Emerging Trends in The Social and Behavioral Sciences

The Politics of Secularism in the United States

Item

Title

The Politics of Secularism in the United States

Author

Campbell, David E.

Layman, Geoffrey C.

Research Area

Class, Status and Power

Topic

Political Power

Abstract

Secularism is on the rise within the United States. Religious nonaffiliation has increased dramatically, while other measures of religiosity—worship attendance, belief in God, belief in scripture—are waning. This essay considers the political implications of this secular turn, in light of the historically high rates of religiosity within the United States. In doing so, we distinguish between passive and active secularism and describe measures of each. Passive secularism refers to not being religious—not identifying with a religion, not attending religious services, and not believing in God. Active secularism is the affirmative adoption of a secular worldview and identity. With this distinction, we can trace Americans' growing political polarization along religious–secular lines. One form of passive secularism—religious nonaffiliation—is often triggered by a backlash to the Religious Right. More generally, both passively and actively secular voters are moving to the political left. Furthermore, while smaller in number than passive secularists, active secularists are highly engaged in politics, comprising a sizeable share of the Democratic Party's activist base.

Related Essays

Understanding American Political Conservatism (Political Science), Joel D. Aberbach

Lived Religion (Sociology), Nancy T. Ammerman

Heuristic Decision Making (Political Science), Edward G. Carmines and Nicholas J. D'Amico

Trends in Religiosity and Religious Affiliation (Sociology), Kevin J. Christiano

Institutional Change in American Religion (Sociology), Casey Clevenger and Wendy Cadge

Patterns of Attachments across the Lifespan (Psychology), Robyn Fivush and Theodore E. A. Waters

Empathy Gaps between Helpers and Help‐Seekers: Implications for Cooperation (Psychology), Vanessa K. Bohns and Francis J. Flynn

Political Advertising (Political Science), Erika Franklin Fowler

Presidential Power (Political Science), William G. Howell

Group Identity and Political Cohesion (Political Science), Leonie Huddy

Political Inequality (Sociology), Jeff Manza

Mysticism (Sociology), Barry Markovsky and Jake Frederick

The Sociology of Religious Experience (Sociology), Douglas Porpora

Religion (Anthropology), Benjamin Grant Purzycki et al.

Atheism, Agnosticism, and Irreligion (Sociology), Buster G. Smith and Joseph O. Baker

Complex Religion: Toward a Better Understanding of the Ways in which Religion Intersects with Inequality (Sociology), Melissa J. Wilde and Patricia Tevington

Identifier

etrds0423

extracted text

The Politics of Secularism in the
United States
DAVID E. CAMPBELL and GEOFFREY C. LAYMAN

Abstract
Secularism is on the rise within the United States. Religious nonaffiliation has
increased dramatically, while other measures of religiosity—worship attendance,
belief in God, belief in scripture—are waning. This essay considers the political
implications of this secular turn, in light of the historically high rates of religiosity within the United States. In doing so, we distinguish between passive
and active secularism and describe measures of each. Passive secularism refers
to not being religious—not identifying with a religion, not attending religious
services, and not believing in God. Active secularism is the affirmative adoption of
a secular worldview and identity. With this distinction, we can trace Americans’
growing political polarization along religious–secular lines. One form of passive
secularism—religious nonaffiliation—is often triggered by a backlash to the Religious Right. More generally, both passively and actively secular voters are moving
to the political left. Furthermore, while smaller in number than passive secularists,
active secularists are highly engaged in politics, comprising a sizeable share of the
Democratic Party’s activist base.

The United States is experiencing a rise in secularism. Most notably, the
last 25 years have seen a dramatic rise in the percentage of Americans who
report having no religious affiliation or, as they have come to be known, the
“Nones.” For decades, the percentage of Americans who indicated that they
have no religion hovered between 5% and 7%. Then, beginning in the early
1990s the percentage of Nones began to rise. According to the General Social
Survey, they grew to 14% by 2000 and 21% in 2014. Other national surveys
show the same trend.
While no other trend is as dramatic as the growth of the Nones, other
indicators such as declines in religious attendance, literal belief in the Bible,
and a firm belief in God also reflect a move away from religion in the
United States or, in other words, a secular turn. In the words of Voas and
Chaves, “The evidence for a decades-long decline in American religiosity is
Emerging Trends in the Social and Behavioral Sciences.
Robert Scott and Marlis Buchmann (General Editors) with Stephen Kosslyn (Consulting Editor).
© 2017 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

now incontrovertible. Like the evidence for global warming, it comes from
multiple sources, shows up in several dimensions, and paints a consistent
factual picture” (2016, p. 1524).
We address the political implications of this secular turn, focusing on both
the political causes and consequences of rising secularism in America. The
growth of secularism cannot be understood without recognizing the inextricable links between politics and religion, and thus secularism, within the
contemporary United States.
This essay will deal with individuals’ beliefs and behavior—private
secularism—rather than Americans’ attitudes toward the separation of
church and state, or public secularism. Private and public secularism are
conceptually distinct, because individuals can believe in a strict separation of
church and state and still be highly religious in their personal lives. Indeed,
over the course of American history this has described many staunch advocates of strict separation. Likewise, it is possible to be personally secular but
nonetheless believe in a public role for religion. Importantly, to be privately
secular does not necessarily mean being hostile to religion.
CONTEXT
Understanding the full significance of the recent secular turn within the
United States requires some context. The United States has long stood
apart from other advanced, industrialized democracies for its unusually
high levels of religiosity (Norris & Inglehart, 2011). The United States
has thus long been a notable exception to what might loosely be called
secularization theory. Although there are different incarnations of the theory,
it essentially holds that as societies modernize, religious beliefs and practices
are undermined and thus decline (Gorski & Altinordu, 2008). In contrast,
Finke and Stark (2005) have argued for a completely different way of
understanding the conditions under which religion flourishes. In their view,
religious involvement is driven by competition between religions, not by
the degree of modernization. Religion, in this view, operates in a market,
and when religious groups must compete with one another for members
they respond by preaching a gospel that draws adherents. According to this
theory, the sort of gospel that draws the largest market share is characterized
by “hellfire and brimstone”—people are attracted to religions that are in
tension with secular society. Contrary to a theory of inevitable secularization
in the wake of modernization, they instead argue that there has been a rise
in the religious involvement of Americans over the course of the nation’s
history.
Putnam and Campbell (2010) show that, contrary to secularization theory,
the United States has not been marching inexorably toward ever-greater

The Politics of Secularism in the United States

3

secularization, but contrary to Finke and Stark, neither is religiosity
always rising. Instead, the course of American religious history is better
characterized as having ebbs and flows. While the 1950s, for example, were
a high period for religion in the United States, it began to decline after the
“shock” of the counter-culture of the 1960s. As attitudes toward sexual
mores changed dramatically, religiosity began to fall. Then, beginning in the
1970s religion began to rebound, owing to an “aftershock” in reaction to the
social earthquake of the 1960s. In the wake of sexual liberalization, many
Americans flocked to religions that maintained traditionalist beliefs. In
particular, evangelical Protestantism grew during this period. Importantly,
white evangelical Protestants also awakened as a political movement, when
prior to this period they had been largely absent from national politics. Thus
was born the Religious Right.
Voas and Chaves (2016) offer a different view, seeking to rehabilitate secularization theory as it applies to the United States. To that end, they have
marshaled considerable evidence to argue that, when it comes to secularization, the United States is not exceptional after all. Specifically, they contend
that the recent secular turn is consistent with secularization theory, because
secularism has grown through generational replacement. As an older cohort
of more religious Americans dies, they are replaced by a younger cohort of
less religious Americans—which is precisely the pattern in Western Europe,
where secularization theory has been most applicable. However, they also
note that secularization was delayed in the United States owing to the past
growth of evangelical Protestantism (i.e., the aftershock to the 1960s).
Voas and Chaves make a compelling case for a slowly rising tide of secularism, although their theory cannot explain the inflection point that marks
the rise of the Nones. Change through generational replacement is gradual,
and is not characterized by a sharp break with the past. Instead, the rise of
the Nones appears to have a political explanation, as a backlash to the emergence of the Religious Right as a political movement, and specifically the
Republican Party’s emphasis on social conservatism (Hout & Fischer, 2002,
2014). Nonetheless, religious nonaffiliation (None-ism) is only one indicator of secularism. While the rise of the Nones is largely due to a reaction
to the political environment, other manifestations of secularism are better
described as causes of political attitudes. In other words, whatever its origin, the “slow and steady” secularism described by Voas and Chaves has
important implications for the American political landscape. Political backlash and generational replacement are not mutually exclusive explanations
for rising secularism, as both can occur simultaneously. To that end, Hout
and Fischer, the originators of the political backlash explanation, also highlight the importance of generational replacement as contributing to the secular turn.

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

PASSIVE AND ACTIVE SECULARISM
There is an important conceptual distinction to be made between two forms
of secularism: passive and active (Layman, Campbell, Green, & Sumaktoyo,
n.d.). Passive secularism refers to “not being religious”—not identifying
with a religion, not attending religious services, and not believing in God.
Passive secularists are thus defined by what they are not. The category of
None exemplifies being passively secular, as these are people who decline
to name a religious affiliation. But religious disaffiliation is only one way
to be passively secular, as there are many people who have a nominal
affiliation to a religious tradition but nonetheless do not exhibit other forms
of religiosity.
Active secularists, however, are defined by what they are. Active secularism
is the affirmative adoption of a secular worldview and identity. In contrast
to a None, an active secularist is someone who, for example, might adopt
a self-description as an atheist. Similarly, active secularists will consciously
adopt secular principles when making ethical judgments, such as the efficacy
of reason, science, and human experience.
While, to our knowledge, we are the first to employ the terms active and
passive secularism in this way, the existing literature nonetheless provides
evidence of the need for such a distinction. For example, most Nones have at
least some belief in God and are thus what Hout and Fischer call “unchurched
believers” (2002) rather than ardent secularists. Far from having rejected religion outright, many Nones move back and forth between identifying and
not identifying with a religion. Lim, MacGregor, and Putnam (2010) demonstrate that when asked about their religious affiliation twice in just under a
year, over one third of Nones in the first interview subsequently reported
having a religion, while another third went from affiliation to nonaffiliation.
There was no evidence that these “liminal” Nones had undergone any sort
of religious conversion or disillusionment; rather, they have an uncertain
self-identity and thus, when asked at any given time, may or may not identify with a religion. These are not people who have affirmatively adopted a
secular worldview.
Similarly, sociologists of religion have written extensively about people
who describe themselves as spiritual but not religious (Fuller, 2001). By
many conventional measures of religiosity, these people would appear to
be “secular,” as they often do not attend religious services frequently or
engage in other outward forms of religious behavior. Yet their comfort with
spirituality suggests that they are far from being actively secular.
We have set about measuring passive and active secularism, to see if this
is a distinction with a difference. Passive secularism is straightforward, as it
is simply the inverse of the conventional measures of religiosity long used

