Social Class and the Stubbornness of Family Myths: How Nonbeliever and Pagan Parents Cope with Intrusions on Parenting by Proselytizing Christian Family Members in the U.S. Bible Belt

My mother takes my kids to church with her. I don’t approve but she does it anyway. She waits until she’s babysitting. They sleep over, and then she tells them, “If you’re sleeping at my house you have to go to church.” I try not to let her have them overnight on Saturdays but it’s hard to make it work on weekdays. So, she keeps taking them to church. (Nadia, eclectic Pagan,2 mother to two kids, ages 4 and 9)

one of the most ubiquitous experiences they faced as parents: unwelcomed proselytizing by some Christian family members. Though nonbeliever and Pagan parents differ ideologically, they both face the shared stigma of parenting on the cultural fringe in the Bible Belt, where evangelical Christianity dictates the religious socialization of children within Christianity. The parents I studied faced a set of tough questions in dealing with proselytizing by family members: What level of harm, if any, might proselytizing pose to children? How should one respond to proselytizers when the proselytizers include loved ones? And, ultimately, what happens when parents try to control proselytizing by family members and fail?
Pursuit of these questions is important to expanding work on religious inequality in general and in expanding research on boundary work and emotion work within microsociology specifically. Furthermore, looking at parents yields a unique focus from the study of the childless. For the religiously marginalized, becoming a parent often catalyzes a need to reflect on identity and behaviors not just for oneself but for the sake of one's children (Manning 2015;McClure 2017).
Simultaneous studies of two marginalized groups of parents allowed for comparison of coping strategies for both the highly-resourced nonbelievers and resourcechallenged Pagans, yielding insights into how social class shapes responses to intrusions on parenting. In this article, I argue that social class impacts one's reliance on a particular emotion management strategy-the family myth-in what initially seems to be a counterintuitive pattern; with more privileged nonbeliever parents clinging more tightly to family myths than their less resourced Pagan counterparts. I situate my work-including this finding-in the tradition of Hochschild and Machung's (1989) classic work on emotion management.

Nonbelievers in the United States
The rise in nonbelievers in the United States in the last decade is substantial. The Pew Forum on Religion and Public Life Survey (2015) found that in 2014, almost onequarter (22.8%) of the U.S. population reported affiliation with no religion (referred to as religious "nones"). While most of the "nones" report belief in deity (Pew Forum on Religion and Public Life Survey 2015), the percentage of Americans who self-identify as agnostic or atheist has simultaneously risen between 2007-2014 from 2.4% to 4.0% and 1.6% to 3.1% respectively. For purposes of this article, I focus on the broader group of nonbelievers, those who espouse general disbelief in deity and hold secular worldviews; a subsection of the diverse "nones." To date, there is little research on nonbelievers as parents. There are a few important exceptions. Using secondary data, Ecklund and Lee (2011) examined how nonreligious scientists at elite universities negotiated their secularism with the expected norm of raising children within conventional religion. The authors found that even atheist and agnostic scientist parents believed religious institutions would provide for their children a necessary foundation for socialization. This study is particularly note-worthy. If any group of parents could feel comfortable raising their children outside of religion, these parents, whose colleagues overwhelmingly also identified as nonbelievers, would have. Yet even these parents chose a conventional path in childrearing, demonstrating the strength and appeal of socializing children in mainstream religion.
Christel Manning (2013Manning ( , 2015 analyzed how unaffiliated (religious "none") parents socialize their children. Manning found that (religious) "none" parents claim parenthood complicates both identity and parenting behaviors. In an empirical study impressive in scope, Manning demonstrates how the Christian hegemony in areas with high numbers of evangelical Protestants complicates and intensifies the parenting choices of "none" parents. Manning argues that becoming a parent places the unaffiliated in a hard spot, forcing them to make tough choices that the conventionally religiously do not have to make. The result is that unaffiliated parents follow one of three trajectories: head back to church, negotiate a newer type of agnostic spirituality, or double-down on nonbelief.
Finally, Joel Thiessen (2016) analyzed how parents who are marginally affiliated with religion as well as those who are nonreligious socialize their children. Thiessen found that nonreligious parents tended to defer to their children in offering them some choice around religion. They neither raised their children in religion nor did they expressly plan to raise them as nonreligious. Thiessen also documents parents' approval of the possibility of their children choosing to become religious in the future, with some caveats.

