Portal de Periódicos da CAPES

Locke’s Problem Concerning Perceptual Error

2008; Wiley; Volume: 77; Issue: 3 Linguagem: Inglês

10.1111/j.1933-1592.2008.00216.x

1933-1592

Antonia LoLordo,

Philosophy and Theoretical Science

Both common sense and philosophers of many different schools endorse the claim that perception can be in error. In particular, Ideas of shape can be false. For perceptions of shape can falsely represent their objects. Consider a time-honoured example: when I look at a square tower from far away, it looks round. Thus, on a theory that holds that perception is a matter of having ideas, I have a false idea of the shape of the tower. Locke does not explicitly endorse this claim. Indeed, as has frequently been noted,11 See, for instance, Michael Ayers, Locke: Epistemology and Ontology (New York: Routledge, 1991), p. 166; G. A. J. Rogers, "John Locke and the Skeptics," in ed. Gianni Paganini, The Return of Skepticism (Dordrecht: Kluwer, 2003), 38ff; and Martha Bolton, "Locke on the Semantic and Epistemic Role of Simple Ideas of Sensation,"Pacific Philosophical Quarterly 85: 301–321, p. 301. See also Margaret Atherton, "'Suppose I Am Pricked With a Pin': Locke, Reid and the Implications of Representationalism,"Pacific Philosophical Quarterly 65: 149–165, p. 155, on Locke's neglect of the argument from illusion. his Essay Concerning Human Understanding22 John Locke, An Essay Concerning Human Understanding, edited by P. H. Nidditch (New York: Oxford University Press, 1975). All parenthetical references are to the Essay. contains surprisingly little discussion of perceptual error. However, given how obvious perceptual error is in everyday life, and given that Locke never denies that there is perceptual error, it is prima facie reasonable to assume that Locke recognizes its existence. Conjoining (1) with two other theses for which the Essay provides significant textual evidence and which are almost universally attributed to Locke raises a problem. For the triad consisting of (1) and the following two theses is inconsistent: Ideas of shape are simple ideas. All simple ideas are true—in fact, real, true and adequate. We shall see what Locke means by this in more detail below, but a brief note is sufficient to show that there is a prima facie conflict. Locke explains that "[w]henever the Mind refers any of its Ideas to any things extraneous to them, they are then capable to be called true or false… the Mind in such a reference makes a tacit Supposition of their Conformity to that Thing" (2.32.2). That is, an idea is true just in case it conforms to the external object it is referred to. Thus, all simple ideas conform to their objects. The inconsistency is obvious enough to make it unlikely that Locke accepted all three theses without ever noticing that they are jointly inconsistent. He must, it seems, have some way out of the inconsistent triad. But what? Three possibilities, corresponding to the denial or benign re-interpretation of each claim, come to mind: "Truth", "reality" and "adequacy" are to some extent technical terms for Locke, and "conform" is clearly a technical term. Thus, one might try to dissolve the inconsistency by understanding conformity in such a way that the idea of the square tower as round does conform to the square shape of the tower.33 The reader may suspect that the way ideas conform to objects is different for primary qualities and secondary qualities, and that it is more difficult for Locke to defend the claim that all ideas of primary qualities conform to their objects. This is why I have chosen error concerning a primary quality as my main example, although I discuss error concerning secondary qualities as well. One could deny that there actually is any such thing as perceptual error. The most plausible way to do this is to attribute all apparent perceptual error to judgment, thus claiming, for instance, that my idea of the square tower as round is not in error although any judgment that the tower is round would be.44 Michael Ayers, who is perhaps the most ardent exponent of reading Locke through the lens of Epicurus and Gassendi, suggests something like this reply in a slightly different context. Ayers, Locke, 1.166–167. This tactic has a long philosophical pedigree. It was a cornerstone of Epicurean canonic,55 Diogenes Laertius, Lives of eminent philosophers, edited and translated by R. D. Hicks (London: Heinemann, 1925), X.31–32 and X.50–52. (X.31–32 is Diogenes' account of Epicurean doctrine; X.50–52 is Epicurus'"Letter to Herodotus"). See also Lucretius, De Rerum Natura, edited by Cyril Bailey (Oxford: Clarendon Press, 1947), IV.469ff. and in the 17th century, Pierre Gassendi took it up and used it as a defense against skepticism.66 See canon I in the "Logica Epicuri" of Pierre Gassendi's Syntagma Philosophicum, in