Changing the cartesian mind: Leibniz on sensation, representation and consciousness

Philosophical Review

Volumn 110, Issue 1, 2001, Pages 31-75

  Simmons, Alison ^a  

^a HARVARD UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 60950544138     PISSN: 00318108     EISSN: 15581470     Source Type: Journal    
DOI: 10.1215/00318108-110-1-31     Document Type: Review

References (105)

1. References to Leibniz's major works and letters are given in the text and notes according to the abbreviations and short titles listed in the Appendix. Translations are my own, but I have benefited from those listed in the Appendix.
2. Güven Guzeldere, "Introduction," in The Nature of Consciousness, ed. Ned Block, Owen Flanagan, and Güven Güzeldere (Cambridge: MIT Press, 1997), 18.
3. Guzeldere G., 1997 "Introduction," The Nature of Consciousness Ibid., 19.
4. Sir William Hamilton, Lectures on Metaphysics and Logic (Boston: Gould and Lincoln, 1860), 1: 251.
5. These included not only Hamilton, but also Eduard Beneke, Gustav Fechner, Johann Herbart, George Lewes, Henry Maudsley, and Hermann Ulrici. See Franz Brentano, Psychology from an Empirical Standpoint, ed. Oskar Kraus, trans. Antos Rancurello, D.B. Terrell, and Linda McAlister (New York: Routledge, 1995; originally published Leipzig, 1874), 1.3.6 and 2.2.
6. French commentators have been more attentive to the innovations of Leibniz's philosophy of mind, and in particular to Leibniz's representational- ism, than their Anglo-American counterparts. See, for instance, Jacques Jalabert, "La Psychologie de Leibniz," Revue Philosophique de la France et de I'É- trangerim (1946): 453-72.
7. References to the works of Descartes, La Forge, Malebranche, and Arnauld are given in the text and notes according to the abbreviations and short titles listed in the Appendix. Translations are my own, but I have benefited from those listed in the Appendix.
8. See Descartes, AT 8A:25;
9. La Forge, Traité 6/35-36, 54/57, 57/58, and 96/76;
10. Malebranche, RVZ-l. 1:380-83/198-99;
11. Arnauld, W735.
12. See Descartes, AT 7:28 and 176, AT 8A:17, AT 11:224 and 342-43;
13. La Forge, Traite 14/39, 56-57/58, 74-90/67-73;
14. Malebranche, RV 3-2.1:415/218;
15. Arnauld, VFI8.
16. For an illuminating discussion of the language, nature, and role of consciousness as it develops among the Cartesians, see Geneviève Lewis, Le prob- leme de I'inconscient et le cartésianism (Paris: Presses Universitaires de France, 1950), 112ff.
17. On memory, see Descartes, AT 3:425 and AT 4:114;
18. La Forge, Traite'85/7;
19. Malebranche, RV 2-1.5:224-29/106-109. On innate ideas
20. see Descartes, AT 7:189 and AT 8B:166-67; La Forge, Traité 143-44/96. Malebranche rejects the doctrine of innate ideas altogether (see 3-2.4, 429-32/226-27);
21. Arnauld has little to say about it.
22. See Descartes AT 3:423, AT 5:222 and AT 7:246, and La Forge, Traite 62/60, 271-72/154 and 351-54/186-87. Arnauld and Malebranche have less to say about these phenomena.
23. My distinction between phenomenal and reflective consciousness is similar to a distinction Tad Schmaltz draws between "direct consciousness" and "reflective consciousness" on behalf of La Forge, Arnauld, and Malebranche in his Malebranche's Theory of the Soul: A Cartesian Interpretation (New York: Oxford University Press, 1996), 21. Schmaltz argues that this distinction is not to be found in Descartes, and that, according to Descartes, consciousness comes in only at the reflective level by a second-order act of the intellect. This seems to me a misreading, for the thoughts of which Descartes suggests we are reflectively conscious (pains) are described as being pre-reflectively felt, and so are surely conscious. Reflection adds something: the recognition that this pain is one that I have not felt before. Rodis-Lewis similarly argues for a distinction between consciousness and reflective consciousness in Descartes in Leproblème de I'inconscient et le cartesienisme, 41-42.