Pagans in the United States
Contemporary (Neo-)Paganism emerged in the United States in the 1960's (Berger, Leach, and Shaffer 2003) stemming from a witchcraft revival in Great Britain in the 1950's (Kelly 1991;Russell 1980). Pagans in the 1950's and 1960's from both Great Britain and the U.S. drew from multiple folklores, particularly the "romantic image of a prehistoric matriarchal culture and its goddess religion" (Jorgensen and Russell 1999). Reflecting the broader politics of the times, Pagan movements included feminist critiques of mainstream religion as patriarchal and oppressive (Berger, Leach, and Shaffer 2003;Christ 1982;Goldenberg 1979).
Pagans make up a small but stable religious minority group in the United States. It is difficult to accurately assess the Pagan population due to a number of factors including: high levels of stigma, discrimination, and even violence against Pagans (Adler 1997;Barker 2003;Melton and Poggi 1992;Scarboro, Campbell, and Stave 1994), a general culture of secrecy within Paganism (Adler 1979), dislike of inflexible categorization by Pagans themselves (Adler 1997;Berger, Leach, and Schaffer 2003), and social scientists' inattention to Pagans as a group worthy of rigorous data collection (McClure 2017). We do have some data to guide population estimates. According to the Pew Forum's Religious Landscape Survey (2015), between 2007-2014, 0.3% of the U.S. population identified as Pagan, with the rate remaining steady during that period. Given their extreme marginalized place in American culture, Pagans, then, are a small but, nonetheless, sociologically significant segment of the population. Scant research exists on Pagan parents and families with a few notable exceptions. Data suggest that Pagan parents experience child custody struggles relating to their marginalized religious identity (Adler 1979;Cookson 1997) and that some social workers hold biases against Pagan parents (Yardley 2008), which makes parenting as a Pagan fraught on multiple fronts. In 2003, Berger, Leach, and Shaffer conducted a first of its kind, "Pagan Census" survey. Some crucial information on Pagan parenthood was gleaned from this larger project. The researchers discovered that 41% of Pagans were parents and that many of these parents sought family-friendly groups for support and struggled with Pagans' open and liberal values regarding sexuality (e.g., tolerance towards homosexuality, bisexuality, and polyamory), as these values were used against Pagan parents by some in the mainstream to question their moral worth (and sometimes even custody) as parents.
Political scientist Barner-Barry (2005) studied Pagans generally and in the process revealed some important insights on the common struggles of Pagan parents. Barner-Barry reported Pagans' perception of fear and threat of prejudice and discrimination looming in their daily lives. Barner-Barry also documented Pagans' distrust in mainstream social institutions to protect their interests. Pagans' reports of their daily struggles conveyed the weight that Christian hegemony carried.
Finally, in a previous article (McClure 2017), I bridged lines of research between nonbelievers and Pagans in the U.S. by comparing the stigma management strategies of highly-resourced nonbeliever parents with resourcechallenged Pagan parents in the Bible Belt. Nonbeliever and Pagan parents engaged in defensive othering, a form of reactionary identity work, to cope with threatening stereotypes (the "militant atheist" and the "hedonistic Pagan") that challenged their moral worthiness as parents. According to Schwalbe, Holden, and Schrock (2000, 425): "Defensive othering is identity work done by those seeking membership in a dominant group, or by those seeking to deflect the stigma they experience as members of a subordinate group. The process… involves accepting the legitimacy of a devalued identity imposed by the dominant group, but then saying, in effect, 'There are indeed Others to whom this applies, but it does not apply to me.'" Defensive othering has been observed among seemingly disparate populations: from homeless men at shelters trying to distance themselves from other shelter residents (Snow and Anderson 2001) to residents of trailer park communities separating themselves from their "criminal" neighbors (Kusenbach 2009). Because defensive othering does not require individuals to combat commonly accepted stereotypes, it allows stigmatized individuals an easier path forward. Defensive othering becomes a loophole of sorts wherein individuals establish moral worthiness while avoiding taking on the harshest penalties of stigma and the arduous task of creating social change via activism.
In relying on defensive othering both nonbeliever and Pagan parents unintentionally reinforced the controlling images they sought to combat and rejected activism as too risky. While both groups seemingly relied on the same coping strategy, their varied social class and cultural standings shaped the ways in which they used defensive othering, with nonbelievers using it mainly to buffer personal and professional relationships and Pagans using it as a matter of economic survival. This study demonstrated the importance of examining how economic and cultural resources interplay with parenthood status to shape stigma management strategies.
In reviewing literatures on nonbelievers and Pagans in the U.S., four significant common themes emerge. First, both nonbelievers and Pagans in the U.S. face ongoing prejudice and discrimination, which trigger the need to manage stigma. Second, becoming a parent creates a unique strain for the religiously marginalized, prompting (re)evaluation of identity and stigma management strategies. Third, under conditions of inequality, society is structured in ways that encourage the marginalized to adopt the norms of the powerful. Finally, geography and cultural context shape how one experiences religious marginalization, such that the Christian hegemony of the Bible Belt intensifies the experiences of parenting outside of the religious mainstream.

Emotion Management
Building off of Goffman's earlier work on identity management (1956), Arlie Russell Hochschild's (1979) work helped to refine theories of emotion management. Among many things, Hochschild argued that when individuals are faced with feelings of dissonance over contradictions between what they believe they are supposed to feel in any given situation (the feeling rules) and what they actually feel, they may seek to manage their feelings, often by suppressing negative feelings, through a number of strategies.
In their classic study published in "The Second Shift" (1989), Hochschild and Machung found that couples often manage emotional dissonance by developing family myths, "versions of reality that obscure a core truth in order to manage a family tension." For example, when division of household labor created acute conflict among couples, wives-who shouldered the bulk of the domestic work-implemented an "upstairs/downstairs" family myth; implying that wives take care of the upstairs of the home (consisting of all living areas of the home) while husbands take care of the downstairs (consisting of the garage and yard). While sounding equitable, this arrangement did not challenge the highly unequal division of labor that existed all along. However, it did provide a path for couples-and particularly wives who feared the consequences of divorce more than husbands-a path to remain married and feel more content with the inequality they did not hold the power to resolve. In other words, while the family myth did not manage to close the inequality gap, it did manage to shift the negative feelings associated with the inequality.
While Hochschild and Machung were referring to a strategy used to cope with gender inequality in household labor, their analysis maps onto inequalities experienced by nonbeliever and Pagan parents in their dealings with some Christian family members. Unlike the couples Hochschild and Machung studied, the parents I studied maintained relatively egalitarian parenting arrangements. The inequality they used myths to obscure concerned religious intrusion by proselytizing family members. As such, I use Hochschild and Machung's work on family myths to situate the findings from my current studies.

Setting
The Bible Belt generally refers to a handful of states in the southeastern United States in which Evangelical Protestants make up a larger portion of the population than other areas of the country. Based on data from the Pew Research Center's Religion and Public Life Survey (2015), states in the U.S. South that hold a higher-than-national average population of evangelical Christians (25%) include: Alabama, Arkansas, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, and West Virginia. The percentage of adults identifying with evangelical Christianity in North and South Carolina-where I conducted the bulk of my interviews-is 35% (and 38% in Georgia and 30% in Virginia, where phone interviewees resided). This situates North Carolina and South Carolina squarely within the Bible Belt; neither constituting the highest concentration of Evangelical Protestants (52% in Tennessee) nor the lowest concentration (27% in Louisiana).

