Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

What Do You Mean, “What Does It All Mean?” Atheism, Nonreligion, and Life Meaning:

What Do You Mean, “What Does It All Mean?” Atheism, Nonreligion, and Life Meaning: Nonreligion is often thought to be commensurate with nihilism or fatalism, resulting in the perception that the nonreligious have no source of meaning in life. While views to this effect have been advanced in various arenas, no empirical evaluation of such a view has been conducted. Using data from the 2008 American General Social Survey (N = ~1,200), we investigated whether atheists, the religiously unaffiliated, and persons raised religiously unaffiliated were more likely than theists, the religiously affiliated, and persons raised with a religious affiliation to report greater levels of fatalism, nihilism, and the perception that meaning in life is self-provided. Results suggested that these groups did not differ with regard to fatalism or nihilism. However, atheists and the religiously unaffiliated (but not persons raised in a religiously unaffiliated household) were more likely to indicate that meaning in life was endogenous—that is, self-produced. While atheists and the nonreligious differed from their counterparts on source of meaning in life, this was not associated with any “penalty” for overall existential meaning. Keywords atheism, meaning in life, nonreligion, nihilism, fatalism, nonbelief A core component of human psychological functioning is the justification for goals and purposes, a sense of control, and construction of meaning and purpose in life, a process that is self-worth. Schnell (2009) identified twenty-six sources of subject to self-perceptions and situational effects (Baggini, meaning in life, ranging from aspects of community and 2004; Baumeister, 1991; Berger, 1967/1990; Park, 2010). togetherness, to nature, generativity, religion, and self- However, for atheists and other nonreligious individuals, it is knowledge. In a nuanced approach, Park (2010, 2013) has sometimes assumed that being without god(s) is the equiva- drawn a distinction between global meaning and situational lent of being bereft of meaning or purpose in life (cf. meaning. Global meaning refers to abstract conceptual sys- Blessing, 2013). This assumption is somewhat odd, given tems and schema such as “religious beliefs . . . fairness, con- that Bering (2002, 2003) suggests humans are evolutionarily trol, coherence, benevolence of the world and other people” hardwired with an existential theory of mind, which is a gen- (Park, 2013, p. 361). Situational meaning refers to ascrip- eral cognitive mechanism compelling humans to find reli- tions of meaning or purpose in specific encounters; the gious or philosophical meaning or purpose in life events meaning or purpose a person ascribes to an event depends on (Coleman & Hood, 2015). Similarly, sociological approaches her or his preconceptions and experiences. to meaning conceive it as a compulsion “. . . to impose a Because meaning and purpose in life can be framed in meaningful order upon reality,” all the while emphasizing numerous ways and are derived from multiple sources and the creation and sustainment of meaning as an inherently circumstances, one could reasonably expect that some social, cultural, and discursive practice (Berger, 1967/1990, sources or paths substantially differ from others. Park and p. 22). In essence, the impetus to construct meaning or pur- pose in life is a quintessential consequence of being human, University of New Brunswick, Saint John, NB, Canada rather than something wholly under the purview of a specific Newfoundland & Labrador Centre for Applied Health Research, religious or philosophical framework. Saint John’s, NL, Canada In this vein, there is no shortage of views on what life 3 Coventry University, UK meaning is, how it “works,” or how we should conceive of it. U.S. Air Force Academy, Colorado Springs, CO, USA Frankl (1959/2006) argued that meaning in life was some- Corresponding Author: thing that each individual seeks and constructs for them- David Speed, Department of Psychology, University of New Brunswick, selves, whereas Baumeister (1991) argued that life meaning 100 Tucker Park Rd., Saint John, New Brunswick, Canada, E2L 4L5. is comprised of four elements: a goal or purpose, values and Emails: David.Speed@med.mun.ca; dspeed@unb.ca Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License (http://www.creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). 2 SAGE Open McNamara (2006) note that religious systems provide pur- deficits (e.g., Barrett, 2012), and popular perceptions remain pose or meaning to individuals. In fact, this is often seen as a that they are likely to be more nihilistic or have viewpoints con- primary function of religion. Other researchers have echoed sistent with fatalism (cf. Blessing, 2013). Other research sug- these sentiments and explicitly recognized that religious sys- gests that nonreligiosity “divest[s] people of certain age-old tems provide strong and coherent sources for meaning in life, pathways to psychological health” (Schumaker, 1992, p. 65). due to the fact that these systems appeal to an “ultimate” Additionally, research shows consistent links between source (Crescioni & Baumeister, 2013; also see Vail et al., meaning in life and psychological well-being (e.g., Krause & 2010). Park and McNamara (2006) suggest that not only are Pargament, 2017; Schnell, 2009; Zika & Chamberlain, 1992; “religious frameworks comprehensive, but they tend to be for a review, see Steger, 2017), as well as meaning in life and much more ‘existentially satisfactory’ than secular explana- religiousness or religiosity (e.g., Ivtzan, Chan, Gardner, & tions such as the cold, hard objectivity of science” (p. 67; Prashar, 2013; Steger & Frazier, 2005; Tiliouine & Belgoumidi, also see Farias, Newheiser, Kahane, & de Toledo, 2013; 2009). These findings align with Schumaker (1992), who Schumaker, 1992, on belief in science in the face of stress found that accounting for the relationship between meaning in and existential anxiety). life and well-being substantially reduces the salutary effects of religiosity (Schumaker, 1992). In other words, some of the benefits associated with religiosity are due to its positive rela- Nonreligion and Meaning in Life tionship with producing meaning in life. There are some major drawbacks of this aforementioned However, other research has shown that the irreligious do research. One is an inability to separate the functioning of not experience deficits in meaning in life (Caldwell-Harris purely secular psychological mechanisms from any specific et al., 2011; Wilkinson & Coleman, 2010), happiness (Speed, religious/spiritual processes (Galen, 2017a, 2017b). This 2017; Speed & Fowler, 2017), or well-being (Galen, 2015; inability may be related to research suggesting that religious/ Streib & Klein, 2013), and that the irreligious do not differ in spiritual activities do not have intrinsic benefits, but rather terms of psychological well-being (Galen, 2015; Streib & benefits that are contingent on an individual’s valuation of Klein, 2013; although cf. Hayward, Krause, Ironson, Hill, & those activities (Speed, 2017; Speed & Fowler, 2017). This Emmons, 2016). Essentially, the literature is incongruous: accords with the simple view that “whatever makes life mean- irreligious persons are ostensibly disadvantaged for psycho- ingful is heavily loaded with whatever people value” (Klinger, logical well-being, but do not seem to report poorer psycho- 2012, p. 29). In other words, some persons may find that with- logical well-being. out a framework centered on the divine or transcendent, they This dissonance within the research is problematic for a are more likely to adopt a nihilistic or fatalistic perspective number of reasons. More than 20% of the American popula- (i.e., a negation of [life] meaning; Crosby, 1988). However, tion identifies as nonreligious (Religious Landscape Study, the conclusion that god(s) or religious frameworks are neces- 2016), and some projections suggest this number could climb sary for meaning is predicated on the idea that religion or to almost half of the U.S. population by the year 2042 (see spirituality intrinsically promote meaning for everyone, which Stinespring & Cragun, 2015, who estimate between 26% and can be read to imply that meaning cannot be internally derived 47%). Furthermore, in adolescents and young adults, we find or generated. The perception that an individual is forced to a generational difference, which suggests children are sig- accept nihilism or fatalism in their worldview because they nificantly less religious than their parents (Thiessen & lack a religious or spiritual schema with which to interpret the Wilkins-Laflamme, 2017; Twenge, Exline, Grubbs, Sastry, world is unsupported by the existing literature (Caldwell- & Campbell, 2015), and have relationships with religion Harris, Wilson, LoTempio, & Beit-Hallahmi, 2011; Coleman ranging from explicit repudiation to quiet apathy (Lee, 2012; & Arrowood, 2015; Langston, 2014). Furthermore, the logic see also Francis & Robbins, 2004). of why religious or spiritual frameworks would promote The relationship between youth and religion can also be meaning in life is highly selective. It could be argued that a confusing when we consider whether a religious home person who surrenders to god(s)’ perceived will would accept environment confers a health benefit (i.e., meaning in life, psy- that there is one path to follow or that there is no meaning chological well-being) or whether an unaffiliated home envi- other than serving god(s). Instead, most discussions surround- ronment confers a health penalty. Research suggests that the ing this topic tend to assume as a default that relatively higher home environment can predict future religious/spiritual identi- religion or spirituality, however measured, implies greater ties (Baker & Smith, 2009; Beit-Hallahmi, 2015; Gervais & meaning (e.g., Crescioni & Baumeister, 2013; Pargament, Najle, 2015), but the health consequences of said home envi- 1997; Park & McNamara, 2006). ronment remain unclear. Studies comparing secular sources The consequences of atheism or being religiously unaffili- and levels of meaning, values, and purpose with religious ated for meaning in life, and for psychological well-being more sources are virtually nonexistent (Koenig, 2012), as are the broadly, are unknown or at the very least not held to be static or consequences of nonreligiosity, a nonreligious upbringing, and linear. A cursory examination of the literature would show that atheism on mental well-being (Galen & Kloet, 2011; Hwang, atheists are characterized as having various psychological Hammer, & Cragun, 2011; Morgan, 2013). Speed et al. 3 The existing literature on the relationship between theo- nature of the data. The exact number of participants fluctu- ated slightly from one analyses to the next, but was approxi- logical beliefs, atheism, and fatalism allows us to broadly mately N = 1,200 (see Table 1 for descriptive statistics). suggest that fatalism entails an acceptance of what is per- ceived as an inevitable or uncontrollable outcome, rather than a lack of meaning in life or a lack of purpose. However, Measures fatalism is negatively related to environmental mastery, Meaning in life. Endogenous Meaning, Fatalism, and Nihilism itself a component of psychological well-being (Greenfield, were measured using a 5-point Likert-type scale (i.e., 1 = Vaillant, & Marks, 2009). Conceptually, a person who strongly disagree, 2 = disagree, etc.), with higher scores indi- would say that there is little that can be done to change the cating greater levels of agreement. The 2008 GSS had an item circumstances of his or her life (a person low on a sense of we described as Endogenous Meaning (“Life is only mean- mastery or internal locus of control) is arguably less likely ingful if you provide the meaning yourself”), and was con- to believe that life, the universe, and personal relationships ceptually similar to other questions dealing with whether life are meaningful in any global or time-invariant sense. meaning was internally constructed (e.g., Coleman, Silver, & While some research shows that mastery is positively cor- Hood, 2016; Goodenough, 1998; Schnell & Keenan, 2011; related with religious attendance and/or religiosity (Ai, Streib & Hood, 2013). The 2008 GSS had two items as proxy Peterson, Rodgers, & Tice, 2005; Ellison & Burdette, measures for meaning in life: Fatalism (“There is little people 2011; Schieman, Pudrovska, & Milkie, 2005), other can do to change the course of their lives”) and Nihilism (“In research finds that this relationship is inconsistent my opinion, life does not serve any purpose”). Although the (Greenfield et al., 2009; Speed & Fowler, 2017). Thus, 2008 GSS had an item that assessed God Meaning to some while there is some evidence that theists and the religiously extent (“To me, life is meaningful only because God exists”), affiliated would be more likely to report greater mastery, we did not consider using it because this question would there is no persuasive evidence that atheists, the nonreli- make little conceptual sense with the definition of atheism gious, and those with a nonreligious upbringing are predis- used in the present study. Conceptually, Endogenous Mean- posed toward fatalism (see also Langston, 2014, who finds ing provides information as to the perceived source of mean- that higher external locus of control scores reflect higher ing in life, while Nihilism and Fatalism should be inversely odds of being a theist). related to whether life has a personalized sense of meaning. The present study investigates the relationship between an internal source of meaning, fatalism, and nihilism in a Religious/spiritual identifiers. We made use of three R/S Identi- nationally representative American sample in a series of nine fiers as they related to meaning in life: (a) Belief Identity, (b) planned analyses. More specifically, we sought to investigate Religious Affiliation, (c) and Religious Upbringing. Belief whether belief in god(s), religious affiliation, or religious Identity was based on the GSS item, “Which best describes upbringing were significant predictors of an internal source your beliefs about Gods?” Response options included (a) I of meaning, fatalism, and nihilism. don’t believe in God now, and I never have; (b) I don’t believe in God now, but I used to; (c) I believe in God now, Method but I didn’t used to; and (d) I believe in God now, and I always have. Persons indicating that they believe in God Participants now were labeled as “Theist,” and persons indicating that We accessed data from the 2008 General Social Survey they did not believe in God now were labeled as “Atheist.” (GSS), which was collected by the National Opinion Belief Identity was coded, Theist = 0 and Atheist = 1. While Research Center (NORC) at the University of Chicago the definitions provided for atheism are occasionally con- (Smith, Marsden, Hout, & Kim, 2013). These data are freely flicting (Hwang et al., 2011), the definition used by the pres- accessible from NORC’s website, which also provides a ent study is consistent with negative atheism (see Bullivant, copy