Suchergebnisse

The proof for "War Is in the Error Term," a piece that appeared in the Summer 1999 issue of International Organization, contains a subtle error. Once the correction is made, there are broader implications for testing theories of war using quantitative studies. Large-n tests to verify the incomplete information explanation for war will be more difficult to perform than originally anticipated.

At least since Thucydides, students of international relations have sought rational explanations for the advent of war. Rationalist explanations assume purposive action; states are said to make reasoned decisions about the use of force. Although rationalist explanations have proven persuasive, durable, and offer the basis for cumulative theorizing, they also imply substantial limits on what we can know about war. I show that the most general rationalist explanation for war also dictates that the onset of war is theoretically indeterminate. We cannot predict in individual cases whether states will go to war, because war is typically the consequence of variables that are unobservable ex ante, both to us as researchers and to the participants. Thinking probabilistically continues to offer the opportunity to assess international conflict empirically. However, the realization that uncertainty is necessary theoretically to motivate war is much different from recognizing that the empirical world contains a stochastic element. Accepting uncertainty as a necessary condition of war implies that all other variables—however detailed the explanation—serve to eliminate gradations of irrelevant alternatives. We can progressively refine our ability to distinguish states that may use force from those that are likely to remain at peace, but anticipating wars from a pool of states that appear willing to fight will remain problematic. For example, we may achieve considerable success in anticipating crises, but our ability to predict which crises will become wars will probably prove little better than the naive predictions of random chance. The need for uncertainty to account for war means that the same conditions thought to account for war must also exist among states not destined to fight. Otherwise, states themselves will differentiate between opponents in a way that either removes the motives for war or restores uncertainty. It has long been accepted that social processes possess an element of uncertainty, but the centrality of uncertainty to rationalist explanations for war means that the advent of war is itself stochastic. War is literally in the "error term."

Kern des Aufsatzes ist die Theorie der Kriegsursachen. Der Verfasser übt Kritik an rationalistischen Erklärungsansätzen, die von zielgerichteten Staatenverhalten ausgehen, die einen vorhandenen Konflikt zum Kriegsausbruch bzw. militärischen Gewalteinsatz führen. Die Grenzen dieses Ansatzes werden diskutiert. Ein Artikel von James Fearon (1995) wird zur Beweisführung des Autors herangezogen, daß für außenstehende Staaten und Beobachter die Faktoren, die bei rationalistischen Erklärungen zum tatsächlichen Kriegsausbruch führen, Krieg als reines Zufallsereignis ("random event") erscheinen lassen. Implikationen für die angewandte und theoretische Kriegsursachen-Forschung werden genannt. (SWP-Ebg)

An example is presented to illustrate potential problems of working with logarithmic transformations to linearize an intrinsically nonlinear power model. Recommendations to avoid the pitfalls identified herein are offered.

In this paper, we consider portmanteau tests for testing the adequacy of multiplicative seasonal autoregressive moving‐average models under the assumption that the errors are uncorrelated but not necessarily independent. We relax the standard independence assumption on the error terms in order to extend the range of applications of the seasonal autoregressive moving‐average models. We study the asymptotic distributions of residual and normalized residual empirical autocovariances and autocorrelations under weak assumptions on noise. We establish the asymptotic behavior of the proposed statistics. A set of Monte Carlo experiments and an application to monthly mean total sunspot number are presented.