The Politics of Secularism in the United States

5

by social scientists. These include the frequency of religious attendance,
frequency of prayer, religious salience, strength of belief in God, and literal
belief in scripture. The absence of a religious affiliation—being a None—also
indicates passive secularism, but because, as explained below, theory
provides reason to think that it is driven by different factors as these other
expressions of religiosity, we keep nonaffiliation separate from the other
measures of passive secularism.
Measuring active secularism is more complicated. Like measures of religiosity, it should incorporate both beliefs and identity. Furthermore, a measure of active secularism cannot reference religion since, by definition, it is
neither the absence of religiosity nor antagonism toward religion. Therefore,
the challenge is to measure secular beliefs and identity without pitting them
against religious counterparts.
We conducted a national survey in 2010–2012 that includes the conventional measures of religiosity described above as well as new measures meant
to tap into active secularism. These new measures include an index of secular beliefs. Respondents indicated their level of agreement with the following
statements; the first three are worded to affirm a secular worldview, while the
last two are worded oppositely:
1. Factual evidence from the natural world is the source of true beliefs.
2. The great works of philosophy and science are the best source of truth,
wisdom, and ethics.
3. To understand the world, we must free our minds from old traditions
and beliefs.
4. It is hard to live a good life based on reason and facts alone.
5. Values are more important than factual evidence in making moral decisions.
To gauge the salience of secularism, we asked respondents how much (if
any) guidance they received from “non-religious beliefs, such as derived
from science or philosophy.” Secular identity was measured by asking
respondents to select which (if any) of the following terms describe them:
ecumenical, mainline, charismatic/Pentecostal, humanist, non-traditional
believer, secular, atheist, fundamentalist, born again/evangelical, agnostic,
spiritual but not religious. Exploratory factor analysis reveals that four of
these terms form a single dimension of actively secular identity: secular,
humanist, atheist, and agnostic.
We have employed structural equation modeling to correct for measurement error, in order to test whether (i) our measures of active secularism
hang together as a single construct and (ii) active and passive secularism
are distinct from one another. We refer interested readers elsewhere for a

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

detailed discussion of this analysis, but the results are easy to summarize:
our measures of active secularism do indeed constitute a single dimension
and, while they are related, active and passive secularism are distinct from
one another (Layman et al., n.d.)
With our measures of active and passive secularists, we can report on how
many of each category exist within the American population. We define passive secularists as people who attend religious services “once or twice a year”
or less, receive no guidance from religion, pray no more than “at least once
a week,” do not view the Bible as the word of God, and are uncertain about
the existence of God.1 Active secularists are those who took the secular position on at least three of the five secular belief items, receive some (or more)
guidance from non-religious beliefs, and chose at least one of the secular
self-identities.
Using these definitions, we find that roughly 24% of the American population are passive secularists, while 9% are active secularists. Is 9% big or small?
It is obviously smaller than the share of passive secularists, or of Nones. On
the other hand, if a denomination, active secularists would be among the
largest in the country—larger than Southern Baptists.2 Because passive and
active secularism are not mutually exclusive, we can also report the degree
of overlap between them. Not surprisingly, the vast majority of active secularists are also passively secular. Yet among passive secularists, only 28%
also score high in active secularism. In other words, nearly three-quarters of
Americans who do not exhibit conventional religiosity have not embraced
an overtly secular perspective. They are better described by what they are
not—religious—than by what they are.
POLITICAL IMPLICATIONS
The distinction between passive and active secularism is important for
understanding the ways that secularism affects, and is affected by, the
political landscape. We hypothesize that there are two different processes at
work. First, building on the work of Hout and Fischer and others, we expect
identification as a None to rise as a reaction to the perception that religion
has become closely associated with conservative politics (Hout & Fischer,
2002, 2014; Patrikios, 2008; Putnam & Campbell, 2010). Second, we expect
political identities to have a mutually reinforcing relationship with both
passive and active secularism. Over time, Democrats and liberals should
1. We measured belief in God on a 100-point scale; respondents placed themselves on a scale indicating their certainty in the existence of God, with 0 representing absolute certainty that there is a God and
1 being absolute certainty that God does not exist. Passive secularists are defined as those who scored
above the mean, which was 19.7.
2. According to the 2014 Pew Religious Landscape Survey, roughly 5% of the U.S. population are
Southern Baptists (http://www.pewforum.org/religious-landscape-study).

The Politics of Secularism in the United States

7

become more secular/less religious, while Republicans and conservatives
become less secular/more religious. As this process unfolds, it results in
an increasing returns process—the more secularism is identified with the
political left, the more Democrats and liberals embrace it; likewise, the more
Republicans and conservatives reject it. Polarization results (Campbell,
Layman, & Green, forthcoming).
THE RISE OF THE NONES
In their seminal article on the rise of the Nones—written in 2002, at a time
when the rise was only first being noticed—Hout and Fischer argued that it
was a backlash to the emergence of the Religious Right as a political movement. A key piece of evidence for this argument is that the Nones are concentrated among political moderates and liberals. While they do not use this
language, Hout and Fischer’s account is consistent with the classic theory of
cognitive dissonance. If religion is associated with conservative politics, then
people who are not conservatives may experience dissonance between their
politics and their religion. Increasingly, they resolve the dissonance by keeping their politics and disavowing their religious affiliation. In their original
2002 paper, Hout and Fischer rely largely on circumstantial evidence to make
their case. In a 2014 update, they revisit the question with a more powerful
research design, as they employ panel data—repeated interviews with the
same people—to show that political backlash is driving the turn away from
religion. Likewise, both Putnam and Campbell (2010) and Patrikios (2008)
use different panel studies to come to the same conclusion.
We pause to note the theoretical significance of the finding that people are
turning away from a religious affiliation as a reaction to the political environment, as this reverses the longstanding assumption that religion precedes
politics. In other words, the causal arrow has been reversed; politics can drive
religion.
While compelling, by relying on observational data these studies cannot
definitively establish causation. To that end, we have designed an experiment to test whether religious disaffiliation can be induced by exposure to
politicized religion. A nationally representative sample of subjects was given
a short pre-test to measure their baseline degree of passive and active secularism. A week later, these subjects were randomly assigned to read a fictionalized but realistic local news story about a nearby congressional race. The
story highlights a Republican and Democrat running for an open seat, with
basic biographical information on both candidates. The stories differed only
in the degree to which one or both of the candidates were associated with
religion, ranging from none to moderate to high. For example, the moderate
religious treatments include photos of the candidates standing in a church,

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

while in the high religion treatments the candidates are pictured engaging
in religious behavior; the no religion treatments feature the candidates in a
generic setting. Similarly, the stories varied the degree to which the candidates are quoted as referencing their own religion, and whether they were
endorsed by faith-based groups.
As hypothesized, the experiment finds that when subjects are exposed to
a Republican candidate who invokes religion, they are more likely to disclaim a religious identity or, in other words, to identify as Nones. Conversely,
there is no effect when a Democratic candidate is associated with religion.
Interestingly, there is also no effect when both candidates are highly associated with religious rhetoric, endorsements, and imagery. An effect appears
only when the Republican candidate’s use of religion exceeds that of the
Democratic candidate. Further analysis reveals that the effect is concentrated
almost entirely among subjects who identify as Democrats. Substantively, the
effect for Democrats is sizeable; the probability that they identify as a None
increases by 13 percentage points. Recall that this is after exposure to merely
one news story about a single congressional race.
Of course, this treatment would hardly be the first time subjects were
exposed to the politicization of religion. Instead, the experimental treatment
brings to mind similar campaigns and candidates previously experienced
by the subjects. Furthermore, this much movement in self-identification
as a None is consistent with the work on the malleability of religious
nonaffiliation. It is unlikely that this one experimental treatment is causing
religious alienation among subjects who are religiously devout. Rather, it
nudges people who are ambivalent about religion toward disclaiming a
religious identity.
This experimental treatment, however, has no effect on other forms of passive secularism—attendance at religious services, prayer, religious salience,
strength of belief in God, and a literal belief in the Bible. This is to be expected,
as it would not be possible for the behaviors (attendance, prayer) to change
in the course of the experiment, and the beliefs (religious salience, belief in
God, belief in the Bible) are not subject to the same ambivalence as religious
self-identity. Likewise, the experiment has no effect on active secularism,
measured as secular beliefs, salience, and identity.
SECULARISM AND POLARIZATION
What, though, happens to passive and active secularism over time in the “real
world”? What relationship do the two forms of secularism have with politics? We are able to answer that with the aforementioned 2010–2012 panel
study. Measuring individual-level change in both political and secular orientations enables a test of whether secularism affects politics, politics affects

The Politics of Secularism in the United States

9

secularism, or if the relationship goes both ways. Political views are measured as party identification, self-described ideology, or attitudes on cultural
issues (abortion and same-sex marriage). This analysis again employs structural equation modeling, and is thus able to correct for measurement error.
First, identification as a None is only ever affected by politics, and not the
other way around. Put another way, while being a Democrat and/or liberal
may lead someone to become a None, being a None does not lead people
toward a Democratic and/or liberal orientation.
Second, there is generally a mutually reinforcing relationship between passive secularism and political views.
In other words, over time, Democrats and liberals are becoming more passively secular while passive secularism also leads to an increase in liberalism.
That is to say, we observe polarization. Importantly, though, passive secularism does not lead to a substantive change in party identification. Being
a Democrat leads to greater passive secularism, but being passively secular
does not lead people to become a Democrat.
Like passive secularism and unlike identification as a None, we again see
a mutually reinforcing relationship between active secularism and political
orientations. Unlike its passive analog, active secularism even has an effect
on party identification, which in the American context is arguably the most
stable of political orientations. Again, this is evidence of polarization along a
secular–religious fault line.3
As suggested by active secularism’s effect on political orientations, especially party identification, active secularists differ significantly from their
passive counterparts. Active secularists are highly engaged in politics—they
are more likely to identify with the Democratic Party, hold ideologically
consistent views, and participate in politics. Passive secularists, in contrast,
are more likely to be on the political sidelines, as evidenced by the fact that
being passively secular does not lead to identification with the Democratic
Party. And while passive secularism does affect liberal ideology and cultural
attitudes, those effects are much weaker than those for active secularism.
Furthermore, passive secularists are less likely than their actively secular
counterparts to be politically active.
In other words, while there has been considerable attention paid to the rise
of what we have called passive secularism, it is actually active secularism that
has the larger effect on the political landscape. Active secularists’ numbers
are fewer, but they are far more engaged in politics. As one indication
of active secularists’ high level of political engagement and strength of
3. As further evidence of the links between the political environment and secularism in its various
forms, Campbell, Green, and Layman find that this “secular polarization” only occurs among people who
perceive religion to be politicized, either those who identify evangelical Christians as “mainly Republicans” or perceive a “a lot” of talk about religion in politics.