24. Descartes makes the same distinction with slightly different terminology at AT 7:8 and 232. The term 'idea' is further complicated by the fact that Descartes sometimes uses it to refer to corporeal images traced in the brain (AT 10:416-17 and 11:176-77).
25. Paul Hoffman has recently argued that feelings like fear are themselves representational, in 'Three Dualist Theories of the Passions," Philosophical Topics 19 (1991): 153-200.
26. Fear, he argues, represents the badness of its intentional object. Margaret Wilson defends a similar position in "Descartes on the Representationality of Sensation," in Central Themes in Early Modern Philosophy, ed. J. A. Cover and M. Kulstad (Indianapolis: Hackett, 1990), 1-22.
27. I think this reading is unlikely, for (a) Descartes sometimes attributes the representation of something as good or bad to internal sensations (in particular, pleasure and pain) rather than to emotions (AT 11:400), and (b) it fails to provide an account of the differences among the emotions, all of which concern the goodness and badness of their intentional objects.
28. Arnauld's distinction is in keeping with a long-standing ambiguity in the Latin verb repraesentare, which is sometimes used to suggest something's standing in the place of something else by imitation or resemblance, but is more often used to suggest something's being presented or exhibited or shown.
29. See Leibniz's letter to Queen Sophie Charlotte, G 6:506/L 551; PNG§4; MS19 and § §29-30;
30. and his letter to R. C. Wagner, G 7:529.
31. Strictly speaking, it is not simply perception that constitutes the monadic essence, but perception and appetition-the active principle of change that propels a monad from one perceptual state to another. Although a thorough account of Leibniz's philosophy of mind would have to include a discussion of appetition, it is beyond the scope of this paper. See, however, note 19 for a bit more on this topic.
32. What of appetition? The monadic essence, recall (n. 18), includes both perception and appetition. Appetition, he tells us, is the "action of the internal principle that brings about the change or passage from one perception to another" (M §15) so that all one can find in a monad are "perceptions and their changes" (M §17). Leibniz might have in mind that there are no non- representational states in the monad, appetition being just the change from one representational state and another. Alternatively, Leibniz might argue that appetitions ultimately derive from the representational content of the monad's first perception, for, as Leibniz frequently suggests, all the subsequent perceptions of a monad are in some way contained in and determined by its original perception (see DM §14 and §29 and M §22). If that's true, then one could in principle derive each appetition, each change of perception, from a monad's original perception. And if that's the case, then it seems reasonable to say that the monad is essentially and fundamentally representational. I am indebted to Michael Delia Rocca (personal communication) for a helpful discussion of this matter.
33. Leibniz discusses the notion of expression further in his essays "What is an Idea," G 7:263-64.
34. which is translated in Philip Wiener's Leibniz Selections (New York: Scribners, 1979), 281-83
35. "Metaphysical Consequences of the Principle of Reason," C 15, which is translated in G. H. R. Parkinson's Leibniz: Philosophical Writings (London: Everyman, 1973), 176-77.
36. See also "A Specimen of Discoveries About Marvellous Secrets," G 7:317;
37. letter to Des Bosses, G 2:311;
38. letter to Arnauld, G 2:114/L 340.
39. For helpful discussions of Leibnizian perception, see Mark Kulstad, "Some Difficulties in Leibniz's Definition of Perception," in Leibniz: Critical and Interpretive Essays, ed. Michael Hooker (Minneapolis: University of Minnesota Press, 1982), 65-78;
40. Robert McRae, Leibniz: Perception, Apperception, and Thought (Toronto: University of Toronto Press, 1976), chap. 3;
41. and Margaret Wilson, "Confused versus Distinct Perception in Leibniz: Consciousness, Representation, and God's Mind" in Minds, Ideas and Objects, ed. Phillip Cummins and Günter Zöller, vol. 2 of the North American Kant Society Studies in Philosophy (Atascadero, Calif.: Ridgeview, 1992), 135-50.
42. Although we sometimes speak of linear perspective drawings and paintings as having a single point of view, I think this has to be understood as elliptical for the painting's representing something as seen from a single point of view.