his Opera Omnia (Lyon: Anisson, 1658), 1.53a. Gassendi clearly has anti-skeptical motivations for holding that all appearances are true, but it is the subject of some dispute whether the same is true for Epicurus. See Stephen Everson, "Epicurus on the Truth of the Senses," in his Companions to Ancient Thought I: Epistemology (Cambridge: Cambridge University Press, 1990). Finally, one could deny that the idea of the round shape of the tower is actually a simple idea. Of course, since examples of apparent perceptual error can easily be found for other qualities and sense modalities, this strategy would have to be generalized to a wide variety of cases. Clues about how a suitably generalized strategy might work are provided by Locke's well-known claim, in response to the Molyneux problem, that "the ideas received by sensation, are often … altered by the judgment, without our taking notice of it" (2.9.8). I shall consider each of these three possible solutions in turn. First, however, I want to dispel a worry that might occur to some readers at this point. I remarked earlier that Locke never really discusses perceptual error. Thus, some readers may worry that Locke simply never noticed the existence of perceptual error, or at least never thought it had any philosophical significance, and hence should not be expected to have a response There are a number of reasons to think that Locke must have noticed the existence of perceptual error and entertained the possibility that it raises philosophically significant issues. First, as I stated above, perceptual error and its close cousin, perceptual relativity, are obvious features of everyday life. It is thus unlikely that Locke never noticed their existence. Second, the nature of Locke's project in the Essay suggests that he must have thought about perceptual error at some point. Locke's over-arching goal is to delineate "the original, certainty, and extent of human knowledge; together, with the grounds and degrees of belief, opinion, and assent" (1.1.2). It is surely relevant to that project to examine the circumstances under which we can arrive at true beliefs based on perception and the ways we can tell when we are in such truth-conducive circumstances. Someone who attempts to ground all knowledge and probability in the ideas received through sense should, surely, be concerned with cases where sensory ideas are misleading or false. Third, the tradition Locke wrote and was educated in was intermittently obsessed with perceptual error. Such concern was primarily, although by no means exclusively, connected with the alleged skeptical implications of such error. Works ranging from Descartes'1stMeditation to Sextus Empiricus's Outlines of Scepticism discussed standard skeptical tropes involving perceptual error, relativity and disagreement: the straight stick that looks bent in water, for instance, and the water that feels warm to one hand and cool to the other.77 René Descartes, Oeuvres de Descartes, edited by Charles Adam and Paul Tannery (Paris: J. Vrin, 1996), VII.118. Sextus Empiricus, Outlines of scepticism, translated by Julia Annas and Jonathan Barnes (Cambridge: Cambridge University Press, 2000), I.35ff. Indeed, Locke himself uses a number of these traditional examples to illustrate or develop88 There is some dispute over whether these are meant as arguments for a distinction between primary qualities and secondary qualities or simply as illustrations of that distinction. For the former, see Margaret Atherton, "Ideas in the Mind, Qualities in Body," in Ideas in Seventeenth Century Philosophy, edited by Philip Cummins and Günter Zoeller (Atascadero, CA: Ridgeview, 1993), 117–118. For the latter, see Peter Alexander, Ideas, Qualities, and Corpuscles (Cambridge: Cambridge University Press, 1985), 124. the distinction between primary qualities and secondary qualities.99 For an account of Locke's use of Pyrrhonian tropes in his discussion of primary and secondary qualities, see Martha Bolton, "Locke and Pyrrhonism: The Doctrine of Primary and Secondary Qualities," in ed. Myles Burnyeat, The Skeptical Tradition (Berkeley: University of California Press, 1983). The fact that the same glass of water may feel warm to one hand and cool to the other, for instance, illustrates or supports the thesis that heat is a secondary quality and that nothing like heat as we perceive it exists in the water (2.8.21). Fourth, Locke notices—and is concerned about—the fact that judgments based on perception can be in error. He thinks, for instance, that it is natural for humans to judge that our ideas of secondary qualities resemble something really in bodies even though this judgment is false: "Men can hardly be brought to think Sweetness and Whiteness are not really in Manna" (2.8.18; cf. 2.8.24). It would be very strange to be concerned