1 Seventh Replies at AT 7.536. I am using the translation in The Philosophical Writings of Descartes, John Cottingham et al., ed. (Cambridge: Cambridge University Press 1984), but my references will all be to pages in vol. 7 of the Adam and Tannery edition of Descartes (AT 7).

2 Second Replies at AT 7.145f. Cottingham's translation goes beyond the Latin in making Descartes say that we are certain of ‘these truths’ and express them with the phrase ‘the fact that.'

3 Harry G. Frankfurt, Demons, Dreamers, and Madmen (Indianapolis: Bobbs-Merrill 1970), 164

4 CB 81. That is, Descartes, Conversation with Burman, John Cottingham, ed. (Oxford: Oxford University Press 1976) piece no. 81 (49-50).

5 John Etchemendy, ‘The Cartesian Circle: Circulus ex tempore,’ Studia Cartesiana 2 (1981) 5-42, at 17

6 For good evidence that Descartes did not confine himself to doubts based on good reasons, see Jeffery Tlumak, ‘Certainty and Cartesian Method,’ in Michael Hooker, ed., Descartes: Critical and Interpretive Essays (Baltimore: The Johns Hopkins University Press 1978) 40-73, at 47f.

7 The difference between ‘Perhaps not-P’ and ‘Perhaps I am wrong in thinking that P’ is widely neglected in the secondary literature. A notable exception is Gareth B. Matthews, ‘Descartes's Cogito and Katz's Cogitations,’ Pacific Philosophical Quarterly 68 (1987) 197-204.

8 Edwin M. Curley, Descartes Against the Skeptics (Cambridge, MA: Harvard University Press 1978), 93. See also 95.

9 Principles 1:45. I reluctantly follow the custom of using ‘clear’ and ‘distinct’ to translate Descartes’ words, though they are certainly incorrect. ‘Vivid’ and ‘clear’ - in that order - would be nearer the mark.

10 Bernard Williams, Descartes: The Project of Pure Inquiry (London: Pelican Books 1978), 183

11 Margaret Wilson, Descartes (London: Routledge and Kegan Paul 1978), 141f.

12 See, for example, A.K. Stout, ‘The Basis of Knowledge in Descartes,’ Mind 38 (1929) 330-42, 458-72; reprinted with omissions in Willis Doney, ed., Descartes: A Collection of Critical Essays (Notre Dame: University of Notre Dame Press 1967) 167-91; Willis Doney, ‘The Cartesian Circle,’ Journal of the History of Ideas 16 (1955) 324-48; Bernard Williams, 191n., accuses himself of having ‘firmly expressed’ the memory answer on 351 of his ‘Descartes, Rene’ in Paul Edwards, ed., The Encyclopedia of Philosophy (New York: Macmillan 1967), but the accusation is false. Perhaps Williams had the memory answer in mind when writing that article, but he didn't get it onto the page.

13 The most influential presentation of the case against it seems to have been Harry Frankfurt, ‘Memory and the Cartesian Circle,’ Philosophical Review 71 (1962) 504-11. It is also rejected in Curley, Descartes Against the Skeptics, 102-4; John Cottingham, Descartes (Oxford: Basil Blackwell 1986), 77f.; Williams, 191-8; Anthony Kenny, Descartes: A Study of his Philosophy (New York: Random House 1968), 187f.; and in many other places.

14 Norman Kemp Smith, New Studies in the Philosophy of Descartes (London: Macmillan 1952), 273ff.; John Etchemendy, The Cartesian Circle.ߣ

15 In the Second Replies at AT 7.141 Descartes writes that the atheist ‘cannot be certain that he is not being deceived on matters which seem to him to be very evident.’ Because of the present tense ‘seem,’ that conflicts with the memory account, the continuing-truth account, and the now standard account which I am going to present. It must be a slip.

16 Descartes writes ‘ … what is required for my being certain about anything,’ but that must be a slip. The conclusion is about what suffices, not what is needed.

17 I was made aware of this point by Rudy Garns, ‘Descartes and Indubitability,’ Southem Journal of Philosophy 26 (1988) 83-100.

18 For example, Curley, Descartes Against the Skeptics, ch. 5; Kenny, Descartes, 192f.; Jeffery Tlumak, ‘Certainty and Cartesian Method,’ 58ff.; Alan Gewirth, ‘The Cartesian Circle,’ The Philosophical Review 50 (1941) 368-95 at 371-3; James Van Cleve, ‘Foundationalism, Epistemic Principles, and the Cartesian Circle,’ The Philosophical Review 88 (1979) 55-91, at 66-71; Frankfurt, ‘Descartes's Validation of Reason,’ American Philosophical Quarterly 2 (1965) 149-56; John Cottingham, Descartes (Oxford: Basil Blackwell 1986), 66-70.

19 That is how Spinoza reads him, in expressing the truth rule thus: ‘The things to which we must necessarily assent when we perceive them distinctly must be true’ (Descartes's Principles 1p14s; Gebhardt I/172; in The Collected Works of Spinoza, vol. 1, E. Curley, ed. [Princeton: Princeton University Press 1985], 256).

20 ‘The crucial issue is whether we can know certain propositions prior to proving God; the observation that there are certain moments when we cannot for the moment doubt is epistemically irrelevant’ (Wilson, 133).

21 Panayot Butchvarov, The Concept of Knowledge (Evanston: Northwestern University Press 1970), 84

22 Strictly, the truth rule could assure me now of truths which I do not now Distinguish but which I know I did or will Distinguish, or which I know someone else did, does or will Distinguish. It releases me from confinement to I-now.

23 I use ‘admits’ to render ‘agnoscat.’ Cottingham uses ‘recognizes,’ which is truth-entailing. Either translation is possible, but mine fits better the logic of the passage. If ‘recognizes’ is right, its truth-entailing element is idle.

24 AT 7.144f. (i) My factual ‘cause for doubting’ replaces Cottingham's normative ‘reason for doubting.’ The Latin is habere causam dubitandi. Either rendering is possible, but mine fits the context better: firmness of conviction militates against causes not against reasons for doubt. (ii) My factual What do we care … ?’ is faithful to the Latin ‘Quid cur am us … ?’ in a way that Cottingham's normative ‘Why should we … ?’ is not.

25 AT 7.146. Where I have ‘these,’ Cottingham has ‘such truths’; but that goes beyond the Latin.

26 Letter to Mersenne, 6 May, 1630, in Descartes: Philosophical Letters, Anthony Kenny, ed. (Oxford: Clarendon Press 1970), 13f.

27 Letter to Arnauld, 29 July, 1648, in Descartes: Philosophical Letters, 236f.

28 I have had helpful comments from David Copp, John Cottingham, Jack Davidson, Gareth Matthews, and a referee for this journal. Don Garrett's suggestions in connection with an early draft of this paper were especially useful. William Alston's careful comments on a later draft transformed my thinking; without his help the paper would have been much worse than it is.