43. In claiming that the mind is first and foremost a representing thing, Leibniz is clearly indebted to Spinoza. Leibniz's suggestion that what is distinctive about mental representation is the simplicity of its subject, however, directly contradicts Spinoza's claim that the mind is a composite collection of ideas {Ethics 2pl5). Because Leibniz's account is more explicit and more fully developed than Spinoza's, and because Leibniz is more insistent on the anti- Cartesian thrust of this conception of the mind, his theory of mind lends itself particularly well to studying the attempt to change the Cartesian conception of mind. For an extensive investigation into the representational conception of mind in Spinoza.
44. see Michael Delia Rocca, Representation and the Mind-Body Problem in Spinoza (New York: Oxford University Press, 1996).
45. For convenience, I employ the Cartesian terms 'mind' and 'thought' rather than the Leibnizian terms 'monad' or 'mental substance' and 'perception', as Leibniz himself does confronting the Cartesians and Lockeans in the passages cited.
46. There are, of course, two other possible responses. The Cartesians respond that, despite appearances, the mind is always consciously thinking;
47. we simply do not always remember our thoughts after the fact. Locke responds by denying that thought constitutes the essence, or essential activity, of the mind: conscious thought is a distinctive ability or operation of the mind;
48. but an ability need not be exercised at all times (Essay concerning Human Understanding, 2.1.10).
49. Leibniz thinks it untenable that consciousness survives the death of the body, as it must if the Cartesian mind is to be immortal. He argues that it is more likely that representation survives the death of the body. If the mind is essentially a representing thing, rather than an essentially conscious thing, the argument goes, there is no reason to suppose that the mind cannot survive the death of the body (PNG §4, M §14 and NE 2.19.4, A/RB 162). Although this argument is not likely to be philosophically compelling today, it does reveal an important theological motivation behind Leibniz's re-conceptualization of the mind.
50. See NE 2.1.19, A/RB 118. This argument depends on conceiving consciousness as a form of second-order reflection on the mental state that is said to be conscious. Thus, in order for mental state A to be a conscious mental state, I must have a reflective thought B of mental state A; but then if mental state B is to be conscious too, I must have a further reflective thought C of it;
51. and so on ad infinitum. While Leibniz takes this as decisive against the Cartesians, it misses its target, for the Cartesians do not, by and large, adopt a second-order theory of consciousness. For an interesting discussion of this argument, see Mark Kulstad, 'Two Arguments on Petites Perceptions," Rice University Studies 63 (1974): 57-68.
52. See NE Preface, A/RB 54; 2.1.13-15, A/RB 115-16;
53. 2.20.6, A/RB 164-65;
54. and 2.21.5, A/RB 173.
55. Leibniz's pre-established harmony is surely indebted to Spinoza's representational parallelism (Ethics 2p7 and 2pl2). The metaphysical details and arguments, however, are quite different. Moreover, while his exhaustive parallelism, according to which everything that happens in the body has a corresponding idea, suggests a commitment to unconscious perceptions, Spinoza is not explicit about this and provides no argument for it. Nor does he advance any particular account of what consciousness is. Leibniz, by contrast, is not only explicit about the existence of unconscious perceptions; he also relentlessly argues for their introduction into the mind and provides explicit account of what consciousness is and why it turns up where it does.
56. These judgments include both judgments about the actual size, shape, and distance of objects and also projective judgments to the effect that there actually are objects out in the world that resemble what is sensorily presented (for example, that there is an apple that is red in just the way it appears to be). For Descartes's fullest discussion of the distinction between sensation and sensory perception see his Sixth Replies, AT 7:436-69. For parallel discussions.
57. see Malebranche, RV 1.10:130/52 and 1.14:155-61/67-70;
58. La Forge, Traite 273-74/154-55; Arnauld, VFI253-54 and Logique 1.11:83-85/58-60.
59. There is interpretive dispute whether there are sensations proper of primary qualities. While I believe the Cartesians, with the exception of Malebranche, clearly commit themselves to primary quality sensations, this is not the place to provide a defense of this claim, and the points I want to make are clearest for the other sensations.
60. RV 1.13:145/62;
61. see also 3-2.7:452/238. La Forge writes: "we can very well say what the idea of round or square represents to us, but we cannot likewise explain what the idea of heat or any other sensible quality makes us perceive" [Traite 278-79/156-57).