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Democratic partisanship, Layman and Weaver (2016) have found that
roughly 56% of delegates to the 2012 Democratic National Convention can
be described as actively secular.
CHALLENGES FOR BOTH PARTIES
These results suggest challenges for both the Republican and Democratic
parties. For Republicans, a rising secular generation is a challenge for a
party so closely associated with religion. Given that religious conservatives
are often described as the party’s “base” it seems likely that this association will endure in the public mind. However, the more the Republican
Party emphasizes religion and social conservatism—especially regarding
LGBTQ rights—the more it will fan the flames of further religious disaffiliation, which makes it still more difficult for the GOP to win their
support.
Given the political profile of secularists, and especially active secularists,
they are a critical constituency for the Democratic Party and a large component of its base. Yet, unlike the way the Republicans have embraced their
base of highly religious supporters, the Democrats have not actively courted
their secularist supporters, perhaps because they fear being perceived as
the “anti-religious” party. After all, even with the secular turn, the United
States remains a highly religious nation. As an illustration of the internal
tensions within the Democratic Party, during the 2012 Democratic National
Convention delegates voted to remove any reference to God from the party
platform. Recognizing that this would only provide fodder for Republican
attacks, Democratic leaders maneuvered to restore God to the platform,
against the objections of many delegates. The Democrats have yet to find
a way to mobilize their actively secular supporters without alienating
religious voters. This is all further complicated by the fact that both African
Americans and Latinos are key components of the Democratic coalition and,
on average, highly religious.
THE FUTURE
What does the sizeable secular population in the United States portend for
the future? Because our measures of active secularism only began in 2010,
we cannot say for sure whether these views are becoming more common,
although it seems likely that they are (e.g., the rising share of Americans
who do not believe in God). Whether increasing or not, though, active secularists are highly engaged in politics and aligned with the Democratic Party.
Their heightened engagement suggests continued conflict, both political and
legal, over the many expressions of religion under government auspices that

The Politics of Secularism in the United States

11

have survived as vestiges of an earlier time when the secular population was
smaller and thus less politically influential. Examples are myriad, but they
include “under God” in the Pledge of Allegiance, religious symbols in public parks, prayers at local government meetings, and so on. Likewise, in the
court of public opinion, we can expect further political disputes over the public role of religion.
The decline in religion—that is, rise in both passive and active
secularism—will likely also have an effect on the nation’s civil society.
Religious congregations had largely held firm as participation in other forms
of civic engagement have fallen (Putnam, 2001), but it now appears that they
too are experiencing the same type of decline. This does not bode well for
philanthropic activity, especially rates of volunteering, as religious activity
is a powerful predictor of civic voluntarism—including for secular causes.
Although it is possible that secular organizations will arise to fill in the gap,
it seems unlikely that such groups will ever be full substitutes for the civic
activity facilitated by religious congregations.
While these are likely to be short-term consequences of the secular turn,
one should not assume that secularization will carry on indefinitely. Recall
that there has not been a steady decline or rise in religiosity in the United
States. Religiosity, and thus secularism, has both ebbed and flowed. Because
many passive secularists, especially the Nones, have not abandoned religious
beliefs altogether, perhaps religious entrepreneurs will find a way to win the
religiously ambivalent back to the pews. Presumably such a reawakening
would require disassociating religion from politics, or at least conservative
politics. On the other hand, perhaps the secular turn has reached the point of
no return. Whichever path the future takes, the implications for the nation’s
politics will be profound.

REFERENCES
Campbell, D. E., Layman, G. C., & Green, J. C. (forthcoming). Putting politics first: the
impact of politics on american religious and secular orientations. American Journal
of Political Science.
Finke, R., & Stark, R. (2005). The churching of America, 1776–2005: winners and losers in
our religious economy (2nd ed.). New Brunswick, NJ: Rutgers University Press.
Fuller, R. C. (2001). Spiritual, but not religious: understanding unchurched america. New
York, NY: Oxford University Press.
Gorski, P. S., & Altinordu, A. (2008). After secularization? Annual Review of Sociology,
34, 55–85.
Hout, M., & Fischer, C. S. (2002). Why more Americans have no religious preference:
Politics and generations. American Sociological Review, 67(2), 165–190.

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Hout, M., & Fischer, C. S. (2014). Explaining why more Americans have no religious
preference: Political backlash and generational succession, 1987–2012. Sociological
Science, 1(October), 423–447.
Layman, G. C., Campbell, D. E., Green, J. C., & Sumaktoyo, N. (n.d.). Active secularism
and the politics of irreligion in the United States.
Layman, G. C., & Weaver, C. (2016). Religion and secularism among American party
activists. Politics and Religion, 9(2), 271–295.
Lim, C., MacGregor, C. A., & Putnam, R. D. (2010). Secular and liminal: Discovering heterogeneity among religious nones. Journal for the Scientific Study of Religion,
49(4), 596–618.
Norris, P., & Inglehart, R. (2011). Sacred and secular: Religion and politics worldwide.
Cambridge, England: Cambridge University Press.
Patrikios, S. (2008). American republican religion? Political Behavior, 30(3), 367–389.
Putnam, R. D. (2001). Bowling alone: The collapse and revival of American community.
New York, NY: Simon & Schuster.
Putnam, R. D., & Campbell, D. E. (2010). American grace: How religion divides and unites
us. New York, NY: Simon and Schuster.
Voas, D., & Chaves, M. (2016). Is the United States a counterexample to the secularization thesis? American Journal of Sociology, 121(5), 1517–1556.

DAVID E. CAMPBELL SHORT BIOGRAPHY
David E. Campbell is the Packey J. Dee Professor of American Democracy
at the University of Notre Dame. His most recent book is Seeking the Promised
Land: Mormons and American Politics (with John Green and Quin Monson). He
is also the co-author (with Robert Putnam) of American Grace: How Religion
Divides and Unites Us, which has received both the 2011 Woodrow Wilson
Award from the American Political Science Association for the best book on
government, politics, or international affairs and the Wilbur Award from the
Religious Communicators Council for the best non-fiction book of 2010. In
addition, Campbell is the author of Why We Vote: How Schools and Communities Shape Our Civic Life, and has published numerous articles in journals such
as the American Political Science Review, American Journal of Political Science,
Journal of Politics, and Foreign Affairs.
GEOFFREY C. LAYMAN SHORT BIOGRAPHY
Geoffrey C. Layman is a Professor of Political Science at the University of
Notre Dame, specializing in political parties, political behavior, and religion
and politics. His first book, The Great Divide: Religious and Cultural Conflict
in American Party Politics examines the growing division of the Democratic
and Republican parties along religious and cultural lines. Layman’s current
research includes work with Thomas Carsey on “conflict extension” and

The Politics of Secularism in the United States

13

polarization in American party politics, and with David Campbell and John
Green on the political causes and consequences of growing secularism in
the United States. Layman has published numerous articles in the leading
journals in political science, including the American Political Science Review,
the American Journal of Political Science, the Journal of Politics, the British
Journal of Political Science, and the Annual Review of Political Science.
RELATED ESSAYS
Understanding American Political Conservatism (Political Science), Joel D.
Aberbach
Lived Religion (Sociology), Nancy T. Ammerman
Heuristic Decision Making (Political Science), Edward G. Carmines and
Nicholas J. D’Amico
Trends in Religiosity and Religious Affiliation (Sociology), Kevin J. Christiano
Institutional Change in American Religion (Sociology), Casey Clevenger and
Wendy Cadge
Patterns of Attachments across the Lifespan (Psychology), Robyn Fivush and
Theodore E. A. Waters
Empathy Gaps between Helpers and Help-Seekers: Implications for Cooperation (Psychology), Vanessa K. Bohns and Francis J. Flynn
Political Advertising (Political Science), Erika Franklin Fowler
Presidential Power (Political Science), William G. Howell
Group Identity and Political Cohesion (Political Science), Leonie Huddy
Political Inequality (Sociology), Jeff Manza
Mysticism (Sociology), Barry Markovsky and Jake Frederick
The Sociology of Religious Experience (Sociology), Douglas Porpora
Religion (Anthropology), Benjamin Grant Purzycki et al.
Atheism, Agnosticism, and Irreligion (Sociology), Buster G. Smith and Joseph
O. Baker
Complex Religion: Toward a Better Understanding of the Ways in which
Religion Intersects with Inequality (Sociology), Melissa J. Wilde and Patricia
Tevington



The Politics of Secularism in the
United States
DAVID E. CAMPBELL and GEOFFREY C. LAYMAN

Abstract
Secularism is on the rise within the United States. Religious nonaffiliation has
increased dramatically, while other measures of religiosity—worship attendance,
belief in God, belief in scripture—are waning. This essay considers the political
implications of this secular turn, in light of the historically high rates of religiosity within the United States. In doing so, we distinguish between passive
and active secularism and describe measures of each. Passive secularism refers
to not being religious—not identifying with a religion, not attending religious
services, and not believing in God. Active secularism is the affirmative adoption of
a secular worldview and identity. With this distinction, we can trace Americans’
growing political polarization along religious–secular lines. One form of passive
secularism—religious nonaffiliation—is often triggered by a backlash to the Religious Right. More generally, both passively and actively secular voters are moving
to the political left. Furthermore, while smaller in number than passive secularists,
active secularists are highly engaged in politics, comprising a sizeable share of the
Democratic Party’s activist base.

