• Choose a category
• All
• Classic-Fiction

By Root 736 0

Sage Publications, 1982), pp. 123-147; Stanley Lieberson, “Small N’s and Big Conclusions: An Examination of the Reasoning in Comparative Study Based on a Small Number of Cases,” Social Forces, Vol. 70, No. 2 (December 1991), pp. 307-320; and Lieberson, “More on the Uneasy Case for Using Mill-Type Methods in Small-N Comparative Studies,” Social Forces, Vol. 72, No. 4 (June 1994), pp. 1225-1237; Irving M. Copi and Carl Cohen, Introduction to Logic, 9th ed. (New York: Macmillan, 1994), chap. 12, “Causal Connections: Mill’s Methods of Experimental Inquiry,” pp. 479-525; Daniel Little, “Evidence and Objectivity in the Social Sciences,” Social Research, Vol. 60, No. 2 (Summer 1993), pp. 363-396; Daniel Little, Varieties of Social Explanation: An Introduction to the Philosophy of Social Science (Boulder, Colo.: Westview Press, 1991), pp. 35-37; and Charles Tilly, “Means and Ends of Comparative Macrosociology,” in Lars Mjoset et al., eds., Methodological Issues in Comparative Social Science (Greenwich, Conn.: JAI Press, 1997), pp. 43-53.

303

John Stuart Mill, as quoted in Amitai Etzioni and Frederick I. Dubow, eds., Comparative Perspectives: Theories and Methods (Boston: Little, Brown, 1970), pp. 207-208.

304

Douglas Dion, “Evidence and Inference in Comparative Case Study,” Comparative Politics, Vol. 30, No. 2 (January 1998), pp. 127-145; Gary Goertz and Harvey Starr, eds., Necessary Conditions: Theory, Methodology, and Applications (Lanham, Md.: Rowman and Littlefield, 2003). Of course, it remains true that a given condition may be necessary or sufficient in one case but not others. We must distinguish here among claims that a variable is necessary or sufficient in a case, in a recurrent conjunction, or in all cases. A claim that a variable is necessary or sufficient to an outcome in a case asserts that the variable was necessary or sufficient in the causal context or background of all the other variables extant in the case. Ultimately, any such claim is untestable, as we cannot re-run the same history while changing only one variable. A claim that a variable X is part of a conjunction, say, XYZ that is necessary or sufficient to outcome Q can be disproved. An instance of Q in which XYZ was lacking can disprove a claim of necessity for XYZ, while the absence of Q in the presence of XYZ would disprove a claim that XYZ was sufficient. Note, however, that cases in which X was present without Y or Z and cases in which Y and Z were present but X was absent cannot disprove claims that X is part of a necessary or sufficient conjunction XYZ. A claim that X is nec essary for all cases of Q is easily disproved by a case of Q lacking X, and a claim that X is sufficient for all Q is disproved by a case with X lacking Q.

305

Insofar as we can tell from reading Mill’s disquisition, he barely touches upon this obstacle. The possibility of “false negatives” from applying the logic of elimination is missed entirely in Cohen and Nagel’s otherwise robust critique of the methods of agreement and difference. On the other hand, the possibility of false negatives and false positives was clearly recognized by Zelditch, “Intelligent Comparisons,” pp. 299, 300, 306; and more recently by Ragin who uses different terms, “illusory commonalities” and “illusory differences.” See Charles C. Ragin, The Comparative Method: Moving Beyond Qualitative and Quantitative Strategies (Berkeley: University of California Press, 1987), pp. 43, 47, 48.

306

The problems for the comparative method created by a “plurality of causes” (which he refers to as multiple causation) plays a prominent role in Ragin’s discussion. See, for example, Ragin, The Comparative Method, pp. x, xii, 15, 20, 25, 37, 39, 43, 46, 47. Ragin does not use the term “equifinality.” The phenomenon of equifinality is noted also in Ragin’s book, Fuzzy-Set Social Science (Chicago: University of Chicago Press, 2000). It was briefly noted earlier by Zelditch, “Intelligent Compromises,” p. 296. Equifinality is also noted and emphasized as an important constraint on developing viable general laws by Benjamin