Study One: Nonbelievers
Sample I wrestled with language describing secular research participants. I originally sought to study atheist parents. It was only after I began interviewing "atheists" that some parents challenged the language as it pertained to themselves. For purposes of this article, I refer to all interviewees as "nonbelievers" as this term best encapsulates the cohesiveness of the interview sample based on responses to the first question I asked of each interviewee, "Do you hold any belief in God or gods?" Though all interviewees answered in the negative to this question, it is worth noting-and consistent with diversity in identifiers other secularism scholars have documented in their research-that participants held complex identities, only some of which I had the opportunity to explore in this project.
Some respondents referred to themselves as atheists, 4 while others referred to themselves as agnostic, secular humanist, or spiritual but not religious. (See Table 1 for sample demographics of nonbelievers.) These labels differ in their meaning, both literally and in terms of emotional impact to identity management. To be agnostic in one's belief is not the same as to be atheistic. In fact, atheist is such a charged term that a full 40% (8) of interviewees brought up their issues with the term "atheist" unprompted. What each nonbeliever interviewee held in common, binding them together as a cohesive sample, is both their parental status and lack of belief in deity. To remain sensitive to the needs of participants to define their identities, I favor use of the umbrella term, "nonbeliever," embraced by prominent secularism scholars Blankholm (2014) and Cimino and Smith (2014). 5 The majority of the nonbelievers in my sample occupied the middle to upper-middle classes, with almost one-third working as a stay-at-home parent and all but two interviewees holding jobs that can be argued to fit into the middle to upper-middle class. Ninety-five percent of interviewees had at least some college experience with 60% holding bachelor's degrees and 30% holding graduate degrees.

Methods
Data from Study One derive from two methods, including twenty months of participant observation of two local secular groups (one atheist, one secular parenting) and twenty in-depth interviews.

Participant Observation
In 2007, I searched for local atheist groups online. With permission from the organizer I joined an atheist social support group. I attended monthly meetings. The group boasted over 200 members, but meetings, which took place at members' homes or in public places such as bars, usually drew 15-40 people. Meetings sometimes involved the showing of a video (e.g., an atheist debating a Christian) with informal socializing afterwards. At meetings I made small talk, engaged in philosophical and political debate, and socialized with atheists. I attended a total of eight meetings before shifting my observations to a newly formed secular parenting group.
Two members of the atheist group I observed started a secular parenting group five months after I began my research. 6 The online description for the parenting group read as follows: This group provides encouragement to parents and children who are naturalists, freethinkers, atheists, secularists, agnostics, secular humanists, and perhaps pantheists and generally spiritual people who submit to no dogma and who are comfortable around people who do not believe in afterlife.
This group grew to over 90 members in the 20 months in which I conducted fieldwork. Its website currently lists 239 members (as of July 2018). The initial plan was for large, monthly gatherings. Smaller playgroup meetings were later added. Playgroups met at parks, children's gyms, or skating rinks to give children a chance to play while the adults socialized. Eventually, group leaders also added book discussion meetings for adults.
I attended 14 monthly meetings, taking extensive fieldnotes after each meeting. Monthly meetings lacked formal structure, but, while children played together in the background, parents discussed parenting practices and the unique struggles they faced as secular parents. I observed nine playgroup meetings and two book discussion meetings. Finally, I also observed a secular parenting workshop led by author Dale McGowan and I attended a service at the local Unitarian Universalist Church after hearing multiple nonbelievers mention it as a welcoming place.

Interviews
In addition to participant observation, I conducted 20 indepth, open-ended interviews with nonbeliever parents. I recruited interviewees from both the atheist group as well as the secular parenting group that emerged partway through my research. I used snowball sampling to expand my interview pool beyond parents accessed through group meetings and to reach a population that is often difficult to access. Interviews took place in coffee shops or parents' homes. Interview durations ran from ninety minutes to three hours in length. I conducted interviews using a loosely-structured interview guide. Interview questions explored the process of coming to nonbelief, parenting issues associated with nonbelief, important lessons secular parents should teach children, the transmission of morality from parent to child, and how parents coped with discrimination.

Sample
Definitions of Paganism vary, but the most generally agreed upon definition is any organized faith outside Abrahamic, monotheistic faith groups such as Christianity, Judaism, or Islam that holds an earth-based set of beliefs and a desire by members to revive or reconstruct historic spiritual traditions (Kermani 2013). In the U.S., Paganism is most commonly associated with Wicca.
As with the nonbeliever parents, Pagan parents used a diverse set of labels to identify themselves (see Table 2 for sample demographics), with an equal amount (30%) identifying as either "Wiccan/Witch/Strega (an Italian form of Wiccan)" or "Eclectic (Pagan)." Other identifiers included: Pagan, Druid, Animist, and Norse (Pagan).
In comparison to the nonbelievers in Study One, Pagan interviewees in Study Two tended to be situated much more precariously in lower social classes, with a full 40% of interviewees unemployed and another 25% holding low-wage hourly service jobs. Other positions held by interviewees included: stay-at-home parents (with one viewing herself as "solidly middle-class" and the other as "working class" 7 ), carpenter, administrative assistant, mechanic, chef, and entrepreneur.

Participant Observation
To access Pagan parents, I joined two local Pagan support groups. The first group focused on matters of Pagan theology, such as the proper role of ritual, the moral code embedded in the "Wiccan Rede," the ethical use of powers gained through personal trauma, and techniques for grounding and shielding (protection measures) while working magic. This group was much less organized than the atheist group. Meetings were prone to cancellation and sparse attendance (3-7 members). All meetings took place in the organizer's apartment and lasted approximately 2 hours. Though this was not a parenting group, I was fortunate to recruit a few Pagan parents for interviews. On one occasion, the organizer held a meeting on the topic of Pagan parenting, specifically to assist me in my research. I observed a total of 8 meetings over a 13-month period.
In the second group, meetings were held every 4-8 weeks at a local bookstore coffee shop, with 4-9 members in attendance at any given meeting. I attended three meetings before it disbanded six months after I joined. This group functioned mainly as a social group, though theology was also discussed. Many members had been friends for years and interacted outside the group.
These two groups were located in North Carolina. I also attended two meetings for a Pagan group in South Carolina to expand my observations. This group met in a restaurant and was more robust in attendance (18-25 attendees). Topics discussed at this group were theological in nature.
Additional data was drawn from multiple sources. 8 I observed two Pagan parenting workshops (both held at different annual gatherings), three Pagan Pride Festivals, one annual Pagan gathering for each of two organizations. I also attended a service at a local Unitarian Universalist congregation where some Pagans attended.

