1 This involves rejecting the Scholastic (Aristotelean) doctrine of abstraction, a point I owe to John Carriero, ‘The Second Meditation and the Essence of Mind’ in Amélie Rorty, ed., Essays on Descartes’ Meditations (Berkeley: University of California Press 1986) 199-221.

2 Queries about Descartes’ notion of idea begin during his lifetime. See especially the First, Third, and Fifth Sets of Objections to Descartes’ Meditations (by Caterus, Hobbes, and Gassendi). Controversy began shortly after Descartes’ death (between Arnauld and Malebranche) and has resurfaced from time to time ever since. For useful studies on early problems with Descartes’ notion of ‘idea,’ see Robert McRae, “‘;Idea” as a Philosophical Term in the Sevententh Century,’ Journal of the History of Ideas 21 (1965) 175-90, and Monte Cook, ‘Arnauld's Alleged Representationalism,’ Journal of the History of Philosophy 12 (1974) 53-62. For two recent runs at the question whether Descartes’ work (mainly on perception) amounts to a form of Representationalism, see: (a) John Yolton, ‘On Being Present to the Mind,’ Dialogue 14 (1975) 373-80; Robert McRae, ‘On Being Present to the Mind: A Reply,’ Dialogue 14 (1975) 664-6; Thomas Lennon, ‘Representationalism, Judgment and Perception of Distance: Further to Yolton and McRae,’ Dialogue 19 (1980) 151-62; and John Yolton, Perceptual Acquaintance from Descartes to Reid (Minneapolis: University of Minnesota Press 1984); and (b) Nancy Maull, ‘Cartesian Optics and the Geometrization of Nature’ in Stephen Graukroger, ed., Descartes: Philosophy, Mathematics, and Physics (Englewood Cliffs, NJ: Barnes and Noble 1980) 22-40; and Ronald Arbini, ‘Did Descartes Have a Philosophical Theory of Sense Perception?’ Journal of the History of Philosophy 22 (1983) 317-37.

3 The analogous strategy for reconstructing Descartes’ concept of mental representation would be to treat clear and distinct ideas as paradigm cases of mental representations, and to set to one side ‘materially false ideas’ as importantly anomalous. But I must save for a separate occasion the task of justifying and applying this strategy to mental representation.

4 This paper is basically an attempt to gain better control over some ideas suggested in my ‘Descartes on Life and Sense,’ Canadian journal of Philosophy 19 (1989) 163-92. I am indebted to John Yolton (personal communication) for pressing me to clarify what I intend by ‘meaning’ when I attribute to Descartes the view that brain events mean something to the mind (Section III), and more generally for his provocative thesis that Descartes’ theory of sense (in humans) is a form of direct realism. Although I cannot agree with Yolton's thesis in its full generality (since I think it underestimates the impact of Descartes’ struggles with sensations), it is plausible when restricted to Descartes’ treatment of sensory representation (Section IV). I am also indebted to Jonathan Bennett and to an anonymous reader for pressing me to clarify the difference between sensing mechanical properties and having sensations (Section II).

5 A note on chronology: in the period 1630-1633, Descartes wrote The World or a Treatise of Light, the Dioptrics, and the Treatise of Man, roughly in that order. More precisely, as his letters to Mersenne at the time indicate, he began work on The World, interrupted it to write the Dioptrics, then returned to The World, expanding it to include a discussion of man. The two parts of The World were published posthumously in 1664 by Clerselier as Le Monde de M. Descartes, au le Traité de la lumière and l'Homme de René Descartes. The Dioptrics was published in 1637 along with the Discourse on Method.

6 See the Appendix for a list of abbreviations used in citing Descartes’ works. In quoting from the Principles, I follow the Miller and Miller translation.

7 In Dioptrics, Discourse 1, he says: ‘ … I would have you consider light as nothing else, in bodies that we call luminous, than a certain movement or action, very rapid and very lively, which passes toward our eyes through the medium of the air and other transparent bodies … ‘ (At VI 84; CSM I 153; Ol-scamp 67; in quoting from the Dioptrics, I follow the Olscamp translation). In Discourse 5, he treats redness as the ratio of rectilineal to rotational tendencies of the motion of light reflected off an object which he assumes to be red (AT VI; Olscamp 94).

8 I have outlined my reasons for this interpretation in ‘Descartes on Life and Sense’ Canadian Journal of Philosophy 19 (1989) 183-7. Although my reading of Descartes’ final assessment of the ontological status of colours, sounds, etc. falls within the mainstream of Descartes interpretation, it does have dissenters. See, for example, E.M. Curley, ‘Locke, Boyle, and the Distinction between Primary and Secondary Qualities,’ Philosophical Review 81 (1972) 438-64, especially 460-4.

9 Wilfrid Sellars makes this point in ‘Berkeley and Descartes: Reflections on the Theory of Ideas’ in Peter Machamer and Robert Turnbull, eds., Studies in Perception (Columbus, OH: Ohio State University Press 1978) 262. I am deeply indebted to this paper.

