, see Michael Schoenhals,
Doing Things with Vòrds in Chinese Politics: Five Studies (Berkeley: Center for Chinese Studies, University
of California, 1992).
29. Wedeen, 1999, 147, italic added.
30. Timur Kuran, "Now out of Never: The Element of Surprise in the East European Revolution of 1989,"
Wbrld Politics , 44 (October 1991), 7-48.
3 1 . Bruce Bueno de Mesquita, Alastair Smith, Randolph M. Siverson, and James D. Morrow, The Logic of
Political Survival (Cambridge: The MIT Press, 2003); Jennifer Gandhi, Political Institutions under Dictatorship
(New York: Cambridge University Press, 2008); Roger B. Myerson, "The Autocrat's Credibility Problem and
Foundations of the Constitutional State," American Political Science Review ; 102 (February 2008), 125-39;
Milan Svolik, The Politics of Authoritarian Rule (New York: Cambridge University Press, 2012).
32. Georgy Egorov, Sergei Guriev, and Konstantin Sonin, "Why Resource-poor Dictators Allow Freer
Media: A Theory and Evidence from Panel Data," American Political Science Review ; 103 (2009), 645-68;
Peter Lorentzen, "China's Strategic Censorship," American Journal of Political Science , 58 (April 2014),
402-14. Some contemporary authors include media censorship as part of propaganda. In this article I use
propaganda for its original meaning, i.e., it propagates rather than withholds a message.

436

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms
 Haifeng Huang

33. Chris Edmond, "Information Manipulation, Coordination, and Regime Change," Review of Economic
Studies , 80 (October 2013), 1422-58.
34. Victor Shih, "Nauseating Displays of Loyalty: Monitoring the Factional Bargain through Ideological
Campaigns in China," Journal of Politics, 70 (October 2008), 1177-92.
35. Haifeng Huang, "Signal Left, Turn Right: Central Rhetoric and Local Reform in China," Political
Research Quarterly, 66 (June 2013), 292-305.
36. Deng Xiaoping, "Speech Made while Receiving Cadres of the Martial Law Units in the Capital at and
above the Army Level," June 9, 1989, available at http://cpc.people.com.cn/GB/69112/69113/69684/69696/
4950038.html, last accessed April 6, 2013.
37. Rose, 1993; Che-po Chan, "The Political Pragmatism of Chinese University Students: 10 Years after the
1989 Movement," Journal of Contemporary China , 8 (1999), 381-403.
38. Chris Buckley, "China Takes Aim at Western Ideas," New York Times , Aug. 19, 2013.
39. In a survey of students at five of Beijing's best universities, only 2.2 percent of the students rated the
political education courses highly, with another 15.8 percent saying they were somewhat good, see Shengluo
Chen, "Survey Study on Chinese University Students' Perceptions of the Political Systems of China and the
United States," Chinese Education and Society, 44 (Mar-Apr/May-Jun 2011), 13-57.
40. The political education courses in the survey university are often divided into several modules, with each
module taught by the same instructor throughout the university. Teacher quality, therefore, is largely controlled.
41. Survey question wordings are in the appendix.
42. In several ways the current survey was already more politically delicate than most academic surveys
in China.
43. Alternatively, the respondents could be asked to report their grades from those courses, but that would
lead to concerns about the anonymity of the survey.
44. Because of the students' relatively young ages, CCP probationary members, who normally become full
members after one year of trial membership, were also counted as members in this survey.
45. Jie Chen and Yang Zhong, "Why Do People 'òte in Semicompetitive Elections in China?" Journal of
Politics, 64 (February 2002), 178-97.
46. These scores are slightly different from those in Table 1 because some students did not report whether
they are party members.
47. Stanley Rosen, "Chinese Youth and State-Society Relations," in Peter Hays Gries and Stanley Rosen,
eds., Chinese Politics: State, Society and the Market (Routledge, 2010).
48. Dingxin Zhao, "The Mandate of Heaven and Performance Legitimation in Historical and Contemporary
China," American Behavioral Scientist, 53 (November 2009), 416-33.
49. The 1999 Falun Gong movement was based on religious/spiritual rather than political grounds. The
Charter 08 movement in 2008 was largely restricted to a few hundred intellectuals. The so-called "Jasmine
Revolution" in 2011 attracted far more police officers (and journalists) on the streets of Beijing and Shanghai
than actual protesters; see, for example, Tania Branigan, "China's Jasmine Revolution: Police but no Protesters
Line Streets of Beijing," The Guardian, Feb. 27, 2011.
50. Xi Chen, Social Protest and Contentious Authoritarianism in China (New York: Cambridge University
Press 2011); Peter Lorentzen, "Regularizing Rioting: Permitting Public Protest in an Authoritarian Regime,"
Quarterly Journal of Political Science, 8 (April 20 13), 127-58; Kevin O'Brien, "Rightful resistance," Wbrld
Politics, 49 (October 1996), 31-55.
51. Chen and Shi, 2001.
52. James W. Tong, "Publish to Perish: Regime Choices and Propaganda Impact in the Anti-Falungong
Publications Campaign, July 1999-April 2000 "Journal of Contemporary China, 14 (August, 2005), 507-23.
53. Lynch, 1999; Kevin Latham, "Nothing but the Truth: News Media, Power and Hegemony in South
China," China Quarterly, 163 (September 2000), 633-54.
54. Chan, 1999; Rosen, 2010.
55. Daniela Stockmann and Mary E. Gallagher, "Remote Control: How the Media Sustain Authoritarian
Rule in China," Comparative Political Studies, 44 (April 2011), 436-67.
56. John James Kennedy, "Maintaining Popular Support for the Chinese Communist Party: The Influence of
Education and the State-Controlled Media," Political Studies, 57(October 2009), 517-36.
57. Wenfang Tang, Public Opinion and Political Change in China (Stanford: Stanford University Press, 2005).
58. Qing Yang and Wenfang Tang, "Exploring the Sources of Institutional Trust in China: Culture,
Mobilization, or Performance?" Asian Politics & Policy, 2 (July/September 2010), 415-36.
59. Stockmann and Gallagher, 2011; Stockmann, 2013.

437

This content downloaded from

109.68.126.147 on Fri, 10 Mar 2023 08:45:04 UTC
All use subject to https://about.jstor.org/terms