Interviews
In addition to participant observation, I conducted indepth, open-ended interviews with 20 Pagan parents. Interviews took place in coffee shops, parents' homes, and organized gatherings. Interview durations ran from ninety minutes to three hours in length. I conducted interviews using a loosely-structured interview guide. Interview questions explored the process of coming to Paganism, parenting issues associated with Paganism, important lessons parents should teach children, the transmission of morality from parent to child, and how parents coped with discrimination.
I recruited interviewees from local Pagan meetings. I also relied on snowball sampling to expand my pool of interviewees beyond group participants. Because Pagan parents were harder to identify and recruit than nonbelievers, I conducted 4 of 20 interviews over the telephone. All interviews were tape-recorded and transcribed verbatim. Parenting couples were interviewed apart from their partners whenever possible. However, I conducted three joint interviews of Pagan couples, each time at the insistence of the male partner. These fathers were suspicious of the research process and felt more comfortable with their female partners present, a trend that did not present itself with nonbelievers.

Analysis
Data were analyzed using the constant comparative method, or grounded theory, as laid out by Glaser and Strauss (1967). This analysis was developed semi-inductively. By following this method, I could see how the parents I studied made sense of their lives, how context shaped their identities as parents, and how they developed strategies to cope with the problems that stemmed from their marginal status in a Christian society.
Grounded theory requires the researcher to rely on an iterative inspection of the data to move the research process forward. I did not wait until all data was collected to begin analysis. Instead, I began coding and writing analytic memos after initial observations and interviews. This organic process allowed me to see emerging analytic themes and go back into the field or reach out to past interviewees when needed to ask further questions until a line of analysis was solidified by the data. Saturation of the sample was also achieved in this manner. Writing of analytic memos was followed by writing of higher-order, integrated memos. This iterative process allows the researcher to focus on one analytic thread closely before stepping back to examine how the threads may be woven together. This (semi)inductive process, while organic, is systematically anchored in the data above all else. 9

Findings: How Nonbeliever and Pagan Parents Respond to Proselytizing
In this section, I present findings from the two studies in two parts. First, I begin by establishing proselytizing by Christian family members as common and often problematic-describing both the content and extent to which proselytizers took aim at nonbeliever and Pagan parents and sometimes their children. Second, I compare how nonbelievers and Pagans utilized a particular emotion management strategy-the family myth. I examine the varying degrees to which they rely on family myth and discuss how this strategy is influenced by socioeconomic factors.

Accounts of Proselytizing by Christian Family Members
Not all nonbeliever or Pagan parents interviewed had family members who proselytized. However, some level of proselytizing by family members was discussed by nearly all interviewees (35 out of 40). Some viewed this proselytizing as welcomed (just 9 of 35; 2 nonbelievers and 7 Pagans) or as mere irritation. Others found it so upsetting they cried during interviews. It was, at the very least, an unwelcome complication in the lives of many parents in my studies.
Proselytizing by family members took various forms: silently leaving religious pamphlets at parents' houses; inviting parents and children to attend church activities; asking permission from parents to bring children to church; bringing children to church without parental permission; directly inviting children to attend summer Bible camp; lecturing parents on moral dangers of parenting outside of Christianity; purchasing religious objects for children; pressuring parents to baptize a child; demanding Christian prayers be recited at their own homes, and sometimes others' homes; using babysitting time to speak to children about Christianity; informing children of beliefs in heaven and/or hell behind a parent's back; telling children that parents are going to hell for lack of (correct) beliefs; and secretly baptizing a child without telling the parents. Even this is not an exhaustive list.
Proselytizers shared their religion within their own homes and sometimes tried to force it into the homes of others. Some proselytizing was aimed at parents; in other cases, proselytizing family members appealed directly to children, either in front of parents or behind their backs. Some proselytizing was perceived by nonbelievers and Pagans as harmless or even positive, while other acts were viewed as disturbing. For purposes of this study, I define these acts, major and minor, welcomed and unwelcomed, as "proselytizing." The overwhelming majority (85%) of nonbelievers and Pagans interviewed were raised Christian. They were raised by Methodist, Catholic, Baptist, Pentecostal, Mormon, Lutheran, Jehovah's Witness, and mixed-faith parents. Their families often still held religious worldviews. Though 15% of respondents' families identified as Hindu (1), Jewish (1), Pagan (3), or secular (1), only proselytizing by Christian family members came up in interviews. Not all parents studied had Christian family members that proselytized (5 of 40 did not), and even for those who did, proselytizers were in the minority of family members, though they nearly always held powerful family roles, such as parents, in-laws, and siblings, causing their proselytizing to loom large over family interactions. 10 In parent group meeting settings, when the topic turned to Christian family members' proselytizing, the conversation took on a serious tone. For example, at one secular parenting meeting, Kim, atheist mother to a toddler, shared with a group of five nonbeliever parents her recent struggles with her Christian mother: In this situation and others, parents listened closely to each other's stories, offering thoughtful advice and support with a sense of care and fragility in dealing with this prickly situation. It was evident that dealing with proselytizing Christian family members was a common, and sometimes serious, problem for parents.
In interviews, the strongest objections to proselytizing arose after family members overstepped bounds, particularly with their children. For others, concern intensified after a child began "mindlessly mimicking" beliefs and rituals of Christian family members. Parents were not so bothered by Christian family members' proselytizing to them ("I don't like it. I'll never like it but I can deal with it.") or even their reciting prayers for a newborn baby ("Mom says blessings over [our baby] but it's not like [our daughter] knows what that is."). However, as their children aged and became susceptible to religion, many parents came to see proselytizing as harmful after all.
Though both nonbelievers and Pagans defined proselytizing Christian family members' behaviors as problematic, they did not do so to the same degree. I observed both intragroup and intergroup differences in assessment of risk to children. Some nonbelievers and Pagans felt proselytizing to be more of an irritation than a source of harm. 11 Others saw it as tantamount to child abuse. However, there also existed substantial difference between nonbeliever and Pagan parents in that nonbelievers were more likely to view proselytizing by Christian family as potentially harmful to children. Their Pagan counterparts were more likely to see any possible harm to be minor and/or reparable. Below, I analyze the parents who viewed proselytizing through the lens of potential harm to children (totaling 26 of the 35 interviewees who had experienced proselytizing by Christian family members).

