See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/254967941

Passive and Assertive Secularism: Historical Conditions,

Ideological Struggles, and State Policies toward Religion

Article in World Politics · August 2007

CITATIONS READS

12 1,409

1 author:

Ahmet T. Kuru
San Diego State University
33 PUBLICATIONS 438 CITATIONS

SEE PROFILE

All content following this page was uploaded by Ahmet T. Kuru on 26 March 2020.

The user has requested enhancement of the downloaded file.

Vol. 59 • July 2007 • No. 4

CO NT ENT S

Classing Ethnicity: Class, Ethnicity, and

the Mass Politics of Taiwan’s Democratic
Transition David D. Yang 503

Civil Society and the Legacies of Michael Bernhard and

Dictatorship Ekrem Karakoç 539

Passive and Assertive Secularism:

Historical Conditions, Ideological
Struggles, and State Policies toward Religion Ahmet T. Kuru 568

Heredity Succession in Modern Autocracies Jason Brownlee 595

Review Article
The Emerging Regional Architecture of
World Politics Amitav Acharya 629

Index to Volume 59 653

The Contributors ii

Abstracts iii

Referees 2006 v
 Passive and Assertive
Secularism
Historical Conditions, Ideological Struggles,
and State Policies toward Religion
By Ahmet T. Kuru*

O n December 11, 2003, the Stasi Commission, including twenty

French academics and intellectuals, submitted a report on secular-
ism to President Jacques Chirac. The French executive and legislators
embraced the commission’s recommendation of a law to prohibit stu-
dents’ religious symbols in public schools. While the primary target of
this new law was the Muslim headscarf, it was also extended to cover
Sikh turbans, Jewish skullcaps (kippot), and “large” Christian crosses. A
week after the Stasi Report was issued, the United States Department
of State released its 2003 Report on International Religious Freedom.
At the accompanying press conference, Ambassador John Hanford an-
swered the following questions:

Question: What was your reaction to President Chirac’s headscarf ban?

Ambassador: [A] fundamental principle of religious freedom that we work
for in many countries of the world, including on this very issue of headscarves, is
that all persons should be able to practice their religion and their beliefs peace-
fully without government interference. . . . President Chirac is concerned to
maintain France’s principle of secularism and he wants that, as I think he said,
not to be negotiable. Well, of course, our hope is religious freedom will be a non-
negotiable as well. One Muslim leader said this is a secularism that excludes too
much. . . . [A] number of countries . . . restrict headscarves . . . where people
are wearing these with no provocation, simply as a manifestation of their own
heartfelt beliefs, that we don’t see where this causes division among peoples.
Question: You’re referring to Turkey, yes?
Ambassador: Turkey would be another country, yes.1

* The author thanks Joel Migdal, Anthony Gill, Stephen Hanson, Reşat Kasaba, Christopher
Soper, Jeremy Gunn, and the anonymous reviewers for their helpful comments.
1
“Release of the 2003 Annual Report on International Religious Freedom,” December 18, 2003,
http://www.state.gov/s/d/rm/27404pf.htm (accessed April 24, 2004).

World Politics 59 ( July 2007), 568–94

 pa s s i v e & a s s ert i v e s e c u l a r i s m 569

As the ambassador stresses, there is a sharp policy distinction be-

tween the U.S., which allows students to wear religious garments and
symbols; France, which bans such symbols in public schools; and Tur-
key, which prohibits them in all educational institutions. What is puz-
zling about these cases is that although each has a different policy on
student displays of religious symbols in schools, all three are “secular
states,” as defined by two main criteria: (1) their legal and judicial pro-
cesses are out of institutional religious control, and (2) they establish
neither an official religion nor atheism.2 Other states have established
religious laws and courts as the basis of their legal and judicial systems
(religious states), have recognized one official religion (states with an
established religion), or have shown official hostility toward religions,
generally by establishing atheism (antireligious states).3 Table 1 differ-
entiates among these four sorts of states in terms of their relationships
to religion.
Despite their secular status, the U.S., France, and Turkey have, in
fact, been deeply concerned with religion and have engaged it on many
fronts. The different approaches of these three states regarding the
wearing of headscarves reflect a broad array of policy differences among
them. Historical and contemporary debates on secularism in all three
cases have pointed to education as the main battlefield in state-religion
controversies.4 The debates on secularism in these countries have gen-
erally focused on schools, since struggling groups desire to shape the
young generation’s worldview and lifestyle. I therefore analyze six of
the most publicly debated state policies toward religion in schools to
reveal general policy tendencies. These are policies on (1) student re-
ligious dress and the display or wearing of religious symbols in public
schools; (2) pledges recited in public schools, (3) private religious edu-
cation; (4) religious instruction in public schools; (5) public funding of
private religious schools; and finally (6) prayer in public schools.
Despite the dynamism of the policy-formation process, states still
follow distinct and relatively stable trajectories in their general policies
2
Many scholars emphasize two other dimensions while defining a secular state: (1) separation of
church/mosque and state and (2) religious freedom. See D. E. Smith, “India as a Secular State,” in
Rajeev Bhargava, ed., Secularism and Its Critics (Delhi: Oxford University Press, 1999), esp. 178–83. A
complete separation is, in fact, neither constitutionally declared in many secular states nor a practical
issue. Religious freedom, by contrast, is both constitutionally declared and practical; yet, it is neither
necessary nor sufficient to be secular for a state to provide religious freedom.
3
By religion, I imply a set of beliefs and practices that refer to supernatural beings, generally God.
In this definition, neither atheism nor an ideology like Marxism is a religion.
4
Stephen V. Monsma and J. Christopher Soper, The Challenge of Pluralism: Church and State in
Five Democracies (Lanham, Md.: Rowman and Littlefield Publishers, 1997); Guy Haarscher, La laïcité
(Paris: puf, 2004); Niyazi Berkes, The Development of Secularism in Turkey (1964; New York: Rout-
ledge, 1998).
570 w o r l d p o li t i c s

Table 1
Types of State-Religion Regimes
Religious State with an Secular Antireligious
State Established Religion State State

Legislation and
judiciary religion-based secular secular secular
State’s attitude officially officially officially officially hostile
toward religions favors one favors one favors none to all or many
Examples Vatican Greece U.S. China
Iran Denmark France North Korea
Saudi Arabia England Turkey Cuba
Number in the world 10 100 95 22
Sources: Constitutions of the example states; David B. Barrett, George T. Kurian, and Todd M.
Johnson, World Christian Encyclopedia: A Comparative Survey of Churches and Religions in the Modern
World (New York: Oxford University Press, 2001); International Religious Freedom Report 2006, U.S.
Department of State, http://www.state.gov/g/drl/rls/irf/ (accessed May 1, 2006); James Edward
Wood, Church-State Relations in the Modern World (Waco: Baylor University, 1998), 81–88.

toward religion. There is a sharp qualitative distinction between state

policies toward religion in the U.S. and those in France and Turkey. In
America, students are allowed to display religious symbols and recite
the pledge of allegiance, which includes the statement “one nation, un-
der God.” In France and Turkey, however, the state pursues totally op-
posite policies on these two points. Even on other policy issues, there is
a more positive tone toward religion in the U.S., in contrast to the two
other cases. Religious instruction in Turkish high schools is directly
related to the state’s desire to control religion and the fact that private
religious education is prohibited. Similarly in France the state funds
religious private schools as long as these schools sign a contract to ac-
cept certain state control over them. On the surface, the ban on school
prayer seems similar in the three cases. Yet an in-depth analysis reveals
a distinction. In France and Turkey the ban is justified mainly on the
grounds that prayer contradicts the principle of secularism and the sec-
ular character of the public school. In the U.S., however, an important
rationale is that school prayer implies a “psychological coercion” over
students with minority religious beliefs.5 Table 2 compares the three
cases in terms of their application of these six policies.
Beyond these specific policies in schools, the three cases also show
two opposite attitudes toward religion in the public sphere. In the U.S.
one finds official public visibility of religion, which is not the case in
5
Lee v. Weisman, 505 U.S. 577 (1992); Santa Fe v. Doe, 530 U.S. 290 (2000).
 pa s s i v e & a s s ert i v e s e c u l a r i s m 571

Table 2
State Policies toward Religion in Schools
Ban on A Pledge
Students’ Referring State
Religious to God Ban on Religious Funding of Ban on
Symbols Recited Private Instruction Religious Organized
in Public in Public Religious in Public Private Prayer in
Schools Schools Education Schools Schools Public Schools
U.S. no yes no no no yes
France yes no no no yes yes
Turkey yes no yes yes no yes

France or Turkey. “In God We Trust” appears on all American currency.