about falsity in perceptual judgments without ever considering whether perception itself could be in error. Thus, there are good reasons to think Locke must have considered perceptual error and its significance. However, two reasons why Locke might not have considered it come to mind, and I would like to rebut them both. The first is that Locke might not discuss perceptual error because he thought that doing so would lead to a form of skepticism that is unprofitable and an unworthy subject of philosophical reflection. It is often said that Locke is simply not interested in skepticism, whether of the Pyrrhonian or the Cartesian variety, or that he does not take it seriously.1010 Richard Aaron, John Locke (Oxford: Clarendon Press, 1971), 240. Peter Alexander, Ideas, Qualities, and Corpuscles, 287–289. Nicholas Jolley, Locke: His Philosophical Thoughts (Oxford: Oxford University Press, 1985), 186ff. Roger Woolhouse, "Locke's Theory of Knowledge," in ed. Vere Chappell, Cambridge Companion to Locke (Cambridge: Cambridge University Press, 168–169). Thus, he might think that we should avoid those philosophical topics that give the skeptic a way in. However, it is simply not true that any attempt to address perceptual error would be fodder for the sort of skepticism about the external world mocked at 4.2.14 or for Pyrrhonian suspension of judgment about the qualities of things. One might instead, for instance, respond to perceptual error by providing a detailed account of the way that reason can correct the senses or that the senses can correct each other. Alternately, one might discuss perceptual error in strictly naturalistic terms. Thus, it is implausible that Locke does not discuss perceptual error because he thought any attempt to do so would lead him towards skepticism about the external world or the qualities of the bodies in it. The second is that Locke might not address perceptual error because he thinks it is so easily corrected as to be trivial. Now it is true that Locke thinks the senses are self-correcting and self-reinforcing: Our Senses, in many cases bear witness to the Truth of each other's report, concerning the Existence of sensible Things without us. He that sees a Fire, may, if he doubt whether it be any thing more than a bare Fancy, feel it too; and be convinced, by putting his Hand in it (4.11.7). However, this does not help dissolve our original inconsistency. Even trivially correctible errors in simple ideas are prima facie inconsistent with the thesis that all simple ideas are real, true and adequate. If we can understand reality, truth and adequacy in such a way that even ideas like the idea of roundness received from the square tower turn out to be real, true and adequate, then we can dissolve our inconsistency. To see whether this suggestion works, we need to examine what Locke means by claiming that all ideas are real, true and adequate and the arguments he offers in defense of this claim. I begin with reality. Ideas are real, on Locke's definition, if they are "such as have a Foundation in Nature; such as have a Conformity with the real Being, and Existence of Things, or with their Archetypes" (2.30.1). For real ideas are "designed to be the Marks, whereby we are to know, and distinguish Things, which we have to do with" (2.30.2). The notion of conformity Locke has in mind here is relatively weak. Ideas can serve as marks, and hence conform or correspond, whether they "be only constant Effects, or else exact Resemblances of something in the things themselves: the reality lying in that steady correspondence, they have with the distinct Constitutions of real Beings" (2.30.2). The two options correspond to the distinction between primary and secondary qualities: "the Ideas of primary Qualities of Bodies, are Resemblances of them, and their Patterns do really exist in the Bodies themselves" (2.8.15) while an idea of a secondary quality is "a bare effect of power" (2.8.25). Thus, Locke claims that ideas succeed as marks of primary qualities just in case they resemble that quality, and they succeed as marks of secondary qualities just in case they are "constant effects" of the relevant quality. I assume that the account of ideas as marks of their causes applies to ideas of both primary and secondary qualities, although Locke does not state this explicitly. Hence, ideas of primary qualities should also be "constant effects" of something in bodies, namely the quality they resemble. In other words, an idea of a primary quality must resemble a quality of the body that caused that idea—and not some other, causally irrelevant, body—in order to be real. If a simple idea is real, it is thereby also true and adequate. An idea is true, in the relevant sense,1111 I discuss the various senses of truth below. if the implicit judgment accompanying it—the