github twitter linkedin email

My research focuses on the use and evolution of language, especially in religious contexts. I am passionate about data visualization, statistics, and recent developments in computational text analysis. My work has appeared in the Journal for the Scientific Study of Religion, Poetics, Sociological Perspectives, and Socius.

Google Scholar | ORCID ORCID iD icon


Xu, Bin and John A. Bernau. 2021. “The Sympathetic Leviathan: Modern States’ Cultural Response to Disasters.” Poetics x:x-xx.

Abstract. The sociology of disaster has paid scant attention to states’ cultural response—states’ effort to use meaningful narratives and symbolic actions to address issues about citizens’ suffering and death to enhance their legitimacy and secure citizens’ support. Our paper starts to address this gap by asking: How do states culturally respond to massive disasters? How effective are their responses? What can explain the efficacy of their cultural responses? We propose a perspective based on the cultural sociology of the state and cultural theories of trauma, theodicy, and performance. We illustrate this perspective in a comparative study of the 2008 Sichuan earthquake in China and the 2005 Hurricane Katrina in the United States. We argue that the efficacy of their cultural responses depended on whether and how effectively they addressed the key components of the meaning structure of disasters through compassionate reaction to citizens’ suffering and death, convincing accounts of states’ accountability, and cogent narratives about the long-term consequences of disasters on citizens. Both states struggled to address these issues at various points in their respective disasters, and their cultural responses were shaped by their political structures. The findings also speak to an enduring debate over whether democratic or authoritarian regimes perform better in disaster responses. We eschew the exclusive focus of the debate on administrative responses and its simplistic correlation between regime type and responses.

Bernau, John A. 2021. “From Christ to Compassion: The Changing Language of Pastoral Care.” Journal for the Scientific Study of Religion 60(2):362–381.

Abstract. The rise of neurology, psychology, and psychiatry over the last 100 years has challenged the clergy’s historical monopoly on dealing with “personal problems” and mental well-being. In this study, I document the changing language of pastoral care by analyzing over seventy years of academic articles in the Journal of Pastoral Care and Counseling (N = 4,054) using structural topic modeling. Ultimately, I reveal a linguistic shift from the universal to the particular as pastoral care professionals drop language of human nature and morality for that of individual narratives. I also find a decline of overtly religious language since the 1950s in favor of a more ecumenical language of spirituality, hope, and presence. Both of these trends take place alongside a push for “evidence- based” pastoral care. Together, these linguistic shifts offer insight into a seventy-year struggle to provide authentic religious care in a world of competing alternatives.

Figure 2. Topic composition and longitudinal trends. (Bernau 2021: 12)

Semenza, Daniel Charles and John A. Bernau. 2020. “Information-seeking in the Wake of Tragedy: An Examination of Public Response to Mass Shootings Using Google Search Data.” Sociological Perspectives x:x-xx.

Abstract. Mass shootings are a highly visible form of violence in the United States, although public response to these events varies considerably. Drawing on social problems and collective threat perception literature, we use search data for all Google-using Americans following mass shootings since 2004 to examine how event attributes such as the number of victims, venue, and type of weapon(s) predict public information-seeking related to gun control and gun rights. The results demonstrate that the number of victims, news coverage, school venue, and the use of certain weapons all significantly increase public interest in gun control and gun rights. These key predictors interact with one another to further influence information-seeking behaviors related to both gun control and gun rights. We conclude with a discussion of our findings and the potential for Google Search data in social science research.

Figure 1. Example of standardizing process for Google search traffic for gun control, 2017. a) Individual month searches return daily values scaled 0 to 100. b) Entire date range search (2004–2018) returns monthly averages scaled 0 to 100. c) Mean replacement preserves variation at both levels. (Semenza and Bernau 2020:7)

Bernau, John A. 2018. “Text Analysis with JSTOR Archives.” Socius: Sociological Research for a Dynamic World 4:1-2. [PDF]

Abstract. I provide a visual representation of keyword trends and authorship for two flagship sociology journals using data from JSTOR’s Data for Research repository. While text data have accompanied the digital spread of information, it remains inaccessible to researchers unfamiliar with the required preprocessing. The visualization and accompanying code encourage widespread use of this source of data in the social sciences.

Figure 1. Trends in American sociological publishing. The left pane depicts annual word frequency totals for three key terms in the American Sociological Review’s publishing history, with yearly totals represented as points and summarized with a smoothed trend line. The right pane depicts article length for American Journal of Sociology articles between 1897 and 2014. Trend lines are based on predicted values of article length as a function of date, number of authors, and their interaction (date × number of authors). Points represent individual articles and are sized/colored according to residuals to emphasize outliers.(Bernau 2018:2)


Copyright © 2021 John A. Bernau