The United States is experiencing a rise in secularism. Most notably, the
last 25 years have seen a dramatic rise in the percentage of Americans who
report having no religious affiliation or, as they have come to be known, the
“Nones.” For decades, the percentage of Americans who indicated that they
have no religion hovered between 5% and 7%. Then, beginning in the early
1990s the percentage of Nones began to rise. According to the General Social
Survey, they grew to 14% by 2000 and 21% in 2014. Other national surveys
show the same trend.
While no other trend is as dramatic as the growth of the Nones, other
indicators such as declines in religious attendance, literal belief in the Bible,
and a firm belief in God also reflect a move away from religion in the
United States or, in other words, a secular turn. In the words of Voas and
Chaves, “The evidence for a decades-long decline in American religiosity is
Emerging Trends in the Social and Behavioral Sciences.
Robert Scott and Marlis Buchmann (General Editors) with Stephen Kosslyn (Consulting Editor).
© 2017 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

now incontrovertible. Like the evidence for global warming, it comes from
multiple sources, shows up in several dimensions, and paints a consistent
factual picture” (2016, p. 1524).
We address the political implications of this secular turn, focusing on both
the political causes and consequences of rising secularism in America. The
growth of secularism cannot be understood without recognizing the inextricable links between politics and religion, and thus secularism, within the
contemporary United States.
This essay will deal with individuals’ beliefs and behavior—private
secularism—rather than Americans’ attitudes toward the separation of
church and state, or public secularism. Private and public secularism are
conceptually distinct, because individuals can believe in a strict separation of
church and state and still be highly religious in their personal lives. Indeed,
over the course of American history this has described many staunch advocates of strict separation. Likewise, it is possible to be personally secular but
nonetheless believe in a public role for religion. Importantly, to be privately
secular does not necessarily mean being hostile to religion.
CONTEXT
Understanding the full significance of the recent secular turn within the
United States requires some context. The United States has long stood
apart from other advanced, industrialized democracies for its unusually
high levels of religiosity (Norris & Inglehart, 2011). The United States
has thus long been a notable exception to what might loosely be called
secularization theory. Although there are different incarnations of the theory,
it essentially holds that as societies modernize, religious beliefs and practices
are undermined and thus decline (Gorski & Altinordu, 2008). In contrast,
Finke and Stark (2005) have argued for a completely different way of
understanding the conditions under which religion flourishes. In their view,
religious involvement is driven by competition between religions, not by
the degree of modernization. Religion, in this view, operates in a market,
and when religious groups must compete with one another for members
they respond by preaching a gospel that draws adherents. According to this
theory, the sort of gospel that draws the largest market share is characterized
by “hellfire and brimstone”—people are attracted to religions that are in
tension with secular society. Contrary to a theory of inevitable secularization
in the wake of modernization, they instead argue that there has been a rise
in the religious involvement of Americans over the course of the nation’s
history.
Putnam and Campbell (2010) show that, contrary to secularization theory,
the United States has not been marching inexorably toward ever-greater

The Politics of Secularism in the United States

3

secularization, but contrary to Finke and Stark, neither is religiosity
always rising. Instead, the course of American religious history is better
characterized as having ebbs and flows. While the 1950s, for example, were
a high period for religion in the United States, it began to decline after the
“shock” of the counter-culture of the 1960s. As attitudes toward sexual
mores changed dramatically, religiosity began to fall. Then, beginning in the
1970s religion began to rebound, owing to an “aftershock” in reaction to the
social earthquake of the 1960s. In the wake of sexual liberalization, many
Americans flocked to religions that maintained traditionalist beliefs. In
particular, evangelical Protestantism grew during this period. Importantly,
white evangelical Protestants also awakened as a political movement, when
prior to this period they had been largely absent from national politics. Thus
was born the Religious Right.
Voas and Chaves (2016) offer a different view, seeking to rehabilitate secularization theory as it applies to the United States. To that end, they have
marshaled considerable evidence to argue that, when it comes to secularization, the United States is not exceptional after all. Specifically, they contend
that the recent secular turn is consistent with secularization theory, because
secularism has grown through generational replacement. As an older cohort
of more religious Americans dies, they are replaced by a younger cohort of
less religious Americans—which is precisely the pattern in Western Europe,
where secularization theory has been most applicable. However, they also
note that secularization was delayed in the United States owing to the past
growth of evangelical Protestantism (i.e., the aftershock to the 1960s).
Voas and Chaves make a compelling case for a slowly rising tide of secularism, although their theory cannot explain the inflection point that marks
the rise of the Nones. Change through generational replacement is gradual,
and is not characterized by a sharp break with the past. Instead, the rise of
the Nones appears to have a political explanation, as a backlash to the emergence of the Religious Right as a political movement, and specifically the
Republican Party’s emphasis on social conservatism (Hout & Fischer, 2002,
2014). Nonetheless, religious nonaffiliation (None-ism) is only one indicator of secularism. While the rise of the Nones is largely due to a reaction
to the political environment, other manifestations of secularism are better
described as causes of political attitudes. In other words, whatever its origin, the “slow and steady” secularism described by Voas and Chaves has
important implications for the American political landscape. Political backlash and generational replacement are not mutually exclusive explanations
for rising secularism, as both can occur simultaneously. To that end, Hout
and Fischer, the originators of the political backlash explanation, also highlight the importance of generational replacement as contributing to the secular turn.

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

PASSIVE AND ACTIVE SECULARISM
There is an important conceptual distinction to be made between two forms
of secularism: passive and active (Layman, Campbell, Green, & Sumaktoyo,
n.d.). Passive secularism refers to “not being religious”—not identifying
with a religion, not attending religious services, and not believing in God.
Passive secularists are thus defined by what they are not. The category of
None exemplifies being passively secular, as these are people who decline
to name a religious affiliation. But religious disaffiliation is only one way
to be passively secular, as there are many people who have a nominal
affiliation to a religious tradition but nonetheless do not exhibit other forms
of religiosity.
Active secularists, however, are defined by what they are. Active secularism
is the affirmative adoption of a secular worldview and identity. In contrast
to a None, an active secularist is someone who, for example, might adopt
a self-description as an atheist. Similarly, active secularists will consciously
adopt secular principles when making ethical judgments, such as the efficacy
of reason, science, and human experience.
While, to our knowledge, we are the first to employ the terms active and
passive secularism in this way, the existing literature nonetheless provides
evidence of the need for such a distinction. For example, most Nones have at
least some belief in God and are thus what Hout and Fischer call “unchurched
believers” (2002) rather than ardent secularists. Far from having rejected religion outright, many Nones move back and forth between identifying and
not identifying with a religion. Lim, MacGregor, and Putnam (2010) demonstrate that when asked about their religious affiliation twice in just under a
year, over one third of Nones in the first interview subsequently reported
having a religion, while another third went from affiliation to nonaffiliation.
There was no evidence that these “liminal” Nones had undergone any sort
of religious conversion or disillusionment; rather, they have an uncertain
self-identity and thus, when asked at any given time, may or may not identify with a religion. These are not people who have affirmatively adopted a
secular worldview.
Similarly, sociologists of religion have written extensively about people
who describe themselves as spiritual but not religious (Fuller, 2001). By
many conventional measures of religiosity, these people would appear to
be “secular,” as they often do not attend religious services frequently or
engage in other outward forms of religious behavior. Yet their comfort with
spirituality suggests that they are far from being actively secular.
We have set about measuring passive and active secularism, to see if this
is a distinction with a difference. Passive secularism is straightforward, as it
is simply the inverse of the conventional measures of religiosity long used

The Politics of Secularism in the United States

5

by social scientists. These include the frequency of religious attendance,
frequency of prayer, religious salience, strength of belief in God, and literal
belief in scripture. The absence of a religious affiliation—being a None—also
indicates passive secularism, but because, as explained below, theory
provides reason to think that it is driven by different factors as these other
expressions of religiosity, we keep nonaffiliation separate from the other
measures of passive secularism.
Measuring active secularism is more complicated. Like measures of religiosity, it should incorporate both beliefs and identity. Furthermore, a measure of active secularism cannot reference religion since, by definition, it is
neither the absence of religiosity nor antagonism toward religion. Therefore,
the challenge is to measure secular beliefs and identity without pitting them
against religious counterparts.
We conducted a national survey in 2010–2012 that includes the conventional measures of religiosity described above as well as new measures meant
to tap into active secularism. These new measures include an index of secular beliefs. Respondents indicated their level of agreement with the following
statements; the first three are worded to affirm a secular worldview, while the
last two are worded oppositely:
1. Factual evidence from the natural world is the source of true beliefs.
2. The great works of philosophy and science are the best source of truth,
wisdom, and ethics.
3. To understand the world, we must free our minds from old traditions
and beliefs.
4. It is hard to live a good life based on reason and facts alone.
5. Values are more important than factual evidence in making moral decisions.
To gauge the salience of secularism, we asked respondents how much (if
any) guidance they received from “non-religious beliefs, such as derived
from science or philosophy.” Secular identity was measured by asking
respondents to select which (if any) of the following terms describe them:
ecumenical, mainline, charismatic/Pentecostal, humanist, non-traditional
believer, secular, atheist, fundamentalist, born again/evangelical, agnostic,
spiritual but not religious. Exploratory factor analysis reveals that four of
these terms form a single dimension of actively secular identity: secular,
humanist, atheist, and agnostic.
We have employed structural equation modeling to correct for measurement error, in order to test whether (i) our measures of active secularism
hang together as a single construct and (ii) active and passive secularism
are distinct from one another. We refer interested readers elsewhere for a

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

detailed discussion of this analysis, but the results are easy to summarize:
our measures of active secularism do indeed constitute a single dimension
and, while they are related, active and passive secularism are distinct from
one another (Layman et al., n.d.)
With our measures of active and passive secularists, we can report on how
many of each category exist within the American population. We define passive secularists as people who attend religious services “once or twice a year”
or less, receive no guidance from religion, pray no more than “at least once
a week,” do not view the Bible as the word of God, and are uncertain about
the existence of God.1 Active secularists are those who took the secular position on at least three of the five secular belief items, receive some (or more)
guidance from non-religious beliefs, and chose at least one of the secular
self-identities.
Using these definitions, we find that roughly 24% of the American population are passive secularists, while 9% are active secularists. Is 9% big or small?
It is obviously smaller than the share of passive secularists, or of Nones. On
the other hand, if a denomination, active secularists would be among the
largest in the country—larger than Southern Baptists.2 Because passive and
active secularism are not mutually exclusive, we can also report the degree
of overlap between them. Not surprisingly, the vast majority of active secularists are also passively secular. Yet among passive secularists, only 28%
also score high in active secularism. In other words, nearly three-quarters of
Americans who do not exhibit conventional religiosity have not embraced
an overtly secular perspective. They are better described by what they are
not—religious—than by what they are.
POLITICAL IMPLICATIONS
The distinction between passive and active secularism is important for
understanding the ways that secularism affects, and is affected by, the
political landscape. We hypothesize that there are two different processes at
work. First, building on the work of Hout and Fischer and others, we expect
identification as a None to rise as a reaction to the perception that religion
has become closely associated with conservative politics (Hout & Fischer,
2002, 2014; Patrikios, 2008; Putnam & Campbell, 2010). Second, we expect
political identities to have a mutually reinforcing relationship with both
passive and active secularism. Over time, Democrats and liberals should
1. We measured belief in God on a 100-point scale; respondents placed themselves on a scale indicating their certainty in the existence of God, with 0 representing absolute certainty that there is a God and
1 being absolute certainty that God does not exist. Passive secularists are defined as those who scored
above the mean, which was 19.7.
2. According to the 2014 Pew Religious Landscape Survey, roughly 5% of the U.S. population are
Southern Baptists (http://www.pewforum.org/religious-landscape-study).

