The simplest variants of the agreement method can only be used in cases where our prior knowledge narrowly limits the possible causes and justifies the belief that they will be sensitive. For example, if the character of a disease is bacterial, the microorganism responsible can be identified by discovering that only one type of microorganism, which is not already known as innocent, is present in a number of cases of the disease. Otherwise, observation of the only common factor in a number of cases of a phenomenon can only be used very timidly to suggest a hypothesis that needs to be tested in another way. Under the tailings method, if we have a number of factors that are assumed to be the causes of a number of effects, and we have reason to believe that all factors, with the exception of a factor C, are causes of all effects, with the exception of one, we should infer that C is the cause of the residual effect. The general nature of Mills` methods in the experimental study can be illustrated by examples of the two simplest methods, methods of concordance and difference. Mills` canon for the agreement method is as follows: “If two or more cases of the phenomenon under review have only one circumstance in common, the fact that all instances are identical is the cause (or effect) of the given phenomenon.” These methods have been criticized on two main points: first, it is claimed that they do not define the conclusions envisaged, so that they are not methods of evidence or conclusive demonstrations; and second, that they are not useful as investigative methods. This criticism has been used to support the general finding that these methods play little or no role in the exploration of nature and that the scientific method requires a radically different description. [iii] Skocpol (1979) argued that there were only two fundamental causes of social revolutions: (i) a rigid agricultural class structure and (ii) a state crisis. With only three cases, she had two degrees of freedom; If it had looked at four cases, it could have concluded that another factor – perhaps ideology, industrialization, leadership or some other factor – was an important cause. [iv] Under the title “The Logic of Counterfactual Analysis in Historical Explanation,” the conference was given on April 15, 2016 at the University of Michigan Department of Political Science. [v] Like Mahoney, I will not deal with probabilistic causality separately in this article. [vi] In the published work, Mahoney (2008) attempts to reconcile INUS causality with the necessary and sufficient logic of the Millian method, although Mackie himself (1965:245-52) proposed INUS conditions as a means of going beyond the inadequacies of Millian`s design. [vii) Ragin (1989) advocates something similar that still falls within the variable paradigm, but a comparison between his approach and critical realism goes beyond the scope of this contribution.
[viii] My representation of combined and conjoue causal may resemble one or more of Mahoney`s variants of “internal case analysis” in which “analysts can.