Many official oaths, including the swearing in of the president, custom-
arily contain the statement “so help me God” and are often made by plac-
ing the left hand on a Bible. Sessions of the U.S. Congress begin with a
prayer by a chaplain, and the sessions of the Supreme Court start with
the invocation: “God save the United States and this Honorable Court.”
Such public religious discourse does not exist in France and Turkey.
These differences point to my central question: why are American
state policies inclusionary toward public visibility of religion whereas
policies in France and Turkey are largely exclusionary? Stated differ-
ently, the dependent variable of this work is the variation in policies on
religion, particularly two opposite policy tendencies, as found in three
secular states.
I argue that state policies toward religion are the result of ideological
struggles. In the three cases it is the struggle between “passive secular-
ists” and “assertive secularists” that has shaped public policies. Passive
secularism, which requires that the secular state play a “passive” role
in avoiding the establishment of any religions, allows for the public
visibility of religion. Assertive secularism, by contrast, means that the
state excludes religion from the public sphere and plays an “assertive”
role as the agent of a social engineering project that confines religion
to the private domain.6 Thus, passive secularism is a pragmatic political
principle that tries to maintain state neutrality toward various religions,
whereas assertive secularism is a “comprehensive doctrine” that aims to
eliminate religion from the public sphere.7
6
See Charles Taylor, “Modes of Secularism,” in Bhargava (fn. 2); Wilfred M. McClay, “Two Con-
cepts of Secularism,” in Hugh Heclo and Wilfred M. McClay, eds., Religion Returns to the Public
Square: Faith and Policy in America (Baltimore: Johns Hopkins University Press, 2002).
7
For comprehensive doctrines, see John Rawls, Political Liberalism (New York: Columbia Univer-
sity Press, 1996).
572 w o r l d p o li t i c s

In France and Turkey the assertive secularists are dominant despite the
challenge posed by the passive secularists. In the U.S., however, the asser-
tive secularists are so marginal that they cannot mount a serious challenge
to the dominant passive secularists; the real struggle occurs between the
two different understandings of passive secularism as elaborated below.
Passive and assertive secularism became dominant in these cases as a result
of particular historical conditions during their secular state-building peri-
ods. In France and Turkey the presence of an ancien régime based on the
marriage of monarchy and hegemonic religion was a crucial reason for the
emergence of anticlericalism among the republican elite. The antagonistic
relations between the republicans and the religious institutions underlay
the historical dominance of assertive secularism. America, however, was
a new country of immigrants that lacked an ancien régime. Therefore
secular and religious elites sought and achieved an overlapping consen-
sus on the separation of church and state at the federal level. The result
was the dominance of passive secularism. This historical explanation
completes my argument summarized in Figure 1.
The article proceeds as follows. First, I discuss alternative theoretical
approaches that would explain the puzzle differently. I then elaborate
my own explanation based on ideological struggles in the three cases.
Finally, I analyze the historical reasons for the dominance of a certain
secular ideology in a particular country.

Alternative Theories: Modernization,

Civilization, and Rational Choice
Modernization theory, the civilizational approach, and rational choice
theory are three important theories that scholars reference in analyz-
ing state policies. Modernization theory has different versions. Some
scholars emphasize the epochal impact of modernization to explain
the transformation of medieval sociopolitical systems into modern sys-
tems.8 They therefore offer important insights about the emergence of
secular states. Yet their broad perspectives do not provide parsimonious
explanations for particular state policies. For that reason, I focus on
the parsimonious version of modernization theory, which claims that
economic development is the determining factor in the transformation
of traditional societies into modern societies.9
8
Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism
(London: Verso, 1998); Ernest Gellner, Nations and Nationalism (Ithaca, N.Y.: Cornell University Press,
1983); Charles Taylor, Modern Social Imaginaries (Durham, N.C.: Duke University Press, 2004).
9
Ronald Inglehart, Modernization and Postmodernization: Cultural, Economic, and Political Change
in 43 Countries (Princeton: Princeton University Press, 1997).
 pa s s i v e & a s s ert i v e s e c u l a r i s m 573

I II III
The Presence or Absence Ideological Struggles with Exclusionary or
of an the Dominance of Inclusionary Policy
Ancien Régime Assertive or Passive Tendencies
Based on Monarchy and Secularism toward
Hegemonic Religion Religion

Figure 1
Explanatory and Dependent Variables

From a sociological perspective, modernization theory includes sec-

ularization theory, which regards religion as a traditional phenomenon
that will eventually decay in social life as a result of the moderniza-
tion process, including industrialization, urbanization, and mass edu-
cation.10 Viewed from the political perspective, modernization theory
also predicts the decline of religion’s political role. According to Pippa
Norris and Ronald Inglehart, the process of modernization includes
“[t]he division of church and state, and the rise of secular-rational bu-
reaucratic states.”11 Modernization theory would explain the variation
in different states’ policies toward religion as a function of various levels
of modernization, which are generally measured by three criteria of hu-
man development—gdp per capita, literacy rate, and life expectancy.
This explanation is, however, not helpful for illuminating policy
tendencies in the three cases under consideration here. According to
undp’s Human Development Index for 2002, the U.S. and France have
close scores and rankings of development: the U.S. (0.939 / 8th) and
France (0.932 / 17th). Turkey, however, has a much lower score and
ranking of development (0.751 / 89th). The first two cases are countries
of high development, whereas Turkey is a country of medium develop-
ment. Modernization theory, therefore, would not successfully explain
why in terms of state policies toward religion, a highly developed coun-
try (France) differs from another highly developed country (the U.S.)
while it is similar to a moderately developing case (Turkey).
Modernization theorists would respond to my criticism by saying
that they provide a general explanation of an international trend of
state-religion relations, rather than an explanation of specific state
policies in a few cases. Some large-N analyses, however, also raise con-
10
Steve Bruce, God Is Dead: Secularization in the West (Malden, Mass.: Blackwell, 2002); Pippa
Norris and Ronald Inglehart, Sacred and Secular: Religion and Politics Worldwide (Cambridge: Cam-
bridge University Press, 2004).
11
Norris and Inglehart (fn. 10), 8, also 208–10.
574 w o r l d p o li t i c s

cerns about the theory. Robert Barro and Rachel McCleary examine
188 states in order to explain why some of them have an official state
religion whereas others do not. Although “[t]he standard view is that
richer countries are less likely to have state religions,” they conclude
that per capita gdp has “insignificant relations with” and “an ambig-
uous effect on the probability of state religion.”12 I in turn analyzed
175 countries in terms of their levels of development and state-reli-
gion regimes, using undp’s Human Development Index for 2002 and
World Christian Encyclopedia’s data set for 2000. Table 3 summarizes
the results of the analysis. Countries with high development have a
much higher percentage (57 percent) of having official religions than
do countries with low development (20 percent). This result is the op-
posite of what modernization theory would predict.
In sum, although modernization is an important factor in the analy-
sis of state-religion relations, its monocausal and linear perspective does
not explain diverse state-religion regimes, let alone specific policy ori-
entations. The emergence of secular states and the making of particu-
lar secular state policies toward religion are complex political processes
that cannot be understood without analyzing ideological struggles.
The second theoretical perspective is a civilizational approach, which
is generally called “essentialism” by its critics.13 This approach focuses
on text-based religious essentials to explain religion’s impact on socio-
political life. According to this approach, “Islam is the blueprint of a
social order. It holds that a set of rules exists, eternal, divinely ordained,
and independent of the will of men, which defines the proper order-
ing of society…. These rules are to be implemented throughout social
life.”14 The civilizational approach argues that there are (1) inherent
distinctions between certain religions/religious communities and that
(2) these religious differences have a direct impact on politics.15
Bernard Lewis, an influential civilizationist, argues that Islam and
Judaism are similar to each other and different from Christianity since
they do not have distinct conceptions of “clergy” versus “laity” or of
“sacred law” versus “secular law.” He defines state-religion struggles as
12
Robert J. Barro and Rachel M. McCleary, “Which Countries Have State Religions?” http://post
.economics.harvard.edu/faculty/barro/papers/state%religion%2001-05.pdf, i, 17 (accessed April 1, 2005).
13
Daniel Varisco, Islam Obscured: The Rhetoric of Anthropological Representation (New York: Palgrave
Macmillian, 2005); Alfred Stepan, “The World’s Religious Systems and Democracy: Crafting the ‘Twin
Tolerations,’” in Arguing Comparative Politics (New York: Oxford University Press, 2001), 213–54.
14
Ernest Gellner, Muslim Society (Cambridge: Cambridge University Press, 1983), 1.
15
Bernard Lewis, “The Roots of Muslim Rage,” Atlantic Monthly (September 1990); idem, What
Went Wrong? The Clash between Islam and Modernity in the Middle East (New York: Perennial, 2003);
Samuel P. Huntington, The Clash of Civilizations and the Remaking of World Order (New York: Simon
and Schuster, 1996).
 pa s s i v e & a s s ert i v e s e c u l a r i s m 575

