Date of Award

August 2020

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Political Science


Colin Elman


cyber command, cyber force, cyber operations, implementation dynamics, military diffusion, military innovation

Subject Categories

Social and Behavioral Sciences


What explains the variation in implementation dynamics for cyber forces across militaries? In other words, as cyber forces emerge in states across the international system, why do some militaries undertake wide-ranging implementation efforts with few alterations to cyber force structure, while implementation in other militaries is characterized by a drawn-out, incremental process entailing several changes in cyber force structure?

Militaries have been building cyber capabilities since the late 1980s; however, formalized military cyber organizations for these capabilities have only recently emerged. These cyber forces—active-duty military organizations that possess the capability and authority to direct and control computer network operations (CNOs) for strategic ends—have received little attention from scholars. Despite the potential impacts cyber forces might hold for international security dynamics, there exists no comprehensive overview of cyber forces and no analysis on the various ways they have been implemented across militaries. Moreover, current explanations drawn from the diffusion of military innovations remain incomplete in explaining the ways in which cyber force structure change over the course of the implementation process.

In this dissertation, I examine the diffusion and implementation of cyber forces and advance a theory of organizational size to account for the varying implementation dynamics across militaries. My dissertation makes two important contributions to the growing literature on cyber conflict. First, I offer a novel typology for categorizing cyber forces and the respective force structures. By classifying cyber forces according to organizational model and scale of command, I identify nine distinct cyber force structures: Subordinated Branch, Subordinated Service, Subordinated Joint, Sub-Unified Branch, Sub-Unified Service, Sub-Unified Joint, Unified Branch, Unified Service, and Unified Joint. The second contribution is empirical: I create the first comprehensive database to catalogue the diffusion of cyber forces and evolution of cyber force structures across state—the Dataset on Cyber Force Structures.

This dissertation also makes three broader contributions to the study of the diffusion of military innovations. First, I show how organizational characteristics mitigate diffusion pressures by constraining or enabling innovation and implementation. This dissertation moves past debates that portray militaries as either change-resistant or innovation-seeking organizations by providing a more nuanced claim: organizational characteristics—such as size—can predispose militaries to pursue certain types of changes while creating resistance to others. As such, this dissertation sheds important light on the ways in which the military organizational factors can shape the agency and decisions of those implementing an innovation principle. Second, I advance a stage-based conception of implementation for diffusion frameworks comprised of five stages: pre-adoption, introduction, modification, expansion, and full implementation. This framework can account for both partial and full adoption and provides a way to assess intermediate changes to an innovation prior to its full institutionalization. As a result, I use this framework to showcase the value of stage-based theorizing. Third, this dissertation introduces new methodological tools for testing stage-based hypotheses about adoption and implementation. In conjunction with qualitative analysis, this dissertation utilizes multistate survival modeling to assess variable effects at each stage of the implementation process. Traditional modeling techniques in the military diffusion literature—such as logistic regressions and basic survival modeling—prove both cumbersome and inadequate for assessing stage-based processes. In using multistate survival modeling, I emphasize the importance of matching methods to conceptual and theoretical assumptions.


Open Access