Religion and Political Decision-Making

Abstract: Influential political theorists suggest that religious differences in political life may be overcome through shared commitment to political processes. In this paper I subject the underlying assumptions of this proposition to empirical inquiry. When faced with substantive conflict over policy outcomes, do religious persons defer to a political process for resolution? And if so, to which political process do they defer? Through a novel interview exercise with 61 respondents from a variety of religious backgrounds, I find a general willingness to defer to a legitimate political process, even if it results in an undesirable outcome that violates religious (or other) political preferences. However, I also find that a political process need not be democratic to be seen as legitimate, and that process preferences do not map onto religious differences.

Author: Michael S. Evans

Keywords: democracy, politics, culture

Notice: The final, definitive version of this article is published in Journal for the Scientific Study of Religion 53(1):145-163, 2014, doi:10.1111/jssr.12088 and is available at www.blackwell-synergy.com. For quoting or citing, please refer to the published version.


Questions about the proper role of religion in public life remain at the forefront of scholarly discussion (Audi and Wolterstorff 1997; Rawls 1997; Habermas 2006), sustained in part by popular concerns over the possibility of a fundamentalist Christian “American theocracy” (Phillips 2006) or the rise of fundamentalist Islam in Europe (e.g. Steyn 2006). In response to widely shared concerns over the potential threat that religious differences pose to democracy, prominent political theorists John Rawls and Jürgen Habermas have proposed a procedural solution: all participants in a democratic society should share an overriding commitment to the political process (Rawls 1997; Habermas 2006, see also McCarthy 1994; Yates 2007). In practice, such a commitment requires religious and non-religious persons alike to defer to the outcome of a legitimate political process, even if that process generates results that go against what they believe or advocate in substantive terms.

But is this reasonable to expect? Will persons with different religious commitments actually defer to political processes to resolve substantive differences over matters of policy? This is a basic empirical question. Unfortunately there is a gap in existing literature where the answers ought to be. Of course many sociologists have documented religious sources of substantive political differences over contentious issues such as abortion (Evans 2002; Hoffman and Johnson 2005), euthanasia (Moulton, Hill, and Burdette 2006), and same-sex marriage (Olson, Cadge, and Harrison 2006; Sherkat, De Vries, and Creek 2010). But little attention has been paid to religion as a source of difference over political processes rather than political outcomes. Some sociologists have studied how people reason through and resolve (or avoid) substantive political differences (e.g. Eliasoph 1998; Perrin 2006), and some political scientists have studied general political process preferences (e.g. Hibbing and Theiss-Morse 2002). However, such studies rarely explicitly engage religion or systematically assess religious differences (but see Djupe and Gwiasda 2010).

This paper begins to fill this gap in the literature by examining how religion matters to political decision-making. Of course it is practically impossible to test how religion matters in all ways to all political decision-making in every setting. In this study I use the term “political decision-making” specifically to refer to situations in which individuals in a democratic society (in this case the United States) assess an issue, evaluate options, and recognize a legitimate resolution to substantive political differences. I report the results from a novel interview exercise delivered to 61 respondents, purposively recruited from different religious traditions, in order to elicit political decision-making over issues such as stem cell research, human origins, environment, and the origins of sexuality. Through qualitative analysis of interview responses, I examine how these respondents confront substantive political differences and recognize a legitimate resolution for resolving those differences, noting especially how religion matters (or not) to such political decision-making. This study thus extends the sociology of religion into the study of political processes, while offering a novel empirical contribution to current theoretical debates over religion in public life.

I find that most respondents navigate the exercise by deferring to political processes that they see as legitimate, even if the outcome of such processes violates their closely-held position on an issue that matters to them. Very few respondents simply want to get their way. But otherwise there is no consistent preference for a particular process that is considered most legitimate. Some respondents always prefer a popular democratic vote, but many respondents oppose majoritarianism entirely, identify circumstances under which they would seek alternatives to majoritarian resolution, or simply respect whatever political process is already in place. However, these differences in political decision-making do not generally align with religious differences in the sample, suggesting that religion matters less than other cultural factors in determining what kinds of processes are seen as legitimate. I conclude by discussing the implications of these findings for the sociology of religion and for democratic theory.

Religious Difference as a Problem for Democracy

The proper role of religion in politics has long been an issue in political philosophy (Audi and Wolterstorff 1997). In recent years, debate over religion has taken on increasing prominence in academic discourse about liberal democracy (e.g. Asad 1993; Eberle 2002; Stout 2004; Habermas 2006; Ratzinger and Habermas 2006; Heclo 2007; Taylor 2007; Mendieta and VanAntwerpen 2011; Calhoun, Mendieta, and VanAntwerpen forthcoming). Many influential democratic theorists do not theorize the role of religion in particular (e.g. Honig 1993). And many democratic theorists subsume religion under other, broader, categories such as “culture” (e.g. Kymlicka 1995). But a growing body of work suggests that religion is an important source of meaning, difference, and solidarity that theories of democracy must take into account (see also Stout 2004; Alexander 2006).

Among influential political theorists who have explicitly engaged religion and democracy, one of the most significant points of debate is the unique threat potential that religious differences might pose to a democratic polity (see, e.g., Audi and Wolterstorff 1997; Rawls 1997; Audi 2000; Connolly 2005; Habermas 2006; Ratzinger and Habermas 2007; Mendieta and VanAntwerpen 2011). The basic concern is that religious differences link political differences to a source of legitimacy that is external to the state and to many of its citizens, potentially leading to the subversion of liberal democracy by religious citizens seeking to achieve their goals by any means necessary. Of course, not all theorists share this concern (see, e.g., Stout 2004 for an alternative view). But worth noting is that this concern is shared across different normative ideals of democracy. Religious differences are not simply seen as a potential threat to deliberation (Habermas 2006), but also as a potential threat to ideals of justice (Rawls 1997) and to respectful contestation (Connolly 2005).

If religion is a potential threat to democracy, what solution would mitigate this threat? Among those democratic theorists who explicitly address religion, there is, unsurprisingly, a wide range of proposed solutions. For example, Robert Audi suggests excluding religion from politics altogether (Audi 2000). William Connolly suggests cultivating an internalized “deep pluralism” (Connolly 2005). And Nicholas Wolterstorff suggests encouraging and fostering religious participation in politics (Audi and Wolterstorff 1997). These few simplified examples of theoretical responses do not do justice to the wide range of responses that are available in the literature, and it is impossible for one article to summarize, much less arbitrate, the many and varied theoretical solutions to the problem of religious difference.

So in this paper I focus on one of the most influential normative responses to the problem of religious difference: the “procedural solution” offered by Jürgen Habermas and John Rawls. Following convention in political theory, I use the term “procedural solution” to indicate that this solution to religious differences depends on process commitments (e.g. to a democratic vote) outweighing substantive commitments (e.g. to a desired policy outcome). In the procedural solution, political participation need only be compatible with “the essentials of public reason and a democratic polity” (Rawls 1997:766). Religious traditions, with their “special power to articulate moral intuitions,” are welcome in politics, but ultimately must defer to “democratic procedure” that “has the power to generate legitimacy precisely because it both includes all participants and has a deliberative character” (Habermas 2006:10,12). In short, if there is a basic conflict between substantive religious commitments and political process, the procedural solution resolves this conflict by requiring religious and non-religious participants alike to defer to a political process.