Table 3
Human Development and State-Religion Regimes
States That Have States That Do Not
Official Religions Have Official Religions Total
High development 31 (57 %) 23 (43 %) 54 (100 %)
Medium development 54 (63 %) 32 (37 %) 86 (100 %)
Low development 7 (20 %) 28 (80 %) 35 (100 %)
Total 92 (53 %) 83 (47 %) 175 (100 %)

Sources: UNDP, Human Development Index 2002, http://hdr.undp.org/reports/global/2004/

pdf/hdr04_HDI.pdf (accessed April 1, 2005); David B. Barrett, George T. Kurian, and Todd M.
Johnson, World Christian Encyclopedia: A Comparative Survey of Churches and Religions in the Modern
World (New York: Oxford University Press, 2001).

a “Christian disease” and secularism as a “Christian remedy.”16 Lewis

specifically claims clear and divergent stands for Christianity and Islam
toward state-religion relations: “The distinction between church and
state, so deeply rooted in Christendom, did not exist in Islam.”17 Lewis
and other defenders of civilizationalism often refer to this well-known
verse of the Bible to prove the compatibility of Christianity and secular-
ism: “Render therefore unto Caesar the things which be Caesar’s, and
unto God the things which be God’s.”18 Samuel Huntington extends
Lewis’s thesis to other religions and cultures: “In Islam, God is Caesar; in
China and Japan, Caesar is God; in Orthodoxy, God is Caesar’s junior
partner. The separation and recurring clashes between church and state
that typify Western civilization have existed in no other civilization”19
The civilizational approach rightly alerts us to the importance of
religion in post–cold war world politics. It focuses our attention on
key theological differences among religions, differences that can have
an impact on individuals’ political preferences in different civilizational
contexts. Beyond these general concerns, however, civilizationalism has
little to say about particular state policies toward religion. It would ex-
plain states’ policies toward religion through their diverse religious back-
grounds. Since it overemphasizes the similarities within the West and
the differences between Western and Muslim countries, civilizational-
ism cannot explain why one “Western” state (France) pursues policies
toward religion that are different from those of another “Western” state
(the U.S.) and similar to those of a “Muslim” state (Turkey).
16
Bernard Lewis, “Secularism in the Middle East” (Chaim Weizmann Lecture, Rehovot, Israel,
1991), 10–12, 26.
17
Bernard Lewis, The Political Language of Islam (Chicago: University of Chicago Press, 1991), 2–3.
18
Luke 20:25, quoted by Lewis (fn. 16), 15.
19
Huntington (fn. 15), 70.
576 w o r l d p o li t i c s

Civilizationalists would reply that my cases do not refute their ar-

gument, because Turkey, with its secular state, is an exception in the
Muslim world. A general survey of the Muslim world, however, also
challenges their claims. Ira Lapidus stresses that religious and political
institutions in the Muslim world have been separate since the eighth
century. At that time, independent Sunni schools of law, Shia sects,
and Sufi orders, in addition to secular military and administrative rul-
ers, challenged and replaced the institution of the caliphate, which
claimed to represent both political and religious authorities.20 Recently,
the U.S. Commission on International Religious Freedom issued a
report on state-religion relations in forty-four Muslim countries. The
report concludes that “the majority of the world’s Muslim population
currently lives in countries that either proclaim the state to be secular
or that make no pronouncements concerning Islam to be the official
state religion.”21 The report emphasizes the diversity of state-religion
regimes in the Muslim world and disproves the alleged unity of Mus-
lim countries. Table 4 summarizes the results of the report.
Critics of civilizationalism stress that this approach has difficulty ex-
plaining not only the Muslim world but also Christian societies. The
civilizational argument about the inherent church-state separation in
Christianity overly romanticizes Christian societies by ignoring at least
three aspects: (1) their historical religious wars and church-state strug-
gles, (2) their substantially diverse state-religion regimes at present,
and (3) their current experience of religiously driven debates on politi-
cal and legal issues, such as abortion, gay rights, and evolution. These
divergences cannot be simply explained by rendering these things onto
Caesar. A more refined civilizationist approach acknowledges the di-
versity among Christian societies but argues that Protestantism is more
compatible with secularism than is Catholicism.22 This approach, how-
ever, is unable to explain the complex relations between the Catholic
church and states, changing Catholic views toward democracy, and the
persistence of established churches in several Protestant countries.23
20
Ira M. Lapidus, “The Separation of State and Religion in the Development of Early Islamic
Society,” International Journal of Middle East Studies 6 (October 1975).
21
Tad Stahnke and Robert C. Blitt, The Religion-State Relationship and the Right to Freedom of Re-
ligion or Belief: A Comparative Textual Analysis of the Constitutions of Predominantly Muslim Countries,
U.S. Commission on International Religious Freedom, March 2005, http://www.uscirf.gov/countries/
global/comparative_constitutions/03082005/Study0305.pdf, 2 (accessed April 1, 2005).
22
Regis Debray, Contretemps: Eloges des idéaux perdus (Paris: Gallimard, 1992), 23; John T. S.
Madeley, “A Framework for the Comparative Analysis of Church-State Relations in Europe,” West
European Politics 26 ( January 2003).
23
Jose Casanova, “Civil Society and Religion: Retrospective Reflections on Catholicism and Pro-
spective Reflections on Islam,” Social Research 68 (Winter 2001).
 pa s s i v e & a s s ert i v e s e c u l a r i s m 577

Table 4
State-Religion Regimes in 44 Muslim Countries
States That Have Islam as States That Do Not Have Islam
Official Religion (Total 22) as Official Religion (Total 22)
Islamic States States with Islam
with dominance as Established Antireligious
of Shari’a law Religion Secular States States
10 12 22a 0

Source: Tad Stahnke and Robert C. Blitt, The Religion-State Relationship and the Right to Freedom of
Religion or Belief: A Comparative Textual Analysis of the Constitutions of Predominantly Muslim Countries,
U.S. Commission on International Religious Freedom, March 2005, http://www.uscirf.gov/countries/
global/comparative_constitutions/03082005/Study0305.pdf, esp. 7 (accessed April 1, 2005).
a
Eleven of these twenty-two states openly declare themselves as secular states in their
constitutions, while the other eleven neither declare Islam as the official religion nor use the term
secular state explicitly in their constitutions.

The main problem of civilizationalism is that it underestimates hu-

man agency. Religious groups generally design their political prefer-
ences regarding sociopolitical conditions. As Anthony Gill points out,
the Catholic church seeks state intervention in order to restrict Prot-
estant proselytism in Latin America, where Catholicism is a dominant
religion, while it asks for more church-state separation and religious
freedom in post-Soviet Russia, where Catholicism remains in the mi-
nority.24 Similarly, an influential Islamic movement, Jamaat-i Islami,
defends an Islamic state in Pakistan, where Muslims are the majority,
whereas it supports the secular state in India, where they are a minority.25
The third and final theory—rational choice theory—differs from
the economic determinism of modernization theory and the religious
determinism of civilizationalism. Instead, it attaches importance to
three factors: individual preferences, their rational calculation, and their
structural constraints.26 Rational choice theory provides significant in-
sights for the analysis of political struggles, particularly actor strate-
gies. I therefore agree with most rational choice theorists’ critique of
civilizationalism as cited above. Moreover, despite having some major
reservations about rational choice theory, I find this perspective useful

