1 Two basic explanatory principles are the principle that perception is a kind of alteration (416b34) and the principle that the sense is potentially what its object is actually (418a4; 425b26).

2 Both cases of apperception are initially discussed in 111.2. At 425b12-26 Aristotle argues that it is by sight that we perceive that we see and at 426b12-427a14 he argues that one faculty must perceive that sweet is different from white; he reconsiders apperceptive Judgments of this form at 431 a20-b2. Perceiving that we perceive is also discussed at de Somno 455a17, and perceiving that one sense object differs from another at de Sensu 447a13-449a22.

3 For the sake of brevity, I shall discuss in this paper Aristotle's theory of perception only as it is found in the de Anima. Elsewhere, I have considered his theory of perception as found in the Parva Naturalia treatises and in de Anima. See ‘An Aristotelian Theory of Consciousness?’, in Ancient Philosophy, Spring 1981.

4 For an insightful treatment of the awareness of perceiving, see L.A. Kosman, ‘Perceiving That We Perceive: On the Soul 111.2,’ in The Philosophical Review, 84 (1975) 499-519. Since I agree with Kosman's treatment of perceptual self-awareness, I shall concentrate on apperceptive Judgments about differences among sense objects in this paper.

5 For the sake of simplicity, I shall ignore differences in detail and group together, under the label ‘the traditional view,’ the commentators who take the general line that the common sense performs a number of functions including the awareness of occurrent perceptions and the recognition of differences among sense objects. This view can be found in the commentaries of Hicks, Rodier and Ross. Aristotle De Anima, comm. by Hicks, R.D. (Amsterdam: Hakkert 1965), 426-52;Google Scholar Aristote Traité De L'Ame, trad. et ann. par Rodier, G. (Paris: Ernest Leroux 1900), II, 265-8;Google Scholar 365-96; Aristotle De Anima, comm. by Ross, David (Oxford: Clarendon Press 1961) 306;CrossRefGoogle Scholar Aristotle Parva Naturalia, introd. and comm. by Ross, David (Oxford: Clarendon Press 1955) 35.CrossRefGoogle Scholar

6 Cf. Rodier, op. cit., 265-8.

7 (1) Aristotle's De Anima Books II and III, trans. and comm. by Hamlyn, D.W. (Oxford: Clarendon Press 1968) 128-9.Google Scholar (2) Hamlyn, D.W. ‘Koinē Aisthēsis,’ The Monist, 52 (1968) 195–209.CrossRefGoogle Scholar

8 See Hamlyn (1), 128-9.

9 ‘Kainē aisthēsis’ is used at de Memoria 450a10 for the faculty that presents the memory image and at de Partibus Animalium 636a27 in a list of faculties belonging to the divine part of humans.

10 This is Hamlyn's translation (op. cit.) (1), which I shall use throughout, unless otherwise noted. For the moment, I propose to accept the text as it is. The exegetical problem concerning the occurrence of kinēsei (a17) will be discussed below.

11 Among the commentators who have found this passage puzzling are: Hicks, op. cit., 428; Ross, op. cit., 270; Torstrik, Adolf Aristotelis De Anima Ill, Libri (Berlin: Weidmann 1862) 162-3.Google Scholar Hamlyn, op. cit. (1), surprisingly finds it unproblematic. He suggests that the reference is to the mental and physical movement involved in comprehending the common objects. This is somewhat reminiscent of Philoponus who takes the referent of kinēsei to be the movement set up in the sense by the external object. Ioannis Philoponi, In Aristotelis De Anima Libros Commentaria, ed. Michael Hayduck (Berlin 1897) 458. It should be noted that on either of these interpretations being perceived by motion would not be a distinctive feature of the perception of the common objects (cf. Hicks, 428).

12 Although the wording of 425a17 does not indicate that kinēsei is being used as an example, the statement that number is perceived by the negation of continuity and by the special objects (tois idiois) at 425a19, the statement that each sense perceives one thing at 425a20, and the use of hoion with kinēseos at 425a21 make it clear that Aristotle is using it in this way.