62. See Descartes, AT 7:87 and AT 11:176;
63. La Forge, Traite 226/134 and 280/157;
64. Malebranche, RV 1.5:70-71/20 and 1.13:145/62;
65. and Arnauld, VFI 163 and 258, Logique chap. 11, 84/59.
66. Descartes, AT 11:5-6, AT 6:85 and 112-13, AT 8-A:320-21;
67. La Forge, Traité 276-77/155-56;
68. Malebranche, RV 1.12:141-42/59-60;
69. Arnauld, VFI 244Ab, 248.
70. Alternatively, sensations might be obscure and confused in the sense that they are mistaken by us to represent something bodily when in fact they do not. On this line of interpretation, it is not that sensations are obscure and confused representations, but rather that we obscurely and confusedly treat them as representations of bodily phenomena.
71. There has been considerable scholarly debate in recent Anglo-American literature whether Cartesian sensations are representational (and if so, how). For a defense of the view that they are nonrepresentational, see Ann Wilbur MacKenzie, "Descartes on Sensory Representation: A Study of the Dioptrics," Canadian Journal of the History of Philosophy, supp. vol. 19 (1990): 109-47.
72. For an argument that they represent only mental phenomena, see Alan Nelson, 'The Falsity in Sensory Ideas: Descartes and Arnauld," in Interpreting Arnauld, ed. E. J. Kremer (Toronto: University of Toronto Press, 1996), 13-32
73. For defenses of the view that Cartesian sensations represent (or perhaps misrepresent) corporeal phenomena, see Lilli Alanen, "Sensory Ideas, Objective Reality, and Material Falsity," in Reason, Will, and Sensation: Studies in Descartes's Metaphysics, ed.John Cottingham. (New York: Oxford University Press, 1994), 229-50;
74. Martha Bolton, "Confused and Obscure Ideas of Sense," in Essays on Descartes' Meditations, ed. Amélie Rorty (Berkeley: University of California Press, 1986), 389-404;
75. Paul Hoffman, "Descartes on Misrepresentation," Journal of the History of Philosophy 34 (1996): 357-81;
76. Alison Simmons, "Are Cartesian Sensations Representational?" Nous 33 (1999): 347-69;
77. Margaret Wilson, "Descartes on the Representationality of Sensation," 1-14.141-42/59-60;
78. Arnauld, VFI244-45, 248.
79. Thanks to Sean Greenberg for persistently pressing me on this point.
80. See Wilson, "Descartes on the Representationality of Sensation," 7-8.
81. Leibniz also claims that sensation involves attention and memory ("[On the Souls of Animals]," G 7:330;
82. NE Preface, A/RB 54; PNG §4, G 6:600/AG 208). I will focus, however, on distinctness.
83. It should be said that Leibniz does offer a straightforward definition of the term 'distinct' in his writings. A distinct concept or idea, he claims, is one that (a) enables its possessor to recognize instances of it and (b) enables its possessor to explain why something counts as an instance of it by offering an analysis of the concept's distinguishing marks (see, for example, Meditations on Knowledge, Truth, and Ideas, G 4:423/AG 24). The cognitive activities in terms of which distinctness is defined here, however, only make sense within the life of a mind or substance that is able to reflect, form concepts, and reason. Indeed, the definition is given for concepts or ideas, not for perceptions generally. Sensations, however, are had by creatures who have no concepts or ideas. The notion of distinctness that is used to distinguish sensation from other lower forms of perception therefore cannot be the one that Leibniz explicitly defines in these passages. Unfortunately, he never properly defines the term 'distinct' as it is used in these contexts, so interpretive work is needed to determine its meaning
84. There is a parallel connection between perceptual indistinctness or confusion and unconsciousness (see NE Preface, A/RB 53 and 55).
85. Or perception that p. As with most of the early moderns, Leibniz does not make a careful distinction between object perception (perception of a tree) and perception of states of affairs (perception that there is a tree outside).