judgment that the idea corresponds to its object—is true (2.32.4). In this sense, all simple ideas are true because they "must be suitable to those Powers, [God] has placed in external Objects, or else they could not be produced in us" (2.32.14). Ideas are adequate just in case they "perfectly resemble those Archetypes, which the Mind supposes them taken from; which it intends them to stand for; and to which it refers them" (2.31.1). All simple ideas are adequate because "being nothing but the effects of certain Powers in Things, fitted and ordained by GOD, to produce such Sensations in us, they cannot but be correspondent, and adequate to those Powers" (2.31.1). An idea is true and adequate, then, if it corresponds to the quality that produced it, and this correspondence is also what makes it real.1212 This is more complicated for complex ideas, and complex ideas can be real without being true and adequate. There are several different ways to understand how an idea can serve as a mark of its cause:1313 Compare Martha Brandt Bolton, "Locke on the Semantic and Epistemic Role of Simple Ideas of Sensation,"Pacific Philosophical Quarterly 85.3: 301–324, p. 308. An idea of an object is real just in case that particular idea-token was caused by that particular object-token; An idea of an object is real just in case that type of idea is always caused by that type of object; or An idea of an object is real just in case that type of idea is caused by that type of object under normal conditions. Which of the three does Locke have in mind? Two considerations speak in favor of (i). First, Locke often suggests that it is trivially or tautologously true that simple ideas are real, writing, for example, that simple ideas "must be suitable to those Powers [God] has placed in external Objects, or else they could not be produced in us" (2.32.14). The claim that all simple ideas correspond to their causes in virtue of having been caused by them is tautologously true only as a claim about particular ideas and particular causes. Second, Locke's discussion of sensitive knowledge offers some—though limited—support to the token-token reading. The general thrust of Locke's discussion is that sensitive knowledge is very strictly limited. When I write with ink on paper, I have "a certainty as great, as human nature is capable of" that "I see white and black, and that something really exists, that causes that sensation in me" (4.11.2). However, I do not know that the paper existed a second ago and I do not have any idea what it is like in itself (4.11.9). This at least suggests that sense perception does not inform us about stable patterns. I said that Locke's discussion of sensitive knowledge offers the token-token reading only limited support. For the ink example continues as follows: … I have … the idea … I call white; by which I know that that quality or accident (i.e. whose appearance before my eyes, always causes that idea) doth really exist" (4.11.2, bold mine). The same passage thus both supports and contradicts the weak, token-token reading of correspondence. Locke's discussion of sensitive knowledge does not provide clear evidence for this reading or any other. Locke's suggestions that it is trivially or tautologically true that simple ideas are real do constitute good evidence for a token-token reading. However, I think that on balance it is clear that Locke is making the stronger claim that idea-types correspond to quality-types in some way. It is easiest to see why this must be so for ideas of primary qualities.1414 In this context, it is worth noting that Locke's suggestions that it is tautologously true that simple ideas are real typically use secondary-quality examples. Primary-quality ideas resemble something in bodies. For instance, my ideas of the shape of square things resemble the shape of square things. But all my ideas of the shape of square things also resemble each other. Similarly, all the shapes of square things resemble each other. Thus, it seems, all my ideas of square shapes must resemble all the shapes of various square things.1515 I do not claim this is conclusive; there are certainly ways of understanding resemblance on which this would not go through. But I do think it is suggestive. The claim that primary-quality ideas resemble qualities really in bodies suggests a type-type reading of correspondence, at least for primary-quality ideas. Both textual and philosophical considerations speak in favor of the type-type reading for ideas of secondary qualities as well. Textually, the claim that simple ideas are "constant effects" bearing a "steady correspondence" with real beings clearly supports a type-type reading (2.30.2). The claim that simple ideas are effects of powers "ordained by our Maker, to produce in us such sensations" (2.30.2) could only fit