Steaming in the Dark?

Abstract

The empirical literature on arms races has often failed to find strong evidence of the existence of arms races among "obvious" dyads. In this article it is argued that if we assume that decision makers use simple rules and concentrate on the weapons stock of their rivals, we can produce highly accurate models. These premises arc tested by examining the size of the British navy from 1860 to 1913. The results offer support for this line of argument.From the late 1400s to the end of World War II, Great Britain was an important, often dominant, naval power. In this vast swath of history, it was often faced with challenges, repeatedly involving itself in what we now call arms races. The purpose of this article is to develop a quantitative depiction of the size of the British navy during one critical period of time: 1860 to 1913. Thus, this study belongs within the arms race literature, although it is an attempt to understand only one side in a set of shifting arms race dyads. And although my goal is to examine one country in one time period, the study will suggest an approach that can be applied to the examination of other arms competitions.I begin by noting how this article will differ from the "classic" quantitative arms race study. Then, the major naval, technical, and tactical challenges facing the British during the time period will be described. Finally, I proceed to examine an hypothesis about British naval behavior during that period in order to gain a better understanding of it. The Arms Race Literature: A Brief OverviewOne might argue that the scientific study of international relations began with the study of arms races. Lewis Richardson ([1919] 1960) used a pair of equations to study the dynamics of arms races, and applied the equations to the study of the military expenditures of the great powers before World War I in an attempt to learn why that war started. Richardson's work was generally unknown until an article by Anatol Rapoport (1957) introduced it to social scientists. It is an understatement to state that subsequently Richardson's work had an impact on international relations; a recent review of the arms race literature (Anderton 1989, 349) claims that there are over 100 published studies that use Richardson-process models to study arms races. Of course, there are other ways to study arms races; Huntington's ([1958] 1971) classic article uses no mathematics or statistical methodology. But it is fair to say that Lewis Richardson's work has served as the dominant paradigm for both theoretical and empirical work on arms races among those who adopt the scientific approach to the study of international relations. There appear to be at least three reasons for the popularity of this type of study: