1. I am greatly indebted to Stanley Tweyman for sparking my interest in this intriguing work. His suggestions and comments on earlier drafts of this paper were extremely valuable. I have also relied upon his book, Scepticism and Belief in Hume's "Dialogues Concerning Natural Religion" (Dordrecht, 1986), and the Introduction to his edition of the Dialogues (London; New York, 1991).
2. References to David Hume's Dialogues Concerning Natural Religion are given first to the Kemp Smith edition (Indianapolis, 1947), and second to the Tweyman edition (above, n. 1); for example, D 227/184.
3. The importance of the irregular argument is also emphasized by Tweyman (above, n. 1) in the Introduction to his new edition of the Dialogues. In his discussion of such arguments in his analysis of part 3, he states that "the task of the commentator is now also seen to be more complicated, inasmuch as an understanding of the Dialogues requires not only seeing the connection between the analogical argument presented in Part 2 and the final pronouncement in Part 12 'that the cause or causes of order in the universe probably bear some remote analogy to human intelligence' (D 227), but also determining the connection (if any) between the natural argument advocated in Part 3 and this final pronouncement." It is to this task that this paper is directed.
4. In the interpretive essay to his edition of the Dialogues, Nelson Pike states that the irregular argument is worthy of consideration, but he is not convinced it is correct. He regards both the regular and irregular arguments to be instances of rational activity, but a different kind of activity is involved in the two cases—the difference lies in the fact that the inference in the irregular argument is drawn directly from the data. However, according to Pike, Philo's 'confession of faith' occurs at D 217-18—the expression of the remote analogy proposition. Hume: Dialogues Concerning Natural Religion (London, 1985)
13. In parts 4 and 5 he disputes the rational nature of the world, and thus an inference to a rational cause, and points out the consequences of holding an anthropomorphic conception of God (God's finitude); in part 6, he proposes that the nature of the world might be organic and not mechanical; in part 7, he offers principles of generation other than reason; and he claims in part 8 that the world need not have a designing principle. In parts 10 and 11, he shows Cleanthes that nothing can be said with respect to God's benevolence.
15. Part 12 is devoted to the correction of the 'undistinguished' doubts, i.e., ones that do not differentiate between what is acceptable and unacceptable in an argument; see Tweyman, Scepticism and Belief (above, n. 1), 122. These doubts have been corrected by the end of the eighth paragraph of part 12: as "there is 'some remote inconceivable analogy' between God and human intelligence, it is obvious that no additional sceptical attack is either anticipated or...