of the questionnaire and the user manual. Because this 2013), which is the most inclusive definition of nonbelief. was secondary data analysis, we did not require the Religious Affiliation was based on the item, “What is Institutional Review Board (IRB) clearance from any institu- your religious preference?” followed by a list of different tions to conduct the present study. The 2008 research year religions. Persons who indicated “None” were labeled as was chosen because it is the most recent year that contained “Religiously Unaffiliated” and persons who indicated any- all variables of interest to our research question. To be thing other than “None” were labeled as “Religiously included in the present study, respondents had to have Affiliated.” Religious Affiliation was coded as Religiously answered all covariate questions, at least one of the outcome Affiliated = 0 and Religiously Unaffiliated = 1. variables, and at least one of the religious/spiritual (R/S) Religious Upbringing was based on the question, “In what identifier variables. Persons answering questions with “I religion were you raised?” followed by a list of different reli- don’t know” or persons who answered by refusing to respond gions. Persons who indicated “None” were labeled Raised were excluded from the analyses to maintain the continuous Religiously Unaffiliated, and persons who indicated anything 4 SAGE Open Table 1. Weighted Descriptive Statistics by Religious/Spiritual Identifiers. M (SD) or percentages within categories Belief identity Religious affiliation Religious upbringing Theist Atheist Affiliated Unaffiliated Affiliated Unaffiliated N = 1,034 115 986 192 1,078 96 Fatalism 1.95/1.01 1.91/1.09 1.97/1.03 1.83/0.92 1.94/1.01 1.94/1.03 Nihilism 1.53/0.73 1.67/0.76 1.53/0.72 1.67/0.79 1.55/0.74 1.61/0.66 Endo. Mean 3.08/1.29 3.95/1.04 3.08/1.30 3.64/1.13 3.16/1.30 3.30/1.24 Age 45.98/16.33 40.01/15.37 46.49/16.39 39.2/14.24 45.73/16.32 40.78/15.05 Income 5.57/4.47 6.94/5.59 5.59/4.51 6.36/5.15 5.71/4.6 5.76/4.93 Sex Female 54.37% 24.34% 54.25% 33.60% 51.62% 42.69% Male 45.63% 75.66% 45.75% 66.40% 48.38% 57.31% Race White 75.97% 82.74% 76.36% 78.15% 75.99% 85.73% Black 14.85% 4.95% 13.92% 11.06% 14.29% 4.88% Other 9.17% 12.31% 9.72% 10.79% 9.72% 9.38% Education <High School 15.58% 6.81% 15.11% 12.66% 14.95% 12.38% High School 49.33% 45.93% 48.05% 52.43% 48.68% 49.53% Junior College 9.88% 1.23% 10.09% 4.11% 8.81% 10.89% Bachelor 16.90% 29.24% 17.87% 19.11% 17.79% 22.76% Graduate 8.31% 16.78% 8.89% 11.68% 9.78% 4.43% Marital status Married 56.54% 43.64% 56.41% 46.41% 55.83% 45.11% Widowed 5.58% 1.70% 5.71% 1.84% 5.36% 1.38% Divorced 11.12% 8.30% 11.18% 9.89% 11.00% 11.77% Separated 3.36% 0.70% 3.41% 0.93% 3.14% 1.75% Never Married 23.39% 45.65% 23.28% 40.93% 24.68% 39.98% Note. Only persons who answered all questions of interest are included in this table. Endo. Mean = Endogenous Meaning. other than “None” were labeled Raised Religiously Affiliated. and a weighting correction for nonresponse rate. To achieve Religious Upbringing was coded as Raised Religiously this weighting, Stata’s survey analysis module was used to Affiliated = 0 and Raised Religiously Unaffiliated = 1. specify the weighting parameters by using the syntax pro- vided by Smith and colleagues (2013). In situations where a Covariates. The present study controlled for sex (male/ stratum only had a single respondent, we used a scaled female), age, race (White, Black, Other; White served as approach to address issues with determining variance. the omitted category), marital status (married, widowed, Hierarchical linear regression was used in all analyses. divorced, separated, never married; married served as the Because we used Stata’s survey analysis module, denomina- omitted category), education (less than high school, high tor degrees of freedom in all regression models were based school, some postsecondary, postsecondary, graduate on the number of strata rather than the number of respon- degree; less than high school served as the omitted cate- dents. To meet the underlying assumption of homogeneity of gory), and income in constant dollars (e.g., 1 = US$10,000; variance, heteroscedastic consistent errors (HC1) were 2 = US$20,000, etc.). Please note that “sex” was used in employed. Multicollinearity was not an issue, as the mean lieu of “gender” as the GSS had asked a dichotomous male/ variance inflation factor did not exceed 1.13 for any model. female question. Nine hierarchical linear regression models were used to assess all three R/S Identifiers (belief, affiliation, upbring- ing) and three outcome variables (fatalism, nihilism, endog- Procedure enous meaning). Each model followed the same pattern. All data analysis was done with Stata 13. The 2008 GSS used a complex sampling methodology to achieve a representative Block 1: R/S Identifiers were added (Belief Identity, sample. Because of this, data were weighted to correspond to Religious Affiliation, and Religious Upbringing). which strata respondents were in, the primary sampling unit, Block 2: Covariates were added. Speed et al. 5 Table 2. Belief Identity Predicting Endogenous Meaning, Fatalism, and Nihilism. B coefficients/robust SE Endogenous meaning Fatalism Nihilism (n = 1,156) (n = 1,165) (n = 1,161) Block 1 Block 2 Block 1 Block 2 Block 1 Block 2 Constant −.07/.05 .25/.22 .01/.04 .51/.24* −.02/.05 .87/.25** Belief ID .64/.10*** .57/.10*** −.07/.15 .11/.14 .17/.11 .19/.12 Sex (M/F) −.05/.08 .00/.07 −.15/.06* Age −.00/.00 .00/.00 −.00/.00 Race (White) Black −.08/.10 .28/.10** −.04/.10 Other .22/.12 .18/.11 .42/.17* Married Widowed .16/.15 −.10/.15 −.01/.11 Divorced −.10/.13 −.22/.11 −.13/.09 Separated .22/.22 −.03/.23 .09/.16 Never Married .20/.10* −.14/.10 .08/.09 <High school High School −.23/.12 −.43/.13** −.57/.12*** Junior College −.32/.15* −.61/.16*** −.54/.13*** Bachelor −.31/.13* −.66/.14*** −.57/.12*** Graduate −.18/.18 −.74/.18*** −.68/.14*** Income .01/.01 −.03/.01*** −.03/.01*** R .04*** .07* .00 .11*** .00 .11*** Note. Belief Identity: 0 = Theist; 1 = Atheist; ID = Identity. *p < .05. **p < .01. ***p < .001. The inclusion of demographic covariates in Block 2, as outcome variables were standardized (West, Aiken, & Krull, opposed to Block 1, was done to avoid issues with suppres- 1996). By standardizing the outcome variables, R/S Identifier sion effects. It is possible that by entering covariates in Block coefficients could be interpreted in terms of how many standard 1 and variables of interest in Block 2, variables of interest deviation units groups differed by (e.g., Table 2, Block 1 for may register as significant only because covariates acted as Endogenous Meaning, Belief ID = .64; Atheists were higher suppressor variables. This potential issue is often overlooked, than Theists by an average of .64 SD units). A consequence of likely because there has been a historical interest in ΔR val- this standardization was that the coefficients became close ues. However, we were interested in whether specific R/S approximations (within 2% in Block 1) of Hedges’ g effect size Identifiers predicted meaning in life, which could be deter- index (.20 = small, .50 = medium, .80 = large). The discussion mined by a simple investigation of the relevant coefficients of effect size is critical when using large databases because sta- (i.e., ΔR were not germane to the overall purpose of the tistical significance is a product of power—with sufficient study). Placing R/S Identifiers in Block 1 as opposed to power any difference will become statistically significant. Block 2 does not change outcomes or analysis, but rather was Consequently, if a coefficient was below a “small” effect size done to avoid a potential problem, at no cost to the overall (i.e., B < .20), then researchers noted that the differences may be integrity of the study. statistically different but not practically significant. Using G*Power (v3.1.9.2), we estimated that there was suf- ficient power to detect small-to-medium differences (d = .28) Results between Theists/Atheists (Power = .82); sufficient power to detect small-to-medium differences (d = .23) between Affiliated/ Belief Identity: Theists Versus Atheists Unaffiliated (Power = .83); and sufficient power to detect small- to-medium differences (d = .30) between Raised Affiliated/ The first set of analyses are displayed in Table 2. Endogenous Raised Unaffiliated (Power = .80). Because it was possible for Meaning was regressed onto Belief Identity in Block 1 each regression model to miss small differences (d = .20), the (n = 1,156), R = .038, F(1, 110) = 39.61, p < .001, and Belief effect sizes associated with R/S Identifiers were still discussed Identity was a significant positive predictor, t = 6.29, p < .001, regardless of whether R/S Identifiers were statistically signifi- B = .83, 95% confidence interval (CI) [0.57, 1.09]. Persons cant. To make this discussion of effect sizes transparent, all who were Atheists were more likely to indicate higher levels 6 SAGE Open Figure 1. Theists and atheists for endogenous meaning (with standard error bars). Note. t = 5.71, p < .001, unstandardized Mdiff = 0.74, approximate Hedge’s g = 0.57. of Endogenous Meaning than persons who were Theists. Religious Affiliation: Religiously Affiliated Versus 2 2 Covariates were added in Block 2, ΔR = .036, R = .074, Religiously Unaffiliated F(13, 110) = 2.25, p = .012, but even with their inclusion, The second set of analyses is displayed in Table 3. Endogenous being an Atheist was associated with significantly higher lev- Meaning was regressed onto Religious Affiliation in Block 1 els of the Endogenous Meaning, t = 5.56, p < .001, B = .56, (n = 1,185), R = .023, F(1, 111) = 16.47, p < .001, and was a 95% CI [0.37, 0.77] (see Figure 1). significant positive predictor, t = 4.06, p < .001, B = .41, Fatalism was regressed onto Belief Identity in Block 1 95% CI [0.21, 0.61]. Covariates were entered in Block 2, (n = 1,165), R = .001, F(1, 111) = 0.21, p = .650, but the model 2 2 ΔR = .035, R = .058, F(13, 111) = 2.06, p = .024, and with did not significantly improve. When covariates were added in 2 2 their inclusion, Religious Affiliation continued to be a signifi- Block 2, ΔR = .105, R = .106, F(13, 111) = 7.18, p < .001, the cant predictor, t = 3.10, p = .002, B = .34, 95% CI [0.12, 0.55]. model significantly improved. However, Belief Identity Ultimately, Religious Affiliation continued to positively pre- remained a nonsignificant predictor in the overall model. dict Endogenous Meaning with the inclusion of covariates Nihilism was regressed onto Belief Identity in Block 1 (n = 1,161), R = .003, F(1, 111) = 2.25, p = .137, but the (see Table 3 and Figure 2). model did not substantially improve. Covariates were added Fatalism was regressed onto Religious Affiliation (n = 1,196) 2 2 in Block 2, ΔR = .109, R = .112, F(13, 111) = 4.60, p < .001; in Block 1, R = .003, F(1, 112) = 2.15, p = .145, but being while the model improved, Belief Identity remained a non- Religiously Unaffiliated was not associated with Fatalism. 2 2 significant predictor of Nihilism. Although the coefficient Researchers added covariates in Block 2, ΔR = .103, R = .106, for Belief Identity was not statistically significant, given the F(13, 112) = 7.33, p < .001, but even with these inclusions, coefficient for Nihilism (B = .19) was close to the cut-off Religious Affiliation was again not associated Fatalism. point of B = .20 for practical significance, researchers con- The relationship between Nihilism and Religious ducted an additional analysis to determine there was a genu- Affiliation (n = 1,191) was investigated next. Nihilism was ine relationship between atheism and nihilism. regressed onto Religious Affiliation in Block 1 of the regres- Noting that Atheists scored significantly higher on the sion model, R = .004, F(1, 112) = 2.81, p = .097, but Religious Endogenous Meaning measure than Theists, and that Affiliation was a nonsignificant predictor of Nihilism. 2 2 Endogenous Meaning itself was positively correlated with Covariates were added in Block 2, ΔR = .103, R = .107, Nihilism scores (unweighted r = .21, p < .001), researchers F(13, 112) = 4.70 p < .001; however, Religious Affiliation did explored the relationship between Nihilism and Belief not predict Nihilism. Identity while controlling for Endogenous Meaning. The analyses revealed that once Endogenous Meaning was con- Subgroup analyses. In a series of unplanned post hoc tests, we trolled for, the relationship that Belief Identity had with expanded on the analyses pertaining to Religious Affiliation Nihilism dropped sharply (cf. B = .19, B = .10). These results and investigated whether specific religious categories were suggest that Atheists are not more likely to score differently associated with differences in meaning in life. Unfortunately, than Theists in terms of Nihilism, especially once Endogenous while the 2008 GSS provided information on religious affili- Meaning is controlled for. ation, most of the category options were underpopulated. Speed et al. 7 Table 3. Religious Affiliation Predicting Endogenous Meaning, Fatalism, and Nihilism. B coefficients/robust SE Endogenous meaning Fatalism Nihilism (n = 1,185) (n = 1,196) (n = 1,191) Block 1 Block 2 Block 1 Block 2 Block 1 Block 2 Constant −.07/.05 .25/.22 .02/.05 .53/.23* −.03/.05 .89/.25** Religious Affiliation ID .41/.10*** .34/.11** −.14/.10 −.07/.10 .18/.11 .16/.10 Sex (male/female) −.09/.08 −.01/.07 −.18/.06** Age −.00/.00 .00/.00 −.00/.00 Race (White) Black −.11/.10 .28/.10** −.05/.10 Other .20/.12 .23/.10* .39/.16* Married Widowed .15/.15 −.08/.15 −.01/.11 Divorced −.10/.13 −.21/.11 −.11/.09 Separated .24/.22 −.02/.23 .09/.16 Never Married .22/.10* −.11/.09 .06/.08 <High school High School −.22/.12 −.41/.12** −.54/.11*** Junior College −.30/.14* −.56/.15*** −.51/.13*** Bachelor −.24/.13 −.64/.13*** −.55/.12*** Graduate −.11/.18 −.69/.17*** −.64/.14*** Income .01/.01 −.03/.01*** −.03/.01*** R .02*** .06* .00 .11*** .00 .11*** Note. Religious Affiliation: 0 = Religiously Affiliated; 1 = Religiously Unaffiliated; ID = Identity. *p < .05. **p < .01. ***p < .001. Figure 2. Religiously affiliated and religiously unaffiliated for endogenous meaning (with standard error bars). Note. t = 3.27, p = .001, unstandardized Mdiff = 0.44, approximate Hedge’s g = 0.34. Consequently, only four categories could be compared (Reli- Fatalism, Nihilism, or Endogenous Meaning. Because the giously Unaffiliated, n = 177; Protestant, n = 601; Catholic, comparisons were unplanned, we used Holm–Bonferroni n = 285; and Other, n = 97). corrected p values in interpreting the coefficients. Results We examined whether moving from the Religiously showed no differences across the Fatalism or Nihilism mod- Unaffiliated group to the Protestant group, Catholic group, or els using these corrected values. However, moving from the Other group was associated with differences or changes in Religiously Unaffiliated group to the Protestant group was 8 SAGE Open Table 4. Religious Upbringing Predicting Endogenous Meaning, Fatalism, and Nihilism. B coefficients/robust SE Endogenous meaning Fatalism Nihilism (n = 1,181) (n = 1,191) (n = 1,186) Block 1 Block 2 Block 1 Block 2 Block 1 Block 2 Constant −.01/.05* .37/.22 .01/.04* .50/.23 −.01/.05* .95/.24 Religious Upbringing ID .12/.12 .09/.11 −.12/.14 −.05/.12 .14/.12 .16/.11 Sex (M/F) −.12/.07 .00/.07 −.19/.06 Age .00/.00** .00/.00** .00/.00** Race (White) Black −.12/.10 .27/.10 −.05/.10 Other .20/.11 .21/.10 .39/.16 Married Widowed .17/.15 −.09/.15 .00/.11 Divorced −.10/.13 −.22/.11 −.12/.09 Separated .22/.22 −.02/.23 .08/.16 Never Married .24/.10 −.12/.09 .07/.08 <High school High School −.21/.11 −.41/.13 −.54/.11 Junior College −.32/.14 −.55/.16 −.52/.12 Bachelor −.23/.13 −.62/.13 −.55/.12 Graduate −.09/.18 −.70/.17 −.62/.14 Income .01/.01* −.03/.01** −.03/.01** R .00 .05** .00 .10*** .00 .11*** Note. Religious Upbringing: 0 = Raised Religiously Affiliated; 1 = Raised Religiously Unaffiliated; ID = Identity. *p < .05. **p < .01. ***p < .001. associated with a significant drop in Endogenous Meaning, onto Religious Upbringing in Block 1 of a regression model, t = −3.78, p < .001, B = −.42, 95% CI [−0.64, −0.20]. R = .001, F(1, 113) = 1.29, p = .259, but did not significantly However, whether this was a product of something special improve on the prediction of Fatalism. Covariates were added 2 2 about the Protestant group in relation to the Religiously in Block 2, ΔR = .104, R = .106, F(13, 113) = 4.69, p < .001, Unaffiliated group is unclear. but even with their inclusion, Religious Upbringing remained a nonsignificant predictor of Nihilism. Religious Upbringing: Raised Religiously Affiliated Subgroup analyses. Following up on the Religious Upbring- Versus Raised Religiously Unaffiliated ing analyses, researchers investigated whether being raised The third set of analyses is displayed in Table 4. Endogenous Protestant (n = 656), Catholic (n = 376), or Other (n = 51) Meaning was regressed onto Religious Upbringing in was associated with differing levels of Fatalism, Nihilism, or Block 1 (n = 1,182) of the regression model, R = .001, Endogenous meaning when compared with persons who F(1, 112) = 0.99, p = .321, but Religious Upbringing was a were raised in a Religiously Unaffiliated home (n = 103). We nonsignificant predictor. Covariates were added in Block used Holm–Bonferroni corrections for these comparisons 2 2 2, ΔR = .044, R = .045, F(13, 112) = 2.84, p = .002, but across the three regression models. However, results showed Religious Upbringing remained a nonsignificant predictor no differences across the Fatalism, Nihilism, or Endogenous of Endogenous Meaning (see Table 4). Meaning models. In other words, being raised in a Catholic, Fatalism was regressed onto Religious Upbringing in Protestant, or Other home was not associated with differ- Block 1 (n = 1,193) of the regression model, R = .001, ences when compared with persons raised in a Religiously F(1, 113) = 0.74, p = .391, and Religious Upbringing was a Unaffiliated homes. nonsignificant predictor of Fatalism. With the inclusion of 2 2 covariates in Block 2, ΔR = .104, R = .105, F(13, 113) = Discussion 7.16, p < .001, Religious Upbringing continued to be a non- significant predictor of Fatalism. The present study investigated whether various aspects of We then investigated the relationship between Religious irreligion predicted fatalism, nihilism, and the perception of Upbringing and Nihilism (n = 1,188). Nihilism was regressed life meaning being internally produced. While the existing Speed et al. 9 literature specifies inherent deficits that atheists, the reli- because the notion that atheists do not derive meaning from giously unaffiliated, and persons raised in religiously unaf- god(s) would be tautological. Moreover, given a definition of filiated homes are likely to face regarding meaning in life, meaning where a deity is inherent, what would differences our results did not produce support for such views. In fact, between atheists and theists actually signify? In such a case, these “identity variables” were largely irrelevant to the sta- one group would, by definition, be excluded from experienc- tistical models. However, the analysis confirmed that athe- ing meaning. ists and the religiously unaffiliated were more likely to indicate that life has endogenously produced meaning—one Making one’s own purpose. A consistent and strong finding of the few persistent differences between the groups we that emerged from the present study was related to Endoge- examined. nous Meaning. Being either an atheist or religiously unaffili- ated was associated with stronger agreement that meaning in life came from within, and these differences persisted despite Are the Irreligious More Likely to Be Fatalists and the inclusion of covariates. Not only were these differences Nihilists? statistically significant, but they were also of practical sig- nificance. The observed effect size for atheists in this regard Our present results do not support the idea that atheists or the was g = .57 (a medium-to-large effect), and the effect size for religiously unaffiliated possess a greater sense of fatalism or the religiously unaffiliated was g = .34 (a small-to-medium nihilism. While it is possible to frame the results from the effect). These findings are consistent with two complimen- present study as running against the existing literature, we tary ideas: Theists and the religiously affiliated are more think that the case for atheist or nonreligious deficits in life likely to believe that life has extrinsically produced meaning meaning was never adequately established to begin with. (possibly due to god or gods), and atheists and the religiously This is because most views on this topic rest on academic unaffiliated are more likely to perceive meaning as a product speculation as to the relationship between atheism, religion, of the self rather than a product of an external source or agent and meaning in life, rather than empirical tests. Our findings (or a relationship with this agent). suggest that atheists/theists and the religiously unaffiliated/ Researchers often start with the perception or assumption affiliated do not systematically differ with regard to fatalism that meaning is exogenous, or that exogenous meaning is and nihilism, constructs that are routinely linked with psy- richer or more fulfilling than endogenous meaning (e.g., chological well-being. Overall, these findings would be con- Park & McNamara, 2006). This approach can, unsurpris- sistent with at least a few other studies that find that atheists ingly, lead to conclusions that atheists have impoverished or the religiously unaffiliated do not suffer from psychologi- levels of meaning in life as compared with theists. If mean- cal deficits because of their position (Galen & Kloet, 2011; ing in life is exogenous, then one would expect atheists and, Speed, 2017). to a lesser extent, the religiously unaffiliated, to report greater levels of fatalism and nihilism. However, despite an ade- quately powered analysis using nationally representative Are Assumptions Underpinning Meaning in Life data, these relationships did not emerge. While several rea- Warranted? sons could be provided to explain these null findings, we In light of our findings, it is instructive to investigate why the think that the simplest explanation is that meaning in life is idea persists that irreligion should reflect differences on multidimensional and varies across social categories. On nihilism, fatalism, and source of meaning in life in general. average, atheists (and to a lesser extent the religiously unaf- In our view, the literature often fails to appropriately distin- filiated), in theory, are less likely to accept an exogenous guish between types of meaning, either in its endogenous source of meaning—and were more likely to believe mean- form (meaning perceived to be internally produced) or exog- ing was internally produced. Again, these findings are con- enous form (meaning perceived to be externally produced). sistent with the idea that “whatever makes life meaningful is It is important to note that exogenous meaning does not nec- heavily loaded with whatever people value” (Klinger, 2012, essarily imply the source is god(s). Granted, this may be a p. 29). Atheists in particular are perhaps less likely to have an “traditional” source of exogenous meaning, but it should not externally grounded meaning in life—but not any meaning be thought of as a sole source as persons may perceive any in life (Jörns, 1997). number of external things provide them within meaning Our findings generally align with Wong (1998), who gen- (e.g., the universe, karma, membership in social groups, rela- erated a personal meaning scale, by asking individuals about tionships, occupation). However, if meaning in life is what makes life meaningful and then factor analyzing the assumed to be given or bestowed on persons from a transcen- responses. An overwhelming number of the items were dent source, then this assumption has a substantial impact on related to goal pursuit; the factor of achievement striving investigating an atheistic worldview. Under such an assump- explained over three times the amount of variance explained tion, it would be technically true to suggest that atheists have by the “religion” factor (see Klinger, 2012, p. 29). Religiously less meaning in their lives than theists, but this is only derived meaning may serve as a central source of meaning 10 SAGE Open only insofar as it is personally and socially valued, and such the 2008 GSS was specifically chosen because it contained valuation itself may be part of socialization and learning pro- relevant questions that allowed us to adequately test hypoth- cesses. This illustrates the issues with assuming that com- eses related to meaning, religion, purpose, and fatalism. A mon beliefs or behaviors are generalizable to all. Exogenous related limitation was the number of atheists (n = 115), reli- meaning certainly exists for many, but that does not mean giously unaffiliated (n = 192), and raised religiously unaffili- that endogenous meaning can or should be ignored. As King, ated (n = 96) available in the dataset. While these numbers Heintzelman, and Ward (2016) suggest, “the experience of allowed for adequately powered analyses, it would be bene- meaning in life may be quite a bit more commonplace than is ficial to have a greater number of these groups represented in often portrayed” (p. 211), and we observe that religion is one future research. Although the GSS is a nationally representa- of many paths for attaining it. tive sample, the final sample size was more characteristic of We can also point out that if exogenous meaning is defined psychological studies than of sociological studies at the as meaning that is not self-made, this definition does not macro level. An overall larger sample size would also be intrinsically implicate religion or theism. In this study, we desirable to attain more precise population parameter esti- were specifically concerned with exogenous meaning-mak- mates, allowing us increased confidence in the presence (or ing processes in the context of theism, that is, specifically absence!) of differences between groups. religious meaning-making. The religiously unaffiliated, and It should be noted that focusing on the self as a source of even atheists, can reference or at least perceive a spiritual or meaning bears a specific relation to the kind of culture found transcendent meaning of external origin, though for atheists, in the United States, which is a notably individualistic one. this would very likely depend on how “spirituality” was In this way, we caution that our findings may only be plau- framed or defined. For example, studies suggest that when sible in the context of postindustrial, Westernized countries. atheists are given the opportunity to list sources of profun- This raises the question of whether the self would be a salient dity and meaning, they often list external sources such as source of meaning for nonreligious or atheistic individuals nature, the universe, social causes, and humanity (Caldwell- elsewhere in the world. There is good reason to think that the Harris et al., 2011; Coleman & Arrowood, 2015; Coleman, self would be less prominent an element of meaning-making Silver, & Holcombe, 2013). Future research should explore in other cultures and countries. As such, a cross-cultural per- the psychological consequences of utilizing these external spective on our topic would be a valuable contribution to the sources of meaning. study of meaning systems, identity, and the production of meaning in life. Future research should also seek to delineate between Limitations, Future Directions, and Conclusion types of religious affiliation and religious upbringing to The present study had several limitations. First, we assessed determine whether meaning in life is predicted by this speci- the perception of the extent to which life meaning is inter- ficity. Unfortunately, while the GSS contained a more spe- nally produced, the perception that life was without mean- cific breakdown of religious affiliations, many of the ing, and the perception that one could not act meaningfully categories were too small to allow for adequately-powered with agency. In other words, “life is only meaningful if you analyses. Nevertheless, it is possible that a more refined provide the meaning yourself” is not equivalent to “I have a analysis of these groups, using a different dataset, would high level of self-produced meaning” or other measures of allow for a better picture of differences to emerge. meaning in life. Ideally, we would have assessed actual lev- Overall, the present study found no support for the con- els of self-produced life meaning, which would have allowed tention that atheists and the religiously unaffiliated were for more robust conclusions. However, the questions used in more prone to nihilism or fatalism than their respective coun- the present study demonstrate some conceptual overlap with terparts. While these groups were more likely to indicate that aspects of the Meaning in Life Questionnaire (cf. Steger & meaning in life was endogenous, these differences did not Frazier, 2005) for nihilism (“In my opinion, life does not seem to coincide with a view that life was uncontrollable or serve any purpose”; cf. “My life has no clear purpose”; without meaning. Religious affiliation and/or belief in god(s) Steger & Frazier, 2005). Generally, our point yet remains is certainly a pathway toward meaning in life; however, no that, to avoid classification and measurement issues with potential source for meaning is also a necessary source for nonreligious individuals and populations, future research meaning. should always seek to discriminate between meaning-mak- ing processes that rely on external versus internal sources. To Authors’ Note the extent that such distinctions are obscured or not accounted All authors are members of the Atheist Research Collaborative. for, this may produce misleading conclusions about the impact of meaning in life on other important psychological Acknowledgments outcomes. Second, given that archival data were used, the present The authors would like to thank Luke Galen for helpful comments study was limited to whichever items were asked. However, on an earlier draft of this manuscript. Speed et al. 11 Declaration of Conflicting Interests Caldwell-Harris, C., Wilson, A., LoTempio, E., & Beit-Hallahmi, B. (2011). Exploring the atheist personality: Well-being, awe, The author(s) declared no potential conflicts of interest with respect and magical thinking in atheists, Buddhists, and Christians. to the research, authorship, and/or publication of this article. Mental Health, Religion & Culture, 14, 659-672. doi:10.1080/ 13674676.2010.509847 Funding Coleman, T. J., III, & Arrowood, R. B. (2015). Only we can save The author(s) received no financial support for the research, author- ourselves: An atheist’s “salvation.” In H. Bacon, W. Dossett, & ship, and/or publication of this article. S. Knowles (Eds.), Alternative salvations: Engaging the sacred and the secular (pp. 11-20). London, England: Bloomsbury. Coleman, T. J., III, & Hood, R. W., Jr. (2015). Reconsidering Notes everything: From folk categories to existential theory of mind. 1. Other scientists have disagreed (e.g., Goodenough, 1998); “It [Peer commentary on the journal article “From weird experi- does no harm to the romance of the sunset to know a little bit ences to revelatory events” by A. Taves]. Religion and Society: about it” (Sagan, 1994, p. 80). Advances in Research, 6(1), 18-22. 