Projects & Publications (updated 1/22):

How International Law Keeps Pace with Technological Change: Evidence from an Experiment (under review)

  • Law is often codified in text. Text is static, however, and new situations and technologies arise for which its writers could never have envisioned. Yet societies often find ways of adapting the law rather than writing it anew. When new situations arise for which the law is technically silent, how do states decide whether these constitute loopholes, lacunae, or lawbreaking? This paper advances a formal theory that shows how ambiguity can actually be advantageous in some situations. Self-interest politicizes the interpretation process, but only at the margins, forcing decisionmakers to make tradeoffs between plausibility and security. The theoretical predictions are tested using a vignette-based laboratory experiment fielded on legal experts in the US and China. The empirical findings reveal how legal agents are constrained by analogical capacity flowing from extant legal text, even when the stakes are high.

Controlling Tomorrow: Explaining Anticipatory Bans on Emerging Military Technologies (under review)

  • The idea that technology perennially outpaces legal rules has in recent years become a trope. This is especially the case in international relations, where law is at its "vanishing point." But states have on numerous occasions succeeded in enacting binding limitations on new military technologies before their emergence. If the future is indeed manageable and weapons development is expensive, anticipatory arms control should be the norm. This article offers a formal explanation, theorizing about the conditions under which anticipatory agreements occur. The model is tested using a survey-based randomized controlled trial alongside extensive archival research into declassified sources. The downstream impact of new technologies on the security environment becomes clearer as they approach technical viability. Under uncertainty, anticipatory bargains are a hedging strategy for decisionmakers with pessimistic imaginations. Mutual pessimism can drive anticipatory cooperation. Importantly, the findings show this to be true even when distrust is high, flexibility provisions are absent, and monitoring is difficult. A sharper focus on anticipatory arms control places default explanations for arms control failure in perspective by highlighting improbable successes.

Mapping State Participation in Military AI Governance Discussions (with Elsa Kania, forthcoming Oxford University Press)

  • Can existing conventions be adapted to govern current advances in artificial intelligence and autonomy in weapons systems? What is the likelihood that states can achieve consensus on the core questions at stake in these debates? To date, mechanisms for arms control and governance have struggled to adapt to changing technological expectations, and metrics to gauge progress have remained inadequate. Drawing on original data, this chapter attempts to quantify the landscape of AI governance discussions under the Convention on Certain Conventional Weapons (CCW). We evaluate a battery of factors – military capabilities, economic interests, diplomatic alignment, and regime type -- in order to understand why some states are more eager than others to engage. We find that the determinants of participation are in fact nuanced, and that no single school of thought can account for the variation. Scholars and policymakers can appreciate why nearly all governments are working to shape the outcome of discussions, even if for varied reasons.

Implausible Deniability? Why Proxy Conflict in Cyberspace May No Longer Pay (Journal of Cybersecurity)

  • It is believed that states can achieve military and foreign policy objectives “on the cheap” by outsourcing cyber operations to willing proxy actors, be they cyber mercenaries, patriotic zealots, pranksters, or simply allies of convenience. By outsourcing, this logic goes, a host government can claim plausible deniability while cashing in on strategic gains. Puzzlingly, proxy-associated behavior across three datasets appears to show that activity associated with outsourcing has flagged. Do cyber proxies still pay? A formal model is used to hypothesize about how new norms of attribution (specifically, the willingness of victims to make accusations on the basis of circumstantial evidence) are developing. Sponsors who learn that they will take the heat regardless have fewer incentives to rely on proxies. Empirical evidence drawn from two cyber incident datasets offers support for this proposition. This should decrease our confidence that plausible deniability is the primary reason why states outsource their cyber operations to non-state hackers. The paper joins an emerging body of research that has questioned the logic of plausible deniability in covert action, including cyber conflict.