into the token-token reading by ascribing a very odd view of divine action to Locke. And Locke entertains the possibility of "all Things, that had the texture of a Violet, producing constantly the Idea" that other men call marigold, (2.32.15), which suggests he thinks that in the normal case things with that texture are "producing constantly" the idea of purple. Philosophically, the claim that all simple ideas are true, real and adequate could not bear the epistemic weight Locke wants it to bear on the trivial, token-token reading. Let me explain what that epistemic weight is. Consider the case of the man who regularly is caused to see yellow by things that cause others to see blue, and who calls those things "blue." Locke explains that his idea of that color is adequate because he would be able as regularly to distinguish Things for his Use by those Appearances … as if the Appearances, or Ideas in his Mind, received from those two Flowers, were exactly the same, with the Ideas in other Men's Minds (2.32.15). In this passage, Locke explains that the idea-type that steadily corresponds with an object-type in my mind need not be the same one that steadily corresponds with that object-type in your mind. The correspondences need not be the same across different perceivers. However, they do need to be stable for each perceiver if they are going to allow us to "distinguish Things for [our] use." Sticking just to ideas of secondary qualities, it seems clear that we would not be able to get around in the world successfully if the same violet that sometimes produced the idea of purple other times produced the idea of green, of the smell of gasoline, or of the taste of soap. For the only way in which secondary-quality ideas tell us about the qualities of things is via patterns of correspondence. To know that snow has a quality such as to produce an idea of whiteness in me tells me nothing (other than that an external world exists) if it does not tell me about the relations between that quality and qualities of previously encountered bodies. The sweetness of sugar, to choose another example, helps me distinguish it as edible only because previous sweet things have been edible. The claim that simple ideas serve as guides for everyday life is central to the Essay. God has given us faculties that suffice for "the comfortable provision for this life and the way that leads to a better" (1.1.5). Our sensory faculties are chief among those faculties, and the sensory faculties deliver simple ideas, singly and in combination. Locke must think that simple ideas mark their objects in something more than the trivial, tautological sense if he is to hold that the various senses are useful to us. I think it is clear, then, that Locke has some sort of type-type correspondence in mind. But is what he has in mind really (ii)? Surely, he would have noticed that the existence of perceptual error cannot be squared with the claim that idea-types are invariably caused by quality-types. This line of thought leads one to (iii): an idea serves as a mark of a quality (and hence is real) just in case ideas of that type are normally caused by qualities of that type. Of course, it is notoriously difficult to explicate what 'normal conditions' are. But surely, it would be better for Locke to rely on a badly explicated notion of normal conditions than to simply ignore the possibility of something as obvious as perceptual error. An account of marking in terms of normal conditions has a great deal of initial plausibility, but, unfortunately, that plausibility cannot survive closer examination. We need to be careful here. It is plausible that Locke would explain the intentionality of simple ideas in terms of what qualities cause them under normal conditions. What makes my idea of the tower seen from a distance the idea of something round is, plausibly, that it is the sort of idea typically gotten from round things. Even more plausibly, what makes my idea of the color of a tomato the idea of red is that it is the sort of idea typically received from things with certain primary qualities. But if what makes my idea the idea of roundness is that such ideas are normally caused by round things, and this idea was not caused by a round thing, then the conclusion that my idea misrepresents the shape of the tower seems inescapable. If an idea marks an object just in case ideas of this type are typically caused by objects of that type, then the idea of the square tower as round will serve as a mark of something round, thus marking something other than its cause and failing to be real. Any account of marking in terms of normal or ideal conditions will lead to the conclusion that only the ideas received in such conditions need be real, and this is incompatible with Locke's claim that all simple ideas are real. So far, we