2. For extensive reviews of the demographic and personality Coleman, T. J., III, Silver, C. F., & Holcombe, J. (2013). Focusing characteristics of the nonreligious and atheists, see Caldwell- on horizontal transcendence: Much more than a “non-belief.” Harris (2012), Galen (2014), Streib and Klein (2013), and Essays in the Philosophy of Humanism, 21, 1-18. doi:10.1558/ Zuckerman (2009). eph.v21i2.1 3. Nihilism is not necessarily equivalent to absence of purpose, in Coleman, T. J., III, Silver, C. F., & Hood, R. W., Jr. (2016). “…If the sense that it can be a broader reference to moral or ontolog- the universe is beautiful, we’re part of that beauty.” – A “neither ical nihilism. Worded as it is, however, the item from the GSS religious nor spiritual” biography as horizontal transcendence. reflects more specifically the concept of “purposelessness.” In H. Streib & R. Hood (Eds.), Semantics and psychology of spirituality (pp. 335-372). Dordrecht, The Netherlands: ORCID iD Springer. doi:10.1007/978-3-319-21245-6_22 Crescioni, A., & Baumeister, R. (2013). The four needs for meaning, David Speed https://orcid.org/0000-0001-7033-2068 the value gap, and how (and whether) society can fill the void. In J. Hicks & C. Routledge (Eds.), The experience of meaning References in life: Classical perspectives, emerging themes, and controver- Ai, A. L., Peterson, C., Rodgers, W. L., & Tice, T. N. (2005). Faith sies (pp. 3-15). Dordrecht, The Netherlands: Springer. factors and internal health locus of control in patients prior to Crosby, D. (1988). The specter of the absurd. Albany: State open-heart surgery. Journal of Health Psychology, 10, 669- University of New York Press. 676. doi:10.1177/1359105305055318 Ellison, C. G., & Burdette, A. M. (2011). Religion and the sense Baggini, J. (2004). What’s it all about? Philosophy and the mean- of control among U.S. adults. Sociology of Religion, 73, 1-22. ing of life. New York, NY: Oxford University Press. doi:10.1093/socrel/srr035 Baker, J. O. B., & Smith, B. (2009). None too simple: Examining Farias, M., Newheiser, A. K., Kahane, G., & de Toledo, Z. (2013). issues of religious nonbelief and nonbelonging in the United Scientific faith: Belief in science increases in the face of States. Journal for the Scientific Study of Religion, 48, 719-733. stress and existential anxiety. Journal of Experimental Social Barrett, J. (2012). Born believers: The science of children’s reli- Psychology, 49, 1210-1213. gious belief. New York, NY: Free Press. Francis, L. J., & Robbins, M. (2004). Belonging without believ- Baumeister, R. (1991). Meanings of life. New York, NY: Guilford ing: A study in the social significance of Anglican identity Press. and implicit religion among 13-15 year-old males. Implicit Beit-Hallahmi, B. (2015). Psychological perspectives on religion Religion, 7, 37-54. and religiosity. East Sussex, UK: Routledge. Frankl, V. (2006). Man’s search for meaning. Boston, MA: Beacon Berger, P. L. (1990). The sacred canopy: Elements of a sociologi- Press. (Original work published 1959) cal theory of religion. Garden City, NY: Anchor Doubleday. Galen, L. W. (2014). Nonreligious and atheist emerging adults. In (Original work published 1967) C. Berry & M. Abo-Zena (Eds.), Emerging adults’ religious- Bering, J. (2002). The existential theory of mind. Review of General ness and spirituality: Meaning-making in an age of transition Psychology, 6(1), 3-24. doi:10.1037/1089-2680.6.1.3 (pp. 237-252). New York, NY: Oxford University Press. Bering, J. (2003). Towards a cognitive theory of existential mean- Galen, L. W. (2015). Atheism, wellbeing, and the wager: Why ing. New Ideas in Psychology, 21, 101-120. doi:10.1016/ not believing in God (with others) is good for you. Science, s0732-118x(03)00014-x Religion & Culture, 23(3), 54-69. doi:10.17582/journal. Blessing, K. (2013). Atheism and the meaningfulness of life. In S. src/2015/2.3.54.69 Bullivant & M. Ruse (Eds.), The Oxford handbook of atheism Galen, L. W. (2017a). Overlapping mental magisteria: Implications (pp. 104-118). Oxford, UK: Oxford University Press. of experimental psychology for a theory of religious belief as Bullivant, S. (2013). Defining atheism. In S. Bullivant & M. Ruse misattribution. Method & Theory in the Study of Religion, 29, (Eds.), The Oxford handbook of atheism (pp. 11-21). Oxford, 221-267. doi:10.1163/15700682-12341393 UK: Oxford University Press. Galen, L. W. (2017b). Which functions assumed to be religious Caldwell-Harris, C. (2012). Understanding atheism/non-belief as and spiritual in nature are ultimately attributable or reducible an expected individual-differences variable. Religion, Brain & to purely secular mechanisms? Religion, Brain & Behavior, 7, Behavior, 2, 4-23. doi:10.1080/2153599x.2012.668395 293-295. doi:10.1080/2153599X.2016.1249919 12 SAGE Open Galen, L. W., & Kloet, J. (2011). Mental well-being in the religious to stressful life events. Psychological Bulletin, 136, 257-301. and the non-religious: Evidence for a curvilinear relationship. doi:10.1037/a0018301 Mental Health, Religion & Culture, 14, 673-689. doi:10.1080/ Park, C. (2013). Religion and meaning. In R. Paloutzian & C. Park 13674676.2010.510829 (Eds.), Handbook of the psychology of religion and spirituality Gervais, W. M., & Najle, M. B. (2015). Learned faith: The influ- (2nd ed., pp. 357-379). New York, NY: Guilford Press. ences of evolved cultural learning mechanisms on belief in Park, C., & McNamara, P. (2006). Religion, meaning, and the Gods. Psychology of Religion and Spirituality, 7, 327-335. brain. In P. McNamara (Ed.), Where God and science meet: Goodenough, U. (1998). The sacred depths of nature. New York, How brain and evolutionary studies alter our understanding of NY: Oxford University Press. religion (pp. 67-89). Westport, CT: Praeger. Greenfield, E. A., Vaillant, G. E., & Marks, N. F. (2009). Do formal Religious Landscape Study. (2016). Pew Research Center’s religious participation and spiritual perceptions have indepen- Religion & Public Life Project. Retrieved from http://www. dent linkages with diverse dimensions of psychological well- pewforum.org/religious-landscape-study/ being? Journal of Health and Social Behavior, 50, 196-212. Sagan, C. (1994). Pale blue dot. New York, NY: Random House. doi:10.1177/002214650905000206 Schieman, S., Pudrovska, T., & Milkie, M. A. (2005). The sense Hayward, R. D., Krause, N., Ironson, G., Hill, P. C., & Emmons, of divine control and the self-concept: A study of race dif- R. (2016). Health and well-being among the non-religious: ferences in late life. Research on Aging, 27, 165-196. Atheists, agnostics, and no preference compared with religious doi:10.1177/0164027504270489 group members. Journal of Religion & Health, 55, 1024-1037. Schnell, T. (2009). The Sources of Meaning and Meaning in Hwang, K., Hammer, J., & Cragun, R. (2011). Extending religion- Life Questionnaire (SoMe): Relations to demographics and health research to secular minorities: Issues and concerns. well-being. The Journal of Positive Psychology, 4, 483-499. Journal of Religion & Health, 50, 608-622. doi:10.1007/ doi:10.1080/17439760903271074 s10943-009-9296-0 Schnell, T., & Keenan, W. (2011). Meaning-making in an atheist Ivtzan, I., Chan, C., Gardner, H., & Prashar, K. (2013). Linking reli- world. Archive for the Psychology of Religion, 33, 55-78. doi:1 gion and spirituality with psychological well-being: Examining 0.1163/157361211x564611 self-actualisation, meaning in life, and personal growth initia- Schumaker, J. (1992). Mental health consequences of irreligion. In tive. Journal of Religion & Health, 52, 915-929. doi:10.1007/ J. Schumaker (Ed.), Religion and mental health (pp. 54-69). s10943-011-9540-2 New York, NY: Oxford University Press. Jörns, K.-P. (1997). Die neuen Gesichter Gottes: Was die Menschen Smith, T. W., Marsden, P. V., Hout, M., & Kim, J. (2013). heute wirklich glauben [What people really believe today: The General Social Surveys, 1972-2012 [machine-readable data New Faces of God]. Munich, Germany: C. H. Beck. file]. Principal Investigator, Tom W. Smith; Co-Principal King, L., Heintzelman, S., & Ward, S. (2016). Beyond the search Investigators, Peter V. Marsden and Michael Hout, NORC ed. for meaning: A contemporary science of the experience of Chicago: National Opinion Research Center, producer, 2005; meaning in life. Current Directions in Psychological Science, Storrs, CT: The Roper Center for Public Opinion Research, 25, 211-216. doi:10.1177/0963721416656354 University of Connecticut, distributor. 1 data file (57,061 logi- Klinger, E. (2012). The search for meaning in evolutionary goal- cal records) and 1 codebook (3,422 pp). theory perspective and its clinical implications. In P. Wong Speed, D. (2017). Unbelievable?! Theistic/epistemological view- (Ed.), The human quest for meaning: Theories, research, and point affects religion-health relationship. Journal of Religion applications (2nd ed., pp. 23-56). New York, NY: Routledge. & Health, 56, 238-257. doi:10.1007/s10943-016-0271-2 Koenig, H. G. (2012). Spirituality and health research: Methods, Speed, D., & Fowler, K. (2017). Empowerment or dependency? The measurements, statistics, and resources. West Conshohocken, religion/religiosity–mastery relationship. Canadian Psychology/ PA: Templeton Foundation Press. Psychologie Canadienne, 58, 333-344. doi:10.1037/cap0000095 Krause, N., & Pargament, K. (2017). Losing my religion: Exploring Speed, D., & Fowler, K. (2017). Good for all? Hardly! Attending the relationship between a decline in faith and a positive affect. church does not benefit religiously unaffiliated. Journal of Applied Research in Quality of Life, 12, 885-901. doi:10.1007/ Religion & Health, 56, 986-1002. doi:10.1007/s10943-016- s11482-016-9495-2 0317-5 Langston, J. (2014). Explaining atheism: Testing Hunter’s Steger, M. (2017). Meaning in life and wellbeing. In M. Slade, L. Durkheimian theory. Interdisciplinary Journal of Research on Oades, & A. Jarden (Eds.), Wellbeing, recovery and mental Religion, 10, Article 10. health. Cambridge, NY: Cambridge University Press. Lee, L. (2012). Research note: Talking about a revolution: Steger, M., & Frazier, P. (2005). Meaning in life: One link in Terminology for the new field of non-religion studies. Journal the chain from religiousness to well-being. Journal of of Contemporary Religion, 27, 129-139. doi:10.1080/1353790 Counseling Psychology, 52, 574-582. doi:10.1037/0022- 3.2012.642742 0167.52.4.574 Morgan, J. (2013). Untangling false assumptions regarding Stinespring, J., & Cragun, R. (2015). Simple Markov model for atheism and health. Zygon, 48, 9-19. doi:10.1111/j.1467- estimating the growth of nonreligion in the United States. 9744.2012.01316.x Science, Religion & Culture, 2(3), 96-103. doi:10.17582/jour- Pargament, K. (1997). The psychology of religion and coping. New nal.src/2015/2.3.96.103 York, NY: Guilford Press. Streib, H., & Hood, R. (2013). Modeling the religious field: Park, C. (2010). Making sense of the meaning literature: An inte- Religion, spirituality, mysticism, and related world views. grative review of meaning making and its effects on adjustment Implicit Religion, 16, 137-155. doi:10.1558/imre.v16i2.133 Speed et al. 13 Streib, H., & Klein, C. (2013). Atheists, agnostics, and apostates. Wong, P. T. P. (1998). Implicit theories of meaningful life and the In K. Pargament (Ed.), APA handbook of psychology, religion development of the personal meaning profile. In P. T. P. Wong and spirituality. Washington, DC: American Psychological & P. S. Fry (Eds.), The human quest for meaning: A handbook Association. of psychological research and clinical applications (pp. 111- Thiessen, J., & Wilkins-Laflamme, S. (2017). Becoming a religious 140). Mahwah, NJ: Lawrence Erlbaum. none: Irreligious socialization and disaffiliation. Journal for the Zika, S., & Chamberlain, K. (1992). On the relation between mean- Scientific Study of Religion, 56, 64-82. doi:10.1111/jssr.12319 ing in life and psychological well-being. British Journal of Tiliouine, H., & Belgoumidi, A. (2009). An exploratory study of Psychology, 83, 133-145. religiosity, meaning in life and subjective wellbeing in Muslim Zuckerman, P. (2009). Atheism, secularity, and well-being: How students from Algeria. Applied Research in Quality of Life, 4, the findings of social science counter negative stereotypes and 109-127. doi:10.1007/s11482-009-9076-8 assumptions. Sociology Compass, 3, 949-971. doi:10.1111/ Twenge, J., Exline, J., Grubbs, J., Sastry, R., & Campbell, W. j.1751-9020.2009.00247.x (2015). Generational and time period differences in American adolescents’ religious orientation, 1966–2014. PLoS ONE, Author Biographies 10(5), e0121454. doi:10.1371/journal.pone.0121454 David Speed completed graduate school at Memorial University of Vail, K., Rothschild, Z., Weise, D., Solomon, S., Pyszczynski, T., Newfoundland and is an assistant professor in psychology at the & Greenberg, J. (2010). A terror management analysis of the University of New Brunswick in Saint John. His primary research psychological functions of religion. Personality and Social interest is in the health of atheists, Nones, and the non-spiritual. Psychology Review, 14, 84-94. doi:10.1177/1088868309351165 West, S. G., Aiken, L. S., & Krull, J. L. (1996). Experimental Thomas Coleman III is a PhD student at Coventry University as a personality designs: Analyzing categorical by continu- part of the Brain, Belief, and Behaviour research group.his research ous variable interactions. Journal of Personality, 64, 1-48. interests include theory of mind and folk psychology in nonreli- doi:10.1111/j.1467-6494.1996.tb00813.x gious populations. Wilkinson, P., & Coleman, P. (2010). Strong beliefs and coping Joseph Langston is an independent academic researcher who stud- in old age: A case-based comparison of atheism and reli- ies the origins of both atheism and religion. He is also a former gious faith. Ageing & Society, 30, 337-361. doi:10.1017/ college lecturer in the sociology of religion. s0144686x09990353 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png SAGE Open SAGE

What Do You Mean, “What Does It All Mean?” Atheism, Nonreligion, and Life Meaning:

SAGE Open , Volume 8 (1): 1 – Jan 20, 2018

Loading next page...