Honing Cyber Attribution: A Framework for Assessing Foreign State Complicity (Journal of International Affairs)

  • Concerns about state-directed cyber intrusions have grown increasingly prevalent in recent years. The idea that state principals can obfuscate their involvement in such attacks by delegating operational tasks to non-state agents poses a particularly significant challenge to international enforcement and remedies. Gaps in international law, coupled with obstacles to detection in such cases, may make it more difficult to bring sponsoring states to justice. This paper offers a roadmap for assessing the propensity ofstates to delegate to non-state actors and correct for false positives in standard (typically more technical) cyber attack attribution methods. I conclude that the conditions under which attacks are likely to have been backed by sponsoring states occupy a much narrower window than conventional wisdom suggests, and that the universe oftransgressors can be identified when standard indicators overlap with specific conditions.

Debate Hall or Echo Chamber: Do Cyber Norms Proposals Work? (working paper available)

  • Where, if anywhere, have state-led cyber norms efforts succeeded? Leveraging an original dataset, this paper provides some of the first empirical traction on this question. Contrary to conventional wisdom that states are notoriously tight-lipped about how they think law and norms should apply to new domains, I catalog more than 10,000 crossnational government expressions about how cyber law and norms apply (collectively more than 10 million words) since 1998. Unsupervised computational techniques are used to provide some of the first evidence about the extent to which states have been talking to or past one another on core issues. The findings suggest that there has been surprising convergence, but always not on the issues or in the direction dominant theory would predict. Moreover, increasingly broad participation seems to suggest that states recognize their shared interest in stabilizing the cyber domain, although most-needed type of engagement -- direct engagement between the US, Russia, and China -- is also least prevalent. By implication, the prospects for cyber norms are neither as promising, nor as dire, as divided commentary tends to predict. Future research can use this dataset to develop new theories about norm emergence, persuasion, diffusion, and internalization.

Respectful Rivalry: Comparative US Media Coverage of Soviet and Chinese Space Programs, 1957–2019 (working paper available)

  • The 50th anniversary of the Apollo 11 landing this year was received with much fanfare. But it is easy to forget that the race to the Moon was couched in superpower rivalry under the specter of nuclear war. Attitudes toward outer space activity are a puzzling exception to the "enemy image" that usually accompanies international rivalry, especially given national security sensitivities inherent to space. Do media narratives shape public opinion, or is the public naturally immune to negative government characterization? Relying on an original dataset of New York Times space coverage spanning 1957-2019 and a randomized survey experiment, I find that news media has only a limited capacity to shape respondent attitudes, and negative framing can backfire. Coverage of both the Soviet and Chinese space program has been surprisingly praiseworthy, even during periods of heightened tension. Moreover, public attitudes are mostly positive toward rival successes even when news coverage is negative, suggesting that elites fight an uphill battle when stoking public fears about space. The findings contribute to scholarly understanding about the role of media in shaping public opinion, and suggest that cooperative space ventures -- which cut costs and enjoy public popularity -- may be more politically expedient than space competition.

Two to Tango: How Outside Options Can Constrain the Use of Caselaw Precedent (working paper available)

  • In international dispute settlement, the use of precedent is theoretically discouraged but widespread in practice. Do all courts use precedent to strengthen their position? In positing explanations, existing theory overlooks jurisdictional strength as a factor. I argue that courts that lack the power to compel appearance must worry about attracting litigant entry before worrying about compliance. Because the use of precedent clarifies claimant expectations, noncompulsory courts should economize their use of precedent to avoid pushing out parties that expect to lose based on the court's track record. Using an original dataset of contentious International Court of Justice cases from 1948-2010, I find indirect support for the theory: the court exhibits marked restraint in noncompulsory settings, but less restraint in compulsory ones. The research highlights how precedent can be a double-edged sword, even when judges know their ideal points will not change.

Innovation and Organizational Politics in the US Air Force, in MILITARY STRATEGY, JOINT OPERATIONS, AND AIRPOWER (Georgetown University Press)