I focus on the procedural solution for four reasons. First, it is one of the most influential normative propositions in democratic theory about religion. This is obviously true in the basic sense that a substantial part of academic debate about religion and democracy engages the procedural solution (see, e.g., Stout 2004; Mendieta and VanAntwerpen 2011; Calhoun, Mendieta and VanAntwerpen forthcoming). But it is also true in the more general sense that the question of whether the procedural solution works, or not, has significant implications for public policy. For example, government funding of “faith-based” organizations in the United States depends on members of such religious organizations setting aside any substantive preference for proselytization in order to comply with settled First Amendment law concerning the establishment of religion (see Sager 2010). If the procedural solution is unrealistic, then such restrictions are also unrealistic (see also Leiter 2012 on religious exemptions under law).

Second, unlike some other theoretical solutions to the problem of religious difference, the procedural solution is amenable to empirical assessment using social scientific methods. Empirical methods alone cannot determine what someone should do. In that sense any normative proposition at least partly remains inviolable by empirical research. But the procedural solution is built on two empirical propositions about conflict between substantive preferences and process preferences (see also Hibbing and Theiss-Morse 2002). The first is that religious differences actually pose a threat by granting special status to substantive preferences, such that religious persons are especially likely to prefer substantive outcomes rather than defer to a political process. The second is that, given the problem presented by the first proposition, a shared commitment to some political process is workable in practice as a solution. Both of these empirical propositions can be evaluated through social scientific methods such as in-depth qualitative interviewing. By contrast, a solution like Connolly’s (2005) “deep pluralism,” while useful for thinking through theoretical implications, is difficult to test through social scientific methods, as it requires simultaneous access to, and comparison of, individual internal states relative to an ideally pluralized internal state. The empirical propositions underlying the procedural solution are not only important to evaluate empirically, they are also possible to evaluate empirically, with the recognition that such evaluation might not be fully dispositive of the normative question.

Third, focusing on the procedural solution helps expand our understanding of whether or not, and how, religious differences actually matter to politics. In addition to testing whether or not religious persons will defer to a political process at all, a social scientific assessment of the procedural solution can test whether or not there are religious preferences for process alternatives. Persons faced with a conflict between substantive and process preferences may well prefer to defer to one kind of process rather than another (e.g. elite command rather than a popular vote). This is key to sorting out the role of religion in political decision-making. If the question is whether substantive differences lead to subverting any legitimate political process, then substantive religious differences obviously could be a source of problems. But if the question is about which political process is the most legitimate among persons generally willing to defer to legitimate political processes to resolve substantive conflict, then it is not obvious how substantive religious differences would matter, if at all.

Fourth, focusing on empirical aspects of the procedural solution informs a point of theoretical difference between Rawls and Habermas. While Rawls and Habermas agree on the normative prescription to defer to legitimate political processes, the extent to which such processes must be explicitly democratic is a matter of dispute. That there is a difference here is not surprising, given the different theoretical starting points and subsequent development of liberal democratic theory between the two theorists (see, e.g., accounts in McCarthy 1994; Yates 2007). For Habermas, democratic process, particularly in its deliberative form, is the guarantee of other kinds of rights and protections in society, so the legitimate political process to which citizens should defer must necessarily be democratic. For Rawls, the legitimate political process to which citizens should defer must be agreed upon, and compatible with a “constitutional democratic regime” and “legitimate law” (Rawls 1997:766), but what counts as a legitimate political process need not be an explicitly democratic process in the more narrow sense that Habermas would require. An empirical inquiry into which political processes people are willing to recognize as legitimate, and to which political processes they are likely to defer, helps clarify which version of the procedural solution is more workable in practice: the Habermas version that requires explicitly democratic processes, or the Rawls version that accommodates a wider variety of political process options.

Evidence from Social Science

So, is the procedural solution realistic? Will religious persons defer to a political process, even when faced with an outcome that violates their substantive preferences? Ultimately, questions about whether religious citizens will defer to a political process, and whether or not such deference requires a democratic process, are open empirical questions for social scientists to pursue. Surprisingly, social scientists have not obliged, despite the importance of this problem for religion and politics.

What social scientists have shown is that religious differences can map onto political differences. For example, the realignment of American religion into liberal and conservative camps reinforces a liberal and conservative political divide (Wuthnow 1988; Hunter 1991). Concerns over particular political issues such as abortion (Evans 2002; Hoffman and Johnson 2005), euthanasia (Moulton, Hill, and Burdette 2006), same-sex marriage (Olson, Cadge, and Harrison 2006), reproductive genetics (Evans 2010), and war resistance (Nepstad 2008) also tend to be religion-specific.

However, there is no consistent finding about the relationship between religious differences and political differences. Political differences sometimes cross religious traditions and denominations (Wuthnow 1988; Hunter 1991). Religious differences are not equally salient for all political issues (see e.g. Sherkat and Ellison 2007). Religious alignment with particular political positions shifts over time (Manza and Brooks 1997; Evans 2002). Consistent patterns of difference in one location, such as the U.S., do not necessarily hold in other locations around the world (Norris and Inglehart 2004; Bean, Gonzalez, and Kaufman 2008). Religious elites, such as clergy, sometimes differ in political views or commitments from the people in the pews (Olson 2000; Djupe and Gilbert 2003). Individual congregations within otherwise unified religious denominations can differ substantially in political commitment and involvement (Djupe and Gilbert 2009). And even small changes in the measurement of religious differences (e.g. through self-identification, see Lewis and de Bernardo 2010) can have significant effects on apparent alignment with political differences (see also Chaves 2010).

But while these studies have helpfully illuminated many aspects of the relationship between religious differences and political differences, the question of whether religious persons will defer to political process remains unaddressed in the social scientific study of religion. Many studies across the social sciences pursue empirical questions about the relationship between religious characteristics and political attitudes and behavior with regard to specific issues and outcomes (e.g. Evans 2002; Olson, Cadge, and Harrison 2006; Wilcox and Robinson 2011, see also Woodberry and Smith 1998). And a few studies have shown how religious institutions foster certain types of democratic participation (e.g. Lichterman 2005; Djupe and Grant 2006; Neiheisel, Djupe, and Sokhey 2008; Shields 2009). However, little attention has been paid to the relationship between religious characteristics and political process preferences. Certainly popular books have expressed concern about, for example, subversion of democratic processes by American evangelical Protestants (e.g. Phillips 2006; Sharlet 2009) or Muslims (Steyn 2006). But even though there is a growing literature on how people resolve political differences in practice (e.g. Eliasoph 1998; Perrin 2006) and what kinds of political processes people prefer (Hibbing and Theiss-Morse 2002), scholarly attention to religious attitudes toward political processes remains minimal (but see Djupe and Gwiasda 2010).

In this paper I aim to fill this gap by examining how religion matters (or not) to political decision-making. I evaluate the empirical presuppositions of the procedural solution as a set of research questions about the process preferences of actual religious (and non-religious) persons. Given a situation where a person’s own position on a controversial issue conflicts with the results of a legitimate political process, how will they resolve the conflict? Will religious (and non-religious) persons defer to such a process? If so, to what kind of process will they defer? And, most importantly for the enduring question about the role of religion in politics, how do religious differences matter to such political decision-making, if at all?

Research Design

I draw the interview data for this paper from a larger research project about how ordinary Americans from a variety of religious backgrounds (including non-religious) understand and make sense out of public debates over “religion and science” issues. A fuller description of the broader interview context and how the elements of that project relate to this paper can be found in the Methodological Appendix.