 
/lp/sage/what-do-you-mean-what-does-it-all-mean-atheism-nonreligion-and-life-aL2WqjH20l
Publisher
SAGE
Copyright
Copyright © 2022 by SAGE Publications Inc, unless otherwise noted. Manuscript content on this site is licensed under Creative Commons Licenses.
ISSN
2158-2440
eISSN
2158-2440
DOI
10.1177/2158244017754238
Publisher site
See Article on Publisher Site

Abstract

Nonreligion is often thought to be commensurate with nihilism or fatalism, resulting in the perception that the nonreligious have no source of meaning in life. While views to this effect have been advanced in various arenas, no empirical evaluation of such a view has been conducted. Using data from the 2008 American General Social Survey (N = ~1,200), we investigated whether atheists, the religiously unaffiliated, and persons raised religiously unaffiliated were more likely than theists, the religiously affiliated, and persons raised with a religious affiliation to report greater levels of fatalism, nihilism, and the perception that meaning in life is self-provided. Results suggested that these groups did not differ with regard to fatalism or nihilism. However, atheists and the religiously unaffiliated (but not persons raised in a religiously unaffiliated household) were more likely to indicate that meaning in life was endogenous—that is, self-produced. While atheists and the nonreligious differed from their counterparts on source of meaning in life, this was not associated with any “penalty” for overall existential meaning. Keywords atheism, meaning in life, nonreligion, nihilism, fatalism, nonbelief A core component of human psychological functioning is the justification for goals and purposes, a sense of control, and construction of meaning and purpose in life, a process that is self-worth. Schnell (2009) identified twenty-six sources of subject to self-perceptions and situational effects (Baggini, meaning in life, ranging from aspects of community and 2004; Baumeister, 1991; Berger, 1967/1990; Park, 2010). togetherness, to nature, generativity, religion, and self- However, for atheists and other nonreligious individuals, it is knowledge. In a nuanced approach, Park (2010, 2013) has sometimes assumed that being without god(s) is the equiva- drawn a distinction between global meaning and situational lent of being bereft of meaning or purpose in life (cf. meaning. Global meaning refers to abstract conceptual sys- Blessing, 2013). This assumption is somewhat odd, given tems and schema such as “religious beliefs . . . fairness, con- that Bering (2002, 2003) suggests humans are evolutionarily trol, coherence, benevolence of the world and other people” hardwired with an existential theory of mind, which is a gen- (Park, 2013, p. 361). Situational meaning refers to ascrip- eral cognitive mechanism compelling humans to find reli- tions of meaning or purpose in specific encounters; the gious or philosophical meaning or purpose in life events meaning or purpose a person ascribes to an event depends on (Coleman & Hood, 2015). Similarly, sociological approaches her or his preconceptions and experiences. to meaning conceive it as a compulsion “. . . to impose a Because meaning and purpose in life can be framed in meaningful order upon reality,” all the while emphasizing numerous ways and are derived from multiple sources and the creation and sustainment of meaning as an inherently circumstances, one could reasonably expect that some social, cultural, and discursive practice (Berger, 1967/1990, sources or paths substantially differ from others. Park and p. 22). In essence, the impetus to construct meaning or pur- pose in life is a quintessential consequence of being human, University of New Brunswick, Saint John, NB, Canada rather than something wholly under the purview of a specific Newfoundland & Labrador Centre for Applied Health Research, religious or philosophical framework. Saint John’s, NL, Canada In this vein, there is no shortage of views on what life 3 Coventry University, UK meaning is, how it “works,” or how we should conceive of it. U.S. Air Force Academy, Colorado Springs, CO, USA Frankl (1959/2006) argued that meaning in life was some- Corresponding Author: thing that each individual seeks and constructs for them- David Speed, Department of Psychology, University of New Brunswick, selves, whereas Baumeister (1991) argued that life meaning 100 Tucker Park Rd., Saint John, New Brunswick, Canada, E2L 4L5. is comprised of four elements: a goal or purpose, values and Emails: David.Speed@med.mun.ca; dspeed@unb.ca Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License (http://www.creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). 2 SAGE Open McNamara (2006) note that religious systems provide pur- deficits (e.g., Barrett, 2012), and popular perceptions remain pose or meaning to individuals. In fact, this is often seen as a that they are likely to be more nihilistic or have viewpoints con- primary function of religion. Other researchers have echoed sistent with fatalism (cf. Blessing, 2013). Other research sug- these sentiments and explicitly recognized that religious sys- gests that nonreligiosity “divest[s] people of certain age-old tems provide strong and coherent sources for meaning in life, pathways to psychological health” (Schumaker, 1992, p. 65). due to the fact that these systems appeal to an “ultimate” Additionally, research shows consistent links between source (Crescioni & Baumeister, 2013; also see Vail et al., meaning in life and psychological well-being (e.g., Krause & 2010). Park and McNamara (2006) suggest that not only are Pargament, 2017; Schnell, 2009; Zika & Chamberlain, 1992; “religious frameworks comprehensive, but they tend to be for a review, see Steger, 2017), as well as meaning in life and much more ‘existentially satisfactory’ than secular explana- religiousness or religiosity (e.g., Ivtzan, Chan, Gardner, & tions such as the cold, hard objectivity of science” (p. 67; Prashar, 2013; Steger & Frazier, 2005; Tiliouine & Belgoumidi, also see Farias, Newheiser, Kahane, & de Toledo, 2013; 2009). These findings align with Schumaker (1992), who Schumaker, 1992, on belief in science in the face of stress found that accounting for the relationship between meaning in and existential anxiety). life and well-being substantially reduces the salutary effects of religiosity (Schumaker, 1992). In other words, some of the benefits associated with religiosity are due to its positive rela- Nonreligion and Meaning in Life tionship with producing meaning in life. There are some major drawbacks of this aforementioned However, other research has shown that the irreligious do research. One is an inability to separate the functioning of not experience deficits in meaning in life (Caldwell-Harris purely secular psychological mechanisms from any specific et al., 2011; Wilkinson & Coleman, 2010), happiness (Speed, religious/spiritual processes (Galen, 2017a, 2017b). This 2017; Speed & Fowler, 2017), or well-being (Galen, 2015; inability may be related to research suggesting that religious/ Streib & Klein, 2013), and that the irreligious do not differ in spiritual activities do not have intrinsic benefits, but rather terms of psychological well-being (Galen, 2015; Streib & benefits that are contingent on an individual’s valuation of Klein, 2013; although cf. Hayward, Krause, Ironson, Hill, & those activities (Speed, 2017; Speed & Fowler, 2017). This Emmons, 2016). Essentially, the literature is incongruous: accords with the simple view that “whatever makes life mean- irreligious persons are ostensibly disadvantaged for psycho- ingful is heavily loaded with whatever people value” (Klinger, logical well-being, but do not seem to report poorer psycho- 2012, p. 29). In other words, some persons may find that with- logical well-being. out a framework centered on the divine or transcendent, they This dissonance within the research is problematic for a are more likely to adopt a nihilistic or fatalistic perspective number of reasons. More than 20% of the American popula- (i.e., a negation of [life] meaning; Crosby, 1988). However, tion identifies as nonreligious (Religious Landscape Study, the conclusion that god(s) or religious frameworks are neces- 2016), and some projections suggest this number could climb sary for meaning is predicated on the idea that religion or to almost half of the U.S. population by the year 2042 (see spirituality intrinsically promote meaning for everyone, which Stinespring & Cragun, 2015, who estimate between 26% and can be read to imply that meaning cannot be internally derived 47%). Furthermore, in adolescents and young adults, we find or generated. The perception that an individual is forced to a generational difference, which suggests children are sig- accept nihilism or fatalism in their worldview because they nificantly less religious than their parents (Thiessen & lack a religious or spiritual schema with which to interpret the Wilkins-Laflamme, 2017; Twenge, Exline, Grubbs, Sastry, world is unsupported by the existing literature (Caldwell- & Campbell, 2015), and have relationships with religion Harris, Wilson, LoTempio, & Beit-Hallahmi, 2011; Coleman ranging from explicit repudiation to quiet apathy (Lee, 2012; & Arrowood, 2015; Langston, 2014). Furthermore, the logic see also Francis & Robbins, 2004). of why religious or spiritual frameworks would promote The relationship between youth and religion can also be meaning in life is highly selective. It could be argued that a confusing when we consider whether a religious home person who surrenders to god(s)’ perceived will would accept environment confers a health benefit (i.e., meaning in life, psy- that there is one path to follow or that there is no meaning chological well-being) or whether an unaffiliated home envi- other than serving god(s). Instead, most discussions surround- ronment confers a health penalty. Research suggests that the ing this topic tend to assume as a default that relatively higher home environment can predict future religious/spiritual identi- religion or spirituality, however measured, implies greater ties (Baker & Smith, 2009; Beit-Hallahmi, 2015; Gervais & meaning (e.g., Crescioni & Baumeister, 2013; Pargament, Najle, 2015), but the health consequences of said home envi- 1997; Park & McNamara, 2006). ronment remain unclear. Studies comparing secular sources The consequences of atheism or being religiously unaffili- and levels of meaning, values, and purpose with religious ated for meaning in life, and for psychological well-being more sources are virtually nonexistent (Koenig, 2012), as are the broadly, are unknown or at the very least not held to be static or consequences of nonreligiosity, a nonreligious upbringing, and linear. A cursory examination of the literature would show that atheism on mental well-being (Galen & Kloet, 2011; Hwang, atheists are characterized as having various psychological Hammer, & Cragun, 2011; Morgan, 2013). Speed et al. 3 The existing literature on the relationship between theo- nature of the data. The exact number of participants fluctu- ated slightly from one analyses to the next, but was approxi- logical beliefs, atheism, and fatalism allows us to broadly mately N = 1,200 (see Table 1 for descriptive statistics). suggest that fatalism entails an acceptance of what is per- ceived as an inevitable or uncontrollable outcome, rather than a lack of meaning in life or a lack of purpose. However, Measures fatalism is negatively related to environmental mastery, Meaning in life. Endogenous Meaning, Fatalism, and Nihilism itself a component of psychological well-being (Greenfield, were measured using a 5-point Likert-type scale (i.e., 1 = Vaillant, & Marks, 2009). Conceptually, a person who strongly disagree, 2 = disagree, etc.), with higher scores indi- would say that there is little that can be done to change the cating greater levels of agreement. The 2008 GSS had an item circumstances of his or her life (a person low on a sense of we described as Endogenous Meaning (“Life is only mean- mastery or internal locus of control) is arguably less likely ingful if you provide the meaning yourself”), and was con- to believe that life, the universe, and personal relationships ceptually similar to other questions dealing with whether life are meaningful in any global or time-invariant sense. meaning was internally constructed (e.g., Coleman, Silver, & While some research shows that mastery is positively cor- Hood, 2016; Goodenough, 1998; Schnell & Keenan, 2011; related with religious attendance and/or religiosity (Ai, Streib & Hood, 2013). The 2008 GSS had two items as proxy Peterson, Rodgers, & Tice, 2005; Ellison & Burdette, measures for meaning in life: Fatalism (“There is little people 2011; Schieman, Pudrovska, & Milkie, 2005), other can do to change the course of their lives”) and Nihilism (“In research finds that this relationship is inconsistent my opinion, life does not serve any purpose”). Although the (Greenfield et al., 2009; Speed & Fowler, 2017). Thus, 2008 GSS had an item that assessed God Meaning to some while there is some evidence that theists and the religiously extent (“To me, life is meaningful only because God exists”), affiliated would be more likely to report greater mastery, we did not consider using it because this question would there is no persuasive evidence that atheists, the nonreli- make little conceptual sense with the definition of atheism gious, and those with a nonreligious upbringing are predis- used in the present study. Conceptually, Endogenous Mean- posed toward fatalism (see also Langston, 2014, who finds ing provides information as to the perceived source of mean- that higher external locus of control scores reflect higher ing in life, while Nihilism and Fatalism should be inversely odds of being a theist). related to whether life has a personalized sense of meaning. The present study investigates the relationship between an internal source of meaning, fatalism, and nihilism in a Religious/spiritual identifiers. We made use of three R/S Identi- nationally representative American sample in a series of nine fiers as they related to meaning in life: (a) Belief Identity, (b) planned analyses. More specifically, we sought to investigate Religious Affiliation, (c) and Religious Upbringing. Belief whether belief in god(s), religious affiliation, or religious Identity was based on the GSS item, “Which best describes upbringing were significant predictors of an internal source your beliefs about Gods?” Response options included (a) I of meaning, fatalism, and nihilism. don’t believe in God now, and I never have; (b) I don’t believe in God now, but I used to; (c) I believe in God now, Method but I didn’t used to; and (d) I believe in God now, and I always have. Persons indicating that they believe in God Participants now were labeled as “Theist,” and persons indicating that We accessed data from the 2008 General Social Survey they did not believe in God now were labeled as “Atheist.” (GSS), which was collected by the National Opinion Belief Identity was coded, Theist = 0 and Atheist = 1. While Research Center (NORC) at the University of Chicago the definitions provided for atheism are occasionally con- (Smith, Marsden, Hout, & Kim, 2013). These data are freely flicting (Hwang et al., 2011), the definition