The Politics of Secularism in the United States

7

become more secular/less religious, while Republicans and conservatives
become less secular/more religious. As this process unfolds, it results in
an increasing returns process—the more secularism is identified with the
political left, the more Democrats and liberals embrace it; likewise, the more
Republicans and conservatives reject it. Polarization results (Campbell,
Layman, & Green, forthcoming).
THE RISE OF THE NONES
In their seminal article on the rise of the Nones—written in 2002, at a time
when the rise was only first being noticed—Hout and Fischer argued that it
was a backlash to the emergence of the Religious Right as a political movement. A key piece of evidence for this argument is that the Nones are concentrated among political moderates and liberals. While they do not use this
language, Hout and Fischer’s account is consistent with the classic theory of
cognitive dissonance. If religion is associated with conservative politics, then
people who are not conservatives may experience dissonance between their
politics and their religion. Increasingly, they resolve the dissonance by keeping their politics and disavowing their religious affiliation. In their original
2002 paper, Hout and Fischer rely largely on circumstantial evidence to make
their case. In a 2014 update, they revisit the question with a more powerful
research design, as they employ panel data—repeated interviews with the
same people—to show that political backlash is driving the turn away from
religion. Likewise, both Putnam and Campbell (2010) and Patrikios (2008)
use different panel studies to come to the same conclusion.
We pause to note the theoretical significance of the finding that people are
turning away from a religious affiliation as a reaction to the political environment, as this reverses the longstanding assumption that religion precedes
politics. In other words, the causal arrow has been reversed; politics can drive
religion.
While compelling, by relying on observational data these studies cannot
definitively establish causation. To that end, we have designed an experiment to test whether religious disaffiliation can be induced by exposure to
politicized religion. A nationally representative sample of subjects was given
a short pre-test to measure their baseline degree of passive and active secularism. A week later, these subjects were randomly assigned to read a fictionalized but realistic local news story about a nearby congressional race. The
story highlights a Republican and Democrat running for an open seat, with
basic biographical information on both candidates. The stories differed only
in the degree to which one or both of the candidates were associated with
religion, ranging from none to moderate to high. For example, the moderate
religious treatments include photos of the candidates standing in a church,

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

while in the high religion treatments the candidates are pictured engaging
in religious behavior; the no religion treatments feature the candidates in a
generic setting. Similarly, the stories varied the degree to which the candidates are quoted as referencing their own religion, and whether they were
endorsed by faith-based groups.
As hypothesized, the experiment finds that when subjects are exposed to
a Republican candidate who invokes religion, they are more likely to disclaim a religious identity or, in other words, to identify as Nones. Conversely,
there is no effect when a Democratic candidate is associated with religion.
Interestingly, there is also no effect when both candidates are highly associated with religious rhetoric, endorsements, and imagery. An effect appears
only when the Republican candidate’s use of religion exceeds that of the
Democratic candidate. Further analysis reveals that the effect is concentrated
almost entirely among subjects who identify as Democrats. Substantively, the
effect for Democrats is sizeable; the probability that they identify as a None
increases by 13 percentage points. Recall that this is after exposure to merely
one news story about a single congressional race.
Of course, this treatment would hardly be the first time subjects were
exposed to the politicization of religion. Instead, the experimental treatment
brings to mind similar campaigns and candidates previously experienced
by the subjects. Furthermore, this much movement in self-identification
as a None is consistent with the work on the malleability of religious
nonaffiliation. It is unlikely that this one experimental treatment is causing
religious alienation among subjects who are religiously devout. Rather, it
nudges people who are ambivalent about religion toward disclaiming a
religious identity.
This experimental treatment, however, has no effect on other forms of passive secularism—attendance at religious services, prayer, religious salience,
strength of belief in God, and a literal belief in the Bible. This is to be expected,
as it would not be possible for the behaviors (attendance, prayer) to change
in the course of the experiment, and the beliefs (religious salience, belief in
God, belief in the Bible) are not subject to the same ambivalence as religious
self-identity. Likewise, the experiment has no effect on active secularism,
measured as secular beliefs, salience, and identity.
SECULARISM AND POLARIZATION
What, though, happens to passive and active secularism over time in the “real
world”? What relationship do the two forms of secularism have with politics? We are able to answer that with the aforementioned 2010–2012 panel
study. Measuring individual-level change in both political and secular orientations enables a test of whether secularism affects politics, politics affects

The Politics of Secularism in the United States

9

secularism, or if the relationship goes both ways. Political views are measured as party identification, self-described ideology, or attitudes on cultural
issues (abortion and same-sex marriage). This analysis again employs structural equation modeling, and is thus able to correct for measurement error.
First, identification as a None is only ever affected by politics, and not the
other way around. Put another way, while being a Democrat and/or liberal
may lead someone to become a None, being a None does not lead people
toward a Democratic and/or liberal orientation.
Second, there is generally a mutually reinforcing relationship between passive secularism and political views.
In other words, over time, Democrats and liberals are becoming more passively secular while passive secularism also leads to an increase in liberalism.
That is to say, we observe polarization. Importantly, though, passive secularism does not lead to a substantive change in party identification. Being
a Democrat leads to greater passive secularism, but being passively secular
does not lead people to become a Democrat.
Like passive secularism and unlike identification as a None, we again see
a mutually reinforcing relationship between active secularism and political
orientations. Unlike its passive analog, active secularism even has an effect
on party identification, which in the American context is arguably the most
stable of political orientations. Again, this is evidence of polarization along a
secular–religious fault line.3
As suggested by active secularism’s effect on political orientations, especially party identification, active secularists differ significantly from their
passive counterparts. Active secularists are highly engaged in politics—they
are more likely to identify with the Democratic Party, hold ideologically
consistent views, and participate in politics. Passive secularists, in contrast,
are more likely to be on the political sidelines, as evidenced by the fact that
being passively secular does not lead to identification with the Democratic
Party. And while passive secularism does affect liberal ideology and cultural
attitudes, those effects are much weaker than those for active secularism.
Furthermore, passive secularists are less likely than their actively secular
counterparts to be politically active.
In other words, while there has been considerable attention paid to the rise
of what we have called passive secularism, it is actually active secularism that
has the larger effect on the political landscape. Active secularists’ numbers
are fewer, but they are far more engaged in politics. As one indication
of active secularists’ high level of political engagement and strength of
3. As further evidence of the links between the political environment and secularism in its various
forms, Campbell, Green, and Layman find that this “secular polarization” only occurs among people who
perceive religion to be politicized, either those who identify evangelical Christians as “mainly Republicans” or perceive a “a lot” of talk about religion in politics.

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Democratic partisanship, Layman and Weaver (2016) have found that
roughly 56% of delegates to the 2012 Democratic National Convention can
be described as actively secular.
CHALLENGES FOR BOTH PARTIES
These results suggest challenges for both the Republican and Democratic
parties. For Republicans, a rising secular generation is a challenge for a
party so closely associated with religion. Given that religious conservatives
are often described as the party’s “base” it seems likely that this association will endure in the public mind. However, the more the Republican
Party emphasizes religion and social conservatism—especially regarding
LGBTQ rights—the more it will fan the flames of further religious disaffiliation, which makes it still more difficult for the GOP to win their
support.
Given the political profile of secularists, and especially active secularists,
they are a critical constituency for the Democratic Party and a large component of its base. Yet, unlike the way the Republicans have embraced their
base of highly religious supporters, the Democrats have not actively courted
their secularist supporters, perhaps because they fear being perceived as
the “anti-religious” party. After all, even with the secular turn, the United
States remains a highly religious nation. As an illustration of the internal
tensions within the Democratic Party, during the 2012 Democratic National
Convention delegates voted to remove any reference to God from the party
platform. Recognizing that this would only provide fodder for Republican
attacks, Democratic leaders maneuvered to restore God to the platform,
against the objections of many delegates. The Democrats have yet to find
a way to mobilize their actively secular supporters without alienating
religious voters. This is all further complicated by the fact that both African
Americans and Latinos are key components of the Democratic coalition and,
on average, highly religious.
THE FUTURE
What does the sizeable secular population in the United States portend for
the future? Because our measures of active secularism only began in 2010,
we cannot say for sure whether these views are becoming more common,
although it seems likely that they are (e.g., the rising share of Americans
who do not believe in God). Whether increasing or not, though, active secularists are highly engaged in politics and aligned with the Democratic Party.
Their heightened engagement suggests continued conflict, both political and
legal, over the many expressions of religion under government auspices that

The Politics of Secularism in the United States

11

have survived as vestiges of an earlier time when the secular population was
smaller and thus less politically influential. Examples are myriad, but they
include “under God” in the Pledge of Allegiance, religious symbols in public parks, prayers at local government meetings, and so on. Likewise, in the
court of public opinion, we can expect further political disputes over the public role of religion.
The decline in religion—that is, rise in both passive and active
secularism—will likely also have an effect on the nation’s civil society.
Religious congregations had largely held firm as participation in other forms
of civic engagement have fallen (Putnam, 2001), but it now appears that they
too are experiencing the same type of decline. This does not bode well for
philanthropic activity, especially rates of volunteering, as religious activity
is a powerful predictor of civic voluntarism—including for secular causes.
Although it is possible that secular organizations will arise to fill in the gap,
it seems unlikely that such groups will ever be full substitutes for the civic
activity facilitated by religious congregations.
While these are likely to be short-term consequences of the secular turn,
one should not assume that secularization will carry on indefinitely. Recall
that there has not been a steady decline or rise in religiosity in the United
States. Religiosity, and thus secularism, has both ebbed and flowed. Because
many passive secularists, especially the Nones, have not abandoned religious
beliefs altogether, perhaps religious entrepreneurs will find a way to win the
religiously ambivalent back to the pews. Presumably such a reawakening
would require disassociating religion from politics, or at least conservative
politics. On the other hand, perhaps the secular turn has reached the point of
no return. Whichever path the future takes, the implications for the nation’s
politics will be profound.

REFERENCES
Campbell, D. E., Layman, G. C., & Green, J. C. (forthcoming). Putting politics first: the
impact of politics on american religious and secular orientations. American Journal
of Political Science.
Finke, R., & Stark, R. (2005). The churching of America, 1776–2005: winners and losers in
our religious economy (2nd ed.). New Brunswick, NJ: Rutgers University Press.
Fuller, R. C. (2001). Spiritual, but not religious: understanding unchurched america. New
York, NY: Oxford University Press.
Gorski, P. S., & Altinordu, A. (2008). After secularization? Annual Review of Sociology,
34, 55–85.
Hout, M., & Fischer, C. S. (2002). Why more Americans have no religious preference:
Politics and generations. American Sociological Review, 67(2), 165–190.

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Hout, M., & Fischer, C. S. (2014). Explaining why more Americans have no religious
preference: Political backlash and generational succession, 1987–2012. Sociological
Science, 1(October), 423–447.
Layman, G. C., Campbell, D. E., Green, J. C., & Sumaktoyo, N. (n.d.). Active secularism
and the politics of irreligion in the United States.
Layman, G. C., & Weaver, C. (2016). Religion and secularism among American party
activists. Politics and Religion, 9(2), 271–295.
Lim, C., MacGregor, C. A., & Putnam, R. D. (2010). Secular and liminal: Discovering heterogeneity among religious nones. Journal for the Scientific Study of Religion,
49(4), 596–618.
Norris, P., & Inglehart, R. (2011). Sacred and secular: Religion and politics worldwide.
Cambridge, England: Cambridge University Press.
Patrikios, S. (2008). American republican religion? Political Behavior, 30(3), 367–389.
Putnam, R. D. (2001). Bowling alone: The collapse and revival of American community.
New York, NY: Simon & Schuster.
Putnam, R. D., & Campbell, D. E. (2010). American grace: How religion divides and unites
us. New York, NY: Simon and Schuster.
Voas, D., & Chaves, M. (2016). Is the United States a counterexample to the secularization thesis? American Journal of Sociology, 121(5), 1517–1556.

DAVID E. CAMPBELL SHORT BIOGRAPHY
David E. Campbell is the Packey J. Dee Professor of American Democracy
at the University of Notre Dame. His most recent book is Seeking the Promised
Land: Mormons and American Politics (with John Green and Quin Monson). He
is also the co-author (with Robert Putnam) of American Grace: How Religion
Divides and Unites Us, which has received both the 2011 Woodrow Wilson
Award from the American Political Science Association for the best book on
government, politics, or international affairs and the Wilbur Award from the
Religious Communicators Council for the best non-fiction book of 2010. In
addition, Campbell is the author of Why We Vote: How Schools and Communities Shape Our Civic Life, and has published numerous articles in journals such
as the American Political Science Review, American Journal of Political Science,
Journal of Politics, and Foreign Affairs.
GEOFFREY C. LAYMAN SHORT BIOGRAPHY
Geoffrey C. Layman is a Professor of Political Science at the University of
Notre Dame, specializing in political parties, political behavior, and religion
and politics. His first book, The Great Divide: Religious and Cultural Conflict
in American Party Politics examines the growing division of the Democratic
and Republican parties along religious and cultural lines. Layman’s current
research includes work with Thomas Carsey on “conflict extension” and

The Politics of Secularism in the United States

13

polarization in American party politics, and with David Campbell and John
Green on the political causes and consequences of growing secularism in
the United States. Layman has published numerous articles in the leading
journals in political science, including the American Political Science Review,
the American Journal of Political Science, the Journal of Politics, the British
Journal of Political Science, and the Annual Review of Political Science.
RELATED ESSAYS
Understanding American Political Conservatism (Political Science), Joel D.
Aberbach
Lived Religion (Sociology), Nancy T. Ammerman
Heuristic Decision Making (Political Science), Edward G. Carmines and
Nicholas J. D’Amico
Trends in Religiosity and Religious Affiliation (Sociology), Kevin J. Christiano
Institutional Change in American Religion (Sociology), Casey Clevenger and
Wendy Cadge
Patterns of Attachments across the Lifespan (Psychology), Robyn Fivush and
Theodore E. A. Waters
Empathy Gaps between Helpers and Help-Seekers: Implications for Cooperation (Psychology), Vanessa K. Bohns and Francis J. Flynn
Political Advertising (Political Science), Erika Franklin Fowler
Presidential Power (Political Science), William G. Howell
Group Identity and Political Cohesion (Political Science), Leonie Huddy
Political Inequality (Sociology), Jeff Manza
Mysticism (Sociology), Barry Markovsky and Jake Frederick
The Sociology of Religious Experience (Sociology), Douglas Porpora
Religion (Anthropology), Benjamin Grant Purzycki et al.
Atheism, Agnosticism, and Irreligion (Sociology), Buster G. Smith and Joseph
O. Baker
Complex Religion: Toward a Better Understanding of the Ways in which
Religion Intersects with Inequality (Sociology), Melissa J. Wilde and Patricia
Tevington


The Politics of Secularism in the
United States
DAVID E. CAMPBELL and GEOFFREY C. LAYMAN

Abstract
Secularism is on the rise within the United States. Religious nonaffiliation has
increased dramatically, while other measures of religiosity—worship attendance,
belief in God, belief in scripture—are waning. This essay considers the political
implications of this secular turn, in light of the historically high rates of religiosity within the United States. In doing so, we distinguish between passive
and active secularism and describe measures of each. Passive secularism refers
to not being religious—not identifying with a religion, not attending religious
services, and not believing in God. Active secularism is the affirmative adoption of
a secular worldview and identity. With this distinction, we can trace Americans’
growing political polarization along religious–secular lines. One form of passive
secularism—religious nonaffiliation—is often triggered by a backlash to the Religious Right. More generally, both passively and actively secular voters are moving
to the political left. Furthermore, while smaller in number than passive secularists,
active secularists are highly engaged in politics, comprising a sizeable share of the
Democratic Party’s activist base.

The United States is experiencing a rise in secularism. Most notably, the
last 25 years have seen a dramatic rise in the percentage of Americans who
report having no religious affiliation or, as they have come to be known, the
“Nones.” For decades, the percentage of Americans who indicated that they
have no religion hovered between 5% and 7%. Then, beginning in the early
1990s the percentage of Nones began to rise. According to the General Social
Survey, they grew to 14% by 2000 and 21% in 2014. Other national surveys
show the same trend.
While no other trend is as dramatic as the growth of the Nones, other
indicators such as declines in religious attendance, literal belief in the Bible,
and a firm belief in God also reflect a move away from religion in the
United States or, in other words, a secular turn. In the words of Voas and
Chaves, “The evidence for a decades-long decline in American religiosity is
Emerging Trends in the Social and Behavioral Sciences.
Robert Scott and Marlis Buchmann (General Editors) with Stephen Kosslyn (Consulting Editor).
© 2017 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

now incontrovertible. Like the evidence for global warming, it comes from
multiple sources, shows up in several dimensions, and paints a consistent
factual picture” (2016, p. 1524).
We address the political implications of this secular turn, focusing on both
the political causes and consequences of rising secularism in America. The
growth of secularism cannot be understood without recognizing the inextricable links between politics and religion, and thus secularism, within the
contemporary United States.
This essay will deal with individuals’ beliefs and behavior—private
secularism—rather than Americans’ attitudes toward the separation of
church and state, or public secularism. Private and public secularism are
conceptually distinct, because individuals can believe in a strict separation of
church and state and still be highly religious in their personal lives. Indeed,
over the course of American history this has described many staunch advocates of strict separation. Likewise, it is possible to be personally secular but
nonetheless believe in a public role for religion. Importantly, to be privately
secular does not necessarily mean being hostile to religion.
CONTEXT
Understanding the full significance of the recent secular turn within the
United States requires some context. The United States has long stood
apart from other advanced, industrialized democracies for its unusually
high levels of religiosity (Norris & Inglehart, 2011). The United States
has thus long been a notable exception to what might loosely be called
secularization theory. Although there are different incarnations of the theory,
it essentially holds that as societies modernize, religious beliefs and practices
are undermined and thus decline (Gorski & Altinordu, 2008). In contrast,
Finke and Stark (2005) have argued for a completely different way of
understanding the conditions under which religion flourishes. In their view,
religious involvement is driven by competition between religions, not by
the degree of modernization. Religion, in this view, operates in a market,
and when religious groups must compete with one another for members
they respond by preaching a gospel that draws adherents. According to this
theory, the sort of gospel that draws the largest market share is characterized
by “hellfire and brimstone”—people are attracted to religions that are in
tension with secular society. Contrary to a theory of inevitable secularization
in the wake of modernization, they instead argue that there has been a rise
in the religious involvement of Americans over the course of the nation’s
history.
Putnam and Campbell (2010) show that, contrary to secularization theory,
the United States has not been marching inexorably toward ever-greater

The Politics of Secularism in the United States

3

secularization, but contrary to Finke and Stark, neither is religiosity
always rising. Instead, the course of American religious history is better
characterized as having ebbs and flows. While the 1950s, for example, were
a high period for religion in the United States, it began to decline after the
“shock” of the counter-culture of the 1960s. As attitudes toward sexual
mores changed dramatically, religiosity began to fall. Then, beginning in the
1970s religion began to rebound, owing to an “aftershock” in reaction to the
social earthquake of the 1960s. In the wake of sexual liberalization, many
Americans flocked to religions that maintained traditionalist beliefs. In
particular, evangelical Protestantism grew during this period. Importantly,
white evangelical Protestants also awakened as a political movement, when
prior to this period they had been largely absent from national politics. Thus
was born the Religious Right.
Voas and Chaves (2016) offer a different view, seeking to rehabilitate secularization theory as it applies to the United States. To that end, they have
marshaled considerable evidence to argue that, when it comes to secularization, the United States is not exceptional after all. Specifically, they contend
that the recent secular turn is consistent with secularization theory, because
secularism has grown through generational replacement. As an older cohort
of more religious Americans dies, they are replaced by a younger cohort of
less religious Americans—which is precisely the pattern in Western Europe,
where secularization theory has been most applicable. However, they also
note that secularization was delayed in the United States owing to the past
growth of evangelical Protestantism (i.e., the aftershock to the 1960s).
Voas and Chaves make a compelling case for a slowly rising tide of secularism, although their theory cannot explain the inflection point that marks
the rise of the Nones. Change through generational replacement is gradual,
and is not characterized by a sharp break with the past. Instead, the rise of
the Nones appears to have a political explanation, as a backlash to the emergence of the Religious Right as a political movement, and specifically the
Republican Party’s emphasis on social conservatism (Hout & Fischer, 2002,
2014). Nonetheless, religious nonaffiliation (None-ism) is only one indicator of secularism. While the rise of the Nones is largely due to a reaction
to the political environment, other manifestations of secularism are better
described as causes of political attitudes. In other words, whatever its origin, the “slow and steady” secularism described by Voas and Chaves has
important implications for the American political landscape. Political backlash and generational replacement are not mutually exclusive explanations
for rising secularism, as both can occur simultaneously. To that end, Hout
and Fischer, the originators of the political backlash explanation, also highlight the importance of generational replacement as contributing to the secular turn.