have considered three possible ways to understand Locke's claim that simple ideas are real in virtue of serving as marks of their causes. Only the second—that an idea of an object is real just in case ideas of that type are always caused by objects of that type—is a viable interpretation. Locke's claim that "simple Ideas are all real" (2.30.2) is indeed a substantive claim, and one that is prima facie in conflict with the existence of sensory error. Given this prima facie conflict, it is helpful to consider the justification Locke offers for his claim that all simple ideas are real. He explains that "in simple Ideas… the mind is wholly confined to the Operation of things upon it; and can make to it self no simple Idea, more than what it has received" (2.30.2). Simple ideas are mere passive receipts. This is supposed to justify the claim that simple ideas "answer and agree to those Powers of Things, which produce them in our Minds" (2.30.2), which in turn is supposed to justify their reality.1616 In contrast, fantastical ideas—like the idea of a unicorn, the result of combining the ideas of a horse and a horn—always derive from some action of the human mind. This justification relies on Locke's over-arching conception of human beings as creatures endowed by God with the faculties necessary for the conduct of life and the preservation of our souls. Since simple ideas are received passively, we cannot misuse our faculties in obtaining them; and since those faculties are God-given, they cannot be misleading when used correctly. Locke is careful to explain how this is consistent with the fact that our senses are imperfect: [W]ere our senses altered, and made much quicker and acuter, the appearance and outward scheme of things would have quite another face to us; and, I am apt to think, would be inconsistent with our being, or at least well-being, in this part of the universe which we inhabit … if by the help of such microscopical eyes (if I may so call them) a man could penetrate further than ordinary into the secret composition and radical texture of bodies, he would not make any great advantage by the change, if such an acute sight would not serve to conduct him to the market and exchange; if he could not see things he was to avoid, at a convenient distance; nor distinguish things he had to do with by those sensible qualities others do… (2.23.12). This would be a natural place for Locke to discuss the fact that the senses occasionally mis-represent their objects as well as representing them less than perfectly, but no such discussion occurs. While predecessors such as Descartes took pains to explain how sensory error is consistent with the fact that God endowed us with senses fit for the preservation of the mind-body composite, Locke does no such thing.1717 See the 6th Meditation: Descartes, Oeuvres, 7.88–89. Simple ideas, then, are real because they are passive products of our God-given sensory faculties. Now that we understand what Locke means by claiming that all simple ideas are real and how he justified that claim, let us return to our example, the square tower that looks round. How does Locke's argument for the reality of simple ideas work in the case of the idea of roundness received from a distant square tower? On his definition of reality, can that idea count as real? It does not do very well in terms of resemblance: to the extent that I can make sense of an idea of roundness resembling something in the world at all, it seems that it would have to resemble a round thing. However, Locke's resemblance claim is notoriously tricky, and a number of scholars have suggested either that Locke gets himself into trouble by saying that ideas of primary qualities resemble something really in bodies or that the notion of resemblance must be understood in some highly non-literal way.1818 Jonathan Bennett, Locke, Berkeley, Hume: Central Themes (Oxford: Oxford University Press, 1971), p. 106, for instance, says that the best way to understand Locke's resemblance thesis is that an idea resembles a quality just in case the quality explains the production of the idea. A. D. Woozley, "Introduction" to An Essay Concerning Human Understanding (London: Collins, 1964), pp. 34–35, says that resemblance is just accurate representation. Thus, one might think the real issue is the resemblance thesis and not the existence of perceptual error. However, consider the second, more general way in which ideas mark their causes. Does the idea of roundness received in those circumstances have a law-like correspondence with the shape of the tower? It is hard to see that it does. After all, while square things may always produce ideas of roundness under certain very specific conditions, the all-important caveat "under certain conditions" appears nowhere in Locke's account. Nor, more impor