1 See, for example, Posterior Analytics 11.19, Physics 1.1.

2 As we'll see, Ross thinks Aristotle himself is unsure of the meaning of ‘incomposite’; for the description of the view as ‘mystifying,’ see Bostock, David Aristotle Metaphysics Books Z and H (New York: Oxford University Press 1994), 244Google Scholar

3 The translation is Ross's, from the Revised Oxford Translation.

4 We don't believe Aristotle's intended distinction between the cognition appropriate to simples and that appropriate to composites can be upheld if we read his claim about contact with simples as amounting only to the claim that while we know these entities without error, we might not know ail there is to know about them. Thus to the suggestion that cognition of simples is like touching in precluding error, but not in knowing all there is to know, we reply that this would be to destroy Aristotle's intended distinction. For if an episode of thought-as-contact leaves more of the object still to know, the nature of the object as a simple is threatened, and there seems no way of precluding error about the remaining unknown feature(s). Put differently, if contact only means ‘correct but partial grasp,’ why is the only contrast with contact complete ignorance (agnoia), as Aristotle claims at 1052al?

5 There is strong evidence in De Anima that Aristotle does connect the issue of the possibility of falsehood with the issue of combination: ‘for falsehood always involves combination (to gar pseudos en sunthesei aei),’ 430bl-2. But here the point seems to be that when a thought is a compound formed by synthesis (i.e. when it's the mental correlate of an affirmation which says one thing of one thing), then there can be falsehood. The contrast is not between these thoughts and thought-constituents, but these thoughts and thoughts that are compound, but not as the result of synthesis. The thoughts that signify essences, for example, will be compound in the sense of including different elements, but these elements won't be put together in the way that involves synthesis (predicating one thing of another).

6 In discussing SI and MI separately, we do not mean to rule out the possibility of seeing both SI and MI elements at work in these passages. Indeed, Philoponus interprets the De Anima passage this way. If we are correct that neither strategy successfully secures the epistemological doctrines, however, it's not clear how combining them will help matters.

The reason we discuss the version of SI defended by Wedin is that it explains why the epistemological doctrines are supposed to follow completely independently from MI, so that it counts as a genuine alternative to it.

7 Philoponus, On Aristotle ‘On the Soul 3.1-8’, Charlton, William trans. (Ithaca: Cornell University Press 2000), 124.Google Scholar

8 Wedin, Michael. Mind and Imagination in Aristotle (New Haven: Yale University Press 1988).CrossRefGoogle Scholar SI is also suggested by W.D. Ross, in his commentary on the Metaphysics, although he takes Aristotle himself to be somewhat unclear about what he means by ‘incomposite.’ See Ross, W.D. Aristotle's Metaphysics (New York: Oxford University Press 1953), 276.Google Scholar

9 Wedin, Mind and Imagination, 131.Google Scholar Perhaps the motivation for this idea is along the lines suggested by G. E. Moore in ‘The Nature of Judgment.’ Moore there argues that unless we grasp the constituents of judgments in a way that doesn't involve judgment, we will be off on a vicious regress that will render judgment itself impossible. See Moore, G.E. ‘The Nature of Judgment,’ Mind 8 (1899), 178.Google Scholar

10 See also Metaphysics XII, 1075a6-7.

11 In the De Anima discussions, the word for simple is ‘adiaireton,’ rather than ‘asuntheton.’ Aristotle seems to go back and forth between these locutions. See (e.g.) Metaphysics 1075a 5-7.

12 For fuller discussion, see Berti, Enrico ‘The Intellection of “Indivisibles” According to Aristotle, De Anima III 6,Google Scholar’ Aristotle on Mind and the Senses, Lloyd, G.E.R. and Owen, G.E.L. eds. (Cambridge: Cambridge University Press 1978) 141-63.Google Scholar

13 The remaining translations from the Metaphysics largely follow Irwin, Terence and Fine, Gail Aristotle Selections (Indianapolis: Hackett 1995).Google Scholar

14 Here we accept (with Frede, Irwin et al.) the suggestion of aisthêtikon, rather than aisthêton, but our view of the passage can be defended without it. See Wedin, Michael Aristotle's Theory of Substance (New York: Oxford University Press 2000), 330-1.Google Scholar

15 Here we have been helped by Wedin, Substance, 327-330.

16 Frede, Michael ‘The Definition of Sensible Substances in Met. Z,’ Biologie, Logique Et Metaphysique chez Aristote, Devereux, D. and Pellegrin, P. eds. (Paris: Editions du CNRS 1990), 121Google Scholar

17 For the suggestion of functional parts, see Wedin, Substance, 338Google Scholar n. 76. Irwin worries that if the definition of the form doesn't include material parts, the response will be ‘irrelevant’ to the opposing view. But if Aristotle secures the conclusion that there cannot be a man without the material parts, either by an entailment relation between functional and material parts, or by his account of the relation between the enmattered kind and individuals falling under it (or both), then Aristotle's response need not be seen as an ignoratio, even though material parts remain outside of the definition of the form. See Irwin, T.H. Aristotle's First Principles (New York: Oxford University Press 1988) 246.Google Scholar

