Porn as a Threat to the Mythic Social Order: Christian Nationalism, Anti-Pornography Legislation, and Fear of Pornography as a Public Menace

ABSTRACT

We propose contemporary support for sweeping anti-pornography legislation is driven, in large part, by Christian nationalism―an ideology seeking to politically enforce a mythic Christian sexual order, but not one’s own sexual behavior. We show Christian nationalism strongly predicts a desire to outlaw pornography, but not actual pornography use. Moreover, Christian nationalism is also the strongest predictor that Americans believe pornography leads to a breakdown of morals and to rape, but is unrelated to views about porn’s individual consequences. Unlike indicators of personal piety, which connect pornography to individual behavior, Christian nationalism relates to pornography almost exclusively as a societal threat.

KEYWORDS:

• Religion
• pornography
• Christian nationalism
• evangelicals
• politics

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes

1. Though notoriously difficult to define, the term “pornography” will be used broadly throughout to refer to any form of media (magazine photos, X-rated films, online videos, etc.) that depicts nudity and/or explicit sex acts and is intended and consumed for the purpose of sexual arousal (see Perry 2019).

2. We do not mean to assert that all collective opposition to pornography is conservative or religious in its motivations. Though both religious and political conservatism have long been predictors of support for anti-porn legislation (Lykke and Cohen 2015; Sherkat and Ellison 1997), anti-porn feminists have also been leading voices in the effort to declare pornography a “public health crisis,” (e.g., Dines 2016) and Whittier (2018) has documented important collaborations between anti-porn feminists and conservatives on several issues related to porn and sexual violence. In this article, we are focusing on the motivations behind the conservative side of that effort.

3. Throughout we will avoid referring to “Christian nationalists” as if they were a coherent and self-aware group. Rather we refer to Christian national-ism as an ideology (a set of beliefs, values, and narratives; see Williams 1996) that conceive of America as wedded to a particularly ethnicized brand of Christianity (Gorski 2017; Whitehead and Perry 2020), and who wish to institutionalize that connection. We propose this ideology, rather than traditional Christian morality, drives many Americans to fear pornography as a social threat and oppose it collectively through political means, rather than merely avoiding it.

4. For more on the General Social Survey, see Davis and Perry (2020). For more on the 2017 Baylor Religion Survey see Whitehead, Perry, and Baker (2018). For more on the 2019 Chapman University Survey of American Fears see Baker et al. (2020). The cooperation rate for the 2019 CUSAF was 53% and five rounds of reminder e-mails were sent to potential respondents from underrepresented demographic groups throughout the survey collection period. All analyses use weighted data: WTSSALL for the GSS models, WEIGHT for the 2017 BRS, and the 2019 CUSAF weight which matches the responses to demographic benchmarks for the United States based on age, education, gender, race/ethnicity, region, and civic engagement.

5. 36.7% of respondents chose the first option, 59.54% chose the second, and only 3.75% chose the final option. We ran a series of robustness checks to ensure our recoding of this variable did not adversely influence our results. Making either of the first two response options = 1 resulted in substantively identical findings, with Christian nationalism positively and significantly associated with the first and negatively and significantly associated with the second. The third option was chosen so infrequently that there is little variation to explain and is also a rare event, making binary logistic regression inadequate. We then used cumulative logit models and the results were again substantively identical to binary logistic regression. However, the proportional odds assumption was violated, so we then used multinomial logistic regression analyses. We found yet again that Christian nationalism is a significant predictor of differences between choosing the first or second response option. It was not significantly associated with choosing the third option, but neither was any other predictor. Again, this is most likely due to how infrequently it was chosen resulting in little variation available to be explained. Results from these models is available upon request. Beyond the methodological reasons for dichotomizing this variable, theoretically we are most interested in predicting which Americans want a complete ban of pornography as this has been the end goal of many espousing Christian nationalism over the last 40 years.

6. We ran a series of robustness checks using the original ordinal coding of this frequency of pornography use variable to ensure our findings using OLS regression techniques are trustworthy. In the cumulative logit models the Christian nationalism measure was non-significant, like in the OLS model. However, the proportional odds assumption was violated. In the multinomial logit models the Christian nationalism measure was not a significant predictor of choosing any of the response options, which is again identical to the OLS model. Results from these ancillary analyses are available upon request.

7. While there might conceivably be some social benefits implied here (e.g., as more people watch porn, perhaps collective knowledge of sex increases; or as more persons watch porn, perhaps this indirectly reduces the numbers of persons who would act out on deviant sexual impulses with real people), the most obvious direct recipient of the benefit is the individual. Whereas the other questions more clearly imply the collective (“breakdown of morals”) or others (rape).

8. For the GSS and BRS models we do not include Black Protestant as a stand-alone category due to possible issues with multicollinearity with the race control variables (Sherkat 2014). We instead aggregate these respondents into the Mainline Protestant category after ancillary analyses demonstrated that on the dependent variables, religious practice, and belief measures these groups were most similar. As a robustness check we also aggregated Black Protestants into the evangelical Protestant category. Whether Black Protestants are considered on their own, aggregated with mainline Protestants, or aggregated with evangelical Protestants, the results on key variables were substantively identical to those presented below.

9. For the 1996 and 2014 GSS we had to limit the racial categories to white, black, and other race due to a change in the racial category question between survey years.

10. We checked interaction terms for survey year*Christian nationalism to determine if Christian nationalism worked differently across years. It was non-significant.

11. Because of the online panel design of the 2019 CUSAF, there were very few missing cases. The total N for the 2019 CUSAF is 1,219, and as Table 4 shows all models have only 28 or 29 missing cases. This is why we do not use MI procedures for the 2019 CUSAF data. The PROC MI procedure in SAS 9.3 generates twenty-five imputed datasets using multiple Markov Chains based on all variables included in the models, resulting in an overall N of 43,975 (1,759 x 25) for the 1996/2014 GSS and 37,525 (1,501 x 25) for the 2017 BRS. The results in each table use the MI ANALYZE procedure in SAS. This procedure combines the results of all twenty-five imputations and generates overall estimates, standard errors, and significance tests.

Additional information

Notes on contributors

Samuel L. Perry

Samuel L. Perry is an associate professor of sociology at the University of Oklahoma. He studies the interplay of religion and cultural power within the empirical contexts of American racial politics, families, and sexuality. He authored Addicted to Lust: Pornography in the Lives of Conservative Protestants and Taking America Back for God: Christian Nationalism in the United States (with Andrew Whitehead).

Andrew L. Whitehead

Andrew L. Whitehead is an associate professor of sociology at Indiana University-Purdue University Indianapolis. His research focuses on American religion, Christian nationalism, gender, sexuality, and disability. He is the lead author of the recent book Taking America Back for God: Christian Nationalism in the United States.