24
Anthony Gill, The Political Origins of Religious Liberty (New York: Cambridge University Press,
2007). See also Stathis N. Kalyvas, The Rise of Christian Democracy in Europe (Ithaca, N.Y.: Cornell
University Press, 1996), 3.
25
Mumtaz Ahmad, “Islamic Fundamentalism: The Jamaat-i-Islami and the Tablighi Jamaat,” in
Martin E. Marty and R. Scott Appleby, eds., Fundamentalisms Observed (Chicago: University of Chi-
cago Press, 1991), 505.
26
Mancur Olson, The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidi-
ties (New Haven: Yale University Press, 1984).
578 w o r l d p o li t i c s

in explaining the importance of religious diversity on the church-state

separation in the American founding period.
Gill is one of the few rational choice theorists who examine the
causes of state policies toward religion. He argues that these policies
differ because of rulers’ varying calculations of opportunity costs based
on their preferences for (1) sustaining political survival, (2) minimiz-
ing the cost of ruling, and (3) succeeding in economic development.27
Gill would argue that state rulers in France and Turkey pursue more
restrictive policies toward religion than do rulers in the U.S. because
such policies help them minimize the opportunity costs concerning
these three issues. The strength of this argument lies in its ability to
explain the strategic flexibility of rulers. However, it is not able to ex-
plain the decisions of an important set of actors, high court judges,
whose primary concerns in deciding state-religion cases are not about
political survival or economic issues. In addition, this approach would
have problems regarding my cases. The ban on wearing headscarves in
schools in Turkey and France, for example, has been politically risky (at
least for the Turkish politicians), and it has created huge ruling costs,
while not helping economic development at all.
My main concern about the rational choice approach is that it largely
takes individual preferences as given. According to this approach, state
rulers and social activists have distinct preferences shaped by their so-
cioeconomic status regardless of their ideology. I argue the opposite.
Although I take individual cost-benefit analysis, structural constraints,
and strategic behaviors seriously, I want to go beyond them by unpack-
ing individuals’ ideological preferences.28 Ideologies are not simple in-
struments for material interests or justifications for already decided
behaviors. They are genuinely important factors that shape individuals’
preferences.29 Ideologies and material conditions are separate but inter-
related. I agree with Max Weber on their importance in the construction
of preferences and interests: “Not ideas, but material and ideal interests,
directly govern men’s conduct. Yet very frequently the ‘world images’ that
have been created by ‘ideas’ have, like switchmen, determined the tracks
along which action has been pushed by the dynamic of interest.”30
27
Gill (fn. 24).
28
See Karl-Dieter Opp, “Contending Conceptions of the Theory of Rational Action,” Journal of
Theoretical Politics 11 (April 1999).
29
I deliberately use the term ideology, rather than culture. Culture is practical and habitual, which
makes it inconsistent and fuzzy. Ideology is, by contrast, a set of ideas that is generally related to a consis-
tent utopia, which make it easier to categorize. Ideologies are “formal, explicit, and relatively consistent”
and “articulated by political elites,” whereas cultures are “informal, implicit, and relatively inconsistent”
and “held by people within a given institutional setting.” Stephen E. Hanson, “Review Article: From
Culture to Ideology in Comparative Politics,” Comparative Politics 35 (April 2003), 356.
30
Max Weber, From Max Weber: Essays in Sociology (New York: Oxford University Press, 1946), 280.
 pa s s i v e & a s s ert i v e s e c u l a r i s m 579

My analysis of alternative approaches is summarized in Table 5

through Mill’s methods of difference and agreement.31 The method
of difference searches for the causes of diverse results in similar cases.
My analysis shows the importance of a dominant ideology for explain-
ing the different policies of the U.S. and France—two secular states,
which are similar in terms of economic development and civilizational
identity but different in terms of dominant ideology. The method of
agreement, by contrast, examines similar results in different cases.
France and Turkey are different regarding their levels of economic de-
velopment and civilization, yet they share a similar dominant ideology,
which explains their similar policies toward religion.
The methods of difference and agreement, however, are not suffi-
cient in themselves to test theories, since they show only correlation
and omit certain variables.32 Therefore, I use the method of process
tracing to analyze causal processes between ideological struggles and
policy formation.33 Beyond this methodological concern, a monocausal
explanation based on ideology would also have empirical problems. State
policies toward religions are too complex to be explained simply in terms
of dominant ideologies. Although my cases reflect two opposite policy
tendencies in general, they also include several policy inconsistencies,
exceptions, and changes. For example, taxpayer money cannot be used
directly to fund religious schools in the U.S., whereas the French state fi-
nances Catholic schools and the Turkish state runs Islamic Imam-Hatip
schools. These paradoxes can be explained only by a process-oriented
explanation of ideological struggles. That is why I do not take coun-
tries as monolithically assertive or passive secularists; instead, I analyze
ideological controversies within them, despite the existence of certain
dominant ideologies. The next section focuses on this issue.

Ideological Struggles and State Policies

In the U.S. supporters of passive secularism are dominant while sup-
porters of assertive secularism (for example, American atheists) con-
stitute a marginal group. Yet there has been a debate between passive
secularism’s two different interpretations—accommodationism and
separationism. The accommodationists generally include conservatives
31
John Stuart Mill, A System of Logic (New York: Longmans, 1961).
32
Stanley Lieberson, “Small N’s and Big Conclusions,” in Charles Ragin and Howard Becker,
eds., What Is a Case? (Cambridge: Cambridge University Press, 1992).
33
James Mahoney and Dietrich Rueschemeyer, eds., Comparative Historical Analysis in the Social
Sciences (New York: Cambridge University Press, 2003); Alexander L. George and Andrew Bennett,
Case Studies and Theory Development in the Social Sciences (Cambridge: mit Press, 2005), 205–32.
580 w o r l d p o li t i c s

Table 5
Methods of Difference and Similarity: U.S., France, and Turkey
Alternative Explanatory Variables Dependent Variable
Economic Dominant State Policies
Development Civilization Ideology toward Religion
U.S. high Western passive secularism inclusionary
France high Western assertive secularism exclusionary
Turkey moderate Islamic assertive secularism exclusionary

supporting the Republican Party, while many separationists are liberals

who support the Democratic Party. The accommodationists regard close
state-religion interactions as compatible with secularism, since that does
not mean an establishment of a particular religion.34 The separationists,
however, see any close entanglement as contrary to the First Amend-
ment and seek an impenetrable “wall of separation” between the state
and religion. Even the U.S. Supreme Court Justices have been divided
along accommodationist and separationist lines. As analyzed by Ken-
neth Wald and Joseph Kobylka, court rulings on significant state-religion
cases from 1943 to 2002 included twenty-eight separationist decisions,
thirty-four accommodationist decisions, and three mixed decisions.35
Based on these ideological views, the accommodationist associations,
such as the American Center for Law and Justice, have supported the
school voucher systems allowing individuals to receive government sub-
sidies for educational expenses to private (including religious) schools.
The separationist associations, such as the American Civil Liberties
Union, however, have regarded school vouchers as unconstitutional
public funding of religious schools. As a result of separationist opposi-
tion, school voucher systems have remained marginal in the U.S. The
separationists have also succeeded in keeping organized prayer and reli-
gious instruction out of public schools, since the Supreme Court’s En-
gel v. Vitale decision in 1962. The accommodationists failed to amend
the Constitution to overrule the court on this issue. Nevertheless, the
accommodationists have managed to create spaces in public schools
for Christian clubs to organize student-initiated religious meetings, in-
34
Due to the page limitation, I am neglecting the differences among conservatives, such as those
between the accommodationists and the Christian Right. I analyze it in detail in my forthcoming book,
“Dynamics of Secularism: State Policies toward Religion in the United States, France, and Turkey.”
35
Kenneth D. Wald, Religion and Politics in the United States (New York: Rowman and Littlefield,
2003), 85–87; Joseph F. Kobylka, “The Mysterious Case of Establishment Clause Litigation: How
Organized Litigants Foiled Legal Change,” in Lee Epstein, ed., Contemplating Courts (Washington,
D.C.: cq Press, 1995), 96, 102–3.
 pa s s i v e & a s s ert i v e s e c u l a r i s m 581

cluding Bible studies and prayer, that take place after class hours. They
have done that by using the discourses of freedom of speech and equal
access. They have also succeeded in keeping the phrase “under God”
in the pledge of allegiance recited in schools, though some, but not all,
separationists have tried to eliminate it.36 Despite their opposite policy
preferences, the accommodationists and separationist are both defend-
ers of passive secularism and oppose assertive secularist exclusion of
religion from the public sphere. For example, both groups are critical of
the French and Turkish ban on headscarves in the public schools.37
In France, assertive secularists are dominant despite the resistance
of passive secularists. In original terminology, the former defend laïcité
de combat (combative secularism), while the latter support laïcité pluri-
elle (pluralistic secularism).38 The assertive secularists aim to confine
religion to the home and to the individual’s conscience, while the pas-
sive secularists try to allow a public role for religion. In short, passive
secularists want to liberalize secularism in France with a new emphasis
on individualism and multiculturalism. The Ligue de l’enseignement
(League of Education), a union of educators with two million mem-
bers, is the main supporter of passive secularism. The league and other
passive secularists have not challenged the absence of prayer or refer-
ences to God in schools. Yet the league proposed reintroducting re-
ligious instruction to end French exceptionalism in Western Europe
on this issue. The proposition remained abortive due to the assertive
secularist opposition. Public funding of private schools, 95 percent of
which are Catholic institutions, were a major issue of controversy be-
tween these two groups. The current modus vivendi between them is
that the state is funding private schools that signed a contract to allow
a certain level of state control, especially over the curriculum.
From 1989 to 2004 these two groups ardently debated the matter of
headscarves in the schools. The assertive secularists, especially the franc-
maçons (Freemasons) and libre-penseurs (Freethinkers), supported the ban.39

36
Mary C. Segers and Ted G. Jelen, A Wall of Separation? Debating the Public Role of Religion (Lanham,
Md.: Rowman and Littlefield, 1998); Stephen V. Monsma, Positive Neutrality: Letting Religious Freedom
Ring (Grand Rapids: Baker Books, 1995); T. Jeremy Gunn, “Religious Freedom and Laïcité: A Compari-
son of the United States and France,” Brigham Young University Law Review 24 (Summer 2004).
37
Author interviews with American academics and social activists, Washington, D.C., Seattle, and
Salt Lake City, May 2005–March 2006.
38
Author interviews with French academics and social activists, Paris and Auxerre, October–
December 2004.
39
Again, because of the word limitation, I am neglecting the alliance between the assertive secu-
larists and anti-immigrant Islamophobics with regard to their support for a ban on headscarves. For
a detailed analysis of this alliance, see Ahmet T. Kuru, “Secularism, State Policies, and Muslims in
Europe: Analyzing French Exceptionalism,” Comparative Politics (forthcoming).
582 w o r l d p o li t i c s

The passive secularists led by the league, however, asked for freedom
to wear headscarves. The French Council of State also supported the
passive secularist perspective by opposing a general ban. It decided that
wearing a headscarf was not inherently incompatible with laïcité (sec-
ularism).40 Between 1992 and 1999 the council overturned forty-one
of forty-nine cases in which students wearing headscarves had been
expelled from school.41 By the early 2000s, to overrule the Council of
State, the assertive secularists pressed for a new law to ban headscarves,
with the support of about three-quarters of the French public, accord-
ing to public surveys.42 In early 2004 they finally succeeded in imposing
a ban on headscarves and other religious symbols in public schools.43
Turkey is another case where assertive secularists are dominant. The
Kemalists, who claim to preserve the legacy of M. Kemal Atatürk, de-
fend the existing dominance of assertive secularism, whereas the pro-Is-
lamic conservatives want to promote passive secularism.44 The assertive
secularists, such as the Republican People’s Party and the majority of
military generals and high court judges, want to confine religion, in
general, and Islam, in particular, to the private sphere. Yet the pas-
sive secularists, including conservative parties (for example, the ruling
Justice and Development Party) and groups (for example, the Gülen
movement), want to allow public visibility of religion.45 The assertive
secularists aim to keep Islam under state control. Therefore, they have
outlawed private Islamic education and teaching the Koran to children
under the age of twelve. They have tried to promote an individualized
version of Islam through religious instruction in schools. Other than
that, public schools are totally secular, in the sense that they do not al-
low prayer or other religious expressions. Although conservative parties
have generally received about 70 percent of the vote in national elections,
40
The French Council of State’s opinion in the Headscarf Case, November 27, 1989, no. 346,893.
Some scholars translate the term “laïcité” as “secularity.” Instead, I prefer the term “secularism,” which
is most commonly used in the literature on state-religion relations in English (for example, by Ameri-
can and Indian scholars).
41
Haut conseil à l’intégration, L’Islam dans la République (Paris: La documentation française,
2001), 66.
42
Jean-Louis Debré, La laïcité à l’école: Un principe républicain à réaffirmer: Rapport de la mission
d’information de l’Assemblée nationale (Paris: Odile Jacob, 2004), 179.
43
Alain Seksig, Patrick Kessel, and Jean-Marc Roirant, “Ni Plurielle ni de combat: La laïcité,”
Hommes & Migrations, no. 1218 (March–April 1999); Valentine Zuber, “La Commission Stasi et les
paradoxes de la laïcité française,” in Jean Baubérot, ed., La Laïcité à l’épreuve: Religions et libertés dans
le monde (N.p.: Universalis, 2004); Joel S. Fetzer and J. Christopher Soper, Muslims and the State in
Britain, France, and Germany (New York: Cambridge University Press, 2005), 73–85.
44
Author interviews with Turkish politicians, academics, and social activists, Ankara and Istanbul,
July–September 2004.
45
Ahmet T. Kuru, “Globalization and Diversification of Islamic Movements: Three Turkish
Cases,” Political Science Quarterly 120 (Summer 2005).
 pa s s i v e & a s s ert i v e s e c u l a r i s m 583

they have been unable to resist such policies to due the assertive secu-
larists’ dominance in the military and the judiciary.46
The main source of tension between these two groups has been the
headscarf controversy.47 The assertive secularists have imposed a ban
on headscarves in all educational institutions. Although passive sec-
ularist politicians passed three pieces of legislation in the 1980s and
1990s permitting the wearing of headscarves at universities, these laws
were either vetoed by the assertive secularist president or struck down
by the Constitutional Court, on the grounds that they were against
laiklik (secularism). The headscarf debate is deeper in Turkey than it
is in France for three reasons. First, in Turkey, unlike in France, the
headscarf is not a symbol of an immigrant religious minority: about 60
percent of women in Turkey wear some sorts of headscarf.48 Second,
the ban in France is confined to public schools, whereas that in Tur-
key encompasses all educational institutions. Last, but not least, weekly
church/mosque participation is only 10 percent in France whereas it is
69 percent in Turkey.49 The exclusion of religious symbols from public
schools is relatively easier in less religious French society, in compari-
son with highly religious Turkish society.
In short, particular ideological struggles between the supporters and
opponents of dominant secular ideologies are the main reason for the
two opposite policy tendencies in my three cases. That still leaves an
important question: why did passive secularism initially become domi-
nant ideology in the U.S., whereas assertive secularism became domi-
nant in France and Turkey? The next section addresses this question.

Ancien Régime, Critical Juncture, and

Ideological Path Dependence
The dominance of either passive or assertive secularism results from
the historical conditions and relations during a country’s state-building
period. In general, the dominance of passive secularism is based on an
46
Devlet ve Din I˙lişkileri, Farklı Modeller, Konseptler ve Tecrübeler (Ankara: Konrad Adenauer Vakfı,
2003); Ahmet T. Kuru, “Reinterpretation of Secularism in Turkey: The Case of the Justice and Devel-
opment Party,” in M. Hakan Yavuz, ed., The Emergence of a New Turkey: Islam, Democracy, and the AK
Parti (Salt Lake City: University of Utah Press, 2006); M. Hakan Yavuz, Islamic Political Identity in
Turkey (Oxford: Oxford University Press, 2003).
47
Nilüfer Göle, The Forbidden Modern: Civilization and Veiling (Ann Arbor: University of Michi-
gan Press, 1996).
48
Ali Çarkoğlu and Binnaz Toprak, Değişen Türkiye’de Din, Toplum ve Siyaset (Istanbul: tesev,
2006), 66. According to the survey of Milliyet and A&G in 2003, this ratio was 64 percent. Milliyet,
May 27, 2003.
49
“Les Français et la prière,” Le Pèlerin Magazine, April 13, 2001; Ali Çarkoğlu and Binnaz Toprak,
Türkiye’de Din, Toplum ve Siyaset (Istanbul: tesev, 2000), 41, 45.
584 w o r l d p o li t i c s

“overlapping consensus”50 (where actors reached an agreement for dif-

ferent reasons) between secular and religious groups, whereas that of
assertive secularism is a product of conflict between them. The consen-
sual and conflictual relationships are linked to these groups’ ideological
perspectives. If secular groups are antireligious (in terms of opposing a
public role for religion) and religious groups try to maintain their estab-
lished status, then they will find themselves in conflict. On the contrary,
if secular groups are not antireligious and religious groups are not trying
to keep an established religion, then consensus may arise. The critical
condition that affects these views is the absence or existence of an an-
cien régime that combines monarchy with hegemonic religion. If such
an ancien régime exists, then, it is hard to convince hegemonic religious
groups to agree to the disestablishment of their religion. Moreover, the
ancien régime also leads the secular elite, who oppose the monarchy in
founding a new republic, to combat the hegemonic religion that justifies
monarchy. In the words of Alexis de Tocqueville, “religions intimately
linked to earthly governments . . . sacrifice the future for the present. . . .
Hence religion cannot share the material strength of the rulers without
being burdened with some of the animosity roused against them.”51
The history of religion-state relations in several countries reflects
the causal relationship between the religious institutions’ alliance with
monarchies and the rise of antireligious views. In certain European
countries, such as Spain and Portugal, anticlericalism emerged as the
republicans’ reaction to the Catholic church’s cooperation with the
monarchy. In the words of Paul Manuel, “Absolute political power and
legitimacy in Portugal and Spain until . . . the modern era were in the
hands of the monarch. . . . The Roman Catholic Church legitimized
the monarch’s claim to divine authority, and, in turn, typically received
royal grants of land, among other goods.”52 In the nineteenth century
the republican elite challenged this “Iberian ancien régime.” As a result
of the Catholic church’s continuing support to the crown and the aris-
tocracy, “the Republicans became staunchly anticlerical.”53
This causal process is not uniquely related to the Catholic church
and is also relevant for other religious institutions. The Orthodox

50
Rawls (fn. 7).
51
Alexis de Tocqueville, Democracy in America, trans. George Lawrence (1835; Garden City, N.Y.:
Anchor Books, 1969), 297. See also Alexis de Tocqueville, L’ancien régime et la révolution (1858; Paris:
GF-Flammarion, 1988).
52
Paul Christopher Manuel, “Religion and Politics in Iberia: Clericalism, Anticlericalism, and
Democratization in Portugal and Spain,” in Ted Gerard Jelen and Clyde Wilcox, eds., Religion and
Politics in Comparative Perspective: The One, the Few, and the Many (New York: Cambridge University
Press, 2002), 74.
53
Ibid., 76.
 pa s s i v e & a s s ert i v e s e c u l a r i s m 585

church in Russia, for example, experienced similar antagonism from

the Bolsheviks due to its identification with the Russian monarchy. In
the Russian ancien régime, “the Russian Orthodox Church was the
established church of the Russian Empire, and the Tsar was its head.”54
That was one of the main reasons that Lenin and other leaders of the
1917 Revolution were hostile to the church, in particular, and religion,
in general. Their atheism was much more antagonistic toward religion
than Marx’s philosophical atheism.55
The existence or absence of an ancien régime is also a crucial fac-
tor in my three cases. Passive and assertive secularism became domi-
nant ideologies during the periods of secular state-building in America
(1776–91, from the Declaration of Independence to the First Amend-
ment); in France (1875–1905, from the Constitutional Laws of the
Third Republic to the 1905 Law separating church and state); and in
Turkey (1923–37, from the foundation of the Republic to the constitu-
tional amendment enshrining secularism as a constitutional principle).
Although passive and assertive secularist ideologies had already been
formulated in the minds and writings of several intellectuals decades
before these periods, it was during the state-building periods for these
three cases that the two ideologies became dominant.
These periods are critical junctures where the secular state replaced
the old types of state-religion regimes and left an ideological and in-
stitutional legacy that has persisted ever since. A critical juncture is a
moment when both agency and structural conditions are available for
a systematic change.56 When the new system becomes consolidated, it
creates ideological and institutional path dependence that then persists
for a long time, even if it becomes inefficient and costly.57 Path depen-
dence does not necessarily claim an inevitable historical determinism.58
Rather, it stresses that ideological and institutional change is possible
54
Harold J. Berman, “Religious Rights in Russia at a Time of Tumultuous Transition: A Historical
Theory,” in Johan D. van der Vyver and John Witte, Jr., eds., Religious Human Rights in Global Perspec-
tive: Legal Perspectives (The Hague: Martinus Nijhoff Publishers, 1996), 287.
55
Ibid., 289.
56
“[C]ritical junctures are moments of relative structural indeterminism when willful actors shape
outcomes in a more voluntary fashion than normal circumstances permit.... Before a critical juncture,
a broad range of outcomes is possible; after a critical juncture, enduring institutions and structures are
created, and the range of possible outcomes is narrowed considerably.” James Mahoney, The Legacies
of Liberalism: Path Dependence and Political Regimes in Central America (Baltimore: Johns Hopkins
University Press, 2001), 7.
57
Paul Pierson, “Increasing Returns, Path Dependence, and the Study of Politics,” American Politi-
cal Science Review 94 ( June 2000); idem, “Not Just What, but When: Timing and Sequence in Po-
litical Processes,” Studies in Comparative Political Development 14 (Spring 2000); Scott E. Page, “The
Types and Causes of Path Dependence,” http://www.bramson.net/academ/public/Page-path%20De
pendence.pdf (accessed September 1, 2005).
58
Kathleen Thelen, “Timing and Temporality in the Analysis of Institutional Evolution and
Change,” Studies in Comparative Political Development 14 (Spring 2000).
586 w o r l d p o li t i c s

but remains very difficult. It requires deliberate political collective action,

as well as necessary structural conditions. I disagree with scholars who
argue that the beginning of path dependence is generally a contingent,
accidental event.59 In the cases discussed here, the path dependence of
passive and assertive secularism began with purposeful political strug-
gle, rather than with the occurrence of coincidental events.
America was a new country of immigrants, where the secular elite
(including Thomas Jefferson, James Madison, and George Washing-
ton) neither focused on the elimination of a monarchy nor perceived
religion as its ally.60 The alliance between the British monarchy and the
Anglican church did not constitute an ancien régime for at least two
reasons. First, the Anglican church was established in only six colonies,
while seven others had either established Congregational churches
or had no established church at all.61 Second, even in those six colo-
nies “the Anglican establishment was during most times nominal and
the church’s control over religious concernments largely ineffective.”62
These conditions affected not only secular elites’ toleration of religion’s
public role but also religious groups’ openness to church-state separa-
tion at the federal level. Because there was no nationwide hegemonic
religion, religious groups were open to such a separation “without
nostalgia for an ancien régime.”63 Moreover, the diversity of compet-
ing Protestant denominations led many religious groups to consider
separation of church and state as a second-best choice as a guarantee of
their own religious freedom.
This historical explanation works better than the famous narrative
about the role of Puritan immigrants in the foundation of religious
freedom in America. The Puritan narrative helps explain why religion
has played an important role in the American public sphere. However,
the fact that the Puritans escaped from persecution in Europe did not
automatically make them promoters of church-state separation or re-
ligious freedom. Several American colonies had established churches
and discriminated against various religious minorities, such as dissent-
ing Protestants, Catholics, Jews, native Indians, and African slaves.64
59
James Mahoney, “Path Dependence in Historical Sociology,” Theory and Society 29 (August 2000).
60
Louis Hartz, The Liberal Tradition in America (1955; San Diego: Harcourt Brace and Company,
1991).
61
Carl H. Esbeck, “Dissent and Disestablishment: The Church-State Settlement in the Early
American Republic,” byu Law Review 30 (2004), 1415, 1457
62
Ibid., 1414.
63
Robert Bellah, “Civil Religion in America,” Daedalus 134 (Fall 2005), 50.
64
Michael W. McConnell, “The Origins and Historical Understanding of Free Exercise of Reli-
gion,” Harvard Law Review 103 (May 1990), 1421–30; Kevin Boyle and Juliet Sheen, eds., Freedom of
Religion and Belief: A World Report (New York: Routledge, 1997), 154–55.
 pa s s i v e & a s s ert i v e s e c u l a r i s m 587

The lack of an ancien régime and the presence of religious diversity

were primary factors in the emergence of secularism and religious free-
dom in the U.S. as a gradually evolving political process.
In the American founding secular rationalists were influenced by
the Enlightenment, while the evangelicals were affected by the Great
Awakening. As mentioned above, the former were not antireligion
and the latter were open to church-state separation. They also had an
ideational common ground based on John Locke’s liberalism and the
thought of some Protestant thinkers (for example, Roger Williams,
John Witherspoon, and Isaac Backus) who favored church-state sepa-
ration.65 These two groups largely agreed on separation at the federal
level as formulated in the First Amendment, and that consensus led
the dominance of passive secularism in the U.S.
In France the ancien régime was based on the marriage between the
monarchy and the Catholic church. Anticlericalism and republican-
ism, therefore, were like twins in the dual fight against the clergy and
the monarchy. Many eighteenth-century French philosophers, such as
Voltaire and Jean-Jacques Rousseau, considered the Catholic church
to be an impediment to their republican project.66 In the late nine-
teenth century Léon Gambetta formulized anticlericalism in France: le
cléricalisme, voilà l’ennemi! (clericalism—here is the enemy!).67 France
experienced several back and forths between disestablishment and
reestablishment of the Catholic church from the 1789 Revolution to
the 1905 Law that separated the church and the state. The Catholic
church, seeking to maintain its hegemonic position, opposed secular-
ism until the end of the World War II.
Unlike in the U.S., in France there was almost no ideational bridge
between secular republicans and conservative Catholics. As a result,
the struggle between these two groups was a zero-sum game, the “war
of two Frances.” One France was the inheritor of the values of the 1789
Revolution: it was republican, anticlerical, and secularist. It included
leftist parties, some civic associations (for example, the Freemasons and
Freethinkers), and religious minorities (the Protestants and Jews). The
other France included the clergy and its conservative supporters in pol-
itics and bureaucracy.68 In the early Third Republic (1875–1905) the
65
Noah Feldman, “The Intellectual Origins of the Establishment Clause,” New York University
Law Review 77 (May 2002).
66
According to Rousseau, Catholicism “is so apparently bad that it is a waste of time to enjoy dem-
onstrating its badness.” Jean-Jacques Rousseau, Du contrat social (Paris: G. F. Flammarion, 2001), 174.
67
Quoted by Mona Ozouf, L’Ecole, l’Eglise et la République, 1871–1914 (Paris: Editions Cana,
1982), 50.
68
Jean Baubérot, Histoire de la laïcité en France (Paris: puf, 2004).
588 w o r l d p o li t i c s