Interview Exercise

I designed an interview exercise that would answer the following research question: do respondents from varying religious backgrounds defer to political processes? To get at the decision-making processes and not just the assessment of outcomes, I defined the scope of the qualitative research question in open-ended terms: given the opportunity to resolve a substantive conflict between their own position and the outcome of a given political process, how would respondents reason through a resolution? And what role, if any, would religion play in their reasoning processes?

To this end I constructed a multi-part interview exercise that established an important issue, posited a conflictual outcome, and offered multiple options for resolving the conflict (see full text below). The most important design choice was identifying a live issue that respondents actually had a stake in considering. In prior components of the interview, respondents had identified an issue that was most important to them. Starting with this issue rather than a preselected issue guaranteed that the proposed substantive conflict was not simply hypothetical, but meaningful to respondents and likely to elicit the kind of reasoning that theory suggests would be required of democratic citizens.

The second most important design choice involved identifying legitimate alternatives through which respondents would navigate and engage in political decision-making. For the paradigmatic “democratic process” I used popular democratic vote (sometimes called “majoritarianism” in scholarly literature), which is unlikely to be seen as anything other than a democratic process. But, as with the “issue” at stake, in this exercise what counted as a legitimate alternative was specified by the respondents, not by the interviewer. As part of an earlier component of the interview, respondents selected an ideal committee for making decisions related to each issue. As a structurally distinct alternative to popular democratic vote (limited membership rather than popular inclusion, appointed authority rather than elected authority, etc), the select committee provided a clearly legitimate alternative that would provoke political decision-making.{1}

These choices resulted in the following text of the question(s), taken directly from the interview guide:

For this part of the interview I’d like you to think not just about what is happening, but what should happen, in your opinion. Let’s go back to your committee on [issue], and let’s say that they came up with a position on [issue]. If this position went against your beliefs on [issue], would you want to vote democratically on the proposal, for example in a state referendum? {2}

Should the committee be allowed to override a democratic vote? Why or why not?

[If no committee override] Should anyone be allowed to override a democratic vote? Why or why not?

[If so] Who?

Note that the multiple legitimate options for resolving a substantive difference are presented from both directions, first as a proposition about democratic vote overriding the committee, then as a proposition about the committee overriding the democratic vote (or, ultimately, anyone overriding the democratic vote). Posing the questions this way helped disentangle reasoning about a particular outcome (e.g. reinforcing one’s own position) from reasoning about a particular process (e.g. favoring democratic vote). It also guaranteed that answers would not just reflect a kind of social desirability bias, given that all respondents inhabit an American society that emphasizes its democratic characteristics.

Sample Description

The sample for this research project was highly purposive, and designed to maximize range within purposive categories (see Weiss 1994:22-24). With the limited target size of the sample (approximately 60 respondents) the goal was not to achieve statistical representativeness, as a random sample might not actually capture enough different cases to derive useful theoretical insight. Instead, because of the religion and science content of the debates, I set purposive recruitment targets for religious affiliation and occupation, and informally sought heterogeneity in other categories. For religious affiliation, the target sample distribution was proportional to general US population, approximately 20% mainline Protestant, 33% evangelical Protestant, 25% Catholic, and 20% Other/Non-religious. For occupation, the target sample distribution was approximately 80% non-scientific/technical and 20% scientific/technical. I classified the former initially using denominational membership (from Steensland et al. 2000), adjusting for self-identification where appropriate (e.g. non-denominational church affiliation, or temporary attendance at new local church). I classified the latter initially using occupational categories from the Bureau of Labor Statistics, adjusting as necessary based on personal knowledge of a given respondent’s specific occupation.

Though limited resources constrained site selection, I recruited respondents at two different sites to prevent the idiosyncrasies of one site from skewing results (Weiss 1994). 75% of respondents came from a Southern California city of over 1.5 million residents that is known for high-tech industries and military presence. The remaining 25% came from a South Florida city of fewer than 200,000 residents that is primarily known as a tourism and retirement destination. In addition to geographic and regional differences, demographic differences are a significant source of heterogeneity between the two sites. For example, the distribution of religious affiliations within the “Other/Non-Religious” category differs substantially between the sites. The South Florida site also skews higher in age, which is unsurprising given its status as a retirement destination.

I recruited the initial set of respondents through contacts known to have access to the purposive target groups, then proceeded using a snowball strategy (Weiss 1994). For the Southern California site, I initially worked through personal acquaintances known to have access to local congregations and non-religious organizations. For the South Florida site, I used public information to identify a preliminary list of local congregational leaders and other civic leaders who might act as initial intermediaries. In both cases I asked these intermediaries to identify potential respondents, then contacted potential respondents directly by email or telephone. From there I continued the snowball strategy, enforcing heterogeneity for religion and occupation as necessary through selective recruitment.


Given the sample size, particularly across multiple locations, there is some inevitable variation from the exact target percentages, with most variation occurring as a slightly higher than expected number of respondents ultimately classified in the Other/Non-religious category. (This variation is actually consistent with recent findings about the percentage of non-religious persons in the US general population, see Hout and Fisher 2002). In one case, time constraints prevented delivery of this exercise. That case is omitted from the sample, leaving 61 valid cases. But the resulting sample generally met the purposive targets for religious affiliation and occupational category within each location as well as within the total sample. The sample also ranged usefully in other categories. Respondents ranged in age from 18 to 79 years (average 40). The sample (with invalid respondent omitted) included 34 women and 27 men. All respondents had completed high school, though postsecondary education ranged from none to PhD, with most respondents having taken at least some college courses and a majority having earned AA/AS or BA/BS degrees. However, while the purposive sample shows substantial range in these categories, the sample is not intended to achieve statistical representativeness, so I do not make general claims about differences within these categories.

Data Analysis

All interviews were digitally recorded and professionally transcribed. As I collected data, I took notes about qualitative features of responses in order to inform a later coding scheme (Glaser and Strauss 1967; Luker 2008). Following generally accepted practices of axial and open coding (see Babbie 1998, esp. chapter 13), after all interviews were complete I manually analyzed the relevant portion of each interview transcript. Responses to this exercise usually ran to multiple paragraphs of transcribed text, and were usually interspersed with interviewer followup comments and questions. Drawing on the interview notes as a guide, I identified important concepts in the interview data, adding to the preliminary coding structure and reviewing prior interviews as necessary when new themes emerged (see Glaser and Strauss 1967).

I did not simply identify and tally “pro-democracy” or “pro-committee” themes. In coding I looked particularly at how people reasoned through their assessments and how they decided (or not) which option was best for resolving substantive conflict. As I report below, respondents often engaged both options, expressed limits on a favored option, or even rejected both options entirely. Likewise, broader categories often contained multiple codes, as people could arrive at the same conclusions using different reasoning (e.g. “democracy as fairness” or “incompetent voters”). I also looked particularly for evidence of religious involvement, either in the explicit use of religious language, distinctive religious approaches, or as patterns across respondents with shared religious commitments. Finally, following qualitative methodologists such as Luker (2008), I also looked at whether or not respondents sought to be consistent in their responses, whether they recognized internal inconsistencies in their answers, and more generally how they evaluated the qualities of their own responses as they responded to increased provocation by the interviewer. All of these kinds of qualitative analysis figure in the findings reported below.

Recall that the motivating concern behind objections to religious involvement in politics is that religious commitments to particular substantive outcomes will compel religious persons to try to get their own way, even at the cost of subverting the common polity. If this is the case, then empirically what we should see is that, on an issue of concern where there the substantive outcome of a political process goes against what religious persons believe or advocate, religious persons will base their political decision-making on the pursuit of substantive goals and will not defer to a political process that does not produce their favored outcome. Non-religious persons, by contrast, will not have such religious commitments to particular substantive outcomes, and will therefore defer to a political process to resolve substantive differences.

