Remarks on the Concept of Distribution in Traditional Logic
David Makinson · Noûs · 1969
In traditional logic, a term is often defined to be distributed in a given categorical proposition if proposition says something about every instance of term, or of its extension, rather t-han merely some of its instances or of its extension. Given definition, subjects of universals and predicates of negatives are classified as distributed, whilst subjects of particulars and predicates of affirmatives are taken to be undistributed. The concept of distribution is then used in formulating rules for valid inference, particularly in syllogistic. As is well known however, concept of distribution has in recent years been subject to a good deal of criticism. It is charged with being obscure to point of incoherency. There is no apparent way of spelling out clearly just what is meant by saying that a proposition is about the whole or only part of extension of a term. In so far as concept can be understood at all, it seems to leave rather problematic traditional claim that predicates of particular negatives are distributed. Thus, it is argued, if concept of distribution must be used in syllogistic, it should be defined in a purely enumerative way: subjects of universals, predicates of negatives. Such a view has been argued at length by P. T. Geach in his Reference and Generality (New York: Cornell University Press, 1962), and more briefly in articles in Mind LXV (1956): 67-74 and LXXVII (1968): 113-114. Related views are also expressed by L. Linsky in his review of Reference and Generality, Mind LXXIII (1964): 575-583, and appear to be widely accepted.