Proselytizing as Potentially Harmful to Children
Ninety percent of nonbeliever parents and approximately two-thirds of Pagan parents interviewed thought that the proselytizing they had experienced by family members could harm their children. They differed in the type and level of harm they imagined proselytizing might cause. In general, nonbelievers feared proselytizing would stifle their child's desire and/or ability to think critically ("If grandma starts telling him Jesus is real, and that Jesus will fix anything for him that's wrong in his life, how will he learn to problem solve?"). Nonbelievers also saw it as harmful to manipulate children's beliefs and behaviors through fear tactics-a practice they associated with some Christian doctrines ("I don't want [my daughter] to be good because she's afraid of, you know, like hell.").
Pagan parents who saw proselytizing as harmful did not want to see their children compelled to embrace a singular worldview ("I want [my kids] to try it all. Christianity, Buddhism, Jewish, whatever helps them. I'm open to it. I don't want, ever, for them to hear that there is only ' one true god,' one way to do religion.") All parents attempted to resist proselytizing by Christian family members by erecting physical or ideological boundaries. Physical boundaries involved either cutting off all interaction between proselytizer and child or attempting to insert oneself between proselytizer and child in interactions. Ideological boundaries included attempts to get proselytizers to separate out the "good" parts of proselytizing from the "bad;" mainly the fear-inducing elements. These first three strategies will be explored in another (as of yet, unpublished) manuscript. For purposes of this article, I turn our attention to an ideological strategy employed by both nonbeliever and Pagan parents but to which they clung to with varying levels of tenacity-the family myth of agreeing to a pact between proselytizer and proselytized to "just never bring up religion."

