This book, a combination of original chapters and reprints of some earlier material, represents a summary of the author's current thinking on the use of counterfactuals in international relations. It is a very rewarding book to read for anyone who prizes breadth and erudition. It constitutes a valuable and quite unusual addition to the methodological literature in international relations, though many scholars will have considerable difficulty understanding how best to utilize it.
Counterfactuals are sometimes dismissed as lacking in analytic weight, because the alternative histories that they posit in the absence or presence of the conditions that they identify as interesting never occurred, and so the validity of the inferences derived from them can never be assessed. Richard Ned Lebow addresses this familiar argument in an intriguing way, pointing out that traditional “factual” comparisons, while they are comparisons to actual cases, are nevertheless built upon layers of observational and theoretical assumptions, rendering them considerably more similar to counterfactuals than one might think at first blush.
Moreover, the author admits that counterfactuals are not intended to bear the full inferential weight of other forms of enquiry. At various points in the book he writes that “I do not use counterfactuals to make the case for alternative worlds, but use the construction of those worlds to probe the causes and contingency of the world we know” (p. 6); “I use counterfactuals to probe the limits of Humean causation and to develop better means of understanding causation in a largely open-ended, non-linear, highly contingent world” (p. 17); and, “[a]s Steven Weber puts it, counterfactuals are better used as ‘mind-set changers’ and ‘learning devices’ than as ‘data points in explanation’” (p. 53).
This heuristic view of counterfactuals both relieves them of an empirical burden that they cannot shoulder well and places them squarely in unfamiliar methodological terrain. The implications merit detailed attention, and here the book, in keeping with its title, tantalizes as much as it informs.
For example, one of the strengths of counterfactuals is that they highlight contingency and chance: In the case of a “minimal rewrite” counterfactual, in particular, one small change is thought to have a significant impact on outcomes. Yet that outcome is contingent on all else remaining equal following the change—a condition that downplays precisely the role of contingency and chance. Lebow concedes as much at times, as for example on page 61 when he suggests minimizing the temporal scope of counterfactuals to limit the number of potential confounders that might arise. But the fundamental tension between the contingency that counterfactuals emphasize and the determinacy that their conclusions require is difficult to square.
Similarly, the book very usefully emphasizes the importance, in individual cases, of nonsystematic causal factors, those that would normally be left to the error term in a statistical equation. In quantitative analyses, such factors are thought to be nothing more than noise; the author (or authors, in some chapters) argue/s, in contrast, that they are essential for understanding outcomes—an alarming claim, which, if true, would render many analyses of cross-case systematic causes largely useless on their own. At the same time, however, systematic factors are also often seen as necessary to explain outcomes, and while the book occasionally emphasizes the need for a better understanding of the interplay of systemic and nonsystemic factors (occasionally rendered, for example, in the form of structure and agency), it focuses largely, if not entirely, on the nonsystemic factors, thereby running the risk of not making a right with a second wrong.
For example, the Cold War case, while both learned and succinct, attributes tremendous weight to Mikhail Gorbachev and Ronald Reagan as individuals without paying much attention to the contexts of their actions. In Gorbachev's case, scholars have argued persuasively that his views were hardly fully formed at the time of his ascent to power, and that interaction with and learning from relevant epistemic communities were as critical as his own receptiveness. In Reagan's case, the old saw that “only Nixon could go to China” may have held true: While the authors argue that George H. W. Bush's prudence was largely a function of his own more complex schema, domestic pressure from hard-line conservative critics—to which Reagan was immune—should not be ignored.
The fifth and sixth chapters are devoted to an unusual exercise, an exploration of the relationship of the belief systems of academics to their attitudes toward counterfactuals. This meta-analysis suggests, among other things, that those scholars least likely to lend credence to counterfactuals also believe in a largely deterministic world and display a need for psychological closure. This immensely clever preemptive rhetorical strike against the critics of counterfactual reasoning is only partly undermined by the subtle irony of the fact that it relies on survey research—and the analysis of systematic effects alone—for its findings.
All in all, the reader, while struck by the potential of counterfactuals to illuminate something interesting about causation, is left wondering what exactly they tell us about the world, versus how much they tell us about what we believe about the world. The strength of a counterfactual lies in its plausibility, but plausibility hinges on our prior (theoretical or quasi-theoretical) beliefs about the world. Lebow argues repeatedly (pp. 14, 56) that international relations has few, if any, theories with predictive value; if that is so, our prior beliefs cannot be worth very much when used to form counterfactuals. This rendering of the process suggests that counterfactuals serve more of a hermeneutical role, helping us to clarify, focus, and interpret our theorizing, than an empirical one—a role that Lebow would seem to support, at least at times (“[c]ounterfactual experiments and simulations can tease out the assumptions on which theories and historical interpretations rest” [p. 42]).
Yet if counterfactuals are to serve as a hermeneutical adjunct to more empirical methodologies, the extent to which they are contrasted with statistical studies of international relations is doubly puzzling. It is puzzling in the first instance because the two would seem to be designed to do different things. It is also puzzling because, within the last decade, statistical methodology has made considerable progress in precisely the areas in which Lebow finds it lacking: counterfactual comparison (matching), contingency and overdetermination (also known as causal complexity), accounting for the many ways in which cases might be nonindependent, regularly reporting substantive as well as statistical significance, and so on.
Nevertheless, the larger point—that the contribution, in particular cases, of idiosyncratic factors that would typically be relegated to the positivist “error term” is not only nontrivial but occasionally crucial—both sustains attention and compels serious thought about how to incorporate such factors into other forms of research. For that reason alone, Forbidden Fruit should be read, and discussed widely, by students of international relations who are interested in pushing the methodology of their subfield forward.