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

PASSIVE AND ACTIVE SECULARISM
There is an important conceptual distinction to be made between two forms
of secularism: passive and active (Layman, Campbell, Green, & Sumaktoyo,
n.d.). Passive secularism refers to “not being religious”—not identifying
with a religion, not attending religious services, and not believing in God.
Passive secularists are thus defined by what they are not. The category of
None exemplifies being passively secular, as these are people who decline
to name a religious affiliation. But religious disaffiliation is only one way
to be passively secular, as there are many people who have a nominal
affiliation to a religious tradition but nonetheless do not exhibit other forms
of religiosity.
Active secularists, however, are defined by what they are. Active secularism
is the affirmative adoption of a secular worldview and identity. In contrast
to a None, an active secularist is someone who, for example, might adopt
a self-description as an atheist. Similarly, active secularists will consciously
adopt secular principles when making ethical judgments, such as the efficacy
of reason, science, and human experience.
While, to our knowledge, we are the first to employ the terms active and
passive secularism in this way, the existing literature nonetheless provides
evidence of the need for such a distinction. For example, most Nones have at
least some belief in God and are thus what Hout and Fischer call “unchurched
believers” (2002) rather than ardent secularists. Far from having rejected religion outright, many Nones move back and forth between identifying and
not identifying with a religion. Lim, MacGregor, and Putnam (2010) demonstrate that when asked about their religious affiliation twice in just under a
year, over one third of Nones in the first interview subsequently reported
having a religion, while another third went from affiliation to nonaffiliation.
There was no evidence that these “liminal” Nones had undergone any sort
of religious conversion or disillusionment; rather, they have an uncertain
self-identity and thus, when asked at any given time, may or may not identify with a religion. These are not people who have affirmatively adopted a
secular worldview.
Similarly, sociologists of religion have written extensively about people
who describe themselves as spiritual but not religious (Fuller, 2001). By
many conventional measures of religiosity, these people would appear to
be “secular,” as they often do not attend religious services frequently or
engage in other outward forms of religious behavior. Yet their comfort with
spirituality suggests that they are far from being actively secular.
We have set about measuring passive and active secularism, to see if this
is a distinction with a difference. Passive secularism is straightforward, as it
is simply the inverse of the conventional measures of religiosity long used

The Politics of Secularism in the United States

5

by social scientists. These include the frequency of religious attendance,
frequency of prayer, religious salience, strength of belief in God, and literal
belief in scripture. The absence of a religious affiliation—being a None—also
indicates passive secularism, but because, as explained below, theory
provides reason to think that it is driven by different factors as these other
expressions of religiosity, we keep nonaffiliation separate from the other
measures of passive secularism.
Measuring active secularism is more complicated. Like measures of religiosity, it should incorporate both beliefs and identity. Furthermore, a measure of active secularism cannot reference religion since, by definition, it is
neither the absence of religiosity nor antagonism toward religion. Therefore,
the challenge is to measure secular beliefs and identity without pitting them
against religious counterparts.
We conducted a national survey in 2010–2012 that includes the conventional measures of religiosity described above as well as new measures meant
to tap into active secularism. These new measures include an index of secular beliefs. Respondents indicated their level of agreement with the following
statements; the first three are worded to affirm a secular worldview, while the
last two are worded oppositely:
1. Factual evidence from the natural world is the source of true beliefs.
2. The great works of philosophy and science are the best source of truth,
wisdom, and ethics.
3. To understand the world, we must free our minds from old traditions
and beliefs.
4. It is hard to live a good life based on reason and facts alone.
5. Values are more important than factual evidence in making moral decisions.
To gauge the salience of secularism, we asked respondents how much (if
any) guidance they received from “non-religious beliefs, such as derived
from science or philosophy.” Secular identity was measured by asking
respondents to select which (if any) of the following terms describe them:
ecumenical, mainline, charismatic/Pentecostal, humanist, non-traditional
believer, secular, atheist, fundamentalist, born again/evangelical, agnostic,
spiritual but not religious. Exploratory factor analysis reveals that four of
these terms form a single dimension of actively secular identity: secular,
humanist, atheist, and agnostic.
We have employed structural equation modeling to correct for measurement error, in order to test whether (i) our measures of active secularism
hang together as a single construct and (ii) active and passive secularism
are distinct from one another. We refer interested readers elsewhere for a

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

detailed discussion of this analysis, but the results are easy to summarize:
our measures of active secularism do indeed constitute a single dimension
and, while they are related, active and passive secularism are distinct from
one another (Layman et al., n.d.)
With our measures of active and passive secularists, we can report on how
many of each category exist within the American population. We define passive secularists as people who attend religious services “once or twice a year”
or less, receive no guidance from religion, pray no more than “at least once
a week,” do not view the Bible as the word of God, and are uncertain about
the existence of God.1 Active secularists are those who took the secular position on at least three of the five secular belief items, receive some (or more)
guidance from non-religious beliefs, and chose at least one of the secular
self-identities.
Using these definitions, we find that roughly 24% of the American population are passive secularists, while 9% are active secularists. Is 9% big or small?
It is obviously smaller than the share of passive secularists, or of Nones. On
the other hand, if a denomination, active secularists would be among the
largest in the country—larger than Southern Baptists.2 Because passive and
active secularism are not mutually exclusive, we can also report the degree
of overlap between them. Not surprisingly, the vast majority of active secularists are also passively secular. Yet among passive secularists, only 28%
also score high in active secularism. In other words, nearly three-quarters of
Americans who do not exhibit conventional religiosity have not embraced
an overtly secular perspective. They are better described by what they are
not—religious—than by what they are.
POLITICAL IMPLICATIONS
The distinction between passive and active secularism is important for
understanding the ways that secularism affects, and is affected by, the
political landscape. We hypothesize that there are two different processes at
work. First, building on the work of Hout and Fischer and others, we expect
identification as a None to rise as a reaction to the perception that religion
has become closely associated with conservative politics (Hout & Fischer,
2002, 2014; Patrikios, 2008; Putnam & Campbell, 2010). Second, we expect
political identities to have a mutually reinforcing relationship with both
passive and active secularism. Over time, Democrats and liberals should
1. We measured belief in God on a 100-point scale; respondents placed themselves on a scale indicating their certainty in the existence of God, with 0 representing absolute certainty that there is a God and
1 being absolute certainty that God does not exist. Passive secularists are defined as those who scored
above the mean, which was 19.7.
2. According to the 2014 Pew Religious Landscape Survey, roughly 5% of the U.S. population are
Southern Baptists (http://www.pewforum.org/religious-landscape-study).

The Politics of Secularism in the United States

7

become more secular/less religious, while Republicans and conservatives
become less secular/more religious. As this process unfolds, it results in
an increasing returns process—the more secularism is identified with the
political left, the more Democrats and liberals embrace it; likewise, the more
Republicans and conservatives reject it. Polarization results (Campbell,
Layman, & Green, forthcoming).
THE RISE OF THE NONES
In their seminal article on the rise of the Nones—written in 2002, at a time
when the rise was only first being noticed—Hout and Fischer argued that it
was a backlash to the emergence of the Religious Right as a political movement. A key piece of evidence for this argument is that the Nones are concentrated among political moderates and liberals. While they do not use this
language, Hout and Fischer’s account is consistent with the classic theory of
cognitive dissonance. If religion is associated with conservative politics, then
people who are not conservatives may experience dissonance between their
politics and their religion. Increasingly, they resolve the dissonance by keeping their politics and disavowing their religious affiliation. In their original
2002 paper, Hout and Fischer rely largely on circumstantial evidence to make
their case. In a 2014 update, they revisit the question with a more powerful
research design, as they employ panel data—repeated interviews with the
same people—to show that political backlash is driving the turn away from
religion. Likewise, both Putnam and Campbell (2010) and Patrikios (2008)
use different panel studies to come to the same conclusion.
We pause to note the theoretical significance of the finding that people are
turning away from a religious affiliation as a reaction to the political environment, as this reverses the longstanding assumption that religion precedes
politics. In other words, the causal arrow has been reversed; politics can drive
religion.
While compelling, by relying on observational data these studies cannot
definitively establish causation. To that end, we have designed an experiment to test whether religious disaffiliation can be induced by exposure to
politicized religion. A nationally representative sample of subjects was given
a short pre-test to measure their baseline degree of passive and active secularism. A week later, these subjects were randomly assigned to read a fictionalized but realistic local news story about a nearby congressional race. The
story highlights a Republican and Democrat running for an open seat, with
basic biographical information on both candidates. The stories differed only
in the degree to which one or both of the candidates were associated with
religion, ranging from none to moderate to high. For example, the moderate
religious treatments include photos of the candidates standing in a church,

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

while in the high religion treatments the candidates are pictured engaging
in religious behavior; the no religion treatments feature the candidates in a
generic setting. Similarly, the stories varied the degree to which the candidates are quoted as referencing their own religion, and whether they were
endorsed by faith-based groups.
As hypothesized, the experiment finds that when subjects are exposed to
a Republican candidate who invokes religion, they are more likely to disclaim a religious identity or, in other words, to identify as Nones. Conversely,
there is no effect when a Democratic candidate is associated with religion.
Interestingly, there is also no effect when both candidates are highly associated with religious rhetoric, endorsements, and imagery. An effect appears
only when the Republican candidate’s use of religion exceeds that of the
Democratic candidate. Further analysis reveals that the effect is concentrated
almost entirely among subjects who identify as Democrats. Substantively, the
effect for Democrats is sizeable; the probability that they identify as a None
increases by 13 percentage points. Recall that this is after exposure to merely
one news story about a single congressional race.
Of course, this treatment would hardly be the first time subjects were
exposed to the politicization of religion. Instead, the experimental treatment
brings to mind similar campaigns and candidates previously experienced
by the subjects. Furthermore, this much movement in self-identification
as a None is consistent with the work on the malleability of religious
nonaffiliation. It is unlikely that this one experimental treatment is causing
religious alienation among subjects who are religiously devout. Rather, it
nudges people who are ambivalent about religion toward disclaiming a
religious identity.
This experimental treatment, however, has no effect on other forms of passive secularism—attendance at religious services, prayer, religious salience,
strength of belief in God, and a literal belief in the Bible. This is to be expected,
as it would not be possible for the behaviors (attendance, prayer) to change
in the course of the experiment, and the beliefs (religious salience, belief in
God, belief in the Bible) are not subject to the same ambivalence as religious
self-identity. Likewise, the experiment has no effect on active secularism,
measured as secular beliefs, salience, and identity.
SECULARISM AND POLARIZATION
What, though, happens to passive and active secularism over time in the “real
world”? What relationship do the two forms of secularism have with politics? We are able to answer that with the aforementioned 2010–2012 panel
study. Measuring individual-level change in both political and secular orientations enables a test of whether secularism affects politics, politics affects