Of course, if persons do not defer to any political process to resolve substantive differences, then the question of which political process is most legitimate is rendered moot. But if they defer, the question of which political process is seen as most legitimate is actually the central question to be answered, as it addresses the issue of which version of the procedural solution is more workable in practice. If what we see empirically is that those who defer to political processes always submit committee decisions to a popular democratic vote or some other explicitly democratic process, and never allow a popular democratic vote to be overridden (except perhaps by some other process that they identify as democratic), then the Habermas solution is workable. Variations from this model (e.g. allowing the select committee to override a popular democratic vote) signal that the Rawls version of the procedural solution may be more workable in practice than the Habermas version. However, it remains a completely empirical question whether religion has any role in whether respondents differentially prefer the popular democratic vote, the alternative select committee, some other process entirely, or none at all.


The summary finding is that almost all respondents use process preferences, rather than substantive preferences, to guide their political decision-making. With a few exceptions, religious and non-religious respondents alike do not simply want to get their way, even when they feel strongly about an issue. Most respondents would rather defer to a legitimate political process than stick to their commitments to a particular policy outcome.

However, respondents engage process preferences in strikingly different ways. While a minority of respondents sees popular democratic vote as the most legitimate resolution to substantive differences, most respondents either express willingness to override popular democratic vote with a committee decision, or prefer never to engage popular democratic vote at all. Respondents display a complex understanding of democracy and legitimacy. Majoritarian processes that produce universally bad outcomes are illegitimate, as is any political process that allows self-interested actors to derail or hijack the policy process. In contrast, expert-driven political processes that exclude incompetent or subversive participants are sometimes seen as more legitimate than popular democratic vote.

But religious differences do not generally map onto process preferences. In general, religious differences in this study do not usefully predict variations in political decision-making among respondents. Even though religious differences are useful predictors of substantive positions on some issues under consideration in this study (e.g. on stem cell research or human origins), religious persons in all categories are generally willing to defer to a legitimate political process. For the respondents in this study, shared commitments to legitimate political processes are not subverted by their substantive religious commitments. The robust qualitative finding is that respondents in all religious categories tend to engage in political decision-making based on process preferences, rather than substantive preferences, regardless of the sources from which their substantive preferences are derived.

Getting Your Way

Given apparent religious differences on substantive issues (e.g. on stem cell research, see Nisbet 2005), one might expect that such substantive differences would translate into different political process preferences. In the scenario that the procedural solution is intended to remedy, religious persons with strong substantive commitments would be willing to subvert legitimate political processes in order to get their way. Political decision-making would, in such cases, be a matter of resolving substantive conflict by pursuing whichever process achieved their desired outcome, and overriding any process that presented an obstacle to achieving their desired outcome. In this study, such a commitment to “getting your way” would show up as respondents switch between multiple legitimate options as necessary to achieve their goal.

However, despite the religious stratification of the sample, only five out of 61 respondents clearly indicated a commitment to getting their way. This minority of respondents did not defer to a political process when faced with a substantive outcome with which they disagreed, and instead based their political decision-making on removing or avoiding political processes that produced outcomes that violated their substantive preferences. Put another way, the majority of respondents are willing to defer to political processes to resolve political differences, irrespective of religious category. But to illustrate by contrast how unified the other respondents are in their approach to process preferences, it is helpful to examine those exceptional respondents who simply want to get their way.

Sterling, a 52-year old evangelical Protestant, provided a clear example of political decision-making in pursuit of substantive goals. Following the interview schedule, I asked Sterling what he thought should happen if his chosen committee on human origins came up with a policy that went against his substantive preference. He immediately replied, “I would want to do anything I could to reverse the decision. So fire the committee, have people vote on it if I think that would work.” Despite recognizing the legitimacy of the committee (after all, he had appointed them), Sterling had no qualms about submitting the issue to popular democratic vote or even eliminating the committee if they went against his preferred outcome. But when asked about the reverse situation, allowing the committee to override the result of a popular democratic vote that went against his preferred outcome, Sterling again immediately replied, “Yeah, I would. I’m an advocate of democracy, but, you know, this is an issue where this particular issue has major consequences on human behavior, human attitudes. It’s important.”

Sterling’s response articulates a substantive commitment to something that is so “important” that it overrides any willingness to defer to a legitimate political process. This would appear to validate the common concern among theorists of religion and democracy about the strength of religious commitments. However, the cases where people want to get their way actually illustrate why this is a flawed view of the relationship between substantive preferences and process preferences. The important question is not whether there are religious differences in substantive preferences (there are), but whether those substantive preferences translate into process preferences.

Arthur, a 30-year old mainline Protestant, takes the opposite substantive position on human origins. But his process preferences do not reflect this difference. Asked about whether popular democratic vote should override the committee’s decision, Arthur replied “If I had the option, I would attempt to salvage [my preferred outcome] as best possible, even though I’m afraid I would have little confidence in the democratic vote. But [yes], if it was the only possible way [to get my way].” Likewise, when asked the question in reverse, Arthur replied “Yes. Because while I do not have megalomania, I believe that the benevolent dictator model is ultimately the best provided you have someone who is truly benevolent and truly all powerful.” For Arthur, as for Sterling, the best process is whatever process helps him get his way.

Similarly, both Holly, a 33-year old Catholic, and Anita, a 57-year old evangelical Protestant, show an unwillingness to defer to the outcome of political processes regarding the issue of sexuality. When asked about overriding the committee, Holly said that she is “open minded to a democratic vote.” But when offered the chance to override the vote, she quite cheerfully replied “Oh yeah, definitely. Of course. I always want my position to win.” Likewise, despite saying moments before that a committee decision about sexuality should be put to a popular vote, Anita expressed a desire to override that vote if it went against her preferred outcome: “Personally, I, – yeah. I would like [the override] to be done. ’Cause deep down in my heart I still believe what I think is correct.” In substantive terms, Holly prefers maximally inclusive policy that does not in any way discriminate based on sexuality. Anita sees homosexuality as immoral and would prefer that it be restricted by policy. But they arrive at the same process preference. They both simply want to get their way.

Felix, a 77-year old evangelical Protestant (and self-identified fundamentalist) epitomizes this kind of substantive commitment. For Felix, we “absolutely” should allow any overrides of decisions that go against his beliefs on human origins: “’Cause if we let it stand then we’ll have – in my opinion, it would have disastrous effects like it is now. It would have a disastrous effect. And so I would do whatever was in my power to change that.” The fact that Felix self-identifies as “fundamentalist” might, by itself, be cause for alarm. But I note that, while one evangelical Protestant who self-identifies as “fundamentalist” shows up here, the other (Norma, a 59-year old evangelical Protestant) does not. And in the context of a 61 person sample, a minority of respondents in each religious category are willing to override processes in order to get their way. For example, even though three respondents who want to get their way are evangelical Protestants, there are 16 other evangelical Protestants in the sample who do not.