secular republicans, such as Minister of Education Jules Ferry, secular-

ized education by excluding thousands of clerical teachers, as well as
closing about fifteen thousand Catholic schools.69 Although the Cath-
olics challenged these policies, they were not effective in party politics
and the parliament.70
Despite the opposition of conservative Catholics, the secularists
passed the legislation that separated church and state in 1905. The bill
was approved by a majority in the National Assembly (341 to 233)
and the Senate (179 to 103).71 Pope Pius X, the French clergy, and the
Catholic press condemned the law. As a result of this severe conflict as-
sertive secularism in France became the dominant ideology.
Turkey, too, had its own ancien régime, which depended on the Ot-
toman monarchy and the hegemony of Islam. Islamic law was in use
in the Ottoman Empire and the ulema were an important element of
the state structure.72 Moreover, the Ottoman sultans claimed to be the
caliphs of all Muslims. The Westernist elite in the late Ottoman and
early Republican era, therefore, regarded Islam as a barrier against their
modernizing reforms. One of their ideologues, Abdullah Cevdet, pro-
posed abolishing all Islamic institutions and importing European civi-
lization “with both its roses and its thorns.”73 M. Kemal Atatürk and
other framers who founded the Republic in 1923 opposed the influence
of Islam and other religions on the public sphere. The Islamists, by
contrast, sought to maintain Islam’s hegemonic status. The Kemalists
embraced European schools of thought, especially positivism, and did
not have an ideational connection with the Islamists.74 Conflict be-
tween these two groups was almost inevitable.
The Kemalists abolished the caliphate, expropriated the proper-
ties of pious foundations, and brought the ulema under state control
by founding the Directorate of Religious Affairs (Diyanet). They also
abolished Islamic law and adopted European laws, including the Swiss
Civil Code, the Italian Criminal Code, and the German Commercial
69
Ozouf (fn. 67), 233–34.
70
Maurice Larkin, Church and State after the Dreyfus Affair: The Separation Issue in France (New
York: Barnes and Noble, 1973); Kalyvas (fn. 24).
71
Othon Guerlac, “The Separation of Church and State in France,” Political Science Quarterly 23
( June 1908).
72
Ejder Okumuş, Türkiye’nin Laikleşme Serüveninde Tanzimat (Istanbul: I˙nsan Yayınları, 1999);
Kemal Karpat, The Politization of Islam: Reconcstructing Identity, State, Faith, and Community in the
Late Ottoman State (New York: Oxford University Press, 2001).
73
Translated and quoted by Bernard Lewis, The Emergence of Modern Turkey (London: Oxford
University Press, 1968), 236.
74
M. Şükrü Hanioğlu, “Garbcılar: Their Attitudes toward Religion and Their Impact on the Of-
ficial Ideology of the Turkish Republic,” Studia Islamica 86 (1997); Şerif Mardin, Religion and Social
Change in Modern Turkey: The Case of Bediüzzaman Said Nursi (New York: State University of New
York Press, 1989).
 pa s s i v e & a s s ert i v e s e c u l a r i s m 589

Code.75 The Kemalists closed down all medreses and created a unified
secular school system. Throughout the 1930s there remained almost no
institution in Turkey that could legally provide education of Islam. The
Kemalists targeted Islamic organizations at the societal level as well, by
outlawing tariqas, closing Sufi lodges, and shutting down the tombs of
Muslim saints.
Finally, in 1937 the Kemalists added the principle of secularism to
the Constitution to seal the secularizing reforms. The Kemalists pursued
these reforms despite the Islamists’ opposition. In this regard, the domi-
nation of assertive secularism in Turkey emerged as a result of the con-
flict between these two groups and the former’s victory over the latter.
Table 6 summarizes my argument on the historical dominance of
passive and assertive secularism in these three cases.
Since the secular state-building period—despite the presence of cer-
tain challenging views and conceptual changes—passive and assertive
secularism have preserved their dominance in these three countries
through ideological indoctrination, institutional socialization, and pub-
lic education. In the U.S. passive secularism has remained dominant
and allowed the public visibility of religion. Yet there have been certain
changes in state policies toward religion, as well as in the meaning of
state neutrality itself. From the late eighteenth to the early nineteenth
century, neutrality meant only the federal government’s neutrality to-
ward Protestant denominations: established churches persisted in cer-
tain states until 1833. From that time until the early twentieth century,
secularism implied neutrality toward various Protestant denomina-
tions at both the federal and the state level. At the time there was a
Protestant semiestablishment that marginalized Catholics. The public
schools, for example, were teaching the Protestant King James version
of the Bible.76 The early twentieth century was the period of the redefi-
nition of neutrality, as Catholics and Jews were incorporated.77 Since
the 1950s there has been a new debate on the meaning of neutrality.
For the separationists neutrality requires state impartiality toward all
faiths and atheism, whereas for many accommodationists it asks state
neutrality only for monotheistic religions.
In France although assertive secularism has ideological roots going
back to the eighteenth century, it became dominant in the early Third
75
Sibel Bozdoğan and Reşat Kasaba, eds., Rethinking Modernity and National Identity in Turkey
(Seattle: University of Washington Press, 1997).
76
Christian Smith, ed., The Secular Revolution: Power, Interests, and Conflict in the Secularization of
American Public Life (Berkeley: University of California Press, 2003); Philip Hamburger, Separation of
Church and State (Cambridge: Harvard University Press, 2004).
77
Jose Casanova, Public Religions in the Modern World (Chicago: University of Chicago Press,
1994), 135-209.
590 w o r l d p o li t i c s

Table 6
Historical Conditions and Relations during Secular State-Building
I II III IV

U.S. the absence secular overlapping dominance

of an ancien groups were consensus of passive
régime and not against between secularism
diversity of religion’s secular and
protestant public role religious
denominations groups
religious
groups were
open to
church-state
separation

France and the presence secular groups severe conflict dominance

Turkey of an ancien were against between of assertive
régime religion’s secular and secularism
based on the public role religious
monarchy and groups
the hegemony religious
of Catholicism groups
and Islam sought to
maintain the
establishment
of Catholicism
and Islam

Republic. Since that time, the assertive secularists have succeeded in

marginalizing religion in the public sphere despite the resistance of
conservative Catholics. The recent headscarf controversy resulted in a
new passive secularist challenge to the dominance of assertive secular-
ism.78 In Turkey the assertive secularists have remained dominant since
the foundation of the Republic despite the Islamist opposition. They
have largely succeeded in minimizing Islam’s public roles in Turkey.
The rise of conservative politicians and social movements, however,
meant a passive secularist challenge to their domination.79
78
Jean Baubérot, Laïcité 1905–2005, entre passion et raison (Paris: Seuil, 2004); Henri Pena-Ruiz,
Qu’est-ce que la laïcité? (N.p.: Folio, 2003) ; John R. Bowen, Why the French Don’t Like Headscarves:
Islam, the State, and Public Space (Princeton: Princeton University Press, 2006).
79
Berna Turam, Between Islam and the State: The Politics of Engagement (Stanford, Calif.: Stanford
University Press, 2006); M. Hakan Yavuz and John L. Esposito, eds., Turkish Islam and the Secular
State: The Gülen Movement (Syracuse: Syracuse University Press, 2003).
 pa s s i v e & a s s ert i v e s e c u l a r i s m 591

Although assertive secularism has similarly been dominant in France

and Turkey, these two still reflect a policy divergence. Certain poli-
cies in Turkey, such as the ban on private religious education, indicate
that the Turkish state has been more exclusionary toward religion than
has been the French state. Again, historical conditions are important
for tracing the reasons for this divergence. The main reason for this
relative policy difference lies in the historical trajectories of democra-
tization and authoritarianism in France and Turkey.80 From the secular
state building in the late nineteenth century to the present, assertive
secularism in France has coexisted with a multiparty democracy and
has gained substantial popular support. It was challenged by monar-
chist Catholic movements and pro-Catholic authoritarian rules, such
as the Vichy regime (1940–44). Moreover, because of French democ-
racy, the opponents of assertive secularism have had the political means
to criticize certain policies, and the assertive secularists have had to
make compromises in their utopian ideological views. In Turkey, by
contrast, assertive secularism was established by an authoritarian re-
gime in the early twentieth century and has been defended since 1950
by several military coups d’état against conservative governments. Un-
der the shadow of the Turkish military, it has been much more diffi-
cult to oppose assertive secularist policies. Turkish assertive secularists,
therefore, have very rarely accepted policy compromises.
In sum, the dominance of either passive or assertive secularism that
emerged in the state-building period has continued to the present day in
the three cases as instances of ideological path dependence. On the one
hand, this ideological dominance has filtered struggles over state policies
toward religion. On the other hand, it has experienced conceptual trans-
formations (for example, in the U.S.) and faced resistance of alternative
religious and ideological groups (for example, in France and Turkey).