18 Wedin, Substance, 337.Google Scholar

19 For the inferential view, see also Hartman, Edwin Substance, Body and Soul (Princeton: Princeton University Press 1977), 65.Google Scholar

20 Charles, David Aristotle on Meaning and Essence (New York: Oxford University Press 2000), 279Google Scholar n. 11

21 The distinction between form and kind is also deployed in this way by Loux, Michael in Primary Ousia (Ithaca: Cornell University Press 1991), 175-8.Google Scholar See also Modrak, Deborah K.W. Aristotle's Theory of Language and Meaning (Cambridge: Cambridge University Press 2001), 188.Google Scholar

22 For a similar treatment of this passage, see Gill, Mary Louise Aristotle on Substance (Princeton: Princeton University Press 1989), 131-3.Google Scholar

23 Thus Frede: ‘What gets defined, on the view in Z 10-11, are no longer composite sensible substances, but their forms or essences, man not in the sense of a composite sensible substance, but in the sense of the form of man. Given this it is no surprise that Aristotle should claim that there is no place in the definition of man for a reference to matter or the material parts’ (Michael Frede, ‘Definition,’122). The view that definitions of forms are free of matter is also defended by Gill, Aristotle on Substance, 127;Google Scholar Lewis, Frank Substance and Predication in Aristotle (Cambridge: Cambridge University Press 1991), 303;Google Scholar Witt, Charlotte Substance and Essence in Aristotle (Ithaca: Cornell University Press 1989) 191-3.Google Scholar For the opposing view, see T.H. Irwin, Aristotle's First Principles, 245-7. The differences between the view we defend and Irwin's view should not be overstated. Irwin argues that definitions of these forms include matter, in a way: ‘Still, the definition includes [material] parts only in a certain way, as defined by their capacities and functions, so that the essence is a formal compound (or materiate form) rather than a material compound.’ But Irwin would reject our claim that the souls of natural substances are incomposites.

24 James Lennox writes: ‘To give a definition of a thing's being is to State “what it is to be” for that thing. Such a definition refers primarily or exclusively to form viewed in abstraction from matter’ (Lennox, James Aristotle: Parts of Animals [New York: Oxford University Press 2000], 151Google Scholar).

25 We borrow this phrase from Plato: Theaetetus, McDowell, John trans. (New York: Oxford University Press 1973), 197.CrossRefGoogle Scholar

26 Halper, Edward One and Many in Aristotle's Metaphysics (Columbus: Ohio State University Press 1989), 222Google Scholar

27 ‘How Simple are Plato's Forms?’ Ancient Philosophy 22 (2002) 277-88.

28 This is Russell's language in The Philosophy of Logical Atomism. Russell, Bertrand The Philosophy of Logical Atomism (Chicago: Open Court 1985), 63.Google Scholar

29 One may worry that by admitting that an entity may have one property, as in SP, we are unable to stop that entity from having an infinite number of properties. For any property F, that a has, it may be thought also to have the property (F v G), and (F v H), and (F v G v H) and so on. Thus SP is an unstable position, entailing as it does that any entity that has one property has an infinite number.

The quick way to rebut this, of course, is to deny disjunctive properties by initially shifting the burden onto those who wish to introduce them. The way to do this is to show how strange and unpalatable such properties are: if one is prepared to accept (F v G) as a property, then what about the equivalent property ∼(∼F & ∼G)? Should one take disjunctive properties seriously, one would presumably have to take negative properties seriously as well. For further discussion, see Mellor, D.H. 'Predicates and Properties,’ in Properties, Mellor and Oliver, eds. (Cambridge: Cambridge University Press 1997), 264-5,Google Scholar and Clapp, Lenny ‘Disjunctive Properties: Multiple Realizations,’ journal of Philosophy 98 (2001), 124-32.Google Scholar

30 Monadology, in Philosophical Essays (Indianapolis: Hackett Publishing Company 1989), §1

31 Monadology, §8

32 Rene, Descartes Rules for the Direction of the Mind, Haldane and Ross, trans. (New York: Dover Publications 1931), 42Google Scholar

33 Wedin, Substance, 341Google Scholar

34 For defense of the view that forms are immaterial and changeless, see Heinaman, Robert ‘Aristotle and the Mind-Body Problem,’ Phronesis 35 (1990) 83–102.CrossRefGoogle Scholar

35 Frede, ‘Definition,’ 128;Google Scholar Halper, One and Many 222

36 Lloyd, G.E.R. Aristotelian Explorations (Cambridge: Cambridge University Press 1996), 67–82Google Scholar

37 See Balme, D.M. ‘Aristotle's use of division and differentiae,’ in Philosophical Issues in Aristotle's Biology, Gotthelf, A. and Lennox, J. eds. (Cambridge: Cambridge University Press 1987), 79;Google Scholar see also Lloyd, Explorations, 49.Google Scholar

38 Lloyd, Explorations, 74Google Scholar

39 We'd like to thank Butler's colleagues at Iowa State, the members (including visitor Robert Bolton) of the 2003 NEH Seminar ‘Aristotle on Meaning and Thought,’ and especially its directors, Mark Wheeler and Deborah Modrak. Finally, this paper would not exist in its present form without the valuable comments of two anonymous referees for this journal. Many thanks to them.