So, rather than indicating some basic threat of theocracy, these few exceptions only reinforce the general willingness of respondents to defer to political processes to settle substantive differences. The vast majority of respondents do not simply want to get their way, whatever their religious category. Rather, respondents tend to sound like Erika, a religiously-unaffiliated 51 year old, who says that she “doesn’t have all the answers” and “doesn’t necessarily know” what policy is best, and “even if I disagreed” would “want the policy to be put in place long enough to see if it was effective or not.” Most respondents react to substantive conflict by deferring to the outcomes of legitimate political processes, even when that violates their substantive preferences, even on issues that matter to them, whatever their substantive position on a given issue. In the remainder of this paper I discuss how this occurs, and what it means for democratic theory.

Democracy and Legitimacy

Given a substantive conflict, most respondents are willing to defer to a political process to resolve that conflict. But focusing on the common willingness to defer conceals important diversity in how respondents engage in political decision-making. Throughout the interviews I found that respondents ranged widely in their understandings of democratic legitimacy. Even though respondents consistently discussed various versions of democratic processes as they decided on a legitimate resolution to substantive conflict, they did not always, or even often, see any particular version of democracy as the most legitimate resolution.

To put this into context in the sample, only about a third of all respondents express a consistent preference for resolving substantive political differences through popular democratic vote. Of those, about half can imagine conditions under which they would not defer to democratic processes such as a popular vote. Even when respondents sometimes appear to be explicitly endorsing democratic processes, they generally describe such support in terms that indicate whether or not democracy is a legitimate solution to substantive political differences.

Some respondents are total supporters of majoritarian democracy. These respondents consistently indicate that substantive differences should always be put to popular democratic vote, and that the outcome of such a process cannot be overridden by anything except another democratic process (e.g. an election). Timothy, a 25 year old evangelical Protestant, typifies this category of responses:

I am generally in support of our system of government here in the United States, and [the committee decision] you’ve advocated there is rather different from our system of government. So in that situation I wouldn’t be in favor of this kind of absolute power given to an oligarchy; however, in our system of government I really don’t think a referendum should be able to overrule what the elected officials do.

Looking only at these total supporters, it is not immediately obvious that majoritarianism might not be considered the most legitimate process. But even total supporters liked majoritarianism for different reasons, suggesting that support for democratic process hinged on its ability to meet their (different) criteria for legitimacy. Miley, a non-religious 19 year old, expressed approval of the democratic vote’s ability to implement the will of the majority, suggesting that “most people have the right idea about things” and that “the majority of people are good and know what’s right.” When asked why a democratic vote should stand “no matter what,” Oscar, a 50 year old Catholic, replied “Well, you got to have some kind of fairness.” And for Solomon, a 43 year old Unitarian, popular democratic vote has virtues that outweigh its drawbacks:

I wouldn’t necessarily enjoy it, I wouldn’t necessarily like it, but yes, I would put it to public vote. Now I know that by doing that I’m respecting the opinions of the majority and the people who I’m dealing with …I would say the public made the decision based on information that they were given, therefore they should be given what they want. Whether it was a stupid choice or not, we’re going to have to live with the consequences of our not knowing.

While these respondents agreed that a popular democratic vote should be used to resolve political differences, they did not agree exactly about why this should be so. Whether it is the will of the majority, “fairness,” “respecting the opinions of the majority,” avoiding “oligarchy,” or some other reason, respondents expressed a willingness to defer to the outcome of a democratic vote because they saw that process as legitimate. As Solomon’s answer shows, it is more important for the process to be legitimate, even when a public draws on bad information to make a “stupid choice.” Like other total supporters, Solomon defers to majoritarian process because he thinks that democratic voting, for all its flaws, is the most legitimate political process for resolving substantive differences. Even among total supporters, then, popular democratic vote is preferred because it aligns with their understandings of legitimacy.

Note, however, that the link between majoritarianism and legitimacy does not appear to be governed by religious commitments for total supporters. No distinctively religious language appears. Respondents engage in political decision-making using general political language, terms, and justifications, and while they often express strong feelings about their support for democracy, they do not express this support using distinctively religious talk.

The Limits of Democratic Legitimacy

The obvious implication of the alignment of democracy and legitimacy is that if democracy fails in some way to be, e.g., fair, respectful of majority opinion, or a bulwark against oligarchy, then support for democracy, even among total supporters, might well be provisional. This potential disconnection between democracy and legitimacy is more apparent in the political decision-making of respondents who expressed a general preference for democracy, but nonetheless indicated ways that majoritarianism might not be legitimate. Like Vicki, a 48 year old mainline Protestant who said “I believe very firmly in the democratic process, but I also believe that sometimes there has to be an intervention,” these respondents imagined and articulated the limits of democratic legitimacy by outlining potential scenarios where majoritarianism might produce an illegitimate outcome. Take, for example, the response of Yuri, a 30 year old evangelical Protestant, to the question about whether anything could override a democratic vote:

I mean 99.9 percent is no, but if it’s some ridiculous thing, like lets – I mean, you know, if we were to vote there is going to be no more tax ever, then I do think that needs to be something that, you know, they could override because that just doesn’t make any sense. And it’s not good for us. So it depends on what that issue is. But most of the times I would lean towards “no” because the people have spoken.

For Yuri, as for many other respondents, even a democratic process such as popular democratic vote may be illegitimate if it is used to do illegitimate things. While Yuri expressed her scenario in terms of economic disaster, others indicated different kinds of scenarios to define democracy’s legitimate limits. Nicole, a 50 year old Catholic, worried that letting a committee override a democratic vote might be “communistic” and “like dictatorship.” Nevertheless, she still indicated that an override would be legitimate if “there’s cause to see that it is not the right thing, like it would end up much worse. Like, a scientist says if you blow this bridge up then, we’re going to get a flood and end up killing all these people, like that.” Similarly, Zoe, an 18 year old atheist, suggested that “if suddenly everyone started believing that we should just kill or the birds or something like that, then obviously [the committee] should be able to say that something’s wrong with America.” And in one of the few uses of explicitly religious examples, Harvey, a 79 year old mainline Protestant, pointed out how democracy could go wrong: “Majority ruling is not always the best thing. After all, a majority of people crucified Jesus Christ, didn’t they? ”

Even respondents who generally supported the option of a popular democratic vote (including those who supported it with exceptions), indicated that they saw no obligation to support an illegitimate democratic process, and would under those circumstances defer to other, more legitimate, processes. While what counted as illegitimate democracy varied, one of the most common concerns was that a democratic process could be subverted by bad actors, as Amanda, a 37 year old mainline Protestant, indicated in her answer:

My first answer and what I think is the most ethical is of course no. We should be governing ourselves and the government works for us, and that’s also the judicial branch. But what happens is, that powerful groups I think maybe manipulate the vote and that sounds kind of conspiracy theorist, but I think it’s true in a lot of ways and so I think that we do have to have those checks and balances. But there probably are times when the popular vote could be over ruled by say for example, our [U.S.] Supreme Court or the state supreme court.

But even in these cases, respondents did not indicate that democratic processes should be entirely removed or changed. Rather, other legitimate political institutions would ideally handle the few exceptions where democracy turned illegitimate. For example, Daniela, a religiously-unaffiliated 33 year old, indicated that if an exception happened, she would be “OK with that kind of thing being overturned by the court system, but not so much by a political action group or by an appointed committee.” However, even while recognizing that deferring to courts might be necessary, respondents still sometimes expressed discomfort moving beyond majoritarianism. Meg, a 63 year old mainline Protestant, questioned how courts could override a popular vote: “It’s like what’s the point of having the people involved if the government that’s supposed to be for the people and by the people doesn’t listen to the people? ” But such concern only reinforces the idea that, for respondents who support majoritarianism, popular democratic vote and legitimacy are nearly synonymous.