"We Just Never Bring Up Religion": The Construction of a Family Myth
So, how did nonbeliever and Pagan parents respond when their best efforts to manage intrusions on parenting by proselytizing family members failed? What did they do when they exhausted (or rejected) strategies of: isolating kids from proselytizers; minimizing time spent between children and proselytizers; and/or asking proselytizers to eliminate perceived fear-inducing aspects of religion from their proselytizing? When these failed, some turned to the creation of a family myth.
When physical and ideological boundaries failed, as they often did, nonbeliever and Pagan parents experienced emotional dissonance. Therefore, some nonbeliever and Pagan parents looked to rewrite the narrative of inequality into one of equality in an attempt to manage dissonance. Interwoven between accounts of Christian family members' proselytizing, nearly one-third of nonbeliever parents and 20% of Pagan parents claimed that there was an unspoken arrangement to "just never bring up religion" in order to maintain harmony. Though apparent contradictions-between the proselytizing that parents claimed occurred and the alleged arrangement to never bring up religion-might appear obvious to readers, most of the parents did not recognize these contradictions; at least not initially. Kristine, atheist mother to a son, described her relationship with her Southern Baptist mother: [My mother] doesn't ever ask any more questions about religion. She won't bring it up. And I think it is because she knows that I am probably doing something that she doesn't want me to do, so she just would rather live in ignorance. And that is sad to me because this is a very important part of my life. And I know she doesn't agree with it, but I don't agree with her going to church three times a week either. You know? So, yeah, we've learned to just let [the topic of religion] lie.
While Kristine described her earlier relationship with her mother as extremely difficult, she presents her current relationship with her mother as sad but stable, since they "just don't talk about it." Some Pagans described a similar relationship with Christian parents. Joel, Pagan animist father of three children, described how his relationship with his "totally Christian" mother evolved over time: I was surprised at her, you know? The [angry] reaction when I told my mom [I was Pagan]. She didn't, she just can't get it. She's totally Christian. She will never change. Ever. So, when I told her she was like, "Oh, I'm so disappointed." It was a bad thing. So, if I tried to talk about it with her she would always change the subject. She doesn't want to talk about it. Because she kept saying "You need to find a church." And I told her, well, I found one, you know? The Church of the Elements [Pagan church]. Well, she found out what it was, so she kept telling me I need to find a church, so I just gave up… Yeah, so now I just don't talk about it with her. And I think, I don't know, but we don't ever talk about [religion] anymore. Though Joel admits his mother was devastated when she found out he was Pagan, he frames his current relationship with her in much the same terms as his nonbeliever counterpart Kristine. After trying to gain acceptance through repeated conversations, both parties settled with a frustrating but seemingly equitable tactic of agreeing to never discuss religion. This informal arrangement appeared to allow nonbeliever and Pagan parents to maintain their marginalized identities, while lowering the chances of hostility and conflict with family members. Yet, it also served as a family myth, one that was more apparent to Pagan parents than nonbeliever parents.
Because nonbeliever and Pagan parents described many instances when their arrangement became one-sidedwith efforts to respect Christians' beliefs left unreciprocated but also unacknowledged-I refer to their continued reliance on the "we just don't talk about religion" strategy as a family myth. Hochschild and Machung (1989) found that couples often develop family myths, "versions of reality that obscure a core truth in order to manage a family tension." While Hochschild and Machung were referring to a strategy used to cope with inequality in household labor, their analysis maps onto inequalities experienced by nonbeliever and Pagan parents in their dealings with some Christian family members. Unlike the couples Hochschild and Machung studied, the parents I studied maintained relatively egalitarian parenting arrangements. The inequality they used myths to obscure concerned religious intrusion by proselytizing family members.
Nonbeliever and Pagan parents were not happy with the agreement to avoid the topic of religion, because it left a proverbial elephant in the room. The more serious problem was that religious family members often failed to respect the agreement. Nonbeliever and Pagan parents cited multiple occasions of Christian family members bringing up religion-after supposedly agreeing not to. The rhetoric and reality of these relationships simply did not match.
Atheist Kristine, mentioned in the previous section, claimed a we-just-never-bring-up-religion arrangement with her mother. However, during the same interview she described multiple examples of how her mother inserted her religion into their relationship recently: repeatedly inviting the family to church for special events; talking to Kristine's son (behind her back) about heaven as the place where people go after death; and buying the family a Bible for Christmas. Kristine also strongly suspected that her mother had baptized her son behind her back ("I'm like 95% sure she did.").
Pagan Joel, also quoted in the previous section, experienced similar contradictions between the claim that "religion never comes up" and recent proselytizing by his Christian mother. After he and his wife claimed (in separate interviews) that religion never came up with Joel's "totally Christian" mother anymore, both told of recent times when Joel's mother had brought up religion. Joel's mother had invited them to church, invited her grandchildren to church without asking their permission, and left religious pamphlets in their home, all in the year prior.
In both Joel's and Kristine's cases, the supposed agreement to never bring up religion was allegedly not respected by their Christian mothers. They were not alone. Almost one-third of nonbeliever and 20% of Pagan parents interviewed cited a similar arrangement based on what one atheist referred to as a "live and let live" attitude.
Nonbeliever and Pagan parents claimed to have kept up their side of the bargain by not discussing their beliefs or practices with Christian family members. This agreement, however, was left unreciprocated. Nonbeliever and Pagan parents-those who claimed that religion never came up with Christian family members-cited the following examples of religious intrusion: • The father of an atheist dad left an anti-atheist ( pro-Christian) book in the guest room after visiting for a weekend. • The mother of a Pagan mom signed her granddaughter up for vacation Bible school without permission. • The mother-in-law of a Pagan woman annually invited her to a birthday party for Jesus. • The step-father of an atheist angrily demanded that the family say grace before every meal eaten in the atheist mother's home.
Examples of proselytizing listed above are perhaps not surprising in the Bible Belt. More surprising was the differing level of commitment that nonbeliever and Pagan parents devoted to the family myth. Consider how Melissa, eclectic Pagan, mother of three, used the family myth strategy: Melissa: I guess, um, that even though my dad and I will never agree [on religion], we just try not to talk about it at all. It's easier that way. It works that way.
Interviewer: So, you don't want to talk about it but he still gives you a hard time about being Pagan, right? (referring to points she made previously in the interview) Melissa: Oh, for sure. He, um, doesn't want my kids to be [Pagan]. He wants them raised like I was (Christian).
Interviewer: So, you try not to bring religion up when you talk but he does anyway?
Melissa: Yeah, he's always going to do that.
Melissa employs the family myth initially but faced with inconsistencies she concedes the strategy as myth.
Here, too, Wiccan mother, Jane, discusses her mother's proselytizing: [My mom] doesn't pressure me. It's not the kind of thing that comes up really.
Interviewer: Didn't you say you were unhappy that she kept trying to get you and your daughter to come to her church and with her discussing her religion with [your daughter]? Jane: I did. I did. I just want to raise [my daughter] free from all that. Interviewer: And your mother won't just let you raise her how you'd like to? Jane: No, she won't. She gets really worked up [about it].
Jane employs the family myth strategy claiming "[Religion]'s not the kind of thing that comes up…" with her mother. Yet, when pressed, she readily admits that her mother does indeed bring religion up. Now compare these to two examples from nonbeliever parents. Ian, atheist father to two children talked about his relationship with his Christian mother: Interviewer: So, your mother wishes you were raising the boys in her religion?
Ian: She won't tell me that to my face. She knows I won't have it. That crosses a line.
Interviewer: You said she took the boys to church with her last month?
Interviewer: Did she ask you first?

Ian:
No. She just took them. She wouldn't ask me because she's not allowed to. That would not be ok. It's a real sore spot for us so we just don't [discuss religion] and that keeps the peace.
Interviewer: So, she brings the kids to church but you two don't discuss religion with each other oneon-one? Ian: (Getting frustrated with me) No, we don't. Why would we even go down that road when she knows what I believe? It's not something we're going to discuss.
Ian clings to the comfort that the family myth of equality provides him even as he provides evidence to the contrary of this supposed arrangement.
Sandy, atheist mother to a newborn daughter, also framed her relationship with both her mother and her step-father, whom she referred to as "fundamentalist Christians," as tense but manageable due to their unspoken arrangement to not bring religion up as a topic of conversation: Sandy: Look, I don't have a great relationship with [my mother and step-father].
Interviewer: Due to differences in your beliefs? Sandy: I mean, yes, but there are other issues too.
Interviewer: So, then how do you deal with those differences when they come up? Sandy: They don't [come up.] That's how. I make sure it never comes up.
Interviewer: Like, never? It never just comes up? Even though they are fundamentalist in their religion?

Sandy: Nope.
Interviewer: What about when your step-father makes you say grace at your house when they're over for dinner? Sandy: That's different. It's not a conversation.
Interviewer: So, it doesn't feel like religion to you? Sandy: No. No. Not religion. Just ritual.
In the interview Sandy spent a great deal of time describing in detail the difficulties she faced from dealing with her mother's and step-father's proselytizing, including their demands that she recite mealtime prayers in her own home when they came to dinner. Yet, near the end of the interview she began to use the family myth rhetoric.
Out of 4 Pagans who reported the "we just don't talk about religion" strategy; all 4 conceded the strategy quickly and easily as unreciprocated by Christian family members when pressed in interviews. There was little expression of frustration or irritation with me in provoking this concession. When the 6 nonbeliever parents who reported reliance on the "we just don't talk about religion" strategy were pressed in interviews to address inconsistencies between rhetoric and reality, only 1 readily conceded this disconnect; leaving the other five putting up some form of resistance to the suggestion of the agreement as family myth. Their resistance took the form of denial but also irritation directed at me for asking the question and seeming frustration that I just couldn't "get it." Continued reliance on this emotion management strategy in the face of evidence to the contrary and the emotional irritability the question evoked in interviewees demonstrates just how emotionally fraught and deeply painful these intrusions on parenting, and the inability to stop them, can be for some parents. When all else failed, a family myth became the next best option to admittance of defeat for some.

