Case Studies and Theory Development in the Social Sciences - Alexander L. George [181]
13
Here and in the next few paragraphs we draw directly from distinctions among covering laws, causal mechanisms, and typological theories suggested by David Dessler (private communication, January 7, 1998).
14
Kenneth Waltz, Theory of International Politics (New York: McGraw-Hill, 1979). For additional discussion, see Chapter 12.
15
See, for example, Terrence J. McDonald, ed., The Historic Turn in the Human Sciences (Ann Arbor: University of Michigan Press, 1996).
16
Harry Eckstein, “Case Studies and Theory in Political Science,” in Fred Greenstein and Nelson Polsby, eds., Handbook of Political Science, Vol. 7 (Reading, Mass.: Addison-Wesley, 1975), pp. 79-138. For further discussion, see Chapter 6.
17
An important example is Robert Bates et al., Analytic Narratives (Princeton, N.J.: Princeton University Press, 1998).
18
Bennett, Barth, and Rutherford, “Do We Preach What We Practice?”
19
We do not necessarily expect individuals to do state-of-the-art work using more than one method in a single research project. There are examples in our field of exceptional and well-trained individuals doing excellent multi-method work, but while we want to encourage this practice, we do not want to set it as the standard expectation for Ph.D. theses, books, or articles. Since it is sufficiently difficult to do cutting-edge work with one method, we suspect most multi-method work will involve collaboration between researchers who are expert at different methods, a practice that deserves encouragement.
20
Gary King, Robert O. Keohane, and Sidney Verba, Designing Social Inquiry: Scientific Inference in Qualitative Research (Princeton, N.J.: Princeton University Press, 1994).
21
Some points of agreement involve fairly standard methodological admonitions: leave a clear and replicable record of your research methods, generate a list of observable implications for alternative hypotheses under consideration, specify what empirical findings would call into question each of these hypotheses, and keep in mind that science is a social enterprise in which no research is perfect and diversity of belief serves as a useful check on individual misperceptions and biases. We also agree that counterfactual analysis can serve as a useful cross-check on theorizing, that reconfiguring one’s theory after seeing some of the data is defensible as long as it leads to new predictions on other data that hold up to additional empirical tests, and that parsimonious theories are desirable but should not be pursued at the cost of oversimplifying a complex world and reducing our ability to produce rich explanations. Most of these points are raised in King, Keohane, and Verba, Designing Social Inquiry, pp. 7-33.
22
Postmodernists would of course disagree with us and with Designing Social Inquiry on the applicability of positivist logic, even broadly construed, to the social sciences.
23
King, Keohane, and Verba, Designing Social Inquiry, p. 3.
24
Ibid., p. 3.
25
Ibid., pp. 85-87.
26
Robert Keohane, “Problematic Lucidity: Stephen Krasner ’s ‘State Power and the Structure of International Trade,’” World Politics, Vol. 50, No. 1 (October 1997), pp. 150-170; and his unpublished paper that focuses on the importance of causal mechanisms in efforts to explain the extinction of dinosaurs: “Dinosaurs, Detectives and Causal Mechanisms: Coping With Uniqueness in Social Science Research,” paper presented at American Political Science Association Annual Meeting, September 4, 1999, Atlanta, Georgia. In the latter paper, Keohane also concedes that “[Ronald] Rogowski was right [in his American Political Science Review (June 1995) critique] to criticize Designing Social Inquiry for not emphasizing sufficiently the importance of elaboration of models and the deduction of implications from them.” Keohane