seeking security: threat perception and policy-making in a dangerous world
When new dangers emerge – as SARS-CoV-2 did in 2020 – or peripheral security concerns become priorities – as terrorism did after Sept. 11, 2001 – what influences the strategies policy-makers pursue in response?
I argue that in these moments, individual policy-makers’ preferences for how their governments should respond matter a great deal because institutional mechanisms for coping with new dangers are often minimal. Instead, these are moments when the cognitive processes associated with threat perception exert significant influence over individual-level preferences for national security strategy. And, because only a handful of individuals shape policy at these moments, these preferences are consequential. The strategies developed in these moments are also consequential because they embody new theories of how to keep the nation secure and can result in substantial redirection of government resources. The origins of these strategies are thus worth understanding.
At the outset of Seeking Security, I develop a theory of the cognitive processes that are engaged when individuals are confronted with potential dangers. I draw on evidence from a wide range of neuroscientific studies to characterize the brain-level basis of threat perception. I show that the brain parses danger into at least three categories: physical harm, loss (material and non-material), and contamination (literal and imagined). Using original survey data, I show that these categories extend to “political” dangers, such as immigration, terrorism, and climate change. With this data, I show that individual perceptions of danger vary in two ways: threat detection (is something dangerous?) and threat classification (how is it dangerous?). Breaking somewhat with the existing literature, I argue that threat classification is the more significant form of variation when it comes to national security decision-making outcomes. In combination with predictions grounded in behavioral biology, I introduce Threat-Heuristic Theory (THT), which links threat classification along the three dimensions of physical harm, loss, and contamination to a small set of predictable preferences for specific danger mitigation strategies. While the mapping between threat classification and mitigation strategies is relatively generic (e.g., preventive aggression as a response to unavoidable attack), I argue that there are national security policy equivalents (e.g., preventive war) and that individual-level preferences for these specific features of national security strategies (e.g., preventive war as defined in the 2002 NSS) can be predicted.
In the second section of the book, I test Threat-Heuristic Theory in the cases of the early Cold War national security strategies (1950-1953) and the 2002 National Security Strategy. Using new corpora in which policy-makers describe and discuss the dangers with which they are concerned prior to the strategy development process, I measure threat classification. I then use that measure to predict the specific features each policy-maker will endorse during the strategy development process. I show that my theory makes predictions that are more accurate than the predictions generated by alternative theories of preferences (e.g., bureaucratic politics). These cases provide both substantive insight into two pivotal moments of American foreign policy decision-making and theoretical purchase on the problem of integrating measures of subjective perception into the study of decision-making.
The last of the book’s empirical chapters looks cross-nationally at the perception of both terrorism and climate change. I use official speeches discussing these dangers at the United Nations to derive a measure of threat classification. I then compare the theory’s predictions to observed counterterrorism and climate-related policy taken at the national level. The book concludes with a discussion of the implications of variability in threat perception for coordinated responses to global dangers. Across the book, I rely on a variety of data sources, including archival material, observational and experimental survey data, interviews, large speech corpora, and functional magnetic resonance imaging data.