There are, of course, legitimate concerns about the referent of the term ‘property’ when Descartes ascribes to the same property formal existence in one substance and objective existence in another. But I do not think we can avoid acknowledging that he does treat properties in this manner. It may help, perhaps, to note that Descartes’ Objective Reality Principle (introduced in Meditation Three) implies that a property cannot exist objectively in a thinker unless it exists formally in a substance. So it is not as if these ‘properties’ actually exist unexemplified, or that they can be thought without their having been exemplified.

10 I will return to this point in II, 5.

11 This passage reworks material from Descartes’ early scientific writings: The World, the Dioptrics, and the Treatise of Man. The latter, written around 1633, explicitly separates the treatment of proper sensibles and common sensibles. The separation is made explicit at AT XI 159; Hall 59.

12 Margaret Dauler Wilson, Descartes (Boston: Routledge & Kegan Paul 1978) 118-19

13 Sellars, ‘Berkeley and Descartes,’ 271

14 Although Margaret Wilson attributes to Descartes the view that the difference is phenomenological (correctly claiming that there is no such thing), her main point is to show that Descartes cannot use this difference to establish the conclusion that colours, sounds, tastes and the like are not real properties of matter. And she is right about that.

15 ‘Are’ instantiated by environmetal physical objects when sensory representation is veridical; ‘can be’ instantiated when representation is non-veridical.

16 This chemical compound conception saddles Descartes with philosophical problems which he never fully resolves. See Sellars’ penetrating discussion in ‘Berkeley and Descartes’ (259-78).

17 This, at least, is one way to make sense of Prine. I 69: ‘ … size, in a body which has been observed, or figure, or motion … are known by him in a manner very unlike that in which he knows, in the same body, what colour is, or pain, or odor, or flavor, or any of the other things which I have said must be referred to the senses’ (AT VIIIA 33-4; HR I 248; CSM I 217-18; MM 31; emphasis added).

18 I owe these examples to Nollaig MacKenzie.

19 Indeed, in some contexts, as Margaret Wilson points out, Descartes even uses the terms ‘represent’ and ‘exhibit’ interchangeably (Descartes, 102, and note 2, 232).

20 Evidently Descartes does not view the brain state as the total efficient cause of any mental state, including sensations. The mind is crucially (causally?) required. This can be seen by noting that, according to Descartes, non-human animals, some of which do have relevantly the same brain states, but which lack minds, do not have sensations.

21 I make no attempt here to identify the historical target (if there is any single one) of Descartes’ attack, and therefore no attempt to evaluate his criticism. David Lindberg's spellbinding work on the history of optics and theories of vision displays vividly the complex, convoluted role played by the notion(s) of form in scientific theorizing about the nature and propagation of light and colour. See especially ‘The Science of Optics’ in David C. Lindberg, ed., Science in the Middle Ages (Chicago: University of Chicago Press 1978) 338-68, and Theories of Vision from Al-kindi to Kepler (Chicago: University of Chicago Press 1976).

22 Robert Stalnaker, Inquiry (Cambridge, MA: M.I.T. Press 1984) 12-13. Stampe's groundbreaking paper is ‘Toward a Causal Theory of Linguisitic Representation’ in P. French, T. Uehling, & H. Wettstein, eds., Contemporary Perspectives in the Philosophy of Language (Minneapolis: University of Minnesota Press 1979) 81-102.

23 Stalnaker, 12

24 This term belongs to Dennis Stampe.

25 Stalnaker, 12

26 Stalnaker, 12-13. I am departing from Stalnaker's treatment here, since he is grinding an axe not directly related to my project.

27 Stalnaker, 13

28 It should be evident from this that the notion of natural indication here specified differs significantly from Fred Dretske's ‘meaning-n’: natural indication accommodates misrepresentation whereas meaning-n does not.

29 Particularly uncontrolled are his speculations about neuromechanics. I hasten to add that his treatment of the mechanisms of depth perception is deeply informed by his optics and by his improved understanding of the physiology of the eye.

30 I am not certain that Descartes thinks there is a function, strictly speaking, because he may not think that the uniqueness requirement is satisfied. More on this shortly.

31 The Dioptrics, Discourse 6

32 ‘Supervenient upon’ in the sense defined by Jaegwon Kim. See his ‘Causality, Identity, and Supervenience in the Mind-Body Problem,’ Midwest Studies in Philosophy 4 (1979) 31-49 and ‘Epiphenomenal and Supervenient Causation’ Midwest Studies in Philosophy 9 (1984) 257-70.

33 This is the source of my doubt that Descartes envisions a function mapping corporeal images onto features of the environment: given M-m pairs, there will be no unique value of f(a).

34 ‘Both’ if one relaxes Stalnaker's requirement that there be a function; ‘neither’ if one does not. The natural choice is the former, treating natural indication as a relation.

35 I am indebted to Romane Clark for the basic idea that direct realist theories of perception are best construed as making a claim about semantics. See his ‘Considerations for a Logic of Naive Realism’ in Peter Machamer & Robert Turnbull, eds., Studies in Perception (Columbus, OH: Ohio State University Press 1978), 525-56.

36 I am indebted to Mildred Bakan for comments on ‘Descartes on Life and Sense’ which forced me to recognize this point.

37 My thinking here is influenced by the hard conceptual work done in Marilyn McCord Adams’ ‘Ockham's Nominalism and Unreal Entities,’ Philosophical Review 86 (1977) 144-76.