Most countries deployed their military in some capacity to combat the COVID-19 pandemic. We present original data on early pandemic-related deployments, identifying seven types of deployment: logistic operations, enforcement, international involvement, border protection, information provision, intelligence operations, and domestic protection. We find that military deployments are shaped by capacity and electoral considerations, even after accounting for cross-country differences in perceptions of the military. Countries with elected leaders were significantly more likely to deploy the military for border protection. Incumbents facing reelection were especially sensitive to electoral concerns, becoming significantly less likely to deploy the military for domestic enforcement when facing an imminent election.

Academic research on civil-military relations often assumes that dangers for democracy and civilian control mainly emanate from the military's predisposition of ‘pushing’ its way into politics. Yet, civilian control frequently is a precondition for governments’ moves of ‘pulling’ the military into roles that may potentially be problematic. These can include the military's involvement in political disputes or internal public security missions. Notwithstanding its empirical relevance, little academic work has been devoted to understanding how ‘pulling’ works. In this article, we aim to provide a first, exploratory framework of ‘pulling’ that captures the dynamics of the military's reactions and indirect consequences for civil-military relations. We identify three analytically distinct phases in which pulling occurs. First, politicians initiate either operational or political pulling moves. Second, we situate the military's reaction on a spectrum that ranges from refusal to non-conditional compliance. This reaction is driven by the military's role conceptions about appropriate missions and their relation to politics. In a third phase, the military may slowly start shifting its role conceptions to adapt to its new roles. We illustrate our argument with case studies of two different instances of pulling: operational pulling in the case of France (2015–19) and operational – then-turned-political – pulling in the case of Brazil (2010–20).

A considerable amount of research within security studies has explored the military's increasingly diverse and multifaceted tasks. However, this debate has been disconnected from the literature on civil-military relations to the effect that we still lack knowledge about how and why these operational tasks have consequences for the relations between the armed forces, civilian authorities, and society at large. In order to provide for a better understanding of these effects, this introduction to the Special Issue debates the concept of operational experiences to capture how the military's routine activities affect the equilibria, logics, and mechanisms of civil-military relations. The article then provides an overview of the Special Issue's six contributions, whose diverse and global perspectives shed light on different aspects of the relationship between military missions and the military's roles in society and politics. Among other factors, they highlight role conceptions – the military's shared views on the purpose of the institution – as crucial in shaping the dynamic relation between what the military does and what place it occupies within the state and society. The article concludes by describing potentially fruitful areas of future research.

Qualitative scholars exhibit a wide range of views on and approaches to causality. While some approaches reject causality from the outset, a large strand of qualitative research in political science and international relations does, however, pursue causal explanation. Qualitative scholars nevertheless disagree about what causality means. Our paper reviews what causality means within different strands of qualitative research and how qualitative scholars engage in causal explanations. We focus particular attention on the fertile middle ground between qualitative research that seeks to mimic the statistical model and research that rejects causality entirely. In broad strokes, we understand views of causality as lying on a spectrum and partly overlapping. Along the spectrum, we identify three main clusters: ‘positivist leaning,’ ‘postpositivist leaning,’ and ‘interpretivist leaning.’ Within each cluster, we identify the main traits and provide illustrative examples. We find merit in each of these three clusters of approaches and in the ongoing dialogue among qualitative scholars of different orientations. Understanding similarities and differences in the way various scholars address causality might encourage some to take steps along the spectrum and expand their repertoires to embrace elements of other approaches. By making these distinctions more explicit, we hope to be able to enhance our understanding of different views of causality and the extent to which they overlap and provide the potential for collaboration.