used by the pres- accessible from NORC’s website, which also provides a ent study is consistent with negative atheism (see Bullivant, copy of the questionnaire and the user manual. Because this 2013), which is the most inclusive definition of nonbelief. was secondary data analysis, we did not require the Religious Affiliation was based on the item, “What is Institutional Review Board (IRB) clearance from any institu- your religious preference?” followed by a list of different tions to conduct the present study. The 2008 research year religions. Persons who indicated “None” were labeled as was chosen because it is the most recent year that contained “Religiously Unaffiliated” and persons who indicated any- all variables of interest to our research question. To be thing other than “None” were labeled as “Religiously included in the present study, respondents had to have Affiliated.” Religious Affiliation was coded as Religiously answered all covariate questions, at least one of the outcome Affiliated = 0 and Religiously Unaffiliated = 1. variables, and at least one of the religious/spiritual (R/S) Religious Upbringing was based on the question, “In what identifier variables. Persons answering questions with “I religion were you raised?” followed by a list of different reli- don’t know” or persons who answered by refusing to respond gions. Persons who indicated “None” were labeled Raised were excluded from the analyses to maintain the continuous Religiously Unaffiliated, and persons who indicated anything 4 SAGE Open Table 1. Weighted Descriptive Statistics by Religious/Spiritual Identifiers. M (SD) or percentages within categories Belief identity Religious affiliation Religious upbringing Theist Atheist Affiliated Unaffiliated Affiliated Unaffiliated N = 1,034 115 986 192 1,078 96 Fatalism 1.95/1.01 1.91/1.09 1.97/1.03 1.83/0.92 1.94/1.01 1.94/1.03 Nihilism 1.53/0.73 1.67/0.76 1.53/0.72 1.67/0.79 1.55/0.74 1.61/0.66 Endo. Mean 3.08/1.29 3.95/1.04 3.08/1.30 3.64/1.13 3.16/1.30 3.30/1.24 Age 45.98/16.33 40.01/15.37 46.49/16.39 39.2/14.24 45.73/16.32 40.78/15.05 Income 5.57/4.47 6.94/5.59 5.59/4.51 6.36/5.15 5.71/4.6 5.76/4.93 Sex Female 54.37% 24.34% 54.25% 33.60% 51.62% 42.69% Male 45.63% 75.66% 45.75% 66.40% 48.38% 57.31% Race White 75.97% 82.74% 76.36% 78.15% 75.99% 85.73% Black 14.85% 4.95% 13.92% 11.06% 14.29% 4.88% Other 9.17% 12.31% 9.72% 10.79% 9.72% 9.38% Education <High School 15.58% 6.81% 15.11% 12.66% 14.95% 12.38% High School 49.33% 45.93% 48.05% 52.43% 48.68% 49.53% Junior College 9.88% 1.23% 10.09% 4.11% 8.81% 10.89% Bachelor 16.90% 29.24% 17.87% 19.11% 17.79% 22.76% Graduate 8.31% 16.78% 8.89% 11.68% 9.78% 4.43% Marital status Married 56.54% 43.64% 56.41% 46.41% 55.83% 45.11% Widowed 5.58% 1.70% 5.71% 1.84% 5.36% 1.38% Divorced 11.12% 8.30% 11.18% 9.89% 11.00% 11.77% Separated 3.36% 0.70% 3.41% 0.93% 3.14% 1.75% Never Married 23.39% 45.65% 23.28% 40.93% 24.68% 39.98% Note. Only persons who answered all questions of interest are included in this table. Endo. Mean = Endogenous Meaning. other than “None” were labeled Raised Religiously Affiliated. and a weighting correction for nonresponse rate. To achieve Religious Upbringing was coded as Raised Religiously this weighting, Stata’s survey analysis module was used to Affiliated = 0 and Raised Religiously Unaffiliated = 1. specify the weighting parameters by using the syntax pro- vided by Smith and colleagues (2013). In situations where a Covariates. The present study controlled for sex (male/ stratum only had a single respondent, we used a scaled female), age, race (White, Black, Other; White served as approach to address issues with determining variance. the omitted category), marital status (married, widowed, Hierarchical linear regression was used in all analyses. divorced, separated, never married; married served as the Because we used Stata’s survey analysis module, denomina- omitted category), education (less than high school, high tor degrees of freedom in all regression models were based school, some postsecondary, postsecondary, graduate on the number of strata rather than the number of respon- degree; less than high school served as the omitted cate- dents. To meet the underlying assumption of homogeneity of gory), and income in constant dollars (e.g., 1 = US$10,000; variance, heteroscedastic consistent errors (HC1) were 2 = US$20,000, etc.). Please note that “sex” was used in employed. Multicollinearity was not an issue, as the mean lieu of “gender” as the GSS had asked a dichotomous male/ variance inflation factor did not exceed 1.13 for any model. female question. Nine hierarchical linear regression models were used to assess all three R/S Identifiers (belief, affiliation, upbring- ing) and three outcome variables (fatalism, nihilism, endog- Procedure enous meaning). Each model followed the same pattern. All data analysis was done with Stata 13. The 2008 GSS used a complex sampling methodology to achieve a representative Block 1: R/S Identifiers were added (Belief Identity, sample. Because of this, data were weighted to correspond to Religious Affiliation, and Religious Upbringing). which strata respondents were in, the primary sampling unit, Block 2: Covariates were added. Speed et al. 5 Table 2. Belief Identity Predicting Endogenous Meaning, Fatalism, and Nihilism. B coefficients/robust SE Endogenous meaning Fatalism Nihilism (n = 1,156) (n = 1,165) (n = 1,161) Block 1 Block 2 Block 1 Block 2 Block 1 Block 2 Constant −.07/.05 .25/.22 .01/.04 .51/.24* −.02/.05 .87/.25** Belief ID .64/.10*** .57/.10*** −.07/.15 .11/.14 .17/.11 .19/.12 Sex (M/F) −.05/.08 .00/.07 −.15/.06* Age −.00/.00 .00/.00 −.00/.00 Race (White) Black −.08/.10 .28/.10** −.04/.10 Other .22/.12 .18/.11 .42/.17* Married Widowed .16/.15 −.10/.15 −.01/.11 Divorced −.10/.13 −.22/.11 −.13/.09 Separated .22/.22 −.03/.23 .09/.16 Never Married .20/.10* −.14/.10 .08/.09 <High school High School −.23/.12 −.43/.13** −.57/.12*** Junior College −.32/.15* −.61/.16*** −.54/.13*** Bachelor −.31/.13* −.66/.14*** −.57/.12*** Graduate −.18/.18 −.74/.18*** −.68/.14*** Income .01/.01 −.03/.01*** −.03/.01*** R .04*** .07* .00 .11*** .00 .11*** Note. Belief Identity: 0 = Theist; 1 = Atheist; ID = Identity. *p < .05. **p < .01. ***p < .001. The inclusion of demographic covariates in Block 2, as outcome variables were standardized (West, Aiken, & Krull, opposed to Block 1, was done to avoid issues with suppres- 1996). By standardizing the outcome variables, R/S Identifier sion effects. It is possible that by entering covariates in Block coefficients could be interpreted in terms of how many standard 1 and variables of interest in Block 2, variables of interest deviation units groups differed by (e.g., Table 2, Block 1 for may register as significant only because covariates acted as Endogenous Meaning, Belief ID = .64; Atheists were higher suppressor variables. This potential issue is often overlooked, than Theists by an average of .64 SD units). A consequence of likely because there has been a historical interest in ΔR val- this standardization was that the coefficients became close ues. However, we were interested in whether specific R/S approximations (within 2% in Block 1) of Hedges’ g effect size Identifiers predicted meaning in life, which could be deter- index (.20 = small, .50 = medium, .80 = large). The discussion mined by a simple investigation of the relevant coefficients of effect size is critical when using large databases because sta- (i.e., ΔR were not germane to the overall purpose of the tistical significance is a product of power—with sufficient study). Placing R/S Identifiers in Block 1 as opposed to power any difference will become statistically significant. Block 2 does not change outcomes or analysis, but rather was Consequently, if a coefficient was below a “small” effect size done to avoid a potential problem, at no cost to the overall (i.e., B < .20), then researchers noted that the differences may be integrity of the study. statistically different but not practically significant. Using G*Power (v3.1.9.2), we estimated that there was suf- ficient power to detect small-to-medium differences (d = .28) Results between Theists/Atheists (Power = .82); sufficient power to detect small-to-medium differences (d = .23) between Affiliated/ Belief Identity: Theists Versus Atheists Unaffiliated (Power = .83); and sufficient power to detect small- to-medium differences (d = .30) between Raised Affiliated/ The first set of analyses are displayed in Table 2. Endogenous Raised Unaffiliated (Power = .80). Because it was possible for Meaning was regressed onto Belief Identity in Block 1 each regression model to miss small differences (d = .20), the (n = 1,156), R = .038, F(1, 110) = 39.61, p < .001, and Belief effect sizes associated with R/S Identifiers were still discussed Identity was a significant positive predictor, t = 6.29, p < .001, regardless of whether R/S Identifiers were statistically signifi- B = .83, 95% confidence interval (CI) [0.57, 1.09]. Persons cant. To make this discussion of effect sizes transparent, all who were Atheists were more likely to indicate higher levels 6 SAGE Open Figure 1. Theists and atheists for endogenous meaning (with standard error bars). Note. t = 5.71, p < .001, unstandardized Mdiff = 0.74, approximate Hedge’s g = 0.57. of Endogenous Meaning than persons who were Theists. Religious Affiliation: Religiously Affiliated Versus 2 2 Covariates were added in Block 2, ΔR = .036, R = .074, Religiously Unaffiliated F(13, 110) = 2.25, p = .012, but even with their inclusion, The second set of analyses is displayed in Table 3. Endogenous being an Atheist was associated with significantly higher lev- Meaning was regressed onto Religious Affiliation in Block 1 els of the Endogenous Meaning, t = 5.56, p < .001, B = .56, (n = 1,185), R = .023, F(1, 111) = 16.47, p < .001, and was a 95% CI [0.37, 0.77] (see Figure 1). significant positive predictor, t = 4.06, p < .001, B = .41, Fatalism was regressed onto Belief Identity in Block 1 95% CI [0.21, 0.61]. Covariates were entered in Block 2, (n = 1,165), R = .001, F(1, 111) = 0.21, p = .650, but the model 2 2 ΔR = .035, R = .058, F(13, 111) = 2.06, p = .024, and with did not significantly improve. When covariates were added in 2 2 their inclusion, Religious Affiliation continued to be a signifi- Block 2, ΔR = .105, R = .106, F(13, 111) = 7.18, p < .001, the cant predictor, t = 3.10, p = .002, B = .34, 95% CI [0.12, 0.55]. model significantly improved. However, Belief Identity Ultimately, Religious Affiliation continued to positively pre- remained a nonsignificant predictor in the overall model. dict Endogenous Meaning with the inclusion of covariates Nihilism was regressed onto Belief Identity in Block 1 (n = 1,161), R = .003, F(1, 111) = 2.25, p = .137, but the (see Table 3 and Figure 2). model did not substantially improve. Covariates were added Fatalism was regressed onto Religious Affiliation (n = 1,196) 2 2 in Block 2, ΔR = .109, R = .112, F(13, 111) = 4.60, p < .001; in Block 1, R = .003, F(1, 112) = 2.15, p = .145, but being while the model improved, Belief Identity remained a non- Religiously Unaffiliated was not associated with Fatalism. 2 2 significant predictor of Nihilism. Although the coefficient Researchers added covariates in Block 2, ΔR = .103, R = .106, for Belief Identity was not statistically significant, given the F(13, 112) = 7.33, p < .001, but even with these inclusions, coefficient for Nihilism (B = .19) was close to the cut-off Religious Affiliation was again not associated Fatalism. point of B = .20 for practical significance, researchers con- The relationship between Nihilism and Religious ducted an additional analysis to determine there was a genu- Affiliation (n = 1,191) was investigated next. Nihilism was ine relationship between atheism and nihilism. regressed onto Religious Affiliation in Block 1 of the regres- Noting that Atheists scored significantly higher on the sion model, R = .004, F(1, 112) = 2.81, p = .097, but Religious Endogenous Meaning measure than Theists, and that Affiliation was a nonsignificant predictor of Nihilism. 2 2 Endogenous Meaning itself was positively correlated with Covariates were added in Block 2, ΔR = .103, R = .107, Nihilism scores (unweighted r = .21, p < .001), researchers F(13, 112) = 4.70 p < .001; however, Religious Affiliation did explored the relationship between Nihilism and Belief not predict Nihilism. Identity while controlling for Endogenous Meaning. The analyses revealed that once Endogenous Meaning was con- Subgroup analyses. In a series of unplanned post hoc tests, we trolled for, the relationship that Belief Identity had with expanded on the analyses pertaining to Religious Affiliation Nihilism dropped sharply (cf. B = .19, B = .10). These results and investigated whether specific religious categories were suggest that Atheists are not more likely to score differently associated with differences in meaning in life. Unfortunately, than Theists in terms of Nihilism, especially once Endogenous while the 2008 GSS provided information on religious affili- Meaning is controlled for. ation, most of the category options were underpopulated. Speed et al. 7 Table 3. Religious Affiliation Predicting Endogenous Meaning, Fatalism, and Nihilism. B coefficients/robust SE Endogenous meaning Fatalism Nihilism (n = 1,185) (n = 1,196) (n = 1,191) Block 1 Block 2 Block 1 Block 2 Block 1 Block 2 Constant −.07/.05 .25/.22 .02/.05 .53/.23* −.03/.05 .89/.25** Religious Affiliation ID .41/.10*** .34/.11** −.14/.10 −.07/.10 .18/.11 .16/.10 Sex (male/female) −.09/.08 −.01/.07 −.18/.06** Age −.00/.00 .00/.00 −.00/.00 Race (White) Black −.11/.10 .28/.10** −.05/.10 Other .20/.12 .23/.10* .39/.16* Married Widowed .15/.15 −.08/.15 −.01/.11 Divorced −.10/.13 −.21/.11 −.11/.09 Separated .24/.22 −.02/.23 .09/.16 Never Married .22/.10* −.11/.09 .06/.08 <High school High School −.22/.12 −.41/.12** −.54/.11*** Junior College −.30/.14* −.56/.15*** −.51/.13*** Bachelor −.24/.13 −.64/.13*** −.55/.12*** Graduate −.11/.18 −.69/.17*** −.64/.14*** Income .01/.01 −.03/.01*** −.03/.01*** R .02*** .06* .00 .11*** .00 .11*** Note. Religious Affiliation: 0 = Religiously Affiliated; 1 = Religiously Unaffiliated; ID = Identity. *p < .05. **p < .01. ***p < .001. Figure 2. Religiously affiliated and religiously unaffiliated for endogenous meaning (with standard error bars). Note. t = 3.27, p = .001, unstandardized Mdiff = 0.44, approximate Hedge’s g = 0.34. Consequently, only four categories could be compared (Reli- Fatalism, Nihilism, or Endogenous Meaning. Because the giously Unaffiliated, n = 177; Protestant, n = 601; Catholic, comparisons were unplanned, we used Holm–Bonferroni n = 285; and Other, n = 97). corrected p values in interpreting the coefficients. Results We examined whether moving from the Religiously showed no differences across the Fatalism or Nihilism mod- Unaffiliated group to the Protestant group, Catholic group, or els using these corrected values. However, moving from the Other group was associated with differences or changes in Religiously Unaffiliated group to the Protestant group was 8 SAGE Open Table 4. Religious Upbringing Predicting Endogenous Meaning, Fatalism, and Nihilism. B coefficients/robust SE Endogenous meaning Fatalism Nihilism (n = 1,181) (n = 1,191) (n = 1,186) Block 1 Block 2 