86. It is a matter of considerable interpretive controversy just what this second-order perception (and so apperception) is supposed to be. One problem is that this and other passages make it sound as if second-order perception is performed by the faculty of reflection. Only rational minds, however, have reflection. If consciousness requires it, then only rational minds are capable of consciousness. Robert McRae defends this interpretive conclusion in Leibniz: Perception, Apperception, and Thought, chap. 3. In his late works, however, Leibniz clearly commits himself to animal consciousness (see NE 2.21.5, A/RB 173). Indeed, in this very passage Leibniz is trying to distinguish the normal (presumably conscious) mental states of animals from those they have when they are occasionally reduced to the state of lower monads (as when in a deep sleep or stupor), that is, unconscious perceptions. Animals, then, are attributed apperception or consciousness. Does this mean that we have to attribute some sort of proto-faculty of reflection to animals, contrary to Leibniz's explicit claim that reflection belongs to rational minds alone? Mark Kulstad defends this interpretive conclusion in Leibniz on Apperception, Consciousness, and Reflection (Munich: Philosophia Verlag, 1991). I am inclined to agree with Kulstad on this issue and claim that animals must have some capacity for reflection, although certainly not of the sort that rational human minds have, for Leibniz's commitment to animal consciousness seems to me difficult to dismiss in the late works.
87. Rocco Gennaro defends a related conclusion, that a special kind of nonreflective apperception is attributed to animals in "Leibniz on Consciousness and Self-Consciousness," in New Essays on the Rationalists, ed. Rocco Gennaro and Charles Huenemann (New York: Oxford University Press, 1999), 353-71.
88. One might object to my reading of this passage as follows: what Leibniz is distinguishing in PNG §4 is not unconsciou perception from conscious perception, but rather conscious perception from self-conscious perception (see, for example, Nicholas Rescher, The Philosophy of Leibniz (Englewood Cliffs, N.J.: Prentice-Hall, 1967), chap. 10).
89. The proper way to read PNG §4, on this view, is not as identifying apperception with consciousness full stop, but rather with consciousness of my internal states. In other words, "consciousness" is grammatically tied in this passage to the direct object "this internal state." This in an untenable suggestion. First, PNG§4 is concerned with distinguishing the mental states of lower monads from the more typical mental states of animals; the distinction he is after is surely between unconscious and conscious perception, not conscious and self-conscious perception. Second, although the French verb s'appercevoir de indeed takes a direct object, so that it is true that it would be impossible simply to say that I am aware without being aware of something, Leibniz rarely follows his innovative nominalization of that French verb, I'ap- perception, with a direct object, and so it is perfectly plausible that apperception is not grammatically joined to "this internal state" in the passage. Furthermore, in the Monadobgy passage that parallels PNG §4, Leibniz distinguishes perception from apperception "or consciousness [full stop]" (M §14).
90. It is not clear just how reflective second-order perception is supposed to introduce consciousness into one's perceptual life: if perceptions are not intrinsically conscious, then why think that adding more of them will produce consciousness? Second-order representations do not normally produce consciousness (for example, a photograph of a painting), so why should they in this case? One might argue on Leibniz's behalf that what is distinctive about second-order perception is that it is reflective and so represents not any old representation but a representational state of itself. That argument only goes so far: a painting like the Arnolfini Wedding that represents a room with a mirror reflecting some part of the room arguably includes a (literally) reflective second-order representation of a representational state of itself, but all without consciousness. At this point, I think Leibniz's only resource is to appeal to the difference between mental representation and corporeal representation generally: the simplicity of the representing subject that affords a single point of view. In the reflective second-order perception constitutive of consciousness, unlike other second-order representations, the perceiving subject must represent a representational state of herself as hers or as a state of the self-same subject. This need not require full-blown concept of the self, something had only by rational minds. If it did, then animal consciousness would be jeopardized. Leibniz maintains that memory serves this psychological unifying function among perceptions (see M§26) and that consciousness involves an "immediate memory" of a perception (see NE 2.27.13, A/RB 238). Consciousness, then, is not just the representation of a representation, but a memory of a representation had by this subject. For more on the connection between memory and consciousness, see Kulstad, Leibniz on Apperception, Consciousness, and Reflection, chap. 1. Whether successful or not, we have in Leibniz one of the first attempts to offer a reductive analysis of consciousness within the confines of a representationalist conception of mind.
91. Montgomery Furth presents such an interpretation in "Monadology, " Philosophical Review 76 (1967): 169-200;
92. reprinted in Leibniz: A Collection of Critical Essays, ed. H. Frankfurt (Notre Dame: Notre Dame Press, 1972), 99-136.
93. see esp. §4.
