Dissertation

Why do capable states sometimes possess seemingly inefficient militaries that leave them vulnerable to security threats? Practitioners have long maintained that a full spectrum combined-arms military is the best defense in an unpredictable and anarchic international environment. Yet capable states often either forgo some vital defense capabilities, like the US omitting minesweepers from its planned 600 ship navy in the 1980s, or overproduce some defense capabilities, like Albania developing coastal patrol vessels with a range of 1,750 miles despite having a coastline that is roughly one tenth that size. Are states with seemingly inefficient militaries simply making mistakes? I argue that vulnerable force structures are not just the result of poor planning or resource constraints; rather, these observed “inefficiencies” are often a strategically motivated decision to specialize one’s force structure. While there are certainly advantages to a diversified military portfolio, states can engage in strategically motivated functional differentiation by specializing their militaries when they engage in cooperative security alignments. When a collection of states facing a similar threat environment are able to minimize the risk of defection and ensure effective coordination, they can engage in a division of labor where each state individually specializes in different military capabilities that, when brought together, still comprise a full spectrum military force. I substantiate these arguments with evidence from a new dataset on the distribution of military capabilities from 1970-2020 and find that 1) states in cooperative security alignments have more specialized militaries, and 2) cooperative security alignments with more closely aligned interests and higher vertical integration have a higher division of labor.

Draft excerpt and slides presented at the Notre Dame Emerging Scholars in Grand Strategy Conference (May 2021)

• Under review
• co-authored with Erik Gartzke, Jon R. Lindsay, and Peter Schram

Abstract

Recent trends increasingly place conflict in a ``gray zone'' between peace and war. Observers interpret gray zone conflicts as deterrence failures. New technologies or tactics---from cyber operations to ``little green men''---reduce the costs or increase the effectiveness of low-intensity aggression. But gray zone conflict could also reflect deterrence success. Credible prospects of retaliation encourage challengers to adopt less effective means of aggression. These dueling ``push-pull'' logics suggest contrasting conflict dynamics impacting stability. We develop a formal model that synthesizes both perspectives by analyzing deterrence success as variable, rather than dichotomous. In the model, the intensity of a challenger's provocation varies inversely with the credibility of the defender's deterrent threat. We empirically analyze Russian gray zone activity since the 1990s. Russia is more restrained, and less effective, against nations in or closer to NATO. The model suggests inherent trade-offs between stability and military potency in limiting the risk of escalation.

Presentation at the annual American Political Science (APSA) Conference (September 2018)

Draft (updated December 2020)

• under review

Abstract

Technological advances have added cyber and space to the more traditional domains of human interaction (land, sea, air). These new domains arguably increase opportunities for conflict across, as well as within, domains. Cross-domain conflict is thus seen by many as an emerging source of international instability. Yet, existing systematic empirical research has little to say about how domains interact. This study introduces a new dataset of the domains in which nations took military action during 412 international crises between 1918 and 2015. Analysis of these data yields several surprises. Far from being rare, cross-domain interactions are the modal form of conflict in crises during this period. Nor is cross-domain conflict "new:" crises that play out in more than one domain were about as frequent (proportionately) in decades past as they are today. Cross-domain crises are also less violent and of no greater duration than crises between belligerents using similar means.

Presentation at the International Studies Association (ISA) Annual Meeting (April 2021)

Draft (updated May 2021)

• co-authored with Shannon Carcelli, Rex Douglass, Erik Gartzke, Jon R. Lindsay, and Thomas Leo Scherer

Abstract

How do international crises evolve? There are significant data and measurement challenges in seeking answers to this critical question. We conceive of international affairs as a strategic chess game between adversaries, which demands a way to systematically measure pieces,moves, and gambits accurately and consistently over different contexts and periods. We develop such a measurement strategy with a crisis-specific ontology of actions and interactions and apply it to the corpus of 470 crisis narratives developed by the International Crisis Behavior (ICB) Project. Our ontology has high coverage, capturing thoughts, speech, actions, and interactions by actors. These data exhibit high accuracy with multiple human codings per sentence of each narrative. We then evaluate the resulting ICBe dataset by comparing it to prominent state-of-the-art datasets using existing strategies for measuring conflict processes. We find that ICBe captures the process of a crisis with greater accuracy and granularity than events or crisis datasets.

• co-authored with Kerry Chavez

Abstract

Research on aerial bombing has primarily focused on its effects. However, less is known about variation in the manner in which that bombing occurs. This paper contributes to research about how states fight by providing the first comprehensive dataset of the entire universe of US cruise missile strikes from 1991-2017 – a type of bombing that has been largely excluded from prior analysis on aerial bombing campaigns. This fine-grained data allows scholars to examine how the type of bombing a state undertakes is affected by international and domestic factors, geography, and tactical considerations. We demonstrate how new information on military platforms, the geography of long-range strikes, and the timing of conflict alters theories about the effects of technological innovation on international conflict.

Presentation at the ISSS-IS Conference (November 2018)

• co-authored with Rex Douglass

Abstract

Military capacity is considered central to national power, yet its measures remain aggregated and generalized. This paper aims to produce a better understanding of military capacity through a nuanced analysis of military technology. The acumen of these technological bundles determine the military systems available to a given country and thus shape military strategy, structure, and success. This paper introduces new data using the DoD's Militarily Critical Technologies List (MCTL) to shed light on the composition of technological bundles as well as empirically identifying military domains based on technological portfolios. This new way of thinking about military technology through empirical validation rather than non-military proxy will shed light on nations' system integration skills, the complementarity and substitutability of military technologies, and the role that alliances, trade ties, and domestic factor endowments play in shaping a military's technological portfolio.

Presentation at ISSS-IS Conference (November 2018)