Other than Harvey’s striking use of Jesus Christ as an example of democracy gone wrong, religious language did not figure in the discussion of the limits of democracy. Most political decision-making involved general political language, with most references involving American political traditions, including America’s perceived enemies and opposites (e.g. communism). And even for Harvey, the religious example served to show the limits of political processes, rather than providing grounds for committing to a particular substantive position.

Legitimacy Without Democracy

Even among supporters of democracy, majoritarianism is not always the most legitimate resolution to substantive conflict. The possibility for non-majoritarian legitimacy is clearest among those who see a popular democratic vote as fundamentally illegitimate. About one in four respondents actually expressed firm opposition to majoritarian resolution of substantive conflict. These respondents rejected the idea of a popular democratic vote, and fully endorsed a committee override of any majority decision.

Many of those opposed to majoritarianism placed their faith in the expertise, whether technical, moral, or procedural, of the committee that they had selected to make policy on a controversial issue. Scott, a 33 year old Catholic, typifies this category of responses:

I respect [the committee] morally and think they have the expertise necessary to understand the facts. Part of the problem is I think there’s definitely in our government there’s the concept of majority rule but minority rights. Anytime you have a democratic decision I think there’s a suggestion that that means the majority makes the decision and that stands and that’s just the way it is. I think [the committee] is an important balance to that …a counterbalance to the fact that if you have large groups of people you get different results to certain questions than if you take two people talking together or an individual making decisions on their own.

Left undescribed by Scott is what these “different results” might be, or why they might be a problem. The answer to this question emerged in other responses. For many other respondents, the problem with democratic majoritarianism is that it invites participation from people who are incapable of making rational, informed decisions to resolve substantive differences. The rule of experts is therefore more legitimate, not because expertise in itself is legitimate, but because such expertise makes experts less susceptible to the influence of emotion or rhetoric. So, rather than only expressing personal faith in expert knowledge as the source of legitimacy, they contrasted the expertise of the committee with the ignorance of the general population. Max’s response captures this sentiment neatly:

I don’t believe people are informed. I don’t believe the average person is informed at all. So I would say, no, I don’t care what the popular thing is. Oh, I do care, I’ll be politically sensitive, but this is still what it is. Just like you wouldn’t want popular opinion at a hospital. You want what the doctor says, not what everybody else feels. Not what all the patients feel. –Max, 40, evangelical Protestant

Similar responses indicated that the problem with majoritarianism is not so much a matter of expertise, but rather a basic concern that people in general seek to pursue their own interests, and therefore cannot be trusted to make good decisions that are in the general interest. For example, Leo, 24, evangelical Protestant, said that “most people vote their self-interest and I don’t trust most people’s self-interests so probably not, no. Despite my political views on such things, I don’t think that that should be up to popular choice.” Having heard from Leo that his politics were broadly libertarian, I fully expected that, even though he would not necessarily demand a popular democratic vote, he would certainly not allow a committee to override a popular democratic vote. But instead his response was consistent with his distrust of democracy and its participants: “I’d say yes, unfortunately, even though that goes against my libertarian politics.” For Leo, as for others with similar responses, a popular democratic vote may be illegitimate, because allowing everyone to pursue their self-interests necessarily subverts the legitimacy of the process.

Whether because of a faith in expertise, a concern over the lack of an informed citizenry, or distrust of self-interested actors, about one quarter of the respondents in this study actively resisted a majoritarian resolution to substantive differences, preferring instead to leave important decisions solely to their chosen committees. While this apparently authoritarian position might be expected to align with more authoritarian forms of religion or religious belief (see, e.g., Canetti-Nisim and Beit-Hallahmi 2007), in this study there is no evidence of religious differences at all among those who reject democratic majoritarianism as a legitimate resolution to substantive differences.

Legitimacy as Precedent

Finally, about one in four respondents offered responses that consistently expressed refusal to override one process with another. At first this appeared to be a source of internal contradiction, as respondents in this category did not consistently prefer either option to resolve substantive differences. But unlike those respondents who simply want to get their way using whatever process will help them best achieve their goals (see above), respondents in this category consistently indicated that they would defer to whatever process that they imagined or assumed could already be in place. For these respondents, precedent determined legitimacy, irrespective of whether or not the process in place was seen as democratic.

Respondents in this category varied in their expressed reasons for supporting existing processes. Some respondents, such as Dwight, indicated that they thought that one process was better than another in the abstract, but acknowledge that in practice whatever process we have agreed to must be legitimate, precisely because we have already agreed to it. Even though he might prefer one process rather than another, Dwight ultimately defers to whatever process is currently set up, because it is something that we have already agreed to do:

[E]ven if they came up with something that supported me, it needs to go through proper political channels anyway. I’m just hoping [for a favorable outcome], really, but in the context of what you said, well no, the rules were that whatever they decided that was the thing. And yeah, I’ll keep working to try and change it if it comes up against it, what I believe. But to be truly right in my beliefs, whatever they come up with should go through the political process after they make their case regardless, but not a popular referendum vote, because I do not find that a good mechanism for change. It’s too much a mechanism against change. [But it] should be respected to the extent the political system is set up to do that. –Dwight, 49, Unitarian Universalist

For Dwight and other similar respondents, precedent is legitimate because it indicates a common commitment to the existing political process. For others, such as Arienne, respecting whatever process is in place is not just an acknowledgement of existing legitimacy, but an opportunity to build legitimacy in whatever process is in place. In the short term, this might run against one’s own substantive preferences. But in the future, substantive preferences could still be realized through a process that others would then respect. The specific form of the process is less important than building legitimacy for whatever process is in place. At first, Arienne indicated that she would accept whatever policy her committee generated. But then, when asked if the committee could override a popular vote, she replied:

A: I guess not, even though it would be really hard. I mean, popular vote is popular vote. And that’s – I guess that’s – I keep going back to Prop 8, but it was hard – it would just be hard to swallow, a hard pill to swallow that people still – yeah.

Q: Should anyone be able to override a democratic vote? Not just your committee, but anyone?

A: No. I don’t think so, as frustrating as it can be sometimes to hear what the popular vote is when it’s not your vote, when it doesn’t match your vote. I think that the changes could come. –Arienne, 32, Catholic

For a few respondents in this category, the legitimacy of precedent is not grounded in common consensus or even the guarantee of a framework for future change. Rather, precedent has legitimacy because it circumscribes political possibilities. Respecting precedent is a way to avoid the pitfalls of self-interested actors who might otherwise seek to change the process in order to suit their own purposes. A typical example comes from Susanne, 51, atheist, who says that the process shouldn’t change “once [the issue]’s in the political system.” For these respondents, it is better to finish how you started, if only to prevent those who only want to get their way from changing the rules.

Finally, in one of the very few explicit uses of religious language in this study, one respondent was perfectly fine with whatever process was already in place, because she saw the existence of such a process as evidence that God had granted that process legitimacy. When asked whether a committee could override a popular democratic vote, Grace, 37, an evangelical Protestant, indicated that it would be fine either way, saying “I think they [the voters] should pray about it and let God decide” but that the outcome could be overridden if it was “blatantly wrong.” This seemed at first like Grace was indicating a willingness to defer to her committee. But when pressed on who would decide whether is something is blatantly wrong, and how the overriding would take place, she replied “I think only God can decide, and He’ll use people in His own timing to do that.” For Grace, the legitimacy of any process comes from God, so the specific form of the political process is simply not all that important.