Block 1 Block 2 Block 1 Block 2 Constant −.01/.05* .37/.22 .01/.04* .50/.23 −.01/.05* .95/.24 Religious Upbringing ID .12/.12 .09/.11 −.12/.14 −.05/.12 .14/.12 .16/.11 Sex (M/F) −.12/.07 .00/.07 −.19/.06 Age .00/.00** .00/.00** .00/.00** Race (White) Black −.12/.10 .27/.10 −.05/.10 Other .20/.11 .21/.10 .39/.16 Married Widowed .17/.15 −.09/.15 .00/.11 Divorced −.10/.13 −.22/.11 −.12/.09 Separated .22/.22 −.02/.23 .08/.16 Never Married .24/.10 −.12/.09 .07/.08 <High school High School −.21/.11 −.41/.13 −.54/.11 Junior College −.32/.14 −.55/.16 −.52/.12 Bachelor −.23/.13 −.62/.13 −.55/.12 Graduate −.09/.18 −.70/.17 −.62/.14 Income .01/.01* −.03/.01** −.03/.01** R .00 .05** .00 .10*** .00 .11*** Note. Religious Upbringing: 0 = Raised Religiously Affiliated; 1 = Raised Religiously Unaffiliated; ID = Identity. *p < .05. **p < .01. ***p < .001. associated with a significant drop in Endogenous Meaning, onto Religious Upbringing in Block 1 of a regression model, t = −3.78, p < .001, B = −.42, 95% CI [−0.64, −0.20]. R = .001, F(1, 113) = 1.29, p = .259, but did not significantly However, whether this was a product of something special improve on the prediction of Fatalism. Covariates were added 2 2 about the Protestant group in relation to the Religiously in Block 2, ΔR = .104, R = .106, F(13, 113) = 4.69, p < .001, Unaffiliated group is unclear. but even with their inclusion, Religious Upbringing remained a nonsignificant predictor of Nihilism. Religious Upbringing: Raised Religiously Affiliated Subgroup analyses. Following up on the Religious Upbring- Versus Raised Religiously Unaffiliated ing analyses, researchers investigated whether being raised The third set of analyses is displayed in Table 4. Endogenous Protestant (n = 656), Catholic (n = 376), or Other (n = 51) Meaning was regressed onto Religious Upbringing in was associated with differing levels of Fatalism, Nihilism, or Block 1 (n = 1,182) of the regression model, R = .001, Endogenous meaning when compared with persons who F(1, 112) = 0.99, p = .321, but Religious Upbringing was a were raised in a Religiously Unaffiliated home (n = 103). We nonsignificant predictor. Covariates were added in Block used Holm–Bonferroni corrections for these comparisons 2 2 2, ΔR = .044, R = .045, F(13, 112) = 2.84, p = .002, but across the three regression models. However, results showed Religious Upbringing remained a nonsignificant predictor no differences across the Fatalism, Nihilism, or Endogenous of Endogenous Meaning (see Table 4). Meaning models. In other words, being raised in a Catholic, Fatalism was regressed onto Religious Upbringing in Protestant, or Other home was not associated with differ- Block 1 (n = 1,193) of the regression model, R = .001, ences when compared with persons raised in a Religiously F(1, 113) = 0.74, p = .391, and Religious Upbringing was a Unaffiliated homes. nonsignificant predictor of Fatalism. With the inclusion of 2 2 covariates in Block 2, ΔR = .104, R = .105, F(13, 113) = Discussion 7.16, p < .001, Religious Upbringing continued to be a non- significant predictor of Fatalism. The present study investigated whether various aspects of We then investigated the relationship between Religious irreligion predicted fatalism, nihilism, and the perception of Upbringing and Nihilism (n = 1,188). Nihilism was regressed life meaning being internally produced. While the existing Speed et al. 9 literature specifies inherent deficits that atheists, the reli- because the notion that atheists do not derive meaning from giously unaffiliated, and persons raised in religiously unaf- god(s) would be tautological. Moreover, given a definition of filiated homes are likely to face regarding meaning in life, meaning where a deity is inherent, what would differences our results did not produce support for such views. In fact, between atheists and theists actually signify? In such a case, these “identity variables” were largely irrelevant to the sta- one group would, by definition, be excluded from experienc- tistical models. However, the analysis confirmed that athe- ing meaning. ists and the religiously unaffiliated were more likely to indicate that life has endogenously produced meaning—one Making one’s own purpose. A consistent and strong finding of the few persistent differences between the groups we that emerged from the present study was related to Endoge- examined. nous Meaning. Being either an atheist or religiously unaffili- ated was associated with stronger agreement that meaning in life came from within, and these differences persisted despite Are the Irreligious More Likely to Be Fatalists and the inclusion of covariates. Not only were these differences Nihilists? statistically significant, but they were also of practical sig- nificance. The observed effect size for atheists in this regard Our present results do not support the idea that atheists or the was g = .57 (a medium-to-large effect), and the effect size for religiously unaffiliated possess a greater sense of fatalism or the religiously unaffiliated was g = .34 (a small-to-medium nihilism. While it is possible to frame the results from the effect). These findings are consistent with two complimen- present study as running against the existing literature, we tary ideas: Theists and the religiously affiliated are more think that the case for atheist or nonreligious deficits in life likely to believe that life has extrinsically produced meaning meaning was never adequately established to begin with. (possibly due to god or gods), and atheists and the religiously This is because most views on this topic rest on academic unaffiliated are more likely to perceive meaning as a product speculation as to the relationship between atheism, religion, of the self rather than a product of an external source or agent and meaning in life, rather than empirical tests. Our findings (or a relationship with this agent). suggest that atheists/theists and the religiously unaffiliated/ Researchers often start with the perception or assumption affiliated do not systematically differ with regard to fatalism that meaning is exogenous, or that exogenous meaning is and nihilism, constructs that are routinely linked with psy- richer or more fulfilling than endogenous meaning (e.g., chological well-being. Overall, these findings would be con- Park & McNamara, 2006). This approach can, unsurpris- sistent with at least a few other studies that find that atheists ingly, lead to conclusions that atheists have impoverished or the religiously unaffiliated do not suffer from psychologi- levels of meaning in life as compared with theists. If mean- cal deficits because of their position (Galen & Kloet, 2011; ing in life is exogenous, then one would expect atheists and, Speed, 2017). to a lesser extent, the religiously unaffiliated, to report greater levels of fatalism and nihilism. However, despite an ade- quately powered analysis using nationally representative Are Assumptions Underpinning Meaning in Life data, these relationships did not emerge. While several rea- Warranted? sons could be provided to explain these null findings, we In light of our findings, it is instructive to investigate why the think that the simplest explanation is that meaning in life is idea persists that irreligion should reflect differences on multidimensional and varies across social categories. On nihilism, fatalism, and source of meaning in life in general. average, atheists (and to a lesser extent the religiously unaf- In our view, the literature often fails to appropriately distin- filiated), in theory, are less likely to accept an exogenous guish between types of meaning, either in its endogenous source of meaning—and were more likely to believe mean- form (meaning perceived to be internally produced) or exog- ing was internally produced. Again, these findings are con- enous form (meaning perceived to be externally produced). sistent with the idea that “whatever makes life meaningful is It is important to note that exogenous meaning does not nec- heavily loaded with whatever people value” (Klinger, 2012, essarily imply the source is god(s). Granted, this may be a p. 29). Atheists in particular are perhaps less likely to have an “traditional” source of exogenous meaning, but it should not externally grounded meaning in life—but not any meaning be thought of as a sole source as persons may perceive any in life (Jörns, 1997). number of external things provide them within meaning Our findings generally align with Wong (1998), who gen- (e.g., the universe, karma, membership in social groups, rela- erated a personal meaning scale, by asking individuals about tionships, occupation). However, if meaning in life is what makes life meaningful and then factor analyzing the assumed to be given or bestowed on persons from a transcen- responses. An overwhelming number of the items were dent source, then this assumption has a substantial impact on related to goal pursuit; the factor of achievement striving investigating an atheistic worldview. Under such an assump- explained over three times the amount of variance explained tion, it would be technically true to suggest that atheists have by the “religion” factor (see Klinger, 2012, p. 29). Religiously less meaning in their lives than theists, but this is only derived meaning may serve as a central source of meaning 10 SAGE Open only insofar as it is personally and socially valued, and such the 2008 GSS was specifically chosen because it contained valuation itself may be part of socialization and learning pro- relevant questions that allowed us to adequately test hypoth- cesses. This illustrates the issues with assuming that com- eses related to meaning, religion, purpose, and fatalism. A mon beliefs or behaviors are generalizable to all. Exogenous related limitation was the number of atheists (n = 115), reli- meaning certainly exists for many, but that does not mean giously unaffiliated (n = 192), and raised religiously unaffili- that endogenous meaning can or should be ignored. As King, ated (n = 96) available in the dataset. While these numbers Heintzelman, and Ward (2016) suggest, “the experience of allowed for adequately powered analyses, it would be bene- meaning in life may be quite a bit more commonplace than is ficial to have a greater number of these groups represented in often portrayed” (p. 211), and we observe that religion is one future research. Although the GSS is a nationally representa- of many paths for attaining it. tive sample, the final sample size was more characteristic of We can also point out that if exogenous meaning is defined psychological studies than of sociological studies at the as meaning that is not self-made, this definition does not macro level. An overall larger sample size would also be intrinsically implicate religion or theism. In this study, we desirable to attain more precise population parameter esti- were specifically concerned with exogenous meaning-mak- mates, allowing us increased confidence in the presence (or ing processes in the context of theism, that is, specifically absence!) of differences between groups. religious meaning-making. The religiously unaffiliated, and It should be noted that focusing on the self as a source of even atheists, can reference or at least perceive a spiritual or meaning bears a specific relation to the kind of culture found transcendent meaning of external origin, though for atheists, in the United States, which is a notably individualistic one. this would very likely depend on how “spirituality” was In this way, we caution that our findings may only be plau- framed or defined. For example, studies suggest that when sible in the context of postindustrial, Westernized countries. atheists are given the opportunity to list sources of profun- This raises the question of whether the self would be a salient dity and meaning, they often list external sources such as source of meaning for nonreligious or atheistic individuals nature, the universe, social causes, and humanity (Caldwell- elsewhere in the world. There is good reason to think that the Harris et al., 2011; Coleman & Arrowood, 2015; Coleman, self would be less prominent an element of meaning-making Silver, & Holcombe, 2013). Future research should explore in other cultures and countries. As such, a cross-cultural per- the psychological consequences of utilizing these external spective on our topic would be a valuable contribution to the sources of meaning. study of meaning systems, identity, and the production of meaning in life. Future research should also seek to delineate between Limitations, Future Directions, and Conclusion types of religious affiliation and religious upbringing to The present study had several limitations. First, we assessed determine whether meaning in life is predicted by this speci- the perception of the extent to which life meaning is inter- ficity. Unfortunately, while the GSS contained a more spe- nally produced, the perception that life was without mean- cific breakdown of religious affiliations, many of the ing, and the perception that one could not act meaningfully categories were too small to allow for adequately-powered with agency. In other words, “life is only meaningful if you analyses. Nevertheless, it is possible that a more refined provide the meaning yourself” is not equivalent to “I have a analysis of these groups, using a different dataset, would high level of self-produced meaning” or other measures of allow for a better picture of differences to emerge. meaning in life. Ideally, we would have assessed actual lev- Overall, the present study found no support for the con- els of self-produced life meaning, which would have allowed tention that atheists and the religiously unaffiliated were for more robust conclusions. However, the questions used in more prone to nihilism or fatalism than their respective coun- the present study demonstrate some conceptual overlap with terparts. While these groups were more likely to indicate that aspects of the Meaning in Life Questionnaire (cf. Steger & meaning in life was endogenous, these differences did not Frazier, 2005) for nihilism (“In my opinion, life does not seem to coincide with a view that life was uncontrollable or serve any purpose”; cf. “My life has no clear purpose”; without meaning. Religious affiliation and/or belief in god(s) Steger & Frazier, 2005). Generally, our point yet remains is certainly a pathway toward meaning in life; however, no that, to avoid classification and measurement issues with potential source for meaning is also a necessary source for nonreligious individuals and populations, future research meaning. should always seek to discriminate between meaning-mak- ing processes that rely on external versus internal sources. To Authors’ Note the extent that such distinctions are obscured or not accounted All authors are members of the Atheist Research Collaborative. for, this may produce misleading conclusions about the impact of meaning in life on other important psychological Acknowledgments outcomes. Second, given that archival data were used, the present The authors would like to thank Luke Galen for helpful comments study was limited to whichever items were asked. However, on an earlier draft of this manuscript. Speed et al. 11 Declaration of Conflicting Interests Caldwell-Harris, C., Wilson, A., LoTempio, E., & Beit-Hallahmi, B. (2011). Exploring the atheist personality: Well-being, awe, The author(s) declared no potential conflicts of interest with respect and magical thinking in atheists, Buddhists, and Christians. to the research, authorship, and/or publication of this article. Mental Health, Religion & Culture, 14, 659-672. doi:10.1080/ 13674676.2010.509847 Funding Coleman, T. J., III, & Arrowood, R. B. (2015). Only we can save The author(s) received no financial support for the research, author- ourselves: An atheist’s “salvation.” In H. Bacon, W. Dossett, & ship, and/or publication of this article. S. Knowles (Eds.), Alternative salvations: Engaging the sacred and the secular (pp. 11-20). London, England: Bloomsbury. Coleman, T. J., III, & Hood, R. W., Jr. (2015). Reconsidering Notes everything: From folk categories to existential theory of mind. 1. Other scientists have disagreed (e.g., Goodenough, 1998); “It [Peer commentary on the journal article “From weird experi- does no harm to the romance of the sunset to know a little bit ences to revelatory events” by A. Taves]. Religion and Society: about it” (Sagan, 1994, p. 80). Advances in Research, 6(1), 18-22. 