To close the list, that is, to show that certain factors are not relevant, we must use an analogue of the compliance method. If, as before, we assume that the total cause of P in F is many factors (X, X′, X”, etc.), but also that P reacts to all these factors in the sense that for each variation, z.B. in X, while X′, X, etc. remain constant, P varies and X, X′, X, X”, etc. are identical to some of the possible causes A, B, C, D, E, if we observe that P remains constant, while, for example, A, C, D and E remain constant, but B varies, we can conclude that B is irrelevant, that none of the X is identical to B. In general, we will look for a condition that is both necessary and sufficient for the phenomenon, but there are variants of methods in which we look for a condition that is only necessary or sufficient. However, in practice, these are conditions that are not necessarily necessary or sufficient, but are rather necessary and/or sufficient for a field, i.e. certain substantive conditions that can be specified more or less precisely. For example, we are not interested in the cause of a particular disease in general, but in what it causes in people who live on earth, breathe air, etc. Again, this is not the cause of hardness in general, but the cause of above-average hardness in iron under ordinary circumstances and at ordinary temperatures. The field in which we seek a cause of a phenomenon must be such that the phenomenon sometimes occurs in this area and sometimes not. We can assume that this field consists of the presence of certain qualities or, at the very least, certain general descriptibility characteristics, and not by a particular place.
The accompanying variation method says that if we find in a number of situations that lead to a particular effect, a certain property of the effect, which varies with the variation in a factor common to those situations, then we can deduce this factor as cause. . . .