But Grace is the exception. As with every other category of responses, respondents generally did not use explicitly religious language in their political decision-making. Instead, as with the other categories of responses, respondents tended to draw on more general political language, often with reference to principles of fairness, justice, inclusion, and respect. As with the other categories, the most interesting finding is that political decision-making does not seem to involve much religion at all.


The empirical findings in this study challenge the underlying assumptions of the procedural solution to the problem of religious difference (e.g. Rawls 1997; Habermas 2006). Recall that the basic concern of many democratic theorists about religion in politics is that religious persons will be so committed to achieving their substantive outcomes that they will disregard political processes in favor of getting their way. The basic finding of this study is that this is simply not the case. With few exceptions, religious and non-religious persons alike are perfectly willing to defer to the outcome of legitimate political processes, even if such outcomes violate their substantive preferences concerning issues that matter most to them. What binds Americans together in common support for legitimate government seems to be a stronger cultural force than religious differences that might separate them.

But the general willingness to defer to a legitimate political process does not mean that all versions of the procedural solution are equally workable in practice. Respondents hold different views of legitimacy that draw on a variety of cultural resources about the nation, government, fairness, and justice. There are many ways that respondents see a process as legitimate, and many different reasons why they are willing to defer. Certainly deference to a legitimate political process is the prevailing mode, whatever the reasoning behind such deference. But respondents evince a willingness to defer to a variety of political processes, whether it is respondents who always prefer a popular democratic vote, respondents who oppose majoritarianism, respondents who identify circumstances under which they would seek alternatives to majoritarian resolution, or respondents who simply respect whatever political process is already in place. On this point, the findings from this exercise suggest that the Rawls version of the procedural solution, with its accommodation of a wider variety of legitimate political process options, is more workable in practice than the more restrictive democratic version offered by Habermas.

But might these differences yet be linked to religion? Are differences in views about political process linked to religious differences? Here again, the answer appears to be no. The robust qualitative finding is that few respondents invoke religious references at all in their political decision-making, much less use religion systematically to justify preferences about process legitimacy. And while broad quantitative inference is ill-advised, given the sample configuration, there is basically no alignment between religious differences and process preferences (see Table 2).


While this research appears to subvert the underlying empirical assumptions of the procedural solution, it also suggests avenues for further investigation. The findings reported here are consistent with a growing literature in political science and sociology showing that processes generally packaged with democracy in theory are, in practice, often disconnected or even independent (e.g. deliberation, see Shields 2009; Klemp 2010; He and Warren 2011). They are also consistent with findings in political science showing that Americans are less concerned with the specific form of governance than they are concerned with the personal burden of participating in governance, preferring instead a sort of “stealth democracy” that operates without demands for individual intervention (Hibbing and Theiss-Morse 2002). As this study cannot definitively show that religion is completely irrelevant to these consistent political preferences, future research should investigate whether religion might show up as relevant in a larger, preferably nationally-representative, sample. Religion might not be the answer to the question in this study, but it might well be the answer to other questions about how various “citizenship practices” (Perrin 2006) are related, or not, to theoretical propositions about democracy.

Finally, the consistent qualitative finding that religion does not play much of a part in the political decision-making in this study suggests some worthwhile avenues of investigation in the sociology of religion. To the extent that religion is a shared source of cultural resources (a kind of “tool kit,” see Swidler 1986), it is also subject to the same individual limitations as other cultural resources, meaning that sometimes religion simply does not figure in social action, even among persons identifiable as religious in a survey setting (Vaisey 2009; Martin 2010). This observation is consistent with existing literature on political variations in congregations (e.g. Djupe and Gilbert 2009) and with recent discussions on what religious “congruence” actually entails (Chaves 2010). Future research in the sociology of religion would do well to consider that assumptions about religious groups may simply reflect how such groups are constituted in public on “a few issues constructed as important by activists with resources” (Evans and Evans 2008) rather than general religious differences that necessarily ramify across processes and issues alike (see also Edgell 2012). At a minimum, further study of religious attitudes about political processes, and how such process preferences might vary from substantive preferences, offers exciting potential for expanding our knowledge of religion and politics.


Thanks to John Evans for helpful comments on an earlier version of this paper, and to the JSSR editor and three anonymous reviewers for their insightful suggestions and direction. Data collection supported in part by the Religious Research Association and the Society for the Scientific Study of Religion.


Alexander, Jeffrey C. 2006. The civil sphere. New York: Oxford University Press.

Asad, Talal. 1993. Genealogies of religion: Discipline and reasons of power in Christianity and Islam. Baltimore, MD: Johns Hopkins University Press.

Audi, Robert. 2000. Religious commitment and secular reason. New York: Cambridge University Press.

Audi, Robert and Nicholas Wolterstorff. 1997. Religion in the public square: The place of religious convictions in political debate. Lanham MD: Rowman and Littlefield.

Babbie, Earl R. 1998. The practice of social research. Belmont, CA: Wadsworth Publishing Company, 8th edition.

Bean, Lydia, Marco Gonzalez, and Jason Kaufman. 2008. Why doesn’t Canada have an American-style Christian right? A comparative framework for analyzing the political effects of evangelical subcultural identity. Canadian Journal of Sociology 33(4):899–943.

Calhoun, Craig, Eduardo Mendieta, and Jonathan VanAntwerpen (eds.). Forthcoming. Habermas and religion. Cambridge: Polity Press.

Canetti-Nisim, Daphna and Benjamin Beit-Hallahmi. 2007. The effects of authoritarianism, religiosity, and “New Age” beliefs on support for democracy: Unraveling the strands. Review of Religious Research 48(4):369–384.

Chaves, Mark. 2010. Rain dances in the dry season: Overcoming the religious congruence fallacy. Journal for the Scientific Study of Religion 49(1):1–14.

Connolly, William E. 2005. Pluralism. Durham, NC: Duke University Press.

Djupe, Paul A. and Christopher P. Gilbert. 2003. The prophetic pulpit: Clergy, churches, and communities in American politics. Lanham, MD: Rowman and Littlefield.

Djupe, Paul A. and Christopher P. Gilbert. 2009. The political influence of churches. New York: Cambridge University Press.

Djupe, Paul A. and J. Tobin Grant. 2006. Religious institutions and political participation in America. Journal for the Scientific Study of Religion 40(2):303–314.

Djupe, Paul A. and Gregory W. Gwiasda. 2010. Evangelizing the environment: Decision process effects in political persuasion. Journal for the Scientific Study of Religion 49(1):73–86.

Eberle, Christopher J. 2002. Religious conviction in liberal politics. Cambridge: Cambridge University Press.

Edgell, Penny. 2012. A cultural sociology of religion: New directions. Annual Review of Sociology 38:247–265.

Eliasoph, Nina. 1998. Avoiding politics: How Americans produce apathy in everyday life. New York: Cambridge University Press.

Evans, John H. 2002. Polarization in abortion attitudes in U.S. religious traditions, 1972–1998. Sociological Forum 17(3):397–422.

Evans, John H. 2010. Contested reproduction: Genetic technologies, religion, and public debate. Chicago: University of Chicago Press.

Evans, John H. and Michael S. Evans. 2008. Religion and science: Beyond the epistemological conflict narrative. Annual Review of Sociology 34:87–105.

Glaser, Barney G. and Anselm L. Strauss. 1967. The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine Publishing Company.

Habermas, Jürgen. 2006. Religion in the public sphere. European Journal of Philosophy 14(1):1–25.