Discussion: How Social Class Shapes Reliance on Family Myths
Many nonbeliever and Pagan parents used a rhetoric of equality to describe a condition of inequality. But, as demonstrated, they differed in their reliance on this rhetoric. Nonbeliever parents appeared to be less willing than their Pagan counterparts to concede their inability to stop Christian family members from proselytizing. When probed about their emotional responses to this situation, nonbeliever parents expressed more anxiety than Pagan parents; even going so far as to express irritation towards me in interviews. This seemingly made the myth of equality more important to them as an emotion management device. Pagans more readily admitted the gap between rhetoric and reality, acknowledging that the "we just don't talk about it" strategy was not consistent with their experiences of proselytizing. If Pagans were more willing to accept this situation, it was perhaps because their tenuous class position made them more dependent on economic resources obtained through family ties than their nonbeliever peers.
The nonbelievers I interviewed occupied the middle to upper-middle class. Their education levels included: high school degree (1), associate's degree (1), undergraduate degrees (12, with two people holding dual bachelor's degrees), master's degrees (2) and PhD's (4). They held careers in: computer programming, accounting, stayat-home parenting (with partner in professional career; totaling 2 self-described stay-at-home fathers and 5 stayat-home mothers), business ownership, law, military service, education, banking, and scientific research. Only one father, a trained lawyer, referred to himself as unemployed.
The Pagans interviewed, on the other hand, tended to occupy poor, working, or lower-middle classes, with just one interviewee describing herself as solidly middle-class (from an inheritance from her father). Pagans' education levels included: GED (1), high school diploma (6), some college experience but no degree (4), associates or 2-year technical degrees (5), and bachelor's degrees (4). They held jobs such as: customer service manager, administrative assistant, carpenter, chef, and mechanic for the military. Two mothers explicitly referred to themselves as stay-at-home moms; with no men referring to themselves as stay-at-home fathers despite the fact that seven male interviewees were unemployed.
One alternate explanation might hold that nonbeliever parents are actually equal in status to their proselytizers due to their social class positioning. Perhaps they merely recognize that they cannot change the behaviors of proselytizing family members and instead choose to allow them to proselytize. This would be a solid interpretation-and admittedly may be the case for a couple of my interviewees-if not for the fact that nonbeliever parents who made use of the family myth did not readily admit that they were allowing proselytizing to occur. Instead, nonbeliever parents used the myth to imagine that proselytizing has (nearly) stopped and that they maintained control over this arrangement by holding up their end of the bargainnot bringing up religion with Christian family members.
One of the major advantages of a family myth is that it allows one to side-step the dissonance stemming from an unequal reality. Therefore, the reliance on family myths seemed to serve as less of a choice and more as a rhetorical strategy used to quell the discomfort and pain of finding one's agency stymied.
The mainly poor and working-class Pagan parents, who often relied on economic and other forms of help from family members, perhaps felt that the risk of proselytizing was minor compared to the benefits that came from maintaining supportive ties with Christian relatives. Onequarter of Pagan parents, unprompted, mentioned family assistance through housing, transportation, free or cheap babysitting, and assistance with groceries and necessities such as healthcare.
Here, Greg, a witch, 12 described his financial dependence on his parents: I wouldn't share my beliefs with my parents-well, my mom in particular. She would not stand for it.
Interviewer: So, you are concerned that if she knew your beliefs you wouldn't get along?
Greg: Yeah, kind of, well it's that and…I mean it would lead to fights but also, we all live together (in his parents' home). So, we need to all get along (his emphasis) Greg, a single father, was employed as a low-wage film processor. Living with his parents was not a lifestyle choice. It was a necessity.
Doreen, an eclectic Wiccan, single mother of a six-yearold, and retail cashier, describes the financial reasons why she needs a relationship with her mother: Though not generalizable to all Pagans interviewed, 25% of Pagan parents, unprompted, mentioned relying on family for material assistance to some degree, and many more did so outside of interviews in group settings. Under these conditions, family relationships, perhaps serve as a lifeline for some Pagan parents.
The more affluent nonbeliever parents could make a different calculation. They had no compelling economic reasons to tolerate proselytizing and thereby risk their child's open-mindedness and ability to think critically. Once level of resources and specific parenting values are taken into account, it makes sense that many more nonbelievers than Pagans would see proselytizing as a threat to their children.
So, then, why might middle to upper-middle class nonbelievers cling more tightly to a family myth of equality than their poor and working-class Pagan counterparts? You might suspect that having more resources would enable nonbelievers to absorb the reality of inequality in this one aspect of their lives better than their Pagan peers. Yet, this did not appear to be the case. Here, a social class analysis may be warranted. Pagan parents were presumably accustomed to experiencing the world as unfair, given their lower social class and occasionally demeaning service-work jobs. Though they actively sought to expand their agency through religion, they did not experience much agency in the rest of their lives. Being dismissed, looked down upon, having to hide one's "true self" and generally being controlled by those with more power might then be reasonably presumed to be an everyday reality for many Pagans; a reality that may have allowed them to see through the family myth more readily than their nonbeliever counterparts.
The upper-middle-class nonbeliever parents were presumably accustomed to having far more control over their daily lives. They often held powerful positions in their careers and leadership roles in groups and organizations of which they were a part. They had ample financial and educational resources to assist them in coping with the stigma of raising children outside of religion, such as homeschooling their children or enrolling them in private Montessori schools where their nonbelief would be respected (McClure 2017). Therefore, perhaps, they believed they were entitled to have their wishes as parents respected. When the boundaries they laid down were not respected by some Christian family members, this likely threatened their sense of control as individuals, and especially as parents. Family myths of equality minimized this identity threat and restored their sense of control.