94. G. H. R. Parkinson advances such an interpretation in "The 'Intellectualization of Appearances': Aspects of Leibniz's Theory of Sensation and Thought," in Leibniz: Critical and Interpretive Essays, ed. Michael Hooker (Minneapolis: University of Minnesota Press, 1982), 3-20.
95. Cf, however, MJ2.19.1, A/RB 161, where Leibniz writes that sensation is "awareness of an external body" and that sleep is the "cessation of sensation.
96. See, for example, NE 2.9.1, A/RB 134 and Preface, A/RB 54. To the extent that perceptual distinctness is sensitive to context and perceiver motivation, we should say that it is context and perceiver relative. This fact makes it clear that it would be a mistake simply to identify perceptual distinctness with any single feature of a perception, such as its size, number, or variation. Size, number, and variation all conspire to ground perceptual distinctness. In the end, however, what it is for a perception to be distinct is not simply for it to have such and such a size, number or variation, but for those things to render the perception capable of being noticed-that is, capable of attracting a second-order perception-under the perceptual circumstances.
97. It is tempting to argue that surely some sensations, like pleasure and pain, are essentially conscious even by Leibniz's lights. Leibniz sometimes suggests such a view. He writes, for instance, that an unconscious monad "would be with-out pleasure and without pain" (NE 2.1.11, A/RB 113) and that the notion of pain includes consciousness (NE 2.21.36, A/RB 162). On the other hand, he explains why an unconscious monad would be without pleasure and pain by claiming not that these are essentially conscious or noticed perceptions, but that they are essentially notable or noticeable perceptions (perception notables) (NE 2.20.1, A/RB 162). What is more, he defines pleasure and pain as follows: "pleasure is a sensation of perfection, and pain a sensation of imperfection, each being notable enough that one can be conscious of them" (NE 2.21.36, A/RB 194). Furthermore, although he grants that hunger is a particularly notable or noticeable perception, "even when one is hungry, one does not think about the hunger all the time, but when one things about it, one is aware of it, for it is a very noticeable disposition" (NE 2.1.19, A/RB 118). It is not clear then that there are any sensations that are essentially conscious for Leibniz.
98. It is not crystal clear just how this is possible, but the burden of proof was not on Leibniz here. Most of his opponents, and especially the Cartesians, would have accepted the claim that the soul is simple despite its having a multitude of perceptual modifications.
99. Robert McRae attributes such a position to Leibniz: "As qualitatively simple these sensible perceptions are not aggregates of insensible perceptions. Rather, they are novel emergents from a mass of insensible perceptions" (Leibniz: Perception, Apperception and Thought, 38).
100. Although Leibniz does not have an efficient causal explanation for why sensations appear as they do, he does have a teleological story to tell about the reason for our having sensations that appear as they do. For an illuminating discussion of this story, see Martha Bolton, "The Explanation of Consciousness in Leibniz's Nouveaux Essais," manuscript.
101. William Lycan, for example, writes: "the mind has no special properties that are not exhausted by its representational properties, along with or in combination with the functional organization of its components. It would follow that once representation itself is (eventually) understood, then not only consciousness in our present sense but subjectivity, qualia, 'what it's like,' and every other aspect of the mental will be explicable in terms of representation together with the underlying functionally organized neurophysiology" (Consciousness (Cambridge: MIT Press, 1996) ,11).
102. See also David Armstrong, The Nature of Mind and Other Essays (Ithaca: Cornell University Press, 1980);
103. Fred Dretske, Naturalizing the Mind (Cambridge: MIT Press, 1996);
104. David Rosenthal, 'Two Concepts of Consciousness," Philosophical Studies 49 (1986): 329-59;
105. Michael Tye, Ten Problems of Consciousness (Cambridge: MIT Press, 1996). As the Lycan passage makes clear, however, many contemporary defenders of a representational theory of mind simultaneously defend some form of naturalism about the mind, which is, of course, quite contrary to Leibnizian metaphysics. The form of naturalism that is most often defended among these philosophers is functionalism, according to which mental states are type-individuated not by their intrinsic features but by their causal relations to other states. Nothing could be more contrary to Leibnizianism, for on Leibniz's view the apparent causal relations of a mental state are specified by the internal features of the mental state itself.