He, Baogang and Mark E. Warren. 2011. Authoritarian deliberation: The deliberative turn in Chinese political development. Perspectives on Politics 9(2):269–289.

Heclo, Hugh. 2007. Christianity and American democracy. Cambridge, MA: Harvard University Press.

Hibbing, John R. and Elizabeth Theiss-Morse. 2002. Stealth democracy: Americans’ beliefs about how government should work. Cambridge: Cambridge University Press.

Hoffmann, John P. and Sherrie Mills Johnson. 2005. Attitudes toward abortion among religious traditions in the United States: Change or continuity? Sociology of Religion 66(2):161–182.

Honig, Bonnie. 1993. Political theory and the displacement of politics. Ithaca, NY: Cornell University Press.

Hout, Michael and Claude S. Fischer. 2002. Why more Americans have no religious preference: Politics and generations. American Sociological Review 67(2):165–190.

Hunter, James Davison. 1991. Culture wars: The struggle to define America. New York: Basic Books.

Klemp, Nathaniel J. 2010. The Christian right: Engaged citizens or theocratic crusaders? Politics and Religion 3(1):1–27.

Kymlicka, Will. 1995. Multicultural citizenship: A liberal theory of minority rights. Oxford: Oxford University Press.

Leiter, Brian. 2012. Why tolerate religion? Princeton, NJ: Princeton University Press.

Lewis, Andrew R. and Dana Huyser de Bernardo. 2010. Belonging without belonging: Utilizing evangelical self-identification to analyze political attitudes and preferences. Journal for the Scientific Study of Religion 49(1):112–126.

Lichterman, Paul. 2005. Elusive togetherness: Church groups trying to bridge America’s divisions. Princeton, NJ: Princeton University Press.

Luker, Kristin. 2008. Salsa dancing into the social sciences: Research in an age of info-glut. Cambridge, MA: Harvard University Press.

Manza, Jeff and Clem Brooks. 1997. The religious factor in U.S. presidential elections, 1960-1992. American Journal of Sociology 103(1):38–81.

Martin, John Levi. 2010. Life’s a beach but you’re an ant, and other unwelcome news for the sociology of culture. Poetics 38(2):229–244.

McCarthy, Thomas. 1994. Kantian constructivism and reconstructivism: Rawls and Habermas in dialogue. Ethics 105(1):44–63.

Mendieta, Eduardo and Jonathan VanAntwerpen (eds.). 2011. The power of religion in the public sphere. New York: Columbia University Press.

Moulton, Benjamin E., Terrence D. Hill, and Amy Burdette. 2006. Religion and trends in euthanasia attitudes among U.S. adults, 1977–2004. Sociological Forum 21(2):249–272.

Neiheisel, Jacob R., Paul A. Djupe, and Anand E. Sokhey. 2009. Veni, vidi, disseri: Churches and the promise of democratic deliberation. American Politics Research 37(4):614–643.

Nepstad, Sharon Erickson. 2008. Religion and war resistance in the Plowshares movement. New York: Cambridge University Press.

Nisbet, Matthew C. 2005. The competition for worldviews: Values, information, and public support for stem cell research. International Journal of Public Opinion Research 17(1):90–112.

Norris, Pippa and Ronald Inglehart. 2004. Sacred and secular: Religion and politics worldwide. Cambridge: Cambridge University Press.

Numbers, Ronald L. (ed.). 2009. Galileo goes to jail and other myths about science and religion. Cambridge, MA: Harvard University Press.

Olson, Laura R. 2000. Filled with spirit and power. Albany, NY: SUNY Press.

Olson, Laura R., Wendy Cadge, and James T. Harrison. 2006. Religion and public opinion about same-sex marriage. Social Science Quarterly 87(2):340–360.

Perrin, Andrew J. 2006. Citizen speak: The democratic imagination in American life. Chicago: University of Chicago Press.

Phillips, Kevin. 2006. American theocracy. New York: Viking Penguin.

Ratzinger, Joseph Cardinal and Jürgen Habermas. 2006. The dialectics of secularization: On reason and religion. San Francisco: Ignatius Press.

Rawls, John. 1997. The idea of public reason revisited. The University of Chicago Law Review 64(3):765–807.

Sager, Rebecca. 2010. Faith, politics, and power: The politics of faith-based initiatives. New York: Oxford University Press.

Sharlet, Jeff. 2009. The family: The secret fundamentalism at the heart of American power. New York: Harper Perennial.

Sherkat, Darren E. and Christopher G. Ellison. 2007. Structuring the religion-environment connection: Influences on environmental concern and activism. Journal for the Scientific Study of Religion 46(1):71–85.

Sherkat, Darren E., Kylan Mattias De Vries, and Stacia Creek. 2010. Race, religion, and opposition to same-sex marriage. Social Science Quarterly 91(1):80–98.

Shields, Jon A. 2009. The democratic virtues of the Christian right. Princeton, NJ: Princeton University Press.

Steensland, Brian, Jerry Z. Park, Mark D. Regnerus, Lynn D. Robinson, W. Bradford Wilcox, and Robert D. Woodberry. 2000. The measure of American religion: Toward improving the state of the art. Social Forces 79(1):291–318.

Steyn, Mark. 2006. America alone: The end of the world as we know it. Washington, DC: Regnery Publishing.

Stout, Jeffrey. 2004. Democracy and tradition. Princeton, NJ: Princeton University Press.

Swidler, Ann. 1986. Culture in action: Symbols and strategies. American Sociological Review 51(2):273–286.

Taylor, Charles. 2007. A secular age. Cambridge, MA: The Belknap Press of Harvard University Press.

Vaisey, Steve. 2009. Motivation and justification: A dual-process model of culture in action. American Journal of Sociology 114(6):1675–1715.

Weiss, Robert S. 1994. Learning from strangers: The art and method of qualitative interview studies. New York: Free Press.

Wilcox, Clyde and Carin Robinson. 2011. Onward Christian soldiers? The religious right in American politics. Philadelphia: Westview Press, 4th edition.

Woodberry, Robert D. and Christian Smith. 1998. Fundamentalism et al: Conservative Protestants in America. Annual Review of Sociology 24:25–56.

Wuthnow, Robert. 1988. The restructuring of American religion. Princeton: Princeton University Press.

Yalçinkaya, M. Alper. 2011. Science as an ally of religion: A Muslim appropriation of “the conflict thesis.” British Journal for the History of Science 44(2):161–181.

Yates, Melissa. 2007. Rawls and Habermas on religion in the public sphere. Philosophy & Social Criticism 33(7):880–891.


{1} Note that respondents could see the committee as an alternative for a variety of reasons. And even in the unlikely case that the committee were seen by respondents to be the same thing as a popular democratic vote, the exercise would provoke an explanation for why the committee was not an alternative. No respondents took this position.

{2} While this example might not be universal, both interview sites (see below) are located in states that have such a referendum process in place, so a state referendum is a sensible example for respondents.


Table 1: Purposive Sample Breakdown (n=61)

Location Mainline Evangelical Catholic Other
Southern California
Scientific Occupation 2 2 4 2
Non-scientific Occupation 6 11 6 13
Scientific Occupation 1 2
Non-scientific Occupation 3 4 3 2

Table 2: Religious Distribution of Process Preferences (n=61)









Existing Process None


Catholic 2 2 1 3 5 13
Ev. Prot. 4 1 3 5 4 2 19
Main. Prot. 5 1 3 3 12
Other/None 5 2 4 4 2 17
Total 11 10 5 15 16 4 61

© Michael S. Evans 2008-2017