Conclusion: Intrusions on Parenting
Nonbeliever and Pagan parents are much like other groups of minority parents. Previous research on intrusion in parenting by powerful outsiders has focused mainly on how bureaucratic government representatives have used their power to shape and limit the lives of marginalized parents. For example, social scientists have studied how the American military enforced government mandates to reshape parenting practices of imprisoned Japanese individuals in America during WWII (Espiritu 2007;Levine 1995); how conditions of immigration family detention centers in the U.S. shape parenting (Ortega, Graybill, and Lasch 2015); how DACA recipients navigate access to familial resources (Gonzales, Terriquez, and Ruszczyk 2014); how social service agencies police and constrain the parenting practices of foster parents (Swartz 2004); how courts have withheld full access to parental rights for gay and lesbian parents (Weston 1997;Sullivan 2004); how government has implemented welfare legislation to promote traditional marriage (Coontz and Folbre 2010); and how courts systematically deny attorney rights to families being evicted from their homes (Desmond 2016). In sum, an examination of how powerful others intrude on the parenting (and family formation) of the marginalized has been well documented in the sociological literature.
As suggested above, studies of intrusion on marginalized parents tend to focus overwhelmingly on the power of government (via legislation, court mandates, military might, and social service policies, etc.) or employers to shape and limit the contours of family formation and parenting. Most of these studies emphasize how the tremendous disparity in power between government and minority parents reinforces preexisting inequalities. Yet intrusion on parents by the privileged is not limited to faceless bureaucracies. It can also take the form of a more intimate dynamic, such as the dynamic of parent and (adult) child. Though nonbeliever and Pagan parents were not subordinated to proselytizing Christian family members in the same manner as individuals are to their government or employers, they faced a complex and difficult emotional relationship to negotiate in trying to enforce their rights as parents. A welfare recipient can distance him-or herself emotionally from a meddling social service worker. In such cases, the impersonal nature of the intrusion can make it easier to deal with emotionally.
This coping mechanism did not work for the parents I studied. When nonbeliever and Pagan parents vented about Christian family members' proselytizing, it was often accompanied by feelings of guilt and anxiety. These parents wanted to protect their children from religious intrusion while maintaining loving relationships (and resources) with key family members. When they did the former, they sometimes sacrificed the latter, and they felt bad about it.
Nonbeliever and Pagan parents found themselves in the uncomfortable situation of having to defend their parenting from loved ones, parents, or in-laws who routinely overstepped appropriate boundaries. This occurred in large part because nonbelievers and Pagans found themselves in a broader culture that devalued their beliefs. The situation was made worse by the strength of Christian hegemony in the Bible Belt South. No matter what defense they mounted, some Christian family members kept trying to recruit them. This reveals much about power, privilege, and Christian hegemony in the Bible Belt. When nonbeliever and Pagan parents failed to follow local norms for the moral socialization of their children, some Christian family members saw the opening to proselytize and felt justified, it seems, in doing so.
The experiences of nonbeliever and Pagan parents show that the reproduction of Christian hegemony relies not only on the powerful working together to maintain privilege, but also, to a smaller degree, on the cooperation of the less powerful. In the present case, nonbeliever and Pagan parents capitulated to Christian hegemony, at least in part, for the sake of family harmony. More aggressive enforcement of physical and ideological boundaries would have been emotionally and-for the Pagans-economically costly. Though I did not directly study proselytizers, I suspect that they knew their intrusions, though clearly unwanted, were hard to resist.
In this study, I have examined parents who are pained by the intrusion on what is expected to be private terrainone's parenting. Flipping the research around to analyze the motivations and strategies of the accused proselytizers could also yield a more holistic understanding of this phenomenon. However, regardless of the individual intent of proselytizers, were it not for the Christian hegemony of the Bible Belt, nonbeliever and Pagan parents would likely have had the privacy in parenting respected that they expected and believed they deserved.

Notes
1 All names of individuals and organizations have been changed to protect the identity of research participants. 2 Iuse the language of "Pagan" and "Paganism" because these are the terms used by the people I studied. Technically, however, "neo-Pagan" and "neo-Paganism" are the more accurate terms for the modern movements of polytheistic, "environmentally-based" religions. 3 I discuss choice of cohering identifier label in the "Sample" subsection of the "Study One: Nonbelievers" section. 4 For discussion of the negative reaction some respondents took with the technical, but politically charged, label of "atheist," see (McClure 2017). 5 The difficulty in finding a cohering identifier label illuminates the complexities of modern secularism wherein beliefs, practices, and issues of power come to a head in a manner that eschews facile labeling. For an examination of the ongoing debates among nonbelievers over labels, I refer you to Blankholm (2014). 6 Organizers of the secular parenting group claimed my interest in the experiences of secular parents helped catalyze the formation of the group. I did not pitch this idea, nor did I have any hand in founding the group, but it is an example of the effect a researcher can have on their participants unwittingly. 7 I did not ask interviewees to choose a social class with which they identified. I extrapolated social class from a combination of education level and job/profession. However, sometimes-such as in the case of the 2 stayat-home Pagan mothers-participants offered a social class with which they believed they belonged. 8 I analyzed archival data. Specifically, I drew data from online group message boards and multiple local news articles published on secular and Pagan parenting, conducting content analysis of stories and readers' comments. Archival data informed-but is not directly incorporated into-the present analysis. 9 For a more thorough explanation of grounded theory analysis, I refer you to Kathy Charmaz's "Constructing Grounded Theory" (2006). 10 It is likely that Christian family members are more likely to proselytize to apostates than to those who are not apostates. However, it is beyond the scope of this article to analyze this possibility. 11 A small minority of parents interviewed viewed proselytizing by family members as harmless to children (2 nonbelievers and 7 Pagans); with Pagans being less likely to imagine potential for harm to children than nonbelievers. Though worthy of notation, I do not have space in this manuscript to analyze this in-depth. 12 Though many associate the identifier "witch" with women, not all Pagans observe such strict gendering of labels. Greg's self-identification was as a "witch."