The Politics of Secularism in the United States

9

secularism, or if the relationship goes both ways. Political views are measured as party identification, self-described ideology, or attitudes on cultural
issues (abortion and same-sex marriage). This analysis again employs structural equation modeling, and is thus able to correct for measurement error.
First, identification as a None is only ever affected by politics, and not the
other way around. Put another way, while being a Democrat and/or liberal
may lead someone to become a None, being a None does not lead people
toward a Democratic and/or liberal orientation.
Second, there is generally a mutually reinforcing relationship between passive secularism and political views.
In other words, over time, Democrats and liberals are becoming more passively secular while passive secularism also leads to an increase in liberalism.
That is to say, we observe polarization. Importantly, though, passive secularism does not lead to a substantive change in party identification. Being
a Democrat leads to greater passive secularism, but being passively secular
does not lead people to become a Democrat.
Like passive secularism and unlike identification as a None, we again see
a mutually reinforcing relationship between active secularism and political
orientations. Unlike its passive analog, active secularism even has an effect
on party identification, which in the American context is arguably the most
stable of political orientations. Again, this is evidence of polarization along a
secular–religious fault line.3
As suggested by active secularism’s effect on political orientations, especially party identification, active secularists differ significantly from their
passive counterparts. Active secularists are highly engaged in politics—they
are more likely to identify with the Democratic Party, hold ideologically
consistent views, and participate in politics. Passive secularists, in contrast,
are more likely to be on the political sidelines, as evidenced by the fact that
being passively secular does not lead to identification with the Democratic
Party. And while passive secularism does affect liberal ideology and cultural
attitudes, those effects are much weaker than those for active secularism.
Furthermore, passive secularists are less likely than their actively secular
counterparts to be politically active.
In other words, while there has been considerable attention paid to the rise
of what we have called passive secularism, it is actually active secularism that
has the larger effect on the political landscape. Active secularists’ numbers
are fewer, but they are far more engaged in politics. As one indication
of active secularists’ high level of political engagement and strength of
3. As further evidence of the links between the political environment and secularism in its various
forms, Campbell, Green, and Layman find that this “secular polarization” only occurs among people who
perceive religion to be politicized, either those who identify evangelical Christians as “mainly Republicans” or perceive a “a lot” of talk about religion in politics.

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Democratic partisanship, Layman and Weaver (2016) have found that
roughly 56% of delegates to the 2012 Democratic National Convention can
be described as actively secular.
CHALLENGES FOR BOTH PARTIES
These results suggest challenges for both the Republican and Democratic
parties. For Republicans, a rising secular generation is a challenge for a
party so closely associated with religion. Given that religious conservatives
are often described as the party’s “base” it seems likely that this association will endure in the public mind. However, the more the Republican
Party emphasizes religion and social conservatism—especially regarding
LGBTQ rights—the more it will fan the flames of further religious disaffiliation, which makes it still more difficult for the GOP to win their
support.
Given the political profile of secularists, and especially active secularists,
they are a critical constituency for the Democratic Party and a large component of its base. Yet, unlike the way the Republicans have embraced their
base of highly religious supporters, the Democrats have not actively courted
their secularist supporters, perhaps because they fear being perceived as
the “anti-religious” party. After all, even with the secular turn, the United
States remains a highly religious nation. As an illustration of the internal
tensions within the Democratic Party, during the 2012 Democratic National
Convention delegates voted to remove any reference to God from the party
platform. Recognizing that this would only provide fodder for Republican
attacks, Democratic leaders maneuvered to restore God to the platform,
against the objections of many delegates. The Democrats have yet to find
a way to mobilize their actively secular supporters without alienating
religious voters. This is all further complicated by the fact that both African
Americans and Latinos are key components of the Democratic coalition and,
on average, highly religious.
THE FUTURE
What does the sizeable secular population in the United States portend for
the future? Because our measures of active secularism only began in 2010,
we cannot say for sure whether these views are becoming more common,
although it seems likely that they are (e.g., the rising share of Americans
who do not believe in God). Whether increasing or not, though, active secularists are highly engaged in politics and aligned with the Democratic Party.
Their heightened engagement suggests continued conflict, both political and
legal, over the many expressions of religion under government auspices that

The Politics of Secularism in the United States

11

have survived as vestiges of an earlier time when the secular population was
smaller and thus less politically influential. Examples are myriad, but they
include “under God” in the Pledge of Allegiance, religious symbols in public parks, prayers at local government meetings, and so on. Likewise, in the
court of public opinion, we can expect further political disputes over the public role of religion.
The decline in religion—that is, rise in both passive and active
secularism—will likely also have an effect on the nation’s civil society.
Religious congregations had largely held firm as participation in other forms
of civic engagement have fallen (Putnam, 2001), but it now appears that they
too are experiencing the same type of decline. This does not bode well for
philanthropic activity, especially rates of volunteering, as religious activity
is a powerful predictor of civic voluntarism—including for secular causes.
Although it is possible that secular organizations will arise to fill in the gap,
it seems unlikely that such groups will ever be full substitutes for the civic
activity facilitated by religious congregations.
While these are likely to be short-term consequences of the secular turn,
one should not assume that secularization will carry on indefinitely. Recall
that there has not been a steady decline or rise in religiosity in the United
States. Religiosity, and thus secularism, has both ebbed and flowed. Because
many passive secularists, especially the Nones, have not abandoned religious
beliefs altogether, perhaps religious entrepreneurs will find a way to win the
religiously ambivalent back to the pews. Presumably such a reawakening
would require disassociating religion from politics, or at least conservative
politics. On the other hand, perhaps the secular turn has reached the point of
no return. Whichever path the future takes, the implications for the nation’s
politics will be profound.

REFERENCES
Campbell, D. E., Layman, G. C., & Green, J. C. (forthcoming). Putting politics first: the
impact of politics on american religious and secular orientations. American Journal
of Political Science.
Finke, R., & Stark, R. (2005). The churching of America, 1776–2005: winners and losers in
our religious economy (2nd ed.). New Brunswick, NJ: Rutgers University Press.
Fuller, R. C. (2001). Spiritual, but not religious: understanding unchurched america. New
York, NY: Oxford University Press.
Gorski, P. S., & Altinordu, A. (2008). After secularization? Annual Review of Sociology,
34, 55–85.
Hout, M., & Fischer, C. S. (2002). Why more Americans have no religious preference:
Politics and generations. American Sociological Review, 67(2), 165–190.

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Hout, M., & Fischer, C. S. (2014). Explaining why more Americans have no religious
preference: Political backlash and generational succession, 1987–2012. Sociological
Science, 1(October), 423–447.
Layman, G. C., Campbell, D. E., Green, J. C., & Sumaktoyo, N. (n.d.). Active secularism
and the politics of irreligion in the United States.
Layman, G. C., & Weaver, C. (2016). Religion and secularism among American party
activists. Politics and Religion, 9(2), 271–295.
Lim, C., MacGregor, C. A., & Putnam, R. D. (2010). Secular and liminal: Discovering heterogeneity among religious nones. Journal for the Scientific Study of Religion,
49(4), 596–618.
Norris, P., & Inglehart, R. (2011). Sacred and secular: Religion and politics worldwide.
Cambridge, England: Cambridge University Press.
Patrikios, S. (2008). American republican religion? Political Behavior, 30(3), 367–389.
Putnam, R. D. (2001). Bowling alone: The collapse and revival of American community.
New York, NY: Simon & Schuster.
Putnam, R. D., & Campbell, D. E. (2010). American grace: How religion divides and unites
us. New York, NY: Simon and Schuster.
Voas, D., & Chaves, M. (2016). Is the United States a counterexample to the secularization thesis? American Journal of Sociology, 121(5), 1517–1556.

DAVID E. CAMPBELL SHORT BIOGRAPHY
David E. Campbell is the Packey J. Dee Professor of American Democracy
at the University of Notre Dame. His most recent book is Seeking the Promised
Land: Mormons and American Politics (with John Green and Quin Monson). He
is also the co-author (with Robert Putnam) of American Grace: How Religion
Divides and Unites Us, which has received both the 2011 Woodrow Wilson
Award from the American Political Science Association for the best book on
government, politics, or international affairs and the Wilbur Award from the
Religious Communicators Council for the best non-fiction book of 2010. In
addition, Campbell is the author of Why We Vote: How Schools and Communities Shape Our Civic Life, and has published numerous articles in journals such
as the American Political Science Review, American Journal of Political Science,
Journal of Politics, and Foreign Affairs.
GEOFFREY C. LAYMAN SHORT BIOGRAPHY
Geoffrey C. Layman is a Professor of Political Science at the University of
Notre Dame, specializing in political parties, political behavior, and religion
and politics. His first book, The Great Divide: Religious and Cultural Conflict
in American Party Politics examines the growing division of the Democratic
and Republican parties along religious and cultural lines. Layman’s current
research includes work with Thomas Carsey on “conflict extension” and

The Politics of Secularism in the United States

13

polarization in American party politics, and with David Campbell and John
Green on the political causes and consequences of growing secularism in
the United States. Layman has published numerous articles in the leading
journals in political science, including the American Political Science Review,
the American Journal of Political Science, the Journal of Politics, the British
Journal of Political Science, and the Annual Review of Political Science.
RELATED ESSAYS
Understanding American Political Conservatism (Political Science), Joel D.
Aberbach
Lived Religion (Sociology), Nancy T. Ammerman
Heuristic Decision Making (Political Science), Edward G. Carmines and
Nicholas J. D’Amico
Trends in Religiosity and Religious Affiliation (Sociology), Kevin J. Christiano
Institutional Change in American Religion (Sociology), Casey Clevenger and
Wendy Cadge
Patterns of Attachments across the Lifespan (Psychology), Robyn Fivush and
Theodore E. A. Waters
Empathy Gaps between Helpers and Help-Seekers: Implications for Cooperation (Psychology), Vanessa K. Bohns and Francis J. Flynn
Political Advertising (Political Science), Erika Franklin Fowler
Presidential Power (Political Science), William G. Howell
Group Identity and Political Cohesion (Political Science), Leonie Huddy
Political Inequality (Sociology), Jeff Manza
Mysticism (Sociology), Barry Markovsky and Jake Frederick
The Sociology of Religious Experience (Sociology), Douglas Porpora
Religion (Anthropology), Benjamin Grant Purzycki et al.
Atheism, Agnosticism, and Irreligion (Sociology), Buster G. Smith and Joseph
O. Baker
Complex Religion: Toward a Better Understanding of the Ways in which
Religion Intersects with Inequality (Sociology), Melissa J. Wilde and Patricia
Tevington



Media

The Politics of Secularism in the United States