My research takes an inductive approach that integrates computational and qualitative methods to explore the institutional dynamics of contested fields. I focus on analyzing semi-legal markets, deviant organizations, and unregulated financial instruments to understand how the absence of legal structure leads to particular types of social control. Additionally, I also maintain a parallel research program devoted to using automated text analysis to understand the various dimensions of political violence.
Commercial Cannabis in The United States
I am particularly interested in understanding the social transformation of the cannabis market in the United States. As an industry that is federally criminalized, but sanctioned as a commercial enterprise through state laws, there is a unique opportunity to explore some of the more vexing questions that remain about how deviant behavior can change political and cultural institutions.
My dissertation project makes use of a number of original data sources. In the first phase of the project, I collected 56 in-depth interviews with cannabis professionals and regulators across California, Arizona, and Texas in order to understand how each of the three types of cannabis programs in the United States (e.g. Recreational, Medical, and Non-Psychoactive) are organizing the commercial cannabis market. These interviews offer insights into the contemporary challenges facing cannabis professionals, and how these challenges are shaping the future of commercial cannabis in the United States. Largely, this data reveals the types of business practices that independent and corporate cannabis companies use to navigate contradictory regulations which situate them in a persistent state of semi-legality. In the second phase of the project, I surveyed cannabis professionals across the United States to understand whether these insights were patterned experiences that revealed broader market contexts, or if they were unique responses to localized criminogenic and legal processes. This survey was funded by the National Science Foundation (Award: 1903986), and the Social and Behavioral Sciences Research Institute at the University of Arizona.
Thus far, two working papers have emerged from this project. The first paper entitled “Surveillance, Social Control, and Managing Semi-Legality in the U.S. Commercial Cannabis Industry” theorizes that commercial cannabis is gray market because it operates as a state-legal enterprise while remaining federally criminalized. With this assumption in mind, this article draws on the interviews from my dissertation to explore how cannabis professionals manage their semi-legal status. The findings in this paper uncover three distinct state rationales (or “governmentalities”) for implementing social control in the cannabis industry. Through preventative surveillance, the state implements oversight practices designed to stop deviance from occuring within the cannabis industry. In contrast, adaptive surveillance informs oversight practices designed to tacitly permit some forms of deviance so long as they establish market boundaries. Finally, integrative surveillance becomes a means for the state to transform previously deviant actors into licensed cannabis industry populations. While on the surface these rationales may seem like a purely top-down form of social structure, they also operate as a heuristic device that cannabis professionals leverage in efforts to identify how they should stay compliant when the laws they are operating under contain ambiguious or unclear mandates. Consequently, the cannabis industry incorporates practices that support, modify, and at times, extend core state surveillance initiatives as a way to signal good faith attempts at compliance and avoid incurring consequences for business strategies that are not completely sanctioned. This paper is currently under review and available upon request.
The second paper entitled “Sumptuary Administration and the Subnational Policy Domain for Commercial Cannabis in the United States” theorizes that despite federal precedent, commercial cannabis continues to expand because cannabis businesses are strategically organizing their labor practices around reshaping state laws that are typically inadequate, incomplete, and/or incapable of fully supporting the market. This mixed-methods article analyzes a custom dataset consisting of all state regulations configuring commercial cannabis in the United States and the interviews gathered in first phase of this project. Using the latent text analysis procedure Structural Topic Models (STM), I draw out prominent themes embedded in these regulations to make comparisons between collections of states (which in this context are considered cannabis “regimes”) that adhere to each of the three types of programs described above. These comparisons are used to distill the logics that organize the commercial cannabis market. Next, I then turn to the interviews in order to evaluate how these logics translate into actual ways that cannabis businesses established absent market institutions in California, Arizona, and Texas as proximate cases of these regimes. The findings of this paper reveal that state governments actually follow the lead of cannabis businesses when crafting policies as opposed to setting the pace of the market. This paper is a working draft and available upon request.
Automated Text Analysis and Political Violence
I maintain a parallel research program that utilizes automated text analysis to explore the causes and consequences of various dimensions of political violence.
With colleagues Andrew P. Davis and Yongjun Zhang, I leveraged Latent Dirichlet Allocation (LDA) topic modeling to demonstrate that despite sharing similar religious motivations, Al-Qaeda in the Arabian Peninsula (AQAP) and the Islamic State of Iraq and Syria (ISIS) deploy different recruitment frameworks that focus on dissimilar goals. This finding challenges a common research practice of uncritically grouping terrorist organizations into the same niche based on abstract categories such as religion or region. We argue that focusing on more fine grained characteristics such as goals, allows for a richer understanding of the motivations and outcomes of terrorist activity. Additionally, this article demonstrates the efficacy of topic modeling for comparative analysis in low-N studies This research was published in Poetics.
With colleagues Jessica Pfaffendorf and Andrew P. Davis, I leveraged Structural Topic Models (STM) to explore the underlying themes of a novel dataset of mass shooter manifestos. Contrary to popular media narratives that frame mass shootings as a consequence of individual circumstances, this article finds a set of distinctly social logics that motivate these acts. These logics largely center on a preoccupation with responding to feelings of exclusion, threats to racial status, and challenges to masculininity. This finding supports and extends social psychological explanations of mass shooter violence. It also offers new evidence for the utility of topic modeling as a confirmatory approach that is useful for augmenting qualitative studies. This research was published in Sociological Inquiry.
Frameworks of Value Among Cryptocurrency Adopters
A previous line of research I pursued explored the boundaries between symbolic and non-symbolic value systems among adopters of cryptocurrencies. I began this project in 2015-2016 by collecting semi-structured interviews with early adopters of the popular cryptocurrency Bitcoin in order to understand how traditional economic value and symbolic values are tethered in an alternative money system that is not sanctioned by any government. I demonstrated how unique value schemas or “extra-institutional logics” emerged out of this community that supported Bitcoin during a market downturn despite being predicated on dissimilar visions about the future of the system. This research is published in Social Currents. Additionally, this project has revealed that embedding into alternative money systems follows a patterned and reflexive process. This research is forthcoming in Sociological Focus.