13 It might be thought that if this is Aristotle's position, he would have extended the argument to include the other koina but it seems equally likely that he would have left the extension to his audience. Moreover, the alternative is to take the use of kinēsei literally and make the perception of all the koina dependent upon movement, but then we must explain why Aristotle does not characterize the perception of all the koina in terms of kinēseis in this passage or elsewhere.

14 See 424b22-425a13. A full discussion of this difficult passage cannot be undertaken here. However, it is hoped that the following discussion will suffice to show that the argument Aristotle gives to show that there are only five senses leaves much to be desired. Aristotle seems to argue from the material constituents of the sense organs to the number of senses. He divides the senses into touch which perceives by direct contact and those senses which perceive through media. The latter (sight, hearing and smell) have organs made of air or water, so apparently more than one organ has the same primary constituent. Aristotle is not sure whether fire belongs to all organs or to none or whether earth belongs to none or to touch only. Clearly the number of elements does not determine the number of sense organs, so if Aristotle's argument turns on the relationship between the number of constituents of the sense organs and the number of senses, as it seems to, it fails. Alternatively, we might suppose that Aristotle is arguing that there are only five because a sixth would be superfluous since five are adequate to perceive the physical reality made up of the four elements, but Aristotle does not put the argument in this way.

15 The argument that there is no special sense for the common sensibles found at 425a14-29 divides into two parts. The first concludes with hoste dēlon … kinēseos at 425a20-1. Houto at 425a21 introduces a further argument, which also concludes with the statement that there is no special sense for the koina.

16 It has been suggested that kata sumbebēkos at 425a15 is a corruption since it appears to contradict ou kata sumbebēkos at a28. Philoponus (op. cit.) denies that Aristotle is here expressing his own view, and Torstrik (op. cit.) wants to emend the first phrase to ou kata sumbebēkos; this leads both to misconstrue Aristotle's argument here, making it turn upon the 11.6 statement that the perception of theta koina is kath'hauta. Other suggestions have been made: Ross (op. cit.) suggests that Aristotle uses kata sumbebēkos initially to stress that ta koina are not special objects and then ou kata sumbebēkos to contrast them with incidental objects like Clean's son. Hamlyn (op. cit.) suggests distinguishing between the perception of ta koina by one sense and all the senses. Neither of these explanations is wholly satisfactory. Nonetheless, retaining kata sumbebēkos as an expression of Aristotle's view seems the most reasonable course at this stage in our investigations — otherwise we run the risk of prejudicing our findings about the nature of this sort of perception.

17 The use of amphoin (of both) makes it clear that Aristotle is using the kath ‘hauta perception of two special objects to explain the kata sumbebēkos perception of a special object.

18 The kath ‘hauto I kata sumbebēkos distinction is found in a number of Aristotle's works. It typically marks the distinction between essential and incidental attributes. See Po.An. 73b4, 74b12, 75a20; Top. 102b3, 103b17; Met. 1018a1, 1025a14, 1026b32, 1059a1, 1065a1. I take this distinction, when applied to perceptual objects, to indicate the causal role of the object in perception. If the object acts directly on the sense, it is perceived kath' hauto. Hamlyn takes 'kath'hauto’ to indicate the definitional dependence of the sense upon its object. ((1), 105) I agree that kath'hauta objects play this role as well but I take it to be a consequence of their causal role.

19 Aristotle describes three kinds of perception as kata sumbebēkos: the perception of the son of Diares (418a20) (or the son of Cleon, 425a25) the perception of sweet by sight (42Sa30), and the perception of a common sensible by a special sense (425a 15). It is easy to see that all three cases fit my characterization of kata sumbebēkos perception. Although Wittgenstein's discussion of seeing as (Philosophical Investigations ll.xi) suggested the formula, ‘perceiving X as Y’ to me, because it seemed to capture the notion of perceiving something over and above the pattern of sensory stimulation, I have adapted this notion to fit Aristotle's conception of incidental perception. I use it to explicate the relationship between the sense object(s) that is (are) directly affecting the perceiving sense(s) and other concurrent perceptions.

20 Since a perceptual faculty and its object are one in the act of perception, a unity in one is ipso facto a unity in the other. Thus, it makes little difference whether we take hēi mia to refer to the unity of the sensible qualities or the unity of the senses.