2. For extensive reviews of the demographic and personality Coleman, T. J., III, Silver, C. F., & Holcombe, J. (2013). Focusing characteristics of the nonreligious and atheists, see Caldwell- on horizontal transcendence: Much more than a “non-belief.” Harris (2012), Galen (2014), Streib and Klein (2013), and Essays in the Philosophy of Humanism, 21, 1-18. doi:10.1558/ Zuckerman (2009). eph.v21i2.1 3. Nihilism is not necessarily equivalent to absence of purpose, in Coleman, T. J., III, Silver, C. F., & Hood, R. W., Jr. (2016). “…If the sense that it can be a broader reference to moral or ontolog- the universe is beautiful, we’re part of that beauty.” – A “neither ical nihilism. Worded as it is, however, the item from the GSS religious nor spiritual” biography as horizontal transcendence. reflects more specifically the concept of “purposelessness.” In H. Streib & R. Hood (Eds.), Semantics and psychology of spirituality (pp. 335-372). Dordrecht, The Netherlands: ORCID iD Springer. doi:10.1007/978-3-319-21245-6_22 Crescioni, A., & Baumeister, R. (2013). The four needs for meaning, David Speed https://orcid.org/0000-0001-7033-2068 the value gap, and how (and whether) society can fill the void. In J. Hicks & C. Routledge (Eds.), The experience of meaning References in life: Classical perspectives, emerging themes, and controver- Ai, A. L., Peterson, C., Rodgers, W. L., & Tice, T. N. (2005). Faith sies (pp. 3-15). Dordrecht, The Netherlands: Springer. factors and internal health locus of control in patients prior to Crosby, D. (1988). The specter of the absurd. Albany: State open-heart surgery. Journal of Health Psychology, 10, 669- University of New York Press. 676. doi:10.1177/1359105305055318 Ellison, C. G., & Burdette, A. M. (2011). Religion and the sense Baggini, J. (2004). What’s it all about? Philosophy and the mean- of control among U.S. adults. Sociology of Religion, 73, 1-22. ing of life. New York, NY: Oxford University Press. doi:10.1093/socrel/srr035 Baker, J. O. B., & Smith, B. (2009). None too simple: Examining Farias, M., Newheiser, A. K., Kahane, G., & de Toledo, Z. (2013). issues of religious nonbelief and nonbelonging in the United Scientific faith: Belief in science increases in the face of States. Journal for the Scientific Study of Religion, 48, 719-733. stress and existential anxiety. Journal of Experimental Social Barrett, J. (2012). Born believers: The science of children’s reli- Psychology, 49, 1210-1213. gious belief. New York, NY: Free Press. Francis, L. J., & Robbins, M. (2004). Belonging without believ- Baumeister, R. (1991). Meanings of life. New York, NY: Guilford ing: A study in the social significance of Anglican identity Press. and implicit religion among 13-15 year-old males. Implicit Beit-Hallahmi, B. (2015). Psychological perspectives on religion Religion, 7, 37-54. and religiosity. East Sussex, UK: Routledge. Frankl, V. (2006). Man’s search for meaning. Boston, MA: Beacon Berger, P. L. (1990). The sacred canopy: Elements of a sociologi- Press. (Original work published 1959) cal theory of religion. Garden City, NY: Anchor Doubleday. Galen, L. W. (2014). Nonreligious and atheist emerging adults. In (Original work published 1967) C. Berry & M. Abo-Zena (Eds.), Emerging adults’ religious- Bering, J. (2002). The existential theory of mind. Review of General ness and spirituality: Meaning-making in an age of transition Psychology, 6(1), 3-24. doi:10.1037/1089-2680.6.1.3 (pp. 237-252). New York, NY: Oxford University Press. Bering, J. (2003). Towards a cognitive theory of existential mean- Galen, L. W. (2015). Atheism, wellbeing, and the wager: Why ing. New Ideas in Psychology, 21, 101-120. doi:10.1016/ not believing in God (with others) is good for you. Science, s0732-118x(03)00014-x Religion & Culture, 23(3), 54-69. doi:10.17582/journal. Blessing, K. (2013). Atheism and the meaningfulness of life. In S. src/2015/2.3.54.69 Bullivant & M. Ruse (Eds.), The Oxford handbook of atheism Galen, L. W. (2017a). Overlapping mental magisteria: Implications (pp. 104-118). Oxford, UK: Oxford University Press. of experimental psychology for a theory of religious belief as Bullivant, S. (2013). Defining atheism. In S. Bullivant & M. Ruse misattribution. Method & Theory in the Study of Religion, 29, (Eds.), The Oxford handbook of atheism (pp. 11-21). Oxford, 221-267. doi:10.1163/15700682-12341393 UK: Oxford University Press. Galen, L. W. (2017b). Which functions assumed to be religious Caldwell-Harris, C. (2012). Understanding atheism/non-belief as and spiritual in nature are ultimately attributable or reducible an expected individual-differences variable. Religion, Brain & to purely secular mechanisms? Religion, Brain & Behavior, 7, Behavior, 2, 4-23. doi:10.1080/2153599x.2012.668395 293-295. doi:10.1080/2153599X.2016.1249919 12 SAGE Open Galen, L. W., & Kloet, J. (2011). Mental well-being in the religious to stressful life events. Psychological Bulletin, 136, 257-301. and the non-religious: Evidence for a curvilinear relationship. doi:10.1037/a0018301 Mental Health, Religion & Culture, 14, 673-689. doi:10.1080/ Park, C. (2013). Religion and meaning. In R. Paloutzian & C. Park 13674676.2010.510829 (Eds.), Handbook of the psychology of religion and spirituality Gervais, W. M., & Najle, M. B. (2015). Learned faith: The influ- (2nd ed., pp. 357-379). New York, NY: Guilford Press. ences of evolved cultural learning mechanisms on belief in Park, C., & McNamara, P. (2006). Religion, meaning, and the Gods. Psychology of Religion and Spirituality, 7, 327-335. brain. In P. McNamara (Ed.), Where God and science meet: Goodenough, U. (1998). The sacred depths of nature. New York, How brain and evolutionary studies alter our understanding of NY: Oxford University Press. religion (pp. 67-89). Westport, CT: Praeger. Greenfield, E. A., Vaillant, G. E., & Marks, N. F. (2009). Do formal Religious Landscape Study. (2016). Pew Research Center’s religious participation and spiritual perceptions have indepen- Religion & Public Life Project. Retrieved from http://www. dent linkages with diverse dimensions of psychological well- pewforum.org/religious-landscape-study/ being? Journal of Health and Social Behavior, 50, 196-212. Sagan, C. (1994). Pale blue dot. New York, NY: Random House. doi:10.1177/002214650905000206 Schieman, S., Pudrovska, T., & Milkie, M. A. (2005). The sense Hayward, R. D., Krause, N., Ironson, G., Hill, P. C., & Emmons, of divine control and the self-concept: A study of race dif- R. (2016). Health and well-being among the non-religious: ferences in late life. Research on Aging, 27, 165-196. Atheists, agnostics, and no preference compared with religious doi:10.1177/0164027504270489 group members. Journal of Religion & Health, 55, 1024-1037. Schnell, T. (2009). The Sources of Meaning and Meaning in Hwang, K., Hammer, J., & Cragun, R. (2011). Extending religion- Life Questionnaire (SoMe): Relations to demographics and health research to secular minorities: Issues and concerns. well-being. The Journal of Positive Psychology, 4, 483-499. Journal of Religion & Health, 50, 608-622. doi:10.1007/ doi:10.1080/17439760903271074 s10943-009-9296-0 Schnell, T., & Keenan, W. (2011). Meaning-making in an atheist Ivtzan, I., Chan, C., Gardner, H., & Prashar, K. (2013). Linking reli- world. Archive for the Psychology of Religion, 33, 55-78. doi:1 gion and spirituality with psychological well-being: Examining 0.1163/157361211x564611 self-actualisation, meaning in life, and personal growth initia- Schumaker, J. (1992). Mental health consequences of irreligion. In tive. Journal of Religion & Health, 52, 915-929. doi:10.1007/ J. Schumaker (Ed.), Religion and mental health (pp. 54-69). s10943-011-9540-2 New York, NY: Oxford University Press. Jörns, K.-P. (1997). Die neuen Gesichter Gottes: Was die Menschen Smith, T. W., Marsden, P. V., Hout, M., & Kim, J. (2013). heute wirklich glauben [What people really believe today: The General Social Surveys, 1972-2012 [machine-readable data New Faces of God]. Munich, Germany: C. H. Beck. file]. Principal Investigator, Tom W. Smith; Co-Principal King, L., Heintzelman, S., & Ward, S. (2016). Beyond the search Investigators, Peter V. Marsden and Michael Hout, NORC ed. for meaning: A contemporary science of the experience of Chicago: National Opinion Research Center, producer, 2005; meaning in life. Current Directions in Psychological Science, Storrs, CT: The Roper Center for Public Opinion Research, 25, 211-216. doi:10.1177/0963721416656354 University of Connecticut, distributor. 1 data file (57,061 logi- Klinger, E. (2012). The search for meaning in evolutionary goal- cal records) and 1 codebook (3,422 pp). theory perspective and its clinical implications. In P. Wong Speed, D. (2017). Unbelievable?! Theistic/epistemological view- (Ed.), The human quest for meaning: Theories, research, and point affects religion-health relationship. Journal of Religion applications (2nd ed., pp. 23-56). New York, NY: Routledge. & Health, 56, 238-257. doi:10.1007/s10943-016-0271-2 Koenig, H. G. (2012). Spirituality and health research: Methods, Speed, D., & Fowler, K. (2017). Empowerment or dependency? The measurements, statistics, and resources. West Conshohocken, religion/religiosity–mastery relationship. Canadian Psychology/ PA: Templeton Foundation Press. Psychologie Canadienne, 58, 333-344. doi:10.1037/cap0000095 Krause, N., & Pargament, K. (2017). Losing my religion: Exploring Speed, D., & Fowler, K. (2017). Good for all? Hardly! Attending the relationship between a decline in faith and a positive affect. church does not benefit religiously unaffiliated. Journal of Applied Research in Quality of Life, 12, 885-901. doi:10.1007/ Religion & Health, 56, 986-1002. doi:10.1007/s10943-016- s11482-016-9495-2 0317-5 Langston, J. (2014). Explaining atheism: Testing Hunter’s Steger, M. (2017). Meaning in life and wellbeing. In M. Slade, L. Durkheimian theory. Interdisciplinary Journal of Research on Oades, & A. Jarden (Eds.), Wellbeing, recovery and mental Religion, 10, Article 10. health. Cambridge, NY: Cambridge University Press. Lee, L. (2012). Research note: Talking about a revolution: Steger, M., & Frazier, P. (2005). Meaning in life: One link in Terminology for the new field of non-religion studies. Journal the chain from religiousness to well-being. Journal of of Contemporary Religion, 27, 129-139. doi:10.1080/1353790 Counseling Psychology, 52, 574-582. doi:10.1037/0022- 3.2012.642742 0167.52.4.574 Morgan, J. (2013). Untangling false assumptions regarding Stinespring, J., & Cragun, R. (2015). Simple Markov model for atheism and health. Zygon, 48, 9-19. doi:10.1111/j.1467- estimating the growth of nonreligion in the United States. 9744.2012.01316.x Science, Religion & Culture, 2(3), 96-103. doi:10.17582/jour- Pargament, K. (1997). The psychology of religion and coping. New nal.src/2015/2.3.96.103 York, NY: Guilford Press. Streib, H., & Hood, R. (2013). Modeling the religious field: Park, C. (2010). Making sense of the meaning literature: An inte- Religion, spirituality, mysticism, and related world views. grative review of meaning making and its effects on adjustment Implicit Religion, 16, 137-155. doi:10.1558/imre.v16i2.133 Speed et al. 13 Streib, H., & Klein, C. (2013). Atheists, agnostics, and apostates. Wong, P. T. P. (1998). Implicit theories of meaningful life and the In K. Pargament (Ed.), APA handbook of psychology, religion development of the personal meaning profile. In P. T. P. Wong and spirituality. Washington, DC: American Psychological & P. S. Fry (Eds.), The human quest for meaning: A handbook Association. of psychological research and clinical applications (pp. 111- Thiessen, J., & Wilkins-Laflamme, S. (2017). Becoming a religious 140). Mahwah, NJ: Lawrence Erlbaum. none: Irreligious socialization and disaffiliation. Journal for the Zika, S., & Chamberlain, K. (1992). On the relation between mean- Scientific Study of Religion, 56, 64-82. doi:10.1111/jssr.12319 ing in life and psychological well-being. British Journal of Tiliouine, H., & Belgoumidi, A. (2009). An exploratory study of Psychology, 83, 133-145. religiosity, meaning in life and subjective wellbeing in Muslim Zuckerman, P. (2009). Atheism, secularity, and well-being: How students from Algeria. Applied Research in Quality of Life, 4, the findings of social science counter negative stereotypes and 109-127. doi:10.1007/s11482-009-9076-8 assumptions. Sociology Compass, 3, 949-971. doi:10.1111/ Twenge, J., Exline, J., Grubbs, J., Sastry, R., & Campbell, W. j.1751-9020.2009.00247.x (2015). Generational and time period differences in American adolescents’ religious orientation, 1966–2014. PLoS ONE, Author Biographies 10(5), e0121454. doi:10.1371/journal.pone.0121454 David Speed completed graduate school at Memorial University of Vail, K., Rothschild, Z., Weise, D., Solomon, S., Pyszczynski, T., Newfoundland and is an assistant professor in psychology at the & Greenberg, J. (2010). A terror management analysis of the University of New Brunswick in Saint John. His primary research psychological functions of religion. Personality and Social interest is in the health of atheists, Nones, and the non-spiritual. Psychology Review, 14, 84-94. doi:10.1177/1088868309351165 West, S. G., Aiken, L. S., & Krull, J. L. (1996). Experimental Thomas Coleman III is a PhD student at Coventry University as a personality designs: Analyzing categorical by continu- part of the Brain, Belief, and Behaviour research group.his research ous variable interactions. Journal of Personality, 64, 1-48. interests include theory of mind and folk psychology in nonreli- doi:10.1111/j.1467-6494.1996.tb00813.x gious populations. Wilkinson, P., & Coleman, P. (2010). Strong beliefs and coping Joseph Langston is an independent academic researcher who stud- in old age: A case-based comparison of atheism and reli- ies the origins of both atheism and religion. He is also a former gious faith. Ageing & Society, 30, 337-361. doi:10.1017/ college lecturer in the sociology of religion. s0144686x09990353

Journal

SAGE OpenSAGE

Published: Jan 20, 2018

Keywords: atheism; meaning in life; nonreligion; nihilism; fatalism; nonbelief

References