Conclusion
This article examines two opposite policy tendencies toward religion as
manifested in three secular states. The American state is generally more
tolerant toward public expressions of religion than are its French and
Turkish counterparts. An example of this policy divergence is the three
states’ policies toward religious attire. In the American legal system,
80
Jean-Paul Burdy and Jean Marcou, “Laïcité/Laiklik: Introduction,” Cahiers d’études sur la Médi-
terranée orientale et le monde turco-iranien, no. 19 (1995), 29; Jean Baubérot, “D’une comparaison:
Laïcité française, laïcité turque,” in Isabelle Rigoni, ed., Turquie, les mille visages: Politique, religion,
femmes, immigration (Paris: Editions Syllepse, 2000).
592 w o r l d p o li t i c s

neither the federal government nor individual states can prohibit reli-
gious symbols in general or symbols of a particular religion by singling
them out.81 In France and Turkey, however, the state has singled out the
wearing of religious attire as grounds for excluding students from pub-
lic schools, without subsuming it under some more general regulation
with a practical rationale (for example, health or security).
I explain this policy difference in terms of ideological struggles.
Such an emphasis on ideologies is a challenge to the mainstream view
in the social sciences, which tends to attach importance to strategic and
instrumental behaviors, while disregarding actors’ ideas. In the U.S. the
accommodationists and separationists defend different interpretations
of passive secularism, though both of them allow public visibility of
religion. In France the assertive secularists, who aim to exclude religion
from the public sphere, have succeeded in establishing their dominance
despite the resistance of the passive secularists. Similarly, in Turkey the
Kemalists have powerfully defended the dominant assertive secular-
ism despite the challenge of the pro-Islamic conservatives, who have
promoted passive secularism. In short, I do not take states as mono-
lithically passive or assertive secularist.82 Nor do I claim an ideological
determinism. Instead, I attach importance to human actors who em-
brace ideologies and struggle to materialize their political agenda.
In a social scientific analysis variables are generally parts of broader
chains. I first took the two opposite policy tendencies as the dependent
variable and tried to explain it in terms of the ideological struggles
between assertive and passive secularists. I then examined dominant
ideologies themselves as a dependent variable and explained it through
historical conditions, particularly the existence or absence of an ancien
régime based on monarchy and hegemonic religion.
The establishment of a new ideological dominance generally re-
quires a long historical process. First, ideologies emerge in the works
of some native thinkers, or they are imported from other intellectually
influential countries. Then, they find certain followers among the elite
through publications and school education. Next, these followers orga-
nize and mobilize to challenge the existing ideological establishment.
Finally, these activists need available structural conditions (such as
wars, economic crisis, or critical elections) to replace the old ideological
dominance. When the activists find convenient conditions, they estab-
81
Derek H. Davis, “Reacting to France’s Ban: Headscarves and other Religious Attire in Ameri-
can Public Schools,” Journal of Church and State 46 (March 2004).
82
For fragmentation of state actors, see Joel S. Migdal, State in Society: Studying How State and
Societies Transform and Constitute One Another (Cambridge: Cambridge University Press, 2001).
 pa s s i v e & a s s ert i v e s e c u l a r i s m 593

lish the new ideology by making it dominant in both abstract institu-

tions (the Constitution, other laws, and the market) and organizations
(schools, courts, and barracks).
Since the dominant ideology plays a crucial role in the formation of
state policies, its change results in substantial policy transformations.
Two recent examples of such change are postcommunist Russia and
postmonarchist Iran. Although the collapse of the Soviet Union and
the Iranian Revolution had multiple causes, ideological transformation
marked their results, in terms of new patterns of policy orientation. The
elimination of the communist ideology in former Soviet republics led
to a set of major policy transformations, including those concerning
state-religion relations. Today, the Russian state is far from being athe-
istic. Instead, it has close relations with the Orthodox church and tries
to please Muslims both inside and outside its territory with particular
policies, such as being an observer of the Organization of the Islamic
Conference. In the aftermath of the Iranian Revolution, Shia political
Islamism replaced the Shah’s secularist ideology. This ideological rup-
ture has caused extensive policy repercussions from foreign affairs to
state-religion relations. Ideological changes may also be more gradual
than these revolutionary examples. In an ideologically divided institu-
tion, such as the U.S. Supreme Court, appointment of even one justice
with conservative or liberal views may result in the change of the court’s
dominant perspective, which will have certain policy implications.
This article relies on an analytical tool that can be used by other
scholars. It reemphasizes the importance of historical comparison to
trace the impact of history on current policymakers via ideological path
dependence. It is an analysis of how the ideological consensus or con-
flict at the critical juncture in a country’s history defines the ideological
constraints under which current struggles occur and policies are made.
Researchers can explore the relationship between historical conditions,
ideological struggles, and state policies in their analysis of other types
of regime change (for example, democratization) or policy formation
(for example, ethnic policies). Moreover, my typology of passive and
assertive secularism provides a conceptual framework with which to
examine other secular states. Scholars of two well-known cases of secu-
larism, India and Mexico,83 for example, may use passive secularism for
the former and assertive secularism for the latter in their analyses.
83
Bhargava (fn. 2); Gary Jeffrey Jacobsohn, The Wheel of Law: India’s Secularism in Comparative
Constitutional Context (Princeton: Princeton University Press, 2003); Roberto Blancarte, “Un regard
latino-américain sur la laïcité,” in Jean Baubérot and Michel Wieviorka, eds., Les Entretiens d’Auxerre:
De la séparation des Églises et de l’État à l’avenir de la laïcité (N.p.: L’aube, 2005).
 594 w o r l d p o li t i c s

My analysis of secularism may prompt further research on the rela-

tionship between secularism, liberalism, and democracy. U.S. policies
seem to be largely liberal and democratic, since they respect individual
rights and, generally, follow people’s demand. French policies that vio-
late individual freedoms, such as the ban on headscarves, are not liberal
but can still be defined as democratic, since 72 percent of the popula-
tion supported the ban.84 Turkish policies, again such as the ban on
headscarves, are neither liberal nor democratic, since only 22 percent
of the population supports it.85 Future studies can elaborate these and
other cases regarding the relationship between these three concepts.
Another related and intriguing issue is the compatibility of secular-
ism and Islam. My approach stresses that such a question is misleading
unless the researcher is aware of passive and assertive types of secu-
larism, as well as diverse interpretations of Islam. Regarding assertive
secularism, the answer is more likely to be negative, since it has had
problems not only with Islam in Turkey but also with Catholicism in
France. Assertive secularism seems to be incompatible with any religion
that has public claims. A practical reflection of this abstract debate is
the popular disdain toward secularism in many Muslim countries. In
the Muslim geography, from Syria to Tunisia, from Saddam’s Iraq to
Algeria, what people have experienced has been assertive secularism
mainly as a result of French colonial and intellectual influence.86 Mus-
lim populations may want to rethink secularism if they recognize an
alternative mode—passive secularism—that tolerates public visibility
of religion. This issue requires further research and separate essays, if
not books.

Debré (fn. 42), 19.

84

Çarkoğlu and Toprak (fn. 48), 75.

85
86
Edward Webb, “Kemalism in More Than One Country? Religious Jacobinism in the Early
Turkish Republic, the French Third Republic, and the Republics of Tunisia and Syria” (Ph.D. diss.,
University of Pennsylvania, 2007); Rachid al-Ghannouchi, “Secularism in the Arab Maghreb,” in
John L. Esposito and Azzam Tamimi, eds., Islam and Secularism in the Middle East (New York: New
York University Press, 2000); Nikki R. Keddie, “Secularism and the State: Towards Clarity and Global
Comparison,” New Left Review (November–December 1997), 25–30.

View publication stats