21 For this reason, I do not see how Hamlyn's claim that the perception of the common objects does not require a unity of the senses ((1). 128) can be right.

22 This is Hamlyn's translation with one change. I have translated ‘aisthēsei’ (b14) as ‘by a sense’ instead of ‘by perception,’ because I think it makes Aristotle's meaning clearer.

23 Kosman, op. cit., argues that Aristotle's failure, to draw this distinction in the case of the apperceptive awareness of perceiving, follows from his conception of the relationship between the perceptual act and its object. Aristotle identifies the perception of X with the actualization of X as a sense object, so the awareness of the perception is ipso facto the awareness of the sense object.

24 Aristotle's theory of perception is formulated to apply to normal percipients. Although discriminating between white and sweet would not be essential to the perceptual experience of a person blind from birth, this case would not count against Aristotle's view, since this person would not be a normal percipient.

25 That is, the perception of the difference is a direct result of the senses’ being affected by the white and sweet object.

26 Cf. de An. 425b16.

27 Presumably, Aristotle thinks that since the sense and the object are one in the act of perceiving, the motions which are the sensings of different objects stand in the same relation to each other as the objects do. Contrary objects would produce contrary motions. Aristotle thinks it is impossible that the same sense should be affected by contrary motions at the same time, because these motions would cancel each other.

28 Cf.de An. 431a24.

29 The two words Aristotle uses here for point are: stigmē (a 10) and peras (a 13). The former is used technically for mathematical points. The latter, not a technical term, typically means end or limit. Ross follows Hick's interpretation, according to which the point divides a line into two segments and is the terminus of one segment and the beginning of the other. The disadvantage of their interpretation is obvious: only two faculties could be so related. In addition to represent the five senses as segments on one line tends to obscure their distinctness. There is a precedent for our interpretation. Alexander Aphrodisiensis takes stigmē to refer to the center of a circle at which radii terminate (Alexandri Aphrodisiensis Praeter Commentoria Scripta Minora: De Anima Liber Cum Mantissa, ed. I. Bruns (Berlin: 1887) 63; Scripta Minora Reliqua 96). Charles Kahn, who does not specifically discuss this analogy, seems to hold a similar view, for he speaks of the faculties as lines which converge at a single center. (Kahn, ‘Sensation and Consciousness in Aristotle's Psychology,’ in Archiv für Geschichte der Philosophie 39 (1966) 43–81Google Scholar).

30 Since, for Aristotle, the faculty as actualized is one with its object, one might be inclined to deny the possibility of its being simultaneously white and black. But we need only recall Aristotle's empiricism to realize that he could not have intended to deny the obvious fact that our visual field at any given moment is an array of colors. Any representation of this situation would involve picturing the faculty of sight as divided in some way. Thus, our interpretation on which one faculty is represented by a number of intersecting lines is consistent with Aristotle's view.

31 The many pronouns and variables in this passage make any interpretation highly speculative. Questions of the sort — what is the referent of tauta (a22)? of hekateron (a23)? of to and to (a25)? — have no definite answers.

32 The perception of the relevant special objects kath'hauta is a necessary condition; this perception may be, and typically is, occurrent, but sometimes the perception of one of the objects kath'hauto is remembered. In the latter case, the resulting perception of the unity of the object or of differences among its qualities is kata sumbebēkos.

33 See Hamlyn, op. cit. (1), 129.

34 Cf. Sorabji, R. ‘Aristotle on Demarcating the Five Senses,’ Philosophical Review, 80 (1971).Google Scholar

35 Seen. 32 above.

36 It is clear from Aristotle's discussion of time in the Physics that the perception of motion involves the awareness of temporal succession:

37 This is not inconsistent with Aristotle's speaking of the incidental perception of the common objects by individual senses. Sight qua sight perceives motion incidentally but perceives it kath'hauto qua one of the senses meeting in hē koinē aisthēsis.

38 Cf. de An. 425b5-9.

39 I would like to thank Mark Kulstad, Ernie Loevinsohn, Bill Tait, Ian Mueller, Charles Kahn and Roger A. Shiner for their helpful criticisms and suggestions.