David R. Davis, Emory University; Kristian Skrede Gleditsch, University of Essex & Peace Research Institute Oslo posted on August 09, 2021 10:05
We were deeply saddened by the passing of our mentor and friend Michael D. Ward on 9 July 2021. Mike had a huge impact on the profession and innumerable people within it and his passing is tremendous loss for the disciple and the many colleagues, students, and friends he touched. We extend our deep sympathies to Mike’s, family Sandra and Chris.
Mike was an innovative scholar who made many significant contributions to the discipline. Mike was a long-standing member of the International Studies Association, and he served in several leadership positions within the organization. He was a vice-president of the ISA from 2002-2004. He served on the ISA Finance Committee 2003-2006, and on the publications committee from 1997-2001, chairing the committee for the last two years. He was also very active in the ISA Scientific Studies of International Processes section, serving on the governing council and as SSIP president. In 1987 he was awarded the ISA’s Karl Deutsch Award, recognizing scholars under 40, or within ten years of defending their dissertation, judged to have made the most significant contribution to the study of International Relations and Peace Research.
Born in Japan to a military family, Mike had a long academic career, at many institutions and multiple countries. He received a PhD in Political Science from Northwestern University in 1977, with a doctoral dissertation on the political economy of inequality which was later published as The Political Economy of Distribution: Equality Versus Inequality (Elsevier-North Holland, 1978), in which he drew attention to the role of international factors and domestic processes on multiple facets of inequality beyond income.
Mike remained at Northwestern, working with Harold Guetzkow on foreign policy decision making and global modelling. This was followed by an extended period at the Science Center in West Berlin, Germany (Wissenschaftzentrum Berlin, WSB) from 1980, a city which Mike already knew from his military service at the frontline of the Cold War in the United States Army between 1970-72. While at the WSB, he worked on the GLOBUS model, an important early effort to model international relations, simulate trends and forecast state behavior in the international system.
Mike returned to the US to take up a tenured position at the University of Colorado, Boulder in 1982. This marked the start of his long and sustained career working closely with PhD students, with a first doctoral dissertation supervisee completing in 1989 (Kun Y. Park). After his wife Sandra started to work for Microsoft in Seattle he moved to the University of Washington. In 2009, Mike moved to Duke University where he remained until formally retiring in 2019. After retiring he maintained an active research agenda through his consulting company Predictive Heuristics.
Mike’s research interests have been diverse and varied, but we would like to highlight a few of the areas where Mike has made significant and enduring contributions. Mike had a long-standing interest in the study of defense spending and arms races. In an early study with Thomas Cusack in 1981, he noted that the traditional Richardson model was unable to provide a satisfactory account of military spending of the USA, USSR, and People’s Republic of China. His first article in the American Political Science Review in 1984 (“Differential paths to parity: A study of the contemporary arms race”) proposed an important way to address the shortcomings of the standard framework, proposing a stock-flow model where states react to both budgets and military personnel. He also carried out important studies of defense spending in a range of individual countries, including France, India, Israel, and South Korea. Unlike many other scholars who examined military expenditures and their effects in ad hoc statistical frameworks, Mike’s work displayed close attention to established models of economic growth and often involved close collaboration with economists. A 1992 article with David R. Davis in the American Political Science Review (“Sizing up the peace dividend: Economic growth and military spending in the United States”) tried to quantify the negative impact of military spending of the economy in the US, using a Feder-Ram model of the effects of government spending on the economy, and then used the results to simulate the likely peace dividends from proposed cuts to the defense budget after the Cold War.
Another well-known area of Mike’s research interest is the role of space and geography in international relations and political science. Although there has been an explosion of interest in space and geographical data in the social sciences over the recent years, Mike was one of the first political scientists to take an interest in the literature on spatial dependence in economics and geography and to understand its wider implications for the study of political phenomena. Following a long-standing collaboration with a geographer John O’Loughlin, Mike was awarded a National Science Foundation grant to study the diffusion of democracy in 1995. The guiding insight in the project was that international factors are likely to have a major impact on the prospects for transitions to democracy and the viability of specific regime types. According to Google Scholar, Mike’s most cited article is “Diffusion and the international context of democratization” in International Organization with Kristian Skrede Gleditsch. His interest in spatial processes and dynamics then led to a more general interest in the study of networks, especially methods to detect latent network structures from data involving applications to international conflict, trade, and outbreaks of civil war and political violence.
During the last decades Mike has been particularly well known for his work on conflict prediction and strong advocacy for the centrality of prediction as a tool for evaluating the accuracy of propositions and the quality of research. A 2010 article with Kristin Bakke and Brian Greenhill (“The perils of policy by p-value: Predicting civil conflicts”, Journal of Peace Research) examined the predictive ability of two well-known models of civil war onset, and demonstrated that models that appeared to fit the data well in-sample, with significant coefficients, often provided poor guidance to predicting events out of sample. Some casual readers might see this as a negative evidence, affirming the old cliché that “prediction is difficult, especially about the future”, and that researchers should not waste their time trying to predict but content themselves with understanding the past in the best possible manner. Mike instead set off on a sustained research agenda to prove such critics wrong and show how conflict prediction could be improved. In this line of research, Mike drew heavily on insights from his previous work such as the importance of space and networks, but also picked up many new tools such as advanced in event history models of heterogeneous populations, Bayesian model averaging, and machine learning. He started a sustained set of cooperative projects with end users, including important work for the Political Instability Task Force, a US government-sponsored research project seeking to examine how best practices in research can inform US national intelligence and government policies. In a 2016 article (“Can we predict politics? Toward what end?,” Journal of Global Security Studies), Mike wrote an impassioned plea for prediction as an end for research and evaluation of learning, including a deliberately provocative “call for less theory in security studies,” advocating instead to “winnow the many, many such ‘theories’ that occupy the world of security studies” through “more predictions” (p. 84).
A number of common themes run through Mike’s scholarship and practice. First is an enduring curiosity, and a willingness to always explore new topics and new approaches instead of adding marginal extensions to past projects, even when such lower-hanging fruit might have clearer pay-offs in the short run. A second is interdisciplinarity, specifically engagement with the developments and work of scholars in many disciplines, including computer science, economics, geography, and statistics. Many of Mike’s research collaborators come from outside the borders of the USA, and his impact and reputation also extend beyond internationally. A third is an emphasis on collaboration and building community. Mike believed in building working groups, integrating numerous scholars and students at many different stages of their careers, and his co-authored scholarship demonstrated that we can learn more as members of a teams than as individual scholars. But what stands out most to us is the limitless generosity and dedication Mike showed to his students and collaborators. To be sure, Mike could be demanding and temperamental, but he spent an enormous amount of time with others, taking a real interest in their work, and making constructive suggestions on how to improve this. We are struck by how many of his former students have started their tributes by stating that they would never have finished their degree without Mike’s help and mentorship
It will take a long time for many of us to process the absence of Mike in our lives and accept that we will never share another laugh or an eureka moment together. However, we will treasure the many memories that we have of Mike, and try to honor his legacy by trying to show the same kind of generosity and mentoring that Mike showered upon us. We plan to honor Mike at a roundtable at the 2022 ISA annual meeting.
David R. Davis, Emory University
Kristian Skrede Gleditsch, University of Essex & Peace Research Institute Oslo